UC-NRLF B 3 112 fiftf? TY1 ^ry^^sv^OL CLL/fi/T r * LIBRARY OF THE UNIVERSITY OF CALIFORNIA. MONOGRAPHS ON APPLIED ELECTROCHEMISTRY EDITED BY VIKTOR ENGELHARDT, Head Engineer and Chief Chemist of the Siemens & Halske A. G. , Vienna. WITH THE COOPERATION OF Dr. E. Abel, Chemist for the Siemens & Halske A. G., Vienna. E. G. Acheson, President of the International Acheson Graphite Co., Niagara Falls, N. Y. Dr. P. Askenasy, Superintendent of the Akkumulatorwerke, Liesing. H. Becker, Publisher of "L'Industrie electro-chimique, " Paris. Dr. W. Borchers, Professor at the Technical High School, Aachen. Sh. Cowper-Coles, Publisher of "The Electrochemist and Metallur- gist," London. Dr. F. Dieffenbach, Professor at the Technical High School, Darm- stadt. Dr. G. Erlwein, Chief Chemist of the Siemens & Halske A. G., Berlin. H. Friberg, Engineer of the Siemens & Halske, A. G., Berlin. H. Gall, Director of the Socie"te d'Electrochimie, Paris. F. E. Giinther, Mining Engineer, Aachen. Dr. F. Haber, Professor at the Technical High School, Karlsruhe. Dr. C. Ha'ussermann, Professor at the Technical High School, Stutt- gart. Dr. R. Hammerschmidt, Electrochemist, Charlottenburg. Dr. G. Hausdorff, Registered Chemist, Essen. Dr. K. Kellner, General Director, Vienna. A. Krakau, Professor of the Electrochemical Institute, St. Petersburg. Dr. H. Landolt, Director of the Society for Electrochemical Industry, Turgi. Dr. M. Le Blanc, Professor at the Technical High School, Karlsruhe. C. Liebenow, Engineer, Berlin. Dr. R. Lorenz, Professor at the Swiss Polytechnic, Zurich. Dr. R. Lucion, Director of Solvay & Co., Brussels. A. Minet, Publisher of " L'l^lectrochimie, " Paris. A. Nettel, Engineer, Berlin. H. Nissenson, Director of Akt.-Ges. of Stolberg & Westfalen, Stolberg. Dr. F. Peters, Instructor at the Royal Mining Academy, Berlin. Dr. W. Pfanhauser, Manufacturer, Vienna. Registered Chemist Dr. O. Prelinger, Chemist of the Siemens & Halske A. G., Vienna. Titus Ulke, M. E., Electrometallurgical and Mining Engineer of the Lake Superior Power Co., Sault Ste. Marie, Ontario. Dr. Th. Zettel, Chief Chemist of Brown-Boveri & Co,, Baden. And other experts. MONOGRAPHS ON APPLIED ELECTROCHEMISTRY. VOL III. THE PRODUCTION OF Chromium and Its Compounds BY THE AID OF THE Electric Current BY DR. MAX LE BLANC. PROFESSOR AND DIRECTOR OF THE PHYSICAL-CHEMICAL (ELECTROCHEMICAL) INSTITUTE OF THE TECHNICAL HIGH SCHOOL, KARLSRUHE. AUTHORIZED ENGLISH TRANSLATION BY JOSEPH W. RICHARDS, M.A., A.C., PH.D. PAST-PRESIDENT OF THE AMERICAN ELECTROCHEMICAL, SOCIETY. PROFESSOR OF METALLURGY AT LEHIGH UNIVERSITY. [^ or THE f UNIVERSITY PUBLISHED BY THE CHEMICAL PUBLISHING COMPANY, 1904. ~r~P 7 " Itni COPYRIGHT, 1904, BY THE CHEMICAL PUBLISHING Co. AUTHOR'S PREFACE TO THE GERMAN EDITION. It has been my endeavor to give in the present Monograph a complete review of what has been published up to the end of the year 1901, in relation to the manufacture of chromium and its compounds with the assistance of the electric current. To accomplish this I undertook a comprehensive study of the literature of the subject, in particular reviewing the German, English and American patent literature. I can scarcely hope to attain the end I had in view, since to get together all the observations which have been made concerning chromium, part of which are in places where they would scarcely be ex- pected, is indeed, at least for the single investigator, an almost impossible task ; I may, however, remark that I have taken special pains in this direction. I shall receive with thanks any information concerning items which I have overlooked and which are deserving of mention. The original publications quoted have been extracted in such manner that reference to the original for further par- ticulars is in most cases unnecessary. Further remarks on this subject are made on page 10. It will be to me a satisfaction if this work proves itself of service to my colleagues in the profession. I thank heartily Dr. J. Erode for his friendly assistance in reading the proof. M. LE BLANC. Karlsruhe. TRANSLATOR'S PREFACE. The study of these attempts to produce chromium and its compounds electrolytically will be profitable not only to the electrochemist particularly interested in chromium, but also to every active electrochemist whose work deals with difficult problems in electrolysis. I wish to acknowledge the assistance furnished by Walter S. Landis, Met. Eng., in the preparation of the manuscript and reading of proofs. JOSEPH W. RICHARDS. Lehigh University. .A O O CONTENTS. I. OBTAINING OF METALLIC CHROMIUM .... i A. By electrolysis of aqueous solutions . '. I B. By the use of high temperatures 32 II. THE OBTAINING OF COMPOUNDS OF CHRO- MIUM WITH METALS 53 A. By electrolysis of aqueous solutions 53 B. By the use of high temperatures 54 III. OBTAINING OF THE COMPOUNDS OF CHRO- MIUM WITH THE NON-METALS 66 A. Carbon compounds 66 B. Silicon compounds 67 C. Phosphorus compounds 71 D. Sulphur compounds 72 E. Oxygen compounds 73 (1) Chromous oxide compounds 73 (2) Chromic oxide compounds 76 (fl) BY ELECTROLYSIS OF AQUEOUS SOLUTIONS 76 (b) BY THE USE OF HIGH TEMPERATURES . . 8 1 (3) Chromic acid compound 86 (a) CHROMATES OF THE HEAVY METALS ... 86 () CHROMATES OF THE ALKALI METALS AND CHROMIC ACID 95 APPENDIX .... .120 I. OBTAINING OF METALLIC CHROMIUM. A. By Electrolysis of Aqueous Solutions. Nearly fifty years ago Bunsen, in his ramblirigs in the field electrochemistry, found a process for the manufacture of me- tallic chromium. 1 At that time the cheap production of chemical compounds on a large scale with the assistance of the electric current was essentially impracticable, and it has only been in the last fifteen years, in fact, that the previously unanticipated development of electro-technology has rendered it possible. I reproduce the most important principles of his manipulation, mostly in his own words, which show an aston- ishing clearness of view considering the time at which it was written, concerning the importance of current density in elec- trolytic operations. " Most important for the chemical action of the current is the density of the current, that is, the current strength divided by the electrode surface at which the electrolysis takes place. With this density increases the power of the current to over- come the chemical affinities. If one conducts a current, for in- stance, of constant strength through a cell of chromic chloride in water, it depends upon the area of the reducing electrode whether one gets hydrogen, chromic oxide, chromous oxide? or metallic chromium. The relative amounts of the constit- uents of the electrolyte through which the current passes are of no less importance. If one increases gradually, for instance, the amount of chromous chloride in solution, keep- ing constant the current strength and the electrode surfaces, a point is soon reached where the elimination of chromous oxide is accompanied by a reduction of the metal and finally is entirely replaced by it." He then describes further on the decomposing cell, one pole of which was the interior surface of a carbon crucible con- tained inside a porcelain crucible and filled with hydrochloric acid, and kept hot in a water-bath, while the other pole was 1 Poggendorff's Ann. 91, 619 (1854). 2 CHROMIUM AND ITS COMPOUNDS a narrow strip of platinum placed inside a clay cell in which the fluid to be decomposed was placed, and the whole stood inside the carbon crucible. This arrangement proved itself as very suitable for the obtaining of high current densities at one electrode, and he recommended it in general for the electrolytic deposition of metals, especially for the reduction of chromium from its aqueous solutions. " Using in such a reduction experiment chromous chloride solution containing some chromic chloride, metallic chromium can be easily obtained in continuous sheets of 50 sq. mm. in size, but quite brittle, the surface lying against the platinum electrode being perfectly white and of metallic lustre. The chromium, which can be obtained chemically pure only in this way, resembles iron very much in its external appear- ance, but it is more permanent in damp air and burns on being heated in air to chromic oxide. Hydrochloric and sulphuric acids dissolve it slowly to chromous salts with evolution of hydrogen. It is scarcely attacked by nitric acid even when boiling." Bunsen gives further details of the experiments in the fol- lowing sentences: "The reduction of the metal takes place therefore out of concentrated chlorous solutions heated to boiling, when each square millimeter of the reducing pole surface receives a current of the absolute intensity 0.067. z If the current density is gradually lowered a point is soon reached where the reduction of metal disappears, and is re- placed by a copious formation of anhydrous chromous-chromic oxide. This oxide can be made only in this manner and can be made in large quantities and is purified by long boiling with aqua regia. It is a very black non-crystalline powder, soluble in no acid, and burning in the air like pyrophoric iron, with lively deflagration, to a green chromic oxide. The composition of this chromous-chromic oxide varies between Cr 2 Cr and Cr 3 Cr" (Cr 4 O 5 and Cr s O 6 . See further under 1 0.67 ampere per square centimeter. r "^ sa*AAJA UA>V CA C/H* %V\l *" BY ELECTROLYSIS OF AQUEOUS SOLUTIONS. 3 chromic oxide)." Bunsen remarks finally that it would cer- tainly be of great interest to prove whether the metal reduced out of the green and the blue salts of chromium is identical in both cases, or u whether the allotropic conditions would make themselves evident also in the deposited metal." At present we recognize as the ground for the division between the green and the blue chromium solutions, the presence in them of different molecules or molecular complexes, and no longer speak of the allotropic form of the metal in these two solutions. 1 The data of Bunsen are on the one hand valuable for the general point of view which they take, and for the conditions which they impose for the separation of chromium, but one misses on the other hand exact numercial data ; his work has the appearance as if he had left over for a more comprehen- sive work the determination of the exact details, as is indi- cated by the following sentence: " I have so far determined the influences on which the decomposing power of the current depends, and have left over further investigations to my stu- dents." However, nothing further on the subject was pub- lished. The next investigations on the obtaining of chromium were a generation later ; they were directed towards the industrial manufacture of chromium and are part of patent literature. The names of Placet and Bonnet are in particular found at the heads of such patents which were concerned principally with the electrolysis of metallic compounds in general, and partly in particular with the electrolytic production of chro- mium. We will consider the contents of the first as far as it refers to chromium. The first patent of these two experimenters is concerned with improvements in electrolysis in general, and in particular with the electrolysis of metallic solutions. 2 It is characterized like others of their patents by its comprehensiveness, by its very 1 R. Whitney : Zeitschr. f. physik. Cheuiie, 2O, 40 (1896). 2 English Patent No. 19, 344, November 27, 1890. 4 CHROMIUM AND ITS COMPOUNDS. general point of view and by its endeavor to extend the limits of the patent as far as possible. They ascribe the poor results in the electrolytic separation of metals, up to their time, to the injurious formation of gas bubbles on the negative electrode. In order to prevent this they recommend keeping the bath under an increase or de- crease of pressure ; if this is done, then can all metals chro- mium is particularly named be obtained electrolytically with the greatest ease, of high purity and of any desired size and thickness, by simply using high current density ! Satis- factory results are also said to be obtained by taking two electrodes of different conductivity, for instance a copper wire for leading in the electricity to the cell and an iron, zinc or platinum electrode for the taking away of the current. Or two electrodes may be used of the same material but of different sizes. In place of the increasing of the external pressure similar good results may be obtained in many cases by increasing the density of the bath by dissolv- ing in it salts of more electropositive metals. The following salts are recommended for this purpose : i. Saturated solution of a mixture of potassium, sodium and ammonium sulphates ; 2. Saturated solution of a mixture of potassium, sodium and ammonium chlorides ; 3. Saturated solution of the sulphate or chloride of the metal to be deposited. A litre each of the three solutions is taken and mixed ; this mixture is a remarkable bath from which the metal can be precipitated pure in any desired thickness! In place of the saturated solutions, normal concentrated solutions may be taken, 120 to 150 grams of each of the three alkaline salts and 150 to 200 grams of the metallic salt to each litre of water. For the refining of metals a mixture is recommended, made by pouring together a litre of each of the saturated solutions of the three alkaline chlorides and three alkaline sulphates as well as water acidified with sulphuric or hydrochloric acids. BY ELECTROLYSIS OF AQUEOUS SOLUTIONS. 5 As electrodes, two plates of the metal to be refined are used. The alkaline sulphates and chlorides may be replaced by all other possible salts ; by alkaline nitrates and phosphates, magnesium chloride and nitrate, etc. According to the nature of the metal the baths should be alkaline, neutral or acid ; especially with chromium they should be strongly acid. For acidifying the baths all possible organic acids may be used besides the inorganic. The addition of acid influences favorably the purity of the deposited metal since any acci- dental suboxide will be dissolved. Warming' of the solution appears advantageous. Either soluble or insoluble anodes may be used, and it is recommended to use diaphragms in order to avoid polarizing action by gas bubbles and the contamination of the bath in consequence of insufficient purity of the anodes, the anodes being immersed in fluids which will combine with the oxygen set free there, for instance sulphuric acid, hydrochloric acid, sodium chloride solution, sulphurous acid, etc. When using insoluble anodes, polarization may be largely avoided by roughening their surface. The quality of the precipitate is improved by the addition (i to 2 grams to 10 litres) of all possible organic compounds, such as gelatine, gum, etc., or sulphur chloride, sodium sul- phide, iodine, arsenic acid, etc. To obtain alloys the following procedure is recommended : After making the bath of great density, for instance, the above-described mixture of the three alkaline sulphates or chlorides, there is added the salt of the different metals which one wishes to alloy, or soluble anodes of the metals concerned are used. Then a feeble current is passed through the solu- tion and gradually increased. Two cases are now possible. If the bath is acid, that metal which is the most electronega- tive is deposited first ; if the bath is alkaline, that metal sepa- rates out first which is the most electropositive ! Later, it is stated, that it is only necessary to increase the strength of the current in order to obtain an alloy consisting of all of the 6 CHROMIUM AND ITS COMPOUNDS. metals contained in the bath. The relative quantities of metal in the alloy may be varied by altering the relative concentra- tions or the surfaces of the soluble metallic anodes. Operating in this way very malleable alloys of zinc, tin or copper with aluminium, chromium, tungsten, and molybdenum have been obtained, also very hard and coherent deposits of chromium with tungsten, iron and molybdenum, and very light alloys of aluminium with a small quantity of chromium,, tungsten, iron and molybdenum. If the ores are sufficiently good conductors of electricity they may be used directly as anodes. In other cases the valu- able metals must be dissolved and can then be obtained in the v pure condition from the impure salts by the methods de- scribed. We reproduce here only those patent claims which relate directly or indirectly to chromium : (1) A method of strengthening the action of the current in electrolysis consisting in subjecting the hot or cold electrolyte to a determined pressure, obtained either by physical means r gas pressure or fluid pressure, or by a greater density of the electrolyte obtained by dissolving therein indifferent material not decomposed by the current, or by using conductors of various cross-section or various conductivities, which hinder the current in its passage and so produce an artifical pressure. (2) An exerting of pressure upon the hot or cold electro- lytic bath containing the compound of the metal to be depos- ited, or in which the anode of the metal concerned is placed. This pressure is obtained by physical means (gas or fluid pressure), and at the same time currents of comparatively high intensity are used for the production of the pure metal. (3) The exerting of pressure upon the hot or cold electro- lytic bath containing different compounds of two, three or more metals or in which anodes formed of these metals are dipped. This pressure is produced by physical means, and such currents are utilized at the same time whose electromo- tive force and intensity can be so regulated that the precipitate BY ELECTROLYSIS OF AQUEOUS SOLUTIONS. 7 of those metals is obtained of which the alloy should consist. (4) The use of anodes of varying section and different con- ductivity as the means for the production of the artificial cur- rent pressure, using at the same time a current of relatively high intensity, the electrolytic bath being either hot or cold and containing the metal or metals to be deposited. (5) The use of hot or cold electrolytic baths whose density and consequently the pressure is increased by the solution of indifferent salts as sulphates, sulphites, phosphates, or oxalates of the alkalies or alkaline earths and, in fact, by all com- pounds of all bases which are more electropositive or more electronegative than the metal to be deposited according as to whether the bath is acid or alkaline, and which contain the compounds of the metal to be deposited, while pure or impure soluble anodes of the metals contained in the bath are used, using at the same time currents of relatively high intensity. (6) The use of cold or hot electrolytic baths whose density is increased by saturating them with (a) the three alkaline sulphates, (ft) the three alkaline chlorides, (c) a sulphate or a chloride or a soluble compound of the metal to be deposited, using at the same time currents of relatively high intensity. (7) The use of cold or warm electrolytic baths whose den- sity may be increased by preparing them from saturated solu- tions of (a) the three alkaline sulphates, (b) the three alkaline chlorides, in which are placed soluble anodes of the metals to be deposited, using at the same time currents of relatively high intensity. (8) The use of electrolytic baths whose density is increased by making them from a saturated solution of the three alka- line chlorides, or a saturated solution of the three alkaline sulphates and to one of these solutions a saturated solution of the chloride or sulphate of any other soluble salt of the metal to be deposited is added, or by using smooth or rough anodes of the metal concerned, using at the same time currents of rela- tively high intensity. 8 CHROMIUM AND ITS COMPOUNDS. (9) The use of electrolytic baths whose density is increased as described above and which contain insoluble anodes placed in porous vessels, which latter contain for the avoidance of polarization one of the following solutions : Sulphuric acid, hydrochloric acid, sodium chloride, calcium chloride, sul- phurous acid or sulphites, using at the same time currents of relatively high intensity. (10) The use of electrolytic baths whose density is increased as before described and in which soluble, roughened anodes are placed in porous vessels for the avoidance of polarization and the arrangement of which is shown in the accompanying plate, for the purpose of avoiding the contamination of the bath by the impurities of these anodes, using at the same time currents of relatively high intensity. (n) The use of electrolytic baths whose density is increased as above described in which roughened soluble or insoluble anodes are contained in porous vessels whose arrangement is shown by the accompanying plate and which contains a de- polarizing fluid in case insoluble anodes are used. The de- polarization fluid or the fluid to be decomposed or both should be automatically renewed, using at the same time currents of relatively high intensity. (12) The use of electrolytic baths whose density is in- creased as above described in which a soluble zinc anode is placed in a porous vessel containing oxalic acid or an alkaline oxalate in order to hinder the zinc salt formed from passing into the bath which is being decomposed, using at the same time currents of relatively high intensity. (13) The use of electrolytic baths whose density is in- creased as before described, into which a soluble or insoluble anode dips and in which, for the improvement of the quality of the metal, there is added per litre of bath i to 2 grams of gelatine, gum, dextrin or similar material, using at the same time currents of relatively high intensity. (16) The use of electrolytic baths whose density is in- creased as before described and into which dip soluble anodes BY ELECTROLYSIS OF AQUEOUS SOLUTIONS. 9 (ores when of sufficient conductivity may be used as anodes) or in which compounds of the metals to be deposited as alloys are dissolved (these compounds may be the ores of such metals if sufficiently soluble). The surfaces of the anodes and the relative quantities in which the compounds are dis- solved vary according to the proportions in which the metals are desired in the alloys, using at the same time a current whose electromotive force and intensity may be so regulated that a precipitate of the metals is obtained of which the alloy .should consist. (17) The alloys: (a) of aluminium, zinc, tin or copper, of chromium, zinc, tin or copper, of tungsten, zinc, tin or copper, of molybdenum, zinc, tin or copper, etc.; (d) of molybdenum and silver, of uranium and silver ; (c) of chromium and tungsten, of iron and chromium, of nickel and chromium, of nickel and molybdenum, etc.; (d) of aluminium and chromium, of alu- minium and tungsten, of aluminium and iron, of aluminium and molybdenum, etc. These alloys can be alloyed with each other. (18) The use of electrolytic baths whose density is increased as before described and in which good conducting ores are used as soluble anodes, using at the same time currents of rel- atively high intensity suitable for precipitating the metal. (19) The use of electrolytic baths whose density is in- creased as before described in which poorly conducting ores are dissolved by being previously transformed into soluble sulphates, chlorides, etc., using at the same time a current of relatively high intensity suitable for precipitating the metal. (20) The use of electrolytic baths whose density is in- creased as before described, in which the compounds of valu- able metals are dissolved, in case these are soluble or trans- formable into soluble compounds, using at the same time currents whose density may be progressively altered in order to precipitate successively the single metals forming the alloys. (21) The use of electrolytic baths whose density is in- creased as before described and in which the alloys of valuable 10 CHROMIUM AND ITS COMPOUNDS. metals dip, if these alloys can serve as soluble anodes, using- at the same time a current whose intensity may be progres- sively altered in order to be able to precipitate successively the single metals forming the alloy. (22) The use of several electrolytic baths in series whose density is increased, or which are subjected to an increased pressure as before described, and in which one or more me- tallic compounds are dissolved or in which one or more solu- ble anodes are placed, according to whether a single metal or alloy is to be produced, using at the same time currents of relatively high intensity, whose intensity and electromotive force may be altered in the case that alloys are to be manu- factured. It is superfluous to criticize the above patent claims since every one, who has only a small knowledge of the subject, rec- ognizes at once that these are not the outcome of actual ex- periments, but are at least for the most part mere fantasies. The further patents of Placet and Bonnet contain similar fantasies, and they are to be found also in the patents of other inventors. An explanation of this lies in the fact that the patent specifications are often written out before the comple- tion and at times even before the beginning of tedious investi- gations, and on this account contain statements which have in no way been controlled by experiment. On this account a certain caution is to be recommended concerning statements in patents. The question may be quite prudently asked as to whether it is worth while to reproduce the contents of such fantastic patents in this monograph. I have answered the question in the affirmative after considerable reflection and acted accord- ingly, for on the one hand the limit separating works worth quoting from others not worth quoting is not always easy to draw and might be decided differently according to the person making the decision, and as many times a small grain of gold may be hidden under much rubbish, and yet in truth serve as a crystallizing point for further investigations and also, however. BY ELECTROLYSIS OF AQUEOUS SOLUTIONS. II l>e highly detrimental to the claims of later investigators. Any one who has had much to do with the taking out of pat- ents has probably had the experience that the remarks of many patentees are so general that they frequently cover more particulars than was intended, and their generalities frequently include in their terms facts which were completely unknown to the patentee, but such generalities are frequently of great hindrance and burdensome to applicants for new pat- ents ; naturally in the second case an exact knowledge of the literature would be of considerable use. This monograph endeavors, in contrast with the yearly re- ports, etc., to give a complete compilation of the subject of such extensiveness as shall make superfluous the necessity of referring to the frequently difficultly accessible original litera- ture. We wish first to find our bearings in the mass of ma- terial to be found in this field, valuable and valueless. On these grounds I did not deem a selection admissible. A second patent 1 of Placet and Bonnet is concerned directly with the electrolytic production of chromium. The inventors state that it is necessary in manufacturing chromium and chromium alloys electrolytically to keep the content of chro- mium in the bath constant during the whole duration of the process. The electrical conductivity of the bath is increased by adding sulphates or chlorides of the alkalies, or alkaline earths; the addition of organic materials like gum and dex- trin is also useful. Regarding the concentration of the chromium salt solution itself, they recommend to use only about one-fifth of the quan- tity of salt necessary to make a completely saturated solution; the solution is preferably to be warmed. The nature of the metallic precipitate is favorably in- fluenced by making the surface of the anode larger than that of the cathode. As anodes, soluble and insoluble materials may be used, but it is, however, always to be recommended that in order to avoid the contamination of the baths and the 1 German Patent No. 66,099, December 5, 1890. 12 CHROMIUM AND ITS COMPOUNDS. polarization resulting from gas bubbles, diaphragms should be used and to place in the anode compartment such fluids as will combine with the oxygen there liberated ; for instance sulphuric acid, hydrochloric acid, sodium chloride solution, sulphurous acid, etc. To keep the chromium content of the bath constant, solid chromium salt is to be added from time to time. A tension of 30 to 40 volts (!) is recommended as particu- larly advantageous for operating the bath. If alloys are to be directly precipitated the salt of the metal to be alloyed is dissolved in the bath just described or the metal to be alloyed is used as a soluble anode in a one-half or three-quarter saturated solution of the alkalies or the alkaline earths and the current used at that tension " which corresponds to the composition of the alloy." For example, if it is wished to alloy iron with chromium it is well to add about one-quarter of the quantity of iron salt to the solution that would be necessary to saturate it, and about one-fifth of that quantity of chromium salt which would like- wise saturate the liquid in the absence of any other salt. To the solution thus formed is added as much alkali or alkaline earth salt as it will take up. " The alloy can either be directly extracted from such a bath or the two metals may be deposited consecutively, but one upon the other and subsequently melted together to the alloy." In both cases the composition of the alloy may be regulated, " in the first case by using a certain determined tension, in the latter by the duration of the action of the current on the bath. At 6 volts tension only iron separates out, at 30 volts only chromium." By using intermediate ten- sions, alloys of any composition desired may be obtained. The latter can also be obtained by precipitating first the iron with a tension of 6 volts, then the chromium at a tension of 30 and finally melting the precipitated metals together. Patent Claim. The electrolytic obtaining of chromium and its alloys, BY ELECTROLYSIS OF AQUEOUS SOLUTIONS. 13 characterized by adding to the solution of chromium salt in about five times the volume of water necessary to dissolve it, materials which will not chemically injure the metal to be precipitated, for instance, alkali or alkaline earth sulphates, chlorides, or similar salts alone or in combination with or- ganic materials, gum, dextrin or similar materials, such ma- terials being added to saturation and the mixture electrolyzed hot or cold, keeping the chromium content of the bath as con- stant as possible. The next patent 1 of the same inventors is entitled : " Im- provements concerning the electrolysis of metals." They refer to their first English patent No. 19,344 of 1890 (page 3) and now propose to avoid the formation of oxide at the cathode by the use of acid salts (bisulphates and phosphates, acetates, etc.) in place of acids. It is particularly difficult to find the correct amounts of acid to add in depositing chromium from the solutions containing 10-15 grams of chromium alum, 10- 15 grams of sodium, potassium or ammonium sulphates or a mixture of these sulphates in 100 grams of water, whereas, on the other hand, they obtain at once a particularly fine chro- mium deposit by using bisulphates in place of the sulphates, as for instance 10-15 grams of bisulphate to 10-15 grams of chromium alum in 100 grams of water. In case the bisul- phates contain free acid the corresponding quantity of normal sulphate is to be mixed with it. Besides chromium, chromium alloys as well as other metals may be precipitated from the corresponding solutions. The patent claim reads : The use of strong polybasic salts alone or mixed with neutral salts in order to obtain electro- lytic precipitates of chromium, aluminium, copper, iron, nickel, cobalt, tungsten, molybdenum, antimony, tin, silver, etc., and their alloys in the above-described way. In two further patents taken out at the same time, Placet and Bonnet protect several modifications of their process for making chromium. 