DOCTORAL DISSERTATION SERIES title OF COPPiU S/l/C$7£ $* AUTHOR pom . cc * UNIVERSITY. p /i DEGREE f it* lJ* ! <*} f. u m s m A*. / DATE. PUBLICATION NO. 7^*- in u UNIVERSITY MICROFILMS Q M MKI A B R H R . / / 7/ MICHIGAN FLOTATION OF COPPER SILICATE BY SELECTED ALKYL SUBSTITUTED POLYHYDROXY NITROSO PHENOLS By Marvin Duane Livingood A THESIS Submitted to the School of Graduate Studies of Michigan State College of Agriculture and Applied Science in p artial fulfillment of the requirem ents for the degree of DOCTOR OF PHILOSOPHY Department of Chemical Engineering 1951 ACKNOW LEDGM ENT The author wishes to express appreciation to P ro fesso r C. C. DeWitt for his guidance and encouragement in the p u r­ suance of this r e s e a r c h objective. His use of new and uncon- vaitional methods of approach to re s e a rc h problems has been a continuing inspiration. Mr. tance. Ishwarbhai A. Mr. P atel gave valuable laboratory a s s i s ­ Hillard Pivnick kindly made available certain o r ­ ganic compounds obtained from Sharpe and Dohme. The De­ partment of Chem istry furnished sev eral organic intermediate compounds used in the work reported in this thesis. ii Marvin Duane Livingood candidate for the degree of Doctor of Philosophy Final examination: 301 Olds Hall. Dissertation: November 26, 1951, 3:00 P. M. , Room Flotation of Copper Silicate by Selected Alkyl Substituted Polyhydroxy Nitroso Phenols Outline of Studies: Major Subject: Minor Subject: Chemical Engineering Mathematics Biographical Items: Born August 15, 1918, Undergraduate Studies, ical College, 1934-38 Corning, Kansas Oklahoma Agricultural and Mechan­ Graduate Studies, Oklahoma Agricultural and Mechanical College, 1938-40. Illinois Institute of Technology, 194041. Michigan State College, 1946-1951. Experience: Graduate Assistant, Oklahoma A. and M. College, 1939-40; Graduate A ssistant, Illinois Institute of Technology, 1940-41; Chemical Engineer, Arzone Products Company, Chicago, 1941; Instructor, Missouri School of Mines and Metallurgy, 1941-46; A ssistant P r o ­ fe sso r of Chemical Engineering, Michigan State College, 1946-1951 (Assigned to Engineering Experiment Station as of 1949). TABLE OF CONTENTS Page I. Introduction A. B. C. D. Field of Thesis Discussion of Flotation Terminology P r io r Art in Flotation of Copper Silicates Scope of This Problem II. Experimental Work A. B. C. D. E. III. Design of Collector Flotation Machine Preparation of Reagents and Ores Experimental Procedure Experimental Data, as Tables and Figures Discussion of Results A. B. C. D. E. F. G. H. I. P ercen t Recovery as a Function of pH P ercent Recovery as a Function of Coll­ ector Concentration Enrichment as a Function of pH Enrichment as a Function of Collector Concentration Comparison of Nitroso Alkyl Resorcinols and Nitroso Resorcinol E thers Pounds Copper Recovered per Pound Collector as a Function of Collector Concentration Improvement Factor as a Function of C o llec­ tor Concentration Improvement Factor as a Function of pH Tests on Commercial Copper Queen Ore IV. Appendix: Conclusions Organic Synthetic Work Bibliography 1 1 2 5 8 10 10 16 19 20 23 63 64 66 68 69 71 74 77 81 86 92 94 116 L IS T O F T A B L E S Page Table I: Table II: Table III: Table IV: Table V: Table VI: Table VII: Flotation of Chrysocolla By Nitroso Alkyl Resorcinols (30-120 mesh feed) 23 Flotation of Chrysocolla By Nitroso R esorcinol E thers (30-120 mesh feed) 24 Flotation of Chrysocolla by Nitroso Alkyl Resorcinols (Minus 120 mesh feed) 25 Flotation of Chrysocolla by Nitroso R esorcinol Ethers (Minus 120 mesh feed) 26 Derived Data on Chrysocolla Flotation by Nitroso Resorcinol E thers 75 Derived Data on Chrysocolla Flotation by Nitroso Alkyl Resorcinols 76 High Concentration Tests 85 Table VIII: Flotation of Copper Queen Ore 87 Table IX: 88 Flotation of Copper Queen Ore (Series Test) L IS T O F F IG U R E S Page 1. Atomic Arrangement of Copper Silicate Structure With 2-N itroso- 4 -Alkyl Resorcinol Attached 11 2. Flotation Cell (Photograph) 17 3. Flotation Cell (Drawing) 18 Percent Recovery and Enrichment as Functions of pH: 4. Minus 120 mesh feed, 0. 2 Lb. /Ton Collector 27 5. Minus 120 mesh feed, 0. 1 Lb. /Ton Collector 28 6. Minus 120 mesh feed, 0.05 L b ./T o n 7. 30-120 mesh feed, 0. 2 Lb. /Ton Collector 30 8. 30-120 mesh feed, 0. 1 Lb. /Ton Collector 31 9. 30-120 mesh feed, 0.05 L b ./T o n Collector Collector 29 32 P ercent Recovery and Enrichment as Functions of Collector Concentration: 10. 11. 12. 13. 14. Nitroso-4-Hexyl Resorcinol, feed NAR-6, Nitroso-4-Octyl Resorcinol, feed NAR-8, Nitroso-4-Decyl Resorcinol, feed -120 mesh 33 -120 mesh 34 NAR-10, -120 mesh 35 Nitroso Hexyl Resorcinol Ether, mesh feed NRE-6, Nitroso Octyl Resorcinol Ether, mesh feed NRE-8, -120 36 -120 37 vi Page 15. 16. 17. 18. 19. 20. 21. Nitroso Decyl Resorcinol Ether, NRE-10, mesh feed Nitroso-4-Hexyl Resorcinol, mesh feed NAR-6, Nitroso-4-Octyl Resorcinol, mesh feed NAR-8, Nitroso-4-Decyl Resorcinol, mesh feed NAR-10, -120 38 30-120 39 30-120 40 30-120 41 Nitroso Hexyl Resorcinol Ether, NRE-6, mesh feed 30-120 Nitroso Octyl Resorcinol Ether, NRE-8, mesh feed 30-120 42 Nitroso Decyl Resorcinol Ether, NRE-10, mesh feed 43 30-120 44 Pounds Copper Recovered per Pound of Collector and Improvement F acto r as a Function of pH: 22. Minus 120 mesh feed, 0.2 L b ./T o n Collector 45 23. Minus 0.1 L b ./T o n Collector 46 24. Minus 120 mesh feed, 0.05 L b ./T o n Collector 47 25. 30-120 mesh feed, 0.2 L b ./T o n Collector 48 26. 30-120 mesh feed, 0. 1 Lb. /Ton Collector 49 27. 30-120 mesh feed, 0.05 L b ./T o n Collector 50 120 mesh feed, Pounds Copper Recovered per Pound of Collector and Improvement Factor as a Function of Collector Cone.: 28. Nitroso-4-Hexyl Resorcinol, feed NAR-6, -120 mesh 51 Vll Page 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. Nitroso-4-Octyl Resorcinol, feed Nitroso-4-Decyl feed NAR-8, -120 mesh 52 Resorcinol, NAR-10, -120 mesh 53 Nitroso Hexyl Resorcinol E ther, NRE-6, mesh feed -120 Nitroso Octyl mesh feed -120 Resorcinol Ether, NRE-8, 55 Nitroso Decyl Resorcinol Ether, NRE-10, mesh feed Nitroso-4-Hexyl mesh feed Resorcinol, NAR-6, Nitroso-4-Octyl mesh feed Resorcinol, NAR-8, Nitroso-4-Decyl mesh feed Resorcinol, NAR-10, -120 56 30-120 57 30-120 58 30-120 59 Nitroso Hexyl Resorcinol Ether, mesh feed Nitroso Octyl mesh feed 54 NRE-6, Resorcinol Ether, NRE-8, 30-120 60 30-120 61 Nitroso Decyl Resorcinol Ether, NRE-10, mesh feed 30-120 62 Solid surface graphs, P ercen t Recovery, Enrichment, Pounds Copper Recovered P e r Pound Collector, and Improvement F acto rs varying with pH and Coll. Cone.: 40. Nitroso Resorcinol E th ers, Minus 120 mesh feed Pocket 41. Nitroso Alkyl Resorcinols, Minus 120 mesh feed Pocket v iii Page 42. Nitroso Resorcinol E thers, 30-120 mesh feed Pocket 43. Nitroso Alkyl Resorcinols, 30-120 mesh feed Pocket INTRODUCTION The large scale recovery of economic minerals from r a n ­ domly mixed aggregates of naturally occurring run-of-mine ore has for many years been the particular field of the mining engin­ eer and more recently of the chemical engineer. ations of flotation, tion, extraction, leaching, The unit oper­ sedimentation, adsorp­ and absorption have all grown from the need of sorting natural mixtures or those mixtures incident to certain manufac­ turing processes. The mining engineer from the first has made use of these unit operations on an empirical basis; the chemical engineer’s contribution to this a re a has been in the form of fun­ damental studies relating to these several unit operations. The process of flotation arose originally from mining engi n- eering practice an d has been developed in this field. nevertheless, It has, been possible for chemical engineers to make con­ tributions to the advancement of flotation practice, pa rticularly where their knowledge of basic chemical principles as applied to surface phenomena come into play. Chemical engineers have also adapted flotation to the separation of industrially occurring mix­ tures of solids [27a], 2 Flotation, als, as applied to the recovery of economic m ateri­ is defined as the selective separation of a mixture of r e l ­ atively finely ground solids suspended in a liquid (generally water) by mechanical agit ation in the presence of bubbles of gases (generally air) and reagents which act as collectors, frothers, or dep ressan ts. To the ground pulp containing the economic and undesir­ able particles a collector is added followed by a period of s t i r ­ ring or conditioning. Directly after the conditioning perio d a suitable frothing agent is added and air bubbles are beaten into the suspension. The collector reacts and/or adsorbs selective­ ly on the surface of the economic p a rtic le s . If one-thj.rd of the area of the desired particles is covered with the collector and one of these particles is thrust inside an air bubble the p a r ­ ticle orients itself so that sthe coated portion rem ains in the air interface of the bubble while the wetted portion is in the water interface. The difference in specific gravity of the p a rtic le and the apparent specific gravity of the water due to the presence of many finely ground p a r t i c l e s urges the a ir bubble with its* min­ eral load to rise to the surface where it may be skimmed off, dewatered and saved. sm elter. It may then be refloated or sent to the The machines used in flotation separation of solids are generally known as flotation cells. There are several types; all of them possess the following parts: (1) an agitator, shaped so a s to mix the contents of the cell thoroughly and keep even coarse p a rtic le s in suspension; (2) a source of air or other gas which terminates close to or around the agitator, the latter being so arranged that air or other gases are finely divided and thoroughly beatenainto athe pulp; (3) provision for an exit open­ ing at the bottom from which tailing or waste is removed; (4) an overflow launder at the top, generally with a stilling section between ait and the main cell body in order to prevent p a rtic le s being c a r r i e d over the launder by mechanical action; and (5) a jneans for introducing the pulped solids and the reagents into the cell. The flotation reagents are grouped generally as frothers, collectors, depressants, and activators. F roth ers are chemicals whose sole function is to produce a stable a ir bubble; they should ideally produce no other influence on the flotation process. Col­ lector sO function to attach themselves to the mineral particle and render it water repellent. Depressants are chemicals which ex­ ert a re v e rse action to that of collectors; they cause a particle to be readily wettable and hence difficultly attachable to an air bubble. Activators alter the surface of a mineral particle, either by cleaning it or by producing a layer of m aterial with g re ate r affinity for the collectors than that of the original min­ eral. The pulp may be prepared by wet grinding or by adding dry ground solids to water and mixing or conditioning the pulp prio r to i t s ’ introduction into the cell. In many cases, a p o r­ tion of some of the flotation reagents is added to the pulp while it is being conditioned. Conditioning may take place for long or short periods of time, and may be applied to wet or d ry ground m aterials. Wet grinding is usually considered ad­ vantageous in that the newly exposed surfaces are cleaned by mechanical action and are hence in a more receptive condition for selective adsorption or possible chemical reaction. advantage of this phenomenon, To take some workers have proposed ad­ ding reagents to the pulp while it is being wet ground [24a, 121], Flotation operations are tested by considering the degree of enrichm ent and the p ercentag e of the desired m ate rial o r ­ iginally present which is recovered in the froth concentrate or the tailings. The degree of enrichm ent may be expressed as the ratio of the p e rc e n t desired m aterial in the concentrate to that in the feed or heads; it may also be e x p re ss e d as the ratio of pounds desirable m aterial per pound undesirable to tthe concen­ tra te tp that in the heads. Secondary c r ite r ia in evaluating flot- 5 ation te sts are the rejection of m aterials from the concentrate which tend to interfere with further flotation or refining of the concentrate, and the production of an easily handled concentrate. Sometimes reagents used in one flotation step may interfere with reagents in a subsequent step for the separation of a different mineral from a com plex mixture. Some reagents are relatively expensive; these m aterials mustsjustify their use by producing proportionately better enrichments and recoveries. Two different c lasses of froth flotation are recognized. one of these, bulk flotation, than one metal, Ir^ several m inerals, containing more are separated from severalaother minerals. If the flotation is highly selective, i. e. , one mineral is separated from another or several closely related m inerals containing only one prim ary metal, differential flotation is said to take place. P r io r Art In Flotation of Copper Silicates From the initial p ate nts of Bessel and Bessel [17] who p ro ­ posed small amounts of various oils or various organic compounds for selectively wetting and floating graphite, and of Everson [42] who described hydrocarbons and sulfonated fatty oils* both inves­ tigators using air or gases introduced by agitation or chemical reaction, the line of development of froth flotation has arisen. 