LIBRARY 
 
 OF THE 
 
 UNIVERSITY OF CALIFORNIA. 
 
 OF" 
 
 ' 
 
ON TWO NEW 
 
 Electrochemical Processes 
 
 FOR THE 
 
 Extraction of Silver and Gold 
 from Their Ores 
 
 THESIS 
 
 Presented in Partial Fulfillment of the Requirements for the Degree ot 
 
 Doctor of Philosophy in the College of Chemistry 
 
 of the University of California 
 
 BY 
 
 MOOSHEGH VAYGOUNY 
 
 BERKELEY, 1905 
 
ON TWO NEW 
 
 Electrochemical Processes 
 
 FOR THE 
 
 Extraction of Silver and Gold 
 from Their Ores 
 
 THESIS 
 
 Presented in Partial Fulfillment of the Requirements for the Degree ol 
 
 Doctor of Philosophy in the College of Chemistry 
 
 of the University of California 
 
 BY 
 
 MOOSHEGH YAYGOUNY 
 
 fi 
 
 BERKELEY, 1905 
 
CONTENTS 
 
 Page 
 
 Introductory 5 
 
 Historical 6 
 
 Theoretical 1 1 
 
 I. Chloridation or Sulphatation 14 
 
 Experimental 21 
 
 Electrolysis 32 
 
 II. Persulphatation 36 
 
 Historical 36 
 
 Theoretical 36 
 
 Experimental 41 
 
 Electrolysis 55 
 
 Resume 59 
 
 Acknowledgements * 61 
 
On Two New Electrochemical Processes 
 
 for the Extraction of Silver and 
 
 Gold from Their Ores 
 
 INTRODUCTORY 
 
 Some two years ago, Mr. Gilbert Gurney, a practical mining 
 man, came to this University with some of the now well- 
 known Tonopah gold and silver ores, with a view to finding 
 out "what could be done with them in the line of a remunera- 
 tive treatment of the same." As Mr. Gurney, upon the writer 
 coming to know him, admitted frankly to have very limited 
 command of the knowledge of chemistry or metallurgy neces- 
 sary to c^pe with the problem in question, the latter, by com- 
 mon consent, soon devolved upon the present writer, while 
 Mr. Gurney himself kindly offered to purvey certain facilities 
 whereby the necessary investigations were to be carried out. 
 
 The following is an account of the results of the various 
 experiments and studies that were undertaken in 'this line 
 during the academic year '03- '04, and continued throughout 
 the year '04- '05. 
 
 It is necessary to state here that nearly all the assays made 
 in connection with the first part (Chloridation method) of this 
 study were made by Thomas Price & Sons, professional as- 
 sayers of San Francisco, through the courtesy of Mr. Gurney. 
 On the other hand, the writer assumes responsibility for the 
 remaining assays, all of which were made at the chemical 
 laboratories of this University. 
 
 t *? Q * > "i > 
 
HISTORICAL 
 
 It seems to be an admitted fact in mining circles that, in 
 spite of the goodly number of the various methods proposed, 
 some in journal and mostly in patent literature, in the last 
 half century or so, the desideratum of an economic, wet process 
 for the simultaneous extraction of the precious metals gold 
 and silver from their sulphide ores is still keenly felt. In- 
 deed, the methods in vogue, even at present, are still some 
 one form or another of smelting, amalgamation, Augustin, 
 or hypo-sulphite processes, all of which entail, at some point 
 or other in the course of their operations, certain costly treat- 
 ments, now in the form of a preliminary roasting, and now 
 in form of constant consumption or rejection of the reacting 
 chemicals, such as salt, copper sulphate, and mercury or its 
 salts, while all, except smelting, are none too efficient, even 
 irrespective of the cost of treatment. 
 
 Although at the time the writer began the studies which 
 form the subject of this paper, and for some time later, he was 
 quite unaware of the various attempts and suggestions made 
 in patent literature in the line of improved methods of hand- 
 ling such precious sulphide ores, and though such patent liter- 
 ature, at best, could constitute anything but reliable informa- 
 tion to guide one's steps by, it is none the less of interest to 
 first give here a cursory account of the most important meth- 
 ods having any direct bearing on those to be advanced in these 
 pages, as far as they have come under the writer's notice. 
 
 While the use of purely "wet" or chemical-reaction methods 
 had been introduced as early as the sixteenth century by the 
 adoption of the well-known Patio process where silver sul- 
 phides are decomposed by means of a solution of copper sul- 
 phate and salt, and amalgamated it seems, however, that 
 Becquerel* was the first to realize (in 1835) the importance 
 of the introduction of some neat and less empirical methods 
 
 *Beoquerel, Comptes Rcndus, vol. 38, p. 109; also, Elements 
 d'Electrochimie, chap. VI. 
 
in the process of silver and gold ore treatment. Becquerel's 
 idea was an entirely new departure in the metallurgical meth- 
 ods then known. He suggested to first oxidize the ores, either 
 by a roasting or by means of a solution of CuSO 4 in presence 
 of a concentrated solution of common salt, and then to recover 
 the metals thus dissolved by using voltaic couples of zinc and 
 iron, for instance. To do this, the less electro-positive metals 
 were dipped in the solution containing the metals to be recov- 
 ered, while the more electro-positive metals were immersed 
 in an ordinary solution of salt, separated from the first by 
 means of a porous membrane. 
 
 Although this process has long since been left out of serious 
 consideration, and for obvious reasons, still, if we remember 
 that not only was this method proposed in those early days 
 when dynamos were unknown things, but that Becquerel had 
 the patience to actually treat many tons of ore in this way, 
 we cannot but admire the insight and courage of that great 
 man. 
 
 The next mention of an improved method of treating sul- 
 phide ores is that of Body, who, in a United States patent 
 issued in 1886, proposed to use a solution of a mixture of 
 ferric salts and common salt as the agent to react upon the 
 sulphides in general, such as those of Ag, Cu, Pb, etc., and 
 the electric current from a dynamo machine as the agent to 
 recover the dissolved metals. For this purpose Body advo- 
 cated the use of a cement tank provided with two quadrangu- 
 lar compartments, one within the other, and separated from 
 each other by means of a porous diaphragm. Towards the 
 upper part of the cell these compartments communicated with 
 each other freely, as the diaphragms did not stand high enough 
 to separate them completely. The inner one of these two 
 compartments was provided with a carbon floor serving as 
 anode, while the cathodes stood upright in the outer com- 
 partment. The ore previously roasted or otherwise oxidized, 
 if necessary being put in the anode or inner compartment, 
 the solution was admitted from below and made to ascend 
 through it, while the entire mass was being kept constantly 
 stirred. During its stay in this compartment, the solution 
 was to dissolve the desired metals by acting upon their sul- 
 phides, thus being itself reduced to a solution of ferrous salts, 
 
but soon oxidized again to the ferric condition by virtue of 
 the chlorine here generated simultaneously with the deposi- 
 tion of the metals on the cathodes outside under the influence 
 of the electric current. To facilitate the deposition of the 
 metals, the solution was not allowed to remain long in this 
 same compartment, but was continuously forced upward and 
 made to overflow into the next or cathode department, and 
 there be directly electrolized and freed from its metals. This 
 done, the solution, which must not only have now been im- 
 poverished of its metals, but also have its iron salts largely 
 reduced to the ferrous condition, was next returned to the 
 anode compartment again, there to have first oxidized the 
 latter to the ferric condition and then to have resumed the 
 attack of the remaining sulphides. 
 
 Needless to say that, in spite of the fact that this process 
 seems to have been widely discussed in those days, it does 
 not seem to have as yet taken a practical form anywhere, 
 and for obvious reasons. Indeed, even if the chemical prin- 
 ciples involved left nothing to be desired, the idea of burying 
 an electrode under tons of non-conducting ore-masses could 
 certainly not stand the test of any practical operation. 
 
 In the same year that Body's patent was issued, Endlich 
 and Muhlenberg* secured a patent claiming the use of "chlor- 
 ine solutions in presence of other metallic chlorides, such as 
 those of iron and copper, with or without the addition of 
 common salt." These inventors, however, did not specify 
 by what means they intended to recover the metals brought 
 into solution; so that their suggestion does not even stand 
 the test of a fair criticism. 
 
 Soon after this, also in the same year, a patent was issued 
 to Cassell, who proposed to use a diaphragm cell, the ores 
 being kept in the anode compartment, somewhat on Body's 
 principle, but using merely a solution of common salt as the 
 lixiviating agent. It was claimed that during the passage 
 of an electric current the chlorine generated at the anode 
 and there held in the solution impregnating the ore, would 
 act upon the metallic compounds and bring about their attack 
 
 *United States Patent Office Gazette, October, 1886. 
 ^United States Patent Office Gazette, October, 1886. 
 
and dissolution. This done, the same electric current was 
 then merely to carry the desired metals over into the cathode 
 compartment, through the diaphragm, and there deposit them. 
 
 It will be seen that in this process the chief defects are, 
 unlike that of Body, in the chemical principles involved more 
 than in practical considerations, since, not to speak of other 
 difficulties, aquous chlorine, upon whose presence the oxida- 
 tion of the sulphides is supposed to depend, is neither an 
 active oxidant of such sulphides nor can it be got in any 
 practically useful concentrations, this gas being only slightly 
 soluble in concentrated brines. 
 
 Some two years later there appeared the patents of Dr. 
 Hoepfner,* whose process seems to have caused much more 
 excitement in certain mining circles than perhaps any of the 
 other similar electro-chemical processes advanced before or 
 since then. Dr. Hoepfner proposed to use solutions of 
 CuCl L , in conjunction with common salt or CaCL and their 
 subsequent electrolysis. Although his chief aim was to treat 
 ores of the sulphides of copper, his process was also expected 
 to extract any silver and nickel which may accompany the 
 copper in the ores. 
 
 In this process the ores were finely crushed and leached 
 with the above solution, in order to decompose the sulphides 
 present and take up their corresponding metals. The solu- 
 tion was then continuously passed through special diaphragm 
 cells, where the silver was either to be deposited on special 
 cathodes by means of a current, or it was simply to be pre- 
 cipitated by copper shot. This done, and the other foreign 
 metals being thrown out by chemical methods, such as by the 
 use of CuO, CaO, KOH, etc., the solution was then elec- 
 trolyzed with a proper voltage so as to recover the excess of 
 copper brought into solution during the extraction of the ore. 
 During this latter operation the remaining cuprous salts were 
 oxidized to cupric salts, and could therefore be used over again 
 to leach the ores with. 
 
 It has been pointed out in all the accounts on this process 
 which has been given a more thoroughgoing, practical test 
 than perhaps any of the others, though without success that 
 
 *Borchers, Electric Smelting and Refining. 
 
 9 
 
one of the principal drawbacks in the method lies in the 
 necessity of having to use diaphragms. Be the case as it 
 may, however, it only stands to reason that such a process 
 could not be used for exclusively silver ores for reasons eco- 
 nomical and practical, as will be indicated further on. 
 
 Besides the above-enumerated more prominent methods, the 
 writer has, through the kindness of Dr. W. J. Sharwood, 
 been able to come to know of a list of not less than fifteen to 
 twenty different patents, issued in this country or abroad, 
 all relating to the extraction of silver and gold by the use 
 of solutions of copper or iron salts, now in the form of sul- 
 phates, now as chlorides; sometimes in conjunction with the 
 use of a strong solution of common salt, sometimes without 
 it ; now by introducing some free acids, and now none, while 
 nearly all prescribe a preliminary roasting and some form or 
 other of a diaphragm cell whenever electrical methods are 
 explicitly stated to be employed in recovering the metals 
 brought into solution. 
 
 Such was found to be the status of the problem of the 
 chemical, wet methods of extracting the precious metals from 
 their ores when the writer first took cognizance of them. 
 It needs no special argument to show that, in face of such 
 accounts of "suggested," but "unhatched," "patent" methods of 
 various degrees of theoretical soundness, but of no positive 
 reliability, the problem could still be fairly considered one 
 awaiting a special study, and a solution, if possible. In- 
 deed, it is a recognized fact that not a few of the patents 
 granted every year in such chemical or metallurgical lines 
 represent mere hastily conceived ideas, unbacked by any sound 
 experimental data, and that the few methods which have found 
 their way into text-books can be called anything but of un- 
 questionable authenticity, they being more or less the results 
 of hearsay or gathered from patent accounts. 
 
 Owing to the above considerations, it will therefore not 
 be improper, in the following pages, to carry on the discus- 
 sion entirely without reference to these patent suggestions. 
 This will be the more logical as the first and essential steps 
 in these studies had been taken, as was mentioned above, 
 long before the writer was aware of the existence of the 
 methods sketched or alluded to. 
 
 10 
 
THEORETICAL 
 
 On first venturing to develop any chemico-metallurgical 
 method of treating ores of the type here under consideration, 
 the first points to be considered are, naturally, relative to the 
 requirements to be fulfilled by the process to be worked out. 
 In the case on hand, what are these requirements? They 
 are obviously that the method should be economical ; it should 
 be a lixiviation process pure and simple, as far as possible; 
 and, if the method is at all to be of wide applicability, it 
 should be capable of extracting both the metals in question- 
 silver and gold as they occur so frequently together. Fur- 
 ther, it must do away with any preliminary roasting, which 
 is more or less troublesome and expensive. On the other 
 hand, the solution used must neither be costly nor have any 
 properties deleterious to health ; the metals dissolved thereby 
 must admit of being readily recovered ; and, finally, the solu- 
 tion should lend itself to being regenerated by some appro- 
 priate and ready means, so as to avoid the necessity of its 
 being constantly renewed ; in other words, the method should 
 be "cyclic." 
 