1 English Patent 22,854, Dec. 31, 1891. 14 CHROMIUM AND ITS COMPOUNDS. The first patent 1 bears the title: " A method for the precipi- tation of chromium with the assistance of electrolytic baths containing chromium salts. The bath used is again given the composition of 10, 15, to 20 grams of chromium sulphate with 100 grams of water to which besides is added so much sulphuric acid that no forma- tion of oxide can take place at the negative electrode. In place of chromium sulphate may also be used chrome alum, in place of sulphuric acid, either nitric, hydrochloric or hydro- fluoric and organic acids or a mixture of the same. First- class precipitates are said to have been obtained for instance by using a bath of 10 to 15 grams of chrome alum, 10 to 15 grams of sodium or potassium sulphate, 5 grams oxalic acid or some other organic acid and 100 grams of water, heated until the originally green solution has become violet and then electrolyzed. The solution of alkaline chromate or bichromate to which some sulphate is added may also be used, and for the accelera- tion of the reaction the addition of a little alcohol is recom- mended. The chromium salt preferably used is the sulphate or a chrome alum in about the following proportions : 10 to 15 grams of chromate or bichromate with 10 to 15 grams of chrome alum and 100 grams of water. In place of the chromate the more costly chromic acid may be used and in place of the chrome alum other chromium salts such as the nitrate. Alkaline chromium baths likewise furnish chromium when electrolyzed, even if the chromium solution gives a precipitate with alkali ; the same is true of baths containing cyanides or sulphocyanides of the alkalies, and finally chromium may be electrolytically precipitated from solutions of the roseo-chrome salts or ammoniacal chrome compounds. For raising the conductivity there may be added to any of these baths 10 to 15 per cent, of alkali sulphate or ammonium sulphate or a mixture of both or alkali nitrates, phosphates, 1 English Patent 22,855, Dec. 31. 1891. U. S. Patent 526,114. BY ELECTROLYSIS OF AQUEOUS SOLUTIONS. 15 etc., so that the bath may consist of, foi instance, 10 to 15 grams of chrome alum or chromium fluoride, 10 to 15 grams of sodium, potassium or ammonium fluorides, 5 to 10 grams of hydrofluoric acid or another acid. In case a heavy precipitate is desired in a short time the bath must be heated to higher temperatures. " In many cases we heat our baths so high that the salts are melted v (see later). The composition of the solution is kept constant either by using soluble anodes or by circulation. The addition of small quantities of alcohol, sugar, glycerin or analogous organic compounds exercise a favorable influence upon the qualities of the precipitated metal. Finally, alloys can be made by using the baths described and adding the salt of the other metal. An alloy may also be obtained by the alternate precipitation of chromium and the other metal and subsequent melting down. The patent claim reads : The use of an electrolytic bath containing chromium salts in the manner and for the purpose described. The second patent 1 has the title : A method for the pre- cipitating of chromium by the use of electrolytic baths con- taining chromic acid salts. The baths described therein need to contain only i to 2 grams of chromic acid in 100 grams of water ; however, with increasing concentration the quantity of deposit increases. The shade of the precipitated metal alters if small quantities 5 to 10 grams per litre of the following acids are added : Phosphoric, sulphuric, oxalic, benzoic, formic, gallic, pyrogallic, picric, purpuric (?) (phenic), salicylic ; all such solutions give white chromium deposits. Acetic and hydrofluosilicic acids give shining but dark de- posits, boric acid dead but very copious deposits, while by adding sulphuric, nitric, hydrochloric, arsenic, tartaric, citric, lactic or tannic acids the metal is more or less dark according to the quantity added. It may be noticed also that the chromic acid necessary for the baths may be manufactured by the electro- 1 English Patent 22,856, Dec. 31, 1901. 1 6 CHROMIUM AND ITS COMPOUNDS. lytic oxidation of chromic salts ; also that alloys of chromium and other metals may be obtained from solutions containing chromic acid and other metals. The claim reads : The use of electrolytic baths containing chromic acid in the manner and for the purpose described. In an article 1 published in 1892, "The Manufacture of Metallic Chromium by Electrolysis," E. Placet presents nothing new. He only repeats that a beautiful lustrous deposit of pure chromium was obtained by electrolyzing an aqueous so- lution of chrome alum to which alkaline sulphates and a small quantity of sulphuric or other acid has been added. Also chromium coatings on bronze, brass, copper and even iron of any desired thickness and resembling oxidized silver can be obtained in this manner. There was shown at a meeting of the Academy, when this article was read, a piece of metallic chromium weighing more than a kilogram, chro- mium alloys and brass accoutrements electrolytically coated with chromium. In reference to the qualities of the metal so obtained, which was then being industrially produced, it was mentioned that it was of a bluish white color, very hard and extraordinarily resistant alike to atmospheric influences as also to concentrated sulphuric acid, nitric acid and caustic potash solution. If the baths are electrolyzed under certain conditions, groups of chromium crystals are obtained which have the form of pine tree branches. The fruitfulness of the two inventors E. Placet and J. Bon- net is not exhausted by the five patents already described ; two years later we run across a new patent 2 worded thus " Im- provements in the electrolytic manufacture of chromium and chromium alloys." They commence with one of the observations made in a former patent (see page 15) according to which they heat the baths in some cases so high that the salts are melted, and give us the following additional information thereon : 1 Comptes rendus, 115, 945 (1892). 2 English Patent 6,751, March 30, 1893. BY ELECTROLYSIS OF AQUEOUS SOLUTIONS. 1 7 We take, they say, for instance 10 to 15 grams of potassium bisulphate, 100 grams of chrome alum and 100 grams of water, bring the salts into solution by heating and electrolyze ; chromium precipitates at once upon the cathode. The con- centration of the bath is kept up by the addition of chromium alum or a concentrated solution of chromium alum and alkali bisulphate ; by doing this the volume increases and on this ac- count an overflow is provided to keep the level of the bath constant. In case only chromium alum is added all the water originally in the solution is soon used up and the bath con- sists only of the salts " reduced either by water or by heat(?)." According to that, water can be dispensed with for the for- mation of the bath and the latter be formed by directly melt- ing down the mixture of salts. If that is so when and where could we add to the bath a little potassium chlorate, boric acid, benzoic acid or analogous material(!) in order to facilitate the precipitation of the metal. In place of the potassium bi- sulphate there may be used as a flux either sodium or am- monium sulphate or bisulphates or the phosphates, borates, chlorates, silicates, chlorides, fluorides, etc., of the alkalies or alkaline earths or mixtures of the same ; in place of the chrome alum also chromic oxide and all other chromium salts which melt easily when mixed with the alkaline earths. On account of the high melting-point of chromium we usually use an electric furnace and in place of pure carbon electrodes we use electrodes made of a mixture of carbon, chromic oxide, chromic salts or ores, a flux as borax, etc., and a reducing agent (like zinc, magnesium, aluminium, etc.) so that the metal is simultaneously reduced and melted. If chromium alloys are to be immediately produced the electrodes may be made of the alloying metal (copper, aluminium, silver, nickel, zinc) or of their oxides. It is finally recommended to introduce into the melting crucible indifferent or reducing gases, as hydrogen, etc., and as the simplest manner of introduction through hollow elec- trodes. 1 8 CHROMIUM AND ITS COMPOUNDS. The patent claim reads : i. The above-described improve- ments in the obtaining of metallic deposits of chromium of any thickness and in considerable quantities for commercial purposes. 2. The above-described improvements in chro- mium coating various articles, among others of carbon fila- ments and rods such as are used in electric lighting. 3. The above-described improvements for the direct electrolytic pro- duction of alloys of chromium and other metals. My opinion of the contents of this patent of Placet and Bonnet is a repetition of what I have said on page 10. In order to give my judgment some experimental justification I have had Mr. Shick produce and electrolyze in my laboratory some of the baths recommended in the patent specifications. I quote some extracts from the report of Mr. Shick : A well-saturated solution of chromic sulphate at the tem- perature of the room, was used and 100 cc. diluted to 600 cc. with water and then sodium chloride added to saturation. A platinum foil was used in each case as cathode. With 40 sq. cm. active cathode surface, using a current density of 0.2 ampere per sq. cm. there was obtained a quite small black precipitate which from its behavior appeared to be chro- mium. With a current density of 0.3 ampere per sq. cm. no precipitate was obtained. A precipitate .did not appear also when the above bath was saturated with sodium sulphate instead of sodium chloride and electrolyzed at 30 and 80 with a current of 0.2 and 0.3 ampere per sq. cm. Another bath was prepared with 15 grains of chromium alum and 15 grams of potassium bisulphate dissolved in 100 grams of water. Also in this case there was obtained with the above current densities and varying temperatures only at the most traces of chromium deposits. Other baths gave no better results. It is sufficient therefore to merely mention the baths used and to notice that in all cases good circulation was obtained by active motion of the cathode. BY ELECTROLYSIS OF AQUEOUS SOLUTIONS. 19 Bath III. Fifteen grams of chrome alum, 15 grams of am- monium fluoride, 10 grams of concentrated hydrochloric acid and 100 grams of water. Bath IV. Fifteen grams of chrome alum., 15 grams of bi- sulphate, 100 grams of water. These statements agree entirely with the results of other investigators who likewise could obtain no chromium deposits in the manner described. 1 According to that there is no longer any doubt that the kilogram of chromium (page 16) exhib- ited at the French Academy by Placet could not have been obtained in the manner stated or that at least conditions of the highest importance for the obtaining of good chromium deposits were not then stated by him. While the processes so far described attempt to obtain me- tallic chromium, and the accessory phenomena of the process were only very slightly or not at all noticed, the patent of the Electro-Metallurgical Company, Limited, 2 about to be de- scribed, lays particular weight upon these phenomena. The patentees start from the undotibtably correct point of view that whoever wishes to reach an economical operation must keep in view two other points besides the consumption of electrical energy, points which have an important influence upon the cost of the metal ; namely, the anodes to be used and the nature and the quantity of the products resulting. Since platinum electrodes are much too expensive to be used for the production of chromium it is preferable to use chromium sulphate solutions for the electrolysis, which permits of the use of lead anodes. The latter are coated over with a layer of peroxide which protects them for a long time from destruc- tion, provided that the sulphuric acid is no more concentrated than is usually employed in lead accumulators. Care must be taken on this point, if a diaphragm is used and sulphuric 1 Cowper-Coles : Chemical News, 81, 16 (1900) ; J. Fe"re*e : Bull. Soc. Chim., [3] 23 1 , 617 (1901). 2 German Patent 105,847, Sept. 7, 1898. The corresponding English Patent No. 18,743, f r 1898, is under the name of Moller and Street. 2O CHROMIUM AND ITS COMPOUNDS. acid is the anode fluid, to bear in mind that the latter becomes more concentrated during electrolysis and must therefore be diluted in a suitable manner. If no diaphragms are used the content in sulphuric acid must in that case be kept low in order to get a precipitate of chromium at all, and therefore there is no need of giving precautions for the preservation of the electrode. The second object, the obtaining of by-products which will have a certain marketable value and at the same time to keep their quantities down to the minimum is obtained by the patentees in the following somewhat complicated manner. Imagine, for instance, an electrolytic apparatus provided with a diaphragm, the anode solution consisting of water acidulated with 5 per cent, of its weight of concentrated sul- phuric acid of 66 B., while the cathode fluid contains to each 1.5 kilograms of water, 1.5 kilograms of sodium-chrome alum and 1.5 kilograms of sodium sulphate. If this is heated to about 90 and electroly zed with a current density of 0.4 ampere per sq. cm. there is obtained 40 grams of chromium per ampere hour. 1 As the electrolysis proceeds the sulphuric acid in the anode compartment becomes more concentrated, as already de- scribed, and also sodium sulphate and chromic sulphate pass into it, the latter being oxidized at the anode to chromic acid. An analysis showed that during the obtaining of 23.5 kilograms of metallic chromium 0.562 kilogram of chromic acid and 0.936 kilogram of sodium sulphate passed into the anode compartment. Since, as may be seen from the formula of chromic sulphate, 294 grams of sulphuric acid are set free at the electrode for 104 grams of chromium precipitated, a works which manufactures a ton of chromium in twenty-four hours would obtain nearly three tons of sulphuric acid of 66 B., but in a diluted condi- tion and mixed with the other by-products mentioned and therefore in a form of little commercial value. In order to 1 Apparently this is a mistake ; in the English patent it is given as 0.2 gram, which Jwould be about 30 per cent, of the theoretical amount. BY ELECTROLYSIS OF AQUEOUS SOLUTIONS. 21 || avoid this and other difficulties the process was operated as follows : The anode liquid after it has become somewhat rich in chromic acid, etc., is treated with gaseous sulphur dioxide whereby the chromic acid is reduced to chromic sulphate ; the fluid is then concentrated to 30 B., by doing which the excess of sulphur dioxide is driven off. Then sodium bi- chromate is added in such quantities that independently of the sodium sulphate formed 2CrO 3 is present for every two equiva- lents of sulphuric acid, 1 and allows this mixture to flow in a more or less thin stream into a lead vessel which contains always an excess of melted sulphur kept at 130. The chromic acid oxidizes the sulphur, and chromic sulphate is formed by the two equivalents of free sulphuric acid alluded to. This with the sodium sulphate formed regenerates the alum. By treating with some water and filtering from the excess of sulphur there is obtained a solution of chromic alum (and sodium sulphate), which can be led into the elec. trolyzing apparatus if necessary. The first treatment with sulphurous acid can be omitted by taking account of the chromic acid present when adding the bichromate. Considering the reactions in the cathode compartment there is naturally a much greater depletion of the chromic sulphate, while the content of sodium sulphate is only slightly altered. It is clear that under such conditions with a cathode solution saturated with sodium chrome alum the quantity of sodium sulphate soon reaches such a point that the operation of the apparatus is interrupted. This interruption in fact occurs very easily as was empha- sized since the content of free acid in the solution is carefully regulated in order to obtain a good output and the quantity of 1 The explanation is not very clear ; the following formula according to which the oxidation apparently proceeds gives the relations between the chromic acid and the sulphuric acid : Na 2 Cr 2 7 + 3H 2 SO 4 + S -= Cr. 2 (SO 4 ) 3 -f Na 2 SO 4 + sH 2 O. 22 CHROMIUM AND ITS COMPOUNDS. p sodium chrome alum and the accessory salts (in the foregoing case sodium sulphate) must stand to each other in the right proportions. If this is not the case, the following occurrences take place : No precipitate of chromium comes down, the bath is too acid. It is then electrolyzed some time, whereby the content in acid diminishes of itself or it is neutralized by means of chro- mium hydroxide or any other base or a carbonate. There results chromic acid at the cathode, a sign that the addition of acid is necessary. There occurs also the necessity of removing the sodium sulphate in excess, which may be most simply done by cooling the hot cathode liquor and crystallizing out the sodium sul- phate. It may be seen that continuous working with a utilization of the by-products may be achieved in this way. If wished, other changes may be introduced. Instead of supplying the apparatus with chromium alum, chromic sulphate may be used in which case the sodium sulphate is superfluous. Further, the acid produced in the anode compartment can be neutral- ized by chromic hydroxide and later the whole anode fluid, if a sufficient migration of chromium has taken place, be brought over to the cathode side. In that case it is previously neces- sary to reduce the chromic acid formed by sulphurous acid and neutralize further with chromic hydroxide. Or the whole of the chromic hydroxide added may be supplied after the re- duction of the chromic acid. Finally it is remarked that the process may also be carried out without diaphragms, but with more difficulty, and that al- kali, earth alkali or chromic sulphides are proposed for the reduction of the chromic acid. The patent claims read : i. Processes for the obtaining of chromium by electrolysis of salts containing chromic sulphate, using lead anodes, and characterized by the reduction of the fluid leaving the anode R /* OF THt UNIVERSITY BY ELECTROLYSIS OF AQUEOUS SOLUTIONS. 23 compartment by sulphurous acid and further treatment with melted sulphur, whereby chrome alum is formed when sodium chromate is added, and the chrome alum being then conducted to the cathode compartment, the cathode fluid being circulated and its temperature kept low in order to avoid its enrichment in sodium sulphate. 2. A method of carrying out the process of Claim i, char- acterized by saturating the fluid coming from the anode com- partment, after its reduction by sulphurous acid, with chromium hydroxide, and thereupon using it for supplying the cathode compartment. After working one's way through this patent specification one has not the feeling that it is a process proved through practical investigations, and which has been found profitable. At this place we may also mention the process of the same company Electro-Metallurgical Company intended to ren- der possible exact control of electrolytic metal depositions and especially chromium deposition. 1 The company has found out in their investigations (see also Bunsen, page i) that using a chromium bath with sulphate salts of known composition and keeping the current strength constant the alteration of the amount of cathode surface changes the nature of the pre- cipitate. " There is obtained, for instance, from the same bath and with the same current strength a precipitate of oxide if a rod 8 mm. diameter is used, but with a rod of 15 mm. diame- ter a fine metallic precipitate which increases in thickness with a good efficiency of deposition. With a rod 18 mm. in diameter a precipitate is obtained which does not increase in thickness, and finally with a rod of 20 mm. in diameter no precipitate is obtained. These figures are only given by way of example ; according to the composition of the bath and its content of free acid, they are quite different." For a given bath which with a certain current strength gives a good precipitate on a rod 15 mm. thick, there may be, therefore, easily found the diameter of a rod which will give 1 German Patent No. 104,793, Sept. 7, 1898. 24 CHROMIUM AND ITS COMPOUNDS. in the same bath a chromium precipitate which does not in- crease in thickness. As long as the composition of the solu- tion suffers no material change, the chromium precipitate upon the first rod will increase in thickness as is desired, while the weight of the second rod remains unchanged. The reverse is true, that an alteration in the weight of the test rod would be a sign that the solution had changed, and that a chromium precipitate upon the first rod was not coming down in the properly prescribed manner. In order then that the test rod which should remain of con- stant weight may act as an automatic signal of the current it is hung from the arm of a balance whose other arm is fur- nished with contacts which ring an electric bell as soon as the test rod loses or gains in weight. To make the bell stop ring- ing and bring the electrolyte again into proper condition the bath must be neutralized if the weight of the rod has decreased, or acidified if it has increased. " The process permits of finding out with exactness the condition of the electrolyte and thereby regulating properly the operation of the apparatus." The patent claim reads : A process for the control of electrolytic depositions of metals characterized by the use of a test rod which is connected in parallel in the precipitating bath, the dimensions of the rod being so chosen that with a proper current density and with normal content of acid the test rod is covered with only a light deposit as may be determined by removing it from the bath or by hanging it from the arm of a balance. In the latter case the other arm of the balance should be connected with an electric bell whereby the increase of acid content of the bath or the diminution of metal in the bath on the one hand, or the decrease of the content in acid or the increase of metal precipitated on the test rod on the other hand, causes a movement of the arm of the balance and the closing of the circuit through the electric bell. The series of patents concerning the electrolytic precipita- BY ELECTROLYSIS OF AQUEOUS SOLUTIONS. 25 tion of chromium from aqueous solutions is herewith ex- hausted. We have to add to them simply two scientific articles on this subject. J. Feree 1 states that he has obtained steel-gray chromium 99.82 per cent, pure by using a platinum cathode and electro- lyzing a solution of 740 grams of water, 100 grams of hydro- chloric acid and 160 grams of crystallized chromic chloride (see also under chromium amalgam.) The current density used is 0.15 ampere per sq. cm., tension 8 volts. He obtained a silver-white precipitate of finer appearance from a solution of chromium-potassium chloride; 266.5 grams CrCl 3 -f 6H 2 O (i molecule) and 223.5 grams KC1 (3 molecules) were dissolved in one litre of water and the cur- rent density and tension used were as above. With 15 sq. cm. cathode surface the weight of the precipitate was 3 grams corresponding to an ampere-efficiency of 45 per cent. The author complains that often processes for the electro- lytic precipitation of chromium are described without exact data being given (compare the foregoing chapter), and states that he has tried in vain to get chromium from a solution of chromium sulphate acidulated with sulphuric acid and satu- rated with alkaline sulphates, as Placet claimed to have done (see page 18). The electrolytically obtained metal is very hard and un- changeable at ordinary temperature. At a red heat, it covers itself with a light layer of green chromic oxide. It is unattacked by either concentrated sulphuric or nitric acid, and is not acted upon by a concentrated solution of caustic potash. Some further data on electrolytically-precipitated chromium has been given by Co wper- Coles. 2 According to this writer, chromium precipitated upon copper can be scratched by the diamond point only when it is weighted with 0.07 gram. Electroplated nickel requires, for instance, o.oi gram, elec- 1 Bull. Soc. Chim. [3], 25, 617 (1901). * Chem. News, 81, 16 (1900). 26 CHROMIUM AND ITS COMPOUNDS. troplated palladium 0.066 gram, a brilliant palladium precip- itate 0.09 gram. Fused chromium is still harder. The capability of reflecting light is soon diminished in the case of chromium by tarnish, which accumulates under the same conditions much stronger than with silver, platinum or palladium. In the finely polished condition its capability of reflection is equal to that of silver. When making this communication Cowper-Coles stated that he had made experiments on the production of chromium alloys electrolytically and had taken out a patent for the same in March, 1884, but I have not been able to discover the same. He manufactured chromium alloys by heating chromium com- pounds with charcoal in a closed crucible and pouring upon the reduced mass 2 J^ parts of copper and then i to i J^ parts of melted tin ; the whole was then granulated, again melted and cast into the desired form. " The plates thus formed were used as anodes in a solution which is made by dissolving i pound of potassium cyanide and i pound of ammonium car- bonate in a gallon of water at 53 C. until a good precipitate was obtained on the cathode, using a low current density." It must be granted that these data are somewhat hazy. Later he reports on several new investigations. He ob- tained a fine, lustrous precipitate of chromium from a solution of 25 parts of chromic chloride in 75 parts of water electro- lyzed at 75 C., with a current density of 40 to 50 amperes per square foot (0.3 ampere per square meter). No metal is obtained from a cold solution, but only a black non-adhering precipitate ; the tension was 4 volts. An evolution of gas took place at both electrodes. Good precipitates could be obtained only by the use of an excess of hydrochloric acid, which dissolving the precipitate of chromic oxide kept the bath clear; according to this the chromium chloride solution before men- tioned would have to be acid. Chromium could not be obtained either in the cold or warm solution containing 100 parts of water to 100 parts of chrome alum and 12 parts of barium sulphate (potassium sulphate), BY ELECTROLYSIS OF AQUEOUS SOLUTIONS. using current densities between 15 and 90 amperes per square foot. The anodes used were carbon and platinum, and the cathode copper. The tension was about 4 volts. Finally, we mention a work of B. Neumann 1 appearing in 1901, in which the different conditions were determined which are of importance for separating out chromium. His investi- gation was very properly based upon determining singly the influences of current strength, density, concentration and tem- perature with different solutions. His data are, on this ac- count, of importance and we will consider them in detail. The experiments were carried out by G. Glaser with the use of diaphragms ; the cathode space contained the chromium solution, and the anode space mineral acid or salt solution. The cathode was ordinary carbon, but the metal deposited adheres also to metals such as platinum, brass or lead ; the anode was, according to the solution, lead, platinum, or car- bon. A strong circulation of the cathode fluid is important for the obtaining of useful deposits because otherwise the solution at the cathode becomes too much impoverished when using high current densities and in that case no metal, but only chromous-chromic oxide is precipitated. Current den- sity in amperes per square centimetre. Product separated out. Ampere output in percentage. 0.009 O.OlS 0.036 0.045 0.072 0.091 0.137 0.182 At first metal then chromous-chromic oxide .... Metal mixed with chromous-chromic oxide Thin layer of metal upon which later is depos- 5-4 23-4 38-4 38.0 38.6 Metal with a small amount of chromous-chromic These results were obtained with a chromic chloride solu- 1 Zeitschr. f. Blektrochemie, 7, 656 (1901). A lecture given at the general meeting of the German Electrochemical Society at Freiburg, Baden. 28 CHROMIUM AND ITS COMPOUNDS. tion containing 100 grams of metallic chromium per litre, electrolyzed at the temperature of the room. If the temperature is varied while the current density re- mains constant it is found that the precipitate remains good up to about 50 but over 50 the chromium deposits as a black powder ; the output changes very little. The alteration of the concentration of the solution gave the following results : Concentration in grams of chromium per litre of solu- tion. The product separated out. Ampere output in percentage. 210 184 158 135 105 79 53 26 Metallic powder mixed with chromous chloride- Metallic powder mixed with chromous chloride- . - 56.6 49.0 38.4 At first metal, then chromous-chromic oxide A trace of chromous-chromic oxide and a strong The current density at which these tests were made is not particularly given; probably it lay between 0.09 and 0.18 ampere per square centimetre. There follows the incomprehensible remark that the ampere output was increased from 51 per cent, up to 89.5 per cent, by changing the anode solution. Or should this read "by changing the cathode solution?" All the preceding measurements were made on chromic chloride solutions ; sulphate and acetate solutions gave simi- lar results except that the values for the separation of metal in the concentration and current density tables are changed. The best proportion for sulphuric acid solutions with 65 to 85 grams of chromium to a liter was at 0.13 to 0.20 ampere per square centimetre current density, using which the current output of 84.6 per cent, was reached. The influence of tem- perature was, as with the chloride, of slight importance ; in BY ELECTROLYSIS OF AQUEOUS SOLUTIONS. 