iraiv <^^ ON inv iL MEMCAL .S€1HI©(Q)L California College of Pharmac^ AN ELEMENTARY TREATISE ON QUALITATIVE CHEMICAL ANALYSIS BY J. F.^aLERS Professou of Chemistry, Mercer University, Georgia California CoHege of Pharmacy REVISED EDITION GTNN AND COMPANY BOSTON • NEW YORK • CHICAGO • LONDON Copyright, 1900, 1909 By J. F. SELLERS Alili RIGHTS RESERVED CINN Sz COMPANY • PRO- PRIETORS . BOSTON . U.S.A. 3 PREFACE So many books on analytical chemistry are already in print that the question may be raised whether it is wise to add still another to their number; and therefore the author desires to present the following reasons which seem to him to justify the publication of the present work. Most writers on analytical chemistry have gone either to the one or the other of two extremes. First, there are those who, like Fresenius or Prescott and Johnson, have endeavored to cover the entire field and to include the whole detail of analytical chemistry. Their works are indispensable to teachers and to students who make chemistry a specialty ; but for beginners, who may not give more than one year of eight or ten hours a week to the subject, they are fq,r too voluminous. On the other hand, there are those whose ardor for brevity has led them to the other extreme of condensing their material into " tables " and " schemes," — by which means they have magnified the empirical and have minimized the rational aspect of the subject, to its considerable detriment as a factor in liberal education. In order to avoid either extreme the writer presents this elementary treatise having these features : — 1. A course short enough to be digested during the time allotted in an ordinary college curriculum, but at iii 39971 IV PBEFA CE the same time intended to magnify the scientific and pedagogical nature of analytical chemistry. 2. A course both practical and progressive, — practical, in that the student can master the methods and principles of chemical analysis, and become a practical analyst ; pro- gressive, in that the chief aim of the book is to prepare the student thoroughly for advanced university work. 3. A course selected from the most recent and approved methods recorded in the best literature and verified by actual application in the author's laboratory. Among some of the improved methods are mentioned : — (a) Reddrop's application of normal solutions to quali- tative analysis. Chemical News, May, 1890. (h) Hofmann's separation of arsenic, antimony, and tin, by modification of Marsh's test. Fresenius' Quali- tative Analysis, 1897 edition, p. 299. (c) Parr's separation of aluminum, chromium, and iron, by means of sodium peroxide. This method commends itself for its accuracy, its briefness and simplicity, and its certainty in detecting aluminum. Other methods depend- ing on sodium hydroxide are defective, in that the reagent itself generally contains aluminuni salts ; sodium peroxide, by reason of its manufacture, does not contain perceptible traces of such salts. Journ. Amer. Chem. Soc., 19, p. 341. (d) Fresenius and E-uppert's separation of barium, strontium, and calcium, by means of the differences of solubility of their nitrates in ether-alcohol. Fres. Qual. Anal., p. 160. (e) Hager's separation of chlorine, bromine, and iodine, by means of the differences of solubility of their silver salts in ammonium " sesqui " carbonate. Fres. Qual. Anal., p. 378. PREFACE V 4. A course free, as is thought wise, from the mechan- ical schemes in qualitative analysis. To this end, many of the usual tables of separation are omitted, and in their place some suggestive hints are given after the list of reactions for each group. Thus the student is expected and encouraged to exercise his judgment in selecting methods of analysis. 5. A course conformable to the modern dissociation theory of solutions. For example, why is the activity of certain acids modified by adding the salts of those acids J or, more specifically, why is the solvent power of acetic acid decreased by adding some sodium acetate ? 6. A course giving more than ordinary emphasis to the spectroscope. Though spectroscopy is not chemical analysis, it possesses superior advantages over the chem- ical methods in these particulars : — (a) Methods of greater exactness and readiness of execution. (b) Methods superior for the preliminary detection of the alkali and alkali-earth metals. This is important, especially when the alkali-earth metals are combined with phosphoric, oxalic, and hydrofluoric acids. (c) Methods superior for detecting certain metals, which, under some conditions, are evasive; e.(/., aluminum, man- ganese, and magnesium. It is obvious that the study of the theory of solution and of spectroscopy may either be taken up in the order of the text or reserved for the last work in the course; and also that these subjects may be omitted entirely if a very elementary course is desired. In the latter case it would be possible also to omit the discussion of the VI PREFACE analysis for the metals of the third group in the pres- ence of phosphoric acid, and those portions of Part II which are printed in small type. The discussion of solutions in the brief space available in this book is necessarily much condensed, and possibly it is somewhat abstract and uninviting ; but in the author's opinion its introduction is desirable. Its purpose is to provide the student of qualitative analysis with the means for a rational interpretation of many apparently irrational reactions, and to help prepare him for the next stage of his chemical education, — namely, the study of quantitative analysis, — where the application of the laws of solutions is more abundant. No other text-book on qualitative analysis, within the author's knowledge, incorporates this dissociation theory of solution ; but its adaptability to qualitative instruction is shown by the fact that during the past half decade many teachers of the subject have devoted more or less time in their lectures to the practical application of the theory. In the preparation of this book the following literature has been consulted : — 1. Many of the smaller text-books on qualitative analysis, including Noyes's, Newth's, and Volhard and Zimmer- mann's. 2. Standard works on general and analytical chemistry, including Watt's Chemical Dictionary ; Eoscoe and Schor- lemmer's Treatise on Chemistry ; Mendeleeff's Principles of Chemistry ; Ostwald's and Nernst's works on physical chem- istry ; Vogel's, Landaur's, and Eoscoe's works on spectrum analysis ; Fresenius' works — the latest editions. 3. Memoirs in American and foreign chemical journals. PREFACE VU Grateful acknowledgment is made to Dr. E. W. Jones of the University of Mississippi, for his painstaking criti- cism of the manuscript of this little book. The author learned the chemical alphabet and received much inspira- tion and encouragement from this excellent teacher. Appreciative mention also is made of the following gen- tlemen : Dr. H. C. White of the University of Georgia, for valuable suggestions as regards the adaptability of the book to elementary college work ; Dr. J. W. Mallet of the University of Virginia, Dr. J. Stieglitz of the Uni- versity of Chicago, and Dr. E. Eenouf of Johns Hopkins University, for opinions concerning modern theories of solution; and Mr. H. V. Jackson of Mercer University, for general assistance. t t^ a Macon, Ga., September, 1900 PREFACE TO SECOND EDITION The more important modifications made in this edition of the book are the appending of 13 pages of reference notes (see p. 163) and tables, revision of several of the processes of separation, and correction of a number of typographical errors. For criticism and proof reading the author is indebted to many of his friends, among whom may be mentioned Professor W. H. Emerson and Dr. G. H. Boggs of the Georgia School of Technology, Dr. J. P. Montgomery of the Mississippi Agricultural and Mechanical College, Dr. Homer V. Black of the University of Georgia, Professor C. W. Steed of Mercer University, Professor G. P. Shingler of Emory College, and Professor Alexander Smith of the University of Chicago. J. F. S. Macon, Ga., May, 1909 CONTENTS PART I — ANALYTICAL OPERATIONS CHAPTEU I. Introduction II. Theory of Analytical Operations . III. Methods of Analytical Separation . IV. Flame Coloration and Spectroscopy V. List and Preparation of Reagents . VI. Systems of Analytical Examination PAGB5 1 5 26 50 65 73 PART II — REACTIONS AND SEPARATIONS VII. Metals of Group I 77 VIIL Metals of Group II 82 IX. Metals of Group III 100 X. Metals of Group IV 114 XI. Metals of Group V 121 XII. Metals of Group VI 126 Xiri. Acids of Group I 130 XIV. Acids of Group II 141 XV. Acids of Group III 148 XVI. The Systematic Procedure of Analysis . . 151 NOTES 163 INDEX 175 ix CHEMICAL ANALYSIS Part I — Analytical Operations CHAPTER I INTRODUCTION The science of chemistry is commonly subdivided, for purposes of convenience in reference and teaching, into several tolerably distinct branches. The usual classifi- cation is into the main divisions of inorganic and organic chemistry, each of which may in turn be further divided into descriptive^ theoretical, and analytical chemistry. Furthermore, analytical chemistry may itself be sepa- rated into the subdivisions of qualitative and quantitative analysis ; the former having for its object the detec- tion of chemical elements and compounds, and the latter the relative proportions of such substances. Analytical chemistry is commonly taught as a dis- tinct branch, but it is not independent of the other divisions of the science ; and hence, in all discussions in this book, both as to theory and manipulation, the presumption is that the student has, in the beginning of the course, a fair knowledge of the elements of general chemistry. It obviously is essential to success in analysis that the analyst should have a clear idea of the operations 1 22 CHEMICAL ANALYSIS involved in his work, as well as of the compounds with which he is dealing ; and therefore, though both manipulation and theory are assumed to have been studied- to some extent in connection with general chemistry, it is deemed well to review many of the ordinary operations from the analytical standpoint. The first part of this book is devoted largely to such a review ; and it is earnestly recommended that it be studied closely, and that all of the experiments there given be carefully performed. It is true that the time spent on this preliminary work will delay somewhat the beginning of actual analysis ; but it is believed that the student will be repaid in the end by the acquisition of a clearer conception of the work and of more skill in the manipulation of apparatus. It should be remembered that it is far easier to form good habits than to correct bad ones ; and so from the beginning the attention of the student should contin- ually be directed to the importance of the following details which, though simple and apparently insignifi- cant, are absolutely essential to continued success in analysis. CARE OF APPARATUS (a) Keep all apparatus clean. This can best be done by cleaning the desk and apparatus before leaving for the day. Of course this does not apply to apparatus connected with unfinished experiments. (b) When vessels containing materials of unfinished experiments are to be set aside, they should be properly labeled. INTRODUCTION 3 (c) Provide towels, clean rags, soap, and a covering for the clothes, either a long apron or a workingman's overalls. (d) Have a place for all reagents and apparatus, and keep them in their place. Reagents for general use should not be kept at the individual desks. This is a source of great annoyance and injustice to one's neighbors. (e) Use all care in keeping the reagents pure. Stoppers should not be placed on the desk while using the bottles, but held between the fingers. No foreign objects should be dipped into the bottles, nor should any excess of reagents be poured back into the bottles. (/) Use small quantities of reagents. It is best to add liquid reagents, drop by drop, with frequent shaking of the test-tube, so that secondary reactions can be observed. LABORATORY NOTES Provide a well-bound notebook for the subject and use it for nothing else. Keep accurate and methodical records of all experiments performed. These records should be made during or immediately following the performance of the experiment, and not transferred or erased afterwards. Original notes of an unsuccessful experiment are more valuable than a well-written description of a successful experiment, if the latter is composed in the absence of the experiment. Some states prescribe by law that chemists, in giving expert testimony before the courts, shall present only 4 CHEMICAL ANALYSIS such data as are recorded in the presence of the experiments. If desirable the original notes may be written on alternate lines or pages, and other notes of interpre- tation added at leisure. But the latter should be recorded with differently colored ink, or otherwise distinguished, in order that the original notes be not confused with subsequent additions. CHAPTER II THEORY OF ANALYTICAL OPERATIONS Nature of Anal3rtical Chemistry. — Analytical chemistry has already been defined as the art of recognizing the elements, or compounds, which may be present in any substance ; and, as the nature of the art implies, it commonly is practiced upon mixtures of one kind or another. Such mixtures may be mechanical only; and in such a case, if the elements of the mixture are sufficiently characterized by their color, crystalline form, or other external properties, it may be possible to identify or even to separate them by purely mechan- ical means. But the mixtures with which the chemist most commonly has to deal are those in which the strictly mechanical element plays a minor part. Such mixtures are produced when, by any appropriate means, two or more substances are brought into such intimate contact that they interpenetrate each other even to their minutest particles — the molecules. We have examples of mixtures of this class in the air, which practically is a homogeneous mixture of its constituent gases and vapors ; in common " solutions^" such as are produced when any suitable material, like salt or sugar, is treated with some liquid which, like water, has the power of "dissolving" the material in. ques- tion ; or in alloys, which are produced when two or 5 6 CHEMICAL ANALYSIS more metals are united by fusion into a mass which is, at all points, of uniform composition. Of these mix- tures, the commonest are the solutions ; and these are so important, from the standpoint of the analytical chemist, that it is desirable to spend some time in a careful study of their properties. Solution. — In a general sense a solution is the prod- uct of the homogeneous absorption of a gas by a gas, or of a gas by a liquid, or of a liquid by a liquid, or of a solid by a liquid; and in recent years the term " solid solution " has been applied to certain homo- geneous solid mixtures of which the alloy mentioned above may serve as the type. But specifically, in speaking of a solution, we have in mind the liquid product of the absorption by a liquid, called the solvent, of a gas, a liquid, or a solid, called the solute. It has been found of all gases, and of some liquids, that they are capable of mixing homogeneously with one another in all proportions ; but, on the contrary, it has not been found possible, under ordinary conditions, to dissolve a gas or a solid in a liquid in any desired proportion. Sooner or later a point is reached where the solvent refuses to take up more of the solute ; and at this point the solution is said to be saturated. In most cases the application of heat to a saturated solu- tion will enable it to absorb more of the solute; and the application of cold will usually result in the sepa- ration of a part of the material already dissolved. In such cases we may recover, a portion of the solute by the mere chilling of its saturated solution; and in cases THEORY OF ANALYTICAL OPERATIONS 7 where the solute is practically as soluble at low tem- peratures as at high ones, we may reach the same end by removing a part or the whole of the solvent by evaporation. It may be mentioned at this point that we have still another means of separating the solute from its solution; viz.