6 It is deemed sufficient to outline the major steps in the development of the flotation p ro cesses which have been extended by this r e s e a rc h on the flotation of copper silicate. works published by Gaudin [47], Wark [121], Standard and Taggart [116] have been consulted as well as the lite ra tu re published since their appearance for information concerning the flotation of copper silicate [35, 41, 45, 48, 95, 99, 100], Other flotation literatu re was consulted which has been of help in visualizing the theory and principles involved [5, 49, 50, 51, 93, 107, 110, 115, 117], 10, 19, 26, 27, 32, 37, The only work reported has been that sof Ayers and coworkers [7], and of Ludt and DeWitt [85], Dean and others [33, Until now, 34] no successful com m er­ cial application appears to have been made of any of the p rin ­ ciples proposed in these sev eral p ro cesses. Ayers [7] proposed to sulfidize mixed copper o re s after bringing the pulp to and maintaining in at pH 4. 8-6. 5. sulfidizing agent was specified as Na 2 S, calcium polysulfides. ated by flotation, The hydrogen sulfide or The metal compounds were then s e p a r ­ presumably by xanthates. Dean and coworkers [33] applied xanthate flotation to the separation of chrysocolla from a synthetic mixture with pegma­ tite (quartz and orthoclase with a little kaolin). The specified conditions were activation with sodium sulfide or hydrogen sul- 7 fide in aqueous pulp at a pH of 4. fide concentration was very critical, ressant. It was reported that sull an excess acting as dep­ It was found necessary to add the collector before the sulfidizing agent when using hydrogen sulfide as a sulfidizer. A recovery of 96% of the copper was made when op­ erating on a feed containing 0. 6% copper with amyl xanthate 0. 2 pound per ton, hydrogen sulfide 0. 2 pound per ton, and pine oil frother as required. In a second Bureau of Mines investigation by Dean and others [34] soap flotation was employed. ific as a collector. Soap was not spec­ The presence of certain metallic ions was found to promote formation of insoluble soaps which were deem­ ed u seless in the collection of the chrysocolla. F re e acids were not collectors for chrysocolla; the pH had to be main­ tained between eight and nine. Under these conditions, using the soap both as the collector and frother in amounts of three to twelve pounds per ton, the best re s u lts were secured on chrysocolla-pegmatite mixtures ground to pass 100 mesh screens. Ludt and DeWitt [85] used alkyl substituted triphenylmethane dyes as adsorbed collectors for from silicates and carbonates. of concentrations, pH levels, separating chrysocolla They investigated a wide range and other variables, and^ conclu- 8 ded that the triphenylmethane dyes studied are reasonably good collectors. It was not considered possible to single out sopti— mum operating conditions. Maximum recoveries of approxi­ mately 76% were reported, with maximum enrichments of 850 percent. It was not considered possible to combine high r e c ­ overies with high enrichments. Scope Of This Problem This re s e a rc h was undertaken with the idea of securing a reasoned approach to the flotation of m inerals. By the ap­ plication of known chemical and engineering principles to the design of a reagent a p ro cess was evolved to effect the sepa­ ration of a desired c lass of m ineral from m aterials naturally associated with it. The selection of a m ineral of potential commercial im por­ tance, concerning whose separation by flotation little had been recorded, was considered desirable. fils these req uirem en ts. Chrysocolla amply ful­ It is p re sent in large quantities .in tailings dumps from American as well as foreign mills. One mine is e stim a te d to have produced for over forty years 22, 000 tons per day of tailings which contain 0. 3% of copper, largely as chrysocolla. per, This re p re se n ts 963, 600 tons of cop­ worth more than $481,000,000 at present market prices. 9 The recovery of any portion of this copper, mined and ground, which need not be should be a very re a l contribution to the extension of our copper re s e rv e s . The problem, then, was to investigate the basic chemi­ cal and engineering principles involved; to design a collecting reagent and a p ro cess which would specifically separate cop­ per silicate, chrysocolla (CuSi0 synthetic mixtures, 3 #2 H2 O), from silica-sand and to demonstrate the success of this separation when the same collector was used to beneficiate a commercial ore. In doing this, it was found necessary to synthesize most of the reagents used. available, These reagents were not commercially nor were they reported in the lite ra tu re . The syn­ thesis of these new compounds is reported in the Appendix. 10 EXPERIMENTAL WORK Design of Collector In designing a collector for the flotation of chrysocolla, consideration was fir s t given to the general physical and chem­ ical properties of t h i s mineral. ted copper silicates. It is one of a series of hydra­ It is said to occur in either green or blue modifications with an earthy overcast [18, 83, 90]. It may occur either in the colloidal or crystalline states. Ideal­ ized chrysocolla is a dihydrated copper metasilicate. The a metasilicate structu re is that of SiC>3 units linked in chains, with the four oxygen atoms tetrahedrally surrounding the sili­ con atoms occurring as two shared and two unshared oxygen atoms. C ross linking may occur between the chains, plates of SiC>3 lattices, minerals. forming from which are formed the mica type There may also be vertical linking of the rem ain­ ing oxygen atom, which accounts for the relatively stable quartz 2 n _ structu res. The copper atoms occur as links between (Si0 units with six-coordination numbers. hardness of chrysocolla (2. 4, 3 )n From the structure and Moh’s scale) as compared with other m inerals of known structure, it seems more likely that 11 Atomic S i l i c a t e \ F I C L HK 1 A rrangem ent S t r u c t u r e tfexyl of With l i e s o r n n o ! Copper 2 -Ni tr oso At t p c It e c1 12 the stru ctu ral units are sim ilar to the (Si3 (Si60 18)12” units [18, 38, 40, 98, 0 101, 111, g) fi — or the 122] in which the metal ions are six -coordinated with the unshared oxygen atoms in the units. These units a re arranged in layers not joined to each other; the la yers ween layers. are linked by the copper ions bet­ It seem s very likely that some of these chains or units may be at least partly incomplete; the existence of varying mineralization grouped somewhat loosely under the chrysocolla heading lends strength to this assumption. It is further suggested that some of the metal ions may exist at the surface without having their coordination possibilities realized. This would occur especially during the grinding process. ure Fig­ 1 shows a typical m etasilicate stru ctu re with a collector molecule attached. In the design of the collector itself, it was recognized that it is necessaryato have a link of some so rt between the m ineral surface and the collector molecule, together with a hydrophobic part which would confer the desired w ate r-re p e llency on the m ineral particles* if the collector were to adsorb and/or reach with the copper silicate. Linkage through the exposed copper atoms in the stru ctu ral units would seem to be most desirable, since the copper-oxygen coordinate linkages are relatively weak compared to those of the silicon-oxygen i 13 bonds in the stru ctu re of copper silicate. The literatu re d is­ closes that linkages might be formed between a carbon-hydrogen compound and a copper atom [2, 3, 4, 22, 24, 35, 39, 43, 44, 52, 53, 54, 57, 58, 59, 60, 72, 74, 76, 81, 82, 86, 89, 92, 97, 105, 108, 112, 120], This indicates that chelation or \ ring formation between substituents on carbon atoms in an aliphatic chain, or between substituents on two adjacent ring carbons in a benzene ring, At this point, Karzeff [14], p ossesses interesting possibilities. the work of Baudisch [11, Baudisch and Heggen [13], 12,] Baudisch and Baudisch and Roth­ schild [15] and others [8] on ortho-nitroso phenols and their substituted homologues was consulted. demonstrated that copper, metallic ions, Cronheim [28] had together with a few other divalent possessed the property of forming inner com­ plexes of good stability with ortho-nitrosophenol or i t s ’ p araalkylated homologues. limited, however, His application of this property was to qualitative detection of Co and Fe++ ions in solution. The relatively easy formation of the chelate link b et­ ween copper ions and ortho-nitrosophenols suggested two des­ irable byproducts. F irst, the presence of a benzene nucleus in the collector molecule would be no hindrance but a definite help in promoting w ater-repellency, provide d the other substituents on the ring did not themselves promote water avidity. The size of the benzene ring may be worked out from the C-C and C=C distances, tively, 1. 54 and 1. 32 %. re s p e c ­ the C radius of 0.771 or 0.665 *k., and the H radius, 0. 3 A. The C-H bond distance is the C-C bond average distance is 1. 08 %. in 1. 39 5L. benzene and This would give a breadth of 1. 39 + 1. 08 + 1. 08 + 0. 3 = 3. 85 5L and a length of 1, 39 + 1.00 + 1.08 + 0. 3 = 4.85 K as the relative in te r f e r ­ ing size of the benzene nucleus without any long chain sub­ stituent. The addition of a chain substituent on the benzene ring in the 4-position would add 1. 54n + 1. 35 & as an addi­ tional length of hydrophobic carbon chain which would s te r ically interfere with a relatively large a re a surrounding sit so far as wetting is concerned. The distance between adja­ cent silicon atoms in a chain-linked m etasilicate structu re / is 5. 2 & and if the chains are edge-linked the distance bet­ ween adjacent silicon atoms is 9.0 &. [40, 101], The extension of this reasoning dictated the choice which was finally made, that of testing the ortho-nitroso-para-alkyl phenols as collectors. This line of work was followed until unexpected difficulties, reported in the Appendix, made the purification of the nitroso derivatives well-nigh impossible. 15 During this period, information became available upon the synthesis and c h a ra c te ristic s of the nitroso-monoalkyl ethers of resorcinol and the program of work was shifted to the testing of the nitroso monoethers of resorcinol and the nitroso-4-alk yl reso rcin o ls . This program gave chemicals pos­ sessing the originally desired properties* Subsequent appli­ cations of such compounds proved successful. The fre sh rupturing of the copper-oxygen coordination bonds might be expected to provide active copper atoms which would more readily enter into the chelation complex with the nitroso-hydroxyl unit. This would provide activity in the solid m ineral analogous to that provided by the metallic cop­ per ions in solutions. This situation would be provided if the ore were milled in the presence of the collector and p re f e r ­ ably in the presence of water to clean clayey m aterial from the p articles. met. In this work neither of these conditions were The ore was milled dry and the collector was not added until the actual flotation operation. In spite of this, demons­ trable visual evidence of the production of the characteristic cojnplex color showed that chelation actually took place in the short contact time of the collector with the pulp in the flota­ tion cell. It seems reasonable to assume that wet milling in 16 the presence of the collector might make the yields even bet­ te r or make possible collector economies. The latter was not investigated for the reason that if the reagent were pro­ perly designed it should be able to actively find its* place on the copper silicate particles. Flotation Machine The flotation cell used in this work was a laboratory size sub-aeration type cell made of Lucite, whose details are shown in the illustrations (Figures 2 and 3). The agitator was driven through a V-belt from an electric motor; pulleys made possible variations in agitator speed. step Froth overflowed from the cell into an eight centimeter Buchner funnel; the filtrate from the Buchner was conducted to a suc­ tion bottle connected to a small vacuum pump. The suction bottle was so arranged that by cutting soff the vacuum, possible to drain the filtrate into a feed bottle. it was Liquid con­ tained in the feed bottle was fed into the cell during a run to maintain constant pulp level. Air was supplied to the bot­ tom inlet of the cell from a low -pressure blower. Reagents were added to the cell from a special titration burette having a long, curving tip which made it possible to introduce por­ tions of solution well below the actual liquid surface into the FIGURE 3 /O O a*. O T A 7~/0 A / C£~CL. 4 10 zone where they would be promptly dispersed into the pulp. P reparation of Reagents and Ores The chrysocolla and sand were dry ground in a one foot by one foot Abbe porcelain ball mill, grinding medium. using pebbles as the The ground m aterials were separated using Tyler standard sieves into two major fractions, that passing 30 mesh but retained on 120 mesh and that passing 120 mesh. Care was taken to avoid overgrindings of the fine m aterial. Sufficient m aterial was prepared at one time to supply the entire block of runs. The chrysocolla w as a selected sample which exhibited the proper earthy luster. sand used was Ottawa silica sand. mixtures, In preparing the synthetic the appropriate quantity of the ground chrysocolla and silica-sand were added separately to the cell and therein. The mixed T o tal charge weight was held constant at 100 grams. The frother used in all runs was steam distilled pine oil. It was added in sm all portions as required. The collectors tested were made up as alcoholic solutions. The appropriate amountwof collector was dissolved in absolute ethyl alcohol to give a solution containing 0. 