29 the most favorable cases the current output rose to 86.2 per cent. The acetate solutions gave imperfect precipitates and very deficient current output. Concerning the properties of the electrolytically precipita- ted chromium it was said that it showed the tendency like nickel to spring off in thin sheets. In spite of that, white de- posits were made on the cathode several mm. in thickness. If the cathode itself is rapidly rotated branch-like deposits several millimeters in length with crystalline projections are formed. On breaking thick pieces, a laminated structure is plainly to be seen, which in fact comes from thin formations of chromous- chromic oxide lying between the laminae. The metal is bright gray, frosted, and resembles white steel. In thin layers it is silvery. Its purity is high, containing only o.i to 0.2 per cent, of iron, while commercial chromium usually has a purity of only 97 to 98 per cent, on an average, though purer kinds may be produced. Electrolytic chromium is hard and brittle, showing, like the Goldschmidt chromium, two different con- ditions an active and a passive which may be artificially pro- duced and destroyed. Taken fresh out of the bath it is active, but by lying in the air becomes gradually passive and can easily be made active again by the methods proposed by Hittorf. While being dissolved periodic phenomena, as far as has been determined, do not take place during solution, which may be explained by the assumption that the periodic phe- nomena are really caused by a foreign element in the chro- mium, and, therefore, do not appear with the very pure elec- trolytic metal. The value of the potential of electrolytic chromium in absolute measure when dipped in normal chromium sulphate solution was 0.258 volt, in normal chromic chloride solution 0.241 volt, while the corresponding values for zinc were 0.524 and 0.503 and the values of cadmium 0.162 and 0.174. 3O CHROMIUM AND ITS COMPOUNDS. Active chromium stands, therefore, in the tension series, as has already been remarked by Hittorf, between zinc and cadmium. The voltage required for decomposition was found for a five times normal sulphate solution 1.90 volts, for an eight times normal chromic chloride solution 1.69 volts. For practical purposes at least double these tensions must be used. In the discussion connected with this communication H. Goldschmidt stated that he likewise had performed ex- periments for the precipitation of chromium in similar appa- ratus. Hydrochloric and also acetic acid solutions were used, the latter with the worst results ; an iron tube was used as cathode, rotated rapidly around its axis, the anode was retort carbon in acidulated chromic chloride or in hydrochloric acid. A small propeller was placed in the cathode fluid which was contained in a porous cell, in order to increase the circulation. With current densities of o.i to 0.4 ampere per square centi- meter of cathode surface there was obtained peculiar ribbed or furrowed layers of metallic chromium several millimeters thick, but very brittle, easily detached, and never quite free from oxygen ; it was particularly lustrous upon the tops of the ribs. " Dr. Neumann has gone decidedly further in his experiments than I, since he has made exact determinations of favorable conditions ; moreover, I obtained much poorer current output than Dr. Neumann. On that account I laid these investigations completely to one side, because calcula- tions showed that even if a higher current output can be ob- tained, these methods would be too expensive for commercial use, at least as far as concerns the manufacture of chromium it- self, in consequence of the complicated apparatus, and the high current consumption caused by the high tension in the bath." A later remark of the same person is worthy of reproduc- tion. " Chromium is, as is well-known, a somewhat noble metal in certain relations since it is more resistant to air than nickel and even better than cobalt. Unfortunately it has not been possible so far to obtain a strongly adhering deposit of BY ELECTROLYSIS OF AQUEOUS SOLUTIONS. 31 electrolytic chromium. If it were only possible to plate metallic articles with a thin sheet of chromium, as is so often done with nickel (recently electrolytic zinc-platings have also been successfully produced), it would be a great ser- vice, I believe, to many branches of technology and also to the artistic industries. It may be perhaps a small step in the direction of reaching that goal to use soluble anodes of chro- mium, with the aid of which a constant electrolytic chromium bath may be produced. The investigations of Dr. Neumann, which we have just listened to, go far in this direction, and which may well form the foundation of a technically useful method of chromium-plating, but will hardly, as before re- marked, serve for manufacturing metallic chromium. Appar- ently a chromium-plating upon iron ought to be just as good as a zinc-plating." In relation to the latter point Ostwald says that even if it was possible to plate iron with chromium, the chromium would scarcely be permanent since it would become active again by contact with this metal. " Therefore the danger of using chromium is that when rain falls upon it hydrogen commences to evolve, and in that case we can carry the chro- mium from our chromium-plated bicycles home in our hand- kerchief." If, however, this observation is worth noticing, the fear of such calamitous results has no foundation. Chromium is even in its active condition more passive than zinc. As already remarked, Hittorf established the fact that chro- mium occurs in two conditions, one active in which it lies be- tween zinc and cadmium, and one passive in which it stands near to silver ; he also gives the conditions under which chromium becomes active or passive. I will content myself in this work with referring to the original article 1 since it is easily accessible and scarcely falls within the limits of our subject, as also the work of Ostwald : " On the Periodic 1 Zeitschr. f. phys. Chem., 25, 729 (1898) ; 3O, 481 (1899) ; 34, 385 (1901). 2 CHROMIUM AND ITS COMPOUNDS. Phenomena when Chromium Is Dissolved in Acids ; m and the communication of Brauer: 2 " On the Electrolytic Behavior of Chromium during Solution in Acids." B. By the Use of High Temperatures. (See also pages 17 and 18.) The obtaining of chromium at high temperatures by the help of the electric current was mentioned in a patent speci- fication 3 not much later than the first patent was taken out for the electric precipitation of chromium from aqueous solutions. The invention is concerned with the melting and the reduc- tion of ores, heated by the assistance of the electric current and used for the production of aluminium and its alloys as well as other metals and their alloys of a more or less refrac- tory character like boron, silicon, calcium, chromium, titanium, etc., iron and steel. If the ores and ordinary oxides are melted and heated by means of the electric current they are decomposed and the oxygen set free violently attacks the vessel and the electrodes, causing considerable expense. This is avoided in Wilson's apparatus by the introduction of a re- ducing gaseous agent into the apparatus in such manner that the oxygen and the injurious constituents combine with the reducing medium so that the electrodes are protected. When treating oxides, hydrogen, hydrocarbons and in par- ticular ordinary illuminating gas, for instance, may be used. The manipulation when producing aluminium and alumin- ium bronze, which were the probable productions to which the invention was applied, was described at length and we need not follow this branch further. It will suffice to repro- produce the most important patent claims relating to our sub- ject. (i) The herein-described process of reducing metals or ores, consisting in heating them to the melting-point by the elec- tric arc passing from the metal or ore to an electrode and in- 1 Zeitschr. f. phys. Chem., 35, 33, 204 (1900). 2 Ibid., 38, 441 (1901). 3 English Patent No. 9,361, June 17, 1890, Thomas Leopold Wilson. BY THE USE OF HIGH TEMPERATURES. 33 troducing a reducing agent between the place at which the oxygen or other corrosive agent is set free, and the incandes- cent surface endangered by the corrosion, in order to protect the latter from destruction, substantially as described. (2) The herein-described process of reducing metals or ores consisting in heating them to melting by an arc passed between the metal or ore and an electrode above and the in- troduction of a reducing agent downwards into the arc be- tween the place at which the oxygen or any other destructive agent is set free and the incandescent electrode surface which is endangered, in order to protect the latter from destruction, substantially as described. (3) The herein-described reduction process for metals or ores, consisting in heating them to melting by means of an electric arc passed between the metal or ore and an electrode above, and the introduction of a reducing agent downwards into the arc by passing it through the electrode shown, whereby the gas distributes itself uniformly around the incan- descent end of the electrode and rising in a protecting stream, surrounds the electrode on all sides, as described in the speci- fications and for the purpose designated. (4) The herein-described reduction process for metals or ores consisting in subjecting them to the heat of the electric arc passing from the positive electrode underneath the metal or ore to the negative electrode above the same, introducing a reducing gas into the arc through the said negative elec- trode, whereby the gas is introduced into the center of the current of gas which is projected from the positive electrode beneath, against the negative electrode, and the gas so protects the electrode from destruction, substantially as described. (5) The herein-described process for the reduction of aluminium and other metals from refractory ores and for the production of alloys of these metals, consisting in the melting of the ores by means of the electric arc in the presence of the melted bath of the base metal to be alloyed and passing the electric current through the said bath as anode to a carbon 34 CHROMIUM AND ITS COMPOUNDS. cathode placed above it, in the presence of a reducing agent which can unite with the oxygen set free or any other corro- sive elements resulting from the decomposition of the ore, substantially as previously described. (6) The improved electric furnace for the reduction of re- fractory ores consisting of the melting crucible connected with the current conductor which serves to conduct the cur- rent to the interior of the crucible, a tube-formed electrode projecting downwards into the said crucible and a source of reducing gas in connection with the electrode through which the gas enters the crucible, in order to surround the white-hot end of the electrode and to protect it, substantially as de- scribed. The patent 1 taken out in the year 1893 y the nrm f Friedrich Krupp, of Essen, has essentially for its principal object the manufacture of chromium ; as the raw material for the manufacture of pure chromium, there is used an anode of impure, carbonaceous, crude metal in the ordinary commercial form ; if an electric current is led through it, using, for in- stance, molten sodium chloride as electrolyte, the anode is changed into metallic chloride by the chlorine evolved, which dissolves in the melt, while at the cathode metallic sodium is evolved. The latter must be rendered harmless since it would precipitate again the metal from the metallic chloride by simple replacement ; to avoid this the patent provides various devices. The negative pole may be surrounded by a clay cell or an unattackable diaphragm may be brought between the two poles, or the electrolysis may be performed in a U-shaped tube of refractory material. The sodium evolved may also be vaporized by the use of a high enough temperature, or changed back again to chloride by the use of a current of chlorine gas. As soon as a sufficient amount of metal has been converted into chloride at the anode any sodium or sodium oxide re- maining at the cathode is removed in a suitable manner, and 1 German Patent 81,225, June i, 1893. BY THE USE OK HIGH TEMPERATURES. 35 by further electrolysis the pure metal or alloy is obtained, while at the same time fresh metal is continually dissolved at the anode and the composition of the bath remains constant. By using impure ferro-chromium as an anode, pure ferro-chro- mium is obtained at the cathode ; the same process may be employed for other metals and alloys. In place of sodium chloride other halogen salts melt- ing at a red heat may be used, and in case they can be ob- tained the halogen compounds of the metallurgical products being electrolyzed. The patent claim reads : An electrolytic process for the manufacture of pure chromium and manganese and pure alloys of the same by electrolyzing compounds rendered fluid by heat characterized by using the carbonaceous metals and alloys of the two named metals as anodes in an easily fusible bath of halogen compounds, and precipitating out pure metals and pure alloys at the cathode. The manufacture of pure chromium does not succeed in the manner described without further precautions, since the metallic impurities of the crude metal will be found for the most part in the metal precipitated at the cathode. The idea of " purity" seems to be somewhat indistinct in the patentee's mind, since the mere absence of carbon seems to be regarded as sufficient warrant for calling the metal pre- cipitated " pure." What has just been said regarding chro- mium itself is true also of its alloys since besides the metals to be alloyed, other metals were separated out of the electro- lyte if present. With constant use the bath would further become very im- pure ; it is not clear that the solution of the anode would go forward easily and without a considerable increase of voltage. Much better results 'may be obtained by the manner de- scribed by Moissan in two communications. 1 He heated a 1 Compt. rend., 116, 349 (1893) and 119, 185 (1894). See also Ann. chim. phys., 78, 559 (,1896). A form of furnace which is described in Compt. rend., 117, 679, in which the raw material is passed into the melt- ing zone through a hollow, suitably inclined, carbon, tube terminating about i cm. underneath the arc. 36 CHROMIUM AND ITS COMPOUNDS. mixture of calcined chromic oxide and carbon in the electric furnace and obtained a shining and completely melted mate- rial. Using 350 amperes and 50 volts, the operation was imished in eight to ten minutes ; the charge had a weight of 100 to no grams. With 30 volts and 50 amperes he obtained a button of TO grams in thirty to forty minutes. These buttons stuck to the crucible and contained more or less carbon according to the composition of the original mix- ture. The carbon content of the pieces analyzed varied be- tween 8.6 and 12.8 per cent. The mass was purified by a further operation : It was broken into large pieces, put into a carbon crucible carefully lined with chromic oxide, and covered with chromic oxide. The mixture was then subjected to the heat of the electric arc by which the oxide upon the surface and also the metal melted and the latter gradually lost its carbon content ; at least in heating in a current of chlorine no traces of carbon remained behind. The natural chrome iron ore, FeO.Cr 2 O , can be reduced in the same manner as chromic oxide forming a chromium- iron alloy ; the material is broken to a coarse powder, an ap- propriate quantity of carbon added and the whole heated in the electric furnace. Moissan operated with 2 kilograms of ore and 1000 amperes of current, at 60 volts, and obtained in a few minutes completely melted lumps of ferro-chrome of the following composition : Chromium 60.9 per cent, iron 31.6 per cent., total carbon 6.1 per cent., silicon i.i per cent. He proposed to use this product for the manufacture of alka- line bichromates by bringing the coarsely pulverized material into a bath of molten potassium or sodium nitrate; there would result insoluble iron oxide and soluble alkaline chromates. In his second communication, Moissan explains that the chromium made in the way just described, " is saturated with oxygen, and, therefore from the standpoint of the metallurgist was burned" He attempted on this account another method of purification, by melting the metal in quantities of one-half to one kilogram with lime, this taking up the larger part of BY THE USE OF HIGH TEMPERATURES. 37 the carbon to form calcium carbide, but leaving in the chro- mium the remainder of it, 1.5 to 1.9 per cent, as beautiful cubical and octahedral crystals, which he could not remove. " We observe, in fact, that the reverse reaction would take place in the presence of the fluid lime and the furnace gases, if the chromium was sufficiently pure. Thus the whole metal would be transformed into the crystallized double oxide of chromium and lime." 1 Finally this easily formed double oxide was used for puri- fication ; a lime crucible was lined with it and the carbonized metal melted in it. After cooling there was obtained a brilliant metal which could be filed and polished easily and showed on analysis no trace of carbon. Borchers remarked .on this process : 2 " Moissan has solved this question of the refining in a very skilful way by bring- ing over for the purpose of refining chromium, the principle of the basic Siemens-Martin furnace for making steel. By heating lime with chromic oxide he has manufactured a basic material which not only serves for oxidation but for the pur- pose of slagging off the impurities. By lining the hearth of the electric furnace with this calcium chromite and melting thereupon the carbonaceous and siliceous metal the impurities are entirely oxidized and removed by the chromic oxide just as occurs to impurities in the Martin steel process by the action of iron oxide. No further data are known about the cost of this process. The commercial practicability is scarcely to be doubted, but the cost could be considerably reduced by producing the crude chromium as a carbonaceous metallic powder either in cruci- ble furnaces or in electrolytic resistance furnaces in order to save the heat necessary for melting it. At this point Moissan remarks that melted lime can also be used for the refining of commercial ferro-chromium in the electric furnace. He has for instance found that commer- 1 Moissan : " Der Elektrische Oefen," 1897, p. 196. ' 2 "Elektrometallurgie," 2nd Ed., p. 347. 38 CHROMIUM AND ITS COMPOUNDS. cial ferro-chromium with 7.3 per cent, of carbon, contained only 5 per cent., after having been melted once under a layer of melted lime and after the second melting only o.i per cent.; by using this commercial product a considerable amount of chromium can be introduced into steel without raising its carbon content. He is of the opinion that these experiments might probably be utilized on an industrial scale. Since the properties of pure chromium are of importance, the following data of Moissan's are repeated, especially since they contradict part of the information found in the text-books. The density is 6.92 at 20 C. It burns quickly but with a much more active discharge of sparks than iron when brought into the blue point of the blowpipe flame ; the residue is a round piece of melted chromic oxide. . It is worth while re- membering, moreover, that the carbonaceous material burns much more difficultly although it melts lower ; the melting- point of pure chromium is above that of platinum. Molten chromium in the electric furnace is a mobile shin- ing fluid similar to mercury ; it can be cast into a mould. By using 1000 amperes and 70 volts, 10 kilograms of (not quite pure) chromium can be melted at one time, in a suitable fur- nace. Quite pure chromium containing no iron exerts no influ- ence on the magnetic needle. It has great capacity for taking a polish, and its hardness is less than that of quartz ; glass is scratched by it with great difficulty. Carbonaceous chromium is much harder. The carbide CCr 4 scratches quartz with difficulty, the carbide C 2 Cr 3 scratches topaz, and the finely granular metal, whose carbon content is between 1.5 and 3. per cent., can only be worked by cutting stones charged with diamond dust. Carbonic acid and moisture have no appreciable influence upon the crude chromium at ordinary temperatures, while pure, well-polished chromium suffers a superficial light oxida- tion. In general, it can be designated as unchangeable in the air. Heated in oxygen to 2000 it burns with evolution of BY THE USE OF HIGH TEMPERATURES. 39 sparks. If filings are heated to 700 in sulphur vapor they glow and are changed to chromic sulphide. If a mixture of pure chromium and carbon is heated, crys- talline needles of the carbide CCr are formed with a violent deflagration, while at the temperature of the electric furnace C 2 Cr s results. Boron and silicon unite with chromium at the heat of the electric furnace to finely crystalline compounds of great hard- ness and resistance to chemical reagents. Gaseous hydrochloric acid acts on it at a red heat, forming chromous chloride. Aqueous hydrochloric acid attacks the metal very slowly when cold, faster when warm ; used as an anode it dissolves even in dilute acids which otherwise would attack it only at boiling. Further data on the behavior of chromium towards acids and its use as electrodes, about its active and passive conditions, etc., were published a year later by Hittorf (see page 29). At 1200 chromium is attacked by a current of hydrogen sulphide and converted entirely into a sulphide of crystalline appearance ; in carbonic acid it covers itself at this tempera- ture superficially with a layer of green chromic oxide mixed with carbon ; carbon dioxide attacks it, forming superficially a layer of Cr 2 O 3 while the metal absorbs carbon, a circum- stance which explains the difficulties which are met with when one tries to refine chromium in the lime crucible. Chromium is not sensibly attacked by melted caustic potash; on the other hand it burns with beautiful illumination in con- tact with melted potassium nitrate or still better at a red heat with potassium chlorate. Finally, the interesting properties of some alloys are men- tioned. Pure copper containing 0.5 per cent, chromium has nearly double as much electrical resistance and is, when well polished, more resistant to damp air than without this addi- tion. Two Russians, A. Koryscheff and S. Demmenie, together, 40 CHROMIUM AND ITS COMPOUNDS. in the year 1896, took out the English patent 1 for a very gen- eral process for the direct manufacture of iron, chromium and analogous metals from their oxygen, sulphur, arsenic, etc., compounds, and since they use electrical energy we will re- produce the more important part of it from their very com- prehensive patent specifications which contain many draw- ings. We need not go further, however, into the considera- tion of their metallurgical furnace likewise protected by En- glish patents. The two authors emphasize that the study of "pyrochem- ical solutions," that is, the solutions in which one body is dis- solved in another rendered fluid by heat, leads to the convic- tion that the dissolved bodies, be they of simple or of combined chemical nature, have acquired a gas-like that is, a very diluted condition. The degree of "dilution" (rarefaction) can be increased in three ways : First, by raising the temperature ; second, by the action of chemical agents ; and third, by the help of electricity. Naturally, all three aids to this end may be used alone or in combination with each other ; so may, for instance, iron oxide dissolved in molten silicates be reduced either by raising the temperature to the dissociation point or at a lower temperature by the help of chemical reagents or by electricity. After this somewhat fantastic presentation they explain the main idea of their invention, that of performing the decom- position or the reduction of metallic compounds by bringing them into the condition of pyrochemical solutions, in order to then subject them to the action of heat, chemical and elec- trical energy ; they then affirm that this pyrochemical pro- cess is applicable to all metals and alloys. Since a pyrochemical solution is aimed at, great care must be taken that the ores, gangue and furnace lining are so chosen that a fluid mass containing the metal will result. Electrical energy can be called upon as explained, as sup- plementary to the influence of heat and chemical action, but 1 English Patent 6,654 ; March 26, 1896. BY THE USE OF HIGH TEMPERATURES. 41 its office when so used is an entirely different one from elec- trolysis ; it acts here upon materials which are already in the condition of "tension," so that only the smallest assistance is required from it in order to complete the dissociation. The electric energy can be used either as dynamic or static. In practice it is advantageous to use transformers for the ob- taining of very high tensions, and to connect the anodes with the slag and the cathodes with the hearth of the furnace which in this case is made of a conductor such as graphite. The current produced in the transformer is so regulated that an "electric" discharge passes from one side of the elec- trodes, and a dark discharge from the other. For this pur- pose suitable resistances are put into the circuit as well as a quick-working interrupter to avoid the development of the small high-tension closing currents. The above will do for characterizing this new invention ; further criticism is not necessary. Of the patent claims, the following concern us. They claim : (i) A new process for the direct production of metals and their alloys from their ores, characterized by the conversion of the whole charge into the melted condition in such manner that the metallic compounds are dissolved " pyrochemically " in the rest of the mass. (5) In the manufacture of metals from their ores according to the methods described under (i), the use of chemical reagents in the form of pyrochemical solutions as a supplement to the thermic energy. (6) For the manufacture or formation of pyrochemical so- lutions (see 5) the introduction of carbon into the charges in quantities of 1/5 to 1/40 of that necessary fof the conversion of the oxygen present in the oxide into carbon dioxide. (7) The combination of the process described under (i) with the use of electrical currents as supplemental to thermic and chemical energy as described. (8) In the process as described the use of electrical energy in the form of dark discharges of very high tension, which are 42 CHROMIUM AND ITS COMPOUNDS. rendered possible practically by the including in the circuit of suitable resistances for the purposes of choking the contact currents of the transformer or by means of the acoustic inter- rupter which produces alternating current of very high tension and extremely high frequency, as described. The German patent of Aschermann 1 is concerned only with the reduction of chromium. An air-tight electrical melting apparatus of cast steel, containing an exactly-fitting graphite crucible to receive the bath into which a movable non-porous electrode passes, is used. It is charged with a mixture of chromic oxide and antimony sulphide, in the proportion of 10:23 or also, not so strongly recommended, however, with a mixture of chromic oxide, sulphur and pure antimony in the proportions of i o : 10 : 23; the apparatus is thereupon closed air- tight and a relatively gentle current of 20 to 2 5 amperes passes through. This suffices to melt the mass and cause the reaction in the crucible. The process proceeds with considerable de- velopment of heat. On opening the apparatus there is found, in the upper part, antimony oxide as an amorphous deposit, a very small part of which may, however, be crystalline, also antimony sulphide and traces of chromic oxide, but on the other hand the crucible contains a molten mass of evidently crys- talline structure consisting of metallic chromium in combina- tion with the excess of antimony. The removal of the anti- mony by repeated remeltings is difficult and costly ; it is much simpler and cheaper to break up the mass and heat it over an open fire. The antimony volatilizes at a white heat and me- tallic chromium remains behind, since chromium itself only volatilizes at temperatures above 2000. It is worth remarking in connection with this process that the electric current seems to be used only for starting the re- action. A small part of the mass is heated by the help of the heat developed by the current to the temperature of reaction and the heat generated by the decomposition then suffices to continue the reaction throughout the whole mass and to bring 1 German Patent 93,744, June 30, 1896. BY THE USE OF HIGH TEMPERATURES. 43 it to fusion. The cost of the electrical energy is, therefore, al. most negligible. If the reaction takes place as is claimed, and if the separation of the chromium from the antimony proceeds in such a simple manner as the description would lead one to believe, then this process must be described as comparatively simple and applicable for the manufacture of large quantities of chromium. The patent claim reads : Obtaining of metallic chromium from chromic oxide by the agency of antimony sulphide in the electric furnace. In a supplementary patent 1 the author extends his process to other metals and to metallic alloys. He makes in this con- nection the following remarks which are of interest, particu- larly for the obtaining of pure chromium and part may be re- garded as an explanation of the process described in the principal patent. " The manufacture of pure metals in electric melting fur- naces from their oxides offers great difficulties if the metals have a great capacity for dissolving carbon. There is then -obtained either a carbide or a metal highly contaminated with carbon. It is necessary to use an easily fusible metal as an addition which will unite with the reduced metal in the nas- cent state and prevent it thereby from taking up carbon. But this alone is not sufficient. To obtain the metal as free as possible from carbon it is further necessary that the reduc- tion proceed quickly and that it be operated by the smallest current possible. Quite aside from the cost of production, this is an important requisite. It is not possible to obtain the reduced metal pure if strong currents are used or by a lengthy application of the arc. The fluxes volatilize and in their place carbon is absorbed by the metal. Finally the metal obtained will volatilize itself, and there will remain behind only carbon in the form of graphite. The metal added has 1 D. R. P. No. 94,405 of 24th of November, 1896. Extreme life until June 29, 1911. The English Patent No. 859, taken out by Aschermann, Jan- uary 12, 1897, contains the substance of both the German patents cited. 