 Now, given the fact that in the entire field of chemistry 
 there is no solution of a known compound that will dissolve 
 the sulphides in question, as such, sufficiently and economically, 
 without the necessary oxidation of the latter, it becomes 
 forced upon the investigator in view of the above condition- 
 ing factor that all roasting operations should be avoided, as 
 far as possible that he choose only such solutions as have 
 naturally oxidizing properties. But all oxidation reactions 
 involve concomitantly a reduction on part of the oxidizing 
 agents which may be employed; hence, if these agents are to 
 be used cyclically, they must necessarily admit of being read- 
 ily oxidized back into their original conditions. Further, 
 since economic operation requires that there should be no 
 decomposition products arising, not only during the attack 
 of the sulphides, but also during such regenerating operations, 
 the compound or compounds to be selected as the solvent agent 
 desired must be fairly stable and be very "labile," or capable 
 of taking up first a higher, then a lower stage of oxidation. 
 Then, too, since the best and neatest method of recovering 
 
 ii 
 
the metals dissolved would be, beyond doubt, one depending 
 upon the use of the electric current, it would be highly desir- 
 able that the compounds selected should be capable of being 
 oxidized by means of this same current, best concurrently 
 with the electro-deposition of the metals. 
 
 If now, with these data on hand, we inquire as to which 
 compound or compounds will meet the requirements imposed, 
 we will notice that there are only two groups of salts that 
 will answer for the purpose at issue, and which are at the 
 same time sufficiently cheap and available. We have: 
 
 I. THK SULPHATES AND THE CHLORIDES 
 
 of those common elements, with varying valencies or degrees 
 of oxidation, as are also sufficiently soluble in water, namely, 
 those of iron, copper, and manganese. 
 
 II. THE SO-CALLED PERSULPHATES 
 
 A comparative view of the desirabilities of the salts of iron, 
 copper, and manganese will, however, show that not all these 
 three classes of bodies are equally desirable. For while a 
 moderately acid solution of the above iron salts may well be 
 electrolyzed without any of the metal separating at the elec- 
 trodes, those of copper and manganese, on the other hand, 
 cannot be so treated with advantage. For, with solutions 
 of copper salts, not only would we have economic difficulties 
 to meet with in the line of the large amounts of copper 
 which would have to be locked up in the stock solutions to 
 be used, but, what is more, such solutions could only be freed 
 from undesirable base metals, such as Sb, As, etc., which 
 would accumulate in them and foul them in time, with both 
 complicated and costly chemical methods of precipitation of 
 these impurities. This would be the more unavoidable, as 
 it would certainly be difficult to avail one's self of the neater 
 methods of simple electrolysis with higher voltages to remove 
 these base metals, lest, in doing so, copper should also be 
 deposited on the cathodes, and thereby cause much trouble. 
 Again, with the salts of manganese, the oxide MnO 2 is likely 
 to appear at one or both electrodes during the electrolytic 
 winning of the metals, and thus give rise to many unnecessary 
 difficulties. 
 
 12 
 
It thus becomes clear that, of the salts of the above three 
 commoner elements, those of iron, when used in conjunction 
 with some free acids, are much to be preferred, since, besides 
 presenting none of the above difficulties, they are also the 
 cheapest and most readily got. 
 
 Leaving the study of the persulphates for the second part 
 of this paper, we may then first consider here the application 
 of the iron salts to the extraction of the precious metals. 
 
 13 
 
I. CHLORIDATION OR SULPHA- 
 TATION 
 
 If we attempt to study, in a preliminary way, the relative 
 adaptations of the chloride and the sulphate of iron by carry- 
 ing out some qualitative experiments with artificially prepared 
 silver sulphide, and with two separate solutions, one made 
 up of about i to 2% FeCl 3 and HC1, the other consisting of 
 about the same amount of Fe 2 (SO 4 ) 3 and H 2 SO 4 , we ob- 
 serve that, so far as their efficiencies in oxidizing the silver 
 sulphide is concerned, whether in the cold or in the warm, 
 there is no appreciable difference to be noticed in favor of 
 one or the other of these solutions. 
 
 Now, seeing that it is a matter of no small economic im- 
 portance whether one deals, in practice, with sulphates or 
 with chlorides, it might seem here, a priori, that a sulphuric 
 acid solution of ferric sulphate should be preferable to one of 
 HC1 and FeCl 3 . This would be especially so, inasmuch as, 
 in practice, such a solution would be expected to form the 
 more soluble Ag 2 SO 4 , instead of the insoluble AgCl, on re- 
 acting with the sulphides of this metal, and that, therefore, 
 it would lend itself better to the subsequent leaching out of 
 the values from the ore masses. 
 
 Such a choice, however, would be objectionable for two 
 main reasons. In the first place, Fe 2 (SO 4 ) 3 having no 
 appreciable solvent action on free gold, its exclusive use would 
 not be in conformity with the desideratum set previously, 
 namely, the simultaneous extraction, as far as possible, of 
 both the silver and the gold contained in ores. In the second 
 place, the use of such simple sulphate solutions would be 
 amiss, even with ores that contain no appreciable amounts 
 of gold but are more exclusively silver ores, because, there 
 being no natural ore, nor a method of ore treatment possible 
 that could avoid sufficiently the introduction of some chlorine 
 compounds during such treatments, there would always be 
 a liability for the conversion into AgCl of much of the dis- 
 
 14 
 
solved Ag 2 SO 4 formed under the oxidizing influence of the 
 ferric sulphate upon the silver sulphides. In fact, such a 
 change would be undesirable not merely owing to the im- 
 practicability thus arising of the removal of the silver values 
 by a simple leaching, but also because the formation of the 
 insoluble AgCl would tend to hinder the complete oxidation 
 of the silver sulphides, owing to a protective coating which 
 this compound would form and envelop the particles of the 
 sulphides unless it be dissolved away and removed as fast as 
 it is formed. 
 
 If, therefore, ferric sulphate is at all to be chosen as the 
 oxidizing agent desired, it is essential to use concurrently 
 with it, on the one hand, some simple, appropriate solvent 
 of AgCl, and, on the other, one of free gold. 
 
 Unfortunately, of all the simpler solvents of AgCl or of 
 gold, there seems to be none known capable of co-existing 
 in solution with ferric salts, let alone the consideration of their 
 cheapness. Thus, cyanides could not be used because of the 
 liability of their giving rise to insoluble ferro or ferri cyanides, 
 thus resulting in loss and in general nuisances. On the other 
 hand, thiosulphates and bisulphites could not be used owing 
 to their being decomposed in presence of ferric salts ; nor, for 
 obvious reasons, could NH 4 OH be used for the purpose under 
 consideration. 
 
 There is, however, but one possible solution to this problem, 
 and that lies in the use of high concentrations of some suitable, 
 soluble chloride, such as common salt, in conjunction with 
 these sulphates. Such chloride solutions, indeed, are especially 
 appropriate, as by their use not only is it possible to bring 
 about the solution of gold readily under the influence of the 
 free chlorine which can be generated and made to saturate 
 the main solution simultaneously with the electro-deposition 
 of the dissolved metals, but particularly because they also 
 afford a ready ^means for the dissolution and removal of the 
 silver values concurrently with the gold. It seems, indeed, 
 a peculiarity of silver chloride that it will go into solution 
 quite appreciably in presence of other chlorides, especially 
 those of the alkali and alkaline-earth metals, and more par- 
 ticularly in free HC1. No doubt the secret and the sole func- 
 tion of the use of such large amounts of common salt in con- 
 
junction with CuSO 4 in the old Patio amalgamation process 
 lies in just this effect of the salt being capable of holding 
 in solution the silver chloride which is formed under the 
 oxidizing effect of the CuSO 4 on the sulphides of this metal, 
 so that the mercury may with readiness first reduce this latter 
 from its solution to the metallic state and then amalgamate 
 with it. 
 
 But now, if such high concentrations of chlorides are to be 
 used, as they needs must, with a solution of ferric sulphate 
 and sulphuric acid, it at once becomes evident that there re- 
 mains no serious advantage in favor of such a solution over 
 one of ferric chloride and either HC1 or H 2 SO 4 . Indeed, 
 so far as economic differences in the use of one or the other of 
 these compounds are concerned, none can exist worth being 
 called serious, inasmuch as, once the necessary supply of 
 either salt is secured and the stock solution made up, the 
 process being cyclic, the concentration of the iron in solution 
 will not diminish sufficiently to necessitate any continued sup- 
 ply of such salts, if, indeed, it be ever necessary to buy these, 
 even in long-continued practice, rather than depend upon the 
 iron contents of the ores themselves to furnish what little 
 amount of iron salt be requisite at any time. 
 
 On the other hand, though there is certainly an apparently 
 decided difference in the cost of the two acids HC1 and H 2 SO 4 , 
 still, it being true that to whatever extent SO 4 radicals are 
 substituted by Cl radicals to that extent will the solvent power 
 of the solution on AgCl be enhanced a fact which alone is 
 of nature to counterbalance what slight difference there may 
 be in original outlays in securing these acids even the use 
 of one or the other of these compounds becomes more or less 
 a matter of choice in a cyclic process like that in question. 
 Indeed, should even a given ore be too alkaline to stand the 
 cost of the HC1 necessary to first neutralize it, it may be treated 
 with H 2 SO 4 in a preliminary way ; or one may even use HC1 
 itself in such extreme cases, allowing this acid to percolate 
 through the mass, and subsequently treating the collected 
 solutions now containing all the calcium as CaCL with 
 H 2 SO 4 , thus rejecting this metal as CaSO 4 by a simple filtra- 
 tion and at the same time regenerating the HC1 which had 
 combined with the calcium. 
 
 16 
 
On the whole, therefore, it seems quite immaterial, for 
 the purpose on hand, whether one uses a solution made up 
 of Fe 2 (SO 4 ) :! and H 2 SO 4 in presence of a cheap chloride, 
 such as NaCl, or whether one chooses a solution of FeCl 3 
 and HC1 or H 2 SO 4 , with the same cheap salt, or of CaCL, 
 to furnish the necessary concentration of chlorine radicals. 
 As for the concentration of the reagent solution in these 
 various compounds, one need not diverge materially from the 
 approximate composition given on a previous page, though 
 the concentration of the common salt (NaCl) or of CaCl 2 
 may best be made as high as 20%. 
 
 But it may be objected here that such strong solutions as 
 the above may not be the best that could be wished from the 
 standpoint of facilities in commercial handling. However, 
 considering the fact that these solutions can be cheaply got, 
 readily recovered, and regenerated indefinitely, coupled with 
 the fact that they possess high electric conductivities, and 
 therefore require less expenditure of power for the electro- 
 lytic work, than were they more dilute, these facts are certainly 
 of a nature to more than counterbalance what little disad- 
 vantage this high concentration ma-y offer. 
 
 Reactions Involved. Before giving quantitative data to 
 show to what extent the use of ferric salts, under proper con- 
 ditions, can be regarded satisfactory for the extraction of the 
 precious metals from their ores, it is of interest to consider 
 here first the mechanisms of the reactions involved in their use. 
 
 In carrying out qualitative experiments with the above solu- 
 tions under the proper conditions, namely, in presence of 
 much NaCl, for instance, we observe, especially if the solutions 
 be warmed, that the Ag 2 S is rapidly decomposed. At the 
 same time silver goes into solution and much of the ferric 
 iron present is reduced to the ferrous condition, while a gray- 
 ish scum of sulphur appears floating on the surface of the 
 solution. The reaction taking place seems therefore to be, 
 with the sulphate of iron : 
 
 Ag|hFe 2 (S0 4 ), = Ag a S0 4 +2Fe.S0 4 +Sj( 
 Ag 2 S0 4 +2NaCl == 2AgCl+Na 2 SO 4 
 
 and with the chloride of iron: 
 
 Ag 2 S+2FeCl 3 +NaCl = 2AgCl-f 2 FeCl 2 +S-fNaCl 
 
 17 
 
Now while this seems to be the simplest way for represent- 
 ing the facts observed and it is indeed the one commonly 
 accepted still it is a question whether the reaction involved 
 is as direct as that shown above. For the same facts may 
 be represented equally as well by the following equations, 
 considering exclusively chloride solutions alone for clearness' 
 sake: 
 
 Ag 2 S+2HCl = 2AgCl+H 2 S 
 and then : n 
 
 H 2 S+2FeCl 3 = 2HCl^FeCl,+ S 
 
 The mere fact that, ordinarily, acids alone have no appre- 
 ciable effect upon the sulphide of silver is certainly no cri- 
 terion against such a view. For if this preceding reaction 
 be regarded as a reversible equilibrium reaction as all re- 
 actions may be so looked upon only with its equilibrium point 
 towards the extreme left, namely, toward the formation of 
 Ag 2 S, it becomes clear that if we only remove the disturbing 
 reaction product, H 2 S, by some appropriate method, the point 
 of equilibrium could be readily shifted to the right, so that 
 AgCl could form freely. The function of the ferric salt 
 would therefore be, in this light, only a secondary one, namely, 
 one of removing this disturbing reaction product, H 2 S, as 
 fast as it is formed under the direct influence of the acids. 
 That this view is more than a mere supposition is evidenced 
 by the fact that when a mixture of Ag 2 S and HC1, for in- 
 stance, is boiled, even in absence of any oxidizing agents, 
 there is a slight decomposition of the sulphide taking place, 
 especially if the acid used be concentrated. It being permissi- 
 ble to regard the effect of boiling as merely facilitating the 
 expulsion of H 2 S, this phenomenon may well be taken as an 
 indirect evidence in favor of the above view. The very fact, 
 indeed, that, even in commercial practice, Ag 2 S is decomposed 
 by concentrated H 2 SO 4 in the absence of oxidants may here 
 be adduced as even a more tangible evidence. 
 
 Again, the fact that a solution of a ferric salt, free from 
 any surplus foreign acids, decomposes Ag 2 S cannot be ad- 
 duced as necessarily favoring the existence of a direct action 
 between these two compounds, inasmuch as all ferric salts are 
 hydrolyzed freely, and are therefore apt to play the role both 
 of acids and of oxidizing agents. 
 
 18 
 
In view of these considerations, it seems that the reactions 
 whereby the oxidation of Ag.,8 takes place may best be ex- 
 plained in the light of the second view, unless, indeed, it be 
 admitted that both the direct and the indirect reactions take 
 place concurrently. 
 