0005 gram of col­ lector per m illiliter of solution. In a few cases, stock sol­ utions of one hundred times this concentration were made due 20 to tem porary lack of alcohol; these solutions were later dil­ uted to the final concentration. kept in tight, The collector solutions were glass stoppered bottles. There was no appar­ ent change in these solutions over periods of time. Reagents used for adjusting pH were C. acid and C. P. potassium hydroxide, P. hydrochloric the latter as a 5% aqueous solution. All pH m easurem ents were made using Hydrion test papers from a set of narrow range papers providing at least two checks on any single pH figure. Experimental Procedure The major portion of the flotation te s ts reported in the fcQlowing tables of data were made using only frother, collector, and pH adjusting agents. was used for all of these runs. A standard procedure Into the clean cell was in­ troduced sufficient water to leave room after introduction of the sample for a froth blanket of five to ten m illim eters. Approximately seventy cubic centim eters of water whose pH had been adjusted to that specified for the run was placed in the feed bottle. turned on, and the The agitator, a ir, and vacuum were sample introduced into the cell. The pH 21 was adjusted at once to that desired, using the minimum pos­ sible quantities of adjusting reagents. A conditioning period of five minutes was then allowed, at the end of which one small portion of pine oil was added. This usually resulted in an immediate dimunition of the natural froth present d u r­ ing all but very acid runs; the froth reappeared after the pine oil had become d ispersed.' lished, When froth was r e e s ta b ­ the collector was added at a regular dosage every thirty seconds from the dispensing burette. The dosages were so chosen that the flotation runs were not less than seven m inutes and not more than ten, minutes. the majority being ten The froth coming over from the beginning of col­ lector addition to the end of the run was collected on one seven centimeter medium filter paper in the Buchner funnel. No attempt was made to separate the froths coming over at the beginning of a run and those produced at the end of the run. Visual observations of the apparent composition and color of the froth were recorded. manually with a small sc ra p e r, The froth was removed care being taken toaremove only froth from the stilling zone. The appearance of the cake after it had been sucked dry on the filte r was recorded. the end of a run, the drain plug at. the bottom of the cell At 22 was removed and the contents allowed to drain into a Mason ja r. The tailings and liquid were kept in these j a r s and ob­ served visually for quantity and color. Both dried concen­ tra te s and tailings were preserved in case doubtful re su lts required check analyses or runs. The froth cake was placed in a stainless steel dish and dried at 115° C. Copper was determined by the volumetric method of P ark. Solutions were restandardized at appropriate intervals and check anal­ yses agreed satisfactorily. Experimental Data Tables I-IV Figures 4-39 23 TABLE I FLOTATION OF CHRYSOCOLLA BY NITROSO ALKYL RESORCINOLS (Feed 0.0456% copper, 30-120 m esh dry ground) 0,2 lb, /Ton Collector Fig, 7 : 0.1 L b./T on Collector Fig. 8 : 0.05 Lb,/T on Collector Fig. Gram s Percent E nrich­ fo Copper Grams Percent Enrich­ % Copper : Grams Percent Enrich­ % Copper : Froth Copper ment Recovered Froth Copper ment R ecovered: Froth Copper ment R ecovered: pH Nitroso 4-hexyl reso rcinol NAR-6 (Fig. 16) 5.510 pH 12 0,445 9.85 54.2 3.250 0,706 15.60 50.8 3,'810 0,613 13,60 52,0 12 pH 8 5.350 0.503 11,20 59.7 3.000 0.739 16,30 49.0 4.300 0.550 12.15 52,2 8 pH 6 4.734 0.429 9,40 44.8 2.520 1.138 25.00 62.0 2.810 0,474 10.50 29.5 6 pH 4 4.4785 0.538 11.80 53.0 2.194 0,890 19,60 42.0 3.840 0,382 8,45 32,0 4 Nitroso 4-octyl reso rc inol NAR-8 (Fig. 17) pH 12 3.4785 0.950 21,00 72.7 3.041 0.901 19.90 67.7 2, 8762 0,982 21.70 62,3 12 pH 8 3. 2585 1.007 22.40 73.0 3.910 0,675 14.90 58,3 3.905 0,489 10.40 42,3 8 pH 6 4.410 0,722 15.95 70.3 3.248 1.038 23.70 77.1 4.214 0.368 8,13 35,2 6 pH 4 4.462 0.754 16.60 74.4 4.178 0.872 19,20 80.5 4,541 0.612 13.72 61,5 4 Nitroso 4-decyl reso rc inol NAR-10 (Fig. 18) 13.320 pH 12 0.337 7.47 99.0 10. 6885 0,428 9,40 95,5 8.540 0,512 11,30 96,8 12 pH 8 9.884 0,460 10,20 99.0 10.147 0.414 9,13 94.2 5.930 0.751 16.60 98,0 8 pH 6 6.880 0.599 13.20 91.0 6.950 0,613 13.50 94.1 6.978 0,459 10,10 70,2 6 pH 4 10.430 0,383 8.45 88.0 7,520 .0 .5 2 6 10.80 87,3 4.658 0, 3678 8,12 37.8 4 24 TABLE II FLOTATION OF CHRYSOCOLLA BY NITROSO RESORCINOL ETHERS (Feed 0.0456% Copper, 30-120 m esh dry ground) 0,2 L b,/T on Collector Fig, 7 : 0.1 L b./T on Collector Fig, 8 : 0,05 lb, /Ton Collector Fig, G ram s Percent E nrich- ft C o p p e r: G ram s Percent E nrich- ft Copper : Gram s Percent E nrich- ft Copper Froth Copper ment Recovered: Froth Copper ment R ecovered: Froth Copper ment Recovered Nitroso hexyl resorcinol ether NRE-6 (Fig. 19) 11,4450 pH 12 pH 0,277 6.10 70,0 9, 3785 0,460 10,20 94,8 6.400 0,430 9,48 60,7 12 pH 8 9,8422 0,407 9,00 88,4 10. 2335 0,432 9,53 97,8 5,500 0,368 8,10 45,5 8 pH 6 10.8785 0,382 8,40 91,2 7. 2550 0,614 13.70 96,5 4,430 0,400 8,83 38,7 6 pH 4 9, 9307 0,168 3.72 37,0 10, 7785 0,384 8,48 91,0 4,860 0,320 7,08 34,3 4 10, 10 8,16 98, 7 72, 7 10, 10,7070 10,00 12,20 12,20 12,20 12,20 12,90 8,8,1616 5,43 5.43 74,2 74,2 57,6 50,0 50,0 55, 55,88 62,2 46,2 46,2 21, 21,75 75 0,537 0,399 11,80 8,85 41,6 44,3 74,5 92,0 0,487 0,487 0,460 0,552 0,552 0,552 0,552 0,582 0,368 0,368 0,245 0,245 3,516 5,036 6,79 12. 20 6, 9200 6,9200 5.6529 4. 1035 4,1035 5785 4,4.5785 4,8325 5,5,6880 6880 4.0168 4.0168 5, 3099 5.4430 0,399 0,383 8,85 8,43 46,6 45, 9 12 8 7 6 5 4 3 0. 503 11, 10 99,9 6,2785 6,2785 0.690 0,690 15.00 15.00 95,4 95,4 3.400 0,504 10,90 37,0 3,613 0,459 9,90 35,4 Nitroso octyl resorcinol ether NRE-8 (Fig. 20) pH 1 2 , 9, 7585 0,460 pH 8 8,9697 0,368 pH 7 pH 6 pH 5 11. 3871 0.306 pH 4 7.5685 0.552 pH 3 Nitroso decyl resorcinol ether NRE-10 (Fig. 21) pH 12 8,8785 pH 8 6,4000 0,534 11, 50 73.5 3,7100 0,904 19,80 74,0 pH 6 4,3780 0.534 11,50 50,2 3,7000 3,7000 0,597 13,10 13,10 48,8 pH 4 4, 1300 0. 522 11,20 46,6 3, 3000 0,3075 6,73 22,4 12 25 TABLE III FLOTATION OF CHRYSOCOLLA BY NITROSO ALKYL RESORCINOLS (Feed 0.0456% copper, -120 mesh dry ground) 0.2 lb. /Tori C oilector‘ ' ' Fig. 4 T 0.1 k / T o n Collector Fig. 5 : 0.05 Lbl /Ton Collector Fig. : Gram s Percent E nrich­ to Copper : Grams Percent Enrich­ fo Copper : Gram s Percent E nrich­ to C o p p e r: : Froth Copper ment Recovered : Froth Copper ment Recovered : Froth Copper ment R ecovered: Nitroso 4-hexyl re so rc ­ inol NAR-6 , (Fig, 10) 41.3706 pH 12 pH 0.0613 1.35 56.0 16. 2235 0. 0854 1.88 30.7 32.1285 0.1288 2.85 91.3 12 pH 8 16.6100 0.184 4.07 67.4 23.5785 0.166 3.67 82.4 18.700 0.0920 2.03 38.0 8 pH 6 18.5110 0.160 3.54 65.3 19. 6985 0. 0859 1.90 37.3 13.729 0.117 2.56 35.3 6 pH 4 18.0040 0.0920 2.03 36.5 16.6265 0. 0883 1.95 32.4 13.433 0,1104 2.44 32.7 4 0.116 3.60 96.3 22. 3277 0.178 3.94 87.7 15. 3785 0.221 4.90 74.8 12 Nitroso 4-octyl re so rc ­ inol NAR-8 (Fig. 11) pH 12 37. 4785 pH 8 20, 2030 0.176 3.91 87.5 15.5135 0.166 3.68 56.8 9.8130 0.141 3.12 30,5 8 pH 6 15. 3785 0.184 4.07 62.3 15.9300 0.153 3.39 54.0 11. 2200 0.123 2.72 30.3 6 pH 4 16. 9000 0.172 3.81 64.0 15.4227 0.151 3.34 51.3 10. 7400 0.133 2.94 31.4 4 Nitroso 4-decyl re so rc ­ inol NAR-10 (Fig. 12) pH 12 21,3200 0.208 4,60 98.0 24.2252 0.184 4,07 98.5 15.8885 0.190 4.22 66.7 12 pH 8 12. 2040 0.190 4,22 51.2 13.8280 0.184 4.07 56.2 7.6785 0.251 5.57 42.6 8 pH G 9, 8885 0.190 4.22 41.5 10.2510 0,208 4.62 47.2 4.7540 0.123 2.71 32.2 6 pH 4 11. 8785 0.194 4.28 50.8 7. 6735 0.190 4.22 32.2 4.6025 0,147 3.26 37.3 4 26 TABLE IV FLOTATION OF CHRYSOCOLLA BY NITROSO RESORCINOL ETHERS (Fee'] 0,0456% copper, -120 mesh dry ground) G ram s Percent Enrich­ % Copper : Gram s Percent Enrich­ fj Copper : Gram s Percent' E nrich- It C o p p e r: Froth Copper ment Recovered : Froth Copper ment R ecovered: Froth Copper ment R ecovered: Nitroso hexyl resorcinol ether NRE-6 (Fig, 13) 35.8385 pH 12' pH 0.098 2.17 77.5 7. 7640 0,1532 3.39 25.9 34.4065 0.110 2.45 82.9 12 pH 8 26.6735 0.1164 2.57 68.7 10.4485 0.1532 3.39 35.1 29. 1385 0.110 2.45 71.0 8 pH 6 16,8128 0,166 3,67 61,4 9.1785 0,1840 4,07 37,2 23.4785 0,147 3.25 76.4 6 pH 4 23. 2785 0,123 2.71 63,0 16,1885 0,1690 3.73 60,2 23,8690 0,1288 2.85 67,7 4 N itroso octyl resorcinol ether NRE-8 (Fig. 14) 50, 9929 pH 12 0,0846 1,87 95,2 48. 8934 0,0772 1,72 83,3 41, 5608 0.0994 2.20 91,2 12 pH 8 22, 8885 0.1840 4.07 93,0 25.3185 0,1715 3,80 95.8 19, 8690 0,2145 4,75 94,0 8 pH 6 21, 0345 0,2165 4,80 99.5 15. 8281 0.2145 4,75 71.2 16, 6945 0, 2580 5.72 95,1 6 pH 4 13, 2551 0,1535 3,40 45,0 13.0450 0.1226 2,72 35.3 12.0940 0,1072 2.37 28,6 4 N itroso decyl resorcinol ether NRE-10 (Fig. 15) pH 12 52, 1728 0,0797 1,76 91,7 37.1141 0.110 2,45 87,8 11,0405 0,2300 5,10 56,1 12 pH 8 16. 8430 0.1840 4,07 64.8 15, 0903 0.276 6.12 91.8 13, 5896 0,1592 3,53 47,75 8 pH 6 16, 3305 0,1535 3.40 55.3 19, 3466 0,2145 4,75 91,4 10,8565 0.094 2,08 22,1 6 pH 4 14,1449 0,0767 1,70 23,9 11,5525 0.2145 4,75 54.8 13, 4785 0,1317 2.92 39,2 4 NRE 8 NRE 6 NAR 6 NAR 10 NRE 8 □ T1AB 8 NRE 6" NRE 10 ° NAR 6 4 6 8 10 12 4 pH 6 8 pH FIGURE 4 0 . 2 Lb. C o l l e c t o r p e r Ton Feed Minus 120 mesh feed 10 12 NAH 10 NHK 10 NAH 8 NRF 1 0 - AH I) NHF. 8 10 4 12 pH 6 8 pH FIGURE 5 0 . 1 Lb. C ollector p e r Ton Feed Minus 120 mesh feed 10 12 NRE 8 A NAR 8 NRE 6 NAR 6 ' /^^N A R 10 NRE 10 si S so - 40 : 30 w _NARJ0 ^ n^ J O IRE b NAR 6 4 6 8 10 12 4 6 pH 8 10 pH FIGIHE 6 O.OS Lh C o l l e c t o r p e r Ton F eed M i n u s 120 mesh f e e d 12 NAR 10 /NRE 10 --k ---/* / / \N A R 10 NXO" - - B / NAR 6 X y 40 NAR NRE 8 NRE 4 6 8 10 12 4 6 10 8 pH pH FIGURE 7 0 . 2 Lb. C o l l e c t o r p e r Ton Feed 30-120 mesh feed 12 NRE 6 NAR 10 NRE 10 NAR 6 VAR NRE 8 cs 50 NRE 8 NRE 6 x NAR 10 8 10 4 12 p8 6 8 pH FIGURE 8 0 . 1 Lb. C o l l e c t o r p e r Ton Feed 30-120 mesh feed 10 12 / / NAR 8 ^ /• NRE 6 *1NAR > 50 ' NRE & NRE 10 4 6 8 10 12 4 pH 6 8 pH FIGURE 9 0 . 0 5 Lb. C ollector p e r Ton Feed 30-120 mesh fee d 10 12 e U_&. 0.05 0.10 0.15 0.20 0.05 , Lh. C ollector/T on 0.10 0.15 0.20 Feed FIGURE 10 Minus NAR 6 120 mesh f e e d 00 oo pH 12 1 60 i 50 £ 40 a 30 12 0.05 0. IQ 0.15 0.20 0.05 Lb. C o l l e c t o r / T o n Feed FIGURE 11 M i n u s 120 mesh f e e d NAR 8 0.10 0.15 0.20 V © © © oo © © to © 2E M• 3 3 »i N9 M z O ?D 3 33 PTJ (% (0 o rr* N9 ft (% 2. 3r * o o P ercent to -* r> R ecovery cn O' oo rr O 1 H o a T .V (V Q. 00 o E n ri chm ent N3 © © 01 © © © © to © S£ to OO oo 90 I c / 12 « U / / /E l // / V " - " * - - P » u V8l W \ \ \ \ \ 80 8 ^O / X 16 14 6 *----------- * » / / 12 ft* ► 8 5° | cc >^7 y ^ / \ 18 / c 40 41 u / V X yT *30 CL 10 £ tE> r* »< 8 «^« b e 6 “ pll 6 20 4 10 0,05 0.10 0.15 2 0.20 0.05 lb . C ollector/T on Feed FIGURE 13 NRE 6 M ib u s 120 mesh f e e d 0.10 0.15 0.20 X •>30 0.05 0.10 0.15 0.20 0.05 Lb. C o l l e c t o r / T o n Fed FIGURE 14 M in u s 120 mesh f e e d NRE 8 0.10 0.15 0.20 0 05 0.10 0.15 0.20 0.05 Uh. C o l l e c t o r / T o n 0 . 10 0 45 0.20 Feed FIGURE 15 M in n s 120 Mesh f e e d NRE 10 w CD «- 30 0.05 0 . 10 0 . 15 6u 0.20 0.05 0.10 0.1 5 0.20 L.b. C o l l e c t o r / T o n Reed FIGURE 16 3 0 - 1 2 0 mesh f e e d NAR 6 w pH 12 0.05 0.10 0.15 0.05 0.20 Lb. C o l l e c t o r / T o n 0.10 0.15 0.20 Feed FIGURE 17 3 0 - 1 2 0 mesh f e e d NAR 8 k * o o to cs o o C n ao © »s3 © O ' n OJ o *— k—» * o • hW ts 3 O 3 (V Co O ' T k-H > 3= 8 33 rj r» o -I O w-> OD o 3 z m ’"*1 (V Percent ^ H 00 R ecovery (A ^ o to E n rich m en t T) (% a. a. o O o o W 1 o to o 00 It O O CO o to © © M © r* o- NRE 6 30-120 mesh ■n «—4 O C rn m © feed o o —< rj r* 3 "1 H 3 3 -n •» a. M © o 4* © R ecovery C n © E n rich m en t © • © tn © © © C n N 3 o Zf P ercent to 0\ © 00 © VO © pH 12 0.05 0.10 0.15 0.20 0.05 Lb. C o lV c t o r /T o n Feed FIGURE 20 NRE 8 0. 10 0.15 0.20 pH oo 12 o On KJ n o CO o — -n V V a. P ercent oo < % o ff -1 R ecovery on O' CO O O © to cr o o oo E nri chm ent 3 r» * eu o O' Ctk 00 140 0 £12 00 •cj o1 0 0 0 u r- 8 00 c 4. 4. E C4. Cc bOO > C U u fc. E N A H . § _ ^ i s * NPE 8 10 400 NAP 8 NHF. 10 M1E 8 NP F - 6 NAH 6 8 10 200 12 pl I FI CUKE 22 0.2 lb AiAR 10 C o l l e c t o r p e r i o n F ood Mi n u s 120 me s h f e e d 1400 w 1200 o l> i> u c 1000 * NAR NRE 1 0 / ^ * ^\ / / \ y NAB 8 600 \ nar 400 a NRE 10 200 t j NRE 8 oNRE 6 NAR fu. 4 pri 6 8 t>h FIGURE 23 0 . 1 Lb. C o l l e c t o r p e r Ton Feed Minus 120 mesh feed 10 12 NhK 1400 1200 1000 800 600 > SHE 8 400 & 200 8 10 N A H 10 NHL y f 8 12 10 12 pH PiI KICCHE 24 0 0 5 Lt> Co t l e c t o r Minns 120 me s h per I on F e e d feed -4 rr >» Ocl (tt. |o;) «j'j oo c~. tO O o Lh 4*. O 1-3 On Copper/!.b. O' C = . o 05 Co 1 l e c t o r O o to o CO it c O' o hector j)^^j uuj jdJ Improvement o o ■0 CO 'T / to O 8t O' 05 •— oo NliF ]() c 1200 c 1000 *■' cj « N HE 6 NhK 6 ——-A ’ar 10 P.F. 10 Nl.F 8 NliK 8 NAH 8 U bOO ------------ » NAIi b 10 FI Gl hK 2b 0 1 1.0 C o i l e r t n r pnr I nn Fp p 8 30 120 mpsh fpprf 12 NAH ]0 1 NAH 8 NAH 8 NAH 10 NUF 6 1000 - NAH 6 Nlih 8 ^NHE 1(1 NAP 6 t 600 NHF 6 NHF 8 - 8 10 NHF 10 400 12 8 pi! 10 12 pH KKii HK 27 0 06 !.»' O o l l e r t o r p e r Ton Feed 30 - ] 20 mesh feed O n O 800 u 600 JL. 400 ° 0. 