44 CHROMIUM AND ITS COMPOUNDS. either none or only a very slight influence upon the course of the reduction; it serves more as a means of protection against the taking up of carbon. If a metallic sulphide is used as the addition, the reaction takes place quickly and with a much smaller expenditure of electrothermal energy. It is therefore the sulphur which influences favorably the course of the reduction." The manufacture of metallic compounds in the way de- scribed is stated to be of great importance. For instance, ac- cording to this process, ferro-chrome can be made by the melt- ing together of 10 parts of iron sulphide and 9 parts of chromic oxide ; also for this reaction a comparatively small amount of electrical energy suffices. Patent Claim. A process of obtaining metallic alloys and pure metals by the electrical heating of a mixture consisting of the oxide of one and the sulphide of another metal or the oxide of the metal and the sulphide of a non-metal, or from one oxide of a non-metal and the sulphide of the metal with the exclusion of the process of manufacture of chromium described in patent No. 93,744- The foregoing compilation exhausts the processes which endeavor to manufacture chromium by the direct application of the electric current. I cannot, however, close this chapter without thinking of the processes which manufacture these metals by the indirect use of the electric current such as by reducing chromium from its compounds by means of alumin- ium, or magnesium or calcium carbide, which substances are indeed to-day manufactured by electrolytic processes. Since, moreover, these processes are likely to become important and are already beginning to be so at present, it is advisable to present the patent literature on this subject, which may not be easily accessible to everybody. The first to be mentioned is the proposition to obtain chrom- ium in the already-mentioned manner, made by J. Vautin. 1 1 English Patent 8,306, April 26, 1894. BY THE USE OF HIGH TEMPERATURES. 45 His patent is dated April 26, 1894. He alludes to the great affinity of the finely divided metallic aluminium for oxygen and sulphur by means of which other metals may be reduced from their oxides and sulphides and in every case from their sulphates while the heat generated in the decomposition keeps in fusion the difficultly fusible metals and brings the reduced metal mostly into a compact molten condition. The metal obtained is free from aluminium and naturally also from carbon. For the reduction there is used a vessel lined with refrac- tory material like magnesia into which the substance to be reduced mixed with very finely divided aluminium is placed. Stress is laid upon the very fine division of the aluminium, since on this depends the obtaining of the reduced metal con- taining no admixture of aluminium: The resulting slag may be in certain cases made more fusible by the addition of fluxes like cryolite and fluorite ; potassium chloride and sodium chloride may also be used. The equation of the reduction is given as Cr a O -f A1 2 = Cr 2 -f A1 2 O 3 , which makes it evident that the normal propor- tions by weight were used for the mixture, but this is not particularly specified. An excess of aluminum will not hurt, since it is easily burned out or volatilized, and in this manner does not get into the reduced metal. In general, an excess of oxide or sulphide is to be recommended. The reaction is com- plete " as soon as the temperature is reached at which the aluminium unites with the oxygen or sulphur." The patent claim reads : (1) The manufacture of chromium, iron, tungsten, titanium, molybdenum, nickel, cobalt, uranium, vanadium, and beryl- lium by the treatment of their oxides and sulphides with aluminium, in the finely divided or powdered condition at high temperatures, as described. (2) In the herein-described process of reducing metals from their oxides and sulphides the use of metallic aluminium in 46 CHROMIUM AND ITS COMPOUNDS. finely divided or powdered condition as the reducing agent, as described. The German patent 1 of the firm of Th. Goldschmidt, of Essen, has, in the main, the same contents except that it is limited especially to the manufacture of chromium. It states that the attempts to reduce the oxides of manganese by alumin- ium have not been practically successful, because the result- ing slag, A1 2 O , with an admixture of manganese oxides, is extremely fluid and highly capable of dissolving the materials of which crucibles are made. The case is not so difficult with chromium, because the chromium-carrying alumina slag does not attack the crucible walls and is so viscous that it will not pass through cracks of considerable size, " but in fact makes its own containing vessel by the formation of a solidified layer or coating." " The process is carried out by mixing the chromium oxide with a small excess above the theoretically required quantity of finely divided aluminium, and this mix- ture heated in a suitable furnace, as for instance a reverberatory furnace or in crucibles, until the reaction begins, whereupon there is formed a metallic mass and a layer of melted alumina lying upon it." Patent Claim. A process for the manufacture of molten reguline metallic chromium characterized by heating a mixture of aluminium with chromic oxide in a crucible or furnace by external heat until the reaction begins, whereby a metallic mass of molten chromium is formed and upon it a layer of melted chro- miferous alumina. In both the patents alluded to, the commencement of the reaction is brought about by heating the whole mixture ex- ternally up to the temperature of reaction. Since the reac- tion is, however, strongly exothermic (evolving heat) it is therefore evidently sufficient to heat only a part of the mass either by the blowpipe flame or an incandescent body heated electrically, or in any other way, and so bring it locally to the 1 German Patent 112,586, Jan. 20, 1895. BY THE USE OF HIGH TEMPERATURES. 47 reaction temperature. The heat set free by the reaction will suffice to propagate it throughout the whole mass. As I have already mentioned, a similar manner of ignition appears in Aschermann's process (page 42). The same idea is again worked out by Goldschmidt and Vautin in two new patents. The first 1 begins with the following sentence : " The object of the present patent is a process for the manufacture of pref- erably difficultly fusible metals and their alloys with each other and with metalloids. These metals or metalloids are separated out of their oxygen and sulphur compoiinds by a process of fusion, either alone or several simultaneously by the action of aluminium or a mixture of aluminium and mag- nesium." It is seen that the process is not here limited to chromium but rendered general, and that magnesium is added to alumin- ium as a means of reduction. Further attention is drawn to the fact that the reaction is exothermic and a new manner of ignition is proposed. It is worth noticing that the external heating of the mixture is now designated as particularly im- practicable. " Since if the mixture is heated all at once to the temperature of ignition it will react with explosive violence and it has been found impossible to manufacture the metals or alloys of the same commercially in this way." It is not necessary to enter further into the details of these patents ; they are clearly mentioned in the patent claims. We may add, however, that continuous working is described, for instance, by taking out the metal formed at the bottom of the reaction vessel from a tap-hole, and allowing the slag to flow from a corresponding opening higher up, and continuing the process in this manner by the addition of new reaction ma- terial. If it is desired to obtain an alloy, one may take a mixture of the oxides of the metals concerned or of their sulphides or evidently adding a part in the metallic form to the mixture ; if 1 German Patent 96,317, March 13, 1895. 48 CHROMIUM AND ITS COMPOUNDS. it is wished that the alloy contain aluminium the latter is used in excess, while by using an excess of the other metallic com- pounds, metals are obtained free from aluminium. The fol- lowing patent claims are reproduced : (1) An apparatus for the manufacture of metals and metal- loids or alloys of the same with each other by reducing their oxygen or sulphur compounds by means of aluminium or aluminium and magnesium, the process being characterized by the fact that the whole mixture to be ignited is not ignited at once but only a part, whereby the exothermic reaction propagates itself from that part to the rest of the mixture without the application of external heat. (2) A method of operating the process described in claim i, consisting in igniting in the melting vessel a small quantity of the reduction mixture and continuing the reaction by the addition of further quantities of the mixture, which may finally be added in the briquetted condition. (3) A method of operating the process described in claim i, characterized by heating the reaction mixture placed in the melting crucible, up to its reaction temperature at one point only, whereby the exothermic reaction propagates itself from this point. (4) In the method of partial ignition described in claims i to 3 the introductory process consisting in bringing into con- tact with the principal mixture and igniting another mixture acting upon the same principle, that is, consisting of alumin- ium with a metallic compound (oxide or sulphide) which is easily ignited and produces a high reaction temperature, 1 and thus communicates the ignition to the principal mixture. We may finally draw our attention to the fact that in the patent specifications the manufacture of manganese is given as an example, while this metal was not producible according to the former patent (see page 45), evident!}' in consequence of 1 In the description a mixture of barium peroxide and aluminium powder is given as an example. BY THE USE OF HIGH TEMPERATURES. 49 the extremely high temperatures obtained by the old method of operation. The second patent already referred to, 1 that of Vautin an- nounced in the year 1896, agrees so exactly with the Gold- schmidt patent that we can immediately pass to its patent claims, which, leaving out unnecessary repetitions, are as fol- lows: (i) to (8) Process for the production of metals and alloys from metallic oxides, sulphides or chlorides either alone or mixed with each other, by the reducing action of aluminium, consisting in heating to the reacting temperature a small quantity of the mixture of aluminium or aluminium-magnes- ium alloy with the metallic compounds, the latter in the finely powdered condition. The temperature attained by the reaction propagates itself successively and automatically to the adjacent parts of the mixture and ignites the whole mass, as above described. (9) The process described for the production of metals or alloys from metallic oxides, sulphides or chlorides alone or mixed with each other, characterized by heating to the reac- tion temperature a small quantity of the mixture of the finely pulverized compound and finely pulverized aluminium or aluminium-magnesium alloy in order to start the reaction, and then adding the principal part of the mixture, so that in consequence of the heat evolved by the reaction of the first part the main reaction is propagated automatically through- out the whole mass, as before described. (10) The process described for the reduction of metals or alloys from metallic oxides, sulphides or chlorides alone or admixed with each other, characterized by covering a small portion of the mixture of the finely pulverized compound with finely pulverized aluminium or aluminium-magnesium alloy with a small quantity of a similar mixture which is more easily started in action, and heating the latter whereby the reaction 1 English Patent 16,685, July 28, 1896. 50 CHROMIUM AND ITS COMPOUNDS. is started and is then propagated to the principal mixture as described. Finally Dollner 1 has patented the use of carbides in place of aluminiiim or magnesium. " The details of the process aside from the replacement of aluminium and magnesium by the carbide are exactly the same as set forth in the above patent specifications Nos. 112,586 and 96,317." The reaction tem- peratures are somewhat lower, and if it is wished to raise them aluminium or magnesium may be mixed in ; if it is wished to diminish them a ballast of negative material can be added, for instance, an excess of the metallic compound to be reduced or a flux, or finally an excess of calcium carbide if it is desired to carbonize the metal produced or to introduce calcium into it. By using a sufficient excess of the oxide or the sulphide, the metal can be obtained more or less free from carbon, at least free from the metallic components of the carbide. It is not possible to produce in this way completely carbonless metals or alloys, and, moreover, the evolution of gas produced by the formation of carbon monoxide may not be advantageous to the process. Patent Claim. (1) A method of carrying out the process of patent No. 96,317, for the manufacture of metals and alloys of the same with each other or with metalloids, using a mixture of oxygen or sulphur compounds of these metals or metalloids with pulver- ized carbides of the earth or alkaline earth metals, with or without the admixture of aluminium or magnesium, and operating the process in such a manner that the mixture is ignited at one point and thus the reaction started which is concerned in the production of those metals or alloys. (2) In the process described in claim i, the starting of the reaction by using as igniting material a mixture of carbides and oxides in composition analogous to the principal mix- 1 German Patent 113,037, Feb. 10, 1897. BY THE USE OF HIGH TEMPERATURES. 51 ture but reacting more violently, and thus transferring the ignition to the principal reacting mass. (3) A method of operating the process of claim i for the manufacture of alloys, characterized by adding the compo- nents in part in the metallic state to the reacting-mixture and not in reducing both of the components of the alloy simul. taneously from their oxygen or sulphur compounds. 1 Two lectures by Dr. Hans Goldschmidt given in iSgS 2 and iQOi 3 contain some further additions concerning the alumino- thermic operations. It is stated that the mixture of aluminium with the oxide to be reduced, called thermite, is theoretically made in quan- tivalent proportions ; though in general an excess of oxide is used since it is thereby possible to oxidize completely all the aluminium so that metal free from aluminium is obtained, in spite of the great alloying tendency of aluminium. " Before all, such metals are to be considered whose produc- tion in the pure state has not been possible by other processes, or even in the electric furnace and whose production now in the pure carbonless condition would offer particular advan- tages to the metallurgist. Chromium and manganese are the most important. The manufacture of these metals is carried out on a large scale, since the consumption has become quite large at several works. The principal consumption is supplied by the Allgemeine Thermit-Gesellschaft at Bssen ; besides them the French So- ciete d'filectro-Chimie in St. Michael de Maurienne (Savoy) has manufactured the metals named in particular chromium, under a license since 1899. The manufacture is carried out in large crucible-shaped vessels in which several hundred weight of the metals are produced in one operation ; this 1 Another method of reduction by means of calcium carbide which shows important variations from the above has so far been only incompletely de- scribed. See Zeitschr. f. Elektrochemie, 7, 451 (1901). 2 Zeitschr. f. Elektrochemie, 4, 494 (1898). 3 Enlarged reprint from Stahl und Eisen, 1901, No. n. 52 CHROMIUM AND ITS COMPOUNDS. operation takes scarcely a half hour in consequence of the rapidity of the reaction. On top of the metallic mass there is found the resulting alumina which, after removal, can be worked up again into metallic aluminium in the most suitable manner. The metal reduced from it, however, contains some chromium and is therefore best used again for the manufacture of chromium, so that we have here a complete cyclical process of technical practicability. The alumina can also be used as such, since it is quite free from water and therefore possesses many advantages over natural emery for use as an abrasive, since the emery on ac- count of its small water content does not possess the great hardness of the corundum made by fusion. The slag is interesting from other considerations. It shows small, red, transparent crystals mostly in small cavities, which are to be regarded as true rubies. On account of their small- ness they do not possess commercial value. Concerning the pure carbonless chromium, in the manu- facture of which pure aluminium must be used, it is principally used in the production of steel. In doing so, it has confirmed the old observation that metals very nearly pure possess other qualities than those contaminated by additions and that by their purity the properties of the alloys are influenced in the foregoing cases quite favorably. Some further communications concerning pure chromium will be given in connection with the chromium alloys at the conclusion of the next section. II. OBTAINING OF COMPOUNDS OF CHROM- IUM WITH THE METALS. A. By Electrolysis of Aqueous Solutions. If a pure solution of chromic chloride is electrolyzed with- out particular precautions, using mercury as the negative and platinum as the positive electrodes, the mercury according to J. Feree 1 takes up only traces of chromium, and there results only a brownish black powder whose formula according to Bunsen lies between Cr 4 O 5 and Cr s O 6 (see however later, under chromium oxides). In order to obtain large quantities of chromium amalgam a solution must be chosen as electrolyte which contains 160 grams of crystallized chromic chloride, 100 grams of concen- trated hydrochloric acid and 740 grams of water, and the cur- rent density must be raised somewhat high. With a current of 22 amperes and a mercury surface of 8.05 sq. cm. that is, a current density of 2.73 amperes per square centimeter 1.5 kilograms of solid chromium amalgam was obtained in an hour. A too feeble current density is to be avoided since with a current density of 0.28 ampere per square centimeter the formation of amalgam does not take place. The amalgam formed is washed quickly with cold water after pouring off the solution, and is then pressed out, after being dried on filter-paper, in a piece of chamois leather. The mercury pressed through the leather contains only traces of chromium, scarcely 0.03 per cent. The amalgam remaining in the leather has the composition Hg 3 Cr. If this amalgam is further dried by laying it upon double folded filter-paper and subjecting it to a pressure of 200 kilo- grams per square centimeter for several minutes, it loses more mercury and becomes a new amalgam having the formula HgCr. While the amalgam Hg 3 Cr is soft, lustrous, only slightly alterable in the air, loses its mercury on heating without melting and easily oxidizes, the amalgam HgCr is also lus- 1 Compt. rend., 121, 822 (1895). 54 CHROMIUM AND ITS COMPOUNDS. trous but harder and more easily alterable and small drops of mercury soon form on its surface. Both amalgams leave a residue of metallic chromium if heated in a vacuum ; the metal is very crumbly, but retains the shape of the original lump. If the amalgam is distilled in a vacuum at tempera- tures below 300 it leaves (which amalgam is not stated) a metallic chromium, the qualities of which are very different from those of the chromium obtained by Deville and Fremy. This chromium is pyrophoric even when cold ; it ignites spontaneously in the air and absorbs at the same time, in a remarkable manner, both nitrogen and oxygen. The heat generated by the reaction brings the whole mass to a bright incandescence. The chromium produced by Moissan by the distillation of the amalgam below 350 ignited in the air only when it was warmed slightly. The pyrophoric chromium resembles in its behavior the pyrophoric manganese obtained in a similar manner ; even when cold it absorbs nitrogen dioxide with incandescence, producing chromic oxide and chromic nitride. When heated in an atmosphere of nitrogen it also glows, forming chromic nitride. Even sulphur dioxide is absorbed in the cold with incandescence, and a slight warming in an atmosphere of carbon dioxide or carbonic acid will suffice to set the chro- mium in active incandescence. After heating, the pyrophoric chromium is less active and almost unchangeable in air. Some further communications on chromium alloys have already been given on pages 5, 10, 14, 15 and 23. B. By the Use of High Temperatures. Besides the processes which have been mentioned in de- scribing the methods for producing pure chromium (pages zfl, 3^ ^et seq\ there are some others which are recommended for the ^manufacture of chromium alloys ; according to Borchers 1 there belongs here the process of Moissan patented in 1 Jahrbuch der Elektrochemie, 2, 183 (1895). 2 German Patent 82,624, Nov. 9, 1894. BY THE USE OF HIGH TEMPERATURES. 55 Borchers says: "The reason why chromium and similar metals which it is desired to obtain free from carbon, cannot be manufactured simply by processes analogous to the manu- facture of aluminium lies in their high melting-point. Where the presence of aluminium is not injurious the manufacture of the more easily fusible aluminium alloys of these metals is recommended after the general idea of the electrolysis of alumina. It is a well-known knack of alloying to take one metal which is not fusible easily or does not dissolve quickly enough in a second metal with which it is to be alloyed, and to dissolve it in a third, which forms a better solvent for it, in order then to bring it in a dissolved form to the second metal. This process has been patented in Germany by Moissan, with his usual good fortune for getting patents for all known things. His patent concerns the difficultly fusible metals and metalloids such as molybdenum, tungsten, uranium, zirconium, vanadium, cobalt, nickel, manganese, titanium, chromium, silicon and boron which are to be dissolved in iron or other metals. He gets these materials at first in solution in alumin- ium or magnesium. In fact Moissan does not say that his process is to be operated electrolytically, but it is in fact operated in this way and has been done so several years, as the author can especially verify. Those metals which are not difficult to separate out electro- lytically, yet are hard to melt, may be prepared in the fluid con- dition from molten electrolytes, if they are received in more easily fusible metals, which is done by making the latter the cathode or by preparing the chromium simultaneously with the other metal." The patent specification describes the bringing the alumin- ium or magnesium alloy in an iron bath, or other suitable material whereby the difficultly fusible elements as chromium, etc., are taken up at once by the latter. The elimination of the aluminium is easy, in general it can be performed by using the oxide of the metal with which one is working. 56 CHROMIUM AND ITS COMPOUNDS. "In order for instance to get chromium-copper alloy a determined quantity of chromium-aluminium is added to molten copper, according to the content of chromium which is wished. The bath is well stirred and if the quantity of aluminium present is too high and must be removed, the ex- cess of aluminium is burned out by the addition of small quantities of cupric oxide, and the aluminium passes into the slag." In other cases the aluminium in excess can be oxidized u by the gaseous oxygen present in the solution," or by a current of air. An alloy of iron with tungsten, molybdenum, and chro- mium is obtainable in the Siemens-Martin furnace or in the Bessemer converter in an analogous manner. Patent Claim. Process for the manufacture of alloys, characterized by in- troducing the difficultly fusible metals molybdenum, titanium, chromium, etc., into metals and alloys in the form of alloys with aluminium. A process for the commercial manufacture of alloys of iron with chromium, etc., in an electrolytic manner, by using a flux consisting of lime, was patented by J. Heibling, of Grenoble, in I895. 1 He uses a shaft furnace, the inner lining being of carbon bricks and the outer of refractory bricks ; a cast-iron plate lies upon the movable bottom of the shaft, which is likewise lined with carbon bricks. Above the latter, which together with the lining of the furnace acts as the negative electrode, is the positive electrode in the form of a carbon rod of suitable dimensions. A carbon crucible is made " by placing in the closed cast- iron form a suitable mixture of pulverized coke or graphite and tar and baking." A mixture of pulverized ore (oxide), carbon and pulverized caustic lime is brought into the furnace, the lime being to the carbon in the proportion of 3 molecules to i. If practicable 1 German Patent 86,503, July 12, 1895. BY THE USE OF HIGH TEMPERATURES. 57 there is added to this mixture a quantity of iron oxide neces- sary to form an iron alloy of a certain proportion, with the carbon necessary for the reduction, in which case the weight of the cast-iron furnace bottom is to be taken into considera- tion. To the whole mass is finally added a suitable quantity of pulverized fluor-spar to combine with the silicon in excess. The action of the electric current is described as follows : The ore and finally the oxide of iron is reduced and forms an iron alloy which dissolves the carbon. The latter unites, however, immediately to form calcium carbide with the cal- cium from the lime (carbonic oxide being formed). The car- bide contains, dissolved, the remainder of the silicon as carbide of silicon and with the latter forms the slag under which the fluid regulus is found. By inclining the movable carbon bottom to one side the alloy can be drawn off. The silicious material serves as a protection for the calcium compound which, without its presence, would be much more easily subject to attack by damp air. An excess of silicon is to be avoided. The author predicts likewise the application of his process in particular to ores poor in the heavy metals. The patent claim reads : Preparation of alloys of iron with manganese, chromium, aluminium, nickel from their ores, in the electric furnace, characterized by the use of lime as a flux. In contrast to the processes so far mentioned which work with large amounts of carbon and afterwards treated by the process of Moissan (see page^g^, furnishing alloys containing carbon, stress is laid by La Societe Neo-Metallurgie Marbeau, Chaplet & Co., 1 upon the obtaining of alloys as free as possible from carbon. For this purpose they also use a "suitable electric furnace" which is apparently lined with a material containing no carbon, and heated with u horizontal arcs in such a manner that contact of the carbon electrodes with the metal and oxides in question, and the consequent carburiza- tion of the final product is avoided." 1 English Patent, 7,847, April 19, 1895. 58 CHROMIUM AND ITS COMPOUNDS. The process is described in detail as follows : A sufficient amount of a more or less easily fusible metal which is to form the alloy either in the metallic form or as an oxide mixed with the necessary amount of carbon for reduc- tion is put in the furnace, heated and, after reduction, melted. To the bath so made is added gradually the already prepared oxide of the heavy metal to be alloyed mixed with the req- uisite amount of carbon for reduction. Another method of procedure can also be used in place of the one already mentioned : "The same alloys can also be ob- tained by mixing the oxide of the difficultly fusible metal with the oxide of the easily fusible metal in sufficient quanti- ties and in the right proportions to form the desired alloy. " To this mixture of the various oxides there is added the proper amount of carbon for reduction, and a suitable amount is charged into the furnace. Under the influence of the elec- tric arc, the oxide of the more easily fusible metal is first re- duced and melted and forms the foundation of the metallic bath by which is gradually accomplished the reduction and fusion of the other oxide (the addition of the mixture of the oxides is dependent on the output of the furnace)." Finally there can be added to the bath of the easily fusible metal, the difficultly fusible metal in the form of a fine powder, which may have been produced by an electrolytic process. The composition of the manufactured alloy can be easily calculated, and the process is applicable for the manufacture of alloys with a high content of the difficultly fusible metal (70 to 80 per cent.) which then in this condition can find ap- plication as additions to other metals. It is quite possible with this process to obtain alloys quite low in carbon, but with certainty not entirely free from carbon. The patent claim reads : i. The preparation of improved alloys, as described, in a suitable furnace heated by horizontal arcs, to avoid the contact BY THE USE OF HIGH TEMPERATURES. 59 of the carbon electrode with the oxides and metals in question and thereby prevent the carburizing of the finished product, as explained. 2. The herein-described improved methods for the prepara- tion of improved alloys as performed in various cases ; as alloys of a more or less difficultly fusible metal, as chromium, tungsten, molybdenum, etc., with a more easily fusible metal, as nickel, cobalt, manganese, iron, copper, aluminium, etc., whereby the alloys named are obtained in the molten metallic condition, with the desired content of each the single con- stituents but particularly with a high content of the difficultly fusible metal. Such alloys are more easily fusible than the more difficultly fusible metal, as explained. 3. Various applications of the above-described method of preparing alloys ; in particular it is possible to add a diffi- cultly fusible metal when molten to another metal, or alloy, and particularly to iron and steel to which they will com- municate the characteristics of the difficultly fusible metal, as explained. A complete exclusion of carbon and other impurities is possible in the process of the Electro-Metallurgical Company, Limited, of London 1 , for the preparation of chromium alloys. In the description the effect of the addition of chromium will be discussed, although it is pointed out that the chromium is to be added in the chemically pure condition, in particular free from carbon, otherwise the property of the alloy will be in- juriously influenced. Further it is remarked that the addition of chemically pure chromium, electrolytically prepared, will improve all metals ; it makes them harder and more resistant to shocks, tension, and friction, as well as the corrosive action of air, dampness, acids, high temperatures, etc. Such chromium alloys are formed only when the chromium is added directly to the molten alloy. Likewise it is possible 1 German Patent 89,348, IJan. 16, 1896 ; English Patent of Placet, 202, 1896. 