^ by the addition to the solu- tion of some material which will decrease the solubility therein of the solute, without changing the identity of the latter. This process of separation is of considerable practical importance, and we shall presently have occa- sion to refer to it again. Experiment 1 (a) Dissolve 5 grams of potassium nitrate in 25 c.c. of dis- tilled water, at a temperature of 15°-25°C. Then add succes- sive portions of 1 gram each, shaking after each addition until all has dissolved that the solution will hold at this temperature. Note the total amount added and then raise the temperature of the solution to about G0°, — as hot as the hand can bear without too much discomfort, — and add more of the finely powdered salt while keeping the solution from cooling. Note the extra amount which is needed at this temperature to saturate the solution. Now cool the solution quickly and note the result. Compare any material which may separate with potassium nitrate. (b) Dissolve 5 grams of common salt in 25 c.c. of distilled water at 15°-25°. Now add successive portions of ^ gram, shaking after each addition until the solution is saturated. Note the total amount dissolved. Raise the temperature as in the preceding part of the experiment, and see whether it is pos- sible to dissolve more salt in the hot solution. Allow any un- dissolved material to settle, and then pour off some of the clear solution into a clean dry test-tube, and cool as much as pos- sible. Note the result. Evaporate a portion of this solution and compare the residue with salt. 8 CHEMICAL ANALYSIS (c) To about 25 c.c. of a clear saturated solution of common salt add 50 c.c. of concentrated hydrochloric acid, stirring all the time. Note the result, allowing the mixture to stand for some minutes. Pour off the clear liquid from any material which may have separated, and press a little of the latter be- tween filter papers, to remove the acid liquor. It will be well to remove the last traces of acid by washing the residue with a little saturated brine. Compare the residue with common salt. It will have been seen, in the performance of these experiments, that the recovered solute is of the same character as the original solute. But there are forms of solution in which this is not the case. Experiment 2 Dissolve a small piece of zinc in dilute hydrochloric acid and evaporate the solution to dryness. Compare the residue with metallic zinc. Solution of this kind may be called chemical solution, in dis- tinction from the simple solutions of Exp. 1. It will be seen that it involves (1) a compound solvent — HCl + water — which itself is a simple solution ; and a solute, Zn ; (2) a chemical reaction between the solute, Zn, and one con- stituent of the solvent, HCl — Zn + 2 HCl = ZnClg + 2 H — in which reaction the identities of the solvent and of the solute are changed ; (3) a simple solution, — ZnCU + water. Chemical solution is usually the result of the mutual reaction between (1) an acid, or a base, and a metal ; (2) an acid, or a base, and a salt; (3) an acid and a base. But it may happen, as when metallic sodium is dissolved in water, that the phenomenon cannot be classified under any of these heads. THEORY OF ANALYTICAL OPERATIONS 9 Simple solution is often a necessary predecessor of chemical solution, as has been seen in Exp. 2 ; and, in general, it prepares the way for chemical action by placing the reagents in close contact. Experiment 3 Mix .5 gram of dry potassium iodide with .5 gram of dry mercuric chloride in a dry mortar, and rub the mixed salts well together with the pestle. Note the result. Add a little water and rub again. Furthermore, simple solution may be necessary to the continuance of chemical action, in order that the products of reaction may be removed from between the reagents. Experiment 4 Add a bit of zinc to 5 c.c. of concentrated sulphuric acid in a test-tube ; leave for a few moments, noting all that happens. Now transfer the contents of the tube to a dish containing 15- 20 c.c. of water. When zinc is treated with conceutrated sulphuric acid,^ chemi- cal action occurs for a short time only, and then ceases entirely. The explanation is probably this : zinc sulphate, insoluble in concentrated sulphuric acid, coats the zinc and prevents further contact of the reagents. The addition of water, in which zinc sulphate is very soluble, removes the coating and permits chemi- cal action to go on once more. Properties of the Solute. — So far in our study of the phenomena of solution, we have considered only those properties of the solute which are associated with its solid condition, — when, in point of fact, it cannot properly be called a solute. Let us now see whether we can discover anything concerning the properties of the true solute, — the body in solution. 10 CHEMICAL ANALYSIS We have seen that a solution which is saturated with a given body at one temperature may acquire the power of dissolving an additional quantity of that body in consequence of an elevation of temperature, and that, on the contrary, it may give up a portion of its solute if its temperature is lowered. That is to say, if we have a " system " consisting of a limited quantity of some saturated solution in contact with an excess of its solute, there will be for any given temperature a concentration of the solution at which there will be a condition of equilibrium between the dissolved and undissolved solute. This condition is entirely analo- gous to that which is observed when a volatile liquid is exposed in contact with a limited volume of air or other gas. In the latter case the liquid will volatilize, — rapidly at first, and afterwards more slowly, — until the concentration of its vapor in the atmosphere to which it is exposed has reached a certain limit which will be dependent on the temperature. With a rise in temperature, more liquid will pass into the state of vapor; with a fall, a portion of the liquid already vaporized will be condensed again. This analogy has been recognized for many years ; but it is now hardly more than a decade since first its completeness was fully demonstrated. Colloids and Crystalloids.^ — It had been shown by Graham (1842)^ that certain colloid solutes, whose solu- tions are not real liquids, but emulsions, cannot pass through porous membranes ^ — such as parchment — and that most crystalloid solutes, whose solutions are real liquids, readily penetrate such septa. He first THEORY OF ANALYTICAL OPERATIONS 11 put separate solutions of a colloid and a crystalloid into separate open cylinders whose bottoms were closed with parchment, and then suspended the cylinders in vessels of water so that the membranes were immersed. After a few hours a large part of the crystalloid had passed through the parchment into the water in the outer vessel; and by renewing this water all of the crystalloid was finally extracted from the cylinder. From the other cylinder, however, no colloid had passed out. Osmosis. — Pfeffer,^ the botanist, in demonstrating and measuring the internal bursting force of plant cells (1877), established the fact that crystalloids, though they do not pass through the so-called " semi-permeable " membranes, — of which protoplasm ^ is a type, — do press strongly against the partition in their futile attempt to penetrate it. Connecting a mercury gauge and ther- mometer with a membrane, composed of a porous cell coated with copper ferrocyanide, and charging this ap- paratus with saccharine solutions of different strengths, he found that different concentrations of solution pro- duced correspondingly different pressures within the apparatus when the temperature was kept constant, and that for any given concentration the pressure varied as the absolute temperature. He showed, therefore, that the relations of concentration, pressure, and tem- perature, which are shown by sugar in its solutions, are identical with those manifested by gases, — of which it will be remembered that the concentration or density of a given mass varies directly as the pressure and inversely as the absolute temperature. To the form 12 CHEMICAL ANALYSIS of tension exercised by the dissolved sugar he gave the name osmotic pressure. Law of Osmotic Pressure. — Van't Hoff^ (1887) found that a large number of solutions behave like that of sugar, and announced the following law: The osmotic pressure of a substance in solution is identical with the pressure which it would exert were it in the form of a gas occupying the same volume {i.e.^ the volume of the solution) at the same temperature.'^ We may conveniently express the simple law which governs the phenomena of gas and osmotic pressures in the following form : — ^, MT MT y= or !) = ——-» p ^ V wherein M represents the number of molecules ^ in a given body of gas, T and p the temperature and pres- sure, and V the volume. Certain gases, such as oxygen, nitrogen, and hydrogen, are obedient to this law within very wide limits; but there are vapors whose behavior with regard to it is apparently anomalous. Evidently V can be made constant, and T and p can be measured with any desired degree of accuracy. And therefore unless there can be a change in the value of M^ any change in T ought to be accompanied by an exactly proportional change in p. Now we find that certain vapors — such as that of ammonium chloride — give greater pressures than can be accounted for by either the value of T, or the value of il[f which is based upon the commonly accepted molecular weight; and, as has been indicated, we find the explanation of this behavior in THEORY OF ANALYTICAL OPERATIONS 13 the fact that the molecule NH^Cl is split up, or " disso- ciated," 1 when we seek to vaporize it, into the smaller molecules NHg and HCl. The analogies between the behavior of gases and substances in solution seem to extend to this phenomenon of dissociation, for it has been observed of many solutes that their osmotic pres- sures are so large as to be accounted for only on the supposition that their molecules are split up in solution and thereby increased in number. Sugar and other bodies of its neutral character obey the simple law as stated above; but acids, bases, and salts in aqueous solution usually exhibit anomalous pressures. Freezing Point Depression. — Moreover, this is not the only evidence which bears upon the question of the dissociation of the molecules of solutes. It is a matter of common knowledge that the boiling and freezing points of aqueous solutions are respectively higher and lower than those of pure water. These relations were studied carefully by Raoult,^ who showed that the phe- nomenon is a general one and that : — (a) When any substances are dissolved in inactive solvents, the changes in the freezing and boiling points of the solvents vary with the amounts of substance dissolved. (b) When equal weights of different substances are dissolved in equal amounts of the same solvent, the changes vary inversely with the molecular weights of the solutes. It was found of many bodies — such as sugar — that equal depressions of the freezing point were pro- duced by the solution of equimolecular proportions in 14 CHEMICAL ANALYSIS water; and in such cases the depressions were exactly in inverse ratio to the molecular weights. In other cases, however, the solutions of equimolecular weights of different substances produced unequal depressions; and the solution of different weights of a given sub- stance produced depressions which were not in exact ratio to the weights so dissolved. In the latter anoma- lous cases the depressions were greater than seemed to be called for by the amount of matter which had been dissolved, as naturally would be the case if the mole- cules of the dissolved substances were dissociated into more numerous and smaller molecules; and the sub- stances which exhibited this behavior were those which show abnormal osmotic pressures, namely, the majority of acids, bases, and salts. In these two pieces of independent evidence we have a strong demonstration of the fact that many sub- stances exhibit, when dissolved in water, a peculiar structural condition in which their molecules are split up into smaller bodies than are indicated by their accepted formulae; and we have to inquire what further evidence we have which will throw light upon the precise nature of these submolecules. We shall find this evidence in connection with the behavior of solutions which are subjected to the passage of an electric current. Electrolytes. — It lias long been known that the con- ductivity exhibited by liquids is unlike that of metallic conductors, in that the latter are not affected chemi- cally by the passage of a current, whereas the former are decomposed with separation at the electrodes — THEORY OF ANALYTICAL OPERATIONS 15 the points where the current enters and leaves the liquid- — of products of varying character. In 1834 Faraday ^ suggested, in explanation of this phenomenon, that the liquid which conducts electricity has in solu- tion a compound whose molecules are divided into freely moving particles, some of which are charged with positive and the rest with negative electricity. He named such compounds electrolytes; and to the hypothetical fragments of their molecules he gave the name of ions. Those which were assumed to be posi- tively charged were called cations, and were either metals, or atom-complexes, like NH^, which react analo- gously to metals. Those bearing a negative charge were termed anions, Sind were such bodies as the halo- gens and acid radicles. The attraction or neutralizing effect which ions of opposite polarities were supposed to exercise upon each other, was held to maintain the identity of the solute until the solution was subjected to the passage of an electric current; whereupon the introduction of electrodes of opposite polarities upset the equilibrium previously existing between the ions and caused them to migrate, — the negative ions going toward the positive electrode, and the positive ions in the opposite direction. The appearance of decomposi- tion products at the electrodes was explained as being due to the union of the ions, upon arrival at those points, to the molecular condition or to compounds with the elements of water. In 1887 it was demonstrated by Arrhenius^ that the solutions which exhibit normal osmotic pressures and freezing point depressions are nonconductors of elec- 16 CHEMICAL ANALYSIS tricity, and that their solutes are not electrolytes. Conversely, the solutions which give abnormal osmotic pressures were proved to contain ionized solutes ; and it was shown, furthermore, by highly accurate experimental methods, that the degree of their con- ductivity is proportional to the amount of dissocia- tion as measured by the osmotic pressure. Between the extremes presented by bodies like sugar, which are characterized by little chemical reactivity and the absence of conductivity and dissociation, and such substances as salts and strong acids and bases, which are distinguished by great reactivity and perfect con- ductivity and dissociation, were arranged the other varieties of chemical compounds, which possess various but proportional activities of the three kinds. With the establishment of these facts the phenom- enon of electrolytic dissociation received a new signifi- cance from the standpoint of analytical chemistry. The behavior of molecules in solution was seen to be chiefly dependent upon their tendency toward or from disso- ciation. The solutions of strongly ionized bodies are characterized rather by the reactions of the ions than by the properties of the undissociated molecules. In the case of sodium chloride, for example, the solution presents certain definite properties which are charac- teristic of the chlorine and sodium ions, and practically none which are characteristic of salt itself. In the case of sugar solutions, on the contrary, such properties as are manifested are those of the sugar molecule alone ; and no indication is to be seen in them of the nature of the constituent elements of sugar. THEORY OF ANALYTICAL OPERATIONS 17 Analytical Significance of Ions. — Borrowing an illustra- tion from Ostwald, let us assume that we have to deal with 50 basic and 50 acidic units of some kind, which may in theory unite to form 2500 distinct compounds with as many sets of distinctly individual properties. Were the analyst compelled to recognize these com- pounds singly, in the solid condition, he obviously would have to be familiar with the properties of each individual among the whole number; and were he to attempt to identify the individuals that might be present in a mixture, the task would be beyond accomplishment. Were the compounds not dissoci- able in solution, his problem would still be scarcely less difficult of solution ; but, being dissociable, his task is made comparatively light. Since the proper- ties of the solution of an ionized compound are merely the sum of the properties of its ions, and since the total number of ions with which we have assumed it necessary to deal is 100, it follows that the knowledge of 100 sets of properties is sufficient for the identifica- tion of any of the 2500 compounds. If, as it some- times happens, the substance under examination is not soluble or readily dissociated, the analyst has only to convert it by appropriate means into a body which 18 soluble and dissociable, and then to determine its nature from the character of the latter substance. Laws of Electrolytic Dissociation. — So far we have con- sidered only the qualitative effects of electrolytic dis- sociation; let us now examine briefly the quantitative effects, which are of no less importance to the analytical chemist. 