Hi! A TREATISE ON CHEMISTRY OF THE UNIVERSITY OF LIF( A TREATISE ON CHEMISTRI BT SIR H. E. ROSCOE, F.K.S. AND C. SCHOKLEMMEE, F.R.S. VOLUME I THE NON-METALLIC ELEMENTS Chymia, alias Alchemia et Spagirica, est ars corpora vel mixta, vet composita, vel aggregata etiam in principia sua resolvendi, aut ex principiis in talia com- binandi." STAHL, 1723 NEW EDITION COMPLETELY KEVISED BY SIR H. E. ROSCOE ASSISTED BY DRS. H. G. COLMAN AND A. HARDEN WITH THREE HUNDRED AND SEVENTY-FOUR ILLUSTRATIONS AND A PORTRAIT OF DALTON Eg&AED BY C. H. JEENS ' \ - THE OF . NEW YORK D. APPLETON AND COMPANY 1903 Authorized Edition. PREFACE TO THE FIRST EDITION IT has been the aim of the authors, in writing the present treatise, to place before the reader a fairly complete, and yet a clear and succinct, statement of the facts of Modern Chemistry, whilst at the same time entering so far into a discussion of Chemical Theory as the size of the work and the present transition state of the science permit. Special atten- tion has been paid to the accurate description of the more important processes in technical chemistry, and to the careful representation of the most approved forms of apparatus employed. As an instance of this, the authors may refer to the chapter on the Manu- facture of Sulphuric Acid. For valuable information on these points they are indebted to many friends both in this country and on the Continent. The volume commences with a short historical sketch of the rise and progress of chemical science, and a few words relative to the history of each element and its more important compounds prefaces the systematic discussion of their chemical properties. For this portion of their work, the authors wish here to acknowledge their indebtedness to Hermann Kopp's classical works on the History of Chemistry. In the part of the volume devoted to the description of the non-metallic elements, care has been taken to - O O f\ O vi PREFACE select the most recent and exact experimental data, and to give references in all important instances, as it is mainly by consulting the original memoirs that a student can obtain a full grasp of his subject. Much attention has likewise been given to the representation of apparatus adapted for lecture-room experiment, and the numerous new illustrations re- quired for this purpose have all been taken from photographs o'f apparatus actually in use. The fine portrait which adorns the title-page is a copy, by the skilful hands of Mr. Jeens, of a daguerreotype taken shortly before Dalton's death. MANCHESTER, July, 1877. PREFACE TO THE THIRD EDITION IN this new, completely revised and reprinted edition I have endeavoured to carry out the aims which were put forward in the preceding preface seventeen years ago. Deprived of the aid of my late friend and colleague, I have been fortunate in securing the help of two of the ablest of my former students, and to them I tender my thanks. How far we have succeeded in bringing this edition up to the level of the science of the present day, it will be for the public to judge. All I can say is that no pains have been spared to do so. H. E. EOSCOE. LONDON, September 29th, 1894. CONTRACTIONS EMPLOYED IN THIS PABT Am. Chem. Journ. = American Chemical Journal. Annalen = Liebig's Annalen der Chemie. Ann. Chim. Phys. = Annales de Chimie et de Physique. Ber. = Berichte der deutschen chemischen Gesellschaft Bull. Soc. Chim. = Bulletin de la Societe Chimique de Paris. Chem. Centr. = Chemisches Centralblatt. Chem. Zeit. = Chemiker Zeitung. Compt. Rend. = Comptes Rendues de 1' Academic des Sciences. Gazzetta = Gazzetta Chimica Italiana. Journ. Chem. Soc. = Journal of the Chemical Society. J. Pr. Chem. = Journal fur praktische Chemie. Journ. Soc. Chem. Ind. = Journal of the Society of Chemical Industry. Monatsh. = Monatshefte fur Chemie. Phil. Mag. = Philosophical Magazine. Phil. Trans. = Philosophical Transactions of the Royal Society. Pogg. Ann. = Annalen der Physik und Chemie (Poggendorf). Proc. Chem. Soc. = Proceedings of the Chemical Society. Proc. Roy. Soc. = Proceedings of the Royal Society. Rec. Trav. Chim. = Recueil des Travaux Chimiques des Pays-Bas. Wied. Ann. = Annalen der Physik (Wiedemann). Zeit. analyt. Chem. = Zeitschrift fiir analytische Chemie. Zeit. angew. Chem. = Zeitschrift flir angewandte Chemie. Zeit. anorg. Chem. Zeitschrift fiir anorganische Chemie. Zeit. phys. Chem. = Zeitschrift fiir physikalische Chemie. Zeit. physiol. Chem. = Zeitschrift fur physiologische Chemie. CONTENTS PAGE HISTORICAL INTRODUCTION 3 GENERAL PRINCIPLES OF THE SCIENCE 41 Properties of Matter . . . 41 Elementary and Compound Bodies ,....., 51 Properties of Gases , ,59 Relation of Volume to Pressure. Boyle's Law .... 59 Relation of Volume to Temperature. Dalton's Law ... 60 Kinetic Theory of Gases 60 Diffusion of Gases 63 Continuity of Gaseous and Liquid States of Matter ... 72 Liquefaction of Gases 74 Boiling Points of Liquids 85 Laws of Chemical Combination 86 Combination by "Weight 87 Combination by Volume 97 Experimental Methods for the Determination of Molecular Weights . 104 Aqueous Solutions 116 Chemical Nomenclature 118 THE NON-METALLIC ELEMENTS ........ 126 Hydrogen 129 Fluorine .142 Fluorine and Hydrogen 147 Chlorine 153 Chlorine and Hydrogen . . . 168 Bromine 188 Bromine and Hydrogen ,193 Chlorine and Bromine . . 198 Iodine 199 Iodine and Hydrogen 206 Iodine Compounds with other Halogens ,212 Oxygen ,214 The Oxides ,..,...,.., 234 Ozone ..... .235 Oxygen and Hydrogen . . ,244 Oxygen and Chlorine . 313 Oxygen and Bromine . . . 329 Oxygen and Iodine 331 CONTENTS Sulphur 336 Sulphur and Hydrogen , . 349 Sulphur and Chlorine 358 Sulphur and Bromine ..... ..... 362 Sulphur and Iodine 362 Sulphur and Fluorine 363 Sulphur and Oxygen 364 Selenium 423 Selenium and Hydrogen . . . 427 Selenium and the Halogens 428 Selenium and Oxygen 431 Selenium and Sulphur. 435 Tellurium 436 Tellurium and Hydrogen 439 Tellurium and the Halogens . 440 Tellurium and Oxygen 442 Tellurium and Sulphur 445 Nitrogen . 446 Nitrogen and Hydrogen 452 Nitrogen and the Halogens . . .477 Nitrogen and Oxygen 481 Nitrogen and Sulphur 516 Nitrogen and Selenium .516 Sulphonic Acids of Ammonia and Hydroxylamine .... 519 The Atmosphere 523 Phosphorus 549 Phosphorus and Hydrogen 566 Phosphorus and the Halogens 573 Phosphorus and Oxygen . . . , . . . . .581 Phosphorus and Nitrogen 609 Arsenic 612 Arsenic and Hydrogen . . . 616 Arsenic and the Halogens 619 Arsenic and Oxygen . 622 Arsenic and Sulphur 629 Arsenic and Selenium 632 Arsenic and Phosphorus 633 Boron 640 Boron and Hydrogen 644 Boron and the Halogens 645 Boron and Oxygen 650 Boron and Sulphur 656 Boron and Nitrogen 657 Boron and Phosphorus 657 Carbon . 658 Carbon and Hydrogen 690 Carbon and the Halogens .705 Carbon and Oxygen . 705 Carbon and Sulphur 732 Carbon and Nitrogen 747 CONTENTS xi PAGE Coal-Gas 769 Wood-Gas 787 Oil-Gas . 788 Water-Gas o 790 Nature of Flame . 792 Silicon or Silicium 800 Silicon and Hydrogen. 804 Silicon and the Halogens 806 Silicon and Oxygen . .814 Silicon and Sulphur . . . . 823 Silicon and Nitrogen 824 Silicon and Carbon 825 CRYSTALLOGRAPHY 827 COMPARISON OF METRICAL WITH ENGLISH MEASURES . . . .874 ADDENDUM 876 INDEX 877 CHEMISTRY CHEMISTRY HISTOEICAL INTEODUCTION IN looking back at the history of our Science, we find that although the ancient world possessed a certain empirical know- ledge of chemical facts derived chiefly from an acquaintance with pharmaceutical and manufacturing art, the power of con- necting or systematizing these facts was altogether wanting. The idea of experimental investigation was scarcely understood, and most of those amongst the ancients who desired to promote a knowledge of Nature attempted to do so rather by pursuing the treacherous paths of speculation, than the safe though tedious route of observation and experiment. They had no idea of the essential differences which we now perceive between elements and compound substances, nor did they understand the meaning of chemical combination. The so-called Aristotelian doctrine of the four elements, Earth, Water, Air or Steam, and Fire, bore no analogy to our present views as to the nature and pro- perties of the chemical elements, for with Aristotle these names rather implied certain characteristic and fundamental properties of matter than the ideas which we now express by the term chemical composition. Thus " Earth " implied the properties of dryness and coldness ; " Water," those of coldness and wet- ness ; " Air or Steam," wetness and heat ; " Fire," dryness and heat. All matter was supposed to be of one kind, the variety which we observe being accounted for by the greater or less abundance of these four conditions which were supposed to be essential to every substance, that which was present in the greatest degree giving to the substance its characteristic pro- perties. To men holding such views, a change of one kind of HISTORICAL INTRODUCTION matter into a totally different kind appeared probable and natural. Thus, the formation of water from air or vice versa is described by Pliny as a usual phenomenon seen in the formation and disappearance of clouds, whilst the ordinary experience, that cold acts as a solidifying and hardening agent, bears out Pliny's view, that rock crystal is produced from moisture, not by the action of heat, but by that of cold, so that it is, in fact, a kind of ice. A transformation of one sort of substance into another quite different thus appears not only possible but probable, and we are not surprised to learn that, under the influence of the Aristotelian philosophy, which throughout the middle ages was acknowledged to be the highest expression of scientific truth, the question of the trans- mutability of the base into the noble metals was considered to be perfectly open. Much light of a very interesting and remarkable character has been thrown upon the origin of alchemy, the artificial pro- duction of the noble metals, by the discovery of an Egyptian papyrus, which contains more than a hundred metallurgical recipes written in Greek, many of which consist of elaborate directions for the falsification of the precious metals. The con- nection between these working notes of a fraudulent Egyptian goldsmith and the dreams of the later alchemists is attested by the reappearance of several of these very recipes in the writings of the following century under the guise of formulae of transmutation. 1 The oldest works of a strictly chemical character date from about the beginning of the third century of our era, and are due to Greek authors resident in Egypt, where our science appears to have had its birth. Among these writers, the most ancient and distinguished of whom is known as Zosimus of Panopolis, the possibility of the transmutation of the metals is fully accepted, and their works consist mainly of directions for the achievement of this object, couched however in language so deeply tinged with religious mysticism, symbolism and metaphor as to be almost unintelligible. 2 They allude to their subject as the " divine art," and it is not until the fourth century that we find in the works of the Byzantine writers the term Chemia, applied to the art which treats of the production of gold and 1 Berthelot, Les Origines de I'Alchimie, p. 3. Introduction A V Etude de la Chimie des Anciens et du Moyen Age, pp. 21, 59. 2 Berthelot, Collection des Anciens Alchimistes Grecs. Part I. THE ARABIAN ALCHEMISTS silver. 1 The fact that all these authors were closely connected with the celebrated schools of Alexandria, the last resting-place of the proscribed secrets of the Egyptian priests, adds to the probability that our science was first practised in Egypt. Plutarch, indeed, states that the old name for Egypt was Chemia, and that this name was given to it on account of the black colour of its soil. The same word was used to designate the black of the eye, as the symbol of the dark and mysterious. It is therefore possible that chemistry originally meant simply the Egyptian or secret knowledge, whilst others identify the name with the Greek %fyu-o9, sap or liquid, the name of the agent of transmutation being applied to the art. The Aristotelian philosophy became known to and was ex- tended by the Arabians, who, in the year 640, overran Egypt, and thence, through Northern Africa, penetrated into Spain. They first became acquainted with chemistry in Egypt, and pre- fixed the Arabic article to the original name, so that the word alchemy has from that time been used to signify the art of making gold and silver. The works of Geber, the most celebrated of Arabian alchem- ists, are handed down to us through Latin translations. In these books we learn that the aim of the science of which Geber treats is the transmutation of the base into the noble metals. He describes many chemical operations, such as filtra- tion, distillation, crystallization, and sublimation; and by these he prepares new substances or purifies the old ones. Bodies such as alum, green vitriol, saltpetre, and sal-ammoniac are em- ployed ; and we find that he was able to prepare nitric acid, or aquafortis, and from it the valuable solvent for gold, aqua regia. It is probable that even sulphuric acid was known to Geber, and certainly a number of metallic compounds, amongst which were mercuric oxide and corrosive sublimate, the preparation of which he describes, were known. Geber was the first pro- pounder of a chemical theory. He asserts that the essential differences between the metals are due to the preponderance of one of two principles, mercury and sulphur of which all the metals are composed. The first principle is characteristic of the truly metallic qualities, whilst the latter causes the peculiar changes noticed when the metals are exposed to heat. The noble metals were supposed to contain a very pure mercury, and are, therefore, unalterable by heat, whilst the base metals contain 1 Kopp, Beitrdge zur Gcschichte der Chemie. 1 Stuck, p. 40. HISTORICAL INTRODUCTION so much sulphur that they lose their metallic qualities in the fire. These constituents may, however, not only be present in different proportions, but also in different degrees of purity or in different states of division ; and thus it might naturally be supposed that, if not by a variation in their relative quantity, at any rate by a change in their condition, such an alteration in the properties of one metal may be brought about as would produce from it some other known metal. Thus gold and silver contain a very pure mercury, which in the one instance is combined with a red and in the other with a white sulphur ; and he explains that the reason why these two metals amalgamate so easily is because they already contain a large quantity of mercury, and are therefore quickly attracted by the liquid metal. Whilst Greece and Italy sank deeper and deeper into bar- barism, arts and science flourished under Arabian dominion, and the academies of Spain were thronged with students from all parts of the Christian world. The knowledge of alchemy spread from this source over Western Europe, and in the thirteenth century we find alchemists of the Arabian school in all the chief countries of Europe. In Christian Spain lived the celebrated Raymund Lully ; in France we hear of Arnold Villanovanus ; Albertus Magnus flourished in Germany. Then Thomas Aquinas, pupil of Albertus, was also an alchemist, as was our own Roger Bacon (1214-1294), who was tried at Oxford for sorcery, and who, to disprove these charges, wrote the celebrated treatise, 1 in which he shows that appearances then attributed to supernatural agency were due to common and natural causes. It was Roger Bacon, from his rare accomplish- ments and learning termed Doctor Mirabilis, who first pointed out the possible distinction between theoretical alchemy, or chemistry studied for its own sake, on the one hand, and prac- tical alchemy, or the striving after certain immediately useful ends, on the other. Although all these men agreed that the transmutation of metals was not only possible but that it was an acknowledged fact, and that for the preparation of gold and silver the philoso- pher's stone was needed, it is difficult, not to say impossible, now to understand their methods or processes, inasmuch as all that they have written on this subject is expressed in the ambiguous and inflated diction of the Byzantine and Arabian authors. The fourteenth century finds the study of alchemy widely 1 Epistola de secretis operibus artis et naturae, et nullitate magice, - TRANSMUTATION spread over the civilized world, and the general attention which the subject attracted gave rise to the discovery of a large number of chemical substances. By the end of the fifteenth century, although the knowledge of chemical facts had continued to in- crease, the old views respecting the ultimate composition of matter were still accepted. In addition, however, to the sulphur and mercury, supposed by Geber to be the universal constitu- ents of matter, we find a third constituent, viz. salt, introduced. We must bear in mind however that these three principles, like the four Aristotelian elements, were not supposed to be identical with the common substances which bear their names. That men of such wide experience and great powers, as the chemists of this period proved themselves to be, could bring themselves to believe in the possibility of the discovery of the philosopher's stone, a substance of such potency that when thrown on the base metals in a state of fusion (moment of pro- jection) it transmutes them into gold and silver, appears to us very remarkable. No one doubted the possibility of such a transmutation, and the explanation may be found in the fact, at that time well known, that the colour of certain metals can be altered by the addition of other bodies. Thus Geber knew that when red copper is melted with tutty (an impure oxide of zinc), the golden-yellow brass is obtained ; and also that other minerals (those which we now know to contain arsenic) give to copper a silver-white colour. Still the difference between these alloys and the noble metals must soon have been discovered, and the possibility of the transmutation lay rather in the notion already alluded to, that the different metals contained the same constituents arranged either in different quantities or in different states of purity. Nor were experimental proofs of this view wanting. Thus Geber believed that by adding mercury to lead the metal tin was formed, and the solid amalgam does closely resemble tin in its appearance. Then again the metallurgical processes were in those days very imperfect, and the alchemists saw proof of their theory in the formation of a bead of pure silver from a mass of galena, or in the extraction of a few grains of gold out of a quantity of pyrites. It was not until the beginning of the seventeenth century that it was proved that galena frequently contains silver, and that traces of gold are often found in iron pyrites. Even so late as 1709 we find Homberg stating that pure silver after melting with pyrites is found to contain gold, and it was only after several chemists had HISTORICAL INTRODUCTION performed the experiment with a like result that the mineral itself was acknowledged to contain traces of gold. Again, it was not at this time recognised that some salts are metallic compounds, and the precipitation of copper from a solu- tion of blue-stone by metallic iron was supposed to be a transmu- tation of iron into copper. These apparent experimental proofs of the truth of the alchemical doctrine were accompanied by a mass of traditional evidence ; that is, of stories handed down from generation to generation, in which cases of the transmuta- tion of metals are circumstantially narrated. Thus the belief in the fundamental principle of alchemy became firmly established. 1 A satisfactory explanation of the belief in the power of the philosopher's stone to heal disease and to act as the elixir vitce, the grand panacea for human ills, is more difficult to find. It may possibly have at first arisen from a too literal interpretation of the oriental imagery found in the early Arabian writers, where, although the peculiar doctrine of elixir vitse is unknown, we find such passages as the following : " If thoti carriest out my prescription with due care thou shalt heal the bad disease of poverty." The Arabians called the base metals " diseased." Thus Geber says, " Bring me the six lepers, so that I may heal them ; " that is, transmute the other six known metals into gold. The belief in the healing power of the philosopher's stone was also much strengthened by the discovery, about this time, of many substances which produce remarkable effects on the human frame, and of these the alchemists of the thirteenth century write in the most exaggerated and exalted terms. The work known by the fantastic title of the Triumph- Wagen des Antimonii, which contains a large amount of accurate information concerning the preparation and the medicinal pro- perties of many of the compounds of antimony, and is ascribed to the authorship of a monk, Basil Valentine by name, who was supposed to have lived at the beginning of the fifteenth century, has been shown by the late Prof. Schorlemmer to be an undoubted forgery dating from about 1600, the information being culled from the works of other writers and thrown into the mystical, semi-religious style suitable to the earlier period. 2 The same appears to be true of the other writings attributed to this author. The man who effected the inestimable union between 1 For further information on this subject Kopp's classical work, Die Geschichte der Chemie, or Thomson's History of Chemistry, may be consulted. 2 See also Kopp, Beitrdge III. 110. UNIV THE MEDICAL CHEMISTS chemistry and medicine was Paracelsus (1493-1541). He not only assumed the existence of three components of all inorganic substances, but he was the first who included animal and vegetable bodies in the same classification, and held that the health of the organism depends on the continuance of the true proportions between these ingredients, whilst disease is due to a disturbance of this proper relation. The era thus inaugurated by Paracelsus continued up to the end of the seventeenth century. Chemistry was the handmaid of medicine, and questions respecting the ultimate composition of matter were considered of secondary importance to those relating to the preparation of drugs. Of the contemporaries of Paracelsus, Agricola (1490-1555) was one of the most dis- tinguished, and his remarkable work De Re Metallica contains a complete treatise on metallurgy and mining, which did much to advance the processes of technical chemistry, many of the methods which he describes being in use even at the present day. Whilst Agricola devoted himself to the study of metallurgy, his countryman, Libavius, greatly assisted the general progress of science, inasmuch as he collected together in writings which are characterised by a clear and vigorous style, all the main facts of chemistry; so that his Alchemia, published in 1595, may be regarded as the first handbook of chemistry. His chief object was the preparation of medicines, but he still maintained the science in its old direction and distinctly believed in the trans- mutation of metals. The first who formally declined to accept the Aristotelian doctrine of the four elements, or that of Paracelsus of the three constituents of matter, was Van Helmont (1577-1644). He denied that fire had any material existence, or that earth can be considered as an element, for it can, he says, be produced from water, but he admitted the elementary nature of air and water, and gave great prominence to the latter in its general dis- tribution throughout animate and inanimate nature. Van Helmont' s acknowledgment of air as an element is the more remarkable, as he was the first to recognise the existence of different kinds of air and to use the term gas. Thus, his " gas sylvestre," which he clearly distinguished from common air, is carbonic acid gas, for he states that it is given off in the process of fermentation, and also formed during combustion, and that it is found in the " Grotto del Cane," near Naples. He also mentions a " gas pingue " which is evolved from dung, and is 10 HISTORICAL INTRODUCTION inflammable. It was Van Helmont who first showed that if a metal be dissolved in an acid it is not destroyed, as was formerly believed, but can again be obtained from solution as metal by suitable means ; and he considered the highest aim of the science to be the discovery of a general solvent which would at the same time serve as a universal medicine, and to which the name of " alkahest " was given. Although Van Helmont accomplished much towards the overthrow of the Paracelsian doctrine, his discoveries of the different gases were forgotten, and even up to the middle of the seventeenth century much divergence in opinion on fundamental questions prevailed. Those who were interested in the connection of chemistry with medicine still believed in the dreams of the alchemist, and held to the old opinions ; whilst those who advancing with the times, sought to further the science for its own sake, or for the sake of its important technical applications, often upheld views more in accordance with those which we now know to be the true ones. Among the names of the men who, during this period, laboured successfully to pro- mote the knowledge of chemistry, that of Glauber (1603-1668) must be first mentioned. He was both alchemist and medicinal chemist, and discovered many valuable medicines. Another name of importance at this epoch is that of N. Lemery (1645-1715). He, as well as Lefebre and Willis, believed in the existence of five elements ; mercury or spirit, sulphur or oil, and salt are the active principles; water or phlegm, and earth are the passive ones. Lemery's ideas and teachings became well known through the publication of his GOUTS de Chymie (1675) which was trans- lated into Latin, as also into most modern languages, and exerted a great influence on the progress of science. In this work the distinction between mineral and vegetable bodies was first clearly pointed out, and thus for the first time the distinction between Inorganic and Organic chemistry was realized. Pre-eminent amongst the far-seeing philosophers of his time stands Robert Boyle (1627-1691). It is to Boyle that we owe the complete overthrow of the Aristotelian as well as the Paracelsian doctrine of the elements, so that, with him we begin a new chapter in the history of our science. In his Sceptical Chymist, 1 1 The Sceptical Chymist or Chemico- physical Doubts and Paradoxes, touching the Experiments whereby vulgar Spagyrists are wont to endeavour to evince their Salt, Sulphur, Mercury, to be the true Principles of Things. First published in 1661 (Boyle's Works, 1772, 1, 458). "THE SCEPTICAL CHYMIST " 11 he upholds the view that it is not possible, as had hitherto been supposed, to state at once the exact number of the elements ; that on the contrary all bodies are to be considered as elements which are themselves not capable of further separation, but which can be obtained from a combined body, and out of which the com- pound can be again prepared. Thus he states, " That it may -as yet be doubted whether or no there be any determinate number of elements ; or, if you please, whether or no all com- pound bodies do consist of the same number of elementary in- gredients or material principles." 1 Boyle, it is clear, was the first to grasp the idea of the distinction between an elementary and a compound body, the latter being a more complicated substance produced by the union of two or more simple bodies and differing altogether from these in its properties. He also held that chemical combination consists in an approximation of the smallest particles of matter, and that a decomposition takes place when a third body is present, capable of exerting on the particles of the one element a greater attraction than the particles of the other element with which it is combined. More, however, than for his views on the nature of the elements, is science indebted to Boyle for his clear statement of the value of scientific investigation for its own sake, altogether independent of any application for the purposes either of the alchemist or of the physician. It was Boyle who first felt and taught that chemistry was not to be the handmaid of any art or profession, but that it formed an essen- tial part of the great study of Nature, and who showed that from this independent point of view alone could the science attain to vigorous growth. He was, in fact, the first true scientific chemist, and with him we may date the commencement of a new era for our science when the highest aim of chemical research was acknowledged to be that which it is still upheld to be, viz., the simple advancement of natural knowledge. In special directions Boyle did much to advance chemical science (his published writings and experiments fill six thick quarto volumes), particularly in the borderland of chemistry and physics ; thus in the investigations on the " Spring of the Air," he discovered the great law of the relation existing between volumes of gases and the pressures to which they are subjected, which still bears his name. Although Boyle was aware of the fact that many metals when heated in the air form calces which weigh more than the metals 1 Boyle's Works, 1772, 1, 560. 12 HISTORICAL INTRODUCTION themselves, and although he examined the subject experimentally with great care, his mind was so much biassed by the views he held respecting the material nature of flame and fire that he ignored the true explanation of the increase of weight, namely, that it is due to the absorption of a ponderable constituent of the atmo- sphere, and looked upon the gain as a proof of the ponderable nature of fire and flame, giving many experiments having for their object the " arresting and weighing of igneous corpuscles." * Similar views are found expressed in his essay " On the mechanical origin and production of fixedness/' 2 written in 1675, where Boyle, speaking of the formation of mercuric oxide from the metal by exposure to the air at a high temperature, says, " chemists and physicians who agree in supposing this pre- cipitate to be made without any additament, will, perchance, scarce be able to give a more likely account of the consistency and degree of fixity, that is obtained in the mercury ; in which, since no body is added to it, there appears not to be wrought any but a mechanical change, and though I confess I have not been without suspicions that in philosophical strictness this precipitate may not be made per se, but that some penetrating igneous particles, especially saline, may have asso- ciated themselves with the mercurial corpuscles." We owe the next advances in chemistry to the remarkable views and experiments of Hooke (Micrographia, 1665), and of John Mayow (Opera Omnia Medico-physica, 1681). The former announced a - theory of combustion, which although it attracted but little notice, more nearly approached the true explanation than many of the subsequent attempts. He pointed out the similarity of the actions produced by air and by nitre or saltpetre, and he concluded that combustion is affected by that constituent of the air which is fixed or combined in the nitre. 3 Hooke did not complete his theory or give the detail of his experiments, but similar conclusions seem to have been independently arrived at by Mayow, who in 1669 published a paper, De Sal-Nitro et Spiritu Nitro-aereo, in which he points out that combustion is carried on by means of this " spiritus nitro-aereus " (another, and not an in- appropriate name for what we now call oxygen), and he also distinctly states that when metals are calcined, the increase of weight observed is due to the combination of the metal with this " spiritus." Mayow was one of the first to describe experiments. 1 Boyle's Works, 3, 706718. 2 Boyle's Works, 4, 309. 3 Micrographia, pp. 103 5. THE THEORY OF PHLOGISTON 13 made with gases collected over water, in which he showed that air is diminished in bulk by combustion, and that the respiration of animals produces the same effect. He proved that it is the nitre-air which is absorbed in both these processes, and that an inactive gas remains, and he drew the conclusion that respiration and combustion are strictly analogous phenomena. There is, therefore, no doubt that Mayow clearly demonstrated the heterogeneous nature of air, although his conclusions were not admitted by his contemporaries. Another theory which was destined greatly to influence and benefit chemical discovery, was advanced about this time by J. J. Becher (1635 1682), and subsequently much developed and altered by G. E. Stahl (16601734). It made special reference to the alterability of bodies by fire, and to the explanation of the facts of combustion. Becher assumed that all combustible bodies are compounds, so that they must contain at least two constituents, one of which escapes during combustion, whilst the other remains behind. Thus when metals are calcined, an earthy residue or a metallic calx remains ; metals are therefore compounds of this calx with a combustible principle, whilst sulphur or phosphorus are compounds contain- ing a principle which causes their combustion. Bodies unalter- able by fire are considered to have already undergone combus- tion ; to this class of bodies quicklime was supposed to belong, and it was assumed that if the substance which it had lost in the fire were again added a metallic body would result. The question as to whether there be only one or several principles of combustibility was freely discussed, and Stahl decided in favour of the former of these alternatives, and gave to this combustible principle the name Phlogiston ((frXoyio-ros, burnt, combustible). An example may serve to illustrate the reasoning of the upholders of the Phlogistic theory. Stahl knew that oil of vitriol is a product of the combustion of sulphur ; hence sulphur is a combination of oil of vitriol and phlogiston. But this latter is also contained in charcoal, so that if we can take the phlogiston out of the charcoal and add it to the oil of vitriol, sulphur must result. In order that this change may be brought about, the oil of vitriol must be fixed (i.e. rendered non- volatile) by combining it with potash ; if then, the salt thus obtained is heated with charcoal, a hepar sulphuris (a compound also produced by fusing potash with sulphur) is obtained. The argument shows that when charcoal is heated with oil of vitriol 14 HISTORICAL INTRODUCTION the phlogiston of the charcoal combines with the oil of vitriol, and sulphur is the result. The phlogiston contained in sulphur is not only identical with that contained in charcoal, but also with that existing in the metals, and in all organic bodies, for these are obtained by heating their calces with charcoal, or with oil or other combustible organic bodies. The amount of phlogiston contained in bodies was, according to Stahl, very small, and the greatest quantity was contained in the soot deposited from burning oil. It was likewise considered that the phlogiston given off by combustion is taken up again from the air by plants ; and the phenomena of fermentation and decay were believed to depend upon a loss of phlogiston which, however, in this case only escapes slowly. Stahl explains why combustion can only occur in presence of a good supply of air, because in this case the phlogiston assumes a very rapid whirling motion, and this cannot take place in a closed space. However false from our present position we see the phlogistic theory in certain directions to be ; and although we may now believe that the extension and corroboration of the positive views enunciated by Hooke and Mayow might have led to a recognition of a true theory of chemistry more speedily than the adoption of the theory of phlogiston, we must admit that its rapid general adoption showed that it supplied a real want. It was this theory which for the first time established a common point of view from which all chemical changes could be observed enabling chemists to introduce something like a system by class- ing together phenomena which are analogous and are probably produced by the same cause, for the first time making it pos- sible for them to obtain a general view of the whole range of chemical science as then known. It may appear singular that the meaning of the fact of the increase of weight which the metals undergo on heating, which had been proved by Boyle and others, should have been wholly ignored by Stahl, but we must remember that he considered their form rather than their weight to be the important and character- istic property of bodies. Stahl also, perhaps independently, arrived at the same con- clusion which Boyle had reached, concerning the truth of the existence of a variety of elementary bodies, as opposed to the Aristotelian or Paracelsian doctrine, and the influence which a clear statement of this great fact by Stahl and his pupils JOSEPH BLACK 15 amongst whom must be mentioned Pott (1692 1777) and Marggraf (1709 1782) exerted on the progress of the science was immense. It is only after Stahl's labours that a scientific chemistry becomes, for the first time, possible. The essential difference between the teaching of the science then and now being that the phenomena of combustion were then believed to be due to a chemical decomposition, phlogiston being supposed to escape, while we account for the same phenomena now by a chemical combination, oxygen or some element being taken up. Thus Stahl prepared the way for the birth of modern chemis- try. It was on August 1st, 1774, that Joseph Priestley dis- covered oxygen gas. Between the date of the establishment of the phlogistic theory by Stahl, and of its complete overthrow by Lavoisier, many distinguished men helped to build up the new science Black, Priestley, and Cavendish in our own country, Scheele in Sweden, and Macquer in France. The classical researches of Black on the fixed alkalis (1754) 1 not only did much to shake the foundation of the phlogistic theory, but they may be described with truth as the first beginnings of a quantitative chemistry, for it was by means of the balance, the essential instrument of all chemical research, that Black established his conclusions. Up to this time the mild (or carbonated) alkali was believed to be a more simple compound than the caustic alkali. When mild alkali (potashes) was brought into contact with burnt (caustic) lime, the mild alkali took up the principle of combustibility, obtained by the lime- stone in the fire, and it became caustic. Black showed that in the cases of magnesia-alba and chalk the disappearance of the effervescence on treatment with an acid after heating, was accompanied by a loss of weight. Moreover, as Van Helmont's older observations were quite forgotten, he was the first clearly to establish the existence of a kind of air or gas, termed fixed air (1752) totally distinct both from common atmospheric air and from modifications of it, by impurity or otherwise, such as the various gases hitherto prepared were believed to be. This fixed air, then, is given off when mild alkalis become caustic, and is taken up when the reverse change occurs. This clear statement of a fact which of itself is a powerful argument against the truth of the theory in which he had been 1 "Experiments upon Magnesia-alba, Quicklime, and other Alkaline sub- stances." Edin. Phys. and Literary Essays, 1755. 16 HISTORICAL INTRODUCTION brought up, was sufficient to make the name of Black illustrious, but he became immortal by his discoveries of latent and specific heats, the principles of which he taught in his classes at Glasgow and Edinburgh from 1763. The singularly unbiassed character of Black's mind is shown in the fact that he was the only chemist of his age who completely and openly avowed his conversion to the new Lavoisierian doctrine of combustion. From an interesting correspondence between Black and Lavoisier, it is clear that the great French chemist looked on Black as his master and teacher, speaking of Black's having first thrown light upon the doctrines which he afterwards more fully carried out. 1 This period of the history of our science has been called that of pneumatic chemistry, because, following in the wake of Black's discovery of fixed air, chemists were now chiefly engaged in the examination of the properties and modes of preparation of the different kinds of airs or gases, the striking and very different natures of which naturally attracted interest and stimu- lated research. No one obtained more important results or threw more light upon the existence of a number of chemically different gases than Joseph Priestley. In 1772 Priestley was engaged in the examin- ation of the chemical effect produced by the burning of com- bustible bodies (candles) and the respiration of animals upon ordinary air. He proved that both these deteriorated the air and diminished its volume, and to the residual air he gave the name of phlogisticated air. Priestley next investigated the action of living plants on the air and found to his astonishment that they possess the power of rendering the air deteriorated by animals again capable of supporting the combustion of a candle. Fig. 1, a reduced fascimile of the frontispiece to Priestley's celebrated Observations on Different Kinds of Air, shows the primitive kind of apparatus with which this father of pneumatic chemistry obtained his results. The mode adopted for generating and collecting gases is seen ; hydrogen is being prepared in the phial by the action of oil of vitriol on iron filings, and the gas is being collected in the large cylinder standing over water in the pneumatic trough ; round this trough are arranged various other pieces of apparatus, as, for instance, the bent iron rod holding . a small crucible to contain the substances which Priestley desired 1 Brit. Assoc. Reports, 1871, p. 189. JOSEPH PRIESTLEY 17 18 HISTORICAL INTRODUCTION to expose to the action of the gas. In the front is seen a large cylinder in which he preserved the mice, which he used for ascertaining how far an air was impure or unfit for respiration, and standing in a smaller trough is a cylinder containing living plants, the action of which on air had to be ascertained. On August 1st, 1774, Priestley obtained oxygen gas by heating red precipitate by means of the sun's rays concen- trated with a burning glass, and termed it dephlogisticated air,, because he found it to be so pure, or so free from phlogiston, that in comparison with it common air appeared to be impure. Priestley also first prepared nitric oxide (nitrous air or gas), nitrous oxide (dephlogisticated nitrous air), and carbonic oxide ; he likewise collected many gases for the first time over mercury, such as ammoniacal gas (alkaline air), hydrochloric acid gas. (marine acid air), sulphurous acid gas (vitriolic acid air), and silicon tetrafluoride (fluor acid air). 1 He also observed that when a series of electric sparks is allowed to pass through ammoniacal gas, an increase of volume occurs, and a combustible gas is formed, whilst on heating ammonia with calx of lead phlogisticated air (nitrogen gas) is evolved. Priestley's was a mind of rare quickness and perceptive powers, which led him to the rapid discovery of numerous new chemical substances, but it was not of a philosophic or delibera- tive cast. Hence, although he had first prepared oxygen, and had observed (1781) the formation of water, when inflammable air (hydrogen) and atmospheric air are mixed and burnt together in a copper vessel, he was unable to grasp the true explanation of the phenomenon, and he remained to the end of his days a firm believer in the truth of the phlogistic theory, which he had done more than any one else to destroy. Priestley's notion of original research, which seems quite foreign to our present ideas, may be excused, perhaps justified, by the state of the science in his day. He believed that all dis- coveries are made by chance, and he compares the investigation of nature to a hound, wildly running after, and here and there chancing on game (or as James Watt called it, " his random haphazarding"), whilst we should rather be disposed to compare the man of science to the sportsman, who having, after persistent effort, laid out a distinct plan of operations, makes reasonably sure of his quarry. In some respects the scientific labours of Henry Cavendish 1 Priestley's Observations on Different Kinds of Air, 1 } 328. HENRY CAVENDISH 19 (1731-1810) present a strong contrast to those of Priestley ; the work of the latter was quick and brilliant, that of the former was slow and thorough. Priestley passed too rapidly from subject to subject even to notice the great truths which lay under the surface ; Cavendish made but few discoveries, but his researches were exhaustive, and for the most part quantitative. His investigation on the inflammable air 1 evolved from dilute acid and zinc, tin, or iron, is a most remarkable one. In this memoir we find that he first determined the specific gravity of gases, and used materials for drying gases, taking note of alterations of volume due to changes of pressure and tempera- ture. He likewise proved that by the use of a given weight of each one of these metals, the same volume of inflammable gas can always be obtained no matter which of the acids be employed, whilst equal weights of the metals gave unequal volumes of the gas. Cavendish also found that when the above metals are dissolved in nitric acid, an incombustible air is evolved, whilst if they are heated with strong sulphuric acid sulphurous air is formed. He concluded that when these metals are dissolved in hydrochloric or in dilute sulphuric acid their phlogiston flies off, whilst when heated with nitric or strong sulphuric acids, the phlogiston goes off in combination with an acid. This is the first occasion in which we find the view expressed that inflammable air is phlogiston a view which was generally held, although Cavendish himself subsequently changed his opinion, regarding inflammable air as a compound of phlogiston and water. The discovery of oxygen by Priestley, and of nitrogen by Rutherford, naturally directed the attention of chemists to the study of the atmosphere, and to the various methods for ascer- taining its composition. Although Priestley's method of estimating the dephlogisti- cated air by means of nitric oxide was usually employed, the results obtained in this respect by different observers were very different. Hence it was believed that the composition of the air varies at different places, and in different seasons, and this opinion was so generally adopted, that the instrument used for such measurements was termed a eudiometer (evSta, fine weather, and /j,rpov, a measure). Cavendish investigated this subject with his accustomed skill in the year 1781, and found that when every possible precaution is taken in the analysis, 1 On Factitious Air. Hon. Henry Cavendish. Phil. Trans. 1766, 141. 20 HISTORICAL INTRODUCTION "the quantity of pure air in common air is -jf," or 100 volumes of air always contain 20'8 volumes of dephlogisticated, and 79*2 volumes of phlogisticated air, and that, therefore, atmospheric air had an unvarying composition. But the discovery which more than any other is for ever connected with the name of Cavendish is that of the composition of water (1781). 1 In making this dis- covery Cavendish was led by some previous observations of Priestley, and his friend Warltire. They employed a detonating closed glass or copper globe holding about three pints, so arranged that an electric spark could be passed through a mixture of inflammable air (hydrogen) and common air, 2 but though they had observed the production of water, they not only overlooked its meaning, but believed that the change was accompanied by a loss of weight. Cavendish saw the full importance of the phenomenon and set to work with care and deliberation to answer the question as to the cause of the formation of the water. Not only did he determine the volumes of air and hydrogen, and of dephlogisticated air (oxygen) and inflammable air (hydrogen) which must be mixed to form the maximum quantity of water, but he first showed that no loss of weight occurred in this experiment and that the formation of acid was not an invariable accompaniment of the explosion. On this important subject it is interesting to hear Caven- dish's own words; in the Philosophical Transactions for 1784, page 128, we read : " From the fourth experiment it appears that 423 measures of inflammable air are nearly sufficient to phlogisticate 1,000 of common air ; and that the bulk of the air remaining after the explosion is then very little more than four-fifths of the common air employed ; so that, as common air cannot be reduced to a much less bulk than that by any method of phlogistication, we may safely conclude when they are mixed in this proportion, and exploded, almost all the inflammable air and about one-fifth part of the common air, lose their elasticity and are condensed into a dew which lines the glass." Since 1,000 volumes of air contain 210 volumes of oxygen and these require 420 volumes of hydrogen to combine with them, we see how exact Cavendish's 1 Phil. Trans. 1784, 119, and 1785, 372, Mr. Cavendish's experiments on air. 2 A similar apparatus (originally due to Volta) was used by Cavendish. The pear-shaped glass bottle with stopcock, usually called Cavendish's eudiometer, would not be recognised by the great experimenter. HENRY CAVENDISH 21 experiments were. " The better," he continues, " to examine the nature of the dew, 500,000 grain measures of inflammable air were burnt with about 24 times that quantity of common air and the burnt air made to pass through a glass cylinder eight feet long and three-quarters of an inch in diameter, in order to deposit the dew .... By this means upwards of 135 grains of water were condensed in the cylinder, which had no taste or smell, and which left no sensible sediment when evaporated to dryness ; neither did it yield any pungent smell during the evaporation ; in short, it seemed pure water." Cavendish then sums up his conclusions from these two sets of experiments as follows : " By the experiments with the globe it appeared that when inflammable and common air are exploded in a proper proportion, almost all the inflammable air, and near one- fifth of the common air, lose their elasticity and are condensed into dew. And by this experiment, it appears that this dew is plain water, and consequently that almost all the inflammable air and about one-fifth of the common air are turned into pure water." Still more conclusive was the experiment in which Cavendish introduced a mixture of dephlogisticated air and inflammable air nearly in the proportions of one to two into a vacuous glass globe, furnished with a stopcock and means of firing by electricity. " The stopcock was then shut and the included air fired by electricity, by which means almost all of it lost ' its elasticity. By repeating the operation the whole of the mixture was let into the globe and exploded, without any fresh exhaustion of the globe." Priestley had previously been much led astray by the fact that he found nitric acid in the water obtained by the union of the gases. Cavendish, by a careful series of experiments, explained the occurrence of this acid, for he showed that it did not form unless an excess of dephlogisticated air was used, and he traced its production to the presence in the globe of a small quantity of phlogisticated air (nitrogen) derived from admixture of common air. He likewise proved that the artificial addition of phlogisti- cated air increased the quantity of acid formed in presence of dephlogisticated air (oxygen), whilst if the latter air were re- placed by atmospheric air no acid was formed, in spite of the large amount of phlogisticated air (nitrogen) present. In this way he showed that the only product of the explosion of pure dephlogisticated with pure inflammable air is pure water. 22 HISTORICAL INTRODUCTION Although Cavendish thus distinctly proved the fact of the com- position of water, it does not appear from his writings that he held clear views as to the fact that water is a chemical compound of its two elementary constituents. On the contrary, he seems to have rather inclined to the opinion that the water formed was already contained in the inflammable air, notwithstanding the fact that in 1783 the celebrated James Watt had already expressed the opinion that " water is composed of dephlogis- ticated and inflammable air." 1 Cavendish's general conclusions in this matter may be briefly summed up in his own words as follows : " From what has been said there seems the utmost reason to think that dephlogisticated air is only water deprived of its phlogiston, and that inflammable air, as was before said, is either phlogisticated water, or else pure phlogiston ; but in all probability the former." To the end of his days Cavendish remained a firm supporter of the phlogistic view of chemical phenomena, but after the overthrow of this theory by Lavoisier's experiments the English philosopher withdrew from any active participation in scientific research. Whilst Priestley and Cavendish were pursuing their great discoveries in England, a poor apothecary in Sweden was actively engaged in investigations which were to make the name of Scheele (1742-1786) honoured throughout Europe. These investigations, whilst they did not bring to light so many new chemical substances as those of Priestley, and did not possess the quantitative exactitude which is characteristic of the labours of Cavendish, opened out ground which had been entirely neglected, and was perhaps unapproachable by the English chemists. Scheele's discoveries covered the whole range of chemical science. A strong supporter of the phlogistic theory, he held peculiar views (see his celebrated treatise Ueber die Luft und das Feuer) as to the material nature of heat and light, and their power of combining with phlogiston, and, like Stahl, he considered modification in the forms of matter to be of much greater importance than alteration in its weight. In experiment- ing upon the nature of common air he discovered oxygen gas independently of, but probably somewhat later than, Priestley. The investigations which led Scheele to this discovery are of interest as a remarkable example of exact observations leading to erroneous conclusions. His object was to explain the part played by the air in the phenomenon of combustion ; and for 1 Letter from Watt to Black, 21st April, 1783. KARL WILHELM SCHEELE 23 this purpose he examined the action exerted by bodies sup- posed to contain phlogiston upon a confined volume of air. Thus he found that when a solution of hepar sulphuris (an alka- line sulphide) was brought into contact with a given volume of air, that volume gradually diminished, the residual air being incapable of supporting the combustion of a taper. The same result was observed when moist iron filings or the precipitate formed by the action of potash on a solution of green vitriol was employed. Scheele argued that if the effect of the combination of phlogiston with air is simply to cause a contraction, the remaining air must be heavier than common air. He found, however, that it was in fact lighter, and hence inferred that a portion of the common air must have disappeared, and that common air must consist of two gases, one of which has the power of uniting with phlogiston. In order to find out what had become of the portion of air which disappeared, Scheele heated phosphorus, metals, and other bodies in closed volumes of air, and found that these act just as the former kind of sub- stances had done. Hence he concluded that the compound formed by the union of the phlogiston with one of the constituents of the air is nothing more nor less than heat or fire which escapes through the glass. In confirmation of the truth of this hypothesis, Scheele believed that he had experimentally realised the decomposition of heat into phlogiston and fire-air. Nitric acid had, in his belief, a great power of combining with phlogiston, forming with it red fumes 5 he found that when he heated nitre in a retort, over a charcoal fire, with oil of vitriol, he obtained, in addition to a fuming acid, a colourless air, which supported combustion much better than common air. This he explained by assuming that when charcoal burns, the phlogiston combines with the fire-air to form heat, which passes into the retort, and is there decomposed into phlogiston, which by combining with the acid gave rise to the red nitrous fumes, and pure fire-air. He conceived that he had brought about the same chemical decomposition of heat by warming black oxide of manganese with sulphuric acid, or, still more simply, by heating calx of mercury ; for here it was clear enough that by bringing heat and calx of mercury together, the phlo- giston combined with the latter, and fire-air was liberated, thus : Heat. Mercury. Phlogiston + Fire-air + Calx of Mercury = Calx of Mercury + Phlogiston + Fire air. 24 HISTORICAL INTRODUCTION In the year 1774 Scheele made his great discovery of chlorine gas, which he termed dephlogisticated muriatic acid ; in the same year he showed that baryta was a peculiar earth ; shortly after- wards he proved the separate existence of molybdic and tungs- tic acids, whilst his investigations of prussian blue led to the isolation of hydrocyanic acid, of which he ascertained the properties. It was, however, especially in the domain of animal and vegetable chemistry that Scheele's most numerous discoveries lay, as will be seen by the following list of organic acids first prepared or distinctly identified by him : tartaric, oxalic (by the action of nitric acid on sugar), citric, malic, gallic, uric, lactic, and mucic. In addition to the identification of each of these as distinct substances, Scheele discovered glycerin, and we may regard him not only as having given the first indication of the rich harvest to be reaped by the investigation of the compounds of organic chemistry, but as having been the first to discover and make use of characteristic reactions by which closely allied substances can be detected and separated, so that he must be considered one of the chief founders of an- alytical chemistry. We have now brought the history of our science to the point at which Lavoisier placed it in the path which it has ever since followed. Before describing the overthrow of the phlogistic theory it may be well shortly to review the position of the science before the great chemist began his labours about one hundred years ago. Chemistry had long ceased to be the slave of the alchemist or the doctor ; all scientific chemists had adopted Boyle's definition, and the science was valued for its own sake as a part of the great study of nature. Stahl had well defined chemistry to be the science which was concerned with the resolution of compound bodies into their simpler consti- tuents, and with the building up of compounds from their elements ; so that the distinction between pure and applied chemistry was perfectly understood. Geber's definition of a metal as a fusible, malleable substance, capable of mixing with other metals, was still accepted ; gold and silver were con- sidered to be pure or noble metals, whilst the other malleable metals, copper, tin, iron, and lead, were called the base metals. Mercury, on the other hand, was thought to be only a metal-like body until it was frozen in 1759. After that date it was con- sidered to be a true metal in a molten state at the ordinary temperature Arsenic, antimony, bismuth, and zinc, from being LAVOISIEK 25 brittle, were classed as semi-metals, and to these well-known bodies were added cobalt in 1735, nickel in 1751, and man- ganese in 1774, whilst platinum was recognised as a peculiar metal in 1750, and molybdenum and tungsten were discovered about 1780. The several metals were supposed to be compounds of phlogiston with metallic calces, whilst sulphur, phosphorus, and carbon were looked upon as compounds of phlogiston with the acids of these elements. Of the simple gases the following were known : inflammable air (hydrogen), supposed to be either pure phlogiston or phlogisticated water ; dephlogisticated or fire- air (oxygen) ; phlogisticated air (nitrogen) ; and dephlogisticated muriatic acid (chlorine). When the metals dissolve in acids the phlogiston was thought to escape (as inflammable air) either in the pure state or combined with water. It was also known that when a metal is calxed, an increase of weight occurs, but this was explained either by the metal becoming more dense, which, in the opinion of some, would produce an increase of weight, or by the absorption of fiery particles, or again by the escape of phlogiston, a substance which instead of being attracted is re- pelled by the earth. In short, confusion and difference of opinion in the quantitative relations of chemistry reigned su- preme, and it was not until Lavoisier brought his great powers to bear on the subject that light was evoked from the darkness and the true and simple nature of the phenomena was rendered evident. In the year 1743 Lavoisier was born. Carefully educated, endowed with ample means, Lavoisier, despising the usual occu- pations of the French youth of his time, devoted himself to science, his genius, aided by a careful mathematical and physical training, rendering it possible for him to bring about a complete revolution in the science of chemistry. Before his time quanti- tative methods and processes were considered to be purely physical, though they now are acknowledged to be chemical, and of all these, the determination of the weights of bodies taking part in chemical change, as ascertained by the balance, is the most important. Others, indeed, before him, had made quantitative investigations. Black and Cavendish almost ex- ceeded Lavoisier in the exactitude of their experiments, but it is to the French philosopher that the glory of having first distinctly asserted the great principle of the indestructibility of matter belongs. Every chemical change, according to him, consists in a transference or an exchange of a portion of the 26 HISTORICAL INTRODUCTION material constituents of two or more bodies ; the sum of the weights of the substances undergoing chemical change always remains constant, and the balance is the instrument by which this fundamental fact is made known. In his first important research (1770) Lavoisier employs the balance to investigate the question, much discussed at the time, as to whether water on being heated becomes converted into earth. For one hundred and one days 1 he heated water in a closed and weighed vessel ; at the end of the experiment the weight of the closed vessel remained unaltered, but on pouring out the water he found that the vessel had lost 17'4 grains, whilst on evaporating the water, he ascertained that it had dis- solved 20*4 grains of solid matter. Taking the excess of 3'0 grains as due to unavoidable experimental errors, he concludes that water when heated is not converted into earth. Shortly after this, the same question was examined independently by Scheele, who obtained the same results by help of qualitative analysis, which showed that the water had taken up a constituent of the glass, viz., the alkaline silicates. When he became acquainted with the novel and unexpected discoveries of Black, Priestley, and Cavendish, a new light burst upon the mind of Lavoisier, and he threw himself instantly with fresh ardour into the study of specially chemical phenomena. He saw at once that the old theory was in- capable of explaining the facts of combustion, and by help of his own experiments, as well as by making use of the experiments of others, he succeeded in finding the correct explanation, destroying for ever the theory of phlogiston, and rendering his name illustrious as having placed the science of chemistry on its true basis. On looking back in the history of our science we find indeed that others had made experiments which could only be explained by this new theory, and in cer- tain isolated instances the true explanation may have previously occurred to the minds of others. Thus in 1774 Bayen showed that calx of mercury loses weight, evolving a gas equal in weight to what is lost, and he concludes that either the theory of phlo- giston is incorrect, or this calx can be reduced without addition of phlogiston. This, however, in no way detracts from Lavoisier's glory as having been the first to carry out the true ideas consist- ently and deliberately through the whole science. It is the systematic application of a truth to every part of a science which 1 (Euvres de Lavoisier, 2, 22. LAVOISIER 27 constitutes a theory, and this it was that Lavoisier and no one else accomplished for chemistry. When a man has done so much for science as Lavoisier, it seems almost pitiful to discuss his shortcomings and failings. But it is impossible in any sketch of the history of chemistry to ignore the question how far Lavoisier's great conclusions, the authorship of which no one questions, were drawn from his own discoveries, or how far he was indebted to the original investigations of his contemporaries for the facts upon which his conclusions are based. The dispute has recently assumed fresh interest. Certain chemists consider that to him alone the foundation of modern chemistry is to be ascribed, both as . regards material and deduction, whilst others, affirming that Lavoisier made use of the discoveries of his predecessors, and especially of the discovery of oxygen by Priestley, without acknowledgment, assert that he went so far as to claim for himself a participation in this discovery to which he had no right whatever, and insist that until he had thus obtained, from another, the key to the problem, his views upon the question of combustion were almost as vague as those of the phlogistonists themselves. To enter into a full discussion of the subject would lead us into a historical criticism which would outrun our space. Suffice it to say that many of the charges which have been brought against Lavoisier's good faith unfortunately turn out upon investigation to be well founded, so that whilst we must greatly admire the clear sight of the philosopher, we cannot feel the same degree of respect for the moral character of the man. His investigations on the phenomena of combustion began in the year 1772. In a first memoir 1 Lavoisier finds not only that ^hen sulphur and phosphorus are burnt no loss of weight occurs, but that an increase of weight is observed. Hence he concludes that a large quantity of air becomes fixed. This dis- covery leads him to the conclusion that a similar absorption of air takes place whenever a body increases in weight by combus- tion or calcination. In order to confirm this view, he reduces litharge with charcoal/and finds that a considerable quantity of air is liberated. This, he asserts, appears to him to be one of the most interesting experiments made since the time of Stahl. Lavoisier's next publication was his Opuscles physiques ct 1 Sur la Cause de I' Augmentation des Poids. (Euvres, 2, 99- 28 HISTORICAL INTRODUCTION chymiques, commenced in 1774. In these memoirs he first examines the kind of air given off in the processes of breathing, combustion, and fermentation. The views which he expresses are similar to those put forward long before by Black, to whom he repeatedly refers as the originator of them, this acknowledg- ment of his indebtedness to the Scottish philosopher being repeated in the letters from Lavoisier to Black which have been already referred to, 1 in one of which the following passage occurs : " Plus confiant dans vos idSes que dans les miennes propres, accoutume d vous regarder comme mon maitre" &c. In the year 1774 he describes experiments on the calcination of lead and tin, which he, like Boyle, heats in closed glass globes : so long as the vessel is closed it does not change in weight, but when the neck of the flask is broken, air rushes in, and the weight increases. He further shows that only a por- tion of the air is taken up by the molten metal, and that the residual air is different from common air, and also from fixed air. From these statements it is clear that Lavoisier considered that the air consists of two different elastic fluids, but that he was not acquainted with Priestley's discovery of oxygen. Nor were his views at this time so precise or well defined as we should gather from reading his papers published in the memoirs of the French Academy for 1774. The explanation is simple enough, inasmuch as owing to the careless and tardy manner in which the memoirs of the French Academy were at that time edited, changes in the original communications were frequently made by the writers before publication, so that the papers printed in the memoirs were corrected to suit alteration in view or in fact which had become known to the authors between the times of reading and of publication. Thus, for instance, it is clear that the paper 2 detailing the results of his experiments on the calcination of the metals above referred to, which was read before the Academy in Nov. 1774, does not express the same views which we find given in the extended description of his experiments contained in the volume of the memoirs for 1774, which however was not published till 1778. So that although Lavoisier in 1774 considered air to be made up of several different elastic fluids, it is certain that he was not then acquainted with the kind of air which was absorbed in calcination, that his views 1 British Association Reports, 1871, 190. 2 Journal de Physique for Dec. 1774. THE DISCOVERY OF OXYGEN 29 on the subject were in reality very similar to those expressed a century before by Jean Key (1630), Mayow (1669), and later, by Pott (1750), and that they were far from being as precise and true as we should gather them to have been from the perusal of his extended memoir, printed in 1778 and corrected so as to harmonise with the position of the science at that date. It is not until we come to a paper, Sur le nature du principe qui se combine avec les me'taux pendant leur calcina- tion, first read in 1775 and re-read on Aug. 8, 1778, that we find a distinct mention of oxygen gas, which he first termed " I'air e'minemment respirable," or " I' air pur," or " I' air vital," and that we see that the whole theory of combustion is clear to Lavoisier. He shows that this gas is necessary for the cal- cination of metals, he prepares it from precipitation per se, as Priestley had previously done, and in the year 1778 we find the first mention of oxygen or the acidifiant principle. The name was given to it because he observed that combined with carbon this substance forms carbonic acid, with sulphur vitriolic acid, with nitrous air nitric acid, with phosphorus phosphoric acid, although with the metals in general it produces the metallic calces. In his Elements de Chimie, published in 1782, we find the following words under oxygen gas : " Get air que nous avons (ttcouwrt presque en mSme temps, Dr. Priestley, M. Scheele et moi." x Now there is no doubt whatever that in October, 1774, Dr. Priestley informed Lavoisier, in Paris, of the discovery he had lately made, and that Lavoisier was at that time unacquainted with the fact that precipitatum per se yields this new gas on heating. Hence we cannot admit Lavoisier's claim to the joint discovery of oxygen, a claim, it is to be remembered, not made until eight years after the event had occurred. In corroboration of this conclusion we find in Priestley's last work, published in 1800, and singularly enough entitled The Doctrine of Phlogiston Established, the following succinct account of the matter. " Now that I am on the subject of the right of discoveries," he says, " I will as the Spaniards say, leave no ink of this kind behind in my ink-horn, hoping it will be the last time I shall have any occasion to trouble the public about it. M. Lavoisier says (Elements of Chemistry, English edition, p. 36) ' This species of air (meaning dephlogisticated) was discovered almost at the same time by Mr. Priestley, M. Scheele, and myself.' The case was this : having made the discovery 1 (Euvres, 1, 38. 30 HISTORICAL INTRODUCTION some time before I was in Paris in 1774, I mentioned it at the table of M. Lavoisier, when most of the philosophical people in the city were present ; saying that it was a kind of air in which a candle, burned much better than in common air, but I had not then given it any name. At this all the company, and Mr. and Mrs. Lavoisier as much as any, expressed great surprise ; I told them then I had gotten it from precipitatum per se and also from red lead. Speaking French very imperfectly, and being little acquainted with the terms of chemistry, I said plomb rouge and was not understood till M. Macquer said, ' I must mean minium! M. Scheele's discovery was certainly independent of mine, though I believe not made quite so early." The two memoirs in which Lavoisier clearly puts forward his views on the nature of combustion and respiration are, first, one read before the Academy in 1775, Sur la combustion en gtne'ral, and second, one entitled Reflexions sur la Phlogistique, published by the Academy in 1783. In the first of these memoirs he does not attempt to substitute for Stahl's doctrine a rigorously demonstrated theory, but only an hypothesis which appears to him more conformable to the laws of nature, and less to con- tradict known facts. In the second memoir he develops his theory, denying the existence of any " principle of combust- ibility," as upheld by Stahl, stating that the metals, and such substances as carbon, sulphur, &c., are simple bodies which on combustion enter into combination with oxygen, and concluding that Stahl's supposition of the existence of phlogiston in the metals, &c., is entirely gratuitous, and more likely to retard than to advance the progress of science. The triumph of the antiphlogistic (Lavoisierian) doctrines was, however, not complete until the discovery of the compound nature of water by Cavendish in 1783 became fully known. The experiment concerning the combination of hydrogen (phlo- giston) and oxygen to form water was at once repeated and confirmed by Lavoisier and Laplace on the 24th June, 1783, and then Lavoisier was able satisfactorily to explain the changes which take place when metals dissolve in acids, and to show that the metals are simple bodies which take up oxygen on combustion, or on solution in acid, the oxygen being derived in the latter case either from the acid or from the water present. Here, again, if we investigate the position occupied by Lavoisier respecting the discovery of the composition of water we shall see CAVENDISH, WATT, AND LAVOISIER 31 that, not content with the glory of having been the first to give the true explanation of the phenomena, he appears to claim for himself the first quantitative determination of the fact, 1 although it is clear that he had been previously informed by Blagden of Cavendish's experiments. 2 The verdict concerning the much- vexed question as to the rival claims of Cavendish, Watt, and Lavoisier, cannot be more forcibly or more concisely given than in the following words of Professor Kopp Cavendish first ascertained the facts upon which the discovery of the composition of water was based, although we are unable to prove that he first deduced from these facts the compound nature of water, or that he was the first rightly to recognise its constituent parts. Watt was the first to argue from these facts the compound nature of water, although he did not arrive at a satisfactory conclusion respecting the nature of the components ; whilst Lavoisier, also from these facts, first clearly recognised and stated the true nature of the components of water. Although at this period the experimental basis of the true theory of combustion was complete, it was some time before the clear statements of Lavoisier were accepted by chemists. Many of those who were most distinguished by their discoveries remained to the last wedded to the old ideas, but by degrees, as fresh and unprejudiced minds came to study the subject, the new views were universally adopted. In considering this great discussion from our present point of view, we cannot but recognise in the phlogistic theory the expression of an important fact, of which, however, the true in- terpretation was unknown to the exponents of the theory. The phlogistonists assert that something which they term phlogiston escapes when a body burns ; the antiphlogistonists prove, on the other hand, that no escape of material substance then occurs, but that on the contrary, an addition of oxygen (or some other element) always takes place. In thus correcting from one aspect the false statement of the followers of Stahl, Lavoisier and his disciples to some extent overlooked an interpretation which may 1 (Euvres, 2, 338. 2 For an exhaustive discussion of this subject we must refer the reader to George Wilson's Life of Cavendish, 1849, as well as to Prof. H. Kopp's Beitrdge zur Geschichte der Chemie. Die JSntdcckung der Zusammensetzung des Wassers. Vieweg und Sohn, 1875. See also Grimaux, Lavoisier, F. Alcan, Paris. Thorpe, Brit. Ass. Reports, 1890, p. 761., and Berthelot, La Revolution Chimique y F. Alcan, Paris, 1890. 32 HISTORICAL INTRODUCTION truly be placed upon the statements of the phlogistonists, for if in place of the word " phlogiston/' we read " energy," this old theory becomes the expression of the latest development of scientific investigation. We now know that when two elements combine, Energy, generally in the form of heat, is usually evolved, whilst in order to resolve the compound into its constituent elements an expenditure or absorption of an equal amount of energy is requisite. The fact that every distinct chemical compound possesses a fixed and unalterable composition, was first proved by the endeavour to fix the composition of certain neutral salts. Bergman from the year 1775, and Kirwan from 1780, were occupied with this experimental inquiry, but their results did not agree sufficiently well to enable chemists to come to a satis- factory conclusion, and it was to Oavendish that we owe the first proof that the combining proportion between base and acid follows a distinct law, whilst to him we also owe the introduction of the word " equivalent " into the science. It is, however, to Richter (1762-1807) that we are indebted for the full explanation of the fact, that when two neutral salts undergo mutual decomposition, the two newly-formed salts are also neutral. He shows in his " Stochiometrie," that the proportions by weight of different bases which saturate the same weight of a given acid will also saturate a different but a constant weight of a second acid. So that if we have determined what weight of a given base is required to saturate a given weight of several different acids, and also if we know the weights of the different bases which are needed for the neutralization of a given weight of any one of these acids, we can calculate in what proportion each of these bases will unite with any one of these acids. Richter also showed that when the different metals are separately dissolved in the same quantity of sulphuric acid, each one takes up the same quantity of oxygen ; or, as we may now express it, the varying quantities of these different oxides which neutralize one and the same quantity of any acid, all contain the same quantity of oxygen. These important observations attracted but little attention or consideration from Richter's contemporaries, all of whom were busily engaged in carrying on the phlogistic war in which he himself took an active part in defence of the older doctrine. The investigations of Richter and his predecessors had reference mainly to the proportions by weight in which acids and bases unite, which, according to Lavoisier's theory, are not "ESSAI DE STATIQUE CHIMIQUE " 33 simple substances, whilst Lavoisier recognised the fact that the elements themselves combine in definite proportions by weight. In opposition to this view of combination in definite unalterable quantities, L. Claude Berthollet published in 1803 his cele- brated Essai de statique Chimique, in which he refers the phenomena of chemistry to certain fundamental properties of matter, endeavouring to explain chemical changes by the motions of the particles of matter on the same principles as Newton's theory of gravitation accounts for the simpler motions of the heavenly bodies. Considering chemical change from this mechanical point of view, Berthollet pointed out the cir- cumstances under which we can accomplish the highest development of the science, namely, prediction of phenomena ; .and, if, in his assumed identity of the laws of gravitation and chemical action, he was mistaken, the aim which he set before himself is that which has remained, and will ever remain, the highest ideal of the science. The influence which Berthollet's views exercised on the progress of the science was less powerful than it otherwise would have been owing to the fact that he, considering chemical combination to be based upon purely mechanical laws, was obliged to admit that an alter- ation of the conditions, such as mass and temperature, must generally produce an alteration in the composition of the chemical compound, and was, therefore, forced to the conclusion that combination may take place, as a rule, between variable proportions of the elements, fixity of proportion being the exception, due to some special physical property of the com- pound containing those proportions, such as insolubility or elasticity. The opposite view that combination only takes place in a small number of definite fixed proportions was defended by his countryman Proust, and this led to a keen debate between the two French philosophers which lasted from the year 1801 to the year 1808. In the end, however, Proust proved conclusively that Berthollet's views were incorrect inasmuch as he showed that when one metal gives rise to two oxides, the weight of the metal which combines with the same quantity of oxygen to form the various oxides is a different but a fixed quantity, so that combination does not take place by the gradual addition of one element, but by sudden increments. This observation ought in fact to have led to the recognition by Proust of the law of combining proportions, but his analyses 4 34 HISTORICAL INTRODUCTION were not sufficiently accurate for this purpose, 1 so that neither Proust nor Richter arrived at the true expression of the facts of chemical combination, and it was reserved for John Dalton, (1766 1844) clearly to state the great law of chemical com- bination in multiple proportions, and to found upon this a theory which fully explains the observed facts. Democritus, and after him Epicurus and Lucretius, had long ago taught that matter is made up of small indivisible particles, and the idea of the atomic constitution of matter, and even the belief that chemical combination consists in the approximation of the unlike particles, had been already expressed by Kirwan in 1783^ as well as by Higgins in 1789. Dalton was, however, the first to propound a truly chemical atomic theory, the only one hitherto proposed which explains the facts of chemical combina- tion in a satisfactory manner. The cardinal point upon which Dalton's atomic theory rests, and in which it differs from all previous suggestions, is that it is a quantitative theory respecting the constitution of matter, whereas all others are simply quali- tative views. For whilst all previous upholders of an atomic theory, including even Higgins, had supposed that the relative weights of the atoms of the various elements are the same, Dalton at once declared that the atoms of the different elements are not of the same weight ; and that the relative atomic weights of the elements are the proportions by weight in which the elements combine, or some multiple or submultiple of these. In 1803 Dalton published his first table of atomic weights of certain elements and their compounds, as an appendix to a paper read before the Manchester Literary and Philosophical Society, Oct. 23, 1803, on the absorption of gases by water and other liquids. As a reason for introducing these numbers, Dalton states that the different solubility of gases in water depends upon the weight and number of the ultimate particles of the several gases. " The inquiry," he continues, " into the relative weights of the ultimate particles of bodies is a subject, as far as I know, entirely new ; I have lately been prosecuting this inquiry with remark- able success. The principle cannot be entered upon in this paper, but I shall subjoin the results as far as they appear ascertained by my experiments." 1 Journal de Physique, 59, 260 and 321. JOHN DALTON 35 Daltoris First Table of the Relative Weights of the Ultimate Particles of Gaseous and other Bodies. Hydrogen ..... 1 Nitrous oxide . . . 137 Azot 4'2 Sulphur 14-4 Carbon 4'3 Hypo-nitric acid . .15-2 Ammonia 5'2 Sulphuretted hydrogen 15 '4 Oxygen 5'5 Carbonic acid . . . 15'3 Water 6'5 Alcohol 151 Phosphorus . . . . 7'2 Sulphureous acid . .19*9 Phosphuretted hydrogen 8*2 Sulphuric acid . . . 25*4 Nitrous gas .... 0*3 Carburetted hydrogen, Ether 9*6 from stagnant water. 6'3 Gaseous oxide of carbon 9*8 Olefiant gas . . . . 5'3 Thus then, at the end of a paper on a physical subject, does Dalton make known a principle the discovery of which at once placed the science of chemistry upon its true basis, and has rendered the name of its discoverer second only to that of Lavoisier amongst the founders of the Science. It is not easy to follow in detail the mental or experimental processes by which Dalton arrived at this great theory. Certain it is, however, that the idea which lay at its foundation had long been in his mind, which was essentially of a mathematical and mechanical turn, and that it was by his own experimental deter- minations, and not by combining any train of reasoning derived from the previous conclusions of other philosophers, that he was able to prove the correctness of his theory. Singularly self-reliant, accustomed from childhood to depend on his own exertions, Dalton was a man to whom original work was a necessity. 1 In the preface to the second part of his New System of Chemical Phil- osophy, published in 1810, he clearly shows his independence and even disregard of the labours of others, for he says " Having been in my progress so often misled by taking for granted the results of others, I have determined to write as little as possible but what I can attest by my own experience." As early as 1802, in an experimental inquiry into the propor- tions in which the several gases constituting the atmosphere occur, Dalton clearly points out " that the elements of oxygen may combine with a certain portion of nitrous gas " (our nitric 1 Lonsdule's Life of Dalton. Longmans, 1874. 36 HISTORICAL INTRODUCTION oxide) " or with twice that portion, but with 110 intermediate quantity," and this observation was clearly the first which led to the possibility of drawing up the table already given. 1 In that table, it will be seen that the relative weights of the smallest particle of nitrous gas is given as 9*3, that of Azot (nitrogen) being 4'2, and that of oxygen 5*5. Dalton clearly intending by this to express that the gas is a compound of one atom of nitro- gen with one atom of oxygen, whilst the substance to which he gives the name of hypo- nitric acid (now called nitrogen peroxide), is a compound in which one atom of nitrogen is combined with two of oxygen, and therefore having the rela- tive atomic weight of 15'2. 2 The first public announcement of the atomic theory, and of the law of combination in multiple proportions upon which it was founded, was, singularly enough, not made by Dalton himself, but by his friend, Professor Thomas Thomson, of Glasgow, who pub- lished in 1807 an account of Dalton's discovery in the third edition of his System of Chemistry. In the following year (1808) Dal- ton made known his own views in the remarkable book entitled A New System of Chemical Philosophy, in which (Part i. p. 213) he says " It is one great object of this work to show the importance and advantage of ascertaining the relative weights of the ultimate particles, both of simple and compound bodies, the number of simple elementary particles which constitute one compound particle, and the number of less compound particles which enter into the formation of one more compound particle." Thomson states that during the years 1803 and 1804 Dalton was occupied with the examination of the composition of the two gaseous hydro-carbons, marsh gas and olefiant gas, and the results of this examination led him to the adoption of the atomic theory. He found that both these bodies consist solely of carbon and hydrogen, and that the first of these gases contains twice as much hydrogen to a given quantity of carbon as the second. Hence he concluded that olefiant gas contains one atom of carbon combined with one of hydrogen, whereas marsh gas consists of one atom of carbon combined with two atoms 1 Manchester Memoirs, 2nd Series, 1, 250. 2 Certain inaccuracies in the values of the weights of some of the compounds occur in this table ; thus, 4 '2 + 5 "5 = 97, whilst 9 '3 appears opposite nitrous oxide. Whether these are merely printer's errors or are to be explained in some other way can now only be conjectured. See Roscoe on Dalton's First Table of Atomic Weights. Manchester Lit. and Phil. Soc. Mem. 3rd Series, 5, 269, 1874-5. DALTON'S ATOMIC THEORY 37 of hydrogen. The same idea and method of investigation he then applied to the oxides of carbon, oxides of sulphur, oxides of nitrogen, to ammonia and other bodies, and he showed that the composition of these might be most simply explained by the assumption that one atom of one element is attached to 1, 2, 3, &c. atoms of another. The novelty and importance of his view of the composition of chemical compounds induced Dalton to introduce a method of graphic representation of the atoms of the elements, and the system he adopted was as follows : Relative Weight Symbol. of the atom. Oxygen O 7 Hydrogen 1 Nitrogen 5 Carbon 9 5 Water O 8 Ammonia 00 6 Carbonic oxide O 12 Carbon dioxide O O 19 Olefiant gas 00 6 Marsh gas 7 Nitrous oxide O 17 Nitric oxide O 12 Nitrous acid f O O I 26 Acetic acid ^ ' O O 26 These atomic weights, it is evident, are far from being those which we now accept as correct, indeed they are different from those given in his first table, for Dalton not only frequently altered and amended these numbers, according as his experi- ments showed them to be faulty, but even distinctly asserts the doubtful accuracy of some. Chemists at that time did not possess the means of making accurate determinations, and when we become acquainted with the rough methods which Dalton adopted, and the imperfect apparatus he had to employ, we can- not but be struck with the clearness of his vision and the boldness of grasp which enabled him, thus poorly equipped, to establish a doctrine which further investigation has only more firmly established, and which, from that time forward, has served 38 HISTORICAL INTRODUCTION as the pole star round which all other chemical theories revolve. Amongst those to whose labours we are indebted for advancing Dalton's atomic theory are Thomas Thomson and Wollaston, but before all, the great Swedish chemist Berzelius. to whom we owe the first really exact values for these primary chemical constants. With remarkable perseverance he ascertained the exact composition of a large number of compounds, and was, there- fore, able to calculate the combining weights of many elements, thus laying the foundation-stones of the science as it at present exists. In 1818 Berzelius published his theory of chemical pro- portions, and of the chemical action of electricity, and in these remarkable works he made use of the chemical symbols and formulae such as we now employ, to denote not only the qualita- tive, but also the quantitative composition of chemical compounds. From this time forward it was satisfactorily proved and generally acknowledged that the elementary bodies combine together either in certain given proportions by weight, or in simple multiples of these proportions ; and, through the researches of Berzelius and others, the list of elements, which at the time of Lavoisier amounted to twenty-three in number, was now con- siderably increased. Next in order comes Humphry Davy's discovery of the com- pound nature of the alkalis (1808), proving that they are not simple substances but oxides of peculiar metals, and thus entirely revolutionizing the views of chemists as to the constitution of a large and important class of compounds, including the salts of the alkaline earths. The discussion in 1810 as to the constitu- tion of chlorine then termed oxygenated muriatic acid decided by Davy and Gay-Lussac in favour of its elementary nature, was likewise a step of the greatest importance and of wide applica- tion. In 1811 iodine was discovered by Courtois, and most carefully investigated by Gay-Lussac, who proved the close analogy existing between this element and chlorine. The dis- covery of many other elements- now opened out fresh fields for investigation, and gave the means of classifying those already known. The names and properties of these will be found in the portions of this book specially devoted to their description. If Dalton, as we have seen, succeeded in placing the laws of chemical combination by weight on a firm basis, to Gay-Lussac belongs the great honour of having discovered the law of the combination of gaseous bodies by volume. In the year 1805 HUMBOLDT AND GAY-LUSSAC 39 Gay-Lussac and Alexander von Humboldt found that one volume of oxygen combines with exactly two volumes of hydrogen to form water, and that these exact proportions hold good at what- ever temperature the gases are brought into contact. This observation was extended by Gay-Lussac, who in 1808 published his celebrated memoir on the combination of gaseous bodies, 1 in which he proves that gases not only combine in very simple relations by volume, but also that the alteration of volume which these gases undergo in the act of combination may be expressed by a very simple law. Hence it follows that the densities of gases must bear a simple relation to their combining weights. The true explanation of these facts was first given by Avogadro in 1811, and his hypothesis is now universally admitted both by chemists and physicists. According to the Italian philo- sopher the number of smallest particles or molecules con- tained in the same volume of every kind of gas is the same, similar circumstances of pressure and temperature being of course presupposed. The discovery by Gay-Lussac of the laws of volume- combination, together with Avogadro's explanation of the law, served no doubt as most valuable supports of Dalton's atomic theory, but the truth of this latter theory was still further asserted by a discovery made by Dulong and Petit in 1819. These French chemists determined the specific heat of thirteen elementary bodies, and found that the numbers thus obtained, when compared with the atomic weights of the same bodies, showed that the specific heats of the several elements are inversely proportional to their atomic weights, or in other words, the atom of each of these elements possesses the same capacity for heat. Although subsequent research has shown that this law does not apply in every case, it still remains a valuable means of controlling the atomic-weight determina- tions of many elements. In the same year a discovery of equal importance was an- nounced by Mitscherlich that of the law of Isomorphism. According to this law, chemically analogous elements can re- place each other in many crystalline compounds, either wholly or in part, without any change occurring in the crystalline form of the compound. This law, like that of atomic heats, has proved of great value in the determination of atomic weights. Gradually the new basis given by Dalton to our science was 1 Memoires d'Arcueil, 2, 207. 40 HISTORICAL INTRODUCTION widely extended by these discoveries and by the researches of other chemists, and a noble structure arose, towards the com- pletion of which a numerous band of men devoted the whole energies of their lives. Especially striking was the progress made during these years in the domain of Organic Chemistry, or the chemistry of the substances found in, or obtained from, vegetable or animal bodies. Dalton had in vain endeavoured to obtain analytical results to prove that the complicated Organic bodies followed the same laws as the more simple Inorganic compounds. It is to Berzelius that we owe the proof that this is really the case, and his exact analyses placed organic chemistry in this respect on a firm and satisfactory basis. There still remained, however, much doubt as to the strict identity of the laws according to which organic and inorganic compounds were severally formed. Most of the compounds met with in mineral chemistry could be easily prepared by the juxtaposition of their constituents ; they were of comparatively simple constitution, and could as a rule be pre- pared by synthesis from their constituent elements. Not so with organic bodies ; they appeared to be produced under cir- cumstances wholly different from those giving rise to mineral compounds ; the mysterious phenomena of life seemed in some way to influence the production of these substances and to preclude the possibility of their artificial preparation. A great step was therefore made in our science when, in 1828, Wohler artificially prepared urea, a body which up to that time had been thought to be a product 'peculiar to animal life. This discovery broke down at once the supposed impassable barrier between organic and mineral chemistry, pointed out the rich harvest of discovery since so largely developed, especially by Liebig, in the synthesis of organic substances, and paved the way to the know- ledge which we have gained, chiefly through the labours of the last-named chemist, that the science of Physiology consists simply in the Chemistry and Physics of the body. GENEEAL PRINCIPLES OF THE SCIENCE i MATTER is capable of assuming three different states or con- ditions : the solid, the liquid, and the gaseous. Of these, the first two have, for obvious reasons, been recognised from the earliest ages as accompanying very different kinds of substances, It is, however, only within a comparatively short time that men have come to understand that just as there are many distinct kinds of solids and liquids, so there are many distinct kinds of gases (Van Helmont). These may, indeed, be colourless and in- visible, but, nevertheless, they can readily be shown to differ one from another. Thus, Black, in 1752, collected a peculiar gas, which wenow know as carbonic acid gas, or carbon dioxide, obtained by the action of dilute acids on marble ; to this gas he gave the name of " fixed air," because it is fixed in the alkaline carbonates, which at that time were called the mild alkalis, in contradis- tinction to the caustic alkalis. This invisible gas does not, like air, support the combustion of a taper, and, unlike air, it renders clear lime-water turbid ; it is also much heavier than air, as can be shown by pouring it downwards from one vessel to another, by drawing it out of a vessel by means of a syphon, or by pouring it into a beaker glass previously equipoised at one end of the beam of a balance (see Fig. 2). That the gas has actually been poured out is seen either by a burning taper being extinguished when dipped into the beaker glass, or by adding some clear lime-water, which then turns milky. In 1766, Cavendish showed that the gas termed by him in- flammable air, and obtained by the action of dilute acids on metallic zinc or iron, is also a peculiar and distinct substance, to which we now give the name of hydrogen gas. It is so much lighter than air that it may be poured upwards, and takes fire when a light is brought in contact with it, burning with a pale blue flame. Soap-bubbles blown with hydrogen ascend in the 42 GENERAL PRINCIPLES OF THE SCIENCE air, and if hydrogen be poured upwards into the equipoised bell- jar hung mouth downwards on the arm of the balance (Fig. 3), the equilibrium will be disturbed, and the arm with the bell-jar will rise. 2 On August 1st, 1774, Priestley heated some red precipi- tate (oxide of mercury) and obtained from it a new colour- less gas called oxygen, and this, although invisible, possesses FIG. 2. properties quite different from those of air, carbonic acid gas, or hydrogen gas. A red hot chip of wood is at once rekindled when plunged into this gas, and bodies such as iron wire or steel watch-spring, which do not burn in the air, burn with brilliancy in oxygen. These examples suffice to show that invisible gases exist which differ in the widest degree from each other, though many more illustrations of the same principle might be given. THE EXPERIMENTAL METHOD 43 3 The method which we have had to adopt in order thus to discern differences between these invisible gases, is termed the Experimental Method. Experiments may be said to be ques- tions put to nature, and a science is termed experimental, as opposed to observational, when we are able so to control and modify the conditions under which the phenomena occur as to produce results which are different from those which are otherwise met with. Chemistry is, therefore, one of several experimental sciences, each of which has the study of natural phenomena for its aim. These sciences are most intimately con- nected, or, rather, the division into separate sciences is quite arbitrary, so that it is not possible exactly to say where the phenomena belonging to one science begin and those appertaining to another science end. Nature is a connected whole, and the divisions which we are accustomed to make of natural phenomena 44 GENERAL PRINCIPLES OF THE SCIENCE into separate sciences serve only to aid the human mind in its efforts to arrange a subject which is too vast in its complete extent for the individual to grasp. Although it may not be possible exactly to define the nature of the phenomena which we class as chemical, as distinguished from those termed physical, it is not difficult, by means of examples, to obtain a clear idea of the kind of observations with which the chemist has to do. Thus, for instance, it is found that when two or more given substances are brought together, under certain conditions they may change their properties, and a new substance, differing altogether from the original ones, may make its appearance. Or, again, a given substance may, when placed under certain conditions, yield two or more substances differing entirely from the original one. in their essential properties. In both these cases the change which occurs is termed a chemical change ; if several distinct substances have coalesced to form one new substance, an act of chemical combination is said to occur ; if one substance is made to yield two or more distinct new bodies, a chemical decomposition has taken place. These acts of chemical union and disruption occur alike amongst solid, liquid, and gaseous bodies ; they are regulated in the first place by the essential nature of the substances, and secondly, by the circum- stances or conditions under which they are brought together. It is also to be observed that these actions of chemical union in the first place do not occur when the component materials, are situated at a distance from each other, close contact being necessary in order that such changes should take place ; whilst secondly, we almost invariably notice that such a combination is attended with an evolution of heat and,, sometimes, of light. 4 Some simple illustrations of chemical action may here be cited : If powdered sulphur and fine copper-filings be well mixed together, a green-coloured powder will result, in which, however, a powerful microscope will show the particles of sulphur lying by the side of the particles of copper. On heating this green powder in a test tube, the mass suddenly becomes red-hot, and, on cooling, a uniform black powder is found. This is neither copper nor sulphur, but a chemical compound of the two, in which no particle of either of the substances can be seen, how- ever high a magnifying power be employed, but from which, by the employment of certain chemical means, both copper and sulphur can again be extracted. Here then we have a case CHEMICAL ACTION 45 of chemical combination. An experiment similar to that made by Priestley when he discovered oxygen, may serve as an illustra- tion of a chemical decomposition. 21 '6 grams l of red oxide of mercury are heated in a small retort provided with a receiver, and a gas delivery tube passing to the top of a graduated cylinder filled with water to the beginning of the graduations, and standing in the pneumatic trough over water (Fig. 4). On heating the red oxide by means of the Bunsen gas flame, it first becomes dark- coloured and then soon begins to decompose into metallic mercury, which collects in small bright drops in the neck of the retort gradually running down in the receiver, and into oxygen gas, which passes through the delivery tube and collects in the graduated cylinder. After the heat has been continued for some time, the whole of the red powder will have disappeared, FIG. 4. .having been changed by heat into metallic mercury and oxygen. On allowing the retort to cool, a volume of 1120 cubic centi- metres of gas has been collected, and this, on application of the red-hot chip test, is shown to be oxygen. If this experiment is properly conducted, the volume of oxygen obtained is always found to be the same from the same weight of oxide (provided the temperature and pressure at which the gas is measured are the same in the different experiments), viz., 1120 cubic centi- metres from 21 '6 grams of oxide. Another interesting case of chemical change is the decompo- 1 For a table of equivalent values of the common English weights and measures with those of the metrical system, see Appendix to this volume. 46 GENERAL PRINCIPLES OF THE SCIENCE sition of water by galvanic electricity as discovered by Nicholson and Carlisle in 1800. For the purpose of exhibiting this we only need to pass a current of electricity from four or six Grove's or Bun sen's elements by means of two platinum poles through some water acidulated with sulphuric acid (Fig. 5). The instant contact is made, bubbles of gas begin to ascend from each platinum plate and collect in the graduated tubes, which at first are filled with the acidulated water. After FIG. 5. a little time it will be seen that the plate which is in con- nection with the zinc of the battery evolves more gas than the one which is in contact with the platinum or carbon of the battery ; and after the evolution has continued for a few minutes one tube will be seen to contain twice as much gas as the other. On examination, the larger volume of gas will be found to be hydrogen, and will take fire and burn when a light INDESTRUCTIBILITY OF MATTER 47 is brought to the end of the tube in which it was collected whilst the smaller volume of gas is seen to be oxygen, a glowing chip of wood being rekindled when plunged into it. 5 In many cases of chemical action, the products are gaseous, whilst one or more of the materials acted upon are solid or liquid. Hence, in these cases, a disappearance or apparent loss of matter occurs. It has, however, been shown by many accurate experi- ments that in these cases the loss of matter is only apparent, so that chemists have come to the conclusion that matter is in- destructible, and that in all cases of chemical action in which matter disappears, the loss is apparent only, the solid or liquid being changed into an invisible gas, the weight of which is, how- ever, exactly identical with that of its component parts. We only require to allow a candle to burn for a few minutes in a clean flask filled with air in order to show that the materials of the candle, hydrogen and carbon, unite with the oxygen of the air to form, in the first place, water, which is seen in small drops bedewing the bright sides of the flask, and in the second, carbon dioxide or carbonic acid gas, whose presence is revealed to us by lime-water being turned milky. The fact that the sum of the weights of the products of combustion (water and carbon dioxide) is greater than the loss of weight sustained by the candle is clearly shown by an experiment made by means of the appa- ratus (Fig. 6), which consists of a tube equipoised on the arm of a balance. In the long vertical tube a taper is placed, the other end of the system being attached to a gasholder filled with water, which, on being allowed to run out, causes a current of air to pass through the tube, and thus maintains the combustion of the taper. The water and carbonic acid gas which are formed are absorbed by the bent tube, which contains caustic potash. After the taper has burnt for a few minutes, the apparatus is disconnected from the gasholder and allowed to vibrate freely, when it will be found to be appreciably heavier than it was before the taper had burnt, the explanation being that the excess of weight is due to the combination of the carbon and hydrogen of the wax with the oxygen of the air. 6 Another series of experiments which also show plainly the fact of the indestructibility of matter, and are of historical in- terest, are those by which it has been clearly demonstrated that the air consists of two different gases, oxygen and nitrogen. The fact of the composite nature of air was proved by Priestley 48 GENERAL PRINCIPLES OF THE SCIENCE in 1772 by setting fire, by means of a burning glass, to charcoal contained in a vessel of air. He showed that fixed air (carbonic acid gas) was produced, and that on the absorption of this fixed air by lime-water, one-fifth of the original bulk of the air disappeared, and a colourless gas remained, which did not support combustion or respiration. It was not, however, till the year 1775, after he had discovered oxygen, that Priest- FIG. 6. ley distinctly stated that this gas was contained in common air, and about the same time Scheele came to an identical conclusion from independent experiments. But the method, by which the existence of oxygen in the air was first demon- strated in the clearest way, is that adopted by Lavoisier, and described in his Traitt de Chimie. 1 Into a glass balloon (Fig. 7) having a long straight neck, Lavoisier brought 4 ounces of pure mercury ; he then bent the neck so that when the balloon rested 1 Part i. , chap. iii. LAVOISIER'S EXPERIMENTS 49 on the top of the furnace, the end of the bent neck appeared above the surface of the mercury contained in the trough, thus placing the air in the bell-jar in communication with that in the balloon. The volume of the air (reduced to 28 inches of mercury and a temperature of 10) contained in the bell-jar and balloon amounted to 50 cubic inches. The mercury in the balloon was now heated, by a fire placed in the furnace, to near its boil- ing point. For the first few hours no change occurred, but then red-coloured specks and scales began to make their appearance. Up to a certain point these increased in number, but after a while no further formation of this red substance could be noticed. After heating for twelve days the fire was removed, and the volume of the air was seen to have undergone a remarkable diminution : the volume, measured under the same conditions as before, having been reduced from 50 to between 42 and 43 cubic inches. The red particles were next carefully collected, and on weighing, were found to amount to 45 grains. These 45 grains were next introduced into a small retort connected with a graduated glass cylinder (Fig. 8), and on heating they yielded 41 J grains of metallic mercury and from 7 to 8 cubic inches of a gas which was found to be pure oxygen. Thus, the whole of the oxygen, whether measured by volume or by weight, which was withdrawn from the air by the mercury, was obtained again when the oxide formed was decomposed by heat. 5 50 GENERAL PRINCIPLES OF THE SCIENCE 7 It is not merely to the investigation of changes occurring in the essential properties of inorganic or mineral matter that the chemist has to direct his attention. The study of many of the phenomena observed in the vegetable or animal world also claim his notice. So much so, indeed, is this the case that the science of physiology has been denned as the physics and chemistry of the body. The simplest as well as the most complicated changes which accompany life are, to a great extent, dependent upon chemical laws, and, although we are still unable fully to explain many of these changes, yet each year brings us addi- tional aid, so that we may expect some day to possess an exact L & FIG. 8. knowledge of the chemistry of life. In order to be convinced that vital actions are closely connected with chemical phenomena, we only need to blow the air from our lungs through clear lime-water to see from the ensuing turbidity of the water that carbonic acid gas is evolved in large quantities during the process of the respiration of animals, and when we further observe that the higher animals are all warmer than surrounding objects, we come to the conclusion that the process of respiration is accompanied by oxidation, and that the breathing animal resembles the burning candle, not only in the products of this combustion, viz., water and carbon dioxide, but in the heat \ OF v /y ELEMENTAEY AND COMPOUND BODIES 51 which that combustion evolves, the difference being that in the one case the oxidation goes on quickly and is confined to one spot (the wick of the candle) whereas in the other it goes on slowly and takes place throughout the body. In like manner the living plant is constantly undergoing changes, which are as necessary for its existence as the act of breathing is for animals. One of the most fundamental of these changes is readily seen if we place some fresh green leaves in a bell-jar filled with spring water and expose the whole to sunlight. Bubbles of gas are observed to rise from the leaves, and these, when collected, prove to be oxygen. In presence of the sunlight the green leaf has decomposed the carbonic acid gas held in solution in the spring water, assimilating the carbon for the growth of its body and liberating the oxygen as a gas. Nor, indeed, are the investiga- tions of the chemist now confined to the organic and inorganic materials of the earth which we inhabit. Recent research has enabled him, in conjunction with his colleague the physicist, to obtain a knowledge of the chemistry as well as of the physics of the. sun and far distant stars, and thus to found a truly cosmical science. 8 It is the aim of the chemist to examine the properties of all the different substances which occur in nature, so far as they act upon each other, or can be made to act so as to produce something different from the substances themselves; to ascertain the circumstances under which such chemical changes occur, and to discover the laws upon which they are based. In thus investigating terrestrial matter it is found that all the various forms of matter with which we are surrounded, or which have been examined, can be divided into two great classes. I. ELEMENTARY BODIES. Elements, or simple substances, out of which no other two or more essentially differing substances have been obtained. II. COMPOUND BODIES, or compounds, out of which two or more essentially differing substances have been obtained. Only twenty-three elements were known during the lifetime of Lavoisier ; now we are acquainted with not less than sixty- nine. Of these, and their compounds with each other, the whole mass of our globe, solid, liquid, and gaseous, is composed, and these elements contribute the material out of which the fabric of our science is built. The science of chemistry has for its aim the experimental examination of the elements and their 52 GENERAL PRINCIPLES OF THE SCIENCE compounds, and the investigation of the laws of their combination one with another. The following is an alphabetical list of the elementary bodies known at present (1893) : LIST OF ELEMENTS.* Atomic weight. Atomic weight Aluminium . . 26-9 Molybdenum . . 95-2 Antimony . . . 119-4 Nickel . . . . 58-6 Arsenic . . . 74-4 Niobium . . 93-5 Barium . . . 135-97 Nitrogen . . 13-94 Beryllium . . 8-99 Osmium . . . . 189-3 Bismuth . . . 206-4 Oxygen . . . . 15-88 Boron .... 10-74 Palladium . . . 104-7 Bromine . . . 79-36 Phosphorus . . 30'8 Cadmium . . . 111-3 Platinum . . . 193-3 Caesium . . . 131-9 Potassium . . . 38-85 Calcium . . . 39-7 Rhodium . . 102-0 Carbon . . . 11-91 Rubidium . . . 84-8 Chlorine ... 3519 Ruthenium . . 103-0 Cerium . . . 139-0 Samarium . . . 149-0 Chromium . . 51-7 Scandium . . . 43-8 Cobalt .... 58-6 Selenium . . . 78-5 Copper .... 62-8 Silver . . . 107-13 Didymium ^' ' 142-5 ' 139-7 Silicon . . Sodium . . . . 28-2 . o 22-87 Erbium . . . 165-0 Strontium . . . 86-8 Fluorine . . . 18-9 Sulphur . . . . 31-82 Gallium . . . 69-4 Tantalum . . . 181-0 Germanium . . 71-8 Tellurium . . . 124-0 Gold 195-7 Thallium 202*6 Hydrogen . . 1-0 Thorium d\J ^ \j . . 230-7 Indium . . . 112-8 Thulium . . . 169-7 Iodine .... 125-91 Tin .... 117-2 Iridium . . . 191-7 Titanium . . . 47-7 Iron 55-6 Tungsten . . . 182-6 Lanthanum . . 137-5 Uranium . . 237-6 Lead 205-4 Vanadium 50'8 Lithium . . . 6-98 Ytterbium . *J \J O . . 172-0 Magnesium . . 24-2 Yttrium . . . . 88-0 Manganese . . 54-6 Zinc . . . 65*0 o Mercury . . . 198-9 Zirconium . . . 90-0 See also Appendix. LIST OF ELEMENTS 53 In addition to the above the existence of many other elements, discovered in certain rare Norwegian and American minerals, has been announced, among which are Decipium, Holmium, Iduminium, Norwegium, and Yttrium-a. These, however, have not as yet been very perfectly investigated, and their atomic weights and chemical relationships remain undetermined. For the sake of convenience it is customary to divide the elements into two classes the Metals and the Non-Metals, or Metalloids, a distinction which was first made about the time of Lavoisier, when only a few elements were known. Now, however, the division is a purely arbitrary one, as it is not possible to draw an exact line of demarcation between these two groups, so that there are cases in which an element has been considered as a metal by some chemists and as a non-metal by others. To the first class belong such substances as gold, silver, mercury, and tin; to the second substances which are gaseous at the ordinary temperature, such as hydrogen, nitrogen, and oxygen, together with certain solid bodies, as carbon and sulphur. The number of metals is much larger than that of the non- metals : we are acquainted with fifty-four metals, and with only fifteen non-metals. 9 In this treatise the elements will be considered in the follow- ing order, although another system of classification, termed the natural system, based upon the chemical properties of the elements, will be discussed and explained in a later volume. The natural system is, however, as yet imperfect, and its value can only be properly appreciated when a fuller knowledge of the properties of the elements has been gained. ARRANGEMENT OF THE ELEMENTS. I. NON-METALS. Symbol. Symbol. Hydrogen . . . . H Nitrogen N Phosphorus . . . . P Fluorine F Arsenic As Chlorine . . . . Cl Bromine . . . . Br Boron B Iodine I Oxygen O Sulphur S Carbon C Selenium . . . . Se Silicon Si Tellurium , . Te 54 GENERAL PRINCIPLES OF THE SCIENCE II. METALS. Symbol. Potassium . . . . K Sodium Na Lithium Li Kubidium .... Rb Caesium Cs Calcium Ca Strontium . . . . Sr Barium , . Ba Beryllium . Magnesium Zinc . . . Cadmium . Lead . . . Thallium Copper Silver Mercury Yttrium Cerium . . Lanthanum Didymium Erbium . . Ytterbium . Thulium . Samarium . Aluminium Scandium , Indium . Gallium . Be Mg Zn Cd Pb Tl Cu Ag Hg Y Ce La Di Er Yb Tu Sa , Al Sc , In , Ga Symbol. Manganese .... Mn Iron ....... Fe Cobalt Co Nickel Ni Chromium . . . . Cr Molybdenum .... Mo Tungsten ..... W Uranium U Tin Sn Titanium ..... Ti Zirconium Zr Thorium Th Vanadium V Antimony Sb Germanium .... Ge Bismuth Bi Tantalum Ta Niobium Nb Gold Au Platinum Pt Palladium Pd Rhodium Rh Iridium . . Ir. Ruthenium Osmium . . Ru . Os Of these elements only four occur largely in the air, about thirty have been detected in the sea, whilst all the sixty-nine are found irregularly distributed throughout the solid mass of our planet. Some are very abundant, and are widely distributed, whilst ARRANGEMENT OF THE ELEMENTS 55 others have hitherto been found only in such minute quantities and so seldom, that even their properties have not yet been satisfactorily examined. Thus oxygen is found throughout the air, sea, and solid earth in such quantities as to make up nearly half the total weight of the crust of our planet, whilst the com- pounds of caesium, although tolerably widely distributed, occur only in very minute quantity, and those of erbium have as yet been met with only in very small quantities, and in very few localities. In order to obtain an idea as to which elements form the main portion of the solid crust of the earth, we may examine the composition of all the different kinds of granitic or eruptive rocks which constitute by far the greater part of the earth's crust. From analyses made by Bunsen we find that all granitic rocks possess a composition varying between the limits given in the following table, so that these numbers give a fair idea of what is known of the average chemical composition of the solid globe. All the other elements occur in quantities less than any of those mentioned in the table. The Composition of the ^Earth's Solid Crust in 100 parts ly weight. Oxygen. . 44'0 to 487 Calcium. . . G'6to0'9 Silicon . . 22-8 36'2 Magnesium. . 27 01 Aluminium. 9'9 61 Sodium . . . 2'4 2'5 Iron ... 9'9 2'4 Potassium . . 17 31 10 In considering for the first time the subject of the elements, the question will at once suggest itself Are these sixty-nine all the elements which make up our earth, or is it likely that other hitherto undiscovered elements exist ? Judging from analogy, remembering what has previously occurred, and looking to the incomplete state of our know- ledge concerning the composition of the earth's crust, we may fairly conclude that it is all but certain that other elementary bodies remain to be discovered. Every improve- ment in our methods of examination leads to the detec- tion either of new elements or of old ones in substances in which they had previously been overlooked. Thus by the methods of Spectrum Analysis no less than five (caesium, rubidium, thallium, indium, and gallium) new elements have been discovered, and the existence of several 56 GENERAL PRINCIPLES OF THE SCIENCE others rendered highly probable especially among the rare earths. By help of this method we are also enabled to come to certain conclusions respecting the distribution and occur- rence of these same elements in some of the heavenly bodies, and we learn that many of the metals, and even non-metals, which are well known to us on the earth, are found in the sun and the fixed stars. The conclusion that the terrestrial ele- ments exist beyond the bounds of our planet is borne out by the chemical examination of the meteoric stones which are constantly falling upon the surface of the earth. In hundreds of these which have been examined, no single case of the discovery of an un- known element has occurred. The substances of which meteorites have been found to consist are iron, nickel, oxygen, calcium, silicon, carbon, and other well-known terrestrial elements. Another question which may here be asked is Are these elements really undecomposable substances ? and to this it may be replied, that so far as our chemical knowledge enables us to judge, we may assume, with a considerable degree of probability, that by the application of more powerful means than are at present known, chemists will succeed in obtaining still more simple bodies from the so-called elements. Indeed, if we examine the history of our science, we find frequent examples occurring of bodies which only a short time ago were considered to be elemen- tary which, upon more careful examination, have been shown to be compounds. 11 A very remarkable fact observed in the case of many elements is that they are capable of existing in more than one distinct condition, presenting totally different physical qualities. One of the most striking examples of these allotropic modifica- tions or conditions of matter (aXXo?, another T^OTTO?, away or mode) occurs with carbon, which exists as Diamond, Graphite, and Charcoal, bodies which as regards colour, hardness, specific gravity, &c., bear certainly but a slight resemblance to each other, but which, when they are burnt in oxygen, all give the same relative weight of the same product, viz., carbonic acid, thereby proving their chemical identity. 12 The Balance. As it is the aim of the chemist to examine the properties of the elements and their compounds, and as the weight-determination of a substance is of the greatest import- ance, it becomes necessary for him to ascertain with great pre- cision the proportion by weight in which these several elements combine, as well as that in which any one of them occurs in a THE BALANCE 57 given compound, and for this purpose the Balance is employed. By means of this instrument the weight of a given substance is compared with the unit of weight. It consists essentially of a light but rigid brass beam (Fig. 9), suspended on a fixed hori- zontal axis situated at its centre ; and this beam is so hung as to assume a horizontal position when unloaded. At each end of the beam scale-pans are hung, one to receive the body to be weighed and the other for the weights. When each pan is equally weighted the beam must still retain its horizontal position, but when one pan is more heavily weighted than the other, the beam will incline on the side of the heavier pan. The balance is, there- fore, a lever with equal arms, and it is evident that the weight of the substance relative to the unit weight employed is the sum of FIG. 9. the weights necessary to bring the balance into equilibrium. The two important requisites in a balance are (1) accuracy, (2) sen- sibility, and these can only be gained by careful construction, It needs but little consideration to see that in a delicate balance the friction of the various parts must be reduced to a minimum. This is usually accomplished by suspending the beam by means of an agate knife-edge, working on an agate plane, whilst the pans are attached to each end of the beam by a somewhat similar arrange- ment shown in Fig. 10. The position of the axis of suspension relatively to the centre of gravity of the beam is likewise a matter 58 GENERAL PRINCIPLES OF THE SCIENCE of consequence. If the axis of suspension and the centre of gravity in a balance were coincident, the beam would remain stationary in all positions in which it might be placed. If the axis of suspension be placed below the centre of gravity the beam would be in a condition of unstable equilibrium. Hence the only case in which the balance can be used is that in which the point or axis of suspension is above the centre of gravity, for in this case alone will the beam return to a horizontal position after making an oscillation, and in this case the balance may be considered as a pendulum, the whole weight of the beam and pans being regarded as concentrated at the centre of gravity. In order that the weight of the substance and the sum of the measuring weights in the scale-pan may be equal, it is evident that the axis of suspension must be exactly in the centre of the beam, or in other words, that the balance must have arms of equal length. It is also necessary that the balance should have great sensibility; that is, that it may be FIG. 10. moved by the smallest possible weight; for this end it is like- wise requisite that the vertical distance of the centre of gravity below the axis of suspension should be as small as possible. As the whole weight of the instrument may be regarded as con- centrated at the centre of gravity, it evidently requires a less force to act at the end of the beam to move the instrument when the distance of the centre of gravity from the point of suspension of the balance is small, than when that distance is greater, inasmuch as in the latter case the weight has to be lifted through a longer arc. The sensibility of the balance is also increased, both by increasing the length of the beam and by diminishing the weight of the beam and of the load. When, however, the beam is made either too long or too light it ceases PROPERTIES OF GASES 59 to be rigid, and a serious source of error is introduced. If great accuracy in the weighing is desired, it is advisable to have recourse to the method of weighing by vibration 1 by which the excursions of the moving beam are accurately observed instead of its approach to the horizontal position. By combining this method of vibrations with that of double weighing, which consists in reversing the position of the weights in the two pans, it is possible for a good balance which is loaded with a kilogram in each pan, to turn with 0*0007 grm. or about the 1400000 th of the weight in either pan. PROPERTIES OF GASES. 13 The chemist has to deal with matter in all its various states ; solids, liquids, and gases alike being the objects of his examination. The study of gases in particular has led to most important results in the theoretical branch of the science, the system of formulas now employed being in fact founded upon observations made upon matter in the state of gas. RELATION OF VOLUME TO PRESSURE. BOYLE'S LAW. 14 The gaseous condition of matter is well defined to be that in which it is capable of indefinite expansion. If a quantity of gas as small as we please is placed in a closed vacuous space, however large, the gas will distribute itself uniformly throughout that space. The relation between the volume and pressure of a gas, the temperature remaining constant, is expressed by the well-known law of Boyle (1662), viz., that the volume of the gas varies inversely as the pressure, or, in other words, the pressure of a gas is proportional to its density ; so that the pressure exerted on the containing vessel by two portions of any given gas is the sum of the pressures which each portion would exert if present by itself. Dalton extended this law, inasmuch as he showed that if different gases, which do not act chemically on each other, are mixed together, the pressure exerted is likewise the sum of the separate pressures of the different gases. 1 See Prof. W. H. Miller, Phil Tram. 1856, 763; and article "Balance," Watts' Dictionary, 1st Edition. 60 GENERAL PRINCIPLES OF THE SCIENCE RELATION OF VOLUME TO TEMPERATURE. DALTON'S LAW. 15 Another simple numerical law, which characterises the gaseous condition, is known as the law of Dalton, 1 but often referred to as the law of Charles or of Gay-Lussac. 2 This states that all gases heated under constant pressure expand by an equal fraction of their volume at centigrade for equal increments of temperature, one volume at becoming 1*3665 at 100 ; so that the coefficient of expansion of gases is O003665 or nearly ^T^ for an increase from to 1 centigrade. If the temperature of the gas be lowered, the volume contracts in the same proportion. When, on the other hand, the volume of a gas is kept constant and the temperature raised, the pressure of the gas increases in this same ratio; so that a mass of gas which at has a pressure of 1, has at 100 a pressure of 1*3665 if it be not allowed to increase in volume. The behaviour of substances in the gaseous state as regards pressure and temperature is distinguished by its simplicity and uniformity from that of the solid and liquid forms of matter. For in the case of solids and liquids the effect on the volume of alteration of pressure as well as of temperature is different for every substance, whilst gases are all uniformly affected. Hence we are led to conclude that the gaseous form of matter is that in which the constitution is most simple, and this result is borne out by many other considerations. THE KINETIC THEORY OF GASES. 1 6 The doctrine that heat is only a kind of motion is one which is now generally admitted, so that a hot body may be regarded as possessing a store of energy, some portion of which at any rate may be made use of to accomplish actual work. The energy of motion is termed Kinetic (from /ctveco, I move), and this energy is communicated when the body possessing it comes to rest by contact with some other body. The other form of energy 1 Dalton, Manchester Memoirs, v. 535 (1801). 2 The experiments of Gay-Lussac were published at a later date than those of Dalton. The results obtained by Charles were never published, but were verbally communicated to Gay-Lussac. Gay-Lussac, Ann. Chim. [1], 43, 137.- Dixon, Memoirs Lit. and Phil. Soc. of Manchester [4], 4, 36. THE KINETIC THEORY OF GASES 61 depending on position with respect to other bodies and not upon the condition of matter is termed Potential energy. It has been shown that in a hot body a very considerable portion of the energy arises from a motion of the parts of the body ; so that every hot body is in motion, but this motion is not one affecting the motion of the mass as a whole but only that of the molecules or small portions of the body. These molecules may consist of a collection or system of smaller parts or atoms which partake as a whole of this general motion of the molecule. The subject of the motion of the smallest particles of matter attracted the attention of the ancients, and Lucretius held that the different properties of matter depended upon such a motion. Daniel Bernoulli was the first to conceive the idea that the pres- sure of the air could be explained by the impact of its par- ticles on the walls of the containing vessel, whilst in the year 1848, Joule l showed that these views were correct, and cal- culated the mean velocity which the molecules must possess in order to bring a.bout the observed pressure. Since the above date, Clausius, Maxwell, and other physicists, have extended and completed the dynamical theory of gases. Many of the phenomena observed in gases and also in liquids, especially diffusion, prove that the large number of small particles or molecules of which these forms of matter are made up are in a constant condition of change or agitation, and the hotter a body is the greater is the amount of this agitation. According to the kinetic theory, these molecules are supposed to move with great velocity amongst one another, and, when not otherwise acted on by external forces, the direc- tion of this motion is a rectilinear one, and the velocity uniform. The molecules, however, come into frequent contact with one another, or as Maxwell describes it, encounters between two molecules occur. In these encounters, and also when the mole- cules strike the surface of the containing vessel, no loss of energy takes place, provided of course that everything is at the same temperature, so that the total energy of the enclosed system remains unaltered. 17 From these principles, assuming simply that the molecules have weight and are in motion, and applying the usual laws of masses in motion, the experimental laws of gases, already alluded to, as well as others, may be deduced. The pressure of a gas is thus due to the impacts of its molecules upon the walls 1 Brit. Assoc. Reports, 1848, 2nd Part. p. 21. 62 GENERAL PRINCIPLES OF THE SCIENCE of the containing vessel ; hence, when twice as many molecules are crowded into a given space, by the compression of the gas to half of its original volume, the frequency of the impacts, and, therefore, also the resulting pressure will be doubled, or, in other words, the pressure is inversely proportional to the volume (Boyle's Law). The temperature is measured by the kinetic energy of the molecules, which is equal to the product of half their mass into the square of their velocity i.e., \ mv 2 . Increase of temperature, therefore, means increase of the velocities of the molecules, and hence if the volume of a gas be kept constant and its temperature raised, both the force of impact of each molecule against the wall of the vessel and the number of impacts per second will be increased, the pressure rising in pro- portion to the square of the velocity, or in other words in direct proportion to the rise of temperature (Dalton's Law). The tem- perature at which the velocity of the molecules, and, therefore, also the kinetic energy, would become equal to nothing, is called the absolute zero of temperature and is found by calculation to be 273 centigrade, temperatures reckoned from this zero being termed absolute temperatures. When two gases are at the same pressure, the total kinetic energy of the molecules in equal volumes must be the same ; if their temperatures are also equal, the kinetic energy of each molecule must also be equal, and it hence follows that the number of molecules in equal volumes of the two gases must be the same. Assuming then that the tem- perature of a gas represents the kinetic energy of its molecules it follows that equal volumes of all gases under similar conditions of temperature and pressure contain equal numbers of mole- cules. This is the statement known to chemists as Avogadro's theory (p. 98), which is thus shown to rest upon a sound physical foundation. 1 8 The velocities of the molecules can be calculated for any given temperature when we know the density of the gas and its pressure at that temperature. As will be seen from the table given below their magnitude is very considerable, and this accounts for the great velocity with which disturbances, such as sound waves or explosions, are propagated through gases. VELOCITY OF MOLECULES OF GASES 63 Velocity at in metres per second. Hydrogen 1843 Ammonia 628 Oxygen 461 Carbon dioxide 392 Chlorine 310 Hydriodic acid 230 These numbers represent the mean velocity of the molecules, since it is supposed that owing to encounters the velocity is not absolutely uniform but varies about an average value; in hydrogen for example at C. and 760mm. some of the molecules are moving at a much slower rate than 1843 metres per second, whilst others are moving much more rapidly. In addition to the energy of translation which the molecule as a whole possesses, the atoms of which it is composed and which are capable of motion relatively to each other have also a certain amount of kinetic energy. When a gas is heated, there- fore, both of these kinds of energy are increased, or in other words the motion of the atoms within the molecule is increased at the same time as the velocity of the molecule itself. Clerk- Maxwell and others have calculated that the actual num- ber of molecules which we must conclude to be present in one cubic centimetre of a gas at the standard temperature and pres- sure is no less than 21 trillions (21,000,000,000,000,000,000 or 21 x 10 18 ) ; the weight of a single molecule of hydrogen being, therefore, about O'OOO 000 000 000 000 000 04 milli- grams or (0-04 x 10- 18 ). DIFFUSION OF GASES. 19 Early in the history of the chemistry of gases it was observed that when gases of different specific gravities, which exert no mutual chemical action, are once thoroughly mixed, they do not of themselves separate in the order of their several densities by long standing. On the contrary, they remain uniformly distributed throughout the mass. Priestley l proved this by very satisfactory experiments ; but he believed that if the different gases were very carefully brought together, the 1 Observations on Air, 2, 441. 64 GENERAL PRINCIPLES OF THE SCIENCE heavier one being placed beneath and the lighter one being brought on to the top without being mixed with the other, they would then, on being allowed to stand, not mix, but continue separate one above the other. Dalton, 1 in 1803 proceeded to investigate this point, and he came to the conclusion that a lighter gas cannot rest upon a heavier, as oil upon water, but that the particles of the two gases are constantly diffusing through each other until an equilibrium is reached, and this without any regard to their specific gravities. This conclusion Dalton regarded as a necessary consequence of his theory of the constitution of matter, according to which the particles of all gaseous bodies exert a repulsive influence on each other, and each gas expands into the space occupied by the other as it would into a vacuum. In fact, however, it does not so expand, for the rate at which a gas diffuses into another gas is many thousand times slower than that at which it rushes into a vacuum. 2 As was usual with him, the apparatus used by Dalton in these experiments was of the simplest kind. It consisted of a few phials and tubes with perforated corks. " The tube mostly used was one 10 inches long and of ^V inch bore ; in some cases a tube of 30 inches in length and J inch bore was used ; the phials held the gases which were the subject of experiment and the tube formed the connection. In all cases the heavier gas was in the lower phial and the two were placed in a perpendicular position, and suffered to remain so during the experiment in a state of rest; thus circumstanced it is evident that the effect of agitation was sufficiently guarded against ; for a tube almost capillary and 10 inches long, could not be instru- mental in propagating an intermixture from a momentary com- motion at the commencement of each experiment." The gases experimented on were atmospheric air, oxygen, hydrogen, nitro- gen, nitrous oxide and carbonic acid; and after the gases had remained in contact for a certain length of time the composition of that contained in each phial was determined, and invariably showed that a passage of the heavier gas upwards and the lighter gas downwards, had occurred. Similar experimental results were also obtained by Berthollet in 1809. 3 The passage of gases through fine pores was likewise observed by Priestley 4 in the case of unglazed earthenware retorts which although perfectly air-tight so as not to allow of any escape by 1 Manch. Memoirs, 1805, p. 259. 2 Phil. Mag. 4, 26, 409. 3 Him. d'Arcueil, 2, 463. < Observations, I loosen) is the following. A slow current of the detonating gas obtained by the electrolysis of water, and 68 GENERAL PRINCIPLES OF THE SCIENCE consisting of 2 volumes of hydrogen to one volume of oxygen, is allowed to pass through a common long clay tobacco-pipe, the gas on issuing from the pipe being collected over water in a pneumatic trough. On bringing the gas, thus collected, in m FIG. 11. contact with a flame it no longer detonates. On the contrary, it will rekindle a glowing chip of wood, thus showing that in its passage through the porous pipe the greater portion of the lighter hydrogen has escaped by diffusion through the pores of the clay, whilst the heavier oxygen has not passed through. DIFFUSION OF GASES 69 A third experiment to illustrate the law of diffusion is one which possesses interest from another point of view, inasmuch as it has been proposed to employ the arrangement for giving warning of the outbreak of the dangerous and explosive gas termed fire-damp by the coal-miners. Fire-damp or marsh gas is lighter than air, and 134 volumes of this gas will diffuse through a porous medium in the same time as 100 volumes of air will do. Hence if a quantity of fire-damp surround the porous plate, the volume within the vessel will become larger, and this increase of volume may be FIG. 12. made available either to drive out water as in the first experi- ment or to alter the level of a column of mercury so as to make contact with a connected battery and then to ring a warning bell. The latter form of apparatus is seen in Fig. 12. Holding a beaker-glass (A) filled with hydrogen or common coal-gas over the plate of porous stucco fastened into the tube- funnel an increase of volume occurs inside the glass tube and a consequent depression of the mercury takes place in the bend of the tube which is sufficient to make metallic contact with a second platinum wire fused through the glass and to bring the GENERAL PRINCIPLES OF THE SCIENCE current to act on the magnet of the electric-bell (B). The other tube is arranged for showing that a dense gas, such as carbonic acid, does not diffuse through the porous septum so quickly as air escapes. By immersing the porous plate in a jar (c) filled with the heavy gas, the volume inside the tube becomes less, the level of the mercury in the bend is altered, contact is again made with the battery, and the ringing of the bell gives notice of the change. 