05 0 10 0 15 12 200 0 20 Lh. 0.05 Co 1 1 < * c to r/T o n Fepd FIGUHK 28 Minus k. 120 m*sh fp o d NAH 6 0 10 0.15 0. % 1400 200 UlOOO m 0.05 0 . 10 0.15 0.20 LI. 0.05 Col 1 f e t o r / I o n 0.10 Fe e d FJGfjRF 29 Minus 120 mes h NAH-8 feed o r ■or 1400 1200 1000 800 O 600 400 200 0. 05 0 . 10 0 . 15 0.20 Lh. 0.05 C o ! 1c c f o r /T'orj Fe e d KIGUHF. 30 Minus 120 mesh NAH 10 feed 0 . 10 0 . 15 1400 1200 ’1000 800 600 400 200 0.05 0. IS 0. 15 Lh. 0 . 10 0.05 0 . 20 Col l e r t . o r / T o n 0 . 15 C. 20 Fp p H FiGUHE 31 Minns 120 me«h feed NHF b cn J400 1200 J000 ^ c. 8 0 0 js ' 600 400 200 0.05 0. JO 0. J5 0.20 0.05 Col J p r t o r / l o n FeeH FIGLHK 32 Mi n us 320 mesh f PPtJ NHF 8 w * C n 1400 1000 u. 800 i 600 400 200 0.05 0 . 10 0 . 20 0 . 15 Lh. 0 . 10 0.05 CoJ)ector/Ton Feed FIGl'HF 33 Mi n u s 120 mes h NHE 10 feed 0 . 15 0.20 1400 1200 ~ 1000 aoo 600 400 200 0.05 0 . 10 0. 15 0.05 0 . 20 L,b . Col 1 e c t o r / T o n KIGl'RF 34 3 0 - 1 2 0 mesh NAH 6 Feed 0 . 10 0 . 15 0.20 1600 01200 U *1000 ° I 600 200 0.05 U. 10 0.15 0.20 l b 0 . u5 (). JO 0 15 0.® Co 1 1 p t t o r / 1 on Fe e d FI GLUE 35 3 0 - 120 mesh NAF 8 feet) oo \ 1 600 1400 1000 m 800 I 600 e 400 200 0.05 ('. 10 0 . 15 0 . 20 1,0. 0.05 Co I I f t t o r f'\ on Fe e d FlGl.UE 36 3 0 - 1 2 0 mesh \ A K - 10 feed 0 . 10 0. 15 0 . 20 V © o cn oo o o l.h Collector/Ion o f>3 o h. o < 2. Copper/l.b. o ^ o o o § s § Improvement O Feerl o o © N J CO © tj o 09 C ollector Factor IbOO 1400 U C u 1000 800 60 0 400 200 0.05 0.10 0.1b 0.20 Lb. 0.05 C o i 1f e t o r / T o n 0.10 0.15 0.20 Fee O o o Lb. Lb. C o C5 C fapper/Lb. 0s o oo o G O Im provem ent C o llecto r o o » N5 O F actor Food o cn C5 O O O O o Z9 lbOO Collar tor/Ton o to o 63 D ISCUSSIO N O F R E S U L T S Examination of Tab I t s I to IV and Figures 4-21, 40, and 43 indicates that every series of tests showed both en­ richment and recovery of copper silicate from the synthetic mixtures. This enrichment and recovery is due to the p re s ­ ence of the collectors tested; blank runs without collectors gave very little enrichment or recovery of copper silicate. Other investigators have suggested that carryover of reagents from one run to the next may influence subsequent recover­ ies. In these experiments, care was taken to flush the flo­ tation cell between runs so that residual material and froth from the previous run were removed. It is believed that the small amount of froth produced in the blank run represents a gravity assistance of flotation which will be discussed in greater detail in connection with the tests upon chrysocolla and commercial Copper Queen ore. This recognition of grav­ ity assistance opens an entirely new field of flotation investi­ gation. To facilitate discussion, referred to, where individual collectors are the following shortened designations are employed; 64 nitroso-4-hexyl resorcinol, NAR-6; nitroso-4-octyl resorcinol, NAR-8; nitroso-4-decyl resorcinol, NAR-10. The nitroso- monohexyl ether of resorcinol will be called NRE-6; The nitroso-monooctyl ether, ether, NRE-10. NRE-8; and the nitroso-monodecyl The collectors as a class will be refe rre d to as nitroso alkyl resorcinols (NAR) and nitroso resorcinol ethers (NRE). Percent Recovery As a Function of pH At Varying Collector Concentrations Nitroso Alkyl Resorcinols: (Tables 1, 41, 43). III, Several general trends are indicated. Figures 4-9, Maximum recoveries are found at pH twelve for the lowest collector concentration, 0. 05 pounds per ton of feed. lector concentrations, the optimum pH value was between six and eight for the NAR-6 collector. collectors, For higher col­ For NAR-8 and NAR-10 optimum pH was between eight and twelve with one exception (Table I and Figure 8). This exception was noted at 0. 1 pound per ton for the NAR-8 collector when operating on 30-120 mesh feed; maximum recovery for this series was obtained between pH four and six. Size of Feed: The NAR-6 collector gave higher recov­ eries when floating minus 120 mesh material than when floating 30-120 mesh feed. NAR-8 collector gave higher max­ imum and lower minimum recoveries for minus 120 mesh feed. NAR-10 collector gave higher recoveries for 30-120 mesh feed. Nitroso Resorcinol Ethers: 40, 42). (Tables II, IV, Figures 4-0, The best pH conditions for varying collector con­ centrations are values between six and eight. With the NRE- 10 collector 0. 2 pounds per ton appears excessive. • The p e r­ cent recovery curves for both minus 120 mesh and 30-120 mesh feed give lower percent recovery at pH values four, six and eight; maximum recoveries develop at pH twelve. The percentage recoveries here recorded using 0. 1 pound of collector per ton of feed for both minus 120 and 30-120 mesh m aterial are much greater than those recorded in flotation literature [37, 88, 107] for single cell flotations. At 0.2 pound of collector per ton the recoveries are satisfactory above pH six for the NRE— 6 and NRE-8 collectors; the NRE10 reagent gave the best results at very high pH values. Size of Feed: The co arser feed gives higher maximum recoveries for each collector and concentration. The NRE-8 and NRE-10 collectors at 0.05 and 0.1 pounds per ton pro­ vide an exception; these concentrations of reagent, finer feed, gave high recoveries. using 66 Percent Recovery as a Function of Collector Concentration at Varying pH Nitroso Alkyl Resorcinols: 12, 16-18, 41, 43), (Tables I, III, Figures 10- The majority of the curves indicate either maxima at 0, 1 pound of nitroso alkyl resorcinol col­ lector per ton of feed, or increasing curves with no demon­ strable maximum but with fair to excellent recoveries at the highest collector concentration, feed. 0, 2 pound per ton of The data indicate that there will be a maximum r e c ­ overy for each collector and pH as the concentration of the collector increases. The recovery in several cases asymp­ totically approaches 95-100%. Chain Length: sidered, When the influence of chain length is con­ it is seen that for the minus 120 mesh feed an in­ crease in chain length indicates a maximum recovery at/ eight carbon atoms for pH values of four, At pH twelve, six, and eight. the NAR-10 gives the maximum recovery. With 30-120 mesh fed, increased chain length at any pH in­ creases the recovery. Size of Feed: 'th ere is little choice between the recov­ ery curves for NAR — 6 using minus 120 mesh and 30 — 120 mesh feed. At eight carbon atoms there appears tosbe better re c ­ overy using the coarser feed and at ten carbon atoms superr o r n u p r i n « i n c r the* coarser feed are readily apparent. Nitroso Resorcinol Ethers: 13-15, 19-21, 40, 42). per ton. (Tables II, IV, Figures Maxima again appear at 0. 1 pound Minimum recoveries are recorded using NRE-6 and NRE-8 collectors at this concentration for minus 120 mesh feed. NRE-8 and NRE-10 collectors indicate gener­ ally increasing recoveries with increasing collector concen­ trations when operating on 30-120 mesh feed. Chain Length: Increased chain length is definitely ben­ eficial at all pH values for minus 120 mesh feed. erating on 30-120 mesh feed, When op­ increased chain length decrea­ ses recoveries for pH four and six; At pH eight and twelve increased chain length has little influence; all recoveries are very good. Size of Feed: higher recoveries. The NRE-6 gives with the coarser feed With the NRE-8 there is little variance of recovery except at pH eight where the finer feed is used. The NRE-10 at the higher pH values of eight and twelve gives better recoveries from the coarse feed; at pH values of four and six better recoveries are obtained with the finer feed. 68 Enrichment as a Function of pH at Varying Collector Concentrations Nitroso Alkyl Resorcinols; 41, 43). (Tables I, III, Figures 4-9, It is shown that variations in pH provide a maxi­ mum enrichment at pH six or eight for both fine and coarse feed. At 0. 05 pound per ton of feed, appears for the NAR-10 collector. an ill-defined maximum The NAR-6 and NAR-8 at this concentration provide inconclusive evidence of maxi­ mum or minimum enrichments when using minus 120 mesh feed. C o arser feed using NAR-6 indicates a maximum at pH twelve; the NAR-8 gives a minimum at pH six. Chain Length : Increased chain length has little influence on enrichment for minus feed, 120 mesh feed. For 30-120 mesh the NAR-8 gives better enrichments than the NAR-6 or NAR-10. Size of feed: Enrichments secured are fair to apoor for the minus 120 mesh feed. For 30-^120 mesh m aterial, very excellent enrichments are with few exceptions secured at all pH values. Nitroso resorcinol ethers: (Tables II, IV, 40, 42). Figures 4-9, Where maximum enrichments are found these occur­ red mostly at pH values of six, infrequently at pH eight. At 69 0. 05 pound of collector per ton fed with the co a rse r feed, enrichment maxima occurred at pH twelve. The presence of maximum enrichments indicates an optimum pH value for both the nitroso alkyl resorcinols and the nitroso resorcinol ethers. Chain Length: Increased chain length shows for the finer feed increased enrichments at the middle collector con­ centration, O.'l pound per ton. tle effect was noted. At other concentrations, For the co a rse r feed, enrichment were noted at all concentrations. li t­ increases in The most spec­ tacular increase appeared in the change from NRE-8 to NRE10 (Table II, Figures Size of Feed: 7-9). The nitroso resorcinol ethers gave with the c o a rs e r feed concentrates comparatively much rich er than did the finer feed. Enrichment as a Function of Collector Concentration at Varying pH Values Nitroso Alkyl Resorcinols: 16-18, 41, 43). Using minus (Tables I, III, 120 mesh feed, Figures 10-12, change in col­ lector concentration influences enrichments but little. aThere a re slight increases with increased concentrations for the three lower pH values of four, six, and eight; at pH twelve 70 the enrichment decreases slightly or remains constant with increased concentration. Using the coarser feed, definite maximum enrichments appear at 0. 1 pound collector per ton for NAR-6 and NAR-8; the effect of increased concentration of NAR-10 is inconclusive. • Chain Length: There is little influence on enrichment for minus 120 mesh feed. Using 30-120 mesh feed, an en­ richment maximum appears at eight carbon atoms length. Size of Feed: from two to five, Enrichments are poor to fair, ranging for the nitroso alkyl resorcinols operating on the finer fe.ed. The co arser feed with these reagents gives excellent enrichments, eight to twenty»five times feed concentration. Nitroso Resorcinol E th ers: (Tables II, IV, 15, 19-21, 40, 42). Figures al3 - Variation in collector concentration for \ the co a rse r feeds indicates maximum enrichments at 0. 1 pound collector per ton feed except at pH twelve and four for the NRE-8 and at pH four for the NRE-10. At the lat­ te r conditions the maximum enrichment is sit 0. 05 pound per ton for NRE-8, pH twelve; at 0. 2 pound per ton for NRE-8, pH four and NRE-10, pH four. For the finer feeds, the NRE-6 and NRE-8 indicate little change in enrichment with 71 concentration at any pH. The NRE-10 indicates a definite enrichment maximum at 0. 1 pound collector per ton. Chain Length: Increased chain length shows little effect at 0. 2 and 0.05 pound per ton; at 0. 1 pound per ton, crease in enrichment is noted using finer feed. er feed, an in­ For c o a rs­ increases in enrichment are given at all concentra­ tions. Size of Feed: Enrichments for the finer feed ranged from 1. 75 to 6. 2 times the feed concentration. The coarser feed gave enrichments ranging from 6. 1 to 19. 8. Comparison of Corresponding Nitroso Alkyl Resorcinols and Nitroso Resorcinol Ethers Variations with Collector Concentration: (Tables I-IV, Figures 10-21, 40-43). In comparing variation of percent recovery and enrichment with collector concentration, it is noted that enrichments are generally somewhat superior for the nitroso resorcinol ethers as compared with the nitroso alkyl resorcinols of the same chain length, on minus 120 mesh feed. when operating For the coarser feed, nitroso a l ­ kyl reso rcin o ls are much superior to the nitroso resorcinol ethers ain enrichment for the two lower homologues, while for the decyl compounds the higher pH values give the ethers 72 a clear superiority. The two lower pH values indicate approx­ imate equality of enrichment. Variations with Chain Length: Percent recoveries are better for the finer feed when NRE-8 and NRE-10 are used. Of the hexyl compounds, at pH twelve the nitroso alkyl r e s o r ­ cinol gives higher percent recoveries but the two classes are about even in percent recovery at the other pH values. the c o a rse r feed, NAR-6. For NRE-6 gives higher recoveries than the The NAR-8 and 10 give better recoveries excepting at pH twelve, where the effectiveness of the two classes of \ compound is about equal. Variations With pH: The finer feed shows slightly higher enrichments for the corresponding nitroso resorcinol ethers as compared with the nitroso alkyl resorcinols.- The nitroso a l­ kyl resorcinols give consistently higher enrichments with coarse feed excepting the NRE-10, which at 0.1 and 0. 2 pound collector per ton are much superior. Variations with Size of Feed: Percent recoveries for the finer feed indicate a great superiority at the two lower concent trations of collector when using nitroso resorcinol ethers, and a moderate superiority for the ethers at 0. 2 pound collector per ton of feed. The NAR-10 using the co arser feed is super­ ior to the NRE-10 at all concentrations. The NRE-8 is a better collector in te rm s of percent recovery when the higher concentration is used; when either 0. 