18 CHEMICAL ANALYSIS As has been said already, different electrolytes have been found to show great dissimilarity in conductivity and ionization, even when dissolved in equimolecular proporjtions. But it also has been found that all are obedient to the same law with regard to the degrees of their dissociation, and that the dissimilarities are accounted for by constants which depend upon the nature of the electrolytes. The observed relations between the amounts of dissociated and undissociated electrolyte in a solution are expressed most simply for binary compounds in the equation a.b = k.c, wherein a represents the concentration of the positive ions, b that of the negative ions, e that of the mole- cules of undissociated material, and k a constant func- tion of the electrolyte. By assuming a value, such as unity, for the total amount of electrolyte in solution, and by representing the amount of dissociated material by «;, and the volume of the solution by v, we may expand this equation to a somewhat more instructive form : — c, concentration of undissociated electrolyte = ; V a and 5, concentrations of the two ions, — either ion = -• V By substitution we obtain the equation in the form = kv. 1-a Inspection of these equations, which are merely the formal expression of observed fact, reveals : — THEORY OF ANALYTICAL OPERATIONS 19 (1) that increase in a (or b) will be accompanied by a.h an increase in the ratio — ? i.e.. the free ions will increase and the molecules will decrease ; (2) that decrease in a (or h) will have the opposite effect, i.e., the free ions will decrease and the mole- cules will increase; (3) that the degree of dissociation may vary in either direction according as k is increased or decreased by variation in the nature of the electrolyte ; (4) that dilution of a solution, and corresponding increase of v, will call for an increase in the propor- tion of dissociated solute, the degree of dissociation approaching totality as its limit, as the dilution is indefinitely increased; (5) that concentration will have the opposite effect, and that the ratio of dissociated to undissociated solute will reach its minimum limit in a saturated solution. Further inspection of the equation a.b = k.c will reveal another fact which is of great practical significance for the analytical chemist. It is evident that in the solution of any given electrolyte, at a fixed tempera- ture, the only possible variants will be a, 5, and c. Let us suppose that it is possible in some way to intro- duce an added quantity of one ion, so that either con- centration a ov h will be increased. This being done, the increase in the product a.h will demand an increase in the value c. But the only way in which c may be increased is through the return from dissociation of a certain proportion of the ions. Assuming the concen- tration h to have been increased, the concentration a 20 CHEMICAL ANALYSIS must be diminished until, by the decrease in a.h and the corresponding increase in c, the original condition of equilibrium has been restored. In case that we are dealing with a saturated solution of the electrolyte, any increase in c will result in supersaturation of the solution ; and we shall see that a portion of the solute may separate in solid form. In fact, we have already seen this in a practical way in Exp. 1, c. In the saturated sodium chloride solution of that experiment, a considerable portion of the solute was present in the form of Na and CI ions; and the re- mainder was present in the molecular condition in quantity sufficient to produce saturation. The addi- tion of concentrated HCl, whose solution is very strongly dissociated, introduced a very large excess of CI ions in the salt solution; and, in consequence, the reunion of sodium and chlorine ions to the molec- ular state was set up and continued until equilibrium had been restored. But as the solution had already been saturated with the molecules of salt, these re- formed molecules were forced to separate in the solid form. If we dissolve together two substances which are dis- sociated more equally, such as KCl and NaCl, we find that less action of this sort takes place ; but when, of our two solutes with a common ion, one is more strongly dissociated than the other, the weaker is forced back to the molecular and inactive condition. The constant k has a very uniform value for neutral salts, but varies considerably for acids and bases, being high for strong acids and low for weak ones. THEORY OF ANALYTICAL OPERATIONS 21 Regarding the dissociation values of k, Ostwald has separated acids, bases, and salts into three classes : Class 1 : Neutral salts, strong acids, and strong bases. The strong acids mentioned are hydrochloric, hydro- bromic, hydriodic, nitric, chloric, and sulphuric ; the strong bases are hydroxides of the alkali and alkali- earth metals. Class 2 : Moderately strong acids and bases. The acids are phosphoric, sulphurous, and acetic ; the bases are the hydroxides of ammonium, silver, and magnesium. Class 3 : Weak acids and bases. The acids are car- bpnic, hydrosulphuric, hydrocyanic, silicic, and boracic ; the weak bases are the hydroxides of the trivalent metals and of those divalent metals not mentioned in Classes 1 and 2. Applications. — This discussion of the theories and laws of electrolytic dissociation enables us to explain many important operations and reactions in analytical chemistry, which otherwise could hardly be understood. A few of the explanations may be conveniently formu- lated by questions and answers : 1. How does ionization aid chemical activity? By dissociation of the solute into its ions, making it possible for them to combine with other ions. 2. How may heat aid chemical activity ? ^ By producing rapid vibrations of the molecules, which thus increases the speed of the reaction. 3. How may dilution aid chemical activity? By expanding the volume, thus decreasing the pree- SLiro^ and increasing the degree of dissociation. 22 CHEMICAL ANALYSIS 4. Why is the activity of an add or a base usually decreased hy adding some salt of that acid or base ? Two examples are given : (1) ^The addition of sodium acetate to acetic acid decreases the solvent power of the acid, since the salt is more strongly dissociated than the acid, and causes a portion of the latter to reassume the molecular condition by increasing the concentra- tion of the CgHgOg ions. (2) The addition of ammonium chloride to ammonia^ water decreases the solvent action of the latter by increasing the con- centration of the NH4 ions, and decreasing the dissociation and activity of the NH4OH. 5. When an excess of a normal salt of a weak acid is added to a solution of a strong acid, why is the activity of the strong acid destroyed, and that of the resulting weak acid greatly weakened? If an excess of sodium acetate is added to a solution of calcium phosphate in very dilute hydrochloric acid, the phosphate will be precipitated in spite of the fact that it is soluble in both hydrochloric and acetic acids. The explanation of this behavior is as follows : ^ Hydro- chloric acid and sodium acetate react to form sodium chloride and acetic acid. The latter, in the presence of the excess of sodium acetate, is forced back into the inactive molecular condition in which it is no longer able to hold the phosphate in solution. 6. Why does the addition of a solvent having an ion in common with that of a solute salt tend to precipitate the solute ? This question already has been answered in the THEORY OF ANALYTICAL OPERATIONS 23 explanation of the precipitation of common salt from its solution by the addition of hydrochloric acid. 7. Wh^ do reagents behave differently towards the same elements in different compounds f For example, hydrogen sulphide precipitates black cupric sulphide from a solution of cupric sulphate, but not from a solution of potassium cuprous cyanide. Another example, silver nitrate precipitates white silver chloride from a solution of potassium chloride, but not from a solution of potassium chlorate. The general answer to the question is that the chemical activity of a compound depends on its dissociated ions, — not on the presence of certain elements. Hydrogen sulphide, HgS, reacts with cupric sulphate, CuSO^, because the latter is ionized into Cu and SO4. Hydrogen sul- phide does not react with potassium cuprous cyanide, K3Cu(CN)4, because the latter gives no free Cu ions, but the molecule is dissociated into the ions, 3K and Cu(CN),. In the second example silver nitrate, AgNOg, reacts with potassium chloride, KCl, because the latter is dissociated into K and CI; but silver nitrate does not react with potassium chlorate, KClOg, as the latter salt is dissociated into K and CIO3. 8. Wh^ do reagents behave alike with various salts of the same metal f When we say of any substance that it is a salt of a certain metal, — such as copper, — we imply that it dissociates in solution with the formation of ions of that metal. These always react alike, no matter 24 CHEMICAL ANALYSIS what the negative ions be with which they are in equilibrium. 9. Whi/ are normal salts usually better precipitants than their corresponding acids or bases f For example, calcium chloride readily reacts with ammonium carbonate, but not with carbonic acid. The following equations illustrate the comparative reactivi- ties of normal salts, acid salts, and acids : (NH4)2C03 + CaCl2 yields an immediate precipitate; H(NH4)C03 + CaCl2 yields a tardy precipitate; HgCOg + CaClg yields no precipitate. Normal salts are most completely dissociated, while weak acids are very slightly dissociated. Acid salts of weak acids partake of the nature of both normal salts and weak acids. As ammonium carbonate is a normal salt, it is moBe completely dissociated than either the acid salt, H(NH4)C03, or the acid HgCOg, — and hence it reacts with calcium chloride more readily. 10. Why does an excess of a strong basic precipitant redissolve many precipitates from salts of weak bases? For example, a weak solution of sodium hydroxide precipitates aluminum hydroxide from a strong solution of aluminum sulphate, but on adding an excess of the precipitant, the precipitate disappears. Two reactions occur here : (a) Aluminum hydroxide is formed : 6NaOH + Al2(S04)3 = 2Al(OH)3 + 3Na2S04; and on adding more sodium hydroxide the white precipitate dissolves, forming sodium aluminate, — (b) Al(OH)3 4-3NaOH = Na3A103 + 3H20. THEORY OF ANALYTICAL OPERATIONS 25 Interpreted in terms of the ionic theory, aluminum being a very weak basic metal, its hydroxide is easily influenced by a strong base. In aqueous solution A1(0H)3 is in equilibrium, being partly dissociated into the ions A1+ and 3 0H~, and, by loss of water, partly into the ions H"*" and A102~. When a strong base like NaOH is added, it neutralizes the acid HAlOg, forming NagAlOg and water. This destroys the equi- librium, and more H+ and A102~ are developed, only to be in turn neutralized by more NaOH. And so the process continues till all of the A1(0H)3 goes into solution as NaoAlOo. CHAPTER HI METHODS OF ANALYTICAL SEPARATION Object of Separation. — It is only in rare cases that the chemist is able to recognize and identify individual elements or compounds in the mixtures which contain them, without having first separated them from the other bodies there present. In some cases, — the mix- tures being purely mechanical, — a mechanical treatment is sufficient to accomplish the separation; in other cases, — as when the substances are present in solution, — it is necessary in addition to make use of chemical or physical processes, by which means the material under examination is converted into such form that the recognition of its elements is possible. We have therefore two classes of separations, — the members of the first class being of a mechanical nature, whereas those of the second are of either physical or chemical character. The principal separations of the first class are brought about by the operations of deeantation^ filtration^ and washing. MECHANICAL SEPARATIONS Decantation. — When we have a mixture of a solid with a liquid in which it is insoluble, or a mixture of two liquids which are mutually insoluble, we may 26 METHODS OF ANALYTICAL SEPARATION 27 separate them by this process, provided that their specific gravities are so different that one of the compounds of the mixture will settle and separate completely from the other. From a mixture of liquid with solid, — for example, water and silver chloride, — we may remove most of the liquid by careful pouring or by suction with a pipette. From a mixture of liquids, — such as water and ether, — we may remove either layer with the pipette, or we may draw off the lower layer by means of a separatory funnel. Though decantation never separates completely, it is convenient for the removal of the bulk of liquids from finely divided precipitates which pass through the filter paper, or from gelatinous precipitates which clog its pores. Separation can be hastened by centrifugal shak- ing of the mixture before decantation. Ezperiment 5 (a) Dissolve a few crystals of silver nitrate in 10 c.c. of water in a test-tube, and then add dilute hydrochloric acid, drop by drop, until, by shaking, the white silver chloride settles beneath a clear liquid. Decant the liquid by pouring it oif with a glass rod held against the edge of the test-tube. Add more water to the solid and decant again by immersing the tip of a pipette in the clear liquid and sucking it off with the mouth (never allow the liquid to rise to the mouth). Close the mouth-end of the pipette with the tongue, lift out the pipette, and when the tongue is removed the liquid will flow out. (b) Mix 5 c.c. each of ether and water in a test-tube by shak- ing vigorously. The lighter ether will rise to the top. Remove either the ether or water with a pipette. Filtration. — Filtration is the separation of a solid residue from a liquid filtrate by means of a porous 28 CHEMICAL ANALYSIS partition impervious to the residue. The partition most frequently used in analytical work is unsized paper, supported in a glass funnel. A circular paper is folded twice^ so as to form the quadrant of a circle, and is then fitted into a glass funnel and dampened, so as to adhere closely to the sides of the funnel. For rapid filtration it is convenient first to fold the paper once across the middle, and then to "plait" it on radial lines, so that it resembles finally a folding paper fan. On opening a paper so folded, it will be seen to fit loosely in the funnel and to leave numerous channels through which the filtrate may escape. To prevent spattering, the beveled tip of the funnel should rest against the inside of the receiving vessel. Three important factors determine the rate and de- gree of separation by filtration, namely: temperature, pressure, and the ratio between the size of the pores of the partition and the size of the particles of the residue. Increase of temperature decreases the adhesion be- tween the molecules of the filtrate and those of the residue, and increases the size of the colloidal granules and crystals of the residue. Pressure needful for filtration is usually obtained through gravity, but sometimes through gravity and suc- tion combined. The most effective method of diminish- ing the atmospheric pressure is the use of the suction pump. A platinum cone should be placed in the apex of the funnel to support the moistened paper, which should be so closely fitted to the sides of the funnel as to leave no air channels. The funnel tube is then to be connected with the filter flask of a suction pump. METHODS OF ANALYTICAL SEPARATION 29 The third factor for effective filtration consists in increasing the size of the particles of residue, or in diminishing the size of the pores of the partition. The particles of the residue are enlarged by heat and by contact with the liquid from which they are formed. Both colloidal and crystalline particles grow when im- mersed in a mother liquor containing smaller particles of the same kind. Hence it is the usual practice to digest the mixture for a short time before filtration. Care should be taken, however, to keep the residue covered with liquid during the digestion, as on exposure to the air many precipitates oxidize and often redissolve in another form. The amorphous variety of ferrous sulphide obtained by precipitation with ammonium sulphide oxidizes in contact with the air to soluble ferrous sulphate. Experiment 6 Arrange filtering apparatus as described and filter the follow- ing mixtures : silver chloride in water, barium sulphate in water (made by adding dilute sulphuric acid, drop by drop, to a dilute solution of barium chloride), and aluminum hydroxide in water (made by adding a very dilute solution of sodium hydroxide to a concentrated solution of alum till a heavy gelatinous mass appears). Try all three mixtures with folded filters, then with creased filters, and finally with the filter pump. After all these trials, if the filtration of either mixture is very slow or the filtrate remains muddy, try decantation first, and then filtering the moist residue with the pump. Washing. — Neither decantation nor filtration will thoroughly cleanse the residue from the filtrate. It is necessary in most cases to wash off the adhering filtrate 30 CHEMICAL ANALYSIS ^ by pouring on water, or water made acid, alkaline, or salt, according to the nature of the residue. When a thorough separation is demanded, the washing should continue till the washings no longer show a trace of the solutes of the filtrate. In many separations, where the residues are liable