22 Effusion of Gases is the name given by Graham to the flow of gases under pressure through a minute aperture in a metallic plate. The law of diffusion is found to hold good with regard to this molecular motion of gases, the times required for equal volumes of different gases to flow through an aperture of a diameter of ^^ of an inch having been found to be very nearly proportional to the square roots of their densities, and the velocity of flow to be inversely as the square roots of their densities. This law, which is true for the flow of all fluids through a small aperture in a thin plate has been applied by Bunsen l for the purpose of determining the specific gravity of gases, the method serving admirably when only small quantities of the gas can be obtained. DEVIATIONS FROM THE LAWS OF BOYLE AND DALTON. 23 These laws are only approximately true, for it is found that no gas exactly follows them under all conditions of temperature and pressure, although under moderate variations of these conditions the deviations are but slight. The ideal gas which would agree with these laws under all conditions is called a perfect gas, and this term is often also applied to gases which approach this state. The effect of high pressures upon certain gases is shown in the following table exhibiting the experi- mental results obtained by Natterer. 2 - l Gasometry, p. 121. 2 Pogg. Ann. 94, 436. DEVIATIONS FROM BOYLE'S LAW 71 Table of Deviations from the Law of Boyle. Pressure in Number of volumes of gas compressed into one volume. Atmospheres. Hydrogen. Oxygen. Nitrogen. Air. Nitric Oxide. 1 1 1 1 1 1 50 50 50 50 50 50 100 98 100 99 100 100 500 396 439 381 396 412 1,000 623 595 519 527 544 1,500 776 590 607 617 2,000 899 641 661 669 3,600 1,040 710 800 : Two factors seem to be involved in producing deviations of this character. In the first place, as the density of the gas 40 30 40 80 P. in Atmospheres A Hydrogen. 120 160 200 240 280 320 B Ethylene. FIG. 13. C Carbon dioxide. increases, the volume occupied by the molecules themselves, which we must suppose to be unalterable by pressure, bears a greater proportion to the distance between them, which can be lessened by compression, and consequently the volume diminishes less rapidly than when the density is smaller. Secondly, as the 72 GENERAL PRINCIPLES OF THE SCIENCE molecules approach one another more closely, they tend to cohere, and the volume is thereby diminished more rapidly than the pressure increases. The actual effect produced is generally due to a combination of these two causes ; thus the volume of carbon dioxide at a temperature of 100 decreases more rapidly than we should expect from Boyle's law until a pressure of 160 atmo- spheres is reached, after which it decreases less rapidly. The behaviour of several gases is shown in Fig. 13, on the preceding page. In every case the value of the product of pressure and volume should remain invariable, according to the law of Boyle, whilst the actual values are seen to deviate considerably from this theoretical constancy. Deviations of a similar character are observed when the density of a gas is increased by lowering its temperature. If any gas be subjected to a greatly increased pressure and its temperature simultaneously greatly reduced, a point is at last reached at which the gas undergoes sudden contraction and becomes converted into a liquid. Just before liquefaction the gas may be considered as the vapour of a liquid, so that carbon dioxide at a high pressure and a low temperature may be compared to steam just about to condense. THE CONTINUITY OF THE GASEOUS AND LIQUID STATES OF MATTER. 24 It is matter of everyday experience that the tension of the vapour of water and of other liquids heated with excess of the liquid in closed vessels, increases in a Very rapid ratio with increase of temperature, and that the density of the steam or vapour in such a case undergoes a similar rapid increase. Thus at 231 the weight of a cubic metre of steam is -gV part of the weight of the same bulk of water at 4, the point of maximum density, the weight of steam at 100, being only TT Vir of that of the same bulk of water, so that at a temperature not very far above 230 the weight of the vapour will become equal to that of the liquid. The result of this must be that under these circum- stances a change from the gaseous to the liquid state is not accompanied by any condensation, and in such a case the distinc- tions we have been in the habit of drawing between these conditions of matter cease to have any meaning. So long ago as ] 822, Cagniard de la Tour 1 made experiments upon the action of 1 Ann. Chim. Phys. [2], 21, 127, 22, 410. THE CRITICAL TEMPERATURE 73 liquids sealed up in glass tubes of a capacity but little greater than that of the liquid. When a tube one-fourth filled with water was heated up to about 360 the water entirely disap- peared, the tube appearing empty, and as the vapour cooled a point was reached at which a kind of cloud made its appearance, and a few moments later the liquid was again visible. Cagniard de la Tour considered that the substance when thus heated assumes the gaseous condition ; but Dr. Andrews 1 has shown that in such an experiment the properties of the liquid and those 01 the vapour constantly approach one another, so that above .a given temperature the properties of the two states can- not be distinguished. Hence it follows that at all temperatures above this particular one no increase of pressure can bring about the change by condensation which we term liquefaction. This temperature is called by Andrews the critical point. For many gases, the critical points are situated at temperatures which render the observation of this fact easy. Thus, for example, the critical temperature for carbon dioxide (or carbonic acid gas) is 30'92, and at all temperatures above this point no condensation from gas to liquid occurs ; so that if the pressure on the gas be gradually increased up to 150 atmospheres a steady diminution of volume occurs as the pressure augments, and no sudden diminution of volume will occur in any stage of it. The temperature may then be gradually allowed to fall until the carbon dioxide has reached a temperature below the critical one when it is found to be a liquid ; thus beginning as a gas and by a series of gradual changes, pre- senting nowhere any abrupt alteration of volume or sudden evolution of heat, it ends by being an undoubted liquid. This clearly shows that the properties of a gas can be continually and imperceptibly changed into those of a liquid. The tension of the substance at the critical temperature is known as the critical pressure. The critical temperatures of a few gases and liquids are given in the following table : 1 Phil. Trans. 1869, Part 2, 575. 74 GENERAL PRINCIPLES OF THE SCIENCE Oxygen -113 Ethylene +13 Carbon dioxide . . . +30'9 Nitrous oxide . . . +36 Chlorine +141 Alcohol +234 Benzene +281 LIQUEFACTION OF GASES. 25 The first instance of a substance, which, under ordinary conditions, is known as a gas, being transformed by pressure into a liquid is chlorine gas. This gas was first liquefied under pres- sure by Northmore l in 1806. Faraday investigated the subject fully shortly afterwards 2 showing that many other gases such as sulphur dioxide, carbonic acid, euchlorine, nitrous oxide, cyanogen, ammonia, and hydrochloric acid can also be reduced to the liquid state. In these experiments Faraday em- ployed bent tubes made of strong glass, in the one limb of which, being closed at the end, materials were placed which on being heated will yield the gas ; the open limb of the tube was then hermetically sealed and the gas evolved by heating the other end. The pressure or tension exerted by the gas itself, when thus generated in a closed space is sufficient to condense a portion into the liquid state. The following table shows the maximum tensions of some of these more readily liquefiable gases at : Table of Tensions. Atmospheres. Atmospheres. Sulphur dioxide . . 1'53 Sulphuretted hydrogen 10' 00 Cyanogen .... 2'37 Hydrochloric acid . . 26'20 Hydriodic acid . . 3*97 Nitrous oxide . . . 32*00 Ammonia . v . . . 4*40 Carbon dioxide . . . 38'50 Chlorine .... 8'95 If, therefore, any of the above gases at be exposed to pres- sures exceeding those given in the table they will condense to liquids, their critical temperatures being all above 0. The liquefaction of gases can be brought about not merely by 1 Northmore, Nicholson's Journal, 12, 368 ; 13, 232. " Phil. Trans. 1823, 160. Ibid, 1823, 189. LIQUEFACTION OF GASES 75 simple exposure to high pressure but also to low temperature ; l thus, if we reduce the temperature of sulphur dioxide, under the ordinary atmospheric pressure, to - 10 it liquefies, and when the temperature sinks to 76, the liquid freezes to an ice-like mass. 26 Until the important researches of Pictet 2 and Cailletet 3 certain gases, such as oxygen, hydrogen, and nitrogen, had resisted all attempts to reduce them to the liquid or solid state, because their critical temperatures had never been reached, and to these the name of the permanent gases was given. The meeting of the French Academy of the 24th December, 1877, was a memorable one. On that day the Academicians were told that Cailletet had succeeded in liquefying both oxygen and carbon monoxide at his works at Chatillon-sur-Seine, and that the former gas had also been liquefied by Raoul Pictet at Geneva. These experimenters soon succeeded in condensing the other gases already named, with the exception of hydrogen, which has, however, since yielded to the lower temperatures produced by later workers, and thus we are able to give experimental proof of the view which has been frequently expressed that all bodies without exception possess the power of cohesive attraction. These important results were arrived at independently by both observers, each having made the question the subject of many years' study and experiment. It is difficult, on reading the description of these experiments, to know which to admire most, the ingenious and well-adapted arrangement of the apparatus employed by Pictet, or the singular simplicity of that used by Cailletet. The latter gentleman is one of the greatest of French ironmasters, whilst the former is largely engaged as a manu- facturer of ice-making machinery, and the experience and practical knowledge gained by each in his own business have materially assisted to bring about one of the most interesting results in the annals of scientific discovery. The process successfully adopted in each case consisted in simultaneously exposing the gas to a very high pressure and to a very low temperature. The increase of pressure was effected by Pictet by evolving the gas in a wrought-iron vessel strong enough to withstand an 1 Faraday, Phil. Trans. 1845, 155. 2 Compt. Rend, 85, 1214, 1220. 3 Ibid. 815. 76 GENERAL PRINCIPLES OF THE SCIENCE enormous tension; whilst in Cailletet's arrangement the same end was brought about by a hydraulic press. For the purpose of obtaining a low temperature the first experimenter made use of the rapid evaporation of liquid carbon dioxide, thus producing a constant temperature of 130. Cailletet, on the other hand, effected the same end by suddenly diminishing the pressure upon the compressed gas. This sudden expansion gives rise to a rapid diminution of temperature caused by the transference of heat into the motion of the particles of the expanding gas (chaleur de detente). So great is the amount of heat thus absorbed that the temperature of the particles sinks below the critical point of the gases, and a condensation occurs, the finely divided liquid oxygen or nitrogen appearing as a mist in the tube. PICTET'S METHOD OF LIQUEFYING GASES. 27 A vertical section and a ground plan of Pictet's apparatus are shown in Figs. 14 and 15. p and p' are two pumps, p' being an exhausting, and p a compressing pump, such as are used in the ice-making machines. These are employed respectively for the volatilization and condensation of liquid sulphur dioxide, contained in the outer inclined double- jacketed tube (R) Fig. 15 ; and both pumps are so arranged that there is the largest possible amount of difference of pres- sure always kept up between the two cylinders. In the tube (R) the pressure is so regulated that the liquid sulphur dioxide evaporates at a temperature of 65. The gaseous sulphur dioxide passes through the pumps and is condensed to a liquid by a stream of cold water which surrounds the reservoir (c) at a tempera- ture of + 25, and under a pressure of 2'75 atmospheres. The liquid then flows back through the small tube (z) into the tube (R). (o) and (o) are two smaller pumps which act upon liquid carbon dioxide which is contained in the tube (s). These pumps are so arranged that the evaporation of the liquid takes place from the tube (s) at a temperature of 140, this being surrounded by the liquid sulphur dioxide, and flowing under a pressure of five atmospheres through the tube (s) into the tube or cylinder surrounding the tube (A). All these portions of the apparatus are made of strong cold-drawn copper tubes able to resist a high pressure. (B) is a strong wrought iron PICTET'S EXPERIMENTS 77 78 GENERAL PRINCIPLES OF THE SCIENCE PICTET'S EXPERIMENTS retort employed for the evolution of the gas about to be con- densed. The gas thus generated passes into the long thin copper condensation-tube (A), four metres in length, which is surrounded by a bath of liquid carbon dioxide at a tem- perature varying from 120 to 140. The end of this condensation-tube is provided with a well-fitted stopcock (v) and a Bourdon's manometer at (m), capable of indicating a pressure up to 800 atmospheres. With this apparatus oxygen was first condensed on the 22nd December, 1877. The following description of the experiment will render intelligible the working of the process : (1.) 9 A.M. The pumps for condensing and rarefying the sulphur dioxide were set to work. (2.) 9.30. The temperature of the upper tube was 55. The pumps for condensing and rarefying the carbon dioxide were started. (3.) 10.40. Temperature inside the tubes 60; pressure, 5 atmospheres. 800 litres of carbon dioxide have been liquefied. (4.) 11.0 The shell containing the chlorate of potash is now heated. (5.) 11.15. The temperature of the carbon dioxide sinks to 130. The pressure of oxygen in the copper tube = 5 atmospheres. The pressure then began gradually to rise, and at last it remained constant, as is seen in the following table : (6.) 12.10 P.M. Pressure of oxygen 50 ats. ; temp, as before. (7.) 12.36 ..... 100 (8.) 12.37 ..... 200 (9.) 12.38 ..... 460 (10.) 12.40 ..... 525 (11.) 12.42 ..... 526 (12.) 12.44 ..... 525 (13.) 1.0 ..... 471 (14.) 1.5 ..... 475 The pressure now remained constant. The whole of the interior of the glass tube was filled with liquid oxygen. On opening the stopcock at the end of the oxygen tube, a lustrous jet of liquid oxygen issued with great violence, whilst around it was a haze of particles of what was taken to be solid oxygen. 80 GENERAL PRINCIPLES OF THE SCIENCE CAILLETET'S EXPERIMENTS 81 The general arrangement and appearance of Pictet's apparatus is shown in Fig. 16 in which the end of the oxygen condensa- tion-tube (A), the stopcock (v), and the manometer (m) are seen. CAILLETET'S PROCESS FOR LIQUEFYING GASES. 28 The apparatus employed by M. Cailletet 1 for the lique- faction of oxygen is shown in Fig. 17. The first part of this apparatus consists of a powerful hydraulic press, the soft steel cylinder (A) of which is fixed by the bands (BB) on to a horizontal cast-iron bed. A steel piston works into FIG. 17. this cylinder through a stuffing-box, and to the end of the piston is attached the screw (F), which can be carried backwards or forwards by turning the wheel (M) either in one direction or the other. The hydraulic cylinder is filled with water from the vessel (G), and this communicates with the interior of the cylinder by a fine opening, which can be perfectly closed at pleasure by means of a conical valve attached to a piston worked by the wheel (o). On withdrawing this piston water flows into the cylinder. The second part of the apparatus is the receiver (Fig. 19). 1 Compt. Rend., 10, 85, 815 ; and Ann. Chim. Phys. [5], 15, 132. 7 82 GENEEAL PRINCIPLES OF THE SCIENCE This consists of a glass tube with reservoir at the lower end firmly bedded into a steel head (B, Fig. 19), sufficiently strong to resist a pressure of 1,000 atmospheres. This receiver is placed in direct connection with the hydraulic pump by means FIG. is. of a flexible metallic tube (TU) of small diameter. A steel head is firmly screwed on to the upper part of the receiver by the screw (E 1 ) and this head carries the glass tube (T), which FIG. 19. contains the gas to be experimented upon. The shape and mode of fixing this tube with its reservoir of gas is seen in Fig. 18; whilst in Fig. 19 the same is shown placed in position with the lower portion dipping into the mercury which fills the CAILLETET'S EXPERIMENTS lower part of the steel receiver. As the glass reservoir is exposed to the same pressure on both its inside and outside surfaces, its dimensions may be made large in spite of the extremely high pressures to which it is subjected in the course of the experiments. The thin tube, on the other hand, which passes out above the steel head of the condenser has of course to support the pressure necessary for the condensation of the gases, and hence it must be made of strong glass with a capillary bore. A glass cylinder (M) resting on tfee iron flange (s) serves to enable the experimenter to surround the tube either with a freezing-mixture or with a warm liquid. When the reservoir FIG. 20. has been filled with the pure dry gas under examination the end of the tube is carefully hermetically sealed, and the whole screwed into position. Water is then forced into the receiver from the hydraulic cylinder ; this forces the mercury into the reservoir, and the compressed gas condenses in the capillary tube, where the changes which occur can be readily observed. The position of the receiver and capillary tube is shown at a and ra, Fig. 17. The pressure is measured by two manometers (N and N 1 , Fig. 17). With this apparatus Cailletet liquefied ethylene at + 4 under a pressure of 46 atmospheres; acetylene under the ordinary 84 GENEEAL PRINCIPLES OF THE SCIENCE temperature at a pressure of 86 atmospheres ; nitric oxide and marsh gas required to be cooled to 11, and these became liquid at the respective pressures of 104 and 108 atmospheres. Oxygen and carbon monoxide remained gaseous at a temperature of 29 under a pressure of 300 atmospheres, as did nitrogen at a temperature of +13 and under a pressure of 200 atmospheres. When, however, this pressure was suddenly reduced a thick mist was formed in the tubes, and this con- densed, forming small drops. Hydrogen, on the other hand, appeared, when the pressure from 300 atmospheres was suddenly removed, in the form of a slight mist, but dried air liquefied under a pressure of 200 atmospheres after it had been well cooled with liquid nitrous oxide. Air has also recently been solidified (Dewar). 29 An apparatus for exhibiting the liquefaction of the difficultly condensible gases has been constructed by Messrs. Ducretet et Cie. of Paris. 1 The condensing arrangements of this apparatus are seen in Fig. 20, the apparatus of the receiver and glass tube and reservoir being identical with that just described (Fig. 20). The piston of the hydraulic cylinder is worked by the lever (L), and by this means a pressure of 200 atmospheres can be reached. In order to increase this pressure up to 300 atmospheres a steel plunger can be slowly forced into the cylinder by means of the first wheel. The second wheel works a second plunger, by the withdrawal of which the pressure can be suddenly diminished, and thus the temperature so reduced that the gas is liquefied by the intense cold produced. The method now adopted for the liquefaction of oxygen, air, nitrogen, &c., is a modification of that of Pictet, the gas being strongly compressed by a powerful pump and simultaneously cooled by liquid ethylene boiling under reduced pressure, by means of which temperatures of -100 to -150 can be maintained. 2 1 Rue des Feuillantines, 89. 2 Wroblewski, Compt. Rend. 98, 304 ; Olszewski, Compt. Rend. 98, 365, 101, 338 ; Dewar, Proc. Roy. Institution, 13, part 3, p. 695. BOILING-POINTS OF LIQUIDS 85 BOILING-POINTS OF LIQUIDS. 30 It is evident from what has been said that the distinction between gas and vapour is only one of degree, for a vapour is simply a gas below its critical temperature. The same laws according to which the volumes of gases vary under change of temperature and pressure apply also to vapours, at any rate when they are examined at temperatures considerably above their points of condensation. When a gas or vapour is near this point its density increases more quickly than the pressure, and as soon as the point is reached the least increase of the pressure brings about a condensation of the whole to a liquid. The tempera- ture at which the liquid again assumes the gaseous form is termed the boiling-point of the liquid, and at this tempera- ture the tension of the vapour is equal to the superincumbent pressure. The following table gives the boiling-points of some well-known bodies under a pressure of 760 mm. of mercury : Table of Boiling -Points. Propionic acid Oxygen . . . . - Nitrous oxide Carbonic acid . - Cyanogen . . . - Sulphurous acid - Ethyl chloride . - Prussic acid . . Ether Carbon disulphide Chloroform . . Bromine .... Alcohol .... Benzene .... Water .... Acetic acid 105 78 21 10 12-5 26'5 34 c -5 43'3 61 63 78-4 80-3 100 118 Butyric acid . . Aniline .... Phenol .... Napthalene . . Phosphorus . . Sulphuric acid . Mercury .... Sulphur .... Stannous chloride Zinc bromide . Zinc chloride . . Cadmium . . . Zinc + 141 162 182 183 218 290 332 350 440 606 650 730 860 1049 31 Liquids possess a notable tension below their boiling- points ; thus water gives off vapour at all temperatures, and even slowly evaporates when in the solid state, for the tension of the vapour coming from ice at 10 is O208 mm. According to the 86 GENERAL PRINCIPLES OF THE SCIENCE experiments of Faraday, there is, however, a limit to evapora- tion ; thus he found that mercury, which gives out a perceptible amount of vapour during the summer, emits none in the winter ; and that certain compounds which can be volatilized at 150 undergo no evaporation when kept for years at the ordinary temperature. The tension of the vapour of a liquid is constant for a given temperature, and this amount, which is always reached when an excess of the liquid is present, is termed the maximum tension of the vapour. Dalton 1 in 1801 discovered that this maximum tension or density of a vapour is not altered by the presence of other gases, or, in other words, that the quantity of a liquid which will evaporate into a given space is the same whether the space is a vacuum or is filled with another gas. The same philosopher also believed that the vapours of all liquids possessed an equal tension at tempera- tures equally distant from their boiling-points. Regnault 2 has, however, shown by exact experiments that the above conclusions can only be considered as approximately true, inas- much as he found that about 2 per cent, more vapour ascends into a space filled with gas than into a vacuum, whilst at considerable but equal distances from the boiling-point the tensions of volatile liquids are by no means equal. LAWS OF CHEMICAL COMBINATION. 32 The composition of a chemical compound can be ascertained in two ways : (1) By separating it into its component elements, an operation termed analysis (ava\vco, I unloose), and (2) by bringing the component elements under conditions favourable to combination, an operation termed synthesis (ffwrldrjiu, I place together). In both of these operations the balance is employed ; the weight of the compound and of the components in each instance must be ascertained, except indeed in the case of certain gases of known specific gravity, when a measurement of the volume occupied by the gas may be substituted for a deter- mination of its weight. It is one of the aims of analytical chemistry to ascertain with great precision the percentage composition of all chemical sub- 1 Manch. Memoirs, 1st series, 5, 535. 2 Memoires de I'Acad. des Sciences, 21, 465. LAWS OF CHEMICAL COMBINATION 87 stances, and this branch of inquiry is termed quantitative analysis, as contradistinguished from that which has only to investigate the kind of material of which substances are com- posed, and which is hence termed qualitative analysis. COMBINATION BY WEIGHT. 33 The first great law discovered by the use of the balance, is that the elements combine with one another in a limited number of definite proportions, this number being almost invariably found by experiment to be a small one. When two elements are brought together under such conditions that they can combine, it is always found that one or more of a small number of com- pounds is produced, the particular substance or substances formed depending upon the special circumstances of the experi- ment. Thus carbon is found to be capable of uniting with oxygen in two different proportions, producing two distinct sub- stances, carbonic acid gas and carbon monoxide, these being the only compounds of carbon with oxygen which are known. Some elements, on the other hand, only form one compound with each other, whilst others again form a larger number. Each one of these compounds is found to have a fixed composition, contain- ing the elements of which it is made up in a definite proportion by weight, and this fixity of composition is used as a character- istic of a chemical compound as opposed to a mere mechanical mixture, the constituents of which may be present in any vari- able proportions. In whatever way the conditions under which the elements are made to combine may be varied, it is always found that they unite in exactly the same ratio, unless, as some- times happens, the changed conditions are favourable to the production of one of the small number of other compounds which can be formed by the same elements. Thus, for instance, the combination of silver with chlorine has been brought about in no less than four different ways, but in every case it was found that the resulting compound contained 107*13 parts of "silver for 35 ' 1 9 of chlorin e. 1 The combination of chlorine with phosphorus, on the other hand, takes place in two distinct ratios, so that when an excess of phosphorus is present, the resulting compound con- tains 10*27 parts of this element for 35*19 parts of chlorine, whilst if the latter be kept in excess, this weight of it only combines 1 Stas, Eecherches, etc. pp. 108, 210. 1865. 88 GENERAL PRINCIPLES OF THE SCIENCE with 6'16 parts of phosphorus. These are, however, the only two compounds of these elements which are known. In like manner hydrogen combines with oxygen to yield water, a substance which contains 88*81 parts of oxygen to 11*19 of hydrogen. If these elements are brought together in proportions differing from those in which they are present in water, the excess of one element remains in the free state ; thus, if 98*8 L parts of oxygen by weight be brought together with 11-19 parts of hydrogen under circumstances in which they can combine, 88'81 parts of the oxygen will combine with all the hydrogen to form 100 parts of water, whilst 10 parts of oxygen remain in the free state. It will therefore be seen that the chemical combination of two or more elements does not result in the production of a series of compounds varying gradually in composition, according to the conditions of the experiment, but yields one or more compounds, each of which contains its constituents in a perfectly fixed and definite ratio. 34 As has been said, the case frequently occurs of two elements uniting to form several compounds, for each of which the law of definite proportion holds good. The special relations which exist between the weights of the two elements entering into combination were first discovered by John Dalton, and indeed form the basis of his atomic theory. Thus the two elements, carbon and oxygen, unite to form two distinct compounds, carbonic oxide gas and carbonic acid gas, and 100 parts of each of these bodies are found by analysis to contain the following weights of the elements : Carbonic Oxide Gas. Carbonic Acid Gas. Carbon .... 42*86 . . . 27'27 Oxygen .... 5714 . . . 7273 100-00 100-00 Knowing these facts Dalton asked himself what was the relation of one element (say of the oxygen) in both compounds when the other element remains constant ? He thus found that, in proportion to the carbon, the one compound contained exactly double the quantity of oxygen which the other contained; thus : LAWS OF CHEMICAL COMBINATION Carbonic Oxide Gas. Carbonic Acid Gas. Carbon ..... lO'O . . . lO'O Oxygen ..... 13'3 . . . 26'6 23-3 36'6 Thus again, analysis showed that two compounds which carbon forms with hydrogen, viz., marsh gas and olefiant gas, have the following percentage composition : Marsh Gas. Olefiant Gas. Carbon .... 74'95 . . . 85'68 Hydrogen . . . 25'05 . . . 14'32 100-00 100-00 Dalton then calculated how much hydrogen is combined in each compound with 10 parts by weight of carbon, and he found that in olefiant gas there are 1*67 parts by weight of hydrogen to 10 of carbon, whilst marsh gas contains 3'34 parts of hydrogen to the same quantity of carbon, or exactly double as much. As another example we may take the compounds of nitrogen and oxygen, of which no less than five are known to exist. The percentage composition of these five bodies is found by experi- ment to be as follows : (1) (2) (3) (4) (5) Nitrogen . . 63-71 46-75 36-91 30-51 25-99 Oxygen . . 36-29 53-25 63-09 69-49 74-01 100-00 100-00 100-00 100-00 100-00 If then, like Dalton, we inquire how much oxygen is con- tained in each of these five compounds, combined with a fixed weight, say 10 parts of nitrogen, we find that this is represented by the numbers 5'7, 1T4, 171, 22'8, and 28'5. In other words, the relative quantities of oxygen are in the ratio of the simple numbers 1, 2, 3, 4, and 5. 35 The above examples illustrate the relations exhibited in the combination of two or more of the elements to form compounds, but a careful examination of the quantitative composition of a 90 GENERAL PRINCIPLES OF THE SCIENCE whole series of chemical compounds leads to a further conclusion respecting the nature of the laws of chemical combination which is of the highest importance. Let us examine the com- position of any given series of compounds as determined by analysis, such as the following : CHLORIDES. Hydrogen Chloride. Chlorine .... 97'24 Hydrogen. . . . 276 100-00 Potassium Chloride. Chlorine .... 47'53 Potassium 52'47 100-00 Sodium Chloride. Chlorine .... 60'61 Sodium .... 39-39 100-00 Silver Chloride. Chlorine .... 2473 Silver . 75-27 100-00 BROMIDES. Hydrogen Bromide. Bromine .... 98'76 Hydrogen. . . . 1'24 100-00 Sodium Bromide. Bromine .... 77'62 Sodium 22-38 loo-oo Potassium Bromide. Bromine .... 67*13 Potassium . . . 32'87 100-00 Silver Bromide. Bromine .... 42'55 Silver . 57'45 100-00 IODIDES. Hydrogen Iodide. Iodine 99'21 Hydrogen. . . . 0'79 100-00 Potassium Iodide. Iodine 76 '42 Potassium 2358 100-00 LAWS OF CHEMICAL COMBINATION 91 Sodium Iodide. Iodine 84'62 Sodium . . . . 15-38 100-00 Silver Iodide. Iodine 54*03 Silver . 45-97 100-00 Arranged in this way we do not notice any simple relation existing between the components of this series, except that the quantity of hydrogen is always smaller than that of the chlorine, bromine, or iodine, whilst the quantity of sodium is always smaller than that of potassium, and this again is less than the quantity of silver. 36 If, however, instead of examining a constant weight of the several compounds we ask ourselves, how much of the one constituent in each compound combines with a constant weight of that constituent which is common to several, we shall obtain at once a clear insight into the law of the formation of the compound. In the series of hydrogen com- pounds for instance, let us calculate (by simple proportion) how much chlorine, bromine, and iodine combine with the unit weight of hydrogen. We then obtain for the composition of these compounds : Hydrogen Chloride. Chlorine . . 3519 Hydrogen . 1-00 Hydrogen Bromide. Bromine . 79 '36 Hydrogen . TOO Hydrogen Iodide. Iodine . 125*91 Hydrogen . 1*00 3619 80-36 126-91 Continuing our calculation, let us next ask how much of the metals, potassium, sodium, and silver, unite with 35 "19 parts by weight of chlorine to form chlorides ; with 79*36 parts of bro- mine to form bromides, and with 125*91 parts of iodine to form iodides. The result is as follows : Potassium Chloride. Chlorine . 3519 Potassium . 38*85 CHLORIDES. Sodium Chloride. Chlorine . 35*19 Sodium 22-87 Silver Chloride. Chlorine . 3519 Silver . . 10713 74-04 58-06 143-03 92 GENERAL PRINCIPLES OF THE SCIENCE Potassium Bromide. Bromine . 79*36 Potassium . 38 -85 118-21 BROMIDES. Sodium Bromide. Bromine . 79*36 Sodium . 22-87 102-23 Silver Bromide. Bromine . 79*36 Silver . . 107'13 186-49 Potassium Iodide. Iodine . . 125'91 Potassium. 38*85 16476 IODIDES. Sodium Iodide. Iodine . . 125'91 Sodium . 22-87 148-78 Silver Iodide. Iodine . . 125*91 Silver . 107*13 233-04 Now for the first time a remarkable relation becomes apparent, for it is clear that the SAME weights of the metals, potassium, sodium, and silver, which combine with 35*19 parts of chlorine to form chlorides, also combine with 79'36 parts of bromine to form the bromides, and with 125*91 parts of iodine to form the iodides. In other words, if we replace the 35*19 parts by weight of chlorine in each of these compounds by 79*36 parts of bromine, we get the bromides, and if by 125 '91 parts of iodine we obtain the iodides of the metals. Hence one and the same weight of metals (38*85 of potassium, 22*87 of sodium, and 107*13 of silver) has the power of forming compounds with the precise quantities of chlorine, bromine, and iodine respectively, which unite with 1 part by weight of hydrogen, to form the hydrides of these elements. These quantities of the elements in question are called equivalent quantities, because they are the amounts of them which will combine with the same weight of some other element. ( 35*19 of chlorine ) 38*85 of potassium combine with^ 79'36 bromine ( 125*91 ,, iodine 22*87 sodium 107*13 silver 1*00 hydrogen tlve1 ?' Similar results are obtained from the examination of the compounds of all the other elements, so that a number may be LAWS OF CHEMICAL COMBINATION 93 -assigned to each element which is termed the combining weight or equivalent weight of the element. 37 Taking an example from another group of chemical com- pounds we find that the well-known oxides of hydrogen, lead, copper, mercury, and cadmium possess the following percentage composition : Water. Hydrogen Oxygen . 1119 88-81 100-00 OXIDES. Lead Oxide. Lead . . 92*82 Oxygen . 718 Copper Oxide. Copper . 79-82 Oxygen . 2018 100-00 Mercury Oxide. Mercury . . . . Oxygen . . . . Cadmium Oxide. 92-60 7-40 100-00 Cadmium Oxygen . 100-00 87-49 12-51 100-00 Whilst the corresponding sulphides exhibit the following composition : SULPHIDES. Sulphuretted Hydrogen. Lead Sulphide. Copper Sulphide. Hydrogen . 6'01 Lead. . . 86'58 Copper . . 6676 -Sulphur . . 93-99 Sulphur . 1342 Sulphur . 33*24 100-00 Mercury Sulphide. Mercury .... 86'20 Sulphur .... 13-80 100-00 100-00 100-00 Cadmium Sulphide. Cadmium . . . 77'73 Sulphur .... 22-27 100-00 38 If, as before, we now compare the quantity of each element united with one and the same weight of oxygen, taking 7"94 parts of this element, because this is the amount of it which combines with one part of hydrogen and is, therefore, the equivalent weight of oxygen, we get the following numbers : 94 GENERAL PRINCIPLES OF THE SCIENCE Water. Lead Oxide. Hydrogen . TOO Lead . . 102*68 Oxygen . . 7'94 Oxygen . 7*94 8-94 110-62 Copper Oxide. Copper . . 31-41 Oxygen . . 7*94 39-35 Mercury Oxide. Mercury . Oxygen . 99-45 7-94 107-39 Cadmium Oxide. Cadmium . . . 55'63 Oxygen .... 7 '94 63-57 And, if we investigate the sulphides, we find that one and the same weight of sulphur, viz. 15'64 parts by weight unites with weights of these elements to form sulphides, which are identi- cal with the amounts that combined with 7 '94 parts by weight of oxygen to form oxides. Thus we have : Sulphuretted Hydrogen. Hydrogen . TOO Sulphur . .15-64 16-64 Lead Sulphide. Lead . . 102*68 Sulphur . 15-64 118-32 Copper Sulphide. Copper . . 31-41 Sulphur . 15-64 47-05 Mercury Sulphide. Mercury .... 99'45 Sulphur .... 15-64 115-09 Cadmium Sulphide. Cadmium . . . 55*63 Sulphur . . . . 15-64 71*27 Hence we see again that the amounts of these elements which unite with an equivalent of oxygen also combine with an equivalent of sulphur, so that 1 part by weight of hydrogen combines with -J I 15'64 of sulphur 102*68 parts by weight of lead combine with 15*64 of sulphur 31*41 copper 99*45 mercury 55*63 cadmium and these are, therefore, the equivalent weights of these elements. DALTON'S ATOMIC THEORY 95 39 When one element combines with another in more than one proportion it is said to have more than one equivalent, and since the amounts of one element which combine with a fixed weight of a second are in a simple ratio to one another, it follows that the several equivalents of an element must also stand in a simple ratio to one another. Iron for example forms several different compounds with oxygen, two of which have the following composition as determined by analysis : Iron Ferrous Oxide. ..... 77-78 Ferric Oxide. 70-00 Oxygen 22-22 30-00 100-00 100-00 Calculating the amount of iron combined with the equivalent (7*94 parts) of oxygen we find (l) (2) Iron 2778 18'52 Oxygen 7'94 7*94 35-72 26-46 These amounts of iron are, however, in the simple ratio of 2 : 3, 27*78 being the equivalent of iron in ferrous oxide and 18*52 in ferric oxide. 40 It will be seen, therefore, that combination always takes place between certain definite and constant proportions of the elements or between multiples of these. 41 At a time when the question of combination in a limited number of definite proportions was still under discussion, John Dalton's speculative mind conceived an hypothesis which clearly explained the law of combination in constant proportions, and solved the question as to the nature of the compounds formed by the union of two or more elements in several different pro- portions. The hypothesis known as Dalton's Atomic Theory may be said to have become one of the most important foundation stones of the science, and to have exerted an influence on its progress greater than that of any other generalization, with, perhaps the single exception of Lavoisier's explanation of the phenomena of combustion, and the discovery of the indestructibility of matter. 96 GENERAL PRINCIPLES OF THE SCIENCE The Atomic Theory, then, follows the doctrines of the Greek philosophers so far as it supposes that matter is not continuous, but made up of extremely small individual particles termed atoms (a privative and re^va) I cut) ; but differs from that of the ancients and becomes truly a chemical atomic theory inas- much as it supposes the atoms of different elements not to possess the same weights, but to be characterised by different weights. Thus the atom of oxygen is 15*88 times as heavy as the atom of hydrogen, and the weights of the atoms of oxygen .and chlorine are as 15 '88 to 35 '19. Dalton assumed, in the second place, that chemical combination consists in the approximation of the individual atoms to each other. Having made these assumptions he was able to explain why combination always takes place between certain amounts of the elements or between multiples of these amounts ; since combination being supposed to take place between some number of atoms of each of the elements which unite, it follows that the amounts which combine must be some finite multiple of the weights of the atoms. It is thus clear that the atomic theory explains the formation of all compounds which are found to exist, but it is equally evident that it in no way decides how many compounds can be formed by any two or more elements. This at present can only be learned by experiment, but we are not without indications that a time approaches when this further problem will receive a theoretical solution. 42 Although the atomic theory satisfactorily explains all the known laws of chemical combination, the actual existence of atoms is far from being thereby positively proved; indeed, from purely chemical considerations it appears unlikely that the question will ever be solved. 1 Nevertheless, there is evidence, gradually becoming more cogent, connected with certain physical phenomena, which compels us to admit a limit to the divisi- bility of matter. The phenomena in question belong to the science of molecular physics, and have reference to such subjects as the capillary attraction of liquids, the diffusion of gases, and the pro- duction of electricity by the contact of metals. Reasoning from facts observed in the study of these subjects, physicists have not only come to the conclusion that matter is discontinuous, and, therefore, that indivisible particles or molecules (molecula, a small 1 Williamson, "On the Atomic Theory," Chem. Soc. Journ., 22, (1869) 328 and 433. COMBINATION BY VOLUME 97 mass) exist, but they have even gone so far as to indicate the order of magnitude which these molecules attain. Thus Sir William Thomson states that in any ordinary liquid or transparent or seemingly opaque solid, the mean distance between the centres of contiguous molecules is less than one hundred-millionth, and greater than the two thousand-millionth of a centimetre. Or, in order to form a conception of this coarse-grainedness, we may imagine a rain-drop or a globe of glass as large as a pea to be magnified up to the size of the earth, each constituent molecule being magnified in the same proportion ; the magnified structure would be coarser-grained than a heap of small shot, but probably less coarse-grained than a heap of cricket-balls. 1 The molecular constitution of matter is likewise an essential condition of the mechanical theory of gases, by means of which nearly every known mechanical property of the gases can be explained on dynamical principles, so that in this direction again, we have a confirmation of the real existence of molecules (p. 60). 43 It will be seen that the atomic theory as proposed by Dai- ton does not provide any means for ascertaining what the relative weights of the atoms really are. We have seen, for instance, that 7'94 parts by weight of oxygen combine with one part of hydrogen to form water, but we cannot draw any conclusion from this fact as to the atomic weight of oxygen, (that of hydrogen being taken as equal to one), until we know how many atoms of each of these elements have taken part in the combination. If the combination is between an equal number of atoms of each element, then the relative weights of their atoms must be as 1 : 7*94, whilst if two atoms of hydrogen have combined with one of oxygen the relation will be as 1 : 15'88. The answer to this question as to the number of atoms of each element between which combination has taken place has been supplied by a study of the laws of combination of gaseous substances. COMBINATION BY VOLUME. 44 The discovery by Gay-Lussac and Humboldt in 1805 of the simple relation existing between the combining volumes of oxygen and hydrogen gases, followed by that of the general law of gaseous volumes enunciated in 1808 by Gay-Lussac alone, 1 Nature, March 31, 1870. 8 98 GENERAL PRINCIPLES OF THE SCIENCE serves as a powerful argument in favour of Dalton's Atomic Theory. This law states that the volumes in which gaseous sub- stances combine bear a simple relation to one another and to the volume of the resulting product. This is true both for elementary and compound gases, the simple relations which exist being illustrated by the following table : 1 vol. of chlorine and 1 vol. of hydrogen form 2 vols. of hydrochloric acid gas. 1 vol. of oxygen and 2 vols. of hydrogen form 2 vols. of steam. 1 vol. of nitrogen and 3 vols. of hydrogen form 2 vols. of ammonia. 1 vol. of oxygen and 2 vols. of carbonic oxide form 2 vols. of carbon dioxide. According to the atomic theory, however, combination takes place between the atoms of which substances are made up, and it hence follows, if we accept this theory, that the number of atoms which is contained in a given volume of any gaseous body, must stand in a simple relation to that contained in the same volume of any other gas (measured under equal circumstances of temperature and pressure). The simplest as well as the most probable supposition respecting this question is that put forward by Avogadro in 181 1, 1 who assumed that equal volumes of all the different gases, both elementary and compound, contain the same number of particles or integrant molecules, and this theory is now generally accepted by physicists, who have arrived at the same conclusion as the chemists have reached by an inde- pendent train of reasoning (p. 61). If we take the simplest case of volume combination, that of one volume (one molecule) of chlorine and one volume (one molecule) of hydrogen uniting to form two volumes (two molecules) of hydrochloric acid gas, it is clear that, since each molecule of hydrochloric acid contains at least one atom of chlorine and one of hydrogen, there are at least twice as many atoms as molecules of these elements present. Hence, to conform to Avogadro' s theory, the integrant molecule of free chlorine and of free hydrogen must consist of at least two atoms combined together, and we shall represent the combination as taking place between one volume (one molecule of two atoms) of chlorine, and one volume (one molecule of two atoms) of hydro- gen, forming two volumes (two molecules) of the compound hydrochloric acid gas. Again, two volumes of steam are formed from two volumes of hydrogen and one volume of oxygen, hence if there are the same number of molecules of steam, of hydrogen 1 Journ. de Phys., par De la Metherie, 73, Juillet, 1811, 58-76. Ibid., Feb, 18U. AVOGADRO'S THEORY 99 and of oxygen in the same volume of each gas, it is clear that in the formation of water from its elements, each molecule of oxygen must be split up into two similar parts. We are thus led to distinguish between the atom and the molecule, the latter term being applied to the smallest particle of an element or com- pound ivhich can exist in the free state. Avogadro's theory refers exclusively to these molecules and states that equal volumes of all gases, measured under the same physical conditions, contain equal numbers of molecules. An immediate consequence of this theory is of the utmost importance. If we weigh equal volumes of two gases, we are obviously weighing equal numbers of their molecules, and the ratio of the weights of the gases will also be the ratio of the weights of their molecules, so that we are thus enabled to deter- mine the relative weights of the molecules of all gases, by simply finding their relative densities. Hydrogen gas is taken as the standard of comparison because it has a lower density than any other gas, and, since it has been shown that its molecule can be divided into at least two parts (p. 98), the weight of its molecule is taken as equal to two. The molecular weight of any gas is therefore equal to twice its density compared with hydrogen; nitrogen, for example, is about 14 times as heavy as hydrogen and hence its molecular weight is about 28, whilst carbon dioxide has a density of 22 and therefore a molecular weight of 44. 45 When the molecular weight of a gas and also its composi- tion, as determined by analysis, are both known, it is possible to calculate what proportion of each of the component elements is present in the molecule. Water for instance contains 88'81 per cent, of oxygen and 1119 of hydrogen, whilst the density of steam is about 8'94, its molecular weight being, therefore, equal to 17*88. If now we calculate how much oxygen and hydrogen are present in 17'88 parts of water, we find that this amount is made up of 15 '88 of oxygen and 2 of hydrogen. Carbon dioxide again has a molecular weight of 43' 7, and contains Carbon .... 27'27 Oxygen . . . 7273 100-00 Hence 437 parts of this gas contain 11-91 of carbon and 31'76 of oxygen. 100 GENERAL PRINCIPLES OF THE SCIENCE A repetition of this process for all the known com pounds of some particular element enables us to ascertain the least amount of that element which is ever found in a molecule of one of its compounds, and to this amount the name of atom is given. A comparison of all the compounds of oxygen, for example, teaches us that the least amount of it ever found in a molecule is 15*88 parts, and this is, therefore, taken as the atomic weight of oxygen. This having been ascertained, we are in a position to say that the molecule of water contains one atom of oxygen, whilst that of carbon dioxide contains two. All the non-metallic elements form compounds which can exist in the state of gas, and hence the atomic weights of all these elements have been found by this method. Many of the metals on the other hand do not form volatile compounds, and the atomic weights of these have, therefore, to be determined by different methods. A discussion of these methods will be found in a later volume (Vol II. Part I. p. 24). In any case it must be remembered that the method described above is not generally capable of great accuracy and only yields an approximate number for the atomic weight, the exact value being found by determining the equiva- lent of the element by an accurate analysis of one of its compounds, and then taking as the exact atomic weight the multiple of this number which approaches most closely to the approximate number obtained from the molecular weights. 46 It will be seen that the determination of the atomic weight is quite distinct from that of the molecular weight. This is well shown in the case of carbon ; a comparison of the gaseous compounds of carbon shows that the atomic weight of this element is about 12, this being the least amount of it which is found in the molecule of one of its compounds ; we are, however, quite ignorant of the molecular iveight of carbon itself, since its density in the state of gas has never been determined. The molecules of some elements contain two atoms, this being the case with hydrogen, oxygen, nitrogen, chlorine and others, whilst the molecules of mercury vapour and of the vapours of some of the other metals consist of single atoms, the molecular and atomic weights being, therefore, identical. 47 For the first time we may now employ chemical symbols, a kind of shorthand, by which we can conveniently express the various chemical changes. To each element we give a symbol, usually the first letter of the Latin, which is generally also that of the English name. Thus O stands for oxygen ; EMPIRICAL AND MOLECULAR FORMULAE 101 H for hydrogen ; S for sulphur ; Au for gold (aumm) ; Ag for silver (argentum). These letters, however, signify more than that a particular substance takes part in the reaction. They serve also to give the quantity by weight in which it is present. Thus O does not stand for any quantity, but for 15*88 parts by weight (the atomic weight) of oxygen; H always stands for one part by weight of hydrogen ; and in like manner S, Au, and Ag stands invariably for 31'28, 195*7, and 107*13 parts by weight of the several elements respectively. By placing symbols of any elements side by side, a combination of the elements is signified, thus : HC1 Hydrochloric acid HI Hydriodic acid HBr Hydrobromic acid HgO Mercuric oxide. If the molecule contains more than one atom of any element, this is indicated by placing a small number below the symbol of the atom of the element, thus H 2 O signifies 17*88 parts by weight of a compound (water) containing two atoms or 2 parts by weight of hydrogen and one atom or 15'88 parts by weight of oxygen. In such a case as this, where the molecular weight and the number of atoms in the molecule are known and expressed in the formula, the latter is said to be a molecular formula and consequently represents such a weight of the substance as will in the state of gas occupy the same volume as two parts by weight of hydrogen. When the molecular weight of the compound to be represented by a formula is not known, it is only possible to express the relative number of the atoms of the constituent elements which are present. A formula of this kind is known as an empirical formula and may be calculated for any substance of which the composition has been determined by analysis. The gas known as ethylene has the following composition as determined by analysis : Carbon .... 85'62 Hydrogen . . 14*38 100*00 In order to find the empirical formula of this substance it is only necessary to divide the percentage of each element by the 102 GENERAL PRINCIPLES OF THE SCIENCE atomic weight of the element, which gives us the ratio of the number of atoms of each of the two elements, and if we express this ratio in the smallest possible whole numbers we have at once the relative numbers of atoms present in the molecule, without, however, having any information as to the absolute number. Percentage Simplest Percentage. Atomic Weight. Ratio. Carbon 85*62 7*19 1 Hydrogen 14*38 14*38 2 The simplest or empirical formula of ethylene is therefore CH 2 . The density of this gas however is found to be equal to 13*91 and its molecular weight is therefore 27*82, its molecular formula being consequently C 2 H 4 . It is usual to represent chemical changes in the form of equations ; the materials taking part in the change being placed on one side and the products formed, which are always equal to them in weight (p. 48), being placed on the other. If we heat potassium chlorate, a substance which has the empirical formula KC1O 3 , it is decomposed into oxygen and potassium chloride and this decomposition is represented by the equation 2KC1O 3 = 2KC1 + 30 2 in which the sign -f connects the two products and signifies " together with." This equation is an expression of the fact, ascertained by experiment, that 121*68 (38*85+35*19-i-3 X 15*88) parts of this salt by weight leave behind on heating 74*04 parts (35*1 9+38*85) of potassium chloride and liberate 47*64 parts of oxygen. Hence it is clear that the quantity of oxygen which is obtained from any other weight of the salt and vice versd can be found by a simple calculation when the equation representing the chemical change is known. To take a more complicated case, when we know that the equation representing the change which occurs when we heat potassium ferrocyanide, the empirical formula of which is K 4 C 6 N 6 Fe, with strong sulphuric acid H 2 SO 4 and water, is the following : K 4 C 6 N 6 Fe+6H 2 S0 4 +6H 2 0-6CO-f2K 2 S0 4 4-3(NH 4 ) 2 S0 4 +FeS0 4 yielding carbon monoxide gas, CO, potassium sulphate K 2 S0 4 . ammonium sulphate (NH 4 ) 2 S0 4 , and iron sulphate FeSO 4 , we CALCULATION OF CHEMICAL QUANTITIES 103 can easily calculate how many grams of carbon monoxide gas, CO, can be obtained from any given weight of the ferrocyanide, K 4 C 6 N 6 Fe, inasmuch as analysis proves that the amounts represented by these formulae are made up as follows : Carbon Monoxide. Ferrocyanide of Potassium. Carbon C 11-91 Potassium K 4 155'40 Oxygen O 15'88 Carbon C 6 71'46 Nitrogen N 6 83'64 27-79 Iron Fe 55'58 366-08 The foregoing equation then shows that 36 6 '08 parts by weight of the ferrocyanide yields 166'74 parts by weight of car- bon monoxide, and hence a simple proportion gives the quantity yielded by any other weight. The illustration is, however, not yet complete ; commercial potassium ferrocyanide contains, as do many crystalline compounds, a certain quantity of water cf crystallization, which is given off when the salt is heated, in consequence of which the crystals fall to a powder. But, as the equation shows, a certain quantity of water takes part in the reaction, and it is, therefore, unnecessary to dry the salt previously if only we know how much water of crystallization it contains. Analysis has shown that the commercial salt has the composition K 4 C 6 N 6 Fe-|-3H 2 O ; hence if we add 3xl7'88, the weight of 3 molecules of water, to 366*08, we obtain the number 419'72 as the weight of the liydratcd salt, which must be taken in order to obtain 166'74 parts by weight of carbon monoxide. As, however, the quantity of a gas is almost always estimated by measuring its volume, and from this volume calculating its weight, it becomes of the greatest importance to know how to calculate the volume of a gas from its weight, or vice versa. This can readily be done, for we know that the density of every compound in the gaseous state is half its molecular weight ; or that every molecule in the gaseous state occupies the volume filled ~by two parts ly weight of hydrogen. One litre of hydrogen gas at and under 760 mm of mercury weighs 0'089872 gram.; hence the weight of a litre of any other gas measured under the same circumstances of temperature and pressure is obtained by multiplying the density of the gas, or half its molecular weight, 104 GENERAL PRINCIPLES OF THE SCIENCE by the above number, or by 0*0899, when absolute accuracy is not required. Thus one litre of carbon monoxide weighs 27*79 - x 0*0899=1*249 grms., and 166*74 grms. of this substance 166*74 occupy a volume of litres at and 760 mm . 1*249 It is now easy to calculate what volume this weight will occupy at any other temperature or pressure, for we know that all gases expand by Y |^ of their volume at 0C. when their temperature is raised 1C. (at constant pressure, Law of Dalton), and that their volume is inversely proportional to the pressure to which they are subjected (Law of Boyle). Hence if the temperature at which the gas was collected were 17C., and if the barometer then stood at 750 mm the volume (v) in litres of the carbon monoxide collected would be 166-74 x (273 + 17) x 760 v = 1*249 x 273 x 750. EXPERIMENTAL METHODS FOR THE DETERMINA- TION OF MOLECULAR WEIGHTS. 48 We have seen above (p. 99) that the density of every com- pound in the gaseous state is half its molecular weight, from which it follows that in order to determine the molecular weight of a substance it is simply necessary to determine its density in the state of gas. This process can be readily applied to such substances as are gases under ordinary conditions of temperature and pressure, and also to such as can be rendered gaseous by moderate increase of temperature, but is of course inapplicable to bodies which decompose on heating or which cannot easily be vapourised. In such cases, however, several methods are available which depend upon the properties of solutions, and hence, since nearly all chemical compounds are soluble in some liquid, the molecular weight of a body can almost always be determined. Special experimental methods have to be adopted for each of these classes of substances. o DETERMINATION OF MOLECULAR WEIGHTS 105 I. DETERMINATION OF THE MOLECULAR WEIGHTS OF PERMANENT GASES. 49 For this purpose it is only necessary to determine the specific gravity of the gas, air being usually taken as the practical unit of comparison. A large glass balloon capable of holding from 1 to 10 litres of the gas, is employed and is first of all freed from air as far as possible by the vacuum pump, and weighed. In weighing a body of such large volume it is essential to make allowance for the buoyancy of the air, since the body to be weighed appears to be lighter than it really is by an amount equal to the weight of the air which it displaces. This is best accomplished by suspending a vessel of similar size and shape to the other arm of the balance, by which arrangement the effect of buoyancy is neutralized, each of the vessels being affected in the same manner, and at the same time the uncertainties of calculation, due to the varying temperature, pressure, and moisture of the atmosphere are avoided, as well as any inaccuracy due to condensation on the surface of the glass. As soon as the weight of the empty vessel has been ascertained, it is removed from the balance and filled with the pure dry gas, the tempera- ture and pressure being carefully observed. The globe is then reweighed with the same precautions as before. One additional correction must be here noticed as it gives rise to a considerable error especially when light gases are being weighed. The capacity of a vacuous globe of glass is found to be perceptibly less than that of the same globe when filled with gas at the pressure of the atmosphere, and hence the globe displaces less air in the former condition than in the latter. The difference between the corrected weights of the globe empty and filled with the gas is equal to the weight (W) of the given volume of gas at the observed temperature and pressure. The same globe is then filled with dry air freed from carbonic acid gas, or else with pure hydrogen, and the weight of an equal volume of the latter (A) under the same conditions thus ascertained. The specific gravity of the gas compared with air is then equal W to -j-. Since air has been found to be 14*39 times as heavy as hydrogen and the molecular weight of a gas is twice its density with respect to hydrogen, it is only necessary to multiply the 106 GENERAL PRINCIPLES OF THE SCIENCE specific gravity by 14'89x2 to obtain the molecular weight of the gas in question. The most accurate determinations of this kind have been made by Regnault, and more recently by Rayleigh, Leduc &c. II. DETERMINATION OF THE MOLECULAR WEIGHTS OF VOLATILE LIQUIDS AND SOLIDS. 50 A detailed account of the various methods proposed for this purpose is to be found in a later volume (Vol. III. Pt. I. pp. 84-112) ; only the two most frequently employed will be here briefly described. The specific gravity of the vapour of a liquid or solid can be determined in two ways ; either by weighing the vapour which occupies a known volume under given conditions of temperature and pressure (Dumas' method) or by measuring the volume occu- pied under given conditions by the vapour of a known weight of the liquid or solid (Gay-Lussac, Hofmann, Victor Meyer). I. METHOD OF DUMAS. 51 For this purpose a thin glass globe is employed of 150 200 cubic centimetres in capacity, having a finely drawn out neck ; the exact weight of the globe, weighed in air and filled with dry air at a certain temperature and pressure, having been found, a small portion of the substance of which the vapour density is to be determined is brought inside, and the globe then heated by plunging it into a water- or oil-bath raised to a temperature at least 30 above the boiling point of the substance ; as soon as the vapour has ceased to issue from the end of the neck, this end is hermetically sealed before a blow- pipe, and the exact temperature of the bath as well as the barometric pressure observed. The bulb thus filled with vapour is carefully cleaned, allowed to cool, and accurately weighed. The point of the neck is next broken under water, which rushes into the globe, the vapour having condensed, and, if the experiment has been well conducted, completely fills it. The bulb is then weighed full of water, and its capacity calcu- lated from the weight of water which has entered. DETEEMINATION OF VAPOUE DENSITY 107 We have now all the data necessary for the determination. In the first place we have to find the weight of the given volume of the vapour under certain circumstances of tempera- FIG. 21. ture and pressure, and we then have to compare this with the weight of an equal volume of hydrogen gas measured under the same circumstances. The following example of the deter- mination of the vapour density of water may serve to illustrate the method : FIG. 22. Weight of globe filled with dry air at 15'5 C Weight of globe filled with vapour at 140 Capacity of the globe . 23*449 grams. 23-326 I78cc. 108 GENERAL PRINCIPLES OF THE SCIENCE As the barometric column (760 mm.) underwent no change from the beginning to the end of the experiment, no correction for pressure is necessary. In order to get at the weight of the vacuous globe the weight of air contained must be deducted from the weight of the globe in air. Now 1 cc. of air at and 760 mm. weighs 0*001293 178 x 273 grams, and 178 cc. of air at 15 '5 would occupy - - _ 288'5 168-4 cc. at 0, so that the weight of this air is 168-4 x 001293 = 0'218 gram ; hence the weight of the vacuous globe is 23-231 (23-449 - 0'218), and the weight of the vapour 23'326 - 23'231 = 0'095 gram. We must now find what 178 cc. of hydrogen at 140 will weigh. One thousand cc. of hydrogen at weigh 0'0899 gram. ; 178 cc. at 140 will con- 178 x 273 117-6 x 0-0899 tract to -prs = 117'6 at , which weigh 1nrtA Txychloride, or atacamite, Cu 2 Cl(OH) 3 , and many others. The chlorides of the alkalis occur in the bodies of plants and limals, and play an essential part in the economy of the animal and vegetable worlds. Chlorine likewise occurs combined with hydrogen, forming hydrochloric acid, a substance which found in nature in small quantities in certain volcanic gases. 1 Opusc. Tome 1, 247. 2 Him. de I'Acad. dcs Science*, Paris, 1785, p. 276. 3 Phil. Trans. 1811, pp. 1 and 32 ; Bakerian Lecture for 1810, read Nov. 10th,. L810. 4 Memoires d'Arcueil. Tome 2, 357. 154 THE NON-METALLIC ELEMENTS 77 Preparation. (1) Chlorine gas is prepared by the action of the black oxide of manganese or manganese dioxide MnO 2 , on strong hydrochloric acid, HC1, thus : 4HC1+ MnO 2 =Cl 2 + MnCl 2 +2H 2 O. This reaction consists in the removal of two atoms of hydrogen in two molecules of hydrochloric acid by union with one atom of oxygen of the manganese dioxide to form water, the two atoms of chlorine being set free ; whilst the manganese mon- oxide MnO, which may be considered as also being formed, dissolves in the two remaining molecules of hydrochloric acid to produce manganese chloride, MnCl 2 , and another molecule of water, H 2 O. When manganese dioxide and cold concentrated FIG. 37. hydrochloric acid are brought together a dark brownish-green solution is formed, and this, on heating, evolves chlorine gas, whilst manganese chloride, MnCl 2 , is formed. There can be little doubt that this dark coloured solution contains a higher and unstable chloride of manganese, probably MnCl 4 , corre- sponding to MnO 2 , which, on heating, decomposes into MnCl 2 + Cl 2 . For this preparation the oxide of manganese should be used in the form of small lumps free from powder, and the hydro- chloric acid poured on, so as about to cover the solid ; on gently heating, the gas is copiously evolved. PREPARATION OF CHLORINE 155 (2) It is often more convenient for laboratory uses to evolve -the hydrochloric acid in the same vessel in which it is acted upon by the manganese dioxide, and to place a mixture of one part of this substance and one part of common salt in a large flask (Fig. 37) containing a cold mixture of two parts of strong sulphuric acid and two of water ; on very slightly warming the mixture, a regular evolution of gas takes place. The change which here occurs was formerly supposed to be represented by the equation : 2NaCl + 3H 2 SO 4 -f MnO 2 =Cl 2 +2NaHSO 4 + MnSO 4 + 2H 2 O. according to which the whole of the chlorine is evolved in the free state. Klason has shown, however, 1 that this is not the case, the correct equation being as follows : 4NaCl+Mn0 2 I 2H 2 SO 4 -2NaHSO 4 +Na 2 SO 4 +MnCU2H 2 O+Cl 2 In order to purify and dry the gas prepared by either of the above methods, it is necessary to pass it, first through a wash bottle (&, Fig. 38) containing water, to free it from any hydrochloric acid gas which may be carried over, then through a second wash bottle (a) containing strong sulphuric acid, to free it from the larger quantity of the aqueous vapour which it takes up from the water, and lastly through a long inclined tube (c) containing pieces of pumice-stone, moistened with strong and boiled sul- phuric acid. The tube (d) which dips under water serves as a safety-tube in case the evolution of gas becomes too rapid, when bhe excess of gas can thus escape. In order to expel the air rhich fills the apparatus, the evolution of the chlorine must allowed to go on until the gas is almost entirely absorbed )y a solution of caustic soda. As the crude black oxide of langanese frequently contains carbonate of lime, the presence )f which will cause the admixture of small quantities of carbon iioxide, CO 2 with the chlorine, it is advisable to moisten the >re before using it with warm dilute nitric acid, which will lissolve out the carbonate of lime, leaving the manganese Iioxide unacted upon ; after well washing, the latter may be ised without danger of this impurity. (3.) By heating a mixture of bichromate of potash (potassium lichromate) and hydrochloric acid, chlorine gas can also be 1 Ber. 23, 334. 156 THE NON- METALLIC ELEMENTS obtained, chromium chloride and potassium chloride being formed ; thus : 14HCl-|-K 2 Cr 2 O 7 =3Cl 2 -f2CrCl 3 +7H 2 O-h2KCl. Chlorine gas is also evolved when an acid is added to an FIG. 38. alkaline hypochlorite or to bleaching powder ; to prepare it from the latter, the bleaching powoler is made coherent by mixing with plaster of Paris or by compressing it to form a solid cake which is afterwards broken into lumps of suitable size. 1 As the Winkler, Ber. 20, 184 ; Thiele, Annalen, 253, 239. PREPARATION OF CHLORINE 157 evolution of the gas takes place at the ordinary temperature, the Kipp's apparatus shown later under sulphuretted hydrogen may be employed, by which a current of the gas can be obtained as required. (4) By passing a mixture of air and hydrochloric acid over heated bricks, a portion of the hydrogen of the hydrochloric acid is oxidised, water and chlorine gas being formed (Oxland, 1847). If the mixture of gases be allowed to pass over a heated surface impregnated with certain metallic salts, especially sul- phate of copper, the oxidation of the hydrogen of the hydro- chloric acid goes on to a greater extent, and by the absorption of the unaltered hydrochloric acid, a mixture of chlorine and nitrogen gases can be obtained. This process, patented by the late Mr. Henry Deacon, of Widnes, is used on a large scale for the eco- nomic production of chlorine and bleaching powder. The singular .and but imperfectly understood decomposition which takes place may be shown on a small scale by the following arrangement : The hydrochloric acid gas is evolved from the common salt and sulphuric acid in the large flask a, Fig. 39, this gas passes into the tube, b, in which are placed pieces of tobacco-pipe moistened with a saturated solution of copper sulphate, the tube being exposed to a gentle heat. As the hydrochloric acid enters "the tube containing the sulphate of copper it mixes with atmospheric air which is driven in by the tube c, from the gas- holder. On passing over the heated copper sulphate, the hydro- chloric acid and the oxygen of the air act upon one another, water and chlorine gas being formed according to the equation 4HC1 + 2 = 2H 2 + 2C1 2 . During the process the sulphate of copper remains unchanged -and may be used for a great length of time. The mixture of chlorine, nitrogen, steam, and any undecomposed hydrochloric acid pass by a bent tube into a bottle, e, containing water, by which the last-named substance is arrested, together with a portion of the steam which is condensed ; the mixed gases, still containing some aqueous vapour, are then passed through a tube, d, containing calcium chloride, by which the gases are completely dried, after which ulb in a cloth during the operation of sealing, as not unfre- [uently the gas explodes. As soon as one bulb is removed a second is introduced, and placed in connection with the evolution lask, and after ten minutes sealed as described. The bulbs rims obtained should be numbered, and the first and last tested >y exposing them to a strong light, and if these explode, all the itermediate bulbs may be considered good. Sixty such bulbs 172 THE NON-METALLIC ELEMENTS may be prepared with the above quantity of acid, and may be kept in the dark for an unlimited time without change. 1 On exposing one of these bulbs to the light emitted by burning magnesium ribbon, or to bright daylight, a sharp explosion occurs and hydrochloric acid is formed (Fig. 47). When the mixture of hydrogen and chlorine is exposed to diffused daylight combination gradually takes place, the rate of formation of hydrochloric acid increasing with an increasing proportion of the more refrangible rays of light. Bunsen and Roscoe 2 found, however, that the action of light is at first very slow and only attains its full activity after a certain length of time. Thus, for example, when a mixture of the gases was exposed to the light of a small petroleum lamp burning at a constant rate it was found that the amount of hydrochloric acid formed in each minute increased for the first nine minutes and then became constant. To this phenomenon they have given the name " photo-chemical induction." From the recent in- vestigations of Pringsheim 3 it seems probable that the phe- nomenon is due to the formation of an intermediate product, for, just as in the case of carbon monoxide and oxygen the mixture of gases undergoes combination far more easily when moist than dry, even a very bright light bringing about an explosion in the latter case only with great difficulty, and hence it appears unlikely that in the combination we have a simple reaction between hydrogen and chlorine, but that inter- mediate compounds are formed by the combined influence of light and moisture. The amount of hydrochloric acid formed during the first period of the action of light would then be too small, because the formation of the acid could only take place after a certain quantity of the intermediate compound had been formed ; after a time, however, the quantity of the intermediate compound formed would be equal to the quantity decomposed, and the amount of the latter in the mixture and of the hydro- chloric acid formed in a given time would remain constant. The first period corresponds to the period during which " induction " takes place, and the second to that in which the light brings about the formation of the product of the reaction at a constant rate. Concerning the nature of this intermediate compound, which is probably present in only extremely minute quantities, nothing 1 Roscoe, Proc. Manch. Lit. and Phil. Soc. Feb. 1865. 2 Phil. Trans. 1857. 3 Ann. Phys. Chem. [2], 32, 384. PREPARATION OF HYDROCHLORIC ACID 173 can be definitely said ; Pringsheim has, however, suggested that the two reactions may be represented by the following equations : (1) H 2 + C1 2 - C1 9 + H 2 (2) 2H 2 + C1 2 O = H 2 O + 2HC1. Combination of hydrogen and chlorine also takes place when the mixed gases are heated to a sufficiently high temperature : in closed vessels explosion occurs between 240 and 270, whereas if the mixture is passed in a stream through the heated vessel the explosion does not take place till the temperature reaches 430 to 440 . 1 86 Hydrochloric acid is also formed by the action of chlorine upon almost all hydrogen compounds ; which are decomposed by it either in the dark or in presence of light ; thus sulphuretted hydrogen, olefiant gas, turpentine (see p. 162), and water, are all decomposed by chlorine, hydrochloric acid being formed. When hydrogen is passed over many metallic chlorides, such as silver chloride, hydrochloric acid is evolved, and the metal produced, thus : 2AgCl + H 2 = 2Ag + 2HC1. By these and other reactions hydrochloric acid is frequently formed ; but none of them serve for the preparation of the gas on a large scale. 87 Preparation. For this purpose six parts by weight of com- mon salt are introduced into a capacious flask, and eleven parts of strong sulphuric acid slowly poured on it through a bent tube- funnel ; the gas, which is at once rapidly evolved, is purified from any sulphuric acid or salt which may be carried over, by passing through a small quantity of water contained in a wash bottle, and it may then either be collected by displacement (like chlorine), or over mercury, or passed into water, as shown in Fig. 53, if an aqueous solution of the acid is needed. The reaction which here occurs is represented by the equation : NaCl + H 2 SO 4 = HC1 + NaHSO 4 . Hydrochloric acid comes off, and a readily soluble acid sulphate of soda, or hydrogen sodium sulphate, NaHSO 4 , is left. If two molecules of salt be taken to one of sulphuric acid, a less easily soluble salt, normal sodium sulphate Na 2 SO 4 , is formed, a greater 1 Freyer and V. Meyer, Zeit. Phys. Chem. H, 28. 174 THE NON-METALLIC ELEMENTS heat being needed to complete the decomposition than when an excess of acid is employed, thus : NaCl + NaHS0 4 = Na 2 S0 4 + HC1. The pure aqueous acid is best prepared for laboratory from pure salt and pure sulphuric acid. 88 Properties. Hydrochloric acid is a colourless gas, which was first liquefied by Davy and Faraday, 1 who estimated the tension of the gas to be 20 atmospheres at 16, 25 atmos- pheres at -- 4, and 40 atmospheres at + 10. The method adopted by Davy for the liquefaction of the gas is shown in Fig. 48. FIG. 48. Into a tube, three times bent at a right angle and closed at one end, are placed, at a, a few small pieces of sal-ammoniac (a com- pound of hydrochloric acid and ammonia) ; some strong sulphuric acid is then poured by means of a bent tube-funnel (d), into the second bend at I. The open end of the tube is next carefully drawn out, thickened, and closed before the blow-pipe, and when the tube is cold, it is so inclined as to allow the acid to flow on to the sal-ammoniac. Hydrochloric acid gas is at once disengaged according to the equation : 2NH 4 C1 + H 2 S0 4 = 2 HC1 + (NH 4 ) 2 SO 4 , and after a time the pressure becomes sufficiently great to liquefy the further portions of the gas which are evolved, and 1 Davy and Faraday, Phil. Trans. 1823, p. 164. PROPERTIES OF HYDROCHLORIC ACID 175 by gentle heat the liquid may be distilled over into the empty limb of the tube. It is a colourless liquid, has a specific gravity of 0'854 at 105, and solidifies at -- 115 to a crystalline mass which melts at 112'5 (Olszewski), and does not conduct electricity. The action of liquid hydrochloric acid upon various substances has been carefully examined by Gore. 1 These experi- ments show that the liquid acid has but a feeble solvent power for bodies in general, and, with the exception of aluminium, the metals are not attacked by it. Hydrochloric acid gas is heavier than air. Its specific gravity, according to the most accurate experiments of Biot and Gay- Lussac, is 1'278 (air = 1) whilst according to Leduc 2 it is 1*2696 or its density is 18'39 (H = 1), the calculated density being 18*095. At 1500 the gas appears to have undergone no change, but at 1700 a considerable amount of dissociation has taken place. 3 The gas fumes strongly in the air, uniting with atmo- spheric moisture, and it is instantly absorbed by water or ice, yielding the aqueous acid. It possesses a strongly acid reaction and suffocating odour, and is not inflammable. A burning candle is extinguished when plunged into the gas, the outer mantle of the flame, before extinction, exhibiting a characteristic green coloration. Direct sunlight has no action on a dry mixture of hydrogen chloride and oxygen, but if more moisture than is necessary for the complete saturation of the gas be present, decomposition gradually takes place, free chlorine and water being formed. The amount of chlorine liberated depends upon the proportion of oxygen present ; thus a mixture of 4 volumes HC1 and 1 volume O only gave 0'34 per cent, of free chlorine after 24 days' exposure, whilst with 8 volumes of oxygen 73'81 per cent, of the chlorine was liberated. It appears that in some cases hypochlorous acid is also formed. 4 89 The composition of hydrochloric acid gas can be best ascertained as follows : Metallic sodium decomposes the gas into chlorine, which combines with the metal to form sodium chloride, and into hydrogen, which is liberated. If a small piece of sodium be heated in a deflagrating spoon until it begins to burn, and then plunged into a jar of hydrochloric acid gas, the combus- 1 Proc. Roy. Soc. 14, 204. 3 Comptes Rend. 116, 968. 3 Langer and V. Meyer, Pyrochem. Unters. (Vieweg), p. 67. 4 McLeod, Journ. Chem. Soc. 1886, i. 591 ; Richardson, Journ. Chcm. Soc. 1887, i. 802. 176 THE NON-METALLIC ELEMENTS tion of the metal (union with chlorine) will go in the gas. In order to show what volume of hydrogen is evolved fro: a given volume of hydrochloric acid gas by this reactio the following experiment may be made with the eudiome tube, the construction of which is clearly seen in Fig. 49. T begin with, both limbs are filled completely with dry mercury, then the end of the tube carrying the stop-cock is connected by a piece of caoutchouc tubing with an evolution flask, from which pure hydrochloric acid gas is being slowly evolved from a mix- ture of dry salt and strong sulphuric acid, care being taken that the air has been driven out. On turning the stop-cock at the top of the tube, and opening the screw-tap on the caoutchouc in the U-tube, the mercury will run out, and dry hydrochloric acid gas FIG. 49. will enter the one limb, whilst air fills the other to the same level. As soon as the gas reaches a mark on the tube indicating that it is two-thirds full of gas the stopcock is closed. A small quantity of sodium amalgam is now prepared by pressing six or eight small pieces of clean cut sodium, one by one, under the surface of a few ounces of mercury contained in a porcelain mortar. The amalgam is then poured into the open limb of the U-tube so as to fill it, and the end firmly closed with the thumb ; the hydrochloric acid gas is now transferred to the limb containing the amalgam, and well shaken so as to bring the gas and amalgam into contact. The gas is next passed back into the closed limb, and the pressure equalised by bringing the mercury in both limbs to the same level, which is easily done by allowing some mercury to flow out by loosening the screw-tap COMPOSITION OF HYDROCHLORIC ACID 177 at the bottom of the U-tube. The hydrochloric acid gas will be completely decomposed by contact with sodium amalgam, chloride of sodium being formed, whilst the hydrogen is left in the ga,seous state. This will be found to occupy exactly half the volume of the original gas, the level of the mercury having risen to a mark previously made and indicating exactly one-third of the capacity of the tube. As the closed limb is provided with a stopcock, the residual gas may be inflamed, and thus shown to be hydrogen. 90 It still, however, remains to ascertain the volume of the chlo- rine which has disappeared. This is done as follows : Two glass tubes about 50 cm. long and 1'5 cm. in diameter, drawn out at each end to fine threads, are filled with the gaseous mixture evolved by the electrolysis of the aqueous acid (see p. 170). The process is conducted exactly as if a bulb were being filled, and the tubes are then sealed up and kept in the dark. When it is desired to exhibit the composition of the gas, one of the tubes thus filled is brought into a dimly-lighted room, and one of the drawn- out ends broken under mercury. No alteration in the bulk of the gas will be noticed. The mercury in which the tube dips is now replaced by a colourless solution of iodide of potassium, and by giving the tube a slight longitudinal shaking, a little of this solution is brought in contact with the gas. No sooner does the liquid enter the tube than it becomes of a dark brown colour, due to the liberation of the iodine, the chlorine uniting with the potassium to form the chloride of that metal. A consequent diminution of bulk occurs which corresponds precisely to the volume of chlorine contained in the tube, and in a few moments the column of liquid fills half the tube, proving that half the volume of the mixed gas con- sists of chlorine, and the other half of hydrogen. We have, however, learnt from the previous experiment that hydrochloric acid gas contains half its own volume of hydrogen, so that we have now ascertained (1) that hydrochloric acid gas is entirely made up of equal volumes of chlorine and hydrogen, and (2) that these elementary components combine together without change of volume to produce the compound hydrochloric acid gas. This fact may be further illustrated by exposing a second sealed-up tube, containing the electrolytic gas, for a few minutes, first to a dim, and then to a stronger daylight. The greenish colour of the chlorine will soon disappear, a gradual combination 13 178 THE NON-METALLIC ELEMENTS of the gases having occurred. On breaking one end of the tube under mercury, no alteration of bulk will be observed, whilst on raising the open end into some water poured on the top of the FIG. 50. mercury, an immediate and complete absorption will be noticed, and the tube will become filled with water. In order to determine with a greater degree of exactitude than is possible by the above methods, the relation existing between the two gases, a quantitative analysis of the chlorine contained in a given volume of the electrolytic gas must be MANUFACTURE OF HYDROCHLORIC ACID 179 made. Two experiments thus conducted gave the following results : I. II. Calculated. Chlorine ... 49*85 ... 50*02 ... 50*00 volumes. Hydrogen ... 50*15 ... 49*98 ... 50*00 100-00 100-00 100-00 Showing that the gas obtained by decomposing aqueous hydro- chloric acid consists exactly of equal volumes of chlorine and hydrogen, or vol. of chlorine weighing 1 = 17*595 J hydrogen 1 = 0'500 1 vol. of hydrochloric acid weighing ... 18*095 91 Hydrochloric acid gas is very soluble in water, and the solu- tion is largely used for laboratory and for commercial purposes, and frequently termed muriatic acid. In order to exhibit the solubility of the gas in water, a large glass globe (Fig. 50) placed on a stand is filled, by displacement, with the gas ; a tube, reaching to the centre of the globe and dipping to the bottom of an equal-sized globe placed beneath, being fixed in a caoutchouc stopper placed into the neck of the upper globe. Between the two globes the tube is joined by a piece of caoutchouc tubing, closed by a screw-tap. When it is desired to show the absorption, the lower globe is filled with water coloured blue by infusion of litmus, the screw-top is opened and a little of the water forced into the upper globe (so as to begin the absorption) by blowing through the side tube into the space above the surface of liquid in the lower globe. As soon as the water makes its appearance at the top of the tube, a rapid absorption occurs, the liquid rushes up in a fountain, and at the same time becomes coloured red. 92 Manufacture of Hydrochloric Acid. This acid is obtained on the large scale as a bye-product in the manufacture of soda-ash, (Cf. Vol. II. Part I. p. 132.) In the alkali works 10 cwt. of salt is introduced into a large hemispherical iron pan, 9 feet in diameter, heated by a fireplace underneath, and covered by a brickwork dome ; upon this mass of salt the requisite 180 THE NON-METALLIC ELEMENTS quantity (10 cwt.) of sulphuric acid (sp. gr. I 1 7) is allowed to run from a leaden cistern placed above the decomposing FIG. 51. pan. Torrents of hydrochloric acid gas are evolved, which collect in the space between the pan and the brickwork dome, whence they pass by a brickwork or earthenware flue into MANUFACTURE OF HYDROCHLORIC ACID 181 upright towers or condensers, built of bricks soaked in tar, or of Yorkshire flags fitted and clamped together. These towers, shown in vertical section in Fig. 51, and in ground plan in Fig. 52, are filled with bricks or coke, down which a small stream of water, from a reservoir at the top of the tower, is allowed to trickle. The gas passing upwards, as shown in the figures by the arrow, meets the water and is dissolved by it ; and as the acid -liquor approaches the bottom of the tower it becomes more and more nearly saturated with the gas. The aqueous commercial acid thus obtained from impure materials is generally far from pure ; it is usually of a yellow i {-r/ ('-', --j * i 'i \\ ,\ I '/ V 1 v ' /'/ \\ V / // \\ ^L.S// /S>J-,'' // ^J_-^'' ,'VVVJ -->' - - Ax S S .'* ' 5 10 15 20 25 SOPeot colour, due to organic matter, and may also contain sulphur dioxide, sulphuric acid, chlorine, iron, and arsenic : this last is often present in large quantities, being derived from the pyrites used in making the sulphuric acid. The presence of arsenic may be detected by Marsh's reaction ; or by the addition of stannous chloride, which produces a brown precipitate of impure arsenic. To remove traces of arsenic, solution of stannous chloride may be added, the precipitate allowed to settle, and the clear liquid re-distilled. Chlorine may be detected by the addition to the diluted acid of pure iodide of potassium and starch solution, when if chlorine be present the blue iodide of starch will be formed. The presence of 182 THE NON-METALLIC ELEMENTS sulphuric acid can be easily ascertained by adding chloride of barium solution to the diluted acid, whilst that of sulphurous acid may be shown by adding zinc to the diluted acid, when sulphuretted hydrogen will be given off and its presence readily FIG. 53. ascertained by its blackening action on lead paper. It is, however, not easy to separate these substances so as to obtain a strong pure acid from one originally impure, and by far the simplest plan is to exclude the foreign matters by employing pure materials to begin with. PROPERTIES OF HYDROCHLORIC ACID 183 93 The pure saturated aqueous acid is a colourless liquid fuming strongly in the air, and freezing when cooled below - 40 to a butter-like mass having the composition HC1 -f 2H 9 O. It is prepared for laboratory use by means of the apparatus shown in Fig. 53. One volume of water at absorbs 503 times its volume of hydrochloric acid gas. The weight and volume of the gas absorbed under the pressure of 760 mm. by one gram of water at different temperatures is given in the following table. 1 Temp. Grms. HC1. Temp. Grms. HC1. .... 0-825 32 .... 0-665 4 .... 0-804 36 .... 0-649 8 .... 0-783 40 ... . 0633 12 .... 0762 44 .... 0-618 16 .... 0-742 48 .... 0-603 20 .... 0-721 52 .... 0-589 24 .... 0-700 56 .... 0575 28 .... 0-682 60 .... 0'561. The weight of gas dissolved under changing pressure (the tempe^- rature remaining constant) does not vary proportionally to the pressure, and, therefore, this gas does not follow Dalton and Henry's law. Thus, for instance, under the pressure of 1 metre of mercury 1 grm. of water dissolves 0'856 grm. of the gas ; according to Dalton and Henry's law the weight of gas absorbed under a pressure of 1 decimetre of mercury should be 0'0856 .grm., whereas it is found to be 0'657 grm. On heating a saturated solution of the gas in water having a specific gravity of T22, hydrochloric acid gas is given off, and the liquid becomes weaker. On the other hand, a weak acid on being boiled loses water and becomes stronger, so that at last both the strong acid and the weak acid reach the same strength, and both when boiled distil over unchanged, provided the pressure does not vary. The aqueous acid, which boils unchanged at 110 under the normal pressure contains 20*24 per cent, of hydrochloric acid HC1. 2 If the distillation proceeds under a greater or less pressure than the normal, distillates of constant composition are obtained, but each one contains a different quantity of hydrochloric acid. This is clearly seen from the following table. 1 Roscoe and Dittmar, Quart. Journ. Chem. 1860, 128. 2 Roscoe and Dittmar . loc. cit. 184 THn NON-METALLIC ELEMENTS Column I. gives the pressure in metres of mercury under which distillation was conducted ; Column II. the percentage of hydrochloric acid (HC1) found in the residual acid. I. II. I. II. I. II. 0-05 23-2 0-8 20-2 17 18'8 O'l 22-0 0-9 19-9 1-8 187 0-2 22 3 1-0 197 1'9 18'6 0-3 21-8 1-1 19'5 2-0 18'5 0-4 21-4 1-2 19-4 21 18-4 0-5 21-1 1-3 19-3 2-2 18'3 0-6 207 1-4 191 2-3 18'2 07 20'4 1-5 19-0 2-4 181 076 20-24 1-6 18-9 2'5 18'0 Here the percentage of the acid and the constant composition obtained by distillation under a pressure of 0'5 decimetre is seen to be 23'2 HC1 ; whereas when the pressure is increased to 2'5 metres the percentage of the acid of constant boiling point is 18'0 HC1. From this it is clear that definite hydrates of hydro- chloric acid (i.e., compounds of hydrochloric acid and water in simple atomic proportions) are not formed on distillation, al- though it happens that by chance the liquid distilling under a pressure of 760 mm. corresponds to HC1 + 8H 2 O. In the same way, if dry air is passed through aqueous hydrochloric acid a part of the acid is vaporized and a residue is obtained which for each given temperature remains of constant composition. An acid weaker or stronger than this ultimately attains this composition. The following table shows the composition of the constant aqueous hydrochloric acids obtained by leading air at given temperatures through the liquid. Temp. Per Cent, of HC1. Temp. Per Cent, of HC1. 25-0 60 ..... 23-0 10 247 70 22-6 20 24-4 80 22-0 30 241 90 21-4 40 23-8 100 207 50 . . . . . 23-4 Hence it is seen that an aqueous acid which boils unaltered under a given pressure, and, therefore, at a constant tempera- ture, contains the same percentage of HC1 as the constant acid THE CHLORIDES 185 obtained by passing dry air through the aqueous acid. Thus the boiling point of the acid under 0*1 metre of pressure, con- taining 22-9 per cent, of HC1, is from 61 to 62 ; and if dry air be passed through an aqueous acid at 62 the constant point is attained when the liquid contains 22*9 per cent, of HC1. 94 The following table gives the specific gravity of solutions of aqueous hydrochloric acids of varying strengths, according to the experiments of Kolb. 1 100 of Aqueous Specific Gravity at Acid contain HC1. 2*22 3-80 6-26 11-02 15-20 18-67 20-91 2372 25-96 29-72 31-50 34-24 36-63 38-67 40-51 41-72 43-09 From these numbers the percentage of any acid of known specific gravity can easily be found by interpolation. 95 The Chlorides. The compounds of chlorine with the metals are formed either by the direct union of chlorine with a metal or by the replacement of the hydrogen in hydrochloric acid by a metal. Certain metals enter very readily into combination with chlorine, heat being always evolved, and the phenomena of combustion frequently observed. Other metals again do not combine so easily. Most of the metallic chlorides are soluble in water ; amongst those insoluble are silver chloride AgCl, mercurous chloride (calomel) HgCl, and cuprous chloride 1 Compt. Rend. 74, 737. 15 1-0116 1-0103 1-0202 1-0189 1-0335 1-0310 1-0581 1-0557 1-0802 1-0751 1-0988 1-0942 1-1101 1-1048 1-1258 1-1196 1-1370 1-1308 1-1569 1-1504 1-1666 1-1588 1-1806 1-1730 1-1931 1-1844 1-2026 1-1938 1-2110 1-2021 1-2165 1-2074 1-2216 1-2124 186 THE NON-METALLIC ELEMENTS Cu 2 Cl 2 . Many metals combine in more than one proportion with chlorine, thus we find : Cuprous Chloride, Cu 2 Cl 2 , Cupric Chloride, CuCl 2 . Mercurous Chloride, HgCl, Mercuric Chloride, HgCl 2 . Tin Bichloride, SnCl 2 , Tin Tetrachloride, SnCl 4 . Platinum Bichloride, PtCl 2 , Platinum Tetrachloride, PtCl 4 . Ferrous Chloride, FeCl 2 , Ferric Chloride, FeCl 3 . The chlorides of the metals are usually prepared by one of the following processes. (1) By acting on the metal with chlorine gas, especially when the anhydrous chloride is required. (2) By the action of chlorine upon metallic oxides, when it drives off the oxygen and unites with the metal to form a chloride. (3) By acting on the metal with hydrochloric acid. (4) By dissolving the oxide, hydrate, or carbonate of the metal in hydrochloric acid. (5) In certain cases, by adding a soluble chloride to a solution of a salt of the metal, when the metallic chloride is obtained as an insoluble precipitate. Chlorine also unites with all the non-metallic elements, and with certain groups of atoms termed radicals, to form chlorides of these elements and radicals respectively, some examples of which are as follows : Non-Metallic Chlorides. Hydrochloric Acid HC1. Chloride of Sulphur SC1 2 . Trichloride of Boron . , BC1 3 . Tetrachloride of Silicon SiCl 4 . Pentachloride of Phosphorus PC1 5 . Chlorides of Inorganic Radicals. Chloride of Sulphuryl S0 2 C1 2 . Chloride of Phosphoryl POC1 3 . Chlorides of Organic Radicals. Chloride of Ethyl C 2 H 5 C1. Chloride of Ethylene C 2 H 4 C1 2 . Chloride of Acetyl C 2 H 3 OC1. Chloride of Cyanogen CNC1. 96 Detection and Estimation of Chlorine. In the free state chlorine gas is recognized by its peculiar colour, its suffocating DETECTION OF CHLORINE 187 smell, and by its bleaching action on organic colouring matters. When present in smaller quantities, its presence may be de- tected by the blue colour which it causes on a paper moistened with a solution of iodide of potassium, KI, and starch paste, owing to the fact that chlorine liberates iodine from its com- pound with potassium, combining with the metal to form the chloride, KC1, whilst the liberated iodine forms a deep blue compound with starch. This reaction is very delicate, but it must be remembered that an excess of chlorine again removes the blue colour, and also that the same effect is produced by bromine, nitrous fumes, ozone, and other oxidising substances. When combined with metals to form chlorides soluble in water, the element is usually detected by the formation of the curdy white precipitate of silver chloride AgCl, on addition of a solution of silver nitrate, AgNO 3 , to that of a soluble chloride, such as KC1, thus : KC1 + AgNO 3 = AgCl + KN0 3 . One part of chlorine in one million parts of water can thus be detected a faint opalescence occurring. The precipitated silver chloride becomes violet-coloured on exposure to light, and is in- soluble in water and dilute acids, especially nitric acid, but readily soluble in ammonia and in solutions of potassium cyanide, and sodium thiosulphate (the so-called hyposulphite of soda). Mercurous nitrate likewise produces in solutions of a chloride a white precipitate of mercurous chloride (calomel), which does not dissolve, but turns black on addition of ammonia. In order to detect a chloride in presence of an iodide and bromide, the dried salt is distilled with potassium chromate and strong sulphuric acid, when chromium oxychloride, CrO 2 Cl. 2 , distils over as a dark red liquid, decomposed by addition of water or ammonia, and yielding a yellow solution which, on addition of hot acetic acid and a soluble lead salt, gives a yellow precipitate of lead chromate ; neither bromine nor iodine forms a similar compound with chromium. Chlorine, when combined to form a chloride, is always estimated as silver chloride, AgCl, and according to Stas 142 '32 of silver chloride contain 35 '19 of chlorine. If the chlorine is present in the free state it can be determined by volumetric analysis (see Chlorimetry, under Bleaching Powder), or it may be reduced by sulphur di- oxide, SO 2 to hydrochloric acid, and then precipitated as silver chloride and weighed. 188 THE NON-METALLIC ELEMENTS The atomic weight of chlorine was first determined by Berzelius, 1 together with that of silver and potassium. Penny 2 and Marignac 3 have also made similar determinations; the latter obtaining the numbers 35*372 and 35*462 by a method similar to that employed by Berzelius. It is to Stas 4 that we owe the most exact determinations. He converted pure silver into silver chloride by four different methods, and found as the mean of closely agreeing results that the relation of the weight of silver to that of the silver chloride it yields is 1 : T3285. It is further known that the relation of the atomic weights of silver and oxygen is 6 '745 6 : 1, this ratio having been also obtained from a large number of experiments, such for example, as the ratios of equivalent quantities of the following : Ag 2 S : Ag 2 S0 4 and AgCl : AgClO 3 . Hence if the atomic weight of oxygen is taken as 15*88 that of chlorine is 3519. BROMINE. Br = 79-36 97 BROMINE does not occur in the free state in nature ; it was discovered in the year 1826 by Balard, 5 who prepared it from the liquor called bittern, remaining after the common salt has crystallised out from concentrated sea-water, in which it occurs, combined with metals to form bromides ; he gave it the name from /fyw/i-o?, a bad smell. Bromine occurs in combination with silver in certain ores, from Mexico, Chili, and Bretagne ; but is found in large quantities (combined with sodium, potassium, magnesium, or calcium, forming bromides) in the water of many mineral springs, some of which contain enough to serve as a source of this element. It is also found, though in very small quantity, in all sea-water, 6 and has been detected in sea- weed from many 1 Pogg. Ann. 8, 1. 2 Phil. Trans. 1839, p. 20. 3 Annalen, 44, 11 ; Bill. Univ. de Geneve, 43, 350. 4 Recherches sur Us Lois des Proportions Chimiques, Bmxelles, 1865. 5 Ann. Chim. Phys. [2], 32, 337. 6 The water of the Dead Sea is said to contain large quantities, of no less than -42 gram, in the litre (Lartet). BROMINE 189 localities, and even in certain marine animals, as well as in English rock-salt. The mineral springs at Kreuznach, Kis- singen, and Schonebeck, and the potash beds of Stassfurt, as well as certain American springs in Ohio and elsewhere, contain considerable quantities of bromine, the commercial article being obtained almost entirely from the two last-named sources. Preparation. In order to detect bromine in a mineral water, or to prepare it in small quantities, the following method is employed. The mother-liquor remaining after the brine from any of the above sources has been well crystallised is treated with a stream of chlorine gas, so long as the yellow colour of the liquid continues to increase in depth. Chlorine has the power of liberating bromine from bromides, itself uniting with the metal, and the bromine being set free, thus : MgBr 2 + Cl a = MgCl 2 + Br 2 . The addition of excess of chlorine is to be avoided, as a com- pound of chlorine and bromine is then formed. The yellow liquid is then well shaken with chloroform, which dissolves the bromine, forming, on standing, a brown solution below the aqueous liquid. On adding caustic potash to this solution the colour at once disappears, the bromine combining to form the bromide KBr, and bromate of potassium, KBr0 3 thus : 3Br 2 + 6KHO = KBrO 3 + 5KBr + 3H 2 O. On concentrating the solution a mixture of these salts remains, and from these the bromine is again liberated by dis- tilling the liquid with black oxide of manganese and sulphuric acid in a tubulated retort. The decomposition which here occurs is similar to that which takes place in the preparation of chlorine, thus : 2KBr + 3H 2 S0 4 + Mn0 2 = Br 2 + 2KHSO 4 + MnSO 4 + 2H 2 O. Dark red fumes of bromine are liberated, and a black liquid condenses in the well-cooled receiver. If the bromine is required to be anhydrous it must be redistilled over concentrated sulphuric acid ; and if iodine is present this must be got rid of previously by precipitation as subiodide of copper. By far the greater quantity of the bromine brought into 190 THE NON-METALLIC ELEMENTS commerce is now manufactured at Stassfurt from the mother- liquor remaining after the separation of the potassium salts contained in the salt deposits, the process adopted consisting in the treatment of the liquors with chlorine under suitable con- ditions. The most recent apparatus for this purpose is shown in Fig. 54, and is so arranged that the process is continuous. The mother-liquor enters the apparatus by the hydraulically sealed pipe a, and by means of the sandstone drum b, and perforated plate e, is distributed equally over the whole area of FIG. 54. the tower A. The latter is filled with balls over which the liquor flows, thus exposing a large surface to the action of the chlorine gas passing through the tower in the opposite direction ; the waste liquor passes away by the pipe d, which is sufficiently large to allow of the simultaneous passage of the chlorine gas to the tower from the generator D. In order to free the waste liquor completely from all traces of chlorine and bromine it is run into the vessel B, which is kept full to the bottom of the pipe d ; to pass away from B the liquor must pass over the sandstone shelves, in the direction shown by the PROPERTIES OF BROMINE 191 arrows, in doing which it is subjected to the action of a current of high pressure steam which is introduced into the apparatus by the pipe g. All the chlorine and bromine are thus driven out of the liquor and rise to the top of B, where they mingle with the chlorine gas passing from the generator to the tower, whilst the now quite innocuous liquor passes away by the pipe i. The bromine vapour and excess of chlorine pass out from the top of the tower by the pipe o, through the condenser p, where most of the bromine condenses, the last traces of the bromine and chlorine being removed by the vessel c, which contains iron filings kept moist by a small stream of water. The bromine thus obtained is purified by redistillation, the small quantities of chlorine present being removed by the addition of potassium- or ferrous-bromide, or by collecting separately the more volatile portions of the distillate, which contain all the chlorine in the form of chloride of bromine. The manufacture of bromine was commenced in Stassfurt in 1865, in which year the quantity obtained only amounted to 25 cwt. ; in 1885 the amount had risen to 260 tons, whilst the quantity manufactured in America during the same year was estimated at 120 tons, and elsewhere 20 tons, giving a total production of about 400 tons. 98 Properties. Bromine is a heavy mobile liquid, so dark as to be opaque except in thin layers. It is the only liquid element at the ordinary temperature except mercury. Its specific gravity at is 3'1883 : it freezes at 7 to a dark brown solid, evaporates quickly in the air, and boils at 59 (Thorpe). Bromine possesses a very strong unpleasant smell, the vapours when inhaled produce great irritation, and affect the eyes very painfully. When swallowed, it acts as an irritant poison, and when dropped on the skin it produces a corrosive sore, which is very difficult to heal. In its general properties, as well as in those of its compounds, bromine closely resembles chlorine, although they are not so strongly marked. Thus it bleaches organic colouring matters, but much less quickly than chlorine does, and it combines directly with metals to form bromides, though its action is less energetic than that of chlorine. It does not combine at all at ordinary temperatures with metallic sodium ; indeed these two substances may be heated together to 200 before any perceptible action commences, whereas bromine and potassium cannot be brought together without combination occurring, sometimes with almost 192 THE NON-METALLIC ELEMENTS explosive violence. l The addition, however, of a drop of water to bromine and clear sodium sets up a lively reaction. If brought into contact with free bromine, starch-paste is coloured orange yellow. The vapour density of bromine at temperatures slightly above its boiling point is somewhat higher than the normal namely, 5*8691 (air = 1), but at 228 the density is 5'5247, showing that at that temperature the bromine molecules contain 2 atoms. 2 Moreover, if bromine vapour be mixed with 10 volumes of air, its vapour density, calculated from that of the mixture, corresponds to the formula Br 2 even at the ordinary temperature. 3 At about 1570 the observed density is about I of that at 228, so that, as in the case of chlorine, the molecules Br 2 are at this temperature partially dissociated into molecules consisting of simple atoms. 4 Bromine is largely employed in the colour industry and also in medicine, whilst smaller quantities are used in analytical and synthetical chemistry, and as a disinfectant. To employ it for the latter purpose, advantage is taken of the fact that the siliceous earth known as " kieselguhr " absorbs as much as 75 per cent, of its weight of bromine, and still retains its solid form ; the product is sold under the name of " bromum solidificatum." Bromine and Water. A definite crystalline compound of bromine and water is obtained by exposing a mixture of the two substances to a temperature near the freezing point. This hydrate consists of Br + 5H O, and undergoes decomposition into bromine and water at 15. The analysis of the compound gave the following results : Calculated. Found (Lowig). Bromine. . . . 79*36 47'00 45'5 5H 2 O .... 89-40 53-00 54'5 168-76 100-00 100-0 From the dissociation pressure, Roozeboom 5 believes that it has the composition Br 4- 4H 2 O or Br 2 + 8H 2 O. The solubility of bromine in water between the tem- 1 Merz und Weith, Per. 6, 1518. 3 Jahn, Ber. 15, 1238. 3 Langer and v. Meyer, Ber. 15, 2773. 4 V. Meyer and Ziiblin, Ber. 13, 405 ; Crafts, Compt. Rend. 90, 183. 5 Bee. Trav. CMm. 3, 73. BROMINE AND WATER 193 peratures of 5 and 30 , 1 is shown in the following table, 100 grams of saturated bromine water containing by weight At 5 . . 3*600 grams of bromine. 10 ... 3-327 15 ... 3-226 20 ... 3-208 25 ... 3-167 30 ... 3126 The solution of bromine in water has an orange red colour ; it soon loses bromine in contact with the air, and bleaches organic colouring matter. Bromine water is permanent in the dark, but on exposure to sunlight it becomes acid from the form- ation of hydrobromic acid and evolution of oxygen. Bromine also dissolves readily in chloroform, carbon disulphide, alcohol, ether, and acetic acid. The atomic weight of bromine has been determined by Marignac and by Stas ; the latter obtained as a mean of a large number of experiments 79*36 as the atomic weight of bromine when that of oxygen is 15*88. BROMINE AND HYDROGEN. HYDROBROMIC ACID. HBr = 80-36. 99 Bromine, like chlorine, forms only one compound with hydrogen, containing one atom of bromine and one of hydrogen, but, unlike chlorine, these two bodies do not unite to form hydrobromic acid when brought together in sunlight. If, however, hydrogen and the vapour of bromine are passed through a red-hot tube containing finely-divided metallic plati- num or pieces of charcoal, 2 a combination occurs of equal volumes of bromine and hydrogen and formation of hydro- bromic acid gas. The combination of these two bodies may be easily shown by passing hydrogen over bromine vapour and lighting the escaping gas, when dense fumes of hydrobromic acid will be noticed. Preparation. Hydrobromic acid gas cannot well be prepared by the action of the ordinary acids on the bromides, as in the 1 Dancer, Journ. Ghem. Soc. 1862, ii, 477. 2 Merz and Holzman, Bcr. 22, 867. 14 194 THE NON-METALLIC ELEMENTS case of hydrochloric acid, owing to the facility with which hydrobromic acid splits up, with formation of free bromine. If, however, phosphoric acid be used, free hydrobromic acid is obtained. Sulphuric acid of sp. gr. 1*41 liberates hydrobromic acid from potassium bromide without simultaneous formation of bromine, and in this manner a dilute solution of the acid may be obtained, which may be concentrated by distillation. 1 One of the best methods of preparing hydrogen bromide is to bring bromine and phosphorus together in presence of a little water, when a violent action occurs, hydrobromic acid gas and phosphoric acid being formed : p + 5Br + 4H 2 = 5HBr + H 3 P0 4 . FIG. 55. In order to prepare the gas a flask provided with a doubly bored caoutchouc cork, Fig. 55, is made use of; through one of the holes a gas delivery-tube is fixed, whilst through the other a stop- pered funnel tube is passed. A mixture of one part by weight of amorphous phosphorus and two parts of water is introduced into the flask, and ten parts of bromine are allowed to fall drop by drop through the stoppered funnel-tube on to the mixture in the flask. As each drop falls in a sudden evolution of gas occurs, accompanied in the first part of the operation by a flash of light, and as soon as a certain amount of hydrobromic acid has been formed the bromine dissolves quietly, and on gently warming the flask hydrobromic acid gas is given off. This is then allowed to pass through a U-tube containing amorphous i Feit and Kubierschky, J. Pharm. [5], 24, 159. HYDROBROMIC ACID 195 phosphorus, to free it from any vapour of bromine, and may be collected in dry stoppered cylinders by displacement or over mercury. The same apparatus serves for preparing a saturated aqueous solution of the gas ; for this purpose the gas delivery- tube is removed and a short tube substituted for it. This passes through a cork fitting in the tubulus of a retort placed in the position shown in Fig. 56 ; the neck of the retort dips under water, and the retort itself serves as a safety tube in case the gas be absorbed so quickly that the liquid rushes back, as then the solution is not sucked back into the flask, but simply rushes into the upper part of the retort. A second method is to pass a current of hydrogen sulphide evolved from one of the continuous apparatus through a layer of bromine contained in a tall cylindrical vessel and covered by a layer of water or hydrobromic acid, when the following reaction takes place : Br 2 + H 2 S = 2HBr + S. The liberated sulphur partially combines with the excess of bromine, forming bromides of sulphur. The gas evolved is washed by passing through a solution of potassium bromide in hydrobromic acid containing a little amorphous phosphorus in suspension, and is thus obtained free from bromine and sulphur- etted hydrogen ; with this method the evolution of the gas can be regulated very exactly. 1 A third method is to pass a mixture of hydrogen and bromine . vapour through a tube in which a platinum spiral is heated to bright redness. For this purpose a glass tube about 18 cm. in length and 15 mm. in width is fitted at each end with a cork carrying a small tube and a piece of stout copper wire ; the ends of the stout wires are joined within the tube by a platinum spiral 1 in. in length, which is maintained at a bright red heat by means of an electric current ; a stream of hydrogen is then bubbled through bromine heated to 60 and passed through the wide tube, the small tube being plugged with glass wool to prevent possible explosions. So long as the hydrogen is in slight excess, the hydrogen bromide is quite free from bromine, and the presence of a small quantity of hydrogen is of no importance for most purposes, especially when the aqueous solution of the acid is required. 2 1 Recoura, Compt. Rend. HO, 784. 2 Newth, Chcm. News, 1891, ii. 215. 196 THE NON-METALLIC ELEMENTS 100 Properties. Hydrobromic acid is a colourless gas, having a strong irritating smell, with an acid taste and reaction. It fumes strongly in the air, and on exposure to a temperature of 73 (obtained by the evaporation of a mixture of ether and solid carbonic acid), it condenses to a colourless liquid, and afterwards freezes at - 87 to a colourless ice-like solid. 1 Like hydrogen chloride, hydrogen bromide is decomposed by sunlight in presence of oxygen and moisture, with liberation of bromine, but the dry mixture of the gases is unaffected. 2 FIG. Pure aqueous hydrobromic acid is colourless, and remains so even when exposed to air ; it fumes when saturated at 0, and then possesses a specific gravity of 178. The weak aqueous acid becomes stronger, and the concentrated acid weaker on distilla- tion, until an acid containing from 47'38 to 47'86 per cent, of HBr, distils over under pressures varying from 0752 to 0762 1 Faraday, Phil. Trans. 1845, p. 155. 2 Richardson, Journ. Chcm. Soc. 1887, i. 804. HYDROBROMIC ACID 197 metres. When the pressure under which the distillation occurs varies, the composition of the constant acid changes as that of hydrochloric acid does, and if a stream of dry air be passed through the aqueous acid a point is reached, different for each temperature, at which the acid no longer undergoes change. Thus the acid which evaporates unchanged in air at 100 contains 49'35 per cent. HBr, whilst that obtained at 16 contains 51-65 per cent of HBr. The variation of the specific gravity of the aqueous acid with the percentage of hydrobromic acid dissolved has been determined by Topsoe, 1 as also by C. R. A. Wright,' 2 who obtained the following numbers: Per Cent. HBr. Spec. Grav. at 15. 10-4 1-080 23-5 1190 30-0 . 1-248 40-8 1-385 48-5 1-475 49-8 1-515 The composition of this gas is analogous to that of hydro- chloric acid, and can be ascertained in a similar way by bringing a given volume of the dry gas in contact with sodium amalgam, when sodium bromide is formed and hydrogen liberated, the volume of which is found to be exactly half that of the original hydrobromic acid gas. 101 The Bromides. These compounds are formed in a similar manner to the corresponding chlorides. They possess an analogous composition with these, and exhibit similar properties. Bromine unites with nearly all the metals, forming bromides, which are also produced by the action of metals on hydrobromic acid, or by the action of vapour of bromine on the metallic oxides, oxygen being liberated. The metallic bromides are nearly all soluble in water, the most insoluble being silver bromide, AgBr, mercurous bromide, HgBr, and lead bromide, PbBr 2 , which latter is slightly soluble. All the bromides are solid at the ordinary temperature, but when heated they fuse and volatilize, some undergoing decomposition others remaining unchanged. They are, however, all decomposed by chlorine, either in the cold or upon heating, a metallic chloride being formed and bromine liberated ; they are .also decomposed 1 Ber. 3, 404. 2 Chem. News, 23, 242. 198 THE NON-METALLIC ELEMENTS by sulphuric and nitric acids, with evolution of hydrobromic acid, which again is partly oxidized, bromine being set free. 102 Detection and Estimation of Bromine. Bromine when in the free state may be recognized by the red colour of its vapour, and by its exceedingly disagreeable odour, and by imparting to starch paste an orange-yellow colour. When present in small quantities it may be detected by shaking up with chloroform or ether, which dissolves it, and acquires thereby a red or brownish colour. Bromine in the state of a soluble bromide may be detected by giving with silver nitrate a yellowish white precipitate of silver bromide, which is insoluble in nitric acid, and dissolves only with difficulty in ammonia, but readily in cyanide of potassium. Also by giving with nitrate (but not chloride) of palladium a reddish- brown precipitate of the bromide ; by its tinging carbon di- sulphide yellow in presence of hydrochloric acid and a drop of sodium hypochlorite ; and by the liberation of bromine on heat- ing with sulphuric acid, with sulphuric acid and manganese dioxide, or with sulphuric acid and potassium bichromate. When the bromine is present as a soluble bromide it is usually estimated by precipitation as bromide of silver, which contains 42*42 per cent, of bromine. In presence of chlorine the two elements are precipitated together by nitrate of silver, the pre- cipitate is then fused and weighed. A portion of it is next ignited in a current of chlorine, when the whole of the bromine is expelled, the residue of silver chloride weighed, and from the weight thus obtained and that of the mixed silver salts, the quantities of chlorine and bromine are calculated. For every 79*36 parts of bromine expelled, 35*19 parts of chlorine have been substituted; or, if a difference of 44*17 is observed, 79*36 parts of bromine must have been present. Hence, for any other difference the weight of bromine is found by multiplying that difference by ^?|= 1'797. 44*17 CHLORINE AND BROMINE. 103 Bromine monochloride, BrCl. When chlorine gas is passed into liquid bromine cooled below 10 it is largely absorbed, a red- dish brown volatile mobile compound of the two elements being formed. 1 It dissolves in water yielding a yellow solution, from 1 Balard, Ann. CMm. Phys. 32 337 : Bornemann. Annalcn, 189, 183. IODINE 199 which, on cooling below 0, a crystalline hydrate BrCl + 10H 2 separates out, which melts at 4- 7. This is possibly simply a mixture of chlorine and bromine hydrates. IODINE. 1=125-91. 104 Iodine was discovered in 1812 by Courtois, 1 of Paris, in the mother-liquors of the soda salts which are prepared from kelp or burnt seaweed. It was afterwards examined by Davy, 2 and much more completely by Gay-Lussac. 3 Iodine derives its name from loeiSys, violet-coloured, owing to the peculiar colour of its vapour, by means of which it was first discovered. Like chlorine and bromine, iodine does not occur in the free state in nature, but is found combined with metals to form iodides, which occur in small quantities, but widely diffused, both in the organic and inorganic kingdoms, having been detected in sea-water, in sea-plants and animals, and in many mineral springs. The quantity of iodine present in sea-water is extremely small, but certain plants and even animals have the power of absorbing and storing up the iodine. The ash of the deep-sea-weed (Fucus palmatus especially) contains more iodine than that which grows in shallow water, and it is from the weed collected on exposed and rocky coasts, as the north and west coasts of Ireland, Scotland, and France, that a large portion of the iodine of commerce is obtained. Of late years, however, the quantity manufactured from kelp has considerably decreased, owing to the discovery of iodine in the crude Chili saltpetre or Caliche NaN0 3 , from the mother liquors of which it is now chiefly obtained ; it is likewise found in combination with silver in a Mexican silver ore, in some specimens of South American lead ore, in certain dolomites, and in small quantities in almost every deposit of rock salt. Iodine has also been found in coal, and it has been detected in some few land and fresh- water plants, and in many sea animals, as in sponges and oysters, and also in cod liver oil. 105 Preparation from Seaweed. The stormy months of the spring are those in which the deep-sea tangle is thrown up on 1 Courtois, Clement, and Desormes, Ann. Chim. 88, 304. 2 Phil. Trans. 1814, 2, 74 and 487. 3 Ann. Chim. 88, 311, 319, and 91, 5. 200 THE NON-METALLIC ELEMENTS the north coasts of France and Ireland and the western coasts of Scotland. The inhabitants collect the weed, allow it to dry during the summer, and then burn it in large heaps. The ash thus obtained is termed kelp in Scotland and varec in Normandy ; it contains from O'l to O3 per cent, of iodine. When the seaweed is completely burnt, and when the ash is fused, a con- siderable fraction of the iodine is lost from volatilization ; hence it is preferable to carbonize the weed in closed retorts, the whole of the iodine remaining in the ash, whilst the tar and ammonia- cal liquor are also recovered ; the carbon remaining after the FIG. 57 lixiviation of the ash resembles animal charcoal, and is used as a disinfectant. In the most recent process for obtaining the iodine from sea- weed the latter is directly lixiviated without previous carboniza- tion, the residual apparently unaltered plants being converted into algin, a substance which resembles gelatin and is employed as a substitute for bladder skins, &c. On lixiviating systematic- ally either the kelp or the carbonized weed, a concentrated solution of the alkaline carbonates, chlorides, sulphates, and a small quantity of sulphites and sulphides, together with the MANUFACTURE OF IODINE 201 iodides and bromides of the alkali metals, is obtained, and from this solution the carbonates, chlorides, and sulphates are allowed to crystallize, leaving the bromides and the iodides in the mother-liquor. This liquor is then treated in several ways in order to obtain the iodine. (1) An excess of sulphuric acid is added to the liquor, when the sulphides and sulphites which it contains are decomposed and the iodine and bromine liberated as hydriodic and hydrobromic acid. In this process the liquor, after any separated crystals of sodium sulphate which may have formed have been taken out, is placed in iron boilers, Fig. 57, surrounded with brick- work, each gently heated by a separate fire to a temperature of 60, and fitted with leaden hoods, which can be lifted off by means of a chain and winch. Each cover is fitted with a leaden pipe (a), and this is connected with a series of glass or earthen- ware condensers, termed udells, fitting one into the other. After the introduction of the liquor, the covers are luted on with clay, the pipes (a) fixed in their receptacles and connected with the condensers. Manganese dioxide is then thrown little by little into the still through the hole (fr), which can be closed by a stopper. The iodine thus liberated condenses in the receivers, 1 and the accompanying water escapes through a tubulus at the bottom of each receiver and runs away along the channel (c). When no more iodine distils over, the leaden pipes are dismounted, and the stills are connected with a second receiver (D). More manganese dioxide is then added, and the bromine, which hitherto has not been liberated, is now disengaged and collects in (D) and in the WoulfFs bottles (E). The decomposition occurring during the formation of the iodine is represented by the following equation : 2NaI + 3H 2 SO 4 + MnO 2 = I 2 + 2NaHSO 4 + MnSO 4 + 2H 2 O. The iodine thus obtained may then be partially purified by resublimation, but even then invariably contains traces of chloride, bromide, and cyanide of iodine. (2) The iodine contained in the liquors, after separation of the crystallizable alkaline salts, may also be liberated by the addition of sulphuric acid containing a considerable quantity of nitric acid. The acidified liquor is then agitated with the most volatile portion of petroleum (petroleum-naphtha, kcrosine), which dissolves the iodine. The petroleum solution of iodine 1 One ton of kelp usually yields 12 Ibs. of iodine. THE NON-METALLIC ELEMENTS is next drawn off from the aqueous liquor and shaken up with an aqueous solution of caustic soda, whereby the iodine is withdrawn from the hydrocarbon and converted into iodide and iodate of sodium. The iodine is then liberated from these salts by the addition of hydrochloric acid, thus : (3) In France the mother liquor is treated with a slight excess of sulphuric acid, filtered from sulphur, diluted to 40 Tw., and saturated with chlorine, care being taken to avoid adding an excess of the latter, which would cause loss owing to the formation of chloride of iodine. The iodine separates out in the solid form and is filtered off, dried, and resublimed. Preparation from Caliche. The crude Chili saltpetre, known as Caliche, now forms the chief source of iodine ; to obtain the FIG. 58. iodine the last mother-liquor obtained in the preparation of the sodium nitrate, which contains about 20 per cent, of sodium iodate, is treated with a solution of sodium bisulphite, obtained by burning native sulphur and passing the products into a solu- tion of sodium carbonate. The iodine separates out in the solid form, and is filtered off and purified by resublimation, the vapours being condensed in a series of udells similar to those shown in Fig. 57. In order to purify the commercial iodine it is washed with a small quantity of water, dried on porous plates, and resublimed. According to Stas, 1 the only mode of obtaining chemically-pure iodine (free from every trace of chlorine and bromine) is to dissolve the commercial resublimed substance in iodide of potassium solution, and then to precipitate the iodine by water. The precipitate is well washed with water and then distilled 1 Rechcrches, p. 136. PROPERTIES OF IODINE 203 with steam, the solid iodine in the distillate collected and dried in vacuo over solid nitrate of calcium, which is frequently changed, and distilled afterwards over solid caustic baryta to remove the last traces of water and of hydriodic acid. 1 06 Properties. Iodine is a bright, shining, crystalline, opaque, blackish-grey solid. The crystals when large possess almost a metallic lustre ; it crystallizes by sublimation in the rhombic system, in the form of prisms or pyramids. Finer crystals are obtained from solution, as by exposing to the air a solution of iodine in ether, or in an aqueous solution of hydriodic acid. The crystals thus obtained have the ratio of their axes represent- ed by the numbers 4 : 3 : 2. The crystal represented in Fig. 58 a was obtained by Marignac from solution in hydriodic acid. Iodine is a heavy substance, having a specific gravity of 4'948 at 17, melting at 114-2 and solidifying at 113'6 (Stas). It boils at 184'35 under 760 mm. pressure, 1 giving rise to a vapour which, seen by transmitted white light, possesses, when chemi- cally pure, a splendid deep blue colour, but when mixed with air a reddish-violet colour (Stas). The specific gravity of iodine vapour was found by Deville and Troost to be 8*72 (air = l), which corresponds to the density, 125'9, proving that the molecule, or two volumes of iodine gas, weighs 125'91 x 2 = 251-82 and that the molecular formula is I 2 . When iodine vapour is heated above 700 its specific gravity begins to diminish until at 1700 it becomes constant, and is half that at 700, the vapour consisting of free atoms. 2 At the ordinary temperature it volatilizes slowly, shining crystals being deposited on the sides of a bottle on the bottom of which a little iodine has been placed. It is a bad conductor of electricity, and possesses a peculiar smell less penetrating than, though similiar to that of, chlorine and bromine. The specific heat of solid iodine is, according to the experiments of Regnault, 0'05412, and that of the liquid 010882. The latent heat of fluidity of iodine is 11 '7 thermal units, its heat of vaporization 23'95 thermal units. When an electric discharge is passed through a heated Geissler's vacuum-tube containing a trace of iodine vapour, a spectrum of bright lines is obtained, characteristic of this 1 Ramsay and Young, Journ. Chem. Soc. 1886, 1, 453. 2 V. Meyer, Bcr. 13, 394, 1010, 1103 ; v. Meyer and Biltz, Ber. 22, 725 ; Meier and Crafts, Compt. Rend. 90, 690 ; 92, 39. 204 THE NON-METALLIC ELEMENTS element. 1 This emission spectrum is, however, not identical with the characteristic absorption-spectrum of iodine, so care- fully mapped by Thalen, 2 and seen when white light is passed through iodine vapour. Salet 3 has recently shown that when an electric current of feeble tension is passed through a Geissler's tube containing iodine, another set of bright bands is obtained, which are identical in position with the dark bands of Thalen's absorption- spectrum, each bright band being replaced by a black band when the vapour is illuminated from behind. 107 In its chemical properties iodine resembles chlorine and bromine ; the two latter elements have the power of displacing iodine from its combination with metals (or electro-positive elements), thus : C1 = 2KC1 I. The compounds of iodine with oxygen (or with electro- negative elements) are, on the other hand, more stable than those of the other two elements. Thus iodine expels chlorine from the chlorates with formation of iodate and free chlorine : These differences are explained when we examine the amount of heat evolved by the several decompositions in question. This heat may be taken as a measure of the relative affinity or power of combination which the elements exhibit towards one another. Thus the heat evolved on the combination of chlorine, bromine, and iodine with hydrogen, or the heat modulus of the reactions according to Julius Thomsen's experiments 4 is : H + Cl . . . 21836 heat units H + Br . . . 8337 H + I . . . - 6060 When these same elements unite with oxygen and hydrogen to form the oxyacids (HC1O 8 , HBrO 3 , HIO 3 ) the heat evolved is as follows : Cl + 3 O + H . . 23761 heat units Br + 3 O + H . . 5344 I + 3 + H . . 43211 1 Plucker and Hittorf, Phil. Trans. 1865, 28. 2 Kon Svenska Acad. Handb. 1869. 3 Phil Mag. [4], 44, 156. 4 Journ. Chem. Soc. 1873, 1188. PROPERTIES OF IODINE 205 Hence we see that as regards affinity for oxygen chlorine stands nearly midway between bromine and iodine, for 43711 + 5344 Iodine dissolves but very sparingly in water, one part being soluble in 5524 parts of water at 10 ; but it dissolves freely in an aqueous solution of potassium iodide, and in alcohol, yielding brown solutions. Tincture of iodine of the pharmacopoeia contains J oz. of iodine, J oz. of iodide of potassium, rectified spirit 1 pint. It is also soluble in chloroform, carbon disulphide, and many liquid hydrocarbons, imparting to these liquids a fine violet or dark-red colour when present in small quantities, and in larger quantities forming a black opaque solution, which in the case of carbon disulphide is diathermous, allowing the invisible heating rays of low refrangibility to pass, though it is opaque to the visible rays. Although iodine, unlike chlorine and bromine, does not combine readily with hydrogen, it unites with many of the metals and non-metals with evolution of light and heat. Thus solid phosphorus, when brought into contact with iodine, first melts and then bursts into flame owing to the heat evolved in the act of combination : and powdered antimony takes fire when thrown into iodine vapour, antimony iodide being produced, whilst if the vapour of mercury be passed over heated iodine, immediate action occurs, the iodides of mercury being formed. When iodine is brought into contact with water and filings of iron or zinc, a violent reaction occurs, colourless solutions of the respective iodides resulting. The action of iodine upon the alkali-metals is analogous to that of chlorine and bromine. Sodium and iodine can be heated together without any alteration, whilst if potassium be employed an explosive combination occurs. Potash at once decolorises a solution of iodine, iodide and iodate of potassium being produced, thus : 3I 2 + 6KHO = 5KI + KI0 3 + 3H 2 O. When acted upon by strong nitric acid, iodine is completely oxidized to iodic acid HIO 3 . The most characteristic property of free iodine is its power of forming a splendid blue colour with starch-paste. This is formed when starch granules are brought into contact with the vapour of iodine, or, better, when a solution of iodine is added 206 THE NON-METALLIC ELEMENTS to starch paste. The blue colour disappears on warming the solution, but reappears on cooling, and its formation serves as a most delicate test for the presence of iodine. 1 In order ,to exhibit this property, a few grains of iodide of potassium may be dissolved in three or four litres of water placed in a large glass cylinder, and some clear, dilute, well-boiled starch-paste added. As the iodine is here combined with the metal, no coloration will be seen, but if a few drops of chlorine water be added, or, better, if a little of the air (containing free chlorine) from a bottle of chlorine water be poured on to the surface of the liquid, a blue film will be formed, which on stirring will impart a blue tint to the whole mass. Iodine both free and in combination is largely used in medicine. The atomic weight of iodine has been very accurately determined in several ways by Marignac and Stas, the mean value obtained from a large number of closely-agreeing experiments being 125-91 (O = 15-88). IODINE AND HYDROGEN. HYDRIODIC ACID. HI = 126-91. 108 Iodine and hydrogen undergo partial combination when they are passed over finely-divided platinum heated to redness, or over charcoal at a bright red heat, 2 forming a strongly acid gas, having properties very similar to hydrochloric and hydro- bromic acid. Hydriodic acid can also be obtained by heating iodide of potassium with phosphoric, but not with sulphuric, acid ; for when this latter acid is used, sulphur dioxide SO 2 and free iodine are formed at the same time, thus : 3H 2 S0 4 + SKI = 2KHS0 4 + I 2 + SO 2 + 2H 2 O. On the other hand, hydriodic acid is easily prepared by allow- ing iodine and phosphorus to act on one another in presence of water, thus : P + 51 + 4H 2 = SHI + H 3 P0 4 . Preparation. For this purpose 1 part by weight of amorphous phosphorus and 15 parts of water are brought together in a 1 Collin and Gaultier de Claubry, Ann. Chim. 90> 87. 2 Merz and Holzmann, Ber. 22, 867. HYDRIODIC ACID 207 tubulated retort or flask, provided with a caoutchouc cork and gas delivery-tube, and to these 20 parts of iodine are gradually added, the contents of the flask during this operation being kept cool by immersing the flask in cold water. When all the iodine has been added, and as soon as no further evolution of gas can be noticed, the flask may be gently warmed. The gas thus obtained may either be received in dry bottles filled with FIG. 59. mercury in the mercurial trough, or it may be collected by displacement, as it is more than four times as heavy as air. If we possess a concentrated solution of hydriodic acid, the gas may be obtained in a still more simple manner. Two parts of iodine are dissolved in aqueous hydriodic acid of specific gravity 1*7, and this solution is allowed to fall, drop by drop, by means of a stoppered funnel-tube, into a flask containing amor- phous phosphorus covered with a thin layer of water. The evo- lution of gas occurs at first without any application of heat being 208 THE NON-METALLIC ELEMENTS necessary, but after a time the flask may be slightly warmed. The apparatus, Fig. 59, is used for the purpose of preparing, according to the above method, a saturated aqueous solution of hydriodic acid. When hydriodic acid is prepared by the foregoing methods it frequently contains phosphuretted hydrogen; the forma- tion of this impurity may, however, be avoided if the iodine is never allowed to come in contact with an excess of phosphorus. According to Lothar Meyer, this may be readily carried out as follows : 100 parts of iodine moistened with ten parts of water are placed in a tubulated retort, the neck of which is inclined upwards ; a thin paste of 5 parts of amorphous phosphorus and 10 parts of water is then allowed gradually to drop in through the tubulus of the retort. For this purpose a dropping funnel is employed, which, in place of a stopcock, is furnished with a glass rod, ground at the lower end to fit into the funnel tube ; by raising the latter, small quantities of the paste are allowed to pass into the retort. The first few drops musi: be added cautiously, waiting each time for the reaction to moderate, as otherwise an explosion may occur ; after a short time, however, larger quantities may be added at once, and the mixture may be completed in a quarter of an hour. The iodine carried over mechanically settles for the most part in the neck of the retort, and may be removed completely in most cases by washing with a little water. 1 The collection of the gas or the preparation of the solution may be carried out in the manner already described. Properties. Hydriodic acid exists at the ordinary temperature and pressure as a colourless gas, having a strongly acid reaction and suffocating odour, and fuming strongly in the air. It can be condensed to a colourless liquid 2 by a pressure of four atmospheres at 0, or by exposure, under the ordinary atmos- pheric pressure, to the low temperature of a bath of ether and solid carbonic acid, and if cooled to - 55 it freezes to a colour- less ice-like solid mass. Its specific gravity (air=l) has been found to be 4*3737, or 62'94 (H = l), thus closely corresponding to its theoretic density, 63'45. Hydriodic acid gas is easily decomposed by heat into iodine and hydrogen, as is seen by the violet colour which appears when the gas is passed through a heated glass tube. A hot 1 Ber. 20, 3381. 2 Faraday, Phil. Trans. 1845, 1, 170. HYDKIODIC ACID 209 metallic wire plunged into the gas also causes an immediate decomposition, violet fumes of iodine making their appearance. 109 The aqueous acid is obtained by passing the gas into water, by which it is absorbed quickly and in large quantities, yield- ing, when kept cold by ice, a solution which is twice as heavy as water, having a specific gravity, according to De Luynes,of T99. A simple mode of preparing a dilute aqueous solution of hydri- odic acid consists in passing a current of sulphuretted hydrogen gas through water in which finely-divided iodine is suspended, the reaction which occurs being as follows On standing, the clear liquid may be poured off from the precipitated sulphur and boiled to expel any trace of sulphuretted hydrogen. It is found that the strongest acid which can in this way be prepared has a specific gravity of I 1 56. On distillation, aqueous hydriodic acid behaves like aqueous hydrochloric and hydrobromic acids, both the strong and weak aqueous acid yielding on distillation in an atmosphere of hy- drogen (to prevent oxidation and liberation of iodine) an acid of constant composition, boiling at 127 (under a pressure of 774 mm.), and containing 57*0 per cent, of hydriodic acid. If dry hydrogen be led through aqueous acids of varying strengths, each will attain the same constant composition at the same temperature ; thus from 15 to 19 the constant acid contained GO'S to 6(V7 per cent, of HI. When the hydrogen was passed through the liquid at 100, the percentage of hy- driodic acid in the constant acid was 58'S, 1 and hence it is seen that no definite hydrate of the acid is obtained by boil- ing, as was formerly supposed. Aqueous hydriodic acid also rapidly undergoes oxidation with liberation of iodine when ex- posed to the air, the colourless solution becoming brown owing to the solubility of iodine in the acid. Gaseous hydrogen iodide is also decomposed by dry oxygen when the mixed gases are exposed to bright sunlight, and differs in this respect from hydrogen chloride and bromide which are only decomposed by oxygen in presence of moisture. 2 no The Iodides. The metallic iodides possess great analogy with the corresponding chlorides and bromides ; they are all 1 Roscoe, Journ. Chem. Soc. 1861, 160. 2 Richardson, Journ. Chem. Soc. 1887, i. 805. 15 210 THE NON-METALLIC ELEMENTS solid bodies, less fusible and volatile than the corresponding chlorides and bromides. Silver iodide, Agl, mercurous iodide, Hgl, and mercuric iodide, HgI 2 , are insoluble in water, and lead iodide, PbI 2 , sparingly soluble, whilst the other metallic iodides dissolve readily in water. Most of the iodides are decomposed on heating, either the metal or an oxide being formed and iodine set free. All the iodides, whether soluble or insoluble in water, are decomposed by chlorine and nitrous acid, the iodine being liber- ated. Some of the insoluble iodides possess a brilliant colour. Thus, on adding a solution of corrosive sublimate (mercuric chloride) to a soluble iodide, a salmon-coloured precipitate is thrown down, which rapidly changes to a brilliant scarlet one of mercuric iodide, HgI 2 , soluble in excess of either reagent ; a soluble lead salt, such as the nitrate or acetate, produces a bright yellow precipitate of lead iodide, PbI 2 ; silver nitrate gives a light yellow precipitate of silver iodide, Agl, insoluble in nitric acid and in ammonia. If a mixture of ferrous sulphate, FeSO 4 , and copper sulphate, CuS0 4 , be added to that of a soluble iodide, a greenish white precipitate of cuprous iodide, Cul, is formed. This reaction depends upon the fact that ferrous sulphate is oxidized to ferric sulphate, Fe 2 (SO 4 ) 3 , whilst cuprous iodide is precipitated, thus : 2CuSO 4 + 2FeSO 4 + 2KI = 2CuI + K 2 SO 4 + Fe 2 (S0 4 ) 3 This reaction serves as a means of roughly separating iodine from a mixture containing chlorides and bromides. The metallic iodides can be prepared by similar processes to those which yield the chlorides arid bromides. (1) By the direct action of iodine on the metal, as in the cases of the iodides of iron and mercury. (2) By the action of iodine on certain of the metallic oxides, hydroxides, or carbonates, as those of potassium, sodium, barium,, calcium, and silver. (3) By the action of hydriodic acid on certain metals, such as zinc, hydrogen being liberated. (4) By the action of hydriodic acid on the metallic oxides, hydroxides, or carbonates. (5) By adding a soluble iodide, such as potassium iodide, to a solution of the salt of the metal, when the metallic iodide is thrown down in the form of a precipitate ; this method, however, can only be used when the iodide required is insoluble. ESTIMATION OF IODINE 211 in Detection and Estimation of Iodine. For the detection of iodine, the starch reaction, the violet-coloured vapours, and the above-mentioned coloured precipitates are sufficient. To estimate iodine in the free state, a standard solution of sulphurous acid is employed, and the point ascertained at which sufficient of this solution has been added to reduce all the iodine to hydriodic acid, thus : I 2 + 2H 2 + S0 2 = 2HI + H 2 S0 4 . The solution of sulphurous acid may be replaced by one of sodium thiosulphate, a substance which reacts with free iodine in the following manner : 2Na 2 S 2 3 + I 2 = 2NaI + Na 2 S 4 6 . For the quantitative determination of iodine in a soluble iodide and for the exact separation from chlorine or bromine, use may be made of the fact that the palladium nitrate, Pd (NO 3 ) 2 , pro- duces with solutions of an iodide, an insoluble precipitate of PdI 2 , which on ignition yields metallic palladium. Iodine when in the form of an alkaline iodide can be weighed also as iodide of silver, when neither chlorine nor bromine is present; 100 parts of silver iodide contain 54*128 parts of iodine. In the case of the insoluble iodides, it is best either to transform them into soluble iodide of sodium by fusing them with carbonate of soda, or to digest them with zinc and dilute sulphuric acid, when hydriodic acid is liberated, thus : 2AgI + Zn + H 2 S0 4 = 2HI + 2Ag + ZnSO 4 . If it is required to determine chlorine, bromine, and iodine when mixed in solution together the following method may be employed : Field has shown 1 that chloride of silver is completely de- composed by digestion with bromide of potassium, the chlorine and bromine changing places ; and that both bromide and chloride of silver are decomposed in like manner by iodide of potassium. Hence, if a solution containing chlorine, bromine, and iodine, be divided into three equal parts, each portion precipi- tated by nitrate of silver, the first precipitate dried and weighed 1 Ohem. Soc. Quart. Journ. 