1 or 0.05 pound collector per ton is used the octyl compounds are roughly equal. The NRE-6 gives a better percent recovery at the two higher concentrations; at 0.05 pound per ton, the NRE-6 and NAR-6 are equally effective. In further consideration of the effect of the several operating variables, length, size of feed, pH, collector concentration, and type of collector used, chain the data presented in Tables I-IV were recalculated to yield athe d e r­ ived factors shown in tables V and VI. After consideration of the dependent resu lts the pounds of copper actually recov­ ered in the froth per pound of collector used were calculated for each run. There was also computed the product of the enrichment and the percent recovery which combines for ex­ amination the separating efficiency and the selectivity into one factor. This product factor is the Improvement Factor; it is dimensionless. Examination of these factors set forth in Tables V and VI indicate certain trends. Variation of the pounds copper collected per pound of collector with pH is not discussed. 74 Anything said for the percent recovery will be equally applicable to the pounds copper recovered per pound of collector since the variation of a factor derived from the percent recovery with pH is proportional at constant concentration to the percent recovery. When the variation of pounds copper recovered per pound of col­ lector with collector concentration is examined, it is seen that significant differences appear in contrast to the variations of percent recovery with collector concentration. Pounds Copper Recovered P e r Pound Collector as a Function of Collector Concentration at Various pH Values Nitroso Resorcinol Ethers: 40, 42). (Table VI, Figures 31-33, When minus 120 mesh m aterial is fed, 37-39, the pounds cop­ per per pound of collector is a maximum at 0.05 pounds collec­ tor per ton of feed. The maximum recovery from 30-120 mesh m aterial as pounds copper per pound of collector is found at 0. 05 or at 0. 1 pound collector per ton feed. There is an indi­ cation that for the higher pH values the maximum pounds of cop­ per recovered per pound of collector shifts toward the higher collector concentrations (Figures 40, 42) with increase in chain length. This is entirely consistent with the fact that the weight of the collector per unit area is greater for the same number of molecules. D E R IV E D D A T A ON CHRYSOCOLLA Minus 120 mesh feed 0.2 #/Ton Coll. 0. 1 #/ To n Coll. 0.05 #/T Fig. 22 Fig. 23 Fig. Impr. Impr. # Cu per # Cu per # Cu per Fact. Fact. # Coll. # Coll. # Coll. Nitroso hexyl resorcinol ether NRE-6 (Figs. 31, 37) pH 12 353 168 236 88 1510 pH 8 313 176 320 119 1294 pH 6 279 225 339 151 1390 pH 4 287 234 550 224 1236 434 424 178 378 760 874 143 354 1652 1715 453 478 650 338 1735 205 153 322 96 522 Nitroso decyl resorcinol ether NRE-10 (Figs. 33, 39) pH 12 418 161 800 215 1023 Nitroso octyl resorcinol ether NRE-8 (Figs. 32, 38) pH 12 pH 8 pH 7 pH 6 pH 5 pH 4 pH 3 pH 8 295 263 837 561 870 pH 6 252 188 834 433 403 pH 4 109 41 500 260 716 75 ATION BY NITROSO RESORCINOL ETHERS 1. r. t. 0.2 #/Ton Coll. Fig. 25 Impr. # Cu per Fact. # Coll. 30-120 mesh feed 0. 1 # / Ton Coll. Fig. 26 Impr. # Cu per Fact. # Coll. : 0. 05 # /Ton Coll. : Fig. 27 : # Cu per Impr. : # Coll. Fact. PH ■ . ' * ■f 3 319 426 865 966 1110 575 12 I 403 795 891 930 831 368 8 1 416 766 880 1320 706 341 6 2 169 1375 830 771 626 243 4 ) 450 352 996 594 491 392 ! 505 1120 795 576 610 680 802 317 118 760 809 340 331 678 525 456 509 568 421 198 847 837 412 387 12 8 7 6 5 4 3 1 456 1100 868 1430 676 403 12 - 335 845 G75 1467 686 383 8 229 578 445 640 646 350 6 212 521 204 151 616 223 4 ) > D E R I V E D D A T A ON C HR Y SOC OLL • 0.2 # /Ton Coll. Fig. 22 # Cu per Impr. # Coll. Fact. Nitroso 4-hexyl r e s o r c ­ inol NAR-6 (Figs. 28, 34) pH 12 Minus 120 mesh feed 0. 1 #/Ton Coll. 0. 05 Fig. 23 # Cu p er Impr. # Cu # Coll. Fact. # Cc 255 75 280 58 166 pH 8 316 275 750 302 69 pH 6 297 232 340 71 63 pH 4 167 74 295 63 59 439 347 800 344 136 pH 8 398 342 518 209 55 pH 6 284 253 492 182 55 pH 4 292 244 468 171 57 446 450 900 400 121 pH 8 233 216 513 228 78 pH 6 189 175 431 218 58 pH 4 232 215 294 135 68 Nitroso 4-octyl r e s o r c ­ inol NAR-8 (Figs. 29, 35) pH 12 Nitroso 4-decyl r e s o r c ­ inol NAR-10 (Figs. 30, 36) pH 12 76 [ }N BY NITROSO ALKYL RESORCINOLS : 0.2 #/Ton Coll. : Fig. 25 Impr. : # Cu per Fact. : # Coll. 30-120 mesh feed 0. 1 #/Ton Coll. Fig. 26 # Cu per Impr. # Coll. Fact. 0. 05 # / Ton Coll. Fig. 27 # Cu per Impr. # Coll. Fact. pH 247 533 463 793 950 707 12 272 688 446 799 954 634 8 204 421 565 1548 539 320 6 241 625 382 824 585 271 4 331 1525 616 1346 1 138 1351 12 352 1633 531 870 773 440 8 318 1120 703 1826 643 284 6 337 1232 734 1543 1122 844 4 446 740 870 897 1768 1094 12 446 1010 858 859 1790 1625 8 411 1200 857 1268 1 281 702 6 397 745 795 942 691 307 4 77 Size of Feed: feed, In the case of the NRE-6 and minus 120 mesh pounds of copper per pound of collector is lower except at pH twelve. The NRE-8 gives more pounds of copper per pound of collector for the finer feed, Nitroso Alkyl Resorcinols; 41, 43). likewise the NRE-10. (Table V, Figures 28-30, 34-36, The pounds of copper recovered per pound of nitroso alkyl resorcinol collector decreases with increasing collector con­ centration above 0.05 pound per ton. Chain Length: Increased chain length improves yield slight­ ly from hexyl to octyl for the finer feed; the decyl yields are about the same as those for the octyl. For the coarser feed, in­ creased chain length increases yield on a rough average of 200 pounds of copper per pound of collector for each increase of two carbon atoms in the substituent chain. Size of Feed: The bulk of the data indicate more pounds of copper recovered per pound of collector for the coarse feed than for the fine feed. Improvement Factor as a Function of Collector Concentration at Varying pH Values In considering this derived variable, it is arb itra rily a s ­ sumed that an Improvement Factor of 400 would be a satisfac­ tory flotation criterion. This corresponds to a 40% recovery with 78 at least tenfold enrichment or a fourfold enrichment with 100% recovery. These lower limits represent the best practice for multiple flotation cells in series-parallel installations [88, 106], The data here reported were secured in a single pass through a single cell. It should also be noted that the standards cited exemplify general practise in flotation of copper ores and do not apply spec­ ifically to the flotation of . copper silicate. No commercial flota­ tion of copper silicate is recorded in the literature. The four references already cited [7, 33, 34, 85] represent the known meth­ ods for flotation of copper silicate as chrysocolla. Nitroso Resorcinol E thers: With this standard in mind, the data reported in Table V and Figures 31-33, 37-39, 40, and 42 indicate that using the finer feed satisfactory Improvement Factors are obtained. The NRE-6 gave Improvement Factors varying only slightly with collector concentration, none exceeding 250. The NRE-8 gave maxima of 478 at 0.05 pound per ton and pH six, and of 545 at 0..2 pound per ton and pH six. ton the improvement factor was 354. At 0. 1 pound per The NRE-10 gave maximum values of Improvement Factor at 0. 1 pound per ton with variation in concentrations. The satisfactory values at 0. 1 pound per ton were 561 at pH eight and 433 at pH six. 79 Using the c o arser feed. Improvement Factors are generally satisfactory on the standard set forth. various pH values, For the NRE-8 at the maximum Improvement F a c to rs are found at 0. 2 pound per ton except at pH 6 where the maximum is 680 at 0. 1 pound per ton. The highest Improvement F actors are 996 at pH twelve and 1, 120 at pH four, both at 0. 2 pound per ton. Only the Improvement Factor at pH four and 0. 1 pound per ton is unsatisfactory for the NRE-8. Using the c o arser feed, the NRE-6 gives unsatisfactory Improvement Factors at 0.05 pound per ton for pH four, eight and for 0. 2 pound per ton, tors are satisfactory. pH four. Maxima are 930 at pH eight, For the NRE-10, and Otherwise these Fac­ noted at 0. 1 pound per ton with variation in collector concentration.Maxima are pH twelve, six, 1, 320 at pH six, 956 at and 771 at pH four. satisfactory Improvement Factors are found ex- \ cep t at pH four for the two lower collector concentrations studied and at pH six for 0.05 pound per ton collector. Maxima appear at 0. 1 pound per ton with collector1 concentration varying; they a re 1,430 at pH twelve, Chain Length; 1,467 at pH eight, and 630 at pH six. Increase in chain length seems to have the effect of decreasing the maximum Improvement Factors obtained at pH six and increasing the maxima obtained at pH eight and 80 twelve. The highest maxima were recorded for the decyl sub­ stitution. Size of Feed; Using the co arser feed, higher improvement F actors are recorded in every instance except that of NRE-8 at 0. 05 pound per ton, where the finer feed gives higher Factors at pH eight and six. Nitroso Alkyl Resorcinols: 41, 43). Using the finer feed, (Table V, 34-36, entirely unsatisfactory Improve­ ment F acto rs were noted in all cases. for NAR-6, Figures 28-30, These ranged from 58 pH twelve and 0. 1 pound per ton to 450 at pH twelve and 0. 2 pound per ton for NAR-10. Maximum Improvement F ac­ to rs appeared either at 0. 1 o r 0. 2 pounds per ton of collector. Using c o arser feed, unsatisfactory Improvement Factors appear only for 0.05 pound per ton and pH four or six. collector concentration varying, With maxima appear always at 0. 1 pound p er ton for pH four and six. For pH eight and twelve, the / maximum Improvement F actor appears at 0. 2 pound per ton for NAR-6; at 0. 1 pound per ton for NAR-8; and at 0. 05 pound per ton for NAR-10. lengths, It is noteworthy that for the two shorter chain the maximum Improvement Factor appears at pH six and 0. 1 pound per ton, while for the decyl compound this pair of con­ ditions produces a much reduced value of 1, 268 while a new max­ im um of 1, 625 appears at pH eight and 0.05 pound per ton. 81 This data suggests a possible relation between increased chain length and collector concentration. Further study is needed to determine just what this relation may be. surfaces, Figures 41 and 43, The three-dimensional give a better visualization of this phenomenon. Chain Length; Increased chain length is of assistance in increasing the Improvement FaPtor; the NAR-8 and NAR-10 col­ lectors gave generally enhanced Improvement Factors over the NAR-6 collector. Size of Feed: The coarse feed gave higher Improvement Factors than did the finer feed* In general, with respect to Improvement Factors, the nitroso alkyl resorcinols operating on coarse feed gave the best results, the nitroso resorcinol ethers on coarse feed gave lower results, while the two classes of compound operating on fine feed gave much less satisfactory results. There is little to choose from between the nitroso resorcinol ethers and the nitroso alkyl resorcinols operating on fine feed. Improvement Factor as a Function of pH With Varying Collector Concentration Nitroso Resorcinol Ethers: 42). (Table VI, When minus 120 mesh material is fed, Figures 22-27, 40, it is noted that maximum Improvement Factors appear for 0.05 pounds collector per ton at pH six for the NRE-6 and NRE-8 collectors. NRE-10 collector, Using the Factor is a maximum at pH twelve; a min­ imum occurs at pH six. Only the Factors for NRE-8 at pH six and eight are above 400; they are 540 and 450; At 0. 1 pound per ton maxima appear at pH eight for NRE-8 and NRE-10. The NRE-6 exhibits a minimum at pH twelve. for NRE-10 are satisfactory, Using 0. 2 pounds per ton, 440 at pH six and 560 at pH eight. maxima appear at pH eight for NRE- 10 and at pH six for NRE-8. at pH four. Only the values The NRE-6 exhibits a minimum Only the value of 480 for NRE-8 and pH six is satisfactory. Using coarser feeds and the nitroso resorcinol ethers, 0.05 pound per ton, at all of the compounds exhibit inflected curves with points of inflection at pH six and eight and maxima at pH twelve. NRE-6, The maxima at pH twelve in descending order are: 570; NRE-8, 490; NRE-10, 400. maxima appear at pH five for NRE-8, eight for NRE-10. pH six for NRE-6, and pH Only the values at pH four and three for NRE-8 and NRE-10 are unsatisfactory. 1,470 for NRE-10. At 0. 1 pound per ton, Highest maximum is At 0.2 pound per ton, the values again ex­ hibit points of inflection which occur at pH six for the NRE-6 and NRE-10 reagents. The NRE-8 reagent exhibits a minimum at pH 83 six. Maximum values are: pH four, 1, 370 at pH four, NRE-8; and 1, 100 at pH twelve, NRE-6; 1,120 at NRE-10. All values are satisfactory. Chain length: Using the finer feed, increased chain length produces a maximum Improvement Factor at 9. 2 pound per ton for NRE-8, at 0.1 pound per ton for NRE-10, per ton for NRE-8. Employing coarser feed, and at 0.05 pound maximum Improve­ ment FaPtors are found at 0. 2 pound per ton for NRE-6, pound p e r ton for NRE-10, at 0. 1 and at 0.05 pound per ton for NRE-6. There is a subsidiary maximum at 0.1 pound per ton for NRE-6. Size of feed: The finer feed gave sm aller values of the Improvement Factor than did the coarser feed except for NRE-8 at 0. 05 pound per ton. Nitroso Alkyl Resorcinols: (Table V, Figures 22-27, 41, 43). The re su lts obtained using finer feed at 0. 05 pound per ton indi­ cate minima at pH four for NAR-6 and NAR-8. pH twelve. Maxima are at These maxima and minima are all below 400. 