to oxidation on exposure to the air, it is necessary to hasten the washing with the pump. Two rather serious difficulties are frequently en- countered both in washing and in filtration ; namely, the clogging of the paper with certain finely divided residues, and the tendency of such residues to pass through the paper. As both difficulties are due to the same cause, the same treatment will correct both. The troublesome residues are colloidal in nature and, as has been stated, their particles unite or coagulate on heating. Hence it is well to subject the mixture to quiet heat — not boiling — long enough to allow the precipitate to settle to the bottom; decant the supernatant liquid into a filter and apply the pump; add warm water to the residue; and, after settling, decant and filter again in like manner. Repeat .the washing by decantation once or twice, and then add the solid to the paper, finally washing directly on the paper. As colloidal substances are somewhat soluble in water, but less soluble in many neutral salt solutions, the latter are frequently used for washing. It is also necessary in special cases to use acid or akaline solu- tions for washing. As regards the effectiveness of washing, Ostwald gives this calculation : " Should the washing liquid amount to nine times as much as the original moisten- METHODS OF ANALYTICAL SEPARATION 31 ing solution, and should 1 gram of foreign substance be mixed with the precipitate to begin with, then after four washings only (J^)* gram, i.e. 0.0001 gram, of the impurity would remain." Experiment 7 (a) Wash the residue of barium sulphate (last experiment) with water till the washings show no white precipitate with drops of barium chloride solution. (b) Precipitate some ferric hydroxide by adding ammonium hydroxide to a boiling solution of ferric chloride till the latter is permanently alkaline. Filter and wash according to directions above. PHYSICAL AND CHEMICAL SEPARATIONS The second class of separations, to which reference was made on p. 26, includes solution, precipitation, evaporation, and ignition. Solution. — Most methods of separation are more or less dependent upon solution as their starting point, and in some cases separation is completed by this oper- ation. Solids composed of two or more substances, like minerals and alloys, may be separated in this manner when only a part of their constituents is soluble. Experiment 8 (a) Finely powder a small piece of dolomite in an agate mortar, and dissolve by warming with dilute hydrochloric acid. The small residue insoluble in the acid is silica, whose separation may be completed by filtration. (b) Dissolve some filings of soft solder in a test-tube, in a mixture of equal parts of nitric acid and water, and when the 32 CHEMICAL ANALYSIS action ceases, add some water. When the white powder has settled, pour oft" the clear liquid into a small dish and evaporate to dryness. Try to dissolve the residue in the test-tube by boiling with' water. Soft solder is an alloy of tin and lead, and by solu- tion of the lead in nitric acid the two metals are separated. The ordinary solvents for solids are water, hydro- chloric acid, nitric acid, and aqua regia. A small portion of the powdered solid is treated in a test-tube with cold water. If the solid does not disappear after shaking the contents several times at intervals, transfer a drop of the solution to a watch glass with a glass rod. Place also a drop of distilled water on the watch glass near that of the solution., Heat the watch glass on an asbestos board or sand bath to dryness. Compare the residues left from the two drops. If both are alike in size, cold water does not dissolve the solid; but if the residue from the supposed solution is larger, the solid is at least partly soluble in water. In either case boil the contents of the test-tube, and if the solid still does not disappear, again evaporate a drop of the solution to dryness on the watch glass. A large residue indicates a partial solubility of the solid in hot water. Treat another portion of the powdered solid in a similar manner in succession with dilute and concen- trated hydrochloric acid, dilute and concentrated nitric acid, and aqua regia. If the substance dissolves com- pletely in any one of the solvents, the solvents following need not be used. Sometimes a solid is composed of different substances which have no solvent in common. One substance may METHODS OF ANALYTICAL SEPARATION 33 dissolve only in water, another only in hydrochloric acid, another only in nitric acid, etc. In such extreme cases it is necessary to separate the mixed solids by means of solvents. Some substances cannot be dis- solved by any of the reagents mentioned. It is neces- sary in such cases to fuse them with an alkaline carbonate in a platinum crucible or foil, and afterwards to digest the fused mass with water. (See directions for fusion, p. 43.) Experiment 9 (a) Using the methods described above, dissolve .5 gram of each of the following substances : cupric sulphate, barium sul- phate, and sand. (b) Mix .5 gram each of the same substances and separate them by solution. Precipitation. — The terms soluble and insoluble, as used in practice, have only relative values ; and they merely indicate considerable differences in degree of solubility, for, in fact, all substances are soluble. We have seen in Exp. 1 that a substance which is in complete solu- tion at one temperature may be rendered less soluble and thrown out of solution by change to another tem- perature. And we also have seen that a material whose solubility appears to be about the same at all tempera- tures may be thrown out of solution by modifying the solvent in such a manner as to decrease its solubility therein. When by either of these means we have forced a solution to give up a part of its solute in solid form, we have performed the operation of precipitation ; and the solid thrown down is called a precipitate. In 34 CHEMICAL ANALYSIS Acetate. Arsenate. Arsenite. Borate. Bromide. Carbonate. Chlorate. Chloride. Chromate. Cyanide. Ferrocyanide. Ferricyanide. Fluoride. Hydroxide. Iodide. Nitrate. Oxalate. Oxide. Phosphate. Silicate. Sulphate. Sulphide. Sulphite. Tartrate. Ag rH(N(M(Neo(NT-ico(NeocoeciH c<3 tH (N IM IM Tjt (M < eoeoTj(N^(NiH tH (N (M 2 1 - - - - - = J w w H ^ c -Si -8 - — - - \ -1 - - ■| (^ - — - - ^ ^5 - - ^ - -s - 5 - c -S - - '^ 1 n. - "- c= - - - c -a - - - *= - e= -1 ^ - *= -1 ^ - - - -8 - - - -a - ^ ' « — — — -J5 -9 _ - I 1 cu Ki 't «o u> N 00 FLAME COLORATION AND SPECTROSCOPY 69 1 s 1 ^ 1 — N - jO - 1^1 2 — 4 - -8 1 I 9 - 60 CHEMICAL ANALYSIS EXPLANATION OF TABLE VII Nos. 1, 2, 3, 4, 5, and 6 represent the discontinuous spectra of ; salts of metals volatile in the flame. The lines and curves in the ! field of ieach spectrum indicate the position and distinctness of ' visible lines. For example, the spectrum of potassium appears on the scale as a strong line between 10 and 20, — more accurately : I7j^ — and another thinner and shorter line between 150 and 160, — more accurately 154. The long curve from 20 to 130 shows that there are many indistinct lines within that area, and the varying heights of the curve indicate the relative distinctness of the lines. Nos. 7, 8, and 9 represent the discontinuous spectra of salts of metals volatilized by the electric spark. Spark spectra are char- acterized by the small number of narrow lines and the absence of indistinct lines. No. 10 represents the absorption spectrum of sunlight, showing the so-called Fraunhofer's lines. Nos. 11, 12, 13, 14, 15, and 16 represent absorption spectra. The shaded parts show the portion of the spectrum absorbed, and the curved margins the relative degrees of absorption. FLAME COLORATION AND SPECTROSCOPY 61 light the test-tube filled with a dilute solution of the substance. Certain portions of the continuous spectrum will now appear dark. Mapping Spectra. — Two methods have been adopted for recording spectra : — 1. Kirchoff and Bunsen's scale,^by which the posi- tions of the lines are recorded on a graduated scale. The conventional practice is to adjust the scale so that the yellow sodium line shall coincide with 50 on the scale. This method is quite simple, and though not so accurate as the other method, it is generally used for chemical analysis. 2. The wave-length method, by which the wave lengths of the colors are calculated from the formula V X = -, in which X, v, and n, respectively, are wave length, velocity of light, and number of vibrations. The unit is one ten-millionth of a millimeter, called an Angstrom. Professor Rowland of Johns Hopkins University, by means of his improved concave grating spectroscope, has compiled an atlas ^ of a large number of spectra recorded in wave lengths. In this elementary book measurements of wave lengths would not be consistent with the character of the work. Hence the use of Kirchoff and Bunsen's scale is recommended. The table on pages 58 and 59 includes some illustra- tions of a method of mapping spectra. Experiment 18 (a) Examine and map the flame spectra of the following salts: sodium chloride, potassium chloride, lithium chloride, barium chloride, strontium chloride, and calcium chloride. 62 CHEMICAL ANALYSIS (6) Examine and map the spark spectra of the following salts: magnesium chloride, zinc chloride, manganese chloride, copper chloride, and bismuth chloride. (c) Examine and map the absorption spectra of the follow- ing inorganic salts : ferric chloride in water, potassium perman- ganate in water, chrome alum in water, and cobalt nitrate in alcohol. ((/) Examine and map the absorption spectra of alcoholic solu- tions of blood and fuchsine, and a water solution of logwood. Special Method for Aluminum (Vogel). — Make a solution of extract of logwood by boiling the chips in water. Place a test-tube containing this extract between the spectroscope and a luminous flame. The right end of the spectrum will be absorbed, the extent of absorption depending on the concentration of the logwood. The boundary between the absorbed and unabsorbed parts of the spectrum is made to coincide with a convenient line on the scale. Now add a few drops of a dilute neutral solution of an aluminum salt. This will cause the boundary line to move to the left in proportion to the concentration of the solution. The aluminum salt solution is made neutral by adding to it, drop by drop, a very dilute solution of ammonia, until a slight but permanent precipitate is produced. Neutral ferric salts give the same reaction, but iron can be tested for in the wet way. In case of a mixture of aluminum and iron salts, the iron can be removed by adding an excess of ammonium sulphocyanate solution and shaking out the ferric sulphocyanate with ether. The colorless aqueous portion is tested for aluminum salts. (See Nos. 13 and 14 on the table.) FLAME COLORATION AND SPECTROSCOPY 63 Special Method for Magnesium. — Make a dilute solution of alcana and record its absorption spectrum. Now add a dilute neutral solution of magnesium chloride. The alcana spectrum will be moved to the left in propor- tion to the concentration of the magnesium-chloride solution. Special Method for Manganese. — Boil the compound with some lead dioxide and a little nitric acid and test for the absorption spectrum of permanganic acid. (See No. 11 on the table.) Special Method for Cobalt (Wolff). — Add ammonium sul- phocyanate to the cobalt-chloride solution and shake with alcohol (amyl preferable) and ether. This dis- solves the cobaltous sulphocyanate, and the solution gives a characteristic absorption spectrum. Special Method for Iodine (Vogel). — Iodine can be tested for with the apparatus here shown : a is the gas flame colored by the iodide ; 6 is a hard glass tube held in position over the smaller tube by a spiral cop- per wire, c; c? is a hard glass test-tube containing at its both tom a mixture, e, of copper oxide and an iodide ; / is a stream of illuminating gas passing through the apparatus and burning at a ; g is the spectroscope ; A is a gas burner. Bromine and chlorine can also be detected in the same manner by using a bromide or a chloride instead of an iodide. The copper iodide, or chloride, or bromide 64 CHEMICAL ANALYSIS escapes with the gas and colors the flame green. The spectroscope shows a number of bands, especially in the green part of the spectrum, which are different for iodine, chlorine, and bromine. CHAPTER V LIST AND PREPARATION OF REAGENTS It is desirable that the student know not only the chemical nature of reagents, but also their proper dilu- tion. From a careful study of the principles of solu- tion and of mass action, the reason for a knowledge of dilution must be obvious. In Exp. 4 an illustration is given of the different effects of concentrated and dilute sulphuric acid on zinc. Furthermore, it is desirable that reagents be so diluted as to give uniform strengths, so that the volume of solu- tion used will be an index of the quantity of reagent employed. Most analysts use an arbitrary system of dilutions that has no special significance, save that it meets the empirical requirements of ordinary analysis. Reddrop (1890) suggested that dilutions be based on the equivalent weights of the reagents. The equiva- lent weight of a substance is its molecular weight divided by the number of its replaceable hydrogen atoms, or those which are the equivalents of hydrogen. For example, the equivalent weight of hydrochloric acid is 36.5, obtained by dividing its molecular weight by its number of replaceable hydrogen atoms : 36.5^1 = 36.5. Likewise, the equivalent weights of sodium chloride, sodium hydroxide, and silver nitrate are, respectively, 65 66 CHEMICAL ANALYSIS 58.5, 40, and 170. The equivalent weight of sul- phuric acid is 49, obtained by dividing its molecular weight by its number of replaceable hydrogen atoms : 98 -^ 2 = 49. The following list gives some equivalent weights : Hydrogen, H 1. Oxygen, O 8. Hydrochloric Acid, II CI 36.5. Nitric Acid, HNO3 63. Sulphuric Acid, n2S04 49. Acetic Acid, HC2II3O2 60. Phosphoric Acid, H3PO4 ....... 32.6. Ammonium Chloride, NH4CI 53.7. Barium Chloride, BaClz 103.8. When equivalent weights are dissolved in equal volumes of water, equal fractional parts of the solu- tions will be equivalent. A normal solution of a sub- stance is one which contains the equivalent weight of that substance, in grams, dissolved in a liter of solution. For example, a normal solution of sodium chloride is its equivalent weight, 58.5 grams, dissolved in sufficient water to produce a liter of solution. Equal volumes of normal solutions whose solutes react with one another should neutralize each other per- fectly, leaving no excess of either reagent, as in the fol- lowing equation : AgNOg + NaCl = AgCl + NaNOg. Experiment 19 Arrahge two burettes. Fill the one with a normal solution of sulphuric acid, and the other with a normal solution of sodium hydroxide. Into a beaker or flask measure out 20 c.c. of the LIST AND PREPARATION OF REAGENTS 67 alkali and add about 1 c.c. litmus solution. Now add the acid from the burette, drop by drop, stirring or shaking constantly, till the blue color changes to red. Compare the volumes of the two solutions used. Note. — The normal solutions in this experiment should be prepared by the instructor. Theoretically, solutions of all reagents should be of equivalent strengths ; but it has not been found practi- cal to give all of them these values because of their lack of uniform solubilities. It is convenient to adopt the normal solution as a standard and to express all dilutions as multiples or fractions of normal. The letter N is used to denote a normal solution, and all variations from normal are expressed in terms of N. For example, the best dilution for sulphuric acid is about five times its normal strength, expressed thus: 5 N H2SO4, or 5 N solution ; and the best dilution for silver nitrate is about one-fifth its normal strength, N N expressed thus : ^ AgNOg, or — solution. It is sufficient for the purposes of qualitative analysis if the strength of the reagents be approximately known, and it will be understood that the strengths given below are only approximate. LIST OF REAGENTS Solutions 1. Acetic Acidj HC2H3O2 + Aq. 1 volume 80% acid to 2 volumes water = 5 N solution. 2. Concentrated Hydrochloric Acid, HCl. Sp. gr. 1.20 = UN solution. 3. Dilute Hydrochloric Acid, HCl + Aq. 1 volume con- centrated acid to 3 volumes water = 5 N solution. 68 CHEMICAL ANALYSIS 4. HydrosulphuriG Acid Gas, HgS. For generating the gas Kipp's apparatus is generally used.^ The generator should be kept in a hood with a good draught. The mate- rials used in the generator are lumps of ferrous sulphide, FeS, and dilute hydrochloric or sulphuric acid. Often when the acid seems exhausted it will renew its activity if it is removed from the generator, and the lumps of ferrous sulphide thoroughly washed with water. 5. Hydrosulphuric Acid Solution, HgS + Aq. The gas is passed into cold water to saturation. The solution of the gas should be kept in the dark or in bottles of deeply colored glass, as sunlight decomposes the acid with sepa- ration of sulphur. 6. Concentrated Nitric Acid, HNO3. Sp. gr. 1.42 = 16 N solution. 