1858, 234. 212 THE NON-METALLIC ELEMENTS the second digested with bromide of potassium, then dried and weighed, and the third digested with iodide of potassium, then dried and weighed, the relative quantities of the three elements may be determined from the following equations : x+y+z=w 186-49 233-04 233-04 where w, w' w", are the weights of the three precipitates, and x, y, and z, the unknown quantities of chloride, bromide, and iodide of silver respectively. The mixture of the three salts of silver may also be treated with a solution of potassium bichromate in sulphuric acid, which converts the chloride and bromide into the soluble sulphate, whilst the iodide is converted into the insoluble iodate. After dilution and filtration the iodate may be reduced and the silver in it determined, whilst the silver originally present as chloride and bromide may also be determined in the filtrate. 1 In this case the equations became : x + y 4 z ^ w Ag Ag Ag where w is the weight of the mixed salts, w p the weight of silver from silver iodate, and w 9 the weight of silver from the chloride and bromide. IODINE COMPOUNDS WITH OTHER HALOGENS 112 Iodine burns with a pale flame in fluorine, yielding a colourless fuming oil, which rapidly attacks glass, and is probably an iodine fluoride. It is most likely identical with the compound obtained by Gore by the action of iodine on silver fluoride. (Moison.) 1 Macnair, Proc. Chem. Soc. 1893, 181. IODINE TRICHLORIDE 213 Two compounds of iodine and chlorine are known : (1) Iodine monochloride, IC1. (2) Iodine trichloride, IC1 3 . They are both obtained by the direct union of chlorine and iodine, the higher chloride being formed when the former element is in excess. Iodine Monochloride, IC1, is prepared (1) by passing dry chlorine gas over dry iodine until the latter is completely liquefied ; (2) according to Berzelius, by distilling 1 part of iodine with 4 parts of potassium chlorate the chlorine evolved by the reaction combining with a portion of the iodine ; x (8) by boiling iodine with strong aqua regia ; after dilution with water the liquid is shaken up with ether in which the chloride of iodine dissolves and remains behind when the ether is evaporated. 2 The product thus obtained is a reddish brown oil, which, on standing, solidifies, forming long well-defined crystals melting at 24 0> 7 ; it boils at 101'3 and has a sp. gr. at that temperature of 2'88196 (Thorpe). It smells like a mixture of chlorine and iodine, bleaches indigo solution, but does not colour starch-paste blue. The following analyses show the composition of this substance : Found Liquid. Solid. Calculated. (Burisen.) (Schiitzenberger.) Iodine . 12591 78*85 77'85 7S"52 Chlorine 3519 2M5 22'15 21'48 161-10 10000 100-00 100-00 A second modification, known as ft iodine monochloride, is obtained by heating the crude monochloride till it is free from trichloride, and then cooling to - 10. It melts at 13-9 . 3 Iodine Trichloride, IC1 3 , is obtained (1) by acting on iodine, gently heated, with a large excess of chlorine ; (2) by treating iodic acid, HIO 3 , with hydrochloric acid ; (3) by heating iodine pentoxide, I 2 5 with pentachloride of phosphorus, PC1 5 . This compound forms long lemon-coloured crystals, and very readily undergoes dissociation. When heated in the air to 25 it decomposes, giving off chlorine gas, forming the monochloride ; but when heated in an atmosphere of chlorine it only de- 1 Thorpe and Perry, Journ. Chem. Soc. 1892, i. 925. 2 Bunsen, Annalen, 84, 1. 3 Tanatar, Journ. Russ. Chem, Soc., 25, 97. 214 THE NON-METALLIC ELEMENTS composes at a much higher temperature, which rises as the pressure of the chlorine is increased. Thus, under a pressure of one atmosphere it decomposes at 67 into the monochloride and free chlorine, and these again unite on cooling to form a yellow sublimate of the trichloride. 1 The composition of the compound is seen from the following analysis : Calculated. Found. Iodine . . . 125*91 54'39 54'34 Chlorine . . 105*57 45*61 45'66 ^31-48 100-00 100-00 Both the chloride and trichloride dissolve in water, ether, and alcohol apparently without decomposition. When either of them is acted upon by a small quantity of an alkali, an iodate and chloride are formed and iodine is liberated. 2 6KHO+5C1I=5KC1+KI0 3 +2I 2 +3H 2 0. They are both very hygroscopic and give off irritating vapours. The formation of both of these compounds can be demon- strated by inverting a small cylinder containing chlorine and bringing its mouth in contact with that of another of the same size filled with hydriodic acid gas. Iodine is first liberated, which combines with the excess of chlorine, the yellow tri- chloride being deposited on the sides of the upper jar, in which chlorine is in large excess, whilst the brown monochloride, mixed with iodine, is formed in the lower jar. 113 Iodine unites with bromine to form a solid, volatile, crystalline compound which is probably the monobromide, and also a dark liquid, possibly the tribromide. These bodies possess properties similar to those of the chlorides of iodine. OXYGEN. = 15-88. 114 Of the elements which occur on our planet, oxygen is the most widely diffused, and is found in the largest quantity. The old crystalline rocks, which constitute the chief mass of the earth's crust, consist of silicates, or compounds of silicon and various metals with oxygen. These rocks contain from 44 to 48 per cent, of oxygen. Water likewise is a compound 1 Brenken, Ber. 8, 487. 2 Philip, Ber. 3, 4. PREPARATION OF OXYGEN 215 of oxygen and hydrogen, containing 88*81 per cent, of the former element. .Oxygen also exists in the free state in the atmosphere, which contains about 21 per cent, of its volume of this gas. Although the absolute amount of free oxygen con- tained in the air is very great, yet the proportion which the free oxygen bears to that in a state of combination is but very small. It has already been mentioned, in the Historical Intro- duction, that the air was believed to be a simple or elementary substance until the investigations of Priestley, Rutherford, and Scheele x showed distinctly that it is a mixture of two different gases, only one of which is capable of supporting combustion and respiration. This constituent of the atmosphere is oxygen, discovered on the 1st of August, 1774, by Priestley, who, by heating "red precipitate" (mercuric oxide) by means of the sun's rays, decomposed it into oxygen and metallic mercury. The discovery of oxygen enabled Lavoisier to put forward the true theory of combustion, and to the body capable of support- ing this combustion was given the name "oxygene" (ofvs sour, and yevvdco I produce), from the fact that the products of combustion are frequently of an acid nature. Preparation. (1) The simplest method of preparing oxygen is to heat mercuric oxide, HgO, in a small retort of hard glass. The oxide decomposes at a red-heat into metallic mercury and oxygen ; 100 parts by weight yield 7 '4 parts by weight of oxygen. 2 HgO=2Hg + 2 . The apparatus in which this decomposition can be shown is seen in Fig. 60. Owing to the comparatively high price of oxide of mercury, this process is only used as a means of illustrating the decomposition. (2) The best and most usual mode of preparing oxygen con- sists in heating potassium chlorate, commonly called chlorate of potash, KC1O 3 . This salt loses the whole (39' 14 per cent, of its weight) of its oxygen, leaving potassium chloride, thus : 2KC10 3 = 2KC1 + 3O 2 . The preparation and collection of the gas, according to this method, may be carried on in the apparatus shown in Fig. 61. 1 It appears that Scheele had prepared oxygen prior to the date of Priestley's discovery, but that these results were not published until after Priestley's experiments had been made known. See Carl Wilhelm Scheele : Nachgelassene Brief e und Aufzeichnungen, edited by A. E. Nordenskiold (Stockholm, 1892). 216 THE NON-METALLIC ELEMENTS The temperature has to be raised much above the melting point of the salt (372), before the evolution of the gas begins ; and after a certain time has elapsed, the fused mass becomes thick, FIG. 60. owing to the formation of potassium perchlorate, KC1O 4 , a part of the evolved oxygen having united with the chlorate, whilst FIG. 61. potassium chloride, KC1, and oxygen are at the same time formed, 1 thus : 10KC10 3 = 6KC10 4 + 4KC1 + 30 2 . 1 Frankland and Dingwall, Journ. Chem. Soc. 1887, 274 ; Teed, Journ Chem Soc. 1887, 283. PEEPARATION OF OXYGEN 217 When more strongly heated, the perchlorate also decomposes into potassium chloride and oxygen. The gas can be obtained at a lower temperature by employ- ing a mixture of potassium and sodium chlorates (Shenstone). (3) In order to obtain the evolution of oxygen at a lower temperature, a small quantity of manganese dioxide is generally mixed with the powdered chlorate ; the oxygen then comes off at about 350, before the salt fuses, and thus the preparation of the gas is greatly facilitated. The manganese dioxide is found, mixed with potassium chloride, in the residue unaltered in composition. 115 In order to prepare oxygen on a larger scale this mixture FIG, 62, , of potassium chlorate and manganese dioxide is heated in a thick copper vessel a, Fig. 62, provided with a wide tube con- nected with the wash-bottle b, containing caustic soda, for the purpose of absorbing chlorine gas, which is always evolved along with the oxygen. It is difficult to give a satisfactory explanation of this peculiar action of the manganese dioxide. It may possibly be due to the fact that certain oxides, such as this one, are capable of undergoing a higher degree of oxidation, but that these higher oxides part very readily with a portion of their oxygen, forming again the lower oxide. In this way such oxides would perform the part of carriers of oxygen, first taking it up and then setting it free. Indeed it has been rendered 218 THE NON-METALLIC ELEMENTS probable that when manganese dioxide is used, potassium permanganate is first formed according to the equation : 2KC10 3 4- 2MnO 2 = 2KMnO 4 + C1 2 + This is borne out by the facts that traces of chlorine are in- variably evolved and that when only a small quantity of manganese dioxide is added, the pink colour of the permanga- nate can be distinctly seen. The latter is then broken up by the action of heat and the chlorine evolved by the further decomposition of the chlorate, into potassium chloride, manga- nese dioxide and oxygen, after which the reaction is repeated. 1 According to Brunck the gas prepared in this manner invariably contains ozone. 2 The decomposition of the chlorate is also aided by many other oxides, the action of which is probably to be explained in a similar manner to that of manganese dioxide. Finely divided platinum (platinum black), kaolin, and other powdered sub- stances also produce the same effect, but the above explanation does not hold good in these cases. 3 It not unfrequently happens that the commercial black oxide of manganese may be accidentally mixed or adulterated with carbon (pounded coal), and this impure material, when mixed with chlorate of potash and heated, ignites, giving rise to even fatal explosions. Hence care should be taken to try any new or doubtful sample on a small scale beforehand by heating it with chlorate of potash in a test-tube. (4) Many other salts behave like potassium chlorate in yield- ing oxygen on heating: among these are the hypochlorites, chlorites, perchlorates, bromates and perbromates, as well as the iodates, periodates, nitrates, nitrites and permanganates; but these compounds are not usually employed for this purpose. (5) Several oxides, such as manganese dioxide, MnO 2 , lead dioxide, Pb0 2 , barium dioxide, Ba0 2 , chromium trioxide, CrO 3 , lose a portion of their oxygen when strongly heated, and all these may, therefore, be used for preparing oxygen. In order to obtain oxygen by heating the first-named oxide, the sub- stance is placed in a strong iron bottle which can be heated in a furnace to bright redness (Fig. 63). The pure manganese 1 McLeod, Journ. Chem. Soc. 1889, 184. 2 Ber. 26, 1790. 3 Veley, Phil. Trans. 1888, i., 271 ; Fowler and Grant, Journ. Chem. Soc. 1890, 272. PREPARATION OF OXYGEN 219 dioxide loses one-third (12'4 per cent.) of its oxygen, being con- verted into the brown oxide, Mn 3 O 4 , thus : 3MnO 2 = Mn 8 O 4 + O 2 . (6) By heating manganese dioxide in a glass flask with sul- phuric acid, one-half of its oxygen is given off, and manganous sulphate, MnS0 4 , is formed. 2MnO 2 + 2H 2 SO 4 = 2MnSO 4 + 2H 2 O + O 2 . (7) Chromium trioxide can also be employed for the prepara- tion of oxygen, but it is not necessary to obtain this substance FIG. 63. in the pure state, for if bichromate of potash, K 2 Cr 2 O 7 , be heated with sulphuric acid, chromium trioxide is formed, thus : K 2 Cr 2 O 7 + 2H 2 SO 4 = 2KHSO 4 + 2CrO 3 + H 2 O. The chromium trioxide is then further decomposed by the action of sulphuric acid with the formation of chromium sul- phate, a decomposition which is rendered visible by the change of colour from the original red to a deep green, thus : 4Cr0 3 + 6H 2 SO 4 = 2Cr 2 (SO 4 ) 3 + 6H 2 4 3O 2 . (8) Oxygen can be obtained by the decomposition of bleach- ing powder (Mitscherlich, 1843 ; Fleitmann, 1865). For this preparation a clear concentrated solution of bleaching powder which contains calcium hypochlorite, CaCl 2 O 2 is placed in a flask, and a few drops of cobalt chloride solution added. An 220 THE NON-METALLIC ELEMENTS oxide of cobalt, which probably has the formula 1 CoO 2 , is pre- cipitated, and on heating the mixture to about 80 a rapid effervescence of oxygen occurs. The cobalt oxide, which is formed, is left unchanged after the operation, and may be employed again ; it probably acts, like the manganese dioxide, by the formation of a higher oxide, which is again quickly reduced, the oxygen being liberated as a gas. Instead of a clear solution, a thick paste of bleaching powder may be used, with the addition of a little cobalt salt and a small quantity of paraffin oil, to prevent the frothing which usually occurs. The best temperature for the evolution of gas is from 7080. CaCl 2 2 = CaCl 2 + O 2 . The same decomposition and the replacement of chlorine for oxygen may be shown in a striking manner by passing chlorine gas, generated in a flask from manganese dioxide and hydro- chloric acid, into a second flask, which contains boiling milk of lime, to which a little cobalt nitrate solution has been added. Oxygen gas is then liberated in the second flask, and may be collected as usual. The following equation explains the replacement, and we see that two volumes of chlorine yield their equivalent, or one volume of oxygen : 2C1 2 + 2Ca (OH) 2 = 2CaCl 2 + 2H 2 + O 2 . (9) Oxygen can also be prepared by the decomposition of sulphuric acid. For this purpose a thin stream of sulphuric acid flows into a platinum retort heated to redness ; the acid splits up into sulphur dioxide, water, and oxygen, yielding 15*68 per cent, of its weight of the gas, or in practice 55 grams, of acid yield 6 litres of gas, thus : 2H 2 S0 4 = 2S0 2 + 2H 2 O + O 2 . The resulting sulphur dioxide and water can be absorbed, whilst the oxygen can be collected in a gas-holder. An apparatus for illustrating this decomposition is described under Sulphuric Acid. In order to prepare oxygen cheaply on the large scale several 1 Vortmann, quoted by McLeod, British Ass. 1892, 669. PROPERTIES OF OXYGEN 221 other processes have been suggested. Amongst them the following is now widely used. (10) When baryta, BaO, is gently heated to dull redness in the air, it takes up an additional atom of oxygen, forming the dioxide, BaO 2 , but at a bright-red heat this parts with the additional atom of oxygen with the reproduction of baryta. By thus alternately varying the temperature, first leading air over the baryta contained in a porcelain tube, and then placing the tube in connection with a gas-holder and raising the temperature, and again repeating the process, a regular pro- duction of gas can be obtained from a small quantity of baryta. 1 This simple method is now carried out on a large scale according to a process patented by the Brin Oxygen Company 2 Oxide of barium, prepared from the nitrate, is introduced in pieces about the size of walnuts into steel or cast-iron retorts. These retorts, placed in a vertical position in a gas furnace, are heated to about 700, and air, carefully purified from moisture and carbonic acid then pumped through them under a pressure of about 15 Ibs. on the square inch. The baryta absorbs the oxygen, becoming converted into peroxide, and as soon as this peroxidation has been carried as far as is economical, the pump is reversed and the pressure in the retorts thus reduced. When the reduction of pressure has reached about 26 28 inches of mercury below the normal, the peroxide begins to give up its oxygen, which passes through the pump and is delivered to a gas-holder. A complete operation lasts about ten minutes, and nearly 140 operations can be conducted per diem. The oxygen is then compressed at 120 atmospheres in wrought iron cylinders. 3 (11) Potassium manganate, K 2 Mn0 4 , loses oxygen when heated in a current of steam, forming caustic potash and lower oxides of manganese, which when again heated absorb oxygen, the manganate being reproduced, so that the same portion may be used over and over again (Tessie du Motay). A modified form of this process in which sodium manganate is used, has been applied by Fontana 4 for the industrial preparation of oxygen. It has also been proposed to utilise calcium plumbate 1 Boussingault, Ann. Chim. Phys. [3] 35. 2 Pat. 157, Oct. 5, 1885. 3 Journ. Soc. Chem. 2nd. 1890, 246. 4 Journ. Soc. Chem. Ind. 1892, 312. THE NON-METALLIC ELEMENTS Ca 2 Pb0 4 , for this purpose. 1 The material is heated at a low temperature in the presence of carbonic acid, which causes its decomposition into calcium carbonate, CaCO 3 , and lead peroxide, Pb0 2 . When the mixture of these is more strongly heated oxygen is first given off and a lower oxide of lead formed, after which the calcium carbonate loses its carbon dioxide. The residual mass, consisting of a mixture of lime and the lower oxides of lead is reconverted into calcium plumbate by heating in a current of air. 116 Properties Oxygen is a colourless, invisible, tasteless, inodorous gas which is slightly heavier than atmospheric air. According to Regnault's experiments 2 the density of the gas is 1-10562 (air=l), one litre at and 760 mm. weighing T43011 grams, whilst according to Leduc 3 the density is 1*10503, and to Rayleigh, 4 T10535, one litre weighing 1*42961 grams. The density of the gas compared with hydrogen has been found by Rayleigh to be 15 '882 (see p. 260) ; its molecular weight is therefore about 31 '76 and its formula O 2 . Oxygen was first liquefied by Cailletet and Pictet in December, 1877. It forms a light blue liquid, boiling at - 181'4, at which temperature the density of the liquid is 1124. The critical temperature of the gas is - 118, the corresponding pressu^^ being 50 atmospheres. As long ago as 1847 Faraday discovered that oxygen is less diamagnetic than air, 5 and it was subsequently shown to be actually paramagnetic. Liquid oxygen is strongly magnetic, and when placed in a cup-shaped piece of rock salt between the poles of a powerful electro-magnet suddenly leaps up to the poles and remains there permanently attached until it evaporates. 6 Liquid oxygen presents the same absorption spectrum as the gas, characteristic bands being present in the orange, yellow, green, and blue, but the bands are much more intense and well marked than those of the gas. It also possesses a measurable thermal absorption and presents a very high resistance to an electric current. When cooled to 210 by its own rapid evaporation, it is no longer capable of supporting combustion or of combining with substances like phosphorus and sodium. 1 Kassner, Chem. 2nd. 1890, 104, 120 ; Le Chatelier, Compt. Rend. H7, 109. 2 Regnault, Mem. Acad. Science, 21, 144 ; Crafts, Compt. Rend. 106, 1662. 3 Leduc, Compt. Rend. 113, 186. 4 Rayleigh, Proc. Roy. Soc. 53, 134. 5 Faraday, Phil Mag. (3), 31, 401. 6 Dewar, Proc. Roy. Soc. 50, 247. PROPERTIES OF OXYGEN 223 Pictet 1 has indeed found that chemical action in general entirely ceases at temperatures approaching 150. Thus sulphuric acid and caustic potash when compressed together at this temperature do not react, although the normal action takes place at 90. In the same manner sodium preserves its metallic lustre in liquid alcohol of 84 per cent, at 78 and does not commence to act upon it until the temperature reaches 48; and alcoholic litmus solution remains blue in contact with solid sulphuric acid at all temperatures below 105, at which reddening suddenly takes place. Oxygen dissolves appreciably in water ; at one volume of water absorbs O04890 volume of oxygen, measured under the normal temperature and pressure. When the tempera- ture rises, the quantity of oxygen absorbed becomes less, according to a complicated law, which is expressed by the empirical formula. 2 G = 0-04890 - 0'0013413t + 0'0000283t 2 - 0'00000029534t 3 . Certain metals also absorb oxygen when in the molten state, and give it off again on solidifying ; thus, melted silver absorbs about ten times its bulk of oxygen, and this is nearly all emitted when the metal cools, giving rise to the peculiar phenomenon of the " spitting " of silver. As oxygen is the constituent of the air which supports com- bustion, it naturally follows that bodies burn in oxygen with much greater brilliancy than they do in common air. A glowing chip of wood, or the red-hot wick of a taper, ignites with a slight detonation when plunged into oxygen gas, and even metals such as iron, which oxidize only slowly in the air, burn brilliantly in oxygen. The following experiments serve to illustrate this property of oxygen : A bundle of thin iron wire, with the ends tipped with lighted sulphur or a burning piece of twine, burns when plunged into a jar of oxygen, forming the black oxide, Fe 3 O 4> which falls down in glowing drops. A piece of watch spring also burns easily with splendid scintillations if held in a flame obtained by blowing a jet of oxygen into the flame of a spirit lamp. An even more striking mode of showing the combustion of iron is to place a heap of cast iron nails on a brick and burn them by means of a blow-pipe fed with oxygen and coal- * Compt. Rend. 115, 814. 2 L. W. Winkler, Ser. 22, 1764 ; 24, 3607. 224 THE NON-METALLIC ELEMENTS gas contained in separate gas-holders. Substances like sulphur and phosphorus, which take fire readily in the air, burn with much greater brilliancy in oxygen ; combination takes place much more rapidly, and, therefore, the temperature reached is much higher in oxygen than in the air, in which, moreover, the inert nitrogen takes up a share of the heat. The best method of exhibiting combustion in oxygen is to place the substance to be burnt in a metal cup, riveted on to an upright stem, carrying a round saucer containing water. As soon as the body has been ignited, a large glass globe filled with oxygen gas is placed over it, so that the cup occupies a FIG. 64. central position in the lower half of the globe, and then the combustion can proceed with great rapidity without fear of the globe being cracked by the heat evolved (Fig. 64). In this way sulphur burns with a bright violet flame, with form- ation of colourless sulphur dioxide gas, SO., : whilst phosphorus, thus burnt, emits a brilliant white light, which vies with sunlight in intensity. In this case the white solid phosphorus pentoxide, P 2 O 5 , is the product of the combustion. 117 An act of chemical union accompanied by the evolution of light and heat, is termed a combustion, and hence oxygen is commonly termed a supporter of combustion, whilst those bodies which thus unite with oxygen are called combustible substances. COMBUSTIONS IN OXYGEN AND IN HYDROGEN 225 A little 'consideration, however, shows that these terms are only relative, and an experiment makes this plain. One of the stop- pered bell-jars (Figs. 65 and 66) is filled with oxygen gas, the other with hydrogen ; two gas-holders, one containing hydrogen, the other oxygen, are provided with flexible gas delivery-tubes at the end of which is fixed a perforated caoutchouc stopper carrying a metal tube with a nozzle. The hydrogen gas is .allowed to escape through the nozzle, then ignited, and the flame of hydrogen plunged into the bell-jar filled with oxygen, the caoutchouc stopper fitting tightly into the tubulus. A flame FIG. 65. of hydrogen burning in oxygen is then seen, the hydrogen being the burning body and the oxygen the supporter of combustion. A stream of oxygen gas is next allowed to issue from the nozzle of the second gas-holder, the stopper of the bell-jar containing hydrogen, is then removed, and the jet of oxygen is plunged into the bell-jar, whilst the flame of a candle is brought at the same instant to the tubulus. On pressing the caoutchouc stopper into its place, a flame, not to be distinguished from that burning in the other bell-jar, is seen, in which oxygen is the burning body, and hydrogen is the supporter of com- bustion. 16 THE NON-METALLIC ELEMENTS Most bodies do not combine with oxygen rapidly enough at the ordinary atmospheric temperatures to give rise to the phenomena of combustion, but require to be heated before this begins. Oxidation is, however, often slowly going on, as in the case of the rusting of metals, the decay of wood and organic bodies. Thus, we come to distinguish between quick and slow oxidations in which the intensity of the heat and light evolved is very different. This slow oxidation frequently occurs in presence of certain finely-divided metallic particles, probably owing to the con- FIG. densation of the gases on the surface or in the pores of the metal. Thus a small quantity of spongy platinum (obtained by heating the double chloride of platinum and ammonium) when held over a jet of coal gas or hydrogen, first becomes red hot, owing to the combustion of the gas occurring on its surface, and afterwards the temperature of the metal ma|[ rise so high that the jet of gas is ignited. In some cases the products of the slow oxidation of the sub- stance are different from those formed by its rapid oxidation or SLOW OXIDATION 227 combustion. When a coil of fine platinum wire rs first heated in a flame and then hung whilst warm over the surface of some alcohol contained in a small beaker-glass, the coil soon begins to glow, and remains red-hot until all the alcohol is consumed, but no flame is seen. Alcohol has the formula C 2 H 6 O, and when it burns with a flame its constituents unite with oxygen to form water, H 2 O, and carbon dioxide, CO 2 . When oxidised at the lower temperature, a peculiar smelling body termed aldehyde is formed, having the formula C 2 H 4 ; hence the oxidation of the alcohol is partial or incomplete. Only two of the hydrogen atoms of alcohol are then withdrawn, water being formed, whilst the volatile aldehyde escapes, giving rise to a peculiar choking smell. The effect of mechanical division on the combustibility of substances, especially of metals, is well known, and advantage is taken of this in the preparation of the various pyrophori. If tartrate of lead be gently heated in a glass tube, the lead is left in a state of very fine mechanical division, and mixed with carbon. After heating, the tube is hermetically sealed, and on cooling it may be opened and the contents shaken out into the air, when the finely divided particles will at once take fire. In the same way, if the oxides of iron, cobalt, or nickel be reduced by hydrogen at a moderate temperature, the metal is formed in a pulverulent state, in which it takes fire spontaneously on exposure to the air. The explanation of this is, that by fine division, the ratio of the surface exposed, to the mass to be heated becomes so great that the heat generated by the oxidation of the surface is sufficient to bring the mass to incandescence. The spontaneous ignition of a mass of inflammable materials like cotton or woollen rags, when mixed with a substance, such as oil, capable of rapidly absorbing oxygen, and thereby generating heat, is one of the most common sources of fire, both in manufactories and on board ship. Similar cases of spontaneous combustion occur in hay-ricks, in which the hay has been put up damp, for moisture greatly assists the process of slow oxidation. Other examples of the same thing are seen in the fires which break out in ships carrying coal, or in heaps of coal or shale ; these seem to be due to the oxidation of the bituminous constituents of the coal, into carbon dioxide and water by the oxygen of the air, which is absorbed by the coal, especially when much broken up, and thus evolve heat 228 THE NON-METALLIC ELEMENTS enough to set the mass on fire. All the supposed cases of spontaneous combustion occurring in the human body have been clearly proved to be mistakes or deceptions, as may be seen by reading Chapter xxv. of Liebig's admirable Letters on Chemistry, in which this matter is fully discussed. 118 Temperature of Ignition. In order that a body may take fire in air or in oxygen, a certain temperature must be reached : this point is termed the temperature of ignition. The tempera- ture at which inflammation occurs varies widely with different substances; thus while the vapour of carbon bisulphide is ignited by bringing in contact with it a glass rod heated only to 149, a jet of coal gas cannot be lighted with a piece of iron at a dull red-heat : and, again, certain substances, such as the liquid phosphuretted hydrogen, or zinc ethyl, only require to be exposed to the air at the ordinary temperature in order to ignite, whilst nitrogen can only be made to unite with oxygen by heating the mixture to the temperature of the electric spark. The temperature at which slow oxidation commences is of course lower than that of ignition ; thus phosphorus begins to enter into slow combustion in the air (exhibiting phospho- rescence) below 10 C. ; but we must heat it up to 60 C., before it begins to burn brightly, or to enter into quick com- bustion. The Davy Lamp. A most striking example of the fact that a certain temperature must be reached before a mixture of in- flammable gas with air can take fire, is seen in the safety lamp for coal mines, invented by Sir Humphry Davy. 1 The principle upon which this depends is well illustrated by holding a piece of wire gauze, containing about 700 meshes to the square inch, over a jet of gas (Fig. 67). If the gas is lit, it is possible to remove the gauze several inches above the jet, and yet the in- flammable gas below does not take fire/ the flame burning only above the gauze. The metallic wires in this case conduct away the heat so quickly that the temperature of the gas at the lower side of the gauze cannot rise to the point of ignition. In a similar way we may cool down a flame so much that it goes out, by placing over it a small coil of cold copper wire whereas it is impossible to extinguish the flame if the coil of wire be previously heated. The " Davy lamp " consists of an oil lamp (Figs. 68 and 69) the top of which is inclosed in a covering of wire gauze, so that the products of combustion of 1 Phil. Trans. 1817, pp. 4577. TEMPERATURE OF IGNITION 229 the oil can escape, while no flame can pass to the outside of the gauze. Hence no ignition is possible, even if the lamp is placed in the most inflammable mixture of fire-damp and air, although the combustible gases may take fire and burn inside the gauze. It is, however, necessary, to be careful that the flame thus FIG. 67. kindled inside the gauze does not heat it up to the point of ignition of the inflammable gas, and especially to avoid placing the lamp in draughts, which might blow the flame against a point of the gauze and thus heat it above the point of safety. FIG. 68* FIG. 69. Indeed, it was pointed out by Davy himself that the lamp is no longer safe if exposed to a draught of air, and several serious accidents have occurred from the neglect of these precautions. It has also been shown that the flame burning inside a wire gauze may be mechanically blown through the gauze by a cur- 230 THE NON-METALLIC ELEMENTS rent or blast of air passing at the rate of eight feet per second, 1 and this has doubtless given rise to many serious accidents. Several modifications of the original Davy lamp have been intro- duced to lessen this danger, but the firing of shots in fiery pits is, in any case, much to be condemned. It is almost unneces- sary to say that the lamp ought not be opened whilst in use in the pit. 119 Heat of Combination. In every chemical reaction the formation of new compounds is accompanied by a change in the energy of the system. Thus when two grams of hydrogen at unite >. with 15*88 grams of oxygen, also at 0, to form 17*88 grams of water at the same temperature, we not only have the material conversion of the 33*3 litres of mixed gases into 17'88 cc. of liquid water, but we find that sufficient heat is evolved to raise the temperature of 67,940 grams of water (at C) one degree. The amount of heat necessary to raise the tempera- ture of one gram of water (at C.) through one degree is called a calorie, and the heat evolved in chemical reactions is measured in terms of this unit. In order to measure the heat-change which accompanies the chemical transformation of a mixture of hydrogen and oxygen into water, oxygen is burnt in hydrogen contained in a platinum globe immersed in water contained in a calorimeter. 2 The latter consists of a gilded brass cylindrical vessel, surrounded by two concentric brass cylinders to prevent loss or gain of heat by radiation. The air in the platinum vessel is displaced by a current of hydrogen, and dry oxygen then introduced by means of a tube and ignited by an electric spark, so that the jet of oxygen continues to burn in the atmosphere of hydrogen which is supplied through another tube. The excess of hydrogen passes out of the globe by a third tube through a weighed calcium chloride tube, which retains the water vapour. The water of the calorimeter is stirred throughout the experiment. The temperature of the water is observed just before the commencement of the combustion and a series of observations is taken at the close of the experiment, the final temperature of the water being calculated from these after allowing for loss of heat by cooling. The amount of water formed is found by displacing the hydrogen by air and weighing the platinum globe after the 1 Galloway, Proc. Roy. Soc. 22, 441. 2 Thomsen, Thermochemischc Untersuchungen 2. 45. HEAT OF COMBINATION 231 experiment, the moisture carried off by the escaping gases being retained by calcium chloride and weighed. An actual experi- ment gave the following results : Initial temperature 16*075. Final temperature (corr.) 19'357. Rise of temperature 3'282. Water value of calorimeter 2,460 grams. Total weight of water formed 2'129 grams. The water value of the calorimeter represents the amount of water contained in it together with the amount which would require as much heat to raise its temperature one degree as the metal work of the calorimeter. The heat developed, therefore, amounts to 2,460 x 3'282 = 8,074 cal. To this must, however, be added the heat required to raise the water produced by the combination to this final temperature, and the latent heat which was absorbed by the water passing off with the hydrogen as vapour, amounting in all to 15 cal. The total heat for 2'129 grams of water is therefore 8,089 units, and hence the heat of formation of water is 8,089 x 17-88 2 . 129 - - 67,934 cal. The mean of a number of experiments gives 67,940 cal. for the production of 17 '88 grams of water. This may be ex- pressed in an equation in the following manner, the chemical formulae being understood to represent simply the quantities of the reacting substances and not the molecular weights : 2H + = H 2 + 67,940. The product of this reaction possesses less energy than its constituents, and the mixed gases are, therefore, said to possess potential energy, which is converted on their combination into the kinetic energy or energy of motion of the molecules of the heated products. In order to reconvert the liquid water into the same weight of the mixed gases this exact amount of energy must be restored to it, and generally, the heat produced (or absorbed) in the formation of a compound is exactly equal to that absorbed (or evolved) by the decomposition of the substance into its original constituents. Most substances are formed with evolution of heat, but in some cases heat is absorbed in the formation of 232 THE NON-METALLIC ELEMENTS the compound so that it possesses more energy than its constituents. This is the case, for example, with carbon bisul- phide, the oxides of chlorine and nitrogen, ozone and many other bodies. Such substances give out heat on decomposition, and consequently can often be made to undergo explosive decomposition. In these cases the heat evolved by the resolu- tion of a small portion of the substance into its constituents is sufficient to bring surrounding parts to the decomposition point, and thus the whole mass rapidly breaks up. Chlorine monoxide, for example, gives out 17,670 cal. on decomposition: C1 2 O = 2 Cl + O + 17,670 cal. and is an exceedingly unstable compound, exploding violently when heated or even shaken. 1 20 It is frequently impossible to ascertain the heat of form- ation of a compound directly, but an indirect determination is rendered possible by the fact that the heat evolved in a chemical reaction depends only on the initial and final states of the system and is independent of the intermediate stages. 1 Thus if we wish to find the heat evolved in the production of a dilute solution of ammonium chloride from the two gases, ammonia and hydrogen chloride, we may either (1), allow these two gases to combine directly and dissolve the product in water ; or (2), dissolve the two gases separately in water and mix the solutions. In both cases we start with ammonia, hydrochloric acid gas, and water, and end with a dilute solution of ammonium chloride, and the amount of the heat-change is found to be the same in both cases. The determination of the heat of formation of hydriodic acid gas from its elements may serve as an instance of the applica- tion of this indirect method. When hydriodic acid, dissolved in water, is decomposed by chlorine we have the reaction : HI Aq + Cl = HC1 Aq + I + 26,000 cal. In this reaction the hydriodic acid has been decomposed and the hydrogen and chlorine have combined to form hydrochloric acid which has dissolved in the water. The second part of this change gives rise to 39,000 cal. H + Cl + Aq = HC1 Aq + 39,000 cal. 1 Hess. Pogg. Ann. 50, 385. HEAT OF COMBINATION 233 From this it follows that the heat absorbed in the decomposi- tion of the aqueous hydriodic acid into hydrogen and iodine is 39,000 26,000 = 13,000 cal. and therefore that the heat of formation of the dilute acid is accompanied by an evolution of 13,000 cal. H + I + Aq = HI Aq +13,000 cal. It has further been found that the solution of hydriodic acid gas in water gives rise to an evolution of 19,060 cal., so that the actual formation of the gas from its elements is accompanied by a heat-change of 13,000 - 19,060 = - 6,060 cal. Gaseous hydriodic acid is, therefore, produced from its elements with absorption of this amount of heat. This branch of chemical science, which is known as Thermo- chemistry, has been studied by many chemists, among whom may be mentioned Andrews, Favre and Silbermann, Julius Thomsen and Berthelot, from whose researches the numbers given in the following table are taken. 1 MOLECULAR HEAT OF FORMATION FROM THE ELEMENTS. HC1 21835 HBr 8337 HI - 6060 O 3 - 29378 H 2 67940 H 2 2 - 22927 C1 2 O - 17670 I 2 5 44961 SO 2 70567 SO 3 102526 NH 3 11910 N 2 -17965 NO . - 21438 NO, 2035 N 2 O 5 13000 PH, 4267 POL 104213 POC1 3 144905 . 308626 As 4 G . . . As 2 O 5 217751 AsR - 43770 B 2 3 314821 CH 4 21637 C 2 H 4 - 2679 C0 2 (a) charcoal . . . 96253 (&) gas carbon . . 95806 (c) diamond . . 93190 (d) graphite . . 92660 CO . 67490 CS, -2581 COC1 2 55183 CC1 4 20843 P 2 O 5 177062 PC1 3 74933 1 These numbers have been recalculated from the original results on the basis of = 15-88. 234 THE NON-METALLIC ELEMENTS THE OXIDES. 121 All the elements with the single exception of fluorine are found to unite with oxygen to form an important class of com- pounds termed oxides, possessing very various properties acccord- ing to the nature of the combining element and the quantity of oxygen with which it unites. In many instances one element is found to combine with oxygen in several proportions, giving rise to distinct oxides. Oxides may be divided into three classes, distinguished as Basic oxides, Peroxides, and Acid- forming oxides. (1.) The basic oxides, such as K 2 O, potassium oxide ; BaO, barium oxide ; Fe. 2 O 3 , ferric oxide, form in combination with water a class of compounds termed hydroxides or Jiydrated oxides, such as caustic potash, KOH ; barium hydroxide or caustic baryta, Ba(OH) 2 ; ferric hydroxide, Fe(OH) 3 ; thus : K 2 + H 2 = 2KOH. BaO + H 9 O = Ba(OH) 2 . Fe 2 3 + 3H 2 = 2Fe(OH) 3 . The characteristic property of these oxides as well as of the corresponding hydroxides is their power of neutralizing acids and forming compounds which are termed salts. (2.) The peroxides contain more oxygen than the basic oxides. A portion of it is loosely combined and is given off on heating ; thus 3Mn0 2 = Mn 3 O 4 + O 2 ; and although they can form hydrox- ides, these peroxides have not generally the power of neutralizing acids and forming stable salts. The following is a list of some of the more important peroxides. Barium dioxide BaO 2 ; potassium tetroxide K 2 O 4 ; sodium peroxide Na 2 O 2 ; manganese dioxide Mn0 2 ; lead dioxide PbO 2 . The term peroxide is a somewhat vague one and is applied to oxides possessing very different properties. (3.) The acid-forming oxides combine with water to form hydrates, which are termed acids ; thus Sulphur trioxide S0 3 yields Sulphuric acid H 2 S0 4 . Nitrogen pentoxide N 2 5 yields Nitric acid HN0 3 . Phosphorus pentoxide P 2 O 5 yields Phosphoric acid H 3 P0 4 . S0 3 + H 2 = H S0 4 . N 2 5 + H 2 - 2HN0 3 . P 2 5 + 3H 2 - 2H 3 P0 4 . THE OXIDES 235 Acids possess a sour taste, turn blue litmus red, and neutralize the basic oxides, which when they are soluble have the opposite property, and turn red litmus blue. Salts may be considered to be acids in which the hydrogen is replaced by a metal. They are obtained by a variety of reactions, of which the following are the most important. (1) When certain metals are brought in contact with an acid; thus : Zn + H 2 S0 4 = ZnS0 4 + H 2 . (2) When a basic oxide, or a hydroxide acts upon an acid or an acid-forming oxide, thus : PbO + H 2 S0 4 - PbSO 4 + H 2 O. Ba(OH) 2 + H 2 S0 4 = BaS0 4 + 2H 2 O. BaO + SO 3 = BaSO 4 Ba(OH) 2 + SO 3 = BaS0 4 + H 2 O. (3) When a carbonate of a metal is acted upon by an acid : CaC0 3 + 2HN0 3 = Ca(N0 3 ) 2 + H 2 O + C0 2 . The division into these three classes of oxides cannot, however, be strictly carried out. Thus, whilst the position of the extreme members of each series, such as the strong bases or alkalis, on the one hand, and the acids on the other, can be sharply defined, it is often difficult to classify the middle terms such as alumina, A1 2 O 3 , manganese dioxide, MnO 2 , and tin oxide Sn0 2 , which act sometimes as weak bases and at other times as weak acids. OZONE. O 3 = 47-64. 122 So long ago as 1785 Van Marum observed that oxygen gas through which an electric spark had been passed possessed a peculiar smell, and at once tarnished a bright surface of mer- cury ; but it was not until the year 1840 that the attention of chemists was recalled to this fact by Schonbein. 1 This chemist showed that the peculiar strongly-smelling substance, to which he gave the name of ozone, from ojo>, I smell, is capable of liberating iodine from potassium iodide, and of effecting many other oxidising actions. Schonbein, moreover, showed that ozone is produced in other ways. 1 Pogg. Ann. 4, 616. 236 THE NON-METALLIC ELEMENTS (1.) It is evolved at the positive pole in the electrolysis of acidulated water. (2.) It is obtained by the slow oxidation of phosphorus in the air. (3.) It is formed by the discharge from an electrical machine through air or through oxygen gas. For many years much doubt existed respecting the exact chemical nature of this oxidising principle. Williamson and Baumert came independently to the conclusion that ozone is an oxide of hydrogen having the formula H 2 O 3 ; while Marignac and De la Rive, as well as Fremy and Becquerel, found that ozone is formed when electric sparks are passed through perfectly FIG. 70. dry oxygen gas. The explanation of these contradictory results lies in the fact that it was found impossible to obtain ozone except in very small quantities, and that an exact investi- gation of its composition is rendered still more difficult by its extremely energetic properties. Further researches, conducted with the greatest care, have, however, shown that ozone is nothing more than condensed oxygen, and the steps by which this conclusion has been arrived at constitute an admir- able example of the successful resolution, by the convergence of many independent investigations, of an apparently insoluble problem. To Andrews l belongs the credit of having first proved that 1 Phil. Trans. 1856, p. 13. OZONE 237 ozone, from whatever source derived, is one and the same body, having identical properties, and the same constitution, and also that it is not a compound of two or more elements, but oxygen in an altered and allotropic condition. If a series of electric discharges be sent through a tube con- taining pure and dry oxygen, only a small portion of the gas is converted into ozone ; but if the ozone is absorbed as soon as it is formed, by a solution of iodide of potassium, for example, the whole of the oxygen can be gradually converted into ozone. In order to obtain the maximum production of ozone, pure oxygen gas is allowed to pass through an apparatus (Fig. 70), which con- sists essentially of an iron tube (BB) turned very truly on its outside, through which a current of cold water can be passed by means of the tubes (cc). Outside this metal cylinder is one of glass (AA) very slightly larger than the iron one. By means of the tubes (DD) air or oxygen can be passed through the annular space between the two cylinders. Part of the outer cylinder at G is covered with tinfoil. The outer tinfoil coating and the inner metal cylinder are connected with the poles of an induction coil at E and F. By this means the oxygen is subjected to a series of silent discharges, by which it is converted partially into ozone. The action of this stream of ozonised oxygen upon a sheet of paper covered with a solution of iodide of potassium and starch is strikingly shown when the paper is held in front of the current of issuing gas. The white surface assumes instantly a deep blue colour. 123 That this ozonisation is accompanied by a change of bulk was shown by Andrews and Tait. 1 These chemists filled a glass tube (Fig. 71) with dry oxygen ; one end was then sealed off, whilst the other ended in a capillary tube, bent in form of a syphon, and containing a liquid, such as strong sulphuric acid, upon which ozone does not act. On passing through the gas a silent discharge, obtained by attaching one platinum wire to one pole of a Ruhmkorff's coil, or to the conductor of a frictional electrical machine, a gradual diminution of volume occurred, but this never reached more than T V*h of the whole. After the ozonized gas was heated to about 300 C. it was found to have returned to its original bulk, and had lost all its active properties. This decomposition of ozone into oxygen can be readily shown by allowing the stream of ozonized oxygen to pass through a 1 Phil. Trans. 1860, p. 113. 238 THE NON-METALLIC ELEMENTS tube heated by the flame of a Bunsen-lamp. Every trace of heightened oxidizing action will have disappeared and the blue iodide of starch will not be formed ; whilst on removing the hot tube an immediate liberation of iodine is observed, if the prepared paper is again brought into contact with the issuing gas. In order to gain a knowledge of the composition of ozone, Andrews introduced into his ozone tube a sealed glass bulb containing substances able to destroy the ozone, such as iodide of potassium solution, or metallic mer- cury. After transforming into ozone as much as possible of the oxygen con- tained in this tube, the bulb filled with the iodide of potassium solution was broken and the iodine liberated by the ozone. On observing the column of sulphuric acid in the syphon tube it was found to have remained un- altered after the ozone had reacted, showing that the change had not been attended with any alteration in volume, whilst on afterwards heating up to 300 C. no further increase in the volume occurred, proving that all the ozone had been decomposed. These facts are explained by the sup- position that, in the formation of ozone, three volumes of oxygen condense to form two volumes of ozone 30 = 2(X EIG. 71. vols. 4 vols. which, when heated, increase m bulk again to form the original three volumes of oxygen, whilst, when acted upon by potassium iodide, one- third of the ozone is spent in liberating the iodine, and the other two-thirds go to form ordinary oxygen thus :- + 2KI H 2 O = O I +2KOH. This supposition has been proved to be correct by Soret,as follows : Many essential oils, such as turpentine and oil of thyme, had FORMULA OF OZONE 239 been observed by Schonbein to possess the property of absorbing ozone without decomposing it, and Soret 1 showed that the diminution in volume which takes place on the absorption of the ozone from a measured quantity of ozonised oxygen by these oils is exactly twice as great as the increase of volume observed when the ozone is decomposed by heating the gas. A series of three experiments proved that for every 19*3 cc. of ozone absorbed by the oil, 9 '47 cc., instead of the exact number 9'(55 cc., of common oxygen were formed on heating. Hence ozone possesses the molecular formula O 3 , three volumes of common oxygen having been condensed to two volumes by the formation of ozone. Soret obtained a confirmation of his results from a totally different point of view. 2 If the density of ozone is one-and-a- half times as great as that of common oxygen, the rate of diffusion (p. 65) will be inversely as the square roots of these numbers ; if, therefore, we know the rate at which ozone diffuses, compared with the ra-te of diffusion of another gas whose density is also known, we can draw conclusions respecting the density of ozone. The gas chosen for experiment was chlorine, and it was found by experiment that 227 volumes of chlorine diffused in the same time as 271 volumes of ozone, or for one volume of ozone there diffused 0*8376 volumes of chlorine ; hence, according to the law of inverse squares of the densities, the density of ozone is 24*8 > for 1 : 0-8376 : : \/3519 : V24.8 whereas from the formula 3 it should be the half of 3 X 15 '88 that is 23-82. Brodie 3 arrived, by a long series of most exact determinations, at the same result, inasmuch as he obtained the ratio of 1 to 2 be- tween the volume of the oxygen used in liberating iodine from potassium iodide and that of the ozone absorbed by turpentine, and also showed that all the oxidising effects of ozone upon the most various substances can be explained upon this basis. These experiments prove conclusively that dry oxygen is con- verted by the action of the silent electric discharge into an allo- tropic modification. But they do not decide the question whether the strongly-smelling body obtained in the electrolysis of water has an analogous constitution, or whether it may not be an oxide 1 Ann. Chim. Phys. [4], 8, 113 ; Phil. Mag. [4] 31, 82, and 34, 26. 2 Ann. Chim. Phys. [4] 13, 257. 3 Phil. Trans. 1872, Part ii. 435. 240 THE NON-METALLIC ELEMENTS of hydrogen. Andrews, however, proved that if such electrolytic oxygen is perfectly dried, it does not lose its powerful smell, and that if the dried gas be then passed through a hot glass tube, the smell, as well as the oxidizing power, altogether disappeared with- out the smallest trace of moisture being formed, and this must have been deposited if the electrolytic oxygen had contained an oxide of hydrogen. 124 Atmospheric Ozone. The difficult question as to whether ozone exists in the atmosphere can scarcely be regarded as settled in the affirmative ; it is, however, certain that hydrogen peroxide (p. 311) is present, and it is very probable that it is accompanied by ozone. The higher oxides of nitrogen, amongst other substances, possess the same, power as ozone of liberating iodine from potassium iodide, and these oxides are certainly formed in the atmosphere by electrical discharges, so that if the ozone be measured, as is usually the case, by the amount of iodine liberated, by observing the variation in tint of the so- called ozone papers, we measure, along with the ozone, the higher oxides of nitrogen. The experiments of Andrews ] have, however, decisively proved that an oxidizing substance does occur in the atmosphere which agrees in many of its properties with ozone. Thus when air at the ordinary temperature was passed over ozone test-papers contained in a glass tube, an indication of ozone was seen in two or three minutes. When the air before passing over the test- paper was heated to 260 C. not the slightest action occurred on the test-paper, however long the current was allowed to pass. Similar experiments made with an artificial atmosphere of ozone, that is, with the air of a large chamber containing a little electrolytic ozone, gave precisely the same results. On the other hand, when air mixed with very small quantities of chlorine or the higher oxides of nitrogen was drawn over the papers, they were generally affected whether the air had been previously heated or not. Houzeau has shown that a neutral solution of iodide of potassium on exposure to air becomes alkaline with the liberation of iodine, an effect which would not be produced by the oxides of nitrogen, and which he believed to be due to the presence of ozone in the air. These effects may, however, possibly be due to hydrogen peroxide. The only perfectly satisfactory distinction between ozone and hydrogen peroxide is that the former produces a 1 Proc. Eoy. Soc. 16, 63. ATMOSPHERIC OZONE 241 deposit of peroxide of silver on a piece of silver foil, and this has never been obtained with atmospheric air. 1 In spite of the uncertainty as to whether ozone really exists in the air, many methods have been given for its determination. Thus Zenger passed 100 litres of air through a dilute solution of hydriodic acid and obtained iodine liberated, which corre- sponded to 0*001 or 0'002 milligram of ozone ; and it is doubt- ful how far the iodine was really liberated by ozone. Certain other observers 2 state that the proportion of ozone in the air stands in a direct relation to the amount of atmospheric elec- tricity present, whilst others 3 again conclude from their ob- servations that under the normal atmospheric conditions the amount of ozone in the air is absolutely constant. The usual method of estimating the amount of ozone present in the air is a very rough one. It consists in exposing to the air papers which have been impregnated with a solution of starch and iodide of potassium, for a given time (and best in the dark), and noting the tint which they assume compared with certain standard tints. The papers prepared according to the directions of Dr. Moffat are those on which most reliance is placed. It has indeed been proposed by Bottger to use papers impregnated with thallious oxide as a test for ozone, as this substance is not permanently changed in tint by the nitrogen oxides, but this suggestion has not been generally adopted, and doubt has been thrown by Lamy on the use of this reagent, as anything more than a qualitative test of the presence of ozone. It is scarcely necessary to remark that in thickly-inhabited districts, especially in towns where much coal is burnt, ozone is almost always absent, as it is reduced to ordinary oxygen by the organic emanations as well as by the sulphurous acid constantly present in such air. Some observers state that in the air of the country, and especially in sea air, the presence of ozone can almost always be recognized, often indeed by its peculiar smell, this being said by them to be the most reliable test for its presence. Respect- ing the variations in the amount of atmospheric ozone in different localities or in different seasons we possess at present no reliable information. 4 1 Fremy, Compt. Rend. 61, 939 ; Schone, Ber. 13, 1503. - Neumann, Pogg. Ann, 102, 614, and Poey, Compt. Rend. 65, 708. :1 Smyth, Proc. Meteoro. Soc. June 16, 1869. 4 Compare Ilosva, Bull. Soc. Chim. 3, 2,377. 17 242 THE NON-METALLIC ELEMENTS A possible cause of the formation of ozone in the air has been pointed out by Gorup v. Besanez, 1 inasmuch as he has shown that an oxidizing substance is invariably formed when water evaporates. This substance is however probably hydrogen peroxide. The production of ozone by the slow oxidation of phosphorus has already been mentioned. Several other substances on oxidation also give rise to a formation of ozone ; thus turpentine and several other essential oils when acted upon by atmospheric oxygen transform a portion of it into ozone. This may be seen by shaking turpentine in a flask containing air or oxygen, when the liquid will exhibit the properties of ozone. Another method by which the active variety of oxygen may be obtained is by acting with strong sulphuric acid upon dry barium dioxide, when oxygen is given off, which is found to contain considerable quantities of ozone. It is also stated that ozone is formed during combustion, and can be recognized by its smell when a current of air is blown through the upper portion of a flame. Properties. Ozone prepared by any of the above methods is a gas possessing a peculiar odour, somewhat resembling that of very diluted chlorine. It has a faint blue colour, which is rendered more evident by compression. Ozone, when dry, may be preserved in sealed glass tubes at the ordinary atmospheric temperature for a very long time, but it changes gradually into common oxygen. Not only is ozone destroyed by heat, but also when agitated strongly with glass in fine fragments (Andrews). It is one of the most powerful oxidiz- ing agents known ; it attacks and at once destroys organic substances such as caoutchouc, paper, &c. One of the most characteristic actions of ozone is its effect on mercury. The metal at once loses its mobility and adheres to the surface of the glass in a thin mirror, and so delicate is this reaction, that a single bubble of oxygen containing -gVth of its bulk of ozone will alter the physical characters of several pounds of mercury, taking away its lustre and the convexity of its surface. In many of its oxidising actions the volume of ozone does not undergo any alteration, one molecule of ozone, O 3 , yielding one molecule of ordinary oxygen, O 2 , and one atom of oxygen being employed for the oxidation. Ozone is converted into ordinary oxygen by contact with certain metallic oxides, such as oxide of silver and manganese dioxide or with 1 Annalen, HI, 232. PROPERTIES OF OZONE 243 platinum black. These substances are not permanently altered by the reaction, which is probably of somewhat the same nature as that by which potassium chlorate is decomposed at low tem- peratures in the presence of certain bodies (p. 217). Some non- metals as well as most metals are at once oxidized in presence of moist ozone : phosphorus to phosphoric acid, sulphides to sulphates, ferrocyanides to ferricyanides, whilst blood is com- pletely decolorized, the albumen being entirely, and the other organic matters being nearly all, destroyed. Ozone has not, however, according to the experiments of Carius, 1 the power it was formerly supposed to possess, of oxidizing nitrogen to nitric acid in presence of water. Ozone is somewhat soluble in water, imparting to water its peculiar odour as well as its oxidizing powers. According to Carius, 1,000 volumes of water dissolve 4*5 volumes of ozone, and it is much more soluble in certain ethereal oils. Ozone forms on condensation an indigo-coloured liquid (Hautefeuille and Chappuis), which boils at 106 (Olszewski), and like liquid oxygen is strongly magnetic. 2 The compression of the gas must be effected slowly, or the heat produced raises its temperature to such a point that the gas is suddenly con- verted into ordinary oxygen with explosion. A considerable amount of heat is evolved in this change, ozone being formed from oxygen with absorption of 29,378 cal. (Berthelot). 30 2 = 20 3 - 2 x 29378. It is found that the change of one allotropic form of a substance into another is always accompanied by either an evolution or an absorption of heat. Schonbein, and certain other chemists, believed that another modification of oxygen besides ozone exists, to which they gave the name of ant-ozone ; the chief peculiarity of this body being its power of combining with ozone to form ordinary oxygen. Further experiments have, however, proved that ant-ozone is nothing more than hydrogen dioxide. 3 1 Liebig, Annalen, 179, 1. 2 Dewar, Proc. Roy. Soc. 50, 261. 3 See Brodie, Phil. Trans, 1862, 837. 244 THE NON-METALLIC ELEMENTS HYDROGEN AND OXYGEN These elements form two compounds, (1) HYDROGEN MONOXIDE or WATER, H 2 O, and (2) HYDROGEN DIOXIDE, H 2 O 2 . WATER. H 2 17-88. 125 The question of the discovery of the composition of water, a substance which up to nearly the end of the last century was considered to be a simple body, has been fully discussed in the FIG. 72. historical introduction. We there learned that Cavendish first ascertained that by the combustion of two volumes of hydro- gen and one volume of oxygen, pure water and nothing else is produced. Warped, however, as his mind was with the phlogistic theory, he did not fully understand these results, and the true explanation of the composition of water was first given by Lavoisier in 1783, when the French chemist repeated and confirmed the experiments of Cavendish. The apparatus, of much historical interest, used by him for proving that hydrogen gas is really contained in water, is seen in facsimile in Fig. 72. WATER 245 The water contained in the vessel a was allowed to drop slowly into the tube, e d, from which it flowed into the gunbarrel, df, heated to redness in the furnace. Here part of the water is decomposed, the oxygen entering into combination with the metallic iron, whilst the hydrogen and some undecomposed steam passed through the worm, s, where the steam was con- densed and the hydrogen was collected and measured in the glass bell-jar, m. The result of these experiments was found to be that 13*13 parts by weight of hydrogen united to 86'87 parts by weight of oxygen, or 12 volumes of oxygen with 22'9 volumes of hydrogen. 1 Cavendish, by exploding air with hydrogen by means of the FIG. 73. electric spark had, on the other hand, come to the conclusion that the relation by volume of the two gases combining to form water was 1 of oxygen to 2 of hydrogen, and this was confirmed in 1805 by the more exact experiments of Gay- Lussac and Humboldt. 2 The formation of water by the combustion of hydrogen in the air can be readily observed by means of the arrangement shown in Fig. 73. The hydrogen is dried by passing through the horizontal tube filled with pieces of chloride of calcium, then ignited at the end of the tube, and the flame allowed to burn under the bell- jar. By degrees drops of water form, these collect on the sides of the glass, and drop down into the small basin placed beneath. 1 Memoire par MM. Meusnier et Lavoisier, annee 1781, p. 269, In le 21 Avril, 1784. Mem. de VAcad. de Sciences, 2 Journ. de Phys. 60, 129. 246 THE NON-METALLIC ELEMENTS Another apparatus for exhibiting the same fact is seen in Fig. 74. It consists of a glass gasholder filled with hydrogen, which is dried by passing through the chloride of calcium tube (b), and then burns under the glass funnel (c). The water formed collects in the tube (e), an aspirator (/) drawing the steam formed by the combustion through the tube (d). 126 Eudiometric Synthesis of Water. The method which Cavendish employed for the purpose of ascertaining the com- position of water is still employed in principle, although the modern processes are much superior in accuracy to the older ones. Bunsen's modification of the method consists in bringing known volumes of the constituent gases successively into a eudio- FIG. 74. meter and allowing these gases to combine under the influence of the electric spark, carefully observing the consequent change of volume. The eudiometer employed is a strong glass tube (c), Fig. 75, one metre in length and 0'025 m. in breadth, closed at the top and open at the bottom, having platinum wires sealed through the glass near the closed end. The tube is accu- rately divided into divisions of length by etching a millimetre scale on the glass, and the capacity of each division of length on the scale is ascertained by a process of calibration, consisting in pouring successively exactly the same volume of mercury into the tube, until the whole is filled with the metal, the height to EUDIOMETRIC SYNTHESIS OF WATER 247 which each volume of mercury reaches being carefully read off on the millimetre scale etched on the glass. The eudiometer containing at the top one drop of water to render the gases moist, is first completely filled with mercury and inverted in the pneumatic trough (d) containing the same metal. FIG. 75. Then a certain volume of perfectly pure oxygen gas, prepared from pure potassium chlorate, is introduced, the volume is read off, and the necessary reductions for temperature and pressure are made. For this purpose a thermometer (6) is hung up near the eudiometer, and the temperature as well as the level of the meniscus of mercury in the tube read off by means of a tele- 248 THE NON-METALLIC ELEMENTS scope placed in a horizontal position at such a distance that the radiation from the observer does not produce any sensible effect on the reading. The pressure to which the gas is subjected is then ascertained by reading off the height of the barometer (a), also placed near the eudiometer, and subtracting from this the height of the column of mercury in the eudiometer above the level of the mercury in the trough, this height being obtained by reading the millimetre divisions at the upper and lower levels of the mercury. The temperature of the mercurial columns in the barometer and eudiometer must also be observed, so that cor- rection may be made for the expansion of the mercurial column the height of which must be reduced to that of a column at. 0C. We have now, taking an actual example : (1) The observed volume of moist oxygen taken from the reading of the upper level of mercury and from the calibration- table of the eudiometer = 399'1. (2) The temperature of the gas = 15 C. (3) The height of the barometer (corrected to C.) = 765 mm. (4) The height of the mercury column in the eudiometer (corrected to C.) = 500 mm. From these data it is easy to obtain the volume of the gas at the normal temperature (0) and under some standard pressure (either 1 m. or 760 mm. of mercury at 0). The gas has, however, been measured in the moist state ; the water vapour present exerts a certain pressure, and thus depresses the column of mercury in the eudiometer and increases the apparent pres- sure of the gas. In order, therefore, to obtain the actual pressure of the dry gas it is necessary to subtract from the- height of the barometer, the column of mercury in the eudio- meter and the vapour pressure of the water at the temperature of the experiment, which may be found in a table of vapour pressures (p. 277). This amounts at 15 to 127 mm. of mercury, so that the true pressure of the gas considered dry is : 765 - 500 - 127 = 252-3 mm. Applying the laws of Boyle and Dalton it appears that 399'1 vols. of gas at 15 and 252'3 mm. pressure will, at and a pressure of 1 metre of mercury, occupy a volume of : 3991 x 273 x 252-3 _ Q5 . 45 288 x 1000 EUDIOMETRIC SYNTHESIS OF WATEft 249 The second part of the process consists in adding a volume of pure hydrogen, care being taken not to allow any bubbles of gas to remain attached to the sides of the tube. The volume of hydrogen added must be such that the inflammable mixture of two volumes of hydrogen and one volume of oxygen shall make up not more than from 30 to 40 per cent, by volume of the whole gas, otherwise the mercury is apt to be oxidized by the high temperature of the explosion. Thus supposing we had five volumes of oxygen, we must add ten volumes of hydrogen to combine with this, and ~= = 28 volumes for the purpose oo of dilution. As soon as the temperature equilibrium has been established, the volume of the mixed gases contained in the eudiometer is again read off with the same precautions, and the temperature and pressure again ascertained as before. This having been accomplished, the open end of the eudiometer is firmly pressed down below the mercury in the trough upon a plate of caout- chouc, previously moistened with corrosive sublimate solution, and held firmly in this position by a stout clamp. By means of an induction coil an electric spark is then passed from one platinum-wire through the gas to the other wire ; the mixed gases are thereby ignited, and a flame is seen to pass down the tube. On allowing the mercury from the trough again to enter freely at the bottom of the tube a considerable diminution of bulk is observed. The eudiometer is then allowed to remain untouched until the temperature of the gas has again attained that of the surrounding air, and the volume, pressure, tension, and temperature are ascertained as before. The volume which has disappeared does not, however, exactly correspond to the true volume of gases which have united, inasmuch as the water formed, occupies a certain although a very small, space. In order to obtain the exact volume of the combined gases, the volume of the explosive gases before the explosion must be multiplied by the number O'OOOo, which represents the frac- tion of the total bulk of the component gases which is occu- pied by the liquid water formed, and this volume must then be added to the observed contraction. For other corrections the article on this subject in Bunsen's Gasometry must be consulted. The following numbers illustrate the course of such an ex- periment : 250 THE NON-METALLIC ELEMENTS Synthesis of water by volume. Reduced to and 1 m. of mercury. Volume of oxygen taken 95 '45 Volume of oxygen and hydrogen . . 557*26 Volume after the explosion .... 27l'06 Hence 286*2 volumes disappeared, or 95*45 volumes of oxygen have combined with 190*75 of hydrogen. Consequently 1*0000 volume of oxygen combines with 1*9963 volumes of hydrogen to form water. FIG. 76. By careful repetition of the above experiments the com- position of water by volume has been ascertained, within very narrow limits, to be in the proportions of one of oxygen to two of hydrogen. Since the knowledge of the exact composition of water by volume is of the greatest importance for the determination of the atomic weight of oxygen, many attempts have been made to devise more accurate methods than that above de- scribed, for ascertaining the exact ratio in which hydrogen and oxygen combine by volume. ELECTROLYSIS OF WATER 251 The most accurate of these is due to Morley, 1 who has re- tained the principle of the eudiometric method, but greatly improved its details. The greatest care was expended in purify- ing the hydrogen gas employed, which was prepared by the electrolysis of pure dilute sulphuric acid. Special precau- tions were taken to free the gas from any traces of oxygen by passing it over heated copper, after which it was care- fully dried and used for the experiment. Portions of each of the pure gases were then brought into the eudiometer and exploded, the oxygen being in excess in some and the hydrogen in other experiments. As soon as a large volume of the gases had been brought into combination by successive introductions followed by explosions, the residue remaining in the eudiometer was analysed and the amount of residual hydro- gen or oxygen, and impurity consisting of atmospheric nitrogen from which it is impossible to free the hydrogen, ascertained. This was then subtracted and the volumes of the two gases thus determined. The average of twenty experiments con- ducted in this manner gave the ratio of 1 : 2'0002. It is clear that if Gay-Lussac's law were strictly applicable, the volumes must be in the simple ratio .of 1:2. The slight deviation from this proportion, which seems to be well established, serves to show that this law, like those of Boyle and Dalton, is only approximately true, and that Avogadro's theory which is founded upon it, is also only an approximate representation of the truth. This deviation must of course vary with the pressure at which the comparison is made, since the two gases deviate to different extents from Boyle's law. Electrolytic Analysis of Water. 127 A convenient form of voltameter for demonstrating the com- position of water by volume is shown in Fig. 76. On passing a current of electricity through the water, acidified with sulphuric .acid, which fills the U-shaped tube, bubbles of oxygen rise from the surface of the platinum plate forming the positive pole, whilst bubbles of hydrogen are disengaged from the negative pole. The gases from each pole are collected separately, and the volume which collects in the tube containing the negative pole is seen to be a little more than double that which collects from the positive pole. On trial the latter is found to be 1 Chem. News, 63, 218. 252 THE NON-METALLIC ELEMENTS oxygen, and the former hydrogen. In this experiment the volume of the oxygen gas is found to be rather less than half that of the hydrogen, because, in the first place, it is more soluble in water than hydrogen, and, secondly, because a portion of the oxygen is converted into ozone, which, being condensed oxygen, occupies a less volume than oxygen in the ordinary form. The fact that ozone is thus produced may be shown by bringing some iodized starch paper in contact with the electro- lytic gas, when iodine will be liberated, and the paper will FIG. 77. at once be turned blue. By raising the temperature of the acidulated water to 100, the solution of the gases is prevented, and at the same time the formation of ozone is avoided, so that the true volume relation is thus much more closely attained. It must be remembered that the decomposition of the water is not directly caused by the passage of the electric current. The sulphuric acid which is present yields the ions H and SO 4 (p. 116), the former of which give up their charges of electricity at the negative pole and escape from the solution, the atoms VOLUMETRIC COMPOSITION OF STEAM 253 combining to form molecules of hydrogen. The ions SO 4 , on the other hand, give up their charges at the positive pole and at once decompose, oxygen being liberated and a fresh quantity of sulphuric acid formed : S0 4 + H 2 = H 2 S0 4 + O. The atoms of oxygen then unite to form molecules which escape from the solution, whilst the regenerated sulphuric acid again undergoes a similar series of changes. The apparatus, the construction of which is plainly shown in Fig. 77, is used for collecting the mixed gases evolved by the electrolysis of water. The mixed gases, thus prepared, combine with explosive violence when a flame is brought in contact with them, or when an electric spark is passed through the mixture. In this act of combination, the whole of the hydrogen and the whole of the oxygen unite to form water : in other words, sub- ject to the correction above referred to respecting the volume of water formed, the total volume of the detonating gas disappears. That this is the case is seen from the following experiments made by Bunsen, in which air was mixed with the electrolytic gas, the mixture exploded, and then the volume of air deter- mined. A second addition of the explosive gas was next made and the volume of air again read off, and the operation repeated a third time. 1 Original volume of air, in which detonating \ gas had been once exploded j After explosion, with 55*1.9 vols. detonating ) gas f Ditto, measured again after 24 hours . . . 112*57 After second explosion with 71 '23 vols. deto- ") ,. I 112'66 nating gas J 128 Volumetric Composition of Steam. Gay-Lussac not only determined the composition of water by volume, but was the first to ascertain that three volumes of the mixed gases com- bine to form two volumes of gaseous steam ; inasmuch as he found the specific gravity of steam to be O6235, the number deduced from the above composition being 0'6221. This fact can be readily shown by exploding some of the 1 Gasometry, p. 65. 254 THE NON-METALLIC ELEMENTS electrolytic detonating gas evolved from the voltameter, Fig. 77 in the eudiometer E Fig. 78 which is so arranged that the pressure on the gas can be altered at pleasure. Surrounding FIG. 78. the eudiometer is a glass tube (T), and between the two tubes a current of the vapour of amyl alcohol, which boils at 132, can be passed from the flask (F), and the vapour, after passing COMPOSITION OF WATER 255 through the tube, condensed in the flask cooled in the trough of water (H). When the temperature of the tube and of the gas has risen to 132, the volume of the gas is exactly read off on the divided scale of the eudiometer, the height of the mercury in the two limbs having been brought up to the same level by means of the reservoir of mercury (M) attached to the iron foot of the eudiometer by the caoutchouc tube (G). The pressure on the gas is now reduced by lowering the level of the mercury and, by means of the induction coil (c), a spark is passed. As soon as combination has taken place, the level of mercury in the two tubes is brought to the same height, and the volume of the gaseous water is accurately read off, the temperature of the whole being still kept up to 132 by the current of amyl alcohol vapour. This volume is found to be almost exactly two-thirds of that of the original mixed gases, and hence we conclude that 2 vols. of hydrogen and 1 vol. of oxygen unite together to form 2 vols. of steam or gaseous water. The Composition of Water and the Atomic Weight of Oxygen, 129 A knowledge of the composition of water by weight is of the utmost importance, because the ratio in which hydrogen and oxygen combine must be known before the exact atomic weight of the latter (hydrogen being taken as the unit) can be deter- mined. The earlier experimenters adopted a very simple method for this purpose. Many metallic oxides such as copper oxide, CuO, when heated in a current of hydrogen lose their oxygen which combines with the hydrogen to form water, the metal being produced. By ascertaining the loss of weight which the oxide thus suffers, and by weighing the water formed, we obtain all the data required for determining the ratio by weight in which the two gases are present in water, inasmuch as water contains no other constituent besides oxygen and hydrogen. This method of determining the synthesis of water by weight was first proposed and carried out in 1820 by Berzelius and Dulong, 1 with the following results : 1 Ann. Chim. Phijs. 15, 386. 256 THE NON-METALLIC ELEMENTS Synthesis of Water ly Weight. No. 1 . Loss of weight of copper oxide. . 8-051 . . Weight of water obtained. 9-052 2 . . 10-832 . . 12-197 3 8-246 9-270 Hence we have the following numbers as representing the per- centage composition of water according to these experiments : Percentage Composition of Water ly Weight (Berzelius and Didong}. No. 1. No. 2. No. 3. Mean. Oxygen . . 88-942 88-809 88-954 88-90 Hydrogen 11-058 11-191 11-046 11-10 100-000 100-000 100-000 100-00 It is thus seen that the separate experiments do not agree very closely amongst themselves, and do not therefore yield us cer- tain information as to the exact proportions by weight in which the gases combine to form water. In the year 1843, Dumas 1 undertook in conjunction with Stas, a most careful repetition of these experiments ; pointing out the following probable sources of error in Berzelius's experiments : (1) The weight of water formed ought either to be ascertained in vacuo or reduced to a vacuum ; this reduction would increase the quantity of water by about 10 to 12 milligrams. (2) The weight of oxygen ought also to be reduced to a vacuum. (3) The hydrogen ought to be much more carefully dried than was the case in the older experiments. (4) Lastly, even supposing that the weights had thus been adjusted, and if the hydrogen had been properly dried, Berze- lius's determinations were made upon too small a scale to ensure the necessary degree of accuracy. A facsimile of the apparatus as used by Dumas is shown in Fig. 79. F is the vessel in which the hydrogen is evolved. E is a funnel with stop-cock, containing sulphuric acid. 1 Ann. Chim. Phys. [3], 8, 189 COMPOSITION OF WATER 257 18 258 THE NON-METALLIC ELEMENTS A is a cylinder filled with mercury under the surface of which dips a safety tube. The first U-tube contains pieces of glass moistened with nitrate of lead. The second U-tube contains glass moistened with silver sulphate. The third U-tube contains in the first limb pumice moist- ened with potash, and in the second limb pieces of solid caustic potash. The fourth and fifth U-tubes contain fused solid caustic potash. The sixth and seventh U-tubes contain fragments of pumice powdered over with phosphorus pentoxide, and are immersed in a freezing mixture. The eighth is a small weighed tube containing phosphorus pentoxide. B is a bulb blown on hard glass containing the dry oxide of copper, furnished with a stop-cock (r) at its upper end, and drawn out so as to pass into the narrow neck of the vessel B t at the lower end. The bulb B can be heated by the Bunsen lamp placed on the sliding holder of the retort-stand. B : is the bulb in which the water, formed by the decom- position, collects. The U-tube placed next to the bulb B, contains pieces of fused caustic potash. The next U-tube contains phosphorus pentoxide and is sur- rounded by a freezing mixture. Next to this is placed a small weighed tube containing phosphorus pentoxide, whilst at the end we find another tube like the last, but not weighed. A cylinder A 2 filled with sulphuric acid, through which the excess of hydrogen gas escapes, completes the arrangement. With .this apparatus Dumas made no less than 19 separate experiments carried out with very great care. The hydrogen evolved from zinc arid dilute sulphuric acid, might contain oxides of nitrogen, sulphur dioxide, arseniuretted hydrogen and sul- phuretted hydrogen. These impurities are eliminated, and the gas at the same time completely dried by passing over the substances contained in the U-tubes ; the nitrate of lead absorbs the sulphuretted hydrogen ; the sulphate of silver de- composes any trace of arseniuretted hydrogen, and the rest of COMPOSITION OF WATER 259 the tubes serve to arrest every trace of carbonic acid and mois- ture, so that the gas passing through the stop-cock r into the bulb B consists of perfectly dry and pure hydrogen. In order to render this certain, the small tube next to the bulb is weighed before and after the experiment, and if its weight remain con- stant we have proof that the gas has been properly dried. A similar small weighed tube serves a like purpose at the other end of the apparatus. Great care must be taken that the oxide of copper contained in the bulb B is perfectly dry, for this oxide being a hygroscopic substance is liable to absorb water from the atmosphere. The weight of the bulb containing the oxide is then accurately deter- mined, and after all the air has been driven out of the U-tubes by the dry hydrogen, the bulb is fixed in its place. The bulb destined to receive the water is also carefully weighed before the experiment, together with the 3 drying-tubes placed beyond it for the purpose of absorbing every trace of aqueous vapour car- ried over by the hydrogen. Then the oxide of copper is heated to dull redness ; the reduction commences, and the formation of water continues for from 10 to 12 hours. After this, the bulb B is allowed to cool in a current of hydrogen ; the apparatus is then taken to pieces, the bulb rendered vacuous and weighed, whilst the hydrogen contained in the bulb and tubes serving to collect the water is displaced by dry air before this portion of the apparatus is weighed. It is clear the weight of hydrogen is not directly determined by this method but that it is obtained as the difference between the weight of water produced and that of the oxygen consumed. As, however, the weight of the hydrogen is only of that of the water formed, it is evident that a percentage error of a given amount on the weight of water will represent a much larger percentage error on the smaller weight of hydrogen. The simplest way of reducing such errors is to arrange the experiment so that a large quantity of water is obtained, for the experimental errors remain, for the most part, constant, and by increasing the quantity of substance experi- mented upon, the percentage error is kept down. For this pur- pose Dumas took such weights of copper oxide as would produce in general about 50 grams of water, so that the experimental error, on hydrogen taken as the unit, is reduced to O'OOo of its weight. In the 19 experiments Dumas found that 840*161 grains of oxygen were consumed in the production of 945*439 260 THE NON-METALLIC ELEMENTS grams of water ; or the percentage composition of water by weight is as follows : Percentage Composition of Water ly Weight (Dumas). Oxygen 88'864 Hydrogen 11136 100-000 In other words two parts by weight of hydrogen combine with 15 '9 6 08 parts by weight of oxygen to form water. This number is in almost exact accordance with the results of the volumetric analysis of Bunsen, and the density deter- minations of Regnault. According to the former, the ratio of the volumes in which the two gases combine is exactly 1 : 2, whilst the latter .found that oxygen is 15 '96 times as heavy as hydrogen. Dumas' results represented the most ac- curate determinations made up to the year 1888. but since that time a number of investigations on the subject has been made with the utmost care and accuracy, the result of which has been to show that the number 15 '96 is undoubtedly too high. The chief sources of error in the experiments of Dumas are (1) the presence of impurities in the hydrogen; (2) the fact that heated copper takes up a certain amount of hydro- gen, forming a hydride of this metal; (3) the action of the hydrogen upon sulphuric acid, used by Dumas in some of his experiments as a drying agent, by which sulphur dioxide is liberated. In the more recent determinations, the results of which are given in the accompanying table, these errors have as far as possible been avoided. Thus Cooke and Richards, Noyes and Reiser, weighed the hydrogen employed, the last-named in the form of palladium hydride, the others as the free gas, together with the water produced by the reduction of copper oxide. Dittmar and Henderson, and Leduc made use of Dumas' original method, whilst Rayleigh weighed both the hydrogen and the oxygen, but not the water. The atomic weight calculated from the volumetric determination of Morley and the density as found by Rayleigh leads to a number which is identical with the mean of those directly obtained (if the result of Reiser's experiments, which seems to be undoubtedly too high, be excluded). It may, therefore, be concluded that two parts by weight of hydrogen PKOPERTIES OF ELECTROLYTIC GAS 261 combine with 15 '88 of oxygen to form 17*88 parts of water, the percentage composition of which is the following : Hydrogen Oxygen . 11186 88-814 100-000 The following table contains a summary of the results obtained by the different investigators who have determined the relative atomic weights, density and combining volumes of these two elements : Name. Date. Atomic Weight. Density. Combining Volumes. Gay-Lussac and Humboldt Dumas 1805 1842 15-96 ... 1 :2 Regnault 1845 15-96 Regnault 1 corrected . . . Ravleiffh 2 1888 ... 15-91 15-884 Cooke and Richards 3 . . . Keiser 4 . 1888 1888 15-869 15-949 Scott^ Rayleigh * 1888-93 1889 15-89 ... 1 . 1-9952-0025 Noyes 6 Dittmar and Henderson 7 . Morley 8 1890 1890 1891 15-896 15-866 15-879 1 2-00023 Leduc 9 . 1891 15-905 1 : 2-0037 Rayleigh 2 1892 15-882 f *i Leduc y 1892 15-880 Experiments with the Detonating Mixture of Oxygen and Hydrogen. 130 In order to exhibit the explosive force of this detonating gas a thin bulb (B), Fig. 80, of a capacity from 70 to 100 cubic centimetres is blown on a glass tube. This is filled with the gas evolved from the voltameter (A) as shown in the figure, and, when 1 Regnault (corrected by Crafts), Compt. Rend. 106, 1662. 2 Rayleigh, Proc. Eoy. Soc. 43, 356 ; 45, 425 ; 50, 448. 3 Cooke and Richards, Amer. Chem,. Journ. 10, 81, 191. 4 Keiser, Amer. Chem. Journ. 10, 249. 5 Scott, Proc. Eoy. Soc. 42, 396 ; Brit. Assoc. Reports, 1888, 631. 6 Noyes, Amer. Chem. Journ. 12, 441. 7 Dittmar and Henderson, Proc. of Glasgow Phil. Soc. 189091 ; Chem. News (1893), 54. 8 Morley, Amer. Journ. of Science [3], 41, 220 ; Chem News, 63, 218, &c. 9 Leduc, Compt. Rend. 113, 186 ; 115, 41, 311 ; 116, 1248. 262 THE NON-METALLIC ELEMENTS full, is placed over the perforated cork (c), through which two insulated copper wires are inserted, these being connected at the extremity by a fine platinum wire. The bulb is then sur- rounded with a protecting cover of wire gauze (G), and a current of electricity passed through the platinum wire, which soon becomes heated to a temperature high enough to cause an instantaneous combination of the oxygen and hydrogen to occur ; a sharp explosion is heard, and the bulb is shattered to fine dust. The amount of the energy thus generated can be easily calcu- lated when the quantity of heat developed by the combination is known. Thus 1 grm. of hydrogen on burning to form water FIG. 80. evolves 33,970 thermal units, or heat sufficient to raise 33,970 grms. of water from to 1. But the mechanical equivalent of heat is 423, that is, a weight of 423 grams falling through the space of 1 metre is capable of evolving heat enough to raise 1 gram of water from to 1. Hence 1 gram of hydrogen on burning to form water, sets free an amount of energy represented by that required to raise a weight of 33,970 x 423 grams or 14,309 kilograms through the space of 1 metre. The gases may, however, be made to combine not only rapidly, as we have seen, but also slowly and quietly. High tem- perature, the passage of the electric spark, and the presence of platinum and other bodies effect the change in the first of these ways. The smallest electric spark suffices to cause the combina- tion of the largest masses of pure detonating gas, because the PHENOMENA OF EXPLOSION 263 heat which is evolved by the union of those particles in whose neighbourhood the spark passes is sufficient to cause the com- bination of the adjacent particles, and so on. In every case a certain minimum temperature, termed the temperature of igni- tion, differing for each gas, must be reached in order that the union shall take place, and the temperature may be so lowered by mixing the detonating gas in certain proportions with inactive gases that the explosive mixture cannot inflame. Thus one volume of detonating gas explodes when mixed with 2'82 vols. of carbon dioxide, with 3*37 vols. of hydrogen, or with 9*35 vols. of oxygen ; but it does not explode when mixed with 2*89 vols. of carbon dioxide, with 3'93 vols. of hydrogen, or with 10*68 vols. of oxygen. 1 When the mixture of electrolytic gases is sealed up in a glass bulb (which can be done without explosion occurring if the side tubes are of capillary bore) and heated in the vapour of boiling sulphur (448) combination begins to take place slowly, the whole of the mixed gases being at length converted into water without any explosion occurring. If on the other hand the bulb be immersed in the vapour of boiling zinc chloride (606) an explosion invariably takes place. This does not occur, however, until the temperature reaches 650730 if the gas is passed through the bulb in a slow stream. The temperature at which combination commences is much influenced by the nature of the surface with which the gas is in contact. Thus if the interior of the bulb be coated with silver, combination begins at as low a temperature as 155 . 2 The following experiments indicate the slow combination of oxygen and hydrogen. If a spiral of clean platinum wire is held for a few seconds in the flame of a Bunsen burner, and then the flame extinguished and the gas still allowed to stream out round the spiral, it will be seen that the spiral soon becomes red hot, either continuing to glow as long as the supply of gas is kept up, or rising to a temperature sufficient to ignite the gas (Davy). A palladium wire acts in a similar way, but wires of gold, silver, copper, iron and zinc produce no action of this kind. A perfectly clean surface of platinum plate also first effects a 1 Bunsen, Gasometry, p. 248. 2 V. Meyer and Krause, Annalen, 264, 85 ; V. Meyer and Askenasy, Annalen, 269, 49 ; V. Meyer and Freyer, Ber. 25, 622 ; Zeit. Phys. Chem. H, 28 ; V. Meyer, Ber. 24, 4233. See also Mitscherlich, Ber. 26, 160. 264 THE NON-METALLIC ELEMENTS slow, but after a time even an explosive combination of the detonating gas (Faraday). The finely-divided metal (spongy platinum) which exposes a great surface to the action of the gas, also induces, at the ordinary temperature, the combination of hydrogen with air or oxygen ; at first a slow combustion takes place, but when the metal becomes red hot, a sudden explosion occurs (Dobereiner). Small traces of certain absorbable gases, such as ammonia, destroy the inflaming power of the spongy platinum, but this power is regained on ignition. The most probable explanation of this property of platinum is that this metal possesses the power of condensing on to its surface a film of hydrogen and oxygen, which gases, when brought under these circumstances into intimate contact, are able to combine at the ordinary atmospheric temperature, and by the heat which their com- bination evolves, to excite the union of the remaining gaseous mixture. The following experiment strikingly shows that a mixture of hydrogen and air becomes inflammable only when a definite proportion between the two gases has been reached. Fig. 81 represents a suspended glass bell-jar closed at the top, and covered at its mouth by a sheet of paper gummed on to the glass. A glass syphon passing through the paper cover is fastened by copper wires to the bell-jar with the longer limb on the outside. By means of a gas-generating apparatus the bell- jar is filled with hydrogen by displacement, a rapid current of the gas being made to pass in through the syphon, the air finding its way out through the pores of the paper. When the bell- jar is full of hydrogen, the vulcanized tube is removed from the end of the long limb of the syphon, and the stream of hydrogen gas which issues from the end (hydrogen being lighter than the air, can be syphoned upwards) is then lighted and is seen to burn with its usual quiet non-luminous flame. After a short time, however, this flame may be seen to flicker, and is heard to emit a musical note which begins by being shrill, but gradually deepens to a bass sound, until, after a time, distinct and separate impulses or beats are heard, and at last, when the requisite pro- portion between the hydrogen and the air which enters through the pores of the paper has been reached, the flame is seen to pass down the syphon and enter the bell-jar, when the whole mass ignites with a sudden and violent detonation. 131 The Phenomena of Explosion in Gases. The phenomena PHENOMENA OF EXPLOSION 265 which accompany the ignition of a detonating gas are of a very interesting and important character. Bunsen, who was the first to investigate this subject, directed his attention mainly to two points, the rate of propagation of the explosion and the pressure produced, both of which have been more fully studied by later investigators. 1 Bunsen, 2 in 1857, attempted to determine the rate at which the explosion is propagated by igniting the gas as it issued from FIG. 81. the end of a narrow tube and measuring the rate at which the detonating mixture had to be supplied to prevent the flame passing back along the tube. In this way he found for hydrogen and oxygen the rate of 34 metres per second. 1 Berthelot, "Sur la force des Matieres Explosives (Paris)," Ann. Chim. Phys. [5], 28, 289 ; Mallard and Le Chatelier, Compt. Rend. 1881, 145, "Combustions des Melanges Gazeux " ; Dixon, Phil. Trans. 184. 97. Phil. Mag. [4], 34, 493. 266 THE NON-METALLIC ELEMENTS In 1881, however, it was found that when the explosive gas is fired in a tube, the rate of explosion increases from its origin until it reaches a certain maximum, after which it remains constant whatever the length of the column of gas may be (Berthelot, Mallard and Le Chatelier). The disturbance by which the ignition is propagated throughout the gas is known as the " explosion wave " ; the maximum rate attained is exceedingly high and has a perfectly definite and constant value for each explosive mixture. The experiment of Bunsen, as will be seen, referred to the initial period of the combination before the explosion wave had attained its characteristic velocity. In order to determine the rate, the detonating gas is brought into a leaden tube, about 9mm. in diameter and 100 metres in length, which is closed at either end by steel stop-cocks. Near one end the gas can be fired by means of an electric spark, and at about four feet from this point an insulated bridge of silver foil is placed across the interior of the tube, a second similar bridge being placed near the second stop-cock. These bridges convey electric currents, and are connected with a delicate chronograph. The gas is fired by a spark, and the explosion after attaining its maximum rate in the first four feet of the tube breaks the first bridge, passes throughout the length of the tube and finally breaks the second bridge, the time which elapses between the two ruptures being recorded by the chrono- graph. In the case of hydrogen and oxygen the enormous velocity of 2,821 metres per second has been found, and the rates in other gases are of the same order of magnitude. The addition of an excess of one or other of the gases or of an inert gas is found to modify the rate of explosion by altering the temperature which is attained and the density of the gases. Thus, in a mixture of 8 vols. of hydrogen with 1 of oxygen the rate is 3,532m. whilst in one containing 1 vol. of hydrogen with 3 vols. of oxygen it is 1,707m. per second. These velocities correspond closely with those which would be attained by sound in the gases concerned at the high temperature produced by the combustion under the circumstances of the experiment (Dixon). The ignition seems indeed to be propagated in somewhat the same manner as a sound-wave. The numbers calculated for hydrogen and oxygen on this supposition agree well with the experimental results obtained with the diluted gases. When the pure detonating gas is employed, however, the calculated results are invariably higher than the experi- OXYHYDROGEN FLAME 267 mental. This is probably due to the fact, that a certain fraction of the gas escapes combustion in the explosion wave, the temperature of which is probably above that at which the dissociation of steam begins. The greater part of this uncom- bined gas undergoes combustion as a secondary reaction after the passage of the wave, but about 1 per cent, entirely escapes and is found in the tube at the close of the experiment (Dixon). According to Bunsen's experiments the pressure produced when electrolytic gas is exploded is equal to 9*5 atmospheres, and a similar result was subsequently obtained by Berthelot. By comparing this result with the pressure as calculated from the heat of combination, Bunsen concluded that only one-third of the total volume of gas is burnt at the highest temperature of the explosion. Later experiments made with more delicate appliances have shown that the pressures produced in the explosion wave, although of exceedingly short duration, are considerably higher in value than those measured by Bunsen ; the pressure in the case of electrolytic gas for example probably exceeds 20 atmospheres. 132 The Oxyhydrogen Flame. By bringing a jet of oxygen gas within a flame of hydrogen gas, burning from a platinum nozzle, a flame of the mixed gases is obtained which evolves but very little light, although it possesses a very high tempera- ture. A watch-spring held in the flame quickly bums with bright scintillations. Platinum, one of the most infusible of the metals, can be readily melted and even boiled, whilst silver can thus be distilled without difficulty. The arrangement of such an oxyhydrogen blowpipe is seen in Fig. 82, the gases being collected separately in the two gas- holders. The nozzle at s (Fig. 83) is screwed on to the tap of the oxygen gasholder, the points a and b serving to keep the oxygen tube in the centre, whilst the hydrogen enters the tube by the opening w, which is connected with the supply of this gas by a caoutchouc tube. The hydrogen is first turned on and ignited where it issues from the point of the nozzle ; the oxygen tap is then gently turned on so that the flame burns quietly. No backward rush of gas or explosion can here occur, for the gases only mix at the point where combustion takes place. If any solid, infusible and non-volatile substance, such as a piece of quick-lime, be held in the flame, the temperature of the surface of the solid is raised to a very high point and an intense THE NON-METALLIC ELEMENTS white light is emitted, which is frequently used, under the name of the Drummond light, for illuminating purposes. In certain metallurgical processes, especially in working the platinum metals, this high temperature of the oxyhydrogen FIG 82. flame is turned to useful account. One of the forms of furnace used for this purpose is shown in Fig. 84. It is built from a block of very carefully burnt lime A, A, which has been cut in half and then each piece hollowed out, so that when brought FIG. 83. together they form a chamber into which the substance to be melted is placed. The upper block is perforated to allow the nozzle (c, Q) of the blowpipe to fit in, and the gases pass from the separate gasholders, into two concentric OXIDATION AND REDUCTION 269 tubes c c and E f E', each provided with a stopcock (o and H), the hydrogen being delivered by the outer and the oxygen by the inner tube. MM. Deville and Debray l have in this way melted 50 kilos, of platinum in one operation, and Messrs. Johnson, Matthey, & Co. melted, by this process, a mass of pure platinum weighing 100 kilos, which was shown at the Exhi- bition of 1862. Since that time the same firm has melted no less than 250 kilos, of an alloy of platinum and iridium for the International Metrical Commission. 133 Alternate Oxidation and Reduction. An interesting experiment exhibiting the increase and loss of weight on the FIG. 84. oxidation of metallic copper, and the subsequent reduction of the copper oxide is carried out as follows : Copper oxide is rubbed up into a stiff paste with gum-water, and the mass rolled into the form of a cylinder about 1 cm. broad, and 3 cm. long, which is then dried and ignited in the air. On heating this in a current of hydrogen, the oxide is reduced to the metal, and a cylinder of porous copper is obtained. A platinum wire is then wound round it, and the cylinder held in the flame of a Bunsen burner, so as to warm it, but not to bring it to a red heat. On now plunging it into a jar of oxygen gas, it is seen to glow from combination with oxygen, and this continues until 1 Ann. Chim. Phys. [3], 56, 385. 270 THE NON-METALLIC ELEMENTS the whole is converted into oxide. When removed from the oxygen, the colour of the metallic cylinder will be seen to be changed from a bright red to the black colour of the oxide. Next let this oxidised cylinder be suspended from one pan of a balance, and a counterpoise placed in the other pan. The cylinder is then removed from the balance, gently heated, and whilst warm pushed up into a wide tube, placed mouth down- wards, through which a current of hydrogen is passing ; the copper oxide is at once seen to glow, water being formed, which condenses and drops down from the sides of the tube, whilst the colour of the cylinder changes from black to the brilliant red of the reduced metal. As soon as the reduction is complete the cylinder is removed from the hydrogen and again hung on to the pan of the balance, when a considerable loss of weight, due to the loss of the oxygen, will be perceived. PKOPERTIES OF WATER. 134 Pure water is a clear, tasteless liquid, colourless when seen in moderate quantity, but when viewed in bulk possessing a bluish green colour, well seen in the water of certain springs, especially those in Iceland, and in certain lakes, particularly those of Switzerland, which are fed by glacier streams This blue colour is also observed if a bright white object be viewed through a column of distilled water about six to eight metres in length, contained in a tube with blackened sides and plate-glass ends. Water is an almost incompressible fluid, one million volumes becoming less by fifty volumes when the atmospheric pressure is doubled ; it is a bad conductor of heat, and probably a non-conductor of electricity. Expansion and Contraction of Water. When heated from to 4, water is found to contract, thus forming a striking exception to the general law, that bodies expand when heated and contract on cooling; on cooling from 4 to it expands again. Above 4, however, it follows the ordinary law, expand- ing when heated, and contracting when cooled. This pecu- liarity in the expansion and contraction of water may be expressed by saying that the point of maximum density of water is 4 C. ; or according to the exact determinations of Joule, 3'945 ; that is, a given bulk of water will at this temperature PROPERTIES OF WATER 271 weigh more than at any other. Although the amount of con- traction on heating from to 4 is but small, yet it exerts a most important influence upon the economy of nature. If it were not for this apparently unimportant property, our climate would be perfectly Arctic, and Europe would in all probability be as uninhabitable as Melville Island. In order better to understand what the state of things would be if water followed the ordinary laws of expansion by heat, we may perform the following experiment, first made by Dr. Hope. Take a jar con- taining water at a temperature above 4, place one thermometer at the top and another at the bottom of the liquid. Now bring the jar into a place where the temperature is below the freezing point, and observe the temperature at the top and bottom of the liquid as it cools. It will be seen that at first the upper ther- mometer always indicates a higher temperature than the lower one ; after a short time both thermometers mark 4 ; and, as the water cools still further, it will be seen that the thermometer at the top always indicates a lower temperature than that shown by the one at the bottom : hence we conclude that water above or below 4 is lighter than water at 4. This cooling goes on till the temperature of the top layer of water sinks to 0, after which a crust of ice is formed ; and if the mass of the water be sufficiently large, the temperature of the water at the bottom is never reduced below 4. In nature precisely the same phe- nomenon occurs in the freezing of lakes and rivers ; * the surface- water is gradually cooled by cold winds, and thus becoming heavier, sinks, whilst lighter and warmer water rises to supply its place : this goes on till the temperature of the whole mass is reduced to 4, after which the surface-water never sinks, however much it be cooled, as it is always lighter than the deeper water at 4. Hence ice is formed only at the top, the mass of water retaining the temperature of 4. Had water become heavier as it cooled down to the freezing point, a continual circulation would be kept up, until the mass was cooled throughout to 0, when solidi- fication of the whole would ensue. Thus our lakes and rivers would be converted into solid masses of ice, which the summer's warmth would be quite insufficient thoroughly to melt ; and hence the climate of our now temperate zone might approach in severity that of the Arctic regions. The following table gives the volume, and specific gravities 1 The point of maximum density of sea-water is considerably lower than that of fresh, and is in fact below C. 272 THE NON-METALLIC ELEMENTS of water for temperatures varying from to 100, according to the most accurate experiments. 1 Tempe- rature. Volume. Specific Gravity. Tempe- rature. Volume. Specific Gravity. 1-000122 0-999878 19 1-00153 0-99847 1 1-000067 0-999933 20 1-00173 0-99827 2 1-000028 0-999972 21 1-00194 0-99806 3 1-000007 0-999993 22 1-00216 0-99785 4 1-000000 1-000000 23 1-00238 0-99762 5 1-000008 0-999992 24 1-00262 0-99739 6 1-000031 0-999969 25 1-00287 0-99714 7 1-000067 0-999933 30 1-00425 0-99577 8 1-000118 0-999882 35 1-00586 0-99417 9 1-000181 0-999819 40 1-00770 0-99236 10 1-000261 0-999739 45 1-00974 0-29035 11 1-000350 0-999650 50 1-01197 0-98817 12 1-000456 0-999544 55 1-01436 0-98584 13 1-000570 0-999430 60 1-01694 0-98334 14 1-000703 0-999297 65 1-01967 0-98071 15 1-000847 0-999154 70 1-02261 0-97789 , 16 1-000997 0-999004 80 1-02891 0-97190 17 1-001162 0-998839 90 1-03574 0-96549 18 1-001339 0-998663 100 1-04323 0-95856 , 135 Latent Heat of Water. In the passage from solid ice to liquid water, we notice that a very remarkable absorption or disappearance of heat occurs. This is rendered plain by the following simple experiment : Let us take a kilogram of water at the temperature 0, and another kilogram of water at 79. If we mix these, the temperature of the mixture will be the mean, or 39'5 ; if, however, we take one kilogram of ice at and mix it with a kilogram of water at 79, we shall find that the whole of the ice is melted, but that the temperature of the resulting 2 kilograms of water is exactly 0. In other words, the whole of the heat lost by the hot water has just sufficed to melt the ice, but has not raised the temperature of the water thus produced. Hence we see that in passing from the solid to the liquid state a given weight of water takes up or renders latent just so much heat as would suffice to raise the temperature of the same weight of water through 79 C. ; the latent heat of water 1 Volkmann, Wied. Ann. 14, 260. PROPERTIES OF WATER 273 is, therefore, said to be 79 thermal units a thermal unit mean- ing the amount of heat required to raise a unit weight of water through 1 C. When water freezes, or becomes solid, this amount of heat, which is necessary to keep the water in the liquid form, and is, therefore, well termed the heat of liquidity, is evolved, or rendered sensible. A similar disappearance of heat on passing from the solid to the liquid state, and a similar evolution of heat on passing from the liquid to the solid form, occurs with all substances ; the amount of heat thus evolved or rendered latent varies, however, with the nature of the substance. A simple means of showing that heat is evolved on solidification consists in obtaining a saturated hot solution of acetate of soda, and allowing it to cool. Whilst it remains undisturbed, it retains the liquid form, but if agitated, it at once begins to crystallize, and in a few moments becomes a solid mass. If a delicate thermometer be now plunged into the salt while solidi- fying, a rapid rise of temperature will be noticed. 136 Freezing Point of Water and Melting Point of Ice. Although water usually freezes at it was observed so long ago as 1714 by Fahrenheit that under certain circumstances water may remain liquid at temperatures much below this point. Thus when brought under a diminished atmospheric pressure, water may be cooled to 12 without freezing ; or if water be boiled in a glass flask, and the neck of the flask be plugged whilst it is hot with cotton wool, the flask and its contents may be cooled to 9 without the water freezing, but when the cotton wool is taken out, particles of dust fall into the water, and these bring about an immediate crystallisation, the tem- perature of the mass quickly rising to 0. Sorby 1 has shown that when contained in thin capillary glass tubes, water may be cooled to 15 without freezing, whilst Boussingault 2 has exposed water contained in a closed steel cylinder to a tem- perature of 24 for several days in succession without its freezing. The melting point of ice under the ordinary atmo- spheric pressure is 0, but this point is lowered by increase of pressure ; thus under a pressure of 81 atmospheres, ice melts at 0'059 and under 16'8 atmospheres, at 0'129 or the melting point is lowered by about 0'0075 3 for every additional atmo- sphere. This peculiarity of a lowering of the melting point under pressure is common to all substances which, like water 1 Phil. Mag. [4], 18, 105. 2 Compt. Rend. 73, 77. 3 James Thomson, Edin. Roy. Soc. Trans, vol. 16, 575, 1849. 19 274 THE NON-METALLIC ELEMENTS expand in passing from the liquid to the solid state, whilst in the case of bodies which contract under like circumstances, the melting point is raised by increase of pressure. Thus Bunsen in the case of paraffin, 1 and Hopkins in the case of sulphur, obtained the following results : Under a pressure of 1 atmosphere, paraffin melts at 46*3 85 48-9 100 49-9 1 atmosphere, sulphur melts at 1 07'0 519 135-2 792 140- 5 From what has been stated we should expect that by increas- ing the pressure upon ice it could be melted, and Mousson 2 has shown that this is the case, for by exposing it to a pressure of 13,000 atmospheres he has converted ice into water at a tem- perature of 18. This lowering of the melting point of ice with pressure explains the fact that when two pieces of ice are rubbed together the pressure causes the ice to melt at the portions of the surface in contact, the water thus formed running away, and the temperature being lowered ; then as soon as the excess of pressure is taken away the two surfaces freeze together at a temperature below 0, one mass of solid ice being produced. This phenomenon, termed regelation, was first observed by Fara- day in 1850, and was afterwards applied by Tyndall to explain glacier motion. 137 The crystalline form of ice is hexagonal, being that of a rhombohedron. Snow crystals exhibit this hexagonal form very clearly ; they usually consist of crystals which have grown on to another crystal in the direction of the three horizontal axes, that so the snow crystal clearly exhibits these three directions, as shown in Figs. 85, 86. Ice is transparent, and when seen in small quantities it ap- pears to be colourless, though large masses of ice, such as icebergs or glaciers, possess a deep blue colour ; like water it is also a bad conductor of heat and a non-conductor of electricity, and becomes electrical when rubbed. Water on freezing increases nearly y 1 ^- of its bulk, or, according to the exact experiments of Bunsen, 3 the specific gravity of ice 1 Ann. Chim. Phys. [3], 35, 383. - Pogg. Ann. 105, 161. 3 Phil. Mag. [4], 41, 165. PROPERTIES OF WATE11 275 at is 0*91674, that of water at being taken as the unit ; or one volume of water at becomes 1*09082 volumes of ice at the same temperature. This expansion plays an important part in the disintegration and splitting of rocks during the winter. Water penetrates into the cracks and crevices of the rocks and on freezing widens these openings; this process being repeated over and over again, the rock is ultimately split into fragments. Hollow balls of thick cast-iron can thus easily be split in two by filling them with water and closing by a tightly fitting screw, and then exposing them to a temperature below 0. FIG. 85. FIG. 86. 138 Latent Heat of Steam. Under the normal barometric pressure of 760 rom water boils in a metal vessel at 100 C. When liquid water is converted ir to gaseous steam, a large quantity of heat becomes latent, the temperature of the steam given off being the same as that of the boiling water, as, like all other bodies, water requires more heat for its existence as a gas than as a liquid. The amount of heat latent in steam, is roughly ascertained by the following experiment. Into 1 kilogram of water at 0, steam from boiling water, having the temperature of 100, is passed until the water boils : it is then found that the whole weighs T187 kilos., or 0'187 kilo, of water in the form of steam at 100 has raised 1 kilo, of water from to 100 ; or 1 kilo, of steam at 100 would raise 5 '36 kilos, of ice-cold water 276 THE NON-METALLIC ELEMENTS through 100, or 536 kilos, through 1. Hence the latent heat of steam is said to be 536 thermal units. Whenever water evaporates or passes into the gaseous state, heat is absorbed, and so much heat may be thus abstracted from water that it may be made to freeze by its own evaporation. A beautiful illustration of this is found in an instrument called Wollaston's Cryophorus, Fig. 87 ; it consists of a bent tube, having a bulb on each end, and containing water and vapour of water, but no air. On placing all the water in one bulb, and plunging the empty bulb into a freezing mixture, a condensation of the vapour of water in this empty bulb occurs, and a corre- sponding quantity of water evaporates from the other bulb to supply the place of the condensed vapour ; this condensation and evaporation go on so rapidly that in a short time the water cools down below 0, and a solid mass of ice is left in the bulb. By a very ingenious arrangement this plan of freezing water by its own evaporation has been practically carried out on a large FIG. 87. scale by M. Carre", by means of which ice can be most easily prepared. This arrangement consists simply of a powerful air-pump (A, Fig. 88), and a reservoir (B), of a hygroscopic substance, such as strong sulphuric acid. On placing a bottle of water (c) in connection with this apparatus, and { on pumping for a few minutes, the water begins to boil rapidly, and its temperature is so lowered by the evaporation that the water freezes to a mass of ice. 139 Tension of Aqueous Vapour. Water, and even ice, con- stantly give off steam or aqueous vapour at all temperatures, when exposed to the air. Thus we know that if a glass of water be left in a room for some days, the whole of the water will gradually evaporate. This power of water to rise in vapour at all temperatures is due to the elastic force or tension of aqueous vapour ; it may be measured, when a small quantity of water is placed above the mercury in a barometer, by the extent to which the vapour thus given off is capable . of PKOPERTIES OF WATER 277 depressing the mercurial column. If we gradually heat the drops of water thus placed in the barometer, we shall notice that the column of mercury gradually sinks ; and when the water is heated up to 100 C., the mercury in the barometer tube is found to stand at the same level as that in the trough, showing that the elastic force of the vapour at that tem- perature is equal to the atmospheric pressure. Water is said to boil in the air when the tension of its vapour is equal to the superincumbent atmospheric pressure. On the tops of mountains, where the atmospheric pressure is less than at the sea's level, water boils at a temperature below 100 : thus at Quito, at a height of 2914 metres above the sea's level, the mean height of FIG. 88. the barometer is 523 mm., and the boiling point of water is 90'l ; that is, the tension of aqueous vapour at 90'l is equal to the pressure exerted by a column of mercury 523 mm. high. Founded on this principle, an instrument has been constructed for determining heights by noticing the temperatures at which water boils. A simple experiment to illustrate this fact consists in boiling water in a globular flask, into the neck of which a stopcock is fitted : as soon as the air is expelled, the stopcock is closed, and the flask removed from the source of heat ; the boiling then ceases ; but on immersing the flask in cold water, the ebullition recommences briskly, owing to the reduction of the pressure consequent upon the condensation of the steam ; the tension of the vapour at the temperature of the water in 278 THE NON-METALLIC ELEMENTS the flask being greater than the diminished pressure. All other liquids follow a similar law respecting ebullition ; but as the tensions of their vapours are very different, their boiling points vary considerably. When steam is heated alone, it expands according to the law previously given for permanent gases ; but when water is present, and the experiment is performed in a closed vessel, the elastic force of the steam increases in a far more rapid ratio than the increase of temperature. The following table gives the tension of aqueous vapour, as determined by experi- ment, at different temperatures measured on the air ther- mometer. Table of the Tension of the Vapour of Water. Temperature Centigrade. Tension in millime- tres of mercury. Temperature Centigrade. Tension in atmosphs., 1 atmosphere = 760 mm. of mercury. -20 0-927 100 1 - 10 2-093 111-7 1-5 4-600 120-6 2 + 5 6-534 127-8 2-5 10 9165 133-9 3 15 12-699 144-0 4 20 17-391 159-2 6 30 31-548 170-8 8 40 54-906 180-3 10 50 91-982 188-4 12 60 148-791 195-5 14 70 233-093 201-9 16 80 354-280 207-7 18 90 525-450 213-0 20 100 760-000 224-7 25 140 Water as a Solvent. Water is the most generally valuable of known solvents. Not only do many solids, such as sugar and salt, dissolve in water, but certain liquids, such as alcohol and acetic acid, mix with it completely. Other liquids again, such as ether, dissolve to a certain extent in water, although they do not mix with it in all proportions. Gases also dissolve in water, some, such as ammonia and hydrochloric acid, in very large quantities, exceeding more than one hundred times the bulk of the water ; others, again, such as hydrogen and nitrogen, WATER OF CRYSTALLIZATION 279 are but very slightly soluble, while carbon dioxide and some other gases stand, as regards solubility, between these extremes. Concerning the nature of solution, whether of solids, liquids, or gases, we know at present but little. The phenomena of solution differ, however, essentially from those of chemical combination, inasmuch as in the former we have to do with gradual increase up to a given limit, termed the point of saturation, whereas in the latter we observe the occurrence of constant proportions in which, and in no others, combination occurs. Solution follows a law of continuity, chemical com- bination one of sudden change or discontinuity. The solubility of solids varies with the essential nature of the solid, with that of the liquid, and with the temperature at which they are brought together ; the same may be said of the solvent action of water upon liquids and upon gases, except that the solubility of gases is also influenced by the pressure to which the gas and the water are subjected. The quantity of any solid, liquid, or gas which dissolves in a solvent, such as water, must be ascertained empirically in every case, as we are unacquainted with any law according to which such solvent action takes place, and we, therefore, cannot calculate the amount. The effect of change of temperature on the solubility on a substance, whether solid, liquid, or gaseous must likewise be determined by experi- ment, but the effect of pressure upon the solubility of gases is given by a simple law, known as the law of Dalton and Henry (p. 284). The solubility of most solids increases with the temperature, a limit being reached at each temperature, beyond which no further quantity of the solid dissolves. When the temperature of such saturated solutions falls, or when the solvent is allowed to evaporate, a portion of the dissolved substance is deposited from solution usually in the form of a solid possessing some definite geometrical form, and termed a crystal, whilst the substance is said to crystallize. Water of Crystallization. Many salts owe their crystalline character to the presence in a solid state of a certain definite number of molecules of water. When this chemically combined water is driven off by heat, the crystal falls to powder, and hence it has been termed water of crystallization. Some salts contain a large quantity of water definitely combined in this form ; thus the opaque white powder of anhydrous alum, K 2 A1 2 (SO 4 ) 4 , unites with no less than twenty-four molecules of 280 THE NON-METALLIC ELEMENTS water to form the well-known transparent octohedral crystals of common alum, K 2 AL,(SO 4 ) 4 + 24H 2 O ; in like manner anhydrous and powdery carbonate of soda, Na 2 CO 3 , when dissolved in water deposits large monosymmetric crystals of common wash- ing-soda, having the composition Na. 2 CO 3 +10 H 2 O. The tem- perature of the solution from which such crystals are deposited materially affects the quantity of water with which the salt combines ; thus, in the case of carbonate of soda, whilst mono- symmetric crystals of the ten-molecule hydrate are deposited at the ordinary temperature, other crystals, having the composition Na 2 CO 3 + 8H 2 O, or again others represented by the formula Na 2 CO 3 + 5H 2 O, are deposited, when crystallization is allowed to take place at higher temperatures. The water in these crystals has a definite vapour tension and is, therefore, given off when the temperature becomes so high that the tension of the water of crystallization is greater than that in the surrounding atmosphere. Different substances lose their water in the air at very different temperatures, and even the molecules of water combined with a single molecule of salt behave differently in this respect. Thus potash alum loses ten molecules of water at 100, but it needs to be heated to 120 in order to drive off a second ten molecules of water, and retains the last four molecules until the temperature rises to 200. Copper sulphate in a similar manner loses four of its molecules of water below 110, whilst the fifth is only driven off at 200. Sodium carbonate, Na 2 CO 3 + 10H 2 O, loses water on simple exposure to the air, the tension of its combined water being greater than that of the aqueous vapour in the atmosphere, and the salt becomes covered with a white powder. Crystals which behave in this manner are said to effloresce. Many salts which do not lose water in the atmosphere do so when placed in dry air. Other solid salts, such as calcium chloride, and potassium acetate, combine with water with such avidity that when left exposed to the air they begin to liquefy from absorption of the atmo- spheric moisture ; the salts are then said to deliquesce. In the year 1840 Dalton observed that different salts, whose water of crystallization has been driven off by heat, dissolve in water without increasing the volume of the liquid, whereas if the hydrated salt is dissolved, an increase of volume occurs which is exactly that due to the water which is combined in the salt. Playfair and Joule l extended these observations, showing 1 Chem Soc. Mem. 2, 477 ; 3, 54, 199 ; Chem. Soc. Quart. Journ. 1, 121. CRYOHYDRATES 281 for instance, that in the case of carbonate of soda, crystallizing with ten molecules of water, and in that of the phosphates and arsenates crystallizing with twelve molecules, the volume of the whole molecule of hydrated salt is the same as that of its water of crystallization would be if frozen to ice. The particles of anhydrous salt would hence appear to occupy the spaces inter- vening between those of the water without increasing its volume. Thus the crystals of common washing soda have the following composition : Na 2 CO 3 = 104-29 10H 2 6 = 178-80 283-09 and 283'09 grams, of these crystals occupy exactly the space of 178'8 grams, of ice. The following table gives the specific gravities of the above mentioned salts, first as observed by experiment, and secondly as calculated upon the above hypothesis, and shows the close agreement of the two sets of numbers. Specific Gravity. Observed. Calculated. Sodium Carbonate . . . . Na 2 C0 3 + 10 H 2 . . 1'454 . . 1-463 Hydrogen Sodium Phosphate Na 2 HP0 4 + 12 H 2 . . 1'525 . . 1'527 Normal Sodium Phosphate . Na 3 P0 4 + 12 H 2 . . 1'622 . . 1'622 Hydrogen Sodium Arsenate . Na 2 H As0 4 + 12 H 2 . 1736 . . 1736 Normal Sodium Arsenate . Na 3 As0 4 + 12 H 2 . . 1'804 . . 1*834 In the case of certain other salts the volume of the crystal was found to be equal to the sum of the volume of the water when frozen and that of the anhydrous salt. 141 Cryohydrates. It has been already mentioned that when a dilute solution is cooled, ice separates out (p. 11 4). If the ice be removed and the cooling continued, a temperature is at length reached at which the whole solution becomes solid. Thus Guthrie, 1 who has investigated this subject, found that when a dilute solution of common salt is cooled down to 1'5, ice begins to separate out, and this formation of ice continues until the temperature sinks to 23, at which the whole mass becomes solid. A concentrated salt solution, on the other hand, deposits at 7 crystals having the composition NaCl + 2H 2 O, and the separation of this compound, in the form of iridescent scales, also goes on until the liquid has cooled down to 23 when, as before, it freezes en masse. 1 Phil. Mag. [4], 49, 1, 206, 266. 282 THE NON-METALLIC ELEMENTS The solid mass thus formed has been called a cryohydrate, and resembles a chemical compound inasmuch as it has a definite composition and a definite melting-point. It has, however, been shown that these so-called cryohydrates are simply mixtures of ice and a solid salt, and that their properties are in all cases the mean of those of their constituents. 1 142 Freezing Mixtures. The solution of a solid in water is generally accompanied by a lowering of temperature, caused by the conversion of sensible into latent heat by the liquefaction of the solid. In the case, however, of 'many anhydrous salts, solution is accompanied by a rise in temperature, which may possibly be caused by the production of a definite chemical compound between the solid and the solvent. By the solution of many salts such a diminution of temperature is effected that this process may be used for obtaining ice ; thus when 500 grams, of potassium thiocyanate are dissolved in 400 grams, of cold water, the temperature of the solution sinks to 20. When common salt is mixed with snow or pounded ice a con- siderable reduction of the temperature of the mass occurs, the two solid bodies becoming liquid and forming a concentrated brine whose freezing-point lies at 23. This solution contains thirty-two parts by weight of salt to 100 parts of water, and in order to bring about the greatest possible reduction in tempera- ture the salt and snow must be mixed in the above proportions. Equal weights of crystallized calcium chloride and snow when mixed together give a freezing mixture whose temperature sinks from to - 45. The temperatures produced in this way are those at which the so-called cryohydrates are formed, since these are the lowest temperatures at which the mass can remain liquid. The Properties of Solutions. Concerning the relation between the dissolved substance and the solvent in a concentrated solu- tion, but little is known. Many such solutions, when cooled, deposit crystals which contain a portion of the solvent combined with the substance which has been dissolved. It has, therefore, been supposed that the solutions in these cases actually contain these compounds, and this seems to be probable at all events at low temperatures. On the other hand, it has been held that no compounds are present in the solutions, but that the com- bination takes place at the moment of separation of the solid substance. 1 Offer, Wien. Akad. Ber. 81, ii. 1058. PROPERTIES OF SOLUTIONS 283 The properties of dilute solutions can be to a large extent explained, as already described (p. Ill), on the supposition that the dissolved substance is not combined with the solvent. In the case of solutions of salts (electrolytes) in water a special hypothesis, that of electrolytic dissociation, has been introduced, according to which the salt is to a large extent split up in the solution into its electrolytic ions. This view is rendered probable not only by the facts already adduced (p. 116), but by the circumstance that almost all the properties of such solutions have been shown to be equal to the sum of the properties of the ions. Each ion has a definite weight, volume, colour, &c., and the effect produced by a salt is found to be equal to the sum of the effects which would be independently produced by the ions into which it is capable of being decom- posed. This has already been explained for the osmotic pressure, freezing-point and vapour pressure of solutions, and also holds for the colour, specific gravity, refractive index, &c. This theory is, therefore, in accordance with the physical properties of such solutions, and it moreover throws great light upon many chemical reactions. Chemical action occurring in dilute solution, according to this view, takes place between the ions of the substances concerned, so that the tests usually applied for the metals and acids are in reality tests for the corresponding ions. Thus, for example, the tests for ferrous salts fail to detect iron in potassium ferrocyanide, K 4 FeC 6 N 6 , although they are given by ferrous sulphate, FeSO 4 . This is due to the fact that in the latter case the ion of ferrous iron is present in the solution, whilst in the former the ions are K and the complex group FeC 6 N 6 . The fact that the same amount of heat is evolved by the neutralisation of any of the strong acids by any of the strong bases also receives a new and interesting interpretation in the light of this theory. Taking the case of the action of hydro- chloric acid on caustic soda, which is usually represented by the equation, NaHO + HC1 = NaCl + H 2 O, we see that the action takes place between the ions in the following way : Na + HO + H + Cl = Na + Cl + H 2 O. 284 THE NON-METALLIC ELEMENTS The result of the reaction is simply the formation of water, since the Na and Cl ions remain dissociated. Exactly the same reaction takes place if another acid and a different base be employed : + - + - + K + HO + H + N0 3 = K + N0 3 + H 2 O. The amount of heat evolved is, therefore, also the same. 143 Absorption of Gases by Water. All gases are soluble to a greater or less degree in water, the extent of this solubility depending upon (1) the nature of the gas, (2) the temperature of the gas and water, (3) the pressure under which the absorption occurs. No simple law is known expressing the relation between the amount of gas absorbed and the temperature. Usually the solubility of a gas diminishes as the temperature increases, but the rate of diminution varies with each gas, so that the amount of gas dissolved in water at a given temperature can be ascer- tained only by experiment. A simple relation has however been found to exist between the quantity of gas absorbed under varying conditions of pressure, the temperature remaining constant. In the year 1803 William Henry * proved that the amount of gas absorbed by water varies directly as the pressure, or, in the words of the discoverer of the law, " under equal circumstances of temperature, water takes up in all cases the same volume of condensed gas as of gas under ordinary pressure. But as the spaces occupied by every gas are inversely as the compressing force, it follows that water takes up of gas condensed by one, two, or more additional atmospheres, a quantity which ordinarily compressed, would be equal to twice, thrice, and so on, the volume absorbed under the common pressure of the atmosphere." Two years after Henry had enunciated this law, Dalton 2 extended the law to the case of mixed gases, proving that when a mixture of two or more gases in given proportions is shaken up with water, the volume of the gas having a finite relation to that of the liquid, the absorptiometric equilibrium occurs when the pressure of each gas dissolved in the liquid is equal to that of the portion of the gas which remains unabsorbed by the liquid ; the amount of each gas absorbed by water from such a mixture being solely dependent on the pressure exerted by the particular gas. This law, termed Dalton 's law of partial pressures. 1 Phil. Trans. 1803, 29, 274. 2 Mane. Memoirs, 1805. SOLUTION OF GASES IN WATER 285 may be illustrated by the following example : if two or more gases which do not act chemically upon each other be mixed together, and the mixture of gases brought into contact with water until the absorptiometric equilibrium is established, the quantity of each gas which dissolves is exactly what it would have been if only the one gas had been present in the space. Thus, for instance, the absorption co-efficient of oxygen at is 0*04890, that of nitrogen at the same temperature being 0*023481. Now 100 volumes of air contain on an average 79*04 volumes of nitrogen and 20'96 volumes of oxygen, hence the partial pressure of the oxygen is 0*2096 of an atmosphere, whilst that of the nitrogen is 07904, and as the solubility of each gas is propor- tioned to its partial pressure, 0*2096 X 0*04890 = 0*010244 will be the proportion of oxygen dissolved, and 0*7904 x 0*023481 = 0*018559 will be the proportion of the nitrogen dissolved, or the per- centage composition of the air dissolved in water will be : Calculated. Found. Nitrogen . . . 64*4 . . . 64*9 Oxygen . . . 35*6 . . . 35*1 100*0 lOO'O Thus the relation between the dissolved gases as found by ex- periment agrees closely with that calculated on the above assumption, and the law of partial pressures is verified. Every absorbed gas which follows the law of pressure will of course be driven out of solution when the pressure on the gas is reduced to zero. This can be effected by removing the super- incumbent pressure by means of an air-pump, by allowing the liquid to come in contact with a very large volume of some other indifferent gas, or, lastly, by boiling the liquid when all the dissolved gas will be driven off with the issuing steam, except where a chemical combination or attraction exists between the gas and the water. Some gases dissolve in water in very large quantities, whereas ad Water. Parts per Million. Grains pel- Gallon. Parts per Million. Grains per Gallon. Total solids .... Nitrogen as nitrites and nitrates 63-0 0*25 4-4 0-017 530 7'8 37-1 0-546 Free ammonia .... Albuminoid ammonia . Chlorine 0-03 0-07 11-4 0-002 0-005 0-8 4-32 0-9 69 0-303 0-063 4-8 Temporary hardness . Permanent hardness. Total hardness o-i 2-4 2-5 7-2 14-4 21-6 1 For the special details of the processes of water analysis, the following works or memoirs may be consulted : Water Analysis, by Wanklyn and Chapman. 1889. Triibner, London. Frankland and Armstrong, Journ. Chem. Soc. 1868. p. 77 ; Frankland, ibid. 109 ; also Journ. Chem. Soc. June, 1876. Percy Frank- land, Agricultural Chemical Analysis, p. 257. Macmillan, 1883. RIVER-WATERS 305 154 River- Waters. The composition of river- water varies considerably with the nature of the ground over which the water runs; thus Thames water contains about 11 grains per gallon of carbonate of lime ; the Trent 21 grains of sulphate of lime, or they are both hard waters, the first temporarily and the second permanently hard. The waters of the Dee and the Don, in Aberdeenshire, draining a granite district, are, on the other hand, soft waters. The composition of these waters is shown in the following table : TABLE GIVING THE COMPOSITION OF CERTAIN RIVER- WATERS. Grains per Gallon. Calcium carbonate . . . Calcium sulphate . . . Calcium nitrate .... Magnesium carbonate . Sodium chloride .... Silica . .... Thames. 10-80 3-00 017 1-25 1-80 0-56 0-27 trace 2-36 Trent. 0-32 21-55 5-66 17-63 0-72 0-50 trace 3-68 Dee. 0-85 0'12 0-36 0-72 0-14 0-06 trace 1-54 Don. 2-23 0-13 1-07 1-26 0-52 0-27 trace 3-06 Ferric chloride and alu- mina Calcium phosphate . . . Organic matter .... Hardness 20-21 50-06 3-89 8-54 14-0 26-5 1-5 3-0 Unfortunately in England, as in other manufacturing and densely populated countries, the running water seldom reaches the sea in its natural or pure state, but is largely contaminated with the sewage of towns, or the refuse from manufactures or mines. So serious indeed is this state of things that steps have been taken to prevent the further pollution of the rivers of the country, and a Royal Commission has already reported and several acts of Parliament have been passed with a view of preventing the evil. The following analyses of the composition of Lancashire rivers, taken from the First Report of the Com- missioners l appointed in 1868, show clearly the pollution which the originally pure waters of the Irwell and Mersey undergo on flowing down to the sea, 1 P. 15. 21 306 THE NON-METALLIC ELEMENTS COMPOSITION OF LANCASHIRE RIVERS. Parts in 100,000. IRWELL. MERSEY. n | , 3 4 Total soluble solids . . . 7-8 55-80 7-62 39-50 Organic carbon .... 0-187 1-173 0-222 1-231 Organic nitrogen 0-025 0-332 0-601 Ammonia 0-004 0-740 0-002 0-622 Nitrogen as nitrates and nitrites 0-021 0-707 0-021 Total combined nitrogen 0-049 1-648 0-023 1-113 Chlorine 115 9-63 0-94 Hardness temporary . . 3-72 15-04 4-61 10-18 Total hardness .... 3-72 15-04 4-61 10-18 SUSPENDED MATTER Organic 2-71 Mineral 2-71 Total 5-42 From these numbers it is seen that the quantities of free ammonia and nitric acid become increased 300 or 400-fold in the river below Manchester, whilst the total combined nitrogen is increased from 0'049 to 1"648. 155 Sea- Water. The amount of solid matter contained in the waters of the ocean is remarkably constant when collected far from land. The mean quantity is about 35*976 grams in 1000 grams of sea- water; the average specific gravity of sea- water is 1-02975 at 0. h l. The Irwell near its source. 2. The Irwell below Manchester. 3. The Mersey, one of its sources. 4. The Mersey below Stockport. SEA-WATER 307 COMPOSITION OF THE WATER OF THE IRISH SEA IN THE SUMMER OF 1870. 1 One thousand grams of sea-water contain Grams. Sodium chloride Potassium chloride Magnesium chloride 26-43918 0-74619 3-15083 Magnesium bromide 0-07052 Magnesium sulphate Magnesium carbonate .... Magnesium nitrate Calcium sulphate 2 06608 traces 0-00207 1-33158 Calcium carbonate Lithium chloride . 