60 CHROMIUM AND ITS COMPOUNDS. to add one or more metals, which facilitate alloying on account of their large capacity for dissolving chromium, as zinc, man- ganese, magnesium, aluminium, antimony, bismuth, palla- dium, etc. Surely this last can only find application on paper. In order to avoid, in this process, the oxidation of the chrom- ium, it is electrolytically or otherwise covered with a more or less thick coat of a protecting metal, as copper, nickel, gold, silver, zinc, aluminium, tin, platinum. The addition of chromium in very small amounts exerts a good influence ; a few tenths of a per cent, makes the noble metals, gold, silver, copper, nickel, tin, as well as lead and zinc, considerably harder and tougher. For the best results the chromium should be added in the proportion of 0.5 up to 15 or 20 per cent. The metal alloys with a larger content of chromium are so extraordinary hard that they can only be worked with the grindstone. Recapitulating, the great advantages of the addi- tion of chromium consist as follows : An addition of chromium from 0.5 to 20 per cent, in copper or copper alloys, as bronze, brass, German silver, etc., im- proves them so that their "fracture" approaches that of steel. A small addition of chromium to the coinage metals makes them noticeably more resistant to pressure and the necessary washing. Finally chromium makes all metals as well as al- loys more resistant against high temperatures, and is valuable for the manufacture of hearth-plates and moulds, also more resistant against acids, alkalies, and other chemical reagents, which is to be provided against in the manufacture of chem- ical apparatus and cooking utensils ; it increases the hardness of metals and their ring, as of bells and trumpets ; it raises the electric resistance of manganese, manganese-iron, nickel- iron, and other metals, serving for the manufacture of resis- tance wire; and raises the anti-magnetic properties of copper, platinum, palladium, manganese, tungsten, and cadmium, a circumstance made use of in the construction of clocks, etc. The patent claim reads : BY THE USE OF HIGH TEMPERATURES. 6l Process for the preparation of alloys of pure chromium, characterized by the fact that the chromium added to the bath is coated previously in an electrolytic or other manner, with a coating of a metal (copper, nickel, aluminium, gold, silver, etc.), which prevents the oxidation of it at the moment of contact with the bath. A modification of this process of the same company is de- scribed in a Russian patent of September 13, 1897, from which the Chemiker-Zeitung 1 gives an extract. According to it the oxidation of the chromium will be prevented by being first brought into contact with a sufficient amount of molten cryolite, borax, or other easily fusible flux sufficient to cover it and then the metal to be alloyed added. In this way alu- minium will readily alloy with chromium, although otherwise even on long heating with molten aluminium a piece of chrom- ium will not dissolve because it has coated itself with oxide. It is remarkable that at these high temperatures the coating of oxide will not easily be reduced by aluminium, although the whole process of Vautin-Goldschmidt (see page ^efand following), rests on this fact ; if anything, the temperatures are still higher. The process can also be worked in this manner : Sufficient borax, cryolite, etc., can be added to form a complete cover- ing to the molten bath and then pieces of the hard metal, or it may be the alloy, thrown into the bath. While passing through the fluid coating the metal will be prevented from oxidizing and any coatings of oxide will be dissolved by the flux. To avoid contamination, the crucible used in melting must be lined with alumina or chromic oxide. For completeness we have an American patent 2 , as well as that of the Societe* Generate de Aciers Fins 3 , both patents concerning the manufacture of chromium-iron alloys. 1 Chemiker-Zeitg., p. 1068(1897). 2 American Patent 567,757, Electrometallurgical Company, Limited, July 8, 1896 ; the Official Gazette of the U. S. Patent Office, 76, 1642, 1896, nearly corresponding to the German Patent 90,746, of the I4th of June, 1896. 3 German Patent 120,310, April 6, 1899. 62 CHROMIUM AND ITS COMPOUNDS. The first patent claim reads : The described improvements in the treatment of iron and steel with pure chromium or similar metals, as molybdenum or tungsten, for the preparation of alloys of these metals, con- sisting therein, that by the addition of a proper amount of aluminium to the molten mass of iron or steel, it is nearly completely deoxidized, the deoxidation, however, by the alu- minium being first only completed after a small part of the chromium or similar metal has been oxidized, the rest alloy- ing with the mass of iron or steel. The other patent claim has the following wording : Process for the manufacture of chrome-steel, characterized by allowing a stream of open-hearth steel to flow over a mix- ture of aluminium powder and chromium-iron alloy in the casting ladle. The ignition of the aluminium by the outflowing stream evolves sufficient heat to bring the surrounding chromium- iron alloy to a quick and complete fusion, so that the chrom- ium can at once combine intimately with the stream of steel, and the cast-steel block which is characterized by great den- sity and freedom from blow-holes also shows a uniform distri- bution of chromium throughout. In concluding the processes for the preparation of chromium alloys the process of Heroult 1 for the preparation of very rich ferro-chrome must be considered. In the reference cited it is stated that chromite is heated in an electric furnace with the addition of a sufficient amount of carbon for reduction and 38 to 30 per cent, of fluxing material (lime, kaolin, or fluorspar). The ferro-chrome formed will, on account of the greater vola- tility of the iron, become richer in chromium the quicker the operation is completed ; the fluxing material permits the regulation of the operation. Quite identical with the previous patent is an American patent 2 , under the name of Vielhomme, whose claim only was accessible to me. It reads : 1 French Patent 284,525, Dec. 30, 1898 ; Chemiker-Zeitg., 694 (1899). 2 U. S. Patent 648,119, July 25, 1899; the Official Gazette of the U. S. Patent Office, 91, 717 (1900). BY THE USE OF HIGH TEMPERATURES. 63 Process for the manufacture of rich ferro-chrome, consisting therein, by the heating of chromite in an electric furnace in the presence of a sufficient amount of fluxing material and powdered coke, to a temperature sufficient for reduction and melting of the ore and the volatilization of the larger amount of iron. The remaining iron then forms with the chromium a rich ferro-chrome alloy. 1 At this place also, use will be made of a very kind letter of Dr. H. Goldschmidt, and reference made also to page^&flmd the literature quoted there, and some general remarks on the preparation of chromium and its alloys will be added. As much as is known chromium and its alloys are produced technically in four ways : i. As ferro-chrome in the blast-furnace ; 2, as ferro-chrome in crucibles ; 3, as ferro-chrome in the electric furnace ; 4, as pure chromium by means of aluminium. The blast-furnace process furnishes no particularly rich chromium product, but only one of 30 to 40 per cent, and may get into a difficult position because at this day higher percentage alloys are preferred. Processes 2 and 3 furnish alloys containing 60 to 63 per cent, of chromium and 8 to 12 per cent, of carbon, calculated on the chromium content. Large quantities are produced in this way in France and Germany. The process of manufac- ture in crucibles in which, as is said, only a single kilogram can be made at one time, is not cheap. The composition of the chromium mixtures are held secret. Finally this process has been somewhat superseded by the electrical process, so that to-day the larger amount of ferro- chrome is produced in the electric furnace, and particular preference for this method is shown in France. In particular the early calcium carbide manufacturers, who had to give up 1 In the Russian Privileg., 351, Sept. 30, 1897 (referred to in Chemiker- Zeitg., 1897, 1068), E. Placet protects a process for the manufacture of chrome-iron or -steel, Tubes or bars on which chromium has been electro- lytically precipitated were melted together or welded or drawn into wires or cemented ; in the last case the outer surface becomes very hard. See also page ii. 64 CHROMIUM AND ITS COMPOUNDS. their production because the patents of Bullier under which they worked were declared legal by the French Court, in opposition to the German Imperial Court, turned to this new opportunity with avidity. The foundation of this process is indeed that given by Moissan (page^j^T 35~ The disadvantage of the ferro-chromium, particularly that made in the electric furnace, which probably contains a few more per cent, of chromium than that made in the crucible, lies in its high and irregular content of carbon, so that for the best qualities of steel alloys, especially if several per cent, of chromium is to be added, it cannot always be used. For instance if it is a question of increasing the content of chromium relatively to the carbon which is added to Martin steel in the prepared condition at the end of the operation so as to provide a charge, for instance, which along with 2.5 per cent, of chromium shall contain 0.2 or scarcely only 0.15 per cent, of carbon, this cannot be obtained without the assistance of carbon-free chromium ; since when using ferro-chromium at least 0.25 to 0.3 per cent, of carbon is carried in, which with the carbon already in the fluid metal raises the percentage to at least 0.3 or 0.35 per cent, in the finished metal. Also in crucible casting even where small additions of chromium are concerned, i to 2 per cent., and where therefore a particularly small carbon content of the alloy added is not of importance, carbon-free chromium is preferred. Further, carbon-free chromium is used for the manufacture of high chromium tool steel containing 6 to 10 per cent, or even more chromium. It is worth mentioning that the addition can be made in the form of chrome-thermite (as the mixture of aluminium and chromium oxide is called). It is used in the manner that the highly heated steel is allowed to flow upon the mixture men- tioned and immediately ignites it. The chromium dissolves in the metal bath, while the slag floats on the top (see page 56). BY THE USE OF HIGH TEMPERATURES. 65 On the grounds before mentioned, in spite of the fact that the ferro-chromium prepared in the electric furnace is much cheaper, pure chromium prepared from chromium-thermite is often preferred. The last ignites itself only at very high tem- peratures as has been stated, and yields a good output only when used in large amounts. The market price of 60 per cent, ferro-chrome, according to the quotations of December 15, 1901, was $350 per metric ton ; it may upon large demands become much lower. It was much higher a short time ago before the introduction of the electrical process. The price of commercially pure chromium is $2 per kilo. In spite of that the product is used by many firms by the car- load. Besides ferro-chromium there is still another rich chromium alloy which finds many applications, the chrome-manganese containing 70 per cent, of manganese and 30 per cent, of chromium. It is used in the manufacture of copper alloys, the so-called chrome-manganin. Chromium alloys directly with copper with great difficulty, it liquating out quite easily so that the chromium-copper with 10 per cent, of chromium made by the alumino-thermite process and used in the manu- facture of chromium-treated copper is very seldom chosen, while the chrome-manganese alloys easily with copper like pure manganese itself and has just recently come into frequent use. In contrast to iron, chromium seems to raise the ductility of copper. A chrome-copper with 10 per cent, of chromium has almost completely the color of pure copper, but is, however, considerably harder. III. OBTAINING OF THE COMPOUNDS OF CHROMIUM WITH THE NON-METALS. A. Carbon Compounds. It has often been stated in the description of the methods serving for the manufacture of chromium, that the chromium often contains carbon. We shall now give information re- garding the obtaining of well-characterized compounds of chromium with carbon, which has appeared in the work of Moissan 1 already referred to (see page 3$, et seq.\ If a crucible containing chromium and a large excess of carbon be heated in an electric furnace using 350 amperes at 70 volts for from ten to fifteen minutes, there is obtained an inco- herent mass in the bottom of the crucible filled with crystals of chromium carbide of the formula C 2 Cr (according to two analyses). This carbide forms in lamellae of very high, somewhat greasy lustre, and is only slowly attacked by dilute hydrochloric acid on long standing, not by concentrated hy- drochloric acid, fuming or dilute nitric acid, or aqua regia. It is slightly attacked by molten caustic potash, but on the contrary molten saltpeter attacks it with ease. Its specific gravity is 5.62. Hot water is not decomposed by it. The formation of another carbide has been observed in a number of preparations of metallic chromium. The surface of the metallic pieces are seen to be covered with brownish red needles, which often reach a length of i to 2 cm, and show the composition CCr 4 (after three analyses). Sometimes this carbide takes the form of shining needles lining the cavities which form in the molten chromium. Its density is 6.75. Both compounds dissolve, on a further raising of the tem- perature, large quantities of carbon, which they then again give up in the form of small shining folia of well crystallized graphite. P. Williams 2 obtained a double carbide of iron and chrom- ium, on heating a mixture of 200 grams of pure calcined chromic oxide, 200 grams of iron and 70 grams of petroleum coke in a carbon crucible for five minutes, using 900 amperes 1 Compt. rend., 119, 185 (1894). 2 Ibid., 127,483(1898). SILICON COMPOUNDS. 67 and 45 volts. The resulting metallic mass had a crystalline appearance and resembled commercial ferro-chromium. On treating from two to three hours on the water-bath with aqua regia it was gradually disintegrated, and the crystals which sepa- rated were retreated anew to separate from them the last traces of iron. They were separated from the remaining impurity, a small amount of graphite, by means of bromoform. The crystals had a metallic appearance, somewhat the color of nickel, and consisted of a conglomeration of tough needles; they scratched glass but not quartz, were not magnetic and had a density of 7.22 at 19. They oxidized at about 500 in a stream of air or steam. Gaseous or liquid acids attack them slowly, nitric acid and aqua regia being without appre- ciable effect. The analysis gave the composition of 3Fe 3 C. 2Cr s C 2 . It was also noticed that an excess of iron carbide easily changed the results of analysis. B. Silicon Compounds. In connection with the description of the manufacture of the hardest possible materials, which will scratch chromium steel, the ruby and even the diamond, such as the silicide and boride of titanium, carbon, etc.; it was mentioned by F. Chap- let 1 that in the same manner also the silicides of the metals, zirconium, chromium, iron, nickel, cobalt, manganese, molyb- denum, tungsten, uranium and vanadium may be pre- pared containing various amounts of nitrogen and carbon. This patent demands more attention on this account. For preparing the silicide of titanium, for instance, the fol- lowing process was proposed : 82 parts of pulverized rutile, 60 parts of sand and 48 parts of pulverized carbon are mixed together. The dry mixture is pressed into a carbon crucible, placed in an electric furnace, built somewhat like Moissan's with movable electrodes, which is closed up and heated for a longer or shorter time according to the strength of the current. There is obtained a gray, fused mass which, after cooling, has a shiny fracture, is of great hardness and has a density of about 4.8 and which can contain according to the duration of 1 English Patent 15,453, August 14, 1893. 68 CHROMIUM AND ITS COMPOUNDS. heating, besides the titanium and silicon, 20 to 25 per cent, of carbon and a varying amount of nitrogen. This compound can also be obtained in a crystalline state if a metal in the molten state is employed as a solvent and is afterwards re- moved by a suitable acid. For further particulars of the process a reproduction of the detailed patent claim will suffice : (1) The manufacture or production of hard materials, which are able to scratch chrome steel, and scratch or cut rubies or diamonds, characterized by subjecting a mixture of the neces- sary materials with or without reducing agents, as desired, in a crucible to the heat of the electric arc, whereby there is ob- tained the desired substances in pure crystallized, amorphous or melted condition, or in the last case dissolved in a suitable metal as solution medium from which it crystallizes out, and then after the solution of the metal in an acid can be obtained in the form of hard crystals. The process can be carried out as follows: i. By a simple direct combination of the mixed materials ; 2. By the direct combination of the mixed materials in the presence of an excess of a metal which can be later removed by the aid of an acid ; 3. By preparation of the material from an oxide or a salt of one of its constituents, by reduction and solution in a metal from which it can be afterwards obtained by removal of the metal essentially as carried out above. (2) The various new hard compounds which have been pre- pared by the above process, as the silicide and boride of titanium, as well as other silicides and borides, the silico- carbide and silico-boride of titanium, and the silicides, boro- silicides and silico-carbides of other metals and metalloids, are able to scratch chromium steel and to scratch or cut rubies or diamonds, and in the pulverized form as abrasive material, glued to paper or cloth, for cutting, polishing or grinding. Moissan 1 gives more exact information regarding the com- mercial methods used in the preparation of chromium silicide. After mentioning briefly that chromium and silicon will com- 1 Compt. rend., 121, 624 (1895). SILICON COMPOUNDS. 69 bine in a reverberatory furnace in a current of hydrogen at a temperature of 1200, he turned his investigations to the elec- tric furnace. If 100 parts of chromium, free from carbon, and 15 parts of silicon are brought together in a carbon crucible, and subjected to a current of 900 amperes at 50 volts for nine minutes, there will be found a silicide with a crystalline fracture embedded in the surplus metal. Or if 60 parts of silica, 200 parts of chromium oxide and 70 parts of sugar carbon be heated together with a current of 950 amperes at 70 volts for ten minutes, there is found a well- fused, brittle and crystalline metallic mass. In the interior are found spaces filled with needles of chromium silicide. This metallic mass is coarsely pulverized and treated for a short time with cold concentrated hydrofluoric acid, until the acid attacks actively, then moderating the action by the addition of water, in order to avoid raising of the temperature and de- composition of the silicide. It must now be washed alternately with water and cold hydrofluoric acid until all action has ceased ; the residue consists of single or coherent small prisms of silicide, which often are contaminated by small amounts of silicon carbide. The chemical properties of the chromium silicide remind one of iron silicide, especially in its behavior towards acids ; if finely pulverized it is attacked by hot strong hydrochloric acid. Nitric acid has but slight effect, while aqua regia de- composes it with the formation of silica. The gaseous hy- drogen acids attack it only at temperatures lying between dark and bright red heats. At 700 gaseous hydrochloric acid transposes both the chromium and silicon into chlorine compounds. Chlorine alone acts at a red heat causing the silicide to glow brightly. Molten potassium nitrate quickly forms chrom- ate and silicate, while caustic potash acts only slowly. The silicide scratches quartz and even corundum with greatest ease; it appears that the majority of the silicides 70 CHROMIUM AND ITS COMPOUNDS. possess a much greater hardness than the corresponding car- bides. For performing the difficult analysis the silicide is fused with a mixture of 2 parts of potassium carbonate and 8 parts of potassium nitrate, and afterwards treated with hydro- chloric acid. The separation of the silica, which was made insoluble by two treatments on the sand-bath, was tedious. The presence of carbo-silicides complicated the analysis. The results found corresponded closely to the formula SiCr 2 . SiCr 2 . Chromium 80.22 79.83 80.36 78.79 Silicon 19.60 21.08 19.92 21.21 Another much softer, but with the exception of hydrofluoric acid also resistant to acids, crystallized chromium silicide of the formula SiCr s was prepared by Zettel 1 by heating in a Perrot furnace and utilizing the heat set free by the reduction of chromium oxide by aluminium, in which he melted 140 grams of copper and 140 grams of aluminium in a refractory crucible and gradually added 200 grams of previously ignited chromic oxide. The mixture was vigorously stirred, and to- wards the end some aluminium filings were added ; the silicon came from the crucible and from the stirrer of like material. In experiments to produce these silicides in the electric fur- nace, mixtures of many chromium silicides were obtained but could not be separated from each other because of their similar chemical behavior. Moissan and Zettel in their investigations used chromium in excess and likewise obtained compounds containing chrom- ium in excess. De Chalmot 2 carried on his researches in quite a different direction ; he heated chromium oxide with wood charcoal and an excess of silica in the electric arc, and obtained large needles of a gray color and brilliant metallic lustre, which were embedded in free silicon and could only be separated from it mechanically with difficulty. Cold hydro- 1 Compt. rend., 126, 833. ' 2 Am. Chem.- Journal, 19, 69 (1897). PHOSPHORUS COMPOUNDS. 7 1 chloric acid and aqua regia had no effect on it, hydrofluoric acid dissolved it, which fact was made use of for separating out free silicon. Two analyses showed a composition correspond- ing to the formula Si 2 Cr, in one of which was found 13. 84 per cent, of free silicon and in the other 4.71 per cent.; the last one had a density of 4.393. The chromium silicide is charac- terized by great tendency to crystallize even in the presence of much free silicon. In a later publication 1 the same author makes a statement regarding the behavior of the above-described chromium-sili- cide towards sulphur. Under 300 almost no reaction took place, while at a higher temperature some silicon disulphide and not free silicon (as with copper silicide, Cu 2 Si) could be detected. The affinity of chromium for silicon is therefore under these conditions somewhat greater than that of copper. C. Phosphorus Compounds. Th. Parker made the observation that in the electrolysis of chloride solutions, if chromium phosphide is used as anode, it was in general not seriously attacked by the chlorine set free, and in 1892* protected the application of chromium phosphide as anode material in the following patent claim : An anode, in cells for the electrolysis of solutions contain- ing chlorides, consisting of or containing chromium phos- phide, essentially as previously described. The patent literature gives several details regarding the preparation of the chromium phosphide ; chromium phosphate is mixed with carbon and if necessary a fluxing material, and decomposed in any furnace at a high temperature, whereby chromium phosphide is formed. Or chromium phosphate or chromium oxide and other phosphates like calcium or alumin- ium phosphate can be heated to the necessary temperature in an electric furnace. The chromium phosphate used as crude material need in no way be pure. 1 Amer. Chem. Journal, 19, 871 (1897). 2 English Patent 6,007, March 28, 1892. 72 CHROMIUM AND ITS COMPOUNDS. Should the chromium phosphide anode be attacked at very high current densities, the chromium which has gone into- solution can be very easily removed. Another process for the preparation of chromium phosphide has been proposed by S. Maronneau 1 . He had found that by the reduction of copper phosphate with carbon in the electric furnace a copper phosphide could be obtained, which was- stable at the temperature of boiling copper. This led him to believe that by the help of copper phosphide, the phosphides of the other metals and also that of chromium could be pro- duced. He used for this purpose a commercial copper phos- phide containing 90 per cent, of copper and 10 per cent, of phosphorus. He mixed 100 parts of this copper compound with 10 parts of chromium filings, which had been made by the process of Moissan. After four minutes heating with a current of 900 amperes at 45 volts, he found a somewhat crumbly metallic material with a crystalline structure from which could be isolated, by the use of concentrated nitric acid, a chromium phosphide of the formula CrP. It occurs as dull gray crystals, is insoluble in all acids with the excep- tion of a mixture of nitric and hydrofluoric acids, and at 15" has a density of 5.71. D. Sulphur Compounds. Mourlot 2 also used the electric furnace in his investigations on the production of the sulphur compounds of chromium. He first heated 100 grams of metallic chromium for from six to eight hours in a reverberatory furnace heated by coke, leading dry hydrogen sulphide over it. The metallic mass so ob- tained showed a crystalline fracture, could be easily ground to a black powder, and from appearances corresponded to the protosulphide CrS. If heated three to four minutes in an elec- tric furnace with a current of 900 amperes at 50 volts, it melted completely and on cooling formed a regulus with a beautiful crystalline fracture, and with cavities containing prismatic 1 Compt. rend., ISO, 656 (1900). * Ibid., 121, 943(1895). OXYGEN COMPOUNDS. 73 needles. Long heating in contact with carbon causes decom- position. This crystalline sulphide has a density of 4.08, and scratches quartz somewhat easily. Fluorine attacks it strongly at a low red heat ; chlorine at 340 with the formation of chromic chloride and sulphurous chloride. Bromine reacts only at higher temperatures, and iodine not at all. Oxygen acts violently at a low red heat with the formation of chromic oxide and sulphur dioxide, assuredly also SCK Also gaseous hydrochloric and hydrofluoric acids act on it, the latter even at low temperatures. Sulphuric acid reacts only when heated ; there results sulphur, sulphurous acid and chromic sulphate. Nitric acid and aqua regia decompose the chromium sulphide at ordinary temperatures. Hydrogen sul- phide is evolved by the action of steam at a red heat, and non- crystallized chromic oxide remains behind. It is easily oxidized by energetic oxidizing agents, also by molten potash. Reducing materials react only with difficulty ; hydrogen not even at 1200. On heating with carbon in the electric fur- nace there is obtained according to the time more or less molten metallic chromium containing more or less sulphur. E. Oxygen Compounds. (1) Chromous Oxide Compounds. The chromous oxide compounds have up to recently not been easily obtainable and at no time have played any im- portant part in chemical technology. They are important because of the great power of reduction they possess, so it would appear desirable to discover a convenient method of preparation in order, if opportunity offered, to be able to make use of them in technical processes. From an American patent of Villon 1 , from which we extract the following from the patent claim, it is seen that active search is being made to make practical this method of reduc- tion, but chromium is too expensive after utilizing its power of reduction for the proposed purpose, to throw aside the use- 1 Gazette U. S. Patent Office, 84, 846 (1898) ; U. S. Patent 608,652, March n, 1897. 74 CHROMIUM AND ITS COMPOUNDS. less chromic salt, so it is desired at the same time to make the chromium, by the help of the electric current, a means of continuously transferring hydrogen. The chromic oxide salt formed by the process of reduction should be reduced back in a simple manner at the cathode. The heretofore-mentioned extract of the patent claim which is concerned with the manu- facture of ethyl alcohol reads : " The herein-described process for the manufacture of ethyl alcohol, consisting in the decomposition of calcium carbide by water, by which acetylene is formed, and in the treatment of the acetylene by chromium-ammonium double sulphate, thus producing ethylene, in the absorption of the ethylene . . and finally subjecting the reducing agent to the action of the electric current either unbrokenly or in subdivided spaces. " I have found no further information. The preparation of chromous oxide salts in the solid form has been protected within a few years by Boehringer and Sons. 1 According to the patent specifications the process is carried out in the following manner : The anode space of a de- composition cell, provided with a diaphragm, is filled with a 30 per cent, sulphuric acid solution. In the cathode space is poured a mixture of 500 parts by weight of chromium sul- phate in 500 parts by volume of water and 250 parts by volume of Concentrated sulphuric acid. This is reduced by the use of lead for electrodes and a high current density of 1000 to 1500 amperes per square metre = from o. 10 to o. 1 5 am- pere per square centimetre. The chromous sulphate falls as a bluish-green, fine-grained crystalline powder, which can be separated from the adhering mother-liquor by filtration and washing with alcohol. It contains one molecule of water of crystallization, does not change in the air, and appears to be identical with the recog- nized salt 2 CrSO 4 .H 2 O which is obtained by treating CrSCK 7H 2 O with concentrated sulphuric acid and which also changes but little in the air. 1 German Patent 115,463, May 5, 1899. * Dammer, Anorgan. Chemie, 3, 546. OXYGEN COMPOUNDS. 75 It dissolves slowly in water to a pale bluish-green solution. If in place of chromic sulphate and sulphuric acid, chromic chloride and hydrochloric acid is used as cathode-electrolyte, and treated as before, chromous chloride is obtained as a green, granular crystalline mass. The patent claim reads : Process for the preparation by electrolysis of chromous oxide salts in the solid form from chromic oxide salts, charac- terized by subjecting concentrated acid chromic oxide salt so- lutions in the cathode compartment of an electrolytic cell to the electric current. In an experiment, to prepare chromous sulphate according to the directions given in the patent claim, Dr. F. Glaser ob- tained the following results in my laboratory: On electrolyz- ing a cathode solution consisting of 200 grams of Cr a (SO 4 ) 3 (calculated on anhydrous salt), 200 cc. of water and 100 cc. of concentrated H 2 SO 4 , for some time, with a current density of o. 10 to 0.15 ampere per square centimetre, there separated out solid chromous sulphate in bluish-green, thread-like crystals. They were quickly filtered and the adhering mother-liquor washed from them with a small amount of sulphuric acid (i : i) and then washed with alcohol and dried between filter- paper. If the mother-liquor in the cathode space was electro- lyzed still further, only a very small amount of fresh chromous salt was formed, although the remaining solution was still quite rich in chromic salt. It appeared meanwhile as if the migrating SO ions increased the solubility of the chromous sulphate ; the addition of concentrated sulphuric acid removed the difficulty. For the further continuance of the process, fresh chromic oxide salt was added from time to time. The chromous salt was not obtained pure at any time, but always mixed with the chromic salt, since it is very easily oxidized, and at ordinary temperatures slowly decomposes water with an evolution of hydrogen. Also during the filtra- tion and sucking dry a continual oxidation took place, raising the temperature somewhat high. In consequence the output was disappointing. 