7. Dilute Nitric Acid, HNO3 -f- Aq. 5 volumes concen- trated acid to 11 volumes water = 5 N solution. 8. Concentrated Sulphuric Acid, H2SO4. Sp. gr. 1.84 = 36 N solution. 9. Dilute Sulphuric Acid, H2S04' 2 H2O + Aq. 1 volume concentrated acid to 6 volumes water = 5 N solution. In diluting the concentrated acid, it should be added to the water very slowly, in a large porcelain dish, and allowed to cool before using. 10. Tartaric Acid, H2C4H4OC + Aq. 1 part crystals to 13 parts water = N solution. The acid decomposes in solution and should be prepared fresh each time. 11. Aqua Regia, HCl -h HNO3. 1 volume concentrated nitric to 3 volumes concentrated hydrochloric acid. This proportion is sometimes varied for specific purposes. As a solvent, just enough of the reagent should be used to dis- solve the substance completely. A large excess must be avoided, since if allowed to remain it decomposes other reagents, while if evaporated certain volatile chlorides, LIST AND PREPARATION OF REAGENTS 69 e.f/., arsenic and mercuric chlorides, are liable to be lost. Prepare aqua regia fresh each time it is needed. 12. Ammonium Chloride, 1^11401 -\-Aq. Use 1 part crys- tals to 4 parts water, and allow it to rise to the natural tem- perature ; then dilute with 1 part water = 5 N solution. 13. Ammonium Carbonate, (NH4)2C03 + Aq + NH4OH. 4 parts solid ammonium carbonate dissolved in 7 parts 5 N NH4OH ; then dilute with 14 parts water = 5 N solution. 14. Ammonium ^^Sesqui^^ Carbonate. Dissolve 1 part solid (NH4)2C03 in 9 parts water and add for each 10 c.c. of the liquid 5 drops of strong ammonia (Hager). Pre- pare fresh each time. 15. Concentrated Ammonium Hydroxide, NH4OH + Aq. Sp. gr. 0.90 = 9N solution. 16. Dilute Ammonium Hydroxide, NH4OH -f Aq. 2 vol- umes concentrated ammonia water (sp. gr. 0.90) to 5 volumes water = 5 N solution. Both concentrated and dilute ammonia attack glass vessels ; and if white flakes appear in the clear solutions, they should be filtered out before use. 17. Ammonium Oxalate, (NH4)2C204- HgO + Aq. 1 part solid crystals to 25 parts water = — solution. 18. Ammonium Sulphide, (^114)28 + Aq -f- NH4OH. Saturate 3 parts 5N ammonia with hydrogen sulphide gas, and then add 2 parts 5N ammonia solution. This reagent should be made frequently, as it decomposes on standing. 19. Yellow Ammonium Sulphide, (NH4)2Sx -{- Aq -\- NH4 OH. Digest a solution of (NH4)2S with a little powdered roll sulphur. An excess of sulphur must be avoided, as it produces the red solution containing higher sulphides. 20. Barium CA^oric^e, BaCV 2 H2O + Aq. 1 part solid crystals dissolved in 10. parts water = N solution. 70 CHEMICAL ANALYSIS 21. Bromine, Br. Should be kept in a dark, glass- stoppered bottle. 22. Dilute Bromine, Br + Aq. Make a saturated so- N lution by shaking an excess of bromine in water = — solution. Stronger solutions of bromine can be made by adding potassium bromide to the water solution. 23. Calcium Hydroxide (lime water), Ca(0H)2 + Aq. Saturate freshly boiled distilled water by shaking an excess of freshly slaked lime in it, and allowing ithe N excess of lime to settle. Filtrate = — solution. 24. Chlorine Water, CI + Aq. Saturate cold water N with chlorine gas = — solution. Should be kept in the o dark and in brown bottle. 25. Cobalt Nitrate, Co(N03)2-6H20 + Aq. 1 part solid crystals to 7 parts water = N solution. 26. Ferric Chloride, FeCla 4- Aq. 1 part solid salt to 20 parts water = N solution. 27. Ferrous Sulphate, FeS04-7H20 + Aq. 1 part solid crystals to 7 parts water = N solution. It is best to pre- pare this solution fresh, when needed. 28. Lead Acetate, Pb(C2H302)2' 3 HgO + Aq. 4 parts solid to 21 parts water = N solution. 29. Magnesium Sulphate, MgS04-7H20 + Aq. 1 part crystals to 8 parts water = N solution. 30. Magnesia Mixture, MgClg + NH4CI + ]SrH:40H + Aq. Dissolve 6 grams magnesium chloride crystals and 165 grams ammonium chloride in 300 c.c. water; then add 300 c.c. 5 N ammonia solution ; and dilute to 1 liter = — ■ solution. 31. Mercuric Chloride, HgCl2 + Aq. 1 part solid salt to 37 parts water = — solution. LIST AND PREPARATION OF REAGENTS 71 32. Hi/drochlorplatinie Acid, HoPtClc'G HgO + Aq. 1 part solid salt to 12 parts water = N solution. The solution can also be prepared by dissolving 0.30 gr. platinum foil in aqua regia, evaporating to dryness, and redissolving in 10 c.c. 5 N HCl. 33. Potassium Chromate, KaCrOi -f Aq. 1 part solid salt to 10 parts water = N solution, 34. Potassium Cijanide, KCN + Aq. 1 part solid salt to 15 parts water = N solution. Prepare fresh for each experiment. 35. Potassium Ferricyanide, K3Fe(CN)6-3H20 + Aq. 1 part solid salts to 9 parts water = N solution. 36. Potassium Ferrocyanide, K4Fe(CN)c*3H20 + Aq. 1 part solid salt to 10 parts water = N solution. 37. Potassium Hydroxide, KOH -f Aq. 1 part solid caustic potash to 3.5 parts water = 5 N solution. 38. Potassium Sulphocyanate, KCNS + Aq. 1 part solid salt to 10 parts water = N solution. 39. Silver Nitrate, AgNOa + Aq. 1 part solid salt to N 30 parts water = — solution. 40. Sodium Acetate, ]SraC2H302-3H20 + Aq. 1 part solid salt to 8 parts water = N solution. 41. Sodium Carbonate, Na2C03' 1 OH2O -|- Aq. 1 part solid crystals to 7 parts water = N solution. 42. Sodium Hydroxide, NaOH -f- Aq. 1 part solid to 5 parts water = 5 N solution. First dissolve the solid base in a little water, allow to cool, and then dilute to the required volume. 43. Sodium Phosphate, HNa2P04'12H20 + Aq. 1 part solid crystals to 8 parts water = N solution. 44. Stannous Chloride, SnCl2*2H20 + Aq. Dissolve 3 parts solid salt in 3 parts 5 N hydrochloric-acid solution, and dilute with 20 parts water = N solution. Pieces of 72 CHEMICAL ANALYSIS granulated tin should be kept in the solution. An excel- lent quality of solid stannous chloride can be made by heating granulated tin with repeated small quantities of concentrated HCl, added at intervals whenever ebullition ceases. When all the tin is dissolved, evaporate to dry- ness on the water bath. Solvents. — 45. Alcohol^ CaHgOH, 95 per cent. 46. Carbon Disulphide, CSg. 47. Ether, (€2115)20, commercial. 48. Ether-Alcohol, 1 volume absolute ether to 1 volume absolute alcohol. 49. Petroleum Ether. 50. Water, distilled. Dry Reagents. — 51. Ammonmm Chloride, NH4CI. 52. Ammonium Carbonate, (NH4)2C03. 53. Cobalt Nitrate, 00(^03)2* 6 H2O. 54. Lead Peroxide, PbOg. 55. Manganese Peroxide, MnOg. 56. Microcosmic Salt, HlSra(NH4)P04-8H20. 57. Potassium Carbonate, K2CO3, anhydrous. 58. Potassium Cyanide, KCN. 59. Potassium Disulphate, HKSO4. 60. Potassium Nitrate, KNO3. 61. Potassium Nitrite, KNOg. 62. Sodium Carbonate, NagCOg, anhydrous. 63. Sodium Tetraborate (borax), NagBiOy * 10 HgO. 64. Sodium Peroxide^ Na202. CHAPTER VI SYSTEMS OF ANALYTICAL EXAMINATION Analysis by the Dry Way. — The chief operations in analysis by this system are the observations of (a) , Oxidation and Reduction, and (h) Flame Coloration. (a) Oxidation and reduction have been explained under ignition operations. These include fusion in crucibles, closed tube reductions, oxidation and reduc- tion with fluxes on a platinum wire, and reduction on charcoal with and without fluxes. (h) Flame colorations have been explained under simple flame colorations and spectroscopy. Though these operations are indispensable to the analyst, they do not constitute an independent system, but are only used for preliminary and confirmatory observations. Analysis by the Wet Way. — Solution is the basis of this system of analysis. Advantage is taken of the following facts : — (a) The metallic ions of most compounds behave alike towards certain reagents, regardless of the acid radicals which may be present. For example, all solu- ble silver salts will give insoluble silver chloride with all soluble chlorides. (b) Differences of solubility of similar compounds of different metals may be utilized in separating them into 73 74 CHEMICAL ANALYSIS groups. For example, silver, lead, and copper sulphides are insoluble in acidified solutions, while some other sulphides, e.g., those of zinc and barium, are soluble under like conditions. Silver and lead chlorides are insoluble, and copper chloride is soluble in acidified solutions. Such differences of solubility afford an easy means of separating and detecting these metals. (c) Physical characteristics, color, odor, etc., are used for detecting individual substances. For example, the soluble salts of both cadmium and copper are precipi- tated by hydrogen sulphide; but as the one sulphide is a bright yellow and the other black, the two can be distinguished. (d) The principles of the periodic law of elements are in part regarded in analytical chemistry. The periodic groups — sodium, potassium, lithium ; barium, strontium, calcium ; chlorine, bromine, and iodine — are also utilized as analytical groups. An ideal natural system of classification would have analytical groups to coincide with periodic groups ; but in this respect analytical classification is somewhat arti- ficial, and depends more upon differences in degree of the physical property of solubility than on chemical properties. For example, magnesium, zinc, and cad- mium belong to the same periodic group, but, by reason of the differences in the solubilities of their sulphides, the three metals are placed in three separate analytical groups. By (a) all salts require two analyses : first, for metals ; and, second, for the acid radicals. By (b) and (c) the metals are divided into groups depending on SYSTEMS OF ANALYTICAL EXAMINATION 75 the insolubility of their chlorides, sulphides, hydroxides, and carbonates. There are six groups of metals : — Group I, whose chlorides are insoluble in aqueous solution ; Group II, whose sulphides are insoluble in acidified (HCl) solution; Group III, whose hydroxides are insoluble in alkaline (NH4OH) solution ; Group IV, whose sulphides are insoluble in alkaline (NH4OH) solution ; Group V, whose carbonates are insoluble in alkaline (NH4OH) solution ; Group VI, which has no group characteristic. The group reagent would be for : — Group I, a soluble chloride (HCl) ; Group II, a soluble acidified sulphide (H2S with HCl) ; Group III, a soluble alkali (NH^OH with NH^Cl); Group IV, a soluble alkaline sulphide [(NH4)2S with NH4OH and NH4CI] ; Group V, a soluble alkaline carbonate [(NH4)2C03 Avith NH4OH and NH4CI] ; Group VI, no group reagent. By (b) and (d) the acids are also divided into groups depending on the insolubility of their barium and silver salts, hence the three groups : — Group I, whose barium salts are insoluble in aqueous solution ; Group II, whose silver salts are insoluble in dilute acid solution ; Group III, which has no group reagent. 76 CHEMICAL ANALYSIS By (b) and ( tube. The black residue may consist of metallic silver and SbAgg. Test for SbAgg by dissolving in tartaric acid, etc. The filtrate may contain the hypothetical acid HgAsOg with excess of AgNOg, which can be tested by producing yellow AggAsOg with dilute NH4OH. Filter the contents of the generating flask, remove the undissolved zinc, and test the residue for tin by dissolving , in a small amount of HCl and then adding HgCl2. The reactions involved in the separation of arsenic, antimony, and tin by Hofmann's method are for : — (a) arsenic, — AsHg + 3 HgO + 6 AgNOg = 3 Agg + G HNOg + H3ASO3; (h) antimony, — SbHg + 3 AgNOg = SbAgg + 3 HNO3; (c) tin, — SnCl^ + 4 H = Sn + 4HC1. Separation of Sub-group B. — Transfer the residue (6), supposed to consist of the members of the sub-group, to an evaporating dish, and boil with HNOg^ (diluted 1 : 2) till the chemical action ceases. Filter. The METALS OF GROUP II . 99 residue {a')'^ may be black HgS or a mixture of HgS and white Hg(N03)2* 2 HgS. Dissolve in aqua regia, boil off the chlorine and excess of acids, and test with SnCl2. The filtrate (a) may contain Pb(N03)2, Bi(N03)3, Cu{N03)2, and Cd(N03)2. Concentrate this filtrate until most of the HNO3 is driven off, add dilute H2S04,2 warm gently, and allow to stand for some time. A white precipitate is PbSO^.^ If lead is present, add an excess of dilute H2SO4 and evaporate till all the HNO3 is expelled. Dilute with water, place aside to enable the precipitate to settle, and filter off the insoluble residue (b'). Test it by boiling with NH^C2H302 and adding K2Cr04. A yellow precipi- tate confirms PbCrO^. If lead is present, use the filtrate from PbSO^; if lead is absent, boil off excess of HNO3 from filtrate (a'). Add NH4OH till alkaline. A blue fluid confirms the presence of copper and a white flocculent precipitate, bismuth. Filter.* Dissolve the residue (c') in a few drops of HCl and test for BiOCP by adding an excess of water. If the filtrate (c') is blue, add a dilute solution of KCN care- fully till the color disappears. Often copper is present in small quantities, and the blue^ color is not distinct. In this event evaporate a small quantity of the filtrate almost to dryness, acidify with very dilute HCl, and add K4Fe(CN)g. A brown precipitate or coloration indicates Cu2Fe(CN)g. If copper is present, add very little dilute KCN solution. If copper is absent, adding KCN is not necessary. Pass HgS. A yellow*^ precipi- tate confirms CdS. CHAPTER IX METALS OF GROUP III : ALUMINUM, CHROMIUM, AND IRON Characteristic : Insolubility of the hydroxides in alkaline (NH4OH) solution in the presence of ammonium chloride. Group Reagent : Ammonium hydroxide with ammonium chloride. REACTIONS Aluminum (salt for study, aluminum sulphate, Al2(S04)3). 1. NH4OH precipitates gelatinous aluminum hydrox- ide, A1(0H)3, soluble somewhat in excess of reagent in the cold, but wholly insoluble if NH4CI is present or if the solution is boiled. 2. NaOH acts like NH4OH, except that A1(0H)3 is completely dissolved in excess of reagent, forming sodium aluminate, NagAlOg. This in turn is recon- verted into insoluble A1(0H)3 if the solution is boiled with NH4CI. 3. BaCOg, suspended in water, precipitates aluminum completely in the cold as A1(0H)3 mixed with a basic salt, probably A1(0H)C03. 4. (NH4)2S precipitates Al(OH)g with evolution of HgS. 5. HNa2P04 precipitates white aluminum phosphate, AlPO^-HgO, soluble in alkalies in the absence of NH4CI and in HCl and HNO3 ; insoluble in HC2H3O2. 6. Spectrum (see Special Method for Aluminum, p. 62). 100 METALS OF GROUP III 101 INFLUENCE OF AMMONIUM SALTS ^ The reactions between many salts and alkalies in the presence of ammonium compounds demand the following further explanation. • Soluble alkalies react with salts of many metals to produce hydroxides of various solubilities. Some of the more insoluble of these behave as weak acids in the presence of strong bases and combine with them to form new classes of soluble salts. The following is a type of this class of reactions : — ZnCla + 2 NaOH = Zn(0H)2 + 2 NaCl ; Zn(0H)2 + 2 NaOH = Na2Zn02 (sodium zincate) + 2H2O. Ammonia behaves much like the other soluble bases, provided certain precautions are observed. The following reactions can occur : — ZnCl2 + 2 NH4OH = Zn(0H)2 + 2 NH^Cl ; Zn(0H)2 + 2NH40H = (NH4)2Zn02(ammonium zincate) + 2 H2O. The last reaction is interrupted in two ways : — First: NH3 splits off easily, thus allowing Zn(0H)2 to be reclaimed. Second: the by-product, NH^Cl, strongly influences the reaction and redissolves the precipitated Zn(H0)2. In order to obtain (Isril4)2Zn02 it is necessary to add NH^OII to a cold solution of ZnCl2, to avoid decomposition of (NH4)2Zn02 into NHg and Zn(0H)2, and to filter oif the solution of NH^Cl. When an excess of NH^Cl or other ammonium salt is added, all of these conditions of solubility are modified. This applies to the reactions with the hydroxides of the alkali metals as well as ammonia. In the case of the hydroxides of the bivalent metals, Fe(0H)2, Zn(0H)2, Mn(0H)2, Co(OH)2, Ni(0H)2, Ba(0H)2, Sr(0H)2, Ca(0H)2, and Mg(0H)2, the tendency of the alkalies to redissolve them is greatly augmented by NH^Cl. Hence these hydroxides are not precipitated in the presence of ammonium salts. In the case of the hydroxides of the trivalent metals, A1(0H)3, Cr(0II)3, and Fe(0H)3, the tendency of the alkalies to redissolve them is counteracted by NH^Cl. 102 CHEMICAL ANALYSIS Hence these hydroxides are completely precipitated in the presence of ammonium salts. Two theories are now sanctioned by good authorities for the interpretation of the influence of ammonium salts in the reactions just mentioned. 1. Double Salts Theory (a) Salts of Bivalent Metals. — The following pairs of reac- tions will explain this theory with reference to the salts of biva- lent metals : — ZnClg + 2NaOH (or NH^OH) = Zn(0H)2 + 2NaCl, Zn(0II)2 + 4NH4C1= ZnClVSNH^Cl (a soluble double salt) + 2NH4OII; MgClg + 2 NaOH = Mg(0II)2 + 2NaCl, . Mg(OH)2 + 3NH4CI = MgCl2- NH4CI + 2NH4OH. (b) Salts of Trivalent Metals. — The application of equations analogous to those of the preceding paragraph would lead us to expect the following reactions to occur with hydroxides of triva- lent metals : — t (NH4)3A103 + 3 HCl + NH.Cl, Al(OH)3 + 4NIl4Cl-:-] or ( AICI3 • NH4CI + 3 NH4OH. But the hypothetical, soluble bodies, (NH4)3A103 and AlClg-NH^Cl, are not known to exist, and their non-existence is taken to explain the failure of the hj'^droxides of trivalent metals to dissolve in the solutions of ammonium salts. 