0-04754 traces Ammonium chloride Ferrous carbonate Silicic acid 0-00044 0-00503 traces 33-85946 For the purpose of controlling the analysis, 1,000 grams of water were evaporated to dryness, and the dry residue weighed. Its weight was found to be 33"83855 grams. The specific gravity of the water at C. was 1*02721, whilst that at 15 C. was 1-02484. Forchhammer found that 1,000 parts by weight of the water of the mid-Atlantic Ocean contained 35*976 parts of dissolved salts, whilst the mean, of analyses of sea-water from different localities gave 34*082 for the total salts in summer and 33*838 in winter. Dittmar, 2 on the other hand, from 77 specimens of sea-water collected on board the Challenger in various parts of the world, concludes that the maximum quantity of salt contained in the water of the Indian Ocean is 33'01, and in that of the North Atlantic 37 '37. In the neighbourhood of the shore or in narrow straits the quantity of saline matter is often much smaller. According to Forchhammer the relation in which the several 1 Thorpe and Morton, Journ. Chem. Soc. 24, 506. ~ Challenger Reports, "On the Composition of Ocean Water." By Professor Dittmar, F.R.S. London, 1884. 308 THE NON-METALLIC ELEMENTS salts stand to one another is a constant one ; and in this conclusion Dittmar agrees. The former calculated the quantity of lime, magnesia, potash, and sulphuric acid present with 100 parts of chlorine, including bromine. Dittmar estimated this element and also soda and carbon dioxide, and then calculated as Forchhammer, reckoning the bromine as its equivalent of chlorine. Their results are : Forchhammer. Cl . . . 100-00 . Br . . . S0 3 . . 11-88 . C0 2 . . - . Dittmar calculated the following average composition of the total saline constituents of sea- water, giving a somewhat different arrangement to the acids and bases from that adopted by Thorpe and Morton. Dittmar. Forchhammer. Dittmar. 99-848 CaO. . . 2-93 . . 3-026 0-3 MgO . . 11-03 . . 11-221 11-576 K 2 . 1-93 . . 2-405 0-276 Na 2 . . . 74-462 Sodium chloride Magnesium chloride Magnesium sulphate Calcium sulphate . Potassium sulphate 77*758 Magnesium bromide . . 0'217 10-878 Calcium carbonate . . . 0*345 4-737 3-600 100-000 2'465 All the elements are doubtless contained in sea-water. In addition to those named above the following have been detected : Iodine, fluorine, nitrogen, phosphorus, silicon, carbon, boron, zinc, cobalt, nickel, copper, strontium, barium, manganese, aluminium, iron, lithium, caesium, rubidium (these three detected spectroscopically), silver, lead, and lastly arsenic, making a total of thirty elements. HYDROGEN DIOXIDE, OR HYDROGEN PEROXIDE. H 2 O 2 = 3376. 156 This body was discovered in 1818 by Thenard, 1 who prepared it, by the action of dilute hydrochloric acid on barium dioxide, thus : Ba0 2 + 2HC1 = H 2 O 2 + BaCl 2 The compound is also easily formed by passing a current of carbon dioxide through water, and gradually adding barium dioxide in very small quantities, 2 thus : Ba0 2 + C0 2 + H 2 = H 2 0, + BaCO 3 . 1 Ann. Ghem. Phys. 8, 306. 2 Duprey, Compt. Rend. 55, 736 ; Balard, Compt. Rend. 55, 758. HYDROGEN PEROXIDE 309 Preparation. Hydrogen dioxide is, however, most generally obtained by decomposing pure barium dioxide with dilute sulphuric acid ; l thus : Ba0 2 + H 2 S0 4 = H 2 2 + BaSO 4 . The pure barium dioxide needed for these experiments is prepared as follows : commercial barium dioxide, very finely powdered, is brought little by little into dilute hydrochloric acid, until the acid is nearly neutralised. The cooled and filtered solution is then treated with baryta- water, in order to precipitate the ferric oxide, manganic oxide, alumina, and silica which are always contained in the impure barium dioxide. As soon as a white precipitate of the hydrated barium dioxide makes its appearance, the solution is filtered, and to the filtrate concen- trated baryta-water is added ; a crystalline precipitate then falls consisting of hydrated barium dioxide. This is well washed and preserved, in the moist state, in stoppered bottles. In order to prepare hydrogen dioxide by means of this substance, the moist precipitate is gradually added to a cold mixture of not less than five parts of water to one part of concentrated sulphuric acid, until the mixture remains very slightly acid. The preci- pitate of barium sulphate is allowed to settle and the liquid filtered. The small trace of sulphuric acid which the filtrate contains can be precipitated by careful addition of dilute baryta solution. When the aqueous solution of the hydrogen dioxide thus prepared is brought over sulphuric acid in vacua, water evaporates, and the solution of the dioxide becomes more con- centrated. If, during the concentration, an evolution of oxygen be noticed, a drop or two of sulphuric acid should be added, as the dioxide is more stable in presence of free acid than when perfectly pure (Thenard). An aqueous solution of the peroxide can readily be prepared for use in cases in which the presence of a soluble sodium salt does not interfere, by dissolving sodium peroxide, Na 2 O , a compound now prepared on a large scale, in cold water and adding a dilute acid. Properties. After the slow evaporation has been carried on for some time a colourless transparent oily liquid is obtained, having a specific gravity of 1/452. This liquid evaporates slowly in vacuo without the residue undergoing any change 1 Thomsen, Ber. 7, 74. 310 THE NON-METALLIC ELEMENTS (Thenard). Hydrogen dioxide does not solidify at 30 ; it possesses no smell, has an astringent, bitter taste, and brought on to the skin produces a white blister, which after a time produces great irritation. It bleaches organic colouring matters like chlorine, although acting more slowly, and is a very unstable compound, easily decomposing into oxygen and water. One volume of the liquid at 14 and under a pressure of 760mm. yields, according to Thenard, 475 volumes of oxygen, whilst the theoretical amount is 5 01 '8 volumes. The decomposition of hydrogen dioxide takes place very slowly at a low temperature, whilst at 20 the evolution of gas becomes plainly visible, and if the concentrated solution is heated up to 100, the separa- tion of oxygen occurs so rapidly as sometimes to give rise to an explosion. A dilute aqueous solution of the dioxide is, how- ever, much more stable, and can even be concentrated up to a certain point by ebullition, a portion of the dioxide passing over undecomposed together with the vapour of water. This explains why in the above decomposition the theoretical quantity of oxygen is not obtained. Hydrogen dioxide is easily soluble in ether, and if an aqueous solution of the compound be shaken up with ether, the dioxide dissolves in it. The ethereal is more stable than the aqueous solution, and it can be distilled without decom- position. Hydrogen dioxide undergoes decomposition in presence of a large number of different solid substances, and with the greater rapidity the more finely divided these substances are. Some of the phenomena which thus present themselves may, to a certain extent, be accounted for, but for others we still need an explanation. Thus, for example, the anhydrous com- pound is decomposed with almost explosive violence into oxygen and water, in presence of finely-divided silver, gold, platinum, and other metals, the metals themselves, however, remaining unaltered. The oxides of these metals also decompose hydrogen dioxide easily, being reduced to the metallic state. The same decomposition also occurs with a dilute aqueous solution of the dioxide, and the reaction may be represented by the following equation : Here we have the remarkable phenomenon of a powerful oxidizing agent exerting a reducing action upon metallic oxides, HYDROGEN PEROXIDE 311 the metal being formed. The explanation of this fact is however, not far to seek. The above-named metals possess only a weak power of combination for oxygen, and their oxides accordingly decompose easily into their elements. When these oxides are brought into contact with hydrogen dioxide, which itself contains one atom of oxygen but feebly united, mutual reduction takes place, the one atom of oxygen in the dioxide combining with one atom of oxygen in the metallic oxide to form a molecule of free oxygen. 1 In the same way we explain the fact that common oxygen is formed when ozonized oxygen is brought into contact with aqueous hydrogen dioxide. Here, too, both bodies contain a loosely combined atom of oxygen, which unite together to form a molecule of free oxygen ; thus : When baryta-water (barium monoxide) is mixed with hy- drogen dioxide a precipitate of barium dioxide separates out ; thus : BaO + H 2 O 2 = Ba0 2 + H 2 O. If the hydrogen peroxide be left in contact with the barium peroxide, oxygen is slowly evolved until the whole of the hydrogen peroxide has been decomposed, the barium peroxide being however left unaltered. According to Schone this is due to the formation of a compound, BaO 2 .H 2 O 2 . This is a white substance, which decomposes in the presence of an excess of hydrogen peroxide into barium peroxide, water and free oxygen. Similar reactions occur with the alkaline hydroxides, and this may explain the instability of hydrogen peroxide in alkaline solution. Hydrogen dioxide also transforms many other basic oxides. especially in presence of an alkali, to peroxides. Thus manganous salts become in this way converted into manganese dioxide. On the other hand, these peroxides in presence of an acid are again reduced by hydrogen dioxide to basic oxide. Thus, if hydrogen dioxide be brought into contact with dilute sulphuric acid and manganese dioxide, oxygen gas is given off, and manganous sulphate is formed ; thus : Mn0 2 + H 2 2 + H 2 S0 4 = MnSO 4 + 2H 2 0+O 2 . 1 Cf. Berthelot, Compt. Rend. 90, 572. 312 THE NON-METALLIC ELEMENTS The decomposition here occurring is similar to that which takes place in the reduction of oxide of silver, and this change is assisted by the presence of the acid, which then combines with the basic oxide to form a salt. Detection and Estimation of Hydrogen Dioxide. In order to detect the presence of hydrogen dioxide in solution, the liquid is rendered acid with sulphuric acid, some ether and a few drops of potassium chromate are added, and the solution well shaken. If hydrogen dioxide be present, the solution assumes a beautiful blue colour, and on allowing it to stand the colour is taken up by the ether and a deep blue layer separates out. This blue compound is probably perchromic acid, and the reaction may, in a similar way, be employed for the detection of chromium. 1 When hydrogen dioxide is added to a solution of iodide of potassium and ferrous sulphate, iodine is set free, as may easily be proved by the formation of the blue iodide of starch (Schonbein). This reaction is so delicate that one part of the dioxide in twenty-five million parts may thus be detected. Other oxidizing agents have the power of liberating iodine from iodide of potassium, but not in presence of ferrous sulphate. For the purpose of determining the quantity of hydrogen dioxide present in a solution, the liquid is acidified with sulphuric acid, and then a standard solution of potassium permanganate added, until the purple tint no longer disappears. The reaction here occurring is thus represented : 2KMn0 4 + 3H 2 S0 4 + 5H 2 O 2 = K 2 S0 4 + 2MnS0 4 + 8H 2 + 50 2 . Hydrogen dioxide occurs in small quantities in the atmosphere and has been found in rain and in snow to the amount of about one mgrm. per litre ; but the above method cannot be used for estimating its quantity in this case, as other substances contained in the air, such as the nitrites and organic matter, act upon the permanganate solution In this case a colorimetric method proposed by Schone 2 may be used. It depends upon the fact that a neutral solution of hydrogen dioxide gradually liberates iodine from a neutral solution of potassium iodide ; thus : 1 Moissan, Compt. Bend. 97, 96 ; Berthelot, Compt. Rend. 108, 24. - Her. 7, 1695 ; Annalm, 195, 228. OXIDES AND OXYACIDS OF CHLORINE 313 The liberation of iodine is accompanied by an evolution of oxy- gen, and is very small in comparison to the amount of hydrogen peroxide decomposed. Solutions containing small known but varying quantities of hydrogen dioxide are first prepared, and to each of these solutions potassium iodide and starch paste are added, the degree of tint which the several solutions attain, owing to the formation of the blue iodide of starch, is then com- pared with that obtained in a similar way when rain-water, &c., is employed. An aqueous solution of hydrogen dioxide is now largely used for the bleaching of silk and wool. For this purpose a liquor is prepared by dissolving sodium peroxide in water and acidulating with sulphuric acid. The sodium peroxide is manufactured by the combustion of sodium and is brought into the market under the name of soda-bleach. Peroxide of hydrogen is also used for the purpose of cleaning and bleaching old and stained engrav- ings and oil paintings, and as an auricome for bleaching dark- coloured hair. OXYGEN AND CHLORINE. OXIDES AND OXY- ACIDS OF CHLORINE. 157 Although chlorine and oxygen do not combine directly, two distinct compounds of these elements may be obtained by indirect means. A third oxide, described by Millon and others as chlorine tri-oxide, has been proved to be a mixture of free chlorine and chlorine peroxide. 1 We are acquainted with no less than four compounds of chlorine with oxygen and hydrogen, which are known as the oxy-acids of chlorine. Of these, the first corresponds to the oxide of chlorine, that is, it is formed by the action of water upon the latter. The following are the com- pounds of chlorine, oxygen, and hydrogen as yet known . Oxides Oxy-acids. Chlorine monoxide, C1 2 O Hypochlorous acid, HC10 Chlorous acid, HC1O 2 Chlorine peroxide, C1O 2 Chloric acid, HC1O 3 Perchloric acid, HC1O 4 1 Garzarolli-Thurnlackh, Annalen, 209, 184. 314 THE NON-METALLIC ELEMENTS Chlorine and oxygen can only be made to combine -togetiiar? in the presence of a basic oxide ; thus, if chlorine gas be led over dry mercuric oxide, chlorine monoxide and mercuric oxy- chloride are formed ; thus : 2HgO + 2Cl 2 = Hg 3 OCl 2 + C1 2 0. The same reaction takes place in presence of water, and in this case a colourless solution of the corresponding hypochlorous acid is formed. C1 2 + H 2 = 2C10H. When chlorine is passed into a cold dilute solution of an alkali such as caustic potash, instead of the free hypochlorous acid the corresponding salt, termed a hypochlorite, is formed ; thus : 2KOH + C1 2 = KOC1 + KC1 + H 2 0. If the solution of the alkali be concentrated, or if it be heated whilst the gas is passed through it, a different reaction takes place, in which a salt called a chlorate is formed ; thus : 6KOH + 3C1 2 = KC1O 3 + 5KC1 + 3H 2 O. From the potassium chlorate thus formed, chloric "acia itself can be obtained, and by reduction of this acid the oxide C10 2 may be prepared. PercHioric'acid is prepared by the further oxidation of chloric acid. The oxides corresponding to chlorous, chloric and perchloric acids, viz., C1 2 O 8 , C1 2 5 , and C1 2 O 7 , have not yet been prepared. The oxides and oxy-acids of chlorine are un- stable compounds, as indeed might be expected, owing to the feeble combining power which chlorine and oxygen exhibit to- wards one another ; in consequence of this they act as powerful oxidizing substances, many of them being most dangerously explosive bodies, which suddenly decompose into their con- stituents on rise of temperature, or even on concussion. It is, however, remarkable that perchloric acid, which contains the most oxygen, is the one which is the most stable. CHLORINE MONOXIDE 315 CHLORINE MONOXIDE OR HYPOCHLOROUS ANHYDRIDE. C1 2 O. 158 So long ago as 1785 Bertliollet noticed that chlorine could be combined with an alkali and yet preserve the peculiar bleach- ing power which had been previously discovered by Scheele, and it is to Berthollet that we owe the practical application of this important property. In his first experiments on this substance he employed chlorine water, but afterwards he absorbed the gas by a solution of caustic potash ; and the liquor thus obtained, called Eau de Javelles from the name of a bleach-works where it 'was prepared, was employed for bleaching purposes on the large scale. Berthollet described these experiments to James Watt, who was at that time staying in Paris, and he brought the news to Glasgow, where Tennant, in 1798, patented an improved process for bleaching, in which lime was employed instead of the potash, as being a much cheaper substance. 1 Up to the year 1809-10, when Gay-Lussac and The'nard propounded the view that chlorine might be considered an element, and when Davy proved that this supposition was correct, the bleaching liquors were supposed to contain oxygenated oiuri- ates of the base. Indeed their constitution remained doubtful until the year 1834, when Balard 2 showed that the alkaline bleaching compounds may be considered to be a mixture or combination of a chloride and a hypochlorite. Eau de Javelles therefore contains potassium chloride and hypochlorite, and bleaching-powder solution the corresponding calcium salts ; solid bleaching-powder has, however, a different constitution (Vol. II., Pt. I, p. 194). Preparation. Chlorine monoxide is obtained, as seen in Fig. 95, by the action of dry chlorine gas upon cold dry oxide of mercury, which is contained in a tube (a b), which should be cooled by means of ice or a stream of cold water. 3 The crystal- lized mercuric oxide can, however, not be used for this purpose, as it is not acted on by dry chlorine, and hence the precipitated oxide must be employed, it having been previously carefully washed and dried at 300-400. The reaction which takes place in this case has already been described. Mercuric chloride, HgCl 2 , is not formed in this reaction, but the oxychloride, HgO,HgCl 2 . 1 Tennant's first patent was declared invalid three years after it had been granted, as it was proved that bleachers in Lancashire and at Nottingham had employed lime instead of potash before the year 1798. 2 Ann. Chim. Phys. 57, 225. 3 V. Meyer. Ber. 16, 2999 ; Ladenburg, Ber. 17, 157. 316 THE NON-METALLIC ELEMENTS 159 Properties. Chlorine monoxide is a brownish yellow coloured gas, which has a peculiar penetrating smell, somewhat resembling, though distinct from, that of chlorine. Its density is 43'5. 1 By exposure to a low temperature the gas can be con- densed, as in the tube (D), Fig. 95, to an orange-coloured liquid, which boils at about 4-5. If an attempt is made to seal up this liquid in the tube in which it has been prepared, or even if the tube in which it is contained be scratched with a file, it de- composes suddenly with a most violent explosion (Roscoe) ; and when poured out from one vessel to another a similar explosion FIG. 95. takes place (Balard). It likewise explodes on heating, but not so violently, two volumes decomposing into one volume of oxygen and two of chlorine. According to Garzarolli-Thurnlackh and Schacherl it does not, contrary to previous statements, undergo decomposition in direct sunlight, and the liquid, if all organic matter be carefully excluded, may be distilled without decom- position. Most easily oxidizable substances and many finely divided metals take fire in the gas and produce an explosion ; the gas is also decomposed in presence of hydrochloric acid into free chlorine and water. C1 2 O - 2C1 2 H 2 0. 2HC1 1 Thurnlackh and Schacherl, Annalen, 230, 273. HYPOCHLOROUS ACID 317 Chlorine monoxide is readily soluble in water, the latter dis- solving 200 vols. or 0-78 of its weight of the gas at 0. The solution has an orange-yellow colour. HYPOCHLOROUS ACID. HC1O. 1 60 The aqueous solution of chlorine monoxide must be con- sidered as a solution of hypochlorous acid, a compound which in the pure state is unknown. Preparation. (1) The solution is best prepared by shaking chlorine-water with precipitated mercuric oxide, when the oxide quickly dissolves and the colour of the solution disappears, thus : HgO + 2C1 2 + H 2 = HgCl 2 + 2C10H. The liquid is now distilled in order to remove the mercuric chloride, and a distillate is thus obtained, which, although it contains only half as much chlorine as the original chlorine water, possesses an equal bleaching power. 1 This fact is expressed in the following equations, which also indicate that the bleaching effect produced by chlorine is in reality due to a decomposition of water, the chlorine combining with the hydrogen and liberating the oxygen. It is, therefore, this latter element which is the true bleaching agent, inasmuch as it oxidizes and destroys the colouring agent. (a) Bleaching action of chlorine water, (b) Bleaching action of the hypochlorous acid formed from the chlorine water, (2) An aqueous solution of hypochlorous acid is also easily obtained by adding to a solution of a bleaching compound exactly the amount of a dilute mineral acid requisite to liberate the hypochlorous acid (Gay-Lussac). For this purpose a dilute nitric acid containing about 5 per cent, of the pure acid is allowed to run slowly from a tap-burette into a filtered solution of common bleaching-powder, whilst the liquid is kept well stirred in order to prevent a local super-saturation, which would 1 Gay-Lussac, Ann. Chim. Phys. 43, 161. 318 THE NON-METALLIC ELEMENTS cause a liberation of the hydrochloric acid of the chloride, and thus again effect a decomposition of the hypochlorous acid into chlorine and water. If this operation be conducted with care, no chlorine is evolved, or at any rate only a trace if a slight excess of nitric acid has been added, and the distillate is perfectly colourless. Boric acid may also be employed with advantage for liberating hypochlorous acid from its salts. 1 (3) Another method of obtaining the aqueous acid is to saturate a solution of bleaching-powder with chlorine, then to drive off the excess of chlorine bypassing a current of air through the liquid, and then to distil. The following equation represents the reaction which here occurs : Ca(OCl) 2 + 2C1 2 + 2H 2 = CaCl 2 + 4C1OH. In place of bleaching-powder baryta- water may be employed when barium hypochlorite is at first formed, and this afterwards decomposed as shown above. 2 (4) Hypochlorous acid is so weak an acid that its salts are decomposed by carbonic acid, so that if chlorine gas is led into a solution of a carbonate, or passed through water containing finely divided calcium carbonate in suspension (1 part to 40 of water), no hypochlorite is formed, but only hypochlorous acid (Williamson) ; thus : CaCO 3 + 2C1 2 + H 2 O = 2C10H + CaCl 2 + CO 2 . Other salts of the alkali-metals act in a similar way when a stream of chlorine is passed through their aqueous solutions ; this is the case with sulphate and phosphate of sodium, in these cases an acid salt is formed ; thus : Na 2 S0 4 + H 2 + C1 2 = NaCl + NaHSO 4 + C10H. Concentrated aqueous solutions of hypochlorous acid have an orange yellow or golden yellow colour, and an odour somewhat resembling that of chloride of lime. Only dilute solutions of hypochlorous acid can be distilled without decomposition ; con- centrated solutions are readily decomposed either on heating or on exposure to sunlight, part splitting up into chlorine and oxygen, whilst another part undergoes oxidation, yielding chloric acid. 1 Lauch, Ber. 18. 2287. 2 Williamson, Chem. Soc. Mem. 2, 234. THE HYPOCHLORITES 319 The hypochlorites, like the acid, are unstable compounds, which in the pure state are almost unknown. Of these the most important is formed when bleaching-powder is dissolved in water as calcium hypochlorite, Ca(OCl) 2 , although it is probably not present as such in the solid substance, and it is to the presence of this compound that the bleaching properties of the solution are due, inasmuch as when either hydrochloric or sulphuric acid is added, a quantity of chlorine eo]tial to that contained in the compound is evolved. In the first case half the chlorine is derived from the hypochlorite, the other half from the hydrochloric acid, which first liberates hypochlorous acid, and then decomposes it into chlorine and water ; thus : (1) 2HCl + Ca(OCl) 2 = (2) Tf sulphuric acid is used, the result is the same, as this acid decomposes the calcium chloride ; thus : CaCl 2 + Ca(OCl) 2 + 2H 2 SO 4 = 2CaSO 4 + 2H 2 O + 2C1 2 . When the solutions of hypochlorites are heated, they undergo decomposition into chlorides and chlorates in the following manner : 3KC10 = 2KC1 + KC1O 3 . Hence when chlorine is passed into a hot solution of an alkali the product is a mixture of chloride and chlorate. CHLOROUS ACID AND THE CHLORITES. 161 This acid, like chlorine trioxide, is not known in the free state, but the chlorites can be prepared by adding potassium hydroxide to an aqueous solution of chlorine peroxide, a mixture of potassium chlorate, and potassium chlorite, KC1O 2 , being obtained. 1 The chlorites of the alkali metals are soluble in water, and from their solutions the insoluble, or difficultly soluble chlorites of silver, AgClO 2 , and of lead, Pb(ClO 2 ) 2 , may be prepared by double decomposition, as yellow crystalline powders. All the chlorites are very easily decomposed. Thus if the lead- 1 Garzarolli-Thurnlackh and v. Hayn, Annalen, 209, 203. 320 THE NON-MET ALLIG ELEMENTS salt is heated for a short time to 100, it decomposes with detonation ; and if it is rubbed in a mortar with sulphur or certain metallic sulphides, ignition occurs. The soluble chlorites possess a caustic taste, and bleach vegetable colouring matters, even after addition of arsenious acid. This latter reaction serves to distinguish them from the hypochlorites. CHLORINE PEROXIDE. C1O 2 . 162 This gas was first prepared and examined by Davy in 1815 ; it was obtained by him by the action of strong sulphuric acid on potassium chlorate. In preparing this substance special precautions must be taken, as it is a highly explosive and dangerous body. Preparation. Pure powdered potassium chlorate is for this purpose thrown little by little into concentrated sulphuric acid contained in a small retort. After the salt has dissolved, the retort is gently warmed in warm water. In this reaction chloric acid is in the first instance liberated, and then decomposes as follows into perchloric acid, chlorine peroxide and water ; thus : 3HC10 3 = HC10 4 + 2C10 2 + H 2 O. Properties. The heavy dark yellow gas thus given off must be collected by displacement, as it decomposes in contact with mercury and is soluble in water ; it possesses a peculiar smell, resembling that of chlorine and burnt sugar. When exposed to cold the gas condenses to a dark -red liquid, which boils 1 at + 9, and at 79 freezes to an orange-coloured crystalline mass. Its vapour density corresponds to the formula C1O 2 , 2 and the double formula, C1 2 O 4 , previously employed is therefore incorrect. The gaseous, and especially the liquid and solid peroxide undergo sudden decomposition, frequently exploding most violently, hence their preparation requires extreme care. According to Schacherl, liquid chlorine peroxide may be distilled without decomposition if every trace of organic matter be excluded. 3 In order to obtain an aqueous solution of the gas, a mixture of potassium chlorate and oxalic acid may be heated in a water bath to 70, the mixture of chlorine peroxide and carbon dioxide 1 Pebal, AnnaUn, 177, 1. 2 Pebal and Schacherl, Annalen, 213, H3. 3 Annalen, 206, 68. CHLORIC ACID AND THE CHLORATES 321 then evolved being passed into water. 1 The following equation represents the action here occurring : 2KC10 Chlorine peroxide can be preserved without change in the dark ; it is, however, slowly decomposed into its elementary con- stituents when exposed to light, and this decomposition takes place quickly and with explosion when an electric spark is passed through the gas, two volumes of the gas yielding one volume of chlorine and two of oxygen. When phosphorus, ether, sugar, or other easily combustible substances are thrown into the gas they take fire spontaneously. This oxidizing action of chlorine peroxide is well illustrated by the following experiments. About equal parts of powdered white sugar and chlorate of potash in powder are carefully mixed together with a feather on a sheet of writing paper, the mixture then brought on a plate or stone placed in a draught chamber, and a single drop of strong sulphuric acid allowed to fall upon the mixture, when a sudden ignition of the whole mass occurs. This is caused by the liberation of chlorine peroxide, which sets fire to a particle of sugar, and the ignition thus commenced quickly spreads throughout the mass, and the sugar is all burnt at the expense of the oxygen of the chlorate. The combustion of phosphorus can be brought about under water by a similar reaction : for this purpose some crystals of chlorate of potassium and a few small lumps of yellow phosphorus are thrown into a test glass half filled with water, and a small quantity of strong sulphuric acid allowed to flow through a tube funnel to the lower part of the glass where the solids lie. As soon as the acid touches the chlorate, chlorine peroxide is evolved, and this gas on coming in contact with the phosphorus oxidizes it, and bright flashes of light are given off. Water at 4 dissolves about twenty times its volume of chlorine peroxide gas, forming a bright yellow solution, whilst at lower temperatures a crystalline hydrate is formed. If this aqueous solution is saturated with an alkali, a mixture of chlorite and chlorate is formed ; thus : 2KOH + 2C1O 2 = KC1O 2 + KC10 3 + H 2 0. 1 Calvert and Davies, Journ. Chem. Soc. 2, 193. 22 322 THE NON-METALLIC ELEMENTS When potassium chlorate is treated with hydrochloric acid a yellow gas is evolved, first prepared by Davy, considered by him to be a distinct oxide of chlorine, and termed Euchlorinc, It has, however, been shown by Pebal 1 that this body is a mixture of free chlorine and chlorine peroxide in varying proportions. This mixture possesses even more powerful oxidizing properties than chlorine itself, and is therefore largely used as a disinfect- ant, being not only very effective in removing by oxidation putrescent matter in the air, but being at the same time very easily prepared. CHLORIC ACID. HC1O 3 . 163 Chloric acid is the most important member of the series of chlorine oxy-acids. It was discovered by Berthollet in 1786, and it is obtained when the lower acids or aqueous solutions of the oxides of chlorine are exposed to light. Preparation. Chloric acid is best prepared by decomposing barium chlorate with an equivalent quantity of pure dilute sulphuric acid (Gay-Lussac, 1814) ; thus: Ba(C10 3 ) 2 + H 2 SO 4 = BaS0 4 + 2HC10 3 . The clear solution of chloric acid must be poured off from the deposited precipitate of barium sulphate, and carefully evaporated in vacuo over strong sulphuric acid. The residue thus prepared contains forty per cent, of pure chloric acid corresponding to the formula HC1O 3 -f 7H 2 O. When attempts are made to concen- trate the acid beyond this point, the chloric acid undergoes spontaneous decomposition with rapid evolution of chlorine and oxygen gases, and formation of perchloric acid. Chloric acid can also be prepared by decomposing potassium chlorate with hydrofluosilicic acid, H 2 SiF 6 , when insoluble potassium fluosi- licate, K 2 SiF 6 , is precipitated, and the chloric acid remains in solution together with an excess of hydrofluosilicic acid. This can be removed by the addition of a little silica and by subse- quent evaporation when the fluorine passes away as gaseous tetrafluoride of silicon, SiF 4 , and the pure chloric acid can be poured off from the silica, which settles as a powder to the bottom of the vessel. 164 Properties. The acid obtained in this way in the greatest state of concentration does not rapidly undergo change at the i Annalen, 177, 1. CHLORIC ACID 323 ordinary temperature, but it forms perchloric acid on standing for some time exposed to light. Organic bodies such as wood or paper decompose the acid at once, and are usually so rapidly oxidized as to take fire. Aqueous chloric acid is colourless, possesses a powerful acid reaction and a pungent smell, and bleaches vegetable colours quickly. It is a monobasic acid, that is, it contains only one atom of hydrogen capable of replacement by a metal, with the formation of salts. The Chlorates. Of these salts potassium chlorate (or chlorate of potash), KC1O 3 , is the most important. It is easily formed by passing chlorine in excess into a hot solution of caustic potash ; thus : 3C1 2 + 6KOH =. 5KC1 + KC1O 3 +3H 2 O. The chlorate is much less soluble in water than the chloride formed at the same time, so that by concentrating the solution the chlorate is deposited in tabular crystals, which may be purified from adhering chloride by a second crystallization. Other chlorates can be prepared in a similar way ; thus, for instance, calcium chlorate is obtained by passing a current of chlorine into hot milk of lime when the following reaction occurs : 6C1 2 + 6 Ca(OH) 2 =Ca(ClO s ) 2 + 5 CaCl 2 + 6H 2 0. All the chlorates are soluble in water, and many deliquesce on exposure to the air. The potassium salt is one of the least soluble of these salts, 100 parts by weight of water at dis- solving about 3'3 parts of this salt, whilst water at 15 dissolves twice this amount. By the action of reducing agents such as nascent hydrogen or sulphur dioxide, chlorates lose the whole of their oxygen and are converted into chlorides. A chlorate is recognized by the following tests : (1) Its solution yields no precipitate with silver nitrate, but on ignition the salt gives off oxygen gas, and a solution of the residual salt (a chloride) gives a white precipitate on addition of silver nitrate and nitric acid. (2) To the solution of the chlorate a few drops of indigo solution are added, the liquid acidulated with sulphuric acid, and sulphurous acid (or sodium sulphite dissolved in water) added drop by drop. If a chlorate be present the blue colour is discharged, because the chloric acid is reduced to a lower oxide. 324 THE NON-METALLIC ELEMENTS (3) Dry chlorates treated with strong sulphuric acid yield a yellow explosive gas (C1O 2 ). The composition of the chlorates has been very carefully determined by Stas l and Marignac. 2 The following numbers give the percentage composition of silver chlorate according to the analyses of Stas : Chlorine 18-5257 Oxygen 25'0795 Silver . 56'3948 100-000 PERCHLORIC ACID. HC1O 4 . 165 This acid was discovered by Stadion in 1816 ; it is formed by the decomposition of chloric acid on exposure to heat or light ; thus : 3HC10 3 = HC10 4 + C1 2 + 20 2 + H 2 0. It is best prepared from potassium perchlorate, which can be obtained in any quantity from the chlorate. We have already remarked under oxygen, that when potassium chlorate is heated the fused mass slowly gives off oxygen, and a point is reached at which the whole mass becomes nearly solid, owing to the formation of perchlorate, together with potassium chloride. 10KC1O = The mass is then allowed to cool, powdered, and well washed with water, to remove the greater part of the chloride formed. In order to get rid of the unaltered chlorate, the crystalline powder is gently heated with hydrochloric acid so long as chlorine and chlorine peroxide gases are evolved : a subsequent washing with water removes the remainder of the chloride, and the pure, sparingly-soluble perchlorate is left. Preparation. In order to prepare perchloric acid, the pure dry potassium salt is distilled in a small retort with four times its weight of concentrated (previously boiled) sulphuric acid. At a temperature of 110 dense white fumes begin to be evolved, whilst a colourless or slightly yellow liquid, consisting of pure perchloric 1 Nouvelles Recherches Chiiniques sur les Lois des Proportions, 208. 2 Bibl. Univ. 45, 347. ANALYSIS OF PERCHLORIC ACID 325 acid,HC!O 4 , distils over(Roscoe). 1 If the distillation be continued, this liquid gradually changes into a white crystalline mass, having the composition HClO 4 -fH 2 O. The formation of this latter body can be readily explained ; a portion of the pure per- chloric acid splits up during the distillation into the lower oxides of chlorine, oxygen and water, which latter combines with the pure acid already formed. When the crystalline hydrate is again heated it decomposes into the pure acid, which distils over, and into an aqueous acid which boils at 203, and therefore remains behind in the retort. This reaction is employed in the prepara- tion of the pure acid, HC1O 4 , as that obtained by the first pre- paration is generally rendered impure by sulphuric acid carried over mechanically. An aqueous solution of the acid may be readily prepared by the addition of hydrofluosilicic acid to a solution of the potassium salt, and is sometimes used for the estimation of potassium. 2 Properties. Pure perchloric acid is a crystalline substance which melts at 15, 3 but is usually obtained as a volatile colour- less or slightly yellow mobile liquid, having at 15'5 a specific gravity of T782. It is strongly hygroscopic, quickly absorbing moisture from the air, and emitting dense white fumes of the hydrated acid. When poured or dropped into water it dissolves, combining with the water so vigorously as to cause a loud hissing sound and a considerable evolution of heat. A few drops thrown upon paper and wood cause an instantaneous and almost explosive inflammation of these bodies ; and if the same quantity be allowed to fall upon dry charcoal, the drops decompose with an explosive violence which is almost equal to that observed in the case of chloride of nitrogen. If the pure acid, even in very small quantity, come in contact with the skin it produces a serious wound, which does not heal for months. Perchloric acid undergoes decomposition on distilla- tion ; the originally nearly colourless acid becomes gradually darker, until it attains the tint of bromine, and at last suddenly decomposes with a loud explosion. The composition of the substance which is here formed is unknown. The pure acid also undergoes spontaneous and explosive decomposition when pre- served for some days even in the dark. The methods employed in fixing the composition of this acid 1 Journ. Chem. Soc. 1863, 82. 2 Zeit. angew. Chem. 1893, 68. 3 Berthelot, Compt. Mend. 93, 240. 326 THE NON-METALLIC ELEMENTS may here be referred to as illustrating the mode by which the quantitative analysis of similar bodies is carried out. A quantity of the pure acid (HC1O 4 ) is sealed up in a small glass bulb (Fig. 96), whose weight has been previously ascertained, and the bulb and acid carefully weighed. The sealed points of the tube are then broken, the acid diluted with water, and the aqueous solution saturated with a slight excess of solution of potassium carbonate. Acetic acid is next added in slight excess, and the whole evaporated to dryness on a water bath. The potassium acetate being soluble, and the potassium perchlorate being insoluble in absolute alcohol, the whole of the latter salt formed by the neutralisation of the acid is obtained in the pure state by washing the dry mass with absolute alchohol and drying the residue at 100. Thus it was found that 0'7840 gram of the acid thus treated yielded 1*0800 gram of potassium salt corresponding to 0'7837 gram of pure perchloric acid, or the FIG. 96. acid under analysis contained 99'85 per cent, of HC10 4 , provided, of course, that the salt obtained really had the composition indicated by the formula, KC1O 4 . In order to obtain evidence on this point, the quantities of oxygen, chlorine, and potassium contained in the salt were determined as follows : (1) 0*9915 gram of the dry salt was mixed with pure dry oxide of iron, and the mixture heated in a long tube of hard glass, the weight of which, when thus filled, was determined. The presence of the oxide of iron enabled the perchlorate to yield up its oxygen at a lower temperature than it would have done if heated alone. The loss of weight which the tube experiences on heating represents the total weight of oxygen contained in the salt ; in this case it amounted to 0*4570 gram. (2) The residue is next completely exhausted with warm water, and the chlorine precipitated in the solution as silver chloride ; in the above analysis 0'7683 gram of pure silver was needed for complete precipitation, and this corresponds to 40 '25 2 gram of chlorine. (3) 0*3165 gram of the salt was next carefully heated with an excess of pure sulphuric acid, and the residue strongly PERCHLORATES. 327 ignited. In this way the potassium perchlorate is converted into sulphate, which was found to weigh 0'2010 gram. From these numbers the percentage composition of the salt can be easily obtained, and the results, as shown below, are found to correspond, within the unavoidable errors of experiment, with the numbers calculated from the formula : Analysis of Potassium Perchlorate. Calculated. Found. Chlorine ... Cl 35'19 25'58 25'46 Oxygen ... O 4 63'52 46'18 46'09 Potassium . K 38'84 28'24 28-58 137-55 100-00 100-13 166 Hydrates of Perchloric Acid. The monohydrate, HC10 4 -f H 2 O, whose mode of formation has been mentioned, is obtained in the pure state by the careful addition of water to the pure acid, HC10 4 , until the crystals make their appearance. This substance, discovered by Serullas, was formerly supposed to be the pure acid ; it melts at 50 and solidifies at this temperature again in colourless needle-shaped crystals, often several inches in length. The liquid emits dense white fumes on exposure to the air, and oxidizes paper, wood, and other organic bodies with rapidity. As has been stated, the monohydrate decomposes at a higher temperature into the pure acid, and a thick oily liquid, which possesses a striking resemblance to sulphuric acid, boils at 203, and has a specific gravity of 1'82. This liquid contains 71 '6 per cent, of HC10 4 , and does not correspond to any definite hydrate. An acid of the same composition, and possessing the same constant boiling point, is obtained when a weaker acid is distilled, the residue then -becomes more and more concentrated, until the above composition and boiling point is reached. Aqueous perchloric acid, therefore, exhibits the same relations in this respect as the other aqueous acids. Perchlorates. Perchloric acid is a powerful monobasic acid, forming a series of salts, termed the perchlorates, which are all soluble in water, and a few of which are deliquescent. Potassium perchlorate, KC1O 4 , and rubidium perchlorate, RbC10 4 , are the least soluble of the salts, one part of the former dissolving in 58, and the latter requiring 92 parts of water at 21 for solution. Both these salts are almost insoluble in 328 THE NON-METALLIC ELEMENTS absolute alcohol, and they may be, therefore, employed for the quantitative estimation of the metals. The perchlorates are distinguished from the chlorates by the following reactions : (1) They undergo decomposition at a higher temperature than the chlorates. (2) They are not acted upon by hydrochloric acid. (3) They do not yield an explosive gas, C1O 2 , when heated with strong sulphuric acid. (4) They are not reduced to chlorides by sulphur dioxide. 167 Constitution of the Oxy -acids of Chlorine. The con- stitution of these acids has frequently been the subject of discus- sion, and cannot as yet be regarded as definitely settled. On the assumption that chlorine always acts as a monovalent, and oxygen as a divalent element, the following formulae are the only possible ones for these acids : Hypochlorous acid, H Cl. Chlorous acid, H O Cl. Chloric acid, H O O Cl. Perchloric acid, H Cl. It has however been found, especially in the case of the carbon compounds, that substances containing oxygen atoms united together in this manner become more unstable as the number of oxygen atoms increases, whilst with the oxy-acids of chlorine the contrary is the case, hypochlorous acid being the most and perchloric acid the least unstable. Another theory is that chlorine behaves as a monad in hypo- chlorous acid, a triad in chlorous acid, a pentad in chloric acid, and a heptad in perchloric acid, the constitutional formulae being as follows : /"^l f~\ TT 0=C1-0 H O \C1 H II 0=C1 H .5 OXY-ACIDS OF BROMINE. 329 This formula for perchloric acid readily explains the existence of the hydrate of perchloric acid, HC1O 4 , H 2 O, the constitution of which would then be represented by the formula O HO, || \C1 H HO/ || O A substance possessing this formula should from analogy be- have as a polybasic acid (compare the phosphoric acids), but hitherto no chemical or physical evidence of the polybasic nature of perchloric acid has been obtained, all the salts cor- responding to the formula, HC1O 4 , and the same holds true for chloric acid. On the whole, however, the balance of the evidence is in favour of the second view, which is also in agree- ment with the results obtained for periodic acid (p. 334). Chlorine peroxide has been definitely proved to have the molecular formula C10 2 , and in this compound chlorine must be regarded as a tetrad ; this fact is also in favour of the view that the valency of chlorine varies in its different compounds with oxygen. OXYGEN AND BROMINE. OXY-ACIDS OF BROMINE. 1 68 No compound of bromine and oxygen has as yet been obtained, but oxy-acids corresponding to those of chlorine are known; viz: Hypobromous acid, HBrO. Bromic acid, HBr0 3 . HYPOBROMOUS ACID, HBrO. This acid together with its salts, termed the hypobromites, are formed, in a similar manner to hypochlorous acid, by the action of bromine on certain metallic oxides (Balard). Thus if bromine water be shaken up with mercuric oxide, and if the yellow liquid 330 THE NON-METALLIC ELEMENTS thus formed be treated successively with bromine and the oxide, a solution is obtained which contains in every 100 cc. 6'2 per cent, of bromine combined as hypobromous acid, the reaction being as follows : HgO + 2Br 2 + H 2 O = 2HOBr + HgBr 2 . The greater part of the hypobromous acid contained in this strong solution is decomposed on distillation into bromine and oxygen. It can, however, be distilled in vacua at a temperature of 40 without undergoing this change. 1 Aqueous hypobromous acid is a light straw-yellow coloured liquid, closely resembling in its properties hypochlorous acid, acting ,as a powerful oxidizing agent and bleaching organic colouring matters. By the action of bromine on lime, a substance similar to bleaching powder is formed and this salt is termed bromide of lime. 2 BROMIC ACID, HBr0 3 . 169 When bromine is dissolved in hot caustic potash or soda, a colourless solution is produced which contains a mixture of a bromide and a bromate ; thus : The difficultly soluble potassium bromate maybe easily separated by crystallization from the very soluble bromide. Potassium bromate is also formed when bromine vapour is passed into a solution of potassium carbonate which has been saturated with chlorine gas. Preparation Free bromic acid is formed when chlorine' is passed into bromine water ; thus : Br 2 + 5C1 2 + 6H 2 O = 2HBr0 3 + 1 OHC1. The acid is, however, best obtained by the decomposition of the slightly soluble silver bromate. This salt is thrown down on the addition of nitrate of silver to a solution of a soluble bromate ; the precipitate thus prepared is well washed with 1 Dancer, Journ. Chem. Soc. 1862, 477. 2 Berzelius, Jahresb. 10, 130. IODINE PENTOXIDE AND IODIC ACID 331 water and then treated with bromine ; bromic acid remains in solution and the insoluble silver bromide is thrown down; thus : 5 AgBrO s + 3Br 2 + 3H 2 O = 5 AgBr + 6HBr0 3 . Properties Obtained according to the foregoing methods, bromic acid is a strongly acid liquid reddening and ultimately bleaching litmus paper. On concentration at 100 the aqueous acid decomposes into bromine and oxygen, and it is at once decomposed by reducing agents such as sulphur dioxide and sulphuretted hydrogen, as also by hydrobromic acid, the following reactions taking place : (1 ) 2HBrO 3 + 5SO 9 + 4H 2 O - Br 2 + 5H 2 S0 4 . (2) 2HBrO 3 + 5SH 2 = Br 2 + 6H 2 O + 5S. (3) HBrO 3 + 5HBr = 3Br 2 + 3H 2 O. Hydrochloric and hydriodic acids decompose bromic acid in a similar manner with formation of the chloride or iodide of bromine. The bromates are as a rule difficultly soluble in water, and decompose on heating into oxygen and a bromide, but unlike the chlorates no perbromate is formed in the process. PEKBROMIC ACID, This substance is stated by Kammerer 1 to be formed by the action of bromine on dilute perchloric acid, the bromine liberating chlorine. Other observers have, however, failed to obtain the substance by this means, and' the existence of the acid and of its salts is, therefore, more than doubtful. 2 OXYGEN AND IODINE. OXIDE AND OXY-ACIDS OF IODINE. 170 Only one oxide of iodine is known with certainty. This is the pentoxide I 2 O 5 , which unites with water to form iodic acid, HIO 3 . Besides these, hyd rated periodic acid, HIO 4 + 2H 2 O is known and hypoiodous acid HIO appears to exist in solution. 1 J. Pr. Chem. 90, 190. 2 Muir, Journ. Chem. Soc. 1876, ii. 469; Wolfram, Annalen, 198, 95; Mac Ivor, Chem. News, 33, 35 ; 55, 203. 332 THE NON-METALLIC ELEMENTS HYPOIODOTJS ACID, HIO. When an alcoholic solution of iodine is treated with freshly precipitated mercuric oxide, a yellow solution is formed which does not turn starch blue at once, but only after a time ; the solution is then found to contains mercuric iodide and iodate. It is probable that the solution contains hypoiodous acid, which soon undergoes decomposition into free iodine and iodic acid. 1 An aqueous solution of iodine also yields with alkalis a solution having a peculiar odour, which possesses bleaching properties, and probably also contains a hypoiodite. 2 Lunge and Schoch, by the action of iodine on slaked lime and water at the ordinary temperature, obtained a substance having a peculiar odour, and resembling chloride of lime in its general properties. It has probably the formula CaOI 2 or Ca(IO) 2 + CaI 2 . 3 IODINE PENTOXIDE, I 2 5 AND IODIC ACID, HI0 3 . 171 This acid was discovered by Davy in the form of potas- sium iodate, which he obtained by the action of iodine on caustic potash ; thus : 3I 2 + 6KOH = 5KI + KI0 3 + 3H 2 O. Preparation. (1) Free iodic acid is best obtained by dissolv- ing iodine in pure boiling concentrated nitric acid, which oxidizes it as follows : 3I 2 +10HN0 3 = 6HI0 3 +10NO + 2H 2 0. For this purpose 1 part of iodine is heated in a retort with 10 parts of the acid until the whole of the iodine is dissolved, and no further evolution of red fumes takes place. The solution is then evaporated and the residue heated to 200 until every trace of nitric acid is removed. The iodic acid thus loses water, and a white powder of iodine pentoxide I 2 O 5 is obtained. It has a specific gravity of 4'487, and when heated to 300 decomposes into iodine and oxygen. This substance is very soluble in water, dissolving with evolution of heat and from the thick 1 Kone, Pogg. Ann. 66, 802 ; Compt. Rend. 63, 968. 2 J. Pr. Chem. [I], 84, 385. 8 Ber. 15, 1883. CONSTITUTION OF IODIC ACID 333 syrupy solution thus obtained rhombic crystals of iodic acid, HIO 3 , are deposited. (2) Iodic acid can also be obtained by the action of dilute sul- phuric acid on barium iodate, which is prepared as follows : the requisite quantity of iodine is dissolved in a hot concentrated solution of potassium chlorate and a few drops of nitric acid added; immediately a violent evolution of chlorine gas com- mences, and, on cooling, the potassium iodate crystallizes out. This salt is then dissolved in water and barium chloride added to the solution, when barium iodate separates out as a white powder. The potassium iodate may also be obtained by very carefully heating a mixture of 2 mols. potassium chlorate and 1 mol. iodine, a simple metathesis taking place. 1 (3) Iodic acid is likewise formed when chlorine is passed into water in which iodine in powder is suspended ; thus : I 2 + 5C1 2 + 6H 2 O = 2HIO 3 + 10HC1. In order to separate the hydrochloric acid which is formed at the same time, precipitated oxide of silver is added until the acid is completely precipitated as the insoluble silver chloride. Properties. Crystallized iodic acid has a specific gravity at of 4'629 ; it is insoluble in alcohol, but easily soluble in water. The concentrated aqueous solution boils at 104, 2 and first reddens, and then bleaches litmus paper. Phosphorus, sulphur, and organic bodies deflagrate when heat^ with iodic acid or with the pentoxide. Sulphur dioxicj^Br sulphuretted hydrogen as well as hydriodic acid reduces iodia acid with separation of iodine ; thus : (1) 2HI0 3 + 5S0 2 + 4H 9 = I, + 5H 9 S0 4 . (2) 2HIO 3 + 5H 9 S = I 2 + 5S + 6H 2 O.~ (3) HI0 3 + 5HI = 3I 2 + 3H 2 O. The lodates. Iodic acid is a monobasic acid, and is distin- guished from chloric acid and bromic acid by the fact that it forms not onty the normal salts, but salts which are termed 1 Thorpe and Perry, Journ. Chem. Soc. 1892, i. 925. 2 Ditte, Ann. Chem. Phys. [4], 21, 5. 334 THE NON-METALLIC ELEMENTS acid- or hydrated-salts. Thus the following potassium salts are known : Normal potassium iodate KI0 3 . Acid potassium iodate KIO 3 , HIO 3 . Di-acid potassium iodate KIO 3 , 2HIO 3 . The normal iodates are chiefly insoluble or difficultly soluble in water; the more soluble are those of the alkali metals. On heating they decompose either into oxygen and an iodide, or oxygen and iodine are given off, leaving a residue of the metal or its oxide ; thus : 2Ba(I0 3 ) 2 = 2BaO + 2I 2 + 5O 2 . 2AgI0 3 = 2Ag + I 2 . In order to detect iodic acid, the solution after acidifying with hydrochloric acid, is mixed with a small quantity of starch paste and then an alkaline sulphite or a solution of sulphurous acid added drop by drop, thus liberating iodine which forms with the starch the blue iodide. Sodium iodate occurs in nature associated with sodium nitrate in Chili saltpetre, and iodic acid is not unfrequently met with in nitric acid prepared from this source. The constitution of iodic acid is not known with certainty. Unlike chloric acid it behaves as a polybasic acid, and the formula H-O-O-O-I, is therefore even more improbable than in the latter case. If, however, we assume that the iodine in this acid is pentavalent and ascribe to it the constitutional formula Q "j;I-0-H, we are still unable to account for the existence of the acid and diacid salts mentioned above, except by regarding them as molecular compounds of the anhydrous salt and the acid. Thomsen l regards the molecular formula of the acid as H 2 I 2 O 6 , and ascribes to it a constitutional formula in which the oxygen atoms are regarded as tetravalent, whilst another formula has been suggested by Blomstrand ; 2 the evidence is however as yet insufficient to decide between the different suggestions. 1 Ber. 7, 112. 2 /. Pr. Chem. [2], 40, 305. PERIODIC ACID 335 PERIODIC ACID, HIO 4 . 172 This substance was discovered by Magnus, 1 and subse- quently investigated by other chemists, especially by Ammer- miiller and Rammelsberg. Normal periodic acid, HIO 4 , is not known. The hydrate H 5 IO 6 or HIO 4 + 2H 2 O is formed either by the action of iodine on aqueous perchloric acid, thus : HC10 4 + 1 + 2H 2 = H 5 I0 6 + Cl, or by the decomposition of silver periodate with bromine. This hydrate is a colourless transparent crystalline deliquescent solid which melts at 133 and at 140 is completely decomposed into iodine pentoxide, water, and oxygen. The aqueous solution has a strong acid reaction and it acts upon reducing agents in a similar way to iodic acid. Periodic acid forms a remarkable series of salts, the com- position of which at first sight seems somewhat complex. Thus we are acquainted with salts having the following general formulas, M representing a monad metal : MI0 4 ; M 4 I 2 9 ! M^KV, MJ 2 O !i; M 5 IO 6 ; M 12 I 2 13 - If, however, we regard iodine as a heptad in these compounds, the hypothetical periodic anhydride, I 2 7 , would have the following constitutional formula : O O II II = 101-0 II II O O the above salts may then be looked upon as derived from acids formed from this anhydride by union with varying numbers of molecules of water, in the manner shown below : I 2 O 7 + H 2 O = 2IO 3 .OH = 2HIO 4 I 9 O 7 + 2H 2 O = IO 9 (OH), O IO,(OH) = H 4 I 2 O 9 I 2 O 7 + 3H 2 O = 2IO 2 (OH), = 2HIO 4 . 2H 9 O I O 7 + 4H 2 O = IO(OH) 4 O IO(OH) 4 = H 8 I 2 O n I.,O 7 + 5H 2 O = 2IO(OH) 5 = 2HIO 4 , 4H 2 O 1 A + 6H 2 = I(OH) fl O I(OH) 6 = H 19 I 9 13 I 2 O 7 + 7H 2 O = 2I(OH) 7 = 2HIO 4 , 6H 2 O. 1 Pogg. Ann. 28, 514. THE NON-METALLIC ELEMENTS No salts are known corresponding to the acid I(OH) 7 , but it will be seen that the composition of the remaining six acids corre- sponds exactly with one or other of the series of salts which has already been prepared. (Compare the constitution of the phosphoric acids.) The best known series of periodates are those derived from the acids HIO 4 , H 4 I 2 O 9 , H 3 IO 5 , and H 5 I0 which are termed metaperiodates, diperiodates, mesoperiodates, and paraperiodatcs respectively. The periodates can be obtained in several ways ; thus if chlorine be allowed to act on a mixture of sodium iodate and caustic soda, a mixture of two sodium paraperiodates (Na 2 H 3 IO 6 , and Na 3 H 2 IO 6 ) and sodium chloride is formed. Another method is to heat barium iodate, which is thus converted into barium periodate, iodine and oxygen. 5Ba(I0 3 ) 2 = Ba 5 (I0 6 ) 2 + 4I 2 + 8O 2 . The barium paraperiodate may be heated to redness without decomposition, whereas the other periodates are decomposed at this temperature with evolution of oxygen. The periodates are, as a rule, but slightly soluble in water ; their solutions give with silver nitrate and nitric acid a pre- cipitate of silver periodate, a different silver salt being obtained according to the proportion of nitric acid present. 1 SULPHUR. S = 31-82. 173 SULPHUR has been known from the earliest times as it occurs in the free or native state, in the neighbourhood of extinct as well as of active volcanoes. It was formerly termed Brim- stone or Brennestone, and was considered by the alchemists to be the principle of combustibility, and believed by them to re- present the alterability of metals by fire. The compounds of this element occur in nature in much larger quantities, and are much more widely distributed than free sulphur itself. The com- pounds of sulphur with the metals, termed sulphides, and those with the metal and oxygen termed sulphates are found in large quantities in the mineral kingdom. The more important com- pounds of sulphur occurring in nature are the following : 1 Kimmins Journ. Chem. Soc. 1887, i. 356 ; 1889, i. 148. OCCURRENCE OF SULPHUR 337 (1) Sulphides. Iron pyrites FeS 2 ; copper pyrites CuFeS 2 . galena PbS ; cinnabar HgS ; blende ZnS ; grey antimony Sb 2 S 3 realgar As 2 S 2 ; orpiment As 2 S 3 . (2) Sulphates. Gypsum GaS0 4 -f- 2H 2 O ; gypsum anhydrite aSO 4 ; heavy spar BaSO 4 ; kieserite MgSO 4 + H 2 O; bitter spar MgSO 4 + 7H 2 O ; Glauber salt Na. 2 SO 4 + 10H 2 ; green vitriol FeS0 4 + 7H 2 O. Volcanic gases almost always contain sulphur dioxide and sulphuretted hydrogen, and when these two moist gases come into contact they mutually decompose with the deposition of sulphur, thus : S0 2 + 2H 2 S = 3S + 2H 2 0, FIG. 97. It is very probable that native sulphur is, in some cases, formed by the above reaction. The apparatus shown in Fig. 97 serves to exhibit this change ; the sulphuretted hydrogen gas evolved in the bottle (C) is passed into the large flask (A) into which is led at the same time sulphur dioxide from the small flask (B). The walls of the large flask are soon seen to become coated with a yellow deposit of sulphur. Sulphur compounds are also found widely distributed in the vegetable and animal world, in certain organic compounds such as the volatile oils of mustard and of garlic, and in the acids occurring in the bile. Sulphur is also found in small quantities in hair, and wool, whilst it is contained to the amount of about 1 per cent, in all the albuminous sub- 23 338 THE NON-METALLIC ELEMENTS stances which form so important a constituent of the animal body. 174 Almost all the native sulphur of commerce comes from Italy, where it is found in the Romagna and in other parts of the country, but especially in very large quantities in the volcanic districts of the island of Sicily, where it occurs in widespread masses found chiefly on the south of the Madonia range stretching over the whole of the provinces of Caltanissetta and Girgenti, and over a portion of Catania. No fewer than 476 distinct sulphur workings exist in Sicily, from which the annual production in the year 1886 amounted to 410,000 tons; of this quantity only 30,226 tons were imported into the United Kingdom, 1 and since the above year the production of the Sicilian mines has not been reduced, although the number of the mines worked has decreased. The deposits of Sicilian sulphur occur in the tertiary formation lying imbedded in a matrix of marl, limestone, gypsum, and celestine. The sulphur occurs partly in transparent yellow crystals termed virgin sulphur and partly in opaque crystalline masses, to which the name of volcanic sulphur is given. Both these varieties are separated from the matrix by a simple process of fusion. The method often described, in which the sulphur ore is repre- sented as being placed in earthenware pots in a furnace, the sulphur distilling out into other pots placed outside the furnace, appears to be unknown in Sicily. In the Romagna an apparatus made of cast-iron and provided with a receiver of the same material is employed, but in Sicily a very simple method of melting out the sulphur has long been, and still continues to be, in vogue. This old process consists in placing a heap of the ore in a round hole dug in the ground averag- ing from 2 to 3 metres in diameter and about one-half metre in depth. Fire is applied to the heap in the evening, and in the morning a quantity of liquid sulphur is found to have collected in the bottom of the hole ; this is then ladled out, the combustion being allowed to proceed further until the whole mass is burnt out. By this process only about one-third of the sulphur contained in the ore is obtained, whilst the remaining two-thirds burns away evolving clouds of sulphurous acid. This rough and wasteful process has been greatly improved by increasing the quantity of ore burnt at a given time, the excavation being made 10 metres in diameter, with 1 J. Soc. Chem. Ind. 1889, 313 ; 1890, 118. EXTKACTION OF SULPHUR 339 a depth of 2J metres, and so arranged (on the side of a hill, for instance) that an opening can be made from the lowest portion of the hole so that the sulphur, as it melts, may flow out. These holes are built up with masses of gypsum and the inside covered with a coating of plaster of Paris (see Fig. 98). The calcaroni, as these kilns are termed, are then filled with the sulphur ore which is built up on the top into the form of a cone, and air channels (b b b) are left in the mass by placing large; lumps of the ore together. The whole heap is then coated over with powdered ore (c c), and this again covered with a layer of burnt-out ore, after which the sulphur is lighted at the bottom. By permitting the heat to penetrate very slowly into the mass, the FIG. 98. sulphur is gradually melted, and running away by the opening (a) at the bottom of the heap, is cast into moulds. By this process, which takes several weeks to complete, the richest ores, con- taining from 30 to 40 per cent., may be made to yield from 20 to 25 per cent, of sulphur, whilst common ores, containing from 20 to 25 per cent., yield from 10 to 15 per cent, of sulphur, the remaining portion of the sulphur being used up for combustion. Other methods of extraction by means of solvents, such as bisulphide of carbon, or by the use of ordinary fuel instead of sulphur itself, as a source of heat, have been proposed, but from the nature of the country and its inhabitants these have not yet proved successful in Sicily, and the raw sulphur still remains the cheapest fuel for the purpose. Large deposits of 340 THE NON-METALLIC ELEMENTS sulphur also exist in Iceland, Japan, and Mexico, some of which are worked commercially. 175 Refining of Sulphur. Commercial Sicilian sulphur con- tains about 3 per cent, of earthy impurities which can be removed by distillation, the arrangement being shown in Fig. 99. The sulphur is melted in an iron pot (M) and runs from this by means of a tube into the iron retort (G) where it is heated to the boiling point ; the vapour of the sulphur then passes into the large chamber (A) which has a capacity of 200 cubic metres. FIG. 100. FIG. 99. In this chamber the sulphur is condensed, to begin with, in the form of a light yellow powder termed flowers of sulphur, just as aqueous vapour falls as snow when the temperature suddenly sinks below 0. After a time the chamber becomes heated above the melting point of sulphur, and then it collects as a liquid which can be drawn off by means of the opening (o). It is then cast in slightly conical wooden moulds, seen in Fig. 100, and is known as roll sulphur, or brimstone. It is frequently also allowed to cool in the chamber and then obtained in large crystalline masses, known in the trade as block sulphur. EXTRACTION OF SULPHUR 341 In France, Germany, and Sweden, sulphur is also obtained by the distillation of iron pyrites, FeS 2 . This method, which was described by Agricola in his work De Ee Metallica, depends on the following decomposition of the pyrites : 3FeS 2 = Fe 3 S 4 +S 2 ; and the change which occurs is exactly similar to that by means of which oxygen is obtained from manganese dioxide ; thus : This decomposition of the pyrites is sometimes carried on in retorts, but more generally a kiln similar to a lime-kiln is employed for the purpose, having a hole at the side into which a wooden trough is fastened. A small quantity of fuel is lighted on the bars of the furnace, and then the kiln is gradually filled with pyrites; a portion of the sulphur burns away whilst another portion is volatilized ; the burnt pyrites is from time to time removed from below, and fresh material thrown on the top so that the operation is carried on uninterruptedly. In this way about half the sulphur which is contained in the pyrites can be obtained, whilst only about one-third of the total sulphur can be got by distilling in iron cylinders. Sulphur is likewise obtained in this country, though in smaller quantities, as a by-product in the manufacture of coal-gas. The impure gas always contains sulphuretted hydrogen, which can be removed by passing the gas over oxide of iron, when iron sulphide is formed. This substance on exposure to air is oxidised with separation of free sulphur, thus : Fe 2 S 3 + 30 + H 2 O = Fe 2 O 3 ,H 2 O + 3S. The mass can then be again employed for the purification of the gas, and this alternate oxidization and sulphurization can be repeated until a product is obtained containing 50-60 per cent, of sulphur. The latter may be separated from the iron oxide by distillation or by treatment with carbon bisulphide in a suitable apparatus. In most cases, however, the spent oxide is employed for the manufacture of sulphuric acid, and is then burnt in kilns similar to those employed for burning pyrites. Another and much more important source from which sulphur is now obtained is the residue or waste in the soda manufac- 342 THE NON-METALLIC ELEMENTS ture ; this consists of calcium sulphide mixed with chalk, lime and alkaline sulphides. The sulphur which this material con- tains was formerly altogether wasted ; now, however, it is economically regained in the alkali works on a very large scale. For this purpose the waste, which consists mainly of calcium sulphide CaS, is treated in the presence of water with car- bonic acid gas, calcium carbonate being thus produced whilst sulphuretted hydrogen is evolved : CaS + H 2 O+CO 2 = CaCO 3 + H 2 S. The gas obtained is then burnt with an insufficient supply of air in a specially devised kiln and the sulphur thus recovered in the free and pure state. 1 176 Properties. Sulphur exists in several allotropic modifica- tions. Thus it can be obtained either crystalline or amorphous, and at least two varieties of the amorphous form may be distin- FIG. 101. guished, one of which is soluble and the other insoluble in carbon bisulphide. A form is also known which is soluble in water. Rhombic or a-Sulphur occurs in nature in large yellow transparent octahedra. Fig. 101 shows the form of the natural crystals of sulphur (a being the simpler, and b the more complicated form) which belong to the rhombic system, and have the following rela- tion of the axes : a :b:c = 0'8106 : 1 : T898. In addition to this primary form, no less than thirty different crystallographic modi- fications are known to exist in the case of sulphur. Crystals of rhombic sulphur have also been found in the sulphur chambers, having been deposited by slow sublimation. The specific gravity of this form of sulphur at 0, is 2*05 2'07 ; it is insoluble in water, very slightly soluble in alcohol, benzene, and ether, but dissolves readily in carbon bisulphide, chloride of sulphur, petro- leum, and turpentine. Artificial crystals of sulphur are best ob- 1 Chance, Journ. Soc. Chem. 2nd. (1888), 7, 162. ALLOTROPIC FORMS OF SULPHUR 343 tained from solution in carbon bisulphide, which dissolves at the ordinary temperature about one-third of its weight of sulphur ; the saturated solution on being allowed to evaporate slowly deposits large transparent octahedral crystals. Very well developed crystals can also be obtained by saturating pyridine with sulphuretted hydrogen gas and allowing the solution to .stand exposed to the air for some time. The sulphuretted hydrogen is partially oxidised by the air, and the sulphur which is liberated crystallises out. 1 a-Sulphur melts at 114*5 (Brodie), forming a clear yellow liquid, which has a specific gravity of T803, and when quickly cooled, solidifies again at the same temperature ; generally, however, it remains liquid at a temperature below its melting point, and then solidifies at a slightly lower temperature, which varies with the point to which the liquid has been heated. 177 The second or monosymmetric modification, known as /3- sulphur, is obtained when melted sulphur is allowed to cool at FIG. 102. the ordinary temperature until a solid crust is formed on the surface. The crust is then broken through, and the portion of sulphur still remaining liquid poured out ; the sides of the vessel will then be found to be covered with a mass of long, very thin transparent crystals having the form of mono- symmetric prisms (Fig. 102), the ratio of the axes of which are expressed by the following numbers : 2 a: b : c = T004 : 1 : 1'004. /3-Sulphur appears to be trimorphous, since no less than three different types of crystals, all belonging to the mono- symmetric system, but having different axial ratios and optical properties, have been observed. 3 Monosymmetric sulphur has a specific gravity of 1'96, its melting-point is 120, and like the rhombic modification, it is readily soluble in carbon bisulphide. 178 According to the conditions under which the passage from the fused to the solid state takes place, sulphur may separate 1 Ahrens, Ber. 23, 2708. 2 Mitscherlich, Pogg. Ann. 24, 264. 3 Muthmann, Zeit. Kryst. 17, 336. 344 THE NON-METALLIC ELEMENTS either in the form of rhombic or of monosymmetric crystals. The rhombic crystals are obtained by placing about 200 grams of sulphur previously crystallized from solution in carbon bisul- phide in a flask provided with a long neck, which is afterwards bent backwards and forwards several times to prevent the entry of floating dust. The sulphur is then melted by placing the flask in an oil bath heated to 120, and when the contents are liquid the flask is immersed in a vessel filled with water at 95. On standing for some time at a temperature of about 90, crystals are seen to form, and when a sufficient quantity have been deposited the flask is quickly inverted ; the portion of sulphur still liquid then flows into the neck and there at once solidifies, leaving the transparent rhombic crystals in the body of the flask. When a large mass of molten sulphur is allowed to cool slowly, rhombic crystals are also formed, and these cannot be distinguished from the natural crystals. Thus Silvestri found such rhombic crystals 5 to 6 centimetres in length in a mass of sulphur which had been melted during a fire in a sulphur mine. When a transparent rhombic crystal of sulphur is heated for some time to some temperature above 97'6, but below its melting-point, it becomes opaque when touched with a prismatic crystal, owing to its being converted into a large number of monosymmetric crystals ; l whereas, on the other hand, a transparent crystal of /3-sulphur becomes opaque after standing for twenty-four hours at the ordinary temperature, having undergone a spontaneous change to the rhombic modification, the crystal having been converted into a large number of minute rhombic crystals. This conversion is accelerated by vibration, as, for instance, when the crystals are scratched, and also when they are exposed to sunlight ; the change is always accompanied by an evolution of heat, 2'25 cal. being liberated by the conversion of 31*82 parts of the monosymmetric into the rhombic variety. A saturated solution of sulphur in boiling carbon bisulphide deposits rhombic crystals on cooling, whilst, on the other hand, solutions in alcohol, ether and chloroform give rise to the monosymmetric crystals. A saturated solution in boiling benzene deposits the /^-modification at temperatures between 75 80, and the a-variety below 22, whilst at intermediate temperatures mixtures of the two are formed. 1 Gernez. Com.pt. Rend. 98, 810 915, 100, 1343. ALLOTEOPIC FOKMS OF SULPHUR 345 179 Amorphous sulphur is known in two forms, one of which is soluble in carbon bisulphide and the other insoluble. The soluble variety is produced by the decomposition of sulphuretted hydrogen water in the air and by the action of acids upon the polysulphides of the alkalis, etc. Sulpkur milk (lac sulphur is), & body known to Geber and now used as a medicine, is sulphur in this form. It is deposited as a fine white powder when two parts of flowers of sulphur are boiled with thirteen parts of water and one part of lime slaked with three parts of water, until the whole of the sulphur is dissolved. The reddish-brown solution thus prepared contains calcium pentasulphide, which is decomposed on the addition of hydro- chloric acid with evolution of sulphuretted hydrogen and deposition of milk of sulphur ; thus : CaS 5 + 2HC1 = CaCl 2 + H 2 S + 4S. The insoluble amorphous modification is known as 50 gallons in capacity ; a small quantity of water having been poured into the globe, a stoneware pot was introduced, and on to this a red-hot iron ladle was placed. A mixture of sulphur and saltpetre was then thrown in to this ladle, and the vessel closed in order to prevent the escape of the vapours which were evolved. These vapours were absorbed by the water, and thus sulphuric acid was formed. This product, from the mode of its manufacture, was termed oil of vitriol made by the bell, as contradistinguished from that made from green vitriol, and it cost from Is. 6d. to 2s. 6d. per Ib. Dr. Roebuck of Birmingham was the first to suggest a great improvement, in the use, instead of glass globes, of leaden chambers, which could be constructed of any wished-for size. 1 For a complete account of the manufacture of sulphuric acid see Lunge's excellent treatise on the subject. Gurney and Jackson, London, 1891. 2 See Dossie's Elaboratory Laid Open, 1758, Intro, p. 44. 378 THE NON-METALLIC ELEMENTS Such leaden chambers were first erected in Birmingham in 1746, and in the year 1749 at Prestonpans in Scotland. The mode of working this chamber was similar to that adopted with the glass globes ; the charge of sulphur and nitre was placed within the chamber, ignited, and the door closed. After the lapse of a certain time, when the greater portion of the gases had been absorbed by the water in the chamber, the door was opened, the remaining gases allowed to escape, and the chamber charged again. The leaden chambers first set up were only six feet square, and for many years they did not exceed ten feet square, but in these all the acid employed in the country was manufactured, whilst much was exported to the Continent, where the chamber acid still goes by the name of English sulphuric acid. The first vitriol works in the neighbourhood of London were erected at Battersea in the year 1772, by Messrs. Kingscote and Walker, and in 1783 a connection of the above firm established works at Eccles, near Manchester. This manufactory, the first erected in Lancashire, contained four chambers, each twelve feet square, and four others, each of which was forty-five feet long and ten feet wide. In the year 1 788 a great stimulus was given to the manufacture of sulphuric acid by Berthollet's application of chlorine, dis- covered by Scheele in 1774, to the bleaching of cotton goods, and, from that time to the present, the demand has gradually extended until it has become enormous and almost unlimited in extent. The next improvement in the manufacture consisted in making the process continuous. The foundations of this mode of manufacture appear to have been laid by Chaptal, and the principle employed by him is that which is at the present day in use. The improvements thus proposed were (1) the introduction of steam into the chamber instead of water, (2) the continuous combustion of the sulphur in a burner built outside the chamber, (3) sending the nitrous fumes from the decomposition of nitre placed in a separate vessel, along with the sulphur dioxide gas and air into the chamber. 203 The theory of the formation of sulphuric acid l in the leaden chamber has been the subject of much discussion, and even at the present day the views of chemists differ. The view originally proposed by Berzelius may be simply expressed by saying that although sulphur dioxide in presence of water or 1 Peligot, 1844, Ann. Chim. Phys. [3] 2, 263 ; R. Weber, 1866, Pogg. Ann. 127, 543 ; Raschig. Annalen, 241, 242 ; Lunge, loc. cit. THEORY OF SULPHURIC ACID MANUFACTURE 379 steam is unable rapidly to absorb atmospheric oxygen, it is able to take up oxygen from such oxides of nitrogen, as N 2 3 or NO 2 . If, therefore, these oxides are present in the chamber they give up part of their oxygen to the sulphur dioxide, and are reduced to nitric oxide, NO. This is, however, able to absorb free oxygen, and is at once reconverted into N 2 O 3 or NO 2 . This continuous reaction may be represented as follows : (1) N0 2 + S0 2 + H 2 = H 2 S0 4 + NO. (2) NO + = N0 2 . Another view, founded upon that of Davy, is due to Lunge. According to him, nitrogen peroxide, NO 2 , is not formed in the chambers, and is therefore not the essential agent which brings about the oxidation of the sulphur dioxide, nor does reduction to nitric oxide, NO, usually occur, the active substance being nitrogen trioxide, N 2 3 , (p. 508). In the first instance, the sulphur dioxide combines with nitrogen trioxide, oxygen and water to form nitrosyl-sulphonic acid, S0 2 (OH)NO , according to the equation (1) : (1) 2S0 2 + N 9 O 3 + 0. 2 + H 2 O = 2SO,(OH)N0 . (2) 2S0 2 (OH)N0 2 + H 2 = 2H 2 S0 4 + N 2 O 3 . This substance on meeting with an excess of water vapour is decomposed into sulphuric acid which falls to the bottom of the chamber, and nitrogen trioxide which is ready to react again (equation 2). The existence of nitrosyl-sulphonic acid is well known to the manufacturers of sulphuric acid, since it is formed as a white crystalline substance when the supply of steam has been insufficient, and is termed by them " chamber crystals." It seems probable that in practice both these reactions occur, the formation of sulphuric acid at the commencement of the chambers being accompanied by the presence of nitric oxide, NO, whilst in the subsequent stages the active substance is nitrogen trioxide, NgOg. 1 It is thus clear that according to either theory nitrous fumes act as a carrier between the oxygen of the air and the sulphur dioxide, so that, theoretically, a small quantity of these fumes will suffice to cause the combination of an infinitely large quantity of sulphur dioxide, oxygen, and water to form sulphuric acid. 1 A complete discussion of this somewhat intricate subject will be found in vol. i. chap. x. of Lunge's Sulphuric Acid and Alkali, 1891. 380 THE NON-METALLIC ELEMENTS Practically, however, this is not the case, because instead of pure oxygen, air must be used, and four- fifths of this consists of nitrogen, which so dilutes the other gases that in order to obtain the necessary action a considerable quantity of these oxides of nitrogen must be added. Besides this, nitrogen has to be constantly removed from the chambers, and in its passage Fia. 112. carries much of the nitrous fume away with it, although most of this can, as we shall see, be recovered and used over again. The above reaction can be illustrated on the small scale by the apparatus shown in Fig. 112, in which sulphur contained in the bulb-tube is allowed to burn in a stream of air, supplied from the double aspirator ; the sulphur dioxide and air pass through the wide glass tube into the large glass globe, but carry in on their way the nitrous fumes generated in the small flask (a), SULPHURIC ACID MANUFACTURE 381 from nitre and sulphuric acid. The flask (b) contains boiling water, from which steam passes into the globe. The outlet tube (c) of the globe communicates with a draught. By alternately increasing and diminishing the supply of sulphur dioxide, the disappearance and reappearance of the red nitrous fumes can be readily shown. If the flask be kept dry whilst the two gases are passed in, the white leaden-chamber crystals are seen to be deposited on the glass. When aqueous vapour is admitted, the crystals dissolve with formation of sulphuric acid and ruddy fumes. FIG. 113. 204 The leaden chambers for the manufacture of sulphuric acid are now constructed of a much larger size than was formerly the case, but vary considerably in different works ; they are frequently 30 meters in length, 6 to 7 meters in breadth, and about 5 meters in height, and have therefore a capacity of from 900 to 1,000 cubic meters (about 38,000 cubic feet). The chambers are made of sheet lead weighing 35 kilos per square meter (or 7 Ibs. to the square foot), and soldered together by melting the edges of the two adjacent sheets by means of the oxyhydrogen blow-pipe. The leaden chamber is supported by a 382 THE NON-METALLIC ELEMENTS wooden framework to which the leaden sheets are attached by strips of the same metal, and the wooden framework is generally raised from the ground on pillars of brick or iron and the whole erection protected from the weather, sometimes by a roof, but SULPHURIC ACID MANUFACTURE 383 at any rate by boarding to keep off most of the rain. The space below the chamber is used either for the sulphur burners or for 11. Decim. 10 I , the concentrating pans. The general appearance or bird's-eye view of a sulphuric acid THE NON-METALLIC ELEMENTS chamber is shown in Fig. 113, whilst the arrangement arid con- struction of one of the most complete forms of sulphuric acid plant now in use in this country is shown in Figs. 114, 115, and 116. Three chambers, termed respectively Nos. 1, 2, and 3 (Fig. 114), are placed side by side supported on iron pillars ten feet high. Each chamber has the dimensions already given, and each, therefore, has a capacity of 38,500 cubic feet. A longitudinal section of the chamber (No. 2) in the direction (DC) is shown in Fig. 115, and a sectional elevation in the direc- tion (AB) is shown in Fig. 116. From this last figure it is seen that the roof of the chamber is not horizontal but slightly slanting so as to enable the rain to run off into gutters placed to receive it. 205 Beginning at the first part of the process we find the pyrites-kilns, or burners placed across the ends of the chamber as seen in plan at A, Fig. 114, in longitudinal section and in elevation at A, Fig. 116, and in cross section at A, Fig. 115. The broken pyrites, FeS 2 , is filled, in moderately sized lumps, into the burners, which have previously been heated to red- ness, and when the burning is once started the fire is kept up by placing a new charge on the top of that nearly burnt out. The ordinary charge for each burner of pyrites, containing about 48 per cent, of sulphur, is 6 to 8 cwt., which SULPHURIC ACID MANUFACTURE 385 is burnt out in twenty-four hours, and the kilns are charged in regular succession, so that a constant supply of gas is evolved during the whole time, whilst the quantity of air which enters the kiln is carefully regulated by a well-fitting door placed below. The hot sulphur dioxide, nitrogen, and oxygen gases, are drawn from the pyrites burners, through the whole system of tubes, towers, and chambers, by help of the powerful draught from a large chimney which is placed in connection with the apparatus. These gases first pass from each kiln 'into a central flue, built in the middle of the kiln, and thence into lal : i 15 Meter. I an upright brick shaft through a horizontal earthenware flue, or cast-iron pipe, into the lower part of the square denitrating tower seen in section at G in Fig. 116. This tower, from a to I, is about 45 feet, or 14 meters, in height ; it is built up, from a to c, to a height of 25 feet, or 8 meters, of lead lined with fire brick, and of this about 15 feet, or 5 meters, from d to e, are filled up with pieces of flint. The object of this Glover's tower, or denitrating tower as it is termed, is to impregnate the sulphur dioxide as it comes from the burners with nitrous fumes derived from a later stage of the ^operation. This is effected by allowing strong nitrated acid to flow down the tower together with a stream of chamber-acid. 26 386 THE NON-METALLIC ELEMENTS Strong sulphuric acid, as we shall see, has the power of absorb- ing nitrous fumes, with formation of nitrosyl-sulphonic acid, S0. 2 (OH)(NO. 2 ), and these are given off again when the acid comes into contact with the hot sulphur dioxide from kilns : m fe the 2S0 2 (OH)(N0 2 ) + S0 3SO 2 (OH) 2 + 2NO. SULPHURIC ACID MANUFACTURE 387 Two reservoirs are placed at the top of the Glover's tower ; one containing the strong nitrated acid, the other containing the chamber-acid. Both the strong nitrated and the chamber-acid are allowed to flow down together over the column of flint stones in given proportions, and when the mixture comes into contact with the upward current of hot sulphur dioxide, the nitrous fumes dissolved in the strong acid are given off and swept away, together with the gases from the burners, direct into the chambers. Although the nitrated acid loses its nitrous fumes when diluted with water, it does not do so in presence of chamber-acid of the ordinary-strength, so that the denit rating effect of the Glover's tower depends on the reducing action of the sulphur dioxide, rather than on any dilution by the chamber- acid. This addition is mainly made for the purpose of cheaply concentrating the chamber-acid, for not only does the strong acid lose its dissolved nitrous fumes, with the production of a considerable amount of acid, but the weak chamber-acid coming in contact with the hot dry gases which enter the tower at a temperature of 340, parts with a large quantity of its water, which goes into the chamber as steam, whilst the concentrated acid, falling to the bottom of the tower, flows into a reservoir, z, Fig. 114, placed to receive it. On issuing from the tower, the gas, having now been cooled by contact with the stream of acid to a temperature of about 60, passes into the cast-iron pipe, x, Fig. 114 (4 feet 6 inches in diameter), whence it is delivered at the further end of chamber No. 1 at a height of 8 to 9 feet above the floor. 206 The supply of nitrous fumes, which is needed to act as carrier of the atmospheric oxygen to the sulphur dioxide, is furnished by the three nitre pots seen in plan and in section in Fig. 117. The charges of 30 Ibs., or 13'5 kilos, of nitrate of soda, and 33 Ibs., or 15 kilos, of sulphuric acid, of sp. gr. r?5, are run into the pots from the outside, and after the lapse of two hours, when each charge is exhausted, the fused bisulphate of soda (technically termed sale nixum) is run off into a pan placed on a platform outside the ove-n, and a new charge introduced. The decomposition of the nitre is accelerated by heat from the pyrites burners, placed below the brick arch which separates them from the pots, and the nitric fumes are gathered into a cast-iron pipe, z, Fig. 117, which discharges its contents into the long horizontal main carrying the products from the pyrites burners into the chamber. Here also the 388 THE NON-METALLIC ELEMENTS nitric acid vapour parts with some of its oxygen and is reduced by the sulphur dioxide to the lower oxide which acts as a carrier between the oxygen and sulphur dioxide. The mixture of oxygen, nitrogen, sulphur dioxide, nitrous fumes and vapour of water now meets with steam introduced into the chamber by the tubes, s s, Fig. 115, and the reaction as already described sets in. Having travelled through the length of chamber No. 1, the gases pass by means of the connecting shaft (v) shown in Fig. 116, into the second chamber, where they likewise meet with steam jets, and having passed through this chamber, and having deposited a further amount of liquid FIG. 117. sulpnuric acid, which falls on the floor of the chamber, the gases are drawn into the third or exhaust chamber by the flue (w) shown in Fig. 116. Here, if the process is properly worked, all the sulphur dioxide is converted into sulphuric acid, and red nitrous fumes must always be visible. For the purpose of determining the proper working of the process the percentage of sulphur dioxide contained in the gases entering the first chamber, and that of the oxygen in the gases leaving the third chamber, is regularly ascertained in carefully managed works. The nitrous fumes having been added in excess of the quantity required to convert the SO 2 into H 2 SO 4 still remain in chamber SULPHURIC ACID MANUFACTURE 389 No. 3, and, in order to absorb these, a Gay-Lussac tower (G", Figs. 114 and 116) is employed, the capacity of which ought to be at least one hundredth part of that of all the chambers. This consists, like the Glover's tower, of a square tower 50 feet in height, made of strong lead (78 Ib.) and lined for 35 feet with 2 inch thick glazed fire tiles, and filled with coke. The exit gases from chamber No. 3 are drawn in at the bottom of this coke column, and escape to the chimney by the exit tube (i, Fig. 116) at the top. In their passage they come in contact with a finely divided shower of strong cold acid (sp. gr. 1*75) obtained by concentrating the chamber-acid. This strong sul- phuric acid absorbs the excess of nitrous fumes which would otherwise pass away up the chimney, and having thus become saturated with nitrous fumes, runs away through the spout Jc into reservoirs for the so-called nitrated acid, built under chamber No. 3, the position of which (mm, Fig. 114) is shown on the plan. From these reservoirs the nitrated acid is allowed to run into one of the cast-iron air boilers (nnn) shown on the plan, whence, by air pressure, it is forced up to the cistern on the top of the Glover's tower for employment in the first part of the process as already described. 207 The continuous process of acid making in the chambers is only carried on until the acid has attained a specific gravity of 1'53 to 1*62, or contains 62 70 per cent, of the pure acid, H 2 SO 4 , inasmuch as an acid stronger than this begins to absorb the nitrous fumes. In order to obtain a stronger acid, either the arrangement of the Glover's tower, as described, is employed or, in works where the Glover is not used, the chamber-acid is run into the leaden concentrating pans (po) placed under chamber No. 2, shown in plan in Fig. 114, and in section in Fig. 115. The flame and heated air from the fires (q) play over the surface of the acid contained in these pans, the water passes away in the form of steam, and the strong acid, remains. By this means the acid can be concentrated until it attains a specific gravity of 1*72, or contains 79 per cent, of pure acid ; beyond this degree of concentration the hot acid begins rapidly to attack the lead of the pans, and it therefore cannot be fur- ther evaporated in them. It is then run off into the acid cooler (p), a leaden trough surrounded by cold water, whence it passes into the strong-acid cisterns (r, Fig. 115). In this form the acid is technically known as B. O. V., brown oil of vitriol, as it is always slightly coloured from the presence of traces of 390 THE NON-METALLIC ELEMENTS SULPHURIC ACID MANUFACTURE 391 organic matter, and it is in this condition that it is very largely sold for a great variety of purposes. 208 In order to drive off the remaining portions of water, the acid must be concentrated or rectified in platinum or glass vessels. A common arrangement for concentrating in platinum stills is shown in Fig. 118, as manufactured by Messrs. Johnson, Matthey, and Co., of London. By means of this apparatus no less than 200 cwt. (10,000 kilos) of brown oil of vitriol can be daily concentrated, yielding a product containing 98 per cent, of real acid. The retort or still (A, Fig. 118) consists of plates of platinum, the joints of which are autogenously soldered. This rests on the iron ring (c). The chamber-acid runs from the stopcock through a platinum tube on to the heated thick bottom of the still, where it is quickly concentrated, whilst the aqueous vapour escapes by the head of the still (L). As soon as the level of the concentrated acid reaches the top of the platinum funnel (D), it begins to flow off by means of the tube (E) into the platinum vessel (F), round which a current of cold water circulates. Having been thus 5 cooled, the acid passes into the stoneware jar (H), surrounded by /water, and thence, by means of the lead or stoneware funnel (i) into the reservoir (K). Messrs. Johnson, Matthey, and Co. < have introduced an im- proved form of platinum concentrating apparatus, by means of which all evaporation in leaden pans is avoided, and thus the operation not only considerably cheapened, but the acid ob- tained in a purer condition. This new arrangement is repre- sented in Fig. 119. A A are pans made of platinum plates, which are corrugated at the bottom, and heated by a fire placed below. In these the concentration proceeds until the acid attains a strength of from 78 to 80 per cent, of H 2 S0 4 . It then runs into the retort (B), also having a corrugated surface, and the perfectly concentrated acid which is thus obtained is cooled by passing through the worm (D), made of platinum tube. In many English works the sulphuric acid is rectified in glass and not in platinum vessels. These glass vessels are large retorts made of well-annealed and evenly-blown glass (Fig. 120 a), of such a size as to contain twenty gallons of the acid. Each retort is placed on an iron sand-bath (&), round which the flames from a fire are allowed to play, but so that the flame does not touch the retort. A glass head (c) fits loosely into the neck of the retort, and through this the aqueous vapour 392 THE NON-METALLIC ELEMENTS 1 SULPHURIC ACID MANUFACTURE 39S carrying with it a little acid fume, passes into a condensing box. The plan of a rectifying house containing twenty-four retorts is shown in Fig. 121. The acid having been concentrated in the leaden pans (A A A), passes along the leaden tubes (BB B)^ from which the retorts are filled by means of the upright leaden tubes (d, Fig. 120), which can be bent so as to discharge the acid into the neck of the retort. After the rectification is com- plete the retorts are allowed to cool for twelve hours, and the acid is then drawn out by means of leaden syphons into the stoneware coolers (i t Fig. 120). In order to effect a continuous rectification in glass vessels, the following arrangement has been adopted in some works. Three of the retorts are placed one above the other, as is shown FIG. 120. in Fig. 122. As soon as the acid in retort (B) has attained a specific gravity of 1-84, the retort is connected with a system of syphon tubes (///), and acid of the specific gravity of T74, and having a temperature of 150, is allowed to run into the upper- most retort (D) by means of the stopcock. This acid gradually passes through the three retorts, D, c, and B, and when it has reached the last one it has attained a specific gravity of 1*84, and is allowed to run off through a cooling chamber Qi) into the carboy. Another recent and efficient system of continuous concentration in glass vessels has been patented by Webb, 1 and improved by Levinstein, 2 by means of which acid containing about 96 per cent, of H 2 SO 4 (sp. gr. 1-84) can be obtained in a 1 Eng. patent, 2,343 (1891) ; 17,407 and 18,891. 2 Eng. patent, 19,213 (1892). 394 THE NON-METALLIC ELEMENTS single operation from acid of sp. gr. T625, containing about 71 per cent, of H 2 SO 4 . 209 According to theory, 100 parts of sulphur burnt should yield 306*25 parts of pure sulphuric acid. In practice, however, this theoretical yield is never attained, and for several reasons ; in the first place because a certain amount of loss must neces- sarily take place in working with such enormous volumes of gas, and in the second place inasmuch as an unavoidable loss occurs in the processes of concentration ; and thirdly, owing to FIG. 121. the fact that an amount of sulphur varying from 2 to 5 per cent, remains behind in the burnt ore, and this amount cannot be accurately allowed for. As a general rule a yield of 270 to 280 parts of pure acid from 100 of sulphur is practically considered about the proper production, so that about 5 per cent, of sulphur is lost on the average, of which, however, only a portion passes out in the gaseous form into the air. In cases where special precautions are taken the yield sometimes reaches from 294 to 297, but when the manufacture is not carefully conducted much more serious losses occur. Thus in his eighth annual report SULPHURIC ACID MANUFACTURE 395 {1871) Dr. R. Angus Smith gives (p. 17) a table, showing the total escape of sulphur acids (calculated as sulphuric acid) from twenty-three chemical works. From this it appears that whilst from some of the works no escape of these acids occurs, the average loss of sulphuric acid in the twenty-three works in question is 7*606 per cent, on the total quantity obtainable from the sulphur burnt, and that the loss in the case of four works actually rises to more than 20 per cent., in one case amounting to an escape of 159 Ibs. of sulphuric acid every hour. Facts like these, says the FIG. 122. inspector, dispose of the argument often used by the manu- facturers, that they require the acid, and that it is to their interest to keep it, and of course condense it to the best of their power. Indeed, certain makers are fully aware that they are allowing sulphuric acid to escape in large quantities, but their reply is that it is cheaper to permit a large escape and work rapidly rather than have large chambers and condense the whole of their gases. By the Alkali Act of 1881 the limit of 4 grains per cubic foot has been set to the amount of sulphuric anhydride which may 396 THE NON-METALLIC ELEMENTS be left in the gases escaping from the chambers. The average actually attained is 1*284 grains per cubic foot. 1 The amount, again, of nitrate of soda or Chili-saltpetre used, varies considerably even in the best works, according to the rate at which the reaction is permitted to proceed, and the complete- ness and rapidity with which the nitrous fumes can be recovered in the Gay-Lussac tower and again brought into the chamber. Manufacturers who employ Glover and Gay-Lussac towers use on an average 3'5 to 6'5 parts of nitrate for every 100 of sulphur burnt, whilst at works where these appliances are not in use the quantity of nitre required may rise to from 12 to 13 parts. The larger the quantity of nitrous fumes present in the chamber, the quicker will be the formation of sulphuric acid, and the proportion of fumes which pays best is a question for the manufacturer in each instance to decide. A certain loss of oxides of nitrogen cannot, of course, be avoided ; the fumes are partly not completely con- densed, and pass out by the chimney, and partly, in all prob- ability, reduced by the sulphur dioxide to nitrous oxide or even to nitrogen, which, as they cannot combine again with the atmospheric oxygen, must escape into the air. In order to convert 100 parts of sulphur into sulphuric acid, about 210 parts of water in the form of steam are needed. This steam is costly in its production, but an attempt made by Sprengel to reduce this item of expenditure by employing a jet of water in the form of spray or in a state of very minute division has proved unsuccessful. 21 o None of these processes yield, it must be remembered, chemically pure acid, inasmuch as, in the first place, the water cannot thus be completely removed, and secondly, because im- purities, such as sulphate of lead, arising from the action of the acid on the leaden concentrating pans, and arsenic derived from the pyrites, are not got rid of by this process of simple concentration. In order to prepare pure sulphuric acid, the commercial product must be distilled in a glass retort until one-third has passed over; then the receiver is changed and the acid dis- tilled nearly to dryness. It not unfrequently happens that in this process the acid bumps violently on ebullition, owing to a small quantity of solid lead sulphate being deposited on the bottom of the retort ; the addition of small pieces of platinum foil or wire stops this to a certain extent, but a better preventive is either to heat the retort at the sides rather than 1 Report of Inspector of Alkali Works, 1892, p. 8. PROPERTIES OF SULPHURIC ACID 397 .at the bottom, or, when the ebullition becomes percussive, to allow the liquid to cool, then to pour off the clear acid, leaving the deposit behind, and to proceed with the distillation of the clarified liquid. A slow current of air passed through the boiling acid has also been found efficacious. 211 Properties. The acid thus purified by distillation still con- tains about 1*5 per cent, of water which cannot be removed by this process. If, however, the distillate be cooled, the pure acid containing 100 per cent, of H 2 SO 4 , separates out in the form of crystals which melt at 10 '5. These crystals when once melted generally remain liquid for a considerable time, even when cooled below their freezing point, the liquid only solidifying when it is agitated or when a small crystal of the acid is added, the temperature then rising to 10'5. The specific gravity of the pure liquid acid is l'^3 7 at 15 compared with water at 4 (or, as it is usually expressed, 15/4), 1 whilst according to Lunge and Naef, 2 it is I '8384. When the pure acid is heated, it begins to fume at 30 inasmuch as it then partially decomposes into water and sulphur trioxide. This dissociation increases with increase of temperature until at 338, the boiling-point of the liquid (Marignac), a large quantity of trioxide is volatilized, so that the residue contains from 98'4 to 98'8 per cent, of the real acid, and then this liquid may be distilled without altera- tion. The vapour of sulphuric acid when it is more strongly heated completely decomposes into water and the trioxide. According to Deville and Troost the vapour density at 440 is 25, whilst for equal volumes of aqueous vapour and sulphur trioxide the calculated vapour density is 17-88 + 79-46 When heated still more strongly, the trioxide thus formed itself splits up into oxygen and sulphur dioxide. This decomposition may be readily shown by allowing sulphuric acid to drop slowly into the platinum flask (a, Fig. 123), which is filled with pumice- stone and heated strongly by the lamp ; the mixture of gases which escapes consists of one volume of oxygen to two volumes of sulphur dioxide, which latter gas is absorbed by passing through water containing caustic soda, the oxygen escaping in the free state, whilst any undecomposed sulphuric acid is con- densed in the U-tube and collects in the flask (d). It has been 1 Marignac, Ann. Chim. Phys. [3], 192 ; Mendelejeff, Ber. 17, 2536. 2 Chem. Industrie, 1883, 37. 398 THE NON-METALLIC ELEMENTS proposed to use this process for the preparation of oxygen on the large scale, as the material is cheap and the sulphur dioxide can again be used for the manufacture of sulphuric acid. 212 When sulphuric acid is mixed with water a considerable evolution of heat takes place and a contraction ensues. The amount of heat which is evolved by mixing sulphuric acid and water has been exactly determined by Thomsen l ; his results are given in the following table in which the columns marked I. FIG. 123. give the number of molecules of water, and II. the heat evolved in calories : I. 49-3 , 99-7 H 2 SO 19-1 IT. i- x H,,0 4 . 6336 9355 11294 13020 14851 16147 200-4 401-7 804-4 1609-7 II. + x H 2 0. 16572 16744 16950 17196 17522 17737 From the above numbers it is seen that the addition of the first molecule produces an amount of heat represented by 6,336 thermal units, or about one-third of the total, whilst the addition of two molecules of water gives off about one-half 1 Therm. Unter. III. 34. PROPERTIES OF SULPHURIC ACID 399 the total quantity. The heat evolved by a further addition of water is less for each successive molecule of water, but it has been found impossible to determine the point at which no further evolution of heat is caused by further dilution. Hydrates of Sulphuric Acid. When a mixture of equal mole- cules of acid and water is cooled down, the mixture solidifies to a mass of prismatic crystals, which possess the composition H 2 SO 4 + H 2 O, and melt, according to Pierre and Puchot, at 7'5. Another hydrate of the formula H 2 S0 4 + 4H 2 O, has been isolated by Pickering 1 from a solution of sulphuric acid in water. It forms large, well-defined crystals and melts at 25. This substance is characterised as a definite chemical compound by the facts that it melts and freezes at a definite temperature, and that its freezing point is lowered by the addition of either of its components. The existence of many other hydrates of the acid has been surmised from the properties of its solution in water, but no others have as yet been isolated. 213 Sulphuric acid is largely used in the laboratory not only for the preparation of most of the other acids, but also in con- sequence of its powerful hygroscopic properties for the purpose of drying gases. To effect this, the gas is best led through tubes filled with fragments of pumice-stone which have been boiled in strong sulphuric acid. The acid is also employed to dry solid bodies, or to concentrate liquids, especially in cases where the application of a high temperature is likely to pro- duce a decomposition of the substance, the bodies to be dried being placed over a vessel containing the acid in a closed space or in a vacuum. Sulphuric acid when concentrated does not act in the cold upon many of the metals, although it does so in some cases when heated. Thus copper, mercury, antimony, bismuth, tin, lead, and silver are attacked by the hot acid, with evolution of sulphur dioxide, but are not acted on by the cold dilute acid ; thus : 2Ag 4 2H 2 S0 4 = Ag 2 S0 4 + SO 2 + 2H 2 O. Gold, platinum, iridium, and rhodium are unacted upon, even by boiling sulphuric acid, and this acid is, therefore, employed in the separation of silver and gold. The more easily oxidizable metals, such as zinc, iron, cobalt, manganese, are dissolved by the dilute acid with evolution of hydrogen and formation of a sulphate, but also act upon the hot concentrated acid with production of sulphur dioxide. 1 Journ. Chem. Soc. 1890, i., 1339. 400 THE NON-METALLIC ELEMENTS Many organic bodies are decomposed by sulphuric acid, which abstracts from them the elements of water. Thus, for instance, oxalic acid, C 2 H 2 4 , by heating with strong sulphuric acid is decomposed into carbon dioxide, CO 2 , carbon monoxide, CO, and water, H 2 O ; and alcohol, C 2 H 6 0, is transformed by means of this acid into ethylene gas, C 2 H 4 , and water, H 2 O. Wood, sugar, and other substances are blackened by sulphuric acid, this body withdrawing from them the hydrogen and the oxygen which they contain with production of water. 214 The Sulphates. The salts of sulphuric acid are termed sulphates, and as this acid is dibasic, like sulphurous acid, two series of sulphates exist, viz., the normal salts, such as Na 2 SO 4 and CaSO 4 , and the acid salts such as NaHSO 4 . Many sulphates occur native, existing as well-known and important minerals ; such are : gypsum, CaS0 4 -f 2H 2 O ; heavy spar,BaSO 4 ; celestine,SrSO 4 ; Glauber's salt, Na 2 SO 4 + 10H 2 O; and Epsom salts, MgSO 4 + 7H 2 O. Most of the sulphates are soluble in water, and crystallize well, and these can be readily prepared by dissolving the metal in dilute sulphuric acid, or the oxide or carbonate if the metal does not readily dissolve. Some few sulphates, viz., calcium sulphate and the sulphates of lead and strontium, are only very slightly soluble, whilst barium sulphate is insoluble in both water and dilute acids. This fact is made use of for the detection of sulphuric acid. A soluble barium salt, usually the chloride, is added to the solution supposed to contain a sul- phate ; if sulphuric acid be present, a heavy white precipitate of barium sulphate, BaSO 4 , falls down, which is insoluble in dilute hydrochloric acid. In order to detect free sulphuric acid, together with sulphates, as for instance in vinegar, which is sometimes adulterated with oil of vitriol, the liquid must be evaporated on a water-bath with a small quantity of sugar. If free sulphuric acid is present a black residue is obtained. Free sulphuric acid is found in the water of certain volcanic districts. It has already been mentioned that sulphur dioxide occurs in volcanic gases, and these when dissolved in water gradually absorb oxygen from the air and pass into sulphuric acid. The Rio Vinagre in South America, which is fed from volcanic springs and receives its name on the account of the acid taste of the water, contains free sulphuric acid. A singular occurrence of free sulphuric acid has been noticed in the salivary glands of certain mollusca; thus, according to SULPHURIC ACID 401 Bodeker and Troschel, those of the Dolium galca contain about 2*47 per cent. 215 The following table by Lunge, Isler and Naef 1 exhibits the percentage of real acid, H 2 SO 4 , contained in aqueous sulphuric acid of varying specific gravities. Degree Twaddell Specific gravity 15/4 in vacua Percentage of H 2 S0 4 Degree Twaddell Specific gravity 15/4 in vacua Percentage of H 2 S0 4 1.000 0-09 99 1-495 59-22 3 1-015 2-30 102 1-510 60-65 6 1-030 4-49 105 1-525 62-06 9 1-045 (5-67 108 1-540 63-43 12 1-060 8-77 111 1*555 64-67 15 1-075 10-90 114 1-570 65-90 18 1 090 12-99 117 1-585 67-13 21 1-105 15-03 120 1-600 68-51 24 1-120 17-01 123 1-615 69-89 27 1-135 18-96 126 1-630 71-16 30 1-150 20-91 129 1-645 72-40 33 1-165 22-83 132 1-660 73-64 36 1-180 24-76 135 1-675 74-97 39 1195 26-68 138 1-690 76-30 42 1-210 28-58 141 1-705 77-60 45 1-225 30-48 144 1720 78-92 48 1-240 32-28 147 1-735 80-24 51 1-255 34-00 150 1-750 81-56 54 1-270 35-71 153 1-765 82-88 57 1-285 37-45 156 1-780 84-50 60 1-300 3919 159 1-795 86-30 63 1-315 40-93 162 1-810 88-30 66 1-330 42-66 165 1-825 91-00 69 1-345 4428 166 1-830 92-10 72 1-360 45-88 167 1-835 93-43 75 1-375 47-47 168 1-840 95-60 78 1-390 49-06 1-8410 97-00 81 1-405 50-63 1-8415 97-70 84 1-420 52-15 1-8410 98-20 87 1-435 53-59 1-8400 99-20 90 1-450 55-03 1.8390 99-70 93 1-465 56-43 1-8384 100-00 96 1-480 57-83 1 Sulphuric Acid and Alkali, vol. i. p. 119 (1891). 27 402 THE NON-METALLIC ELEMENTS FUMING SULPHURIC ACID. 216 This substance, which is a solution of varying quantities of sulphur trioxide in sulphuric acid, was known before the sulphuric acid manufactured from sulphur, being termed Nord- Tiausen sulphuric acid, from the fact that it was prepared at Nordhausen in the Hartz, by heating roasted green vitriol. Preparation. (1) When green vitriol or ferrous sulphate, FeSO 4 + 7H 2 O, is roasted in the air it loses water and becomes oxidized to a basic ferric sulphate, Fe 2 S 2 O 9 , which is then FIG. 124. further heated in clay retorts, as shown in Fig. 124, when the following decomposition takes place : Fe 2 S 2 O 9 = 2S0 8 + Fe 2 3 . The sulphur trioxide thus formed, partly combines with the water which is still present to form sulphuric acid, whilst the other portion of the trioxide dissolves in the sulphuric acid thus produced. Fuming sulphuric acid is now almost entirely prepared in Bohemia in the works of J. D. Starck. The solution of green vitriol obtained by the oxidation of the pyrites is evaporated down, and the residue ignited, care being taken that the " vitriol- stone " thus obtained is as free as possible from ferrous sulphate* inasmuch as if this body be present sulphur dioxide is formed FUMING SULPHURIC ACID 403 in the subsequent distillation, and this carries away with it large quantities of the easily volatile trioxide ; thus : 2FeS0 4 = Fe 2 3 + SO 3 + SO 2 . The more completely the vitriol-stone is oxidized, the larger is the yield of fuming acid, which on an average amounts to from 34 to 50 per cent. In some works the green vitriol is allowed to crystallize out, and the oxidized mother liquors alone used for the production of the fuming acid. Fuming sulphuric acid was formerly chiefly used in the arts for dissolving indigo ; at present, however, it is largely employed in the preparation of the soluble sulphonic acids of colouring matters, and in the manu- facture of artificial alizarin. For this purpose an acid is needed which contains more trioxide than the commercial substance, and hence it is prepared specially by the manufacturers themselves by heating the fuming acid in cast-iron retorts and absorbing the trioxide, which is given off, in another portion of the acid in well-closed receivers. (2) Sulphur trioxide is formed, as has been already stated (p. 374), when a mixture of oxygen and sulphur dioxide is passed over heated platinum sponge. It has been proposed to employ this reaction for' the preparation of common sulphuric acid on the large scale, the sulphur dioxide being obtained by the combustion of sulphur, or the roasting of pyrites, and this together with air passed over heated platinized asbestos, the fumes of the trioxide being collected in water. It was found that this process cannot be practically carried out, inasmuch as the platinum soon loses this peculiar property, probably owing to the fact that dirt and particles of dust collect on the surface of the metal. For the production however of a strongly fuming acid, the following process, according to Winkler, 1 answers well. Common sulphuric acid, as has been shown, decomposes on heating, into aqueous vapour, sulphur dioxide, and oxygen. If the mixture of gases be washed by sulphuric acid in order to remove the water and particles of dust, and then the mixture of dioxide and oxygen passed over heated platinized asbestos the trioxide is formed and may be collected in sulphuric acid. Properties. Fuming sulphuric acid is a colourless, thick, oily liquid when pure, but is generally coloured slightly brown from the presence of organic matter. It has a specific gravity of from 1*86 to T89, and evolves on exposure to the air dense white 1 DingL Polyt. Journ. 218, 128. 404 THE NON-METALLIC ELEMENTS fumes, inasmuch as the volatile trioxide escapes and combines with the aqueous vapour of the air to form sulphuric acid. When the fuming acid is cooled, white crystals of the com- pound, H 2 SO 4 + SO 3 , separate out. This substance melts, ac- cording to Marignac, at 35, fumes strongly in the air, and decomposes easily on heating into its constituents. The name disulphuric acid or pyrosulpkuric acid, H 2 S 2 O 7 , has been given to this substance as it forms a series of very stable salts ; thus sodium disulphate, Na 2 S^O 7 , is obtained by heating the acid sodium suh>hate, HNaSO 4 , so long as water is given off; thus : SQ fONa S 2 \ONa When still more strongly heated this salt decomposes into the normal sulphate and sulphur trioxide. Sulphur trioxide and sulphuric acid also unite together to form tdra-sulphuric acid, 3 SO 3 + H 2 SO 4 = H 2 S 4 13 , an oily liquid, and another compound, having the composition S0 3 + 3H 2 S0 4 , a transparent crystalline mass melting at 26. CHLORIDES AND BROMIDES OF SULPHURIC ACID. CHLOKOSULPHONIC ACID, OR SULPHTJRYLHYDROXYCHLORIDE. = 115-65. 217 Williamson first obtained this substance by the direct union of hydrochloric acid and sulphur trioxide. 1 It is, however, best prepared by the distillation of a mixture of concentrated sulphuric acid and phosphorus oxychloride, thus : Hence it is seen that chlorosulphonic acid may be considered to be sulphuric acid, in which the group hydroxyl, OH, is replaced by chlorine. It is a colourless liquid, fuming strongly in the 1 Proc. Roy. Soc. 7, 11. SULPHURYL CHLORIDE 405 air, having a specific gravity of 1*766 at 18, and boiling at 158. Its vapour decomposes on heating, and at 216 its vapour density is found to be 32'8, or the dissociation is nearly perfect. This decomposition is probably represented by the equation : l 201. S0 3 H = S0 2 + C1 2 + H 2 + S0 3 . When thrown into water it decomposes with explosive violence, forming hydrochloric and sulphuric acids, and when added to strong sulphuric acid, disulphuric acid and hydrochloric acid are formed, thus : OH r S0 2 gl H + s02 )OH r r 4 ! SULPHURYL CHLORIDE. SO \ ^i =133*96. ( vl 218 This body was first obtained, mixed with ethylene di- chloride by Regnault in the year 1838 by the action of chlorine upon a mixture of olefiant gas and sulphur dioxide. 2 Sulphuryl chloride is also formed when a solution of the two gases in anhydrous acetic acid is allowed to stand. It is, however, most readily obtained by heating chlorosulphonic acid in closed tubes at a temperature of 180 for 12 hours; 3 thus : r OH ci r OH. or by saturating camphor with sulphur dioxide and then passing in chlorine, the camphor remaining unaltered. 4 Sulphuryl chloride is a colourless liquid boiling at 70, possess- ing a strongly pungent odour, fuming strongly in the air and having a specific gravity of 1'659 at 20; it decomposes in presence of a small quantity of water into chlorosulphonic acid and hydro- chloric acid, and with an excess of water into sulphuric acid and hydrochloric acid ; thus : S0 2 f 2H 2 = S0 2 1 Herrmann and Kochlin, Bcr. 18, 604. 2 Ann. Chim. Phys. (2) 69, 170. 3 Behrend, Bcr. 8, 1004. 4 Schulze, J. Pr. Chem. (2) 23, 351. 406 THE NON-METALLIC ELEMENTS The vapour density of sulphuryl chloride is normal at 100, but falls to one half of this at 442, at which temperature it is completely dissociated into sulphur dioxide and chlorine. 1 The affinity between the sulphur dioxide and chlorine is very weak, and the latter is very readily withdrawn by any com- pound having an attraction for chlorine ; sulphuryl chloride is, therefore, frequently used as a chlorinating agent in organic chemistry 2 DISULPHURYL CHLORIDE. S 2 O 5 C1 2 = 213-42. 219 The chloride of disulphuric acid was first prepared by Rose 3 by the action of chloride of sulphur on sulphur trioxide, thus : S 2 C1 2 + 5SO 3 = S 2 5 C1 2 + 5SO 2 . The same compound has also been obtained by Michaelis 4 by heating sulphur trioxide with phosphorus oxychloride ; thus : 6S0 3 + 2POC1 3 =3S. 2 5 C1 2 + P 2 5 . It is likewise formed when common salt is heated with sulphur trioxide, and when this latter substance is brought in contact with sulphuryl chloride ; thus : SOJ C1 ' ^* \ 01. It is colourless fuming liquid, boiling at 146, and having a specific gravity at 18 of 1*819. Water decomposes it into sulphuric acid and hydrochloric acid. SULPHUR OXYTETRACHLORIDE. S 2 O 3 C1 4 = 252*04. 220 Millon first obtained this substance by the action of moist chlorine upon sulphur or chloride of sulphur. It is best pre- pared by cooling down a mixture of chloride of sulphur and chlorosulphonic acid to 15, and then saturating the liquid with chlorine, when the following decomposition takes place : S0 3 HC1 + SC1 4 = S 2 3 C1 4 + HC1. 1 Herrmann and Kbchlin, Ber. 16, 602. 2 Armstrong, Proc. Chem. Soc., 1891, 60. 3 Pogg. Ann. 44 291. 4 Zeitsch. Chcm, (2) 7, 149. SULPHUR SESQUIOXIDE 407 Sulphur oxytetrachloride forms a white crystalline mass, which has a very pungent smell and attacks the mucous membrane violently. It dissolves in water with a hissing noise, forming hydrochloric, sulphuric, and sulphurous acids. Exposed to moist air, it deliquesces with evolution of chlorine, hydrochloric acid, and sulphur dioxide, leaving a residue of thionyl chloride and disulphuryl chloride. When the compound is heated it partially sublimes in fine white needles, whilst another portion decomposes into sulphur dioxide, chlorine, thionyl chloride, and disulphuryl chloride. When kept in closed tubes this body liquefies with the formation of thionyl and sulphuryl chlorides ; thus : S 2 3 C1 4 = SOC1 2 + S0 2 C1 2 . An oxy-chloride, S 2 OC1 4 , is formed as a red liquid, boiling at 60, by the action of chloride of sulphur on sulphuryl chloride. 1 SULPHURYL BROMIDE. SO 2 j ^' = 222-30. 221 Bromine combines with sulphur dioxide, forming a volatile white solid crystalline mass, which when acted upon with silver oxide forms sulphur trioxide; 2 thus: S0 2 Br 2 + Ag 2 O = S0 3 + 2AgBr. Some doubt has been thrown on the existence of this com- pound by later observers. 3 SULPHUR SESQUIOXIDE. S 2 3 = 111-28. 222 So long ago as the year 1804 Buchholz found that when sulphur is heated with fuming sulphuric acid an intensely blue- coloured solution is formed, and in the year 1812 F. C. Vogel showed that this blue body is also produced by the action of sulphur on sulphur trioxide. In later years this subject has frequently attracted the attention of chemists, but the nature of the blue substance remained unexplained until R. Weber 4 showed that it consists of a new oxide of sulphur. In order to prepare this substance, carefully dried flowers of sulphur are added, in small quantities, to recently prepared and liquid sulphur trioxide, a fresh quantity of sulphur only being added when that already present has entered into combination. 1 Ogier, Compt. Rend., 94, 446 ; Bull. Soc.Chem. 37, 293. 2 Odling, Journ. Chcm. Soc. 1855, 2. a Michaelis, Lehrbuch, i., 733 (5th Edition). 4 Pogg. Ann. 156, 531. 408 THE NON-METALLIC ELEMENTS In order to moderate the reaction, the test-tube in which the solution is made must be placed in water at a temperature of from 12 to 15. The sulphur on falling into the trioxide dis- solves in the form of blue drops which sink down to the bottom of the test-tube and then solidify. As soon as a sufficient quantity of this substance has been formed, the supernatant sulphur trioxide is poured off and the residue removed from the test-tube by very gently warming it. Sulphur sesquioxide forms bluish-green crystalline crusts, in colour closely resembling malachite. At the ordinary temperature it slowly decomposes into sulphur dioxide and free sulphur, and this decomposition takes place more readily when the substance is warmed ; thus : 2S 2 3 = 3S0 2 + S. This compound dissolves in fuming sulphuric acid, giving rise to a blue solution which on the addition of common sulphuric acid gradually changes to a brown. Water decomposes the sesqui- oxide with the separation of sulphur and the formation of sulphuric acid, sulphurous acid, and thiosulphuric acid. HYPOSULPHUROUS ACID. H 2 S 2 O 4 . 223 This compound was discovered by Schiitzenberger 1 who gave it the name of hydrosulphurous acid and assigned to it the formula H 2 SO 2 , its sodium salt being formulated as NaHSO 2 . Bernthsen, 2 however, subsequently showed that the simplest formula of the sodium salt is NaS0 9 , the molecular formula being probably double this, Na 2 S 2 O 4 , and the free acid H 2 S 2 4 . The zinc salt of the acid is obtained by the action of metallic zinc upon an aqueous solution of sulphur dioxide contained in a closed vessel ; no evolution of hydrogen takes place, but the zinc dissolves and a salt of hyposulphurous acid is formed, the reaction being the following (Bernthsen) : Zn + 2S0 2 = ZnS 2 O 4 . The liquid thus obtained, which was looked upon by Schiitzenberger as the free acid, possesses powerful reducing properties. It bleaches organic colouring matters more quickly than sulphurous acid and precipitates the metals silver and 1 Compt. Rend. 69, 169. 2 Annalcn, 208, 142 ; 211, 285. HYPOSULPHUEOUS ACID 409 mercury from solutions of their soluble salts. None of the salts have been prepared pure, but a solution of the sodium salt may be obtained by the action of zinc on a solution of acid sodium sulphite, the liquid, which is contained in a well closed bottle, being kept well cooled by cold water. The following reaction takes place (Bernthsen) : 4NaHSO 3 + Zn = ZnSO 3 + Na 2 SO 3 + H 2 O + Na 2 S 2 O 4 . A portion of the zinc sulphite is converted into a basic salt and sulphur dioxide thus set free, which is then reduced in its turn by the excess of zinc present. The greater portion of the sodium sulphite crystallizes out along with the sulphite and basic sulphite of zinc, but some still remains in solution. In order to remove this, the mother liquor containing the hypo- sulphite is poured off into a flask and three or four times its bulk of strong alcohol added ; the alcoholic solution, on standing well corked, deposits a second crop of crystals of zinc-sodium sulphite, and the supernatant liquid on again being poured off into a well-stoppered flask, crystallizes into a mass of colourless crystals of the hyposulphite which only needs to be pressed between blotting paper or between a cloth and dried in a vacuum. These crystals however only contain about 40 per cent, of sodium hyposulphite, much of the sulphur being present as sulphate, sulphite and thiosulphate. (Bernthsen.) Sodium hyposulphite is also formed when a current of electricity is passed through a solution of acid sodium sulphite, the h}^drogen, which is evolved at the negative pole, reducing the salt. Sodium hyposulphite is employed by the dyer and calico-printer for the reduction of indigo, as it possesses the same reducing properties as the free acid. In the moist state or in solution when exposed to the air it absorbs oxygen quickly and changes at once into sodium metabisulphite. Na 2 S 2 4 + O = Na 2 S 2 5 . The aqueous solution decomposes, even when not exposed to the air, with the formation of sodium thiosulphate. In order to prepare hyposulphurous acid, a dilute solution of oxalic acid is added to a solution of a hyposulphite ; a yellow liquid is then obtained which soon decomposes, thiosulphuric acid being formed, and this being very unstable decomposes into sulphur and sulphur dioxide. 410 THE NON-METALLIC ELEMENTS The ease with which hyposulphurous acid undergoes decompo- sition accounts for the fact that the existence of this compound was so long overlooked. Berthollet showed so long ago as 1789 that iron dissolves in aqueous sulphurous acid without any evolution of gas, and Fourcroy and Vauquelin found in 1798 that zinc and tin also act in a similar manner. That a lower product of oxidation of sulphur is thus formed was even then known, but up to the time of Schiitzenberger's discovery it was supposed that the product of the reaction was thiosulphuric acid. 224 According to Schtitzenberger's formula the acid may be looked upon as derived from the unknown anhydride SO, H 2 SO 2 = H 2 O + SO ; whilst, if Bernthsen's view be correct, it must correspond to the anhydride S 2 O 3 ; H 2 S 2 4 = H 2 + S 2 3 . Bernthsen has succeeded in showing that for every two atoms of sulphur in the form of hyposulphite*, one atom of oxygen ig required to oxidise the substance to a sulphite, which may be effected by means of an ammoniacal solution of copper sulphate, and three atoms of oxygen to oxidize it to a sulphate, a reaction which may be realised by the use of iodine solution. Both of these results are in agreement only with the formula H 2 S 2 O 4 , as will be seen from a comparison of the following equations : (1) S 2 3 + O = 2S0 2 2SO + 2O = 2SO 2 (2) S 2 O 3 -f 3O - 2SO 3 2SO + 40 SULPHUR HEPTOXIDE, S 2 7 , AND PERSULPHURIC ACID, H 2 S 2 8 . 225 When a mixture of dry oxygen and sulphur dioxide is subjected to the silent electrical discharge, combination takes place and sulphur heptoxide is produced. 1 A substance of the same composition is formed at the positive pole during the electrolysis of sulphuric acid of 40 per cent, and 1 Berthelot, Ann. Chim. Phys. (5) 14, 345, 363. PERSULPHURIC ACID 411 also when hydrogen peroxide is mixed with concentrated sulphuric acid. The anhydrous substance is a viscid liquid which solidifies at forming granules, needles or scales. It is readily volatile and gradually decomposes on preservation, rapidly on heating, into sulphur tri oxide and oxygen. When brought into water, it decomposes with evolution of oxygen, according to the equation : 4H 2 + 2S 2 O 7 = 4H 2 SO 4 + 2 . The corresponding acid has not been isolated, but its salts have been prepared by Marshall l and also studied by Berthelot. 2 The potassium salt, K 2 S 2 O 8 , may be obtained by passing a current of 3 to 3J amperes through a saturated solution of potassium hydrogen sulphate. The solution is placed in a platinum basin, cooled externally by a current of water, and connected with the positive pole of the battery ; the negative pole consists of a platinum wire and is placed in a porous cell filled with dilute sulphuric acid and suspended in the solution of potassium hydrogen sulphate. The potassium salt separates out in the course of 24 to 48 hours as a mass of white crystals. These may be recrystallised by dissolving in warm water and allowing to cool, and then form large, tabular, apparently asymmetric crystals. One hundred parts of water at dissolve 177 parts of the salt, the solution being neutral to test paper. On preservation decomposition commences after some time and becomes rapid when the solution is heated, oxygen being evolved and potassium sulphate formed. When the dry salt is heated it loses oxygen and sulphur trioxide and is converted into the sulphate : 2K 2 S 2 8 = 2K 2 S0 4 + 2S0 3 + 2 . The molecular formula of the salt 3 is found, from a determina- tion of the electrical conductivity of its dilute aqueous solution, to be most probably K 2 S 2 O 8 and not KSO 4 . The solution of potassium persulphate has strongly oxidising properties ; thus ferrous sulphate is rapidly converted into the ferric salt, solutions of silver, copper, manganese, cobalt and nickel salts, all yield precipitates of the higher oxides on warming, hydrochloric i Journ. CJiem. Soc. 1891, i. 771, a Ann. Ghim. Phys. [6] 26, 526. 3 Lowenherz, Ckem. Zcit., 16, 838. 412 THE NON-METALLIC ELEMENTS acid evolves chlorine, potassium iodide is decomposed with liberation of iodine, and alcohol is converted into aldehyde. The ammonium salt, (NH 4 ) 2 S 2 O 8 , is prepared in a similar manner to the potassium salt ; it forms lozenge-shaped, ap- parently monosymmetric tablets, and is much more soluble than the potassium salt, 100 parts of water at dissolving 58'2 parts. The barium salt, BaS 2 8 + 4H 2 O, is freely soluble in water, whilst the lead salt, PbS. 2 O 8 -|-3H 2 0, is deliquescent. All the persulphates appear to be soluble in water, the potassium salt being less soluble than any other which has yet been examined. THIOSULPHURIC ACID. H 2 S 2 3 . 226 This compound is better known under its old name of " hyposulphurous acid," with which name however, we now designate the body obtained by the reduction of sulphurous acid. Thiosulphuric acid has not been prepared in the free state, but it forms a series of stable salts which are known as the thiosulphates (hyposulphites). When dilute sulphuric acid is added to a solution of a thiosulphate, the solution remains, to begin with, perfectly clear ; but sulphur soon begins to separate out as a white, very finely divided powder, and the solution is found to contain sulphurous acid, sulphur dioxide being also evolved ; the thiosulphuric acid undergoes on liberation the following decomposition : This decomposition is quantitative when the sulphur dioxide is removed from the solution, but when it is retained, a limit short of complete decomposition is attained, and other products, among which is pentathionic acid, are formed. 1 Thiosulphuric acid is formed in small amount by the action of sulphurous acid upon flowers of sulphur at the ordinary tem- perature, more rapidly at 80 90. 2 The thiosulphates are formed in various ways : thus, for instance, sodium thiosulphate (formerly called hyposulphite of soda), which was first prepared by Chaussier in 1799, but after- 1 Colefax, Journ. Chcm. Soc. 1892, i. 176. 2 Jo urn. Chem. Soc. 1892, i. 199. THIOSULPHURIC ACID 413 wards more carefully examined by Vauquelin, is formed when sulphur dioxide is passed into a solution of sodium sulphide; thus : (a) SO, + H 2 O + Na 2 S = Na 2 S0 3 + SH 2 . (b) SCX + 2SH 2 = 2H 9 O + 3S. (c) Na 2 S0 3 + S = Na 2 S 2 O 3 . The same salt is also formed according to equation (c) when a solution of sodium sulphite is boiled with flowers of sulphur and this is the method by which the salt is usually prepared (Vauquelin). Sodium thiosulphate is also formed when iodine is added to a solution of sodium sulphite and sodium sulphide ; thus : 1 Na 2 S0 3 + Na 2 S + 1 2 = Na 2 S 2 O 3 + 2NaI. Thiosulphates are also produced by the oxidation of sulphides, either by the oxygen of the air or by means of suitable oxidising agents. These various methods of preparation point out that thiosul- phuric acid is formed by the addition of sulphur to sulphurous acid, just as sulphuric acid is formed by the addition of oxygen to the same substance. Thiosulphuric acid may, therefore, be regarded as sulphuric acid in which one atom of oxygen is r OTT replaced by sulphur, and its formula is accordingly S0 2 -< ^TT' The decompositions which its salts undergo bear out this inter- pretation of its composition. 2 Thus the decomposition of the free acid into sulphur and sulphurous acid has already been mentioned ; and when a solution of sodium thiosulphate is treated with sodium amalgam, sodium sulphite and sodium sulphide are formed (Spring); thus: Na 2 S 2 3 + 2Na = Na 2 SO 3 + Na 2 S. Again, when silver thiosulphate is warmed with water, black sulphide of silver separates out, and the solution contains free sulphuric acid ; thus : And, moreover, when a solution of sodium thiosulphate is treated 1 Spring, Her. 7, 1157. 2 Schorlemraer, Journ. Chcm. Soc. 1869, 256. 414 THE NON-METALLIC ELEMENTS with cobalt chloride, black sulphide of cobalt is precipitated ; thus : S0 2 + CoCl + H 2 = S0 + CoS + 2NaCl. This formula is also borne out by the fact that the isomeric sodium potassium sulphites (p. 372) are converted by the action of ammonium polysulphides into isomeric thiosulphates, which differ in solubility and specific gravity, and yield different products when acted upon by ethyl iodide. 1 Sulphite. Thiosulphate. X OK X OK S0 2 < SO / J \Na x SNa /ONa /ONa S0 2 < S0 2 \K The soluble thiosulphates generally crystallise well, and contain water of crystallisation, the last molecule of which is very difficult to remove by heat, being generally lost at such a high temperature that the decomposition of the salt has already com- menced. Hence it was formerly supposed that the thiosulphates all contained hydrogen. The thiosulphates exhibit a great tendency to form double salts ; those of the thiosulphates insoluble in water are found to dissolve in an aqueous solution of sodium thiosulphate, which also has the power of dissolving other insoluble salts, such as silver chloride, silver bromide, silver iodide, lead iodide, lead sulphate, calcium sulphate, &c ; thus : .O.+ AgCl= J Sodium silver thiosulphate forms distinct crystals, having the composition AgNaS 2 3 -f H 2 O, and these are distinguished by possessing a sweet taste. The use of sodium thiosulphate for fixing prints in photography, first suggested by Sir John Herschel, probably depends on the formation of this salt. The silver chloride with which the photographic paper is impregnated when exposed to the light becomes blackened, the chloride 1 Sch wicker, Ber. 22, 1733. DITHIONIC ACID 415 undergoing a chemical change, after which it is insoluble in sodium thiosulphate. In order, therefore, to fix such a photo- graphic print, it is only necessary, after exposure to light, to soak the paper in a bath of the thiosulphate ; the unaltered chloride of silver dissolves, and the picture, on washing, is found to be permanent. Several other complex thiosulphates of silver and sodium are also known. The thiosulphates are distinguished from the sulphites, inasmuch as when dilute hydrochloric acid or sulphuric acid is added to their solution not only is sulphur dioxide given off as a gas, but free sulphur is deposited as a white powder. On adding a thiosulphate solution to a silver or lead salt, a white precipitate soluble in excess of the thiosulphate solution is produced, whilst with a mercuric salt, a white precipitate of the insoluble thiosulphate is first thrown down, but this quickly becomes dark, and finally black, from its decomposition into a metallic sulphide and sulphuric acid. Solutions of mercurous salts give, with the thiosulphates, dense black precipitates, and when a thiosulphate is boiled with an ammoniacal solution of a ruthenium salt, the solution becomes of such an intensely dark red colour that in the concentrated condition it appears almost black. Ferric chloride added to a solution of a thio- sulphate gives a dark violet coloration, which soon disappears, sulphur being deposited. DITHIONIC ACID. H 2 S 2 6 . 227 This acid, formerly called hyposulphuric acid, was dis- covered by Welter and Gay-Lussac in 1819. The manganese salt of the acid is prepared by passing sulphur dioxide into water containing manganese dioxide in suspension ; thus : 2 S0 2 + Mn0 2 = MnS 2 O 6 . At the same time a portion of the manganese is converted into manganese sulphate ; thus : In order to obtain the dithionate free from sulphate, advantage is taken of the solubility of barium dithionate in water ; baryta 416 THE NON-METALLIC ELEMENTS water, Ba (OH.) 9 , is added until all the metal is precipitated as manganese hydroxide, Mn (OH) 2 , and all the sulphuric acid is thrown down as insoluble barium sulphate, BaSO 4 ; on evapora- ting and cooling, barium dithionate, BaS 2 6 +2H 2 0, crystallises out, and when this is decomposed by the requisite quantity of dilute sulphuric acid, a solution of dithionic acid is obtained. This solution may be concentrated in vacuo over sulphuric acid until it has a specific gravity of T347, but on attempting to concentrate it further, the acid is resolved into sulphur dioxide and sulphuric acid ; thus : H 2 S 2 6 = S0 2 + H 2 S0 4 . Dithionic acid is also formed in small quantity by the oxidation of sulphurous acid by means of potassium permanganate. 1 In order to obtain the salts of this acid we may add the cor- responding base to the acid solution, or they may be obtained more simply by adding a soluble sulphate to barium dithionate. Only the normal salts are known, most of which crystallise well, and they contain, with the exception of the potassium salt, water of crystallisation. Their aqueous solutions are not oxidised in the cold either by atmospheric oxygen, by nitric acid, or by potassium permanganate ; though when they are heated with these oxidising agents they are converted into the sulphates. On heating they decompose partially at 100, and entirely at a higher temperature, into sulphur dioxide, and a sulphate which remains behind ; when sodium amalgam is added to a solution of sodium dithionate, two molecules of sodium sulphite are formed (R. Otto) ; thus : SO 2 'ONa Na.S0 2 .ONa. + 2Na = S0 2 'ONa Na.S0 2 .ONa. The dithionates are distinguished from the thiosulphates, inas- much as they evolve sulphurous acid when heated with hydro- chloric acid without separation of sulphur, whilst the solution afterwards contains a sulphate. TRITHIONIC ACID. H 2 S 3 6 . 228 In 1842, Langlois obtained the potassium salt of this acid 1 Ann. CMm. Phys. (3) 55, 374. TRITH10NIC ACID 417 by gently heating a solution of acid potassium sulphite with sulphur ; thus : S 2 + 6KHS0 3 = 2K 2 S 3 6 + K 2 S 2 O 3 + 3H 2 O. The same salt is also produced when a solution of potassium thiosulphate is saturated with sulphur dioxide ; thus : 3SO 2 + 2K 2 S 2 O 3 = 2K 2 S 3 O 6 + S. Further changes also occur, the trithionate combining with the sulphur which is liberated to form tetra- and pentathionate of potassium. (Debus.) The potassium salt is moreover formed when potassium silver thiosulphate is heated with water; thus : When iodine is added to a solution of sodium thiosulphate and sodium sulphite, sodium trithionate is likewise formed (Spring) ; thus : Na 2 S 2 O 3 + Na 2 SO 3 + I 2 = Na 2 S 3 O 6 + 2NaI ; whilst a solution of the trithionate treated with sodium amalgam or caustic potash, decomposes again into sulphite and thiosulphate. In order to prepare the free trithionic acid, hydro-fluosilicic acid is added to a solution of the potassium salt, when the insoluble hydro-fluosilicate of potassium is precipitated. The aqueous acid thus obtained has no smell, but has a strong acid .and bitter taste ; it may be concentrated in a vacuum up to a certain point, but it is an unstable compound, and at the ordinary temperature easily decomposes even in dilute solution into sulphur, sulphur dioxide, and sulphuric acid. The only one of the trithionates which is well known is the potassium salt. This on heating decomposes into sulphur, sulphur dioxide and potassium sulphate and its solution undergoes the same decom- position on standing : K 2 S 3 6 = K 2 S0 4 + S0 2 + S. In this case again secondary reactions occur, tetra- ana penta- thionates being formed by the combination of the trithionate with the sulphur. Its solution is not precipitated by barium 28 418 THE NON-METALLIC ELEMENTS chloride in the cold, though on heating barium sulphate separates out and sulphur dioxide is evolved, whilst silver nitrate gives a yellow precipitate which very quickly becomes black on standing. Sulphuretted hydrogen has no action upon solutions of the free acid but decomposes the potassium salt with formation of sulphate and thiosulphate of potassium and deposition of sulphur (Debus) : 2K 2 S 3 6 + 5H 2 S = K 2 S0 4 + K 2 S 2 O 3 + 5H 2 O -f 8S. The constitutional formula (p. 423) of potassium trithionate is probably the following (Debus) : K.S0 2 .0 KO.SCL TETRATHIOXIC ACID. H 2 S 4 6 . 229 Fordos and Gelis in 1843 first prepared this acid and its salts. They obtained the sodium salt by adding iodine to an aqueous solution of sodium thiosulphate ; thus : (ONa T 2 ) S I * = S0 2 O Salts of this acid are also formed by the action of copper sulphate upon barium thiosulphate and by the action of sul- phuric acid upon a mixture of lead thiosulphate and peroxide. In order to prepare the free acid, iodine in excess is added very gradually to thiosulphate of barium suspended in a very small quantity of water, the iodide of barium and excess of iodine being removed by shaking up the semi-solid mass with strong alcohol, leaving a white crystalline mass of barium tetrathionate, BaS 4 6 . This may then be dissolved in a small quantity of water and recrystallised, whilst from this pure salt trhe acid may be prepared by adding exactly sufficient sulphuric acid to decompose it completely. Tetrathionic acid is a colourless, inodorous, very acid liquid, which, when dilute, may be boiled without undergoing decom- TETRATHIONIC ACID 419 position, but in the concentrated state is easily decomposed into sulphurous and sulphuric acids and sulphur. The tetrathionates are all soluble in water, but their solutions cannot, as a rule, be evaporated without decomposition into sulphur and a trithionate. A solution of the potassium salt decomposes gradually on pre- servation, sulphur dioxide being evolved but no sulphur deposited. The decomposition proceeds in two stages, tri- and pentathionate of potassium being first formed, and the tri- thionate then decomposing in the usual way into potassium sulphate, sulphur dioxide and sulphur which converts a portion of the trithionate into pentathionate. (Debus.) (1) 2K 2 S 4 6 = K 2 S 3 6 4K 2 S 5 6 . (2) 3K 2 S 3 6 = 2K 2 S0 4 + 2SO 2 + K 2 S 5 O 6 . Dry potassium tetrathionate undergoes no alteration on pre- servation. Sodium amalgam decomposes the compound into two molecules of thiosulphate, and the same decomposition occurs on the addition of potassium sulphide ; thus : K 2 S 4 6 + K 2 S --= 2K 2 S 2 3 + S. Potassium tetrathionate does not combine with ordinary sulphur but combines with nascent sulphur to form a pentathionate. Thus the free acid is decomposed by an excess of sulphuretted hydrogen into water and sulphur : H 2 S 4 6 + 5H 2 S = 6H 2 + 9S. When however a slow current of sulphuretted hydrogen is passed into a solution of the potassium salt, the sulphur combines with a portion of the salt and forms potassium pentathionate. Sulphur dioxide reacts with potassium tetra- thionate to form a trithionate and a pentathionate, the latter being probably produced by a secondary reaction from thio- sulphuric acid or its anhydride : PENTATHIONIC ACID. H 2 S 5 O 6 . 