76 CHROMIUM AND ITS COMPOUNDS. In contrast to the extract from the patent literature, the salt obtained was easily oxidizable. The content of water was not investigated. Also the reference as given in Dammer, 1 is that the salt changes slightly in the air (as the salt with 7H 2 O) not that it is unaltered. For continuous electrolysis it must be observed that from time to time concentrated sulphuric acid must be added to the cathode fluid, and the anode fluid moreover diluted, or it may be partially replaced by water, or the conductivity will reach a very low value. (2) Chromic Oxide Compounds. O) BY ELECTROLYSIS OF AQUEOUS SOLUTIONS. Bunsen says that by the electrolysis of chromic salt solu- tions under certain conditions, not metallic chromium but a chromous oxide compound appears at the cathode (see page 2). Lately, however, J. Feree 2 has proven that there must be some error here. The important results of his investigation are as follows : The oxide of Bunsen is not formed when a mercury cathode is used in an acid solution of chromic chloride, no matter what current density is used. With weak current densities only hydrogen is evolved at the cathode ; again with a current density of 0.025 ampere per square centimetre of mercury sur- face no precipitate appears. If a neutral solution of chromic chloride is electrolyzed,. even at high current densities, there appears no precipitate of chromium, so that the earlier published investigations on the preparation of chromium amalgam electrolytically, from neu- tral solutions, evidently must have been in error. On the contrary there is formed in the electrolyte a black powder with the qualities which Bunsen ascribed to his chromous- chromic oxide. It can be easily obtained in large quantities if a platinum cathode is used, by the electrolysis of a solution containing 160 grams of crystallized chromic chloride per liter. 1 Loc. cit. 2 Bulletin Soc. Chim. de Paris [3], 25, 620 (1901). OXYGEN COMPOUNDS. 77 Returning to the opinion of Bunsen, it has been shown qualitatively that the material was pure chromium hydroxide ; a certain amount which had lain for some time in a vacuum over sulphuric acid was put in a tube, which was exhausted and sealed. When the lower part of the tube was heated in an oil-bath, it was observed that at 80 water vapor was evolved which condensed in the colder part of the tube ; at 400 the black contents of the tube glowed strongly and be- came green. The vacuum was not changed, for on heating one place to redness it flattened out; the change had also taken place without the admission of oxygen. From the experiment it may be concluded that no evolu- tion of oxygen had taken place during the change, yet the latter fact was proved by a separate investigation. The analysis of the material gives the formula Cr 2 O 3 .H 2 O, and the transformation consisted only in the change of the black modification of the material into the green oxide. The brownish black powder is formed also by the electrol- ysis of a neutral chrome alum solution. It is insoluble in acids but is easily attacked by nitric acid and potassium chlorate (mixture of both ?). On being heated in the air it changes to the green modification. In the above-described preparation of chromic oxide the current exerts neither an oxidizing nor a reducing influence ; it acts by forming on the cathode an excess of OH ions, that is to say,f produced alkali, which in turn reacts on the dis- solved chromium salt and forms the insoluble oxide. It is therefore clear that the electrolysis of an acid solution would be without result, and also that the material of the cathode was without influence at least as far as the description, accord- ing to which mercury or platinum cathode may be used in- differently^allows us to infer. Chromic oxide can also be prepared by using the reducing effect of the^current on an alkaline chromate solution ; this is the reaction which occurs in the spontaneous working of the chromic acid element, by which reaction electrical energy is 78 CHROMIUM AND ITS COMPOUNDS obtained. The achievement of a high current output in the technical operation of this process lies in keeping within cer- tain working conditions which G. Street has made the object of several patents. 1 If a cold alkali chromate solution is electrolyzed between platinum electrodes only a very moderate amount of oxide is obtained, the amount being.only slightly increased on heating the electrolyte up to 75. If a mercury cathode is substituted for the platinum cathode^ and the temperature of the electrolyte is kept between 70 and 80, chromic oxide is formed at once which remains sus- pended in the electrolyte. The process is carried out as fol- lows : An aqueous 10 per cent, (weight) solution of chromate is electrolyzed between a temperature of 70 and 80, pref- erably without the use of a diaphragm. An electromotive force of 4 volts and a current density of 0.05 ampere per square centimetre of cathode surface and 4 amperes per square centi- metre of anode surface should be used. It is known that the sodium amalgam formed effects the re- duction of the chromic acid to chromic oxide, though doubt- less some part of the chromic acid is transformed primarily into chromic oxide ; the advantage which the mercury shows over the platinum lies in the large overvoltage which is required to set free hydrogen from mercury cathodes. The chromic oxide separated out no longer dissolves in the soda solution at the increased temperature. A circulation of the electrolyte increases the efficiency of the reaction ; this circulation may be accomplished by heating as well as by the rising column of gas on the electrodes. The chromium content can be quantitatively separated out as chromic oxide, at the same time the whole of the sodium being obtained in the form of caustic soda solution. Bichrom- ates can be used in the place of chromates. If one starts with chromate, then bichromate can be added to the bath in such measure as the oxide is separated out ; by combination 1 German Patent 109,824, June n, 1899. OXYGEN COMPOUNDS. 79 with a corresponding amount of alkali the bichromate is changed to chromate. These additions of bichromate are limited only by the concentration of soda which is to be at- tained in the valuable soda solution by-product. Instead of using only one cell the electrolyte can be circulated through a battery of cells. The chromic oxide can be separated from the solution by simple decantation. The remark that the platinum anode could be replaced by mercury ought not to be taken in earnest. Since in this case, as was remarked, there is formed an insoluble mercury chrom- ate, which would entail a high electromotive force were it also, as remarked, possible to decompose it into mercury and chromic oxide by a mere reversal of current, but which as we will soon see is not at all the case. Patent Claim. Process for the electrolytic preparation of chromic oxide from alkali chromates, characterized by the application of mercury as cathode material. In a note Strameo 1 mentions that he has tested and con- firmed Street's remarks relative to the better working of mer- cury electrodes ; on the contrary he found that by the use of two mercury electrodes the change of mercury chromate to mercury and chromic oxide did not take place ; there ap- peared only mercury, and the simultaneous formation of chromic oxide first began after the complete disappearance of the mercury chromate. It is also to be noticed further on in the remarks, that along with the usual product, the hexahydrate, also the tetra- hydrate Cr 2 O(OH) 4 , the so-called Guignet's green is formed, which is insoluble in acids and extraordinarily valuable as pigment, but the amount of it if the directions of Street are followed is only i to 2 per cent. It is formed more abun- dantly if the solution is kept rich in alkali and poor in chromate. 1 L'Electricita 2O, 627 (1901). 80 CHROMIUM AND ITS COMPOUNDS. Finally the possibility is referred to of preparing all chrom- ium salts by adding the desired acid in the manner of the above method. For instance if a nitric acid solution of so- dium chromate is electrolyzed, the chromium will be com- pletely changed to chromium nitrate, along with which there is present in the solution only sodium nitrate. But it should be here noted that this " possibility" is not well founded, be- cause the greater part of the oxide will be again oxidized to chromic acid at the anode. Only so long as one, like Street, works with an alkaline solution in which the chromic oxide is insoluble, can the regeneration to chromate practically be neglected, in case the aid of a diaphragm is not employed. Regarding the many theoretical phantasies to be found in this article we need not pay them further attention. On the above subject there are still several interesting ob- servations made by Reese 1 on the reducing action of electro- lytically generated hydrogen on chromic acid. He arranged the two platinum electrodes one above the other, the cathode on the bottom of a burette, and the anode at the point, so that the hydrogen had to pass through the whole length of the solu- tion, while the oxygen could escape directly. Using perfectly pure chromic acid as electrolyte, and electrolyzing for twenty- four hours, with a strong evolution of hydrogen, no trace of reduction was noticeable, but on the addition of a small amount of a sulphate or sulphuric acid, reduction began in a short time, but only up to a certain amount, which was in one case 2.85 per cent, (reckoned on the whole solution). A like behavior was noticed on the addition of sulphuric acid, potas- sium sulphate and chromium sulphate. A solution of pure potassium bichromate acted exactly like the solution of chromic acid ; reduction first began on the addition of sulphuric acid or sulphates. The author thinks the fact that chromic acid may be reduced to a varying degree according to its content of sul- phuric acid or sulphates, may perhaps serve to determine the 1 Amer. Chetn. Journal, 22, 162 (1899). OXYGEN COMPOUNDS. 8 1 amounts of those substances present or the relative strength of the acids (?). It has almost the appearance as if there is here a catalytic acceleration, and it may be of interest to know that work will begin in my laboratory on a closer study of these facts. (See also the remarks of Buff further on.) () BY THE USE OF HIGH TEMPERATURES. In order to investigate the behavior of chromic oxide at high temperatures, Moissan 1 put ordinary, pure, anhydrous, pulverized chromic oxide under the electrodes of an electric furnace and sprung an arc over it. The carbon electrodes were previously heated to a high temperature in a current of nitrogen. By the use of 30 amperes at 55 volts he obtained a black, shiny wart-shaped mass, studded with groups of small dark crystals, which when treated with potassium nitrate and pre- cipitated with mercuric nitrate showed an average content of 51.9 per cent, of chromium, corresponding closely to the formula Cr a O . These crystals were very hard and left a green streak on porcelain. Molten chromic oxide combines very easily with calcium oxide, and Moissan has obtained many of these double oxides by heating a mixture of quicklime and chromic oxide in vari- ous proportions in a limestone furnace, using the purest pos- sible carbon electrodes and a current of 1000 amperes at 50 volts. Along with dark green, needle-like crystals, often sev- eral millimeters long, and which were not investigated fur- ther, were found towards the termination of the experiment, in hollows in the molten cake, small, yellow, lamellar, trans- parent crystals, which were slowly decomposed by water. The result of an analysis corresponded very nearly to the formula Cr 2 O 3 .4CaO. (See also page 3$) An experiment of Moissan is also worth mentioning, in which he added a little chromium oxide to aluminium oxide, and melted the mixture in a carbon crucible with 25 to 30 1 Ann. de Chimie et de Physique [7], 4, 136 (1895). 82 CHROMIUM AND ITS COMPOUNDS. amperes at 50 volts. By this treatment small red crystals of ruby separated from the mass, which, however, were very much less beautiful than those of Fremy and Verneuil. The operation lasted only ten to fifteen minutes, and furthermore, since at this temperature the aluminium is easily volatilized, so it appears quite possible to obtain beautiful ruby crystals in this way without much trouble. In connection with this work, Dufau 1 has later made a thorough investigation on the formation of double oxides of chromium. Moissan, in his work, made use of a " furnace of calcium carbonate for crucibles," in the interior of which he put the mixture of pure anhydrous oxide, which had just before been calcined in a Pernot furnace. In order to avoid the influence of the furnace material, the lime, the bottom of the cavity was covered with a layer of the oxide, mixed with chromic oxide, on which the mixture was poured, and which was then surrounded with a wall of those oxides. According to the investigations of Le Chatelier 2 , it was noticed that at about 900 chromic oxide changed into another form with a noticeable evolution of heat, this new form being insoluble in acids and at 440 was attacked by neither sul- phur, hydrogen sulphide, chlorine, or oxygen. This change in the ordinary chromium oxide took place in the whole of subsequent investigations. For the preparation of a double oxide of chromium-magne- sium a mixture of 150 grams of Cr a O 3 and 40 grains of MgO was put in the furnace and heated for ten minutes with a cur- rent of 30x3 amperes at 45 volts. After cooling there was found a greenish brown, crystalline, partly melted mass, which was attacked slowly by warm, concentrated hydro- chloric acid with an evolution of chlorine, which denoted the presence of chromate, which fact was also easily established by treating the broken-up mass with boiling water. If this mass is treated with nitric acid until no more dissolves, then 1 Ann. de Chimie et de Physique, [7], 12, 257 (1897). 2 Bulletin Soc. Chim. [2], 47, 303. OXYGEN COMPOUNDS. 83 the crystallized residue forms a double oxide, contaminated by lamellar chromic oxide, the whole remaining stable at the highest temperatures used. By repeated trituration (levi- gation) separation can be obtained with a fair degree of ease. There remains a dark green, nearly black, crystalline powder, which under the microscope appears to consist of small octa- hedral crystals, the thinnest being of a beautiful green color by transmitted light and without action on polarized light. Their density was 4.6 at 20, their hardness higher than that of quartz, streak bright green ; their analysis agreed closely to the formula Cr 2 O 3 .MgO. A second experiment, using 50 volts and 1000 amperes, fur- nished a brown, well-fused, very hard mass, which was only attacked by acids with difficulty ; the carefully separated crystals were also octahedral, and showed the above-described characteristics. Before analysis the finely ground powder was dried in a vacuum over sulphuric acid ; then 200 to 400 mm. of the material were placed in a platinum crucible with ten times the amount of a mixture consisting of 4 parts of potas- sium nitrate and i part of potassium carbonate, and fused. The decomposition took place very regularly in more or less time, according to the nature of the oxides. When the de- composition was thought complete the crucible was heated to redness and allowed to cool. By treating the contents with water the chromate went into solution, while the other oxide remained behind as a carbonate. With magnesium chromite especially, the decomposition was somewhat long, over half an hour. The estimation of the dissolved chromate and the precipitated magnesium oxide was performed in the usual manner. The magnesium chrom- ite showed characteristic reactions. Boiling sulphuric acid dissolved simultaneously both the chromium and the magne- sium ; likewise hydrofluoric and hydrochloric acids acted in the same way, but more slowly; on the contrary boiling nitric acid was without action, as also chlorine, bromine, and iodine. As already seen in the analysis, oxidizing agents acted only 84 CHROMIUM AND ITS COMPOUNDS. slowly ; even oxygen at redness ; on the contrary, there is always formed in the preparation of these double oxides in the electric furnace a considerable amount of magnesium chrornate. Other like combinations could not be obtained in the electric furnace in spite of variation of the current strength. For the preparation of a double oxide of chromium and calcium, a mixture of 115 grams of calcined chromic oxide and 45 grams of unslaked lime was heated in an electric fur- nace by 700 amperes at 50 volts. The crystalline product showed, besides yellow laminae, resembling those described by Moissan (see preceding), needles several millimeters long, of a dark color, and occurring in radiating groups. With 300 amperes at 45 voks only the needles and not the yellow laminae were obtained. In order to isolate the needles, the crystallized parts lying loose in the furnace were treated with concentrated hydro- chloric acid. A strong reaction took place with a simultaneous brown then green coloration of the solution and an evolution of chlorine. After this had stopped the treatment was con- tinued with boiling hydrochloric acid until no more would dissolve. Finally the adhering crystallized chromium oxide was separated by repeated trituration (levigation). The resi- due consisted of dark green, prismatic needles, with metallic luster, transparent thin sections appearing under the micro- scope a beautiful green. Their density was 4.8 at 18, their hardness close to 6, their powder bright green. The analysis gave results corresponding closely to the formula Cr z O .CaO. This double oxide is somewhat resistant to chemical reagents; fluorine will not attack it when cold ; however it will with a slight rise of temperature with a strong glowing and the evo- lution of a thick yellow smoke. The action of chlorine at a bright red heat slowly forms calcium chloride, while the chromium oxide does not appear to be affected. Bromine and iodine do not react. The calcium chromite is easily suscepti- ble to oxidation ; a part is changed to chrornate when it is OXYGEN COMPOUNDS. 85 powdered and heated to 100 in air. In oxygen it burns quickly at the temperature at which glass softens. It is unaf- fected by solutions of hydrofluoric and hydrochloric acids and by concentrated nitric and sulphuric acids. Anhydrous hy- drofluoric acid forms calcium fluoride and amorphous chromic oxide at a red heat, hydrochloric acid gas performing a simi- lar reaction with glowing. Of oxidizing agents potassium chlorate is the most effective. Collectively, the earlier investigations of Moissan can be stated thus : With the application of 300 amperes at 45 volts the chromite, Cr a O 3 .CaO, is formed; with 700 amperes at 50 volts, and so at a higher temperature a second compound is also formed, Cr a O .zj-CaO, which predominates when 1000 am- peres at 50 volts is used. Strontium chromite is not stable at high temperatures ; it appears to be more easily oxidized than calcium chromite. If an intimate mixture of strontium oxide and chromic oxide is subjected to the heating action of an arc of 50 amperes at 50 volts or one of 300 amperes at 55 volts, a green crystalline mass is obtained, which appears to consist of only chromate and crys- tallized chromium oxide. Stronger electric currents have the same effect. The following, on the contrary, were the results with barium oxide, in which a very high chromium-content double compound was formed. On a ten-minute treatment of a mixture consisting of equal parts of both oxides, with a cur- rent of 300 amperes at 50 volts, there was found in the fur- nace a fused green mass with a crystalline fracture, and which was, as before, treated with hydrochloric acid; a lively reaction also took place here, coloring the solution brown and then green, with an evolution of chlorine. After treating until no more would dissolve, there remained a mass of crystals. If the solution is shaken up, dark shining crystals of the double oxide first settled down and on them the green laminae of crystallized chromic oxide; by repeated trituration they could be separated. This double oxide formed small, black, shining 86 CHROMIUM AND ITS COMPOUNDS. crystals, which made a brown powder of a slightly greenish shade ; under the microscope they appeared transparent and hexagonal. They were somewhat harder than quartz; at 15 their density was 5.4. They were unattacked by hydrofluoric hydrochloric, nitric, or sulphuric acids. Chlorine and bromine acted slowly at red heat with the formation of the corresponding barium haloid, while the chromium oxide did not appear to have been affected ; iodine had no effect. In oxygen barium chromate was formed below a fed heat with a bright incandescence; it was on the whole easily oxidizable like calcium chromite. Water and sulphur vapors were without effect even at a bright red heat ; anhy- drous, gaseous hydrofluoric and hydrochloric acids reacted very slowly at redness with the formation of the respective barium compounds. The results of an analysis showed the compound to agree very closely with the formula 4Cr 2 O 3 .BaO. (3). Chromic Acid Compounds. O) CHROMATES OF THE HEAVY METALS. An American patent 1 has been granted to Arthur B. Browne, in which the electric current is used in so far as concerns the preparation of the necessary alkaline chromate. The patent itself was not accessible to me, so a reproduction of the patent claims must suffice. 2 Patent Claim. (i) The process of manufacturing chromate of lead which consists in first producing a solution of a soluble salt of lead, as the nitrate, acetate, or chloride, simultaneously preparing a solution of chrome alum, precipitating chrome hydrate therefrom, mixing said chrome hydrate with an excess of caustic alkali to redissolve it, mixing the resulting strongly alkaline solution with a solution of common salt, passing an electric current through combined solutions to produce a mix- ture of bichromate and chromate of potash, mixing said com- 1 U. S. Patent 538,998, April 18, 1894. * Official Gazette of the U. S. Patent Office, 71, 871 (1895). OF THE UNIVERSITY ^k OF OXYGEN COMPOUNDS. *| 87 bined bichromate and chromate of potash with said solution of a soluble salt of lead to precipitate lead chromate and wash- ing it with water, refiltering it to remove the water, and finally drying the resulting product. (2) The process of manufacturing chromate of lead which consists in first producing a solution of a soluble salt of lead, as the nitrate, acetate or chloride, simultaneously preparing a solu- tion of chrome alum, precipitating chrome hydrate therefrom, filtering said chrome hydrate and washing it, then mixing said filtered chrome hydrate with an excess of caustic alkali to re- dissolve it, mixing the resulting strongly alkaline solution with a solution of common salt, passing an electric current through said combined solutions to produce a mixture of bichromate and chromate of potash, mixing said combined bichromate and chromate of potash with said solution of a soluble salt of lead to precipitate lead chromate, filtering said lead chromate and washing it with water, refiltering it to re- move the water and finally drying the resulting product. (3) The process of manufacturing chromate of lead, which consists in first producing a solution of a soluble salt of lead, as the nitrate, acetate or chloride ; simultaneously producing a solution of a mixture of bichromate and chromate of potash, by passing an electric current through an alkaline solution of chrome hydrate ; mixing said solution of a mix- ture of bichromate and chromate of potash with said solution of a. soluble salt of lead, to precipitate lead chromate ; filter- ing said lead chromate and washing it in water, refiltering it to remove the water, and finally drying the resulting product. In the patent specifications of one English and two German patents of C. Luckow, which are concerned with the electro- lytical production of difficultly soluble salts and oxides, we find special data upon the manufacture of neutral and basic lead chromates. We will examine the patents chronologically, and take out of them what would interest the electrochemist. In the first patent 1 Luckow will practically utilize u the theory of electrolytic and hydrolytic dissociation " in the 1 German Patent 91,707, Dec. 4, 1894. 88 CHROMIUM AND ITS COMPOUNDS. manufacture of difficultly soluble salts or oxides, from solu- ble metallic anodes ; he employs for this purpose only electro- lytes, which, as he says, may be considered as dissociated either electrolytically or hydrolytically or electrolytically and hydrolytically. The maximum of dissociation appears to be different for different salts, fluctuating, however, for the in- vention concerned somewhere between 0.3 and 3 per cent, of dissolved salt (calculated on anhydrous salt); the use of 1.5 per cent, of dissolved salt usually suffices. Better results were obtained in this way with the use of considerably less current consumption than was possible with the electrolytic processes hitherto proposed for the same or similar purposes. Luckow apparently only slightly understood this new theory at that time ; the maximum of dissociation of the salts in question (sodium chlorate, sodium carbonate, etc.), is at in- finite dilution, while the maximum of conductivity in the same case, if one chooses them for the maximum economy of energy, lies with the much higher concentration. Further Luckow says that the desired reactions take place much better if a mixture of salts is used, and mixtures of about 80 to 99.5 of the one to 20 to 0.5 of the other by weight are the most suitable. For the preparation of salts in general, a salt which forms an easily soluble compound with the anode, composes 80 per cent, of the mixture, and the re- maining 20 per cent, of such acids which will form a diffi- cultly soluble combination with the anode material. In most cases salts of the alkalies (ammonia included) and the alkaline earths are suitable. The electrolyte should be "as far as possible neutral, that is, slightly acid, slightly alka- line, or neutral," and its composition should be kept as nearly constant as possible ; also the dissolving anode should be re- newed when the consumption has reduced it to 1.5 milli- metres thickness. " For electrodes there are used for each square metre of anode surface eight positive and nine negative electrodes, which are hung 15 mm. apart in a space 200 mm. high and OXYGEN COMPOUNDS. 89 300 mm. wide, and whose method of suspension allows a sur- face contact of 4 square metres with the electrolyte. The material of the anode varies according to the desired product; the material of the cathode is preferably the same as the anode, but it can, however, be different from it. It is advis- able to hang the cathode in a bag of thin woven material. " The electromotive force required depends on the mix- ture of salts ; for weak acid electrolytes it is between 1.3 and 1.5 volts, for neutral about 1.8 volts, and for feebly alkaline about 2 volts." About 0.005 ampere per square centimetre is recommended as the current density. The temperature should not rise above 50. To avoid interruptions in running, the purest possible materials should be used. The following special directions are given for the manufacture of lead chromate : Neutral Lead Chromate. For electrolyte use a i y 2 per cent, aqueous solution of a mixture of 80 parts by weight of sodium chlorate with 20 parts by weight of sodium chromate. The anode is made of soft lead, the cathode of hard lead. The electrolyte is neutral. The tension is 1.8 volts. The current density is 0.5 ampere per square decimeter. During electrolysis the electrolyte is kept neutral, and water and chromic acid must be carefully introduced. Acid Lead Chromate. The electrolyte consists of a i y 2 per cent, aqueous solution consisting of a mixture of 80 parts by weight of sodium chlo- rate and 20 parts by weight of sodium bichromate. The electrolyte is slightly acid. The electromotive force is 1.5 volts. The current density is 0.5 ampere per square decimeter. During electrolysis the electrolyte is kept feebly acid and water and chromic acid must be carefully added. The patent claims in question read : (i) Process for the production of salts and oxides, which are insoluble or soluble with difficulty in water or dilute, 90 CHROMIUM AND ITS COMPOUNDS. neutral or slightly acid or slightly alkaline salt solutions, by means of electrolysis from a suitable metallic anode, charac- terized by the use of electrolytes, which have been brought into the condition of electrolytic or hydrolytic, or hydrolytic and electrolytic dissociation near the neutralization point, and by supplying the materials separated out in the electrolysis to keep the composition of the bath as near as possible constant, for the purpose of exerting a good influence on the consump- tion of current and course of the reactions. (4) The process characterized by claim i applied to the production of neutral lead chromate as follows : A neutral solution of a mixture of sodium chlorate and sodium chromate is subjected to electrolysis between lead electrodes, with a cur- rent density of 0.5 ampere per square decimeter, with addi- tions of water and chromic acid. (5) The process characterized by claim i applied to the production of acid lead chromate as follows : A slightly acid solution of a mixture of sodium chlorate and sodium bichrom- ate is subjected to electrolysis between lead electrodes, with a current density of 0.5 ampere per square decimeter, with additions of water and chromic acid. In the second patent 1 the process in question is elaborated. In the previously patented process, the electrodes, particularly the anode, was made of massive material, and the insoluble or slightly soluble compounds formed did not stick on the anode during the operation, as they were formed, but they rolled off the surface of the anode in dense clouds into the depths of the bath, where they collected together on the bot- tom in the form of an extremely fine powder ; concerning the obtaining of chemical compounds, it is just the opposite of that which is desired in the operation of the manufacture of the electrodes for electrical accumulators. In the improved process metallic conducting frames were prepared in such a manner as to hold on them insoluble or slightly soluble oxides, salts, metals or ores, in a finely divided state as the 1 German Patent 105,143, Sept. 4, 1895 (compare also the supplemen- tary German Patent 99,121, May 13, 1897). OXYGEN COMPOUNDS. 