2. Ionic Theory Another explanation of the part played by ammonium salts is based upon the simple ionic principle that the addition of an ion in common with one in the solute decreases the dissociation of the latter. (a) Salts of Bivalent Metals. — When NaOH or NII^OH is added in excess to the solution of a bivalent metal, a precipitate METALS OF GROUP III 103 is formed which readily dissolves on the addition of NH^Cl. The explanation is that the addition of the common ion, NH^, sup- presses the negative ion, OH, thus driving the dissociated ions into undissociated and inactive molecules of NH^OH. As the hydroxides of the bivalent metals in question are usually moder- ately well dissociated, NH^Cl would not only suppress the free OH ions of any excess of NH^OH, but also those of the hydrox- ides themselves. Of course it must be understood that if NaOH is used, this reaction first occurs : — NaOH + NH.Cl = NaCl + NH^OH. For application of the principle the important case of magne- sium salts with NH^OH and NH^Cl is considered : — Substituting in the equation a'b = ck,^ the equation NH, X OH = NH.OH x k, or ^^^' ^^,?/^ = k, is obtained. NH^, OH, NH^OH, and k are, respectively, the positive and nega- tive ions, the undissociated molecules, and the ionization constant for NH^OH. Now when NH^Cl is introduced the number of NH^ ions is greatly increased, and the result is to suppress the OH ions. By letting x = number of NH^ ions added, and y = those of the undissociated molecules of NH^OH (resulting from the addition of NH^Cl), the equation becomes : — (NH, 4- X -y) (OH - y) _ NH4OH + y This decreases the number of OH ions and leaves only (OH — y) for unit volume. It is owing to this disappearance of OH ions that Mg(0H)2 is not precipitated, — too little molecular Mg(0H)2 being formed to oversaturate the solution. (&) Salts of Trivalent Metals. — When NaOH, NH^Cl, and AlClg are brought together the following reactions probably occur : — AICI3 4- 3NaOH = Al(OH)3 -h 3NaCl ; NaOH-l-NH^Cl = NaCl + NII.OH. 104 CHEMICAL ANALYSIS Now A1(0H)3, unlike Mg(0H)2 and hydroxides of some biva- lent metals, is a very weak base and, consequently, very poorly dissociated. Hence," when an excess of NH^Cl is added, it sup- presses only the OH ions of NH^OH — not those of A1(0H)3. Thus NH4CI not only does not affect the precipitated hydroxide but also destroys the power of NH^OH to dissolve it. Chromium (salt for study, chromium sulphate, Cr2(S04)3). 1. NH4OH Ogives reactions similar to 1, under alumi- num. 2. NaOH gives reactions similar to 2, under aluminum. 3. BaCOg gives reactions similar to 3, under alumi- num, except that the precipitation requires more time for its completion. 4. (1014)28 gives reactions similar to 4, under alumi- num. 5. Fusion with fusion mixture on a platinum foil, or with sodium dioxide, Na202, on thick silver foil, gives a soluble yellow mass, containing sodium chromate, NaaCrO^. 6. Na202 heated with a solution of a chromium salt gives yellow Na2Cr04. 7. Borax bead with both oxidizing and reducing flames gives a yellow-green coloration of sodium chromium metaborate, NaQCr2(B02)i2- Reactions 5 and 6 illustrate the conversion of chromium as a base-producing element to chromium as an acid-producing element. There are two classes of chromium compounds derived from the two oxides CrgOg and CrOg. Chromic oxide, CrgOg, is basic and forms salts with acids: CrgOg -\- 6 HCl = 2 CrClg + 3 H2O. By oxidation CrgOg is changed to chromium trioxide, CrOg, which is an anhydride and forms METALS OF GROUP III 105 salts with bases: CrOg + 2NaOri = N'a2Cr04 + HgO. The oxi- dation of CrgOg in solution may be accomplished by Na202 or by hydrogen dioxide, HgOg. CrOg is a strong oxidizing agent and is easily reduced to CrgOg by various reagents, namely, HgS, SOg, HCl, and many organic compounds. If HgS is passed through an acidified (HCl) solution of potassium dichromate, KgCrgO^, there will be a change of color from red to green : K2Cr207 + 3H2S + 8 HCl = 2CrCl3 + 2KC1 + 3S + 7H2O. In this case chromium is changed from the acid to the basic condition. CHEMISTRY OF BORAX BEADS Borax (sodium tetraborate, NagB^O^- IOH2O), like NagCOg, is both an inactive flux, as in normal sodium borate, NagBOg, and an active chemical agent, as in boric acid, HgBOg. (See Fusion, p. 43.) When borax is heated, it loses its water of crystallization and fuses to a clear bead on the platinum wire. If a metallic oxide or salt is fused with the clear bead, a double borate is formed, which is often colored. The reaction can be easily understood by a review of the principal hydroxyl acids of boron. Water unites with boric oxide, B2O3, increasing in an arithmet- ical progression to form a systematic chain of polyboric acids, of which the following are important in this connection : — 2 BgOg + HgO = H2B4O7, dihydroxyl tetraboric acid ; 2B20g + 2H2O = 4HBO2, metaboric acid; 2 BgOg + 3 HgO = HgB^Og, hexahydroxyl tetraboric acid ; 2 BgOg + 4 H2O = 2 H4B2O5, diboric acid ; 2 BgOg + 5 H2O = HjqB^Ou, dekahydroxyl tetraboric acid ; 2B20g + 6H2O = 4 HgBOg, (normal) orthoboric acid. From these equations it can be seen that any succeeding acid in the list can be formed from the next preceding by the addition of one molecule of water ; and the reverse is also true, that any higher acid can be formed from the next lower. The corresponding salts of these acids are formed in a similar manner, provided the factor to be added is a metallic 106 CHEMICAL ANALYSIS oxide instead of water. The most important salts are given for illustration : — NagB^O,, sodium tetraborate (borax) ; NagB^O^ + NagO =4]SraB02, " metaborate; NagB^O^ + 3 NagO = 2 Na^B^O^, " diborate ; NagB^O^ + 5 NagO = 4 NagBOg, " ortlioborate. This further explains the statement above, that borax is both an inactive flux, like NagBOg, and an active chemical agent, like HgBOg. It is equivalent to four molecules of boric acid with one mole- cule of water replaced by sodium oxide, and the other five mole- cules of water displaced : — 4 HgBOg + NagO = Na^B^O^ + 6H2O. Thus it is in part a salt and in part an anhydride. If an oxide of a heavy metal is substituted for NagO, double borates are formed : — NagB^O^ + CoO = N'a2Co(B02)4, sodium cobaltous metaborate ; Na2B407 -I- 3 CoO = Na2Co3(B205)2, " cobaltous diborate ; NagB^O^ + 5 CoO = Na2Co5(B03)4, " cobaltous orthoborate. Or, in the case of the triad element chromium : — 3Na2B407 + CrgOg = NagCr2(B02)i2» sodium chromic metaborate, etc. The metaborate, then, is the first product formed by adding a small quantity of the oxide to an excess of borax, while the ortho- borate is the last formed by adding a larger amount of the oxide. What the actual composition of a given bead, made without weighing its components, may be can only be determined by a quantitative analysis. Probably every bead contains more or less of each of a large number of double borates. However, as the colors can be seen best by using small quantities of the oxides with a large excess of borax, the metaborates predomi- nate, and as such the beads are usually represented. METALS OF GROUP III 107 The same laws which apply to the formation of polyborates apply to the formation of polyphosphates and silicates. Hence the vast number of natural and artificial silicates can be traced to their corresponding acids. Ferrous Iron, Fe" (salt for study, ammonium ferrous sulphate, (NH4)2Fe"(S04)2). 1. NH4OH and NaOH precipitate ferrous hydroxide, Fe (011)2, which oxidizes quickly to brown ferric hydroxide, Fe(0H)3. NH4CI partly prevents the pre- cipitation by NH4OH, and partly that by NaOH. 2. (NH4)2S precipitates black ferrous sulphide, FeS. 3. K4Fe(CN)g precipitates white potassium ferrous ferrocyanide, K2Fe"Fe(CN)g, which rapidly oxidizes to Prussian blue. 4. K3Fe(CN)g precipitates TurnbuU's blue, ferrous ferricyanide, Fe3''(Fe(CN)g)2. 5. Borax bead with the oxidizing flame gives a yellow coloration, NaQFe2'"(B02)i2 ; with the reducing flame, a green coloration, Na2Fe''(B02)4. Ferric Iron, Fe'" (salt for study, ferric chloride, FeClg). 1. NH4OH and NaOH precipitate brown ferric hydrox- ide, Fe(0H)3, insoluble in excess of reagents. NH^Cl does not prevent the precipitation either by NH^OH or by NaOH. 2. BaCOg precipitates a brown basic salt, Fe20(C03)2. 3. H2S reduces ferric to ferrous salts and precipi- tates free sulphur: — 2 FeCl3 + H2S = 2FeCl2 + 2HC1 + S. 108 CHEMICAL ANALYSIS The precipitate of sulphur formed iu this reaction is sometimes mistaken for the sulphides of certain members of Group II. This confusion will not occur if it is recalled that the members of Group II are all colored, whereas a precipitate of finely divided sulphur is white. ^ 4. (NH4)2S reduces ferric to ferrous salts and precipi- tates ferrous sulphide : — 2FeCl3 + 3 (NH4)2S = 2FeS + GNH^Cl + S. 5. K4Fe(CN)g precipitates Prussian blue, ferric ferro- cyanide, Fe/"(Fe(CN)e)3. 6. K3Fe(CN)g gives a brown coloration. 7. KCNS gives a deep-red coloration, ferric sulpho- cyanate, Fe(CNS)3.i 8. Borax bead gives the same results as with ferrous salts. PKOCESS OF SEPARATION The separation of the members of this group is based upon the facts that Cr(0H)3 is oxidized to soluble NagCrO^ by means of Na202 or by fusion with the mixture of Na2C03 and KNO3, and that A1(0H)3 is soluble in an excess of NaOH. Two methods of mak- ing the separation are given, of which the first and simpler is to be employed in the absence of phosphoric, boric, silicic, and hydrofluoric acids; whereas the sec- ond and more complicated method is to be followed in the presence of these bodies.^ Boil off all traces of H2S, testing for its removal by holding above the liquid a strip of paper moistened with AgN03 or Pb(N03)2. It must be di'iven off METALS OF GROUP III l09 completely; since, if it were allowed to remain, the members of Group IV would be precipitated out of due course upon the addition of NH4OH, the precipitant for the members of Group III. Next test for iron by adding K3Fe(CN)g to a small portion of the solution; and in case it is present add a few drops of IINO3 and boil until the reaction for ferrous compounds disappears. The oxidation of iron to the ferric state at this point is necessary, since, if left in the ferrous state, it would not be precipitated by NH^OH in presence of NH4CI. The solution should now be tested for oxalic acid or other organic matter by evaporating a small portion of the solution to dryness, and heating the residue in a closed tube connected with a small rubber delivery tube, through which any gas that may be evolved can be conducted into lime water. A charred residue in the closed tube and a white precipitate in the lime water indicate the presence of organic compounds. ^ If such are found, evaporate the whole solution to dryness and heat the residue with the addition of a few drops of sulphuric acid, until the organic matter is thoroughly decomposed. Cover the residue with 1 In testing for organic matter blackening is not conclusive. Many inor- ganic salts, among them those of iron, nickel, cobalt, and manganese, blacken when heated. Furthermore, a failure to blacken is not an evidence of the absence of organic compounds, since some of them which contain a large per cent of oxygen — oxalates in particular — do not char, but give off all their carbon as oxides. The lime-water test, too, is not absolute, though more reliable than that by charring. Compounds evolving oxides of sulphur also whiten lime water. But both tests are good signs ; and as gentle ignition is also the means of eliminating silicic acid, it is best to evap- orate and ignite the solution even when the presence of organic acids is doubtful. V " "' 110 CHEMICAL ANALYSIS concentrated HCl, evaporated almost to dryness, add water and a few drops of concentrated HNO3, and boil. Filter the solution from the separated carbon and silica. Phosphoric acid^ must next be tested for by warm- ing a small portion of the filtrate with an excess of HNO3 solution of (NH4)2Mo04. Should it be present, barium must also be tested for at this stage by making a small portion alkaline with NH^OH, acidifying with HC2H3O2, and testing for barium with K2Cr04. Next add NH^Cl,^ boil, and add NH^OH till its odor persists. Filter quickly while hot. Reserve filtrate (a) for subse- quent groups. Redissolve residue (a) in least quantity of HCl, nearly neutralize with Na2C03, transfer to a stoppered flask, and add when cold a large excess of suspended BaCOg.^ Shake from time to time, and filter after 15 minutes. If phosphoric acid is present, combine filtrates (a) and (h)^ ; if absent, test filtrate {b) for man- ganese by evaporating to dryness and fusing with NagCOg and KNOg. Residue (h) may consist of basic salts of Group III, if phosphoric acid is absent ; or, if present, it may also contain phosphates of metals of Groups III, IV, V, and of magnesium. In Absence of Phosphates Method 1. Thoroughly wash the residue {b) and trans- fer to a test-tube. Add a small quantity of water and some bits of Na202, and boil till effervescence ceases. Filter and wash. The residue {c) may be brown Fe(OH)g, whose identity can be confirmed by dissolving in dilute HCl and testing with K4Fe(CN)6. The filtrate {e) may contain yellow NagCrO^ and NagAlOg. METALS OF GROUP III 111 The following equations explain the formation of the soluble salts : — 3 NaaOa + 2 Cr(OH)3 = 2 Na2Cr04+ 2 H2O + 2 NaOH ; 6 NaOH + 2 A1(0H)3= 2 Na3A103+ 6 II2O. Divide the filtrate (c) into two parts. Acidify the one with HCgHgOg and test for chromium by adding Pb(C2H302)2. Acidify the other part with dilute HCl, and while boiling test for aluminum by adding an excess of NH^OH, — or to the second part of the filtrate add some NH^Cl and boil. After cooling, a gelatinous precipitate confirms presence of aluminum. Method 2. Dry residue (h) and fuse in a platinum crucible or foil with an excess of fusion mixture. The cooled mass is triturated in a mortar, — preferably a glass one, — is digested with water for fifteen minutes, and then is filtered. The residue {c) is tested for iron, and the filtrate {c) for chromium and aluminum, as in Method 1. In case of doubt test the solution for aluminum with the spectroscope. In Presence of Phosphates Dissolve residue {h) in a little HCl, nearly neutralize by cautious addition of Na2C03, add NaC2H302 and HC2H3O2, and boil and filter. The filtrate (c) may con- tain phosphates of the metals of Groups IV and V, and of magnesium. The residue {c) may consist of the phosphates of aluminum, chromium, and iron. To the filtrate (c) add dilute FeClg, drop by drop, until a red coloration appears. At this point, precipitation of 112 CHEMICAL ANALYSIS FePO^ is completed. The red color indicates the formation of Fe(C2Hg02)3, an excess of which would redissolve the FePO^. Hence the solution should now be boiled in order to change any excess of Fe(C2H302)3 to an insoluble basic acetate, FeO(C2H302). It is neces- sary to filter the mixture with the pump while hot, as FeO(C2H302) redissolves to Fe(C2H302)3 on cooling. The filtrate (d) may contain the chlorides of Groups IV and V, and of magnesium. This filtrate should be combined with filtrate (a) and afterwards tested for sub- sequent groups. The residue ( This reaction occurs only with salts of HgCOg, not with the free acid, 130 ACIDS OF GROUP I 131 2. HCl and other acids, excepting HgS and HCN, decompose carbonates with evolution of COg. This gas is readily soluble in water ; and in dilute solutions of the carbonates, it may not be formed in sufficient quantity to oversaturate the solution and escape. From concentrated or hot solutions, it escapes with efferves- cence. Being heavier than air, it may be detected by decantation into a test-tube containing lime water, its presence being shown by the appearance of a milky precipitate : — Ca(0H)2 + CO2 = CaCOg + H2O. An excess of COg will dissolve the precipitate first formed. Sulphur dioxide, SO2, will also produce a white precipitate with lime water, but can usually be detected by its odor. Silicic Acid, H4Si04 (salt for study, sodium silicate, Na4Si04). 1. BaCl2 precipitates white barium silicate, BagSiO^, decomposed by dilute HCl, with separation of gelati- nous 1148104. 2. HCl added, drop by drop, to the solution of a sili- cate, precipitates gelatinous H^SiO^, which, on evapora- tion to dryness, is decomposed with the formation of silicic anhydride, SiOg. 3. Fused with NagCOg on a platinum foil until bub- bles of gas cease to escape, most insoluble silicates are changed by metathesis to sodium silicate and a metallic carbonate or oxide. If the fused mass is then boiled with dilute HCl and filtered, the filtrate will contain 132 CHEMICAL ANALYSIS the chloride of the metal and the residue will consist of H^SiO^: — (a) BagSiO^ + 2Na2C03 = 2BaC03 + Na^SiO^ ; (h) BaCOg + 2HC1 = BaCla + H2O + COg; (e) Na^SiO^ + 4HC1 = H^SiO^ + 4NaCl. 4. HF in an aqueous solution, or in gaseous form, decomposes SiOg with evolution of silicon tetrafluoride, SiF^: — SiOg + 4HF = SiF^ + 2H2O. If a silicate is mixed with three parts of NH^F or five parts of CaFg, moistened with concentrated HgSO^, and then heated till fumes cease to escape, the silicic acid is decomposed and expelled : — (a) H2SO4 + 2NH4F = (NH4)2S04 + 2HF; (h) 6HF + Na2Si03 = Na2SiFg + 3 H2O ; (c) Na2SiF6 + H2S04 = Na2S04 + 2HF + SiF4. 