230 Wackenroder, in 1845, was the first to draw attention to the existence of this acid in the liquid obtained by passing 420 THE NON-METALLIC ELEMENTS sulphuretted hydrogen into a solution of sulphur dioxide ; the clear liquid obtained after the removal of the precipitated sulphur was looked upon by him and his successors as a solu- tion of the acid in water. The reaction was represented by the equation : 5H 2 S + 5S0 2 = H 2 S 5 O 6 + 5S + 4H 2 O. Later observers succeeded in preparing crystalline pentathion- ates from this liquid and thus placed the existence of the acid beyond doubt. The investigations of Debus 1 have shown that the liquid in question, known as Wackenroder's solution, has a very complex composition, which varies according to the method of preparation The liquid is best prepared by passing sulphuretted hydrogen slowly for two hours into 480 cc. of a saturated solution of sulphurous acid at a temperature near 0, allowing the whole to stand for two days in a closed bottle and repeating the operation until all the sulphurous acid has disappeared. The liquid is then filtered from the sulphur which has separated out and is thus obtained as a semi-trans- parent milky fluid. The solution contains oily drops of sulphur in suspension and a considerable amount of sulphur in a soluble form (p. 346), which can be precipitated by the addition of potassium nitrate or by concentrating the fluid on the water - bath. The concentration of the clear liquid cannot be carried further than a sp. gr. of 1*32 in this way without evolution of sulphur dioxide and deposition of sulphur, but may be continued over sulphuric acid in vacuo beyond this point. An analysis of the solution thus obtained (sp gr. 1-32) shows that its composition is approximately that of pentathionic acid, H 2 S 5 O 6 . When however the liquid is carefully treated with about one third of an equivalent of caustic potash and then allowed to evaporate spontaneously, a mixture of potassium tetra- and pentathlon ates is obtained which can be separated by recrystallisation from lukewarm water. The mother liquor on evaporation yields a salt which contains more sulphur than a pentathionate and is probably potassium hexathionate (p. 422). In addition to these substances, Wackenroder's solution, pre- pared in the manner described, contains sulphuric acid and traces of trithionic acid, the amount of which becomes much 1 Journ. Ghem. Soc. 1888, 278 357 (where references to the literature of the subject will be found). PENTATHIONIC ACID 421 greater when an insufficient quantity of sulphuretted hydrogen is used. The pentathionates are decomposed by caustic alkalis, sulphur being deposited and a tetrathionate formed, and they are there- fore best prepared from Wackenroder's solution by treating it with the acetate of the metal and allowing the solution to evap- orate spontaneously. The potassium salt, 2K 2 S 5 O 6 +3H 2 O, crys- tallises in short rhombic prisms or six-sided plates. It forms a clear, neutral solution in two parts of water and can be recrys- tallised from water at 50 made acid with a little sulphuric acid. The solution decomposes on standing, more rapidly on heating, sulphur being deposited and a tetrathionate formed, but is more stable in the presence of a little sulphuric acid ; The salt itself decomposes on keeping if any moisture be present, but if dried by washing the fine powder with dilute alcohol, and then preserved over sulphuric acid, it remains unaltered for years. On ignition it is converted into potassium sulphate, sulphur dioxide and sulphur : Potassium pentathionate is also formed when a slow current of sulphuretted hydrogen is passed into a solution of the tetra- thionate (p. 419). The barium salt, BaS 5 O 6 + 3H,O, a.nd the copper salt, CuS 5 O 6 + 4H 2 O, are also crystalline. The pentathionates may be distinguished from the lower members of the series by the facts that ammoniacal silver nitrate solution produces a brown coloration which rapidly becomes darker, a black precipitate being finally formed, and that caustic potash produces an immediate precipitate of sulphur, tetrathionate being also formed, whilst ammonia only gives this reaction on standing. They are not affected by hydrochloric acid, ferric chloride or barium chloride. A solution of pure pentathionic acid, prepared by the decom- position of the potassium salt, is converted by sulphuretted hydrogen into water and sulphur : H 2 S 5 8 + 5H 2 S = 6H 2 + 10S, THE NON-METALLIC ELEMENTS whilst the potassium salt yields a thiosulphate and a trithi- onate : 3K 2 S 5 O 6 + 3H 2 S = K 2 S 2 O 3 + 2K 2 S 3 O 6 + 3H 2 O + 10S. Sulphur dioxide reacts both with the acid and its salts to form trithionic acid and probably thiosulphuric anhydride : HEXATHIONIC ACID, H 2 S O 6 . 231 This substance is only known in the form of its potassium salt obtained from the mother liquors of potassium pentathionate. This salt, which has not been obtained in a perfectly pure state, forms a warty crust and yields an aqueous solution, which de- composes even in the presence of acid. It reacts with caustic potash and ammoniacal silver nitrate solution in the same way as the pentathionates, but gives an immediate precipitate of sulphur with ammonia. FORMATION OF WACKENROBER'S SOLUTION. It seems probable that in the preparation of Wackenroder's solution, the direct product is tetrathionic acid : H 2 S + 3S0 2 = H 2 S 4 6 . This substance is then acted upon by sulphurous acid with formation of trithionic acid and thiosulphuric anhydride, S 2 O 2 (p. 419). Moreover the tetrathionic acid is converted by the sulphuretted hydrogen into water and sulphur, and a number of secondary reactions then go on, by means of which the thio- sulphuric anhydride gives up some of its sulphur to the tetra- thionic acid, forming pentathionic acid, whilst the trithionic acid combines with nascent sulphur, forming tetra, penta-and hexathionic acids. When the action of the sulphuretted hydrogen is allowed to continue until all action is at an end, water and sulphur are found to be the final products : SELENIUM 423 CONSTITUTION OF THE ACIDS OF WACKENRODER'S SOLUTION. Potassium pentathionate is oxidised by bromine according to the following equation : K 2 S 5 6 + 4Br 2 + 6H 2 = 2KBr + 2S + 3H 2 SO 4 + 6HBr. Hence two of the sulphur atoms are differently combined in the molecule from the other three ; these two atoms are pre- cipitated in the above reaction as free sulphur, whilst the other three are converted into sulphuric acid and are therefore probably already combined with oxygen in the molecule of pen- tathionic acid and in those of tetra- and trithionic acid, which stand in so close a relation to it. Trithionic acid moreover strongly resembles sulphurous acid in its power of combining with sulphur to form other acids and therefore probably contains the group K.S0 2 , which is characteristic of sulphurous acid. The following formulae exhibit all these relations and are also in -accordance with the various decompositions of these substances ; they may therefore be assigned to the salts of these acids (Debus) : K. SO 2 . O Potassium Trithionate ... | KO. S0 2 . S KS. S0 2 . O Potassium Tetrathionate. . . KO. SO 2 . S KS 2 . S0 2 . O Potassium Pentathionate. . . KO. S0 2 . S KS 3 . S0 2 . O Potassium Hexathionate . . KO. S0 2 . S SELENIUM. Se = 78-5. 232 We owe the discovery of this element to Berzelius. 1 He first found it in the deposit from the sulphuric acid chambers at Gripsholm in Sweden, in the year 1817, and it is to him that we are indebted for the knowledge of its most important com- 1 Schweigg Journ. 23, 309, 430 ; Fogg. Ann. 7, 242 ; 8, 423. 424 THE NON-METALLIC ELEMENTS pounds. The name selenium is derived from %\ijvrj, the moon, on account of its analogy with the element tellurium (tellus, the earth), discovered shortly before. Although selenium is somewhat widely distributed, it occurs only in small quantities. It is found together with sulphur in the islands of Volcano and Hawai, and occurs, chiefly combined with certain metals, at Clausthal and Zorge in the Harz, in La Plata, the Argentine and Mexico, as the mineral clausthalite, PbSe ; a selenide of copper and lead, PbSe + Cu 2 Se; lehrbachite, PbSe + HgSe; selenide of silver, Ag 2 Se ; selenide of copper, Cu 2 Se. We also find it as onofrite, HgSe + 4HgS, in Mexico ; whilst eucairite, Cu. 2 Se -f Ag 2 Se, and crookesite (CuTlAg). 2 Se, occur at Skrikerum in Sweden. Selenium is also found in very small quantity in many other minerals, especially in certain iron-pyrites and copper-pyrites, and where these are used for the manufacture of sulphuric acid, a red deposit containing selenium is found in the chambers. Preparation. In order to prepare selenium from this deposit, it is mixed with equal parts of sulphuric acid and water to a thin paste and then boiled ; nitric acid or potassium chlorate being added until the red colour disappears. In this way a solution of selenic acid, H 2 SeO 4 , is obtained, which is diluted with water, filtered, and then heated with a quarter its volume of fuming hydrochloric acid until three-quarters of the liquid has evaporated. By this process chlorine is evolved and selenious acid, H 2 SeO 3 , formed. The cold solution is then poured off from the solid matter and saturated with sulphur dioxide, when selenium separates out as a red powder. The selenium thus obtained contains lead and other metals, from which it may be separated either by distillation or by fusing it with a mixture of nitre and sodium carbonate, by which means sodium selenate is formed, and this is then again treated with hydrochloric acid and sulphur dioxide as above described (Wohler). 1 Selenium may also be easily obtained from the chamber deposit by heating it on a water-bath with a concen- trated solution of cyanide of potassium until it assumes a pure grey colour. On the addition of hydrochloric acid to the filtered solution selenium is deposited in cherry-red flakes. This also contains both copper and lead, and these impurities are removed either by the process as described above or by evaporating the selenium to dryness with nitric acid and reducing the aqueous 1 Handbook of Inorganic Analysis (1854), 154. PKOPERTIES OF SELENIUM 425 solution of selenium dioxide by means of sulphur dioxide (Nilson). Properties. Selenium, like sulphur, exists in several allotropic modifications, one class being soluble, the other insoluble in carbon bisulphide (Berzelius, Hittorf). Soluble selenium is obtained as a finely divided brick-red amorphous powder, when a cold solution of selenious acid is precipitated by a current of sulphur dioxide ; and as a black crystalline powder when this gas is passed through a hot solution. Other reducing agents, such as iron, zinc, stannous chloride or phosphorous acid, also precipitate soluble selenium from solutions of selenious acid. Selenium crystallizes from solution in carbon bisulphide in small dark-red translucent crystals, which are isomorphous with the monosymmetric form of sulphur, and have a specific gravity of 4'5. T Like sulphur, selenium is polymorphous, the crystals which are deposited from a cold saturated solution differing in type from those just described, although belonging to the same system. A third variety, isomorphous with the crystals of tellurium and belong- ing to the hexagonal system, is obtained at temperatures above 130. 2 When selenium is fused and allowed to cool quickly it solidifies to a dark brownish-black glassy translucent amorphous brittle mass, which is also soluble in carbon bisulphide, and has a specific gravity of 4'3. Soluble selenium has no definite melting point, softening gradually on heating. The insoluble or metallic selenium is obtained by cooling melted selenium quickly to 210 and then keeping the melted mass at this temperature for some time. The selenium at length solidifies to a granular crystalline mass, the temperature rising suddenly in the act of solidification to 217. The solid mass thus obtained has a specific gravity of 4'5, and is insoluble in carbon bisulphide. This change from the soluble to the insoluble condition also takes place, but more slowly, at lower tempera- tures ; thus, if a mass of soluble selenium be placed in an air-bath at 100, the change to the insoluble variety takes place gradually, and the temperature rises up to 217. If a concentrated solution of potassium or sodium selenide be exposed to the air, black selenium separates out in microscopic crystals which have a specific gravity of 4 '8, and are likewise insoluble in carbon bisulphide. The insoluble or metallic selenium has a constant melting point of 217, and when quickly cooled is converted 1 Kammelsberg, Ber. 7, 669. 2 Muthmann, Zeit. Kryst, 17, 336. 426 THE NON-METALLIC ELEMENTS into amorphous soluble selenium. The change of vitreous into crystalline selenium is accompanied by an evolution of 2'79 omits of heat. Both modifications of selenium are soluble in chloride of selenium and separate out from this solution in the form of metallic selenium. In addition to these forms, a modification of selenium is also known which is soluble in water. 1 This colloidal selenium is formed when a current of sulphur dioxide is passed into a solution of selenious acid. It is a dark red powder which is completely soluble in water, forming a red fluorescent solution, but w r hich gradually becomes insoluble on preservation. The solution may be boiled without undergoing any change, but the selenium is deposited on the addition of acids or salts. The solution on spontaneous evaporation deposits the selenium as a red transparent film. According to the experiments of Carnelley 2 and Williams, selenium boils at 680, and forms a dark-red vapour which con- denses either in the form of scarlet flowers of selenium, or in dark shining drops of the melted substance. As in the case of sulphur, the vapour density of selenium diminishes very rapidly with the temperature ; thus at 860 the vapour density is 110-7, whilst at 1420 it has a density of 81'5, closely corresponding to the normal vapour density of 78'5. Metallic selenium conducts electricity, and exposure to light increases its conducting power. 3 The peculiar effect of light is best exhibited on selenium which has been exposed for a considerable time to a temperature of 210, until it has attained a granular crystalline condition. When selenium in this condition is heated its electrical resistance is increased, whilst on exposing it to the action of diffused daylight, the electrical resistance instantly diminishes ; this, however, is only a temporary change, for on cutting off the light, the electrical resistance of the selenium slowly increases, and after a short time reaches the amount exhibited before the exposure. This remarkable pro- perty of selenium may possibly be made use of for photometrical purposes. When selenium is heated in the air it burns with a bright 1 Schulze, Jr. Pr. Chem. [2], 32, 390. 2 Journ. Chem. Soc. 1879, i. 563. 3 Sale, Proc. Roy. Soc. 1873, 21, 283 ; W. G. Adams, Proc. Roy. Soc. 1875, 23, 535 ; W. Siemens, Berlin, Ber. Akad. 1875. S. Bid well, Phil. Mag. (5), 30 178. SELENIURETTED HYDROGEN 427 blue flame, forming an oxide of selenium to which is due the characteristic odour, resembling that of rotten horse-radish, noticed when selenium burns. The emission spectrum of selenium is seen when a small bead of the element is held in .a non-luminous gas flame ; it is a channelled spectrum highly -characteristic and beautiful, consisting of a very large number of bright bands, which in the green and blue are arranged at regular intervals. 1 According to Salet, 2 selenium, like sulphur, gives two emission spectra, one consisting of lines and the other of bands. The absorption spectrum of selenium has been examined by Gernez. 3 The atomic weight of selenium has been determined by Pettersson and Ekman, who determined the composition of the oxide Se0 2 and of silver selenite Ag 2 Se0 3 and obtained the number 78*5 ; 4 their numbers agree with the older determina- tions of Berzelius. 5 SELENIUM AND HYDROGEN. HYDROGEN SELENIDE OR SELENIURETTED HYDROGEN. H 2 Se. 233 This gas is formed when selenium vapour and hydrogen -are heated together, and the amount which is thus produced is a function of the temperature. The quantity formed increases when the temperature is raised from 250 to 520, but above this point the amount gradually diminishes. When selenium is heated in a closed tube filled with hydrogen, it sublimes in the cool part of the tube in the form of beautiful glittering crystals, and these increase in number until the whole of the selenium has volatilized. The formation of these crystals depends upon the decomposition by heat of the seleniuretted hydrogen which is formed ; for when selenium is heated in a tube filled with an indifferent gas, only red amorphous selenium is found to sublime. Seleniuretted hydrogen is easily obtained by the action of dilute hydrochloric acid on potassium selenide, K 2 Se, or on iron .selenide, FeSe, which is prepared by adding selenium to heated 1 W. M. Watts, Index of Spectra, p. 56. 2 Compt. Rend. 73, 559 and 742. 3 Compt. Rend. 74, 1190. 4 Ber. 9, 1290. 5 Pogg. Ann. 8, 21. 428 THE NON-METALLIC ELEMENTS iron. It is also formed when organic materials such as colo- phonium are heated with selenium. It is a colourless, inflam- mable gas, possessing a smell which, to begin with, resembles that of sulphuretted hydrogen, but afterwards is found to have a much more persistent and intolerable odour, a small quantity affecting the mucous membrane in a remarkable degree, attacking the eyes, and producing inflammation and coughing which last for days. Berzelius describes the effects as follows: 1 "In order to become acquainted with the smell of this gas I allowed a bubble not larger than a pea to pass into my nostril ; in consequence of its action I so completely lost my sense of smell for several hours that I could not distinguish the odour of strong ammonia, even when held under my nose. My sense of smell returned after the lapse of five or six hours, but severe irritation of the mucous membrane set in and lasted for a fortnight." In order to ascertain the composition of seleniuretted hydrogen metallic tin is heated in a measured volume of the gas, when tin selenide is formed, and a volume of hydrogen is liberated equal to that of the original gas. Hydrogen selenide is more soluble in water than the corre- sponding sulphur compound, yielding a colourless solution which reddens blue litmus paper, colours the skin a reddish -brown tint, and possesses the foetid odour of the gas. Exposed to the air, the aqueous solution absorbs oxygen with the separation of red selenium. When added to solutions of salts of most of the heavy metals, it produces precipitates of the insoluble selenides in an analogous manner to sulphuretted hydrogen. Sulphur also decomposes it, forming sulphuretted hydrogen and precipitating selenium. SELENIUM AND CHLORINE. 234 Selenium, like sulphur, forms several compounds with chlorine. SELENIUM MONOCHLORIDE. Se 2 Cl 2 . When a current of chlorine is passed over selenium, the latter melts and is converted into a brown oily liquid, which is selenium 1 Lehrbuch, 5 Aufl. 2, 213. SELENIUM MONOBROMIDE 429 monochloride. It is more readily prepared by passing hydro- chloric acid gas into a solution of selenium in fuming nitric acid. 1 Selenium monochloride is slowly decomposed by water according to the equation : 2Se 2 Cl 2 + 3H 2 = H 2 SeO 3 + 3Se + 4HC1. SELENIUM TETRACHLORIDE. SeCl 4 . This body is obtained by the further action of chlorine upon the monochloride, as well as when selenium dioxide is heated with phosphorus pentachloride (Michaelis) ; thus : 3Se0 2 + 3PC1 5 = 3SeCl 4 + P 2 O 5 + POC1 3 . The tetrachloride is a light yellow solid body, which on heat- ing volatilizes without previously melting, subliming in small crystals. It also crystallizes from solution in phosphorus oxychloride in the form of bright shining cubes. It dissolves in water with formation of hydrochloric and selenious acids ; thus : SeCl 4 + 3H 2 = 4HCl + H 2 SeO s . When its vapour is heated its density diminishes, dis- sociation taking place. 2 SELENIUM AND BROMINE. SELENIUM MONOBROMIDE. Se 2 Br 2 . 235 Equal parts of bromine and selenium combine together with evolution of heat to form a black semi-opaque liquid, having a specific gravity at 15 of 3'6. It has a disagreeable smell resembling that of chloride of sulphur, and colours the skin a permanent red-brown tint. On heating it is decomposed, and when brought into contact with water, selenium, selenious acid and hydrobromic acid are formed ; thus : 2Se 2 Br 2 + 3H 2 O = 3Se + H 2 SeO 3 + 4HBr. 1 Divers and Shimose, Ber. 17, 866. 2 Chabrie, Bull. Soc. Ghim. [3], 2, 803. 430 THE NON-METALLIC ELEMENTS SELENIUM TETKABROMIDE. SeBr 4 . This compound is formed by the further action of bromine on the monobromide. It is an orange yellow crystalline powder, best obtained by adding bromine to a solution of selenium monobromide in carbon bisulphide. It is very volatile, vapor- ising between 75 and 80 with partial decomposition and subliming in black six-sided scales. It possesses a disagree- able smell similar to that of chloride of sulphur, decomposes in contact with moist air into bromine and the monobromide^ and dissolves in an excess of water with formation of hydro- bromic and selenious acids. The chlorobromides of selenium, SeCl 3 Br and SeClBr 3 , are orange yellow crystalline substances (Evans and Ramsay). SELENIUM AND IODINE. SELENIUM MONO-IODIDE. Se 2 I 2 . 236 The compounds of selenium and iodine resemble those of selenium with chlorine and bromine. The mono-iodide is ob- tained when the two elements are brought together in the right proportions (Schneider), and forms a black shining crystalline mass, melting between 68 and 70 with the evolution of a small quantity of iodine. When more strongly heated, it decomposes into iodine which volatilizes, and selenium which remains behind, and is decomposed by water in a similar way to the corresponding bromide. SELENIUM TETRA-IODIDE. SeI 4 . It is a granular dark crystalline mass, which melts, at from 75 to 80, to a brownish-black liquid, translucent in thin films ; when more strongly heated it decomposes into its elements. SELENIUM AND FLUORINE. 237 When the vapour of selenium is passed over melted fluoride of lead, a fluoride of selenium sublimes in crystals. SELENIOUS ACID 431 These are soluble in hydrofluoric acid, and are decomposed by water (Knox). When selenium is exposed to gaseous fluorine it fuses and finally takes fire, a white crystalline substance being formed. This is soluble in hydrofluoric acid, but is decomposed by water (Moissan). 1 SELENIUM AND OXYGEN. OXIDES AND OXY- ACIDS OF SELENIUM. 238 Only one oxide of selenium, the dioxide. SeO 2 , is with cer- tainty known to exist in the free state. A lower oxide is stated by Berzelius to be formed when selenium burns in the air, and is probably the cause of the peculiar smell then observed, which is so penetrating that if 1 mgrm. of selenium be burnt in a room the smell is perceptible in every part. Selenium also forms two oxy-acids selenious acid, H 2 SeO 3 and selenic acid, H 2 SeO 4 . SELENIUM DIOXIDE. SeO 2 . 239 When selenium is placed in a bent tube, as shown in Fig. 125 and strongly heated in a current of oxygen, contained in the gasholder and dried by passing over pumice-stone saturated with sulphuric acid, it takes fire and burns with a bright blue flame, a white sublimate of solid selenium dioxide being deposited in the cool part of the tube. Thus obtained it forms long four- sided, white, needle-shaped crystals, which do not melt when heated under the ordinary atmospheric pressure, but evaporate, when heated to about 300, yielding a greenish yellow-coloured vapour possessing a powerful acid smell (Berzelius). SELENIOUS ACID, H 2 SeO 3 . 240 Selenious acid is formed when selenium is heated with nitric acid, or when five parts of the dioxide are dissolved in one part of hot water. On cooling, clear, long, colourless, prismatic, nitre-shaped crystals of selenious acid separate out. These have a strong acid taste, and, when heated, decompose into selenium 1 Ann. Chim. Phys. [6] 24, 239. 432 THE NON-METALLIC ELEMENTS dioxide and water. If sulphur dioxide be allowed to pass into a hydrochloric acid solution of selenious acid, selenium is deposited as a red powder. This decomposition takes place but slowly in the cold and in absence of light ; but when the liquid is heated or in the sunlight, the change occurs quickly. Organic substances also bring about this reduction, and the colourless solution of the pure acid soon becomes tinged when exposed to the air, owing to the presence of the dust in the atmosphere. Selenious acid is distinguished from sulphurous FIG. 125. acid by this reaction, for sulphurous acid gradually absorbs oxygen from the air. Selenious acid is a dibasic acid, and forms not only acid and normal salts, but also salts, containing acid selenites united with selenious acid ; thus, HKSeO 8 + H 9 SeO 3 . The normal selenites of the alkali metals are easily soluble in water ; those of the alkaline earth metals and the heavy metals are insoluble ; whilst all the acid salts are soluble compounds. When a selenite is heated on charcoal in the reducing flame, a characteristic horse-radish-like smell is emitted, and when SELENIC ACID 433 heated in a glass tube with sal-ammoniac, selenium sublimes. The selenites are further distinguished by the fact that red selenium is precipitated when sulphur dioxide is led into their solution in water or in hydrochloric acid. The salts are very poisonous. SELENYL CHLORIDE, OR SELENIUM OXYCHLORIDE. SeOCl 2 . 241 This chloride of selenious acid is formed by the action of selenium dioxide on selenium tetrachloride (Weber) ; * thus : SeO+2SeCI 4 = 2SeOCl 2 . and when the oxide is heated with sodium chloride : 2 2Se0 2 + 2NaCl = SeOCl 2 + Na 2 SeO 3 . It is a yellow liquid which fumes on exposure to air, and which, when cooled below 10, deposits crystals having a specific gravity of 2'44. It boils at 179'5 (Michaelis), and decomposes with water in the same way as all acid chlorides. When mixed with thionyl chloride, selenium tetrachloride and sulphur dioxide are formed ; thus : SeOCl. 2 + SOC1 2 = SeCl 4 + SO 2 . SELENYL BROMIDE. SeOBr 2 . Formed by the action of selenium tetrabromide on selenium dioxide. The two substances are melted together, and the compound, on cooling, crystallizes in long needles. SELENIC ACID. H 2 Se0 4 . 242 This acid, discovered in 1827 by Mitscherlich, is formed by the action of chlorine on selenium or on selenious acid in presence of water ; thus : Se + 3C1 2 + 4H 2 = H 2 SeO 4 + 6HC1. By the same reaction selenites may be converted into selenates. Bromine may for this purpose be employed instead of chlorine. 1 Pogg. Ann. 108, 615. 2 Cameron and Macallan, Gkton. News, 59, 267. 29 \ 434 THE NON-METALLIC ELEMENTS Potassium selenate is also obtained when selenium is fused with nitre. In order to prepare selenic acid, a solution of sodium selenite is treated with silver nitrate ; and to the precipitate obtained, suspended in water, bromine is added (J. Thomsen) ; thus : Ag 2 SeO 3 + H 2 O 4 Br 2 = 2AgBr + H 2 SeO 4 . The aqueous solution of selenic acid can be concentrated by evaporation in the air, but it cannot be thus completely freed from water, as at a temperature of about 280 it begins to de- compose into oxygen, selenium dioxide and water. If heated in vacuo to 180 and then cooled, the whole solidifies to a crystalline mass, consisting of the pure acid. 1 The acid crystallizes in long hexagonal prisms, similar to those formed by sulphuric acid and melts at 58. It combines eagerly with water, forming a monohydrate which melts at 25 and, like sulphuric acid, chars many organic substances. The specific gravity of the solid acid is 2*6273 whilst that of the liquid at 15 is 2-3557. The concentrated solution of the acid is a colourless, very acid liquid, which is miscible with water in all proportions, heat being thereby evolved. The heated aqueous acid dissolves gold and copper with formation of selenious acid, whilst iron, zinc, and other metals dissolve with evolution of hydrogen and pro- duction of selenates. This acid is not reduced either by sulphur dioxide or by sulphuretted hydrogen, and differs in this respect, therefore, remarkably from selenious acid. When boiled with hydrochloric acid it decomposes with the evolution of chlorine and formation of selenious acid ; thus : H 2 Se0 4 + 2HC1 = H 2 SeO 3 + C1 2 + H 2 O. This mixture dissolves gold and platinum, these metals com- bining with the chlorine thus set at liberty. The selenates exhibit the closest analogy with the sulphates so far as regards amount of water of crystallization, crystalline form, and solubility. Barium selenate, like the sulphate, is com- pletely insoluble in water, and is employed for this reason in the quantitative determination of selenic acid ; it is, however, distinguished from barium sulphate inasmuch as when boiled with hydrochloric acid the insoluble selenate is decomposed into 1 Cameron and Macallan, Chcm. News, 59, 219. SELENOSULPHURIC ACID 435 the soluble selenite, whereas barium sulphate remains unchanged. All the other selenates are also reduced to selenites by means of hydrochloric acid, and this reaction serves as a means of recognizing these compounds. SELENIUM AND SULPHUR. 243 Several substances have been at various times described as compounds of selenium and sulphur, 1 but they have all proved to be mixtures of the two elements. 2 SELENOSULPHUR TRIOXIDE. SeS0 3 . It has long been known that when selenium is dissolved in fuming sulphuric acid, a beautiful green colour is produced, and it has been proved that this is due to the formation of the above compound. This substance is best prepared by dissolving selenium in freshly distilled well-cooled sulphur trioxide. The compound then separates out in the form of tarry drops which soon solidify to prismatic crystals, having a dirty green colour, and yielding a yellow powder when broken up. The compound dissolves in sulphuric acid with a green colour, and is decomposed on addition of water with separation of selenium and formation of sulphuric, sulphurous and selenious acids. On heating, the body does not melt but decomposes into selenium, selenium dioxide, and sulphur dioxide (Weber). 3 f OH SELENOSULPHURIC ACID. SO 2 -j g e jj This compound, which corresponds to thiosulphuric acid, was discovered by Cloez, 4 and like this latter acid is not known in the free state. Its potassium salt is obtained when a solution of sulphurous acid is mixed with one of potassium selenide, or, together with the salts of selenotrithionic acid, when selenium 1 Rathke, J. Pr. Chem. 108, 235; Ditte, Compt. Rend. 73, 625, 660; Gerichten, Ber. 7, 26. 2 Divers, Chem. News, 51, 24 ; Bettendorf and von Rath, Pogg. Ann. 139, 329. 3 Pogg. Ann. 156, 513. 4 Bull. Soc. Chim. 1861, Il2. 436 THE NON-METALLIC ELEMENTS is dissolved in a solution of normal potassium sulphite. 1 The selenosulphates are isomorphous with the thiosulphates, and are decomposed by all acids, even by sulphurous acid, with the separation of red selenium. SELENOTRITHIONIC ACID. H 2 SeS 2 O 6 . The potassium salt of this acid is formed when a solution of potassium selenosulphate is mixed with an excess of normal potassium sulphate, and a concentrated solution of selenious acid, thus : QO J S( M The potassium salt forms monosymmetric prisms and is stable in the air. According to Schulze, 2 the free acid is formed when selenious acid is treated with an excess of sulphur dioxide. The same chemist has found that when selenious acid is kept in excess, an acid of the formula H 2 SSe 2 O 6 is obtained. SulpJioselenoxytetrachloride, C1SO 2 (OSeCl 3 ), is obtained by the action of selenium tetrachloride upon chlorosulphonic acid as a mass of white crystals 3 ; it melts at 165 and boils at 183. TELLURIUM. Te = 124 (?). 244 Tellurium occurs in small quantities in the free state in nature, and by early mineralogists was termed aurum paradoxum or metallum problematum, in consequence of its metallic lustre. In 178 2 native tellurium was more carefully examined byMiiller von Reichenstein. He came to the conclusion that it contained a peculiar metal, and at his suggestion, Klaproth 4 in 1798 made an investigation of the tellurium ores confirming the views of the former experimenter that it contained a new metal, to which he gave the name of tellurium from telhis, the earth. 1 Schaffgotsch, Pogg. Ann. 90, 66 ; Rathke, J. Pr. Chem. 95, 1 ; Uelsmaun, AnnalenUQ, 122. 2 J. Pr. Chem. [2] 32, 405. 3 Clausnizer, Ber. H, 2007. 4 Crell. Ann. 1, 91. TELLURIUM 437 Berzelius 1 in 1832, made a more exhaustive investigation of tellurium ; he likewise considered the substance itself to be a metal, but its compounds were found to correspond so closely with those of sulphur and selenium, that tellurium was placed in the sulphur group. Tellurium belongs to the rarer elements. It occurs in Transylvania, Hungary, California, Virginia, Brazil, and Bolivia, in small quantities in the native state, but it is generally found in combination with metals as graphic tellurium, or sylvanite (AgAu) Te 2 ; black or leaf tellurium, or nagyagite (AuPb) 2 (TeSSb) 3 ; silver telluride, Ag 2 Te ; tetradymite, or bismuth telluride, Bi 2 Te 3 , &c. Preparation. In order to prepare pure tellurium, tellurium- bismuth containing about 60 per cent of bismuth, 36 per cent, of tellurium, and about 4 per cent, of sulphur, is mixed with an equal weight of pure carbonate of soda, and then the mixture rubbed up with oil to a thick paste, and heated strongly in a well-closed crucible. The mass is then lixiviated with water, and the filtered solution, which contains sodium telluride and sodium sulphide, is exposed to the air, when the tellurium grad- ually separates out as a grey powder, which after washing and drying may be purified by distillation in a current of hydrogen (Berzelius). In order to obtain the tellurium from graphic, or from black tellurium, the ore is treated with hydrochloric acid in order to free it from antimony, arsenic, and other bodies. The residue is then boiled with aqua regia, and the filtrate eva- porated to drive off the excess of nitric acid ; ferrous sulphate is then added, which precipitates the gold, and the tellurium is thrown down in the filtrate by means of sulphur dioxide (v. Schrotter). The crude tellurium is best purified by treating it with aqua regia, expelling the excess of nitric acid by means of hydro- chloric acid and then diluting somewhat so that the tellurous acid remains in solution whilst the chloride of lead is precipitated. From the filtrate, the tellurium is precipitated by means of sulphurous acid. The material thus obtained, which may contain traces of selenium, lead and other metals, is then fused in small portions with potassium cyanide, the mass extracted with water in the absence of air, and the tellurium thrown 1 Pogg. Ann. 28, 392 ; 32, 1 and 577. 438 THE NON-METALLIC ELEMENTS down from the clear filtrate by a current of air. It is finally melted and distilled in a current of hydrogen. 1 Properties. Tellurium is a silver-white body possessing a metallic lustre, and crystallizing in rhombohedra. Tellurium is a very brittle substance, and can therefore be easily powdered. Its specific gravity is 6*24 : it melts at 452 (Carnelley and Williams), and boils at a still higher temperature, and may accordingly be easily purified by distillation in a stream of hydrogen gas. Amorphous tellurium is obtained when a solution of tellurous acid is precipitated by means of sulphurous acid. On heating it is converted with evolution of heat into the crystalline variety. 2 Tellurium burns when heated in the air with a blue flame, evolving white vapours of tellurium dioxide. It is insoluble in water and carbon bisulphide, but dissolves in cold fuming sul- phuric acid, imparting to the solution a deep-red colour, which is probably due to the formation of a compound analogous to sulphur sesquioxide, namely, STe0 3 , the tellurium being pre- cipitated on the addition of water. On heating the sulphuric acid solution, the tellurium is oxidized to tellurous acid, sulphur dioxide being given off. In the same way it rapidly undergoes oxidation in the presence of nitric acid. Hydrochloric acid does not attack tellurium ; caustic potash dissolves it with formation of potassium telluride and tellurite. When heated nearly to the melting-point of glass, tellurium emits a golden-yellow vapour, which gives an absorption spec- trum, consisting of fine lines stretching from the yellow to the violet. 3 The emission spectrum of tellurium has been mapped by Salet 4 and Ditte. 5 According to Deville and Troost the vapour of tellurium possesses a specific gravity of 9*0 at 1390, which number corresponds to a molecular weight of about 259. 6 The mole- cule of tellurium therefore contains two atoms and has the formula Te 2 . 245 Atomic Weight of Tellurium. Considerable doubt exists as to the value of the atomic weight of tellurium. The earlier observers obtained the number 128, and this was confirmed in 1 Brainier, Monatsh. 10, 414. 2 Fabre and Berthelot, Compt. Rend. 104, 1405. 3 Gernez, Compt. Rend. 74, H90. 4 Compt. Rend. 73, 559, 742. 5 Compt. Rend. 73, 262. 6 Compt. Rend, 56, 871. TELLURETTED HYDROGEN 439 1879 by Wills, 1 who analysed, the double bromide of ' potassium and tellurium K 2 TeBr 6 , and converted the element into tellurous acid by means of nitric acid and of aqua regia. The close relation of tellurium to sulphur and selenium however makes it highly probable, from considerations connected with the arrangement of the elements according to the periodic system (Vol. II., Part 2, p. 503), that the atomic weight of tellurium is less than that of iodine (125*9). Brauner 2 has therefore thoroughly re-examined the question, but finds that all reliable methods of determination (analysis of tellurous acid and of tellurium tetrabromide) lead to about 126'5. He expresses the opinion that the substance at present known as tellurium contains some substance of higher atomic weight. In this lie is con- firmed by the fact that when the tellurium is previously sub- limed in hydrogen its atomic weight is found to be considerably lower (126'5) than when it is sublimed in any indifferent gas (136'7). He supposes therefore that the impurity is partially removed by the hydrogen. TELLURIUM AND HYDROGEN. TELLURIUM HYDRIDE, OR TELLURETTED HYDROGEN. H 2 Te. 246 This compound, discovered by Davy in 1810, is a colour- less gas possessing a foetid smell similar to that of sulphuretted hydrogen. It is formed in small quantities when tellurium is heated in hydrogen gas. If this is allowed to take place in a sealed tube, the same phenomenon presents itself as is observed with selenium, and the tellurium sublimes in long glittering prisms. In order to prepare telluretted hydrogen, tellurium is heated with zinc, and the zinc telluride thus formed, decomposed by hydrochloric acid ; thus : ZnTe + 2HCl-Zn01 2 + H 2 Te. Tellurium hydride is easily combustible, burning with a blue flame. It is soluble in water, and this solution absorbs oxygen from the air, tellurium being deposited. Like sulphuretted hydrogen, telluretted hydrogen precipitates many of the metals from, their solutions in the form of tellurides. The soluble tellurides, suck as those of the alkaline metals, form brownish 1 Journ. Chcm. Soc. 1879, i. 704. ' 2 Monatsh. 10, 411. 440 THE NON-METALLIC ELEMENTS red solutions from which tellurium is deposited on exposure to the air. The density of tellurium hydride, like that of seleniuretted hydrogen, has not been as yet directly determined, but Bineau has shown that both gases, when they are heated with certain metals, give up all their selenium or tellurium, leaving a residue of hydrogen which occupies the same volume as the original gas. Hence each molecule of these gases contains two atoms of hydrogen combined, as analysis shows, with 78'5 and about 124 parts of these elements. Consequently tellurium hydride contains one part by weight of hydrogen and about sixty-two parts by weight of tellurium. TELLURIUM AND CHLORINE. TELLURIUM BICHLORIDE. TeCl 2 . 247 This compound is formed together with the tetrachloride when chlorine is passed over melted tellurium. It may be separated from the less volatile tetrachloride by distillation; and thus obtained, it forms an indistinctly crystalline, almost black mass which gives a greenish yellow powder, melts at 175, boils at 327 , 1 and yields a deep red-coloured vapour, having the density 6*89. 2 Water decomposes the compound with separation of tellurium and formation of tellurous acid ; thus : 2TeCl 2 + 3H 2 = Te f H 2 Te0 3 + 4HC1. TELLURIUM TETRACHLORIDE, TeCl 4 , Is formed by the further action of chlorine on the preceding compound. It is a white crystalline body which melts at 224 , 3 forming a yellow liquid, which when more strongly heated becomes at last of a dark red colour and boils at 380 , 4 without decomposition. It is extremely hygroscopic and is decomposed when thrown into cold water, an insoluble oxychloride being 1 Carnelley and Williams. Journ. Chem. Soc. 1879, i. 563 ; 1880, i. 125. 3 Michaelis, Ber. 20, 2488. 3 Carnelley and Williams, Journ. Chem. Soc. 1880, i. 125. 4 Michaelis, Ber. 20, 2491. BROMIDES OF TELLURIUM 441 formed together with tellurous acid. Hot water, on the other hand, gives rise only to the formation of the latter compound. Like the corresponding tetrachlorides of sulphur and selenium, it forms crystalline compounds with a large number of metallic chlorides. The vapour density of the tetrachloride is 9*2 (Michaelis). TELLURIUM AND BROMINE. TELLURIUM DIBROMIDE, TeBr 2 , Is best obtained by heating the tetrabromide with tellurium. On heating it volatilizes in the form of a violet vapour, which condenses to black needles, melting at 280, and again solidify- ing to a crystalline mass ; it boils at 339. TELLURIUM TETRABROMIDE, TeBr 4 . In order to prepare this compound, finely divided tellurium is added to bromine which has been cooled down to 0< It is a dark yellow solid body which can be sublimed without de- composition. It dissolves in a small quantity of water with a yellow colour, but when added to a large quantity of water the solution becomes colourless from the formation of hydrobromic and tellurous acids. It boils at 420. It forms double salts with the bromides of the alkali metals. TELLURIUM AND IODINE. TELLURIUM DI-IODIDE. TeI 2 . Is obtained as a black crystalline mass by heating iodine and tellurium together in the proper proportions. TELLURIUM TETRA-IODIDE, TeI 4 ; Is formed by the action of hydriodic acid on tellurous acid ; thus : H 2 TeO 3 + 4HI = TeI 4 + 3H 2 O. 442 THE NON-METALLIC ELEMENTS It forms an iron-gray crystalline mass which melts when gently heated, and when heated more strongly decomposes with separa- tion of iodine. It is but slightly soluble in cold water, and is decomposed by boiling water. TELLURIUM AND FLUORINE. TELLURIUM TETRAFLUORIDE. TeF 4 , Is prepared by heating the dioxide with hydrofluoric acid in a platinum retort. It distils over as a colourless, transparent, very deliquescent mass. When a solution of tellurous acid in hydrofluoric acid is con- centrated, crystals of the formula TeF 4 + 4H 2 O separate out. Powdered tellurium becomes incandescent on exposure to fluorine gas, the tetrafluoride being formed. 1 TELLURIUM AND OXYGEN. OXIDES AND OXYACIDS OF TELLURIUM. 248 Three oxides of tellurium are known, TeO, Te0 2 and Te0 3 as well as the acids, H 2 Te0 3 , and H 2 Te0 4 corresponding to the last two of these. TELLURIUM MONOXIDE, TeO. 249 This substance is left behind as a brownish black amor- phous mass when tellurium sulphoxide, TeS0 3 , is heated to 230 in vacuo, sulphur dioxide being evolved. On heating in the air or on exposure to moist air it is gradually oxidised. Sulphuric acid forms with it a red solution, from which a crystalline mass of tellurium sulphate, Te(S0 4 ) 9 , soon separates. The monoxide when heated in hydrochloric acid gas is converted into tellurium dichloride. TELLURIUM DIOXIDE, Te0 2 , 250 Occurs in the impure state in nature as tellurite or tellu- rium ochre, at Facebay in Transylvania. It is formed by the 1 Moissan, Ann. Ghim. Phys. [6] 24, 239. 2 Divers and Shimose, Ber. 16, 1004 ; Journ. Chem. Soc. 1883, i. 319. TELLUKOUS ACID 443 combustion of tellurium in the air, and separates out in small octahedra when tellurium is dissolved in warm nitric acid. It is only very slightly soluble in water, and the solution does not redden blue litmus paper. On heating it melts to a lemon- yellow liquid which boils on further heating without decom- position, whilst on cooling it solidities to a white crystalline mass. Although this is an acid-forming oxide, it also exhibits basic properties, inasmuch as it combines with certain acids forming an unstable class of salts, which are decomposed by water. The nitrate on ignition leaves a residue of TELLUROUS ACID, H 2 Te0 3 , 251 Is obtained by pouring a solution of tellurium in dilute nitric acid into water. It separates out in the form of a very voluminous precipitate, which when placed over sulphuric acid dries to a light white powder. It is but slightly soluble in water, and the solution possesses a bitter taste. Tellurous acid like sulphurous acid, is dibasic, and therefore forms two series of salts : thus we have, normal potassium tellurite, K 2 Te0 3 , and acid potassium tellurite, KHTe0 3 . Other more complicated series of salts exist such as : K 2 Te,0 5 , K 2 Te 4 9 , K 2 Te 6 O 13' The tellurites of the alkali metals are soluble in water, and are formed by the solution of the acid in an alkali, or by fusing the dioxide with an alkali. The tellurites of the alkaline earth metals are only slightly soluble, and those of the other metals are insoluble in water, but soluble in hydrochloric acid. Tellurous acid is converted into telluric acid by the action of potassium permanganate both in acid and alkaline solution. The amount of tellurous acid in a solution may be estimated in this manner. 2 TELLURIUM TRIOXIDE. TeO 3 . 252 This oxide is prepared by heating crystallized -telluric acid to nearly a red heat. If it is heated too strongly, a small quantity of dioxide is formed with evolution of oxygen, but this can be separated by treatment with hydrochloric acid, in which it dis- 1 Klein, Ann. Chim. Phys. [6], 10, 108. 2 Brauner Monatsh. 12, 29. 444 THE NON-METALLIC ELEMENTS solves whilst the trioxide is insoluble. Tellurium trioxide is an orange yellow crystalline mass, which, when strongly heated decomposes into oxygen and the dioxide. TELLURIC ACID. H 2 TeO 4 . 253 When tellurium or the dioxide is fused with carbonate of potash and saltpetre, potassium tellurate is formed ; thus : Te 2 + K 2 C0 3 + 2KN0 3 - 2K 2 TeO 4 + N 2 + CO. The same salt is produced when chlorine is passed through an alkaline solution of a tellurite ; thus : K,Te0 3 + 2KOH + C1 2 = K 2 TeO 4 + 2KC1 -f H 2 O. On dissolving the fused mass in water, and adding a solution of barium chloride, the insoluble barium tellurate is precipitated ; this is purified by washing with water, and afterwards decom- posed by the exact amount of sulphuric acid necessary. The clear acid solution, filtered from the sulphate of barium, gives on evaporation crystals of the hydra ted acid having the composition H 2 Te0 4 + 2H 2 O. These are sparingly soluble in cold, but readily soluble in hot water. When the crystals are heated to 160 they lose their water of crystallization, and the telluric acid remains as a white powder nearly insoluble in cold, but readily dissolving in hot water, forming a solution which deposits the crystalline hydrate. The Tellurates. Amongst the tellurates only those of the alkali-metals are more or less readily soluble in water, those of the remaining metals being either sparingly soluble or insoluble in water, although generally dissolving readily in hydrochloric acid. Certain of the tellurates are found to exist in two modifica- tions, viz. : (a) As colourless salts, soluble in water or in acids. (&) As yellow salts, insoluble in water and in acids. Besides these modifications we are acquainted not only with normal and acid tellurates, but with several other series of acid salts ; thus we have : (1) Normal potassium tellurate, K 2 TeO 4 + 5H 2 O, obtained upon evaporation of a solution, either in the form of crystalline crusts, or as a gum-like residue, both being soluble in water. THE TELLURATES 445 (2) K. 2 TeO 4 + TeO, + 4H..O, a salt sparingly soluble in cold, but dissolving freely in hot water, and crystallizing to a woolly mass. (3) K 2 TeO 4 + 3TeO 3 +4H 2 O (or 2KHTeO 4 + 2H 2 TeO 4 + H 2 O), a salt sparingly soluble in water, obtained by adding nitric acid to salt No. 2. The yellow modification of this salt is obtained when salt No. 2 is heated to redness. On adding water to the residue, normal potassium tellurate dissolves out, and a tetratellurate remains behind, being insoluble in water and in dilute acids. Barium tellurate, BaTeO 4 -f 3H 2 O, is a white powder, not pre- cipitated in dilute solutions as it is not quite insoluble in water. The di- and tetra-tellurate of barium, as well as calcium and strontium tellurates, are similar white precipitates, whilst magnesium tellurate is rather more soluble. All the other tellurates are insoluble in water. When a tellurate is heated to redness, oxygen is evolved and a tellurite formed, and this reduction also occurs, with the evolution of chlorine, when a tellurate is heated with hydro- chloric acid ; thus : K 2 Te0 4 + 2HC1 = K 2 TeO 3 + H 2 O + CL, 254 Tellurium Oxy chloride, TeOCl 2 . This substance 1 is obtained by heating the compound of tellurium dioxide with hydrochloric acid, Te0 2 + 2HC1, above 90, Te0 2 , 2HC1 =TeOCl 2 + H 2 O. On further heating it decomposes into the tetrachloride and the dioxide. 2TeOCl 2 = TeCl 4 + Te0 2 . Tellurium Oxylromide, TeOBr 2 . This compound is obtained in a similar manner to the oxychloride and is a faintly yellow coloured mass. (Ditte). TELLURIUM AND SULPHUR. TELLURIUM BISULPHIDE, TeS 2 . 255 A disulphide, TeS 2 , and a trisulphide, TeS 3 , were pre- pared by Berzelius by the action of sulphuretted hydrogen upon solutions containing an alkali tellurite and telluric acid 1 Ditte, Compt. Rend. 83, 336, 446. 446 THE NON-METALLIC ELEMENTS respectively. Both of these compounds yield up nearly all their sulphur to carbon bisulphide and are therefore probably mixtures. 1 Salts of the formula 3K 2 S, TeS 2 (potassium thio-tellurite) and K 2 TeS 4 (thio-tellurate) are however known. TELLURIUM SULPHOXIDE. TeS0 3 . 256 This compound is obtained by the direct union of tellurium with sulphur trioxide. 2 It is a red amorphous, transparent solid, which melts at 30. On heating for some time to 35, it becomes of a light reddish brown colour. Water decomposes it with formation of tellurium and sulphuric acid, along with other products. On heating to 230, it loses sulphur dioxide and forms tellurium monoxide. NITROGEN. N = 13-94. 257 Dr. Rutherford, Professor of Botany in the University of Edinburgh, showed in the year 1772 that when animals breathe in a closed volume of air, it not only becomes laden with impure air from the respiration, but contains, in addition, a constituent which is incapable of supporting combustion and respiration. He prepared this constituent by treating air in which animals had breathed with caustic potash, by means of which the fixed air (carbonic acid) can be removed. The residual air was found to extinguish a burning candle, and did not support the life of animals which were brought into it. 3 In the same year Priestley found that when carbon is burnt in a closed bell-jar over water, one-fifth of the common air is converted into fixed air which can be absorbed by milk of lime; a residual (phlogisticated) air incapable of supporting either combustion or respiration being left. Priestley, however, did not consider that this air was a constituent of the atmosphere, and it is to Scheele that we owe the first statement, contained in his treatise on Air and Fire, that the " air must be composed of two different kinds of elastic fluids." The constituent known as mephitic or phlogisticated air was first considered to be a 1 Becker, Annalen, 180, 257. 2 Weber, J. Pr Chem. [2], 25, 218 ; Divers and Shimose, Ber. 16, 1009. 3 Rutherford, De aerc Mephitico. Edinb. 1772. PREPARATION OF NITROGEN 447 simple body by Lavoisier, who gave to this gas the name azote (from a, privative, and farf, life, by which it is still usually designated in France). Chaptal first suggested the name nitrogen, which it now generally bears (from vlrpov, saltpetre, and yevvdw I give rise to), because it is contained in saltpetre. Nitrogen is found in the free state in the atmosphere, of which it forms four-fifths by bulk, and occurs also in combi- nation in many bodies such as ammonia, in the nitrates, and in many organic substances which form an essential part of the bodies of vegetables and animals. 258 Preparation. The simplest method for preparing nitrogen is to remove the oxygen from the air. This can be done in a variety of ways : (1) A small light porcelain basin is allowed to swim on the FIG. 126. water of a pneumatic trough, a small piece of phosphorus brought into the basin and ignited, and the basin then covered by a large tubulated bell-jar (see Fig. 126). The phosphorus burns with the deposition of a white cloud of phosphorus pentoxide, which, however, soon dissolves, whilst on cooling, one-fifth of the contents of the bell-jar is found to be filled with water. The colourless residual gas is nitrogen ; this may be easily proved by first equalising the level of the water inside and outside the bell-jar, after which, on opening the stopper and plunging a burning taper into the bell-jar, the flame is seen to be instantly extinguished. The nitrogen thus obtained is never perfectly pure, as it always contains small quantities of oxygen which have not been removed by the combustion of the phosphorus. In presence of aqueous vapour, phosphorus 448 THE NON-METALLIC ELEMENTS slowly absorbs the oxygen of the air, at temperatures above 15, whilst the sulphides of the alkali metals as well as moist sulphide of iron (obtained by heating flowers of sulphur with iron filings) act in a similar way. (2) In order to obtain pure nitrogen, air contained in a gas- holder is allowed to pass through tubes, T and T', Fig. 127, containing caustic potash and sulphuric acid for the purpose of purifying the air from carbon dioxide and drying it. The air thus purified is passed over turnings of pure metallic copper contained in a long glass tube (e f) which is heated to redness in a charcoal furnace ; copper oxide is thus formed, and pure nitrogen passes over and is collected in the pneumatic trough. Fro. 127. (3) Copper also rapidly absorbs oxygen from the air at the ordinary temperature in presence of a solution of ammonia ; in order to obtain nitrogen by this method a slow current of air is passed through a tall cylinder containing copper turnings over which a solution of ammonia is allowed to drop continuously. The oxygen is absorbed and the resulting nitrogen, after washing with water to remove traces of ammonia, is collected in the usual manner. The last traces of oxygen are best removed by a solution of chromous chloride, CrCl 2 , and this method, when carried out with other precautions, yields abso- lutely pure nitrogen. 1 (4) Pure nitrogen can also be prepared by heating a con- 1 Threlfall, Phil. Mag. [5], 35, 1. PREPARATION OF NITROGEN 449 centrated solution of ammonium nitrite, the following reaction taking place : It is more convenient to employ a mixture of potassium nitrite and ammonium chloride, which by double decomposition yield potassium chloride and ammonium nitrite, the latter then decomposing into nitrogen and water. As potassium nitrite frequently contains free alkali it is advisable to add some potassium bichromate in order to neutralise the latter ; a solution of one part of potassium nitrite, one part of ammonium FIG. 128. chloride, and one part of potassium bichromate in three parts of water gives good results. (5) By action of chlorine upon ammonia, pure nitrogen is also formed ; thus : 8NH 3 + 3C1 2 = N 2 + 6NH 4 C1. The chlorine evolved in a large flask passes into a three- necked Woulffe's bottle containing a strong aqueous solution of ammonia. The nitrogen gas which is here liberated is collected in the ordinary way over water, as shown in Fig. 128. Care must, however, be taken in this preparation that the ammonia 30 450 THE NON-METALLIC ELEMENTS is always present in excess, otherwise chloride of nitrogen may be formed, and this is a highly dangerous body, which explodes most violently (p. 477). (6) On heating ammonium bichromate, nitrogen gas, chromium sesquioxide and water, are formed ; thus : (NH 4 ) 2 Cr 2 7 = N 2 + 2 3 + 4H 2 0. Nitrogen may be obtained by this reaction at a cheaper rate by heating a mixture of potassium bichromate and sal-ammoniac instead of the ammonium bichromate, which is a somewhat expensive salt ; thus : K 2 Cr 2 7 + 2NH 4 C1 = N 2 +Cr 2 3 + 2KC1 + 4H 2 O. 259 Properties. Pure nitrogen is a colourless, tasteless, inodor- ous gas, which is distinguished by its inactive properties ; hence it is somewhat difficult to ascertain its presence in small quantities. As has been said, it does not support combustion, nor does it burn nor render lime-water turbid. It combines directly with but very few non-metals, although indirectly it can easily be made to form compounds with most of these elements, and many of its compounds, such as nitric acid, ammonia, chloride of nitrogen, &c., possess characteristic and remarkable properties. It combines directly with a number of the metals, such as calcium, barium, and magnesium, yielding compounds termed nitrides. The specific gravity of nitrogen is 0*97209, and one litre at 0, and under the normal pressure, weighs 1-25749 grm. (Rayleigh). Nitrogen was first liquefied by Cailletet by compressing it to 200 atmospheres and allowing it suddenly to expand. It has since been obtained in much larger quantities by Wroblewski l and Olszewski 2 in the manner already described (p. 84). Under a pressure of 33 atmospheres at 146 (its critical temperature) it forms a colourless liquid, which boils under atmospheric pressure at 194 and has a sp. gr. of 0'885. If allowed to evaporate in a vacuum the temperature is further lowered and at 214 the nitrogen solidifies, forming a crystalline mass. If compressed nitrogen be cooled by means of boiling oxygen and the pressure then somewhat diminished, 1 Compt. Rend. 97, 1553 ; 98, 982 ; 102, 1010 ; Monatsh. 6, 204. 2 Compt. Rend. 99, 133 ; 100, 350 ; Ann. Chim. Phys. 31, 58. PROPERTIES OF NITROGEN 451 the nitrogen separates out in crystalline flakes resembling snow. Nitrogen is but slightly soluble in water, one volume of water absorbing only '02348 of the gas at 4, and a smaller quantity at higher temperatures. 1 The co-efficient of the solubility of nitrogen (c) from 0-20 may be found by the following inter- polation formula : c = 0-023481 - The gas is rather more soluble in alcohol than in water. There are two characteristic spectra of nitrogen both obtained bypassing the spark from an induction coil through a Geissler's tube containing a small quantity of highly rarefied nitrogen gas. The nitrogen spectrum commonly obtained in this way is a channelled one, exhibiting a large number of bright bands especially numerous in the violet. If the spark is produced by high tension, as when a Leyden jar is used, a spectrum of numerous fine lines distributed throughout the length of the spectrum will be obtained (Plucker). Under ordinary conditions nitrogen is incombustible, but a mixture of nitrogen and oxygen can be made to ignite under certain circumstances. Thus if a powerful current of electricity be passed through the primary of a large induction coil, an arching flame is seen to issue from each secondary pole, provided these are not too far apart, the two flames joining in the centre ; the flame is due to the combination of the nitrogen and oxygen of the air to form oxides of nitrogen. When the flame has once been formed the distance between the secondary poles may be considerably increased without extinguishing the flame, and the latter may then be blown out and reignited by a taper. The flame does not extend to the surrounding air because heat is absorbed in the combination, and this can only therefore occur when energy is supplied to the mixture from an external source. 2 Assimilation of Free Nitrogen ~by Plants. As already men- tioned nitrogen forms an essential constituent of a very large number of animal and vegetable substances and is necessary for the maintenance of animal and vegetable life. It was how- ever for a long time believed that members of the vegetable 1 Winkler, Ber. 24, 3605. 2 Spottiswoode, Proc. Roy. Soc. 31, 173 ; Crookes, Chcm. News, 1892, i. 302. 452 THE NON-METALLIC ELEMENTS kingdom were unable to take up the free nitrogen of the air, and that they were dependent for their supply of this element on combined nitrogen contained in the atmosphere and in the soil chiefly in the form of nitric acid and ammonia. About the year 1886 it was shown that certain leguminous plants, such as the white lupine, when grown in air free from ammonia and other nitrogen compounds, contain more nitrogen than was originally present in the seed and in the soil in which they were sown, and they must, therefore, have obtained the excess from the nitrogen of the air ; since then it has been shown certain algae, fungi and mosses behave in a similar manner, and it is not improbable that further investigation will show that this property is possessed by many other varieties. The assimilation is brought about by the action of certain micro-organisms, which absorb the free nitrogen from the air, forming compounds of nitrogen which are then assimilated by the plant. 1 Other classes of micro-organisms also play a considerable part in the assimilation of nitrogen by plants, as the latter are incapable of directly assimilating ammonia, although they can take up nitric acid ; the micro-organisms in the soil bring about the conversion of ammonia into nitric acid, one variety of these converting the former into nitrous acid, and another converting the nitrous into nitric acid. 2 NITROGEN AND HYDROGEN. AMMONIA. NH 3 = 16-94. 260 Ammonia is found in the atmosphere in combination with carbonic acid forming a small but essential constituent of the air. It likewise occurs in combination with nitric and nitrous acids in rain-water, and, especially as sal-ammoniac, NH 4 C1, and as sulphate of ammonia, (NH 4 ).,SO 4 , deposited on the sides, the craters, and in the crevices of the lava streams of active volcanoes, as well as mixed with boric acid in the fumaroles of Tuscany. Many samples of rock-salt also contain traces of ammoniacal 1 For further details see Lawes and Gilbert, Phil. Trans. 180 (B) 1 ; Journ. Roy. Agnc. Soc. 1891, 657 ; Proc. Roy. Soc. 46, 85 ; Schloesing, Compt. Rend. 113, 776 ; Marshall Ward, Nature, 49, 511. 2 Warington, Journ. Chem. Soc. 1891, i. 484. AMMONIA 453 compounds, and all fertile soil contains this substance, which is likewise found, although in small quantities, widely distributed in rain and in running water, as well as in sea- water, in clays, marls, and ochres. Ammoniacal salts are also found in the juices of plants and in most animal fluids, especially in the urine. Ammonia was known to the early alchemists in the form of the carbonate under the name of spiritus salis urince. In the fifteenth century it was known that the same body may be obtained by the action of an alkali upon sal-ammoniac ; and Glauber, in consequence, termed this body spiritus volatilis salis armoniaci. Sal-ammoniac, which was known to Geber, appears to have been brought in the seventh century from Asia to Europe, and was known under the name of sal-armon- iacum. It is possible that this sal-ammoniac was derived from the volcanoes of Central Asia. Geber, however, describes the artificial production of the salt by heating urine and common salt together. In later times, sal-ammoniac was brought into Europe from Egypt, where it was prepared from the soot obtained by burning camel's dung. Its original name was altered to sal-armoniacum, and then again changed to sal- ammoniacum. This last name served originally among the Alexandrian alchemists to describe the common salt (chloride of sodium) and native sodium carbonate, which was found in the Libyan desert in the neighbourhood of the ruins of the temple of Jupiter Ammon. Boyle says in his " Memoirs for the Natural History of Human Blood : " l " Though the sal-armoniac that is made in the East may consist in great part of camel's urine, yet that which is made in Europe, and commonly sold in our shops, is made of man's urine." Later on, sal-ammoniac was obtained by the dry distillation of animal refuse, such as hoofs, bones, and horns ; the carbonate of ammonia thus obtained being neutralized with hydrochloric acid. From this mode of preparation ammonia was formerly termed spirits of hartshorn. Up to the time of Priestley, ammonia was known only in the state of aqueous solution, termed spirits of hartshorn, or spiritus volatilis salis ammoniaci. Stephen Hales, in 1727 observed that when sal-ammoniac is heated with lime in a vessel closed by water, no air is given out, but, on the contrary, water is drawn into the apparatus; Priestley, in 1774, repeated this experiment, with the difference, however, that he used 1 Boyle, op. 4, 597, 1684. 454 THE NON-METALLIC ELEMENTS mercury to close his apparatus. He thus discovered ammonia gas, to which he gave the name of alkaline air. He also found that, when electric sparks are allowed to pass through this alkaline air, its volume undergoes a remarkable change, and the residual air is found to be combustible. Berthollet, following up this discovery in 1785, showed that the increase of volume which ammonia gas thus undergoes is due to the fact that it is decomposed by the electric spark into hydrogen and nitrogen. This discovery was confirmed, and the composi- tion of the gas more accurately determined by Austin (1788), H. Davy (1800), and Henry (1809). It was shown by them that, in the reaction above described, two volumes of ammonia are resolved into three volumes of hydrogen and one of nitrogen. It has been proved by Donkin that ammonia can be syntheti- cally prepared by the direct combination of its elements, 1 the silent electric discharge being, for this purpose, passed through a mixture of nitrogen and hydrogen. If a mixture of 3 vols. of hydrogen and 1 voJ. of nitrogen is subjected to the action of the electric discharge in a eudiometer, containing above the mercury a little dilute sulphuric acid, the whole volume finally disappears, the two gases slowly combining to form ammonia, which dissolves in the acid as fast as it is formed. It is like- wise obtained together with nitrous acid by passing nitrogen over a mixture of platinum black and alkali. 2 Ammonia is also formed : (1) By the putrefaction or decay of the nitrogenous con- stituents of plants and animals. (2) By the dry distillation of the same bodies ; that is, by heating these substances strongly out of contact with air. (3) By the action of nascent hydrogen on the salts of nitric or nitrous acid. It is to the first of these processes that we owe the existence of ammonia in the atmosphere, whilst the second serves for the production of ammonia and its compounds, especially of sal- ammoniac, on the large scale. At the present day almost all the sal-ammoniac and other ammonium salts are prepared from the ammoniacal liquor which is obtained as a by-product in the manufacture of coal-gas. Coal consists of the remains of an ancient vegetable world, and contains about 2 per cent, of nitrogen, some of which, in 1 W. F. Donkin, Proc. Roy. Soc. 21, 281. 2 Loew, Ber. 23, 1443. PREPARATION OF AMMONIA 455 the process of the dry distillation of the coal carried on in the manufacture of coal-gas, is obtained in the form of ammonia dissolved in the water and other products formed at the same time. In order to prepare sal-ammoniac from this liquor, which contains very little if any free ammonia, but chiefly the sulphide, carbonate, sulphite and thiosulphate of ammonia, it is boiled with milk of lime to liberate the whole of the ammonia. This ammonia distils over, and the distillate is neutralized with hydrochloric acid, sal-ammoniac being formed as follows : NH 3 + HC1 = NH 4 C1. The solution is then evaporated to dryness and the salt purified by sublimation. In most cases, however, the ammonia is passed into sulphuric acid, thus forming ammonium sulphate, (NH 4 ) 2 S0 4 , which is much more readily purified, and from which most of the other ammoniacal compounds are prepared. Preparation. If any one of these ammoniacal salts be heated with an alkali, such as potash or soda, or with an alkaline earth, such as lime, the ammonia is set free as gas. In order to prepare the gas it is only necessary, therefore, to heat together sal-ammoniac and slaked lime ; thus : 2NH 4 C1 + Ca(OH) 2 = 2NH 3 + CaCl 2 + 2H 2 0. In order to ensure the decomposition of all the sal-ammoniac, a large excess of lime is usually employed. One part by weight of powdered sal-ammoniac is for this purpose mixed with two parts of caustic lime slaked to a fine dry powder ; these are well mixed together, and then introduced into a capacious flask, placed on a piece of wire gauze and heated by a Bunsen-lamp. The ammonia gas, which comes off when the mixture is gently heated, is then dried by allowing it to pass through a cylinder filled with small lumps of quick-lime, or by placing a layer of the latter over the mixture of ammonium chloride and slaked lime in the flask, and the gas thus dried may be collected either over mercury, or, like hydrogen, by upward displacement in an inverted dry cylinder as shown in Fig. 129, inasmuch as this 8'47 gas is fTTocn = 0'586 times as light as air, one litre weighing at and under 760 mm. pressure, O76193 grams. 456 THE NON-METALLIC ELEMENTS 261 Properties. Pure ammonia is a colourless gas, possessing, like its aqueous solution, a peculiar pungent alkaline odour and caustic taste. In the solid state, however, it possesses but a very faint smell. Ammonia gas turns red litmus-paper blue, like the alkalis, neutralizes acids, and forms with them a series of stable compounds, termed the ammonium salts. It is a very stable compound, but at 1300 gradually decomposes into its constituents. 1 Ammonia is not combustible, and a flame is extinguished if FIG. 129. plunged into the gas. If, however, ammonia be mixed with oxygen, the escaping gas maybe ignited, and burns with a pale yellow flame, with formation of water, nitrogen gas, and nitric acid, HNO 3 . Another method of showing the combustibility of ammonia is to put a jet of this gas into the air holes of an ordinary Bunsen -burner, in which a flame of coal-gas is already burning ; the flame becomes at once coloured yellow, and in- creases greatly in dimensions. A third experiment of this nature is, to allow a stream of oxygen gas to bubble through a small quantity of strong aqueous ammonia placed in a flask and warmed as shown in Fig. 130 ; on bringing a light in contact 1 Crafts, Compt. Rend. 90, 309. PROPERTIES OF AMMONIA 457 with the mixed gases issuing from the neck of the flask they will be seen to burn with a large yellow flame. Ammonia gas was first liquefied by Faraday, in 1823, by heating a compound of silver chloride with ammonia, placed in one limb of a strong hermetically sealed bent tube whilst the other limb was placed in a freezing mixture. The compound of silver chloride and ammonia is obtained by saturating dry precipitated silver chloride with ammonia gas: it has the formula AgCl(NH 3 ) 2 and fuses at 38, whilst at about FIG. 130. 115 it begins to part with its ammonia. The gas thus collects in the tube until the pressure is reached under which it begins to condense as a clear, highly refracting liquid. When the silver chloride cools, the ammonia is again absorbed, the original com- pound being re-formed. Liquid ammonia is also easily obtained by leading the gas into a tube plunged in a freezing mixture composed of crystallized calcium chloride and ice, and having a temperature of 40 J and forms a colourless highly refracting liquid, boiling at 33'7 (Bunsen). When the temperature of the liquid is lowered to below - 75 in a bath of solid carbonic 458 THE NON-METALLIC ELEMENTS acid and ether placed in vacuo, a mass of white translucent crystals of solid ammonia is obtained (Faraday). The co-efficient of expansion of liquid ammonia is 0*00204, and, therefore, larger than that of most liquids having a higher boiling point ; its specific gravity compared with water at J is Q'6234. 1 The tension of liquid ammonia at is 4*4 atmospheres, at 15'5, 6'9 atmospheres, and at 28, 10 atmospheres. FIG. 131. The condensation of ammonia by pressure and the production of cold by its evaporation can easily be shown by the following experiment. The apparatus required for this purpose consists essentially of two strong glass tubes (a and &, Fig. 131), which are closed below and are connected together by the tubes (c c) and (d d). The tube (d d) ends at (I) in a narrower tube (mm), which is at this point melted into the tube (a). The tube 1 Joly, Annalen, 117, 181. PEOPERTIES OF AMMONIA 459 (a) is three-fourths filled with an alcoholic solution of ammonia saturated at 8, and then placed in the cylinder (A). The syphon tube (g) and the tube (//), which reach to the bottom of the cylinder, are fixed in position through the cork. In order now to perform the experiment, the cylinder (A) is nearly filled with warm water ; the glass stopcock (h) is opened, and the tube (6) placed in ice-cold water. The water contained in the flask is now quickly boiled, and thus the water in (A) is rapidly heated to 100, and the ammonia gas driven out of solution until by its own pressure it liquefies in (6). As soon as the condensation of liquid ammonia ceases, the ebullition is stopped and a portion of the hot water is with- drawn from the cylinder by means of the syphon (g), cold water is allowed to enter the cylinder, and after a while this is replaced FIG. 132. by ice-cold water. The cylinder (B) is now removed, when the liquefied ammonia begins to evaporate and is again absorbed by the alcohol, though only slowly. But, on closing the stopcock (h\ the gas above the alcohol is quickly absorbed, and thus the equilibrium is disturbed. The ammonia now passes rapidly through the tube (m m), and is absorbed so quickly that the liquid ammonia in (b) begins to boil, by which the temperature is so much lowered, that if a test-tube containing water is placed outside (b) it is soon filled with ice. Ammonia may be used for the artificial production of ice. For this purpose an apparatus (Fig. 132) has been invented by M. Carre. It consists of two strong iron vessels connected by a vent-pipe of the same metal. The cylinder (A) contains water saturated with ammonia gas at 0. When it is desired to procure 460 THE NON-METALLIC ELEMENTS ice, the vessel (A) containing the ammonia solution, which we may term the retort, is gradually heated over a large gas-burner. The ammonia gas is thus driven out of solution, and as soon as the pressure in the interior of the vessel exceeds that of seven atmospheres, it condenses in the double- walled receiver (B). When the greater portion of the gas has thus been driven out of the water, the apparatus is reversed, the retort (A) being cooled in a stream of cold water, whilst the liquid which it is desired to freeze is placed in a cylinder which fits into the interior portion of the hollow cylinder. A reabsorption of the ammonia by the water now takes place, and a consequent evaporation of the liquefied ammonia in the receiver. This evaporation is accompanied by the absorption of heat which becomes latent in the gas. Thus the receiver is soon cooled down far below the freezing point, and the liquid contained in the vessel (D) is frozen. For the production of larger quantities of ice, a continuous ammonia freezing machine of more complicated construction, but arranged on the same principle, has been devised by M. Carre, in which 10 kilos, of ice can be prepared by the combustion of 1 kilo, of coal. 1 262 Ammonia gas is very soluble in water ; one gram, of water absorbs at and under normal pressure, 0*875 grms. or 1148 cc. of the gas. The solubility at different temperatures and under a pressure of 760 mm. of mercury is given in the following table (Roscoe and Dittmar) : Temp. Gram. Temp. Gram. Temp. Gram. Temp. Gram. 0-875 16 0-582 32 0-382 i 48 0-244 2 0-833 18 0-554 34 0-362 50 0-229 4 0-792 20 0-526 36 0-343 52 0-214 6 0-751 22 0-499 38 0-324 54 0-200 8 0-713 24 0-474 40 0-307 56 i 0-186 10 0-679 26 0-449 42 0-290 12 0-645 28 0-426 44 0-275 14 0-612 30 0-403 46 0-259 The same observers have found that the absorption of ammonia 1 Ice-making machines depending on the same principle are now in use in which liquefied sulphur dioxide, or ether, are employed instead of aqueous ammonia. 2 Journ. Chem. Soc. 1860, 128. SOLUBILITY OF AMMONIA 461 in water does not follow the law of Dalton and Henry at the ordinary atmospheric temperature, inasmuch as the quantity absorbed does not vary directly as the pressure. Sims l has shown that at higher temperatures the deviations from the law become less until at 100 the gas follows the law, the quantity FIG. 133. absorbed being directly proportional to the pressure, taken of course under pressures higher than the ordinary atmospheric pressure, inasmuch as boiling water does not dissolve any of the gas under the pressure of one atmosphere. 1 Journ. Chem. Soc. 1862, 17. 462 THE NON-METALLIC ELEMENTS In order to show the great solubility of ammonia gas in water the same apparatus may be employed which was used for exhibiting the solubility of hydrochloric acid in water (Fig. 133), the lower balloon being filled with water slightly coloured with red litmus solution. The liquor ammoniae of the shops, a solution of the gas in water, is prepared (as shown in Fig. 134) by passing the gas, which FIG 134. has been previously washed, into a flask containing water kept cool by being placed in a large vessel of cold water, considerable heat being evolved in the condensation of the gas. The commercial liquor ammonise is now frequently prepared directly from the ammoniacal liquor of the gas-works instead of from sal-ammoniac. For this purpose the liquor is heated together with milk of lime in an iron boiler having a capacity COMPOSITION OF AMMONIA 463 of 1,000 gallons. The gas which is evolved is first passed through a long system of cooling tubes before it enters the washing and condensing apparatus. The condenser consists of a, series of tubes filled with charcoal, by means of which any remaining empyreumatic impurities are removed. By this method, if the system of tubes be sufficiently long, and by the use of a sufficient number of wash-bottles, a perfectly pure liquor ammonise can be obtained. The fact that great heat is evolved in the production of the saturated solution of ammonia is rendered evident by the following experiment. - If a rapid current of air be passed through a cold concentrated solution of ammonia, the gas will be driven out of solution, and an amount of heat will be absorbed exactly equal to that which was given off when the solution of the gas was made, in consequence of which the temperature of the liquid will be seen to fall below - 40. A small quantity of mercury may thus be frozen. Ammonia is very soluble in alcohol, and this solution, which is frequently used in the laboratory, is most conveniently prepared by gently warming a concentrated aqueous solution of ammonia and passing the gas thu$ evolved into alcohol.-- This method may also be employed for preparing the gas in place of heating sal-ammoniac with lime, or the concentrated solution may be allowed to drop from a tap funnel on to lumps of caustic soda placed in a cylindrical glass vessel, by which means a regular and continuous evolution of the gas is obtained. . The following table gives the percentage of ammonia con- tained in aqueous solutions of different specific gravity (Carius) Specific Gravity. Per Cent. NH 3 . 0-8844 ' 36-0 0-8864 35-0 0-8976 30-0 0-9106 25-0 0-9251 20-0 0-9414 15-0 0-9593 10-0 0-9790 5-0 263 Composition of Ammonia. In order to determine the composition of ammonia the arrangement shown in Fig. 135 is employed. The ammonia gas is placed in the closed limb of 464 THE NON-METALLIC ELEMENTS the syphon eudiometer, after which the mercurial column in both limbs is brought to the same height, the volume accurately read off, and a series of electric sparks from the induction coil allowed to pass through the gas until its volume undergoes no further alteration. The tube and gas are next allowed to cool, and the pressure in both limbs again adjusted. On the volume being again measured it is seen to have loubled. Oxygen is next added in such proportion that the mixture shall contain no more than 35 per cent, of the explosive mixture of oxygen and hydrogen (2 volumes of hydrogen to 1 of oxygen), and an electric spark is passed through the mixture. From the alteration of volume which takes place, the proportion of hydrogen to nitrogen can readily be deduced, as is seen from the following example : Volume of ammonia 20*0 nitrogen and hydrogen ... 40 '0 After the addition of oxygen 157*5 After the explosion 112*5 Hence 45 volumes have disappeared, of which 30 consisted of hydrogen ; consequently two volumes of ammonia contain three volumes of hydrogen and one volume of nitrogen. That the relation between hydrogen and nitrogen in ammonia gas is in the proportion of three volumes of the former to one of the latter can be shown by the following experiment thus clearly described by Prof. Hofmann : l A glass tube for holding chlorine, having a small stoppered portion separated from the rest of the tube by a glass stopcock and used for receiving solution of ammonia, and admitting it, drop by drop, to the chlorine, constitute the requisite apparatus (Fig. 136). The glass tube is from 1 to 1*5 metre long, sealed 1 Introduction to Modern Chemistry. COMPOSITION OF AMMONIA 465 at one end, open at the other, and marked off, by elastic caoutchouc rings slipped over it and clipping it firmly, into three equal portions. The apparatus is thus employed. The long chlorine-tube having been filled with lukewarm water and inverted over a pneumatic trough, with its mouth immersed below the water-level, is filled with chlorine in the usual way (see Fig. 137). When full, it is still allowed to stand for about FIG. 136. fifteen minutes over the chlorine delivery-tube, so that its interior surface may be quite freed from the chlorine- saturated water that would else remain adherent to it. The stopcock is now closed and the chlorine thus shut in the tube. This is then removed from the trough, and turned round so that the stoppered end is uppermost. The small space between the stopcock and the end is next two-thirds filled with a strong solution of ammonia, and a single drop of the ammonia- solution is suffered to fall into the chlorine tube, the stopcock being opened for a moment for this purpose (Fig. 138). The entrance of this drop into the atmosphere of chlorine is marked by a small, lambent, yellowish-green flame at the point where the drop enters the gas. Drop by drop, at intervals of a few seconds, the ammonia solution is allowed to fall into the chlorine-tube, the ammonia of each drop being converted, at the instant of its contact with the chlorine, into hydrochloric acid 31 466 THE NON-METALLIC ELEMENTS and nitrogen with a flash of light and the formation of a dense white cloud. The addition of ammonia must be continued till the whole of the chlorine present is supplied with hydrogen at the expense of ammonia. To insure the ammoniacal solu- tion being added in excess, a column of three or four centimetres is abundantly sufficient. The result is that the hydrochloric acid formed combines with the excess of ammonia to form a compound, which makes its appearance as a white deposit lining the interior of the chlorine tube. This deposit, being soluble, is readily washed down and dissolved by agitating the liquor in FIG. 137. the tube, which now contains the whole of the nitrogen separated. We are now sure of two points, viz., that the whole of the chlorine has been converted into hydrochloric acid at the expense of the ammonia ; and, that we possess within our tube the whole of the nitrogen thus set free. It becomes our next object to withdraw the excess of the ammonia. For this purpose dilute sulphuric acid, which fixes ammonia, is introduced by means of the portion of the tube previously employed to admit ammonia. COMPOSITION OF AMMONIA 467 The nitrogen, being thus freed from all intermixed gaseous bodies, has only now to be brought to mean atmospheric temperature and pressure in order that it may be ready for measurement. To equalize the pressure within and without the tube, the bent syphon tube (Fig. 139) is employed. One end of this communicates with the interior of the tube, while the other plunges beneath the surface of water subject to atmospheric pressure. That this pressure exceeds that of the gas in the tube is at once seen by the flow of water through the syphon into FIG. 138. the tube. As the water-level in the tube rises, the nitrogen, previously expanded, gradually approaches its normal volume, which it exactly attains when the flow ceases, showing the pressure within and without to be in equilibrium. Both tem- perature and pressure being now at the mean, all the requisite conditions are fulfilled for obtaining an exact knowledge of the true volume of nitrogen; and this, on inspection, is found to exactly fill one of the three divisions marked off at the outset on our tube. Now, bearing in mind that we started with the three divisions full of chlorine, and that we have saturated this chlorine with hydrogen supplied by the ammonia ; bearing in 468 THE NON-METALLIC ELEMENTS mind', moreover, that hydrogen combines with chlorine, bulk for bulk, it is evident that the one measure of nitrogen which remains in the tube has resulted from the decomposition of a quantity of ammonia containing three measures of hydrogen. It is, therefore, clearly proved by this experiment that ammonia FIG. 139. is formed by the union of three volumes of hydrogen with one volume of nitrogen (Hofmann). Another method of demonstrating the same fact is by the electrolysis of a strong solution of ammonia. For this purpose, the solution, mixed with a little ammonium sulphate to increase its conductivity, is introduced into a Hofmann apparatus (Fig. 76, p, 250), and subjected to the action of a moderately strong current of electricity ; hydrogen is evolved at the neg- COMBINATION OF AMMONIA WITH ACIDS 469 ative pole and nitrogen at the positive pole, the volume of the former being three times as large as that of the latter. 264 Detection and Estimation of Ammonia. The method adopted for the detection and estimation of small traces of ammonia with Nessler's reagent has already been described under Natural Waters (see p. 300). If the quantity of ammonia or of ammonium salt be larger, it may be detected by the peculiar smell of the gas, by its alkaline reaction, and by the formation of white fumes in presence of strong hydrochloric acid. These fumes consist of ammonium chloride NH 4 C1, and are formed by the direct combination of the ammonia and hydrogen chloride : If, however, the gases are mixed together in a perfectly dry condition, no reaction takes place, but on addition of a little moisture combination immediately ensues. When an am- moniacal salt is present, the ammonia must be liberated by heating the solid salt, or its solution, with a caustic alkali. For the estimation of ammonia in quantities larger than those for which Nessler's method is applicable, it is usual to distil the ammonia either into hydrochloric or sulphuric acid of known strength, and then to ascertain, by volumetric analysis with a standard solution of alkali, the amount of acid remaining free, or into hydrochloric acid of unknown strength to which a solution of chloroplatinic acid, H 2 PtCl 6 , is added. On evaporat- ing the resulting solution to dryness on a water bath, and exhausting with alcohol, an insoluble yellow precipitate of am- monium platinochloride, (NH 4 ) 2 PtCl 6 , is left, and this can either be collected on a weighed filter, or it may be ignited and the quantity of the metallic platinum remaining weighed, from which the weight of ammonia is calculated. In addition to its use in the laboratory and as a means of obtaining artificial cold, ammonia is also largely employed for the preparation of alum, carbonate of soda, aniline colours, and in the manufacture of indigo. 1 Combination of Ammonia with Acids. Mention has been frequently made in the foregoing pages of the combination which takes place between ammonia and acids. Thus ammonia unites with hydrochloric acid to form the compound NH 3 ,HC1, 1 On the Manufacture of Liquor Ammonise see Lunge's work on Coal- Tar and Ammonia, p. 667. London, 1887 470 THE NON-METALLIC ELEMENTS or NH 4 C1, and with hydrobromic acid to form NH 3 HBr, or NH 4 Br, whilst with sulphuric acid it yields 2NH 3 ,H 2 SO 4 or (NH 4 ) 2 SO 4 . In a similar manner ammonia combines with almost all other acids yielding compounds containing the group of atoms NH 4 . Thus the compound NH 4 C1 may be regarded as hydrochloric acid, HC1, in which the hydrogen atom has been replaced by the group NH 4 , just as potassium chloride is hydro- chloric acid in which the hydrogen atom is replaced by a potassium atom. The two substances do in fact bear a strong resemblance to each other both in their chemical and physical properties, and all the other substances obtained by the com- bination of ammonia with acids are similarly related to the salts of potassium. The groups of atoms NH 4 is therefore a never varying con- stituent in a series of compounds, and behaves in these compounds as though it were a simple substance : to such a group the name of "compound radical" is given, and as in this case the group in combination has the same effect on the properties of the substance as a metal, it is termed ammonium, the compounds with acids being known as the ammonium salts, as they correspond to the salts of potassium, sodium, &c. They will be described together with these in Vol. II., Part I. HYDRAZINE OR Di AMIDE, N 2 H 4 . 265. This compound was first obtained by Curtius in 1887 by the action of hot dilute acids on triazo-acetic acid, a sub- stance described in Vol. III., Part II. (2nd edition) p. 107, which has the composition C 3 H 3 N 6 (COOH) 3 , and contains the group N = N three times. 1 It has since been obtained from other organic compounds containing two nitrogen atoms combined together ; among these may be mentioned amidoguanidine, which has the constitution NH 2 .C(NH).NH.NH 2 . As this substance is obtained without difficulty from ammonium thio- cyanate, it affords the best means for the preparation of hydra- zine in quantity, and a description of the preparation may be therefore shortly given here, although the intermediate com- pounds will not be described till later. Ammonium thiocyanate, NH 4 CNS, is heated for some time at a constant temperature of 170 180, and the residue, which consists chiefly of guanidine thiocyanate, treated first with 1 Curtius, Bev. 20, 1632. HYDRAZINE 471 strong sulphuric acid, then with a little fuming sulphuric acid, and the mixture after cooling mixed with nitric acid of sp. gr. 1'5, and the whole poured into water. Crude nitro-guan- idine, NH 2 .C(NH).NH.NO 2 , separates out, and is at once treated with zinc dust and just sufficient dilute acetic acid for its reduction. The solution obtained by the reduction of 208 grams of nitroguanidine is then evaporated to 1200 cc., mixed with a solution of 260 grams of caustic soda in 500 cc. of water, and boiled for 8 10 hours. The cooled liquid is poured off from the sodium hydrogen carbonate which separates out, and mixed with 260 cc. of concentrated sulphuric acid ; the greater part of the hydrazine separates out as the sulphate, which is quite pure after a single recrystallization. 1 Free hydrazine has not up to the present been obtained in a pure condition, owing to the great readiness with which it combines with water to form a hydrate; hitherto it has not been found possible to remove the water completely by the action of even the strongest dehydrating agents. If a mixture of the hydrate and anhydrous baryta is heated in a sealed tube at 170, and the latter opened after cooling, a gas is given off which has an extremely penetrating odour, fumes in the air, and probably consists of free hydrazine. 266. Hydrazine Hydrate, N 2 H 4 ,H 2 O, is best obtained by dis- tilling hydrazine sulphate with a solution of caustic potash in a silver retort, connected without rubber or cork to a silver condensing tube ; the distillation is continued till the last drop has passed over, and the distillate subjected to fractional dis- tillation, by which the hydrazine hydrate is readily separated from the excess of water. Hydrazine hydrate forms a strongly refractive almost odour- less caustic-tasting" liquid, which fumes strongly in the air, but may be kept unaltered in closed vessels ; it boils without decom- position at 118*5 under 739*5 mm. pressure, and solidifies in a mixture of solid carbonic acid and ether, but melts again below -40. It has a sp. gr. of 1*0305 at 21, and a vapour density at 100 under diminished pressure, agreeing with the formula N 2 H 4 .H 2 O, whilst at 170 it is completely split up into hydrazine and water ; if the temperature be further increased the vapour density also increases, until at very high temperatures the latter corresponds with a formula double that observed at 100. The cause of this abnormal behaviour has not yet been ascertained. 1 Thiele, Annalen, 270, 1. 472 THE NON-METALLIC ELEMENTS In aqueous solution a determination of the molecular weight gave numbers corresponding to the formula N 2 H 4 .2H 2 O. Hydrazine hydrate when hot attacks glass strongly, and also quickly destroys cork and indiarubber ; it is a very strong poison for lower organisms. 1 It is the strongest reducing agent known, quickly precipitating all the more easily reducible metals from their solutions even in the cold. It is a very strong base, and like ammonia unites with acids to form well-defined salts, most of which are readily soluble in water. Unlike ammonia, how- ever, it forms more than one series of salts, giving with hydro- chloric acid for example two salts, N 2 H 4 .HC1 and N 2 H 4 .2HC1, and with hydriodic acid three, N 2 H 4 .H1, N 2 H 4 .2HI, and 3N 2 H 4 .2HI. 2 These will be described, together with the ammonium salts, in Vol. II., Part I. AZOIMIDE OR HYDRAZOIC ACID, N 3 H. 267. This interesting substance, was like the foregoing, dis- covered by Curtius, who obtained it by the action of nitrites on derivatives of hydrazine, the reaction being analogous to that of nitrites on ammonia. As already mentioned, the interaction of the latter substances leads to the formation of nitrogen, according to the equation NH 4 C1 + NaN0 2 = N 2 + 2H 2 O + NaCl, and the formation of azoimide is represented in a similar manner by the equation NH 2 N x | + NO.OH = |! >NH +2H 2 0; NH 2 N/ hydrazine itself may be used if only a dilute solution of azoimide is required, 3 but as a rule it is preferable to employ one of its organic derivatives, in which one of the hydrogen atoms is replaced by an organic radical. Curtius in his ex- periments employed hippurylhydrazine, C 9 H 8 NO 2 HN.NH 2 , which by the action of nitrous acid yields hippurylazoimide, /K C 9 H 8 NO 2 .N/ 1 1 ; the latter when treated with acids or alkalis is 1 Loew, Ser. 23, 3203. 2 Curtius and Jay, J. Pr. Chem. [2] 39, 33 ; Curtius and Schultz, J. Pr. Chem. (2), 42, 521. 3 Ber. 26 1263. AZOIMIDE 473 converted into azoimide and hippuric acid. 1 The former distils over with the water on boiling the solution, which must first be acidified if an alkali has been employed for the hydrolysis. Numerous derivatives of azoimide in which the hydrogen atom has been replaced by an organic radical had long been known, the most readily prepared of which is the phenyl X N derivative, C 6 H 5 .N<( ]| (Vol. III., Part III., 2nd edition, p. 325), X -N this substance being known as triazobenzene or diazoben- zeneimide. This compound itself cannot be converted into azoimide by the action of acids or alkalis, but Noelting, Grandmougin and Michel have found 2 that if certain acid radicals such as the group NO 9 be introduced into the phenyl group, the resulting azoimide may be in many cases converted into the mother substance by the action of alkalis. Tilden and Millar have shown 3 that diazobenzeneimide is converted /N into a nitro-derivative, C 6 H 4 (NO 2 )NC [[, by the action of hot nitric acid of sp. gr. 1*4, and that this compound on boiling with alcoholic potash and subsequent acidification yields azoimide, a dilute solution of which is obtained by distilling the liquid. Another convenient source of azoimide is the crude amido- guanidine obtained in the preparation of hydrazine (p. 470) ; this substance is converted by nitrous acid into diazoguani- dine nitrate, NH 2 .C(NH).NH.N:N.NO 3 , which on boiling with alkalis yields azoimide and cyanamide; the former may be isolated by acidification and distillation in the manner already described. A further very interesting synthesis of azoimide from purely inorganic sources has also been described by W. Wislicenus, 4 who obtained its sodium salt by the action of nitrous oxide on sodamide : NaNH 2 + N 2 O = N 3 Na + H 2 O. The water formed acts on a further molecule of sodamide yielding caustic soda and ammonia. The sodamide is obtained by passing ammonia over metallic sodium at a temperature of 150 250, and as soon as all metallic sodium has disappeared, 1 Ber. 23, 3023. 2 Ber. 24, 2546 ; 25, 3328. 3 Journ. Chem. Soc. 1893, i. 256. 4 Ber. 25, 2084. 474 THE NON-METALLIC ELEMENTS the stream of ammonia is replaced by one of nitrous oxide, and continued till ammonia is no longer evolved ; the product is then dissolved in water and distilled with dilute sulphuric acid. The dilute solution obtained by any of the above processes may be concentrated by fractional distillation until the solution contains 91 per cent, of azoimide, and the remainder of the water is then removed by calcium chloride. Pure azoimide forms a colourless mobile liquid which has a most pene- trating, unbearable odour, boils without decomposition at 37, and dissolves readily in water and alcohol. When brought in contact with a hot body it explodes with extreme violence, giving a bright blue flash ; the explosion also sometimes takes place at the ordinary temperature, rendering the substance an extremely dangerous one to work with. Thus on one occasion O'Oo gram was introduced into a barometric vacuum, and exploded with such violence that the glass and mercury were reduced to dust and spread over the whole of a large room, and in another case 0*7 gram exploded when taken out of a freezing mixture, breaking all the bottles in the neighbourhood, and somewhat severely injuring one of the investigators. 1 The aqueous solution of azoimide behaves as a strong acid and readily dissolves zinc, iron, magnesium, and aluminium with evolution of hydrogen and formation of salts of the metals. It also gives salts with the metals of the alkalis and alkaline earths and forms a silver salt, AgN 3 , and a mercurom salt, HgN 3 , both of which are extremely explosive, but in other respects closely resemble the corresponding chlorides. The aqueous solution of the acid, as will be seen from the pro- perties already given, behaves in a similar manner to aqueous hydrochloric acid. Hence the group N 3 must be itself electronegative, and is analogous in its chemical properties with the halogens. The salts formed by azoimide with the metals of the alkalis of the alkaline earths are not nearly so explosive as those formed with the heavy metals, and azoimide itself when in dilute aqueous solution may with reasonable care be handled without danger. Other compounds of nitrogen and hydrogen have been prepared by Curtius ; these consist of the salts formed by ammonia and hydrazine with azoimide, and will be described with the other ammonium and hydrazine salts. 1 Curtius and Radenhausen, J. Pr. Chem. [2], 43, 207. HYDROXYLAMINE 475 HYDROXYLAMINE, NH 3 O. 268. This compound was discovered in 1865 by Lossen, l but until 1891 was only known in the form of salts or in aqueous solution. Lossen obtained it by the action of nascent hydrogen on nitric oxide, the reaction taking place as follows : 2NO+6H=2NH 3 O. Since that time it has been obtained in other ways, such for example as the reduction of nitric acid by metals under suitable conditions, 2 as a product of decomposition of the fulminates 3 (Vol. III., part L, p. 524), and by the action of sulphuretted hydrogen on silver nitrite. 4 By the interaction of nitrites and sulphites under suitable conditions, salts of hydroxylaminedisulphonic acid (p. 520) are formed, and these on heating with water are converted into hydroxylamine sulphate. To prepare it by this, the most con- venient method, a concentrated aqueous solution of sodium hy- drogen sulphite (2 mols.) is added to a similar solution of sodium nitrite (1 mol.), the temperature not being allowed to rise much above ; a sufficient quantity of potassium chloride is then added to convert the whole of the disulphonic acid into the less soluble potassium salt, which separates out in compact crystals. The latter are then heated in neutral aqueous solution at 100 for a long time, or for a short time at 130, and the resulting hydroxylamine sulphate separated from the sparingly soluble alkaline sulphates by fractional crystallisation. 5 For a longtime free hydroxylamine was only known in solution, but in 1891 the anhydrous compound was prepared almost simultaneously by Lobry de Bruyn, 6 and Crismer. 7 The former prepared it by dissolving hydroxylamine hydrochloride in absolute methyl alcohol, adding a solution of sodium methylate, NaOCH 3 , in the same solvent, separating the sodium chloride formed, and distilling off the greater portion of the methyl alcohol under 100 mm. pressure ; the residue is then distilled in small portions under 40 mm. pressure with the addition of a little vaseline to prevent frothing. As soon as solid 1 Annalen Suppl, 6, 240. 2 Journ. Chem. Soc. 1883, i. 443. 3 J. Pr. Chem. [2], 25, 233 ; Ber. 19, 993. 4 Journ. Chem. Soc. 1887, i. 48. 5 Raschig, Annalen, 241, 161 6 Rec. Trav. Chim. 10, 100 ; H, 18. 7 Bull. Soc. Chim. [3], 6, 793. 476 THE NON-METALLIC ELEMENTS hydroxylamine passes over, the receiver is changed and cooled to 0, care being taken that the hydroxylamine vapour is not exposed to air at 60 70 for any length of time, as violent explosions then take place. Crismer prepared the base by heating a double compound which it forms with zinc chloride, ZnCl,, 2NH 2 OH. Hydroxylamine forms white inodorous scales or hard needles, has a sp. gr. of about 1'3, melts at 33'05, and boils at 58 under 22 mm. pressure, the density of its vapour agreeing with the formula NH 3 O. When heated to 90 100 it decomposes, and detonates at a higher temperature. It inflames in a current of chlorine gas and combines violently with bromine and iodine, but without evolution of light. In the pure state it is stable, but in presence of alkali it gradually decomposes, the alkali dissolved from the less resistant forms of glass being sufficient to bring about this change. Strong oxidizing agents such as potassium permanganate or bichromate decompose it with production of flame or explosion, and sodium likewise attacks it with produc- tion of flame. On exposure to the air it liquefies owing to the absorption of moisture, and then undergoes oxidation with form- ation of a solid substance containing nitrous acid and ammonia ; it dissolves readily in water and to a less extent in ethyl and methyl alcohol and in boiling ether, and separates from the last- named solution in acicular crystals on cooling. The solutions have a strongly alkaline reaction, and cause the separation of the more easily reduced metals from their salts. From a solution of copper sulphate it precipitates red cuprous oxide, this reaction being sufficiently delicate to recognize one part of hydroxylamine in 1 00,000 parts of water. If sodium nitroprusside be added to a neutralized solution of hydroxylamine, and then a little caustic soda, the whole assumes on boiling a beautiful magenta red colour. 1 When hydroxylamine or its salts are treated with nitrous acid decomposition takes place rapidly, nitrous oxide being evolved. The reaction proceeds in two stages, hyponitrous acid being first formed according to the equation : HO.NH 2 + ON.OH = HO.N.-N.OH + H 2 0. This substance then splits up into its anhydride, nitrous oxide, and water. 2 The salts of hydroxylamine will be more fully described after 1 Angeli Gazzetta, 23, ii, 102. Wislicenus, Ber. 26, 771 ; Thnm, Monats. 14, 294. NITROGEN CHLORIDE 477 the ammonium salts in Vol. II., Part I. They all decompose on heating with effervescence, the nitrate yielding nitric oxide and water. NH 3 0, HN0 8 = 2NO + 2H 2 O. The behaviour of hydroxylamine towards organic substances shows that it must contain the hydroxyl group OH, and it is there- fore ammonia in which one atom of hydrogen is replaced by that group, its constitutional formula being N^-H COMPOUNDS OF NITROGEN WITH THE ELEMENTS OF THE CHLORINE GROUP. NITROGEN CHLORIDE, NC1 3 . 269. This dangerous body was discovered by Dulong l in 1811, who, notwithstanding the fact that he lost one eye and three fingers in the preparation of this body, yet continued its in- vestigation. A similar accident happened in 1813 to Faraday and Davy, who had, however, been made aware of the ex- plosive properties of this substance. " Knowing that the liquid would go off on' the slightest provocation, the experimenters wore masks of glass, but this did not save them from injury. In one case Faraday was holding a small tube containing a few grains of it between his finger and thumb, and brought a piece of warm cement near it, when he was suddenly stunned, and on returning to consciousness found himself standing with his hand in the same position, but torn by the shattered tube, and the glass of his mask even cut by the projected fragments. Nor was it easy to say when the compound could be relied on, for it seemed very capricious ; for instance, one day it rose quickly in vapour in a tube exhausted by the air-pump, but on the next day, when subjected to the same treatment, it exploded with a fearful noise and injuring Sir H. Davy." - The compound is formed by passing chlorine into a warm solution of sal-ammoniac, or when a solution of hypochlorous acid is brought into contact with ammonia (Balard). If a galvanic current be passed through a concentrated solution of sal-ammoniac, chloride of nitrogen is formed at the positive pole (Bottger, Kolbe). The liquid obtained by the action of 1 Schweigg Journ. 8, 302. - Gladstone, Life of Faraday, p. 10. 478 THE NON-METALLIC ELEMENTS chlorine on a solution of ammonium chloride has been carefully examined by Gattermann, who finds that it is not a homogeneous compound, but is a mixture of more or less completely chlorinated ammonias. If however it be mixed with a little water and chlorine passed over it for half an hour, it is converted into the completely chlorinated compound NClg. 1 The analysis of the liquid was carried out by placing a weighed quantity in water, and carefully adding ammonia solution which slowly converts it into nitrogen and ammonium chloride, the chlorine in the solution being then estimated as silver chloride. Chloride of nitrogen is a thin yellowish oil, which evaporates quickly on exposure to the air, has a sp. gr. of about 1*6, and possesses a peculiar smell, the vapour attacking the eyes and mucous membrane violently. Gattermann has shown that FIG. 140. it does not readily explode spontaneously, and may even be subjected to operations such as washing without much risk of explosion provided direct sunlight is excluded. When heated by itself to 95, or if it is brought in contact with certain bodies, such as phosphorus or turpentine, it explodes with great violence, giving out light, and pulverizing any glass or porcelain vessels in which it may be contained. In cold water it undergoes spontaneous decomposition with the evolution of chlorine, nitrogen, hydrochloric acid, and nitrous acid. The following method is employed for preparing this dangerous 1 JSer. 21, 751. NITROGEN CHLORIDE 479 substance in small quantities, and for showing its explosive properties without risk. A flask of about two litres capacity, having a long neck, is filled with chlorine, and placed mouth downwards in a large glass basin filled with warm saturated solution of sal-ammoniac as shown in Fig. 140. Below the neck of the flask is placed a small thick leaden saucer, in which the chloride of nitrogen is collected. The solution of sal-ammoniac absorbs the chlorine, and, as soon as the flask is three parts filled by the liquid, oily drops are seen to collect on the surface inside the flask. These gradually increase, and at last drop one by one down into the leaden saucer. When a few drops have collected, the leaden saucer may be carefully removed by a pair of clean tongs, another being placed in its stead. A small quantity of the chloride of nitrogen may be exploded by touching it with a feather moistened with turpentine attached to the end of a long rod. Another drop of the oil may be absorbed by filtering paper, and when this is held in a flame a loud explosion like- wise ensues. FJG. Hi. The force of the explosion may be rendered still more evident by placing the flask in a strongly constructed box with glass sides, and when the drops of chloride of nitrogen collect on the surface of the solution, passing in some turpentine from a stoppered funnel ; the latter is kept outside the box and is con- nected with it by means of rubber and glass tube the end of which dips under the neck of the flask. When the turpentine reaches the surface of the solution in the flask a bright flash is seen, and a violent thunderlike explosion occurs, completely shattering the flask. 1 The formation and properties of chloride of nitrogen may be also exhibited in the following way. A solution of sal- ammoniac, saturated at a temperature of 28, is brought into a glass basin (A, Fig. 141), and a cylinder (B), the lower end of 1 V. Meyer, Ber. 21, 26. 480 THE NON-METALLIC ELEMENTS which is closed by a piece of bladder, is also filled with the solution and placed upright in the basin. A layer of oil of turpentine is then poured on to the top of the liquid in the cylinder, and a platinum plate (a) in contact with the positive pole of a battery of six cells placed in the cylinder, whilst the negative pole (ft) is placed under the bladder in the basin. Yellow oily drops soon begin to form on the surface of the positive pole, and these gradually become detached from the pole, and rise in the liquid until they come in contact with the layer of turpentine, when they explode. BROMIDE OF NITROGEN. When bromide of potassium is added to chloride of nitrogen under water, potassium chloride and bromide of nitrogen are formed. The latter is a dark red very volatile oil, possessing a powerful smell, and is as explosive as chloride of nitrogen (Millon). IODIDE OF NITROGEN. 270 When iodine is brought into contact with aqueous or alcoholic ammonia, a black powder is formed which, when dried, decomposes spontaneously with a very violent detonation either when touched, or when slightly heated, violet vapours of iodine being emitted. This compound gradually decomposes under cold water, pure iodine being left behind. Warm water facilitates this decomposition, and the body explodes violently when thrown into boiling water. The composition of this black powder appear to vary accord- ing to the process employed in it preparation. Serullas, 1 who first prepared this body, precipitated an alcoholic solu- tion of iodine with aqueous ammonia ; the compound thus obtained possesses, according to Gladstone, 2 the composition NHI 2 ; but according to Stahlschmidt, 3 its composition is represented by the formula NI 3 , a body having the former composition being obtained when alcoholic solutions of iodine and ammonia are mixed. Bunsen, on the other hand, found that under these circumstances, and when absolute alcohol is used, a precipitate having the composition N 2 I 3 H 3 , or NH 3 ,NI 3 is formed, and that on precipitating an aqueous solution of 1 Ann. Chim. Phys. 42, 200. 2 Joum. Chem. Soc. 1852, 34. 3 Pogg. Ann. 119, 421. OXIDES OF NITROGEN 481 chloride of iodine with ammonia a black powder is obtained, which consists of NH^NIg. 1 These observations render it probable that under different circumstances at least two distinct iodides of nitrogen are formed, one being derived from ammonia by the replacement of the whole, and the other by the replacement of a portion of the hydrogen by iodine, thus : (a) 4NH 3 + 3I 2 = NI 3 + 3NH 4 L (V) 3NH 3 + 2I 2 = NHI 2 + 2NH 4 L Iodide of nitrogen would thus appear to be capable of combining with ammonia, giving rise to the compounds described by Bunsen. Szuhay 2 has recently shown that the compound obtained by adding an excess of aqueous ammonia to a strong solution of iodine in concentrated aqueous potassium iodide has the composition NHI 2 , its formation being represented by equa- tion (&) above. When suspended in water and mixed with an ammoniacal solution of silver nitrate it yields a black substance having the composition NAgI 2 , which is likewise explosive, .and is decomposed by boiling water and dilute acids. If it be treated with a solution of a soluble cyanide the silver may be replaced by other metals, but the compounds thus formed have not been isolated ; they regenerate the compound NAgI 2 on addition of a silver salt. The compound NHI 2 therefore possesses acid properties. NITROGEN AND OXYGEN. OXIDES AND OXY-ACIDS OF NITROGEN. 271 As already mentioned nitrogen burns in oxygen when both gases are heated to a sufficiently high temperature (p. 451), and direct combination also takes place more slowly when a series of electric sparks is passed through fairly dry air, whilst if much moisture is present nitric acid is produced. When a mixture of nitrogen and oxygen is passed over platinum black at 250, oxides of nitrogen are also formed, 3 and they are further produced when carbon is burnt in highly compressed air. 4 We 1 Annalen, 84, 1 2 Ber. 26, 1933. 3 Loew, Ber. 23, 1443. 4 Hempel, Ber. 23, 1457. 482 THE NON-METALLIC ELEMENTS are acquainted with five oxides of nitrogen, 1 and three oxygen acids corresponding to the oxides numbered 1, 3, and 5. Oxides. Acids. 1. Nitrous Oxide, or N ) n A . , HO.N \ Nitrogen Monoxide N / e lcl ' HO.N } 2. Nitric Oxide, or -^-Q Nitrogen Dioxide 3. Nitrogen Trioxide -^-Q ! O Nitrous Acid . . TT | J J 4. Nitrogen Peroxide, or -^Q Nitrogen Tetroxide 5. Nitrogen Pentoxide Mr . 2 \ O Nitric Acid ... T| r O IM u ) n ) NITRIC ACID. HN0 3 . 272 In Geber's tract, De Inventione Veritatis, we find the following description of a mode of preparing nitric acid or aqua- fortis : " Sume libram unam de vitrioli de cypro. et libram salis petrse, et unam quartam aluminis Jameni, extrahe aquam (the acid) cum rubidine alembici." That is, by strongly heating a mixture of saltpetre, alum, and sulphate of copper, the nitric acid distils over, owing to the decomposition of the saltpetre by the sulphuric acid of the other salts. Nitric acid was commonly prepared and used as a valuable reagent by the alchemists, especially as a means of separating gold and silver. The method of preparation which we now use from nitre and oil of vitriol appears to have been first employed by Glauber, for long afterwards the acid thus obtained was called Spiritus nitri fumans Glanberi. The first theory respecting the composition of nitric acid was proposed by Mayow in 1669. 2 He believed that the acid contained two components, one derived from the air and having a fiery nature, and the other derived from the earth. More than a century later, in 1776, Lavoisier showed that one constituent of nitric acid is oxygen, but he was unable to satisfy himself as to the nature of the other components, and it was reserved for Cavendish 3 to prove the exact composition 1 According to Berthelot and Hautefeuille and Chappuis, a sixth oxide, pos- sessing the composition N 2 6 , exists. It is formed by the action of an electric discharge on a mixture of nitrogen peroxide and oxygen. (See Ann. Chem. Phys. [5], 22, 432 ; Compt. Rend. 92, 80 > 134 > 94, 1111, 1306.) 2 Mayow, De sal-nitro et spiritu nitri aereo. 3 Phil. Trans. 1784, p. 119. Do. 1785, p. 372. NITRIC ACID 483 and mode of formation of this acid or its salts by the direct combination of oxygen and nitrogen gases in presence of water or alkaline solutions. Priestley had already observed that when a series of electric sparks was made to pass through common air included between short columns of a solution of litmus, the solution acquired a red colour and the air was diminished in volume. Cavendish repeated the experiment, using lime-water and soap-lees (caustic potash) in place of the litmus, and he concluded that the lime-water and soap-lees became saturated with some acid formed during the operation. He proved that this was nitric acid by passing the electric discharge through a mix- ture of pure dephlogisticated air (oxygen) and pure phlogisticated air (nitrogen) over soap-lees (caustic potash), when nitre (potas- sium nitrate) was formed. Cavendish J clearly expresses his FIG. 142. views in the following words : " We may safely conclude that in the present experiments the phlogisticated air was enabled, by means of the electric spark, to unite to, or form a chemical combination with, the dephlogisticated air, and was thereby reduced to nitrous acid, which united to the soap-lees, and formed a solution of nitre ; for in these experiments those two airs actually disappeared, and nitrous acid was actually formed in their room." The apparatus which he employed, represented in Fig. 142, consisted of a bent syphon-tube containing, in the bent portion, the mixture of gases, and mercury in the limbs, the open ends dipping under mercury in the glasses. Nitric acid is also formed when various bodies are burnt in a mixture of oxygen and nitrogen. Thus, if three to five volumes of the detonating mixture of oxygen and hydrogen be mixed with one volume of air in a eudiometer over mercury, and 1 Phil. Trans. 1785, p. 379. 484 THE NON-METALLIC ELEMENTS an electric spark passed through the mixture, instantaneous combination takes place and nitric acid is formed-, the surface of the mercury becoming covered with crystals of mercurous nitrate. 1 In order to exhibit the direct combination of oxygen and nitrogen, it is only necessary to allow the sparks from an in- duction-coil to pass between two platinum wires placed in the interior of a large glass globe containing dry air, as shown in Fig. 143. Red fumes of nitrogen peroxide are rapidly formed, and their presence may be distinctly recognized by plunging a piece of iodized starch paper into the globe, when the blue iodide of starch will at once be produced. On pouring a few FIG. 143. drops of water into the globe and shaking it up, the red fumes are absorbed and nitrous and nitric acids formed, as may be shown by the acid reaction of the liquid. In a similar manner, if a flame of hydrogen be allowed to burn in a large flask into which oxygen is led, but not in such quantity as to displace the whole of the air, nitric acid is formed in large quantity. 2 The same acid is also produced when ammonia burns in oxygen. The formation of nitric acid by the discharge of electricity through the air accounts for the pres- ence of this acid in the atmosphere. Another and more productive source of the nitrates has yet to be described. When nitrogenous organic matter is exposed 1 Bunsen, Gasometry, p. 58. 2 Kolbe, Annalen, 119, 176 ; Hofmann, Ber. 3, 663. PREPARATION OF NITRIC ACID 485 to the air the nitrogen assumes the form of ammonia ; but when alkalis, such as potash, soda, or lime, are present, a further slow oxidation of the nitrogen takes place, and nitrates of these metals are formed. Hence these nitrates are widely diffused in all surface soils, especially in hot countries such as India, where oxidation takes place quickly. Soil in the neighbourhood of the Indian villages, which contains con- siderable amounts of potash, thus becomes rich in nitre or potassium nitrate, KNO 3 , originating from the decomposition of the urea of the urine. It is from this source that the largest quantity of nitre imported into this country is obtained. Another nitrate, lime-saltpetre or calcium nitrate, Ca(NOg). 2 , is often found as an efflorescence on the walls of buildings, such as stables, or cellars of inhabited houses, and this source was made use of during the French revolution for the manufacture of nitre. Chili saltpetre, or sodium nitrate, NaN0 3 , occurs in large deposits in the province of Tarapaca, a rainless district on the Peruvian coast lying near the 20th degree of south latitude. The nitrate of soda is found mixed with chloride of sodium and other salts, probably pointing to the fact that the locality has been covered by the sea, and that the nitrate is a product of the decomposition of sea-plants and animals. Preparation. In order to prepare pure nitric acid, saltpetre which has been previously well dried is pla'ced in a tubulated retort together with an equal weight of concentrated sulphuric acid. On heating the mixture, the volatile nitric acid passes over, and is collected in a well-cooled receiver, see Fig. 144, hydrogen potassium sulphate (commonly called bisulphate of potash) remaining behind in the retort ; thus : KN0 3 + H 2 S0 4 = HN0 3 + KHSO 4 . The distillate thus obtained has a yellowish colour, caused by the presence of nitrogen peroxide, and it is not free from water, as the concentrated acid, when heated, decomposes partially into peroxide, oxygen and water. In order to purify the acid thus obtained it must be again distilled with its own volume of con- centrated sulphuric acid, and the distillate freed from traces of the peroxide by gently warming the acid, and leading a current of dry air through it until it is cold. Thus prepared it contains from 99'5 to 99'8 per cent, of the anhydrous acid, HN(X. 486 THE NON-METALLIC ELEMENTS 273 Properties. Nitric acid is a colourless liquid fuming strongly in the air. It possesses a peculiar though not very powerful smell, and absorbs moisture from the air with the greatest avidity. Nitric acid is an extremely corrosive sub- stance, which, when brought in contact with the skin, produces painful wounds, being used in surgery as a powerful cautery. The dilute acid acts less energetically, and colours the skin, nails, wool, silk, and other organic bodies of a bright yellow tint. When pure concentrated nitric acid is heated, it begins to boil at 86, and becomes of a dark-yellow colour owing to the decomposition of a portion of the acid into nitrogen FIG. 144. peroxide, oxygen, and water. As soon as about three-fourths of the acid has distilled over, the residue becomes colour- less, and then contains only 95*8 per cent, of acid. 1 If the distillation is pushed further, the boiling point continually rises, a strong acid distils over, and the residue becomes constantly weaker until it contains 68 per cent, of acid, when the liquid is found to boil unaltered at 120'5 under the normal atmospheric pressure, yielding an acid of the above constant composition, and with a specific gravity of T414 at 15'5. This constant acid is always obtained, whether a stronger or a weaker acid be subjected to distillation. If this acid of constant composition be distilled under an increased, or under a diminished pres- 1 Roseoe, Journ. Chem. Soc. 1861, i. 147. PROPERTIES OF NITRIC ACID 487 sure, the composition of the residual acid again undergoes a change, until for each pressure a constant boiling point is reached. Thus, under the pressure of l"22m. of mercury, an acid containing 68'6 per cent, of HNO 3 distils without alteration, whilst under a pressure of 0'070m. an acid distils over at a tem- perature of from 65 to 70, having a constant composition of 66'7 per cent. When a current of dry air is passed through aqueous nitric acid, either a stronger or a weaker acid is volatilized, according to the concentration or the temperature of the acid, until at length a residue is obtained which volatilizes unchanged. Thus, when the experiment is made at 100, the residual acid contains 66*2 per cent. ; when at 60, 64 '5 per cent. ; and at 15 the residual acid contains 64'0 per cent, of HNO 3 . From this it will be seen that nitric acid behaves in a similar way in this respect to hydrochloric and the other aqueous acids. When the concentrated acid is mixed with water an in- crease of temperature and a contraction of bulk is observed. This attains its maximum when one molecule of the acid is mixed with three molecules of water (Kolbe). The following table gives the specific gravities of aqueous acids at and 15 i 1 Per cent. Sp. gr. Sp. gr. HN0 3 . at 0. at 15. 100-0 1-559 1-530 90-0 1-522 1-495 80-0 1-484 1-460 70-0 1-444 1-423 60-0 1-393 1-374 50-0 1-334 1-317 40-0 1-267 1-251 30-0 1-200 1-185 20-0 1-132 1-120 15-0 1-099 1-089 10-0 1-070 1-060 5-0 1-031 1-029 It has been already remarked that concentrated nitric acid begins to decompose, at a temperature of 86, into water, oxygen, and nitrogen peroxide. If the acid be more strongly heated in closed glass tubes this change takes place so rapidly that at 260 the whole of the nitric acid is thus decomposed (Carius). 1 Kolbe, Ann. Chim. Phys. [4], 10, 140. 488 THE NON-METALLIC ELEMENTS In order to exhibit this decomposition by means of heat, the same apparatus serves which was employed for the decomposi- tion of sulphuric acid, Fig. 145. Strong nitric acid is allowed to fall on the hot pumice-stone contained in the platinum flask. Immediately red vapours are emitted, and these are condensed in a U-tube placed in a freezing mixture to a brown liquid, NO 2 , whilst the cylinder placed over the pneumatic trough becomes filled with a colourless gas which can be easily shown to be oxygen (Hofmann). Fm. 145. Pickering has succeeded in isolating two hydrates of nitric acid, having the composition HN0 3 , H 2 O and HNO 3 , 3H 2 O respectively. The former separates in large fairly transparent crystals, and the latter in smaller, more opaque, gritty crystals. 1 274 Commercial Manufacture. In order to prepare nitric acid on the commercial scale, sodium nitrate is substituted for nitre, as it is much cheaper, and the salt is decomposed with sulphuric acid as before. The proportions of these two substances em- ployed are not the same in all works. If one molecule of 1 Journ. Chem. Soc. 1893 i. 441. MANUFACTURE OF NITRIC ACID 489- 490 THE NON-METALLIC ELEMENTS sulphuric acid and two of sodium nitrate be taken, the follow- ing are the reactions. In the first place we have : H 2 S0 4 + NaN0 3 = NaHSO 4 + HNO 3 . When the heat is raised, the acid sodium sulphate acts upon a second molecule of sodium nitrate ; thus : NaHS0 4 + NaN0 3 = Na. 2 S0 4 + HNO 3 . In this case, however, a part of the acid is decomposed owing to the high temperature, and nitrogen peroxide is evolved in the form of red fumes which dissolve in the concentrated acid, giving it the red appearance usually noticed in the strong commercial product. When a large excess of sulphuric acid is employed, a certain quantity of acid sodium sulphate is formed, which lowers the melting point of the residual mass so that it can be withdrawn from the retorts in the fused state, whereas in the other case the residue can only be removed in the solid state after the cylinder has been cooled. The ordinary com- mercial acid has a specific gravity of from 1*38 to T41, and is usually prepared by means of chamber (sulphuric) acid ; but if a more concentrated acid is required a stronger sulphuric acid must be employed. The strongest nitric acid occurring in com- merce has a specific gravity of 1'53, and this is obtained by distilling well-dried Chili saltpetre with sulphuric acid having a specific gravity of T85. The retorts in which nitric acid is usually prepared on the large scale in England consist of cast- iron cylinders built in a furnace in such a way that they may be heated as uniformly as possible, as shown in Fig. 146. Some manufacturers cover the upper half of the cylinder with fire- bricks in order to protect the iron from the action of the nitric acid vapours. This, however, is unnecessary, if the retorts are so thoroughly heated that no nitric acid condenses on the surface of the iron. The ends of the cylinders which are not exposed to the action of the flame are closed by plates of Yorkshire flag cemented on to the iron with a mixture of iron filings, sulphur, sal-ammoniac, and vinegar. In the upper part of one of these flags is a hole, through which the sulphuric acid is intro- duced after the charge of Chili saltpetre. The hole is then closed by a clay plug. A similar hole in the other flag is furnished with a bent earthenware tube (c) passing into a series of large Woulffe's bottles (bfy, one placed behind the other, containing MANUFACTURE OF NITRIC ACID 491 small quantities of water in which the nitric acid condenses, and from which the acid is withdrawn by leaden syphons. The last of these Woulffe's bottles is placed in connection with a tower filled with coke, down which a current of water runs. Any uncondensed nitrogen peroxide passes up this tower, and, coming in contact with the water and the oxygen of the air, is FIG. 147. oxidized to nitric acid. When the operation is complete, one of the flags is removed and the residual fused sulphate of soda scraped out. A usual charge for one retort is 305 kilos of Chili saltpetre and 240 kilos of strong sulphuric acid. The mixture is heated uniformly for about eighteen hours, and the quantity of water placed in the Woulffe's bottles is such that a yield of 363 kilos of nitric acid, of specific gravity 1*35, is obtained, 492 THE NON-METALLIC ELEMENTS whilst 295 kilos of fused sodium sulphate remain behind in the cylinder. In a German factory, where the strongest nitric acid is made, a cast-iron vessel is employed for its generation, the construction of which is seen in Fig. 147. The charge, in this case, consists of 700 kilos of sulphuric acid, of specific gravity 1*84, and 600 kilos of Chili saltpetre. The retort is placed in connection with two series of receivers, twenty-five in number According to this process 100 parts of Chili saltpetre, containing 96 parts of pure nitrate, yield 68 parts of nitric acid, of specific gravity 1*5, and 17 parts of weaker acid. This corresponds to about 96 per cent, of the theoretical yield. A new arrangement has lately been devised, whereby a more rapid generation and condensation is possible than by the older processes. According to this plan, proposed by Guttmann, a charge of 610 kilos can be worked off in 10 11 hours with a yield of strong acid only 3 7 per cent, below the theoretical amount, whilst if the fumes are condensed by water and the weak acid is used, there is absolutely no loss in the process. The charge of sodium nitrate and sulphuric acid is brought into an iron retort similar to that shown in Fig.147, bat which allows of the spent sodium bisulphate being drawn off at the bottom ; in order to reduce the quantity of nitrogen peroxide in the condensed acid to a minimum, a current of air is injected into the gaseous products, the mixed gases then passing into a very carefully made system of earthenware pipes arranged in a some- what similar manner to the well-known atmospheric condensers of the gas-works, which are cooled by water. Each set of upright pipes is connected with a main, into which the acid falls, and flows away into a receiver placed below. The uncondensed gases pass away into a tower built of earthenware, and filled with perforated plates of special construction, down which water is continuously allowed to trickle. The arrangement of the perforated plates is the invention of Prof. Lunge. The wash tower is of great value even if the weak acid be thrown away, inasmuch as without it the draught becomes too great and serious escape of nitrous gas takes place. This process is particularly useful where a very strong acid is needed, as for example in the preparation of nitro-glycerine and gun-cotton. 1 Commercial red nitric acid always contains chlorine, and sometimes iodine in the form of iodic acid, originating from the 1 Journ. Soc. Chem. 2nd. 1893, 203. NITRIC ACID 493 Chili saltpetre. In addition it also contains nitrogen peroxide, iron oxide, sulphuric acid, and sodium sulphate, which have been mechanically carried over. In order to purify the acid, it must be distilled in glass retorts, chlorine and nitrogen peroxide coming over in the first portion. As soon as the acid distillate is free from chlorine the receiver is changed, and the liquid may be distilled until only a small residue is left, containing the whole of the iodic acid, sulphuric acid, and sodium sulphate. Concentrated, as well as dilute nitric acid is largely used in the arts and manufactures. Large quantities are employed in the manufacture of the various coal-tar colours, of nitro-glycerine, of gun-cotton, of sulphuric acid, and of nitrate of silver, which is largely used for photographic purposes. The acid is also used in large quantities for the preparation of certain nitrates, especially lead nitrate, iron nitrate, and aluminium nitrate, all of which are employed in the processes of dyeing and calico-printing, whilst the nitrates of barium and strontium are used for pyrotechnic purposes. In the laboratory it is an indispensable reagent, and is used in the preparation of a large number of inorganic and organic substances. Nitric acid is a monobasic acid forming a series of salts which are termed the nitrates. These are almost all easily soluble in water, and as a rule crystallize well. They may be obtained by neutralizing the acid with an oxide or a carbonate, and ;are almost all formed by dissolving the metal in nitric acid. In this case the metal is oxidized at the expense of a portion of the acid which, according to the concentration or the temperature, is reduced to N0 2 , N 2 O 3 , NO, N 2 O, and even to nitrogen and .ammonia. Several other bodies, such as sulphur, phosphorus, carbon, .and many organic substances, are easily oxidized, especially by the concentrated acid. In order to exhibit this action, some nitric acid may be poured upon granulated tin, which is then oxidized with the evolution of dense red fumes, whilst a white powder of tin oxide is deposited. Turpentine when poured into the concentrated acid is likewise oxidized with .almost explosive violence, light and heat being evolved. In like manner ignition may take place when straw or sawdust becomes impregnated with the strong acid. Other organic bodies treated with nitric acid undergo no .apparent alteration. Thus, for instance, with cotton-wool no ignition or evolution of red fumes occurs. If, however, the 494 THE NON-METALLIC ELEMENTS cotton-wool after having been thus soaked in strong nitric acid is washed and dried, it is found to possess very different properties from ordinary cotton, although in appearance it can hardly be distinguished from it. Cotton- wool, or cellulose, has the formula C 12 H 20 O 10 , whilst after treatment with nitric acid it consists of nitro-cellulose, gun-cotton or collodion-cotton. This consists of cellulose in which some of the hydroxyl groups varying in number from four to six according to the strength of nitric acid employed, are replaced by NO 3 , the changes which have occurred being represented by the equation : C 12 H 20 10 + 4HN0 3 = C 12 H 16 6 (N0 3 ) 4 + 4H 2 O, or C J2 H 20 10 + 6HN0 3 = C 12 H U 4 (NO 3 ) 6 + 6H 2 0. Nitric acid acts in a similar way on many other organic bodies. 275 Action of Nitric Acid on Metals. As already mentioned, nitric acid dissolves a large number of metals with formation of nitrates. Hydrogen is not evolved at the same time as is the case with sulphuric and hydrochloric acids, but in its place lower oxides of nitrogen and even nitrogen itself and ammonia are formed. The explanation usually given of this change is that hydrogen is first evolved, but that it at once acts on the excess of nitric acid present, forming water and the lower oxides of nitrogen. Thus, for example, the formation of nitric oxide by the action of copper on nitric acid is supposed to take place in the two following stages : Cu + 2HN0 3 = Cu(NO 3 ) 2 + 2H, According to Veley, however, this explanation is not correct, inasmuch as pure copper, mercury and bismuth do not dissolve in pure dilute nitric acid, but dissolve readily when nitrous acid is present, the change at any moment being directly proportional to the mass of nitrous acid in the solution, and the more rapid the greater the proportion of the former to the latter. He there- fore believes that the reaction is started either by traces of nitrous acid already present, or by impurities in the metal inducing alocal galvanic current ; the first product of the reduction of the nitric acid is nitrous acid, and the production of lower oxides of nitro- gen he regards as due to the subsequent changes occurring be- tween nitrous acid and cuprous and cupric nitrate or nitrite in DETECTION OF NITEIC ACID 495 presence of an excess of nitric acid, the nitrous acid being decomposed as fast as it is formed. 1 Detection and Estimation. The presence of nitric acid or of its salts is easily ascertained. Thus for instance, if we heat a few drops of not too dilute nitric acid with copper turnings, brownish red vapours of nitrogen peroxide are emitted. The nitrates give the same reaction if they, or their concentrated aqueous solutions, are treated with sulphuric acid and copper. In order to detect nitric acid, or a nitrate in a very dilute solution, a cold solution of ferrous sulphate is added to the liquid, and about an equal volume of strong sulphuric acid is then allowed to flow slowly down to the bottom of the inclined test-tube, care being taken that the two liquids do not mix. If nitric acid be present it is liberated by the action of the sulphuric acid, and at once oxidises the ferrous sulphate where it comes in contact with it, nitric oxide being simultaneously given off. The latter at once unites with the excess of ferrous sulphate forming a dark brown compound, the solution of which is seen as a dark-coloured ring at the point where the dense sulphuric acid joins the lighter aqueous solution. Another very delicate test for the presence of nitric acid is aniline. In order to apply this test, ten drops of aniline are brought into 50 cc. of a dilute sulphuric acid containing 15 per cent, of pure acid, and 0'5 cc. of this solution is poured on to a watch-glass together with 1 cc. of concentrated sulphuric acid. If a glass rod moistened with the solution under exami- nation be now brought in contact with the edge of the liquid in the watch-glass, a red streak will be produced if a nitrate be present, and the colour will increase in intensity until the whole liquid becomes red. If a larger quantity of nitric acid be present, the whole mass will assume more or less of a brown tint (C. D. Braun). The organic base called brucine, bearing a close resemblance to strychnine, serves as a test for nitric acid even more delicate than aniline. If to half a drop of a solution of one part of nitric acid to 100,000 parts of water, one or two drops of a solution of brucine be added, and then a few drops of concentrated sulphuric acid, a distinct pink coloration will be observed if the solution be viewed against a white ground. 2 In order quantitatively to determine the amount of nitric acid 1 Proc. Roy. Soc. 46, 216 ; 52, 27 ; Phil. Trans. 1891 (A), 312 ; Journ. Soc. Chem. Ind. 1891, 204. 2 Reichardt, Jahresbcricht, 1871, 893. 496 THE NON-METALLIC ELEMENTS contained in potassium- or sodium-saltpetre, the well-dried sub- stance is heated to dull redness for half an hour with freshly- ignited and finely-powdered quartz or silica, SiO 2 . The nitrates are thus completely decomposed, whilst any sulphates or chlorides which may be present undergo no change. The decomposition which here takes place may be represented as follows : 2KN0 3 = K 2 -t 2N0 2 + O. -Oxygen and nitrogen peroxide are evolved, whilst the potash combines with the silica to form silicate of potash. From the loss of weight thus ensuing the amount of nitre present can easily be calculated. Another good method, which is particularly useful in the determination of the nitrates contained in drinking water, depends upon the fact that a thin zinc plate, which has been covered with a deposit of spongy metallic copper by dipping it into a solution of copper sulphate, on being heated with water containing nitrates reduces them to ammonia, zinc hydroxide and free hydrogen being at the same time formed (Gladstone and Tribe) ; thus : KN0 3 + 8H = NH 3 + KOH + 2H 2 O. The ammonia thus obtained is distilled over into an excess of hydrochloric acid and determined in the usual way, or if present in only small quantities by nesslerisation (see p. 300). A further method, proposed by Lunge, consists in mixing the nitrate with sulphuric acid and shaking in a graduated tube with mercury; the whole of the nitrogen is liberated in the form of nitric oxide, the volume of which is measured. The reaction which takes place is represented by the following equation : 3Hg + 3H 2 S0 4 + 2HN0 3 = 3HgS0 4 + 4H 2 + 2NO. A special apparatus known as the nitrometer has been designed by Lunge for carrying out this reaction l which is especially useful in the determination of the nitrogen contents of organic nitrates such as the nitro-celluloses. 1 Ber. 11, 434. NITROGEN PENTOXIDE 497 AQUA REGIA. 276 This name is given to a mixture of nitric and hydro- chloric acids which is frequently employed for dissolving the noble metals, such as gold and platinum, as well as many metallic ores and other bodies. A method of preparing this substance was described by Geber in his work De Inventions Veritatis, by dissolving sal-ammoniac in nitric acid ; and he states that the liquid thus obtained has the power of dissolving gold and sulphur. The name, aqua regia, is first found in the writings of Basil Valentine. He, like Geber, prepared it by dissolving four ounces of sal-ammoniac in 1 Ib. of aqua-fortis; he also states that strong aqua regia can be obtained by mixing hydrochloric and nitric acids. The solvent power of aqua regia depends upon the fact that, on heating, this mixture of acids evolves chlorine ; thus : HN0 3 + 3HC1 = 2H 2 + NOC1 + CL, The compound nitrosyl chloride, NOC1, which is liberated at the same time is described on page 510. NITROGEN PENTOXIDE, N 2 O 5 = 107-28. 