 As regards the mechanism whereby the AgCl goes into so- 
 lution in presence of the foreign chlorides, there seems to be 
 conclusive evidence that this takes place by virtue of a ten- 
 dency the former compound has to give rise, with these 
 chlorides, to double salts analogous to double iodides and 
 double cyanides. Such a combination of AgCl and NaCl is 
 readily obtained in crystalline form whenever the conditions 
 are such that NaCl will begin to separate, such as by concen- 
 trating a solution containing these two compounds, by evapo- 
 ration, or by adding concentrated acids to it, especially HCl r 
 In either case, the first crop of crystals that separate will 
 contain practically all the silver chloride. These crystals are 
 visibly cubic and look exactly like so many crystals of pure 
 NaCl, and, strange enough, they do not even seem to be 
 affected by light. They are, however, readily decomposed 
 into flocculent AgCl and NaCl when thrown into pure water. 
 
 Unfortunately it is very difficult to form a definite idea as 
 to the exact composition of the double salt represented in these 
 crystals, as they do not seem to be of any fixed composition, 
 but rather individuals produced by the simultaneous and iso- 
 morphous crystallization of both a double salt and of inde- 
 pendent molecules of NaCl. Although treatises on chemistry* 
 do not seem to make specific mention of a double salt of silver 
 and sodium chlorides, still, as they allude in a general way 
 that AgCl can form double salts of the composition AgClMCl, 
 it seems that the compound here under consideration may best 
 be represented as 
 
 AgCl.NaCl+.rNaCl 
 
 If this be correct, and it seems to be strengthened by the 
 fact that several attempts to obtain this compound in a form 
 sufficiently free from extra NaCl for analysis have resulted 
 in failure, then the salt here under consideration would rep- 
 
 *Watt's Dictionary of Chemistry (Muir and Morley) ; Laden- 
 burg's Handivoerterbuch der Chcmie. 
 
 19 
 
resent a definite case of crystallization where double salt for- 
 mation and isomorphism may coexist. 
 
 There remains now to consider the mechanism of the oxi- 
 dation of gold. 
 
 It must be admitted here that, in spite of the assertions of 
 certain authorities,* no appreciable solution of gold can take 
 place in solutions of ferric iron, whether in form of chloride 
 or sulphate, except in presence of some foreign oxidizing 
 agent, such as chlorine, nitrates, and even ordinary atmos- 
 pheric oxygen. This is certainly not surprising, seeing that 
 such an attack of gold by a ferric salt would give rise con- 
 comitantly to so much of a ferrous salt, and that this latter 
 would naturally tend to react backwards and cause the re- 
 precipitation of gold by virtue of the well-known effects of 
 ferrous salts on solutions of gold. 
 
 It is evident, therefore, that the essential condition for the 
 successful attack of free gold, along with the oxidation of the 
 sulphides of silver and other metals, lies in the removal of 
 the ferrous salts formed during such reactions, as represented 
 on a previous page. 
 
 Such an oxidation of ferrous salts can of course be brought 
 about best by means of the chlorine which will tend to be 
 liberated at the anodes during the electro-deposition of metals, 
 and with which the main solution may be made to be saturated 
 in actual practice. The reactions here involved would then be: 
 
 Fed., + Cl = FeCl 3 
 3 FeS0 4 + Cl, = Fe 2 (S0 4 ) 8 + Fed, 
 
 in both of which cases the ferric salts regenerated can then 
 react upon gold to the extent that there is free chlorine left on 
 hand in the solution, and on the sulphides to the extent that 
 there is any iron left in the ferric condition. 
 
 Now a critical consideration of the above reactions involved 
 in this alternate oxidation and reduction of the iron salts will 
 reveal the fact that while in purely chloride solutions the 
 iron will always remain combined with chlorine that is, it 
 will be present as chloride with solutions, on the other hand, 
 
 *Comey's Dictionary. 
 
 IMcIlhiney, Am. Jour, of Science, '96, vol. ^"152, p. 293. 
 
 20 
 
where the iron may have been originally introduced as a 
 sulphate there will be a continuous tendency for them to 
 have this sulphate of iron converted into the chloride form. 
 It is, indeed, clear that every time Fe 2 (SO 4 ) 8 is reduced to 
 FeSO 4 , the subsequent oxidation of this latter by means of 
 chlorine will convert part of the iron present into FeCl :? , as 
 is shown in the second reaction above ; so that when this 
 process is repeated many times there will be more and more 
 sulphate of iron disappearing, while chloride of iron will take 
 its place. 
 
 Owing to this consideration, it is important to note here 
 that the method of ore treatment here discussed should be 
 properly regarded as a "chloridation" method, at least for 
 clearness' sake, even though the iron originally introduced 
 may have been in the sulphate form, or the acid in the 
 stock solution may be H 2 SO 4 , and not HC1. 
 
 We may now pass to a quantitative study of the extent of 
 suitability of such chloride solutions of iron salts to the treat- 
 ment of some actual ores and the conditions of operation. 
 
 EXPERIMENTAL 
 
 EXPERIMENTS WITH ORE NO. I 
 
 In the following experiments the ore used was obtained 
 from Tonopah, Nevada. Ore No. I was a grayish quartzose 
 ore, containing some galena, pyrite, calcite, silver sulphides, 
 and gold, with a gangue consisting mainly of quartz. It also 
 contained a notable proportion of metallic iron, introduced 
 into the ore owing to improper grinding. 
 
 Experiment I. In 'this experiment the ore was ground to 
 80 mesh to start with, owing to its reputation of being very 
 refractory, and, instead of treating it with a stock solution 
 already made up, it was thought best to utilize the metallic 
 iron in the ore by dissolving it in an acid. As it was of 
 importance to first study the effect of sulphate solutions, the 
 acid chosen here was H 2 SO 4 . When the action had ceased, 
 an excess of this acid and NaCl was added, so that the result- 
 ing solution contained about .5% metallic iron, 2% H 2 SO 4 , 
 and 20% NaCl, with a total volume of I liter for an ore mass 
 
 21 
 
of I kilo. The mixture was put into a large flask, chlorine 
 was introduced into it so as to oxidize the FeSO 4 formed 
 under the influence of the H 2 SO 4 upon the metallic iron in 
 the ore, and a mixture of ferric sulphate and chloride was 
 thus obtained in the ratio of 5 \2. 
 
 In order to facilitate the reactions and shorten the time of 
 treatment, heat was then applied to the mass, which was kept 
 nearly boiling for three hours, care being taken that there 
 was always an abundance of chlorine on hand in the solution 
 throughout this period. At the end of this time the mass 
 was filtered and well washed with a strong solution of com- 
 mon salt until a few drops of fresh teachings showed no 
 turbidity on being diluted with water, which meant, of course, 
 that all the silver chloride held in the solution wetting the 
 ore mass had' been washed out. (It is hardly necessary to 
 say here that it was essential to use a concentrated solution of 
 salt for this operation, because, had ordinary water or even 
 a very dilute solution of salt been used, the liquors' impreg- 
 nating the ore mass would be diluted thereby, and thus much 
 silver chloride might be precipitated and left back in the ore.) 
 
 When this was done, the mass was further washed with 
 pure water to remove the excess of salt solution wetting it, 
 and lastly it was dried and sent to professional assayers as 
 tailings. 
 
 The following are the results of the assays of the sample 
 before and after treatment : 
 
 RESULTS OF ASSAY, EXPERIMENT I. ORE NO. I 
 
 Au. Ag. 
 
 Sample Oz. per ton. Oz. per ton. 
 
 Original .8 84.7 
 
 Tailings .35 . 4.8 
 
 This corresponds to about : 56% extraction of gold 
 and 944% extraction of silver. 
 
 It will be seen from these data that there was much room 
 for improvement as regards the extraction of the gold values. 
 
 Having, however, obtained similar results in a second test, 
 under similar conditions, it was then thought interesting to 
 study the effect of the finer crushing of the ore upon the 
 extraction of the values. 
 
 22 
 
Experiment II. Another lot of the same ore was therefore 
 crushed to 120 mesh and assayed for the second time. The 
 treatment carried out in this experiment was identical to the 
 above in details as to the composition of solution, length of 
 time of boiling or treatment, and, of course, as to care in the 
 method of washing the tailings. Only it was carried out on 
 a lot of 1/2 pound, with the corresponding amount of solution, 
 instead of I kilo, of ore, as was done in the first experiment. 
 
 RESULTS OF ASSAY. EXPERIMENT II. ORE NO. I 
 
 Aii. Ag. 
 
 Sample Oz. per ton. Oz. per ton. 
 
 Original 1.03 86.2 
 
 Tailings .125 i.i 
 
 This corresponds to about: 88% extraction of gold 
 and 98.7% extraction of silver. 
 
 It will now be noticed, on comparing the results of these 
 two experiments, that a good deal of the gold values of this 
 ore are very closely locked up in the ore gangue, thus necessi- 
 tating very fine crushing, and that, so far as the silver values 
 are concerned, So-mesh crushing is practically as good as 
 I2o-mesh. 
 
 Experiment III. Since in the above experiments the solu- 
 tion had to be kept boiling during the treatment, and since 
 the keeping warm of a mass of ore is quite an item in large- 
 scale practice, the necessity of some experiments to study the 
 effect of treatment in the cold became now quite evident. 
 The following test was therefore carried out to elucidate this 
 point. 
 
 The ore treated was y% pound of the I2o-mesh sample. The 
 details of treatment were identical to those in the above tests, 
 except that they were carried out in the cold, and the time of 
 treatment was prolonged to two days, with occasional shaking. 
 
 RESULTS OF ASSAY. EXPERIMENT III. ORE NO. I 
 
 Au. Ag. 
 
 Sample Oz. per ton. Oz. per ton. 
 
 Original . 1.03 86.2 
 
 Tailings .575 18.7 
 
 This corresponds to: 44% extraction of the gold 
 and 79% extraction of the silver. 
 
 23 
 
These data clearly showed the necessity of further prolonga- 
 tion of time of cold treatment, and consequently the following 
 test was then made : 
 
 Experiment IV. The conditions in this experiment, too, 
 were identical to those of the preceding experiment, except 
 that hydrochloric acid, instead of sulphuric, was used to dis- 
 solve the free iron in the ore mass, whereby all the iron was 
 put in the chloride form on chlorinating it, instead of part of 
 it staying in the sulphate condition. The time of treatment 
 was extended to three days instead of two, as in the preceding 
 test. 
 
 RESULTS OF ASSAY. EXPERIMENT iy. ORE NO. I 
 
 Au. Ag. 
 
 Sample Oz. per ton. Oz. per ton. 
 
 Original 1.03 86.2 
 
 Tailings .325 6.6 
 
 This corresponds to: 69% extraction of gold 
 and 92.4% extraction of silver. 
 
 From these results it can be seen that a three days' treatment 
 of the ore with a cold solution will extract highly satisfactory 
 percentages of silver values. 
 
 It might be argued here that the favorable difference ob- 
 tained by this last test might have been due not to the pro- 
 longation of the time of treatment, but rather to the fact that 
 in using hydrochloric acid instead of sulphuric, more ferric 
 chloride, and therefore more chlorides, were introduced into the 
 solution than in the first case. 
 
 Experiment V . In order to determine to what extent such 
 may have been the case, a special experiment was carried out 
 subsequently under the same conditions as with the preceding 
 test, only substituting H 2 SO 4 for HC1. The following were 
 the results obtained: 
 
 EXPERIMENT v. 
 
 Au. Ag. 
 
 Sample Oz. per ton. Oz. per ton. 
 
 Original 1.03 86.2 
 
 Tailings .39 5.6 
 
 Corresponding to about: 93-5% silver extraction 
 
 62.1% gold extraction. 
 
 24 
 
It becomes evident from these results that the divergence 
 between the results of the two preceding tests must in reality 
 have been chiefly due to the difference in the length of time 
 of treatments rather than to the influence of the increase in 
 the percentage of iron present as chloride in the testing 
 solution. 
 
 In order to further test this method of extraction of precious 
 metals, another ore was now taken up and experimented with. 
 
 EXPERIMENTS WITH ORE NO. 2 
 
 This ore, also obtained from Tonopah, was a reddish, oxi- 
 dized ore, containing much ferric oxide, some silver sulphides, 
 and gold, but very little galena or lime. It was crushed to 
 So-mesh before the following tests were made with it. 
 
 Experiment VI. This experiment was carried out under 
 conditions similar to those of Experiment i, only 500 grams 
 of ore being taken instead of I kilo. 
 
 Although in this ore there was practically no metallic iron 
 to depend upon to furnish the ferric salt required for the 
 attack of the sulphides, there was, however, a sufficiency of 
 Fe 2 O 3 , occurring as such in the ore, to serve for that purpose 
 by going into solution when the mixture of H 2 SO 4 and NaCl 
 was added. The iron thus brought into solution amounted to 
 about .4% by weight of the solution. 
 
 RESULTS OF EXPERIMENT VI 
 
 Au. Ag. 
 
 Sample Oz. per ton. % Oz. per ton. 
 
 Original .55 49.45 
 
 Tailings (hot treatment) .18 .5.5 
 
 These correspond to an extraction of about : 89% silver 
 
 and 67.3% gold. 
 
 Experiment VII. In order to determine to what extent 
 these results could be affected by adopting the conditions of 
 Experiment IV, namely, by a cold, three-day treatment in 
 presence of chlorides alone, a second test was made under 
 the latter conditions, with the following results : . 
 
 25 
 
EXPERIMENT VII 
 
 Au. Ag. 
 
 Sample Oz. per ton. Oz. per ton. 
 
 Original .55 4945 
 
 Tailings (cold treatment) .075 7.3 
 
 Corresponding to an extraction of: 85.7% silver 
 
 86.4% gold. 
 
 Although no further experiments were carried out with this 
 ore to study the effect of a finer crushing than 8o-mesh, or 
 of a more prolonged treatment, it is safe to say, however, that 
 had these measures been resorted to, the above results 
 would probably have been much improved upon, especially 
 in view of the fact that it was noticed, after the above tests 
 were made, that the ore experimented with was not quite of 
 8o-mesh crushing, but that fully 15% would not pass through 
 such a sieve. 
 
 It may have been noticed, from the descriptions of the above 
 ores, that they were both remarkably free from any large 
 amounts of the sulphides of "base" metals, such as antimony, 
 lead, etc. The question may therefore arise here as to whether 
 the method of ore treatment under consideration would at 
 all successfully apply to ores containing much base sulphides, 
 and, if so, to what extent and under what conditions. 
 