6. Metaphosphate bead dissolves the metallic parts of the silicates, but not the Si02, which remains floating in the fused bead. As SiOg is not affected, the outline of the particle of the silicate remains intact, giving rise to the so-called "skeleton bead." ANALYSIS OF SILICATES There are two classes of silicates important in analytical chem- istry — silicates decomposed by acids, and those not decomposed by acids : — First class : Silicates decomposed by acids. This is not a very numerous class, composed for the most part of the soluble alkali metal silicates and a few less soluble single and double silicates of other metals. The analysis of this class is quite simple. This ACIDS OF GROUP I 133 is accomplished by treatment of the silicates with HCl, which by- metathesis form soluble chlorides of the metals and colloidal silicic acid. Second class : Silicates not decomposed by acids. This consti- tutes by far the more numerous class, including the natural sili- cates. Many natural silicates contain the alkali metals combined with other metals. The varieties of feldspar are representatives of this kind. The analysis of silicates riot decomposed by acids is usually conducted by one of three methods : — Method 1. — Fusion with alkali-metal carbonates. By metath- esis, soluble silicates and carbonates of the metals of the original silicates are formed, — which resulting salts are then decomposed by HCI (see reactions above). Finely powder the silicate, mix with about 3 parts of NagCOg, or fusion mixture, and heat to quiet fusion in a platinum crucible or foil. When cool, boil the mass in water. Filter, evaporate to a small bulk, and add concentrated HCl. HgSiOg will precipitate as a gelatinous mass. If it is desired to test for the presence of alkali metals in the silicate, this method cannot be used, as the carbonates of these metals are added as a flux. Method 2} — (Method of J. Lawrence Smith.) Fusion with NH4CI and CaCOg. An insoluble silicate like feldspar, contain- ing alkali metals, may be converted into soluble alkali-metal chlorides and some insoluble hydroxides, by heating to redness in a covered platinum crucible with 1 part NH^Cl and 8 parts powdered CaCOg. In all fusions it is necessary for both the substance and the flux to be reduced to very fine powders, and intimately mixed. Method 3. — Fusion with BaO. Fuse in a platinum crucible a mixture of 1 part of the powdered silicate and 4 parts BaO. Digest the mass in a little water to detach it from the crucible, and then dissolve in HCl. Add NH^OH till alkaline, filter, evaporate to dryness, and ignite. 134 CHEMICAL ANALYSIS Sulphurous Acid, H2SO3 (salt for study, sodium sulphite, NaaSOs). 1. BaClj precipitates white barium sulphite, BaSOg, soluble in dilute HCl. 2. Nascent hydrogen reduces sulphites to sulphides, which are decomposed by an excess of HCl with evolu- tion of HgS, detected by its odor, or with Pb{C2H302)2- 3. HgS decomposes sulphites with separation of sulphur. 4. HCl decomposes sulphites with evolution of SOg, detected by its odor and by the production of a white precipitate of calcium sulphite with lime water. Sulphuric Acid, H2SO4 (salt for study, sodium sulphate, Na2S04). 1. BaCl2 precipitates white barium sulphate, BaSO^, insoluble in water or acids ; decomposed by fusion with NagCOg in a platinum crucible or foil, sodium sulphate and barium carbonate being formed : — NaaCOg + BaSO^ = Na2S04 + BaCOg. In like manner, the other insoluble sulphates, SrS04, CaSO^, and PbS04, are decomposed by fusion with Na2COg or by boiling with its solution. 2. Pb(C2Hg02)2 precipitates white lead sulphate, PbS04, almost insoluble in dilute HNOg ; soluble in hot con- centrated HCL' 3. Fused with NagCOg on charcoal, sulphates are reduced to sulphides. If the mass is moistened with very dilute HCl and placed on a bright silver coin, the latter will be stained black. ACIDS OF GROUP I 135 Phosphoric Acid, H3PO4 (salt for analysis, sodium phos- phate, HNagPOi). 1. BaCl2 precipitates white barium phosphate, HBaPO^, — or Ba3(P04)2, if the solution contained a normal phos- phate, — soluble in HCl and HNO3. 2. MgSO^ in presence of NH^OH and NH^Cl precipi- tates white crystalline ammonium magnesium phos- phate, NH^MgPO^, soluble in acids. (Compare with NH^MgAsO^.) 3. (NH4)2Mo04 in HNO3 solution precipitates, in the cold, yellow ammonium phospho-molybdate, (Mo03)i2*(-^H4)3P04. (Compare with behavior of the same reagent toward arsenates.) 4. FeCl3 in presence of NaC2H302 precipitates yellow ferric phosphate, FePO^, soluble in strong acids and excess of FeCl3; insoluble in HC2H3O2. Boric Acid, H3BO3 (salt for study, borax, Na2B407). 1. BaCl2 precipitates white sodium barium borate, Na2Ba5(B03)4, soluble in acids, except H2SO4. 2. H2SO4 precipitates from hot solutions of borates, on cooling, crystalline boric acid, H3BO3. 3. Alcohol, added to free boric acid or to a borate with concentrated HgSO^ and then kindled, burns with a green flame, especially upon stirring the mixture. 4. Turmeric paper, immersed in a slightly acid (HCl) solution of boric acid or a borate and then dried, shows a reddish tint which is turned blue by NaOH. 136 CHEMICAL ANALYSIS Oxalic Acid, H2C2O4 (salt for study, sodium oxalate, Na2C204). 1. BaCl2 precipitates from neutral solutions white barium oxalate, BaCgO^, somewhat soluble in dilute NH4CI and many organic acids; soluble in HCl and HNO3. 2. Lime water and soluble calcium salts precipitate white calcium oxalate CaCgO^, soluble in HCl and HNO3; insoluble in organic acids. 3. Concentrated H^SO^, heated with oxalic acid or an oxalate, removes water, and the compound is decom- posed into CO2 and CO : — H2C2O4 H- H2SO4 = CO2 + CO + H2S04-H20. If in sufficient quantity the CO gas can be burned with its characteristic blue flame. 4. Heating decomposes all oxalates with formation of carbonates or oxides of the metals, and evolution of CO or CO2. Tartaric Acid, H2C4H4O6 (salt for study, potassium tar- trate, K2C4H4O6). 1. BaClg (or, better, CaClg) from neutral solutions precipitates white barium (or calcium) tartrate, soluble in acids, except H2SO4. 2. AgNOg precipitates white silver tartrate, Agfi^fif^^ soluble in NH^OH. On warming this solution, black metallic silver is deposited. If the Ag2C4H40g be carefully redissolved in the least possible amount of NH^OH, and if this solution be heated gently ACIDS OF GROUP I 137 in a test-tube, a mirror of metallic silver will be depos- ited on the walls of the tube. AgNOg precipitates AggC^H^Og only from neutral solutions. This reaction distinguishes tartaric from most other organic acids. 3. Heated in a closed tube, tartrates char and emit inflammable vapors with the odor of burnt sugar. Commingled with the carbon residue is also a carbon- ate, detected by effervescence on adding HCl. Hydrofluoric Acid, HF (salt for study, ammonium fluoride, NH4F). 1. BaCl2 precipitates white barium fluoride, BaFg, soluble with difliculty in HCl and HNO3. 2. Concentrated H^SO^ mixed to a paste with powdered fluorides and warmed in a platinum vessel expels gas- eous HF : — 2NH4F + H2SO4 = (NH4)2S04 + 2HF. If the vessel is loosely covered for an hour with a watch-glass which previously has been coated with wax through which some lines have been cut with a sharp instrument, the lines will be seen to have been etched into the glass upon removal of the wax. The reaction involved is identical with No. 4, under silicic acid. DETECTION OF THE ACIDS OF GROUP I The analysis for acids cannot be made by following a systematic scheme of separation, such as is used in the analysis for metals ; on the contrary, the presence or absence of each acid must be established chiefly by individual tests applied to the original material. 138 CHEMICAL ANALYSIS For convenience the members of Group I may be classified as follows : — Sub-group J_ H2Cr04, HgCOg, H^SiO^, H2SO3. These acids are decomposed, in solution, by HCl and Sub-group ZT— H2SO4, H3PO4, H3BO3, H2C2O4, H^C^H^Oq^ HF. These acids are not decomposed by HCl or H2S. Neutralize a small portion of the original solution, and add some BaClg (or Ba(N03)2, if metals of Group I are present). A precipitate confirms the presence of one or more acids of Group I. Divide a larger portion of the original solution into four parts : — Part I^ for chromic add. — A yellow color indicates chromic acid, confirmed by acidifying with HC2H3O2 and adding Pb(C2H302)2- Part II, for carbonic acid, — Add HCl and warm. An effervescence of an odorless gas indicates the presence of CO2. Confirm by testing with lime water. Part in, for sulphurous acid. — Add HCl and warm. Effervescence with odor of burning sulphur indicates the presence of SO2, confirmed by passing the gas through lime water. Part IV, for silicic acid. — Add dilute HCl, drop by drop. A gelatinous precipitate indicates H^SiO^, con- firmed by evaporating to dryness and testing with the metaphosphate bead. If any of these acids are present, they must be removed from solution before testing for the members of Sub-group II. H2Cr04 is destroyed by HgS, in ACIDS OF GROUP I 139 presence of HCl ; HgCOg and H2SO3 are driven off by boiling with HCl; and 1148104 is removed by evapo- ration with HCl. The solution, thus freed of members of Sub-group I, is now neutralized exactly with NH4OH, — free of (NH4)2C03, — and its examination is continued as follows : — To a small portion BaClg is added. If no precipitate is formed, all members of Sub-group II are absent. If a precipitate is formed which dissolves on the addition of HCl, H2SO4 is absent, but other members may be present. If a precipitate is formed which does not dissolve in HCl, HgSO^ (possibly other acids) is present. In either of the latter cases it is necessary to test indi- vidually for the remaining acids of the group in small portions of the solution. Part I, for phosphoric acid, — Add a few drops of the solution to a strong HNO3 solution of (NH4)2Mo04, and warm gently. A yellow crystalline precipitate confirms the presence of H3PO4. Part II, for boric acid. — Acidify some of the solution with HCl and test with turmeric paper. Evaporate another portion almost to dryness, add alcohol and con- centrated H2SO4, and kindle. A green flame confirms the presence of H3BO3. Part III, for oxalic acid. — Add lime water and boil the white precipitate with HC2H3O2. If the precipitate fails to dissolve, it confirms the presence of H2C2O4. Part IV, for tartaric acid. — Neutralize the solution and add CaCl2. If a white precipitate occurs, filter, dry the residue, and heat in a closed tube. Charring with 140 CHEMICAL ANALYSIS the odor of burnt sugar, and effervescence of the residue with HCl, confirm the presence of HgC^H^Og. Part Vy for hydrofluoric acid. — Evaporate the solution to dryness, transfer the residue to a platinum crucible, add concentrated HgSO^, and cover with a watch-glass. If the gas etches the glass cover, the presence of HF is confirmed. CHAPTER XIV ACIDS OF GROUP II: HYDROCHLORIC, HYDROBROMIC, HYDRIODIC, HYDROCYANIC, HYDROFERROCYANIC, HYDROFERRICYANIC, SULPHOCYANIC, AND HYDRO- SULPHURIC ACIDS Characteristic : Insolubility of their silver salts in dilute nitric acid. Group Reagent : Silver nitrate. REACTIONS Hydrochloric Acid, HCl (salt for study, sodium chloride, NaCl). 1. AgNOg precipitates white silver chloride, AgCl, insoluble in dilute acids; soluble in KCN, NH^OH, and in boiling solution of ammonium "sesqui" car- bonate. 2. Pb02 or Mn02 with concentrated H2SO4 expels chlo- rine gas, detected with starch-KI paper. 3. K2Cr207 with concentrated H2SO4 gives red fumes, condensing to a brown liquid, chromic oxy- chloride, Cr02Cl2, changing to yellow (NH4)2Cr04 on the addition of NH^OH. The dry chloride should be triturated with K2Cr207 crystals, and dis- tilled with concentrated HgSO^ in a small retort (25 c.c). 141 142 CHEMICAL ANALYSIS Hydrobromic Acid, HBr (salt for study, potassium bromide, KBr). 1. AgNOg precipitates yellow silver bromide, AgBr, insoluble in dilute acids and in ammonium "sesqui" carbonate; soluble in NH^OH and KCN. 2. PbOg with concentrated HgSO^ expels brown vapors of bromine, identified by their odor and color. 3. 'Kji^rjdrj with concentrated H2SO4 expels bromine, which is decolorized by NH^OH, forming NH^Br. 4. Chlorine liberates bromine, detected in small quan- tities by coloring carbon disulphide or chloroform brownish-red. Mix the bromide solution with about 1 c.c. of CSg, then add dilute chlorine water, drop by drop, and shake well. The globules of CSg will assume a reddish tint. An excess of chlorine should be avoided, lest it combine with bromine to form colorless bromine chloride, BrCl. Hydriodic Acid, HI (salt for study, potassium iodide, KI). 1. AgNOg precipitates yellow silver iodide, Agl, insoluble in dilute acids, NH^OH and ammonium "sesqui" carbonate; soluble in KCN. 2. PbOg with concentrated HC2H3O2 liberates violet iodine, turning starch paper blue. 3. KgCrgO^ with concentrated H2SO4 liberates iodine. 4. Chlorine water liberates iodine, turning starch paper blue. An excess of chlorine will decolorize the paper by formation of iodine chloride, ICl. 5. KNO2 in concentrated H2SO4 liberates iodine. Into a clear solution of starch paste and an iodide, dip ACIDS OF GBOUP II 143 a glass rod moistened with a solution of KNO2 in con- centrated H2SO4. The liquid in contact with the rod becomes blue. It is necessary to keep the reagent cold, as iodized starch becomes colorless in hot water. Hydrocyanic Acid, HCN (salt for study, potassium cya- nide, KCN). 1. AgNOg precipitates white silver cyanide, AgCN, soluble in excess of KCN, forming the salt KAg(CN)2. AgCN is also soluble in NH^OH and boiling HCl. 2. FeSO^, with a few drops of FeClg, added to tlie solu- tion of a cyanide in weak NaOH, precipitates a bluish- green mixture of ferrous ferric hydroxide, Fe302(OH)^, and Prussian blue. Fe302(OH)4 can be dissolved with dilute HCl, leaving the Prussian blue intact. 3. (NH4)2S3,(a few drops) and a drop of NaOH added to a cyanide solution, form ammonium sulphocyanate, NH^CNS, on heating. Evaporate the solution to dry- ness and test by dissolving in dilute HCl and adding FeClg solution. A deep red coloration shows the pres- ence of HCNS, derived from HCN : — (NH4)2S^ + 4 KCN = 4 KCNS + (NH4)2S(rc - 4). 4. HNaCOg heated with a cyanide expels HCN gas, identified by its odor and the rose color of its flame. Hydroferrocyanic Acid, H4Fe(CN)Q (salt for study, potas- sium ferrocyanide, K4Fe(CN)e). 1. AgNOg precipitates white silver ferrocyanide, Ag4Fe(CN)g, soluble in KCN; insoluble in NH^OH and HNOg. 144 CHEMICAL ANALYSIS 2. FeClg precipitates Prussian blue (see Iron, p. 107). 3. CuSO^ precipitates brown cupric ferrocyanide, Cu2Fe(CN)6 (see Copper, p. 84). Hydroferricyanic Acid, H3Fe(CN)6 (salt for study, potas- sium ferricyanide, K3Fe(CN)g). 1. AgNOg precipitates orange-red silver ferricyanide, Ag3Fe(CN)6, soluble in NH^OH and KCN; insoluble in HNO3. 2. FeSO^ precipitates TurnbuU's blue (see Iron, p. 107). Sulphocyanic Acid, HCNS (salt for study, potassium sulphocyanate, KCNS). 1. AgN03 precipitates white silver sulphocyanide, AgCNS, soluble in NH^OH; insoluble in dilute HNO3. 2. FeClg acidified with HCl gives a deep red colora- tion of Fe(CNS)3 (see Iron, p. 108). Hydrosulphuric Acid, H2S (salt for study, sodium sul- phide, NagS). 1. AgN03 precipitates black silver sulphide, Ag^S. 2. Na2FeNO(CN)5 (sodium nitro-prusside) added to alkaline (NaOH) solution of a sulphide gives a brilliant red-violet tint. 3. Fused with NaOH, insoluble sulphides form NagS ; and on dissolving the mass in a little water, the solu- tion will tarnish a bright silver coin brown. 4. HCl sets free H2S from all soluble, and from many insoluble sulphides ; recognized by its odor and by its 1 ACIDS OF GROUP II 145 power of blackening paper moistened with a solution of DETECTION OF THE ACIDS OF GROUP II The separation and identification of the acids of this group are accomplished by the following means : — (a) The removal of HgS by means of a solution of ZnSO^ in NaOH. (b) Hager's method of detecting HCl, HBr, and HI in the presence of each other ; based upon the different degrees of solubility of AgCl, AgBr, and Agl in ammo- nium "sesqui" carbonate and NH^OH. (c) The detection of HON in the absence of H4Fe(CN)6, H3Fe(CN)6, and HCNS, by the precipita- tion of Prussian blue from a solution of a cyanide by FeSO^, FeClg, and NaOH. (d) The detection of HON in the presence of H4Fe(CN)6, H3Fe(CN)6, and HCNS, by the evolution of HCN on distilling with HNaCOg. HgS must first be tested for in a small portion of the original solution, preferably by adding HCl, boiling, and noting whether any gas is given off which causes lead acetate paper to blacken. If found, it must be removed from the remainder of the solution before test- ing for the other members of the group, since its pres- ence would hinder their detection. Therefore, treat a sufficient portion of the solution with a solution of ZnSO^ in an excess of NaOH, which will precipitate the H2S as ZnS. Reject the precipitate, and divide the filtrate, or portion of the original solution if HgS is absent, into three parts. 