277 This substance, which is commonly called nitric anhydride, was discovered in 1849 by Deville, 1 who obtained it by leading perfectly dry chlorine gas over dry silver nitrate contained in a U-'tube placed in a water-bath. The reaction begins at 95, and when cooled to 60, the decomposition of the nitrate goes on regularly. The pentoxide is collected in a bulb tube surrounded by a freezing mixture. The following equation represents the reaction which takes place : 4AgN0 3 + 2C1 2 = 4AgCl + 2N 2 O 5 + O 2 . In the preparation of the substance all joints of cork and caout- chouc must be avoided, and the parts of the glass apparatus must be connected either by fusion, or by placing the end of one tube inside the other and closing the space between the tubes with asbestos, the pores of which are filled up with melted paraffin. Nitrogen pentoxide can be also prepared still more simply 1 Ann. Chim. Phys [3] 28, 241. 88 498 THE NON-METALLIC ELEMENTS from pure perfectly anhydrous nitric acid by withdrawal from this substance of the elements of water. For this purpose nitric acid is distilled two or three times with concentrated sulphuric acid to remove all water, and is then intro- duced into a retort to the neck of which a long glass tube is fused to serve as a condenser. Phosphorus pentoxide is next gradually added until the mixture has a syrupy consistency, the retort being cooled with ice-water during the addition ; if the nitric acid has been properly dehydrated no hissing noise is heard on addition of the anhydride. The reaction which here takes place may be thus represented : 2HN0 3 + P 2 5 = N 2 6 + 2HP0 3 . By gently heating the retort, a deep orange-coloured distillate is obtained, which on standing separates out into two layers. The upper, or lighter layer is then poured into a thin stoppered tube and cooled down by plunging the tube into ice-cold water. Crystals of the pentoxide soon separate out, and these may be purified by pouring off the orange-coloured liquid from which they are deposited, melting the crystals at a moderate heat, again allowing them to deposit, and finally pouring off the mother liquor. Properties. Nitrogen pentoxide is a white colourless solid, crystallizing in bright rhombic crystals, or in six-sided prisms derived from these. When heatd to 15 20 the crystals become of a yellowish colour, and molt at about 30 to a dark yellow liquid, which decomposes between 45 and 50 with the evolution of dense brown fumes. When suddenly heated, the pentoxide decomposes with explosive violence into nitrogen peroxide and oxygen, and this sudden decomposition occurs sometimes even at ordinary temperatures, if the crystals have been kept for some time. The lower the temperature is kept the longer does the substance remain unaltered, and below 30 it may be sublimed in a closed vessel, depositing in crystals in the cool part of the tube. In dry air the pentoxide volatilizes very quickly, whilst in moist air it deliquesces with formation of nitric acid. Thrown into water, it dissolves with evolution of heat, forming nitric acid ; thus : The pentoxide possesses very powerful oxidizing properties. NITROUS OXIDE 499 Thus, if brought in contact with sulphur it forms white vapours, which condense to a white sublimate of nitrosulphonic anhy- dride, S 2 O 5 (NO 2 ) 2 . Phosphorus and potassium burn with brilliancy in the slightly warmed anhydride. Charcoal does not decompose even the boiling anhydride, but when ignited and brought into the vapour it burns with a brilliant light. When brought in contact with nitric acid, the anhydride com- bines to form the compound N 2 O 5 + 2HN0 3 . This substance, which is a liquid at the ordinary temperature, possesses at 18 a specific gravity of 1'642, and solidifies at 5 to a crystalline mass. It forms the heavy layer obtained in the preparation of the pentoxide, and decomposes with explosion when heated. Its formation is perfectly analogous to that of disulphuric acid, and the constitution is probably represented by the formula N0 2 NO OH. O NO OH. NITROGEN MONOXIDE OR NITROUS OXIDE, N 2 O = 43*76. 278 This gas is formed by the action of easily oxidizable sub- stances, such as potassium sulphide, moist iron filings, the sul- phites, and other bodies upon nitric oxide, and according to these methods it was first prepared by Priestley in the year 1772. It is moreover formed when zinc and other metals are dissolved in very dilute nitric acid, and by the action of sulphur dioxide on nitric oxide in presence of moisture 1 Preparation. In order to prepare the gas we do not, how- ever, usually employ any of these methods, but we have re- course to the decomposition which ammonium nitrate undergoes on heating. This salt splits up into water and nitrous oxide gas ; thus : NH 4 N0 3 = N 2 O + 2H 2 O. It is best, before the experiment, to melt the nitrate, in order to free it from moisture ; and the powdered dry substance is then introduced into a flask furnished with a cork and delivery tube. The flask must be heated gently until a regular evolution of gas begins, and then the flame moderated, as sometimes, if the 1 Lunge, Ber. 14, 2196. 500 THE NON-METALLIC ELEMENTS heat applied be too great, the decomposition takes place so violently, with evolution at the same time of nitric oxide, that an explosion may occur. In order to free the gas from traces of nitric oxide it can be shaken up with a solution of ferrous sulphate, which combines with the latter gas ; whilst in order to remove traces of chlorine derived from the chloride of ammo- nium, which the commercial nitrate often contains, it must be allowed to stand over a solution of caustic potash or soda. These precautions are especially needed when the gas is used for inhaling. A regular stream of the gas may also be obtained by heating sodium nitrate with a slight excess of ammonium sulphate at 240 . 1 FIG. 148. As nitrous oxide is somewhat soluble in cold, but not nearly so soluble in hot water, it is best to fill the pneumatic trough with warm water before collecting the gas. The arrangement used for this purpose is seen in Fig. 148, and requires no further explanation. 279 Properties. Under ordinary circumstances nitrogen mon- oxide is a colourless gas possessing a pleasant smell and sweet, agreeable taste, and having a specific gravity of 1*52 (Colin). Its solubility in water between and 25 is represented by the formula c = 1-30521 - 0-045620 + 0-0006843* 2 ; or its coefficients of absorption are as follows : 1 W. Smith, Jaurn. Soc. Chem. Ind., U, 867, 12, 10. 2 Carius, Annalen, 94, 140. NITROUS OXIDE 501 5 10 15 20 25 1-3052 1-0954 0'9196 07778 0'6700 0*5962 It is still more soluble in alcohol, one volume of this liquid absorbing, according to the experiments of Carius, a quantity of the gas found by the formula c = 4-17805 - 0-0698160^ + 0'0006090 2 . Nitrous oxide was first liquefied by Faraday in 1823 by heating nitrate of ammonium in a bent tube (Fig. 149). The liquid is now prepared on the manufacturing scale by compressing the gas into strong cylinders ; it is a colourless very mobile liquid, which has a specific gravity of 0'9369 1 at and boils at - 87'9 under 767"3 mm. pressure. For experimental purposes the liquid may be transferred to open vessels by placing the cylinder with the valve downwards, and carefully opening the latter, the liquid rushing out in a fine stream through the nozzle. For, although the liquid boils at a temperature more than 80 degrees below the freezing-point of water, it may be kept for more than half an hour in tubes or other open vessels. .Like other condensed gases, liquid nitrous oxide has a very high coefficient of expansion; one volume of the liquid at becoming 1'1202 volumes at 20, whereas one volume of the gas at becomes only 1*0732 volumes when raised to 20. A drop of the liquid brought on to the skin produces a blister, and when water is thrown into the liquid it at once freezes to ice, at the same time producing a dangerously explosive evolution of gas. Phosphorus, potassium, and charcoal do not undergo any change when thrown into liquid nitrous oxide, but if a piece of burning charcoal be thrown on the liquid, it swims on the surface and continues to burn with great brilliancy. On pouring a little mercury into a tube containing the liquid nitrous oxide, the metal solidifies, whilst at the same moment a piece of ignited charcoal may be seen to be brilliantly burning on the surface of the liquid. 1 Andreef, Annalcn. 110. H- 502 THE NON-METALLIC ELEMENTS When liquid nitrous oxide is poured into carbon bisulphide the two liquids mix, and if the mixture be brought under the receiver of an air-pump the temperature sinks to 140. If a tube filled with liquid nitrous oxide be dipped into a bath of solid carbon dioxide and ether, and if this mixture be allowed to evaporate in vacuo, the liquid nitrous oxide freezes to colourless crystals, whose tension is less than one atmosphere. Poured into an open vessel, liquid nitrous oxide cools down by evaporation to a temperature of 100, and if the alcoholic thermometer be taken out of the liquid, a portion of the ad- hering substance solidifies and the temperature sinks to 115. Solid nitrous oxide in the form of snow has also been prepared by Wills, 1 who obtained it by a modification of Thilorier's method of obtaining solid carbon dioxide. Gaseous nitrous oxide, like oxygen, supports combustion vigorously. A red-hot splinter of wood rekindles when brought into the gas, a watch spring burns with bright scintillations, and a bright flame of sulphur continues to burn with a brighter flame. If, however, the sulphur be only just kindled, the flame is ex- tinguished on bringing it into the gas, as then the temperature of the flame is not sufficiently high to decompose the gas into its constituents. All combustions in this gas are simply combustions in oxygen, the burning body not uniting with the nitrous oxide, but with its oxygen, the nitrogen being liberated. Potassium and sodium also burn brightly in the gas when slightly heated, with formation of the peroxides of these metals, and these again when more strongly ignited in the gas yield the nitrates of the metals. The very remarkable effects on the organism produced by the inhalation of nitrous oxide, first observed by Davy, have been further investigated by Hermann. 2 The first effects noticed are singing in the ears, then insensibility, and, if the inhalation be continued, death through suffocation. In the case of small animals, such as birds, fatal effects are observed in 30 seconds, and in rabbits after the expiration of a few minutes. If, however, air be again allowed to enter the lungs as soon as insensibility has set in, the effects quickly pass away and no serious results follow. When a mixture of four volumes of this gas and one volume of oxygen is breathed for from one-and-a-half to two minutes, a curious kind of nervous ex- citement or transient intoxication is produced, without loss of 1 Journ. Ohem. Soc. 1874, p. 21. Jahresberieht, 1865, 662. NITROUS OXIDE 503 consciousness, and this soon passes off without leaving any evil consequences. Hence this substance received the name of laughing-gas. Nitrous oxide is now largely employed as an anaesthetic agent instead of chloroform in cases of slight surgical operations, especially in dentistry, where only a short period of unconsciousness is needed. Care must, however, be taken that for these purposes the gas is free from chlorine and nitric oxide The composition of nitrous oxide may be ascertained in various ways ; thus, a given volume of the gas is brought into a bent glass tube over mercury in the upper part of which (Fig. 150) a small piece of sodium is placed. The lower and open end of the tube is then closed under the mercury by the finger, and the part of the tube containing the sodium heated with a lamp. After the combustion, the tube is allowed to cool, and the volume of the residual gas measured. This is found to be the same as the original volume taken, and to consist entirely FIG. 150. of nitrogen. Now, as 22 '3 litres of nitrous oxide are found by experiment to weigh 43*76, and 22'3 litres of nitrogen are known to weigh 27'88, it is clear that the difference, or 15'88, is due to the oxygen. Hence we see that the molecule of nitrous oxide consists of two atoms of nitrogen combined with one of oxygen. The same result is attained when a spiral of steel wire is placed in a given volume of the gag and heated to redness by a galvanic current. The iron then burns in the gas, and a volume of nitrogen remains equal to that of the nitrous oxide employed. By means of eudiometric analysis we may likewise determine the composition of the gas, and for this purpose the nitrous oxide must be mixed with hydrogen and the mixture exploded by an electric spark, when water is formed and nitrogen gas is left behind ; thus : N O -f H, - N -f- H,0. 504 THE NON-METALLIC ELEMENTS HYPONITROUS ACID, H 2 N 2 O 2 . 280 The salts of this acid were first obtained by Divers by the reduction of an aqueous solution of potassium nitrate with sodium amalgam ; the first product of the reaction is potassium nitrite, which on further reduction yields, in addition to some hydroxylamine, potassium hyponitrite, which was found to have the empirical composition KNO. 1 It is also obtained by the electrolysis of a solution of potassium nitrite, 2 and by the action of nitric oxide on alkaline ferrous or stannous hydroxides, 3 and is always isolated by the addition of silver nitrate which yields silver hyponitrite as a yellow precipitate. The best method of preparation of hyponitrites is however by the action of alkalis on the salts of hydroxylaminesulphonic acid, HO.NH.SO 3 H (p. 321). It has already been stated that dilute acids convert this substance into hydroxylamine sulphate, but with alkalis no trace of hydroxylamine is formed, the sole products being a sulphite, a hyponitrite, and nitrous oxide, the latter being a decomposition product of the hyponi- trites. 4 2HO.NH.S0 3 K + 4KOH = K 2 N 2 O 2 + 2K 2 S0 3 + 2H 2 O. The formation of hyponitrites from derivatives of hydmxylamine shows that in these salts the oxygen atom must be beTween the nitrogen atom and that of the metal N.O.K; nitrogen is, however, never known to act as a monad element, and it is, therefore, probable that the hyponitrites have the double formula N.O.K .O.K This is further confirmed by the fact that the vapour density of ethyl hyponitrite corresponds to the formula (C 9 H 5 ) 9 N 2 2 . 5 Hyponitrous acid is further obtained by adding a solution of sodium nitrite to one of hydroxylamine sulphate and rapidly heating to 60, the following reaction taking place : HO.NH 2 + ON.OH = HO.N : N.OH + H 2 O. 1 Divers, Proc. Roy. Soc. 19, 425. 2 Zorn, Ber. 12, 1509. 3 Divers and Haga, Journ. CKem. Soc. 1885, i. 561 ; Dunstan and Dymond, Journ. Chem. Soc. 1887, i. 646. 4 Divers and Haga, Journ. Chem. Soc. 1889, i. 760. 5 Zorn, Her. H, 1630. NITRIC OXIDE 505 The acid quickly splits up into nitrous oxide and water, but if silver nitrate solution be at once added a considerable quantity of silver hyponitrite is thrown down. 1 Free hyponitrous acid has not been prepared, as when it i& liberated from its salts by addition of acids it very rapidly splits up into its anhydride (nitrous oxide) and water. Even the salts split up readily in a similar manner yielding nitrous oxide and alkali. According to Berthelot, the reason that nitrous oxide does not combine with bases to form hyponitrites is that the difference between the heat of formation of hyponitrous acid in solution (-57*4 cal.) and that of nitrous oxide ( 20'6 cal.) is greater than the heat of neutralization of the acid. An aqueous solution of hyponitrous acid may be obtained by the action of the calculated amount of dilute hydrochloric acid on the silver salt; it is colourless and strongly acid, and is stable towards acids and alkalis even on boiling. On titration with potash in presence of either phenolphthalei'n or litmus it remains acid until the acid salt is formed and then becomes alkaline, but it does not expel carbon dioxide from the alkaline carbonates. In acid solution it is converted by potassium per- manganate quantitatively into nitric acid, but in alkaline solu- tion yields nitrous acid. It neither decolorises iodine nor liberates the latter from solutions of potassium iodide. 2 With lead acetate solution the salts give a white precipitate which becomes dense and yellow after a time, and with silver nitrate a yellow precipitate, which dissolves in acids, but is reprecipitated by ammonia. The acid solution does not colour ferrous sulphate, but on addition of strong sulphuric acid the black coloration characteristic of nitric oxide is observed. NITROGEN DIOXIDE, OR NITRIC OXIDE, NO = 29 - 82. 281 This gas was first observed by Van Helmont, who included it under the term gas sylvestre. It was afterwards more fully investigated by Priestley, who named it nitrous air (see " Historical Introduction "). Preparation. The gas is formed when nitric acid acts on certain metals such as copper, silver, mercury, zinc. &c., as also upon 1 Wislicenus, Ber. 26, 772 ; Thum, Monatsh. 14, 294. 2 Thum, Monatsh 14, 294, 506 THE NON-METALLIC ELEMENTS phosphorus and some other easily oxidizable substances. It is usually prepared by dissolving copper foil or copper turnings in nitric acid of specific gravity 1'2, washing the gas by passing it through water and caustic soda, and collecting it over cold water in the pneumatic trough. The reaction occurring in this case is expressed as follows : 3Cu + 8HNO 3 = 2NO + 3Cu(N0 8 ) 2 + 4H 2 O. The gas thus obtained is, however, not pure, as it invariably contains free nitrogen and nitrous oxide, the quantity of the latter gas increasing with the amount of copper nitrate which is formed. 1 In order to obtain a much purer gas the nitric oxide must be passed into a cold concentrated solution of ferrous sulphate or chloride, with which it forms a singular compound, to be described hereafter, and which dissolves in water with formation of a deep blackish-brown solution. On heating this solution, nitric oxide is given off, but this gas still contains one-500th of its volume which is not absorbable by ferrous salts. 2 The almost pure gas can also be obtained by heating ferrous sulphate with nitric acid, or by heating a mixture of ferrous sulphate and sodium nitrate with dilute sulphuric acid ; or by allowing a concentrated solution of sodium nitrite to drop into a hydrochloric acid solution of ferrous sulphate or chloride. 3 The pure gas may be prepared by acting on mercury with a mixture of sulphuric and nitric acids, 4 a reaction which has been made use of by Lunge in his nitrometer for the estimation of the oxides of nitrogen in sulphuric acid. Properties. Nitric oxide is a colourless gas having a specific gravity of 1*039 (Berard). It was first liquefied by Cailletet in November, 1877, by exposing the gas to a pressure of 104 atmospheres at a temperature of 11. It forms a colourless liquid which boils under 7T2 atmospheres pressure at 93'5 (critical temperature), and at 153'6 at atmospheric pressure. 5 When mixed with an excess of oxygen it yields almost entirely nitrogen peroxide, but, according to Lunge, when nitric oxide is in excess, nitrogen trioxide is also formed ; with an excess of oxygen and moisture it is chiefly converted into nitric acid, whilst with even an excess of oxygen in presence of concentrated 1 Ackworth, Journ. Ohem. Soc. 1875, 828. 2 Leduc, Compt. Rend., 116, 323. 3 Thiele, Annalen, 253, 246. 4 Emich, Monatsh. 13, 73. 5 Olszewski, Compt. Rend. 100, 940. NITRIC OXIDE 507 sulphuric acid it yields solely nitrosylsulphuric acid (p. 516), and no nitrogen peroxide or nitric acid. 1 On exposure to air, this gas at once combines with the atmospheric oxygen, with evolution of heat and formation of red fumes of higher oxides. No combination occurs when the perfectly dry gas is mixed with dry oxygen. 2 Nitric oxide is extremely stable when heated, not being com- pletely decomposed until the temperature of melted platinum is reached. 3 Its formation from its elements is attended by absorption of heat (p. 232), and when exposed to the shock from an ex- plosion of mercuric fulminate, 4 it is resolved into its elements. This property is also shown by other substances which are formed with absorption of heat such as carbon bisulphide and acetylene. If a spiral of iron wire be heated to redness in this gas by means of a galvanic current, the iron burns brilliantly so long as any nitric oxide remains undecom posed, and after the combustion the residual gas is found to consist of nitrogen exactly equal in volume to one-half of the gas employed. When a stream of the gas is passed over heated potassium this metal takes fire and burns brilliantly ; whereas metallic sodium, even wiien heated with a spirit-lamp, remains unaltered in the gas. Phosphorus also burns with a dazzling brilliancy in nitric oxide, but only when it is brought into the gas already brightly burning. The flame of feebly burning phosphorus, as well as that of sulphur and of a candle, are on the other hand extinguished on plunging them into nitric oxide, because the temperature of these flames is not sufficiently high to decompose this gas into its elementary constituents. If a few drops of carbon bisul- phide be poured into a long glass cylinder filled with nitric oxide vapour, and the cylinder well shaken so that the* vapour of the bisulphide is well mixed with the gas, the mixture burns with a splendid blue and intensely luminous flame, which is characterised by its richness in the violet or chemically active rays. So intense is this light for the violet and ultra-violet rays, that a lamp in which the two gases are burnt has been constructed for the use of photographers. 5 The name nitrogen dioxide has been given to this gas because for the same quantity of nitrogen it contains twice as much 1 Ber. 18, 1384. 2 Baker, Proc. Chem. Soc. 1893, 129. 3 Enrich, Monatsh. 13, 78. Berthelot, Compt. Rend. 93, 613. 5 Sell, Ber. 7, 1522. 508 THE NON-METALLIC ELEMENTS oxygen as nitrous oxide or nitrogen monoxide. Its density, however, which remains constant down to - 70 , 1 is about 14, and the gas has therefore the formula NO, and consequently possesses a simpler constitution than nitrous oxide. The chemical and physical properties of nitric oxide bear out this view. Thus, it does not condense under circumstances which effect the lique- faction of nitrous oxide ; it is also much more stable than this latter gas, so that it follows a law which we find to hold good with regard to analogous gaseous bodies, viz. that those possess- ing the simpler constitution are much less easily condensible, and much less easily decomposable, than those of more compli- cated constitution. NITROGEN TRIOXIDE OR NITROUS ANHYDRIDE, N 2 O 3 . = 75*52, 282 When starch, sugar, arsenious anhydride and other easily oxidisable bodies are heated with nitric acid, red fumes are given off, which consist of a varying mixture of nitric oxide, nitrogen trioxide and nitrogen peroxide, the relative propor- tions varying with the strength of the acid employed. The largest proportion of trioxide is obtained by acting on arsenious anhydride with nitric acid of specific gravity 1*35, or on starch with an acid of specific gravity I'.S.S. 1 A red gas is thus formed, which on passing through a freezing mixture condenses to a deep blue mobile liquid, and does not solidify at - 90. During the past few years doubts have arisen as to whether nitrogen trioxide exists in the gaseous condition. Witt, 2 Ramsay, and others 3 maintain that the red fumes having the composition N 2 O 3 consist in reality of a mixture of equal volumes of NO and N % 2 (or N 2 O 4 , see p. 513). Ramsay and Cundall have shown that when nitrogen peroxide and nitric oxide are mixed over mercury no contraction of volume takes place, as would be the case if any nitrogen trioxide were formed, and they conclude from the density of the gas itself that it does not contain any N ? O 3 . Lunge on the other hand has urged that whereas nitric oxide when mixed with an excess of oxygen yields nitrogen peroxide almost entirely, the red fumes having the composition N 2 O 3 even when mixed with a large excess of 1 Dacconio and V. Meyer, Annalen, 240, 326. 2 Lunge, Ber. H, 1641. 3 Witt, Ber. H, 756 ; 12, 2188 ; Ramsay and Cundall, Journ. Chem. Soc. 1885, i. 187, 672 ; Geuther, Annalen, 245, 96. NITROUS ACID 509 oxygen at temperatures varying from 4 to 150 are only con- verted to a comparatively small extent into the peroxide ; hence. he concludes that the dissociation of the trioxide into dioxide and peroxide is only a partial one. He further makes the objection to Ramsay and Cundall's experiment that nitrogen peroxide is not unacted upon by mercury, and that no valid conclusions can be drawn from experiments made with gases confined by that liquid. 1 Ramsay concludes that even liquid N 2 O 3 dissociates partially at as low a temperature as 90, and can only exist dissolved in O NlTROXYPYROSULPHURIC ACID, S0 2 JON0 2 . When sulphur trioxide and nitric acid are mixed together in the cold, a thick oily liquid is formed, from which the above 1 Ber. 7, 1075. NITBILOSULPHONIC ACID 519 compound crystallizes out under certain conditions of concen- tration. It is soluble without decomposition in warm dilute nitric acid, and on cooling the liquid, crystals of this substance separate out, containing one molecule of water of crystallization. SULPHONIC ACIDS OF AMMONIA AND HYDROXYLAMINE. 293 In the year 1845 a number of crystalline salts derived from acids containing both nitrogen and sulphur were obtained by Fremy 1 by the interaction of nitrites and sulphites. He only determined their empirical formula?, but they were further in- vestigated by Glaus, 2 who showed that they are sulphonic acids, that is, compounds containing the group SO 2 OH. He, however, wrongly regarded a number of them as sulphonic acids of the hypothetical substance NH 5 , whereas the later investigations of Berglund 3 and especially of Raschig 4 have shown that all these compounds are sulphonic acid derivatives of ammonia, hydroxyl- amine and the hypothetical dihydroxylamine, NH(OH) 2 . Potassium nitrilosulphonate , N(SO 3 K) 3 + 2H 2 O. This com- pound, which was termed by Claus potassium trisulphammonate, is obtained by adding an excess of a concentrated solution of potassium sulphite to a similar solution of potassium nitrite, and separates out as a crystalline precipitate which may be recrystallized from alkaline solution. The reaction which takes place may be represented as follows : KN0 2 + 3K 2 SO 3 + 2H 2 O = N(S0 8 K) S 4 4KOH. The reaction is remarkable inasmuch as one of the most powerful alkalis is produced by the interaction of two neutral salts. Potassium nitrilosulphonate crystallizes with 2 mols. H 2 0, and forms thin silky rhombic needles, which are decomposed in the following manner by boiling water : N(S0 3 K) 3 + 2H 2 O = NH 2 SO 3 H + K 2 S0 4 + KHSO 4 . Hence only two-thirds of the sulphur is precipitated as barium sulphate on adding barium chloride to the hot solution, amido- sulphonic acid, which will be subsequently described, not being affected by barium chloride. 1 Annalen, 56, 315. 2 Annalen, 152, 336, 351 ; 158, 52, 194. 3 Lunds Universitas Arskrift, 12 and 13. 4 Annalen, 241, 161 . 520 THE NON-METALLIC ELEMENTS Potassium imidodisulphonate, NH(SO 3 K) 2 . When the nitrilo- sulphonate is boiled for only a short time with water, the reaction does not proceed so far as shown in the above equation, the chief product being potassium imidodisulphonate : N(S0 3 K) 3 + H 2 O = NH(SO 3 K) 2 + KHS0 4 . This is more readily obtained by moistening the nitrilo- sulphonate with dilute sulphuric acid, allowing to stand for a day, and recrystallizing from dilute ammonia. It forms mono- symmetric crystals, and is decomposed by boiling water into amidosulphonic acid and potassium sulphate. A large number of other imidodisulphonates have been pre- pared by Divers and Haga ; l the most interesting of these are ammonium imidodisulphonate, NH(S0 3 NH 4 ) 2 , and basic am- monium imidodisulphonate, NH 4 .N(SO 3 NH 4 ) 2 , which were first prepared by Rose in 1834 by the action of ammonia on sulphur trioxide and termed parasulphatammon and sulphatam- mon respectively. Free imidodisulphonic acid has only been obtained in aqueous solution. Amidosulphonic actW,NH 2 .S0 3 H. This acid is, as already men- tioned, the final product of the action of boiling water on the salts just described. It is a very stable compound sparingly soluble in water, and forms colourless rhombic prisms. Its potassium salt, NH 2 .SO 3 K, crystallizes in very similar forms. Hydroxylaminedisulphonic acid, HO.N.(S0 3 H) 2 . This acid is not known in the free state, but its potassium salt is formed by the action of potassium hydrogen sulphite on potassium nitrite in aqueous solution at 0. The reaction proceeds more readily with the sodium salts, which are mixed in the proportions required by the equation : NaN0 2 + 2NaHS0 3 = HO.N(S0 3 Na) 2 + NaOH. The sodium salt is however readily soluble in water, and it is therefore best converted into the potassium salt by adding an equivalent quantity of a concentrated potassium chloride so- lution. Potassium hydroxylaminedisulphonate, HO.N(SO 3 K) 2 , 2H 2 O, separates out on standing in compact crystals, which may be readily separated from the small quantity of slender needles of potassium nitrilosulphonate simultaneously produced. It 1 Journ Ghem. Soc. 1892, i. 943. DIHYDROXYLAMINESULPHONIC ACID 521 forms monosymmetric crystals, which are very slightly soluble in water ; when boiled with the latter or with acids it is first converted into hydroxylaminesulphonic acid and then into hydroxylamine, which combines with the sulphuric acid simul- taneously produced to form the sulphate. This method is now employed for the preparation of hydroxylamine (p. 475). Hydroxylaminesulphonic acid, HO.NH.SO 3 H, like amido- sulphonic acid, is stable in the free state and forms a syrupy liquid. The potassium salt, HO.NH.S0 3 K, is best obtained by the partial hydrolysis of the disulphonate with hot water, and may be purified by quick recrystallization from that liquid. When treated with alkalis it yields salts of hyponitrous acid (p. 504). NitrosoliydroxylaminesidpTionic acid, or Dinitrososulplionic acid, H 2 SN 2 O 5 . The alkali salts of this acid, which are colourless crystalline substances, are formed when nitric oxide is passed into an alkaline solution of a sulphite. The free acid is unknown, as all acids, even carbonic acid, decompose the salts into sulphates and nitrous oxide. When reduced with sodium amalgam the salts are converted into a hyponitrite and sulphite, from which it appears that they contain the group N 2 O 2 com- bined with both potassium and the sulphonic acid group. These reactions are satisfactorily explained by the constitutional formula suggested by Raschig, viz. ON.N(OK)SO 3 K, according to which the potassium salt is a basic salt of hydroxylamine- sulphonic acid, in which one of the hydrogen atoms has been replaced by the group NO. Dihydroxylamincsulphonic acid, (HO) 2 N.S0 3 H,is unknown,but its basic potassium salt is sometimes formed in the preparation of hydroxylaminedisulphonates from potassium nitrite and hydrogen sulphite. It is a crystalline unstable substance which is decomposed by acids with evolution of nitrous oxide. It is probably identical with Freiny's potassium sulphazinite. Fremy also described a salt under the name potassium sulphazinate, which according to Raschig is identical with another derivative of dihydroxylamine, obtained by gradually adding a solution of potassium hydrogen sulphite with con- stant shaking to a solution of potassium nitrite, till the solution forms a magma of crystals. It has the composition KHN 2 O 3 (SO 3 K) 2 , decomposes on heating with evolution of red fumes, gives off nitrous oxide on addition of acids, and decom- poses in aqueous solution into hydroxylaminedisulphonic acid and THE NON-METALLIC ELEMENTS potassium nitrite. It is probably formed from two potassium salts of dihydroxylaminesulphonic acid with elimination of water in the following manner : S0 3 K.N + gQ>N.S0 3 K = S0 3 K.N.O.N.S0 3 K + H 2 OK OH. In addition to the foregoing Raschig 1 has investigated a series of salts known as the sidphazotinates, which appear to be derived from an acid having the constitution (S0 3 H) 2 :N/\N:(S0 3 H) 2 . NATURE OF THE REACTION BETWEEN NITRITES AND SULPHITES. 294 Nitrous acid is usually supposed to have the constitutional formula O : N.OH, but it is not unlikely that in aqueous solution it may further combine with another molecule of water, having then the constitution N(OH) 3 . The reactions which occur with sulphites may then be readily explained as consisting in the substitution step by step of the group S0 3 H for the hydroxyl group with elimination of water and the successive formation of dihydroxylaminesulphonic acid, hydroxylaminedisulphonic acid, and nitrilotrisulphonic acids, or rather their potassium salts, as shown in the following equations : N(OH) 3 + H.S0 3 K = (HO),N.S0 8 K + H 2 O. (HO) 2 N.S0 3 K + HS0 3 K = HO.N(SO,K), + H 2 O. HO.N(S0 3 K) 2 + HS0 3 K = N(S0 3 K) 3 + H 2 O. AMIDE AND IMIDE OF SULPHURIC ACID. 295 When ammonia is allowed to act on a solution of sul- phuryl chloride in chloroform, it yields a mixture of the amide and imide of sulphuric acid, S0 2 (NH 2 ) 2 and SO 2 : NH, together with another substance which is probably imidosulphurylamide NH(S0 2 NH 2 ) 2 . S0 9 C1 2 + 4NH 3 = 2NH 4 C1 + SO 2 (NH 2 ) 2 S0 2 C1 2 + 3NH 3 = 2NH 4 C1 + SO 2 NH. 1 Annalen, 241, 232. THE ATMOSPHERE 523 The three compounds are separated from one another and from the ammonium chloride simultaneously formed by taking advantage of the different properties of their silver salts, which after separation are converted into the free amides by the action of hydrochloric acid. Sulphaniide, S0 2 (NH 2 ) 2 , forms large colourless crystals, which are extremely soluble in water, and on heating soften at 75 and melt at 81. It is converted by alkalis into salts of amidosulphonic acid (sulphaminic acid), NH 2 .S(X,OH (p. 520), and yields a silver compound, SO 2 (NHAg) 2 , which is a white amorphous powder. Sulphimide, S0 NH, is formed together with sulphamide in the manner stated above, and is also obtained when the latter is heated above its melting point. It has not been obtained pure ; its aqueous solution is fairly stable at the ordinary tempera- ture, but on boiling is converted into ammonium hydrogen sulphate. Its silver compound, SO 2 NAg, crystallizes from hot water in long needles, which require 500 600 parts of cold water for their solution. 1 THE ATMOSPHERE. 296 The term air was used by the older chemists in the general sense in which we now employ the word gas, to signify the various kinds of aeriform bodies with which chemistry has made us familiar. At the present day, however, we confine the signification of the word air to the ocean of aeriform fluid or the atmosphere (CLT/JLOS, vapour, and crfalpa, a sphere) at the bottom of which we live and move. Of the existence of an invisible gaseous envelope lying above the solid mass of the earth's crust, we become aware by the resistance offered to our bodies when we pass rapidly from place to place, as well as by the effects produced by the motion of the particles of the atmosphere which we term wind. The most convincing proof of the existence of the air is how- ever given by showing that air has weight. This can readily be done by hanging a large thick-walled glass globe, closed with a cork through which a tube passes, furnished with a stopcock, on to one end of the beam of a balance, and placing 1 Traube, Ber. 26, 607. 524 THE NON-METALLIC ELEMENTS weights in the opposite pan until the arrange- ment is in equilibrium. On exhausting the globe by placing it in communication with an air-pump, and again weighing the globe par- tially freed from air, the weight will be seen to be considerably less than that which it possessed before the evacuation. A knowledge of the composition of the at- mosphere forms the beginning of the present epoch of chemical science, experiment having shown, in opposition to the older views, that the air is not a simple body, but consists mainly of two different kinds of air or gases, oxygen and nitrogen. Although some of the ancients, especially Vitruvius, appear to have held the view that the air possesses weight, yet it is to Torricelli that we owe the first distinct proof that this is the case. In the year 1640 a Florentine pump- maker observed that his lift-pumps would not raise water to a height greater than thirty-two feet, and consulted his great townsman Galileo as to the cause of this phenomenon. Galileo does not appear to have given the correct solu- tion, as he compared the water column to an iron rod hung up by one end, which, when long enough, will at last break with it own weight. Torricelli, however, in 1643, made an experi- ment which gave the true explanation of the pump-maker's difficulty. Filling with mercury a glass tube three feet in length, and closed at one end, but open at the other, he closed the open end with his finger and inverted the tube in a basin filled with mercury. The mer- cury then sank in the tube to a certain level, whilst above this level there was an empty space, which is still called the Torricellian vacuum. Above the mercury in the basin was water, and Torricelli then raised the tube FIG. 153. so that the open end came into the water. The mercury then flowed out and the water rushed up, completely filling the tube. Fig. 153 represents THE ATMOSPHERE 525 the actual tubes employed by Torricelli, photographed from the original instruments placed in the Science Loan Exhibition at South Kensington. The rise of mercury or water in a vacuous tube is caused by the pressure of the atmosphere ; the water is, however, 13*5 times lighter than the mercury, and hence the column of the former liquid which is supported by the atmo- spheric pressure is 13 '5 times as high as that of the latter liquid. Thus was the barometer discovered, though this name was first made use of by Boyle. 1 Hearing of Torricelli's discovery Blaise Pascal resolved to put this theory to a further test. If, argued he, the suspension of the mercury in the barometric tube is due to the pressure or weight of the air, the mercurial column must sink when the barometer is taken to the top of a mountain, owing to the pressure on the mercury being lessened. Unable to try this experiment himself Pascal instructed his brother-in-law, Perier, to ascertain whether this is so or not, and on September 19, 1648, Perier took a barometer to the summit of the Puy-de-D6me, and showed that the mercury sank as he ascended, proving conclusively the correctness of Torricelli's explanation. A simple arrangement enables us to reproduce this experiment in the lecture-room ; a barometer tube filled with mercury is inverted over mercury contained in a trough, when the mercury will be seen to sink to a certain level, the space above this being vacuous. A tubulated receiver furnished with a tight-fitting caoutchouc stopper is then brought over the tube, and the air pumped out from the interior of the receiver. As the pressure of the air is by degrees removed, the level of mercury in the tube will gradually become lower, until at last it will nearly, but not quite, reach the level of the mercury in the trough. On opening the stopcock the air will rush in, and the level of the mercury in the tube will rise until it has attained its former elevation. 297 Following then the laws of gravitation, the air forms part of the earth's body, and accompanies the solid and liquid portions in their axial and orbital motions. The absolute height to which the atmosphere extends above the earth's surface has not been ascertained with accuracy. As its density is not 1 See New Experiments on Cold, published 1664-5, Boyle's Works (Edn. 1772), vol. ii. p. 487- "The barometer, if to avoid circumlocutions I may so call the whole instrument, wherein a mercurial cylinder of 29 or 30 inches is kept .suspended, after the manner of the Torricellian experiment." 526 THE NON-METALLIC ELEMENTS uniform, but diminishes as the distance from the earth's surface increases, the exact point at which the atmosphere terminates is difficult to determine. The height is certainly not uniform, inasmuch as owing to the variation of gravitation at the poles and at the equator, and owing also to the different velocities of rotation as well as to changes of temperature, a column of polar air is considerably shorter than a column of equatorial air. The atmospheric pressure at the sea's surface would naturally be constant if it were not that owing to the variations in the solar radiation, the temperature, and, therefore, the pressure of the air, undergoes frequent alterations. These irregular varia- tions necessitate our reading off the height of the barometer whenever volumes of gas have to be measured. There is no doubt that the atmosphere has a definite limit, and, from observations of the time during which the twilight extends to the zenith, it appears that the atmosphere reaches in a state of sensible density to a height of from forty to forty-five miles above the earth's surface. The relation between this height and the diameter of the earth may be illustrated by the state- ment that if a globe of one foot in diameter represents the earth, a film of air T V f an mc ^ m diameter will represent the atmosphere. If the air were an incompressible fluid, instead of being an elastic one, and if it had throughout the density which it possesses at the sea's level, the height of the atmosphere would be 10513 X 0760 = 8360 meters, or 5'204 English miles. As, however, the air is elastic, it diminishes in density as the distance from the earth's level increases ; thus at a height of 5528 metres, the air expands to twice its volume ; whilst at a height of twice 5528 metres the density of the air is only J of that which it possesses at the sea's level. At greater elevations the volume increases in the following ratios: Geographical Miles. Volume. o-ooo i 0-587 2 1-174 4 1-761 8 2-348 16 2-935 32 3-522 64 THE ATMOSPHEKE 527 According to the corrected determinations of Regnault (p. 105),. one litre of dry pure air at 0, and under the pressure of 760 mm. at the latitude of Paris weighs 1*29349 gram, whilst according to Lasch l the weight at Berlin is 1*293635 gram, or almost exactly T ^ of the weight of water. It must be remembered that since the composition of air varies within certain narrow limits, the weight of a given volume of it must also vary slightly. The weight of the air at the level of the sea in our latitude is equal to that of a column of mercury at of a height of 760 millimetres, and this is taken as the normal barometric pressure. Hence, as one cc. of mercury weighs 13*596 grams, the pressure exerted by the air on one square centimetre of surface at the sea's level will be 13*596 X 76 = 1033*3 grams (or nearly 15 Ibs. on every square inch). In common with all bodies at the earth's surface, the human frame has to support this weight, but, under ordinary circumstances the pressure is exerted in all directions, and it is not felt. If, however, the pressure in one direction be removed, as when the hand is placed over the open end of a cylinder from which the air is being pumped out, the weight of the air is at once perceived. As the air follows Boyle's law, its density being directly proportional to the pressure to which it is sub- jected, it follows that when the height above the sea level increases by equal intervals, the density of the air decreases in a geometric ratio. The difference in height of two stations in the same vertical line is therefore in the ratio of the difference between the logarithms of the barometric readings at the two stations, and, if the temperature of the two stations be the same, we need only multiply the difference of the logarithms of the two readings (reduced to for the expansion of mercury) by the number 18363 to obtain the elevation in meters. The average or mean annual temperature of the air, like its density, is not the same throughout the mass. It diminishes as the elevation above the earth's surface increases, so that at a certain height, differing for different latitudes, a line is reached at which the mean temperature of the air does not rise above the freezing point. This is called the line of perpetual snow. At latitude 75 it reaches the sea level, in the latitude of 60 it is found at a height of 3,818 feet, whilst under the equator the snow-line exists at a height of 15,207 feet above the sea. 1 Pogg. Ann. Erganzungsbd. 3, 321. 528 THE NON-METALLIC ELEMENTS The height of the snow-line is also affected by local causes and is found to vary considerably even in the same latitude. Owing to the unequal heating effect produced by the sun on different portions of the earth's surface, great variations are observed in the temperature of the atmosphere in different places, and these give rise to those motions of the atmosphere which are termed winds. Winds may either be caused by local alterations of temperature confined to narrow limits, as with the land and sea-breezes of our coasts, or they may be produced by a general unequal diffusion of heat over the sur- face of the globe, as with the so-called trade-winds, which are caused by the temperature of the air in the equatorial zones being higher than that of the air in the polar regions. 298 Air was first liquefied by Cailletet, 1 and liquid air has since been more closely investigated by Wroblewski 2 and Dewar. 3 It has not a constant boiling point, as the nitrogen boils off more rapidly than the oxygen, the ebullition proceeding somewhat roughly. Dewar has also succeeded in obtaining air in the solid state, but it is as yet doubtful whether this is solid throughout, or consists of a mixture of solid nitrogen and liquid oxygen, the latter gas not having yet been solidified when in the pure condition THE COMPOSITION OF THE ATMOSPHERE. 299 Atmospheric air, in addition to oxygen and nitrogen, contains as normal constituents, aqueous vapour, carbon dioxide, ammonia, and perhaps ozone. Other gases and vapours do indeed occur in different places, under a variety of circum- stances, and in varying quantities. Furthermore certain sub- stances, such as common salt, ammonium nitrate, and some other salts, occur as finely-divided solid particles, together with other minute floating particles of animal, vegetable, and mineral origin. The discovery of the composition of the atmosphere has been described in the Historical Introduction. We saw there that we owe to Cavendish the first exact determination of the relation existing between the two important constituents oxygen and nitrogen. 4 " During the last half of the year 1 Compt. Rend 85, 1270. 2 Monatsh. 6, 204 ; Compt. Rend. 102, 1010. 3 Proc. Royal Institution. 4 Phil. Trans. 1783, p. 106. "An Account of a new Eudiometer." COMPOSITION OF THE ATMOSPHERE 529 1781," he writes, " I tried the air of near sixty different days in order to find whether it was more phlogisticated at one time than another, but found no difference that I could be sure of, though the wind and weather on these days was very various, some of them being very fair and clear, others very wet, and others very foggy." 1 This result was founded on a long series of experiments, for seven or eight analyses of air collected on the same day were made by different processes, so that altogether Cavendish cannot have made fewer than 400 determinations of the composition of atmospheric air. Ex- periments were likewise made to see whether London air differed from that of the country, and slight differences were sometimes found in favour of London air, in Marlborough Street, sometimes in favour of country air, in Kensington. On taking a mean of the numbers no difference whatever was perceptible, the result of all his experiments being that 100 volumes of air contain 20'83 parts by volume of dephlogisticated air or oxygen. The constant results thus obtained by Cavendish led several chemists, such as Prout, Dobereiner, and Thomson, to maintain that the air is a chemical compound of one volume of oxygen with four volumes of nitrogen. Against this assumption John Dalton protested, 2 insisting that the air is merely a mechanical mixture of constant composition, and contending that, because nitrogen is lighter than oxygen, the relative amounts of the two gases must vary at different heights above the earth's surface, the oxygen diminishing and the nitrogen increasing as we ascend. This view was, however, shown to be erroneous by Gay-Lussac and Thenard, who collected air in a balloon at an elevation of 7,000 metres, and found it to contain exactly the same proportional quantity of oxygen as that collected at the same time in Paris and analysed in the same way. Their results have since been corroborated by the more exact in- vestigations of Brunner, who analysed the air collected at the top and at the bottom of the Faulhorn and found in each case exactly the same proportion between the oxygen and the nitrogen. The very important question whether the composition of the air undergoes variation, under varying conditions of time and place and what is the percentage of oxygen it contains was 1 Phil. Trans. 1783, p. 126. 2 Manchester Memoirs, 2nd series, vol i. p. 244. 35 530 THE NON-METALLIC ELEMENTS further investigated by many chemists. Gay-Lussac and Humboldt in Paris found that the air contained from 20'9 to 21 '1 per cent, of oxygen; Davy in London obtained from 20'8 to 21-1 per cent.; Thomson in Glasgow 211, and Kuppfer in Kasan 21 '1 per cent, of oxygen. It was, however, necessary that more accurate methods of analysis should be employed. 300 Two processes are now used, viz. (1) measurement of the volumes of the component gases, (2) determination of their weight. The first of these processes, or the eudiometric method, has been practised by Regnault, Bunsen, Lewy, and Angus Smith, whilst the latter method has chiefly been used by Dumas and Boussingault in their celebrated research carried out in the year 1841. 1 In their analyses of air by the latter FIG. 154. method, the two French chemists employed an apparatus shown in Fig. 154. The large balloon (v) was rendered as perfectly vacuous as possible, and brought in connection with the vacuous tube (a b), containing metallic copper reduced by means of hydrogen, and placed in a furnace in which it could be heated to redness by means of charcoal or gas. At the other end this tube was connected with the tubes c and B, and with the bulbs A ; these last contained caustic potash, and the others pumice-stone moistened with strong sulphuric acid for the purpose of taking up all moisture from the air, as well as of abstracting from it the whole of its ammonia and carbon dioxide. As soon as the tube a b had been heated to dull red- 1 Ann. Chim. Phys. [3], 3, 257. COMPOSITION OF THE ATMOSPHERE 531 ness the stopcocks r were opened so as to let the air pass slowly into the apparatus. Entering the tube a I, and coming in contact with the glowing copper, the whole of the oxygen was absorbed, only the nitrogen going over into the vacuous globe. When the experiment was complete, the stopcocks were closed, the tube and balloon detached from the apparatus and each accurately weighed. Both tube and balloon were then again rendered vacuous, and weighed a third time. The tube contain- ing the copper oxide, which had been weighed empty to begin with, is thus weighed full of nitrogen, after which the nitrogen is withdrawn by the air-pump and its weight again determined. The following details of an actual experiment may illustrate the method : Grams. Vacuous tube containing copper before the ) experiment j Tube filled with nitrogen and copper after the j 651-415 experiment ) Vacuous tube after experiment 651*346 Balloon containing nitrogen at 19 and under ( 1403-838 pressure 7627 mm ) Balloon, vacuous, at 19'4 and 7627 mm. . . 1391-554 Hence the weight of oxygen is found to be 3*680 grams, whilst the weight of nitrogen in the balloon was 12*304 grams, and the weight of nitrogen in the tube 0'069, giving a total of 12*373 grams. Or the percentage composition is : Oxygen 22'92 Nitrogen 77'08 100-00 Dumas and Boussingault used two balloons, with which they obtained the following results : Percentage by weight of oxygen. With small With large 1841. balloon. balloon. April 27 . 22-92 . . . 22'92 ,,28 23-03 . . . 23-09 29. 23-03 . 23-04 Mean . 22*993 23'016 532 THE NON-METALLIC ELEMENTS The mean of these determinations is 23*005 parts by weight of oxygen and 76' 99 5 of nitrogen. These experiments were subsequently repeated by other chemists, with the following results : Mean Percentage by weight of oxygen. Lewy, in Copenhagen, 1841 22'998 Stas, in Brussels, 1842 23100 Marignac, in Geneva, 1842 22'990 According to these experiments, therefore, the air contains about 23 parts by weight of oxygen and 77 parts by weight of nitrogen. If the composition of air by weight be calculated from its volumetric composition by the aid of the known specific gravities of nitrogen and oxygen, the numbers obtained are : Oxygen 23'2 Nitrogen 76'8 100-00 and these numbers have also been obtained experimentally by a gravimetric method by Leduc. 1 301 This method, although capable of giving exact results, requires large apparatus and good air-pumps and balances. It can only be carried on in a laboratory, and it necessitates the employ- ment of large volumes of air. The eudiometric, or volumetric method is less difficult and tedious, so that the determinations may be repeated by thousands and require a much smaller volume of air. This method is liable to so small an experimental error that, with an observer well skilled in its use, it never reaches the T^^TF^ P art > and may sometimes not exceed the woW P art of the whole. The process depends on the well-known fact that when oxygen and hydrogen gases are mixed and fired by an electric spark, they unite to form water in the proportion of one volume of the former to two volumes of the latter. The volume of the liquid water formed (p. 249) is so small in proportion to that of the constituent gases that, except in cases of very exact estimations, it may be altogether neglected. Hence if 1 Compt. Rend. 113, 129. COMPOSITION OF THE ATMOSPHERE 533 we bring together a given volume of air and more hydrogen than is needed to combine with the oxygen in it, and if an electric spark be passed through the mixture, one-third of the observed contraction will be due to the oxygen. In order to obtain by this method exact results l a very carefully calibrated eudiometer is employed 1 meter long and about O025 wide, and the observations are conducted in a space within which the changes of temperature are as small and as gradual as possible. The air for these determinations is collected either in small flasks of about half a liter in capacity, the necks of which have been previously elongated before the blowpipe, or in long tubes, the ends of which have been drawn out. Inside the flask or tube a small piece of fused chloride of calcium is placed for the purpose of absorbing the ammonia, and a similar piece of fused caustic potash to absorb the carbonic acid, and both substances are allowed to crystallize on the sides of the glass by the addition of a drop of water. It is quite necessary to remove the carbonic acid of the air previous to analysis. Even if the quantity present were only 0'05 per cent, of the total volume, it would produce an appreciable error in the oxygen determination, carbonic acid when exploded with an excess of hydrogen in presence of the detonating mixture of oxygen and hydrogen being decomposed into an equal volume of carbonic oxide, while an equal volume of hydrogen disappears, so that the volume of combined gas would be 0'05 per cent, too large. In carrying out this method with exactitude the same mani- pulatory precautions have to be attended to which have already been described in connection with the determination of the composition of water (p. 246). The following numbers' 2 show the approximation obtained by Bunsen in two analyses of the same sample of air collected in Marburg on 9th January, 1 846 : Volume. Pressure Temp. Vol. at at 0. C. and 1 m. pressure. Air employed .... 841 -8 O'olOl 0'3 428'93 After addition of hydrogen 10517 07137 0*3 74977 After the explosion . . 878'8 0'5460 0'3 480'09 1 For details on these points Bunsen's Gasometry may be consulted. 2 Bunsen's Gasometry , 71. 534 THE NON-METALLIC ELEMENTS Composition of Air in 100 parts by volume. Nitrogen 79'030 Oxygen 20'970 100-000 Air employed 859*3 After addition of hydrogen 1051 '9 After the explosion . . . 870*3 0-5225 0-7079 0-5317 0-6 448-00 0-6 743-01 0-6 461-72 Composition of Air in 100 parts ly volume. Nitrogen 79*037 Oxygen 20'963 100-000 Bunsen, who has brought all the processes of gas analysis to a marvellous degree of perfection, points out 1 that in normal determinations of the composition of the air still greater precision may be attained by repeating, several times, and at regular intervals, the observation of the height of the mercury in the eudiometer. From the agreement between the reduced volumes which are read off, the point in the series of observa- tions is found at which the temperature has been most constant. As an example of such an accurate analysis Bunsen gives the following : - Vol. Press. Temp. Vol at C. and 1 meter pressure. (6fc 0' 7 8 754-9 755-0 755-2 05045 0-5046 0-5047 15-4 15-4 15-5 360-52 ) 360-63 > 360 -62 36070 } After addition j Jl jj of Hydrogen. ) . " J 904-0 904-6 904-9 0-6520 0-6521 0-6518 15-8 16-0 16-0 557-20 ) 557-24 \ 557 '20 557-17 ) After the Ex- ( plosion. ) 5 732-3 732-5 732-7 0-4781 0-4777 0-4777 16-1 161 16-1 330-64^ 330-45 V 330 -54 330 -54 J Hence Nitrogen 79*036 volumes. Oxygen 20'964 100-000 1 Loc. cit. 77. COMPOSITION OF THE ATMOSPHERE 535 As the result of twenty-eight analyses Bunsen found the .average percentage of oxygen to be 20 '9 24 volumes, whilst the lowest percentage found was 20*840. Regnault, 1 using a different volumetric method, has analysed a very large number of samples of air collected in a uniform manner in various quarters of the globe, according to instructions which he had given. The error of two analyses made on the same sample rarely reached 0*02 per cent. In more than 100 samples of air collected in or near Paris, Reg- nault found a maximum amount of 20*999, a minimum of 20*913, and a mean of 20*96 percentage of oxygen. The differ- ence of 0*086 per cent, is too large, according to Regnault, to be due to errors of experiment, and it must, therefore, be ascribed to variations in the composition of the air occurring from day to day. Air collected from other localities gave Regnault the following results : Percentage of oxygen. 9 samples from Lyons gave 20*918 to 20*966 30 samples from Berlin 20*908 20'998 10 Madrid 20'916 20*982 23 Geneva and Chamounix . . 20*909 20*993 17 Toulon Roads and Mediter- ) 5 Atlantic Ocean 20*918 20*965 2 Ecuador 20*96 2 Summit of Pichincha . . . 20*949 20*988 2 Antarctic Seas 20*86 20'94 The conclusion which Regnault draws from these determi- nations, and which all subsequent observers have confirmed, 2 is, that the atmosphere shows perceptible, though very small, -alterations in the amount of oxygen at different times and in different localities. This variation ranges from 20*9 to 21*0 per cent., but from special unknown causes the amount of oxygen seems sometimes to sink in tropical countries as low as 20*4 per cent., as was seen in the Bay of Bengal on March 8, 1849. 302 Dr. R. Angus Smith 3 has extended our knowledge of the variations which the percentage of oxygen undergoes in the air of towns and that of closed inhabited spaces. He 1 Ann. Chem. Phys. [3], 36, 385. 2 Morley, Amer. Journ. of Science, 22, 429 ; Chem. News, 45, 283 ; Hempel, Ber. 18, 267, 1800 ; 20, 1864. 8 On Air and Rain, Longmans, 1872. 20 ' 912 - 20 ' 982 536 THE NON-METALLIC ELEMENTS finds that the percentage of oxygen in air from the sea shore, and from Scotch moors and mountains, is as high as 20*999. In the free air of towns, and especially during foggy weather, it may sink to 2082. In inhabited rooms and crowded theatres, the percentage of oxygen may sink sometimes to 20*28 (Lewy), whilst according to the very numerous (339) analyses of Angus Smith, 1 the percentage of oxygen in mines does not average more than 20*26. The exact composition of the air at high elevations has been investigated by Frankland, 2 who has shown that, as far as nitrogen and oxygen are concerned, the composi- tion of the air up to an elevation of 14,000 feet is constant, and that the variations exhibited in air collected at the above height fall within the limits noticed by former experimenters. 303 That the air is a mechanical mixture and not a chemical combination of oxygen and nitrogen is seen from the following facts : (1) The quantities of nitrogen and oxygen in the air do not present any simple relation to the atomic weights of these elements, and, indeed, the proportions in which they are mixed are variable. (2) On mixing oxygen and nitrogen gases mechanically in the proportion in which they occur in air, no contraction or evolution of heat is observed, and the mixture behaves in every way like air. (3) When air is dissolved in water, the proportion between the oxygen and nitrogen in the dissolved air is quite different from that of the undissolved air, the difference being in strict accordance with the laws of gas-absorption on the assump- tion that the air is a mixture. When water is saturated with air at any temperature below 30, the following is the proportion of oxygen and nitrogen contained in the dissolved and the original air : Air dissolved Air undissolved in water. in water. Oxygen . . . 35*1 20*96 Nitrogen . . . 64*9 79*04 100-0 100-00 If the air were a chemical combination of oxygen and nitrogen, such a separation by solution would be impossible. 1 On Air and Rain, p. 106. 2 Journ. Chcm. Soc. 13, 22. COMPOSITION OF THE ATMOSPHERE 537 (4) When liquefied air is allowed to boil, the nitrogen passes off much more rapidly than the oxygen, which could not take place if the two gases were chemically combined. In order to show the composition of the air, the apparatus Fig. 155 is often used. This consists of a calibrated and divided glass tube filled to a given point with air over mercury. Into this is introduced a small piece of phosphorus supported upon a copper wire. Gradually all the oxygen is absorbed and the mercury rises in the tube. After a while the volume of residual nitrogen is read off, and, corrections having been made for tem- FIG. 155. FIG. 156. perature and pressure, it is found that 100 volumes of the air contain about 21 volumes of oxygen. Another less exact but more rapid method of exhibiting the same fact is carried out by help of the arrangement shown in Fig. 156. In the beaker glass (c) is placed the iron stand (d) carrying the iron cup (e), containing a small piece of phosphorus. Over this stand is placed the tubulated cylinder (a). The upper part of this cylinder is graduated into five equal divisions, and water is poured into the beaker glass until the level reaches the first division. The phosphorus in the cup is ignited by dropping down on it a chain which has been heated in a 538 THE NON-METALLIC ELEMENTS flame. The phosphorus then burns, the fumes of phosphorus pentoxide are absorbed by the water, and, when the gas has cooled, and the pressure been equalized by bringing the level of the water outside up to that inside the cylinder, four-fifths of the original volume of the air remain unabsorbed. SUBSTANCES PRESENT IN SMALLER QUANTITY IN THE ATMOSPHERE. The carbonic acid, aqueous vapour, organic matter, and the other constituents of the atmosphere vary in amount in different places and at different times much more than the oxygen and nitrogen. 304 Atmospheric Carbonic Acid. Reference has already been made to the part played by atmospheric carbon dioxide (carbonic acid). The whole vegetable world depends for its existence on the presence of this gas, which serves, in the sunlight, as the chief food of plants. The normal amount of carbonic acid existing in the air was formerly supposed to be 4 vols. in 10,000, but the investigations of the last 25 years have shown that the methods formerly in use gave too high results and that the amount is nearly 3 vols. in 1 0,000.! This number is the average of a large number of analyses of country air, but it appears that the amount is slightly greater in the night than in the day, probably owing to the in- fluence of vegetation ; over the sea the amount is also 3 vols. in 10,000, but no difference is observed between the day and night values. 2 In dull cloudy weather the amount is somewhat larger than in bright fine weather, and variations are also observed according to the direction of the wind, which when it has come over a large extent of land causes an increase in the quantity of carbonic acid. In large towns where much coal is used the amount of car- bonic acid may rise as high at 6'0 and even 7'0 vols. per 10,000, and at high elevations the proportion of carbonic acid is generally, but not invariably greater than at lower levels. In closed inhabited spaces the volume of carbonic acid proceeding from respiration and from the combustion of illuminating materials is usually much higher than in the open air, 3 and the proportion is much higher in mines than in the air above ground. 1 For a full discussion of the results of the various investigators see Blochmann, Annalen, 237, 39. 2 Thorpe, Jnurn. Chem. Soc. 1867, 189. 3 Roscoe, Journ. Chem. Soc. 1867, 189. ATMOSPHERIC CARBONIC ACID 539 Air containing more than 7'0 vols. of carbonic acid per 10,000 is as a rule harmful for respiration; this is, however, probably not due to the carbonic acid itself, but to the other organic impurities which are formed along with it especially during respiration, for the deleterious properties of such air remain after the removal of the former gas. As the quantity of carbonic acid serves as the readiest and most reliable test of the healthiness or otherwise of an atmo- sphere, it becomes a matter of importance to ascertain its amount with accuracy. The methods in use for this purpose are (1) the gravimetric and (2) the volumetric method. In the first FIG. 157. of these l the carbonic acid is absorbed from a known volume of air, freed from ammonia and aqueous vapour, drawn through weighed tubes containing caustic potash. This necessitates the passage of not less than forty litres of air drawn by means of the aspirator (v, Fig. 157), filled with water, through the tubes, in order that a sufficient weight of carbonic acid for an exact weighing may be obtained. The tubes A and B are filled with pumice-stone moistened with sulphuric acid. The air is thus dried and freed from ammonia, c and D contain moist but solid caustic potash for the absorption of the carbon-dioxide. E and F are also drying- tubes, prepared like the tubes A and B, and 1 Saussure, Pogg. Ann. 19, 391. 540 THE NON-METALLIC ELEMENTS serving to hold back any moisture which the dry gas might have taken up from the tubes c and D. The following example of a determination made by this plan illustrates the process : Gravimetric Determination of Carbonic Acid in London Air, Feb. 27, 1857. Volume of aspirated air at 8 and under ) __4o.o r* 772'5 mm. of mercury ) Weight of absorbed carbonic acid . . . =0'0308 grm. Hence 10,000 volumes of air contain 3'7 volumes of carbonic acid. 305 A much more convenient process is the volumetric one proposed by Pettenkofer. 1 For this a volume of about 10 litres of air only is needed, a glass cylinder closed by a caoutchouc cap being employed, and no balance being required. The method depends upon the fact that a solution of hydrate of baryta of known strength when shaken up with a closed volume of air containing carbonic acid, abstracts the whole of that carbonic acid from the air with formation of insoluble carbonate of barium ; thus : Ba(OH) 2 + C0 2 = BaC0 3 + H 2 0. The quantity of baryta in excess which remains in solution, after shaking up with the air, is ascertained by adding to an aliquot portion of the milky fluid a standard solution of oxalic acid or sulphuric acid until the alkaline character of the baryta water disappears. When this point is reached, the whole of the residual soluble baryta has been neutralised by the oxalic acid. The baryta and oxalic acid solutions are made of such a strength that equal volumes exactly neutralise each other, and so that 1 cc. of baryta solution will precipitate exactly 1 mgrm. of car- bonic acid. The following example will serve to explain this process : Volumetric Determination of Carbonic Acid in Manchester Air, Nov. 10, 1873. Volume of air employed, 10*80 litres at 7 and under a pressure of 765 mm. of mercury. 50 cc. of standard baryta solution (1 cc. = 1 mgm. C0 2 ) was 1 Journ. Chem. Soc. 1858, 292. ATMOSPHERIC MOISTURE 541 shaken up with this air. Of this, after the experiment, 25 cc. needed 22 cc. of standard oxalic acid (1 cc. = 1 mgm. CO 2 ) for complete neutralization. Hence 6 cc. of baryta solution were neutralized by the carbonic acid in 10*8 litres of air; or 10,000 volumes of air contain 2*85 volumes of carbonic acid. To obtain very accurate results by Pettenkofer's method, which is in any case more reliable than the gravimetric method, a number of precautions must be taken. For these reference may be made to Blochmann's paper. 1 306 The moisture contained in the air is liable to much more extensive changes than even the carbonic acid. Amongst the circumstances which affect the atmospheric moisture, distance from masses of water, and the configuration of the land, seem the most important. A given volume of air cannot take up more than a certain quantity of aqueous vapour at a given temperature, and then the air is said to be saturated with moisture. The weight in grams of water capable of being taken up by 1 cubic meter of air, at different temperatures, is given in the following table : C. Grams. 10 2-284 4-871 5 6795 10 9-362 C. Grams. 15 12-746 20 17-157 25 22-843 30 30-095 C. Grams. 35 39-252 40 50-700 100 588-73 This quantity is wholly dependent on the temperature of the air, being represented by the tension of the vapour of water at that temperature. Thus the weight of aqueous vapour which can be taken up by 1 cubic metre of air at 10, at which temperature the tension of its vapour is 9163 mm., is obtained as follows : 1 cubic metre of aqueous vapour at and 760 mm. weighs 304*75 grams. Hence one cubic metre of air needs for satura- tion the following quantity of aqueous vapour : 804-75 x 273 x 9-163 283 X 760 9*82 grm. When the temperature of saturated air is lowered, the aqueous vapour is precipitated in the form of rain, snow, or hail, accord- ing to the temperature of the air during, or after, the deposition. If one cubic mile of air saturated with water at 35 be cooled to 1 Loc. cit. 542 THE NON-METALLIC ELEMENTS 0, it will deposit upwards of 140,000 tons of water as rain, for one cubic metre of air at 35 is saturated when it contains 39'25 grams of aqueous vapour, whereas as it can hold only 4*87 grams in the state of vapour. It seldom happens that the air is completely saturated with moisture, and as seldom that the amount of moisture sinks below y 1 ^ of the saturating quantity. Even over the sea the air is never completely saturated with moisture. Large amounts of watery vapour are, however, driven by the winds into the interior of the continents, and more in summer than in winter, when the land is colder than the sea, and when, therefore, the aqueous vapour is more easily con- densed, and does not so readily penetrate for great distances. The following mean tensions of aqueous vapour in the air at different places exhibit this fact clearly : Mean Tension of Aqueous Vapour in mm. Mean yearly. January. July. London .... 8'6 ... 5'5 ... 12'2 Utrecht .... 7'6 ... 4'8 ... 10'2 Halle 7'5 ... 4-5 ... 11-6 Berlin 7'3 . . . 4'3 . . . 1M Warsaw .... 6'9 ... 3'4 ... 117 St. Petersburg . . 5'7 . . . 27 ... 10'5 Kasan 5'0 . . . To . . . 9'8 Barnaul .... 4'8 ... 1'4 ... 11'3 Urtschusk . . . 4'0 . . . 0'4 . . . 11 -3 In order to determine the amount of moisture in the air, we may employ either a chemical or a physical method. According to the first, a known volume of air is drawn through weighed tubes containing hygroscopic substances, the increase in weight of these tubes giving the weight of the aqueous vapour. Thus 43'2 litres of London- air at 8 and 772'5 mm., when passed through drying tubes, deposited 0*241 grm. of water. In the second method Hygrometers are employed ; of these Regnault's dew-point hygrometer is the best. 1 For the physical determinations of atmospheric moisture, works on Hygrometry must be consulted. 307 Ammonia is another important constituent of the air, originating in the decomposition of nitrogenous organic matter. The relative proportion in which this substance is contained in 1 Ann. Chem. Phys. [3] 15, 129. ATMOSPHERIC AMMONIA 543 the atmosphere is extremely small, and probably very varying, inasmuch as it is not present as free ammonia, but combined with atmospheric carbon dioxide and other acids, and these ammoniacal salts are washed down by the rain or absorbed by the earth. This constituent plays an important part in vege- tation, for it is from it that unmanured crops derive the chief portion of the nitrogen which they require for the formation of seed and other portions of their structure. According to the experiments of Goppelsroder, 1 the rain water collected at Basel contained the following amounts of ammonia in parts per million : 1871. Minimum. * Maximum. January . ... 4'6 .... 7'8 February ... 3'2 .... 6'5 March .... 3'8 .... 18'2 April 3'2 . e . . 6'8 May 3'2 .... 14*8 June 3-2 .... 9-1 Small quantities of nitrous and nitric acids are also found in the atmosphere, in combination with ammonia ; their formation is probably due partly at any rate to the passage of electric discharges through the air, as the quantity of these compounds found in rain water falling during thunderstorms is greater than that occurring in ordinary rain. Ozone also is supposed to occur frequently in small quantity, but as already stated, this cannot yet be regarded as definitely proved (p. 240). The following observations have been made by Bechi 2 on the amount of ammonia and nitric acid contained in the rain water falling in Florence and at Vallambrosa in the Apennines 957 metres above the sea-level for one square hectometer of surface. Florence. Vallambrosa. 1870. 1871. 1872. Rain in cubic metres . 9284 10789 12909 . 20278 Ammonia in grams . . 13236 10572 12917 . 10433 Nitric acid in grams . .15728 9153 13057 . 11726 Tuxen has also published tables showing the amount of nitric acid and ammonia in the rain falling in Denmark during the years 1880 to 1885. 3 1 J. Pr. Chem. [2] 4, 139 and 383. 2 Ser. 8, 1203. 3 Journ. Chem. Soc. 1892, ii. 234. 544 THE NON-METALLIC ELEMENTS 308 Atmospheric Organic Matter. The atmosphere also, of course, contains gases arising from the putrefactive decomposition of organic substances. These gases do not remain in the air any length of time, but undergo pretty rapid oxidation. The particles of dust which we see dancing in the air as motes in the sunbeam are partly organic and partly inorganic. Bechi found that a thousand litres of rain water which fell in November 1870 in a garden in Florence contained 4' 123 grams of total solid residue, of which one-half consisted of organic bodies and ammoniacal salts and one quarter of gypsum and common salt. Amongst the organic substances the germs of plants and animals always occur, as has been proved by the classical labours of Pasteur. 1 These bodies are the propagators of fermentation and putrefaction, and air which has been freed from these particles either by nitration through asbestos or cotton-wool, or by igni- tion (Pasteur), or by subsidence (Tyndall), may be left in contact for any length of time with liquids such as urine, milk, or the juice of meat, without these organic liquids undergoing the slightest change. Air which has thus been filtered is termed by Tyndall optically pure. When a ray of light is allowed to pass through air thus freed from solid particles no reflection is noticed, and the space appears perfectly empty, the motes which in ordinary air reflect the light being absent. (See also p. 546.) The organic nitrogen contained in the air, probably chiefly contained in such germinal bodies, has been quantitatively determined by Angus Smith 2 in the form of ammonia. He obtained the following results : 1 kilogram of air contains of organic nitrogen calculated as ammonia Grams. Innellan (Frith of Clyde) . . . Oil London 0'12 Glasgow 0-24 Manchester 0*20 Near a midden 0*31 The volatile organic products arising from putrefaction, which are always present in the air, appear to exist in larger quantities in marshy districts than elsewhere, and in all probability they are the cause of the unhealthiness of such situations. The unpleasant odour invariably noticed on entering from the fresh air into a 1 Ann Chim. Phys. [3] 64, 5. 2 Air and Earn, 438, VENTILATION 545 closed inhabited space is also due to the presence of the same organic putrescent bodies, whilst the oppressive feelings which frequently accompany a continued habitation of such spaces do not proceed from a diminished supply of oxygen, or an increase in the atmospheric carbonic acid, but are to be ascribed to the influence of these organic emanations. 309 Hence the subject of ventilation is one of the greatest consequence to well-being as well as to comfort, and it is neces- sary to provide for a continual renewal of the deteriorated air. Fortunately this renewal takes place to a considerable extent in a room, even when doors and windows are shut, by what may be called the natural means of ventilation, by the chimney, by cracks and crevices in doors and windows, and especially through the walls. Almost instinctively man appears to have chosen FIG. 158. porous building materials, thus permitting by gaseous diffusion an exchange of fresh for deteriorated air. The well-known unhealthiness of new and damp houses, as well as of those built of iron, is to a great extent to be attributed to the fact that the walls do not permit a free diffusion to go on. The fact that gases readily pass through an ordinary dry brick- or sandstone-wall, is clearly shown by the following experiment proposed by Pettenkofer. (A) Fig. 158 is a piece of wall built of ordinary brick or sandstone 82 centimetres in height, 40 cm. broad, and 13 cm. thick. On each side of the wall two rectangular plates of iron (C) are fixed, and the whole of the outside of the wall is then covered over with a coating of tar, and thus made air-tight. A tube (c c') is soldered into a hole in the centre of each iron plate. If 36 546 THE NON-METALLIC ELEMENTS a candle-flame be held in front of the opening of the tube at one side of the wall, and a puff of air be blown from the lungs through the open end of the tube at the other side of the wall, the candle will at once be blown out ; whilst if the one tube be connected by a caoutchouc tube to a gas jet, and the coal gas be allowed to pass through the tube, a flame of gas can, in a few seconds, be lighted at the open end of the opposite tube. If the bricks or stones of the experimental wall be well wetted, it will be found very difficult to blow out the candle as described. BACTERIOLOGY OF AIR. 310 The air always contains, as has been stated, a certain number of micro-organisms. These belong to the groups of moulds, yeasts, and bacteria. One litre of air contains on an average from four to five microbes. A pure unfiltered river water, on the other hand, contains from 6,000 to 20,000 in one cc., and a fresh undisturbed soil about 100,000 microbes in one cc. 1 The atmosphere is therefore relatively poor in micro- organisms. Their great storehouse in Nature is the soil, and those present in the air are almost entirely derived from this source. The majority of the air microbes consist of the spores of moulds and yeasts. The greater number of the bacteria found belong to the group of the micrococci, and many of these are characterised by the production of pigment when grown on culture media. Sarcina forms are constantly met with. The bacteria, whether bacilli or cocci, are almost entirely saprophytic organisms i.e. organisms which are not disease-producing. It is only occasionally that pathogenic or disease producing forms are found, as for example the pus cocci. The spores of moulds are so light that the isolated cells can float freely in the atmosphere. The conditions are otherwise with regard to bacteria. These are not found isolated in the air, but aggre- gated in small groups and adhering to particles of dust. Dust is the vehicle by which they are transmitted to the air, and the bacteria therefore belong to the more ponderable elements of the air-dust. Thus for instance, if the dust in a room be stirred up, large numbers of bacteria will be found in the air ; but after the dust has once again settled to the ground, the bacteria 1 Fliigge, Grundriss der Hygiene. Leipzig. 1889. BACTERIOLOGY OF THE AIR 547 disappear in great part, leaving the lighter free spores of the moulds in the air. Bacteria do not pass into the air from a moist surface or from the surface of water. Indeed the air of sewers is found to contain fewer organisms than the outside air, the damp walls and the sewage retaining the organisms which are not carried about by wind. 1 It is only when the surface containing them becomes dry that the wind is able to carry up dust and with it bacteria. The factors favouring the distribution of bacteria in the air are dryness of the soil and wind currents ; the factors hindering their presence are moistness of the soil and a still atmosphere. The number of microbes present in the air will, accordingly, vary with the atmospheric conditions. The air contains few microbes after a prolonged fall of rain, and many after prolonged heat and dry weather. The spores of moulds, however, form an exception to this rule, as they are most abundant in the air during damp weather, moisture favouring their production. The less dust there is in the air, the less the number of bacteria, and vice versa. The air over the open sea and on high mountains contains few or no bacteria ; the air of a town more than the air of the country. Drying being essential to the passage of microbes into the air, the forms most largely present will be those least affected by dryness, e.g. moulds and their spores. In the open air from 10 to 20 times as many moulds are found as bacteria. On the other hand drying is fatal to a large number of bacteria, and especially to the pathogenic organisms, e.g. Koch's comma bacillus, glanders bacillus, &c. It would therefore seem that the dangers of air infection have been over-estimated. The air of enclosed spaces and dwelling-houses is richer in microbes than the open air. It thus appears that the most dangerous factor, hygienically, is not the air itself nor the gases and sewage emanations that may be present in it, but the dust to which bacteria cling. The dust of dwellings may, therefore, become an important factor" in carrying infectious microbes, more especially by fragments of clothing, linen, &c., from sick people and their attendants. Cornet found that the dust of dwellings contains the tubercle bacillus, and that such dust can produce an infec- tion with tubercle.' 2 The dust in the air of dwellings and 1 Petri, Zeitschrift fur Hygiene, 1887. 2 Zeitschrift fur Hygiene, Bd. v. 1888. 548 THE NON-METALLIC ELEMENTS hospitals is infectious for animals, and may produce tubercle, malignant oedema, tetanus, or septic peritonitis. 311 The great majority of air microbes are however sapro- phytic, and amongst them always occur the organisms to which fermentation and putrefaction are invariably due (Pasteur, Tyndall.) A knowledge of their nature and action is important for industries in which fermentation is an essential feature, as the presence of strange forms may be injurious to the processes involved. The number of these special forms present in the air also varies according to the locality, weather, and season of the year. There are fewer in the open air than in dwellings, in cold weather than in hot weather, and in winter than in summer. Pasteur, and later Hansen, found a relatively small number of saccharomyces (yeast-plant) in the open air. The germs of the alcoholic fermentation are however already present on the surface of fruits, &c., and do not require to be introduced from the air. 1 The methods of examining the air for microbes fall into two groups : I. The methods of Miquel and their modifications. The air is filtered through glass-wool or soluble filters, and the filters are then distributed in flasks of nutrient bouillon. II. The methods in which Koch's nutrient gelatin is used directly or indirectly. 2 The air may be aspirated over nutrient gelatin in Hesse's tubes, 3 or through soluble or insoluble filters (sugar, sand, &c.). In the latter case the filters and the microbes are distributed on gelatin culture plates. One of the best me- thods is that devised by Petri, 4 in which the air is aspirated through previously sterilised sand filters. By means of the above methods a qualitative and quantita- tive examination of the air microbes can be carried out. For other points concerning the hygienic relations of the air, Dr. Renk's book should be consulted (" Die Luft," ffandbuch 1 Hansen's important investigations, and the zymotechnical methods employed by him are fully described in Jorgensen's book, Die Mikroorganismen der Gahrung Industrie (Berlin, 1893). The book also contains references to all the most important papers. The micro-organisms of the air have been specially studied by Miquel, and a full account of his results is contained in the issues of the Annuaires de Montsouris (1879, 1884, 1886, &c.). 2 Welz, " Bacteriologische Unterschung der Luft," Zeitsehrift fur Hygiene, Bd. xi. 1891. Also Uffelmaii, " Luft Untersuchungen," Archiv fur Hygiene, Bd. viii. 3 Mittheil. d. k. Gesundheitsamt. 1884. 4 Zeitsehrift filr Hygiene, Bd. iii. 1887. PHOSPHORUS 549 d. Hygiene, Leipzig, 1886), and also the monograph on "Air" in A Treatise on Hygiene and Public Health, vol. i. (London : J. & A. Churchill, 1893). PHOSPHORUS, P = 30-8. 312 A considerable amount of uncertainty surrounds the dis- covery of phosphorus, inasmuch as several chemists have claimed the first preparation of this body, while each has contradicted the other in a variety of ways. It seems, however, tolerably certain that phosphorus was first prepared (1674) by the alchemist Brand, of Hamburg, who obtained it from urine in a process which had been previously made use of for the purpose of preparing a liquid supposed to have the power of turning silver into gold. By a secret process, Brand succeeded in preparing phosphorus from this liquid, and he is said to have sold the secret of the manufacture to Krafft, from whom it appears that Kunkel learnt what he knew, and published in the year 1678 a pam- phlet on this remarkable product. 1 In these early days phosphorus was a very costly body, being valued as one of the most remarkable and interesting of chemical substances. Krafft exhibited it as one of the wonders of nature to various crowned heads, amongst others, in the year 1677, to King Charles II. of England. Robert Boyle became acquainted with its existence without, as he tells us, having been informed by Krafft of the mode of preparation except so far as that it was obtained from an animal source, and he succeeded in the year 1680 in the preparation of phosphorus, as Kunkel and Brand had done before him, by strongly heating a mixture of evaporated syrupy urine and white sand in an earthenware retort. 2 The difficulty of thus preparing phosphorus was considerable, and so many chemists failed in the attempt that the price, as late as the year 1730, was extremely high, ranging from ten to sixteen ducats the ounce. Gahn, in 1769, discovered the existence of calcium phosphate in bones, but it was not until this fact was published by Scheele in 1771 that 1 " Oeffentliche Zuschrift vom Phosphor Mirabile und dessen leuchtenden Wunderpilulen. " - Boyle, Phil. Trans., 1693 " A Paper of the Hon. Robert Boyle, deposited with the Secretary of the Royal Society on the 14th of October, 1680, and opened since his death." 550 THE NON-METALLIC ELEMENTS phosphorus was obtained from bone-ash, which has from that time invariably served for its preparation. The name phosphorus (<P OH O HO P/ | HO/ | X O OH P O OH Other chemists however believed that in the phosphoric acids phosphorus behaves as a pentad, and represented them by the following formulae : Orthophosphoric Acid. Pyrophosphoric Acid. Metaphosphoric Acid HO. >P = H0\ HO/ | O = P = O HO-^P = O HO/ HO X | OH >P = HO/ Since it has been shown that phosphorus forms a stable pen- tafluoride, the vapour density of which remains constant through a considerable range of temperature and corresponds to the molecular formula PF 5 , the pentad nature of phosphorus in many of its derivatives has been generally admitted, and the 592 THE NON-METALLIC ELEMENTS second series of formulae for the phosphoric acids is now almost universally accepted. Poisonous Action of the Acids of Phosphorus. When ad- ministered in an uncombined condition, the various oxides of phosphorus produce apparently the same symptoms which follow the administration of other mineral acids. Sufficient data do not exist as to the specific physiological action of all these compounds. In the case of ortho- meta- and pyrophosphoric acids it would appear that the first, when in combination with inactive bases, acts as a perfectly inert body. The second possesses some activity as a poison, while the pyro-salts when introduced directly into the blood are found to be very powerful poisons (Gamgee). ORTHOPHOSPHORIC ACID, H 3 PO 4 . 340 In order to prepare this acid, red phosphorus is heated in a retort with common concentrated nitric acid. The phosphorus is oxidized at the expense of the nitric acid, and red fumes are slowly evolved. When the phosphorus has dissolved, the residue is evaporated in a porcelain dish, and the concentrated solution repeatedly treated with nitric acid, in order to oxidize com- pletely any phosphorous acid which may have been formed. As soon as the further addition of nitric acid is unaccompanied by the evolution of red fumes, the operation is concluded, and the residue only requires to be evaporated again, in order to get rid of the excess of nitric acid (v. Schrotter). Common phos- phorus was formerly employed instead of red phosphorus for this purpose ; but this undergoes oxidation much more slowly than the red variety, inasmuch as it melts forming round globules, which are only slowly attacked by the nitric acid. In addition to this, when common phosphorus is employed, weak nitric acid can alone be used, as the strong acid is apt to produce an explosion when brought in contact with it. The residue obtained in the preparation of hydriodic acid by means of iodine and phosphorus consists of a mixture of phosphoric and phosphorous acids, containing a small quan- tity of hydriodic acid. In order to prepare pure orthophosphoric acid from this residue, it may be heated with a little fuming nitric acid, and filtered, in order to separate it from the solid iodine which is liberated. More nitric acid is then added, in PHOSPHORIC ACID 593 order to oxidize phosphorous acid, and the liquid is evaporated to a syrupy consistency. Orthophosphoric acid is prepared on the large scale from bone-ash, which consists chiefly of tri-calcium phosphate, together with a small quantity of magnesium phosphate and calcium carbonate. According to Liebig, equal weights of sulphuric acid and bone -ash are taken ; the sulphuric acid is diluted with ten times its weight of water, and is allowed to remain in contact with the bone ash for some time. The acid solution is then filtered through linen, and the filtrate evaporated to a small bulk. On the addition of strong sul- phuric acid, the calcium, still present in solution, is precipitated as gypsum. The clear solution is then poured off, evaporated to dryness, and freed from an excess of sulphuric acid by ignition. The residue is free from lime and sulphuric acid, but it contains small quantities of magnesia, which can only with difficulty be removed. In order to prepare a pure acid from bone-ash, the powdered ash is dissolved, in the smallest possible quantity of nitric acid, and to the clear liquid a solution of acetate of lead is added. A precipitate of lead phosphate falls down, which must be warmed for some time with the liquid, in order to free the precipitate from any calcium phosphate which is thrown down with the lead salt. It is well washed with boiling water, and decomposed by sulphuretted hydrogen. According to Berzelius's method, the lead salt is decomposed by dilute sulphuric acid, and the excess of acid removed by ignition of the evaporated filtrate. The residue is then dissolved in water and freed from traces of lead by sulphuretted hydrogen. Another method which may be employed is to dissolve the bone-ash in its own weight of hydrochloric acid of specific gravity 1*18, diluted with four times its weight of water. To this solution one-and-a-half parts of dry sodium sulphate is added, whereby a precipitate of gypsum is produced ; this is then filtered off, and the boiling solution neutralized with sodium carbonate ; the solution is again filtered, to separate any calcium carbonate which may fall down, and the whole precipitated with barium chloride. The precipitate thus formed consists of a mixture of barium sulphate and barium phosphate and this is decomposed by one part of sulphuric acid, having a specific gravity of 171 (Neustadt). A further method is to treat calcium phosphate with hydro- 39 594 THE NON-METALLIC ELEMENTS fluoric acid in a leaden dish ; the excess of acid is then evapor- ated off and the solution filtered from separated calcium fluoride and evaporated. 1 Commercial phosphoric acid frequently contains arsenic acid, derived from the sulphuric acid or hydrochloric acid employed in its manufacture. In order to free it from arsenic, it must be dissolved in water, sulphur dioxide led through the warm solution, in order to reduce the arsenic acid to arsenious acid, then the solution boiled to remove the excess of sulphur dioxide, and sulphuretted hydrogen passed through, by which means the whole of the arsenic is precipitated as the insoluble trisulphide. 341 Phosphoric acid is extremely soluble in water ; it has a pleasant purely acid taste, and is perfectly free from smell. When the aqueous acid is evaporated down until the residue possesses the composition, H 3 P0 4 , it presents the appearance of a thick syrup, from which, on standing, a crystalline mass is deposited. If a crystal of this acid be dropped into a freshly- prepared solution of the requisite strength, crystals begin to form at once, and soon spread throughout the mass. These crystals belong to the rhombic system, forming six-sided prisms terminated by six-sided pyramids, which melt at 38'6. 2 The crystallized acid may be heated to 160 without undergoing any alteration, but above this temperature it loses water, and at 213 is completely converted into pyrophosphoric acid, H 4 P 2 O 7 . This substance, in its turn, loses water when heated to redness, with formation of metaphosphoric acid, HPO 3 . The table on the following page gives the variation of the specific gravity, with the percentage composition of aqueous solutions of ortho-phosphoric acid : 3 342 The Orthophosphates. Orthophosphoric acid, being tri- basic, forms three classes of salts according as one, two, or three atoms of hydrogen are replaced by their equivalent of metal. Thus we know three orthophosphates of sodium : Trisodium or normal sodium phosphate, Na 3 P0 4 + 2H 2 0. Hydrogen disodium phosphate, HNa 9 PO 4 + 1 2H 2 O. Dihydrogen sodium phosphate, H 2 NaPO 4 + H 2 O. Of the normal salts those of the alkalis with the exception of the lithium salt are easily soluble in water and their solutions 1 Nicolas, Oompt. Rend. HI, 975. 2 J. Thomsen, Ber. 7, 997. 3 John Watts, Chem. Neics, 12, 160. THE ORTHO PHOSPHATES 595 have a strong alkaline reaction. The normal orthophosphates which are insoluble in water are easily soluble in dilute acids by which means they are converted into the soluble hydrogen Specific Gravity. Per Cent. PA. Specific Gravity. Per Cent. P 2 5 . Specific Gravity, Per Cent. PA 1-508 49-60 1-328 36-15 1-144 17-89 1-492 48-41 1-315 34-82 1-136 16-95 1-476 47-10 1-302 33-49 1-124 15-64 1-464 45-63 1-293 32-71 1-113 14-33 1-453 45-38 1-285 31-94 1-109 13-25 1-442 44-13 1-276 31-03 1-095 1218 1-434 43-95 1-268 30-13 1-081 10-44 1-426 43-28 1-257 29-16 1-073 9-53 1-418 42-61 1-247 28-24 1-066 8-62 1-401 41-60 1-236 27-30 1-056 7-39 1-392 40-86 1-226 26-36 1-047 6-17 1-384 40-12 1-211 24-79 1-031 415 1-376 39-66 1-197 23-23 1-022 3-03 1-369 39-21 1-185 22-07 1-014 1-91 1-356 38-00 1-173 20-91 1-006 0-79 1-347 37-37 1-162 19-73 1-339 36-74 1-153 18-81 orthophosphates. These latter salts are readily obtained from the normal compounds; even carbon dioxide brings about the change ; thus, if this gas be led into a solution of trisodium phosphate the following reaction takes place : Na 3 PO 4 + C0 2 + H 2 O = HNa 2 PO 4 + HNaCO 3 . The hydrogen disodium phosphate which is here formed together with hydrogen sodium carbonate is the common phosphate of soda of the shops. This salt, although according to its constitution it must be considered as an acid salt, inasmuch as it still contains hydrogen replaceable by a metal, has a slightly alkaline reaction, and, like other soluble hydrogen orthophosphates, is easily obtained by adding a solution of soda to phosphoric acid until the liquid has a weak alkaline reaction. The hydrogen disodium orthophosphate is converted on heating, with loss of water, into the pyro phosphate. The dihydrogen orthophosphates of the alkalis are soluble in 596 THE NON-METALLIC ELEMENTS water, and possess a slight acid reaction. Dihydrogen potassium phosphate, H 2 KPO 4 , forms large well defined crystals, and this salt may be heated to a temperature of 400 without losing water. At a higher temperature, however, one molecule of water is driven off and potassium metaphosphate, KPO 3 , is formed. The orthophosphates can readily be recognized by the follow- ing reactions. The normal as well as the acid salts give with nitrate of silver a yellow precipitate of silver orthophosphate, Ag 3 P0 4 ; thus: Na 3 PO 4 + 3AgNO 3 = 3NaN0 3 + Ag 8 PO 4 . HNa 2 PO 4 + 3AgNO s = 2NaNO 8 + HNO 3 + Ag 3 PO 4 . H 2 NaP0 4 + 3AgN0 8 = NaNO s + 2HNO 3 + Ag 3 PO 4 . In the case of common phosphate of soda and nitrate of silver, we have the singular fact of two neutral or slightly alkaline solutions, when mixed, yielding a strongly acid liquid. When to a solution of an ortho-salt a mixture of sal-ammoniac, ammonia, and magnesium sulphate solutions is added, a crystal- line precipitate of ammonium magnesium phosphate, (NH 4 )MgP0 4 + CH 2 5 is thrown down. In order to detect orthophosphoric acid in a substance insoluble in water, the body may be dissolved in nitric acid and an excess of a solution of molybdate of am- monia in nitric acid added to the liquid. If phosphoric acid be present, this solution on slightly warming, or on standing in the cold, yields a dense yellow precipitate. The composition of this precipitate is approximately represented by the following formula : 14Mo0 3 + (NH 4 ) 3 PO 4 + 4H 2 O. METAPHOSPHORIC ACID, HP0 3 . 343 This modification of phosphoric acid was discovered by Graham in 1833. It is obtained when a solution of phosphoric acid is heated until the residue does not give off any more water. The acid thus prepared solidifies on cooling to a soft pasty mass which on exposure to the air readily absorbs moisture and deliquesces. The glacial phosphoric acid of the METAPHOSPHORIC ACID 597 shops is metaphosphoric acid which usually contains soda as impurity. 1 Metaphosphoric acid is also formed when crystalline phosphorous acid is heated in a sealed tube with bromine (Gustavson) ; thus : H 3 P0 3 + Br 2 = HP0 3 + 2HBr. When phosphorus pentoxide is allowed to deliquesce in moist air or when it is dissolved in cold water metaphosphoric acid is formed. This aqueous solution on standing at the ordinary temperature of the air, gradually undergoes change with forma- tion of common phosphoric acid and this conversion takes place quickly without the formation of the intermediate pyrophos- phoric acid if the liquid be boiled (Graham). Metaphosphoric acid is volatile at a bright red heat and when heated with sulphates expels sulphuric acid from them, for although sul- phuric acid is a stronger acid than phosphoric acid, the former is more easily volatile than the latter. An aqueous solution of metaphosphoric acid is also obtained by passing a current of sulphuretted hydrogen gas through a liquid containing lead metaphosphate in suspension. The solution of metaphosphoric acid is distinguished from those of the other two modifications inasmuch as it produces a white precipitate with solutions of calcium chloride, barium chloride and albumen. 344 The Metaphosphatcs. The salts of metaphosphoric acid are obtained by neutralizing the aqueous solution of the acid by a base or by heating a dihydrogen orthophosphate ; thus : KH 2 P0 4 = KP0 3 + H 2 0. No less than five distinct modifications of the metaphosphates are known to exist. 2 (1) tfonometaphosphates. Of this class only those of the alkali metals are known, such as KPO 3 . The monometaphos- phates are remarkable as being insoluble in water. The potas- sium salt is formed, as above shown, when dihydrogen potassium phosphate is heated. The monometaphosphates are distinguished from the other modifications inasmuch as they do not form any double salts. (2) Dimetaphosphates. These salts are formed when aqueous phosphoric acid is heated to a temperature of 350 (Fleitmann) 1 Brescius, Zeit. Analyt. Chem. 6, 187. 2 Maddrell, Mem. Chem. Soc. 3, 373. 598 THE NON-METALLIC ELEMENT or 316 (Maddrell) with the oxides of zinc, manganese, or copper. If the copper salt be then decomposed by potassium sulphide a soluble potassium dimetaphosphate, K 2 P 2 O 6 , is obtained and if sodium sulphide be employed a soluble sodium dimetaphosphate is in like manner produced. In addition to the dimetaphosphates containing only one metal, double salts such as CuK 2 (P 2 O 6 ) 2 can be prepared. Only the dimetaphosphates of the alkali metals are soluble in water and are crystallizable ; the others are insoluble or only very slightly soluble (Fleitmann). (3) Trimetapliosphates. The sodium salt, Na 3 P 3 O 9 , is obtained together with the monometaphosphate when microcosmic salt, (NH 4 )HNaP0 4 , is gently heated until the fused mass becomes crystalline (Lindbom). By double decomposition other trimeta- phosphates can be obtained from this salt. These are all soluble in water, including the silver salt, and they form double salts such as NaBaP 3 9 . The silver salt may be obtained in the form of large transparent monosyin metric crystals by allowing a mixture of the sodium salt and nitrate of silver solution to stand for some days. In the same way the crystalline lead salt may be prepared by the substitution of nitrate (but not of acetate) of lead for the silver salt. (4) Tctrametaplwsphates. The lead salt, Pb 2 P 4 O 12 , is formed by treating oxide of lead with an excess of phosphoric acid and heating up to a temperature of 300. If this is then decom- posed by sodium sulphide the sodium salt is obtained as a tetrametaphosphate. This, however, is not a crystalline salt but forms with a small quantity of water a viscid elastic mass and on the addition of a larger quantity of water a gum-like solution which will not pass through a filter. The tetrameta- phosphates of the alkalis produce viscid precipitates with the soluble salts of the alkaline earths. If sodium dimetaphosphate is fused with copper dimetaphosphate and the mixture allowed gradually to cool a double compound having the composition CuNa 2 P 4 O 12 is formed (Fleitmann and Henneberg). (5) Hcxametapliosphates. The sodium salt, Na 6 P 6 O 18 , is obtained when fused sodium metaphospliate is allowed to cool slowly. It is a crystalline mass, deliquesces on exposure to the air, and produces with barium chloride a flocculent precipitate, and with the salts of the heavy metals gelatinous precipitates. It also forms characteristic double salts in which fine equiva- lents of the monad metal are replaced by the corresponding PYROPHOSPHORIC ACID 599 amount of a dyad metal, the calcium salt, for example, having the composition Ca" 5 Na 2 (PO 3 ) 12 or (NaPO 3 ) 2 (Ca"P 2 O 6 ) 5 . The metaphosphates which are soluble in water have a neutral or slightly acid reaction. When their solutions are boiled they are converted into orthophosphates. All the metaphosphates undergo this change on boiling with nitric acid or when they are fused with an alkali. The different varieties of metaphosphates are derived from acids, all of which possess the same composition, but differ, as the double salts show, from one another in molecular weight. Compounds of this description are termed polymeric bodies. If we assume the constitution formula now usually accepted for metaphosphoric acid, namely HO P~O O the constitution of these hypothetical acids may be represented graphically as follows : O OH V = P OH P and O = P OH HO O OH. Dimetaphosphoric Acid. Trimetaphosphoric Acid. PYROPHOSPHORIC ACID, H 4 P 2 O 7 . 345 Common phosphate of soda, or hydrogen disodium phos- phate, HNa. 2 PO 4 , when heated to a temperature of 240 loses water, and is converted into sodium pyrophosphate (Graham) : 2HNa 2 PO 4 = H 2 O + Na 4 P 2 O 7 . The salt thus obtained dissolves in water, but does not again form an orthophosphate, and is distinguished from the original salt inasmuch as its solutions yield on the addition of silver nitrate, a white, and not a yellow, precipitate. This fact was first observed by Clark, of Aberdeen, in the year 1828. 1 It is, however, to Graham that we are indebted for a knowledge of the fact that 1 Edinburgh Journal of Science, 7, 298. 600 THE NON-METALLIC ELEMENTS the heated sodium salt as well as the silver salt obtained from it is derived from an acid having the composition H 4 P 2 O 7 , and that this can be obtained from orthophosphoric acid by heating it for a considerable length of time to a temperature of 215. Pyrophosphoric acid is also formed when equal molecules of ortho- and metaphosphoric acids are heated together at 100 . 1 /OH HO V /OH HO P=0+H0 P< = >P P< || X OH HCK|| || X OH 00 00 Pyrophosphoric acid forms either a soft glassy mass (Graham) or an opaque indistinctly crystalline mass (Peligot). An aqueous solution of pyrophosphoric acid is obtained by precipitating the sodium salt with a solution of acetate of lead and decomposing the well-washed lead pyrophosphate with sul- phuretted hydrogen. The acid solution undergoes no change at the ordinary temperature, even after standing for some time, but when heated it is converted into orthophosphoric acid. Pyrophosphoric acid may be distinguished from the ortho- modification inasmuch as its solution produces a white granular precipitate with silver nitrate, and from the meta- variety inas- much as it does not produce a precipitate either with a solution of chloride of barium or with one of albumen. 346 Pyrophosphates. These salts are prepared from the mono- hydrogen orthophosphates by heat, or by neutralizing a freshly prepared solution of the acid by means of a base. Both normal pyrophosphates, such as Na 4 P 2 O 7 , and acid or hydrogen pyro- phosphates, such as H 2 Na 2 P 2 O 7 exist ; the first have an alkaline reaction, the second a slightly acid one. The pyrophosphates of the alkali metals are soluble in water, those of the other metals insoluble, but many of them dissolve in an excess of sodium pyrophosphate. The pyrophosphates in solution remain un- altered in the cold and even on' heating do not change, but when boiled with an acid are decomposed, the orthophosphates being formed. The same change takes place on fusion with an alkali. TetrapJwsphates. A sodium salt having the composition Na 6 P 4 O 13 , was obtained by Fleitmann and Henneberg by fusing 1 Geuther, J. Pr. Chcm. [2], 8, 359. ESTIMATION OF PHOSPHORIC ACID 601 sodium hexametaphosphate with pyrophosphate or orthophos- phate, in quantities represented by the following equations : Na 6 P 6 O 18 + 3Na 4 P 2 7 = 3Na 6 P 4 13 ; or Na P 6 18 + 2Na 3 P0 4 = 2Na 6 P 4 13 . The salt thus obtained can be crystallized from solution in warm water and gives with silver nitrate a white precipitate, Ag 6 P 4 13 which does not dissolve in an excess of the sodium salt. The constitution of the pyrophosphates, or the diphosphates as they may be called, and of the tetraphosphates may be exhibited as follows : Diphosphoric Acid. Tetraphosphoric Acid. OH OH /PO OH I 0< X PO-OH >PO OH \ \ \PO-OH >PO-OH I o< OH \PO-OH I OH 347 Quantitative Determination of Phosphoric Acid. Phos- phoric acid is best determined in the soluble phosphates by adding to the solution a mixture of sal-ammoniac, ammonia and magnesium sulphate solutions, when a precipitate of ammonium magnesium phosphate occurs, NH 4 MgP0 4 4- 6H 2 O. After this has stood for some time, the solution is filtered off, and the precipitate washed with dilute ammonia, dried, converted by ignition into magnesium pyrophosphate, Mg 2 P 2 O 7 , and weighed. The insoluble phosphates must be converted into soluble salts before the determination of the phosphoric acid. If they are soluble in nitric acid, the method proposed by Sonnenschein may be adopted ; namely, to precipitate the nitric acid solution of the phosphate by an excess of ammonium molybdate dis- solved in nitric acid. On standing for some time at a moderate temperature, a yellow precipitate, containing all the phosphoric acid, is formed ; the excess of molybdenum salt is removed by washing with water, the precipitate dissolved in ammonia, and the phosphoric acid precipitated by magnesium sulphate as described above. 002 THE NON-METALLIC ELEMENTS If the phosphates will dissolve in acetic acid the phosphoric acid may be precipitated with uranium acetate as uranium phosphate. This method may also be employed for the volu- metric determination of phosphoric acid, the point of complete precipitation of the phosphoric acid being ascertained by the addition of a drop of the solution to a drop of a solution of ferro- cyanide of potassium with which the slightest excess of uranium acetate produces a brown colour. Metaphosphates and pyrophosphates must first be converted into orthophosphates before precipitation. In like manner phosphorus itself as well as the hypophosphites and phosphites may also be quantitatively determined in the form of phosphoric acid by previously oxidizing them with nitric acid. HALOGEN DERIVATIVES OF PHOSPHORIC ACID. 348 The hydroxyl groups contained in the three modifications of phosphoric acid may be replaced, as is the case with other hydroxyacids, by chlorine or bromine thus giving rise to the oxy- chlorides and oxybromides of phosphorus, as they are commonly termed. PHOSPHORUS OXYFLUORIDE, POF 3 = 103-4, was first obtained by Moissan 1 by exploding a mixture of phosphorus trifluoride and oxygen and is also prepared by heating a mixture of two parts of cryolite and three parts of phosphorus pentoxide in a brass tube, 2 or by dropping phosphorus oxychloride on to dried zinc fluoride, 3 or by the action of anhydrous hydrogen fluoride on phosphorus pentoxide. 4 It is a gas which fumes in the air and condenses to a liquid at 50 or at 16 under a pressure of 15 atmospheres ; it solidifies at a very low temperature. PHOSPHORUS OXYCHLORIDE OR PHOSPHORYL CHLORIDE, POC1 3 . = 152-25. 349 This compound was discovered in the year 1847 by Wurtz, 5 who obtained it by decomposing the pentachloride with the requisite quantity of water : PC1 5 + H 2 = POC1 3 + 2HC1. 1 Compt. llend. 102, 1245. 2 Thorpe and Harably, Journ. Chem. Soc. 1889, i. 759. 3 Bull. Soc. Chim. (3), 4, 260. 4 Bull. Soc. Chim. (3), 5, 458. 5 Ann. Chim. Phys. [3] 20, 472. PHOSPHORUS OXYCHLORIDE 603 The best method of preparing this substance is the one pro- posed by Gerhardt ; l namely, by heating dried oxalic acid with phosphorus pentachloride, when the following reaction takes place : PC1 5 + H 2 C 2 O 4 = POC1 3 + 2HC1 + CO 2 + CO. Instead of oxalic acid, boric acid may be employed, thus : 3PC1 5 + 2B(OH) 3 = 3POC1 3 + B 2 O 3 + 6HC1. Pure phosphorus oxychloride may also be easily obtained by heating the pentachloride and the pent oxide together in sealed tubes in the proportion of 3 molecules of the former to one molecule of the latter : 2 3PC1 5 + P 2 5 = 5POC1 3 . The same compound is also obtained by distilling phosphorus pentoxide with common salt : 3 2P 2 O 5 + SNaCl = POC1 3 + 3NaP0 3 . Phosphorus oxychloride is also formed as a product of decom- position in the preparation of many acid chlorides. Several ex- amples of this mode of formation have already been mentioned and many more will have to be described. Phosphorus oxychloride is a colourless mobile liquid boiling at 107-2 and having a specific gravity of 1-7118 at (Thorpe). When strongly cooled it solidifies in the form of tabular or needle-shaped crystals which melt at -1'5 (Geuther and Michaelis). It fumes strongly in the air and has a very pene- trating and acrid smell resembling that of phosphorus trichloride. According to Cahours the specific gravity of the vapour is 5'334. The oxychloride when thrown into water sinks, and slowly dissolves with the formation of phosphoric and hydrochloric acids : POC1 3 + 3H 2 O = PO(OH) 3 + 3HC1. When brought in contact with many metallic chlorides it forms crystalline double compounds. 1 Ann. Chim. Phys. [3] 44, 102. 2 Gerhardt and Chiozza, Annalen, 87, 290. 3 Kolbe and Lautemami, Annalcn, 113, 240. 604 THE NON-METALLIC ELEMENTS PYROPHOSPHORYL CHLORIDE, P 2 O 3 C1 4 . 350 When the four hydroxyl-groups in pyrophosphoric acid, H 4 P 2 O 7 , are replaced by chlorine, pyrophosphoryl chloride is formed. It was first prepared by Geuther and Michaelis, 1 by the action of nitrogen peroxide on phosphorus trichloride. The reaction which takes place in this case is a somewhat compli- cated one, nitrogen, nitrosyl chloride, phosphoryl chloride, and phosphorus pentoxide being formed. Pyrophosphoryl chloride is a colourless strongly fuming liquid boiling between 210 and 215 and then undergoing partial decomposition into pentoxide and common oxychloride : At 7 the specific gravity of the liquid is 1*78, it decomposes violently in contact with water without sinking in it, and in this decomposition orthophosphoric acid and not pyrophosphoric acid is formed. When it is treated with pentachloride of phosphorus, phosphoryl chloride is produced : P 2 (XC1 4 + PC1 5 = 3POC1 3 . By the action of phosphorus pentoxide on phosphorus oxy- chloride, Gustavson 2 obtained a syrupy mass which he regarded as metaphosphoryl chloride PO 2 C1. Hambly 3 has however shown that this is a mixture of at least two compounds, one of which is pyrophosphoryl chloride ; the other constituent has a constant composition but cannot from the analysis have a simpler formula than P 7 O 15 C1 5 and is probably itself a mixture. PHOSPHORUS OXYBROMIDE, OR PHOSPHORYL BROMIDE. POBr 3 = 28476. 351 This body is formed by the action of a small quantity of water upon the pentabromide, but it is best prepared by distil- ling pentabromide of phosphorus with oxalic acid (Baudrimont). It forms a mass of flat tabular crystals which have a specific gravity of 2'822 (Hitter), melt at 46 and boil at 195. Water decomposes the oxybromide into phosphoric and hydrobromic acids. 1 Ber. 4, 766. 2 a\ 4, 853. 3 Journ. Chcm. Soc. 1891, i. 202. SULPHIDES OF PHOSPHORUS 605 PHOSPHORYL BROMOCHLORIDE. POBrCl 2 . = 196-42. When phosphorus trichloride is allowed to fall drop by drop into absolute alcohol, C 2 H 5 (OH), the first action that takes place is the formation of the compound (OC 2 H 5 )PC1 2 . This body is decomposed by the addition of bromine into ethyl bromide, C 2 H 5 Br, and phosphorus oxybromochloride, POBrCl 2 . This latter compound is a highly refractive liquid, boiling at 136 and having at a specific gravity of 2'049 (Menschutkin). When the liquid is cooled it solidifies in the form of tabular crystals melting at 11, and probably isomorphous with the crystals of phosphoryl chloride and phosphoryl bromide (Geuther and Michaelis). PHOSPHORUS AND SULPHUR. 352 The compounds formed by these two elements were formerly divided into two classes according as they contained a greater number of atoms of phosphorus or sulphur. Two com- pounds of the former class have been described, namely, sulphur tetraphosphide SP 4 and sulphur diphosphide SP 2 , but later researches 1 have shown that these substances are in reality simply solutions of sulphur in phosphorus, for no heat is evolved in their formation from the elements, and a current of an indifferent gas removes the whole of the phosphorus at temperatures below its boiling point. The sulphides of phosphorus are solid bodies formed when phosphorus and sulphur are gently heated together, the reaction being accompanied by the evolution of heat. If common phosphorus is employed for this purpose violent explosions may occur, and hence it is advisable to employ red phosphorus. Even in the latter case, when finely divided sulphur, such as flowers of sulphur, is employed, the reaction is often very violent, and for this reason the sulphides of phos- phorus are best prepared by mixing the necessary quantity of red phosphorus with small lumps of roll sulphur. The mixture is effected in a flask, the cork loosely placed in, and the flask then heated on a sand bath by means of a Bunsen flame, 1 Schultze, J. Pr. Chem. [2] 22, H3 ; Ber. 16, 2066 ; Isambert, Compt. Rend. 96, 1499, 1628, 1721, 606 THE NON-METALLIC ELEMENTS until the reaction begins, when the flame is removed. After the flask has cooled it may be broken to obtain the solid mass which is then preserved in dry well-closed bottles (Kekul4). TETRAPHOSPHORUS TRISULPHIDE, P 4 S 3 . This compound is obtained as a yellow mass, crystallizing from solution in carbon bisulphide or phosphorus trichloride in the form of rhombic prisms. It melts at 166 (Ramme), forming a reddish liquid, which boils about 380 (Isambert). When heated in a current of carbon dioxide it sublimes at 260 and condenses to form crystals, which appear to belong to the regular system. The specific gravity of its vapour is 7'90. It is very easily inflammable and is slowly decomposed in contact with boiling water with formation of sulphuretted hydrogen, phos- phuretted hydrogen, and phosphorous acid : P 4 S 3 + 9H 2 = 3SH 2 + 3P(OH) 3 + PH 3 . TETRAPHOSPHORUS HEXASULPHIDE, P 4 S C , Is obtained according to Isambert 1 by heating the calculated quantities of the elements in an atmosphere of carbon dioxide. It crystallizes in thin needles, and boils at 490, the vapour density corresponding to the above formula. TRIPHOSPHORUS HEXASULPHIDE, P 3 S 6 . This is obtained in crystalline needles, melt-ing from 296 298, by heating yellow phosphorus and sulphur with carbon bisulphide for some hours to 210. The above molecular formula was established by a vapour density determination (Ramme). 2 PHOSPHORUS PENTASULPHIDE, P 2 S 5 . 353 In order to prepare this compound in the pure state, the sulphide prepared by Kekule's method is distilled in a current of carbon dioxide. A pale yellow crystalline mass is thus obtained, and frequently distinct crystals. 3 When a mixture of yellow 1 Compt. Rend. 102, 1386. 2 Ber. 12, 940, 1350. 3 Carl Meyer and V. Meyer, Ber. 12, 610. THIOPHOSPHORIC ACID 607 phosphorus and sulphur in the proper proportions is heated with carbon bisulphide to 210, fine, pale, yellow crystals separate out (Ramme). Phosphorus pentasulphide melts at 274 276, boils at 518 (Goldschmidt), and yields a brown vapour having a specific gravity of 7'67 (C. and V. Meyer). Water decomposes this substance as follows : P 2 S 6 + 8H 2 = 2PO(OH) 3 + 5H 2 S. Phosphorus pentasulphide is often used in making organic preparations for the purpose of replacing oxygen in organic compounds by sulphur. Thus, for instance, if common alcohol, C 2 H.OH, be heated with this body, thioalcohol or mercaptan, C 2 H 5 SH, is formed. PHOSPHORUS SULPHOXIDE, P 4 O 6 S 4 , Is obtained by carefully heating phosphorous oxide with the calculated quantity of sulphur to 150 170, the product when heated in vacuo subliming in colourless strongly refractive crystals. It melts at about 102, and boils at 295, yielding a vapour whose density corresponds to the formula P 4 O 6 S 4 . It rapidly deliquesces in the air and is quickly dissolved by water, form- ing sulphuretted hydrogen and metaphosphoric acid, which is eventually converted into orthophosphoric acid : 6H 2 = 4HP0 3 + 4H 2 S. MONOTHIOPHOSPHORIC ACID, H 3 PSO 3 . 354 This substance may be regarded as phosphoric acid in which one atom of sulphur replaces one atom of oxygen. It is however, not known in the free state. The sodium salt, Na 3 PSO 3 , is produced when the chloride (described below) is heated with caustic soda. The salt forms distinct crystals, which have an alkaline reaction, and are decomposed by the weakest acids, the thiophosphoric acid which is thus liberated being at once decomposed into sulphuretted hydrogen and phosphoric acid ; thus : PS(OH) 3 + H 2 = PO(OH) 3 + SH 2 . Dithiophosphoric acid, H 3 PS. 2 O 2 , and trithiophosphoric acid, H 3 PS 3 0, are also known in their salts. 1 1 Kubierschky, J. Pr. CJtcm. [2], 31, 93. 608 THE NON-METALLIC ELEMENTS THIOPHOSPHORYL FLUORIDE, PSF 3 . 355 This compound is obtained by the action of thiophos- phoryl chloride on arsenic fluoride, but is best prepared by heating a mixture of carefully dried lead fluoride and phos- phorus pentasulphide to 170 250, the following reaction taking place : P 2 S 5 + 3PbF 2 = 3PbS + 2PSF 3 . At the ordinary temperature it is a transparent colourless gas, which condenses to a liquid under a pressure of 10 11 atmospheres. It does not attack dry glass, but is spontaneously inflammable in air or oxygen, and is decomposed by heat or the electric spark with separation of sulphur and phosphorus and formation of phosphorus fluorides. The products obtained by the action of oxygen consist of phosphorus pentafluoride, phos- phorus pentoxide and sulphur dioxide, and with water, of sulphuretted hydrogen, phosphoric and hydrofluoric acids. 1 THIOPHOSPHORYL CHLORIDE, PSC1 3 . This compound, which corresponds to oxychloride of phos- phorus is best obtained by acting upon pentachloride of phos- phorus with phosphorus pentasulphide (Weber) : P 2 S 5 = 5PSC1 3 . It is a colourless, mobile, very refractive liquid, whi(5h fumes strongly in the air, possesses a powerful pungent odour, boils at 125, and has a specific gravity at of 1-16816 (Thorpe). The specific gravity of the vapour according to Cahours is 5 '878 at 298. It is decomposed by water into hydrochloric acid and thiophosphoric acid, which is then further decomposed as described above. THIOPHOSPHORYL BROMIDE, PSBr 3 . According to Baudrimont this body is obtained by distilling the tribrornide of phosphorus with flowers of sulphur. It is however best prepared by dissolving equal parts of sulphur and 1 Thorpe and Rodger, Journ. Chcm. Soc. 1889, i. 306. PHOSPHOROUS DIAMIDE 609 phosphorus in carbon bisulphide, and adding gradually to the well-cooled liquid eight parts of bromine. On distillation the carbon bisulphide first comes over and then the thio-brornide. This is purified by shaking it up several times with cold water ; a hydrate is formed in this way having the composition PSBr s + H 2 O, which on being warmed to 35 separates into its constituents. The anhydrous compound is obtained by the evaporation of the solution in carbon bisulphide as a yellow liquid, which, when touched by a solid body, at once solidifies to a crystalline mass. This compound can also be obtained in the crystalline state from solution in phosphorus tri- bromide, when it separates out in regular octahedra, having a yellow colour and melting at 38. When heated with water it is slowly decomposed, and on distillation is resolved partly into sulphur and into a compound, PSBr 3 PBr 3 , which boils at 205 . 1 PYROPHOSPHORYL THIOBROMIDE, P 2 S 3 Br 4 . 356 This compound, the sulphur analogue of pyrophosphoryl chloride, is obtained by pouring carbon bisulphide over phos- phorus trisulphide and adding to it, drop by drop, a solution of bromine in carbon bisulphide. 2 It forms a light yellow oily liquid which fumes in the air, has an aromatic and pungent smell, and on heating decomposes into the ortho- compound and phosphorus pentasulphide ; thus : PHOSPHORUS AND NITROGEN. 357 No compound of phosphorus and nitrogen is known, but derivatives of phosphorous and phosphoric acids containing this element have been prepared. Phosphorous diamide HO.P(NH 2 ) 2 is obtained by the action of ammonia on a solution of phosphorous oxide in ether or benzene. It is a white powder which dissolves in water instantly, with such violence that it becomes incandescent. When treated with moderately dilute hydrochloric acid, a violent reaction occurs with liberation of non-spontaneously inflammable phosphine, 1 Michaelis, Annalcn, 104, 9. 2 Michaelis, loc. cit. 40 610 THE NON-METALLIC ELEMENTS separation of free phosphorus, and formation of a solution of ammonium chloride, phosphorous and phosphoric acids. 1 Amidophosphoric acid or Phosphamidic acid, NH 9 .PO(OH) 2 . The potassium salt of this acid is obtained by the action of potash on the corresponding phenyl salt. The free acid, which is prepared by decomposing the silver salt with sulphuretted hydrogen and adding alcohol to the filtered solution, crystallises in colourless microscopic crystals, which have a sweet taste, and are not hydrolysed by caustic alkalis. It forms both normal and acid salts. 2 Diamidophosphoric acid, (NH 2 ) 2 PO.OH, is obtained in a similar manner to the foregoing compound. 3 It is a crystalline substance and is converted by nitrous acid, first into monamido- phosphoric acid and then into orthophosphoric acid. In addi- tion to the normal silver salt, (NH 2 ) 2 PO.OAg, it forms a remarkable compound of the formula (NHAg) 2 P(OAg) 3 , derived from the unknown acid (NH 2 ) 2 P(OH) 3 . 358 When dry ammonia is passed over phosphorus penta- chloride as long as it is absorbed, a white mass is obtained consisting probably of a mixture of the compound PC1 3 (NH 2 ) 2 and NH 4 CJ. No means of separating these are known, but if the product be treated with water ammonium chloride dissolves, and the chloramido-compound is decomposed with loss of hydro- chloric acid, and is converted into phosphamide PO(NH)NH 2 . This remains behind as a white insoluble powder, which is slowly converted by the boiling liquid into acid ammonium phosphate : PO(NH)NH 2 + 3H 3 = PO(OH)(ONH 4 ) 2 . Phospham, P 3 H 3 N 6 (?). If the product of the reaction of ammonia and phosphorus pentachloride is heated in absence of air until no further fumes of ammonium chloride are evolved, a light white powder having the above composition remains behind to which the name phospham has been given. 4 It is insoluble in water and does not melt at a red heat, but oxidises on heating in the air with evolution of white fumes. If it be moistened with water and then heated, metaphosphoric acid and ammonia are formed : P 3 H 3 N 6 + 9H 2 O = 3PO 3 H + 6NH 3 . 1 Thorpe and Tutton, Journ. Chem. Soc. 1891, i., 1027. 2 Stokes, Amer. Chem. Journ. 15, 198. 3 Stokes, Ber. 27, 565. 4 Liebig and Wohler, Annalcn, H, 146 ; Hofmann, Ber. 17, 1909. IMIDODIPHOSPHORIC ACIDS 611 When fused with potash it decomposes with evolution of light and heat : P 3 H 3 N 6 + 9KOH + 3H 2 = 3PO(OK) 3 + 6NH 3 . Its exact molecular weight has not yet been ascertained, but from analogy with the corresponding phosphorus chloronitride it has probably the molecular formula given above. Phosphoryl nitride, PON, is obtained by heating phosphamide to redness in absence of air, or by subjecting the product of the reaction of ammonia and phosphorus oxychloride at to the same treatment. It forms a white amorphous powder, which melts at a red heat, and resolidifies to a glassy mass. It is not acted upon by nitric acid, but when fused with caustic alkalis is converted into ammonia and potassium orthophosphate. PON + 3KOH = PO(OK) 3 + NH 3 . IMIDODIPHOSPHORIC ACIDS. 359 The action of ammonia on phosphorus oxychloride was investigated many years ago by Gladstone l and by Schiff. 2 The former succeeded in obtaining from the product a number of acids, which he regarded as pyrophosphaminic acids, or acids derived from pyrophosphoric acid by the replacement of hydroxyl by the group NH 2 . The later investigations of Mente 3 have however shown, that these acids are in reality formed from diphosphoric acid by the replacement of one or more oxygen atoms by the imido group NH and they may therefore be termed imidodiphosphoric acids. The constitution of these acids is probably as follows : / NH \ Imidodiphosphoric acid, HO.PO< >PO.OH, \ O / X NH \ Diimidodiphosphoric acid, HO.PO( >PO.OH, / NH \ Diimidodiphosphaminic acid, HO.POc )PO.NH 2 . 1 AnnaUn, 76, 74 ; Journ. Chem. Soc. 1850, 121 ; 1851, 135, 353 ; 1864, 215 ; 1866, 290 ; 1868, 64, 261 ; 1869, 15. 2 Annalen, 101, 299 ; 102, 113 ; 103, 169. 3 Annalen, 248, 232. 612 THE NON-METALLIC ELEMENTS In addition to these Mente has also prepared nitrilotrimeta- P o/ OH phosphoric acid, which has the constitution Phosphoryl triamide, PO(NH 2 ) 3 has been described by Schiff, but later investigators have failed to obtain it, his compound being probably impure diimidodiphosphaminic acid. Phosphorus chloronitride, P 3 N 3 C1 6 . This substance was first obtained by Lie big and Wohler among the products of the reaction of ammonia and phosphorus pentachloride, and has been more closely investigated by Gladstone * and Wichelhaus. 2 It is best prepared by subliming a mixture of one part of phosphorus pentachloride and two parts of dried ammonium chloride, purifying the sublimate by washing with Avater and distilling in a current of steam. It crystallises in thin trans- parent six-sided rhombic tablets, melts at 114 and boils at 250; its vapour density is 12'05, corresponding to the formula given above. It is insoluble in water, but is slowly decomposed by it with formation of diimidodiphosphoric acid. The substance PNC1 2 obtained by Besson by heating the additive compound PC1 5 , 8NH 3 , is probably identical with this substance. Besson has obtained phosphorus bromonitride in a similar manner ; it forms refractive apparently rhombohedral crystals, melts at 188-190, and begins to sublime at 150 in vacuo. 3 ARSENIC. As = 74-4. 360 The yellow and red sulphides of arsenic, now termed orpiment and realgar, were known to the ancients, although they did not distinguish between them. Aristotle gave to them the name of cravSapd^rj, which was also applied to cinnabar and red lead, and Theophrastus mentioned them under the name of White arsenic or arsenious oxide, As 4 6 , is first distinctly 1 Journ. Chem. Soc. 1869, 15. 2 Ber. 3, 163. 3 Compt. Rend. 114, 1264, 1479. ARSENIC 613 mentioned in the writings of the Greek alchemist Olympiodorus, who describes it under the name of white alum, and gives a recipe for its preparation from the sulphide by roasting in the air. 1 The later alchemists were all acquainted with these sub- stances. Thus, for instance, in the works attributed to Basil Valentine they are described as follows : " In its colour the arsenicum is white, yellow, and red ; it is sublimed by itself without any addition, and also with addition according to manifold methods." The alchemists made use of arsenic especially for the purpose of colouring copper white (see p. 7). The change thus brought about was believed to be the beginning of a transmutation although Albertus Magnus was aware that on strongly heating the alloy, the arsenic is volatilized. He describes this fact in his work " De rebus metallicis " as follows : " Arsenicum aeri con- junctum penetrat in ipsum, et convertit in candorem ; si tamen diu stet in igne, aes exspirabit arsenicum, et tune redit pristinus color cupri, sicut de facile probatur in alchymicis." Free arsenic was known to the Greek alchemists, who obtained it as a sublimate capable of turning copper white, and hence looked upon it as a kind of mercury. 2 That a metal-like substance is contained in white arsenic was probably known to Geber, but Albertus Magnus was the first to state this distinctly : " Arsenicum fit metallinum fundendo cum duabus partibus saponis et una arsenici." Metallic arsenic was considered by the later alchemists and chemists to be a bastard or semi-metal, and was frequently termed arsenicum rex. It was, however, Brandt who in the year 1773 first showed that white arsenic is a calx of this substance. After the overthrow of the phlogistic theory the views concerning the composition of white arsenic were those which are now held, namely, that it is an oxide of the elementary substance. Arsenic occurs in the free state in nature, usually in mam- millated or kidney-shaped masses, which readily split up into laminae. Occasionally, however, native arsenic is met with in distinct crystals. It occurs in large quantity at Andreasberg in the Harz, in Joachimsthal in Bohemia, at Freiberg in Saxony, at ZimeofT in Siberia, in Borneo, and at Newhaven in the United States. Arsenic occurs much more commonly in a state of combination in many ores and minerals, of which the following are the most 1 Berthelot, Introduction ti I'titude, &c., p. 68. 2 Ibid. p. 99. 614 THE NON-METALLIC ELEMENTS important : arsenical iron, FeAs 2 ; tin white cobalt (CoNiFe)As 2 ; arsenical nickel, NiAs ; arsenical pyrites or mispickel, Fe 2 S 2 As ; realgar, As 2 S 2 , and orpiment, As 2 S 3 . Less frequently we find white arsenic or arsenious oxide as arsenite, As 4 O 6 , and several salts of arsenic acid, such as Pharmacolite (HCaAs0 4 ) 2 + 5H 2 O, Cobalt bloom Co 3 (AsO 4 ) 2 + 8H 2 O, Mimetesite 2Pb 3 (AsO 4 ) 2 + Pb 2 (P0 4 )CL Small quantities of arsenic also occur in many other minerals. Thus, for instance, it is contained in almost all specimens of iron pyrites so that it is often found in the sulphuric acid which is manufactured from pyrites, and in the various preparations for which this acid serves. Especially remarkable is the occurrence of arsenic in almost all mineral waters, in which of course it is only contained in traces (Will, Fresenius). Similarly it has been detected in sea-water. 361 Preparation. The arsenic occurring in commerce is either the natural product, which is never quite pure but contains iron and other metals mixed with it, or, more generally, that obtained by heating arsenical pyrites in earthenware tubes in a furnace. These tubes are 1 metre in length and about 32 cm. wide. In the open end of this earthenware tube another is placed, made of sheet iron, about 20 cm. long, so that half of the iron tube is inside the earthenware one. The arsenic sublimes into the iron tube, from which it is obtained by unrolling the sheet iron. Prepared by this process, arsenic forms a compact, brittle, crystalline mass having a strong metallic lustre. The decom- position which takes place is represented by the equation : FeSAs = In Silesia it is prepared from arsenious oxide, which is heated with charcoal in an earthen crucible covered with a conical iron cap, into which the arsenic sublimes. In order to purify the commercial arsenic it is sublimed with the addition of a small quantity of powdered charcoal. On the small scale, arsenic may be purified by introducing the mixture into a glass flask which is then placed in a large crucible, sur- rounded by sand and heated to redness. As soon as the sublimation begins, a loosely-fitting stopper of chalk is placed in the neck of the flask, a second crucible is placed over the first, and the whole heated until the arsenic is sublimed into the PROPERTIES OF ARSENIC 615 upper portion of the flask. In this way rhombohedral crystals of arsenic are obtained, which have a bright metallic lustre and are isomorphous with tellurium and antimony (Mitscherlich). Properties. Arsenic has a steel-grey colour. Its specific gravity at 14 is 5'727, its specific heat 0'083 (Wiillner and Bettendorf), and it is a good conductor of electricity. If pure arsenic is quickly sublimed in a stream of hydrogen gas it is deposited in the neighbourhood of the heated portion of the tube in crystals, but, at a little distance, as a black glittering mass, and, still further on, as a yellow powder. The last two modifications of arsenic are both amorphous ; the first has a specific gravity of 4713, the last of S'70. 1 When heated to 360 they are both transformed into the crystalline variety. According to Berthelot and Engel amorphous and crystalline arsenic evolve the same amount of heat on oxidation. 2 It was formerly supposed that arsenic could not be melted, for when heated under ordinary circumstances to about 450 it passes at once from the solid to the gaseous state. Landolt 3 has, however, shown that under an increased pressure it melts at 500. On cooling, the fused mass forms a dense crystalline solid which has, at 19, a specific gravity of 5*709. Its melting point lies between those of antimony and silver (Mallet). The vapour of arsenic is of a lemon yellow colour and smells disagreeably of garlic. It is, however, still uncertain whether this smell is due to the element itself, or to a low oxide which has not yet been isolated. The specific gravity of the vapour at 8.60 was found by Deville and Troost to be 10 '2, whilst Meyer and Biltz 4 obtained the values 5'45 at 1714, and 5'37 at 1736. The molecule of arsenic at these high temperatures therefore consists of two atoms, whilst at 860 the vapour density points to the presence of four atoms in the molecule. Arsenic oxidizes somewhat rapidly in moist air at the ordinary temperature, becoming covered with a blackish-grey coating. Heated in oxygen it burns with a bright white flame, forming arsenious oxide, which is also produced when arsenic is heated in the air ; at the same time the alliaceous smell of its vapour is perceived. In the act of combination with oxygen, 1 gram of arsenic evolves, according to Thomsen, 1031 thermal units, whilst by its combination with chlorine, in which gas finely powdered arsenic takes fire spontaneously, it evolves 953 thermal 1 Geuther, Annalen, 240, 208. 2 Compt. Rend. HO, 498. 3 Jahrbuchf. Min. 1859, 733. 4 Ber. 22, 725. 616 THE NON-METALLIC ELEMENTS units. It is easily oxidized by nitric acid, and also by con- centrated sulphuric acid with evolution of sulphur dioxide. It combines with various non-metals, and with most of the metals. It stands in such close proximity to the latter class of elements, especially in its physical properties, such as lustre, specific gravity, &c., that some chemists have placed it amongst them. Metallic arsenic is chiefly used for the purpose of hardening lead in the manufacture of shot. The atomic weight of arsenic was first determined by Berzelius, 1 who heated 2'203 grm. of As 4 O 6 with sulphur, and obtained 1-069 of SO 2 , hence As = 74'44. By the analysis of the tri- chloride Dumas 2 obtained As = 74'39, whilst Kessler 3 by oxi- dizing As 4 O 6 to As 2 O 5 found As = 74'7. The most accurate value of the atomic weight is probably 74'4. ARSENIC AND HYDROGEN. These two elements unite to form a gaseous compound, AsH 3 , and a solid compound, As 2 H 2 . HYDROGEN ARSENIDE, OR ARSINE, AsH 3 = 77*4. 362 The existence of this gas was first noticed by Scheele in 1775 on treating a solution of arsenic acid with zinc. He found that the gas thus evolved deposited white arsenic on burning, and explained this as being due to the fact that the inflammable air had dissolved some arsenic. Proust showed in 1799 that the same gas is given off when arsenious acid and dilute sulphuric acid are brought together in the presence of zinc, and also when sulphuric acid is allowed to act on arsenical metals. The gas which is thus given off is a mixture of hydrogen and arsine. In order to prepare hydrogen arsenide in the pure state, zinc arsenide, As 2 Zn 3 , must be decomposed by dilute sulphuric acid, 4 thus : As 2 Zn 3 + 3H 2 S0 4 = 2AsH 3 + 3ZnS0 4 . Zinc arsenide is obtained by heating zinc and arsenic together in a closed crucible, when heat enough is evolved to melt the mass. The greatest care must be taken in the preparation of this gas, as it is extremely poisonous, a quantity no larger than 1 Pogg. Ann. 8, 1. 2 Annalen, 113, 29. 3 Pogg. Ann. 95, 204 ; 113, 134. 4 Soubeiran, Ann. Chim. Phys. [2], 23, 307 ; 43, 407. ARSINE 617 one bubble having been known to produce fatal effects. Gehlen lost his life in this way in the year 1815. Arseniuretted hydrogen, as the gas was formerly called, is also formed by the electrolysis of arsenious and arsenic acid (Bloxam) and in small quantity when arsenic is boiled with water. 1 363 Properties. Arseniuretted hydrogen has a very peculiar and disagreeable smell, and, according to Dumas, possesses a specific gravity of 2'695. It liquefies at 40 (Stromeyer), but it does not solidify when cooled to a temperature of 110. Arsine is formed from its elements with absorption of heat (p. 232), and like many other compounds of the same class can be made to undergo explosive decomposition into its elements when exposed to the shock produced by the detonation of fulminate of mercury, although it does not explode when heated. 2 It burns with a pale bluish flame, emitting dense clouds of arsenious oxide. If a cold piece of white porcelain be held in the flame, metallic arsenic is deposited as a brown or black shining mirror, and when the gas is passed through a glass tube which is heated in one or in several places by means of a gas flame, the arsenic is deposited near the heated portions of the tube in the form of a bright shining mirror, the hydrogen being liberated. This decomposition takes place at 230 . 3 Whenever hydrogen is liberated by means of an acid from any liquid containing arsenic in solution, traces of arseniuretted hydrogen are evolved. This may be easily detected either by the smell or by the above-mentioned reactions, which are of such delicacy that O'Ol mgrm. ( T ^Vo^ f a grai* 1 ) can with certainty be recognized (Otto). Hydrogen is evolved during the growth of moulds and of cer- tain fungi, and it is possible that if arsenic compounds are present where such growths are going on, arseniuretted hydrogen may be evolved. 4 This may, perhaps, explain the evil effects noticed when arsenical wall papers are employed. 5 At the same time it must be remembered that in these cases arsenic doubtless finds its way into the system in the form of dust, which in such rooms invariably contains it. When arseniuretted hydrogen is led over heated oxide of copper, water and copper arsenide are formed. In this way 1 Cross and Higgin, Ber. 16, 1198. 2 Berthelot, Gompt. Rend. 93, 613. 3 Brunn, Ber. 22, 3205. 4 Selmi, Ber. 7, 1642 ; Giglioli, Ber. 14, 2295. 5 Fleck, Dingl. Polyt. Journ. 207, 146. 618 THE NON-METALLIC ELEMENTS arseniuretted hydrogen ean easily be determined quantitatively. When metals such as tin, potassium, or sodium are heated in the gas the arsenides of the metals are formed, free hydrogen, which occupies 1J times the volume of the original arseniu- retted hydrogen, being generated. When potassium and sodium are heated in the gas, the compounds AsK 3 and AsNa 3 are formed. When decomposed by dilute acids these substances yield arseniuretted hydrogen which is purer than that obtained from zinc arsenide (Janowsky). Arseniuretted hydrogen when passed into a dilute solution of a gold or silver salt precipitates the metal, the arsenic entering into solution in the form of arsenious acid : 2AsH 3 + 12AgNO 3 + 6H 2 = 2H 3 As0 3 -f 12HNO 3 + 12Ag. A very small quantity of arseniuretted hydrogen can, in this way, be detected by the precipitation of finely divided silver from the clear solution, the liquid becoming acid at the same time. When on the other hand, arsine is passed into a strong solution of silver nitrate, a yellow coloration Is produced which is due to the formation of a double salt, As 3 Ag,3AgNO 3 : AsH 3 + 6AgNO 3 = As 3 Ag,3AgNO 3 + 3HNO 3 . On the addition of water this substance is decomposed with formation of arsenious acid and nitric acid, metallic silver being precipitated. 1 One volume of water absorbs about five volumes of the gas, and the solution on exposure to the air deposits arsenic. Chlorine decomposes arseniuretted hydrogen with great violence, bromine and iodine also act upon it less energetically. Sulphu- retted hydrogen does not act upon the pure gas, but in the presence of oxygen or at a temperature above 230 the yellow sulphide of arsenic is formed. 