0. 1 pound per ton, At the NAR-6 compound gives a maximum at pH eight while the NAR—8 and NAR-10 give maxima at pH twelve. All values are below 400 except NAR-10 at pH twelve, actly. Using 0.2 pound per ton, 400 ex­ maxima appear at pH eight for NAR-6; at pH eight or twelve for NAR-8; and at pH twelve for 84 NAR-10. Only the value of 450 for pH twelve and NAR-10 is satisfactory. Employing co arser feed, at 0.05 pound per ton the NAR-6 and NAR-8 show maxima at pH twelve, the NAR-10 at pH eight. v Only the minimum value found at pH six for NAR-8 is unsatis­ factory. Maxima are: 1, 620, NAR-10; 1, 350, NAR-8; 700, NAR-6. The concentration of 0. 1 pound per ton gives three very sim ilarly formed curves. All maxima occur at pH six; they are NAR-6; and 1, 270, 1,820, NAR-8; 1,550, unsatisfactory values were found. ma are 700 at pH eight, and 1, 200 at pH six, Chain length; NAR-fO. At 0. 2 pound per ton, NAR-6; 1, 630 at pH eight, No maxi­ NAR-8; NAR-10. Increased chain length on fine feed gave in­ creasing maximum Improvement Factors for 0. 2 and 0.1 pound per ton; at 0. 05 pound per ton, ment Factor at NAR-8. there was a maximum Improve­ On coarse feed, increasing chain length gave maximum Improvement F actors at 0. 2 and 0.1 pounds per ton for NAR-8, and at 0. 05 pound per ton for NAR-10. Increased chain length first increases, maximum at pH six and 0.1 pound per ton. then decreases, the The maximum at 0. 05 pound per ton shifts from pH twelve to pH eight at ten carbon atoms. A maximum at 0. 2 pound per ton and pH eight 85 f irs t in creases, then d ecreases and shifts to pH six at ten carbon atoms length (Figure 43). Size of Feed: The finer feed gave slightly higher Improve­ ment F actors with nitroso resorcinol ethers than with nitroso alkyl reso rcin o ls. The re v e rse of this was true for co arser feeds. In testing the flotation reagents at increased concentrations, te sts were made at a copper concentration ten tim es that used in the main se rie s of experiments. These tests (Table VII) indicated that using the NRE-10 reagent, fair enrichments and recoveries were given for the conditions used. Consideration of the complete se rie s of tests run on more dilute ore mixtures in­ dicates that the NRE-10 collector is not the best collector. TABLE VII HIGH CONCENTRATION TESTS P ercent Pounds per Ton Percent EnrichCopper____________Collector_______Recovery_____________________ ment 0. 456 0 .1 35. 0 6. 0 4. 85 0. 456 0. 5 40. 0 6. 0 5. 00 0. 456 0. 5 53. 0 8. 0 5. 20 86 Commercial Copper Queen Ore The s e rie s of te sts made on Copper Queen (Arizona) ore are recorded in Tables VIII and IX. on dry ground, The firs t se rie s , nondeslimed ore of the two sizes used in the te sts on chrysocolla, indicate that the finer grind is required to release and float the copper content of the ore. made with the ore diluted one hundred tim es, undiluted. made Tests were ten tim es, and Silica sand was again u sed as the diluting medium. The experim ents using the c o a rse r ground ore gave data indicating that as the concentration of feed increased, the yield decreased even with doubled collector concentration. This might have been due to insufficient collector concentration; it might also have been due to sliming difficulties. Slimes were noted in large proportions in the froth and in the tailings from both fine and coarse feed runs. twelve, Highly alkaline conditions, pH gave the best re su lts for the coarse se rie s. Two te s ts made on pure deslimed ore at pH gave reco veries of 77. 3% and 56. 8%. tions of 1. 0 pound per ton of feed were used. High collector concentra­ The effect of using 0. 2 pound per ton of the re ssa n ts in addition to pH adjusting reagents was: nide (22.3%, six and eight following dep­ Sodium cya­ 1.07 enrichment); sodium carbonate (24.1%, 1.07 TABLE VIII FLOTATION OF COPPER QUEEN ORE (Reagent, Lb. /Ton pH Collector Feed Size Feed % Cu i-H • o 0. 1 8. 0 -120 0. 1 12. 0 100. 0 6. 53 None 0.015 . 95. 4 8. 18 None -120 0. 015 87. 4 8. 19 None 8. 0 30-120 0. 015 29. 3 10. 20 None 1 12. 0 30-120 0. 015 41. 8 9. 08 None 0. 2 8. 0 30-120 0. 150 .36. 5 4. 10 None 0. 2 12. 0 30-120 1. 500 22. 3 1. 09 None 0. 2 8. 0 30-120 1. 500 29. 0 1. 33 None 1.0 8. 0 -120* 0. 922 56. 8 2. 15 None 1. 0 6. 0 -120*' 0. 922 77. 3 2. 02 None 0. 2 8. 0 -120 1. 500 22. 3 1.07 NaCN 0. 2 8. 0 -120 1. 500 24. 1 1. 07 Na2C 0 3 0. 2 • 0.015 Depressant Lb. / Ton CM -120 Percent E nrich­ Recovery ment CM 6. 0 NRE-10) 8. 0 -120 1. 500 20. 9 1. 11 Na Silicate 0. : 0. 2 8. 0 -120 1. 500 18. 9 1. 10 Soda Lime 0. 2 0.1 0 . O 9 O *Deslimed ore 88 T A B L E IX F L O T A T IO N O F C O P P E R Q U E E N ORE SE R IE S T E S T S ( E a c h o f t h e r u n s l i s t e d b e l o w f o l l o w s t h e o n e l i s t e d a b o v e it w it h o u t r e m o v a l o f p u lp r e m a i n i n g f r o m p r e v i o u s r u n . S ilic a s a n d a d d e d to k e e p s o l i d s c o n t e n t o f p u lp a p p r o x i m a t e l y c o n ­ sta n t) Percent R ecovery E n r ic h - Im p rovem en t m e n t ______ F a c t o r -120* 0 .0 1 5 41. 6 1 1 . 30 470 8 .0 -120* 0 . 015 38. 1 1 0 .0 0 381 0 . 1. 8 .0 -120* 0 . 015 48. 8 12. 10 593 0. 1 8. 0 -1 2 0 * 0 . 015 61. 4 1 0 . 00 614 0. 1 8 .0 -1 20 * 0 .0 1 5 44. 0 1 5 . 00 674 0. 1 0. 1 o •00 Lb. /T o n __ F eed Feed C o l l e c t o r _________S i z e _____ % C U * N o t d e s l i m e d b ut e s p e c i a l l y t r e a t e d to p r e v e n t o v e r g r i n d ­ in g b e lo w - 1 2 0 m e s h . 89 enrichm ent); sodium silicate soda lim e (18. 9%, (20.9%, 1.11 enrichm ent); 1. 10 enrichm ent). and These d e p re ssa n ts did not function well. A s e r ie s run was set up, in which stepwise additions of Copper Queen o re w ere made in the sam e m anner as in a batch Standard run. The f i r s t cell charge was one gram of Copper Queen and ninety-nine g ra m s of silic a been dry ground so as to p a ss At the end of the f ir s t run, sand, both of which had 120 m esh without overgrinding. a second gram of Copper Queen o re and sufficient silic a sand to keep the solid-liquid ratio approxim ately constant were added. then made in the tim e s. sam e way. A second flotation was This procedure was repeated four Table IX re c o rd s the r e s u lts of th is te s t. It is noted that b o th the enrichm ent and p e r c e n t reco v ery re a c h a m ini­ mum at the in c re a se it is second te s t. ir r e g u la r ly . In succeeding step s these two f a c to r s If the Im provem ent F a cto r is calculated seen to in c re a s e steadily from the second step. Each of the en rich m en ts and re c o v e rie s w ere calculated on the amount of new o re fed in that step. T hese r e s u lts , tog ether with the initial s e r ie s of te s ts on the fin er o re diluted ten tim es with silica, suggest that som e­ thing o ther than o rd in ary flotation is taking place which re q u ire s 90 further investigation. It has already been suggested that this is a gravity assistan ce. colla and chalcopyrite, The two m inerals floated were chrysowhose specific gravities were closest (2. 24 and 4. 1) to the apparent specific gravity of the aerated suspension of solids existing in the cell during flotation (approx­ imately 1. 2’). The dark brown color c h aracteristic of chelate form ation of copper complexes was demonstrated, showing that this action was present. The presence of gravity assistan ce might well increase the time available for reaction of the collector molecules with the m ineral p articles by increasing both the time of residence in the upper p art of the liquid zone and the effective concentra­ tion of the desired p articles in this zone. General Summary of Discussion The te s ts on c h ry so co lla-silica sand m ixtures and on Cop­ per Q ueen-silica sand m ixtures here discussed have shown that many interesting problem s rem ain to be solved in the general field of selective flotation. F u rth er work employing the nitroso reso rcin o l eth e rs and the nitroso alkyl reso rcino ls which would m ore clearly indicate the effect of increased chain lengths over those studied would be desirable. The effect of very greatly 91 increased collector concentrations might well be investigated. An exact determination of the ratio of specific mineral surface to molar quantity of collector should be made for the several important copper minerals. It is also suggested that the ef­ fect of wet grinding in the presence of these reagents be studied. 92 CONCLUSIONS The following conclusions are drawn from the data p re­ sented. (1) decyl It is ethers of shown that the nitroso monohexyl, resorcinol and the nitroso-4-hexyl, octyl, and octyl, and decyl resorcinols function as specific collecting agents for chrysocolla in synthetic mixtures of chrysocolla with quartz sand. It is postulated that attachment of the collector is due to chelate ring formation between the nitroso and hydroxyl groups of these compounds with the copper atoms in the copper silicate lattices. (2) of feed, It is shown that pH, and length of the collector concentration, fineness substituent chain affect in varying degrees the flotation of chrysocolla from silica sand. (3) When operating on a commercial copper ore, ing copper sulfides and chrysocolla, contain­ it is shown that this ore may be effectively concentrated by the use of the resorcinol monodecyl ether collector. (4) A technique has been developed which allows higher recoveries when floating commercial ores containing copper silicate. Evidence is presented which indicates that an increase in the apparent specific gravity of a m ineral suspension m a te r­ ially affects the flotation recoveries and enrichments. 94 APPENDIX Synthesis of 4-Alkyl-2-Nitroso Phenols The se rie s of compounds f irs t chosen for synthesis was that indicated above. A literatu re survey indicated that several methods existed for preparing various alkyl phenols. Among these wasathat of Tzukervanik and Tambovtsevna [119] who employed normal and iso-alkyl chlorides in the presence of molal quantities of aluminum chloride. McGreal and Niederl [87] condensed molar mixtures of alcohols and phenols by refluxing at 180° for several hours in presence of 1. 5 moles of finely powdered anhydrous zinc chloride. te rtia ry hexyl, heptyl, These investigators studied and octyl alcohols, matic and methylcyclohexanols. as well as some a ro ­ Their method was reported p re ­ viously by Liebmann [84], Dianin [36] heated for two days one part of methyl hexyl ketone with four p a r ts phenol and three p arts HC1 to obtain 4 -s e c — octyl phenol. Koenigs [80] reacted eqiiimolal mixtures of phenol and com m ercial amylene with a mixture of concentrated sulfuric acid and glacial acetic acid. Gurewitsch [56] derived p— te rtia ry alkyl phenol from phenol and t-am yl chloride in presence of fe rric chloride. Huston and Hsieh [70] condensed te rtia ry alcohols with phenols in inert solvents such as petrole­ um ether by adding half-m olar quantities of aluminum chloride. Huston and Meloy [71] condensed methyl dipropyl carbinols with phenol using the selected octyl alcohol as solvent with ap­ proximately one-third mole of aluminum chloride. Guile, Bailey, Curtis, Huston, and Esterdahl [68] reported the conden­ sation of secondary alcohols with phenol in presence of alumi­ num chloride. Chichibabin [25] alkylated phenol by using 400 gram s of phosphoric acid, density 1. 85-1. 87, per gram mole of alkylated m aterial (phenol) with a slight excess of the chosen alcohol. Half of the acid was added to each component and the two solutions slowly mixed, then warmed and agitated. After 8-10 hours the acid was separated and washed with ether; the product was purified with sodium hydroxide and vacuum distilled. A United States Patent issued to Putnam, Button, and Perkins [104] claimed the alkylation of phenols having the 4-position free by using te rtia ry alkyl halides in presence of a catalyst such as aluminum chloride and a liquefying solvent such as ex­ cess phenol. A sim ilar patent was issued to Perkins [102], Huston [65] claimed in a patent the reactifcm of phenol with monohydric aliphatic . alcohols in presence of sufficient quantities of 96 anhydrous aluminum chloride to effect a condensation. Hedrick [64, 69] prepared the te rtia ry heptyl phenols by condensing the alcohols with phenol in presence of aluminum chloride. Guile [55, 67] prepared the corresponding te rtia ry octyl phenols by a sim ilar method.- The several authors cited n o ted that te rtia ry alcohols or halides were most reactive. It was also recorded that p rim ary alcohols often underwent isom erization with branching during or after the introduction of the alkyl group. In order to obtain norm al-chain substitution, it was necessary to introduce a ketonic group by alkylating with acid chlorides by a method sim ilar to that of Sandulesco and G irard [109], Aluminum chloride or zinc chloride may be used as a catalyst [96]. The alkyl phenyl ketone is then reduced to the R-CH 2 “ grouping by a suitable method such as the Huang-Minlon modi­ fication of the Wolff7 Kishner reduction [63], The use of te rtia ry substituents was dictated by their superior reactivity and known stru ctu re after introduction. The synthesis firs t chosen for the te rtia ry alcohols was that of Huston and Bailey [66] which consisted of adding one mole of aliphatic acid to 3. 