146 CHEMICAL ANALYSIS Part I, for HCl, HBr, and iTZ— Acidify with HNO3 and add AgNOg. Filter and reject the filtrate. Boil the residue with 100 parts of a solution of ammonium " sesqui " carbonate. Decant the clear supernatant liquid, add more ammonium " sesqui " carbonate, and again boil and decant. The decanted liquid may con- tain AgCl, which can be determined by acidifying with HNO3. The residue from which the liquid has been decanted may consist of AgBr and Agl. Treat it with a dilute solution of NH^OH (5 per cent ammonia water) and filter. The filtrate may contain AgBr, detected by acidifying with HNO3. "^^^ residue may be Agl, indicated by its yellow color. For a further confirma- tion of the three halogens (consisting of the AgCl and AgBr precipitates from the ammoniacal solutions and the undissolved Agl) each can be fused with NagCOg, boiled with water, and filtered : — 2 AgCl + Na2C03 = 2NaCl + AggCOg, etc. The filtrates can be tested for the individual halogens as follows : — (a) Solution of NaCl. Evaporate to dryness and heat with concentrated HgSO^ and PbOg. The evolved chlo- rine can be detected by its odor, its bleaching moistened litmus paper, or its effect on starch-KI paper. (b) Solution of NaBr. Evaporate to dryness and heat with concentrated H2SO4 and PbOg. The evolved bromine can be detected by its odor or by its color. {c) Solution of Nal. Neutralize with HNO3 and add some drops of starch paste and chlorine water. A blue solution confirms presence of the iodide. i ACIDS OF GROUP II 147 Part II for ff^FeiCN),, HsFe^CN)^, and EONS,— Neutralize with HNO3 and divide into two small parts. Pour one part into a test-tube and shake the tube so that its sides will be moistened with the liquid. Hold- ing the tube obliquely, add a few drops of dilute FeClg solution so that they will run down the sides of the tube. A red coloration indicates the presence of HCNS. If H4Fe(CN)g is present, Prussian blue will be formed also, but the red color can be seen commingled with the blue. Add more FeClg. The formation of Prussian blue confirms presence of H4Fe(CN)g. To the second smaller part add FeSO^. The formation of Turnbull's blue confirms the presence of H3Fe(CN)g. Part III, for HCN. — li }1^Yq{C^)^, H3Fe(CN)g, and HCNS are absent, add NaOH, FeSO^, a few drops of FeClg, and HCl in excess. Formation of Prussian blue confirms the presence of HCN. If H4Fe(CN)6, H3Fe(CN)6, and HCNS are present, add some solid bicarbonate of sodium, HNaCOg, to the neutral solution in a test-tube, and boil. The odor of bitter almonds indicates the presence of HCN. Con- firm by kindling the gas. It should burn with a rose- tinted flame. HCN is a deadly poison ; do not inhale. CHAPTER XV ACIDS OF GROUP HI : NITRIC, CHLORIC, AND ACETIC ACIDS No group characteristic. No group reagent. REACTIONS Nitric Acid, HNO3 (salt for study, potassium nitrate, KNO3). 1. Heated on charcoal, nitrates deflagrate with igni- tion, giving off COg : — 2KNO3 + C = 2KNO2 + CO2. Use small quantities of the nitrate in performing this experiment. 2. Heated with KCN in a platinum crucible or foil, nitrates deflagrate with ignition and detonation : — KNO3 + KCN = KNO2 + KCNO. 3. Mixed with copper filings and heated with con- centrated H2SO4, nitrates give red fumes of NO2. 4. If a concentrated solution of FeSO^, free of ferric salts, be carefully added to the cold solution of a nitrate in concentrated H2SO4, so that the two solu- tions form distinct layers, a brown ring will be formed at their junction, (FeS04)2NO : — (a) 2HN03 + 6FeS04 + 3H2S04 = 3Fe2(S04)3 + 4H20 + 2N0; (b) 2FeS04 + NO = (FeS04)2NO. ACIDS OF GROUP III 149 5. Brucine dissolved in concentrated H2SO4 gives a deep red color with nitrates. Touch the edge of the dissolved brucine with a glass rod moistened with nitrate solution ; a distinct red ring will bound the rod. 6. Reduced with zinc dust and HgSO^, nitrates yield nitrous acid, HNO2, detected by starch-KI paper. 7. NaOH with zinc dust and iron filings, on heating, reduces nitrates and sets NHg free : — HNO3 + 8 H = NH3 + 3 H2O. Chloric :At:id, HCIO3 (salt for study, potassium chlorate, KCIO3). 1. Heated on charcoal, chlorates deflagrate with vivid ignition, giving off COg : — 2KCIO3 + 3C = 2KC1 + 3CO2. 2. Heated with KCN in a platinum crucible, chlorates deflagrate with ignition and detonation : — . KCIO3 4- 3 KCN = KCl + 3KCN0. As HCIO3 gives up more oxygen than HNO3, the chemical action in Reactions 1 and 2 is necessarily more vigorous than in those under HNO3. Therefore, use very small quantities of chlorate. 3. Concentrated H2SO4 (a few drops), added with a pipette to a watch-glass containing a chlorate solution, liberates chlorine peroxide : — 3 KCIO3 + 2 H2SO4 = KCIO4 + 2 CIO2 + H2O + 2HKSO4. The peroxide is characterized by a disagreeable odor and a yellow coloration ; also by bleaching a blue solution of indigo. Neither heat nor large quantities of reagents should be used. 4. Brucine behaves very much alike towards nitrates and chlorates. 150 CHEMICAL ANALYSIS Acetic Acid, HC2H3O2 (salt for study, sodium acetate, • NaC2H302). 1. Heated to redness, acetates are decomposed with the formation of carbonates and of acetone, CgHgO, a liquid of penetrating, pleasant, ethereal odor : — 2NaC2H302 = Na2C03 + CgHgO. 2. FeClg, a few drops, added to a neutral acetate solution, produces a deep red coloration, due to the formation of ferric acetate, Fe(C2H302)3. On boiling, the solution is decolorized, and brown basic ferric acetate is precipitated. 3. Heated with concentrated H2SO4 and alcohol, ace- tates yield ethyl acetate, (C2H5)C2H302, characterized by its pungent ethereal odor : — NaC2H302 + C2H5OH = (C2H5)C2H302 + NaOH. DETECTION OF THE ACIDS OF GROUP III Individual tests must be made for the three acids of this group, in separate portions of the original solution. If iodides or bromides are present, they must be removed from the portion which is to be tested for HNO3 by adding HgClg solution and filtering, rejecting the precipitate. Otherwise, they would give a dark coloration on the addition of H2SO4. In testing for HNO3, Reactions 3 and 4 are to be used. In testing for KCIO3, Reaction 3 is to be employed. If H3PO4 is present, it is to be removed before testing for HC2H3O2, since it forms insoluble FePO^ with FeClg. (See p. 111.) Use Reaction 2 in testing for HOgHgOg. CHAPTER XVI THE SYSTEMATIC PROCEDURE OF ANALYSIS PRELIMINARY TESTS The physical properties of the substance under exami- nation — color, odor, whether solid or liquid, etc. — are first to be noted. Solids.i — If the substance is a solid, apply the follow- ing tests to small portions : — (a) Blowpipe flame on charcoal (see Tables IV and V, pp. 48, 49). (b) Heating in a closed tube (see Table II, p. 45). (c) Fusion with borax bead (see Table III, p. 47). (d) Flame coloration on a platinum wire (see Table VI, p. 51). (e) Spectra (see Table VII, p. 58). A larger portion of the solid is to be used for solu- tion, in preparation for the analysis by the wet way. First, treat it with water, determining whether the whole or only a part dissolves. If it be insoluble, or only partly soluble, divide the mixture into three parts, two small and one large, which may be numbered respectively 1, 2, and 3. To 1, in a test-tube, add some dilute HCl and boil. If still insoluble or partly insoluble, add an equal volume of concentrated HCl and boil again. If soluble, then treat 3, the largest portion, with concentrated HCl and 151 152 CHEMICAL ANALYSIS boil. If insoluble or partly insoluble, treat 2, first with dilute, then with concentrated HNO3. If soluble, treat 3 in like manner. If insoluble or partly insoluble in HNO3 as well as in HCl and water, combine the strong HCl and HNO3 mixtures, 1 and 2, and boil. If soluble, treat 3 with aqua regia. If insoluble or partly insoluble, recall which of the four solvents — water, HCl, HNOg, or aqua regia — dissolved the substance most ; and treat the larger portion, 3, with that solvent. Filter. Fuse the residue ^ with fusion mixture on a platinum foil or in a platinum crucible, and boil the mass with water. Sometimes this solution can be added directly to the filtrate without precipitation. Generally, how- ever, a precipitate will be formed. In order to deter- mine this, take small portions of both liquids and mix them. If no precipitate forms, combine the whole of both solutions. If a precipitate forms, separate analyses must be made of the two solutions.^ Liquids are to be tested with litmus paper to determine whether they are neutral, acid, or alkaline; and, also, small portions are to be evaporated to dryness on the wjiter bath. No residue being left, a neutral reaction indicates that only water is present; whereas an acid or alkaline reaction indicates the presence of a volatile acid or of ammonia. If a residue is left on evaporation: — (a) A neutral reaction indicates the presence in solu- tion of a neutral salt. (b) An acid solution may be either (1) a free acid, (2) an aqueous solution of certain normal salts likej SYSTEMATIC PROCEDURE OF ANALYSIS 153 FeClg or CuSO^, which have acid reactions, (3) certain acid salts like bisulphate of potassium, HKSO^, or (4) an acid solution of certain salts. (c) An alkaline solution may contain (1) a free alkali, (2) certain normal salts like Na2C03 which have an alka- line reaction, or (3) an alkaline solution of certain salts. In theory an aqueous solution of a basic salt should react alkaline ; but as the metals which form basic salts have not a very pronounced metallic character, their alkalinity is either too weak to be detected by litmus or, being very weak, is neutralized by water. SYSTEMATIC ANALYSIS FOR METALS If the solution 1 is neutral or alkaline,^ add dilute HCl till acid ; if acid, boil off the excess of acid, and when cold add dilute HCl. If a precipitate is formed, filter, and wash the residue with cold water. Analyze the residue for members of Group I, according to the direc- tions on p. 80. Acidify the filtrate strongly with more HCl, warm to about 70°, and pass a constant stream of HgS through it for about fifteen minutes. Then cool and dilute the solu- tion, and, before filtering, pass HgS again till saturation is completed. If a precipitate is formed, filter, wash, and analyze for members of Group II, as directed on p. 94. Boil off all traces of H2S from the filtrate ; test for ferrous iron^with K3Fe{CN)g ; and, if it be present, add a few drops of HNO3, ^^^^ ^^^^ until the iron is wholly oxidized to the ferric state. Unless the preliminary 154 CHEMICAL ANALYSIS examination has indicated conclusively whether organic matter or phosphates are absent or present, it will be necessary to test for them at this point, as is directed on p. 109. According as they are absent or present, follow the instructions given for the precipitation and separation of the members of Group III, on pp. 110 and 111. Boil off all excess of NH^OH from the filtrate ^ which is to be examined for Groups IV, V, and VI ; add (NH4)2S in moderate excess, and if a precipitate is formed, filter and wash thoroughly. Examine it for members of Group IV according to the directions given on p. 118. To the filtrate which may contain Groups V and VI add NH4CI, NH4OH, and (NH4)2C03 in quantity suffi- cient to precipitate completely any members of Group V which may be present. Warm the mixture gently ; and if a precipitate has formed, filter and wash with ammo- niated water, rejecting the washings. Examine it for members of Group V, according to the directions given on p. 123. Concentrate the filtrate which is to be examined for Group VI, and add small amounts of (NH4)2S04 and (NH4)2C204, to remove any traces of Ca and Ba which may be present. Filter and reject the precipitate, if one be formed ; and examine the filtrate for members of Group VI, as directed on p. 127. Concentrate some of the original solution, and test for ammonium salts with NaOH. SYSTEMATIC PROCEDURE OF ANALYSIS 155 SYSTEMATIC ANALYSIS FOR ACIDS It will have been observed that the preliminary exami- nation and the results of the analysis for metals throw much light upon the nature of the acids which may be present in the material which is being analyzed. For example, the presence of tartaric acid may be indicated by the result of heating in a closed tube; nitrates or chlorates show their presence by deflagration, when heated on charcoal; carbonates, sulphites, sul- phides, and cyanides are detected upon the addition of HCl, the reagent for the metals of Group I, by efferves- cence with or without characteristic odor. Furthermore, the results of the analysis for metals will show, according as Or and As are found absent or present, whether chromic, arsenious, and arsenic acids are absent or possibly present. But in addition to these indications there are others, depending upon the nature of the metals present in a substance and upon the character of the solution of that substance, which may show conclusively whether certain acids or groups of acids are absent or present. If, for example, a metal of Group I is present in a neutral or acid solution, it is fair to presume that no acid of Group II can be present, since the salts of Ag, Pb, and Hg^ with such acids are almost universally insoluble, either in water or acids. If, on the other hand, a metal of Group V be found present in a neutral solution, it is presumable that no acid of Group I will be present, since the combinations between metals of Group V and acids of Group I are all practically insoluble in water. 156 CHEMICAL ANALYSIS It will be seen, therefore, that a knowledge of the solubilities which are shown in Table I, p. 34, will save much time and labor by aiding in the interpretation of the results of the analysis for metals in the manner already shown, and by diminishing the number of acids for which individual tests must be made. In proceeding to the systematic examination for acids it is desirable to remove any heavy metals which may be present, since they are liable to obscure the reactions expected from the reagents for the acids. Accordingly, if the original substance is soluble or partly soluble, remove the heavy metals by boiling the solution with a small excess of NagCOg and filtering. If the substance is insoluble or partly insoluble, fuse the insoluble mass with Na2C03 in a platinum foil or crucible, boil with water, and filter. By metathesis all the heavy metals become carbonates, and the alkali metals form soluble salts with the acids. The filtrate from either method of double decomposition can now be analyzed for acids. Fusion decomposes HgCgO^, HgC^H^Og, HCIO3, and HC2H3O2 ; but salts of these acids are soluble in water or solvent acids, and must be sought for in the portion soluble without the aid of fusion. Of course it is necessary to test for HgCOg in the original substance before NagCOg is added. Neutralize a small portion of the solution of the alkali salts with dilute HNO3, and heat till all COg is expelled. Add BaClg. A precipitate indicates the presence of members of Group I. Divide a larger portion of the solution into three parts, and test for SYSTEMATIC PROCEDURE OF ANALYSIS 157 HgCrO^, H2SO3, and H^SiO^, as directed for Sub- group I, Group I, p. 138. If any members of Sub-group I are present, add dilute HCl to another portion of the alkali salts solu- tion, and pass H2S till the liquid smells of it. Boil off excess of HCl and HgS, and divide into six parts. Examine for H2SO4, H3P04,H3B03,H2C204, U^C^U^O^, and HF, as directed for Sub-group II, Group I, p. 139. To a small portion of the alkali salts solution add a solution of ZnSO^ in NaOH. If a white precipitate occurs, treat a larger portion of the solution in like manner, and filter, rejecting the white residue. Divide the filtrate, or a portion of the alkali salts solution, if H2S is absent, into three parts; and analyze for HCl, HBr, HI, HCN, H4Fe(CN)6, H3Fe(CN)6, and HCNS, as directed for Group II, p. 145. Divide another portion of the alkali salts solution into three parts, and test for HNO3, HCIO3, HC2H3O2, as directed for Group III, p. 150. 158 CHEMICAL ANALYSIS 1^1 l|.a o K en K»- ih I. s SI c5% a ^ w psW S « te =s W to ajO Si, is-dW S5 o ;3^ S8 c8 O W c3 «0 C^ V ^ 6 W WW SEPARATION OF METALS 159 o o-6te ■ 03 «s I? "o O e«5H 2 "te- • ^- .gig llg| no /-) "^ ra O O 1°. g -SSI. Residue (c): CoS, NiS Confirm Co with borax bead. Dissolve in aqua regia, expel CI, almost neutralize with NaoCOg, and add KCN. 'Confirm Ni with NaBrO. ^ .^^^-^ 13 2 ^.-SO -^^ 0^ 0^ 2 ©I'd OD ^ k„ ^ ^ ^H-( o 0) ^ « g-CM'd »3 o • a -id do oo od o i" 1-id do do dd dd dd ©© ^•1 °o ^-c-. a N CiC^ C1O0 qoi is-- 00l>. CJCO ^00 S8S S»& CI S-5 is- gC5 cod dd t^^ g"^ §- dd dd §«■ ©© dd II 1= as 00 iH l-HCO (MtH So" coo -^ t- So 'c^S io" S?00 qco si ^© IS-" i" g-^ d<6 ^-^ g-*" dd i::i<6 J?^ CO^ dd dd o ©'©' CO CO iH '3 4) a 2^. eo^§" [:| oc? £ii io' CO S^ g- deo ©o §- g^ 8^' dd dd dd dd o© dd fi^ c£ O W ^ c^- oo ■^00 sl ^?? ■*.'-i oqq 88 ^g] o"o" II 11 I©" li co"^ i r oo odd. ^- OO 9(6 cod dd ©d dd dd H S33 r3 £■2 ^ CO s tt-i s S CO coo So" io" 8o S'^ S2 8i 15" 3f2 Its S ^ co8 -f © 8S t3 o o a do do c>ci f]« 00 d cod dd dd ^-^ TfO dd i-!d TjJd J/} 1 5^ <0 t^ m S"5- -fia •^"^ s» si si si ^J2 8©" is- 8©" M t^ -1^ do oo dd i"" CO 00 j;o dd oo dd o© d© o© ©© 11 H 1^ pq li Sco t-co lom ^;:5 ^CO T(