 To answer this question, the following experiments, under- 
 taken with the above point in view, may be cited. 
 
 EXPERIMENTS WITH ORE NO. 3 
 
 This ore was obtained from Idaho. It was a dark gray, 
 almost black, mass, chiefly made up of sulphides of antimony 
 and lead, with much silver, but very little gold. The gangue 
 was silicious, with practically no lime or iron oxide. Besides 
 the above sulphides, there was a small amount of chalcopyrite 
 and pyrite seen scattered through the mass. The ore was 
 coarsely crushed when received. An assay gave the following 
 results : 
 
 Original ore: 161.8 ounces silver and .12 ounces gold per ton. 
 
 Experiment VIII. Although difficulties in treating this ore 
 in the cold might have been expected a priori, still it was 
 
 26 
 
thought of interest to carry out some preliminary experiments 
 with it under conditions as nearly economic as possible, name- 
 ly, in the cold and with ordinary crushing. 
 
 First Treatment. The ore was crushed only to 4O-mesh 
 to start with, and, knowing it to contain not much iron to 
 depend upon to supply the necessary amount of oxidant by 
 going into solution during the treatment, the mass was treated 
 with a stock solution containing 2% FeCl :l , and H,SO 4 and 
 NaCl to the extent of 3 and 20% respectively. The weight 
 of ore taken being 300 grams, the volume of solution used 
 was therefore 300 c. c., that is, in the same relative proportion 
 as in all the previous experiments. The resulting mixture 
 of ore and solution was kept in an open vessel for three days 
 at room temperature, but without introducing any free chlorine 
 into it. 
 
 Twice during this period, however, part of the solution 
 was decanted off and replaced with an equal volume of the 
 fresh stock solution so as to facilitate the reactions by keeping 
 a good supply of ferric iron on hand and, at the same time, 
 by removing the reaction products, especially the ferrous iron, 
 which would form and accumulate in the solution to a great 
 extent. 
 
 After this period the mass was thrown on a filter paper 
 in a funnel, leached, washed first with strong salt solution, 
 then with pure water, and finally carefully dried and assayed. 
 The following were the results thus obtained: 
 
 ~ (100.1 oz. silver per ton 
 
 lailingfs, ore No. 3, first treatment ( 
 
 I .06 oz. gold per ton 
 
 If we compare these results with those of the assay of the 
 original ore, we see that this corresponds to an extraction of 
 only about 38.2% of the silver and 50% of the gold present. 
 
 Second Treatment. To see to what extent a prolongation 
 of treatment could improve matters, the above tailings were 
 treated with a fresh lot of the stock solution and allowed to 
 remain in contact with this latter for three days more, part of 
 the solution being again decanted off and renewed twice during 
 this period. The results obtained were : 
 
 ( 82.5 oz. silver per ton 
 Above tailings, after second treatment { ^ " r . , 
 
 j Traces of gold per ton 
 
 27 
 
Which corresponds to an extraction of about 49% of the 
 silver 'present and practically all the gold. 
 
 It is to be noted here that while thus far the extraction 
 of the silver cannot be called satisfactory, yet that of the gold 
 is strikingly good. To what was the latter fact due? 
 
 Although it is not known what was the exact state of 
 combination of the gold originally present in the ore, it is \\-\ 
 unlikely, however, that this metal was found there in some 
 readily soluble form, such that it was gradually removed dur- 
 ing the successive renewals of the solution by decantation or 
 filtration. The fact that the ferrous iron which was always 
 present in the solutions during the treatments, in larger or 
 smaller amounts did not hinder such a removal of the gold 
 must have been due first to the extreme smallness of the 
 amount of this metal present in the ore, and then to the fact 
 that the precipitation of gold by means of ferrous salts 
 not quantitative when other oxidizing agents, such as fen ir 
 salts or atmospheric oxygen, are also at hand. 
 
 Third and Fourth Treatments. In order to discover whether 
 finer crushing would help the extraction of the remaining 
 silver materially, the above (second) tailings were now 
 crushed down to a fineness of 70 meshes and then subjected 
 to a third treatment with the same stock solution for thr *e 
 more days. 
 
 However, as even the tailings thus resulting assayed 63 
 ounces of silver per ton (corresponding to a total extraction 
 of 61%), a fourth treatment of the same tailings, covering 
 three more days, was next undertaken, this time crushing the 
 ore down to 100 meshes and increasing the percentage of 
 ferric chloride in the solution to 5%. This last treatment, 
 however, did not materially alter the preceding results, as the 
 tailings now indicated a total extraction of only 63% of the 
 silver, originally in the ore, during the four treatments, cover- 
 ing a total period of twelve days. 
 
 This naturally rendered it necessary to use the agency of 
 heat so as to accelerate the reactions, and thereby determine 
 whether the fault lay in the process itself or was due to the 
 character of the ore. 
 
 Before doing .this, however, it was thought interesting to 
 decide first whether it was the length of time of treatment 
 
that governed the degree of extraction of the silver in the 
 above experiment, or whether the latter depended mainly upon 
 'the fineness of division of the ore. 
 
 Experiment IX. To answer this question, another experi- 
 ment was therefore undertaken, this time with a fresh lot of 
 ore, crushed to loo-mesh to start with, and with a solution 
 containing Fed., to the extent of 5%, as in the preceding 
 test. The results obtained after a four days' treatment in 
 the cold, with a daily decantation and partial renewal of the 
 solution, showed almost exactly the percentage of extraction 
 arrived at at the end of the twelve days in the above-repeated 
 tests with coarser crushing. 
 
 This proved definitely that the ore was of an exceptionally 
 refractory character, so far as the silver contents were con- 
 cerned. 
 
 The necessity of "forcing" the decomposition of the sul- 
 phides in this ore by carrying the treatment in the warm, 
 especially if finer crushing was to be avoided, became the more 
 apparent at this juncture, as it was noticed at the end of the 
 fourth day, even in the preceding test, that the further action 
 of the solution upon the ore became extremely weak. 
 
 Experiment X. A fresh 3OO-gram lot of the 4O-mesh ore 
 was now taken and a series of tests made with it, by boiling 
 the mass in a flask in presence of a solution of 3% H.,SO 4 , 
 20% NaCl, and 5% Fed,, for the total period of 22 hours. 
 The treatments lasted from three to six hours each, the solu- 
 tions being, of course, renewed for each new test. The fol- 
 lowing is a tabular statement of the data thus obtained : 
 
 EXPERIMENT X. HOT TREATMENT OE ORE NO. 3 
 
 Ag. An. Per cent 
 oz. per ton. oz. per ton. extr. silver 
 
 Original ore 161.8 .12 
 
 Tailings, ist treatment, lasting 5 hrs.* 55.2 65.9 
 
 Tailings, 2nd treatment, lasting 3 hrs. 36.1 777 
 
 Tailings, 3rd treatment, lasting 3 hrs. 30.0 81.5 
 
 Tailings, 4th treatment, lasting 6 hrs. 12.2 92.5 
 
 Tailings, 5th treatment, lasting 5 hrs. 9.0 94.5 
 
 At the end of these 22 hours of treatment, the solutions 
 collected were analyzed (approximately) and found to have 
 
 *The solution was once decanted and renewed at the end of the 
 first 2 hours. 
 
 29 
 
extracted antimony to the extent of 3% of the weight of the 
 ore, and as much as 2% of lead, besides some copper, and 
 silver as shown above. 
 
 From these results it appears clearly that "Chloridation" 
 will succeed even with ores as refractory and "base" as that 
 under consideration, provided the agency of heat is introduced 
 to accelerate the decomposition of the sulphides.* 
 
 It is now of interest to consider what results may be ex- 
 pected of this method of ore extraction as applied to very poor 
 ores or tailings. 
 
 EXPERIMENTS WITH ORE NO. 4 
 
 This ore was a reddish, oxidized mass containing much 
 limestone, ferric oxide, copper carbonate to the extent of 
 .5%, some silver, and some gold. 
 
 Experiment XL As the ore contained much CaCO 3 and 
 Fe 2 O 3 , it was thought to utilize these by using a rather strong 
 solution of HC1 alone for the lixiviating agent, and thereby 
 omitting the introduction of the chlorides of sodium and iron. 
 
 Two hundred grams of the ore were therefore taken and 
 treated with the above acid, so that when all the carbonates 
 had been completely decomposed, there was an excess of 10% 
 HC1 left free in the total volume of 200 c.cs. of solution. 
 
 Although the ore contained a small percentage of MnO 2 , 
 which, reacting upon the HC1 present, gave some free chlorine, 
 nevertheless, in order to insure the attack of the gold, the mass 
 was gently chlorinated and kept at room temperature for three 
 days, with occasional shaking. The following are the results 
 thus obtained : 
 
 EXPERIMENT XI 
 
 Au. Ag. 
 
 Sample Oz. per ton. Oz. per ton. 
 
 Original .09 4.2 
 
 Tailings .02 1.4 
 
 Which correpsonds to about: 66.6% silver extraction 
 
 and 77-8% gold extraction. 
 
 In order to ascertain whether or not these low results were 
 
 *The possibility of giving a preliminary roasting to such ores ii 
 lieu of heating the solution has not yet been considered. 
 
referable merely to the conditions of experimentation, several 
 other experiments were carried out under varying conditions 
 as to the composition of the solution, using H 2 SO 4 and NaCl, 
 etc., but it was found that in nowise could the above results 
 be improved. Of course, the possibility of using heat in con- 
 nection with the treatment of such low-grade ores in practice 
 being out of the question, no attempt was made to test the 
 ore with hot solutions. It is well to say here, however, that 
 just as it is likely that such hot treatments would give better 
 extractions with the above ore, as with the others, by disinte- 
 grating the coarser grains in it to whose presence these low 
 results are probably due just so it is probable that, had these 
 coarser particles been ground finer, the extraction would have 
 been improved, even with ordinary cold treatments. Never- 
 theless, it seems doubtful if tailings of the type of the above, 
 namely, those rich in Fe 2 O 3 , can ever be totally deprived of the 
 precious metals, owing to the fact that such ores are likely 
 to retain by occlusion, as is well known, considerable amounts 
 of chlorides, those of silver and gold among others, in a man- 
 ner difficult, if not impossible, to obviate. 
 
 Besides the above ores, others were experimented with, 
 some mere tailings, others fairly rich in silver and gold. With- 
 out going into a detailed account of these various experiments, 
 the following conclusions, based upon the experience thus far 
 gained by the writer, may be safely drawn : 
 
 1. The Chloridation method, as above described, may well 
 be depended upon to extract the precious metals from their 
 ordinary ores of good grade in a very satisfactory manner, 
 especially as regards the extraction of the silver values. 
 
 2. The thoroughness of extraction of the gold values de- 
 pends entirely upon the degree of oxidation of the reacting 
 solution by means of chlorine. 
 
 3. In many cases, with silver ores carrying gold, a suffi- 
 cient percentage of this latter metal is also extracted, besides 
 the silver, to pay a large share, if not the whole, of the ex- 
 penses of treatment, and this with no more care than is to be 
 given for the extraction of silver alone. 
 
 4. With very low-grade ores, or tailings rich in oxide of 
 iron, the highest extractions can only be had economically by 
 fine crushing. 
 
5. With very base ores of good grade the process is too 
 slow to be paying if carried out in the cold. In such cases, 
 unlike with ordinary ores, the agency of heat, must be resorted 
 to, either by way of warming the solutions or by giving a 
 preliminary roasting to the ores. 
 
 6. In practice the solution used in any case may be either 
 salt and sulphuric acid, or salt (or calcium chloride) and 
 hydrochloric acid, or hydrochloric acid alone, or any other 
 suitable combination, according to local economic facilities. 
 As to the supply of iron salts needed, it is probable that it 
 will seldom be necessary to introduce. such salts into the main 
 solution from without, since ordinarily ores will be found to 
 contain enough of the compound of this metal in a soluble 
 form to furnish all the oxidizing agents necessary to enable 
 the solution to do its work properly, once these salts of iron 
 are brought to the ferric condition. 
 
 ELECTROLYSIS 
 
 There remains now to consider the electrolysis of the solu- 
 tions, with a view to recovering the precious metals dissolved 
 thereby, oxidizing the ferrous salts, and therefore regenerating 
 the ferric salts and the solution as a whole. 
 
 So far as the recovery of the metals is concerned, it seems 
 to be safe to say that, in view of the many vain attempts made 
 by the writer to obtain good adherent deposits on the cathodes, 
 it is useless to trouble one's self by trying to get any such 
 deposits in large-scale practice. 
 
 Without going into any unnecessary accounts as to the 
 details of these various attempts, it is sufficient to say here that 
 if the solution be kept in motion during the electrolysis, which 
 is a necessity for continuous work, it is found that a current 
 density of .1 ampere per square decimeter, with a voltage of 
 1.5 to 2 volts, and with platinum electrodes, will not deposit 
 the values on the cathode at all; whereas this same current 
 will succeed in precipitating all the values if the solution 
 be kept quiescent. The reason for this peculiarity lies, 
 undoubtedly, in that the deposit being slimy or spongy 
 and easily rubbed off the surfaces of the cathodes by the 
 
 32 
 
solution in motion, this latter thus has a stronger ten- 
 dency to re-dissolve the values than has a current of the given 
 density to precipitate them. By increasing the current den- 
 sity, however, the deposition is carried to completion, although 
 as slimy as ever. 
 
 If, however, a very small percentage (the fraction of a per 
 cent) of glue or any mucilaginous substance is added to the 
 solution, the character of the deposit obtained is wholly 
 changed. Instead of the loose and black precipitate on the 
 cathode, one now gets, as a rule, especially if the solution be 
 warmed, a good, firm, and whitish gray deposit, which takes 
 a high polish when rubbed with the handle or back of a knife. 
 Moreover, it now matters no longer whether the solution is 
 left quiescent or stirred vigorously or even kept boiling. The 
 deposit is always good, provided the current density is proper. 
 Thus, by reason of the very fact that this deposit sticks firmly 
 to the cathode, it would seem as if with a small "trick" like 
 this one might get good and firm deposits and thus stop the 
 values, once deposited on the cathodes, from falling off into 
 the solution and there being subject to redissolution. 
 