91 soluble material, so that the insoluble or slightly soluble end- products would remain upon or in these electrode frames dur- ing the operation. The preparation of the electrodes and the operation of the process is the same as in the preparation of electrodes for storage cells whose ribs and grating frames are filled with metallic dust, or coated with insoluble or difficultly soluble oxides or salts as the effective (active) material. For electrodes a special woven wire net with the narrowest possible meshes was employed ; meanwhile the accumulator industry has invented a number of new practicable forms of electrodes which can be used for this purpose. To protect the metallic ribs in case they are made of a metal easily alterable, it is possible to polish them with graphite or to platinize them. After the pasty mass has been kneaded or pasted into the grooves, it is surrounded with a tightly stretched bag of filter cloth or like material. The electrodes should be spaced 20 millimeters apart. Concerning the electrolyte, according to patent No. 91,707, one anion causes the solution of the massive anode, forming an easily soluble salt of the anode metal, the other transforms the anode metal into the desired insoluble or difficultly solu- ble compound ; " it is self-evident that the first anion performs the function of preventing the precipitated insoluble or slightly soluble compound formed on the anode from adher- ing to it. Which constituent must have predominating in- fluence is determined by the nature of the product to be ob- tained." This remarkable fact deserves explanation. How is it for instance that by the electrolysis of a mixture of sodium chromate and chlorate between lead electrodes a beautiful precipitate of lead chromate is formed in the electrolyte, and the anode dissolves easily, while the solubility of the anode weakens and it is finally covered over with an adhering coat when the pure sodium chromate solution alone is used? 92 CHROMIUM AND ITS COMPOUNDS. Perhaps this action depends on the increase of solubility which lead chromate would certainly show in a solution con- taining a large excess of sodium chlorate. It is accomplished by the formation of a non-dissociated lead chlorate. This in- creased solubility allows a somewhat relatively increased super- saturation of chromate of lead, and so prevents the formation of a dense adhering coat of lead chromate directly on the anode. With these explanations, advanced reservedly, agrees the fact that the current density is normally low ; the formation of an adhering coat on the anode is to be feared using a high density. An exhaustive experiment which meanwhile has been worked in my laboratory, has furnished another explana- tion for the above phenomena. The results may be found in Zeitschr. f. Elektrochemie, p. 255 (1902). In the more generalized process the salt forming the easily soluble compound does not predominate so strongly, it only composes one-half to two-thirds of the mix- ture of salts. The temperature is usually kept below 50, without, however, excluding higher temperatures. The current density is, as in the older process, preferably about 0.005 ampere per square centimetre, but the electro- motive force is higher, lying between 2 and 4 volts. Finally there remains several words to be said on the gen- eral method of operation, etc., in so far as they were not stated in the first patent specification. During the electrolysis and the separation of the ingre- dients out of the solution to form the desired product, water, chromic acid, etc., must be added in such a manner that the concentration and the neutrality of the solution is not ap- preciably changed. On the other hand, the accumulation of impurities and deleterious substances is to be avoided. The introduction of a current of air or steam in fine division producing a good stirring without bringing the elec- trolyte into too strong agitation has a good effect. Basic lead chromate (and lead) may be prepared in the fol- lowing manner : OXYGEN COMPOUNDS. 93 " The electrolyte used is a i */ to 2 per cent, aqueous solution of a mixture, consisting of two-thirds sodium chlorate and one-third sodium chromate. The electrolyte is made slightly alkaline with sodium hydroxide. The anode frame consists of platinized hard lead and the cathode frame of either hard or soft lead. The filling for the anode frame is composed of lead dust or litharge, that of the cathode frame of red lead or Phcenicochroite (basic lead chromate). Should the cathode charge not liberate during the electrolysis a sufficient amount of chromic acid (which may be expected) this must be added to the electrolyte." The difficulty with this process plainly lies in converting the lead powder or litharge completely to chromate ; small pieces are completely surrounded by chromate and insulated from further action. At least to convert this final portion a disproportionate current consumption is necessary, if a pure material is desired. Also many difficulties are introduced in the same process in making a product of a certain sized par- ticle, as demanded by the trade, upon which depends the shade or the facility of reaction, as with lead peroxide ; this condition is considered in patent No. 91,707. Patent Claim. The extension of the process protected in patent No. 91,707 to the manufacture from finely divided metals and ores of insol- uble or slightly soluble oxides or salts at both poles, thereby characterized by the filling or spreading out of these materials in or on electrode frames of metallic conducting material and subjecting them to the action of the electric current in a highly dilute electrolyte conformable to patent No. 91,707. The English patent 1 corresponds to the two German patents in contents. Some special data is given on the preparation of lead and copper chromates. The color of the lead chromate depends on the bath ; a neu- tral electrolyte, consisting of a solution of alkaline nitrate, acetate, or chlorate with alkaline chromate or chromic acid 1 English Patent 14,801, Aug. 6, 1895. 94 CHROMIUM AND ITS COMPOUNDS. yields with lead anodes a beautiful chrome yellow, which be- comes brighter if the solution begins to become acid, and redder if it becomes alkaline. Slightly acid solutions yield, therefore, a fiery citron yellow, and slightly alkaline solutions, a beautiful chrome orange, which changes to a chrome red with increasing alkalinity. For the production of the bright chrome yellow, care must be taken that the solution at the anode does not become too acid, or this may become covered with an insoluble coat of oxide or peroxide ; the gradual addition of the necessary amount of chromic acid or chromate is to be recommended. For the production of chrome orange a mixture of potassium perchlorate and chromate is used. Chrome red is obtained by the addition of more alkali. The cost of manufacture of 100 kilograms of pure chrome yellow is $17.50 to $20*00, and of 100 kilograms of pure chrome orange about $12.50. Copper chromate can be pro- duced in the same manner by using copper anodes. The electrolyte consists of a mixture of either an alkaline nitrate^ sulphate, chloride, or chlorate with chromic acid or a soluble chromate. If the solution is neutral or slightly acid a brownish red chromate is precipitated, if feebly alkaline, a green. By increasing the acidity of the solution, the brownish red will become brighter, until finally with too much acid the copper no longer precipitates. With increasing alkalinity the green color becomes deeper but much duller. A very dilute solution of a mixture of six parts of sodium chlorate with one part of sodium bichromate is suitable for the preparation of the red chromate, while for the preparation of the green chromate a dilute solution of a mixture of 4 parts of sodium or potassium chloride with i part of sodium chromate, to which before the beginning of the electrolysis a little alkali is added, can be recommended. The necessary chromic acid for the operation of the process must be con- tinually supplied. OXYGEN COMPOUNDS. 95 The current density must be so regulated that no free oxy- gen is evolved on the anode ; with too high current densities the desired product is not precipitated fine enough. The amount of free alkali or free acid should not be above i y? per cent, by volume of the solution, and the concentra- tion of the salt should remain between 0.3 and 3 per cent, by weight. Great stress is laid on these conditions. A greater dilution is not advisable because of the smaller conductivity of the solution ; a lesser is likewise not, because in those kinds of solutions, particularly if they contain too much alkali or acid, the anodes are oxidized " with quickly decreasing rapidity and more superficially and irregularly ; " moreover, the product suffers and tends to become crystalline. The product is obtained as a pure, amorphous and finely divided powder only when the solution is made up within the above given limits. Patent Claim. (1) The herein-described process for the production of in- soluble or practically insoluble oxides and salts from very dilute neutral, slightly acid or alkaline solutions essentially as described. (2) In the electrolytic production of insoluble or practically insoluble salts or oxides, the simultaneous use of an anode of the desired metal or containing it, and very dilute neutral, or slightly acid or alkaline electrolyte, essentially as described. As far as I know, in the attempts to manufacture chrome colors on a large scale by these methods, serious technical difficulties have arisen. Whether their solution has been un- successful or if finally the processes have proven unprofitable, I know nothing of there lately having appeared on the market large quantities of electrolytically prepared chromium colors. (b) CHROMATES OF THE ALKALI METALS AND CHROMIC ACID. Already in the year 1886, F. Fitzgerald attempted to solve the problem of transforming chromic oxide compounds into chromic acid compounds by the aid of the electric current, in 96 CHROMIUM AND ITS COMPOUNDS. a cheaper way than formerly. Evidently he had in view the treatment of the chromic oxide solutions found in chromic acid cells after their exhaustion, in that he wished to produce simultaneously at the cathode metallic zinc from a part of the zinc sulphate occurring in the solutions. In the patent spec- ifications 1 is given several detailed statements. A solution containing zinc and chromic sulphate (or chromic chloride) is electrolyzed preferably by the use of lead anodes, although all other less attackable conductors as platinum and carbon (? !) are not excluded. Lead, zinc, or the like can be used as cathodes, and are usually suspended in an acid zinc sulphate solution or a solution of ammonium alum in which zinc oxide is dissolved. A diaphragm separates the anode and cathode solutions. The electrolysis is continued until the dark-colored anode solution has become bright red. The metallic zinc precipitated on the cathode can be melted and in this way separated from the chromic oxide precipitated there at the same time. The proposed separation of zinc in this manner must have been attended with considerable difficulty and expense. In order to avoid the inconvenient transportation of large quantities of fluid, the solutions were, after exhaustion, run into a large reservoir, precipitating therein chromium and zinc oxides by lime, magnesia, or the like, the supernatant liquid run off, and the solid residue taken to the chromic acid regeneration plant. There the oxides were mixed together and either dissolved together by a dilute mineral acid, or first treated with ammonia or a boiling solution of fixed alkali, to bring the zinc into solution, and then to dissolve the chrom- ium, a weak acid or the cold solution of a fixed alkali was poured over. The solutions were then respectively oxidized and reduced in the following manner. The patent claim reads : ( i ) The herein-described process for the electrolytic regenera- tion or production of dissolved chromic acid from chromic 1 English Patent 5, 542, April 21, 1886. OXYGEN COMPOUNDS. 97 salts, which result by the reduction of this acid when it has been used for the production of oxygen for galvanic elements, or chlorine for bleaching or other purposes. (2). The herein described process for the electrolytical re- generation or production of dissolved chromic acid from the exhausted or reduced solutions of this acid produced in the manufacture of chlorine or oxygen, consisting in treating the solution with a base which precipitates the chromic oxide and also zinc oxide, redissolving the precipitated oxide and converting it into chromic acid by electrolysis. A note of K. F. Smith 1 shows the preparation of alkali chromates directly from chromite, in which the chromite is intimately mixed with molten potash and the mixture elec- trolyzed (see also page 36). Although he has used this method only for the analytical determination of chromium in chromite, the method is of great practical interest and serves as an intimation of the production of chromates by the aid of the electric current on a large scale. In the first experiments a current of one ampere was used. After fifteen minutes the oxidation of the mineral appeared complete. A stronger current worked poorly: "Among other things, a stronger current acted principally on the iron oxide in the ore and affected the potash, so that a larger or smaller amount of the metal concerned was precipitated. The above-mentioned phenomena appeared after a partial de- composition of the mineral so that some chromite always re- mained undecomposed." Finally good results were obtained with the following method of operation : 30 to 40 grams of caustic potash were melted in a nickel crucible i T/Q inches high and 2 inches in diameter, and heated until the excess water was driven off. The crucible was then placed on a heavy copper ring, which was connected to the anode of a battery, the finely ground mineral (o.i to 0.5 gram) was placed upon the fluid potash and a platinum rod was put in for cathode. The 1 Berlin, Ber., 24, 2182 (1891). 9$ CHROMIUM AND ITS COMPOUNDS. crucible was kept covered during the analysis, and warmed gently. Before stopping the determination the current should be reversed, because metallic iron is precipitated on the cathode and surrounds small particles of the ore, protecting them from being attacked. By reversing the current the lat- ter were set free and exposed to oxidation. We have already shown that it is not possible to use more than one ampere for the above quantity of ore ; after thirty to forty minutes the ore is decomposed, even when it is very resistant. When decomposition is complete hydrochloric acid will dissolve the contents of the crucible completely. Other- wise the decomposition is not complete. The chromate can be determined in the aqueous alkaline filtrate by the usual methods. A treatise by C. Haussermann 1 is concerned with the electrolytic oxidation of alkaline chromic oxide solutions. Haussermann puts a solution of chromic oxide in an excess of concentrated caustic soda solution, in a porous clay vessel, forming an anode compartment, and then puts the latter in a glass vessel filled with water, which forms the cathode space. The anode is a platinum sheet, the cathode an iron sheet each 7 X 12 square centimeters effective surface. With a current strength of 2 amperes the electromotive force finally dropped to 5 volts. Sodium chromate and some oxygen were formed at the anode, while at the cathode caustic soda and hydrogen. The investigation was performed without external heating and was interrupted before the complete oxidation of the chromite and showed a yield of 0.563 grams of the original sodium compound transformed into the higher oxide form per ampere hour, corresponding to a current output of 42 per cent. Further, this method of procedure was applied to the pre- paration of bichromates from neutral chromates, in which the anode space was charged with a solution of 58 grams of Na 2 CrO 4 in 0.5 liter of water. With a current strength of 1 Dingler's Polytechnisches Journal, 288, 161 (1893). OXYGEN COMPOUNDS. 99 between 2 and 3.5 amperes the electromotive force dropped finally to 6 volts ; after eight and one-half hours the experi- ment was interrupted. An evolution of oxygen and an odor of ozone was noticed at the anode, and at the cathode a strong evolution of hydrogen ; the cell was not warmed. The solu- tion in the anode space after some time became red, that in the cathode space yellow ; this last appearance was "ascribed to a wandering or diffusion of the neutral chromate." To prevent misunderstanding it was remarked that the chromate ion, CrO 4 , which is negatively charged travels toward the anode under the influence of the current. Only by diffusion can a small amount of chromate get into the cathode space. An analysis showed at the end of the electrolysis in the cathode solution 14 grams of caustic soda and only 0.4 gram of neu- tral chromate ; the anode liquor after evaporation left 40 grams of crystallized Na 2 Cr 2 O 7 .2H 2 O. In case the soda liquor is returned to the circuit of manufacture to lixiviate the roasted material the chromate admixture does no injury. Excess of caustic soda which was formed in the previously used lixiviation of the roasted material containing calcium chro- mate by aqueous soda solution, does not act injuriously, and therefore the solution obtained by lixiviation of the roasted mass and containing sodium chromate along with caustic soda, can after concentration without further preparation be sub- jected to electrolysis and worked over into bichromate. To be quite clear, let the old process for the preparation of bichromate be briefly compared with the electrolytic process of Haussermann. It should be first noticed that Hausser- mann left undecided the practical utilization of the transfor- mation of sodium chromite into chromate in the electrolytic manner, and was concerned only with the conversion of chromate into bichromate. According to the old methods of Haussermann as we have recently described, the roasted material containing the calcium chromate is treated with an aqueous solution of sodium car- bonate. 1 Often sodium sulphate solution is used, thus pre- 4 may have been employed. 100 CHROMIUM AND ITS COMPOUNDS. cipitating insoluble calcium carbonate and forming soluble sodium chromate, along with some free caustic soda. The solution is concentrated after nitration, and just enough sul- phuric acid is added to neutralize the caustic soda, and trans- form the neutral chromate into bichromate. With the proper concentration conditions the greater part of the sulphates are precipitated in the anhydrous form while the bichromate re- mains in solution. According to the new process the alkaline chromate liquor is placed in the anode compartment of a cell, and the cathode space is filled with pure water. 1 By electrolysis bichromate is formed, and on the cathode side caustic soda solution, of which not an inconsiderable part gets over to the anode side on account of the large velocity of migration of the OH ions. After the completion of the electrolysis the cathode solu- tion or at least a part of it shall find application in the leach- ing of the roasted material in place of a part of the soda so- lution, where it then transforms the calcium chromate into calcium hydrate and dissolved sodium chromate, which latter is again placed in the anode compartment of the electrolytic cell for further oxidation. The bichromate solution in the anode compartment, com- pletely free from sulphates, may be used directly for oxidation or worked up into solid bichromate. It is perhaps advisable to explain just what part the electric current plays in the transformation of chromates into bichro- mates in the anode compartment. Aside from some acces- sory processes we do not find an oxidation taking place, but oxygen is evolved quantitatively. Thereby the electrolyte would become acid, and we know that acid hinders the re- action. The current decomposes, therefore, in this view, neu- tral salt only into base and acid. Incidentally it may be re- marked that according to the investigations of Ostwald 2 chromic acid has the formula H 2 Cr a O 7 . 1 Naturally not to be recommended in practice ; it is customary to use a not too concentrated solution of sodium carbonate. 2 Zeitschr. f. Physikalische Chemie, 2, 78 (i! OXYGEN COMPOUNDS. IOI Haussermann has claimed a great deal when he said : 41 After the preceding, there can be no doubt that the applica- tion of electrolysis to the manufacture of alkali bichromates involves an important step forward, and that the process de- scribed, which is protected by patents, will become an im- portant factor in the commercial world." Nothing has been done with the patent, and there is noth- ing known of a practicable application of the process. The electrolysis of the alkali chromates and the like is, as mentioned before, not new, for Buff J decomposed dissolved neutral potassium chromate in 1856. He was influenced to take up these experiments because of a statement of Geuther, who claimed to have found that by the electrolysis of aqueous solutions of pure chromic acid, hydrogen and chromium were precipitated at the cathode, and oxygen at the anode to the extent of a third or more greater than should be expected from the amount of electricity passing. Buff could not sub- stantiate these claims ; the oxygen evolved always corre- sponded to Faraday's law and also the hydrogen when using a ten per cent, solution of chromic acid. By the electrolysis of a five per cent, solution of pure chromic acid reduction in- versely proportional to the current density was observed. The addition of sulphuric acid increased the reduction. By the electrolysis of acid potassium chromate, the same phenomena was observed as with chromic acid. These results agreed in part with the much later ones of Reese (see page 80). The formation of bichromate at the anode by the electroly- sis of a neutral chromate was not established ; it can only be supposed to have taken place from the remark : 2 " the solution in the region of the positive pole gradually became red and gave an acid reaction." 1 Ann, derChemie und Pharmacie, 1O1,-I (1857). 2 In an article by Merges [Cotnpt. rend. 87, 15 (1878)] the remark is found that by the electrolysis of potassium chromate, bichromate is formed at the anode ; and it was also noticed that chromic acid in a dilute solution was partly reduced by the electric current to chromic chromate. 102 CHROMIUM AND ITS COMPOUNDS. Potassium bichromate, according to the method of R. Lorenz, 1 can be made directly from ferrochrome, by the help of the electric current. He made the ferrochrome the anode in a caustic potash solution, and for cathode used a porous copper oxide plate. On the application of two volts terminal tension, dense red streams ran down off the anode and oxygen was not evolved, even with a considerable rise of tension. Since the ferrochrome consisted of about equal parts of iron and chromium, it might be expected that along with the red chromate potassium ferrate would be formed. This was not the case. The clear red color proved that ; the iron col- lected as hydroxide on the bottom of the vessel and was scarcely "detectable in the solution. Two years later, Heibling obtained a patent in England, for in the main the same idea as Lorenz had. 2 He used as anode an alloy of iron and 'chromium containing at least 55 per cent, of chromium, and a carbon or iron cathode in an electrolyte consisting of alkali salts, preferably an alkaline chloride. He surrounded the electrodes with bells for the separate collection of the hydrogen and chlorine, which later formed the materials for the manufacture of pure hydrochloric acid. The amount of chlorine set free at the anode corre- sponded to the loss of current occurring in the manufacture of chromate. The iron of the anode was changed to oxide. The electromotive force required for the electrolysis is 1.5 volts, and to produce i kilogram of crystallized sodium bi- chromate (Na 2 Cr 2 O 7 -f 2H 2 O) required an energy expenditure of two kilowatt hours. For the preparation of i kilogram of crystallized sodium bichromate there was consumed about 1 100 ampere hours, corresponding to a current output of 80 per cent. It is also worth noticing that in case the anode contains less than 55 per cent, chromium, both iron and chromium will be simultaneously oxidized without the formation of 1 Zeitschr. f. anorgan. Chem., 12, 396 (1896). 2 English Patent 4,624, Feb. 24, 1898. OXYGEN COMPOUNDS. 103 chromate. If the alloy contains crystals (of nearly pure chromium ?) in its interior, at the end of the operation these can be fused together with the use of graphite into a valuable by-product ; in this way chromium practically free from iron can be obtained. The patent claim following will furnish the remaining in- formation. 1. Apparatus for the electrolytic production of alkali bichromates, consisting of electrolytic cells, containing the solution of an alkaline salt, as a chloride, nitrate or caustic alkali, and anodes of chromium or a chromium-iron alloy with a content of over 55 per cent, of chromium and any suitable cathodes. The cells are arranged in a descending series so that the electrolyte can flow from one cell to the other, be- cause of this difference of level, essentially as described. 2. Apparatus for the electrolytic production of alkaline bichromates, consisting of electrolytic cells, containing a solution of an alkali salt as a chloride, nitrate or caustic alkali, and anodes of chromium or a chromium-iron alloy con- taining over 55 per cent, of chromium and any suitable cathodes. The individual electrodes are surrounded by sepa rate collecting bells, which are intended for the collection of the gases chlorine and hydrogen at the two poles to utilize them for the production of pure hydrochloric acid, essentially as described. 3. In the apparatus described for the production of alkaline bichromates electrolytically, the arranging of a number of carbon anodes either in each cell or in each division of many cells for the purpose of facilitating the separation of chlorine from the electrolytes. With the exception of Fitz-Gerald (page 95) all inventors have attempted the production or regeneration of chromates in alkaline solutions. Since at the present time large amounts of sodium chromate in sulphuric acid solution are used for oxidation, the preference for performing the regeneration in alkaline solution would add the loss of sulphuric acid to that of the soda. 104 CHROMIUM AND ITS COMPOUNDS. These and other disadvantages (high tension) were as stated in the patent claim 1 of the Hochst Color Works " to be avoided by a rational process which in a large plant has already been proved to be of high commercial importance. Starting with a solution of the chromium salt in the desired acid, as chrom- ium sulphate in sulphuric acid, and always with the solu- tion of the highest electrical conductivity, 100 grams of Cr o O 3 and 350 grams of H 2 SO can be diluted to a liter. Both the anode and cathode spaces of the lead : lined vessel, provided with a diaphragm, are filled with the solution (lead plates serve for electrodes). By the action of the current chromic acid is formed on ?he anode side, and hydrogen is evolved on the cathode side. A change takes place in the concentration of the sulphuric acid on both sides, decreasing on the cath.or side and increasing on the anode side. The oxidized so* ,^10^ can be used in manufacturing without further treatment, going back again to chromium oxide, when it comes back to the cathode side, the old cathode liquor being now placed in the anode compartment. The cathode solution has by this second operation become richer in sulphuric acid than the anode solution, but on the passage of the current this excess is transferred from the first to the second. NQ f. umulation of sulphuric acid, etc., takes place in this cyeir operation in which the solutions are brought alternately jn the anode and cathode compartments, and these solutions can be, used for a long time as an excellent carrier of r^vgen, with practically no loss or change in composition. The electronic? ve force with a current density of 0.03 ampere per square Centimeter is not quite 3^ volts at 50." '--., The patent claim reads : Process for the regenejff- e ^i of chromic acid fro pi solutions of oxidized chromium --jo ccmy an elects 1 y tic methoc 1 ")iar- acterized by the use of I, o gra on of a c 2 O 3 <iiic c :ide sa ts corresponding acid, as ci ' ; ed. ..ulplr nnot* sulphuric aciu, as 1 German Patent 103,860, June 12, 1898; identic?- 1 With the American Patent 630,612, which bears the lames of Le Blanc a tfReisenegger. OXYGEN COMPOUNDS. IO5 both anode and cathode solutions, but alternated in this way, that the solution containing the chromic acid after passing through its stage of usefulness and having been converted to chromic oxide again, is returned to the cathode side, the previous cathode solution being placed in the anode compart- ment. An interesting article by Regelsberger, on the regeneration of chromic acid from materials containing chromic oxide, ap- peared in 1899 in the Zeitschrift fur angewandte Chemie* After a short review of all the previously proposed processes for the manufacture of chromic acid, Regelsberger described some investigations he himself had made on the electrolytic regeneration of chromic acid. ^e first thought to oxidize solid residues containing chromic oxiuc in an alkaline solution, by the oxidizing influence of chloride of calcium, similarly to what Dercum 2 had done for the chromium solutions. The results were unfavorable, and it appeared advisable to combine the. formation of the oxidation reagents (of the chloride of lime) and the regeneration of the chromic acid in one process, so a solution of chloride was electrolyzed ? an auxiliary electrolyte, and in this solution chromium h oxide was slowly introduced. With a suffi- ciently lon^ sSage of the current he found that a complete transformation into chromate of the added chromic oxide had taken place, while a corresponding oxidation was not to be detected when using --. 1 kaline sulphates ; the reaction only began c >r the addition of a certain amount of chloride. By the-'addition of chromic" chloride alone only chlorine and metallic Chromium resulted ; but on the addition of a larger amount of chloride oxid^tlfvn^tQ chromic acid began. From this was crutful, unless l^Lrin the best current out- put^ .;* much clrsually present a sim, continuously dissolved i almost .- , it might be used in s .ine chlorides in amount equal to th/e chromium resi^nes a' given time interval, in 1 Page i:r the purpose o r 2 Engli: ou t of the S0- b - T 5, 1898. 