3 moles of methyl magnesium bromide in ether solution and refluxing on the water bath. It was also found possible to use the appropriate methyl-alkyl ketone with 97 a corresponding saving in methyl magnesium bromide require­ ments [55, 67], A typical alcohol synthesis from ketones was as follows: In a 5 liter three necked flask with dropping funnel, sealed s tir r e r , ting tube, mercury and reflux condenser bearing a D rierite protec­ was placed 98 grams (4 moles)' Grignard grade mag­ nesium turnings and a few crystals of iodine. The flask was warmed until the iodine began to vaporize and then cooled. A solution of 548 grams (4 moles) n-butyl bromide in 500 cubic centim eters anhydrous ether was made. Thirty cubic centimet­ ers of this mixture was added directly to the magnesium. the reaction started, After 200 cubic centim eters of anhydrous ether was added followed by 475 cubic centimeters of the butyl brom­ ide solution, added dropwise over four hours* time. The r e ­ mainder of the butyl bromide solution was diluted with 300 cub­ ic centim eters of ether and added dropwise. When reaction was complete as possible (a few chips of magnesium still remained) 232 grams (4 moles) of redistilled and dried acetone were ad­ ded in 300 cubic centim eters of ether. a dirty gray color. Reaction mixture turned The reaction was hydrolyzed in a 12 liter flask on 2 kilograms of ice, 300 grams ammonium chloride and 600 cubic centim eters of water. The ether layer was dried 98 over sodium sulfate and the main portion of ether distilled off at atmospheric p ressure. After removal of the remaining sol­ vent and bromide by distillation at the water pump, high vacuum distillation with a short fractionating column gave 173 grams (1.5 moles) water white product boiling at 59-60.5° (25 mm.) Preparation of the 2-methyl-2-hydroxy-nonane was carried out by the method of Huston and Bailey [66], of theory, Yields were 50% of pale yellow m aterial with a faint pleasant odor. Boiling point, 47° at 7 mm. \ The alkyl phenols were prepared by a modification of the method of Bennet and Reynolds [16], who state that alcohols react with hydrogen bromide and the resulting bromide condenses with phenol immediately when the alcohol and hydrogen bromide are made toareact in phenolic solutions. Tertiary butyl brom­ ide reacts quantitatively with phenol in a few minutes at 90° but the reaction is slow at 50°. A little zinc chloride greatly accelerates the reaction with te rtiary and secondary bromides. The method as finally adapted consisted in causing 48% hydro­ gen bromide to react with alcohol in a large excess of phenol at tem peratures well above the melting point of phenol (41 The proportions used in the first experiment are as follbws: A 500 m illiliter three necked flask provided with a mercury C .) 99 sealed s t i r r e r , an offset reflux condenser, placed on a water bath. bath tem perature. added and melted. bromi-de. was The empty flask wassheated to water Ninety-four grams (1 mole) of phenol was There was then added 1.36 gram (0.01 mole) freshly fused zinc chloride, methyl-2-nonanol, and a stopper, 17. 2 gram s (0. 1 mole) 2- and 16.85 gram s (0. 1 mole) 48% hydrogen The la tter was added in three portions with only a short interval between additions. The m aterial was refluxed at the highest heat of the water bath forafive hours; it became deep red almost immediately. One hundred cubic centim eters of water were then added as diluent. There appeared a brown upper layer of about 100 cubic cen tim eters which was treated with ca. 400 cubic centim eters of hot water. the upper layer before shaking. The water was Upon shaking a cafe-au-lait emulsion formed which gradually broke giving a brown lower layer. This was separated after standing overnight and further treated with 200 cubic centim eters of hot water added in four portions. There resulted a cafe-au-lait emulsion which the second addition resolved into a brown upper layer and a cloudy white lower layer. with The upper layer was washed twice more 200 cubic centim eter portions of hot water added as above. The final product was 25 cubic centim eters (23. 3 grams) of a 100 viscous, dark brown liquid, insoluble in water. soluble in acetone and ether and The yield was 99% of theoretical. This m aterial was later subjected to vacuum distillation using a plain distillation head; product was taken as boiling at 171° at 7 mm. theory, pressure. The yield of distillate was about 70% of some m aterial being coked in the distillation flask. \ There was also prepared by this method 2-methyl-2-phydroxy-phenyl octane using dimethyl n-hexyl carbinol made available through the Chemistry Department, College. Michigan State Distillation of this m aterial resulted in a dark brown polymerized residue and little distillation product. Alkylation using 2-methyl-2-hexanol prepared by Guile’s method proceeded s more satisfactorily. 124-128° C. Yield of distilled product boiling between at 5 mm. was 40.1 grams (20.5% of theory). Twenty-five grams of additional higher-boiling material, ing at 128-140°, were collected, boil­ and some m aterial was lost by polymerization. At this stage, the nitrosation procedure of Cronheim [28] was applied to the para-t-decyl phenol prepared as above. In a glass stoppered bottle was placed 23.4 grams (0. 1 mole) p a ra -te rtia ry decyl phenol with 30 cubic centimeters of glacial acetic acid. This was diluted with 50 cubic centimeters of water and sodium acetate added to bring the pH to 4. 2 (meas­ ured by a Beckman meter). The buffered emulsion was now added to a solution of 17. 3 grams (0. 25 mole) sodium nitrite and 12.5 grams (0.05 mole) cupric sulfate in 500 cubic cen­ tim eters of water. The water solution was deep green but after the alkyl phenol had been added and the mixture let stand for twelve hours, in the aqueous layer. there was a reddish brown dispersion Attempts to purify this m aterial after it had stood for five days were indifferently successful. ta rry upper layer, presumably containing the desired n itro so -p ara-tertiary decyl phenol, The orthp- was hard to filter. By tria l and e rro r it was found that this m aterial was soluble in diethyl ether and slightly soluble in petroleum ether. Ethyl alcohol or ethyl alcohol-chloroform m ixtures also dissolved the m aterial somewhat. with ether, Extraction of the reaction mixture evaporation of the ether, and attempted crystalliz­ ation from 2:3 ethyl alcohol-chloroform was unsuccessful. The best product which could be secured was a ta rry am or­ phous clump of m aterial. Cronheim’s article postulated that this was the dimolecular copper chelate or inner complex which should appear as a powdery crystalline m aterial. It was considered possible that impure alkyl phenol (the decyl phenol used had not been distilled) was responsible, so further 102 runs were planned. In all, there were nitrosated by this p ro ­ cedure the following compounds: octyl phenol from Rohm and Haas company (stated to be a highly branched material); vac- uun distilled 2-methyl-2-p-hydroxyphenyl nonane; *Oronite’ p a ra -te rtia ry -te tra d e c y l phenol (‘Oronite* alkyl phenol No. Oronite Chemical Company); 14, p-cresol; p -te rtia ry amyl phneol (Eastman Kodak Company); 2-methyl-2-p-hydroxyphenyl heptane; 2-methyl-2-p-hydroxyphenyl hexane, prepared in these labora­ tories; p -te rtia ry butyl phenol. A ttem p ts’to separate and purify the copper complexes from these reaction m ixtures as the free ortho-nitroso- paraalkyl phenols were very indifferently successful. In every case the purification procedure of Cronheim gave deep-colored solutions and the lightsyellow or green petroleum ether solu= tions c h aracteristic of the two isom eric forms of the orthonitroso phenols were obtained. line m aterial was ever secured. No visible quantity of cry stal­ The ta rrin e s s of the phenolic layers in the reaction m ixtures prevented isolation of the des­ ired compounds. This line of work was finally abandoned. Communication with Cronheim [29] revealed the fact that he had never obtained more than tra c e s of the compounds reported. Indirect methods of nitrosation were considered. The oxidation of ortho-am inorpara alkyl phenols was found to be 103 one possible route. The amino alkyl phenols could be prepared by alkylating blocked ortho aminophenol or by first alkylating phenol, nitrating the alkylate and reducing this m aterial. Ortho aminophenol was first alkylated by the following procedure: in a S00 cubic centim eter three necked flask p ro ­ vided with a m ercury sexled s t i r r e r and reflux condenser were placed 3 6. 1 gram s (0. 25 mole) 2-methyl-2-nonanol and 51. 6 gram s (0.25 mole) 48% hydrogen bromide. This was heated on the water bath for fifteen minutes and 27. 3 grams (0. 25 mole) ortho aminophenol added in small portions over five min­ utes, at the end of which 3. 4 gram s (0. 025 mole) zinccchloride were added. turned black. hours. White fumes evolved and the ortho aminophenol The mixture was heated with stirrin g for four Upon cooling the mixture separated into a clear upper layer and an opaque brown lower layer which contained a gray crystalline m ass, easily filterable on a Buchner funnel. was taken as unreacted ortho aminophenol. This The mother liquor was treated to recover the p a ra -te rtia ry decyl aminophenol. Recovery was unsuccessful. The nitration route was then entered by the following pro­ cedure, a modification of Galloway’s method [46], lite r three-necked flask withsthermometer, In a two glass s tir r e r and dropping funnel, the whole in a good ice bath, 114 gram s 70% nitric acid (sp. gr. 1,41) and 115 cubic centi-* m eters of water which were cooled to 4° C. nonyl phenol (264,4 gram s, were placed P a ra -te rtia ry 1, 15 mole) in 560 cubic centimet­ e rs of reagent grade benzene were then added through the dropping funnel over two hours. tained at 6-9°, The tem perature was main­ Reaction was stirre d one more hour in the cold and removed to a separatory funnel. The upper, reddish brown benzene layer was washed once with 250 cubic centimet­ e rs warm water. The benzene was then distilled off at atmos­ pheric p re ssu re . The product was vacuum distilled using a plain stillhead. A principal fraction of 218 gram s boiling at 165-8° (6 mm. ) was taken; i t was orange. The flask tem per­ ature then rose and a second fraction of 35. 5 gram s boiling at 168-185° (6 mm.) was taken which was slightly darker in col­ or. About 40 grams of very dark, remained in the flask. resinous, hard m aterial The combined frac tions were 84% aof theoretical yield. Octyl phenol from Rohm and Haas and nonyl phenol from Sharpies Chemical Company, chains, both having highly branched side were nitrated by the above procedure. The form er gave 262 grams of product boiling at 155-9° at 7 mm. with sm all amounts of low and high boiling m aterial. together Yield 105 was 87. 5% of theory. The latter gave 140 grams of product boiling at 165-172° and 99 gram s of product boiling 172-6° at 7 mm. yellow, phenols, This was 89% of theory. slightly viscous liquids, Both these products were more mobile than the original with a nitrobenzene or petroleum odor. Reduction of the 2 -n itro -4 -te rtia ry nonyl phenols was accomplished by several methods. There were available the methods of Bar anger [9] who used sodium hydro sulfite in alk­ aline solution; Anish [6] also used this method specifically to prepare 2-am ino-4-alkyl phenols from the 2-nitro compounds. Sidgwick and Callow [114] proposed the action of 4. 5 moles of sodium bisulfite in 25% aqueous solution with addition of zinc dust to the boiling solution. Buck and Ide proposed the use of tin and hydrochloric acid [23], Sodium hydrosulfide in alco­ holic or aqueous suspension [113]; ammonia and hydrogen sul­ fide on alcoholic solutions of the nitro compound [113]; ferrous sulfate added to an aqueous ammoniacal solution of the nitro compound [113]; zinc metal in presence of calcium chloride sol­ ution [21][20]; iron filings in presence of calcium chloride sol­ ution [124]; and reduction with platinum or platinum black [1] have been proposed. Of these methods, f ir s t attempted, reduction with alkaline hydrosulfite was A mixture of 40 gram s (1 mole) sodium hyd­ roxide in 720 gram s water was heated to 80-85° with 26. 5 gram s (0. 1 mole) 2 -n itro -4 -te rtia ry nonly phenol, ution resulting. a clear sol­ To this was added 75 gram s (0. 35 mole) sod­ ium hydrosulfite. Upon stirrin g a black floe appeared. Upon 48 hours* standing the floe w as orange with a black overluster. F iltratio n after 72 hours gave an amorphous flocculent cake whose surface was black, on exposure to a ir. the interior orange-red turning black The cake was washed by refiltering the light orange filtrate with addition of 500 gram s water contain­ ing tra c e s of hydrosulfite. placed in a desiccator. The cake- was then sucked dry and The yield was not m easured but was close to quantitative. A modification of Buck and Ide*s method for m -chlorobenzaldehyde [23] was applied to 2-n itro -4 -nonyl phenol. ice-cooled bwaker, In an 67.5 gram s (0.3 mole) stannous chloride dihydrate and 100 cubic centim eters (1. 2 mole) concentrated hydrochloric acid were cooled to less than 5°. To this was added 26. 5 gram s (0. 1 mole) 2-nitro-4-nonyl phenol and the mixture s tirre d vigorously. Fumes of hydrogen chloride soon appeared and the reaction ran smoothly. Unfortunately the m aj­ or p art of this mixture was lost in purification. 107 A tria l of Sidgwick and Callow’s method [114] using 26.5 grams (0. 1 mole) of 2-n itro -4 -nonyl phenol with 49. 26 grams (0. 