 Unfortunately, however, sometimes, in spite of the best in- 
 tentions and care on the part of the operator, the solution 
 refuses to give good deposits, even with the addition of glue. 
 And then the deposit is again slimy and black. For these 
 and other reasons liable to cause possible hitches in the work 
 of electrolysis, in practice it might be better to have recourse 
 to some mechanical device whereby the slimy metals on the 
 cathodes may be continuously removed before they have any 
 chance to fall off into the solution. 
 
 A handy method in point is one described by Tomassi.* 
 
 Briefly described, it lies in that the cathodes are made 
 of circular plates, only half of each of which dips into 
 the solution. They are, further, so arranged that they can 
 rotate at a slow rate, say once or twice a minute. In conse- 
 quence of this rotation there is always some new surface being 
 covered with slimy metal deposit, while other portions emerge 
 from the solution already so charged. In its motion the ex- 
 ternal part of a plate comes to rub itself against some rubber 
 
 *Comptcs Rcndus, Academic des Sciences de Paris, vol. 122, p. 1476. 
 
 33 
 
pads provided with long grooves leading to a suitable recep- 
 tacle near by. Along these grooves, serving as channels, the 
 slimes, loosened by the friction between rubber pads and 
 cathode plates, are drifted by means of small jets of a dilute 
 solution of salt, and thus carried into the receptacle where the 
 values are thus collected. 
 
 As for the materials of which the electrodes may best be 
 made in connection with this work, it is obvious that carbon 
 or graphite would be the most suitable as regards durability, 
 desirability, and cheapness, especially as no other cheap metal 
 could stand unattacked in such solutions of ferric salts. 
 
 As regards the oxidation of the ferrous salts in solution 
 or the regeneration of the ferric salts, the writer is in position 
 to state here that, in order to bring this about by means of 
 the electric current, there is absolutely no necessity of using 
 any diaphragm cells of any sort. If the current density be 
 properly adjusted and the metals deposited or satisfactorily 
 removed from solution, whether by using glue or by means 
 of the above, or any other, mechanical contrivance, the oxida- 
 tion of the ferrous salts accumulated in the solution during the 
 latter 's passage through the ore mass can go on concurrently 
 with the electro-deposition of the metals, so that ultimately 
 there are not only no ferrous salts left unoxidized, but what 
 is more, the solution will be found to be charged with free 
 chlorine. It is important to call special attention to this fact, 
 inasmuch as certain authorities and all the inventors of 
 "patent'' methods, who have so much as specified the use of 
 the electric current for the regeneration of the solution, have 
 persistently thought it to be impossible not only to thus re- 
 oxidize the iron salts in solution, but have actually feared that, 
 instead, the higher oxides of this and other metals dissolved 
 might be reduced to lower ones; and they have, therefore, 
 advocated the use of some form or other of porous diaphragms 
 to separate the cathodes from the anodes in the electrolytic 
 cells. 
 
 When it is remembered that diaphragms are not the best 
 things that could be wished for or even allowed in large-scale 
 practice, and in solutions of the character here under conside i- 
 tion, the importance of the foregoing observation can scarcely 
 be lost sight of. 
 
 34 
 
In actual practice, what should be the mechanical and othr; 
 arrangements to be adopted in treating ores on a serious scale 
 by the electro-chemical "Chloridation" method here advanced? 
 
 It would certainly be idle to enter here into any details of 
 the possible contrivances which might be devised and put into 
 use under such circumstances. Large-scale practice alone \\ i!l 
 decide questions of that character. 
 
 Suffice it to say here, however, that in putting the method 
 into practical use, the governing principles of operation w : !! 
 be the following: 
 
 The ores, suitably crushed, are to be kept in wooden vats 
 and treated with the stock solution best by simple leaching. 
 
 The teachings thus collected are made to pass continuously 
 through simple electrolytic vats where, on the one hand, the 
 dissolved metals are recoverd and, on the other, the solution 
 is reoxidized, or "regenerated," under the influence of the 
 electric current. The number of cells and electrodes, the cur- 
 rent strength, and the flow of the solution so arranged that, 
 by the time the solution leaves the last cell, it is satisfactorily 
 impoverished of ;i:s precipitable metals and completely oxidized, 
 chlorinated, in other words, fully regenerated; which done, 
 it is ready to be led over into the ore vats again, allowed to 
 leach through the ore masses, then passed through the electro- 
 lytic cells, and finally returned to the ore vats over again, and 
 so on, cycle after cycle. 
 
 When a given batch of ore is sufficiently impoverished of 
 its metals, it is treated with pure water so as to displace and 
 recover the stock solution impregnating the exhausted ore. 
 
 To what extent this latter operation may be economically 
 carried out, and what other modifications, in the line of sec- 
 ondary treatments, may be introduced into the process as a 
 whole, will be alluded to in the second part of this paper. 
 
 35 
 
II. PERSULPHATATION 
 
 HISTORICAL 
 
 Unlike solutions of cupric and ferric salts, and in spite of 
 the fact that they have come latterly to be regular articles of 
 commerce, persulphates do not seem to have as yet been used, 
 or even suggested, as direct agents for extracting metals from 
 their ores. There seems to be one allusion made in the litera- 
 ture on the subject* to the effect that persulphates could be 
 used in conjunction with cyaniding as a substitute for oxida- 
 tion by aeration, which is well known to be essential for the 
 hastening of the solvent action of the cyanides upon gold. 
 Although it appears even that a patent has been issued to a 
 German company to cover this particular use of persulphates 
 along with cyaniding, yet nothing seems to be known, in prac- 
 tical mining circles, of this method of ore treatment having 
 ever been put into actual, large-scale operations anywhere. 
 
 Considering the fact, however, that cyanides are decomposed 
 by persulphates, t it is not at all unlikely that this "patent" 
 suggestion, like so many others, has never- as yet left the con- 
 fines of patent specifications. 
 
 THEORETICAL 
 
 In the method of metal extraction here to be described, the 
 chief point of interest lies in that both the oxidation of the 
 sulphides of silver and of gold is brought about under the 
 influence of one and the same compound, namely, the persul- 
 phate used, under proper conditions, but without any necessary 
 dependence upon the generation of chlorine by electrolysis. 
 
 *Elbs, Ztsch. Angew. Chemic, '97, p. 195. 
 ^Journal Am. Chem. Soc., '97, vol. 19, p. 900. 
 JN. Tarugi, Centrlbl., '03, p. 616. 
 
 36 
 
In bringing this double effect about, advantage is taken of 
 the fact that persulphates decompose chlorides, setting chlorine 
 free, ready to attack gold, especially in presence of free acids, 
 and that, on the other hand, they attack sulphides, readily 
 converting them into sulphates, while they themselves break 
 down to simple sulphates. 
 
 The reaction taking place in the first case seems to be of the 
 type : 
 
 2NaCl+K 2 S 2 O 8 +H 2 SO 4 = Na 2 SO 4 +K 2 SO 4 +H 2 SO 4 +Cl2 
 
 the decomposition of the chloride being referable to the forma- 
 tion, first, of free persulphuric acid, thus : 
 
 K a S 3 O 8 +H a SO 4 = K 2 SO 4 +H 2 S,O S 
 
 which then reacts upon the chlorides in a manner now well 
 known : 
 
 2NaCl+H,S,O H = Na 2 SO 4 +H 2 SO 4 +Cl 2 
 
 While with the ordinary concentrations of acids and' persul- 
 phates the attack of chlorides can be called all but rapid at 
 best a fact which is rather fortunate from the practical 
 standpoint, seeing that the nature of the work to be done by 
 the chlorine thus liberated, namely, the solution of gold, re- 
 quires only a slow but continued oxidizing effect that of the 
 sulphides, on the other hand, is much more readily accom- 
 plished. In fact, this decomposition goes on readily, even in 
 neutral or slightly alkaline solutions, though it is certainly 
 more vigorous in the presence of acids. The reactions here 
 in question seem to take place as follows : 
 
 Ag 2 S+K 2 S 2 O 8 = Ag 2 SO 4 +K 2 SO 4 +S 
 
 though much of the sulphur forming the metallic sulphides 
 is oxidized completely over to sulphuric acid. 
 
 These being the mechanisms and character of the re- 
 actions immediately involved in the use of solutions of 
 persulphates for the dissolution and extraction of the pre- 
 cious metals, it is evident that, as with the chloridation 
 method, so also here it is essential for the successful work- 
 ing of the present method in actual operations to have these 
 solutions charged with a soluble chloride. The introduc- 
 tion of such chlorides is, indeed, important in this case, not 
 
 37 
 
merely because it is necessary to have some chloride pres- 
 ent in the solution so as to react with the persulphates 
 and liberate the chlorine necessary to dissolve the gold in 
 the ore, but also because, AgCl being no more soluble in 
 simple persulphate solutions than in ordinary solutions of 
 iron salts, there is here the same danger of much silver 
 being lost in consequence of the unavoidable presence of 
 chlorine compounds occurring as impurities in ores, unless 
 some appropriate solvent of AgCl is used here as with the 
 chloridation method. 
 
 But there is even a more grave reason for which such 
 a solvent of silver should be used in this connection, and 
 that is that silver salts form an insoluble precipitate of 
 peroxide of silver (AgD), in presence of persulphates, which 
 is only decomposed and dissolved satisfactorily by strong 
 chloride solutions. 
 
 In endeavoring to select an appropriate solvent for these 
 insoluble silver compounds, it was at first thought that, 
 unlike with the previous method, ammonia would be a de- 
 sirable substance to use in conjunction with the present 
 method of ore extraction, inasmuch as it would act as a 
 very efficient solvent of any silver chloride which might 
 form in the course of the treatment, while, at the same time, 
 the action of persulphates upon the sulphides would not 
 thereby be impaired in any way. This belief had, however, 
 soon to be abandoned. It was, indeed, observed that, far 
 from being a suitable substance for the purpose, ammonia 
 was decidedly objectionable, inasmuch as it was found to 
 be rapidly decomposed into free nitrogen and water when- 
 ever any appreciable amount of silver and a persulphate 
 were found in solution with it. This oxidation of ammonia 
 under these circumstances is due, doubtlessly, to a catalytic 
 reaction depending upon the formation, primarily, of the 
 peroxide AgO, which then attacks, more or less violently, 
 any free ammonia which may be present in the solution, 
 and causes its complete decomposition in a well-known 
 manner.* 
 
 The reaction is, indeed, so characteristic that the writer has 
 
 *Dammer's Handbuch Anorganischen Chemie. 
 
 38 
 
found it to be a very handy way for detecting the presence 
 of silver salts in chloride solutions, which is the more useful 
 as it is oftentimes impossible to detect silver in such solutions, 
 whether by any other simple precipitation method or by a mere 
 dilution of the solution with the hope of thus diminishing 
 the solvent power of the liquid on AgCl and thereby causing 
 a turbidity of this compound to appear. In carrying out this 
 test, a strong solution of a persulphate, preferably the am- 
 monium salt, is added to the solution to be tested, and then 
 this latter is made strongly alkaline with NH 4 OH, when, if 
 silver be present, a more or less rapid evolution of nitrogen 
 gas will be observed to take place.* 
 
 The use of ammonia as a solvent of insoluble silver salts 
 being thus out of the question, and none of the other non- 
 chloride solvents being sufficiently stable in presence of per- 
 sulphates, the choice of concentrated solutions of a chloride 
 thus becomes a necessity. 
 
 Now, as it is remembered that the use of persulphates as 
 reagents for treating ores of the precious metals was suggested, 
 on an earlier page of this paper, on the ground that they were 
 good oxidants, capable of being regenerated by the electric 
 current; and, further, as it is realized here that, though a 
 small percentage of a chloride in concentrated solutions of 
 sulphates is very advantageous for the electrolytic formation 
 or regeneration of persulphates, higher percentages of chlo- 
 rides are quite injurious to such regeneration, the question 
 may arise here as to whether the means above stated for hold- 
 ing silver salts in solution in actual practice would not render 
 the process as a whole inoperative, at least as a "cyclic" meth- 
 od. It is, indeed, only too evident that, an essential condi- 
 tion for the formation of persulphates by means of the electric 
 current being a high anodic current density, or better, a high 
 rate of discharge, per unit of area and time, of SO 4 ions at the 
 anodes, the introduction of unduly large proportions of a 
 non-sulphate electrolyte, such as NaCl for instance, will tend 
 to diminish the yield of the current in persulphate formation. 
 
 *This catalytic reaction has also been studied lately and similarly 
 explained by Hugh Marshall (7. Chem. Soc., '01, Abstr. ii, p. 156). 
 
 $Elbs, Ztschr. fur Elektrochemie, '95, p. 245. 
 
 39 
 
How such a serious difficulty may be overcome in practice 
 will be discussed further on. Suffice it to say here only that, 
 were there even no way of surmounting this difficulty, the fact 
 that persulphates are becoming more and more readily avail- 
 able as regular chemicals of commerce, so that they could be 
 economically secured for practical usage, this fact justifies the 
 study here of their suitability for the extraction of the precious 
 metals from their ores. 
 
 As for the question which may arise here regarding the 
 advantages such a method may have, at best, over the chlori- 
 dation method previously described, the following points may 
 be cited in answer : 
 
 1. Theoretically the use of a persulphate should be prefer- 
 able to that of ferric salts alone, because, unlike the latter, 
 whose reaction products, namely, the ferrous salts, are no- 
 torious for their tendency to reprecepitate the precious metals 
 from their dissolved state, persulphates break down merely 
 to sulphates, that is, salts having no such injurious effects 
 upon the solutions of these metals. 
 
 2. For this reason principally, simple gold ores could be 
 treated by persulphatation to a better advantage than by chlo- 
 ridation, or perhaps by any other wet method. 
 
 3. In practice one may have to treat ores that are too rich 
 in limestone to be economically handled by the chloridation 
 process, inasmuch as they would first have to receive a pre- 
 liminary acid treatment before they could be subjected to the 
 action of ferric salts, lest these latter should be precipitated 
 out and the process rendered more or less inoperative. Such 
 ores should be amenable to persulphatation to a better advan- 
 tage, seeing that persulphates can decompose sulphides of 
 silver, as was stated above, in neutral or slightly alkaline media 
 as well as in acid. 
 