106 CHROMIUM AND ITS COMPOUNDS. doing which naturally the chromic oxide must be well dis- tributed." The oxidation product was principally bichromate and may be allowed to accumulate in the solution without decreasing greatly the current output, although doing so perceptibly. By the use of KC1 as the auxiliary electrolyte, potassium bi- chromate can be finally crystallized out in the purest form from the hot solution, while in the case of the sodium bichro- mate solution the largest part of sodium chloride must be pre- viously removed by evaporation. By not quite clear theoretical considerations, Regelsberger attempts then to prove "that with equal returns in formation of chromate the highest current consumption (in ampere hours) takes place when using a chromic salt, a considerably smaller (4/7 of the first) by the use of chromium hydroxide, and the smallest (3/7 of the first) by the presence of a quan- tity of base equivalent to the bichromate solution." In relation to the method of procedure, a diaphragm is not necessary. For larger experiments (40 amperes and over) a rectangular metal box was used, whose under part was roof- shaped and fastened with a flange to the upper part, which is lined inside with vulcanized rubber. The lower part serves as a cathode, opposite to which a roof-shaped anode of platinum netting hangs. Air is blown in to increase the cir- culation. For small experiments the following apparatus serves : two vertical platinum electrodes in a glass vessel. No other suitable anode material than platinum has been found ; PbO 2 has too low a conductivity, and carbon is too strongly attacked. The results of the experiments are given as fol- lows : (i). 400 ccm. of saturate. ri of chromilution -f- 100 grams of NaCl was diluted to 450 ccmy an elect-- -Jyi. of chrome alum solution containing 4.20 gra on of a c 2 O 3 iiic s slowly added. 12.84 ampere-hours applied. >n.lplr )mot* sulphce 3.55 volts at 49 to 66, and electrodes 15 U3; identic?-* With -.'rent density at the anode, which had 24 square lanc a> t^ eisen ^face, 0.178 OXYGEN COMPOUNDS. 107 ampere per square centimeter, on the cathode with 94 square centimeters surface, 0.0454 ampere. The current output of K 2 Cr 2 O 7 was 34.2 per cent. Almost all the Cr z O 3 was used up. 2. 200 cc. of a 25 per cent, solution of KC1 with a gradual addition of 62.6 cc. of neutral chrome alum solution, con- taining 1.9 grams of Cr 2 O 3 . 5.6 ampere-hours applied. A dark brown flocculent precipitate came down first, and all ex- cept a few flakes disappeared ; there was no Cr 2 O 3 present at the end. The current output of K 2 Cr 2 O ? corresponded to 36 per cent. 3. 25 per cent. KC1 solution gradual addition of chromic chloride solution (from 100 grams of chromic hydroxide = 50 grams Cr 2 O 3 dissolved in HC1 and neutralized as far as possible by repeated evaporations). 32.5 ampere-hours. 4 volts at 60. The current output of K 2 Cr 2 O 7 represented 35 per cent. ; nearly half had crystallized out. 4. 25 per cent, solution of KC1, to which was gradually added 50 grams of chromic hydroxide which had been pre- cipitated with 21 grams of caustic lime. 29.5 ampere-hours. Anode current density 0.754 ampere per square centimeter, 4.45 volts at 48. Current output of K 2 Cr a O 7 61.9 per cent. A part had crystallized out. 5. Saturated NaCl solution to which was added a solu- tion of 50 grams of chromium hydroxide and 12. 75 grams of NaOH. 36 ampere-hours. Anode current density 0.692 ampere per square centimeter, 5.2 volts at 78. 63 per cent, output. In the last two cases, the formation of a dark brown precipitate consisting of Cr 2 O 3 and CrO 3 was particularly noticeable. In reference to the practical application of the method, I will say that it is doubtful, unless the oxidation in an acid solution does not usually present a simpler method of proce- dure. However, it might be used in some special cases, for instance, if the chromium residues are available in the solid form or if for the purpose' of purifying, a precipitation of the chromium out of the solution as chromic oxide is to be 108 CHROMIUM AND ITS COMPOUNDS. recommended. The case may also be imagined where the oxidation of the alkaline electrolyte would follow advan- tageously the ignition of chromic oxide with lime as an auxiliary operation. On the whole it can be said that the outlook for the prac- tical application of the alkaline oxidation, even with cheap electrical energy is at present, very limited. For performing the oxidation in acid solution, Regelsberger first used only platinum anodes and even had little success with the use of a diaphragm. He first obtained satisfactory results in a sulphuric acid solution by the use of lead anodes. According to Elbs 1 a 70 per cent, current efficiency could be obtained by the use of freshly ignited platinum anodes and a low current density. The application of electrolysis, relative to economy of voltage, should occur in warm solutions ; also raising of the temperature seems to have a good influence on the current output itself. The alternation of the cathode and anode solutions as in the patent of Hochst Color Works (see page 104) is emphasized as practical. An addition of iron to the solution must be avoided, as it causes a loss of current, which at least would not be the case if the iron possessed an oxidizing effect in the application of the oxidized solution. The output with the application of a pure chrome alum solution with or without free sulphuric acid for the electro- lyte, went up to 92.5 per cent., 70 per cent, of the added amount being oxidized, and with 80 to 86 per cent, efficiency of oxidation. The chrome alum solution used carried about 30 grams of Cr 2 O 3 per liter. As soon as the larger part of the chromium was oxidized, there occurred a strong evolution of oxygen, this serving as a sign for the renewal of the solution or a more active circu- lation. With each electrolysis there formed a small amount of a yellowish white slime, composed of. lead sulphate and lead chromate, and a brownish coat on the anode, which scarcely 1 Zeitschr. f. Elektrochemie, 6, 388 (1900). OXYGEN COMPOUNDS. 1 09 influenced the electromotive force, since it was easily shaken off. With an anode current density of 0.065 to o.ojo ampere per square centimeter, a somewhat higher cathode density, and a total current of 42 amperes, three volts tension was used, tem- perature 75 ; to use this voltage at 30 temperature, not over 0.03 ampere per square centimeter could be exceeded. The apparatus consisted of a lead box. In the box is found the diaphragms, and the cathodes of any suitable conducting mate- rial. The anode may be either the lead box, or may consist of perforated lead plates surrounding the diaphragms. The power consumption may be reckoned as 8 to n kilo- watts for each 100 kilograms of Na 2 Cr 2 O y produced per 24 hours. To this must be added the cost of the simple appara- tus, the diaphragms being the chief consideration according to Regelsberger, but which is not the case at present, and finally the regulation of the apparatus, which may be assumed for the most part simple and automatic. "It is therefore very apparent that the cost of the regeneration of 100 kilo- grams of Na 2 Cr 2 CT should not be over $5.00. Of course cheap electrical energy is assumed. Regelsberger compares this cost of $5.00 with the commercial value of 100 kilograms of bichromate, worth at present $13.50, not counting in the last the required amount of sulphuric acid to start the oxidizing action which up to the present was always lost.'' The comparison would have been more exact if made with the former regeneration cost, which did not amount in toto to more than $5.00. Regelsberger closes with the words : " The process for the 'oxidation in acid solutions' will doubtless be of great importance, as soon as the question of diaphragms, which to-day is a serious problem in the electro- chemical industry, has been satisfactorily solved," etc. A little later there came into general notice, a patent 1 of 1 German Patent 109,012. January 13, 1897. Identical with the Austrian patent 5658, of v. Erggelet, dated April 19, 1901. 110 CHROMIUM AND ITS COMPOUNDS. Darmstadter's, which apparently solves the problem of the re- generation of chromic acid in a highly simple and elegant manner. Darmstadter puts the substance to be oxidized into a solution, for instance, of chromium sulphate in sulphuric acid, and electrolyzes. " In case there is sufficient circulation every trace of chromic acid formed must immediately be oxi- dized ; an accumulation of it is not possible, and therefore any too strong action impossible," which in the usual oxida- tion methods leads easily to a combustion of the substance being oxidized. In this way chinon can easily be prepared from aniline ; this latter compound is brought directly into the cold bath, consisting of chromium sulphate and sulphuric acid, and electrolyzed with continual stirring. The change takes place without noticeable loss. In an analogous manner acetic aldehyde may be prepared from ethyl alcohol. Also the process is said to be particularly advantageous for the oxidation of anthracene or naphthaline .; these materials are put without further precautions into the chromium solu- tion acidified with sulphuric acid, and electrolyzed with good circulation. " There results a good output of phthalic acid or anthrachinone." In the same manner the most varied other soluble or insolu- ble organic compounds may be oxidized. The high current output of this process is a further advan- tage. If chromic sulphate is oxidized in acid solutions with- out simultaneously introducing reducing substances, the cur- rent output decreases the more chromic acid there is formed. It is quite otherwise with the process in question ; an almost quan- titative current output is here obtained. " Since in this case every trace of chromic acid is changed instantly back into chromic sulphate there is always a large excess of the latter present whereby the electrolytically produced oxygen is very completely utilized." OXYGEN COMPOUNDS. Ill Diaphragms are according to the patent specification not necessary, because the chromic acid is utilized immedi- ately after its production for oxidation and no trace reaches the cathode. A. diaphragm is used only in such cases where the organic oxidation product is decomposed by the cathodic hydrogen. It is at times used when producing insoluble materials in order to hinder a deposition upon the electrodes, being placed in a particular part of the electrolytic bath which itself is separated from the electrode space by a filter cloth or the like. The substance to be oxidized can be put in a separate ves- sel, which must be in direct and continual communication with the electrolytic bath, between which as above a piece of filter cloth is interposed. In relation to the details of operation, is appended : " When the oxidation of the organic material and the regeneration of the chromium solution takes place in separate vessels, the solutions employed should be as concentrated in chromic oxide as possible, containing 10 to 12 per cent, or more thereof, so that it will be seldom necessary to transfer the heavy solution from the electrolytic bath to the oxidation ves- sel, and to perform other heavy operations as infrequently as possible. With the process in question this is for evident reasons not necessary, and it suffices completely if the solu- tion contains 4 to 5 per cent, of chromic oxide or less. The content of sulphuric acid is made high 20 to 25 per cent. in order to increase the conductivity of the electrolyte." u The current density may vary within wide limits for an equally good output is obtained with currents from 0.005 to 0.05 and even higher ampere per square centimeter." The electromotive force varies between 2.5 and 4 volts. Patent Claim. i. Process for the oxidation of organic substances by means of chromic acid, characterized by the reactions taking place 112 CHROMIUM AND ITS COMPOUNDS. in the electrolytic bath, in such manner that the reduction pioduct of the chromic acid is continually re-transformed into chromic acid by the action of the electric current. 2. A manner of carrying out the process of oxidation pro- tected by claim i, consisting in treating the organic substance to be oxidized, in a part of the cell separated from the elec- trode compartment by a sieve or filter cloth. 3. A manner of carrying out the oxidation process pro- tected in claims i and 2, consisting in bringing the substance to be oxidized in a separate vessel whose contents are in direct communication with the contents of the bath, instead of in a part of the bath partitioned off. I have already given a criticism of the preceding process in a lecture entitled : u The Electrolytic Regeneration of Chromic Acid and the Production of Acid Resisting Diaphragms." * I said therein: "He (Darmstadter) takes the point of view, that the chromic acid yields its oxygen up immediately to the substance to be oxidized, so that the reducing action of the cathode does not enter into the question, and the oxidized substance itself is little or not changed by the cathode. Aside from the latter assumption we must on the grounds of ex- perience also regard this process as very little hopeful, since the action of the oxidizing agent on organic substances is in general not instantaneous, and from previous investigations (see further on) on the great facility of reduction of the chromic acid. On reading the patent specification, it might be believed, that the problem is simply solved when he speaks of the obtaining of a practicable output of phtalic acid or anthrachinone, from the oxidation of respectively napthaline and anthracene. According to our investigations the process is not applicable for the direct production of just these materials; we have not once been able to prove the formation of phtalic acid qualitatively, with assurance. The following experiments had for their object the settling of the question as to whether it was promising to work with- 1 Zeitschr. f. Elektrochemie, 7, 290 (1900). OXYGEN COMPOUNDS. 1 13 out diaphragms. If the chromium sulphate solution acidified with sulphuric acid is electrolyzed without 'diaphragms, an oxidizing action of the current is noticed, Li out the output is very unsatisfactory. There was obtained in" an investigation with a sulphuric acid solution containing 100 grams of chromic oxide per liter and using 0.03 amjjere per square cen- timeter and lead electrodes, at ordinary temperatures, a cur- rent output of less than 10 per cent, for the solution contained at the end of the investigation about 6 grams of chromic acid per liter, and this large loss in spite of the small concen- tration of the chromic acid formed. If the current density at the cathode is raised, a larger output is obtained with a simultaneous increase of voltage ; but a practical application of the process is not to be thought of. Changing the mate- rial of the cathode had no beneficial effect. In reference to the process 1 6f FitzGerald (page 95) for the simultaneous separation of 2ihc and chromic acid, I should say: Theoretically such 'a 'combination of double working appears pretty, and it pefhaps 1 directs the attention to making such combinations in m&hy electrolytic processes; we have it also in the decomposition of the 1 alkaline chlorides, but here it lies in the nature of the operation. In practice such an artificial combination cannot often be arranged, since for practical use there is always the danger that in obtaining one material the other must always be' produced, even when its production is unprofitable. In such changeable combinations great precautions must be taken in their installation. The proposition of FitzGerald is naturally unusable in the electrolytic regeneration of chromium on a large scale ; since where would the large quantity of zinc sulphate come from? It is not a long step, however, to replace the zinc sulphate solution by sulphuric acid. If we 'do this we quickly observe an enrichment in sulphuric acid at the anode side 'and an im- poverishment at the cathode side. It would therefore 'be necessary in this process to precipitate the excess^of sulphuric acid out of the chromium solution- from time to time 'by lime 114 CHROMIUM AND ITS COMPOUNDS. and to replace by fresh concentrated acid the dilute sulphuric acid on the cathode side, which is contaminated by the chro- mium which has diffused over. I spoke further in the lecture quoted of the German patent 103,860 (see page 104) invented by me and belonging to the Hochst Color Works ; and will make some further remarks : The current output is principally dependent upon the per- centage of chromic oxide in solution and how far it is to be oxidized, as well as on the quantity of organic substances in the solution and how far it must be burned by the oxidizing action of the current, and finally on the properties of the electrodes and the temperature. In general 70 to 90 per cent, output can be counted on, on a large scale. The tension changes with the conductivity of the solution used, the cur- rent density, the distance of the electrodes apart, the temper- ature of the solution and the thickness of the diaphragms used. If one has, for instance, cheap electrical power no par- ticular value would attach to the least possible working ten- sion, but other considerations would be of first importance in obtaining the best commercial result. Usually the tension is held between 2.7 and 4 volts. Only lead anodes are used, which quickly, cover themselves with a layer of peroxide. According to Regelsberger (see page 106) the output is considerably larger with lead electrodes than with platinum, which is to be ascribed to the over- voltage, which must be used to evolve hydrogen on the per- oxidized plates and which thus allows of the conversion of trivalent chromium ions into the hexavalent, without free oxygen escaping in considerable amount. Another way is perhaps worth mentioning by which chro- mium solutions are oxidized, although it has not proven itself of practical value. Lead peroxide oxidizes chromic oxide to chromate, and a charged accumulator plate placed in chromic sulphate solution discharges itself spontaneously, oxidizing the solution. The discharged plate can now be put into storage-battery sulphuric acid, and charged, using a lead plate OXYGEN COMPOUNDS. 115 as cathode. By allowing the plate to alternately discharge in the chromic solution and be recharged in sulphuric acid, the putting in of a diaphragm is unnecessary, and the oxidation may be done with a low tension. Unfortunately the plate gives out quickly. Summing up, I remarked that a rational process for the elec- trolytic regeneration of chromic acid without a diaphragm did not exist. We are therefore restricted to them and will con- sider finally the attempts which led me to the production of a good and cheap diaphragm. Lastly concerning the cost, an average of 350 kilowatt-hours will produce 100 kilograms of CrO 3 in solution from chromic oxide, according to the data given. In conclusion I refer to the data of Regelsberger (page 109), in which it perhaps might be remarked that the hope is entertained that the partial ac- cumulation of a sinking fund would materially reduce the cost. A short time after this lecture, Darmstadter published a second process 1 for the regeneration of chromic acid. He mentions the process protected by the German Patent 103,860 (page 104), and believes that the difficulties there avoided by the changing of the anode and cathode solutions could be avoided by taking into consideration ordinary hydro- diffusion. The sulphuric acid formed in the cathode space by the elec- trolysis as well as that formed by the oxidation of the chromic sulphate tends to distribute itself equally in both compart- ments, and therefore to act against the accumulation of the acid in the cathode compartment. Likewise the chromic sulphate, which, for like reasons, accumulates in the cathode space and decreases in the anode space, tends by diffusion to become of equal concentration in both spaces. If now the current density is so regulated that the transference of the sulphuric acid as well as the chromic sulphate in the one direction is compensated by the transference in the other 1 German Patent 117,947, Nov. 3, 1899. 1 1 6 CHROMIUM AND ITS COMPOUNDS. direction, one has a simple means at hand for overcoming the aforesaid difficulties without it being necessary to frequently renew the solution in the cathode compartment. As the velocity of hydro-diffusion is dependent on the difference of concentration, and on the surface and penetra- bility of the diaphragm, so is (I remark further according to the specifications) the current density regulated to these conditions, that is it may be so much the greater, the greater the diaphragm surface and the greater the differences of con- centration of the sulphuric acid and chromic sulphate in the two compartments. The current strength corresponds to the changes in the latter during the electrolysis. If changes in concentration occur during the electrolysis, that do not injure anything, they must after ending be equalized. Under certain conditions it is practicable to so operate the process that one constituent maintains its concentration con- stant, while that of the other changes. If the highest allowable current densities are to be em- ployed then care must be taken to keep the difference in con- centration very large, that is the content of sulphuric acid in the anode compartment very high, the content of chromium sulphate as low as possible. To attain this with continuous driving, one must resort to the addition of fresh solution mixed with a large amount of solution which has been al- ready oxidized, so that the solution in the anode compartment is always to a great extent oxidized and contains a relatively large quantity of free sulphuric acid and a small amount of chromic sulphate. If the solution to be oxidized is passed through the anode compartments of a number of baths, using the apparatus in series, then it is recommended that a smaller current density calculated on the diaphragm surface should be used in the first than in the last. Patent Claims. i. Process for the oxidation of acid chromium salt solutions OXYGEN COMPOUNDS. 117 in electrolytic baths with diaphragms, characterized by the regulation of the current density, so that the content of free acid or chromium salt or both in the cathode space during the course of the electrolysis, is not appreciably altered. 2. A method of operating the process of claim i, wherein for the furtherance of the therein expressed purpose, the addi- tion of fresh and the abstraction of oxidized solution is so reg- ulated that in the anode compartment there is always found a solution which for the most part is oxidized, so that it con- tains proportionately a large quantity of free acid and a pro- portionately small quantity of unoxidized chromium salt. 3. A method of operating the process of claim i, whereby when passing the solution to be oxidized through a series of several baths, varying current densities calculated on the diaphragm surface are used in the single baths for the at- taining of the desired object. The carrying out of the general idea mentioned in the pre- ceding patent, to make the concentration changes through diffusion occasioned by the action of the current retrogressive, appears hopeless in practice because the velocity of diffusion is so small. Above all things it must be mentioned that the application of this idea in the above case, seems to be wholly out of place. For if there is a noticeable diffusion of sulphuric acid from the anode into the cathode compartment, and of chromic sulphate in the opposite direction chromic acid formed on the anode side will wander over to the cathode side where its concentration is almost zero, and so correspond- ingly diminish the output. If the chromic acid concentra- tion on the anode side is kept as high as possible right from the beginning, as is especially recommended, the loss will naturally be so much the greater, until it happens that the chromic acid formed at the anode becomes so much smaller in comparison with the quantity of current sent through, the more chromic acid is already present. On the whole the process must be characterized as impracticable. An English patent 1 of Schneider, of a later date concerns 1 English Patent 19,029, Oct. 24, 1900. Il8 CHROMIUM AND ITS COMPOUNDS. itself very closely with the contents just described of the Ger- man patent of Darmstadter. Schneider also uses the aid of diffusion in order to make the diffusion changes, which dur- ing the electrolysis of chromium solution take place on both sides of the diaphragm, again retrogressive, to which he adds at least that in general only small current densities may be employed. He also mentions that using a solution contain- ing 100 grams of chromic sulphate and 100 grams of sulphuric acid per liter, and a current density of 0.0025 to 0.0050 am- pere per square centimeter (according to the nature of the diaphragm) no change of concentration of sulphate or acid took place in the cathode space before or after the electrolysis ; also working with 0.03 ampere current density, it made itself noticeable and it could therefore, in this case if the vessel stood long enough without current, be made retrogressive. As a means of increasing this back diffusion it is recom- mended as in the German Patent 117,949, to use only such solutions as anode electrolytes which are already in the greater part oxidized, and according to that regulate the addition of new solution and the abstraction of old in the anode compart- ment, or better still to mix the anode and cathode solutions. The latter could be done in this manner, by allowing fresh solution to run into the electrolyzing vessel until the level of the fluid stood above the edge of the diaphragm, which is usually lower than the edge of the vessel, thus allowing un- hindered diffusion of both fluids, which finally is also facili- tated by stirring. The practical application was carried out in the following manner. By using a solution containing 100 grams of chromic sulphate and 50 grams of sulphuric acid per liter, and a current density of 0.02 to 0.03 ampere per square centi. meter, it is found that after about half of the chromic sulphate had been oxidized at the anode, the acid concentration at the anode had risen to 100 to 150 grams per liter, and a corre- sponding fall had taken place at the cathode. At this time enough fresh chromic solution was put into the vessel OXYGEN COMPOUNDS. 119 until the edge of the diaphragm was covered, when a mixture of the anode and cathode fluids will take place in several hours, or in a shorter time by the aid of stirring, and the dif- ference of concentration will be equalized, upon which the electrolysis may be continued after the removal of the upper layer of solution. If the solution contains 100, 150 or more grams of sulphuric acid per liter, then such a mixing will be required only at longer intervals, because the content of sulphuric acid in the cathode solution will not be reduced to such a small value in so short a time. The electrolysis of sulphate solutions is also recommended because of the possibility of using lead electrodes. The patent claim reads : 1. A method for the equalization of the concentration changes occurring in the electrolysis of chromium oxide solu- tions acidified by sulphuric acid, when a diaphragm is used, consisting in mixing the anode and cathode fluids, either by increasing the diffusion through the diaphragm in relation to the action of the electrolytic migration or by free diffusion or by the aid of artificial means. 2. A method of equalization (as above), consisting in mix- ing the solutions, by the aid of diffusion through the dia- phragms, and increasing this diffusion velocity in relation to the transfer velocity of the sulphuric acid by a regulation of the current density, essentially as described. 3. A method for the equalization (as above), consisting in mixing the solutions by the help of diffusion through the diaphragm, and admitting a continuous or intermittent stream of new solution in the anode compartment, drawing off a corresponding amount of old solution, essentially as described. 4. A method for the equalization (as above), consisting in mixing the solutions by the aid of diffusion through a dia- phragm, and from time to time interrupting the electric cur- rent. 120 CHROMIUM AND ITS COMPOUNDS. 5. A method for the equalization (as above), consisting in the direct mixture of the solutions of both compartments. 6. A method for the equalization (as above), consisting in directly mixing the solutions, by allowing fresh solution to flow in until the edge of the diaphragm is covered, allowing the whole to stand for some time, and then drawing off a sufficient amount of solution, essentially as described. 7. A method for the equalization (as above), consisting in a direct mixing of the solution, while stirring, essentially as described. In passing judgment upon this last process the following with reference to page 105 is added : If there occurs a mixing of the anode and cathode fluids in any way, either by free or limited diffusion, or diffusion facilitated by stirring, there re- sults under all conditions under the assumption that an in- fringement of patent 103,860 (page 104) is to be avoided a certain amount of the chromic acid formed at the anode going to the cathode, where according to the investigations men- tioned on page 101, it is again reduced. This process is less valuable than that of patent 103,860. APPENDIX. 1. The chromium carbide C 2 Cr , described on page 66, can also be made according to a contribution of Moissan 1 by heat- ing a mixture of equal parts of chromic oxide and calcium carbide (200 grams was taken) in an electric furnace for five minutes, using 900 amperes at 45 volts. Moissan found as the products of the reaction, a metallic regulus covered with crystalline needles, underneath a covering of melted calcium carbide, and by analysis corresponding to the above formula, and containing traces of iron and calcium. 2. On the preparation of acid lead chromate according to the process of Luckow (page 89), J. Rontschewsky remarks in a paper, 2 that the production of the lead chromate may be combined with the electrolytic obtaining of zinc in one process, if a diaphragm is employed. 1 Compt. rend., 125, 841 (1897). ' 2 Zeitschr. f. Elektrochemie, 7, 29 ( 1900). INDEX OF AUTHORS. Aschermann 42, 43, 47 Boehringer & Sons 74 Bonnet 3, TO, u, 13, 16, 18 Borchers 37, 54 Browne 86 Buff 81, 101 Bullier 64 Bunsen i, 2, 3, 23, 53, 76, 77 Chaplet 67 Cowper-Coles 19, 25 Darmstadter t no. 112, 115, 118 De Chalmot 70 Demmenie 39 Dercum 105 Deville 54 Dollner ' ;" '. 50 Dufau 82 Elbs i 108 Electro-Metallurgical Co 19, 23, 59 Erggelet, von 109 Fe*re*e 19, 25, 53, 76 FitzGerald 95, 103, 113 Fremy 54, 82 Geuther 101 Glaser, F 75 Glaser, G 27 Goldschmidt, H 30, 51, 63 Goldschmidt, Th 46, 47, 49, 6 1 Haussermann 98, 99, 101 Heibling 56, 102 Heroult . 62 Hittorf 29, 31, 39 Hochst Color Works 104, 108, 114 KoryschefF 39 Krupp 34 L,e Blanc 104, 112 Le Chatelier 82 Lorenz 102 I/uckow 87, 88, 120 Maronneau 72 122 INDEX. M oiler 19 Moissan 35, 36, 37, 38, 54, 55, 57, 64, 66, 68, 70, 81, 82, 85, 120 Merges lor Mourlot 72 Neumann ..'.... 27, 30 Ostwald . ' 31, 100 Parker 71 Placet 3, 10, u, 13, 16, 19, 25, 59, 63 Reese . . . 80, 101 Regelsberger . , . , . '. . '. 105, 106, 108, 109, 115 Reisenegger 104 Rontschewsky 120 Schneider 117 Smith '. 97 Societe" d'lectro-Chimie . '..'.' 51 Socie'te Ge'ne'rale de Aciers Fins 61 Socie'te Neo-Metallurgie Marbeau, Chaplet & Co . 57 Strameo 79 Street . . . . . . .... . ' ! ! 19, 78, 80 Vautin 44, 47, 49, 61 Verneuil 82 Vielhomme , 62 Villon 73 Williams 66 Wilson 32 Zettel 70 UNIVERSITY