2 SOLID HYDROGEN ARSENIDE. As 2 H 2 . 364 This substance is formed as a brown silky mass when sodium arsenide, AsNa 3 , is decomposed by water. 3 The brown powder produced by leaving the gaseous compound in contact with moist air, or by decomposing one part of arsenic and five of zinc by hydrochloric acid, which was formerly supposed to 1 Poleck and Thiimmel, Ber. 16, 2438. 2 Brunn, Ber. 22, 3202. * Janowsky, Ber. 6, 220. ARSENIC TRICHLORIDE 619 be this substance, has been shown to be nothing but the element in a finely divided state. According to Ogier l a solid hydride of the formula As 2 H is formed by the action of the electric current on arsine. ARSENIC AND FLUORINE. ARSENIC TRIFLUORIDE, AsF 3 = 131-5. 365 This compound was obtained by Unverdorben 2 by distil- ling a mixture of four parts of arsenious oxide and five parts of fluor-spar with ten parts of sulphuric acid. It is a transparent colourless liquid boiling at 63 and having a specific gravity of 2'73 ; it fumes strongly in the air, has a pungent and powerful odour, and when brought in contact with the skin produces serious wounds which only heal after a long time (Dumas.) It attacks glass, and is decomposed by water into arsenious and hydrofluoric acids. It is soluble in ammonia, and forms with this gas a crystalline compound. Arsenic Pent a fluoride is not known in the free state. Marignac obtained a double compound AsF 5 + KF in colourless crystals by dissolving potassium arsenate in hydrofluoric acid. ARSENIC AND CHLORINE. ARSENIC TRICHLORIDE. AsCl 3 = 180. 366 This is the only known compound of arsenic and chlorine. Glauber first prepared this substance ; his work " Furni novi philosophici," published in the year 1648, contains the follow- ing recipe : " Ex arsenico et auripigmento to distil a butter or thick oil. As has been described under antimony, so likewise from arsenicum or auripigmentum with salt and vitriol can a thick oil be distilled." Preparation. In order to prepare arsenic trichloride accord- ing to this plan, 40 parts of arsenious oxide must be heated with 100 parts of sulphuric acid to the boiling point of water in an apparatus which is connected with a well-cooled receiver. Small pieces of fused chloride of sodium are then carefully 1 Compt. Ecnd. 89, 1068. - Pogg. Ann. 7, 316 ; see also Moissan, Compt. Rend. 99, 874. 620 THE NON-METALLIC ELEMENTS thrown in. The decomposition which takes place is as follows : 12HC1 + As 4 6 = 4AsCl 3 + 6H 2 0. The water, which is formed at the same time, remains behind in combination with the sulphuric acid, whilst the trichloride distils over. The same compound is easily obtained by passing dry chlorine over heated arsenic which burns to chloride of arsenic. In order to purify it from excess of chlorine it must be rectified over some more arsenic. Properties. Arsenic trichloride is a colourless oily liquid and has a specific gravity of 2*205 (0/4). It solidifies at 18 in pearly needles 1 and boils at 130'2 (Thorpe), evolving a colour- less vapour, which has a specific gravity of 6*3 (Dumas). It is an extremely powerful poison, and evaporates in the air with the emission of dense white fumes. When arseniuretted hydro- gen is led into the liquid, arsenic separates out (Janowsky) ; thus : AsCl 3 + AsH 3 = As 2 + 3HC1. When brought in contact with a small quantity of water, star-shaped crystalline needles of arsenic oxychloride, As(OH) 2 Cl separate out (Wallace). In contact with a large quantity of water, it decomposes into arsenious oxide and hydrochloric acid, and when the solution is distilled, arsenic -trichloride comes over together with the vapour of water ; this explains the fact that the hydrochloric acid prepared from arsenical sulphuric acid invariably contains arsenic. Arsenic trichloride absorbs dry ammonia, forming a solid compound having the composition AsCl 3 +3NH 3 , which dissolves in alcohol, being deposited from this solution in white crystals. 2 ARSENIC AND BROMINE. ARSENIC TRIBROMIDE, AsBr 3 = 312-4. 367 In order to prepare this compound, powdered arsenic is added to a solution of one part of bromine in two parts of carbon bisulphide until the solution becomes colourless. Then bromine and arsenic are added alternately until the colour of the first 1 Besson, Compt. Rend. 109, 940. 2 Wallace, Phil. Mag. [4], 16, 358. ARSENIC TEI-IODIDE 621 disappears ; the clear liquid is poured off, and the bisulphide of carbon allowed to evaporate spontaneously. Arsenic tribromide forms colourless deliquescent crystals which possess a strong arsenical odour (Nickles), and melt at about 20. It has a specific gravity of 3'66, and boils at 220. By the action of water, it is decomposed in a similar way to the chloride. ARSENIC AND IODINE. AESENIC DI-IODIDE, AsI 2 = 326'2. 368 When arsenic tri-iodide is heated with arsenic together with a little carbon bisulphide in a sealed tube, arsenic di- iodide is formed, and the same compound is produced when arsenic is heated with iodine in the proper proportions at 230 in a sealed tube. It crystallises from carbon bisulphide in cherry-red, brittle prisms, and easily oxidises in the air. When heated with water or alkalis it turns black, arsenic and arsenic tri-iodide being formed. 1 ARSENIC TRI-IODIDE, AsI 3 = 452-1. When arsenic and iodine are brought together they com- bine with considerable evolution of heat. For the purpose of preparing the tri-iodide a method is adopted similar to that em- ployed for the preparation of the tribromide. It is obtained in the form of bright red hexagonal tables, which have a specific gravity of 4 '39. It may also be prepared by passing hydriodic acid into arsenic trichloride, when hydrochloric acid is evolved, and the .iodide separates out in the form of crystals, or by adding a hot solution of arsenious oxide in hydrochloric acid to a con- centrated solution of potassium iodide. 2 When it is heated in the air or in oxygen it burns with forma- tion of arsenious oxide. When ammonia is passed through its solution in ether or benzene, a compound of the formula 2AsI 3 + 9NH 3 is formed. On heating with alcohol to 150, ethyl iodide, C 2 H 5 I, is formed. 3 It is used in medicine as a remedy for certain skin diseases. Arsenic Penta-iodide, AsI 5 , has been prepared 4 by heating the 1 Bamberger and Philipp, Ber. 14, 2643. 2 Ibid. 3 Ibid. 4 Sloan, Chcm. News, 46, 194. 622 THE NON-METALLIC ELEMENTS tri-iodide with iodine at 150. It is a brown crystalline mass which melts at 70, has a sp. gr. of 3 93 and is soluble in water and alcohol. The solutions deposit the tri-iodide on standing. OXIDES AND OXYACIDS OF ARSENIC. 369 Arsenic unites with oxygen in two proportions producing two acid-forming oxides, the composition of which corresponds with the oxides of phosphorus : viz. Arsenious Oxide, As 4 O 6 . Arsenic Pentoxide, As. 2 O 5 . ARSENIOUS OXIDE, on ARSENIC TRIOXIDE, As 4 O 6 = 3611. Arsenious oxide has long been known under the names of white arsenic and arsenious acid. In the writings of Basil Valentine the name Htittenrauch, or furnace-smoke, is given to this substance because it is obtained by roasting arsenical pyrites, and is emitted during the process in the form of a white smoke which condenses to a white powder. Arsenious oxide is prepared on the large scale in many metallurgical processes by the roasting of arsenical ores. The vapours of the oxide which are given off are condensed in long passages or chambers called poison chambers (Giftkanale), or in towers termed poison-towers (Giftthiirme) in the form of crude flowers of arsenic or poison-flour (Giftmehl). For the pre- paration of white arsenic, arsenical pyrites are usually employed, It is obtained as a by-product in the roasting of cobalt ores, which are employed in the manufacture of smalt. The crude sublimate obtained in the Freiberg works contains about 75 per cent, of arsenious oxide. Open roasters are now supplanting the muffle furnaces in which the operations were formerly con- ducted. The hearth of such a furnace is about four metres in length and about 2'8 metres in breadth, and in it 900 kilos, of the ore can be roasted at once ; four charges are made during the day, and the white powder which comes off collects in long underground passages of some 200 metres in length. Large quantities of white arsenic are manufactured in the Harz, and also in Devonshire and Cornwall. At the Great Devon Consols and other mines in England 400 tons are manufactured monthly by roasting tin-ore containing arsenical pyrites, and also the grey ARSENIOUS OXIDE 623 flue deposits of the tin mines. In the year 1872 in England no less than 5,171 tons of white arsenic were made. 1 Of late years Oxland's self-acting calciner has been much used for the manufacture of arsenious oxide from the Cornish and Devonshire ores. This furnace consists of an iron tube, from three to six feet in diameter, and thirty feet long, set at an inclination of from half to one inch per foot, varying according to the nature of the ore. This tube is heated by a fire placed at its lower end, whilst at its upper it is placed in connection with the flues in which the white arsenic is deposited. The tube is made to revolve by suitable machinery at the rate of about one revolution in four minutes, and the crushed ore is admitted in a regular stream through a feed-pipe at the back end of the tube. Great economy of fuel is effected by this furnace ; indeed, if properly worked with a good ore, the heat of combustion of the sulphur and arsenic is itself sufficient to carry on the process. One such cylinder turns out upwards of twenty- five tons of ore per diem, and the calcined product contains less than 0'5 per cent, of arsenic. A part of the arsenious oxide comes into the market in the form of a white crystalline powder, the rest in the form of arsenic glass, or amorphous arsenic obtained from the powder by resublimation. For this purpose the arsenic powder which is not white enough to be sent into the market is in the Ger- man manufactories placed in iron pots heated by a furnace and cylinders placed over them, in which the arsenious oxide con- denses in the vitreous form. The pots hold 4J cwt. of the crude arsenious oxide, and this is sublimed in from ten to twelve hours. In order to produce a pure product two sublima- tions are necessary. In this operation care must be taken that a reduction to the state of metallic arsenic does not occur, as this not only colours the arsenic-glass of a dark tint, but is apt to form a fusible alloy with the iron of the pots, and thus to destroy them ; if this occurs the oxide then falls into the furnace and escapes into the air, and this is a continual source of danger to the workmen employed in the operation. In England a common reverberatory furnace is used for the resublimation of the crude white arsenic, but, to prevent dis- coloration by smoke, either coke or anthracite is used (Phillips). Properties. Arsenious oxide possesses no smell, but has a weak metallic sweetish taste ; it forms a colourless and odourless vapour 1 Phillips' Metallurgy, p. 463. 624 THE NON-METALLIC ELEMENTS which has a density of 19 7' 7 at temperatures varying from 570 (Mitscherlich) to 1560 (V. and C. Meyer). 1 Hence its molecular formula is As 4 O 6 . The vitreous modification of arsenious oxide is translucent or transparent, and perfectly amorphous. Its specific gravity is 3*738. On heating, it melts without volatilization at a temperature of about 200. When kept for any length of time it becomes opaque, being changed into a porcelain-like mass. This change is due to the passage from the amorphous or vitreous to the crystalline condition ; the change commences at the outside of the mass, and gradually penetrates into the interior. Arsenious oxide is slightly soluble in water. It is deposited on cooling from a hot saturated solution in transparent regular octahedra. It is very much more soluble in hydrochloric acid than in water, and it may be easily obtained from the solution in the form of large crystals. It also occurs in this form as arsenic-bloom, being founco, I write, indicates its use. In the year 1779 Scheele showed that molybdenum-glance is totally different from graphite, and that this latter body when treated with nitric acid is converted into carbonic acid, so that it must be looked upon as a kind of mineral carbon. Graphite occurs in nature tolerably widely distributed. It is usually found in lumps or nodules in granite, gneiss, and other crystalline rocks. The best graphite for the purpose of 1 " The pencil represented below, is made, for writing, of a certain kind of lead (which I am told is an artificial substance termed by some, English antimony,) sharpened to a point and inserted in a wooden handle." 668 THE NON-METALLIC ELEMENTS making black-lead pencils, was that formerly found exclusively at Borrowdale in Cumberland, in green slate. These mines are however, now almost exhausted, not having been worked for the last forty years. In the sixteenth and seventeenth centuries they were so productive as to yield an annual revenue of 40,000, although they were only worked a few weeks in the year for fear of exhausting the mine. Graphite is also found at Passau in Germany, in Bohemia and in Styria. It likewise occurs in many places in the United States, the deposits at Sturbridge, Mass., and at several localities in New York being large enough to yield a considerable supply. By far the largest mine in the United States is the " Eureka Black-Lead Mine " at Sonora in California. The graphite here forms a layer of some twenty to thirty feet in thickness. It is so pure that it may be obtained in large blocks. In the year 1868 not less than one million of kilos, were raised each month. 1 Large quantities of graphite are also found in Ceylon. In Southern Siberia this substance occurs in considerable quantities in the Batougal mountains, and is largely exported to Europe. Graphite commonly occurs in compact foliated or granular masses, but occasionally in small six-sided tables, which accord- ing to Kenngott belong to the hexagonal, but according to Nordenskjold to the monosymmetric system. It has a steel-gray colour, an unctuous touch, and it is so soft that it gives a black streak on paper. Its specific gravity varies from 2'015 to 2'583 r and this considerable variation is due to the fact that almost all natural graphite contains more or less impurity which, when the graphite is burnt, remains behind as ash and consists of alumina, silica, and ferric oxide, with small traces of lime and magnesia. 2 It usually contains 0'5 to T3 per cent, of hydrogen, 3 a fact which seems to point to its organic origin. Graphite is a good conductor of heat and electricity, whilst the diamond is a non-conductor. The mtumescent graphite obtained by Moissan commences to burn in oxygen at 575, but the temperature of ignition varies for the different specimens of natural graphite. The specific heat at ordinary temperatures is 0-202 (Regnault), whilst at 978 it is 0'467 (Weber). The artificial production of graphite by melting cast iron containing a large proportion of carbon, and allowing it to cool 1 Chem. News, 1868, 299. 2 Mene, Compt. Rend. 64, 1019 and 1867. 3 Regnault, Ann. Chim. Phys. [2], 1, 202. GRAPHITE 669 slowly, was first observed by Scheele in 1778. Cast iron contains a certain amount of carbon combined with iron; in the molten condition, however, it can dissolve a much larger quantity of carbon, up to as much as 4 per cent, of its weight. This excess crystallizes out as graphite when the metal cools. The coarsely crystalline gray pig-iron owes its peculiar pro- perties as well as its appearance to the presence of graphite, and when this form of iron is dissolved in acid, scales of graphite remain as an insoluble residue. Graphite also occurs in many meteoric masses, as, for instance, in the meteorite which fell in 1861 at Cranbourne near Melbourne, and this meteoric graphite is, according to Berthelot, identical in properties with iron-graphite. We may thus conclude that the meteoric mass in which it has been found has been exposed to a very high temperature. According to Wagner, the black deposit which is formed by the spontaneous decomposition of hydrocyanic acid, CNH, contains graphite, as may be seen by washing the deposit with strong nitric acid, when the insoluble scales of graphite are left behind. 1 Another remarkable mode of production of graphite, most likely from cyanogen compounds, was first observed by Pauli 2 in the manufacture of caustic soda from the black-ash liquors. These liquors are evaporated to a certain degree of consistency, and Chili saltpetre is added to oxidise the sulphur and cyanogen compounds present. Torrents of ammonia are thus evolved, and a black scum of graphite is observed to rise to the surface. When the vapour of chloride of carbon is led over melted cast iron, ferric chloride is evolved, and carbon dissolves in the iron until it becomes saturated, after which hexagonal plates of graphite separate out. 3 The carbon which occurs in crystalline boron remains as amorphous carbon when the diamond boron is heated to red- ness. When it is heated to whiteness it takes the form of graphite. Both diamond and amorphous carbon are converted into graphite at the temperature of the electric arc, and when the arc is maintained between two poles of amorphous carbon the negative pole is found to be largely converted into graphite. Many specimens of graphite possess the remarkable property 1 Wagner's Jahresb. 1869, p. 230. 2 Phil. Mag. [4], 21, 541. 3 H. Deville, Ann. Ghim, Phys. [3], 49, 72. 670 THE NON-METALLIC ELEMENTS of swelling up and leaving a very voluminous residue of finely- divided graphite, in the same manner as mercuric sulphocyanide (Pharaoh's serpents) when moistened with strong nitric acid and then strongly heated. This property, however, is not common to all specimens of graphite, and it has even been proposed to divide the known graphites into two classes, those which give this nitric acid reaction being recognized as graphite, whilst those which do not, such as the graphite of the electric arc and that obtained from cast iron, are distinguished as graphitite. 1 The intumescent variety may be prepared by fusing platinum in a carbon crucible in an electric furnace. When the ingot is treated with aqua regia a residue of graphite of this kind is left. It is also found in the interior of an ingot of cast iron which has been cooled by water, whilst the external layers only contain the non-intumescing variety. When intumescence takes place oxides of nitrogen and a little carbon dioxide are given off and a residue of pure graphite left, which is scarcely altered by further treatment with nitric acid. The intume- scence is, therefore, probably due to the sudden evolution of gas produced by the action of traces of nitric acid remaining in the graphite upon a little amorphous carbon entangled among the plates of graphite. 2 Graphite in its chemical relations occupies a position totally distinct from that of all other forms of carbon. When graphite is repeatedly treated with fuming nitric acid and potassium chlorate it is converted into a compound which contains both hydrogen and oxygen, and is known as graphitic acid or graphitic oxide (p. 729), whereas under the same treat- ment diamond is unaltered and amorphous carbon completely dissolved. This reaction, as has been pointed out, provides a means of distinguishing graphite from the other allotropic forms of carbon, and is the basis of Berthelot's method for estimating the proportions of the three varieties in a mixture. When the electric arc is maintained between carbon poles, it is observed that, however powerful the current, a maximum degree of brightness is attained. This seems to indicate that the carbon of the pole volatilises at the temperature corre- sponding to this brightness. By heating a portion of the pole to this maximum temperature and then shaking it off into a 1 Luzi, Ber. 24, 4085 ; 25, 216, 1378 ; 26, 890. 2 Moissan, Compt. Rend. H6, 608. GRAPHITE 671 calorimeter it has been found that the temperature of volatili- sation of carbon is about 3,500, and this is, therefore, the maximum temperature which can be attained in an electric furnace in which carbon poles are employed. 1 396 As already mentioned, many specimens of graphite in- tumesce when moistened with nitric acid and strongly heated. It is also found that when finely-powdered graphite is heated with a mixture of one part of nitric and four parts of strong sulphuric acid, or when a mixture of fourteen parts of graphite and one part of potassium chlorate is warmed with twenty-eight parts of strong sulphuric acid, the graphite assumes a purple tint, but on subsequent washing returns to its original colour. It contains now oxygen, hydrogen, and sulphuric acid, and when it is heated to redness swells up with a copious evolution of gas, and then falls to an extremely finely-divided powder of pure graphite, which has a specific gravity of 2'25. 2 This process is employed for the purpose of purifying natural gra- phite. With this object it is first ground, and the powder well washed in long troughs in order to remove as much as possible of the earthy matrix with which it is mixed. The graphite thus obtained is pure enough for many uses ; but if it is required in the pure state, the powder must be treated with potassium chlorate and sulphuric acid as above described. The fine powder is then thrown upon water, on the surface of which it swims, whilst the earthy matters sink to the bottom. The foliated graphite answers best for this purpose, amorphous graphite being more difficult to purify. This variety may, however, also be rendered pure if a small quantity of fluoride of sodium be added to the mixture as soon as the evolution of chlorine and its oxides has ceased, the object of this addition being to remove the silica as silicon tetrafluoride. Graphite is employed for a great variety of purposes. The Cumberland black-lead pencils, the first of their kind, were originally manufactured by cutting slips of graphite out of the solid block. Experiments were afterwards made for the purpose of making use of the graphite powder, which was fused with sulphur or antimony. The pencils thus prepared were, however, hard and gritty. A remarkable improvement in the manufac- ture, used up to the present day, was introduced by Comte. The powdered graphite is mixed with carefully-washed clay, and the 1 Violle, Cmn.pt. Rend. 115, 1273. 2 Brodie, Ann. Chim. Phys. [3], 45, 351. 672 THE NON-METALLIC ELEMENTS mixture placed in short iron cylinders having an opening at the bottom. The semi-solid mass is then pressed through the hole and assumes the form of a fine thread, which can be cut up into the required lengths and used for pencil-making. Another important application of graphite is for the prepara- tion of the black-lead crucibles largely used in metallurigical operations, especially in the manufacture of cast steel. The Patent Plumbago Crucible Company at Battersea employ Ceylon graphite for the manufacture of their crucibles, whilst the graphite employed by Krupp of Essen in the manufacture of the cruci- bles in which his celebrated cast steel is melted is obtained from Bohemia. The finely-ground graphite is well mixed with Stourbridge fire-clay and water, so as to obtain a homogeneous mass ; the water is then pressed out and the mass formed into blocks, which have to lie for many weeks. The plasticity of the mass is thus much increased. The crucible is next moulded by hand on a potter's wheel, or sometimes formed in a mould ; it is then slowly dried, and afterwards ignited in a pottery furnace, being placed in saggers in order to prevent the com- bustion of the graphite. These crucibles are good conductors of heat, they do not crack readily on change of temperature, and they likewise possess a clean surface, so that the metal can be poured out completely. Finely-divided graphite purified according to Brodie's process is largely used for polishing gunpowder, especially the large- grain, blasting, and heavy ordnance powder. This coating of black lead gives a varnish to the corn and prevents it from absorbing moisture. The explosive force of the powder is, however, somewhat diminished by this coating of graphite, for Abel has shown that the explosive force of unpolished powder being 107'6, the same powder polished with common graphite has an explosive force of 89 '9, and when varnished with Brodie's graphite of 99*7. The particles of the finely-divided and purified graphite adhere together when they are brought into close contact, as by great pressure, and the mass thus obtained may be used for pencil-making and for other purposes. The pressed graphite is found to conduct electricity very much more readily than the ordinary graphite, and better than gas-coke. Thus, according to Matthiesen, its conducting power is eighteen times greater than that of natural graphite, and twenty-nine times as great as that of the dense coke used for the Bunsen's batteries. From this AMORPHOUS CARBON 673 property, graphite powder is used largely in electro typing, the moulds upon which it is desired to deposit the metal being covered with a fine coating of powdered graphite. This acts as a conductor of electricity, and a uniform coating of the metal is deposited upon it. Graphite is used not only for the purpose of preventing the rusting of iron objects, but in some cases also instead of oil for lessening the friction in running machinery. 397 Amorphous Carbon. In early ages the attention of chemists was attracted to charcoal from the fact that it is a body which cannot be acted upon by any solvent. The supporters of the phlogistic theory made this substance a special study, because they believed that it contained more phlogiston than any other known body. This modification of carbon is produced when substances which contain that element are strongly heated in the absence of air. The amorphous carbon thus obtained has received various names which indicate its origin or mode of production, but the different varieties are identical in chemical properties. The chief forms which are thus dis- tinguished are (1) lamp-black, (2) gas-carbon, (3) charcoal, (4) animal charcoal and (5) coke. The last three of these usually contain mineral matter which was originally present in the wood, bone or coal from which they are derived. The sub- stances known by these names also generally contain small amounts of both hydrogen and oxygen. (1) Lamp-black. The luminosity of flame is probably due to the presence in it of intensely heated particles of carbon, which are deposited as soot on any cold surface held in the flame. When the burning substance is rich in carbon, the flame smokes even without its being cooled, and it does so the more strongly the smaller the supply of air. This fact is made use of for the purpose of preparing finely-divided amorphous carbon or lamp- black. In the manufacture of lamp-black, tar, resin, turpentine or petroleum is burnt in a supply of air insufficient to burn it completely, the smoky products of this imperfect combustion being allowed to pass into large chambers hung with coarse cloths, on which the lamp-black is deposited. The finest kind of lamp-black is obtained by suspending metallic plates over oil-lamps, or revolving over them metallic cylinders, on which the soot is deposited. It is purified by heating it in closed vessels, and is used for preparing Indian ink and in calico- printing for producing gray shades ; while common lamp-black 44 674 THE NON-METALLIC ELEMENTS is employed as a black paint and for manufacturing printer's ink. Soot or lamp-black is, however, not pure carbon, but contains about 80 per cent, of that element along with oily and fatty matters and a small amount of mineral matter, for it always contains appreciable quantities of hydrocarbons arising from the incomplete combustion of the tar. In order to remove these impurities it is not sufficient to ignite the lamp- black strongly. It must be heated to redness in a current of chlorine for a considerable length of time, the hydrogen then combining with the chlorine, and the carbon remaining unacted upon. (2) Gas carbon, which, next to lamp-black, is the purest form of amorphous carbon, is formed in the preparation of coal-gas, and probably owes its origin to the decomposition of the gaseous compounds of hydrogen and carbon, which are evolved, by the intensely heated walls of the retort. It is found as a deposit in the upper portion of the retort in the form of an iron-gray mass, so hard that it strikes fire like a flint. The portions of this carbon deposited on the sides of the retort contain no hydrogen, and have a specific gravity of 2*356, whereas those lying further from the surface of the retort contain some hydrogen. Gas carbon is also obtained by passing olefiant gas, C 2 H 4 (which is one of the chief constituents of coal-gas), through a red hot porcelain tube. This form of carbon conducts heat and electricity well, and is used for the preparation of the carbon cylinders or plates employed in Bunsen's battery, and the carbon-poles for the electric light. (3) Charcoal. This substance is obtained in the pure state by heating pure white sugar in a platinum basin. The carbon thus obtained is purified by ignition in a current of pure chlorine. It is tasteless and possesses no smell. It is a good conductor of electricity, and has a specific gravity of T57, Like the other modifications of carbon it is infusible, and is in- soluble in every solvent. Pure sugar-charcoal is used as a reducing agent, especially in the preparation of volatile metallic chlorides, and it is peculiarly valuable inasmuch as its freedom from silica prevents the formation of volatile tetrachloride of silicon. Amorphous carbon is converted by many oxidising agents into complex soluble compounds, among which the most important are oxalic acid, C 2 H 2 O 4 , and mellitic acid, C 12 H 6 O 12 CHARCOAL 675 (Vol. III. Part v. p. 374). Both of these acids are formed by the action of alkaline potassium permanganate solution on amor- phous carbon, 1 whilst mellitic acid is also formed by the oxida- tion of wood charcoal with concentrated sulphuric acid, 2 carbon dioxide, and sulphur dioxide being evolved at the same time (p. 367), and when a current of electricity is passed through a solution of caustic potash between carbon poles. 3 The aluminium salt of mellitic acid, known as honey-stone, occurs in seams of brown coal and is possibly formed by oxidation from that material. The specific heat of amorphous carbon (wood charcoal) is 0-241. On the large scale, wood charcoal is prepared in the same way from wood as coke is from coal. Charcoal-burning is a very old process, and the simple methods which were originally adopted are carried on up to the present day. The method consists in allowing heaps of wood covered with earth or sods to burn slowly with an insufficient supply of air. It is usual to build up large conical heaps with billets of wood placed verti- cally and covered over with turf or moistened soil, apertures being left at the bottom for the ingress of air. and a space in the middle serving as a flue to carry off the gases (Fig. 186). The pile is lighted at the bottom, and the combustion proceeds gradually to the top. Much care is needed in regulating the supply of air, and the consequent rate of the combustion. One hundred parts of wood thus treated yield, on an average, sixty-one to sixty-five parts by measure, or twenty-five parts by weight, of charcoal. In Austria and Sweden charcoal is made from long logs of fir-wood, which are placed horizontally in a rect- angular pile (Fig. 187). In countries like our own where wood is scarce, charcoal is obtained from small wood or sawdust by a more modern process. It consists in the carbonisation of the wood in cast-iron retorts, and in this process not only is charcoal obtained, but the volatile products, especially wood spirit, and pyroligneous acid, as well as tar, are collected. Good charcoal possesses a pure black colour and a bright glittering fracture. When struck with a hard object it emits a sonorous tone, and it burns without smoke or flame. The specific gravity of charcoal, when the pores are filled with 1 Schnlze, Ber. 4, 802, 806. 2 Verneuil, Compt. Rend. 118, 195. 3 Bartoli and Papasogli, Gazetta, J3, 37. 676 THE NON-METALLIC ELEMENTS air, varies between 0*106 (ash charcoal) and 0'203 (birch char- coal). Such charcoal will swim on water, but it sinks when the pores which were filled with air become filled with liquid. Charcoal rapidly absorbs gases and vapours (Saussure), and FIG. 186. possesses the remarkable property of precipitating certain sub- stances from solution, and absorbing them in its pores (Lowitz, 1790). Animal charcoal possesses this absorptive power in a much higher degree than common charcoal. FIG. 187. The properties and chemical composition of charcoal vary much according to the temperature to which the wood is heated. According to Percy, 1 wood becomes perceptibly brown at 220, whilst at 280 it becomes after a time a deep brown- 1 Metallurgy, Fuel, p. 107. CHARCOAL 677 black, and at 310 it is resolved into an easily pulverisable black mass. Charcoal made at 300 is brown, soft, and friable, taking fire easily when heated to 380 ; whilst that prepared at a high temperature is a hard, brittle substance, which does not take fire till it is heated to about 700. The proportion of carbon contained in charcoal prepared at different temperatures varies considerably. That prepared at the lowest point contains much more of the volatile constituents of the original wood than the charcoal made at a red-heat, though, as a matter of course, the yield of charcoal is greater at the low temperature. According to Violette 1 100 parts of buckthorn wood yield the following amounts of charcoal : At 250 50 parts of charcoal containing 65 per cent, carbon 300 33 73 400 20 80 1500 15 96 Thus charcoal, like lamp-black, as ordinarily prepared, never entirely consists of pure carbon (Davy). The following table gives the composition of charcoal obtained at different temperatures. 270 350 432 1023 1100 1250 1300 1500 Over 1500 Carbon . . . 70-45 76-64 81-64 81-97 83-29 88-14 90-81 94-57 96-51 Hydrogen . . 4-64 4-14 1-96 2-30 1-70 1-41 1-58 074 0-62 Oxygen with ) some nitro- > 24-06 18-61 15-24 1413 13-79 9-25 6-46 3-84 0'93 gen . . . ] Ash ... 0-85 0-61 1-16 1-60 1-22 1-20 1-15 0-66 1-94 100-00 100-00 100-00 100-00 10000 ioo-oo|ioo-oo 100 00 100-00 I The following results were obtained by Faisst. 2 Beech charcoal. Carbon 85'89 Hydrogen 2'41 Oxygen and nitrogen . 1*45 Ash 3-02 Water 7'23 Hard charcoal, Light charcoal, made in iron from wood- cylinders, gas-works. 85-18 87'43 2-88 2-26 3-44 0-54 2-46 1-56 6-04 8-21 100-00 100-00 100-00 Ann. Chim. Phys. [3], 32, 305. 2 Wagner, Jahresb. 1855, p. 457. 678 THE NON-METALLIC ELEMENTS (4) Animal Charcoal, which is obtained by the carbonisation of animal materials, differs from that prepared from vegetable sources, inasmuch as it contains considerable quantities of nitro- gen. Bone-Hack is obtained by charring bones in iron cylinders or retorts, and is formed as a by-product in the manufacture of animal, or Dippel's oil (Oleum animale Dippelii). Dry bones contain about 70 per cent, of inorganic material, consisting chiefly of calcium phosphate. This inorganic matter remains behind with the charcoal when the bones are charred, and thus the charcoal is deposited in an extremely finely-divided con- dition. It is probably for this reason that bone-black possesses a greater absorptive and decolourising power than wood charcoal. The presence of the phosphate of lime, however, prevents the application of this decolourising power of the charcoal to cases of acid liquids which dissolve the phosphate. In such cases it is necessary to make use of blood-charcoal. This is obtained by evaporating four parts of fresh blood with one part of carbonate of potash, and heating the residue in a cylinder. The charred residue is then boiled out with water and hydrochloric acid, and again heated to redness in a closed vessel. The addition of potash serves the purpose of making the charcoal as porous as possible. The following is an analysis of a good sample of dried animal charcoal. 1 Carbon 10'51 Calcium and magnesium phosphates, calcium fluoride, &c 80-21 Calcium carbonate 8'30 Other mineral matter . 0'98 100-00 398 The power possessed by wood charcoal and, in a much higher degree, by animal charcoal, of withdrawing many sub- stances from their solution, seems to be almost entirely of a physical nature. Thus, if a solution of iodine in iodide of potas- sium be shaken up with finely-divided charcoal, the iodine is completely withdrawn from solution. In the same way certain metallic salts, especially basic salts, are decomposed when their solutions are filtered through charcoal. This power of withdrawing bodies from solution is exerted most effectually upon colouring 1 Thorpe's Dictionary of Applied Chemistry, Vol. I. 171. ANIMAL CHARCOAL 679 matters and astringent principles. Thus, if a quantity of red wine (claret or port), or a solution of indigo in sulphuric acid, be shaken up and gently warmed with a quantity of freshly-ignited bone charcoal and the mixture filtered, the liquid which comes through is colourless. In the same way, if alcohol containing fusel oil be shaken up with animal charcoal, the characteristic smell of the fusel oil disappears. Wood char- coal is largely used for the latter purpose, whereas animal charcoal is employed for decolourising the juice of raw sugar. Animal charcoal is employed in a similar way in the purifica- tion of many organic compounds, although it not unfrequently happens that the compounds themselves are attracted by the porous charcoal. Occasionally this property is made use of for the separation of compounds which thus adhere to the charcoal, as, for instance, the alkaloids, from other bodies, these compounds being afterwards dissolved from the charcoal by the addition of hot alcohol. Like many other porous bodies both of these forms of charcoal possess the property of largely absorbing gaseous bodies. This power depends upon the fact that all gases condense in greater or less degree on to the surface of solid bodies with which they come in contact, and as charcoal is very porous, or possesses a very large surface to a given mass, its absorbent power is proportionately great. Char- coal, when exposed to the air, condenses large quantities of the latter upon its surface. This may be easily shown by attach- ing a piece of metal to it and sinking the mass in a cylinder filled with water. If the cylinder be now placed under the receiver of an air-pump and the air exhausted, a rapid stream of bubbles will be seen to rise in brisk effervescence from the charcoal. The remarkable absorptive power of charcoal for certain gases is also well illustrated by inserting a stick of recently calcined charcoal into a tube filled over mercury with dry ammonia gas. The gas is so quickly absorbed by the charcoal that the tube soon becomes filled with mercury. Another mode of showing a similar absorption of sulphuretted hydrogen gas is to plunge a small crucible filled with freshly- ignited and nearly cold powdered charcoal into a jar of sul- phuretted hydrogen. This gas is then absorbed by the charcoal in such quantity that if it be removed when saturated and plunged into a jar of oxygen the charcoal will burst into vivid combustion, owing to combination occurring between the ab- sorbed sulphuretted hydrogen and oxygen gases. 680 THE NON-METALLIC ELEMENTS The absorptive power of wood charcoal for gases was first investigated by Saussure. In his experiments he made use of beech-wood charcoal which had been recently heated to redness and then cooled under mercury in order to remove the air from its pores. The following numbers were obtained by him : 1 volume of charcoal absorbs at 12 and under 724 mm. the following (Saussure) : Vols. Vols. Ammonia 90 Ethylene .... 35 Hydrochloric acid . . 85 Carbon monoxide . 9'42 Sulphur dioxide ... 65 Oxygen ..... 9'25 Sulphuretted hydrogen 55 Nitrogen .... 6*50 Nitrogen monoxide . . 40 Hydrogen .... 1*25 Carbon dioxide ... 35 Hunter, 1 who also made experiments on the same subject, found that one volume of charcoal absorbed the following quantities of gas at the temperature and under a pressure of 760 mm. Absorption of gases by charcoal (Hunter). Vols. Vols. Ammonia .... 171'7 Phosphine .... 691 Cyanogen .... 107*5 Carbon dioxide . . 677 Nitrous oxide . . . 86'3 Carbon monoxide . . 21*2 Methyl chloride . . 76'4 Oxygen 17 '9 Methyl ether . . . 76'2 Nitrogen 15'2 Ethene 74'7 Hydrogen .... 4'4 Nitric oxide . . . 70*5 The differences observed between these two series of experi- ments with the same gases are to be explained by the fact that the charcoals employed were not of equal porosity. Both series, however, show that the more readily the gas is condensible the more it is absorbed by charcoal, which seems to show that the gases condensed by charcoal undergo at any rate a partial lique- faction. This view is rendered possible by the observation of Melsen, 2 that when dry hydrogen is brought into contact with charcoal saturated with chlorine a considerable quantity of 1 Phil. Mag. [4], 25, 364 ; 29, 116. 2 Compt. Rend., 76, 81, 92. ABSORPTIVE POWER OF CHARCOAL 681 hydrochloric acid is formed, even when the experiment is carried on in complete darkness ; and that when charcoal saturated with chlorine is brought into a Faraday's tube and the other limb placed in a freezing mixture, liquid chlorine is obtained. In a similar way ammonia, cyanogen, sulphur dioxide, sulphuretted hydrogen, and hydrobromic acid have been liquefied. Wood charcoal, like bone charcoal, has the power of absorbing the unpleasant effluvia evolved in the processes of decay and putrefaction as well as the moisture from the air. Stenhouse, 1 who has investigated this subject, has shown that charcoal not only absorbs these gases and effluvia, but has the power, espe- cially in contact with air, of oxidising and destroying them, inasmuch as when absorbed by charcoal these substances are brought into such close contact with the atmospheric oxygen, which is also absorbed by the charcoal, that a rapid oxidation is set up, and the odoriferous products of decomposition are instantly resolved into carbon dioxide and water, and other simple compounds. This property is retained by the charcoal for a long time, and when it has been lost it can be renewed by ignition. Hence charcoal niters are largely used for preventing the foul sewer gases from polluting the air of the streets and houses, and charcoal respirators and ventilators have been pro- posed by Stenhouse as protections against the ingress of dele- terious gases into the lungs. For the same reason, trays filled with heated wood charcoal, placed in the wards of hospitals or other infected apartments, have proved very effective in absorb- ing noxious emanations. Charcoal filters are also largely em- ployed for filtering water for drinking purposes, as in its passage through the charcoal the water is decidedly improved in quality, not only organic and soluble colouring matters being removed as well as all suspended matter, but the water under- going aeration. Such filters do not, however, necessarily free the water from bacterial life ; but may, on the other hand, increase its amount unless the filter is frequently sterilized. (See ante, p. 301.) (5) Coke. This substance remains behind when bituminous coal is heated to redness in absence of air. Coke is obtained as a by-product in the manufacture of coal-gas, but it is also specially manufactured in coke ovens, and sometimes by burn- 1 On Charcoal as a Disinfectant ; Proc. Hoy. Inst. 2, 53 ; also Pharm. Journ. 16, 363 682 THE NON-METALLIC ELEMENTS ing coal in heaps and stopping the combustion at a certain stage by quenching with water. When prepared by heating in covered ovens or kilns, the coke is harder, more lustrous, and less combustible than that obtained by burning in heaps. It is termed hard-coke or engine-coke, and is largely used for iron smelting, whereas the other variety is termed soft-coke or blacksmith's coke, and is the more combustible. Coke takes fire at a much higher temperature than common coal, and, when burning, gives rise to a very high temperature, but without the elimination of smoke, as it consists almost entirely of carbon. Coke not only contains the inorganic material, or ash, present in the coal from which it is manufactured, but in addition small quantities of hydrogen, oxygen, and nitrogen, as the following analyses show : COKE ANALYSES. First Sample. Second Sample. Carbon 91'30 91'59 Hydrogen O33 0'47 Nitrogen and oxygen . 217 2*05 Ash 6-20 5-89 The amount of coke produced in Great Britain during 1885 was 11,077,375 tons. 1 COAL. 399 When vegetable matter decays in absence of air and under water, or in the earth, it undergoes a change similar to that which occurs when it is heated. Water, carbon dioxide, and marsh gas are given off, and the residual material becomes richer in carbon. The fact of the occurrence of marsh gas and carbon dioxide as products of decomposition is rendered evident by their presence in a highly compressed condition in the coal measures at the present day, from which the former is evolved as fire-damp in enormous quantities which often produce fatal accidents. It is in this way that the coals of various kinds, lignites and peats, have been formed. Their composition com- pared with that of cellulose is shown in the following table, the 1 Thorpe's Dictionary of Applied Chemistry, vol. ii. 162. COAL 683 amount of mineral matter which is left behind as ash on com- bustion having been subtracted : Oxygen Carbon. , Hydrogen. and Nitrogen. Cellulose 50-00 6-00 44-00 Irish peat 60-02 5-88 34-10 Lignite from Cologne . . 66-96 5-25 2776 Earthy coal from Dax . . 74-20 5-89 19-90 Cannel coal from Wigan . 85-81 5-85 8-34 Newcastle Hartley . . . 88-42 5-61 5-97 Welsh anthracite . . . 94-05 3-38 2-57 The above table shows that coal is a less pure form of carbon than wood charcoal. It consists of the more or less altered remains of a vegetable world which once flourished at various points on the earth's surface. The plants of the coal formation consist of Calamites, the representatives of the living Equisetums, of Lepidodendra, arid SigillariaB, which were the principal forest- trees, and which, though of gigantic dimensions, were true cryptd- gams, represented by the living Lycopods and Selaginellse, and an important group of Conifers and Cycads, as yet not well understood, and chiefly known through their seeds, which are numerous and varied. Besides these arborescent types, ferns .formed an abundant undergrowth, many of these also having been representatives of the living tree-ferns. In the passage into coal the original woody fibre has not only undergone a loss of hydrogen and oxygen but it has at the same time become bituminised, so that for the most part all vegetable structure has disappeared, and the coal possesses a fatty lustre and coarse slaty fracture. There are many different kinds of coal containing more or less of the hydrogen, oxygen, and nitrogen of the original woody fibre. Cannel coal and boghead coal contain the most hydrogen, and anthracite contains the least, whilst the various kinds of bituminous coals lie between these extremes. Anthracite Coal of all coals contains the largest percentage of carbon, and has therefore, undergone the most complete change from woody fibre. It is found in the oldest deposits of the car- boniferous series, especially in Wales, Pennsylvania, and Rhode 684 THE NON-METALLIC ELEMENTS Island, and in smaller quantities in France, Saxony, and South- ern Russia. It has a conchoidal fracture, possesses a bright lustre, often sub-metallic, an iron-black colour, and is frequently iri- descent. It burns with a smokeless flame. Anthracite gradu- ally passes into bituminous coal, becoming less hard and containing more volatile matter. Its specific gravity varies from 1-26 to 1'8. The Bituminous Coals consist of a large number of varieties differing considerably from one another in their chemical com- position, as also in their products of decomposition by heat. They have the common property of burning with a smoky flame when placed in the fire, and yielding on distillation volatile hydrocarbons, tar or bitumen, whence their name is derived. The most important kinds of bituminous coals are (1) caking coal, which softens and becomes pasty or semi-solid in the fire, and yields, when completely decomposed, a greyish-black cellu- lar mass of coke ; (2) non-caking coal, agreeing with the last- named variety in all its external characters and even in its chemical composition, but burning freely without softening, and without any appearance of incipient fusion ; the residue which it yields is not a proper coke, being either in powder or in the form of the original coal. Cannel Coal, sometimes called parrot coal. This is a variety of coal differing from the preceding in texture, and yielding usually more volatile matters and being therefore specially employed for the purpose of gas-making. Cannel coal is more compact than bituminous, possesses little or no lustre, does not show any banded structure, breaks with a conchoidal fracture and smooth surface; and has a dull black or grayish-black colour. Its name is derived from the fact that small fragments, when lighted, will burn with flame, and hence it was termed candle- or cannel-coal. 400 The tables on pages 685 to 687, give examples of the composition of the different forms of coal. Connected with the true coals is a peculiar form, termed Boghead coal or Torbane Hill mineral, which was first found near Bathgate in Linlithgowshire, and has since been observed in other places, especially in New South Wales. This deposit occurs in the carboniferous formation, but it is not pro- perly speaking a coal, but belongs to the class of bituminous shales. Its specific gravity is lower than that of true coal, and COMPOSITION OF COALS 685 rH O . <*- 1 CO O 1 1 P ^ 5 CO ^ o I>- 1 iO 1 1 1:- l> - CC OS ^ 10 'CO 10 a 5 l>- X . OS CJ^ iO CO OS oo oo CO s G<| iO -^ (M CO i>- P 1 a rH OS O CO rH rH 1 o co 1 CO O 10 O Ci O CN CO O O 5 CO 10 O S ^ i>- G^ (M 00 X C CO rH (M Ci D CO iO 5 iO (M * rH - (M 0( D O (M r ( g OS rH rH O O CO Ir- 00 OS C 5 ^ CO 10 ^ 00 CO >O O O (M rH (M 1 1 0? .- rH rH t> pQ Q f g c oc i>- i>- ^ c f rH rH IO (M -"^ rH T P iO (M P g 00 'ft CO rH CJ O CO iO -* O u (M CO O ^ "S CO rH P rH (M o rH r-i rH r H rH i 1 rH C5 g d W & OS OS rH (Mr (M CO O - rH CO rH CO CO CO (M O X X iO 10 -^ rH 00 C D CO' OS CO co co ^ c, D CO ^ JH 00 O rH IO C, OS ir- (M co S 9 i z 3 S S rf - c8 a r/) ^S n * 13 o> C3 iJIj ^ ' QQ 2 .2" 2 5 C -> $ '1 ^ . 3 ^ bo &^ ^ ^ H P P S . O2 O -^ O2 CO G 1 >! l S P ^ .2 H W ^ 686 THE NON-METALLIC ELEMENTS a B O g (8 i- 1 OX C > r 1 c CO CO C CO CO C D CO *O ^O -* X o ^ i^ do O X i>- t^ LJ iO O CO 1 *O rH 1 CO 1 o 1 1 2 . CiOi^-COOCOiOOfi O 1C O Oi O *O "* X ^ (NOCMrHCOrHCM^G ^ CO (M O P 00 lOCO^r ((M^OrHiO O X i>- >O Ci .t X !> OOOrHOOOO 1 1 1 d O ~- iO O CO O p co p cp cp -^ (5^1 Cq O5 >O G CO rH I>- O 1>- X *O ^ r rH rH rH r 7> O H=( i^ - rH i- 5 X X H CO CO 5 s ^^iOCO^iOOCOw i O(NrHOiOCO- 1 ^O<5' CO^^CO^iO^Or- 1 D^ X CO i -^ 1>- O XXXJ>-I^XXXO( H CO CO CO D X X X > o of CO O P 1^ 1 1 1 <5- O O ^I>-^OXOOil>- *O CO r-H CO (N (5-rH-^CO rH 1>- ^O 00 G^l ^O ^* 00 O^ O^ CO rHO(3 ^^ G^ rH X CO 1 X Ci O5 C^l (N rH rH O O O rH a 1 XOOrnOCOX(M ^iOOrH^SrH H 1 rH ^ 6 C rH S w* iO^O^O0OXl:-X COCOCO(M. CO O5 X G^ ^1 rH .t^ rH -^ Oi CO CO ^-1 CO l CO CO ^ rH rH rH rH rH rH g $' ^ ^ 1 'I J OQ -S 3 -r if 5 i ^ ^ 8 ^ '6. S Q *S * ^r3 ? .5 s 5 i 2 , ^ ? -5 g 00 >-s L^ "^ O ^ O ~ QJ P^ 1 C H H H C C C H J, H i or CH 3 . CH 2 . CH 2 . CH 3 CH or CH 3 .CH. CH 3 and in fact two and only two compounds of this composition have been experimentally obtained. Substances of this kind, which have the same percentage composition but different physical and chemical properties, are known as isomerides. Such cases are occasionally met with in the compounds of the other ele- ments, as for example in the case of the potassium sodium sulphites (p. 373), but they are of far more frequent occurrence among the carbon compounds. Thus three different hydro- carbons having the formula C 5 H 12 are known, their formulae being (1) CH 3 .CH 2 .CH 2 .CH 2 .CH 3 (2) CH 3 .CH 2 .CH (3) CH, \ CH, CH 3 C CH and the number of possible isomerides rapidly increases with the rise in the number of carbon atoms, no less than 799 different isomerides having the composition C 13 H 26 being theoretically possible. All the hydrocarbons derived in this manner from marsh gas have the general formula C n H 2n+2 , and are usually termed the paraffins. CONSTITUTION OF HYDROCARBONS A different series of hydrocarbons, known as the olefmes, and having the general formula C n H 2n , contains two carbon atoms united together by two combining units, the simplest represen- tative being ethylene, C 2 H 4 , the constitution of which is represented by the formula HTT \ / H/ \H orCH 2 = CH 2 . The hydrogen atoms of this hydrocarbon may be replaced by other carbon atoms in exactly the same manner as described under methane, giving rise to new hydrocarbons, of which the following may be taken as examples : Propylene. o-Butylene. -Butylene. CH 3 .CH = CH 2 CH 3 .CH. 2 .CH = CH 2 CH 3 .CH = CH.CH 3 Another series, somewhat similar to the foregoing, having the general formula C n H 2n _ 9 , contains two carbon atoms united together by three combining units. The first member of this series, known as acetylene, has the constitution CH = CH. In a fourth series of hydrocarbons the carbon atoms are united together in such a manner that they form a closed chain. The most important member benzene, contains a closed chain of six carbon atoms, its constitution being represented as follows : H H C C H H C C H C H It will be observed that in this formula only three valencies of each carbon atom are represented as saturated ; the manner in which the remaining valencies are disposed is still a matter of discussion. 694 THE NON-METALLIC ELEMENTS Almost all the other compounds of carbon may be regarded as derived from one or other of these numerous hydrocarbons ; thus, for example, chloroform, CHC1 3 , is marsh gas in which three of the four hydrogen atoms have been replaced by chlorine, and alcohol, C. 2 H 6 O, is ethane, C 2 H 6 , in which one atom of hydrogen has been replaced by the compound radical hydroxyl : CH 3 .CH 3 CH 3 .CH 2 .OH Ethane. Alcohol. Most of the substances occurring in the animal and vegetable kingdoms belong to the group of carbon compounds, and in addition to these an immense number of other derivatives con- taining this element have been prepared artificially, so that the total number of carbon compounds known is greater than that of all the other elements put together. For this reason they are separately treated of under the head of Organic Chemistry, which is now usually defined as the Chemistry of the Hydrocarbons and their derivatives. In this portion of the work only the simpler hydrocarbons, containing one or two atoms of carbon, will be considered. METHANE, METHYL HYDRIDE, MARSH GAS, OR FIRE-DAMP. CH 4 = 15-91. 402 This gas is found in the free state in nature, and its occurrence was observed in early times. Thus Pliny mentions the combustible gaseous emanations which occur in several dis- tricts, and Basil Valentine remarks upon the outbreaks of flame which occur in mines, and which are preceded by a suffocating damp or vapour. He does not consider that this vapour is com- bustible, but rather believes that the flame was emitted by the rocks for the purpose of destroying this poisonous vapour. Marsh gas, like some other combustible gases, was not dis- tinguished from inflammable air, or hydrogen, until Volta in the year 1776 showed that inflammable air, when burnt, required only one-fourth of the volume of oxygen which was needed for the complete combustion of marsh gas, and that in this latter case alone was carbonic acid formed. In the year 1785 Ber- thollet proved that this gas contains both carbon and hydrogen, but it was at that time not distinguished from olefiant gas METHANE 695 (or ethene, C 2 H 4 ). In 1805 William Henry clearly pointed out the difference between these two gases. We have already seen that marsh gas occurs free in nature. It is evolved, together with other hydrocarbons, in large quantities in petroleum springs. The holy fire at Baku on the Caspian Sea, which has been burning from the earliest historical times, is due to marsh gas, mixed, according to Hess, with small quantities of nitrogen, carbon dioxide, and the vapours of petro- leum. The gas which is evolved from the mud volcanoes of Bulganak in the Crimea has been shown by Bunsen l to consist of pure marsh gas. The gases which escape in large quantities from the oil springs in Butler County, Pennsylvania, contain, according to Sadtler's analyses, 2 marsh gas and its homologues, together with hydrogen. These gases are collected and carried by pipes to the rolling mills at Pittsburg, a distance of fifteen miles, where the gas is employed as a fuel. Enormous quantities of marsh gas, or fire-damp, as it is termed by the miners, are evolved in coal-pits, due in all proba- bility to a slow decomposition of the coal. Reservoirs of this gas in a highly compressed state are often met with pent up in the crevices and cavities of the coal measures. Some beds of coal are so saturated with gas that when they are cut it may be heard oozing from every pore of the rock, and the coal is called by the colliers singing coal ; in other cases the gas escapes by what are termed blowers, and the mixture of gases frequently collects in the old workings or un ventilated portions of the pit. .Not unfrequently fire-damp bursts forth in large quantities from the seams of coal, or from the strata of clay which divide them. This is the frequent cause of the terrible accidents which sometimes, in spite of all care, will occur. The Lundhill colliery explosion in 1857 was one of the most calamitous on record. The sudden escape of gas from a blower in a neighbouring col- liery is thus described : " The fire-clay of the floor of the seam was seen to heave at different points along the face, and presently large fractures were made in it, through which gas was ejected with great violence, and with a sound very similar to the rushing of steam at a high pressure from a boiler. After the explosion at Lundhill, the pent-up gas still issued within the mine under such pressure as to support a column of water thirty feet 1 Gasometry, p. 147. 2 Amer. Philos. Soc. 1876 ; see also Laurence Smith, Ann. Chim. Phys. [5], 8, 566. 696 THE NON-METALLIC ELEMENTS high." The outburst of gas appears, sometimes at least, to be connected with a rapid fall in the barometer, the reduced atmospheric pressure enabling the gas to force its way out. 1 If he should escape from the effect of the explosion, the miner has still to fear its result, inasmuch as the gas in exploding renders ten times its own bulk of air unfit for respiration, and the after- damp or vitiated atmosphere produced by the explo- sion contains carbon dioxide sufficient to render it irrespirable. Hence the difficulty of descending into the pit after the explo- sion without proper precautions, or until a sufficient amount of ventilation has been re-established. The only satisfactory means of guarding against these sudden outbreaks in fiery pits is the establishment of a thorough and perfect system of ventilation, by means of which such an amount of air is brought into all parts of the workings as to render the formation of the in- flammable mixture difficult or impossible, even when a sudden outbreak of gas occurs. The escape of marsh gas at the surface of the ground in the neighbourhood of the coal measures is frequently observed. This was first noticed by Thomson, at Bedley, near Glasgow, where a flame, once lighted, burnt for many weeks in succession. A similar case has been noticed by Pauli near St. Helen's. The following analyses by Graham show the composition of fire-damp : 2 Sp. Gr. Marsh gas. Nitrogen. Oxygen. Five-quarter seam, j fl g 2 Gateshead colliery ) Bensham seam, ) n-^97 n-fi Hebburn colliery f Killingworth colliery (V6306 82'5 16'5 I'O Another instance of the formation of methane by the slow decomposition of vegetable matter, somewhat similar to that taking place in the coal seams, occurs in ponds or marshes, whence one of the names of the gas is derived. The gas- bubbles which rise when a stagnant pool containing decom- posing leaves and vegetable matter is stirred, consist essen- tially of marsh gas, which is mixed with carbon dioxide and nitrogen. The gas collected by Bunsen, in July 1848, from a pond in the botanical gardens of Marburg, contained, after the absorption of carbon dioxide by caustic potash, the following : Methane 48'5 Nitrogen 51*5 1 Scott and Galloway, Proc. Roy. Soc. 20, 292. 2 Phil. Mag. 28, 437. PREPARATION OF METHANE 697 Methane also invariably occurs amongst the products of the dry distillation of organic bodies, and hence it is present in very considerable quantities in coal gas. 403 Preparation. (1) In order to prepare marsh gas an intimate mixture of one part of dried acetate of soda with four parts of soda-lime (a mixture of caustic soda and lime) is heated. This is best accomplished in a tube of hard glass closed at one end, and fitted with a delivery-tube at the other, or, in place of this, an iron tube or a copper flask may be employed. In order to prepare the gas as pure as possible, the mixture must only be heated to the point at which the gas begins to be evolved, but even with all care it is impossible to avoid the presence of some free hydrogen and some ethylene. This latter impurity may, however, be removed by passing the gas through a U-tube containing pumice-stone soaked in strong sulphuric acid. In a sample of the gas thus prepared and purified, Kolbe l found 8 per cent, of hydrogen. The formation of methane from acetic acid is shown by the following equation : C 2 H 8 O 2 Na + NaOH = Na 2 CO 3 + CH 4 . (2) Chemically pure methane is obtained from zinc methyl, Zn(CH 3 ) 2 , which is decomposed by water as follows : 2 Zn(CH 3 ) 2 + 2H 2 = Zn(OH) 2 + 2CH 4 . (3) The pure gas is most readily obtained by the action of a zinc copper couple on a mixture of equal volumes of methyl iodide and alcohol, the reaction being assisted by gentle warming. 3 (4) Marsh gas can be obtained synthetically by passing a mixture of sulphuretted hydrogen and the vapour of carbon bisulphide over red-hot copper : 4 2SH 2 + CS 2 + 8Cu = CH 4 + 4Cu 2 S. (5) The same gas is likewise formed when a mixture of carbon monoxide and hydrogen is exposed to the action of the electric induction spark : 5 1 Ausfuhrl. Lehrb. Org. Chemie, i. 275. 2 Frankland, Phil. Trans. 1852 [2], 417. 3 Gladstone and Tribe, Journ. Chem. Soc. 1884, i. 154. 4 Berthelot, Compt. Rend. 43, 454. 6 Brodie, Proc. Roy. Soc. 21, 245. THE NON-METALLIC ELEMENTS Properties. Methane is a colourless odourless gas, which has a specific gravity of 0'559 ; it is very sparingly soluble in water, its solubility from 6 to 20 being obtained from the equation : c = 0-05449 - 0-001 1807* + 0-000010278* 2 . It dissolves more readily in alcohol, the solubility between 2 and 24 being found from the equation : c = 0-522586 -0-0028655^ + 0-0000142^. The gas was first liquefied by Cailletet 1 in 1877 ; the liquid boils at 155 to 156under the ordinary pressure, and at 73*5 under a pressure of 56'8 atmospheres, 2 and has at 164 a specific gravity of 0'415. 3 It burns with a slightly luminous flame, the illuminating power, when tested under the condi- tions usually observed in testing coal-gas, being equivalent to five candles as compared with an illuminating power of 15 20 candles given by ordinary coal gas. 4 When burnt in such a manner that the temperature of the flame is very high, as in a regenerative burner, the illuminating power is much greater. When mixed with twice its volume of oxygen it explodes in contact with a flame, the detonation being even more violent than in the case of a mixture of hydrogen and oxygen. Its heat of combustion per molecule in grams at 18 is 211,930 cal. Methane is an extremely stable compound and when exposed to a temperature at which hard glass softens is only de- composed to a very slight extent. When the sparks from a strong induction coil are passed through the gas it is partially dissociated into carbon and hydrogen, acetylene being also formed. ETHYL HYDRIDE OR ETHANE, C 2 H 6 = 29-82. 404 This gas is invariably present in the gaseous discharge accompanying petroleum in the oil springs of Pennsylvania, and is dissolved in considerable quantities in the liquid hydro- carbons. 5 Preparation. (1 ) Ethane is readily obtained by treating ethyl 1 Jahresb. 1877, 221. 2 Wroblewski, Jahresb. 1884, 197. 8 Olszewski, Jahresb. 1887, 72. 4 Wright, Journ. Chem. Soc. 1885, i. 200. 6 Ronalds, Journ. Chem. Soc. 1851, 54. ETHANE AND ETHYLENE 699 i iodide with zinc and water under pressure at a temperature of 150 ; x thus: Zn + C 2 H 5 I + H 2 O = ZnO + HI + C 2 H 6 . (2) The same gas is produced when a galvanic current is passed through a concentrated solution of potassium acetate : 2 2C 2 H 3 2 K + H 2 = C 2 H 6 + K 2 CO 3 + CO 2 + H 2 . Carbon dioxide and ethane are evolved at the negative pole, whilst hydrogen is set free at the positive pole. Properties. Ethane is a colourless and odourless gas, slightly soluble in water, but more soluble in alcohol. According to Schickendantz, its coefficient of absorption in water is : c = 0-094556 - 0-0035324* + 0'00006278. 2 It has a specific gravity of 1*036, and condenses to a liquid under a pressure of 40 atmospheres at 4. It burns with a luminous flame, the illuminating power being about half that of ethylene burnt under similar conditions. 3 The heat of combus- tion (for 1 mol. in grams) is 370,440 cal. ETHYLENE, ETHENE, OR OLEFIANT GAS, C 2 H 4 . Density =27'82. 405 This gas appears to have been discovered by Becher, who obtained it by heating alcohol with sulphuric acid. His obser- vations were, however, considered to be erroneous up to the time of Priestley, who, in his Experiments and Observations on Air, mentions that Ingenhouss had seen such a gas prepared by a certain Enee in Amsterdam. The properties of olefiant gas were accurately studied in the year 1795 by Deimann, Paets van Troostwyk, Bondt, and Lauwerenburgh. 4 These Dutch chemists found that the gas obtained from alcohol and sulphuric acid was totally different from ordinary inflammable air, and that it contained both hydrogen and carbon. The difference between marsh gas and olefiant gas was first pointed out by William Henry 5 in 1805, and his view of the composition of the gas was borne out by the subsequent experiments of Dalton, Davy, and Berzelius. In those days, marsh gas and olefiant gas were the 1 Frankland, Journ. Chem. Soc. 1850, 263. 2 Kolbe, Annalen, 69, 257. 3 Frankland, Journ. Chem. Soc. 1885, i. 23;. 4 Crell. Ann. 1795. 5 Nicholson's Journal, 1805. 700 THE NON-METALLIC ELEMENTS only hydrocarbons known. s that their specific gravities being- very different, they were termed the light- and the heavy- carburetted hydrogen respectively. In his History of Chemistry 1 Thomas Thomson states that it was by the inves- tigation of the chemical composition of these two gases that Dalton was led to the recognition of the laws of combination in multiple proportions, and to the conception of his atomic theory,, from noticing that for the same quantity of hydrogen heavy carburetted hydrogen contains twice the quantity of carbon that is contained in marsh gas. Preparation. In order to prepare olefiant gas, alcohol i& heated with strong sulphuric acid. The method proposed by FIG. 188. Erlenmeyer and Bunte 2 is the best. Twenty-five grams of alcohol and 150 grams of sulphuric acid are brought into a flask of from two to three liters in capacity (see Fig. 188), and the mixture heated till the evolution of gas begins. By means of a funnel-tube, furnished with a stopcock, ft, a mixture of equal volumes of alcohol and sulphuric acid is then allowed to drop into the flask. To purify the gas thus obtained it must be first washed through concentrated sulphuric acid, and afterwards through caustic soda, contained in the Woulff s bottles, c and d. It may then be collected over water and preserved in a gas holder. Properties. Ethylene is a colourless gas, possessing a peculiar 1 Vol. ii. p. 291. 2 Annalen, 168, 64. ACETYLENE 701 ethereal smell ; it is only slightly soluble in water, but is more so in alcohol; its solubility in water between 5 and 21 is expressed by the following equation (Pauli) : c = 0-25629 - 0-00913631* + 0'000188108/! 2 . Its specific gravity is 0'9784. It condenses under a pressure of 45 atmospheres to a colourless liquid, which solidifies to a crystalline mass at 181, melts again at 169 and boils at - 105. The liquid has a specific gravity of 0'335 at 8. Ethylene is readily inflammable, and burns with a brightly luminous flame, the illuminating power being about 70 candles. 1 The heat of combustion (for 1 mol. in grams at 18) is 333,350 cal. When mixed with three times its volume of oxygen and ignited it explodes with extreme violence. One very characteristic property of ethylene is its power of uniting with an equal volume of chlorine to form a heavy colourless liquid termed ethylene dichloride, C 2 H 4 C1 2 , or, Dutch liquid, it having been first observed by the four Dutch chemists already named. From this property, indeed, the gas derives its name. It was originally termed gaz hitileux, but this name was afterwards changed by Fourcroy to gaz oUfiant. When brought in contact with strongly ozonised oxygen, ethylene de- tonates very powerfully. In order safely to exhibit this property, a current of the gas is allowed to pass through a wide tube about 10 mm. in diameter, whilst ozonised oxygen is allowed to enter into this by means of a narrow tube, which is inserted into the wider tube to a distance of one centimeter. As soon as the ozo- nised oxygen comes in contact with the olefiant gas a detonation occurs, usually accompanied with the formation of white fumes. 2 ACETYLENE OR ETHINE, C 2 H 2 = 25-82 406 This gas was discovered and its composition determined in 1836, by Edmund Davy, 3 who prepared it by treating with water the black mass obtained in the manufacture of potassium. The existence of this gas was afterwards observed by some other chemists, but it was not until the year 1859 that Berthelot 4 investigated it completely. Acetylene is produced by the incomplete combustion of many volatile organic substances, especially of ethylene, coal-gas, and other hydrocarbons, as well 1 Frankland, Journ. Chem. Soe. 1885, i. 237. 2 Houzeau and Renard, Compt. Rend. 76, 572. 3 Reports of British Association, 1836, p. 62. 4 Ann. Chim. Phys. [3], 57, 82. 702 THE NON-METALLIC ELEMENTS as the vapours of alcohol, ether, &c. Acetylene is also formed when the vapours of these organic liquids are passed through red-hot tubes. Acetylene is remarkable as being the only hydrocarbon which has been obtained by the direct union of its elements. These only combine together at the highest temperature which can be artificially produced. In order to prepare acetylene in this way the electric arc, obtained by the passage of a powerful current between two poles of gas carbon is employed. These carbon poles are fitted through apertures in a globular glass vessel, through which a slow current of pure hydrogen is allowed to pass. Acetylene can also be prepared in any wished -for quantity from ethylene dibromide. If this substance be heated with an alcoholic solution of caustic potash, the following reactions take place : (1) C 2 H 4 Br 9 + KOH = C 2 H 3 Br + KBr + H 2 0. (2) C 2 H 3 Br+KOH = C 2 H 2 +KBr + H 2 0. To remove any vapours of the very volatile bromethylene which may be carried over from this operation, the gases evolved from the boiling liquid are allowed to pass through a second flask containing a boiling alcoholic solution of caustic potash. Another method of preparation consists in allowing water to drop on to barium carbide, which is prepared by heating to redness a mixture of precipitated barium carbonate, magnesium powder and gas carbon. The gas thus obtained only contains 2 3 per cent, of hydrogen and no appreciable quantities of other hydrocarbons. 1 Calcium carbide (p. 703) is also decomposed by water with formation of pure acetylene. 2 Properties. Acetylene is a colourless difficultly condensible gas, having a specific gravity of O92, with a very unpleasant penetrating smell. It is usually stated that the smell observed when a Bunsen burner " burns down " is due to the presence of acetylene, but according to V. Meyer the odour of pure acetylene is quite distinct from this. At 18 it dissolves in its own volume of water, whilst alcohol dissolves six times its volume of the gas. Acetylene burns with a strongly luminous and smoky flame. It acts as a poison when it comes in contact with the blood, combining with the haemo- globin. 3 Like ethylene it combines directly with chlorine and 1 Maqueime, Compt. Rend. 115, 558. - Moissan, Compt. Rend. 118, 501. 3 Bistrow and Liebreieh, Bcr. 1, 220. ACETYLENE 703 bromine forming with the latter element acetylene dibromide, C 2 H 2 Br 2 , as well as the tetrabromide, C 2 H 2 Br 4 . Very characteristic derivatives of acetylene are the explosive compounds which it forms with certain metals or metallic salts. Thus, if the gas is allowed to pass over fused potassium, hydrogen is evolved, and bodies having the composition C 2 HK and C 2 K 2 are formed. Both of these substances are black powders, which are decomposed in contact with water with explosive violence, acetylene being reproduced. A similar calcium compound, C 2 Ca, is obtained by heating a mixture of lime and sugar charcoal in the electric furnace : 1 CaO + 3C = C 2 Ca + CO. This substance, when treated with water, yields lime and acety- lene. When the gas is passed through an ammoniacal solution of a cuprous salt, a dark blood-red precipitate is formed, having the composition Cu 2 C 2 or Cu 2 C 2 ,H.,0. 2 This reaction is a very characteristic one. It serves as a test for the presence of acetylene, and is so delicate, that by its means ^^ part of a milligram of acetylene can be with certainty detected (Berthelot). For the purpose of exhibiting its action with cuprous chloride the flame of a large Bunsen burner is allowed to burn down, the air-openings at the bottom are then nearly closed, and when the maximum amount of gas is burning, a large dry balloon is brought over the tube of the burner, so that the incompletely burnt gas is brought into the middle of the vessel. After some minutes the balloon is removed and a few drops of ammoniacal solution of cuprous chloride, Cu 2 Cl 2 , are poured in, the balloon being shaken so as to bring a thin film of the liquid over the whole of the interior surface. This is then seen to be covered with a coating of the red acetylene-copper compound. If a current of acetylene be led into an ammoniacal solution of a silver salt, C 2 Ag 2 or C 2 Ag 2 .H separates out in the form of a white precipitate. This body, like the copper compound, explodes on heating or by percussion. On the addition of acids to these compounds, acetylene is evolved. Hence they may be employed for the separation of the gas from a gaseous mixture, as well as for the purpose of obtaining it in a chemically pure state ; thus : C 2 H 2 Cu 2 O + 2HC1 = C 2 H 2 + Cu 2 Cl 2 + H,0. 1 Moissan, Compt. Rend. 118, 501. 2 Blochmann, Ber. 7, 274 ; Keiser, Amer. Chem. Journ. 14, 285 ; Plimpton, Proc. Chem. Soc. 1892, 109. 704 THE NON-METALLIC ELEMENTS To determine the quantity of acetylene in coal-gas, or other mixture of gases, these are passed through ammoniacal silver nitrate, which gives a precipitate of silver acetylide mixed with metallic silver. The former is converted by hydrochloric acid into silver chloride, which may be separated by treatment with ammonia, the filtrate precipitated with nitric acid, and the silver chloride weighed ; 1 gram, of silver chloride corresponds to 0'09 gram, or 87 03 cc. of acetylene. 1 A simple method for preparing acetylene is as follows : A bent funnel (see Fig. 189) is placed over a burner, in which the FIG. 189. flame is burning down. Connected with the neck of the funnel are two cylinders, a and b, containing an ammoniacal solution of nitrate of silver, and the products of the combustion are drawn through the liquids contained in the cylinders by means of an aspirator, c. By the addition of hydrochloric acid to the precipitate, pure acetylene can be obtained, and the chloride of silver which is precipitated only requires to be again dis- solved in ammonia in order to fit it to be employed a second time for the same purpose. If the compound C 2 Cu 2 be allowed to remain in contact with zinc and aqueous ammonia, the nascent hydrogen evolved 1 Lewes, Journ. Chem. Soc. 1892, i. 324. CARBON MONOXIDE 705 by the action of the zinc combines with the acetylene and forms ethylene. Platinum-black when brought in contact with a mixture of acetylene and hydrogen also brings about the same combination, both ethylene and ethane being formed. 1 Acetylene is formed from its elements with absorption of heat, and, like many other similar endothermic compounds, can be made to undergo explosive decomposition when subjected to the shock given by the explosion of fulminate of mercury. (Of. p. 734.) CARBON AND THE HALOGENS. 407 Carbon combines with fluorine directly to form carbon tetrafiuoride, CF 4 , the reaction taking place spontaneously at the ordinary temperature with the less dense forms, whilst the denser varieties must be heated to 50 100. The remaining halogens, however, do not combine directly with carbon, but compounds of carbon with these elements can be prepared by indirect methods. Thus the ultimate product of the action of chlorine on methane is carbon tetrachloride, CC1 4 , the four atoms of hydrogen being replaced by chlorine, which also com- bines with the hydrogen set free, forming hydrochloric acid. CH 4 + 4C1 2 = CC1 4 + 4HC1. These compounds are described in the volumes relating to Organic Chemistry. CARBON AND OXYGEN. 408 Carbon forms two oxides, both of which are gaseous : Carbon monoxide, CO. Carbon dioxide, CO 2 . CARBON MONOXIDE OR CARBONIC OXIDE, CO = 2779 This compound, which is commonly known as carbonic oxide, was first obtained by Lassone by heating zinc oxide with charcoal. 2 He found that a combustible gas was thus given off which burned with a blue flame, and when mixed with air did not explode, as was usually the case with inflammable air. Lavoisier (1777) obtained the same gas by heating alum and 1 V. Wilde, Ber. 7, 353. 2 Mem. Paris Acad. 1776. 46 706 THE NON-METALLIC ELEMENTS charcoal together, and found that on combustion it yielded carbonic acid. In spite of these observations, the gas was for a long time mistaken for hydrogen, and Priestley in 1796 showed that iron scale (oxide of iron), when heated with well- calcined charcoal, gives out an inflammable air, whereas according to Lavoisier's theory it ought only to give carbonic acid. This fact was, in Priestley's opinion, opposed to the antiphlogistic system, whilst it supported the view that the oxides contain water, and that inflammable air is phlogisticated water, and this was corroborated by the fact that when steam is led over red- hot charcoal it is phlogisticated to inflammable air. This con- clusion was one which upholders of the Lavoisierian system found difficult to disprove, and they were driven to assume that hydrogen was still contained even in the most strongly heated charcoal ; this fact, however, Priestley most satisfactorily proved to be incorrect in his last work, The Doctrine of Phlogiston Established, published in 1800. In the same year Cruikshank was engaged with the exami- nation of this same gas, which he obtained by heating carbon with different metallic oxides. From its comparatively high specific gravity he concluded that this gas was not a hydro- carbon, as others had assumed it to be. When burnt with oxygen it yielded no water, and nothing but an almost equal volume of carbon dioxide, whilst the oxygen which was needed for its combustion was less in volume than that contained in the carbonic acid gas formed. Hence he concluded that it must be an oxygen compound, and therefore gave to it the name of " gaseous oxyde of carbone." Clement and De*sormes soon after confirmed Cruikshank's results ; they determined the composition of carbonic oxide more accurately than he had done, and found that it is likewise formed \vhen carbon dioxide is led over red-hot charcoal. 409 Preparation. Carbon monoxide can be prepared in various ways. (1) It is formed when zinc oxide, ferric oxide manganese dioxide, and many other oxides are heated with charcoal ; it is also formed when chalk (calcium carbonate) magnesite (magnesium carbonate) and other carbonates are heated with metallic zinc or iron filings; the decomposition which takes place in these cases is represented by the follow- ing equations : CaCO. + Zn = CaO + ZnO + CO. CARBON MONOXIDE 707 (2) When carbon dioxide (carbonic acid gas) is passed through a long tube filled with charcoal and heated to redness : C + CO 2 = 2CO. (3) When oxalic acid or an oxalate is heated with concen- trated sulphuric acid, a mixture of equal volumes of carbon monoxide and carbon dioxide is evolved ; thus : The carbon dioxide may readily be separated from the carbon monoxide either by passing the mixed gases through a solution of caustic soda or by collecting the mixture over water rendered alkaline by this substance. (4) Pure carbonic oxide is formed when formic acid or a formate is heated with concentrated sulphuric acid : CH 2 2 = CO + H,0. On the other hand, carbonic oxide can be converted into formic acid by heating it with caustic potash, potassium formate being produced (Berthelot) : (5) Carbon monoxide can be readily prepared in quantity by heating finely powdered potassium ferrocyanide with from 8 to 10 times its weight of strong sulphuric acid ; l in this reaction, potassium sulphate, ammonium sulphate, and iron sulphate are formed, the following equation representing the decomposi- tion : K 4 FeC 6 N 6 + 6H.,S0 4 + 6H 2 O = 6CO + 2K 2 SO 4 + 3(NH 4 ) 2 S0 4 + FeSO 4 . The water required for the reaction is contained in the ferrocyanide itself, which crystallizes with three molecules of water, and also in the commercial acid, which invariably con- tains some. According to Grimm and Ramdohr 2 it appears that sulphur dioxide and carbon dioxide are evolved in the beginning of the reaction, but afterwards pure carbon monoxide is given off. (6) Carbonic oxide may be readily prepared by passing carbon 1 Fownes, Phil. Mag. 24, 21. 2 Annalen, 98, 127. 708 THE NON-METALLIC ELEMENTS dioxide over zinc dust heated in a glass tube to rather less than a red heat, the resulting gas being washed through caustic soda. 1 This reaction is sometimes made use of for the preparation of the gas on the large scale. Properties. Carbon monoxide is a colourless, tasteless gas, which possesses a peculiar though slight smell, and has a specific gravity of 0'9678 (Craikshank), 0'96702 (Leduc). 2 The relative density remains constant up to 1200, but at 1690 the gas partially decomposes with formation of carbon dioxide, free carbon being deposited. 3 It has been liquefied by Cailletet, Wroblewski, and Olszewski ; its critical temperature is 139 5, the corresponding pressure being 35*5 atmospheres, it boils at -190 and solidifies at -211 (Cailletet). 4 In water it is only very slightly soluble, its absorption coefficient, according to Bunsen and Pauli, being obtained from the following equation c = 0-032874 - 0-00081632* + 0-000016421* 2 . It is, however, easily soluble in an acid or ammoniacal solu- tion of cuprous chloride, Cu 2 Cl 9 . 410 Carbon monoxide easily burns in the air with a bright blue flame, carbon dioxide being produced. The lambent flame observed on the top of a large coal fire is due to the combustion of this gas. When it is mixed with oxygen and ignited by the application of a flame or by an electric spark, a violent explosion occurs. If however the mixture of carbonic oxide and oxygen be thoroughly freed from moisture by long-continued preserv- ation over phosphorus pentoxide, it is found to be impossible to produce explosion by passing an electric spark through the gap. The addition of a trace of water-vapour or of gases, which, by reaction with oxygen, are capable of producing water, such as sulphuretted hydrogen, hydrogen, hydrocarbons, etc., immediately renders the mixture explosive. The combination of carbon monoxide with oxygen therefore does not appear to take place directly, and it is probable that the oxidation of the gas is effected by means of the water-vapour present, which acts as a carrier of oxygen, the following reactions taking place : = C0 2 +H 2 . = 2H 2 0. 1 Noack, Bcr. 16, 75. 2 Compt. Rend. 115, 1072. 3 Meyer and Langer, Ber. 18, 134 c. 4 Compt. Rend. 99, 706 ; 100, 350 ; Monatsh. Ohem. 6, 204. CARBON MONOXIDE 709 This view of the chemical changes which occur when carbon monoxide burns is supported by the fact that the rapidity of explosion of a mixture of oxygen with carbon monoxide in a tube one meter long is greater with a large quantity of aqueous vapour than when only a trace is present. 1 This property of carbon monoxide may be readily demonstrated by plunging a flame of this gas burning in air into a jar of oxygen in which some strong sulphuric acid has been shaken up ; the flame is at once extinguished. When carbon is burned in a free supply of ordinary air or oxygen, carbon dioxide is formed. If, however, every trace of moisture is as far as possible removed both from the carbon and from the oxygen, it is found that the carbon may be heated nearly to redness without taking fire, the oxidation taking place without the evolution of light, and that carbon monoxide is almost the sole product. In order to obtain oxygen sufficiently dry for experiments of this kind, it is necessary to pass the gas through sulphuric acid and then allow it to remain for some weeks in carefully closed flasks, containing a considerable amount of phosphorus pentoxide. 2 Carbon monoxide is also largely formed by the oxidation of carbon when heated in air a,t a comparatively low temperature. These facts render it probable that in the combustion of carbon carbon monoxide is the first product, and is then oxidised to carbon dioxide by the oxygen of the water-vapour present. Carbon monoxide combines directly with metallic nickel and iron to form very remarkable substances of the composition Ni(CO) 4 and Fe(CO) 5 . 3 These substances readily decompose when heated into the metal and carbon monoxide. They will be described under the compounds of the metals. Carbonic oxide is a very poisonous gas, inasmuch as it com- bines with the haemoglobin of the blood : small animals die almost instantly when placed in the gas, and when even small quantities are inhaled severe headache, giddiness, and insensi- bility readily occur. The accidents, as well as suicides, which occur from burning charcoal in a chauffer in a small room, are due to the inhalation of this gas formed by incomplete com- bustion, and deaths occurring from sleeping upon lime-kilns and 1 Dixon, Phil. Trans. 1884, ii. 617. a Baker, Journ. Chem. Soc. 1885, i. 349 ; Phil. Trans. 1888, A. 571. 3 Journ. Chem. Soc. 1890, i. 749 ; 1891, 1090 ; Compt. Rend. H3, 679 ; Proc. Chem. Soc. 1891, 126. 710 THE NON-METALLIC ELEMENTS brick-kilns are probably also produced by this gas. The poisonous nature of the gas from red-hot charcoal was known and expatiated upon as early as the year 1716, when F. Hoffman published his work, Considerations on the Fatal Effects of the Vapour from Burning Charcoal. The remarkable poisonous action of carbonic oxide appears to depend upon the fact that the whole dissolved oxygen is thereby expelled, the blood acquiring a light, purple-red colour. The absorption spectrum of the carbonic-oxide-haBmoglobin is dis- tinguished by the fact that the bands between D and b are situated nearer to b than is the case with oxyhsemoglobin (Hoppe- Seyler), and that, unlike the latter, it remains unchanged in presence of reducing agents. This test serves as a means of detecting cases of poisoning with the gas. Carbonic oxide may also be detected in small quantities in the air by passing the air to be examined through blood and then making a spectroscopic examination of the latter. The gas is either estimated by explosion with oxygen or more generally by absorption with acid cuprous chloride. The composition of carbon monoxide is determined by mixing the gas with oxygen, exploding the mixture, and measuring the carbon dioxide formed. It is thus ascertained that one volume of the gas unites with half a volume of oxygen to form one volume of carbon dioxide. Now the latter is known to contain its own volume of oxygen (p. 714), and it hence follows that carbon monoxide contains half its own volume of oxygen. The molecular weight of the gas is shown by its density to be about 27'79, and this amount of it therefore contains 15'88 of oxygen and 11*91 of carbon, the formula of the gas being CO. * CARBON DIOXIDE on CARBONIC ANHYDRIDE, C0 2 = 43'67. 411 This gas belongs to the class of acid-forming oxides for- merly termed acids, and is still best known under its old name of carbonic acid. It has already been stated in the historical introduction that this gas was first distinguished from common air by Van Helmont, who termed it gas sylvestre. He obtained it by the action of acids on alkaline or calcareous substances, and showed that it is also formed by the combustion of charcoal, and in the fermentation and decay of carbonaceous matter, and that it likewise occurs in the mineral water at Spa, in the Grotto del Cane near Naples, and in other localities. Van Helmout describes the suffocating action it exerts on animal CARBON DIOXIDE 711 life as well as its effects in extinguishing flame. Fr. Hoffmann made further observations on the gas contained in effervescing mineral waters, and he states that this is frequently given off in such quantity that when the water is enclosed in bottles these are burst by the force of the gas. He also shows that this substance, to which he gave the name of spiritus mineralis, has the power of reddening certain blue vegetable colouring matters, and hence he considers it to be a weak acid. Although many other chemists investigated the properties of this gas, it was not until the time of Black that it was dis- tinctly shown to differ essentially from common air. Black (1755) proved that this substance is a peculiar constituent of the carbonated or mild alkalis, being, in them, combined or fixed in the solid state, whence it was termed by him fixed air. In the year 1774 Bergman published a complete history of this peculiar air, to which he gave the name acid of air, because of its occurrence in the atmosphere. Its chemical nature was first properly explained by Lavoisier, who showed that, whilst mercuric oxide heated alone gives off pure oxygen gas, fixed air is evolved when it is heated with carbon, proving that this latter gas is an oxide of carbon. Carbon dioxide is a body which is widely distributed in nature ; as we have already seen, it forms a small but constant and essential constituent of the atmosphere ; it is likewise invariably contained in soil, being one of the chief products of the decay of all organic substances. From the soil it is taken up by rain and spring water, and it is to this substance that the latter, to a great extent, owes its fresh and pleasant taste. It occurs in chalybeate and acidulous waters in large quantities, whilst in both ancient and modern volcanic districts it is emitted in very large volumes from the fumeroles and rents in the ground, for instance in the old craters in the Eifel, at Brohl on the Rhine, as well as in the Auvergne, particularly in the neighbourhood of Vichy and Hauterive, where the gas is actually employed for the manufacture of white lead. Especially remarkable for the evolution of this gas in very large quantities is the Poison Valley in Java, which also is an old crater, and the Grotto del Cane near Naples, which is of such a construction that the heavy carbonic acid gas, entering from the fissures in the floor of the cave, at a depth of from two to three feet below the mouth of the cave, collects up to this depth, and small animals such as dogs, when thrown into the cave, respiring the impure 712 THE NON-METALLIC ELEMENTS air, fall down, whilst a man breathing the pure air above this level is unaffected by the gas. The carbonic acid contained in the air is derived from a variety of sources ; it is formed by the respiration of man and animals, as well as in the act of combustion of organised material, and in its decay and decomposition. The amount of atmospheric carbonic acid varies between certain narrow limits, but on an average reaches 3 volumes in 10,000 volumes of air. In the presence of the sunlight, plants have the power, through their leaves, of decomposing this carbonic acid, taking up the carbon to form their own tissue, and eliminating the oxygen gas ; hence the amount of carbonic acid in the air does not increase beyond the limits named (Saussure). Carbon dioxide is an acid-forming oxide giving rise to a series of salts termed the carbonates, many of which occur in nature as minerals. Amongst these is especially to be men- tioned calcium carbonate, CaCO 3 , which occurs in two distinct crystalline forms as calc-spar and arragonite, whilst it is found in a massive crystalline form in marble and limestone ; calcium carbonate also forms the chief constituent of the shells of mollusca and foraminifera, the remains of which constitute the chalk formation as well as the greater part of all the lime- stones. The double carbonate of magnesium and calcium (MgCa)CO 3 also occurs in large masses as dolomite. Amongst other naturally occurring carbonates may be mentioned mag- nesite, MgCO 3 ; iron spar or spathic iron ore (FeMnCaMg) CO 3 ; witherite, BaC0 3 ; strontianite, SrCO 3 ; and calamine, ZnCO 3 . 412 Preparation. (1) In order to prepare carbon dioxide a carbonate, such as marble or chalk, is brought into a gas- evolution flask, and dilute hydrochloric acid poured upon it, when the gas is rapidly evolved with effervescence : CaC0 3 + 2H01 = C0 2 + CaCl 2 + H 2 0. The gas thus obtained invariably carries over small quantities of hydrochloric acid vapour with it, from which it may be freed by passing through a solution of bicarbonate of soda. In order to obtain a constant stream of carbon dioxide the same apparatus may be employed which was made use of for the preparation of sulphuretted hydrogen gas (Fig. 100). (2) Another method of obtaining a constant current is to PROPERTIES OF CARBON DIOXIDE 713 pour concentrated sulphuric acid over chalk, and add a very small quantity of water: CaC0 3 + H 2 S0 4 = C0 2 + CaS0 4 + H 2 O. The dilute acid cannot be employed for this purpose, since the calcium sulphate formed, which is soluble in the con- centrated acid, does not dissolve in the dilute acid, and thus prevents its further action on the chalk. (3) The gas thus obtained from chalk possesses a peculiar smell, which is due to the presence of small quantities of volatile organic matter always contained in the chalk. In order to prepare a very pure gas, sodium carbonate may be decomposed with pure dilute sulphuric acid ; thus : Na 2 CO 3 + H 2 SO 4 = CO 2 + Na,SO 4 + H 2 O. (4) Carbon dioxide may be obtained on a large scale for the preparation of bicarbonate of soda, white lead, and other com- mercial products by " burning " limestone, or even by the combustion of charcoal or coke. 413 Properties. Carbon dioxide is a colourless gas possessing a slightly pungent smell and acid taste. It does not support either combustion or respiration; hence a flame is extinguished when plunged into the gas, and animals thrown in become insensible and suffer death from suffocation. At the same time, however, it does not exert a poisonous action on the animal economy, as indeed may be gathered from the fact that it is constantly taken into the lungs and emitted from them ; it destroys life, however, because it does not contain any free oxygen. According to Berzelius, common air containing Vth of its volume of carbon dioxide can be breathed without pro- ducing any serious effects ; but from Angus Smith's later experiments 1 it appears that when air contains only 0*20 per cent, by volume of this gas, its effect in lowering the action of the pulse is rendered evident after the respiration has continued for about an hour. It seems, therefore, premature to say that the smallest increase of atmospheric carbonic acid may not be productive of hurtful results. In close spaces inhabited by man the quantity of carbon dioxide is naturally much larger than in the open air. The oppressive feeling which respiration in such air is apt to produce is not however in general due to the presence of this gas, but rather to the volatile organic emana- 1 Air and Rain, p. 209, "On Some Physiological Effects of Carbonic Acid." 714 THE NON-METALLIC ELEMENTS tions, the presence of which is indicated by the unpleasant smell observed on entering such rooms from the fresh air. Carbon dioxide possesses, according to Regnault, a specific gravity of T529, and hence, being much heavier than air, it may be poured like water from one vessel to another; this fact can be strikingly exhibited by bringing a lighted taper into the vessel into which the carbon dioxide has been poured, when it will be instantly extinguished. This property is made use of to test the presence of the gas in old wells, cellars, and coal-pits, where it frequently accumulates, and is termed choke damp. Before the workman descends it is usual to lower down a burning candle, and thus ascertain whether the air is pure enough for respiration. Air containing 4 per cent, of carbon dioxide extinguishes a candle-flame, but it will support respira- tion for a short time. Composition of Carbon Dioxide. When carbon burns- to the dioxide, the volume of the gas which is formed is the same as that of the oxygen which is needed to produce it. This fact may readily be shown by help of the apparatus which is shown in Fig. 108, under sulphur dioxide, the experiment being con- ducted as there described, except that a small piece of freshly heated charcoal is placed in the small combustion pan instead of sulphur. The molecular weight of the gas, calculated from its density, is 43*67, and this amount of it therefore contains 31'76 parts of oxygen and 11*91 parts of carbon, the molecular formula being CO 2 . Dumas and Stas l have accurately ascertained the atomic weight of carbon by a series of careful experiments. For this pur- pose they made use of the apparatus which is shown in Fig. 190, in which they burnt diamond and purified graphite in a stream of oxygen gas. The carbon dioxide formed was completely absorbed by means of caustic potash, and the composition of the carbon dioxide calculated from the weight of it formed, together with the weight of carbon burnt. In order to carry out this investigation it was necessary, in the first place, to in- sure the absence of every trace of carbon dioxide from the oxygen gas employed. The oxygen used was, therefore, collected in a large Woulff's bottle (a) and preserved over water rendered alkaline by caustic potash. The gas was driven out from this gas-holder in the course of the experiment by means of a dilute caustic potash solution (&). The oxygen then passed through a 1 Ann. Chim. Phys., [1] 40, 991- ATOMIC WEIGHT OF CARBON long wide tube filled to the point (c) with pumice-stone moistened with strong caustic potash solution, whilst at (d) it came in contact with pieces of dry solid caustic potash, and at (e) with glass moisten- ed with boiled sulphuric acid. In order to insure the absolute dryness of the gas it passed through a U tube, marked (/), filled with pumice-stone moist- ened with sulphuric acid. This tube was weighed before and after the ex- periment. The pure dry oxygen then passed into a porcelain tube, which was heated to redness in a tube-furnace, the substance undergoing combustion be- ing placed at (y) in a platinum boat. The boat containing the substance was accurately weighed both before and after the experiment, inasmuch as even the purest diamond and graphite always leave on combustion traces of inorganic ash, and this must, of course, be sub- tracted from the total amount of substance taken. The other end of the tube contained, at (A), a quan- tity of perfectly oxidised copper scales, CuO, which served for the purpose of oxidizing to carbon dioxide 716 THE NON-METALLIC ELEMENTS any monoxide which might be formed. The carbon dioxide then passed through the tube (i) containing pumice-stone moistened with sulphuric acid, and then into two Liebig's potash-bulbs (k and /). In order to be certain that the whole of the carbon dioxide was absorbed, and that no moist air passed away from the potash -bulbs, the excess of oxygen was passed through two tubes (m and n), the first of which contained pumice-stone moistened with potash, and the last solid potash. In this way five combustions of natural graphite were made, four of artificial graphite, and five of diamond. As a mean of the closely agreeing results, it was found that 800 parts of oxygen combined with Natural Graphite. Artificial Graphite. Diamond. 299-94 299-95 300'02 The mean of these numbers is 299*97. Hence 2 atoms of oxygen, or 31*76 parts by weight, combine with 1T9 parts by weight of carbon. Similar numbers have been obtained by other chemists, and the most accurate value is probably 11*91. This number is taken as the atomic weight of carbon, because it is the least amount ever found in a molecule of one of its compounds. 414 Liquefaction of Carbon Dioxide. Carbon dioxide can be liquefied both by cold and by pressure. Faraday was the first to obtain this result by pouring some sulphuric acid into the closed limb of a bent tube made of strong glass, whilst over it he pushed a piece of platinum foil on which a lump of carbonate of ammonia was placed. After closing the open end of the tube the sulphuric acid was cautiously allowed to flow over the car- bonate, and in this way the carbon dioxide which was evolved was condensed by its own pressure. Liquid carbon dioxide pre- pared in this way cannot be employed for further experiments, as on endeavouring to open the tube it usually bursts with a loud explosion. Gore, 1 however, made an ingenious arrange- ment in which the carbon dioxide is generated in a small bent tube closed with a stopper of gutta-percha, to which is fixed a small glass cup. In this way the solvent action of the liquid on other bodies was first examined. In order to condense carbon dioxide on a larger scale Thilorier 2 constructed an apparatus which consisted of a strong cast-iron cylinder, termed the generator, capable of being closed hermeti- 1 Phil. Trans. 1861, p. 83. 2 Thilorier, Annalen, 30, 122. LIQUEFACTION OF CARBON DIOXIDE 7r cally, the original form of which is shown in Fig. 191. In this cylinder he placed bicarbonate of soda, whilst sulphuric acid was placed in a separate vessel lowered into the cylinder. The cylinder was closed by means of the screws at the top, and then inclined so that the sulphuric acid came in contact with the bicarbonate of soda, evolving carbon dioxide in such quantity that it was liquefied by its own pressure. The liquid carbon dioxide was then separated from the sulphate of soda formed at the same time, by distilling it into a similar cast- iron vessel, termed the receiver. Owing to the brittle nature and non-homogeneity of the cast-iron of which the cylinders were made, and in consequence of the enormous pressure which they had to withstand, it was feared that serious acci- FIG. 191. dents might occur in the production of the liquid according to Thilorier's plan, and Hare suggested that the apparatus should be made of wrought iron. Not long after this suggestion had been made an accident occurred : one of Thilorier's cylinders exploded with fearful force whilst it was being charged, and a young chemist of the name of Hervey lost his life. The wrought- iron apparatus, which was employed after this time, is shown in Fig. 192, the construction having been much improved byMareska and Donny. 1 Instead of wrought iron they employed a leaden cylinder, surrounded by another one made of copper, this, in its turn, being bound round with strong hoops of wrought iron. In order to prepare liquid carbon dioxide with this apparatus, 1 Mem. cour. et mem. Sava?its itrang. Acad. Roy. Eruxelles, 18. 718 THE NON-METALLIC ELEMENTS the requisite quantity l of bicarbonate of soda is placed in the generator, and a quantity of dilute sulphuric acid, sufficient for its decomposition, is poured into the narrow copper cylinder. The apparatus is closed with a strong screw-tap, and then made to revolve on its axis so that the acid comes slowly in contact with the carbonate. For the purpose of distilling the carbon dioxide, which is formed by this decomposition, into the horizontal receiver, a copper tube is screwed on, connecting the one vessel with the other. When the screw-taps are opened the FIG. 192. liquid dioxide distils over, the receiver being kept cool and the generator being placed, at the end of the operation, in hot water. By repeating this operation a large quantity of liquid carbon dioxide can easily be obtained. The apparatus, however, which is now most commonly em- ployed for the preparation of liquid carbon dioxide on the small scale or for lecture purposes is that made by Natterer 2 of Vienna and Bianchi of Paris. In this apparatus, the general arrange- ment of which is seen in Fig. 193, carbon dioxide, generated by 1 2| Ibs. of bicarbonate of soda, 6 Ibs. of water, and ]^ Ib. of oil of vitriol. 2 J. Pr. Chem. 35, 169. LIQUEFACTION OF CARBON DIOXIDE 719 FIG. 194. FIG. 193. 720 THE NON-METALLIC ELEMENTS the action of dilute sulphuric acid on bicarbonate of soda, is con- densed by means of a force-pump into a wrought-iron pear-shaped vessel, seen in section in Fig. 194, furnished with suitable cocks, t, Fig. 194, the air which was contained in the flask being first of all displaced by the gas. During the operation the reservoir (V) is surrounded by a freezing mixture contained in a vessel, screwed on to the plate a b, Fig. 193. 415 Liquid carbon dioxide is now a commercial article and sold in iron bottles containing 8 kg., being mostly prepared in certain large breweries from the carbon dioxide evolved in the process of fermentation. It is also made from the natural gas evolved from certain springs, especially at Burgbrohl on the Rhine, and from the gas prepared from marble or sodium bicarbonate. The liquefaction is accomplished by compressing the gas by means of powerful pumps into recipients kept cool by a stream of water. It is used for producing cold, as well as high pressures, and therefore employed in the manufacture of cast steel. It is also employed for beer-engines, in the manu- facture of salicylic acid, and for the production of aerated waters. Liquid carbon dioxide is a colourless very mobile liquid slightly soluble in water, upon the surface of which it swims. Its specific gravity is 0'9951 at -10, O9470 at 0, and O8266 at -f 20 . 1 These numbers show that liquid carbon dioxide expands more upon heating than a gas, and its coefficient of expansion is, therefore, larger than that of any known body. The boiling-point of liquid carbon dioxide is 78*2 under a pressure of 760 mm., and its tension at different temperatures is given in the following table : Temperature. Pressure in mm. of mercury. Temperature. Pressure in mm. of mercury. - 25 13007*02 + 15 39646-86 - 15 17582-48 + 25 50207-32 - 5 23441 "34 + 35 62447-30 + 5 30753-80 + 45 76314-60 Liquid carbon dioxide conducts electricity badly ; it does not act as a solvent on most substances, does not redden dry litmus paper, and possesses no very striking chemical properties (Gore). When the stopcock of a vessel containing liquid carbon dioxide is opened, a portion of the liquid which rushes out solidifies in the form of white snow-like flakes (Thilorier). This is caused by the absorption of heat, produced by the rapid 1 Andreef, Annalen, HI. SOLID CARBON DIOXIDE 721 evaporation of another portion of the liquid. For the purpose of obtaining larger quantities of this carbon dioxide snow, the apparatus shown in Figs. 195 and 196, suggested by Natterer, is employed. It consists of two brass cylinders, AB and CD, one of which can be fixed inside the other, each cylinder being fastened to a hollow handle. The liquid carbon dioxide is allowed to pass quickly through the tube d (the position of which inside the box is shown in Fig. 196), from the nozzle n, Fig. 194, of the reservoir. A portion of the liquid undergoes evaporation and passes out as gas through the fine holes in the handles, whilst the larger portion remains behind as a solid finely divided crystalline mass, which is a very light sub- stance, capable of being pressed together like snow. It evapor- ates much less quickly than the liquid dioxide, because it is much colder, and in addition it has to take up the amount of heat necessary for liquefaction. Notwithstanding its excessive- ly low temperature it may be touched without danger, the gas which it constantly emits forming a non-conducting at- mosphere round it. If, however, it be pressed hard upon the skin, solid carbon dioxide pro- duces a blister exactly like that produced by contact with a FlG8< 195 and 196 red-hot body. This snow-like mass may readily be pressed into cylindrical wooden moulds. The cylinders thus obtained have a density of about 1'2, and require a considerable time for evaporation in the air. 1 When solid carbon dioxide is mixed with ether, and the mixture brought 1 Landolt, Ber. 17, 309. 47 722 THE NON-METALLIC ELEMENTS into the vacuum of an air-pump, the temperature sinks to 110. A tube containing liquid carbon dioxide brought into this solidifies to a transparent ice-like solid mass (Mitchell and Faraday). When liquid ammonia is placed over sulphuric acid in a vacuum and allowed to evaporate, such a diminution of tem- perature is obtained that gaseous carbonic acid may be liquefied. Exposed to a pressure of from 3 to 4 atmospheres, this solidifies, and the transparent mass thus obtained, when pressed with a glass rod, separates into cubical masses. 1 It is very remarkable that solid carbon dioxide melts at 65, that is at a temperature lying much above its boiling point at the atmospheric pressure. A process has been patented for carrying out the manufacture of solid carbonic acid by a similar method on the large scale. The solid material may be transported in bags or casks, or in the iron chamber in which it is produced. 2 In the paragraph (24) concerning the continuity of the gaseous and liquid states of matter, it has been pointed out that a certain temperature may be reached for any given gas, above which this gas cannot be liquefied, however much the pressure may be increased. This temperature is termed by Andrews 3 the critical point, that of carbon dioxide being found by him to be 30*92, whilst according to more recent observations it is 31*35, the pressure at this temperature, or the critical pressure, being 72*9 atmospheres. 4 In order to determine this point, Andrews employed the apparatus shown in Figs. 197 and 198. The dry and pure gas is contained in the glass tube a b, which is closed at a and open at b, the gas being shut in by the thread of mercury c. This tube is firmly bedded in the brass end-piece d, and this, in its turn, is bolted on to the strong copper tube which carries a second end-piece at its lower extremity. Through this the steel screw e passes packed with leather washers to render the cylinder perfectly air-tight. The cylinder is completely filled with water, and the pressure on the gas is increased up to 400 atmospheres by turning the steel screw e into the water. The closed and capillary end of the tube containing the compressed gas is sometimes surrounded with a cylinder into which water of a given temperature is brought. A second modification of the apparatus is shown in Fig. 198. The capillary pressure-tube is bent round so as to render it possible to place it in a freezing 1 Loir and Drion, Compt. Rend. 52, 748. 2 English Patent 13,684, 1891. 3 Phil. Trans. 1869, partii, 575. 4 Amagat, Compt. Rend. 114, 1093. LIQUID CARBON DIOXIDE mixture under the receiver of an air-pump. The action both of great pressure and great cold upon the condensed gas can thus be examined. If the screw be turned round when the gas possesses a temperature below the critical point, liquid carbon dioxide is formed when a certain pressure is reached, and a layer of this liquid can be distinctly seen lying with the gas above it. If the same experiment be repeated at a temperature above the critical one, no liquid is seen to form even when the pressure is increased to 150 atmospheres. The volume gradually diminishes, no line of demarcation or other alteration in appearance can be observed, and the tube appears as empty as it did before the gas was submitted to pressure. If the tem- perature be now lowered below 31-35, the whole mass becomes liquid, and when the pressure is dimin- ished begins to boil, two distinct layers of liquid and gaseous matter being again observed. Thus, whilst at all temperatures below 31 '35 carbon dioxide gas cannot be converted into a liquid without a sudden condensa- tion, at temperatures above this point gaseous carbon dioxide may, by the applica- tion of great pressure and subsequent cooling to below the critical point, be made to pass into a distinctly liquid condition without undergoing any sudden change such as is observed in the case of ordinary liquefaction. 416 Carbon dioxide is a very stable body, requiring for its decomposition an extremely high temperature. When passed over pieces of porcelain in a porcelain tube, heated to a tempera- ture of. from 1200 to 1300, it decomposes to a small extent FIGS. 197 and 198. 724 THE NON-METALLIC ELEMENTS into carbon monoxide and oxygen (Deville) ; and the same decomposition is brought about by the electric spark (Dalton and Henry). In this way only a small portion of the gas is decomposed, as when a certain quantity of oxygen has been formed this again combines with the carbon monoxide. Under favourable circumstances the recombination of the gases, accom- panied by the passage of a flame, can be actually observed to take place periodically. 1 If, however, hydrogen, or mercury, or any oxidizable body be present, the whole of the carbon dioxide is converted into monoxide (Saussure). This decomposition is best shown by allowing the electric spark to pass by means of iron poles through a measured volume of carbon dioxide ; after the action is completed, the volume of carbon monoxide formed is found to be exactly equal to that of the dioxide taken, thus : 2C0 2 = 2CO + O 2 , the oxygen being completely absorbed by the metallic iron (Buff and Hofmann). A piece of burning magnesium wire or ribbon continues to burn when plunged into carbon dioxide, the oxide of the metal being formed and carbon liberated : When carbon dioxide is passed over heated potassium or sodium, the carbonates of these metals are formed, and carbon is set free ; thus : Liquid carbon dioxide is also attacked by the alkali metals (Gore). Carbon dioxide is soluble in water, its solubility beng re- presented by the following equation : c = 1-7967 - 0-07761* + 0-0016424* 2 , or one vol. of water at dissolves 1*7967 vols. of carbon dioxide. 5 1-4497 10 1-1847 15 1-0020 20 0-9014 1 Hofmann, Ber. 23, 3303. CAEBONIC ACID 725 It is more easily soluble in alcohol of specific gravity 0*792, the coefficient of solubility being given by the following equa- tion : c = 4-32955 - 0-09395t + 0'00124t 2 . When the pressure is much smaller than that of the atmo- sphere, carbon dioxide follows Dalton and Henry's law of the absorption of gases in water, but deviation is observed from this law at higher pressures, increasing gradually as the pressure is increased. The following table shows the volume of carbon dioxide, measured at and 760 mm. dissolved by one volume of water at (Column S), the pressure being given in the column (P). 1 P S S P in atmospheres Vols. at 1 1-797 1-797 5 8-65 1-730 10 16-03 1-603 15 21-95 1-463 20 26-65 1-332 25 30-55 1-222 30 33-74 1-124 Q If the law of Henry and Dalton were correct, the value of would be constant. It appears from the table, however, that this fraction gradually decreases in value, the solubility increasing at a slower rate than the pressure. CARBONIC ACID AND THE CARBONATES. 417 The aqueous solution of carbon dioxide contains in solu- tion a dibasic acid, termed carbonic acid, CO(OH) 2 . This acid does not exist iri the pure concentrated state, and in this respect resembles sulphurous acid and other acids whose corresponding oxides are gaseous. Its existence is, however, ascertained by the fact that the aqueous solution of carbon dioxide reddens litmus paper, whereas dry carbon dioxide, whether in the gaseous or liquid state, produces no such action. It has also been observed that if the water be saturated with carbon dioxide under pressure, and if the pressure be then at once removed, 1 Wroblewski, Compt. Rend. 94, 1355. 726 THE NON-METALLIC ELEMENTS the gas quickly makes its escape by effervescence, the bubbles being so minute that the whole liquid, during the disengage- ment of the gas, appears milky. If, however, the liquid be allowed to remain in contact with the gas for some consider- able time after saturation, before the pressure is removed, the gas escapes, on removal of the pressure, in large bubbles which adhere chiefly to the sides of the glass vessel in which the liquid is contained. The difference may be explained by the fact that, to begin with, the carbon dioxide is mechanically dissolved in the water, but that after remaining in contact with the water for some time an actual combination takes place between these two substances, which may be represented by the following equation : /OH. O = C = O + H-O-H = O=C If a piece of porous substance like sugar or bread be brought into a liquid, such as soda-water or champagne, which has been saturated under pressure with carbon dioxide, and which has been exposed to the air for some little time, so that the effer- vescence due to the diminution of pressure has ceased, a strong renewal of the effervescence is observed. Water which contains a small quantity of common salt in solution dissolves carbon dioxide more easily than common water. This depends on the fact that a chemical decomposition takes place, a part of the common salt, NaCl, being converted by the carbonic acid present into acid sodium carbonate, NaHCO 3 , free hydrochloric acid, HC1, being liberated. The presence of the latter acid may easily be shown by the addition to the liquid of a small quantity of ultramarine, this substance losing its blue colour in presence of hydrochloric acid, whilst a solution either of common salt or of carbonic acid fails to act upon it. If a current of carbon dioxide be passed through a solution of chloride of lead, an insoluble chlorocarbonate of lead is formed 1 : /Cl Pb< /Cl \0v 2Pb< +C0 2 +H 2 O= >qO+2HCL Na , I give rise to). Carbonyl chloride is, at the ordinary temperature, a gas possessing a peculiar, very unpleasant, pungent smell, and having a specific gravity of 3'4604 (Thomson). At a low temperature it can be condensed, forming a colourless liquid, boiling at +8 01 and having a specific gravity at of 1*432. The compound is decomposed by water into hydrochloric acid and carbon dioxide ; thus : For the purpose of preparing pure carbonyl chloride the method proposed by Wilm and Wischin 2 is the best. The mixed gases issuing at about the same rate are brought into a large glass balloon having a capacity of about ten litres ; from this balloon the mixed gases pass into a second one, which, like the first, is exposed to sunlight. It is best to employ a slight excess of chlorine, this being afterwards got rid of by passing the gas through a tube filled with lumps of metallic antimony. The gas thus purified can be liquefied by passing into a tube surrounded by ice, or, better, by a freezing mixture. Carbonyl chloride is also formed when a mixture of twenty parts of trichloromethane (chloroform), four hundred of sulphuric acid, and fifty of potassium dichromate are heated together on a water-bath (Emmerling and Lengyel), thus : 2CHC1, + K 2 Cr 2 7 + 5H 2 S0 4 = 2COC1 2 + 2KHSO 4 + Cr 2 (SO 4 ) 3 + C1 2 + 5H 2 O. Carbonyl chloride is also obtained when carbon tetrachloride is acted upon by fuming sulphuric acid. 3 1 Emmerling and Lengyel, Annalen, suppl. Band 7, 101. 2 Annalen, 147, 147. 3 Ber. 26, 1993. 732 THE NON-METALLIC ELEMENTS It is the chloride of carbonic acid, just as sulphuryl chloride, SO 2 C1 2 , is the chloride of sulphuric acid, and when treated with ammonia is converted into the amide of carbonic acid, known as carbamide or urea, CO(NH 2 ) 2 (p. 739). Carbonyl chloride is largely used in the preparation of colour- ing matters, for which purpose it is sometimes made by passing equal volumes of carbon monoxide and chlorine over heated animal charcoal ; under these circumstances combination takes place in the dark. Carbonyl chloride is also formed by the spontaneous oxidation of chloroform when it is exposed to the light in the presence of air. Chloroform which is to be used as an anaesthetic should therefore be kept in the dark, and the bottles containing it should be kept filled, as the presence of carbonyl chloride in it renders it extremely dangerous if used for inhalation. CARBON OXYBROMIDE OR CARBONYL BROMIDE, COBr 2 = 186-51. 421 When a mixture of carbon monoxide and bromine vapour is exposed to the light, combination takes place slowly, but the gas thus prepared retains a yellow colour. Potash decomposes it with the formation of potassium carbonate and potassium bromide. 1 Carbonyl bromide has also been obtained in the impure state by the oxidation of bromoform. 2 CARBON AND SULPHUR. CARBON BISULPHIDE, CS 2 = 75'55. 422 This compound was accidentally discovered by Lampadius in 1796, by heating pyrites with charcoal. In their investiga- tion of carbonic oxide in the yea.r 1802, Clement and Desormes wished to ascertain whether charcoal invariably contained combined hydrogen ; they examined the action of sulphur on red- hot charcoal, and obtained the same liquid which had been pre- viously discovered by Lampadius. This liquid they first believed to be a compound of hydrogen and sulphur, but they soon con- vinced themselves that it only contained carbon and sulphur. Notwithstanding these experiments, the nature of the compound remained doubtful, until Vauquelin ascertained that its vapour, 1 Schiel, Annalen, suppl. Band 2, 311. 2 Emmerling, Ber. 13, 874. CARBON BISULPHIDE 733 passed over red-hot metallic copper, is converted into carbon and copper sulphide. Carbon bisulphide is prepared on the large scale by passing the vapour of sulphur over red-hot charcoal. For this purpose a large upright cast-iron cylinder (Fig. 200), ten or twelve feet long and one to two feet in diameter, is employed. This cylinder is placed above a furnace and surrounded by brickwork, and at the same time it is provided with a lid to admit of the whole being filled with charcoal. A second opening (a), fur- nished with a hopper, exists at the bottom of the cylinder, and this serves to bring the sulphur into the apparatus. The sulphur evaporates, and in the state of vapour combines with the red-hot carbon, impure carbon bisulphide distilling over by FIG. 200. the tube (c), and collecting in the vessel (d) under water. The tubes (e) serve as condensers, to separate the vapour of carbon bisulphide from sulphuretted hydrogen formed during the re- action, owing to the presence of hydrogen in the charcoal, the sulphuretted hydrogen being absorbed by passing over the layers of slaked lime contained in the purifier (/). In actual work the yield of bisulphide is about 20 per cent, below the theoretical amount. The crude substance invariably contains sulphur in solution, from which, however, it may be separated by distillation ; but other sulphur compounds, which impart to the crude material a most offensive odour, are also contained in the distillate. In order to remove these impurities, different processes are in use. The substance was formerly purified by frequent re-distillation, over oil or fat, by which means 734 THE NON-METALLIC ELEMENTS the disagreeably smelling compounds were held back. Another means of purification employed is that of shaking the liquid with mercury, and allowing it to remain for a long time in contact with corrosive sublimate in the cold, and then distilling it off white wax. Pure carbon bisulphide is a colourless, mobile, strongly re- fracting liquid, possessing a sweetish srnell not unlike that of ether or chloroform, having a specific gravity at of 1*29232 (Thorpe) and boiling at 46. It solidifies at 116,remelting at 110 (Wroblewski and Olszewski). When carbon combines with sulphur to form carbon bisulphide, heat is absorbed which is given out again when it is decomposed. Like many sub- stances of this class, it can be caused to undergo explosive decomposition, the heat thus evolved being sufficient to propa- gate the explosion throughout the mass of the gas. When a small quantity of fulminating mercury is exploded in a glass tube containing vapour of carbon bisulphide, the vapour is decomposed into carbon and sulphur, which are deposited on the sides of the tube. 1 The same decomposition can be brought about by the explosion of the yellowish-brown powder which is formed by the action of sodium on carbon bisulphide, and has the formula C 5 S 2 . Pure carbon bisulphide is exceedingly inflammable, taking fire in the air at 149, according to Frankland, and burning with a bright blue flame. A mixture of one volume of carbon bisulphide vapour and three volumes of oxygen detonates with great violence on inflammation, so that great care is needed in the manufacture, and frequent and severe explosions are caused by its inflammability. A mixture of the vapour of carbon bisul- phide and nitric oxide burns with a very bright blue flame, par- ticularly rich in chemically active rays. Carbon bisulphide is powerfully poisonous, and its vapour soon produces fatal effects on small animals exposed to its action. It not only acts as a poison when inhaled in large quantity, but it produces very serious effects upon the nervous system when inhaled for a considerable time even in very small amount. The vapour of carbon bisulphide acts as a powerful anti- putrescent, and Zoller 2 has shown that meat and other putrescible bodies may be preserved fresh for a long time if kept in an atmosphere containing the vapour of this compound. Carbon bisulphide is largely used in the arts, especially in the 1 Thorpe, Journ. Chem. Soc. 1889, i. 220. 2 Ber. 10, 707. THIOCARBONIC ACID 735 indiarubber and woollen manufactures, in the first case as a solvent for the caoutchouc, and in the second as a means of regaining the oil and fats with which the wool has to be treated. As it dissolves iodine in large quantity, but does not dissolve appreciably in water, it is employed for the purpose of determin- ing the amount of moisture contained in commercial iodine. A remarkably characteristic reaction of carbon bisulphide is that it possesses the power of combining with triethylphosphine, P(C 2 H 5 ) 3 , a derivative of common alcohol, to form a solid compound, crystallizing in magnificent red crystals, and having the composition P(C 2 H 5 ) 3 CS 2 . Other compounds of carbon and sulphur have been described, but these consist for the most part of amorphous brown sub- stances, and are probably mixtures. A compound tricarbon disulpkide, C 3 S 2 , has recently been described by Lengyel, 1 who prepared it by passing the vapour of carbon bisulphide over an electric arc passing between carbon poles. It is a deep-red liquid, which is slowly volatile at the ordinary temperature, the vapour causing a copious flow of tears. It distils partly unchanged under diminished pressure, but is partly converted at the same time into a black solid amorphous modification of the same composition. It combines with bromine to form a yellow compound of the composition C 3 S 2 Br 6 , which has a not unpleasant aromatic smell. THIOCARBONIC ACID, H 2 CS 3 . 423 The salts of this acid, which is sometimes termed sul- phocarbonic acid, are formed by processes analogous to those by which the carbonates are produced. This is shown by the reaction, discovered by Berzelius, in which carbon bisulphide is brought in contact with a solution of the sulphide of an alkali-metal ; thus : CS 2 + Na 2 S = Na 2 CS 3 , corresponding to CO 2 + Na 2 O = Na. 2 C0 3 . When alcohol is added to the solution thus obtained, the thio- carbonate separates out in the form of a heavy liquid of a slightly brown colour, which on the addition of cold dilute hydrochloric acid decomposes into free thiocarbonic acid. This forms a 1 Her. 26. 2960. 736 THE NON-METALLIC ELEMENTS yellow oil, possessing a most disagreeable penetrating odour ; when slightly heated it is resolved into carbon bisulphide and sulphuretted hydrogen. The thiocarbonates of the alkali-metals and those of the alkaline earths are soluble in water. The soluble thiocarbonates give a brown precipitate with copper salts, a yellow precipitate with dilute silver nitrate solution, and a red precipitate on the addition of a lead salt. These precipitates rapidly become black, owing to the formation of the corresponding sulphides. THIOCARBONYL CHLORIDE, CSC1 2 = 114-11. 424 This compound was obtained by Kolbe 1 by acting for some weeks with dry chlorine gas upon carbon bisulphide. According to Carius, 2 it can be easily obtained by heating phosphorus pentachloride and carbon bisulphide together in sealed tubes at 100, when the following decomposition takes place : PCL + CS 2 = PSC1 3 + CSCly It is most readily obtained by acting on the compound CSC1 4 described below, with tin and hydrochloric acid, and forms a strongly smelling liquid which boils at 73'5, has a specific gravity of 1 '5 085 at 15, and is only slowly attacked by water. 3 The compound CSC1 4 is obtained by the action of chlorine on carbon bisulphide in presence of a little iodine, and is a most unpleasant-smelling liquid, boiling with slight decomposition at 149. CARBONYL SULPHIDE OR CARBON OXYSULPHIDE, COS = 59*55. 425 This gas, discovered by Than, 4 is formed when a mixture of sulphur vapour and carbon monoxide is passed through a moderately heated tube. It is also obtained by numerous other reactions, such for example as the action of sulphur trioxide 5 or chlorosulphonic acid 6 on carbon bisulphide. It is usually prepared by the action of dilute acids on potassium thiocyanate, 1 Annalen, 45, 41. 2 Annalen, 112, 193. 3 Klason, Ber. 20, 2378. v 4 Annalen Suppl. 5, 236. 5 Armstrong, Ber. 2, 712. 6 Dewar and Cranston, Zeitsch, Chem. 1869, 734. CARBONYL SULPHIDE 737 the thiocyanic acid first formed splitting up into carbonyl sulphide and ammonia HCNS + H 2 = COS + NH 3 , the latter combining with the excess of acid present. To obtain the pure gas the following method is employed : To a cold mixture of five volumes of sulphuric acid and four volumes of water such a quantity of potassium thiocyanate (sulphocyanate of potassium) is added that the mass just re- mains liquid. The evolution of gas, which commences without heating, is regulated either by cooling or gently warming the mixture, and by care a constant current of gas can thus be obtained. The decomposition which takes place is as follows : NCSK + H 2 + 2H 2 S0 4 = COS + KHSO 4 + (NH,)HSO 4 . The gas invariably contains the vapours of hydrocyanic acid, HCN, and bisulphide of carbon. The first is removed by passing the gas through a tube filled with cotton-wool which has been rubbed in oxide of mercury, and the second by passing it over pieces of cut caoutchouc contained in a tube placed in a freezing mixture. According to Hofmann, the vapour of carbon bisulphide cannot be thus completely re- moved, but it may be got rid of by passing the gas over cotton-wool saturated with an ethereal solution of triethyl- phosphine, (C 2 H 5 ) 8 P. Carbonyl sulphide is a colourless gas, having a specific gravity of 2*1046, and a peculiarly resinous smell, at the same time resembling that of sulphuretted hydrogen. It becomes liquid at under a pressure of 12'5 atmospheres, and the liquid when poured out solidifies. It is very inflammable, taking fire even when brought in contact with a red-hot splinter of wood, and burning with a blue slightly luminous flame. A mixture of one volume of the gas with one and a half volumes of oxygen in- flames with slight explosion, burning with a bright blue flame ; if, however, it be mixed with seven volumes of air, the mixture burns without explosion. A platinum wire heated to whiteness by the electric current decomposes the gas completely, without alteration of volume, into sulphur and carbon monoxide. Carbonyl sulphide is soluble in water ; one volume of the gas dissolves in an equal volume of water, and imparts to the solution its own peculiar smell and taste. The sulphur- 48 738 THE NON-METALLIC ELEMENTS waters of Harkany and Panic! in Hungary possess similar properties, and probably contain, as other sulphur-waters may do, carbonyl sulphide. The aqueous solution gradually decom- poses with formation of sulphuretted hydrogen and carbon dioxide thus : COS + H0 = SH + C0 2 . The gas is slowly absorbed by aqueous solutions of caustic alkalis with formation of a carbonate and a sulphide : COS + 4KOH = K 2 CO 3 + K 2 S + 2H 2 0. Alcoholic potash also absorbs it rapidly. CARBAMIC ACID. CO 426 This acid, the monamide of carbonic acid, is not known in the free state, although we are acquainted with its salts and its ethereal salts. Ammonium car bam ate is formed when dry carbon dioxide is brought into contact with dry ammonia (J. Davy, H. Rose) : It is best prepared by passing a rapid stream of the two gases into well-cooled absolute alcohol. A deliquescent crystal- line powder is formed, which smells of ammonia and is readily soluble in water. 1 Ammonium carbamate is also contained in the commercial carbonate of ammonia, and can be obtained from this source by allowing the commercial substance to stand in contact with saturated aqueous ammonia for forty hours at a temperature of from 20 to 25; on cooling, the carbamate separates out. The specific gravity of the vapour of ammonium carbamate between 37 and 100 is 0'892 (Naumann) ; hence it appears that this body is completely dissociated on evaporation into carbon dioxide and ammonia. If the gaseous mixture thus obtained be allowed to cool, ammonium carbamate is again formed, but only slowly, owing to the fact that the combination is not a direct one, an intramolecular change taking place. 2 When brought in contact with acids, ammonium carbamate evolves carbon dioxide, whilst with alkalis it evolves ammonia. 1 Basarow, J. Pr. Chem. [2], 1, 283 ; Mente, Annalen, 248, 234. 2 Naumann, Annalen, 150, 1- CARBAMIDE OR UREA 739 The sodium and potassium salts have also been prepared, and also a calcium salt which has probably the constitution \ ( ).Ca(OH). Carbaminic chloride or chloroformamide, NH 2 .COC1, is pre- pared by passing carbonyl chloride over ammonium chloride. It usually exists as a very pungent-smelling liquid, but it may be obtained in long prisms melting at 50; the liquid boils at 61 62. CARBAMIDE OR UREA, CO(NH 2 ) 22 . 427 Urea, which is largely contained in urine and in other liquids of the animal body, was first described in the year 1773 by H. M. Rouelle as Extractum saponaceum urince. It was, how- ever, more accurately investigated in the year 1790 by Fourcroy and Vauquelin. The discovery by Wohler, in the year 1828, 1 that when an aqueous solution of ammonium cyanate is heated, this compound undergoes a molecular change and is converted into urea, is one of the most important discoveries of modern chemistry, inasmuch as it was the first case in which a compound formed in the animal body was prepared from its inorganic constituents : NC(ONH 4 ) = CO(NH 2 ) 2 . Urea is also obtained by the action of ammonia on carbonyl chloride 2 : COC1 2 + 2NH 3 = CO(NH 2 ) 2 + 2HC1. It was obtained first in this manner by John Davy, mixed with ammonium chloride, in 1812, sixteen years before Wohler's synthesis, but he did not recognize its identity, although he determined the relative volumes of the gases entering into combination 3 It is likewise produced by heating ammonium carbamate or common carbonate of ammonia to a temperature of from 130 to 140 (Basarow) : ro 2 _ po 2 _L H o J i ONH 4 - > i NH 2 4 1 Wohler, Pogg. Ann [1828], 12, 253. 2 Natanson, Annalen, 98, 287 ; Neubaner, Annaleit, 101, 342. 3 Phil. Trans. 1812 ; see also Matthews, Chem. Newx, 1890. 740 THE NON-METALLIC ELEMENTS Urea is found in the urine of all mammalia, especially that of the carnivora, and also in small quantities in that of birds and reptiles, as well as in the excreta of the lower animals. It forms an important constituent of the vitreous humour of the eye, and invariably occurs in small quantity in blood, and some- times in perspiration and pus. In order to prepare urea from the urine the liquid is evaporated down on a water-bath to dryness, the residue treated with alcohol, the alcoholic solution evapo- rated, and the residue again treated with absolute alcohol. On evaporating this last solution, urea, slightly coloured with extractive matter, remains behind. A second method is to evaporate down urine to a syrupy consistency, and then add pure nitric acid, when nitrate of urea separates out as a crystalline powder, which is decomposed on the addition of potassium carbonate, with formation of potassium nitrate and free urea. These bodies can then be easily separated by treatment with alcohol, in which the urea dissolves. A third method is that recommended by Berzelius : a concentrated solution of oxalic acid is added to the evaporated urine, when a precipitate of the insoluble oxalate of urea is thrown down ; this is boiled with chalk, when the pure urea is left in solution. The preparation of urea from ammonium cyanate is, however, by far the best. For this purpose potassium cyanate is first prepared by heating dried potassium ferrocyanide with potassium bichromate in the manner described on p. 761. This may be then dissolved in water, and an equivalent quantity of ammonium sulphate added, or the requisite quantity of the latter may be added to the mother liquors obtained in preparing the cyanate. Potassium sulphate is formed, the greater part of which crys- tallizes out, the small portion still remaining in solution being got rid of by evaporation and subsequent crystallization. The liquid is then evaporated to dryness, and the urea contained in the dry residue is extracted with boiling alcohol, and may be further purified by recrystallization from amyl alcohol. 1 Urea dissolves in its own weight of cold water and in every proportion in boiling water. It also dissolves in five parts of cold and one part of boiling alcohol, but is almost insoluble in ether. It crystallizes in long striated quadratic prisms, and possesses a cooling taste resembling that of nitre. On heating, it melts at 132, and decomposes when heated to a higher point. Although urea does not possess an alkaline reaction, it 1 Erdmann, Ber. 26, 2438. SALTS OF UREA 741 combines easily with acids, and forms a series of salts which crystallize well, and of which the following are the most important : Hydrochloride of urea, CO(NH 2 ) 2 HC1. This salt is formed when dry hydrochloric acid acts upon urea. In consequence of the heat which is evolved in the act of combination, the com- pound appears in the form of a yellowish oil, but this crystal- lizes on cooling. On addition of water it is resolved into its constituents. Nitrate of urea, CO(NH 2 ) 2 ~NO 3 H. This is a very charac- teristic compound, being soluble in water but almost insoluble in nitric acid. Hence if pure nitric acid be added to a toler- ably concentrated solution of urea, a crystalline precipitate falls, which on recrystallization from solution in hot water is deposited in prismatic crystals. Oxalate of urea, 2CO(NH 2 ) 2 ,C 2 4 H 2 . This salt separates out in the form of monosymmetric tablets when a solution of urea is mixed with a concentrated solution of oxalic acid. Urea not only unites with acids but also with a variety of salts and oxides. Thus, for instance, it forms with common salt the compound CO(NH 2 ) 2 -f NaCl 4- H 2 0. This substance crystallizes from aqueous solution in large glittering rhombic prisms. Urea also forms similar compounds with certain other chlorides and nitrates. When treated with a solution of caustic potash in the cold, it does not yield any ammonia, but on warming it is gradually resolved into ammonia and carbon dioxide. When heated with water above 100 it is also decomposed, and it is at once acted upon by nitrous acid or anhydride, with liberation of the nitrogen of both compounds in the free state and formation of water if the solution be warm : CO(NH 2 ) 2 + N 2 3 = C0 2 + 2N 2 + 2H 2 0. In the cold ammonium carbonate is also formed 1 : 2CO(NH S ) 2 + N 2 S = (NH 4 ) 2 C0 3 + 2N 2 + C0 2 . Neutral potassium permanganate solution has no action on urea in the cold, and very little at 100, but the acidified solution causes the rapid evolution of 1 volume of nitrogen and 2 volumes of carbon dioxide at 100. Solutions of urea evaporated with silver nitrate yield ammonium nitrate and silver cyanate. 1 Glaus. Ber. 4, 1440 742 THE NON-METALLIC ELEMENTS 428 In order to detect the presence of urea in a liquid it is evaporated down on a water-bath, the residue extracted with alcohol, the liquid evaporated, and a few drops of pure nitric acid added to the residue. If urea is present the nitrate of urea is precipitated, and this is recognised, inasmuch as it consists of microscopic crystalline scales forming rhombic or six-sided tables, having an angle of 82, as measured by means of the microgoniometer. It has already been stated that urine when not sterilized de- composes with separation of ammonium carbonate. This change occurs in in presence of certain micro organisms, notably micro- coccus urine and bacillus coli communes which can be separated from the liquid by nitration. If the filter containing these micro-organisms be now washed with water and dried at a temperature not exceeding 40, they retain their activity, and if this paper be coloured with turmeric it may be used for detecting the presence of urea in neutral liquids. In about ten to fifteen minutes after the paper is moistened with a neutral solution of urea a portion of the urea is converted into ammonium carbonate, and the paper becomes of a brown tint. In this way one part of urea can be detected in 10,000 of liquid. 1 Urea is the last product of the action of the atmospheric oxygen on the nitrogenous components of the human body, and the quantity which is excreted within a given time gives us a measure of the changes which are going on ; for although there are other nitrogenous products contained in the urine, yet their quantity is so small in comparison to that of the urea that in most cases they may be disregarded. For the medical man and the physiologist it is therefore of the greatest conse- quence that a quick and accurate method should exist for determining the amount of urea in the urine. One of the best processes for this purpose is that proposed by Liebig. Jt de- pends upon the fact that when mercuric nitrate is added to a solution of urea, a white insoluble precipitate falls down, possessing the composition 2CO(NH 2 ) 2 + 3HgO + Hg(N0 3 ) 2 . It is convenient that each cc. of the mercuric solution should pre- cipitate O'Ol gram, of urea, and a solution of this strength is obtained by dissolving 71 '48 grams, of metallic mercury in nitric acid, evaporating to drive off as much of the acid as possible, and diluting the solution to one liter. Before this solution is added to the urine it is necessary to precipitate the phosphatesand sulphates which the latter con tains in solution. For 1 Musculus, Compt. Rend. 78, 132. DETERMINATION OF UREA 743 this purpose two volumes of urine are mixed with one volume of a mixture of equal volumes of baryta-solution and cold saturated solution of barium nitrate. The standard mercury solution, is then added by a burette to 15 cc. of the filtrate, corresponding to 10 cc. of urine, until no further precipitate occurs. The end of the reaction is easily ascertained by adding a solution of sodium carbonate to a drop of the liquid, which assumes a yellow colour as soon as a slight excess of mercury is present. One cc. of the mercury solution corresponds to 0*01 gram, of urea. According to Bunsen's method, the urine is heated with a solution of barium chloride in dilute ammonia to a temperature of 230, and the barium carbonate which separates out is weighed : CO(NH 2 ) 2 + BaClg -f- 2H 2 = BaC0 3 -f- 2NH 4 C1. A third method for the determination of urine, known as the Davy-Knop method, depends on the fact that urea when brought into contact with an alkaline hypochlorite or hypo- bromite evolves pure nitrogen, the carbon dioxide wliich is formed being absorbed by the free alkali : CO(NH 2 ) 2 + 3NaOBr = CO 2 + N 2 + SNaBr + 2H 2 0. For this purpose a freshly prepared alkaline solution of sodium hypobromite is prepared by dissolving bromine in excess of caustic soda, and this is mixed with a solution of urea, the nitrogen gas which is evolved by this reaction being collected in a graduated tube. For the details of this process reference must be made to the original papers. 