 On the other hand, it is only fair to admit here that, with 
 very "base" ores, persulphatation cannot be expected to be 
 nearly as useful a method as chloridation, owing to the fact 
 that the satisfactory oxidation of such ores by wet methods 
 such as these, requiring, as they do, the heating of the reacting 
 solutions, persulphates would be rapidly decomposed under 
 these conditions and hence be useless as economic. agents for 
 treating this type of ores. 
 
 40 
 
To what extent the above premises may be justified will 
 now be studied in a quantitative way. 
 
 In the following experiments the persulphate worked with 
 has been the potassium salt throughout. This choice was 
 made for two chief reasons. In the first place because sodium 
 persulphate, whose use would suggest itself first, is unfor- 
 tunately not to be had on the market here, nor can it be made 
 readily in the solid condition. In the second place because 
 the ammonium salt, though it is of all the persulphates the 
 one most readily made or got, is not adapted for experiments 
 of this kind, let alone for large-scale use, owing to the fact 
 that it is rather unstable and easily decomposed into free 
 nitrogen or its oxides by auto-oxidation,* and especially under 
 the influence of the chlorine, which would tend to be liberated, 
 whether during the treatment of the ores or during the electro- 
 deposition of the metals. 
 
 As none of these disadvantages is seriously shared with by 
 potassium persulphate, hence its selection for the following 
 study. 
 
 EXPERIMENTAL 
 
 EXPERIMENTS WITH NON-ALKALINE ORES 
 
 As both the decomposition of sulphides and the solution 
 of gold can be best brought about by persulphates when some 
 free acid is present, it is obvious that whenever the use of 
 acid solutions is permissible they should be resorted to in 
 practice. In fact, there is a second advantage gained in such 
 a practice, namely, that under these conditions there are also 
 more or less ferric salts introduced into the solution through 
 the agency of the acidity of the liquids upon the compounds 
 of iron which may be in the ores, and hence the oxidizing 
 power of the solution is thereby enhanced. 
 
 The following experiments were carried out with a view to 
 studying the effect of persulphates under these conditions: 
 
 ORE NO. 2 
 
 This ore was that already described on a previous page. 
 As it represented more nearly the average type of good ores 
 
 *Marshall, /. Chem. Soc. (London), '01, abstr. ii, p. 156. 
 
 41 
 
ordinarily met with, it was thought best to study it first in 
 this connection. 
 
 In order to form an idea as to the possibilities of the method 
 in a general way as a means of precious metal extraction, a 
 set of experiments was undertaken in a preliminary way. 
 
 Experiment XII. The solution used in these tests was made 
 up of 10% H 2 SO 4 , 20% NaCl, and 2% K,S 2 O 8 . The tests 
 were carried out on a lot of 300 grams of ore, with the usual 
 corresponding bulk of solution, namely, 300 c.cs. in the pres- 
 ent case. The mass was put into a bottle and left standing in 
 the cold during a period of 3 days, with frequent shaking. 
 At the end of each 24 hours, however, it was thrown on a 
 filter, leached, washed, and dried, exactly as with the tests 
 previously described in connection with the chloridation meth- 
 od, and finally a sample was taken and assayed. This done, 
 the remaining tailings were treated with a solution of the 
 same concentration as above mentioned, for a second and third 
 time, only using a fresh solution each time. 
 
 The following were the results obtained : 
 
 EXPERIMENT xii 
 
 Au. Ag. Per cent Extr. 
 
 Sample oz. per ton. oz. per ton. Gold. Silver. 
 
 Original .55 49.45 
 
 Tailings after ist day's treatment .092 7.5 83.3 84.9 
 
 Tailings after a 2nd day's treatment .032 4.5 94.2 90.9 
 
 Tailings after a 3rd day's treatment .031 3.9 94.2 92.1 
 
 These results clearly show a decided improvement over 
 those obtained previously by chloridation alone. 
 
 As it is stated here that during these treatments there was 
 a strong odor of chlorine evolved by the solution, it may be 
 lightly thought a priori that the thoroughness of the extraction 
 of gold is due to the presence of this gas in solution. As it is' 
 remembered, however, that this same gas was also present in 
 the solution with the chloridation tests carried out with the 
 same ore, and even in larger amounts, it becomes evident that 
 the present remarkably high degree of extraction of the gold is 
 due to more than the mere presence of chlorine in solution. It 
 seems, indeed, that the chlorine on hand in the solution in 
 the present experiments being nascent, while that in the chlo- 
 ridation tests was molecular, the favorable difference here 
 
 42 
 
observed should be referred to this particular fact of the chlo- 
 rine being here in a nascent state. If so, this would constitute 
 another advantage of persulphatation over chloridation proper, 
 or perhaps any other form of chlorination, especially as a 
 method of gold extraction. 
 
 Now, since the percentage of free acid used in the above 
 preliminary tests was somewhat unduly too high from the 
 practical standpoint, it was of interest to study the effect 
 of varying the amount of the free acid in the solution on the 
 extraction of the metals, with a view to using lower percentages 
 in practice. 
 
 Experiment XIII. A new set of tests was therefore under- 
 taken, using two solutions, one containing 3%, the other 5% 
 H 2 SO 4 , with 2% of K 2 S 2 O S and the normal amount (20%) 
 of NaCl in each. The treatment, carried out as before in all 
 other respects, lasted only 24 hours, and gave the following 
 results for the tailings assayed : 
 
 XTIT 
 
 Au. Ag. 
 
 Sample Oz. per ton. Oz. per ton. 
 
 Tailings (3% H 2 SO 4 ) .3 5.75 
 
 Tailings (5% H 2 SO 4 ) .22 5.1 
 
 From these results it would appear that the chief function 
 of the acid lies in that it improves the extraction of gold 
 alone, while that of silver is left practically constant. 
 
 Experiment XIV. In order to decide whether this rela- 
 tionship was merely fortuitous or not, another set of experi- 
 ments was carried out, this time using a solution of only i% 
 K 2 S 2 O 8 , but extending the time of treatment to 48 hours, with 
 a daily decantation and renewal of the solutions, the acidity 
 in the latter being again 3% and 5% respectively. 
 
 EXPERIMENT XIV 
 
 Au. Ag. 
 
 Sample Oz. per ton. Oz. per ton. 
 
 Tailings (3% H 2 SO 4 ) .30 9.0 
 
 Tailings (5% H 2 SO 4 ) .163 8.4 
 
 Whence it is clearly seen that in reality the chief function 
 of the acid in these solutions lies in that it facilitates the dis- 
 
 43 
 
solution or extraction of the gold rather than that of the 
 silver values. In other words, the alteration of the concen- 
 tration of the acidity in these solutions affects the extraction 
 of the first metal rather than that of the second. And the 
 general results of these experiments have been corroborated 
 in a third set of tests. 
 
 This point having been thus definitely ascertained, it became 
 next important to see what effect the variation of the concen- 
 tration of the persulphate would have upon the extraction 
 of these precious metals. 
 
 Experiment XV. For this purpose two tests were made, 
 using two solutions of i% and 2% K 2 S 2 O 8 respectively, while 
 both were acidified to the extent of 3% H 2 SO 4 .* The results, 
 after 24 hours' contact of the ore with the solutions, were : 
 
 EXPERIMENT XV 
 
 Au. Ag. 
 
 Sample Oz. per ton. Oz. per ton. 
 
 Tailings (i% K 2 S 2 O 8 ) .30 13.3 
 
 Tailings (2% K 2 S 2 8 ) .30 5.7 
 
 Strange as it may seem, these results clearly indicate that 
 the variation in the concentration of the persulphate in solu- 
 tion affects especially the extraction of silver alone. In other 
 words, the function of the persulphate would seem to be espe- 
 cially one of oxidation of the silver compounds. 
 
 Experiment XVI. In order to ascertain that this was not 
 either a mere fortuitous coincidence, a new test was carried 
 out using a solution containing 3% H 2 SO 4 and 2% K 2 S 2 O 8 , 
 but extending the time of treatment to 48 hours. This was 
 done so as to be able to compare it with the results of a pre- 
 vious test (given in table preceding the last), where only 
 i% K 2 S 2 O 8 was used, with also 3% H 2 SO 4 and a 48-hour 
 treatment. The following is a statement of the results of these 
 two tests: 
 
 *In this and in all the subsequent experiments, as in those pre- 
 ceding, the treatment was carried out in the cold, and the solution used 
 contained the normal amount (20 per cent) of salt, unless otherwise 
 stated. 
 
 44 
 
EXPERIMENT XVI 
 
 Au. Ag. 
 
 Sample Oz. per ton. Oz. per ton. 
 
 Tailings (i% K 2 S,O 8 ) .3 9.0 
 
 Tailings (2% K 2 S 2 O 8 ) .27 4.8 
 
 Whence it is clearly seen that in reality the concentration 
 of the persulphate affects principally the oxidation of the silver 
 sulphides. 
 
 From these various experiments, then, it becomes unmis- 
 takably evident that the best conditions of operation in treat- 
 ing such ores would be to use higher concentrations of both 
 acids and of persulphate as far as practicable, say 5% of the 
 former and 2% of the latter, since solutions containing larger 
 amounts of the one would be inconvenient and uneconomical, 
 while with the other difficult to obtain, K 2 Sv,O 8 being only 
 slightly soluble in aquous media. 
 
 In order to determine how favorably the present method 
 would compare, under such average conditions as to the con- 
 centration of the acid and persulphate in solution, with the 
 chloridation method in extracting the precious metals from 
 the ore under consideration, another experiment was now 
 undertaken. 
 
 Experiment XVII. In this test the concentration of the 
 solution in acid and persulphate was 5% H 2 SO 4 and 2% 
 K 2 S 2 O 8 respectively, wjth the usual percentage of common 
 salt. Since at no time during the treatment with such a per- 
 centage of acid in solution would there be the large excess of 
 chlorine on hand which was the case with the chloridation test 
 of the same ore as previously described, it was thought best 
 to equalize the conditions by having recourse to a continuous 
 percolation of the solution through the ore mass, as would be 
 the case in actual, large-scale practice. To do this, the ore 
 (200 grams) was put in a cylindrical funnel, the lower (inner) 
 orifice of which was covered with a small piece of filter paper, 
 and the solution was poured on the mass and let' leach gradu- 
 ally until one charge had entirely disappeared. When this 
 was done, a lot of fresh solution of the same percentage com- 
 position as the above was poured on the mass and again al- 
 lowed to leach, the charge of solution being then renewed 
 
 45 
 
as before, and so on. It must be noted here, however, that 
 the first lot of solution passing through the ore mass having 
 been analyzed and found to have extracted iron from the ore 
 to the extent of .4%, all subsequent, newer additions of solu- 
 tions were treated with the corresponding amount of FeCl 3 , 
 so as to render the conditions more nearly as they would be 
 in actual practice. 
 
 When ten such additions of solution and thorough perco- 
 lation of same were completed, which required about a week 
 in all, the last teachings were tested for silver, and they being 
 found free of this metal, the extraction was called complete. 
 Fresh water was now added to the mass to displace the saline 
 solutions impregnating the ore before assaying it.* This done, 
 the ore was taken out, dried, and assayed, giving the follow- 
 ing results: 
 
 EXPERIMENT xvn 
 
 Au. Ag. 
 
 Sample Oz. per ton. Oz. per ton. 
 
 Original .55 49.45 
 
 Tailings .04 2.7 
 
 These correspond to an extraction of about : 94.6% silver 
 
 and 92.7% gold. 
 
 These results make it clear, therefore, that, as compared 
 with the chloridation method at least as applied to this par- 
 ticular ore the persulphatation method is far more efficient. 
 
 Having obtained such encouraging results with this ore, it 
 was thought of interest to extend this study to richer and 
 more exceptional ores. 
 
 ORE NO. I 
 
 Experiment XVIII. 300 grams of ore No. I, described on 
 a former page, was taken, and, after attacking it first with a 
 simple solution of H 2 SO 4 to dissolve the metallic iron contained 
 
 *It is of great importance, from the practical standpoint, to note 
 here that when a weight of solution one-half that of the ore used was 
 collected, it was noticed that the teachings now became remarkabjy 
 free from acid or salt, so much so, indeed, that silver nitrate would 
 give but a slight turbidity with the percolating liquid, while tasting 
 could not at all detect the presence of any saline matter in solution. 
 
 46 
 
in it, enough of K 2 S 2 O 8 was added to oxidize completely the 
 ferrous sulphate thus formed to the ferric condition, which 
 done, common salt, more H 2 SO 4 and K 2 S 2 O 8 were added to 
 the resulting solution, so that the latter now contained 20%' 
 NaCl, 10% H 2 SO 4 , and 2% K,S 2 O 8 . The experiment was 
 carried out exactly as was done with the first persulphatation 
 tests with ore No. 2 (page 42), that is, the ore was treated 
 first for 24 hours, then it was filtered, washed, dried, and 
 assayed as usual, after which it was again treated for a sec- 
 ond 24 hours and assayed, and then treated for a third 24 
 hours and assayed for a third and last time. 
 
 The following is a statement of the results obtained): 
 
 EXPERIMENT xvm 
 
 An. Ag. Per cent Extr. 
 
 Sample oz. per ton. oz. per ton. Gold. Silver. 
 
 Original 1.03 86.2 
 
 Tailings after ist 24 hrs.' treatment .66 32.00 36.0 62.9 
 
 Tailings after a 2nd 24 hrs'. treatment .4 13.2 61.2 84.7 
 
 Tailings after a 3d 24 hrs'. treatment .4 7.5 61.2 91.3 
 
 Experiment XIX. These results not being quite satisfac- 
 tory, a second set of experiments was undertaken, this time 
 working with three separate lots of the ore. These were 
 treated with a solution and in a manner identical to the above, 
 only one being assayed at the end of the first 24 hours, a sec- 
 ond at the end of 48 hours, and the third at the end of 72 
 hours, the solutions, with the last two, being merely decanted 
 off and renewed once every 24 hours, instead of filtering, 
 drying, and re-treating them, as was done in the preceding 
 experiment. The results now obtained were as follows : 
 
 EXPERIMENT xix 
 
 Au. Ag. Per cent Extr. 
 
 Sample oz. per ton. oz. per ton. Gold. Silver. 
 
 Original 1.03 86.2 
 
 ist tailings (24 hrs'. treatment) .57 34.5 44.7 60.0 
 
 2nd tailings (48 hrs'. treatment) 31.2 63.8 
 
 3d tailings (72 hrs'. treatment) .4 25.5 61.2 70.7 
 
 Leaving the low figures of the first and second tests aside, 
 that the third test, lasting three days, should also thus fall so 
 far below the degree of extraction previously obtained with 
 the chloridation method (page 24) with this same ore becomes 
 now quite unaccountable. The difficulty grew more mysteri- 
 
 47 
 
cms, indeed, as it was especially noticed during the lapse of 
 the experiment that in all these tests the solutions were always 
 kept well oxidized, as was evidenced by the weak, but dis- 
 tinct, odor of chlorine which evolved from them. 
 
 Experiment XX. Thinking that matters might perhaps 
 improve if the solution were left in contact with the ore during 
 the three days without changing it, or again, perhaps by pro- 
 longing the time of treatment, two more tests were carried 
 out under these new conditions, keeping the percentage com- 
 position of the solutions used as that in the preceding experi- 
 ments. These, however, gave no better results, as can be seen 
 from the following data: 
 
 EXPERIMENT xx 
 
 Au. Ag. 
 
 Sample Oz. per ton. Oz. per ton. 
 
 Tailings (3 days' treatment) .6 2I -75 
 
 Tailings (4 days' treatment) .5 21.00 
 
 In order to see how far the use of HC1, instead of H 2 SO 4 , 
 would improve the results, another experiment was now under- 
 taken, which was also prolonged for 4 days, with the follow- 
 ing results: 
 
 EXPERIMENT XXI 
 
 Au. Ag. 
 
 Sample Oz. per ton. Oz. per ton. 
 
 Tailings (with HC1, 4 days) .6 14.7 
 
 These results were certainly not very much better, as they 
 still left the high extractions obtained with the Chloridation 
 method unattained. 
 
 Experiment XXII. It was finally thought that perhaps a 
 treatment under the conditions of Experiment XVII (page 
 46), namely, resorting to a continuous percolation of the solu- 
 tion through the ore mass for a long period, would give better 
 results. A new test was therefore started under these condi- 
 tions, using a solution of about the same percentage composi- 
 tion, in every respect, as that of the preceding experiments, 
 the total duration of the treatment being 7 days. Unfor- 
 tunately, however, even the results thus arrived at were still 
 far from being satisfactory, as can be seen from the following: 
 
 48 
 
EXPERIMENT xxn 
 
 Au. Ag. 
 
 Sample Oz. per ton. Oz. per ton. 
 
 Tailings (by percolation) .33 14.1 
 
 Having thus studied all the various conditions as to the 
 concentration of the reagents in the solution, the modes and 
 the durations of treatment, with a view to attaining more satis- 
 factory results, and having thereby ascertained that none of 
 these could have been responsible for the low extractions ob- 
 tained by the present method, as against those obtained by 
 chloridation, it was now thought that perhaps the sole expla- 
 nation of the difficulty might lie in the relative abundance 
 of chlorine held in solution in these two methods. It is indeed 
 a matter of fact that while in the experiments with ferric salts 
 alone there was always used a large abundance of chlorine 
 in solution, in the present tests, on the other hand, there was, 
 comparatively speaking, only a small excess of chlorine on 
 hand at any time. Now since when sulphides are decomposed 
 and oxidized by means of chlorine, the oxidation of the sulphur 
 in combination takes place more completely, resulting in the 
 formation of H,,SO 4 , when a large excess of this gas is on 
 hand, whilst it only results in the separation of spongy sul- 
 phur when the chlorine on hand is not sufficiently abundant, 
 it naturally seemed that the low extraction of the precious 
 metals with the persulphatation method, as applied to this 
 ore, may have perhaps been due to the formation of notable 
 amounts of just such spongy sulphur, which might then have 
 coated the particles of the sulphides in the ore, and thus pre- 
 vented their further decomposition. 
 
 Now, if this were so, it is clear that the question would be 
 to remove this sulphur by some appropriate method, and in 
 some appropriate manner, whether by a complete oxidation 
 or by a mere dissolution and removal of this element from 
 the ore mass. 
 
 However, as the complete oxidation of sulphur to sulphuric 
 acid could best be brought about, from the practical stand- 
 point, by the introduction of sufficiently large excesses of 
 chlorine gas into the solution, and as even such a procedure 
 could not always be brought about in large-scale work, it be- 
 comes clearly desirable to adopt some lixiviation method 
 
 4Q 
 
whereby the sulphur may be removed from the ore mass by 
 the use of a simple, convenient solvent of that element, and 
 subsequently got rid of or recovered therefrom as far as 
 possible. 
 
 Now, of all the simple solvents of sulphur, the one which 
 would lend itself best to this purpose is certainly a solution 
 of a fixed alkali, such as NaOH, and for good reasons. In- 
 deed, not only could such a solution be economically made by 
 a simple electrolysis of a sulphate, say Na 2 SO 4 , in a diaphragm 
 cell, but, what is of special value, after the formation of the 
 desired amount and concentration of the free alkali, and the 
 subsequent use of the latter to dissolve the sulphur in ores as 
 desired, the impure, sulphuretted alkaline solution thus result- 
 ing could then be treated with the acid solution which had 
 formed and accumulated in the anode compartment of the 
 electrolytic cell during the electrolysis of the Na 2 SO 4 , and 
 thus made to yield the sulphur it had extracted from the ore 
 either as H 2 S or as free, precipitated sulphur, and at the same 
 time to regenerate the original Na 2 SO 4 , which can then be 
 electrolyzed again and used as desired. 
 
 How well would such alkaline solutions answer for the pur- 
 pose on hand? 
 
 Experiment XXIII. In order to determine this point, the 
 first tailings (3-day treatment) of Experiment XX were taken 
 up, treated with a 10% NaOH solution, and left over night. 
 The following day the mass of ore and solution was filtered 
 off and washed with pure water till it was free from alkali. 
 This done, it was now treated with the ordinary stock solution 
 of acid, salt, and 2% K 2 S 2 O 8 , and left stand 2 days in the 
 cold, with occasional shaking. 
 
 At the expiration of this period, the mass was filtered, 
 washed, dried, and assayed, giving the following results. 
 
 EXPERIMENT XXIII 
 
 Au. Ag. 
 
 Sample Oz. per ton. Oz. per ton. 
 
 Original ore 1.03 86.2 
 
 Original tailings (Exp. XX) .6 21.75 
 Tailings after second treatment, 
 
 subsequent to the treatment 
 
 with NaOH .11 3.00 
 
 So 
 
Which corresponds to an extraction of about : 96.9% silver 
 
 89.3% gold. 
 
 It thus becomes strikingly evident that the use of NaOH 
 in facilitating the further attack of an ore by oxidizing agents 
 is of a great and undeniable value. On the other hand, that 
 it is no less undeniable that the action of NaOH in bringing 
 about this remarkable condition of affairs is in reality due to 
 its power to merely dissolve considerable amounts of spongy 
 sulphur and help to remove it out of the ore mass, thus laying 
 bare the remaining particles of sulphides subject to the further 
 attack of the main oxidizing solution, is proven by the fact 
 that the now impure, brownish solution of NaOH, recovered 
 after the treatment of the ore with it and filtration of same, 
 gave a decided precipitate of yellow, flocculent sulphur, when 
 neutralized with H 2 SO 4 , along with a little H 2 S. 
 
 In fine, then, it is evident that even such extreme ores as 
 the one under consideration may be treated by the use of a 
 persulphate alone, irrespective of the introduction of any chlo- 
 rine gas from without, provided the use of an alkaline solution 
 is resorted to.* 
 
 EXPERIMENTS WITH OTHER ORES 
 
 Besides the above, two new samples of high-grade ores 
 were treated similarly so as to give a rigorous test to the 
 method here advanced. 
 
 Ore No. 5. This ore was obtained from Mexico, and rep- 
 resented a specimen sample of very high grade, consisting 
 chiefly of the simple sulphide of silver or Argentite (Ag 2 S), 
 a little ruby silver (Ag 3 SbS 4 ), gold (not free milling), and 
 a gangue of quartz, with some chalco-pyrite disseminated 
 through it. 
 
 Experiment XXIV. In order to form an idea as to the 
 degree of extraction which could be obtained with coarser 
 crushing, the ore was first ground to 5O-mesh only. The test 
 was carried out on a 2OO-gram lot by the continuous percolation 
 method, as previously described, using a solution made up 
 of 12%. HC1, i% FeCl 3 , and 2% K 2 S 2 O 8 , but with no NaCl 
 
 *It is scarcely necessary to note here that this same modification 
 in the method of ore extraction may be introduced also, if necessary, 
 in connection with the chloridation method, described in Part I of this 
 paper. 
 
 51 
 
whatever, as it was intended to have the large percentage of 
 HC1 used take its place. 
 
 The ore being unusually rich in silver, the complete removal 
 of this metal required fully 15 days, when the final washing 
 with pure water was given and the mass dried and assayed. 
 The following were the results obtained : 
 
 EXPERIMENT XXIV 
 
 Au. Ag. 
 
 Sample Oz. per ton. Oz. per ton. 
 
 Original 3.00 680.00 
 
 Tailings (5O-mesh) 1.32 102.00 
 
 Corresponding to an extraction of aboutf: 85% silver 
 
 and 57% gold. 
 
 Experiment XXV. In order to ascertain to what extent 
 finer crushing would help matters, these tailings were now 
 ground down to 70 meshes, and again percolated with the 
 solution. 
 
 Although this new treatment showed some more extraction 
 of values, as could be seen by a dilution of the percolates, yet 
 as the effect stopped being noticeable within less than even 24 
 hours, and in a manner which could not well be taken to mean 
 that all the remaining 102 ounces of silver per ton, for in- 
 stance, had so soon been extracted from the mass, it was 
 then washed with water, and a 10% solution of NaOH was 
 next allowed to percolate through it so as to dissolve the sul- 
 phur, which must have accumulated in the ore, and thus to 
 free the particles of .the ore from this encumbrance and facili- 
 tate their further decomposition by the solution of persulphate. 
 This done, the ore was again washed with water, and then 
 treated anew with the stock solution, when a large abundance 
 of silver and gold was immediately observed to have gone 
 into solution again. When this action (lasting over 3 days) 
 was also over, the ore was washed, dried, and assayed, as usual, 
 with the following results: 
 
 EXPERIMENT xxv 
 
 Au. Ag. 
 
 Sample Oz. per ton. Oz. per ton. 
 
 Tailings, after second treatment 
 
 (subsequent to treatment with 
 
 NaOH) .72 34.8 
 
 52 
 
These correspond to an extraction of about: 95% silver 
 
 76% gold. 
 
 Although, as it is here seen, the silver extraction was very 
 satisfactory considering the extremely rich character of the 
 original ore, there was still room for improvement as regards 
 the extraction of the gold values. 
 
 Experiment XXVI. A third experiment was now under- 
 taken, using the same solution as in the above tests and with 
 the above tailings crushed down to 100 meshes. The results 
 arrived at were: 
 
 EXPERIMENT XXVI 
 
 Au. Ag. 
 
 Sample Oz. per ton. Oz. per ton. 
 
 Tailings (third treatment) .48 24.0 
 
 Corresponding to an extraction of about: 96.5% silver 
 
 and 84% gold. 
 
 No further experiments were carried out with this ore, as 
 a further extraction of the precious metals was now becoming 
 increasingly difficult. 
 
 Ore No. 6. This ore was an average one, also obtained 
 from Mexico, consisting chiefly of argentite, galena, chalco- 
 pyrite, and some gold, with a gangue of quartz. 
 
 Experiment XXVII. Knowing it to be an ore of the same 
 type as the preceding one, this sample was crushed to 70 
 meshes to start with, and the experiment carried out under the 
 same conditions, in all respects, as with the above ore. 
 
 Owing to its being more of an average type, the extraction 
 of this ore was complete within a week (3 days prior to treat- 
 ment with NaOH and 3 days subsequent to it), when an assay 
 gave the following results : 
 
 EXPERIMENT xxvn 
 
 Au. Ag. 
 
 Sample Oz. per ton. Oz. per ton. 
 
 Original 1.9 106.00 
 
 Tailings .5 5.3 
 
 Which corresponds to an extraction of about: 95% silver 
 
 and 73% gold. 
 
 Although the extraction of gold is still thus seen to leave 
 much to be desired, no attempts were made to repeat the ex- 
 
 53 
 
periment with a finer crushing of the ore, as it was rather 
 certain that better results could thus be obtained, seeing that 
 the ore represented, as was alluded to, the same type of ore 
 as the preceding one. 
 
 EXPERIMENTS WITH ALKALINE ORES 
 
 There remains now to consider the treatment of alkaline 
 ores by means of a persulphate in neutral solutions, so as to 
 determine to what extent such ores may be amenable to a 
 treatment without necessitating the preliminary neutralization 
 of their carbonates with correspondingly large amounts of acids. 
 
 Ore No. I. As ore No. I, previously described, contained 
 a large percentage of CaCo 3 , corresponding to as much as 
 15% H 2 SO 4 , it was thought proper to study it in this con- 
 nection first. 
 
 One hundred grams of this ore, powdered, as in all the pre- 
 ceding experiments with the same ore, to I2omesh, was put 
 in a percolating cylinder and the mass treated with a solution 
 of 20% NaCl and 2% K 2 S 2 O 8 alone. 
 
 After 10 days of continuous percolation, necessitating ten 
 renewals of solution, the mass was leached with pure water, 
 dried, and assayed, giving the following results : 
 
 EXPERIMENT XXVIII 
 
 Au. Ag. 
 
 Sample Qz. per ton. Oz. per ton. 
 
 Original 1.03 86.2 
 
 Tailings .52 7.4 
 
 Which corresponds to an extraction of about : 91.4% silver 
 
 and 49.6% gold. 
 