1 ( NTTOTT HYDROXYCARBAMIDE, CO T 429 For the preparation of this compound, a solution of hy- dro xylamine nitrate in absolute alcohol is cooled to 10, and to this a concentrated solution of potassium cyanate is added. On standing, hydroxycarbamide separates out in white needles which melt at 128130. ISURETINE, CH 4 ON 2 . This isomeride of urea is formed when an alcoholic solution of hydroxylamine is warmed with a concentrated solution of 1 Hiifner, J. Pr. Chem. [2], 3 ; Russell and West, Journ. Chcm. Soc. 1874, 249. 744 THE NON-METALLIC ELEMENTS hydrocyanic acid to a temperature of from 40 to 50. On evaporating the solution isuretine separates out in the form of long colourless rhombic crystals which have an alkaline re- action, and melt at 105. This substance forms with acids a series of easily crystallizable salts, and its constitution is represented by the following formula : NH 2 .CH:N.OH. BIURET, C 2 O 2 H 5 N 3 . 430 This compound, discovered by Wiedemann, 1 is formed when urea is heated for some time to 150 160. The decom- position which takes place is represented as follows : co NH 2 / co/ X NH 2 X NH 2 . When the residue is heated with water, cyanuric acid remains behind, and biuret separates out on cooling the solution. This may be purified by dissolving in hot water and reprecipitating with dilute ammonia. Biuret forms long white needle-shaped crystals, 100 parts requiring for their solution 6,493 parts of water at 15, whilst at 106, the boiling-point of a saturated solution, 222 parts only are needed. 2 When a few drops of cupric sulphate and then an excess of caustic soda are added to a solution of biuret in water, the liquid assumes a colour varying from red to violet, according to the quantity of copper salt added. By means of this reaction it is easy to show the passage of urea into biuret ; it is only necessary to heat a small quantity of urea in a test-tube until ammonia begins to be evolved, the melted mass being poured into hot water and treated as described. 1 Annalen, 68, 323. 2 Hofmanii, Ber. 4, 262. CARBONYLDIUREA 745 CARBONYLDIUREA, C 3 H 6 N 4 O 3 . 431 When urea is heated to 100 with carbonyl chloride in a sealed tube, this substance is formed ; thus : CO It is a white powder, which separates from boiling water as a crystalline powder. On heating biuret with carbonyl chloride a substance very similar to the one now described is obtained, to which the name of carbonyldi-biuret has been given, 1 and which has the follow- ing constitution : CO(NH.CO.NH.CO.NH 2 ) 2 . OXYTHIOCARBAMIC ACID, CO I 432 Berthelot first obtained the ammonium salt of the above acid. This salt separates out in crystals, when dry carbonyl sulphide is brought in contact with ammonia. The acid itself is not known. . When heated in a closed tube the ammonium salt is con- verted into urea 2 : THIOCARBAMIC ACID, CS | 433 The constitution of this compound, originally described under the name of hydro thiosulphoprussic acid, 3 was first pointed out by Debus. 4 The ammonium salt of this acid is formed by the 1 E. Schmidt, J. Pr. Chem. [2], 5, 35. 2 Kretzschmar, J. Pr. Chem. [2], 7, 474. 3 Zeise, AnndUn, 48, 95. 4 Annalen, 73, 26. 746 THE NON-METALLIC ELEMENTS union of carbon bisulphide with dry ammonia in presence of absolute alcohol ; the salt separates out, after some time, in prismatic crystals. When hydrochloric acid is added to its aqueous solution, free thiocarbamic acid separates out. This substance is an oil at the ordinary temperature, but below 10 it forms a crystalline mass. 1 It has a smell resembling that of sulphuretted hydrogen, possesses an acid reaction, and is easily decomposed into sulphuretted hydrogen and thiocyanic acid, NCSH. THIO-UKEA, OR THIOCARBAMIDE, CS(NH 2 ) 2 . 434 Reynolds 2 was the first to prepare this analogue of urea. It is formed by a reaction analogous to that by which common urea is obtained. A molecular change takes place in ammonium thiocyanate NCS(NH 4 ) when heated to 140, corresponding to that observed in the case of ammonium cyanate. At the same time a quantity of guanidine thiocyanate is formed, while a portion of the thiocyanate remains unchanged. 3 In order to separate these three bodies, the melted mass is treated with two- thirds its weight of cold water, when the greater portion of the thio-urer, remains behind. This is then dried on a porous plate, and purified by recrystallization from hot water. According to Glaus, it is not necessary to prepare am- monium thiocyanate for this purpose, but a solution of the crude salt obtained by dissolving carbon bisulphide in alcoholic ammonia may be employed ; this is quickly evaporated, until ammonia, ammonium sulphide, and carbon bisulphide are abun- dantly evolved, and then the residue treated as above described. As long as thio-urea is not perfectly pure it crystallizes in silky needles, but in the pure condition it forms magnificent large rhombic prisms or thick tables. It dissolves in eleven parts of cold water, and on heating with water to 140 is reconverted into ammonium thiocyanate. Heated by itself to 170 180 it is converted into guanidine thiocyanate and ammonium thio- carbonate. Like common urea, thio-urea forms compounds with acids, salts, and oxides. A characteristic compound is the nitrate CS(NH 2 ) 2 , NO 3 H, which crystallizes exceedingly well. Seleno-urea, CSe(NH 2 ) 2 , has also been prepared. 4 1 Mulder, J. Pr. Chem. 101, 401, and 103, 178. 2 Journ. Chem. Soc. 1855, 1. 3 J. Volhard, Ber. 7, 92, and J. Pr. Chem. [2], 9, 6. 4 Verneuil, Compt. Rend. 99, 1154. CYANOGEN COMPOUNDS 747 CARBON AND NITROGEN. CYANOGEN COMPOUNDS. 435 The history of these compounds commences with the dis- covery of Prussian blue, made accidentally early in the eighteenth century by a colour-maker of the name of Diesbach. The fact was shortly afterwards communicated to the alchemist Dippel, who ascertained the conditions under which the formation of the colouring matter takes place. The method for preparing the colour was, however, first published by Woodward, who states that it was obtained by calcining the alkali obtained by heating equal parts of cream of tartar arid saltpetre with ox-blood. The residue of the calcination was then lixiviated, and green vitriol and alum added to the solution, whereby a greenish precipitate was thrown down, which, on treatment with hydrochloric acid, yielded the blue colour. 1 About the same time John Brown found that animal flesh could be employed instead of blood, and Geoffroy, in 1725, showed that other animal matters could be used for the same purpose. Up to this time very remarkable opinions were held concerning the composition of this colouring matter. Geoffroy, for example, assumed that the colour was due to metallic iron which, owing to the presence of the alum, was in an extremely fine state of division. In 1752 Macquer found that Prussian blue could be manufactured without the use of alum, and that when the colouring matter is boiled with an alkali, oxide of iron remains behind, and a peculiar body is formed which is found in solution. To this substance (ferrocyanide of potassium) the name of phlogisticated potash was given, and it was believed that the iron contained in the Prussian blue was coloured blue by a peculiar combustible substance. During the years 1782-5 Scheele occupied himself with the investigation of this compound, and he showed that when phlogisticated potash is distilled with sulphuric acid, a very volatile and inflammable body is formed, which is soluble in water, and possesses the property, when treated with alkalis and green vitriol, of forming a very beautiful blue colour. To this body the name of prussic acid was given. In 1787, the investigation of this compound was taken up by Berthollet, who proved that iron was contained in the prussiate of potash as an essential con- stituent, together with alkali and prussic acid ; and that prussic 1 Phil. Trans. 1724. 748 THE NON-METALLIC ELEMENTS acid, the salts of which yield ammonia and carbonic acid as pro- ducts of decomposition, consists of carbon, nitrogen, and hydrogen. These results were afterwards confirmed by the investigations of other chemists, but it is especially to the labours of Gay-Lussac that we owe a clear statement of the true composition of this acid and its salts. His researches are of great theoretical interest, as it is in them that we find the fact, for the first time , clearly pointed out, that a compound substance may exist which is capable of acting in many respects as if it were an element, Gay-Lussac 1 showed in 1811 that prussic acid is a compound consisting of hydrogen combined with a radical containing carbon and nitrogen, to which he gave the name of cyanoyem (from Kvavos dark blue and yewdco I produce). He proved, moreover, that the salts obtained by the action of prussic acid upon a base are compounds of this cyanog&ne with rnetals. In corroboration of this view, in 1815 he was able to show that this radical cyanogen can be prepared and is capable of existing in the free state. By the name of compound radical we signify a group of atoms which plays the part of a single atom, or, employing Liebig's classical definition, we may say that cyanogen is a radical, because (1) " It is a never-varying constituent in a series of compounds : (2) "It can be replaced in these compounds by other simple bodies : (3) " In its compounds with a simple body, this latter may be easily separated or replaced by equivalent quantities of other simple bodies." " Of these three chief and characteristic conditions of a com- pound radical, at least two must be fulfilled if the substance is to be regarded as a true compound radical." 2 In addition to cyanogen, a number of other compound radicals are known, some of which have already been mentioned. These radicals are distinguished one from another, as the elements are, by their quantivalence. Thus, nitrogen peroxide NO 2 is a monad radical occurring in nitric acid and in many of its derivatives, and to it the name of nitroxyl has been given. Its constitution is represented as follows : Annales de Chimie, 77, 128 ; 95, 136. 2 Annalen, 25, 3. CYANOGEN GAS 749 indicating that the two atoms of dyad oxygen are connected with the nitrogen each by two of the combining units of the pentad atom of the latter. Hence, one combining unit of the nitrogen remains free, and the group acts consequently as a monad radical. Sulphur dioxide or sulphuryl, SO 2 , serves as an instance of a dyad radical. This radical is contained in sulphuric acid, and a large number of compounds derived from it. It is termed a dyad because, of the six combining units of the sulphur atom, four only are saturated by combination within the molecule, and its constitution may, therefore, be represented by the following formula : ,O Many phosphorus compounds contain a triad radical, phos- phoryl PO, which, however, is not known in the free state. It contains the pentad phosphorus atom combined with the dyad oxygen atom, and its constitution can be represented as follows : Cyanogen, ON, is a monad radical, as it contains the triad atom of nitrogen and the tetrad atom of carbon ; its constitution is represented thus : N = C The radical cyanogen, to which the symbol Cy. instead of the formula ON is sometimes given, combines like chlorine with hydrogen and the metals, and its compounds may, there- fore, be compared with those of chlorine, to which they have the strongest resemblance in their physical and chemical properties : Hydrocyanic acid, CyH. Hydrochloric acid, C1H. Potassium cyanide, CyK. Potassium chloride, C1K. Potassium cyanate, CyOK. Potassium hypochlorite. C1OK. Free cyanogen, Cy 2 . Free chlorine, C1 2 . CYANOGEN GAS, OR DICYANOGEN, C 2 N 2 or Cy 2 . 436 This gas, which was discovered by Gay-Lussac, is formed when the cyanides of mercury, silver, or gold are heated. For 750 THE NON-METALLIC ELEMENTS this purpose it is usual to employ mercuric cyanide, which decomposes thus : Hg(CN) 2 = Hg + C 2 N 2 . The mercury salt placed in a tube of hard glass fitted with a cork and gas-delivery tube, or in a small hard glass retort, is heated to dull redness, and the gas collected over mercury. The heat of combination of mercuric cyanide is large, and a very high temperature therefore is required to bring about this decomposition. If mercuric chloride be mixed with the cyanide, the cyanogen comes off at a lower temperature, as the mercury then combines with the mercuric chloride, forming mercurous chloride, and the whole reaction, represented by the equation Hg (ON), + HgCl 2 = (CN) 2 + 2HgCl, takes place with slight evolution of heat. Cyanogen gas can also be obtained by other reactions. For instance, it is formed when oxalate of ammonium is heated with phosphorus pentachloride ; thus : C 2 4 (NH 4 ) 2 = c 2 N 2 + 4H 2 0. It is also formed by gently igniting an intimate mixture of two parts of well-dried ferrocyanide of potassium, Fe(CN) f; K 4 , with three parts of mercuric chloride. 1 According to Ber- zelius, however, the gas thus prepared contains free nitrogen, and he recommends the employment of potassium cyanide in place of the ferrocyanide. 2 There is no doubt that in this reaction mercuric cyanide is first obtained, and this is then decomposed as already described. Cyanogen is most readily obtained by gradually adding a concentrated solution of potassium cyanide to a solution of two parts of crystallized copper sulphate in four parts of water, and then warming; cyanogen is evolved and cuprous cyanide simultaneously separates out from the solution. If the cuprous cyanide be washed by decantation, and then treated with ferric chloride, the remainder of the cyanogen is evolved. 3 Cyanogen gas is found in small quantities in the gases pro- ceeding from the blast furnaces, 4 and it is likewise formed when a mixture of ammonia and coal-gas is burnt in a Bunsen 1 Kemp, Phil. Mag. 23, 179. 2 Berzelius, Jahresb. 24, 84. 3 Jacquemin, Ann. Chim. Phys. [6], 6, 1-40. 4 J. Pr. Chem. [1], 42, 145. CYANOGEN GAS 751 burner. 1 The following equation probably explains the reaction which takes place in the latter case : 4CO + 4NH 3 + .0 2 = 2C 2 N 2 + 6H 2 O. Cyanogen is a colourless gas possessing a peculiar pungent odour resembling that of peach kernels. It is poisonous, and burns, when brought in contact with flame, with a charac- teristic purple-mantled flame, with formation of carbon dioxide and nitrogen. When cyanogen gas is mixed with an excess of oxygen and an electric spark passed through the mixture, an explosion occurs, and, on cooling, the gas possesses exactly the same volume as before the explosion ; thus : C 2 N 2 + 20 2 = N 2 + 2C0 2 . Cyanogen is decomposed when subjected to the action of electric sparks, carbon being deposited, and small quantities of paracyanogen formed. The specific gravity of the gas is 1*806 (Gay-Lussac) ; when exposed to pressure or to cold, cyanogen condenses to a colour- less liquid which boils at 20'7, and possesses at 17 a specific gravity of 0'866. When cooled to a still lower temperature the liquid freezes, forming a crystalline mass which melts at 34'4. 2 According to Bunsen 3 the tension of cyanogen gas at different temperatures is as follows : Temperature. Tension in atmospheres. -207 1 -10 1-85 2-7 + 10 3-8 15 4-4 20 5-0 At the ordinary atmospheric temperature, one volume of cyanogen dissolves in four volumes of water, and in twenty - three volumes of alcohol. The aqueous solution soon becomes turbid, with separation of a brown powder known as azulmic acid, and the solution contains ammonium oxalate, together with smaller quantities of ammonium carbonate, hydrocyanic acid, and urea (Wohler). The formation of these products of decomposition will be explained later (Vol. III., Part ii., 2nd 1 Levoir, Jahresb. 76, 445. 2 Davy and Faraday, Phil Trans. 1823, 196. 3 Pogg. Ann. 46, 101. 752 THE NON-METALLIC ELEMENTS edition, pp. 139, 343). Cyanogen gas when passed over heated potassium unites directly with the metal, forming potassium cyanide. Paracyanogen, (CN) n . In the preparation of cyanogen from mercuric cyanide a brown powder remains behind. This sub- stance possesses the same percentage composition as cyanogen, but its molecular weight is unknown ; when it is heated in a current of carbon dioxide or nitrogen it gradually volatilizes, being converted into cyanogen gas. HYDROCYANIC ACID, OR PRUSSIC AciD, 1 HCN. 437 We owe the discovery of hydrocyanic acid to Scheele in 1782. It was subsequently examined by Ittner, 2 and by Gay- Lussac ; the former of these chemists preparing it for the first time in the anhydrous condition. 3 For this purpose he heated mercuric cyanide with hydrochloric acid in a flask, leading the gas, which was evolved, through a long tube, half filled with pieces of marble in order to retain any vapours of hydrochloric acid which might be carried over, the second half of the tube being filled with chloride of calcium for the purpose of absorbing any aqueous vapour. The gas, thus dried, was passed into a receiver placed in a freezing mixture, where it was condensed. According to this method only two-thirds of the theoretical amount of acid are obtained, as the mercuric chloride unites with a portion of the prussic acid to form a compound which is decomposed only at a high temperature. If, however, sal- ammoniac be added, this salt combines with the mercuric chloride, and the whole of the hydrocyanic acid is set free. 4 Pure hydrocyanic acid is also obtained by leading sulphu- retted hydrogen gas over dry mercuric cyanide heated to about 30. The salt is placed in a long horizontal tube of which the front portion is filled with carbonate of lead for the purpose of retaining the excess of sulphuretted hydrogen ; as soon as this begins to turn black the evolution of sulphuretted hydrogen is stopped (Vauquelin). 5 Hydrocyanic acid is generally prepared from ferrocyanide of potassium, which, as Scheele showed, is decomposed by dilute sulphuric acid. A process described by Wohler enables 1 The name acide prussique was first used by Guyton de Morveau. 2 Beitrage z. Geschichte d. Blausaure, 1809. 3 Ann. Uhim. 78 128, and 95, 136. 4 Bussy and Buignet, Ann. Ohim. Phys. [4], 3, 232. 5 Annalen, 73, 218. HYDROCYANIC ACID 753 us to prepare any desired quantity, either of the aqueous acid of different strengths, or of the anhydrous compound. A cold mixture of 14 parts of water and 7 parts of concen- trated sulphuric acid is poured upon 10 parts of coarsely powdered ferrocyanide of potassium contained in a large retort the neck of which is placed upwards in a slanting direction. If the anhydrous acid be required, the vapour is allowed to pass through cylinders or U-tubes filled with calcium chloride and surrounded by water heated to 30, after which the gas, thus dried, is condensed by passing into a receiver surrounded by ice or a freezing mixture. To prepare the aqueous acid the drying apparatus may be omitted and the gas evolved from a flask as shown in Fig. 201, passed through a Liebig's condenser, and then FIG. 201. led into the requisite quantity of distilled water. The action of the dilute sulphuric acid upon the prussiate of potash is represented by the following equation : 2Fe(CN) 6 K 4 -!-3H 2 S0 4 = 3K 2 S0 4 + Fe(CN) 6 K 2 Fe + 6HCN. As is seen from this equation, only half of the cyanogen con- tained in the ferrocyanide is obtained as hydrocyanic acid, but notwithstanding this loss the method is simpler and cheaper than those previously described. Amongst the various other methods which have been described for the preparation of this acid, one proposed by Clarke may be mentioned, inasmuch as by this method a dilute acid may readily be prepared having an approximately known strength. For this 49 754 THE NON-METALLIC ELEMENTS purpose nine parts of tartaric acid are dissolved in sixty parts of water and the solution poured into a iiask so that it is nearly rilled. Four parts of potassium cyanide are then added, and the flask closed by means of a cork. The mixture is then well shaken, and allowed to stand for some time until the whole of the cream of tartar (acid tartrate of potassium) has separated out, after which the dilute hydrocyanic acid is poured off. This should contain 3*6 per cent, of hydrocyanic acid, and only a small amount of cream of tartar in solution ; the reaction being represented by the following equation : KCN + C 4 H 6 6 = HCN + C 4 H r) KO c . Hydrocyanic acid is also formed when a series of electric sparks is passed through a mixture of acetylene and nitrogen ; L thus : C 2 H 2 + N 2 = 2HCN. It is also obtained by the passage of the silent discharge through a mixture of cyanogen and hydrogen. 2 438 Properties. Pure anhydrous hydrocyanic acid is a colour- less very mobile liquid possessing a characteristic smell resembling that of bitter almonds, and producing when inhaled in small quantities a peculiar irritation of the throat. Its specific gravity is 07058 at 7 and 0'6969 at 18 (Gay-Lussac) ; it boils at 261, and forms a colourless vapour which possesses the specific gravity 0'947. At 15 the liquid solidifies to a mass of colour- less feathery crystals. If a drop of the liquid acid be brought on a glass rod into the air, evaporation takes place so quickly, and so much heat is thereby absorbed, that a portion of the liquid freezes. Hydrocyanic acid is miscible in all proportions with water, alcohol, and ether. A very singular phenomenon is observed when hydrocyanic acid is mixed with water, inasmuch as a diminution of temperature occurs without any increase of the volume of the liquid ; on the contrary, a diminution of volume takes place, and this is the greatest when equal volumes of water and acid are mixed. In this mixture the proportions correspond to the relations, 3H 2 O + 2HCN, and, in this case, the diminution in bulk which occurs is from 100 to 94 (Bussy and Buignet). The anhydrous acid as well as its concentrated aqueous solution is easily inflammable, and burns with a beauti- ful violet flame. When not absolutely pure, it decomposes very 1 Berthelot, Com.pt. Rend. 77, 1141. 2 Boillot, Conipt. Rend. 76, 1132. HYDKOCYANIC ACID 755 readily on exposure to light, with separation of a brown body, and formation of ammonia, whilst, in the case of the pure dilute acid, formic acid, ammonia, and other bodies are produced. The tendency to decomposition is much diminished by the presence of traces of mineral acid or of formic acid. Concentrated mineral acids, as well as boiling alkalis, decompose hydrocyanic acid into ammonia and formic acid, water being taken up ; thus : HCN + 2H 2 = NH 3 + HO.COH. On the other hand, hydrocyanic acid and water are formed when the ammonium salt of formic acid is quickly heated. Pure hydrocyanic is one of the most powerful and rapid of known poisons. When a small quantity of the vapour of the pure substance is drawn into the lungs instant death ensues ; small quantities produce headache, giddiness, nausea, dyspnoea, and palpitation. A few drops brought into the eye of a dog kill it in thirty seconds, whilst an internal dose of 0'05 grain is usually sufficient to produce fatal effects upon the human subject, but cases have been known in which O'l grain has been taken without death ensuing. As antidotes, ammonia and chlorine water have been proposed, and these appear to be efficacious, although we are unable to explain their mode of action, for ammonia under ordinary conditions only forms ammonium cyanide, and chlorine cyanogen chloride, both of which are bodies as poisonous as prussic acid itself. Prussic acid is used as a medicine, and is a constituent of several officinal preparations, such as laurel water, bitter-almond water, &c., \\Iiich are obtained by distilling the leaves of the common laurel, or bitter almonds with water. These plants do not contain the prussic acid ready formed ; but, in common with most of the plants of the same family, contain amygdalin, a complicated compound which, under certain cir- cumstances, splits up into sugar, oil of bitter almonds, and prussic acid. 439 To estimate the quantity of hydrocyanic acid contained in these preparations, an exce&s of potash solution is added to a measured or weighed quantity of the liquid, and then by means of a burette a solution of nitrate of silver containing 6 '3 gram, in one liter is dropped in until a permanent precipitate appears. Each cc. corresponds to two milligrams of anhydrous prussic acid. In this reaction the double cyanide, AgCN f KCN, is formed, which is not decomposed by alkalis, and is soluble in water. 756 THE NON-METALLIC ELEMENTS Hence as soon as exactly half the quantity of prussic acid present is converted into silver cyanide, one drop more of the silver solution will produce a permanent precipitate of silver cyanide. 1 To detect hydrocyanic acid, as in cases of poisoning, the suspected matter is acidulated with tartaric acid, and the prussic acid distilled otf by means of a water-bath. The distillate is made alkaline with caustic soda, and a mixture of a ferrous and ferric salt (a solution of ferrous sulphate oxidized by exposure to the air) is added, and then an excess of hydrochloric acid. Prussian blue remains undissolved if prussic acid be present. If the quantity contained be very small, the solution appears first of a green colour, and, on standing, deposits dark blue flakes. When dilute prussic acid is mixed with yellow ammo- nium sulphide, and the liquid evaporated to dryness over a water-bath, ammonium thiocyanate is formed. The presence of this body is made known by means of ferric chloride, which pro- duces in a solution of thiocyanate a deep blood-red coloration. 2 THE CYANIDES. 440 Hydrocyanic acid turns blue litmus paper red, but it is so weak an acid that its soluble salts are decomposed by the carbonic acid of the air, and they therefore smell of hydrocyanic acid, and have an alkaline reaction even when their solution contains excess of hydrocyanic acid. Cyanides can be prepared in a variety of ways. Carbon and nitrogen do not unite together even under the action of the electric spark, but if these elements are heated in presence of an alkali, a cyanide is formed Thus, for instance, potassium cyanide is obtained when nitrogen is led over a heated mixture of carbon and potash : N 2 + 4C + K. 2 C0 3 = 2KCN + SCO. If, instead of potash caustic baryta is taken, barium cyanide is obtained : N 2 + 3C + BaO = Ba (CN) 2 + CO. According to Kuhlman, ammonium cyanide is formed when ammonia is passed over red-hot charcoal : 2NH 3 + C = NH 4 CN + H 9 . 1 Liebig, Annalen. 71, 102. 2 Liebig, Annalen, 61, 127. THE CYANIDES 757 Cyanides are also easily formed when nitrogenous organic bodies such as hoofs, clippings of hides, wool and blood, are heated with an alkali such as potash. The potassium cyanide thus obtained is prepared on the large scale, and forms the starting-point of the cyanogen compounds. As this substance is difficult to purify on account of its great solubility, it is con- verted into ferrocyanide of potassium or yellow prussiate of potash, which is used for the preparation of the other cyanogen compounds. p The cyanides of the alkali metals, and those of the metals of the alkaline earths, are soluble in water, and smell, as has been already stated, of hydrocyanic acid, as they are decom- posed by the atmospheric carbonic acid. They are, therefore^ just as poisonous as the free acid itself. The cyanides of the other metals, with the exception of mercuric cyanide, are insoluble in water. They dissolve, however, in the cyanides of the alkali metals, with the formation of soluble double cyanides. These double cyanides may be divided into two distinct classes. The compounds of the first class are easily decomposed by dilute acids with the formation of an insoluble metallic cyanide, and free hydrocyanic acid ; and possess all the usual properties of the cyanides ; of these, silver-potassium cyanide and nickel-potassium cyanide may serve as examples. AgCN.KCN + HNO 3 = AgCN + HCN + KNO 3 . Ni(CN) 2 . 2KCN + 2HC1 = Ni(CN) 2 + 2HCN + 2KC1. . The double cyanides of the second class, although containing, like the others, two different metals, possess wholly distinct properties. The best known of these are the yellow prussiate of potash or ferrocyanide of potassium, Fe(CN) 2 4KCN, and the red prussiate of potash or ferricyanide of potassium, Fe(CN) 3 3KCN. If to a solution of the first of these salts dilute hydro- chloric acid be added, no hydrocyanic acid is evolved, but a w r hite crystalline precipitate is obtained having the composition H 4 Fe(CN) 6 . This compound is a powerful acid, to which the name of fcrrocyanic acid has been given, and in which the four atoms of hydrogen can be either partially or wholly replaced by metals. In the same way the potassium can also be replaced in the red prussiate of potash by hydrogen, and thus fcrricyanic acid, H 3 Fe(CN) 6 , can be obtained. Similar double compounds are known, especially those of cobalt, manganese, platinum, and of the metals allied to it. 758 THE NON-METALLIC ELEMENTS In these compounds the second metal does not act as a base, but has combined with the cyanogen to form an acid radical (for example, Fe(CN) 6 ), which then unites with hydrogen and metals in the same manner as other acid radicals such as chlorine. (See also Vol. II., under Iron.) 441 Detection and determination of cyanogen and the cyanides. The detection of cyanogen in the cyanides, such as those of the alkalis and alkaline earths, which are easily decomposed by dilute acid, is simple enough. If the solution is not already alkaline, a solution of caustic soda is added, and next, a few drops of a solution of ferrous sulphate (green vitriol) which has been partially oxidized by exposure to the air ; an excess of hydro- chloric acid is then added, when a precipitate of Prussian blue is thrown down. In this reaction the ferrous salt in the presence of caustic soda and a cyanide gives rise to the form- ation of sodium ferrocyanide, which then undergoes the usual reaction with the ferric salt which is present. Cyanides, which do not evolve hydrocyanic acid on the addition of a dilute acid, may be decomposed by fusing them with dry carbonate of soda. The fused mass is then dissolved in water, and the filtered liquid treated according to the above-mentioned process. In the soluble cyanides, as well as in those which are de- composed by dilute acids, the quantity of cyanogen can readily be determined by precipitation with nitrate of silver. A precipi- tate of cyanide of silver is obtained, which, after washing and drying at 110, is weighed. This method, however, cannot be employed in the cases of mercuric cyanide, potassium ferro- cyanide, and analogous substances, in which cases it is necessary to obtain hydrocyanic acid from the compound by some ap- propriate method, such as treatment with sulphuretted hydrogen, distillation with dilute sulphuric acid, &c. Concerning the further details of the qualitative and quantitative analysis of such compounds we must refer to treatises on analytical chemistry. COMPOUNDS OF HYDROCYANIC ACID WITH THE HYDRACIDS OF THE CHLORINE GROUP OF ELEMENTS. 442 Although hydrocyanic acid possesses weak acid properties, yet it behaves in many respects as a weak base which, like ammonia, is capable of direct union with certain acids. This property depends upon the fact that its chemjpal constitution is CYANOGEN CHLORIDE 759 analogous to that of ammonia, for hydrocyanic acid is formed when ammonia is heated with chloroform under pressure; thus : CHC1 3 + 4NH 3 = NCH + 3NH 4 C1. According to this mode of formation, hydrocyanic acid may be considered to be ammonia in which the three atoms of hydrogen have been replaced by the triad radical methenyl OH, whence it might be termed methenylamine. Hydrocyanic acid forms two compounds with hydrochloric acid. The monohydrochloridc HCN,HC1 is prepared by passing hydrogen chloride into anhydrous hydrocyanic acid at 15 and subsequent warming in a sealed tube to 35-40. It is a crystalline very hygroscopic substance, which is decomposed by water into ammonium chloride and formic acid. 1 The sesquihydrochloride 2HCN,3HC1 is obtained by passing hydrogen chloride into a mixture of anhydrous hydrocyanic acid and ethyl acetate, and forms a very hygroscopic crystalline mass, which yields with water the same products as the previous compound. 2 Similar derivatives have been obtained with hydrobromic acid, having the composition HCN,HBr and 2HCN,3HBr respectively. The hydriodide HCN,HI has also been prepared. Hydrocyanic acid also combines with metallic chlorides to form crystalline compounds, as SbCl 5 + 3HCN, SnCl 4 + 2HCN and TiCl 4 -I- 2HCN. COMPOUNDS OF CYANOGEN WITH THE ELEMENTS OF THE CHLORINE GROUP. CYANOGEN CHLORIDE, C1CN = 61-04. 443 Berthollet obtained this compound in 1787 by the action of chlorine on hydrocyanic acid, and believed it to be oxy- genated hydrocyanic acid. Its true composition was recognised by Gay-Lussac in the year 1815. According to Wohler 3 this compound is easily obtained by passing chlorine gas into a saturated aqueous solution of mercuric cyanide containing some of the solid salt, until the liquid is saturated with the gas, and the air which remains above the liquid is replaced by chlorine. 1 Gautier, Ann. Chim. Phys. [4], 17, 129. 2 Claisen and Matthews, Ber. 16, 309. 3 Ann. Chim. Phys. 95 ? 200. 760 THE NON-METALLIC ELEMENTS The vessel is then well closed and placed in a dark room till all the mercuric chloride has dissolved, or all the chlorine has combined. In order to remove any free chlorine which may remain the solution is shaken up with mercury, and the liquid heated in order to expel the cyanogen chloride, the volatile product being condensed by passing through a bent tube plunged into a freezing mixture. According to Gautier l the liquid chloride is best obtained by leading chlorine into a mixture of one part of hydrocyanic acid and four parts of water contained in a retort or flask which is connected with an inverted Liebig's condenser. As soon as the liquid has attained a green colour, the current of chlorine is stopped, and an excess of mercuric oxide and calcium chloride is added to the liquid, which is well cooled down in a freezing mixture. The chloride of cyanogen is next distilled off and condensed in a well-cooled receiver. At the ordinary temperature cyanogen chloride is a gas which has a very penetrating odour and causes a copious flow of tears ; it is very readily condensed to a liquid, and freezes at 18, forming colourless prisms, which according to Regnault 2 melt at 7, the liquid formed boiling at 127, whilst according to Wurtz 3 it melts at 12 to 15 and boils at 15*5. Its vapour density is 2*13. 4 When allowed to stand it is partially converted into the polymeric cyanuric chloride C 3 N 3 C1 3 (p. 766). CYANOGEN BROMIDE, CNJBr. 444 This compound, which was discovered in the year 1827 by Serullas, 5 is formed by the action of bromine on hydrocyanic acid, or on the metallic cyanides. If bromine is added drop by drop to a well-cooled aqueous solution of potassium cyanide, crystals separate out, which consist of a mixture of cyanogen bromide and potassium bromide. 6 When these crystals are heated to a temperature of from 60 to 65, cyanogen bromide sublimes in the form of delicate transparent prisms, which soon pass into the cubical form. It melts at 52 and boils at 61'3 under 750 mm. pressure, 7 has an exceedingly pungent smell, and acts very powerfully upon the eyes. It is also poisonous ; one grain 1 Bull. Soe. Chim. [2], 403 2 Jahresb. 1863, 70. 3 Annalen, 79, 284. * Bull. Soc. Chim. [2], 4, 105. 5 Ann Chim. Phys. 34, 100. 6 Langlois, Ann. Chim. Phys. [3], 61, 482. 7 Mulder, Bee. Trav. Chim. 4, 151 ; 5, 85. CYANIC ACID 761 dissolved in a small quantity of water brought into the oesophagus of a rabbit produced instant death (Serullas). According to Bineau the specific gravity of its vapour is 3'607. When not absolutely pure cyanogen bromide is converted at 130 to 140 into cyanuric bromide, CgNgBrg. 1 CYANOGEN IODIDE, CNI. 445 This compound was discovered by Davy in the year 1816, 2 and is easily formed by the action of iodine on cyanide of mercury and other metallic cyanides, whilst it frequently occurs as an impurity in commercial iodine. According to Liebig it can be best obtained by dissolving iodine in a warm concentrated solution of potassium cyanide until the liquid solidifies on cooling to a crystalline mass. Cyanogen iodide forms long delicate white needles, easily soluble in alcohol and ether, and from these solutions the compound crystallizes in four-sided tables. It is sparingly soluble in water. It melts at 146'5, but begins to volatilize at a much lower temperature. It has a smell similar to that of the bromide, and is equally poisonous. CYANIC ACID, NCOH. 446 Cyanic acid was first observed by Vauquelin in the year 1818, although it was first distinctly recognised as a peculiar acid and its properties more exactly investigated by Wohler in 1822. Its salts are formed when cyanogen gas is led into an alkali; thus: When potassium cyanide is melted in the presence of air, or, better, with the addition of an easily reducible oxide or peroxide, potassium cyanate is formed. The most convenient method of preparing the cyanates is by the action of potassium bichromate on potassium ferrocyanide. For this purpose 200 grams of completely dehydrated potassium ferrocyanide are mixed whilst still warm with 150 grams of fused potassium bichromate, and the whole heated in an iron dish. The black product is then quickly extracted with a 1 Eghis, Zeitsch. Chem. [2], 5, 376 ; Ponamarew, Ber. 18, 3261. 2 Gilb. Ann. 54, 384. 762 THE NON-METALLIC ELEMENTS mixture of 800 cc. of 80 per cent, etrryl alcohol and 100 cc. of methyl alcohol, potassium cyanate separating from the nitrate on cooling. 1 Cyanates are readily decomposed by dilute acids, but the cyanic acid at the same time takes up water, carbon dioxide and ammonia being formed. Therefore on adding a dilute acid to potassium cyanate effervescence takes place, carbon dioxide being given off. This has a pungent smell, due to the presence of a trace of cyanic acid. By acting on a cyanate with dry hydrochloric acid, cyanic acid is set free. It at once combines, however, with hydrochloric acid to form the compound HC1 NCOH, a colourless liquid, fuming in the air. It is impossible to isolate cyanic acid from this compound, because it at once changes into the polymeric cyanuric acid C 3 N 3 (OH) 3 (p. 766). The only reaction by which cyanic acid can be obtained is by the decomposition of cyanuric acid by heat, when it is resolved into three molecules of cyanic acid, the vapours of which must be condensed by means of a freezing mixture. Cyanic acid is a colourless mobile liquid having a most pungent smell. On taking the vessel containing it out of the freezing mixture the liquid soon becomes turbid and hot, and with a crackling noise, or if in large quantity with explosive ebulli- tion, is soon converted into a white porcelain-like mass which is a polymeric modification, called cyamelidc, the molecular weight of which is unknown. On heating cyamelide it is reconverted into cyanic acid. Of the cyanates the most interesting salt is the ammonium cyanate, NCO(NH 4 ), obtained as a white crystalline mass by mixing the vapour of dry cyanic acid with dry ammonia. The freshly prepared aqueous solution gives the reactions of a cyanate and of ammonia, but on standing for some time, or on heat- ing, the ammonium cyanate is transformed into the isomeric carbamide or urea, CO(NH 2 ) 2 . The dry salt also undergoes the same transformation on heating, and the change may also be shown by heating a solution of potassium cyanate to which an equivalent quantity of ammonium sulphate has been added. THIOCYANIC ACID, NCSH. 447 This acid, to which the name of sidphocyanic acid has also been given, was mentioned by Bucholz in 1798, but first 1 Bell, Glum. News, 32, 49 ; Erdmann, Ber. 26, 2438. THIOCYAX1C ACID 763 carefully examined by Porret in 1808. He obtained its potas- sium salt by boiling a solution of potassium sulphide with Prussian blue. Its quantitative composition was ascertained by Berzelius in 1820. 1 Salts of this acid are formed by the direct union of sulphur with a cyanide. Thus, for instance, potassium thiocyanate is easily obtained by melting together potassium cyanide and sulphur or dried potassium ferrocyanide with potassium carbonate and sulphur. The ammonium salt is obtained by warming hydrocyanic acid with yellow sulphide of ammonium ; thus : 4 S * + HCN = Ammonium thiocyanate is easily obtained in large quantities by warming a mixture of alcoholic concentrated ammonia and bisulphide of carbon which has been allowed to stand for a con- siderable time. Ammonium thiocarbonate and thiocarbamate are formed, and these are decomposed on heating, with evolu- tion of sulphuretted hydrogen (Millon) ; thus : 4 = CS For this purpose it is best to take 350 400 grams of carbon bisulphide, 600 grams of 95 per. cent, alcohol, and 800 grams of ammonia (sp.gr. 0'912). As soon as the carbon bisulphide has dissolved, the solution is concentrated, and on cooling the salt separates out in crystals. 2 If a solution of mercuric nitrate be added to a solution of ammonium or potassium thiocyanate, a heavy white precipitate of mercuric thiocyanate, (CNS).,Hg, is thrown down. This substance when dry is easily inflam- mable, and burns with a pale sulphur flame, leaving a very voluminous residue behind. This salt is used for the prepara- tion of the so-called Pharaoh's Serpents. Aqueous thiocyanic acid may be readily obtained by decom- posing barium thiocyanate with dilute sulphuric acid. This solution may be distilled under diminished pressure, the acid coming over with the first portion, and if dried over calcium chloride forms a yellowish oil which soon decomposes. 3 On 1 Schweig. Journ. 31, 42. 2 Claus, Annalen, 179, 112. 3 Klason, J. Pr. Chem. [2], 36, 57. 764 THE NON-METALLIC ELEMENTS boiling the aqueous solution, the greater part of the acid is converted with absorption of water into carbonyl sulphide and ammonia, whilst at the same time some hydrocyanic acid is formed. Thiocyanic acid and its soluble salts are coloured blood-red by ferric chloride, ferric thiocyanate being formed. The soluble thiocyanates produce with silver nitrate a precipitate of silver thiocyanate which is insoluble even in presence of the stronger acids. A solution of potassium or ammonium thiocyanate of known strength is, therefore, used for the purpose of determining the quantity of silver contained in a solution, by adding some nitric acid and ferric sulphate and then dropping in a standard solution of ammonium thiocyanate from a burette until the formation of a slight red colour shows that all the silver has been thrown down. 1 CYANOGEN SULPHIDE, OR THIOCYANIC ANHYDRIDE, (CN) 2 S. 448 In order to prepare this compound, silver thiocyanate is added to an ethereal solution of iodide of cyanogen, the solution evaporated, and the residue treated with hot carbon bisulphide. 2 If the solution be cooled to 0, cyanogen sulphide separates in rhombic tables, which possess a smell like that of cyanogen iodide. They melt at 60, but when heated above 30 they begin to volatilize. They are very soluble in water, alcohol, and ether, and are decomposed by caustic alkalis as follows : K) CN) S ,CN\ .... H/ ( Kj" k Kj C H 2' Cyanogen Selenide, (CN) 2 Se, is obtained by adding dry silver cyanide to a solution of selenium bromide in bisulphide of carbon. It crystallizes in tables, which are decomposed by water into selenium, selenious acid, and hydrocyanic acid. 3 CYANAMIDE, CN.NH 2 . 449 This body was discovered by Bineau, 4 who prepared it by the action of dry ammonia on gaseous cyanogen chloride. The solid mass thus obtained was supposed by him to be the 1 Volhard, J. Pr. Chem. [2], 9, 217. 2 Linnemann, Annalen, 70, 36. 3 Schneider, Pogg. Ann. 129, 364. 4 Ann. Chim. Phys. [2], 67, 368, and 70, 251. CYANAMIDE 765 chloride of cyanammonium, C1CN(NH 3 ), but Cloez and Can- nizzaro 1 showed in 1851 that this substance consists of a mixture of ammonium chloride and cyanamide : C TT 2 N-^H (H H Cyanamide is most readily obtained, however, by the action of mercuric oxide upon a cold, not too dilute, solution of thio- urea 2 : CS(NH 2 ) 2 + HgO = CN(NH 2 ) + HgS + H 2 O. The mercuric oxide must be very finely levigated, or a dense precipitated oxide may be employed ; this is suspended in water and gradually added to the solution of thio-urea ; an excess of the oxide must most carefully be avoided. The end of the reaction may be easily recognised, a drop of the solution being brought on to filter paper, and then a drop of ammoniacal silver solution added ; this produces a black spot as long as thio-urea is present. As soon as all the thio-urea has been decomposed the solution is filtered and evaporated down on a water-bath, and the residue treated with ether ; on evapora- tion of the ethereal solution, pure cyanamide is left behind. It forms small colourless crystals, melting at 40, which deliquesce on exposure to the air, and are volatile with steam. If a few drops of nitric acid be added to its solution, cyanamide is trans- formed into urea : CN.NH 2 + H 2 =CO(NH 2 ) S . Cyanamide acts as a weak base; the hydrochloride CN.NH 2 , 2HC1 is obtained as an easily crystallizable powder when hydrochloric acid gas is passed into a solution of cyanamide in anhydrous ether (Drechsel). The hydrogen of the cyanamide may also be replaced by metals ; thus, for instance, if an am- moniacal silver solution be added to an aqueous solution of cyanamide, an amorphous yellow precipitate, CN.NAg 2 , is thrown down, which crystallizes from hot ammonia in micro- scopic needles. In the same way, if one part of sodium be dissolved in 15 parts of absolute alcohol, and to the cold 1 Compt. Rend. 32, 62. 2 Volhard, J. Pr. Chem. [2], 9, 6 ; and Drechsel, J. Pr. Chem. [2], 9, 284. 766 THE NON-METALLIC ELEMENTS liquid an alcoholic solution of cyanamide be added, a light crystalline powder falls down, possessing, according to Drechsel, the composition CN.NHNa. When cyanamide is strongly heated it undergoes polymerisa- tion, first forming dicyanamide, C 2 N 4 H 4 , and finally cyanuramide, C 3 N 6 IV POLYMERISATION OF CYANOGEN COMPOUNDS. 450 Mention has frequently been made of the fact that many of the cyanogen derivatives readily undergo polymerisation that is, are converted into compounds having the same percentage composition but a higher molecular weight. Thus cyanogen chloride readily passes into cyanuric chloride C 3 N 3 C1 3 , cyanic acid into cyannric acid C 3 N 3 (OH) 3 , and cyanamide into dicy- anamide C 2 N 4 H 4 and cyanuramide C 3 N 6 H 6 . These compounds have very different properties from the cyanogen derivatives from which they are produced, and contain in most cases a nucleus of carbon and nitrogen atoms united together in such a manner as to form a closed chain. Thus the cyanuric com- / N -\ pounds all contain the nucleus C\ ;N, and from them \N C/ other substances containing the same nucleus can be prepared. A very large number of compounds containing similar closed chains of carbon and nitrogen atoms are now known, and they form a very important division of the carbon compounds usually considered under the head of Organic Chemistry; the poly- merised cyanogen compounds will therefore be more fully treated of together with these in a later volume. GUANIDINE, CH 5 N 3 . 451 This powerful base was first obtained by Strecker 2 by acting with potassium chlorate and hydrochloric acid on guanine, C 5 H 5 N 5 O, a compound contained in guano. It can also be ob- tained by various other reactions. For instance, it is produced by heating to 100 chloropicrin (nitrochloroform), CC1 3 NO.,, a substitution product of marsh gas with an alcoholic solution of ammonia ; 3 thus : CC1 3 N0 2 4- 3NH 3 - CH 5 N 3 .HC1 + 2HC1 + HN0 2 . 1 Annalen, 122, 22. 2 Annalen, H8, 151. 8 Hofmann, Zeitsch. Chem. [2], 2, 1073, and 4, 721. GUANIDINE 767 As an excess of ammonia must be used, no nitrous acid is liberated, water and free nitrogen being formed ; for this reason it is necessary to use strong glass tubes in order to withstand the accumulated pressure. Hydrochloride of guanidine is also formed when an alcoholic solution of cyanamide is heated with sal-ammoniac 1 : CN 2 H 2 + NH 4 C1 = CN 8 H 8 01. It is, however, not necessary to use pure cyannmide, but cyano- gen chloride may be heated with alcoholic ammonia, or, still better, cyanogen iodide may be employed, as this latter forms iodide of ammonium, which is easily soluble in alcohol. 2 When ammonia acts upon carbonyl chloride urea is formed, and at the same time small quantities of cyanuric acid, ammelide, and guanidine. 3 The best method, however, of preparing guanidine is from ammonium thiocyanate, 4 which must be heated for twenty hours to a temperature of 190, when the thio-urea which is formed decomposes, with evolution of sulphuretted hydrogen into cyanamide, and this is decomposed by the ammonium thiocyanate still present ; thus : CN(NH 2 ) + CN SNH 4 = CNH(NH 2 ) 2 .CNSH. A residue of guanidine thiocyanate is obtained, whilst large quantities of ammonium thiocarbonate sublime. Other guanidine salts can be obtained from the thiocyanate as well as the free base. In order to prepare these, 100 grams of the salt are dissolved in lukewarm water, and to the concentrated solution 58 grams of pure potassium carbonate are added, the solution evaporated to dryness, and the residue treated with alcohol in order to dissolve the potassium thiocyanate, whilst carbonate of guanidine, (CH 5 N 3 ) 2 .CO 3 H 2 , remains behind, and can be purified by recrystallization from aqueous solution. The carbonate yields, on treatment with acids, the other guanidine salts, of which the nitrate, CH 5 N 3 .NO 3 H, crystallizing in tablets, is remarkable from its slight solubility in water. The sulphate forms crystals tolerably soluble in water, and yields the free base when treated with baryta-water ; the solution may then be evaporated in a vacuum, and the base obtained in the form 1 Erlenmeyer, Zeitsch Chcm. [2], 7, 28. 2 Bannow, Bcr: 4, 161. 3 Bouchardat, Compt. Rend. 69, 961. 4 Volhard, J. Pr. Chem. [2], 9, 6, and Delitsch, ibid. 1. 768 THE NON-METALLIC ELEMENTS of a crystalline mass having a caustic taste and strongly alkaline reaction, which on exposure to the air quickly absorbs moisture and carbonic acid. When guanidine is heated with dilute sulphuric acid, it is resolved into urea and ammonia ; thus : C(NH)(NH 2 ) 2 + H 2 = NH 3 + CO(NH 2 ) 2 . A cold solution of potash decomposes guanidine thiocyanate in a similar way ; thus : CNH(NH 2 ) 2 'NSCH+KOH = CO(NH 2 ) 2 4-NCSK + NH 3 . These reactions, as well as all the modes of formation of guanidine, show that it possesses the following constitution: NH 2 I C = NH NH 2 . Nitroguanidine, NH : C(NH 2 )NH.NO 2 , is obtained by adding guanidine thiocyanate to concentrated sulphuric acid, then mixing with fuming sulphuric acid, and after cooling adding fuming nitric acid to the solution. When the resulting liquid is poured into a large volume of water, nitroguanidine separates out, and after recrystallization forms needles which melt at 230, with evolu- tion of ammonia. 1 Amidoguanidine, NH : C(NH 2 )NH.NH 2 , is prepared by the careful reduction of the preceding compound with zinc dust and the theoretical quantity of dilute acetic acid, the whole being carefully cooled. It is a crystalline compound soluble in water, and is converted by boiling with acids or alkalis into hydrazine (p. 470), for the preparation of which it is used. PHOSPHORUS TRICYANIDE, P(CN) 3 . 452 In order to prepare this compound a mixture of phos- phorus trichloride and dry silver cyanide is heated in closed tubes for several hours to a temperature of 120-140. When cold the tubes are opened, the excess of phosphorus trichloride driven off by moderate heating, and the residue brought into a tubulated retort, and heated in an oil-bath to 130-140 in a 1 Thiele, Annalen, 270, 1- X : ARSENIC TRICYANIl)Ev. v 769 stream of carbon dioxide. The phosphorus tricyanide sublimes in long glistening white needles or thick plates. It takes fire when slightly warmed, and burns in the air with a bright white flame ; it is decomposed by water into hydrocyanic and phos- phorous acids ; 1 thus : P(CN) 3 + 3H 2 O = P(OH) 3 + 3HCN. ARSENIC TRICYANIDE, As(CN) 3 . This substance is obtained by the action of finely divided arsenic on cyanogen iodide in presence of carbon bisulphide. It forms yellowish microscopic crystals, and is rapidly attacked by atmospheric moisture, and almost instantaneously by water, with formation of arsenious and hydrocyanic acids. 2 BORON CARBIDE OR CARBON BORIDE, BC OR B 2 C 2 . This is prepared by heating a mixture of boric anhydride and carbon in an electrical furnace, and forms a graphite-like powder which melts at a very high temperature, burns with difficulty in oxygen, is insoluble in the usual solvents, but is decomposed by fusion with alkalis. 3 Another boride of carbon, having the composition CB 6 , is prepared by heating sixty-six parts of amorphous boron and twelve parts of carbon, obtained from sugar, in an electrical furnace, or by dissolving the two elements in iron, copper, or silver at a very high temperature, and removing the metal by treatment with aqua regia. It forms very hard black lustrous crystals, having a density of 2*41, which are attacked by chlorine below 1,000, and by oxygen very slowly at that temperature ; like the previous compound, it is decomposed by fusion with alkalis. The powder is sufficiently hard to cut the diamond, but more slowly than diamond dust.* COAL-GAS. 453 It was noticed so long ago as 1726 that when coal is heated in a closed vessel an inflammable gas is evolved. In that year Stephen Hales published his Vegetable Staticks, in which 1 Wehrhane and Hiibner, Annalen, 128, 254, and 132, 277. 2 Guenez, Compt. Rend. 114, 1186. 3 Miihlhausen, Zeit. Anorg. Chem. 5, 92. 4 Moissan, Compt. Rend. 118, 556. 50 770 THE NON-METALLIC ELEMENTS lie states that by the distillation of 128 grains of Newcastle coal he obtained 180 cubic inches of an inflammable gas which weighed 51 grains. Bishop Watson, in his Chemical Essays, describes experiments made on coal-gas, arid mentions that it does not lose its illuminating power when it is passed through water. The first to apply these facts practically to the manu- facture of coal-gas was William Murdoch, a Scotchman living at Redruth in Cornwall. He distilled coal in an iron retort, and lighted his house with the gas which he thus manu- factured. Murdoch was afterwards employed in the celebrated engine works of Boulton and Watt at Soho near Birmingham. Whilst there he improved his process for the manufacture of gas, and in 1798 the Soho factory was for the first time lighted with coal-gas. In 1802 there was a public display of gas illumina- tion at Soho in honour of the Peace of Amiens ; and during the next three years Murdoch's process became so far perfected that in 1805 the large cotton mill of Messrs. Phillips and Lee in Manchester was lighted with gas. The success of this undertaking attracted the attention of several men of ability, especially Dr. William Henry and Mr. Clegg. To the former we owe the first accurate investigation of the chemical composition of coal-gas ; to the latter we are in- debted for many of the mechanical inventions still in use for its preparation and purification, without which the present enormous extension of the manufacture would have been impossible. The streets of London were not lighted with gas until 1812, and it was not introduced into Paris until after the peace of 1815. When coal is subjected to dry distillation at a temperature of a cherry-red heat, various products are formed. These may be divided into three classes. 1. Coal-gas, a mixture of many gaseous compounds. 2. Coal-tar, a thick, oily, strongly smelling liquid. 3. Gas-liquor or Ammoniacal Liquor, an aqueous distillate containing ammonium carbonate and sulphide, and other products in solution. The apparatus needed in the manufacture of coal-gas is: (1) the retorts in which the distillation occurs ; (2) the condensers in which the liquid products of the distillation are condensed and separated from the gaseous products ; (3) the washers and scrubbers in which the last portion of the liquid products and the ammonia are removed ; (4) the purifiers in which the remain- COAL-GAS 771 ing gaseous impurities are removed ; (5) the gas-holders in which the gas is stored, and whence it is distributed through the gas- mains and pipes to the place where it is to be burnt. FIG. 202. FIG. 203. FIG. 204. FIG. 205. FIG. 206. FIG. 207. 454 The Retorts originally introduced by Clegg in 1812 were made of iron, but these soon gave place to retorts made of fire-clay or fire-brick. Several common forms are shown in 772 THE NON-METALLIC ELEMENTS Figs. 202, 203, 204. At one time the retorts were always closed at one end, and this is still very often the case, especially in smaller works, but in the larger works longer retorts open at each end are now usually employed, these being known as through-retorts. The first-named retorts vary from 8 10 feet in length, 14 18 inches in breadth, and 12 -16 inches in height, and contain a charge of from 1 J 3 cwt. of coal. An iron mouthpiece, shown in section in Fig. 205 and in end elevation in Fig. 206, is attached to the open end by means of bolts, and when the retort has been charged the end of the mouthpiece is closed by an iron lid (Fig. 207), held in position by means of a holdfast and screw. An upright wide tube (d), cast COAL-GAS 773 on to the mouthpiece, carries away the gases, which pass up the vertical ascension pipes H H H H, Fig. 208, and thence by the dip pipes n n, Fig. 209, into the horizontal hydraulic main G G, Fig. 208. The retorts are arranged so that a number, varying usually from five to ten, are heated by means of one furnace, a bed of five retorts being shown in elevation in Fig. 208 and in FIG. 209. section in Fig. 209. The through-retorts are similar in shape, breadth, and height to those already described, but are double the length, and contain a correspondingly greater weight of charge. Mouthpieces, ascension pipes, &c., are attached in this case to each end of the retort, so that the gases formed can pass out at either end. Such retorts are frequently capable of car- 774 THE NON-METALLIC ELEMENTS bonising 24 cwt. of coal, and yielding 12,000 cubic feet of gas, per twenty-four hours. 455 The Condensing Apparatus. Large quantities of vola- tile products make their way from the retorts into the hydraulic main as gases, but are there condensed into liquids. The end of the upright pipe from each retort (n, Fig. 209), termed the dip-pipe, passes for about one to three inches under the surface of the liquid in the hydraulic main ; so that the gas as it is evolved can readily pass, but all entrance of air into the main when the retort is opened is prevented. The tarry products as well as the ammoniacal liquor, collecting in the hydraulic main, run off when they reach a certain height by a pipe which com- municates with the tar-well, where all the liquid products of the distillation are allowed to collect. Owing to the several water-joints which are necessary in various parts of the apparatus, the back pressure on the gas in the retorts is very considerable, and much loss of gas ensues owing to the necessarily porous nature of the fire-clay retort. To obviate this loss, and also to avoid the decomposition of the dense hydrocarbons into gas-carbon and marsh gas which takes place when the coal is distilled under pressure, an exhauster worked by an engine is employed to draw out the gas from the retorts, and pass it on through the condensers and purifiers, so that in the hydraulic main the pressure is always less than that of the atmosphere ; care must however be taken that the vacuum is not so great that air is drawn into the hydraulic main through the dip-pipe when the retort lids are opened for drawing the coke or charging with fresh coal. The gas after leaving the retort houses passes through the atmospheric condensers, which consist generally of a series of iron pipes arranged either vertically or horizontally. An ex- ample of a vertical condenser is shown in Figs. 210, 211. The gas in passing through the pipes is gradually reduced to the atmospheric temperature, and the more readily condensable hydro-carbons, &c v separate out in the form of tar ; a consider- able quantity of water is likewise condensed which dissolves a portion of the ammonia, sulphuretted hydrogen, carbonic acid, and other impurities present in the gas, yielding ammoniacal liquor. These fall to the lowest portion of the apparatus, and then pass away through siphons into the tar-well. After leaving the condensers the gas passes to the exhausters, and from this point onwards is under pressure instead of vacuum, COAL-GAS 775 arid is forced through the remainder of the apparatus as this throws a considerable amount of back pressure. 456 A considerable quantity of tarry matter is still retained in suspension in the gas, and to remove this and the remainder of the ammonia it is passed either through washers or scrubbers, or through both these successively. Numerous varieties of washers are employed, but the principle of almost all of these FIG. 210. FIG. 211. is the same, and consists in dividing the gas up into a number of small streams and forcing these through water or the weak liquor which separates out in the hydraulic main. By this means the tarry matter and most of the ammonia are removed, and the solution of ammonia simultaneously combines with a portion of the sulphuretted hydrogen, carbonic acid, and carbon bisulphide in the gas. To completely remove the 776 THE NON-METALLIC ELEMENTS ammonia, the gas is passed through the scrubbers, which consist of iron towers filled with coke down which water is allowed to trickle, thus exposing a very large wetted surface to the gas. The construction of these scrubbers is shown in Fig. 212, the gas entering at the bottom of the tower and passing away from the top, whilst the liquor formed falls to the bottom, and thence by means of siphons to the storage well. FIG. 212. 457 Purification. After leaving the scrubbers the gas still contains the following impurities : sulphuretted hydrogen, 1 2 per cent, or 600 1,200 grains per 100 cubic feet ; carbonic acid, 13 per cent, or 8002,400 grains per 100 cubic feet ; carbon bi- sulphide, 15 50 grains per 100 cubic feet, and about 7 8 grains per 100 cubic feet of other sulphur compounds. Carbonic acid COAL-GAS 777 acts injuriously on the gas, inasrnuch as it seriously deteriorates its illuminating power, and the sulphur compounds when burnt give rise to the formation of sulphur dioxide, and eventually of sulphuric acid, which acts prejudicially on metal and leather work exposed to its influence. In order to remove these FIG. 213. impurities as far as possible the gas is passed through layers of various solid absorbents which are placed in trays in large cast- iron boxes, generally termed the purifiers, the construction of which is shown in Fi^s. 213, 214. FIG. 214. Where no attempt is made to remove the sulphur compounds other than sulphuretted hydrogen, the gas is first passed through a series of boxes (usually three or four) containing hydrated oxide of iron, Fe 2 O 3 xH 2 O, which absorbs the sulphur- etted hydrogen, forming a sulphide of iron as follows : Fe 2 03,H 2 + 3H 2 S = Fe 2 S 3 + 4H 2 O. 778 THE NON-METALLIC ELEMENTS As soon as the gas passing the last box is found to contain sulphuretted hydrogen, the first box of the series is charged with fresh oxide of iron, and this then made the last of the series. The gas, after being thus freed from sulphuretted hydrogen, is passed through another series of purifiers contain- ing layers of moist slaked lime which absorbs the carbon dioxide forming calcium carbonate. When the " spent " oxide of iron is allowed to remain in a moist condition exposed to the air it gradually undergoes oxida- tion, and is reconverted into ferric oxide with separation of sulphur 2Fe 2 S 3 + 3O 2 = 2Fe 2 O 3 + 3S 2 and may then be again placed in the boxes to remove sulphur- etted hydrogen, and this process of conversion into sulphide and revivification continued until the spent oxide contains 50 60 per cent, of free sulphur, when it is sold to the sulphuric acid manufacturer. If a quantity of oxygen, equal to about half that of the sulphuretted hydrogen in the crude gas, be admitted into the purifiers, the above reaction goes on in the boxes simultane- ously with the formation of sulphide of iron, so that a por- tion of the spent oxide is revivified as fast as it is formed, and the boxes then last for a much greater length of time without changing, thus effecting a considerable saving in the cost of labour. It is immaterial for the purpose whether the oxygen be added in the pure state or in the form of air, but when the latter is used the inert nitrogen added to the gas causes some diminution of the illuminating power, and therefore necessitates the use of a certain amount of richer coal to main- tain the standard illuminating power. On the other hand, air is obtained without cost, whilst the use of oxygen necessitates the employment of a separate plant for preparing it ; each gas manager has therefore to decide which plan will be the less costly under his conditions of working. It must further be remembered that it is almost impossible to keep air entirely out of the gas, and there is always a certain amount, varying considerably in different works, already in the gas at the inlet to the purifiers, and the average amount of this should be determined before deciding the amount of air or oxygen it is necessary to add for revivification. As already stated, the above process does not affect any of the COAL-GAS 779 sulphur compounds other than sulphuretted hydrogen, the carbon bisulphide, &c., being allowed to remain in the gas. In many works, however, these impurities are also removed as far as possible, although the complete elimination has not yet been effected on the large scale. In certain towns a limit is placed on the amount of sulphur which may be present in the gas supplied from the works; thus in London the maximum allowed in winter is 22 grains per 100 cubic feet, and in summer 17 grains, and the average amount actually found in London gas is generally about 10 15 grains. To carry out the more complete purification one of two processes is now usually employed. In the first, known as the Beckton system, the gas passes first through two boxes con- taining moist slaked lime which absorbs the carbonic acid ; the sulphuretted hydrogen is also absorbed, but as the carbon dioxide comes forward it decomposes the calcium sulphide which had been formed, and the sulphuretted hydrogen is driven forward into the next two boxes, which contain oxide of iron, where it is completely absorbed. The gas next passes through two lime boxes which have previously been saturated with gas contain- ing sulphuretted hydrogen but free from carbon dioxide, and therefore contain chiefly sulphides of calcium. This acts upon the carbon bisulphide contained in the gas, forming probably among other compounds calcium thiocarbonate, CaCS 3 ; the reaction is however very complicated, and at present very incom- pletely understood. The gas then passes through boxes contain- ing Weldon mud, which consists chiefly of a compound of lime and manganese dioxide, and this absorbs any traces of sul- phuretted hydrogen which may have passed the other purifiers. In the second, or ail-lime system, the gas is passed through a set of boxes, three or four in number, containing slaked lime throughout; the first two boxes absorb the greater portion of the carbon dioxide, whilst the sulphuretted hydrogen is ab- sorbed mainly in the last two boxes, and the sulphides of calcium formed act simultaneously on the carbon bisulphide in the gas. To prevent the passage of any traces of sulphuretted hydrogen, the gas after leaving the lime purifiers passes through " catch boxes " containing oxide of iron. The great objection to the last-named system is that the " spent lime " removed from the boxes has scarcely any com- mercial value, and possesses an extremely objectionable odour, due chiefly to the presence in it of calcium sulphide, which 780 THE NON-METALLIC ELEMENTS is decomposed by the carbon dioxide and moisture in the air, giving off sulphuretted hydrogen. It has been found, however, that the addition of a small quantity of oxygen or air to the gas at the inlet to the purifiers renders the spent lime almost odourless, and at the same time assists the purification, the charges lasting for a longer period without changing, and the process having at least as great an effect on the carbon bisul- phide as the former plan, provided the amount of oxygen be not excessive. If, however, too much oxygen be added to the gas, the carbon bisulphide is practically unacted upon, as the sul- phides are oxidised by the excess of oxygen before they have an opportunity of acting on the bisulphide. Further, if an excess of oxygen is accidentally admitted to purifiers wnich have been previously absorbing the bisulphide, this latter is rapidly given off again, and the gas may then be found to contain far more sulphur at the outlet than at the inlet of the purifiers. After leaving the purifiers the gas passes through the meter to the gas-holders, where it is stored and distributed. 458 The illuminating power and also the yield of gas per ton vary very much with the nature of the coal employed in the manufacture, and also with the conditions under which the distillation takes place. Cannel coal yields the largest quantity of gas, possessing a high illuminating power, whilst ordinary bituminous coal heated in the same manner may yield nearly as much gas, but its illuminating power is greatly inferior to that obtained from cannel. Bituminous coals are, as a rule, used as the chief basis, but as the illuminating power of the gas they yield is not, for the most part, sufficiently high, a certain proportion of cannel is added, in order to maintain the standard illuminating power, which varies very much in different localities. Of late years, however, the price of cannel having greatly increased, many attempts have been made to find a cheaper enriching agent, and several processes are now successfully at work. In many cases the enriching agent is oil-gas, obtained by the dry distillation of the heavier hydrocarbon oils (p. 788) ; in others the vapour from the most volatile portion of petroleum, boiling below 100, is mixed with the gas as it passes from the purifiers to the* holders. Another plan consists in preparing water-gas by a separate plant (p. 790), and carburetting this by mixing it with the vapour of oil, and strongly heating the whole, which is then passed in the requisite proportions into the coal-gas. BUNSEN'S PHOTOMETER 781 459 Determination of the illuminating power of coal-gas. The commercial value of coal-gas depends of course chiefly upon its illuminating power, and it is of great importance to have a ready method of determining the latter. At the present time coal-gas is largely used for heating as well as illuminating purposes, and its heat of combustion is, therefore, also of im- portance, but it has been found that within certain limits the illuminating power and calorific value are very nearly propor- tional to one another, and that, therefore, from a determination of the one the value of the other may be approximately judged. In determining the illuminating power, it is first necessary to fix upon the standard of light to be employed, and in this country a sperm candle of definite construction and consuming 120 grains of sperm (7 '79 grams) per hour is employed. This standard is not altogether satisfactory, but hitherto no other standard has been found to meet with general approval. To compare the value of the coal-gas flame and that of the candle, Bunsen's bar photometer, or some modification thereof, is employed, in which advantage is taken of the fact that the intensity of the illumination from a luminous point is inversely proportional to the square of the distance of the illuminated surface from that point. All that is necessary is to ascertain the distances at which the candle and standard gas flame both produce the same illuminating effect, the illuminating power being then calculated from these observa- tions. In Bunsen's photometer the candles and gas flame are fixed at the end of a graduated bar, usually sixty inches in length, and the illuminated surface consists of a diaphragm of paper, Fig. 215, which, with the exception of a small circle in the centre, has been painted with a solution of spermaceti in benzene, thus rendering the disc, with the exception of the central portion, ' *' 1G - 21 5 - transparent. If, therefore, the disc is more strongly illuminated on one side than the other, a difference is observed between the two portions of the disc, but when both sides are equally illuminated this difference disappears. The disc is fixed at right angles to the bar in a carriage which can be moved in either direction along the bar, and at the back of the box containing the disc are placed two inclined mirrors which enable the observer looking at the diaphragm 782 THE NON-METALLIC ELEMENTS in a direction at right angles to the bar to see simultaneously a reflection of both sides of the disc. The disc carriage is then moved along the bar until the -point is found at which the whole surface of the disc on both sides appears to be equally illuminated. In making the test two candles are usually employed, as these do not vary so much as a single candle. They are fixed in a balance in order that the rate at which the sperm is con- sumed may be accurately ascertained, as this very rarely takes place at the specified rate of 120 grains an hour, and a correction must be made accordingly. The gas burnt is passed through an experimental meter, which is so regulated that the gas is consumed at the rate of five cubic feet an hour, this being the rate agreed upon at present as the standard rate of consumption. The temperature and height of the barometer are read off at the time of testing, and a correction made, the standard tem- perature being taken as 60 F. and the standard pressure as thirty inches of mercury. As a rule a test is allowed to extend over ten minutes, and the average of the readings taken, these being made once each minute. The illuminating power of the same gas varies very much with the nature of the burner employed, and it is therefore necessary to use a standard burner. The same burner cannot however be universally employed, as a burner which gives good results with the quality of gas in one town may give less satisfactory results with the higher or lower quality in other places, and the particular burner to be used must be decided in each locality according to the local circumstances. In London, where the statutory quality of the gas is sixteen candles when burnt at the rate of five feet per hour, the burner employed is an Argand, known as the London Argand, in which the number of holes in the steatite burner, height and width of the glass chimney, &c., are accurately defined. This burner is so con- structed that the requisite amount of air for consuming sixteen- candle gas at the rate of five feet per hour is admitted ; arid if, therefore, this burner is used for determining the illuminating power of gas of either a decidedly lower or higher quality, the result obtained is too low, for in the first case the inferior quality of gas, containing less heavy hydrocarbons, requires less air for its combustion, and the flame is, therefore, cooled and its illumin- ating power reduced by the excess of air drawn into the burner above the requirements of the flame. With the higher quality of COMPOSITION OF COAL-GAS 783 gas the air supply is insufficient, and this again causes a reduction in the illuminating power. Correct results may, however, be obtained with this burner by burning the gas at such a rate that the height of the flame in the chimney is about three inches, as it is with 16-candle gas at 5 feet per hour, and then making a correction for the consumption of gas. Thus, in one case a poor quality of gas gave an illuminating power of 17'30 candles with a consumption of 6'5 feet per hour, the true illuminating- power calculated for the 5-feet rate being therefore 13'19 candles, whilst when burnt at the rate of 5 feet per hour the same gas only showed an illuminating power of 9*35 candles. 1 Another means of testing the quality of coal-gas is Lowe's jet photometer, which depends upon the fact that the pressure required to give a fixed height of flame in gas issuing from a small jet varies inversely as the illuminating power. This test is only of an approximate nature, but is very valuable in the gas manufacture, as by its means the quality of the gas passing through the apparatus can be roughly ascertained at any moment. 460 Composition of Coal-gas. Purified coal-gas is a mixture of combustible gases and vapours, which may be divided into two classes, viz. : (1) Those which burn with a luminous flame and are termed the illuminating constituents ; (2) those which burn with a non-luminous or scarcely luminous flame, and are termed the diluents. The most important of the first class, which are also termed heavy hydrocarbons, are : Ethylene, C 2 H 4 , Acetylene, C. 2 H 2 , Propylene, C 3 H 6 , Allylene, C 3 H 4 , Butylene, C 4 H 8 , Benzene-vapour, C 6 H 6 . To the second class belong hydrogen, carbon monoxide, and marsh gas. In addition to these constituents, coal-gas usually contains small quantities of atmospheric nitrogen and oxygen as well as the small quantities of sulphur compounds, which, as already mentioned, cannot be altogether removed. The influence of the composition of the coal upon that of the gas obtained by its distillation is seen in the following table showing the composition of gas from bituminous and from cannel coal. 1 A Board of Trade Committee is at present (1894) sitting to consider the question of the standards of light ; their report is shortly expected. 784 THE NON-METALLIC ELEMENTS COAL-GAS FROM CANNEL COAL. Manchester Manchester Gas. (Eunsen 0 xOoo FIG. 270. FIG. 271. in fluor-spar, when it usually occurs with the cube, as seen in Fig. 269. The last holohedral form of the regular system is the Hexa- kisoctohedron (Fig. 270), or the forty-eight sided figure. This form is represented by a : ma : na, or m n. Of these, the forms 3 3 / 2 and 402 most frequently occur, although generally met with in combinations. The form 3 3 / 2 is found in garnet ; 4 2 844 CRYSTALLOGRAPHY in fluor-spar. A combination of this form with the cube is seen in Fig. 271. The hexakisoctohedron, m n, is the most complicated of all the holohedral forms of the regular system, and the other six forms may be represented as special cases of this more general expression. This is seen, if special values are given to m and n ; thus, m = n; m = 1 ; m = oo, &c. The relations between this and the other forms of the regular system are shown by the following diagram : FIG. 272 Hemihedral forms of the Regular System. The tetra- hedron (Fig. 238) is one of the most important of these FIG. 273. FIG. 274. V FIG. 275. FIG. 276. forms. This is derived by the extension of the alternate- faces of the octohedron until the other faces disappear. The formula of the tetrahedron is ~, and the two forms (p. 833) THE REGULAR SYSTEM 845 are further distinguished by the signs + (Fig. 238) and (Fig. 240). A combination of the two tetrahedra is seen in Fig. 273 ; Figs. 274 and 275 are combinations of the tetrahedron and cube, and Fig. 276 is a tetrahedron on which the faces of the rhombic dodecahedron appear. The following substances crystallize in tetrahedra: sodium sulphantimoniate, sodium D FIG. 277. FIG. 278. chlorate, cuprous chloride, and fahl-ore. Combinations of two tetrahedra are frequently seen in boracite. In this case the two tetrahedra are equally developed, so that the crystal appears to be an octohedron, but it can be distinguished from this by the fact that four of its faces are bright, whilst the other four FIG. 279. FIG. 280. are non-reflecting. The triakistetrahedron is derived from the icositetrahedron, and therefore exists in several modifications. The general formula is m | m - and each form can have a positive or negative position. These occur as complete forms, and also in combination, as in the case of fahl-ore (Figs. 277 and 278). 846 CRYSTALLOGRAPHY The hemihedral forms of the triakisoctohedron are termed deltoid-dodecahedra (Figs. 279 and 280). Their general formula is and they may be either positive or negative. FIG. 281. FIG. 282. The hexakisoctohedron in like manner yields hemihedral forms, termed Hexakistetrahedra (Figs. 281 and 282). These FIG. 283. FIG. 284. forms are found in combination in fahl-ore and boracite, and alone in diamond. FIG. 285. FIG. 286. FIG. 287, Another class of hemihedral forms is also derived from the hexakisoctohedron. These are termed dyaJdsdodecahedra, or THE HEXAGONAL SYSTEM 847 trapezoid dicositetrahedra (Figs. 283 and 284). In order to dis- tinguish these forms from those of the hexakistetrahedron the faces are represented by the same general formula, but bracketed thus, ("~s~) i They occur in cobalt glance and iron pyrites. The pentagonal dodecahedron is the hemihedral form of the tetra- kishexahedron, and therefore has the formula . The form occurs most frequently in pyrites and cobalt glance (Figs. 286). The combination shown in Fig. 287 is also ^~ 285 and found in the same substance. In addition to the forms already mentioned, another hemi- hedral form, the pentagonal icositetrahedron, and a tetartohedral form, the tetrahedral pentagondodecakedron, are also derived from the hexakisoctohedron, II. THE HEXAGONAL SYSTEM. 491 This system is characterised by one principal plane of symmetry and six secondary planes of symmetry making angles of 30 with one another. The principal axis of symmetry is chosen as the principal crystallographic axis, and is placed OOP FIG. 289. jooP FIG. 290. vertically; three of the secondary axes of symmetry, which lie in a plane perpendicular to the vertical axis being taken as secondary crystallographic axes. These are all equal, but are not equal to the vertical axis in length, and meet in a point making angles of 60. Each face of the simplest pyramid (Fig. 289) cuts the prin- cipal or vertical axis and two of the secondary axes, but runs 848 CRYSTALLOGRAPHY parallel to the third. The formula for the pyramid is there- fore a : a : oo a : c, in which the letter c refers to the principal axis, or abbreviated P. The vertical axis of an acute pyramid is of course longer than the secondary axes, whilst that of an obtuse pyramid is shorter. Thus, for example, the relative length of the axes in the case of the acute pyramid of quartz is I'l : 1, whilst in the case of the obtuse pyramid of beryl it is 0-498 : 1. If, in the pyramid, the length of the principal axis be reduced to 0, a face known as OP will be produced, which is parallel to the plane containing the three secondary axes, whilst if the length of this same principal axis be increased indefinitely, the pyramid becomes an open six-sided prism (Fig. 290), the symbol of which is a : a : oo a : oo c, or x P. This prism may occur FIG. 291. along with the face OP in the closed combination shown in Fig. 290, OP, oo P. The prisms and pyramids, of which each face cuts two of the secondary axes at equal distances, and is parallel to the third, so that the secondary axes pass through the angles (Figs. 289, 290), are termed forms of the first order. A second series of similar forms occurs in this system, which may be derived from the forms of the first order by rotating them through an angle of 30, so that the secondary axes cut the centre of the six sides of the base of the pyramid. Such prisms and pyramids are termed forms of the second order. Fig. 291 exhibits the relation between these two series of forms. It will be seen that in the pyramid of the second order which is there represented in section, each face cuts one of the secondary axes at the THE HEXAGONAL SYSTEM 849 distance a, and the other two axes at the distance 2a, The formula of the pyramid of the second order is accordingly 2a : a : 2a : c, or P2, whilst that of the corresponding prism of the second order is 2a : a : 2a, oo c, or oo P2. In the same way faces may occur which cut the vertical axis at a greater or less distance than c from the centre, so that the general formula for the pyramid of the first order is mP, whilst that of pyramids of the second order is mP2. In the case of the combination of several pyramids in which the value of m varies, these values, in accordance with the general law (p. 833), exhibit a simple ratio, such as 1 : 2, 1 : 3, 1 : 5, etc. Combinations of pyramids of the first and second orders also occur ; when they are equally developed, the combination assumes FIG. 292. the form of a symmetrical twelve-sided pyramid, the formula for which is a : na : pa : me, abbreviated into mPn (Fig. 292). This is in reality the most general form possible in the system from which all the other forms- may be derived. Thus when n = l and p= oo , it becomes an hexagonal pyramid of the first order, whilst when p = oo it becomes a pyramid of the second order. The closed holohedral forms of the hexagonal system do not occur frequently. On the other hand, combinations such as shown in Fig. 293 are found in quartz; Fig. 294 exhibits a form found in calc-spar, and Fig. 295 a combination seen in apatite. The hemihedral forms in the hexagonal system are numerous and important, occurring even more frequently than 55 850 CRYSTALLOGRAPHY the holohedral forms. The most important of these are the rhombohedra (Fig. 296), which are derived from the hexagonal oP OOP FIG. 294. FIG. 295. FIG. 293. pyramid by extending the alternate faces until they produce a p closed form (Fig. 297). These faces have the symbol or R FIG. 296. FIG. 297. Gale-spar, iron-spar, and sodium nitrate occur in the form of simple rhombohedra, The rhombohedron, like the tetrahedron, FIG. 298. FIG. 299. FIG. 300. may occupy two positions according as the one or the other set of alternate faces of the six-sided pyramid is extended. These THE HEXAGONAL SYSTEM 851 are shown in Figs. 298 and 299 ; the two rhombohedra are distinguished by the signs + and . Combinations of the rhom- bohedron also frequently occur. One of these is represented by Fig. 300, which exhibits a combination of several positive and FIG. 301. FIG. 302. FIG. 303. negative rhombohedra, occurring in chabasite. Fig. 301 repre- sents a combination of two rhombohedra which are both positive, but differ in axial ratio, a form which occurs in calc- spar. A combination of the rhombohedron with a prism of the first order is seen in Fig. 302 ; this form has also been observed in calc-spar. A combination of a rhombohedron with a prism of the second order is shown in Fig. 303. This has been observed in dioptase. A second hemihedral form of this system is the scalenohedron. This is obtained from the symmetrical twelve-sided pyramid, by extend- ing the alternate pairs of planes, thus giving rise to a positive ( -f ) and negative ( ) scalenohedron. If we suppose a rhombohedron to be placed within a scalenohedron, it is termed the inscribed rhombohedron, and the lateral edges of both forms will be seen to have a similar position with regard to the axes. It is clear that the scalenohedron may be supposed to be derived from such a rhom- bohedron by an elongation of the axis to the distance n, lines being drawn from the end of this axis to the lateral solid angles of the rhombohedron, as seen in Fig. 304. Although in this way an infinite number of scaleno- hedra may be formed, we find that only those forms actually FIG. 304. 352 CRYSTALLOGRAPHY exist in which the length of the primary axis bears some simple ratio to that of the inscribed rhombohedron. The general symbol for the scalenohedron is + mB, n . The axis in the case of the commonest scalenohedron occurring in calcite is three FIG. 305. FIG. 306. FIG. 307. times as long as that of the inscribed rhombohedron , hence the symbol for this scalenohedron is + R 3 . Tetartohedral Forms. In addition to some other forms of hemihedry, a number of tetartohedral forms occur in this system. Such forms, shown in Figs. 306 and 307, occur in the FIG. 308. case of quartz. Two forms, known as trigonal trapezohedra, are possible, both derived from the twelve-sided pyramid, and these two are not congruent, so that when they occur as in Figs. 306 and 307 on different crystals one of these appears as the THE QUADRATIC SYSTEM 853 reflected image of the other, or one is right-handed and the other left-handed. Some cases of hemimorphism also occur in this system, a well-known instance of which is seen in crystals of tourmaline, seen in Fig. 308. III. THE QUADRATIC, TETRAGONAL, OR PRISMATIC SYSTEM. 492 As in the hexagonal system, the principal axis of sym- metry is chosen as the principal crystallographic axis, whilst two FIG. 309. of the secondary axes of symmetry serve as the other two crys- tallographic axes. These two are in one plane, at right angles FIG. 310. FIG. 312. to each other and to the principal axis, and equal to each other but not to the principal axis. This system differs from the hexagonal in only possessing four secondary planes of symmetry making angles of 45. The most general form of the system 854 CRYSTALLOGRAPHY is the ditetragonal pyramid (Fig. 309), which has the symbol nPm, and from which all the other forms may be derived. It coP coP oP Pco FIG. 313. FIG. 314. FIG. 315. has an octagonal base, and is bounded by sixteen triangular faces. The various forms are related much in the same way as in the hexagonal system. FIG. 316. FIG. 317. FIG. 318. FIG. 319. The simplest form of this system is that of the four-sided pyramid with a square base. This may be either of the first order, a : a : c or P, or of the second order, a : oo a : c or P x , FIG. 320. FIG. 322. derived from the former by rotating it through 45. These square pyramids are distinguished as acute and obtuse (Figs. 310 and 311). When several of these square pyramids occur THE QUADRATIC SYSTEM 855 together in a combination, the lengths of the various principal axes, in accordance with the general law (p. 833), stand in a simple ratio to one another. This is seen in Fig. 312, which represents a crystal of nickel sulphate. From this it will be seen that the principal axis of the pyramid P is exactly twice as long as that of the pyramid J P. A combina- 7^7 P 17 ** FIG. 323. / FIG. 324. FIG. 325. tion of pyramids of the first and second order is shown in Fig. 313. This form is likewise found in nickel sulphate. When the principal axis equals 0, the basal terminal planes OP make their appearance, and when it is infinitely prolonged we obtain the faces of the prism of the first order, oo P, or those of the secondary prism, oo Poo . See Figs. 314 and 315. Combinations belonging to the quadratic system are seen in FIG. 326. FIG. 327. FIG. 328. Figs. 316 and 317 ; the first of these is found in zircon, the second in crystals of calcium copper acetate. Fig. 320 represents another common form of this system, the crystals of potassium ferro- cyanide. The usual form of tin-stone (stannic oxide) is seen in Fig. 321, that of mellite in Fig. 322. Figs. 323 to 325 exhibit the more complicated forms of nickel sulphate. A remarkable form of the quadratic system occurs in the 856 CRYSTALLOGRAPHY mineral leucite. This is a combination of P with 4P2, their forms being equally developed (Fig. 326). This form can only with difficulty be distinguished from that of the regular icositetrahedron (202). So close indeed is the resemblance of these forms that leucite was long supposed to crystallize in this form of the cubic system, and the name leucitohedron was given to it. We owe the discovery of the true crystalline form of the mineral to Vom Rath. 1 The combination, Fig. 327, clearly shows the quadratic form, inasmuch as small faces of a third pyramid as well as those of a prism occur. The twin crystal i i ooP : f j FIG. 329. FIG. 331. (Fig. 328) occurring in the case of leucite also shows that the form is a quadratic one. Hemihedral forms can be derived from the quadratic prism by the same process as that by which the tetrahedron is derived from the octohedron. The forms thus obtained are termed quadratic sphenoids. These can occur in two positions, shown in Figs. 329 and 330; and they are distinguished from the regular tetrahedra inasmuch as each face is a scalene and not an equilateral triangle. They are not often found in nature ; Fig. 331 shows a combination which, occurs in the case of Epsom salts and of zinc vitriol. 1 Pogg. Ann. Erganz. Bd. 6, 198. THE RHOMBIC SYSTEM 857 IV. THE RHOMBIC SYSTEM. 493 This system is characterised by three secondary planes of symmetry all at right angles. The three secondary axes of symmetry are chosen as the crystallographic axes, and they are all unequal and at right angles. It is usual to take as the principal axis that which lies in the direction in which the crystal is most fully developed or in which most modifications are observed. This is called (c), and placed vertically. In the following figures the longer of the other two axes, termed the macrodiagonal (&), runs from back to front of the observer, and the shorter, the lr achy diagonal (c), from right to left. The arrangement of the horizontal axes is purely conventional, and is frequently the reverse of that just described. The fundamental form of this system is the rhombic pyramid Fig. 332, having the formula a : b : c or P. Secondary pyramids also occur, and these are distinguished by the signs ~ or " placed over the P according as the length of the intercept on the macro- or brachy- diagonal is increased. The amount of this altera- tion is represented by a number placed after the letter P, so that the symbol Pn denotes a pyramid the faces of which cut the principal axis and the brachydiagonal at the normal distances c and a, and the macrodiagonal at a distance of n times b ; whilst Pn signifies that the brachydiagonal is cut at a distance of n times a. Faces also occur which would cut the principal axis at a distance which is some multiple of the normal length c y and this is represented by a letter placed before the symbol P. Thus the symbol 3P2 signifies a form each face of which cuts the macrodiagonal at the distance b, the principal axis at a distance of 3 times c, and the brachydiagonal at a distance of twice a from the point of intersection of the axes. In addition to these series of pyramids, a number of prisms are also possible, the faces of which cut two of the axes and are parallel to the third. If the faces are parallel to the principal axis, the form is known as a vertical prism oo P, if to the macro- or brachy-diagonal as a macro- or brachy-dome, Poo or ]Poo , these last two being nothing more than horizontal prisms (Figs. 334 and 338). 858 CRYSTALLOGRAPHY Faces which are parallel to two of the axes but cut the third are known as terminal planes or pinacoids. Of these there are three, OP, usually called the basal plane, oo Poo or the macropinacoid, and x f x the brachypinacoid (Fig. 339). P FIG. 333. FIG. 334. FIG. 335. A form termed the right rhombic prism is formed by the com- bination of the three terminal planes, and is found in nature in crystals of anhydrite, CaSO 4 . These various forms are summarised with their symbols in the following table, in which a = brachy diagonal, b = the macro- diagonal, and c = the vertical axis. oP ""^^ FIG. 336. FIG. 337. FIG. 338. FIG. 339. Rhombic Pyramids. a : b : me, abbreviated into mP = primary rhombic pyramid, na : b : me mf n = brachypyramid. a : nb : me mPn = macropyramid. Vertical Prisms. a : b : oc c, abbreviated into ocP = rhombic prism (primary), na : b : occ ,, ocPn = brachyprism. a : nb : a c ex Pn = macroprism. THE RHOMBIC SYSTEM 859 Domes or Horizontal Prisms. oc a : b : me, abbreviated into mf ex = brachydome. a : or b : me mPoc =macrodome. Terminal Planes. , i . , i , A r> f basal terminal plane = 1 a:ab: * c or pinacoid. -5 f brachy terminal plane aPK or pinacoid -fs f macro-terminal plane < P * orpinacoid P Sulphur is one of the most important substances crystallizing in this system. It occurs in rhombic pyramids and in numerous combinations. Fig. 332 exhibits the fundamental form of FIG. 340. FIG. 341. FIG. 342. sulphur, in which the ratio of the axes a : b : c 0'811 : 1 : 1*898, and Figs. 333 and 334 various combinations which are found on crystals of this element. Zinc sulphate is found in the forms represented in Figs. 335, 336, 337. Barium formate, crystal- lizing in a combination of a prism with a dome, is shown in Fig. 338. Fig. 339 represents a crystal of uranium nitrate, possessing two basal terminal faces. Potassium sulphate crystal- lizes in several interesting forms. Of these, Figs. 340 and 341 may at once be recognized as belonging to the rhombic system, whilst Figs. 342 and 343 might at first sight be mistaken for hexagonal forms. The most important hemihedral forms of the rhombic system .are the rhombic sphenoids. The faces of the two forms of these 860 CRYSTALLOGRAPHY sphenoids (shown in combination Figs. 344 and 345) are equal in number, and exactly similar in form, but when the + and forms occur on separate crystals, they are developed on opposite sides, so that the one may be regarded as a reflected image of the other. Fig. 344 is the double sodium and ammonium salt of ordinary tartaric acid, which is termed dextro-tartaric acid, from its hav- ing the power of turning the plane of polarization to the right. iC" p * <: . op . -, . l P y 2Po> ^J P2 Z^> \ coP 9 Poo s* OOP \ ooPoo W OOPOD OOPOD FIG. 343. FIG. 344. FIG. 345. Fig. 345 is a salt of the same bases combined with a very similar acid, differing only from ordinary tartaric acid in its power of rotating the plane of polarization to the left. Another in- teresting example of a rhombic sphenoid is found in the ordinary crystals of magnesium sulphate, MgSO 4 + 7H 2 O. In these the domes at the summits of the vertical prism are placed in opposite directions. V. THE MONOCLINIC OR MONOSYMMETRIC SYSTEM. 494 This system has only one plane of symmetry, and there- fore only one axis of symmetry. This is chosen as one of the crystallographic axes, and is known as the axis of symmetry, or more generally as the orthodiagonal. This axis is supplemented by two others which lie in the plane of symmetry, and are, therefore, at right angles to the orthodiagonal. These intersect at an angle (/:?) which is not a right angle, and one of them is usually referred to as the primary axis, the other being known as the clinodiagonal. The position of the crystal, like that of crystals of the rhombic system, is purely a matter of convention. In the following figures the orthodiagonal runs from back to front of the observer, so that the plane of symmetry is parallel to the plane of the paper, whilst the clinodiagonal runs from right to left. Many authors, on the other hand, make THE MONOSYMMETRIC SYSTEM 861 the orthodiagonal run from left to right, and place the primary axis vertical. The symbols referring to the clinodiagonal are distinguished by being enclosed in parentheses. The complete description of each form of this system includes the axial ratio and, in addition, the angle of intersection of the primary axis with the clinodiagonal. Thus, for instance, the data of the crystals of felspar are a : b : c = 1'519 : 1 : 0*844 ; @ = 63 53'. The fundamental form of this system, since there is only one plane of symmetry, consists of a hemipyramid made up of four faces. Thus in Fig. 346, in which the plane of symmetry is parallel to the plane of the paper, the two upper right-hand faces and the two lower left-hand faces (marked -}-) belong to one of these hemipyramids, whilst the other two pairs of faces (marked ) belong to the other. The complete octohedron shown in the figure is, therefore, made up of two crystallographically distinct forms, which are known as -f-P and P. Each of these two hemipyramids may of course occur by itself in combination with other forms, so that very frequently only four pyramid faces are found on complex crystals of this system (Figs. 348, 349). The pinacoids, prisms, and domes, which are known as the ortho- and clino-domes, are derived from the pyramids in precisely the same way as in the rhombic system. The prism and the clinodome comprise four faces, whilst the orthodome, owing to the presence of only one plane of symmetry, consists of only two faces, and is really a hemidome + P or Poo . The following figures represent the forms of a variety of well-known chemical com- pounds crystallizing in this system. FIG. 346. FIG. 347. FIG. 348. FIG. 349. Figs. 347, 348, and 349 give the crystalline form of sodium acetate; Fig. 350 that of cane-sugar; Fig. 351 that of nickel potassium sulphate ; Fig. 352, ferrous sulphate (green vitriol), FeSO 4 + 7H 2 O. In this latter form the primary axis cuts the 862 CRYSTALLOGRAPHY clinodiagonal at an angle of 75 C 40', and the relation of the axes a : b : c = 0'8476 : 1 : 1-267. Gypsum, CaSO 4 + 2H 2 O, fre- quently occurs in fine monosymmetric crystals. The form of FIG. 350. FIG. 351. FIG. 352. gypsum is shown in Figs. 353 and 354 ; the relation of these axes is as a: b :c = l'4504 : 1 :0'6003, with an inclination of 80 57'. FIG. 353. FIG. 354. FIG. 355. Some substances, such as felspar, crystallize in forms which are apparently very different (Figs. 355 and 356) owing to the greater or less development of certain faces. Crystals of Glauber's FIG. 356. FIG. 357. FIG. 358. salt and oxalic acid are often found developed in the direction of the orthodiagonal (Figs. 357 and 358). Hemimorphous forms are not uncommon in this system, only THE ASYMMETRIC SYSTEM two faces of the clinodome being found, both of them on the same side of the plane of symmetry. Dextro-tartaric acid not unfrequently crystallizes in the simple form Fig. 358. More usually, however, the four-faced angles at the front are replaced by a dome (Fig. 360), whilst in the case of the laevo-tartaric acid the corresponding angles at the back are similarly replaced, as shown in Fig. 361. FIG. 360. FIG. 361. FIG. 362. It is a singular fact that hemimorphous forms of this kind generally occur in the case of bodies which are optically active that is, which possess the power of rotating the plane of polari- zation of light. They are accordingly found (Fig. 362) in the case of cane-sugar, which also possesses this optical property. VI. THE TRICLINIC OR ASYMMETRIC SYSTEM. 495 This system is entirely without any plane of symmetry, and the choice of axes is, therefore, quite arbitrary. Any one face of the crystal is chosen as the pyramid face, and three unequal axes cutting one another at angles a, 0, 7 are then adopted. One of these is known as the primary axis, the other two being dis- tinguished as brachy- and macro-diagonals. Five values have to be given in order completely to determine an asymmetric form, viz. a : b : c, the relative length of the axes, and the three acute angles (a, /3, 7) which they form with each other. The possible forms belonging to this system correspond closely with the forms of the rhombic system. In addition to the faces P of the fundamental pyramid, the following faces of secondary pyramids occur, m P, Pn, Pn, m Pn, m Pn, together with the surfaces of the prisms, ocP, ocPn, ocfn. Then we 864 CRYSTALLOGRAPHY have the domes, in P a, m P a, and the basal end-faces, P, oc Pa, afroc, a P a. Since there is no plane of symmetry, each crystallographically complete form consists of a pair of parallel faces, so that an entire octohedron would be made up of four independent pyramids, and each prism or dome of two independent forms. FIG. 363. oP FIG. 364. These forms having similar and parallel faces are represented by accentuated letters. Thus P' represents a face of the funda- mental form, which occurs in the upper front of the crystal to the right, whereas the similar and parallel face ,P is one which occurs in the lower part of the crystal to the left. Thus the symbol oc /P indicates the two faces of the prism x> P, one of FIG. 365. FIG. 366. FIG. 367. which occurs above to the left ; m y P' a denotes the faces of a brachydiagonal dome which lie above to the right and below to the left. The direction in which crystals belonging to the triclinic system are placed is a purely conventional one. Thus Fig. 363 may in the first place be regarded as a combination of the forms ocPx), ocpoc, OP; or secondly as oc/P, ocP',, OP. ISOMOEPHISM 865 The following figures exhibit some commonly occurring asymmetric forms : Copper sulphate, CuSO 4 + oH 2 O, Fig. 364. Calcium thio- sulphate, Fig. 365. Albite, Figs. 366 and 367. Other common substances crystallizing in this system are axinite, boric acid, cupric selenate, manganese sulphate, potas- sium anhydrochromate, and sodium succinate. ISOMORPHISM. 496 In his celebrated system of classifying minerals devised in 1801, Haiiy propounded the principle that a difference in the primary form of two crystals invariably indicates a difference of chemical composition, whilst identity of form, except in crystals of the regular system, equally implies identity of composition. It was however soon found that his principle could not be applied in all its generality, for Leblanc had shown so long ago as 1787 that crystals of an identical form could be obtained from solutions containing sulphate of copper and sulphate of iron mixed in very different proportions. He likewise states that crystals of alum, a sulphate of aluminium and potassium, may frequently be found to contain a considerable quantity of iron, although no alteration in the crystalline form can be noticed. Vauquelin, too, showed in 1797 that common potash alum may contain large quantities of ammonia, and yet the crystalline form of the substance does not undergo any change. It was also well known that many minerals which are identical in crystalline form may possess a very different chemical com- position. Thus in certain specimens of red silver ore, arsenic is present as an essential constituent, whilst in others antimony takes its place. In like manner, the common garnet, crystalliz- ing in the regular system, sometimes contains much iron and little aluminium, and sometimes large quantities of aluminium and little iron. Berthollet considered facts like these to be in accordance with his views on chemical combination, but Proust explained them by supposing that we have here to deal not with chemical compounds but rather with mechanical mixtures. In the year 1816 Gay-Lussac made the remarkable observa- tion that when a crystal of common potash alum is hung up in a saturated solution of ammonia alum it grows exactly as if it had been placed in the solution from which it was originally 56 866 CRYSTALLOGRAPHY obtained. From this fact he drew the conclusion that the molecules of these two alums possess the same form. Later on, in 1819, Beudant noticed that if solutions containing two of the following salts, sulphate of zinc, sulphate of iron, or sulphate of copper, be crystallized, the crystals which are deposited always possess the form of one of these salts, although they contain a considerable quantity of the other salt, which, when crystallized by itself, possesses a totally different form. In order to explain these and similar well-recognised facts, Haiiy threw out the notion that certain bodies possess the power of crystallization to such a degree that even when present in small quantities they compel other bodies to adopt their crystal- line form. Clear light was thrown on this subject by the researches of Mitscherlich, the results of which were communicated to the Berlin Academy in 1819. Mitscherlich showed that the com- pounds of various elements possessing a similar constitution have also identical crystalline form, as ascertained by the measurement of their angles. The first substances examined by Mitscherlich were the arsenates and phosphates of sodium potassium, and ammonium. He showed that the crystals of the phosphate and arsenate of the same metal riot only contain the same quantity of water of crystallization, and crystallize in the same form, but that when the two salts are mixed in varying quantities the crystals which such a solution deposits are of the same form as those obtained from solutions of the pure salts, whilst the proportion of each ingredient found in the crystal deposited from the mixed solution varies according to the propor- tions in which the ingredients were mixed. Hence Mitscherlich concluded that analogous elements or groups of elements can replace one another in compounds without any alteration of crystalline form. Such substances are said to be isomorphous (t'0-0?, equal to, fioptyij, shape). A large number of other com- pounds were shown by Mitscherlich to conform to the above law. Subsequent investigations have, however, proved that the angles of isomorphous crystals are not absolutely identical, but that each substance differs slightly from the other in this respect. When several of such substances are crystallized from solution together, the angles of the crystal deposited are the mean of those of the pure substances. 497 The following minerals serve as an excellent illustration of the law of Isomorphism : ISOMORPHISM 867 Apatite, Ca 3 (PO 4 ) 2 + Ca 2 1 |* Pyromorphite, Pb 8 (PO 4 ) 2 + Pb 2 j ^ Mimetesite, Pb 3 (As0 4 ) 2 + Pb 2 j Sj 3 * Vanadinite, Pb 3 (V0 4 ) 2 + Pb 2 j ^ 4 The isomorphous chemical elements in these minerals are : (1) Phosphorus, Arsenic, Vanadium. (2) Calcium and Lead. (3) Chlorine and Fluorine. FIG. 368. They all crystallize in hexagonal prisms, seen in Figs. 369 to 371, derived from the fundamental form P. (Fig. 368). The values of the terminal solid angle a, the lateral solid angle ft ooP FIG. 369. FIG. 370. and the length of the primary axis (c), that of the secondary axes being taken as the unit, are found to be : c. a. Apatite . . . 07321 142 20' Pyromorphite . 07362 142 15' Mimetesite . . 07392 142 70' Vanadinite . 07270 142 30' 0. 80 25' 80 44' 80 58' 80 0' Not urifrequently both chlorine and fluorine occur in the same specimen of apatite ; and phosphorus and arsenic, not unfrequently, replace one another in pyromorphite. -868 CK YST ALLOG R APH Y Another well-marked case is that of the rhombohedral carbonates of the magnesium class of metals, calcium, mag- nesium, iron, zinc, and manganese. These minerals all crys- tallize in similar rhombohedra, and the several metals can replace one another in varying proportions without any change of form occurring. It has, however, been observed that the angles of the different rhombohedra are not exactly equal, but that the different members of the series vary in this respect by one or two degrees ; thus : Angle on termnl. edge. Calcium carbonate, or calc-spar, CaCO 3 . . 105 5' Calcium magnesium carbonate, or dolomite, (CaMg)CO 3 106 15' Magnesium carbonate, or magnesite, MgCO 3 . . 107 25' Ferrous carbonate, or spathic iron ore, FeCO 3 . . 107 0' Zinc carbonate, or calamine, ZnCO 3 . . 107 40' Manganese carbonate, or diallogite, MnC0 3 . . 106 5' The replacement in minerals of varying proportions of the different isomorphous compounds is well illustrated by the following percentage analysis of spathic iron ore : Ferrous oxide, FeO 45'55 Manganous oxide, MnO 12*50 Lime, CaO T57 Magnesia, MgO T80 Carbonic acid, CO 2 38'58 100-00 If we now divide the percentage of oxide by its combining 45*55 weight thus ' for ferrous oxide, etc., etc., we obtain the following numbers respecting the proportion between the number of equivalents of the several constituents present : FeO = 0-6372, MnO-0'1773, CaO = 0'0282, MgO = 0'0448, or summing these we have 0'8875. This is, however, very nearly the proportion which must be present in order to unite with OQ.ro \^= = 0'8833 equivalents of CO 2 , the difference being owing to 4o'o7 .errors of experiment. In other words, the number of molecules of the basic oxides present is the same as that of the carbon -dioxide ; hence the formula for this spathic iron ore is (Fe, Mn, ISOMORPHISM 869 Ca, Mg)CO 3 , signifying that the relative quantities of the metals in question present are indeterminate, but are in the aggregate such as are needed to combine with CO 2 . Hence we have the following as the composition of the mineral : Ferrous carbonate, FeCO 3 . . . 73'39 Manganese carbonate, MnCO 3 . . . 20*24 Calcium carbonate, CaCO 3 . . . 2*80 Magnesium carbonate, MgCO 3 . . . 3'95 100-38 Fahl-ore offers another striking example of isomorphism. The simple formula for this mineral is. 2Cu 2 S -f Sb 2 S 3 , but usually part of the copper is replaced by iron, zinc, silver, or mercury, and some of the antimony by arsenic. This is seen by the following analysis : Sulphur, S. . 25-48-s- 31'82 = 0'801 ^ 0-304 0-651 100-00 If the several percentage weights be divided by the corre- sponding atomic weights, and if the resulting quotients of the isomorphous constituents be added together, the following result is obtained : 8 = 0-801 (Sb, As) = 0-304 (Cu, Fe, Zn, Ag) = 0-651 These numbers have the ratio 5'2 : 2 : 4'2, or allowing for experimental error, 5:2:4, and the formula of Fahl-ore is therefore : 2[(Cu,Fe,Zn,Ag) 2 S] + (Sb,As) 2 S 3 . Antimony, Sb . 17-76 +119-4 =0-149) Arsenic, As . 11-55-s- 74-4 =0-155 J Copper, Cu . 30-73-h 62-8 = 0-489 \ Iron, Fe . 1-42-s- 55-6 = 0-025 / Zinc, Zn . 2-53-T- 65 = 0-039 f Silver, Ag . 10-53-s- 107-13- 0-098 J 870 CRYSTALLOGRAPHY DIMORPHISM AND TRIMORPHISM. 498 Even before Haiiy started the idea that bodies which crystallize in different forms differ also in chemical composition, Vauquelin had noticed that titanic acid occurs as rutile and anatase, two minerals possessing distinct crystalline forms. In like manner, Klaproth pointed out that hexagonal calc-spar is the same chemical compound as rhombic arragonite. These exceptions to Haiiy's law were then explained by the presence in the compound of some impurity which has the power of altering the crystalline form. Thus, arragonite was found to contain small quantities of strontium carbonate, a mineral which is found crystallized in the same form, and this small proportion was supposed to exert so powerful an action as to compel the calcium carbonate to assume a rhombic form. In 1821, Mitscherlich proved that this property of crystallizing in two distinct forms is common to many bodies both elementary and compound, and he termed such bodies Dimorphous. Other substances, again, are capable of existing in three distinct crystalline forms. These Mitscherlich termed Tri- morphous substances. If, lastly, two substances exhibit a double isomorphism, they are said to be Isodimorphous. The trioxides of arsenic and antimony serve as a striking example of isodi- morphism. For a long time these compounds were only known to occur in two forms which were not isomorphous, and this was the more remarkable as their elementary constituents exhibit such a close analogy. It was afterwards found that arsenious oxide, As 4 O 6 , which usually crystallizes in regular octohedra, is occasionally met with in rhombic crystals, exactly identical in form with those in which antimonious oxide, Sb 4 O 6 , commonly occurs in nature. A mineral consisting of this latter oxide, and called senarmontite, was next discovered. The crystals of this are octohedral, so that the isodimorphism of these substances is now completely proved, especially since it has been found possible to produce the octohedral crystals of antimonious oxide artificially. Dimorphous and trimorphous bodies are not only distinguished by their difference in crystalline form. The other physical properties, such as specific gravity, hardness, refractive power, etc., are all different. One of the best examples of a trimorphous compound is titanium dioxide, TiO 2 . The commonest form of this compound is rutile. This mineral crystallizes in the quad- DIMORPHISM AND TRIMORPHISM 871 ratic system with the combinations P, oc P, P 30, and x P x ; the relations of the axes are a : b = 1 : 0*6442 ; its specific gravity is 4*2494. The second form of titanium dioxide is anatase, Fig. 372 ; it likewise crystallizes in the quadratic system, but the relation of the axes is in this case quite different FIG. 372. from that in the case of rutile, viz. a :b= 1 : 1*777. Anatase crystallizes as a rule in the fundamental form P. The specific gravity of anatase is 3'826. The third form is brookite; it crystallizes in flat rhombic prisms, and possesses a specific gravity of 4*22. THERMAL AND OPTICAL RELATIONS OF CRYSTALS. 499 Crystals which belong to the regular system being equally developed in all directions expand when heated equally in every direction. On the other hand, crystals belonging to the hexagonal and quadratic systems expand differently in the directions of the primary and of the secondary axes. This is clearly seen in the following table, which gives the coefficients of expansion of substances crystallizing in these two systems. 1 Primary axis. Secondary axis. Quadratic System. Tinstone . . 0*0004860 0*0004526 Zircon . . 0*0006264 0*0011054 Hexagonal System. Qu&rtz . . 0*0008073 0*0015147 Tourmaline 0*0009369 0*0007732 Crystals belonging to the other systems possess coefficients of expansion which differ for each of the three directions of the 1 Pfaff, Pogg. Ann. 104, 171. 872 CRYSTALLOGRAPHY axes. Hence whilst the angles of substances crystallizing in the regular system remain constant under change of temperature, those of crystals belonging to the other systems undergo small deviations with alteration of temperature. The same relation is exhibited by crystals with regard to the conduction of heat as holds good in the case of expansion. The conducting power of crystals of the regular system is the same in all directions. Those belonging to the quadratic and hexa- gonal systems conduct equally in two directions, and unequally in the third, whilst crystals belonging to the other systems con- duct differently in every direction. This difference in the con- ducting power of crystals can be well shown by covering a face of the crystals with a thin coating of wax, allowing the wax to solidify and then bringing the point of a hot needle or other pointed hot body against the wax coating. If the crystal conduct equally, the wax will melt in a circle of which the hot point is the centre. If the conduction be unequal, the melted wax will be seen to assume an oval form. Pyro-electric Action of Crystals. Certain hemimorphous or tetartohedral crystalline forms when heated exhibit a peculiar development of electricity, one end of the crystal, or fragment of crystal, becoming negatively electrified, whilst the other end ex- hibits positive electricity. Amongst these crystals, tourmaline exhibits this property in a very high degree. Boracite, cane- sugar, topaz, and silicate of zinc are crystals which exhibit pyro- electrical reactions. 500 Optical Properties of Crystals. The relations of crystals to. light are of great importance, as enabling us to determine crys- talline forms in cases in which the usual methods either give uncertain results or fail entirely. Transparent crystals belonging to the regular system exert no peculiar action on a ray of light ; they behave in this respect like glass or any other amorphous substance. The incident ray gives rise to one refracted ray, and crystal possesses one refractive index. 1 Crystals belonging to the hexagonal and quadratic systems are doubly refractive. Each incident ray separates into two re- fracted rays, one being bent more out of its course than the other. This phenomenon of double refraction is, however, only observed 1 Cases have indeed been- observed by Brewster in which a crystal belonging to the cubic system exhibits double refraction, but this is due to unequal tension in the mass of the crystal and resembles the case of double refraction by unequally heated or compressed glass. OPTICAL PROPERTIES OF CRYSTALS 873 when the incident ray falls on the crystal in such a direction as to make an angle with the primary axis. If the ray pass into the crystal parallel to this axis, no double refraction is observed. Crystals belonging to the hexagonal and quadratic systems are accordingly called uniaxial crystals, whilst those of the rhombic and the inclined systems are termed biaxial crystals, because- they possess two directions in which the ray of light may fall without producing the effect of double refraction. Polarization by Absorption in Crystalline Media. Doubly re- fracting crystals, especially tourmaline, have the power of only allowing rays of light to pass through them when those rays are polarized in a given direction with regard to their optic axes. Thus, if a ray be allowed to pass through a plate of tourmaline cut with its faces parallel to the optic axis (the axis c of the hexagonal system), it will refract doubly, but the ordinary ray will be completely absorbed, and only the extraordinary ray will pass through. This ray is polarized in a plane perpendicular ta the axis of the crystal, and so that a second plate of tourmaline cut in a similar way and placed with its axis in a direction at right angles to that of the first will not allow any light to pass, as in Fig. 374, though when the two axes are placed parallel to each other the polarized ray will be transmitted by both crystals, as in Fig. 373. The first crystal used to polarize the ray is termed the polarizer, whilst the second, used to examine the direction of the polarization, etc., is termed the analyzer. The property of circular polarization is possessed not only by crystals but also by many organic liquids, such as solution of sugar, tartaric acid, and many essential oils. In these substances the plane of polarization is rotated sometimes to the right, some- times to the left hand. The specific rotary power of any sub- stance is the angle through which a column of the substance of unit length rotates the plane of polarization of a beam of polarized light. 874 WEIGHTS AND MEASURES I ?l s'S g 8i c?s ^ A II ^i s S P3 f!l v^J O S.? O ^ r- 2 rH r^j p^ ^ s l s I- rt ^ EH W I PH II g E h l Id i! CO O CO rH OS 1 ""w 'tO C*^ rH Tf GO CN ^ ^ rH CO OS ^ c3 O O S^ rH CO QO (N ^ , rH ^fl O ~>* OOoScNrHCCOO *- S t^ rH CO H jjljjlll wl (M 1^ i-H O * l-H o o 1 ^ ,= " 3f b O GN 1 II c ^_- a = ScOrH-oicSoO ' S *' t-. **< oo b b en .-, T^ CO OO rH CO * ^~ O O O O * CO 00 rH CO 03 ! OS * 1>- CO 00 C is OS CO II II I-H O I-H w o II 03 o jS ^ rl w OS O OS ^1 O O O O PH si CO O O r^ ^_, O OS OS OS 71 O O O CO CO CN Ij OOOOOOSOSGNOO (NCOOCOOSOSCMO CO Is CO CN O O CO CO CO CO CN d5 ^ i-H rH o !i | ppCOCNOOpOOOS o "sj* S CO p GO fC bbbcocNooboo CO OS O O CO OS CO !> p tj- OS i t . CN CD ill II t-( ^ '" 22 "S "1 - ' H fl "S 2 '~ H 03 li|l Hit IS -2 ej oj O S 03 goSSrHKWS ^^w WEIGHTS AND MEASURES 875 Isg oo gi- ll ,5 is I C M -r n Gallons = 8 Pints = 277-27384 Cubic Inches. Pints = 34-6 Cubic Inches * is ot^t^co rH t^ C t^ t^ i COlOCOr lOO OCOlOCOi lO OOOCOkOCOrHCO ppppCOkOCOrH OOOOOCOkOCO CO O CO CO t-( O OO * (^ O O O O Oi OO Tji 1 .' Cyanogen iodide, 761 Cyanogen selenide, 764 Cyanogen sulphide, 764 Cyanuric derivatives, 766 D. DALTON, 34-38 Dalton's law of gases, 60 ; deviations from, 70 Dalton's law of partial pressures, 284 Davy lamp, 228 Davy, Sir Humphry, 38 Deliquescence, 280 Density of gases, 103 Dialysis, 819 Diumide, 470 Diamidophosphoric acid, 610 Diamond, 658 ; artificial production of, 661 ; sources of, 660 ' Dicyanamide, 766 Dicyanogen, 749 Diffusion of gases, 63 ; illustrations of, 67 Dihydroxylaminesulphonic acid, 521 Di imidodiphosphaminic acid, 611 Di-imidodiphosphoric acid, 611 Dilute solutions, properties of, 111 Dimetaphosphates, 597 INDEX Dimorphism, 870 Dioptase, 821 Dipeiiodates, 336 Di phosphoric acid, 601 Dissociation, 109 Dissociation, electrolytic, 116 Distilled water, 290 " Distorted crystals, 835 Disulphuryl chloride, 406 Dithionic acid, 414 Dithiophosphoric acid, 607 Double cyanides, 757 Double refraction, 872 Dowson gas, 791 Drammond light, 268 Dulong and Petit's law of specific heats, 39 hydrogen, 267; Smithells' researches on, 799 ; structure of, 792 Flame of Bunsen burner, 797 Flame of burning hydrocarbons, 793 Flint, 817 Fluoboric acid, 646 Fluorides, 152 Fluorine, atomic weight of, 147 ; de- tection of, 151 ; isolation of, 143 : occurrence of, 142 ; properties of, 146 Formulae, chemical, 100 ; constitu- tional, 128 Freezing mixtures, 282 Freezing point, depression of, by dis- solved substances, 314 Fuming sulphuric acid, 402 E. EARTH'S crust, composition of, 55 "Eau de Javelles," 315 Efflorescence, 280 Effusion of gases, 70 Electrolytes, 116 Electrolytic dissociation, 116 Elements, allotropic modifications of, 56 ; Aristotelian doctrine of, 3 ; atomic weights of, 52 ; Boyle's view of, 11 ; classification of, 53 ; com- bining weights of, 93 ; definition of, 51 ; distribution of, 54 ; equivalents of, 92 ; Geber's views of, 7 ; list of, 52 Elixir vitse, 8 Emerald. 821 Enstatite, 821 Equations, chemical, 102 Equivalents, 92 Ethane, 6S8 Ethene, 699 Ethine, 701 Ethylene, 693, 699 ; preparation of, 700 ; properties of, 701 Ethyl hydride, 698 Euchlorine, 322 Explosions in coal mines, 695 Explosion in gases, phenomena of, 265 F. FELSPAR, 821 Ferricyanic acid, 757 Ferrocyanic acid, 757 Fire-damp, 695 Flame, Davy's researches on, 793, 794 ; dissection of, 799 ; Frankland's re- searches on, 794 ; Lewes' researches on, 795, 797 ; luminous, rendered non-luminous by addition of other gases, 797 ; nature of, 792 ; oxy- G. GARNET, 821 Gas-carbon, 674 Gas-liquor, 454, 770 Gaseous and liquid states, continuity of, 72 Gases, absorption of by water, 284 ; critical point of, 73 ; critical pres- sure of, 73 ; density of, 103 ; deter- mination of molecular weight, 105 ; diffusion of, 63 ; effusion of, 70 ; ex- plosion of, 265 ; kinetic theory of, 60 ; liquefaction of, 74 ; molecular weight of, 99 ; properties of, 59 ; relation of volume to pressure, 59 ; relation of volume to temperature, 60 ; transpi- ration of, 66 ; van Helmont's inves- tigations of, 9 Gay-Lussac, 38 Gay-Lussac's law, 60 Gay-Lussac tower, employment in sul- phuric acid manufacture, 389 Geber, researches of, 5, 7 Generator gas, 790 Glauber, 10 Glover's tower, employment in sulphuric acid manufacture, 385 Goniometer, 837 Graphite, 667 ; artificial production of, 668 ; properties of, 668, 670 ; uses of, 671 Graphitic acid, 729 Graphitite, 670 Graphon, 730 Guanidine, 766 H. HALLOYSITE, 822 Halogens, the, 142 Hard water, 298 Hartshorn, spirits of, 453 Heat of combination, 230 INDEX 883 Heat of liquidity, 273 Heavy carburetted hydrogen, 700 Hemihedral crystals/ 833 Hemimorphous crystals, 833 Hemitropic crystals, 835 Hexagonal system of crystallography, 847 Hexametaphosphates, 598 Hexathionic acid, 422 Higgins, 34 Holohedral crystals, 833 Hooke, 12 Humboldt, 39 Hydrated oxides, 234 Hydrazine, 470 ' Hydrazine hydrate, 471 Hydrazoic acid, 472 Hydriodic acid, 206 ; preparation of, 206 ; properties of, 208 Hydrobromic acid, 193 ; preparation of, 193 ; properties of, 196 Hydrocarbons, constitution of, 690 Hydrochloric acid, 168 ; composition of, 175 ; formation of, 169 ; manu- facture of, 179 ; occurrence of, 168 ; preparation of, 173 ; properties of, 174 ; properties of aqueous solutions of, 118 Hydrocyanic acid, 752 ; detection and estimation of, 756, 758 ; preparation of, 752 ; properties of, 754 Hydrocyanic acid hydriodide, 759 Hydrocyanic acid hydrobromide, 759 Hydrocyanic acid hydrochloride, 759 Hydrofluoric acid, 147 ; preparation of, 148 ; properties of, 150 ; vapour density of 150 Hydrofluosilicic acid, 806 Hydrogen, 12$ ; absorption of, by metals, 136 ; experiments with, 139 ; explosion of, with chlorine, 169 ; ex- Elosion of, with oxygen, 261 ; lique- iction of, 135 ; occurrence of, 129 ; preparation of, 129 ; properties of, 134 ; spectrum of, 139 Hydrogen arsenide, gaseous, 616 ; solid, 618 Hydrogen bromide, 193 Hydrogen chloride, 168 Hydrogen dioxide, 308 ; detection and estimation of, 312 ; preparation of, 309 ; properties of, 309 Hydrogen disulphide, 357 Hydrogen fluoride, 147 Hydrogen iodide, 206 Hydrogen monosulphide, 349 Hydrogen monoxide, 244 Hydrogen peroxide, 308 Hydrogen persulphide, 355 Hydrogen phosphide, gaseous 566 ; liquid, 570 ; solid, 573 Hydrogen selenide, *27 Hydrogen sulphate, 377 Hydrogen tellnride, 439 Hydrogenium, 138 Hydrographitic acid, 730 Hydrosulphurous acid, 408 Hydrothiosulphoprussic acid, 745 Hydroxides, 234 Hydroxycarbamide, 743 Hydroxylamine, 475 Hydroxylaminedisulphonic acid, 520 Hydroxylaminesulphonic acid, 521 Hygrometry, 542 Hypobromous acid, 329 Hypochlorous acid, 317 Hypochlorous anhydride, 315 Hypoiodous acid, 332 Hyponitrous acid, 504 Hypophosphites, 582 Hypophosphoric acid, 586 Hypo phosphorous acid, 581 Hyposulphites, 408 Hyposulph'urous acid, 408 I. ICE, artificial production of, 458 ; crys- talline form of, 274 ; melting point of, 273 Ignition, temperature of, 228 Imidodiphosphoric acid, 611 Imidodisulphonates, 520 Imidosulphurylamide, 522 Indestructibility of matter, 47 Induction, photochemical, 172 Introduction, historical, 3 lodates, 333 lodic acid, 332 Iodide of nitrogen, 480 Iodides, 209 Iodine, 199 ; atomic weight of, 206 ; detection and estimation of, 211 ; occurrence of, 199 : preparation of, 199 ; properties of, 203 Iodine bromide, 214 Iodine fluoride, 212 Iodine monochloride, 213 Iodine pentoxide, 332 Iodine trichloride, 213 Isometric system of crystallography, 840 Isomorphism, 39, 865 Isuretine, 743 J. JET, 688 K. KlESELGUHR, 814 Kirwan, 32, 34 884 INDEX L. LAMP-BLACK, 673 Laughing-gas, 503 Lavoisier, 24-42 Law of partial pressures, 284 Laws of chemical combination, 86 Lefebre, 10 Lemery, 10 Leucite, 821 Libavius, 9 Light carburetted hydrogen, 700 Lignite, 688 Lime light, 268 Liquefaction of gases, 74 Liquefied gases, tension of, 74 Liquid and gaseous states, continuity of, 72 Liquidity, heat of, 273 Liquids, boiling points of, 85 Liquor ammonise, 462 Lucifer matches, 564 Lully, Raymond, 6 M. MACLES, 834 Marggraf, 15 Marsh-gas, 694 Marsh's test for arsenic, 637 Matches, 565 Matter, different states of, 41 Matter, indestructibility of, 47^ Mayow, 12 Mesoperiodates, 336 Metaboric acid, 654 Metals, transmutation of, 5 Metaperiodates, 336 Metaphosphoric acid, 591, 596 Metarsenates, 629 Metasilicic acid, 820 Metathioarsenates, 632 Methane, 694 ; preparation of, 697 ; properties of, 698 Methenylamine, 759 Methyl hydride, 694 Metrical system, comparison of, with English measures. 874, 875 Microbes in the atmosphere, 546 Milk of sulphur, 345 Milk quartz, 815 Mimetisite, 867 Mineral waters, 296 Mitscherlich, 39 Moisture, atmospheric, 541 Molecular motion, 61 Molecular weight, determination of, in solution, 111 Molecular weight, experimental deter- mination of, 104 Molecular weight of gases, 99 Molecule, definition of the term, 98 Molecules, size of, 97; velocity of, 61, 63 Monoclinic system of crystallography, 860 Monometaphosphates, 597 Monosy in metric system of crystallo- graphy, 860 Monothiophosphoric acid, 607 Muriatic acid, 168 Muscovite, 821 N. NATROLITE, 822 Natural waters, 289 Nitric acid, 482 ; action of metals on, 494 ; detection and estimation of. 495 ; formation of, 482 ; hydrates of, 488 ; manufacture of, 488 ; prepara- tion of, 483 ; properties of, 486 r properties of aqueous solutions of, 486 Nitric anhydride, 497 Nitric oxide, 505 Nitrilosulphonates, 519 Nitrilotrimetaphosphoric acid, 612 Nitrites, interaction with sulphites,. 522 Nitrogen, 446 ; assimilation of, by plants, 451 ; combustion of, 451 ; discovery of, 446 ; liquefaction of, 450 ; preparation of, 447 ; properties of, 450 ; spectrum of, 451 Nitrogen bromide, 480 Nitrogen chloride, 477 Nitrogen dioxitte, 505 Nitrogen iodide, 480 Nitrogen mjmoxide, 499 Nitrogen pentoxide, 497 Nitrogen peroxide, 512 Nitrogen selenide, 516 Nitrogen sulphide, 516 Nitrogen tetroxide, 512 ^Nitrogen trioxide, 508 Nitroguanidine, 768 Nitrometer, Lunge's, 496 Nitrosohydroxylaminesulphonic acid, 521 Nitrosulphonic acid, 516 Nitrosulphonic anhydride, 499, 518 Nitrosulphonic chloride, 5 1 7 Nitrosyl bromide, 511 Nitrosyl chloride, 510 Nitrosyl tri bromide, 511 Nitrous acid, 509 Nitrous anhydride, 508 Nitrous oxide, 499 Nitroxypyrosulphuric acid, 518 Nomenclature, chemical, 116 0. OCCLUSION, 137 Oil gas, 788 INDEX 885 Olefiant gas, 699 defines, 693 Oliviue, 821 Opal, 816 Optical relations of crystals, 872 Organic constituents of water, 300 Organic matter, atmospheric, 544 Organic substances, first artificial pre- paration of, 40 Orpiment, 630 Ortho-arsenates, 628 Orthoboric acid, 650 Orthoclase, 821 Orthophosphoric acid, 591, 592 Orthosilicic acid, 819 Orthothioarsenates, 632 Osmotic pressure, 112 Oxides, 234 Oxy-acids of chlorine, constitution of, 328 Oxygen, 214 ; atomic weight of, 255 ; discovery of, 15 ; explosion of, with hydrogen, 261 ; occurrence of, 214 ; preparation of, 215 ; properties of, 222; separation of, from nitrogen, 289 Oxy-hydrogen flame, 267 Oxynitrosulphonic anhydride, 518 Oxythiocarbamic acid, 745 Ozone, 235 ; atmospheric, 240 ; mole- cular formula of, 239 ; production of, 235 ; properties of, 242 P. PARACELSUS, 9 Paracyanogcn, 752 Paraffins, 692 Paraperiodates, 336 Parasulphatammon, 520 Parrot coal, 684 Peat, 688 Pentathionic acid, 419 Perbromic acid, 331 Perchlorates, 327 Perchloric acid, 324 ; hydrates of, 327 Periodates, 335 Periodic acid, 335 ; hydrates of, 335 Peroxides, 234 Persulphuric acid, 410 Petalite, 821 Pharaoh's serpents, 763 Phenakite, 820 Phlogistic theory, 13, 23, 31 Phosgene gas, 731 Phospham, 610 Phosphamide, 610 Phosphamidic acid, 610 Phosphine, 566 Phosphites, 585 Phosphoniuin'bromide, 570 Phosphonium iodide, 570 Phosphoric acid, 590 ; determination of, 601 Phosphoric acids, constitution of, 591 ; poisonous action of, 592 Phosphorous acid, 584 Phosphorous anhydride, 583 Phosphorous diamide, 609 Phosphorous oxide, 583 Phosphorus, 549 ; allotropic modifica- tions of, 554 ; amorphous, 561 ; black, 564 ; detection of, 559 ; dis- covery of, 549 ; luminosity of, 556 ; manufacture of, 551 ; metallic, 563 ; molecular weight of, 555 ; occurrence of, 550 ; octohedral, 554 ; proj evties of, 554 ; poisonous action of, 560 ; red, 561 ; rhombohedral, 563 Phosphorus bromonitride, 612 Phosphorus chlorobromide, 579 Phosphorus chloronitride, 612 Phosphorus di-iodide, 580 Phosphorus oxybromide, 604 Phosphorus oxybromochloride, 605 Phosphorus oxychloride, 602 Phosphorus oxyfluoride, 602 Phosphorus pentabromide, 579 Phosphorus pentachloride, 575 Phosphorus pentafluoride, 573 Phosphorus pentasulphide, 606 Phosphorus pentiodide, 580 Phosphorus pentoxide, 587 Phosphorus suboxide, 581 Phosphorus sulphides, 606 Phosphorus sulphoxide, 607 Phosphorus tetroxide, 586 Phosphorus tribromide, 578 Phosphorus trichloride, 574 Phosphorus tricyanide, 768 Phosphorus trifluoride, 573 Phosphorus trifluorodibrouiide, 579 Phosphorus trifluorodichloride, 578 Phosphorus tri-iodide, 580 Phosphoryl bromide, 604 Phosphoryl bromochloride, 605 Phosphoryl chloride, 602 Phosphoryl fluoride, 602 Phosphoryl nitride, 611 Phosphoryl triamide, 612 Phosphuretted hydrogen, 567 Photochemical induction, 172 Photometer, 781 Plastic sulphur, 345 Plumbago, 667 Polarization, circular, 873 ; in crystal- line media, 873 Portable gas, 788 Potash mica, 821 Potassium hydroxylaminedisulphonate, 520 Potassium imidodisulphonate, 520 Potassium nitrilosulphonate, 519 Potassium sulphaziuate, 521 Pott, 15 Priestley, 15-22 Prismatic system of crystallography, 853 88(5 INDEX Producer gas, 790 Proust, 33 Prussic acid, 752 Pyro-arsenates, 629 Pyroboric acid, 654 Pyro-electric action of crystals, 872 Pyrographitic acid, 730 Pyrographitic anhydride, 730 Pyromorphite, 867 Pyrophosphaminic acids, 611 Pyrophosphoric acid, 591, 599 Pyrophosphoryl chloride, 604 Pyrophosphoryl thiobromide, 609 Pyrothioarsenates, 632 QUADRATIC system of crystallography, 853 Quartz, 814 ; coloured varieties of, 815 R. RADICALS, compound, 470 Rain-water, 295 Raoult's method of determining mole- cular weights, 114 Realgar, 630 Red phosphorus, 561 Regelation, 274 Regular system of crystallography, 840 Reinsch's test for arsenic, 640 Rhombic system of crystallography, 857 Richter, 32 River water, 305 Rose quartz, 816 Ruby sulphur, 630 SAFETY LAMP, 228 Sal ammoniac, 453 Saline waters, 297 Salts, 235 Sand, 815 Sceptical Chymist (Boyle), 10 Scheele, 22 Sea-water, 306 Selenic acid, 433 Selenious acid, 431 Selenium, 423 ; action of light on, 426 ; atomic weight of, 427 ; preparation of, 424 ; properties of, 425 Selenium chlorobromide, 430 Selenium dioxide, 431 Selenium fluoride, 431 Selenium moniodide, 430 Selenium moriobromide, 429 Selenium monochloride, 428 Selenium oxychloride, 433 Selenium tetrabromide, 430 Selenium tetrachloride, 429 Selenium tetriodide, 430 Seleniuretted hydrogen, 427 Selenosulphur trioxide, 435 Selenosulphuric acid, 435 Selenotrithionic acid, 436 Selenourea, 746 Selenyl bromide, 433 Selenyl chloride, 433 Serpentine, 821 Silica, 814 Silica, amorphous, 816 Silicates, 820 ; artificial production of, 822 ; classification of, 821 Silicic acid, 819 Silicious waters, 297 Silico-bromoform, 812 Silico- chloroform, 810 Silicofluoric acid, 806 Silicofluorides, 808 Silicotbrmic anhydride, 811 Silico -iodoform, 813 Silicomethane, 804 Silicon, 800 ; amorphous, 801 ; atomic weight of, 803 ; crystallized, 802 ; mixed halogen derivatives of, 813 ; occurrence of, 800 ; preparation of, 801 ; properties of, 801 Silicon carbide, 825 Silicon carbonitride, 824 Silicon carboxide, 825 Silicon chlorohydrosulphide, 824 Silicon dioxide, 814 Silicon disulphide, 823 Silicon hydride, 804 Silicon nitride, 824 Silicon oxychloride, 822 Silicon oxysulphide, 823 Silicon subsulphide, 823 Silicon tetrabromide, 811 Silicon tetrachloride, 808 Silicon tetrafluoride, 806 Silicon tetra-iodide, 812 Silicon tribromide, 812 Silicon trichloride, 810 Silicon tri-iodide, 812 Silico-oxalic acid, 813 Smoky quartz, 815 Soap test for hardness of water, 299 Soda-bleach, 313 Soft water, 298 Solutions, electrolytic dissociation of, 283 ; properties of, 282 Specific heats, Dulong and Petit's law of, 39 Specific rotatory power, 873 Spirits of hartshorn, 453 Springs, thermal, 296 Spring-water, 296 S ahl, 13 Steam, latent heat of, 275* ; volumetric composition of, 253 Suffioni, 651 Sulphamide, 523 INDEX 887 Sulphatammon, 520 Sulphates, 400 Sulphazinates, 521 Sulphazotinatt s, 522 Sulphimide, 525 Sulphites, 372 Sulphites, interaction with nitrites, 522 Sulphites, isomerism of, 373 Sulpho-arsenates, 632 Sulpho-arsenites, 631 Sulphocyanic acid, 762 Sulphoselenoxytetrachloride, 436 Sulphur, 336 a-Sulphur, 342 )8-Sulphur, 343 7-Sulphur, 345 S-Sulphur, 347 Sulphur, allotropic modifications of, 342 ; atomic weight of, 349 ; colloi- dal, 346 ; crystalline forms of, 342 ; detection and determination of, 348 ; flowers of, 340, 346 ; manufacture of, 338 ; milk of, 345 ; occurrence of, 336 ; properties of, 342 ; plastic, 345 ; recovery of, from waste pro- ducts, 341 ; spectrum of, 348 Sulphur, vapour, molecular weight of, 347 Sulphur dichloride, 360 Sulphur dioxide, 364 ; bleaching action of, 369 ; detection and estimation of, 370 ; liquefaction of, 368 ; prepara- tion of, 366 ; properties of, 367 Sulphur diphosphide, 605 Sulphur fluoride, 363 Sulphur heptoxide, 410 Sulphur hexiodide, 363 Sulphur moniodide^-362 Sulphur monobromide, 362 Sulphur monochloride, 359 Sulphur oxytetrachloride, 406 Sulphur sesquioxide, 407 Sulphur subiodide, 363 Sulphur tetrabromide, 362 Sulphur tetrachloride, 360 Sulphur tetraphosphide, 605 Sulphur trioxide, 374 Sulphuretted hydrogen, 349 ; action of on metals, 355 ; composition of, 353 ; preparation of, 350 ; properties of, 352 Sulphuretted waters, 297 Sulphuric acid, 377 ; action of, on metals, 399 ; action of water on, 398 ; fuming, 402' ; history of, 377 ; hy- drates of, 399 ; manufacture of, 381 ; properties of, 397 ; purification of, 396 ; rectification of, 391 ; specific gravity of aqueous solutions of, 401 ; theory of the formation of, 379 Sulphuric anhydride, 374 Sulphurous acid, 371 Sulphurous anhydride, 364 Sulphuryl bromide, 407 Sulphuryl chloride, 405 Sulphuryl hydroxychloride, 404 Symbols, chemical, 100 Synthesis, definition of, 86 T. TALC, 821 Tellurates, 445 Telltiretted hydrogen, 439 Tellurium, 436 ; atomic weight of. 438 ; preparatiop of, 437 ; properties of, 438 Tellurium dibromide, 441 Tellurium dichloride, 440 Tellurium di-iodide, 441 Tellurium dioxide, 442 Tellurium disulphide, 445 Tellurium hydride, 439 Tellurium monoxide, 442 Tellurium oxybromide, 445 Tellurium oxychloride, 445 Tellurium sulphate, 442 Tellurium sulphoxide, 446 Tellurium tetrabromide, 441 Tellurium tetrachloride, 440 Tellurium tetrafluoride, 442 Tellurium tetriodide, 441 Tellurium trioxide, 443 Tellurous acid, 443 Temperature, absolute zero of, 62 Tension of aqueous vapour, 276 Tetartohedral crystals, 833 Tetragonal system of crystallography, 853 Tetrametaphosphates, 598 Tetraphosphoric acid, 601 Tetraphosphorus trisulphide, 606 Tetrathionic acid, 418 Thermal relations of crystals, 871 Thermal springs, 296 Thermo-chemistry, 233 Thioarsenates, 632 Thioarsenites, 631 Thiocarbamic acid, 745 Thiocarbamide, 746 Thiocarbonic acid, 735 Thiocarbouyl chloride, 736 Thiocyanic acid, 762 Thiocyanic anhydride, 764 Thionyl bromide, 374 Thionyl chloride, 373 Thiophosphoric acid, 607 Thiophosphoryl bromide, 608 Thiophosphoryl chloride, 608 Thiophosphoryl fluoride, 608 Thiosulphuric acid, 412 Thiourea, 746 Thomson, 38 Torricellian vacuum, 524 Transmutation of metals, 5 Transpiration of gases, 66 888 INDEX Tricarbon disulphide, 735 Trichlorosilicomethane, 810 Triclinic system of crystallography, 863 Tridymite, 815 Timetaphosphates, 598 Trimorphism, 870 Triphosphorus hexasulphide, 606 Trithionic acid, 416 Trithiophosphoric acid, 607 Turf, 688 Twin crystals, 834 U. UN i AXIAL crystals, 873 Urea, 739 ; detection and estimation of, 742 ; preparation of, 740 ; properties of, 740 ; synthesis of, 739 Urea hydrochloride, 741 Urea nitrate, 741 Urea oxalate, 741 V. VACUUM, Torricellian, 524 Valency, 127 Valentine, Basil, 8 Van Helmont, 9 Vanadinite, 867 Vapour density, determination of, 106 Vapour pressure, diminution of, by dis- solved substances, 115 Ventilation, 545 Volume, combination by, 97 284 ; alkaline, 297 ; analysis of, 300 ; bacteriological examination of, 300 ; carbonated, 297 ; Cavendish's investi- gations of, 20 ; chalybeate, 297 ; composition of, by volume, 250 ; com- position of, by weight, 255 ; dis- covery of the composition of, 20, 244 ; dissolved gases in, 291 ; distillation of, 290 ; electrolytic analysis of, 251 ; eudiometric synthesis of, 246 ; ex- pansion and contraction of, 270 ; freezing point of, 273 ; hard, 2C8 ; latent heat of, 272 ; mineral, i96 ; natural, 289 ; organic constituents of, 300 ; permanent hardness of, 299 ; point of maximum density of, 271 ; properties of, 270 ; purification of, 290 ; rain, 295 ; river, 305 ; saline, 297 ; sea, 306 ; silicious, 297 ; soft, 289 ; solvent action of, 278 ; s]>ecih'e gravity of, at different temperatures, 272 ; spring, 296 ; sulphuretted, 297 ; synthesis of, by volume, 250 ; tem- porary hardness of, 299 Water-gas, 790 ; carburetted, 791 Water of crystallization, 103, 279 ; volume occupied by, 281 Weight, combination by, 87 Weights of elements in compounds, relations of, 87-95 White arsenic, 622 Willis, 10 W r ollaston, 38 Wollastonite, 821 Wood-charcoal, 674 Wood-gas, 787 Wood-tar, 788 W. WACKENRODER'S solution. 420, 422 Water, 244 ; absorption of gases by, ZEOLITES, 822 Zosimus of Panopolis, 4 (V) - Myv " I / / 3 see. UNIVERSITY OF CALIFORNIA LIBRARY