 From these results it will be seen that, as might have been 
 expected, the silver values of the ore are satisfactorily extracted 
 by such a treatment with a neutral solution of potassium per- 
 sulphate, although the extraction of the gold values leaves still 
 much to be desired. Was this latter due to the extremely fine 
 state of division of this metal in the ore, or was it merely due 
 to the extreme slowness of the liberation of chlorine in such a 
 neutral solution of a persulphate? 
 
 54 
 
Unfortunately, as the writer has been unable to obtain other 
 alkaline, but fairly good, ores to test by this modified method, 
 he is not in a position to answer the foregoing question defi- 
 nitely. It may, nevertheless, be well to note here that ore 
 No. 4, which was almost as rich in carbonates as the above ore, 
 was also thus treated and found to give an extraction of about 
 the same order as when treated with the Chloridation method, a 
 fact which, considering the inherently difficult character and 
 extremely low grade of the ore, would seem to speak quite 
 favorably of the feasibility of economically treating certain 
 excessively alkaline ores of the precious metals by means of 
 a neutral solution of a mixture of salt and a persulphate, as 
 against their treatment by the chloridation method, which 
 would necessarily require a preliminary treatment of such 
 ores with an acid, a necessity which might not always be 
 trifling. 
 
 ELECTROLYSIS 
 
 So far as the electrolytic recovery of the precious metals 
 from these persulphate solutions is concerned, no detailed men- 
 tion need be made here of the various steps to be followed 
 in this operation, as it is, in all essential details, identical with 
 that mentioned in the first part of this paper in connection 
 with the chloridation method. 
 
 As regards the electrolytic regeneration of the main reagent, 
 the persulphate, however, things are unfortunately not quite 
 as simple. Indeed, the essential conditions for the electrolytic 
 formation of persulphates being principally dependent upon 
 the concentration of SO 4 ions transferred to the anodes (plat- 
 inum) per unit of time, the unavoidable introduction of large 
 amounts of a non-sulphate, such as NaCl or HC1, into the 
 solution is quite incongruous with the purpose at issue. For, 
 as has been mentioned on a former page, though a little amount 
 of a chloride is highly favorable to persulphate formation, yet 
 when higher concentrations of such a non-sulphate salt are 
 used, the yield of the electric current in persulphate formation 
 decreases very rapidly, owing chiefly to the fact that these 
 foreign compounds also take part in the transfer of electricity 
 through the solution, thus decreasing the rate of discharge 
 
 55 
 
of SO 4 ions at the anodes, and consequently the yield to the 
 current in persulphates. 
 
 In view of this fact, it becomes apparent that, in order to 
 obtain satisfactory yields in the formation of persulphates, 
 either the current density at the anodes must be increased 
 proportionately; or some method must be resorted to whereby 
 the bulk of the chlorides may be removed from the solution 
 before attempting to regenerate, or rather, reform the per- 
 sulphate consumed during the treatment of the ore ; or, again, 
 one must strive to separate some of the sulphate in solution 
 by some appropriate means, convert it into a persulphate, and 
 then return it to the main solutions. 
 
 Now a consideration of these possible methods of arriving 
 at the goal set forth will show that the first method, or that 
 depending upon the increase of anodic current density with 
 solutions of a mixture of sulphates and chlorides, is not the 
 best that could be desired, from the practical standpoint, and 
 for two reasons. First, this is so because the current densities 
 used in ordinary practice, even with pure sulphate solutions 
 (namely, 50 to 500 amperes per square dm.), requiring, as 
 they do, voltages not below 5 volts, sometimes even as high 
 as 7 or 8 volts,* any further increase of the current densities, 
 which would be necessary to offset the injurious effect of the 
 presence of chlorides, would perhaps mean doubling or even 
 trebling the latter figures for the potentials required for the 
 purpose on hand, thus resulting in the decrease of the rende- 
 ment in persulphates per unit of power. . In the second place, 
 the regeneration of the persulphates needed by direct electrol- 
 ysis of a solution of sulphates in presence of much chlorides 
 is not desirable, because, the only metal admissible to serve 
 as anodes for such a purpose being the costly platinum, and 
 this metal being far from unattacked when used as anodes 
 in chloride solutions, an economic difficulty thus looms up 
 in the way of any continued practical operations in such a line 
 of work. 
 
 Further, owing to the fact that no special, economical 
 method can be devised whereby the dissolved chlorides could 
 be adequately removed from at least a part of the stock solu- 
 
 *Friedberger and Muller, Ztsch. fur Blektrochemie, '02, p. 231. 
 
 56 
 
tion before the electrolysis of this latter is carried qut in at- 
 tempting to reform the persulphate consumed, this second 
 method must also be considered impracticable, if not impossible, 
 for large-scale practice. 
 
 This leaves only the third method above suggested as a 
 means of regenerating the necessary persulphate, namely, 
 that consisting in the removal of part of the sulphate in solu- 
 tion from an aliquot part of the latter, converting it into a 
 persulphate in a special electrolytic cell, and finally returning 
 it to the main solution as needed. 
 
 A careful consideration of this possibility will show that 
 not only could such a method be the most rational and feasible, 
 but that the question thus viewed admits of more than one 
 solution. Indeed, since the problem is one of separating a 
 part of the sulphate from the stock solutions by crystallization, 
 it is clear that this may be brought about in two ways, to wit : 
 either by concentrating, by evaporation, an aliquot part of 
 the solution until the sulphate alone begins to separate the 
 original solution being more nearly saturated with respect 
 to this salt than with respect to the chloride in solution or 
 by taking advantage of the decrease in the solubility of certain 
 sulphates by lowering the temperature of their solutions. 
 
 Thus, if the sulphate worked with in practice be the potas- 
 sium salt,* part of the liquors holding this latter in solution 
 may be shunted off best directly after the recovery of the 
 precious metals therefrom and a desired amount of the sul- 
 phate made to separate out in the solid form, either by con- 
 centrating the solution with respect to this salt by evaporation, 
 or by cooling it sufficiently, according to practical facilities. 
 In either case, the solution being more nearly saturated with 
 respect to the sulphate than the chloride present, the former 
 salt will naturally separate out first, in form sufficiently pure 
 and free from admixed chlorides to be redissolved in a fresh, 
 sufficiently chloride-fre'e solution of the same sulphate, and 
 electrolyzed in special electrolytic cells, with a view to convert- 
 ing it, at least in part, to a persulphate, which may then be 
 removed from the solution and returned to the stock liquors. 
 
 While, with potassium sulphate chosen as the starting 
 material to work with in practice, the evaporation method 
 
 *Same would be true with (NH 4 ) 2 SO 4 . 
 
 57 
 
seems to be much more likely to be practicable, with the 
 sodium salt (Na 2 SO 4 ), on the other hand, the separation of 
 the solid Na 2 SO 4 from the shunted solutions by cooling seems 
 to be equally likely to meet with success in practical operations. 
 Indeed, the very rapid decrease of solubility of the Na 2 SO 4 
 with the lowering of the temperature of its solutions, and the 
 very slight decrease under the same conditions of the solu- 
 bility of NaCl which may be present in the same solution, 
 render this method highly useful as a means of bringing about 
 the separation of the former salt from a concentrated solution 
 of both these compounds in a form sufficiently free from the 
 latter for all purposes concerned, and that with only a slight 
 lowering of the temperature of the solution. 
 
 But to what extent can this happy coincidence be turned 
 into practical use, supposing Na 2 SO 4 to be chosen as the 
 starting material to be converted into the corresponding per- 
 sulphate? Would not the difficulty of preparing the persul- 
 phate of sodium in solid form owing to its excessive solu- 
 bility stand in the way of the sulphate of this element being 
 used, in actual commercial practice, as the starting material 
 needed ? 
 
 While not prepared to make light of this difficulty, the writer 
 is of the opinion that it is not absolutely necessary to prepare 
 this persulphate in the solid form, but that the sulphate of 
 sodium (whether original or recovered from the shunted parts 
 of the stock solutions) may be dissolved in water and then 
 converted, to any desired extent, into a solution of the corre- 
 sponding persulphate, which solution may then be returned, 
 in body, to the main or stock liquids. Nor is it, perhaps, true 
 that sodium persulphate cannot be obtained in the solid form, 
 although all authentic accounts of the authorities who have 
 dealt with the subject from a purely scientific standpoint either 
 seem to categorically deny its feasibility,* or at least make no 
 mention of any one having met with success in that line. In- 
 deed, according to Lowenherz, the persulphate of sodium can 
 be obtained without any special difficulty, provided the current 
 density is increased to unusually high limits. 
 
 *Marshall, /. Soc. Chem. Industry, '97, p. 396 ; Elbs, Ztsch. fur 
 Elektrochemie, '95, p. 245. 
 
 Lowenherz, Wagner's Jahrsbr. Chem. Technol, '95, p. 366. 
 
 58 
 
As the account of Lowenherz's work is got from his patent 
 specification, however, arid as no other authority seems to 
 have thus far proven it to be false or correct, it would be 
 unsafe to draw any foregone conclusions, in positive terms, 
 for or against such a possibility. Personally, the writer re- 
 grets that he has had no opportunity, as yet, to verify or dis- 
 prove the contentions of the patent.* 
 
 Just what should be the exact form of the cells and other 
 appliances to be used in this operation of regenerating the 
 persulphates it would be unnecessary to discuss here. The 
 question belongs, strictly, to the domain of practical, large- 
 scale operation, and could therefore be best studied and decided 
 upon under such conditions. Suffice it to say in this connec- 
 tion, however, that the contrivances and methods to be used 
 in practice need be neither complicated, involved, nor costly, 
 since one has to deal with only a rather limited volume of so- 
 lutions at a time, from which to crystallize the needed amount 
 of the sulphate, and then to "persulphate" the latter by elec- 
 trolysis. In other words, these contrivances need not be ma- 
 terially different from those made use of in the ordinary com- 
 mercial method of the preparation of persulphates. 
 
 RESUME 
 
 Two independent cyclic methods have been studied and ad- 
 vanced, in the present work, for the simultaneous winning of 
 the precious metals from their sulphide ores. 
 
 The first, or "chloridation," process depends upon the use 
 of acid solutions of ferric chloride or sulphate, in presence 
 of high concentrations of soluble chlorides, especially hydro- 
 chloric acid or sodium chloride, which act as solvents of silver 
 chloride. With sodium chloride used as a solvent, it was 
 observed that, under certain conditions, an insoluble complex 
 salt of silver and sodium chlorides would separate in form 
 
 *It is important to note here further that besides the above alkali 
 sulphates, that of iron, and especially of aluminum, could also be 
 used in this connection in converting them to their corresponding 
 persulphates. 
 
 $ See foregoing references. 
 
 59 
 
of definite isometric (cubic) crystals. Attempts were made to 
 obtain this salt in condition sufficiently free from extra sodium 
 chloride for analysis, but without success. This salt seems 
 to be best represented by the formula AgCl.NaCl-f xNaCl, 
 where x represents the number of molecules of extra sodium 
 chloride which will unavoidably crystallize isomorphously with 
 the double salt AgCl.NaCl. This may therefore be regarded 
 as an interesting case where double salt formation and iso- 
 morphism may coexist. 
 
 Various ores were treated by this process, and quantitative 
 data obtained showing an average extraction, under proper 
 conditions, of over 95%. silver and from 60% to 88% gold, 
 varying with the character of the ore. 
 
 The metals dissolved were recovered by electrolysis. It 
 was found that, under proper conditions, the reoxidation of 
 the ferrous salts formed during the decomposition of the sul- 
 phides, and the consequent regeneration of the ferric salts, 
 could be readily accomplished without the use of any dia- 
 phragms whatever. This is of special interest, as in all the 
 electro-chemical processes heretofore suggested for this pur- 
 pose the use of some diaphragm has been thought to be an 
 unavoidable, though troublesome, necessity. It was further 
 found that there was no difficulty in carrying the reoxidation 
 of the solution so far as to even charge the latter with much 
 free chlorine, which could then act as an efficient solvent of 
 gold. 
 
 A very small percentage of glue added to the solution was 
 found to give good, coherent deposits of the metals. 
 
 The second, or "persulphatation," method studied depends 
 upon the use of persulphates, also in conjunction with con- 
 centrated solutions of chlorides. Various experiments were 
 carried out to study the effects of varying the concentration 
 of acids and persulphate in solution. The best results were 
 obtained with about 2% persulphate of potassium and 5% 
 sulphuric acid. The extractions correspond to over 95% silver 
 and from 70% to 92% gold. 
 
 It was observed that an intermediate leaching of the ore 
 masses with a solution of an alkali, such as sodium hydrate, 
 would facilitate the further attack of the sulphides. This 
 was found to be due to the fact that such alkalies dissolve 
 
 60 
 
away the sulphur and help to remove this element, which 
 otherwise forms a coating around the particles of the sulphides 
 and thus hinders their further decomposition. The reaction 
 giving rise to the formation of free sulphur may be represented 
 thus: 
 
 Ag 2 S+K 2 S 2 8 = Ag 2 S0 4 +K 2 S0 4 +S 
 
 One important advantage of this second process was found 
 to lie in the fact that it could be used with ores that are too 
 highly alkaline with carbonates to admit of the expenses of a 
 preliminary neutralization of the ore masses with acids, which 
 would be necessary if the first process were to be used. 
 
 A test has been suggested, in this connection, to detect small 
 amounts of silver held in solution by virtue of the presence 
 of chlorides. This test depends upon the use of a persulphate 
 in presence of free ammonium hydrate, when, if silver is pres- 
 ent, an abundant evolution of free nitrogen gas takes place. 
 This is due to a catalytic action depending upon the inter- 
 mediate formation of peroxide of silver and its immediate 
 action upon the free ammonia present. 
 
 61 
 
ACKNOWLEDGEMENTS 
 
 The writer wishes to extend here his grateful thanks to 
 Professors Edmond O'Neill and W. B. Rising, and to Dr. 
 F. G. Cottrell and all the members of the Faculty of this Depart- 
 ment, for assistance, advice, and kind considerations shown 
 to him while engaged in the foregoing studies. 
 
 62 
 
REC'D LD 
 
 OCT211957 
 
YC 68574