45 mole) of sodium bisulfite in 147 grams water was made. This mixture was heated to 85° and portions of zinc dust ad­ ded. Boiling wa s induced but there was litile apparent reac­ tion with the ta rry phenol layer, and the unreacted zinc dust formed large lumps with the phenol. Purification was found difficult and was abandoned in favor of the first two methods. Reduction with iron filings was next essayed. 70. 5 grams of 2--nitro-4-nonyl phenol (0. 265 mole) was added to 125 cubic centimeters of 50% ethyl alcohol in a flask provided with a wing s tir r e r . Iron filings (44. 3 grams, 0. 795 mole) were added fol­ lowed by 3. 2 cubic centimeters concentrated hydrochloric acid dissolved in 25 cubic centimeters of 50% alcohol. was refluxed with stirring for three hours. The mixture It was then filtered after cooling on a 12 centimeter Buchner funnel. The filtrate was neutralized with 1. 53 grams sodium hydroxide in 25 cubic centim eters of water, poured through the cake. The cake was washed with small portions of 50% alcohol. These alcohol washings were distilled at atmospheric pressure, resulting sin a yellow distillate which on cooling deposited a dark colored, viscous lower layer. A sim ilar lower layer was derived when the iron cake remaining in the original flask was 108 extracted with ether, resulting in a dark brown solution. This solution when distilled yielded a gummy dark brown material which poured very slowly. The united viscous brown layers were submitted to vacuum distillation at 5 mm pressure. C ar­ bonization of the flask contents and repeated plugging of a dry ice-acetone cooled cold trap resulted. The material retained in the cold trap was strong smelling and evaporated readily on warming to room temperature; it very probably contained some 2-nitroso-4-alkyl phenol. Treatments of portions of the brown muck from the original reaction with petroleum ether accompanied by acidification resulted in a green petrol­ eum ether layer and a colorless water-alcohol lower layer. Treatment of the green petroleum ether layer with aqueous sod­ ium hydroxide gave a yellow water layer and a brown-green petroleum ether layer. This behavior is characteristic of the presence of a metal inner complex of ortho-nitrosophenols. It was noted that the loss in iron filings was almost exactly that required to form the dimolecular ferrous iron complex postulated by Cronheim [28]. Treatment of portions of the original muck by Cronheim’s procedure for releasing the free orthonitrosophenol gave colored organic salt solutions but no crystalline or amorphous free precipitates. 109 Both the irori-hydrochloric acid and the use of alkaline hydrosulfite seemed to show promise. It was planned to oxi­ dize the amino compounds to the nitroso compounds using p e r­ acetic acid according to the methods of D'Ans and Kneip [31] and P rilaschajew [103], P relim inary tria ls of the two methods were made with indications of success. amorphous, Products were still and quantitative resu lts were not obtained. oxidative procedures based on catalysis [123], idation [118], and on C aro’s acid [74, 94, Other electrolytic ox­ 123] were consid­ ered but not used. Synthesis of Resorcinol Monoalkyl Ethers The difficulty of arriving at pure free nitroso derivatives of 4-alkyl phenols led to a reexamination of other hydroxybenzene compounds which might be alkylated. [78], Henrich [61], Here the work of Kietabl Henrich and Eisenach [62], and of Kharman, Ghatyas and Shternov [77] on nitroso resorcinol monoalkyl ethers were consulted. Kietabl proposed the treatm ent of r e s ­ orcinol with sodium ethoxide to form the sodium salt, followed* by reaction with ethyl iodide to form the ethyl ether. Kharman and his coworkers dissolved resorcinol in ethyl alcohol, the alkyl chloride o r iodide, added then dropped in potassium hydroxide 110 solution to effect the condensation. A modified reaction according to Kietabl was run as fol­ lows: in a one lite r flask with s t i r r e r and 40 centimeter West condenser were placed 27.5 gram s (0.25 mole) Eastman K0 dak re se a rc h grade resorcinol, .3:1 ethyl alcohol. There was then added 16.81 grams (0.30 mole) potassium hydroxide, black. dissolved in 400 cubic centim eters whereupon the solution became Sodium iodide (10.9 gram s, 0.06 mole) was added, then after the initial warming due to potassium hydroxide ad­ dition had abated, 44. 59 gram s (0. 30 mole) octyl chloride was added through the condenser. ing was begun. half hours. After five to ten minutes heat­ The reaction was refluxed for four and one- P a rt of the alcohol-water was distilled off with a vertical distilling trap. The reaction mixture was then drowned in two lite rs of water; the residue before drowning wws about 100 cubic centim eters in volume, standing a dark, dark brown color. oily layer separated. Upon The pH of the solution was eight; upon neutralization the oily layer was somewhat aug­ mented. The oil was taken up in a solution of 10 grams (0. 25 mole) sodium hydroxide in one liter of water. solved as a dark re d — brown solution. All the oil d is­ Neutralization with hyd­ rochloric acid regenrated the same dark-red oil. This wa s Ill taken up in ether and treated with Norit A. Filtration yielded a clear red solution which was concentrated by evaporation. / Yield was 6. 8 gram s, 12. 2% of theory. To obtain a better yield, ution concentration. it was decided to raise the sol­ In a flask provided with a s tir r e r , 55 gram s (0.5 mole) of resorcinol were poured into 50 cubic cen­ tim eters butyl alcohol with stirrin g . Sodium hydroxide 20. 2 gram s (0. 505 mole) was suspended in the 150 cubic centimet­ ers of hot butyl alcohol. The suspension was added to the warmed resorcinol solution. Boiling began; after a minute's refluxing 7.49 gram s (0.05 mole) sodium iodide and 110. 55 gram s (0. 75 mole) octyl chloride were added through the condenser. Refluxing was continued (100 cubic centim eters of alcohol were drawn off through the distilling trap) u n til a sample;gave a pH of 5-6 whendrowned in water, indicating no unreacted alkali. T|ie solution on cooling had a red-gold clear and a pasty lower layer of about the from sodium chloride formation. from the liquid. upper layer volume to be expected The paste was separated Treatment of the paste with water gave a dark re d — brown solution with small additional amounts of clear re d — gold organic layer floating on top. The united org­ 112 anic layers were washed with 1, 100 gram s of water to remove dissolved butyl alcohol. The organic layers were then steam distilled to remove unreacted resorcinol and butanol. Remain­ ing organic layer was f irs t washed with portions of 10% sod­ ium hydroxide until no further solution could be detected (vir­ tually all of the m aterial dissolved). The united sodium hyd­ roxide washes were neutralized with hydrochloric acid; care was required to insure complete neutralization of the greasy organic layer which released caustic gradually. The emul­ sion formed on neutralization broke at pH 7. 5 Organic layer was taken as crude monooctyl resorcinol ether. gram s, 44% of theory. Yield 48. 7 Some m aterial was lost in processing the flask residues by the same procedure. A sim ilar reaction using normal hexyl bromide and half the previous quantity of butyl alcohol as solvent yielded 40 gram s (21% of theory) of distilled monohexyl resorcinol ether boiling at 140-150° (5 m m .). cous m aterial, This was light yellow, very v is­ darkening in air. A second portion of the monooctyl resorcinol ether was made according to Kharman, Ghatyas and Shternov. One mole (119 grams) of resorcinol was dissolved in 225 cubic centimet­ e rs of ethyl alcohol in a one liter flask provided with s tir r e r , 113 reflux condenser, and dropping funnel. octyl chloride was added, One mole (150 grams) the mixture heated to refluxing, and one mole (56. 1 grams) potassium hydroxide in 168 gram s of water dropped in slowly over about three hours. Separation was effected by a simple extraction procedure sim ilar to that used above. Water was added to the cooled mixture which was then placed in a separatory funnel. There was added 750 cubic centim eters aof diethyl ether followed by 400 cubic cen tim eters 6N sulfuric acid to insure acid extraction. ious precipitate of potassium sulfate formed. lay ers were removed from the solid. A cop­ The two liquid The acid layer was extracted with a second portion of ether. The combined ether extracts were washed repeatedly with cold water until no r e a c ­ tion was given with fe rric chloride solution. The ethereal layer was extracted repeatedly with 5% sodium hydroxide. The alkaline extract was shaken with ether; the washed alkali was acidified with 50 cubic centim eters 6N hydrochloric acid. Separated oil was shaken out with ether, a Evaporation of ether gave about 25 cubic centim eters of heavy red oil. Distillation of this m aterial through a fifteen centim eter open fractionating column with an adiabatic jacket gave 20 cubic centim eters (14.5 gram s, 7% theory) clear red product boiling 155° (5 mm.) 114 A sim ilar synthesis was used for monodecyl resorcinol ether. Decyl bromide was synthesised by the method given in Organic Syntheses, Coll. Vol. I [75] for normal octyl bromide. The yield of monodecyl reso rcino l ether from 0. 25 mole of reso rc in o l and 0. 3 mole decyl bromide was 10 gram s, 16% of theory, v is­ boiling 162° at 5 mm. It was a red liquid, cous and darkening to purplish red with storage. The compounds monohexyl, finally selected for nitrosation were the monooctyl, p repared as given. and monodecyl ethers of resorcinol, To this ples of 4-hexyl,4-octyl, se rie s was added re se a rc h sam ­ and 4-decyl-resorcinols made able from Sharp and Dohme, Inc. The procedure used was that of Kietabl [19], nitrosation, follows: avail­ A typical as c a rrie d out on the monooctyl ether, was as one gram (0. 00434 mole) of monooctyl ether was dissolved in 2 cc. of alcohol absolute; one cc. of glacial acet­ ic acid (0.0166 mole) was added followed by 0.428 gram sod­ ium n itrite (0.0063 mole) dissolved in one gram i'he acetic acid -1° C. solution of the ether was cooled to and the sodium nitrite drop by drop. of water, less than solution added very slowly, The ch aracteristic brown color indicating f or ­ mation of the nitroso compound appeared immediately. The reaction mixture was allowed to stand for two hours under continual cooling in ice-salt mixture. At the end of this time, the precipitated nitroso compound was removed with a filter stick, hours. ial, pressed dry gently and put in a desiccator for two No further attempt wa s made of purifying the m ater­ which was dissolved in absolute alcohol for immediate use. Yields were as follows: 26. 1%; nitroso monooctyl ether, ether, monohexyl resorcinol ether, 64. 7%; nitroso monodecyl 56. 3%; nitroso-4-hexyl resorcinol, octyl resorcinol, 87. 5%; nitroso-4- 74,0%; and nitroso-4-decyl resorcinol, 116 BIBLIOGRAPHY 1. Adams, Roger, Cohen, F. L. , and Rees, Chem. Soc. £9, 1093-99 (1927). 2. Albert, A. , and Gledhill, 529-33 (1947). 3. Andrews, L. J. and Keefer, 71, 2379-83 (1949). 4. Ibid., 5. Anon., 6. Anish, A. W ., 14, 1947). 7. W. S. , R. Biochem. M. , J. J. J. Am. 41, Am.Chem. Soc. 2379-80. p . Eng. Min. J. 130, 506 (1930). United States Patent 2, 429, 178 (October Ayer, F. , Crowfoot, A. , Herivel, H* T. , and Sorenson, P. M. United States Patent 1,916,196 (July 4, 1933). 8. Baeyer, 9. B aranger, Adolf and Knorr, P ., Bull. 11. Baudisch, $X , Science 92, 12. Baudisch, O ., J. 13. Baudisch, O ., and Heggen, 239-45 (1942). 14. Baudisch, (1912). O ., 15. Baudisch, (1915). O ., and Rothschild, M. E ., Chim. Barthelemy, Bennet, G . 1935, 131. R. Soc. Edouard, 10. 16. O.W ., Eng. Am. Min. 336 Chem. and Karzeff, A., Ber. [4], J. £9, 135, 3037 (1902). 1213-22 (1931). 401 (1934). (1940). Soc. 63, Arch. Nicholas, and Reynolds, 35, S ., F. Biochem. £, Ber. 1164 Ber. M. , 622 (1941). J. 45, 48, 1662 Chem. Soc. 117 17. B essel Gebrudern, German Patent No. Patent No. 39, 369 (1887). 18. Billiet, V ., Verh. Konn. Vlaamsche, Acad. Wetensch. L e tte ren Schoone Kunsten Belgie, Klaase Wetensch. 4, No. 1, 58 pp. (1942). In Chem. Z ertr. 1943, II, 13. 19. Booth, R. 28, 1951) B. 20. Brand, and Mahr, 21. Brand, K. (1928). 22. Bryson, A. and Dwyer, Wales 74, 107-9 (1940). 23. Buck, J. S. and Ide, W. S ., Organic Syntheses Coll. Vol. II p. 130,John Wiley and Sons, New York, 1943. 24. Calvin, M. , and Melchior, 3270-3 (1948). 24a. Kt and Hill, J. J. , and Moderson, Chapman, 1914. G. A ., B rit. German Patent 650, 498 (Feb. J. P rakt. Chem. 131, 119(1931). A. J. , P rakti Chem. J. 120, I C O F. P. , J* P roc. Roy. Soc. N. S. In Chem. A bstr. 3 5 , 450 (1940). N. C. , J. Am.- Chem. Soc. United States Patent 1, 102, 874, E . , ‘ Compt. July 7, Chichibabin, 2G. C hristm an, Ludwig, 1934). United States Patent 1, 960, 526 (May 29, 27. C hristm an, Ludwig, 1936). United States Patent 1, 952, 907 (Mar. J. 28. Cro nheim, G. 29. Cronheim, G. , P rivate communication (1949). 30. Cronheim, G ., and Wink, 31. D’Ans, and Kneip, ■, J. Ind. Eng. Org. Chem. A ., Chem. 1239-42 (1934). C lem m er, J. B ., Rend. 198, 70, 25. 27a. A. W ., 42 (1877); Wm. , Ber. 12, 41, 41-4 (1948). 1-20 (1947). Anal. 48, 27, Chem. 14, 1142-45 (1918). 447 (1942). 118 32. Dean, R. S. and Hershberger, R ., Amer. Inst. Min. Met. Engrs. Technical Publication No. 605 (1935). In A. I. M. M.. E. Transactions 134^ 81 (1939). 33. Dean, R. S. and others* United States Bureau of Mines Report of Investigation No. 3357 (1938). 34. Dean, R. S. and others. United States Bureau of Mines Report of Investigation No. 3419 (1938). 35. DeWitt, C. C. and Batchelder, Soc. 61, 1250 (1939). 3G. Dianin, (1914). 37. Dugann, 38. Emeleus, H, J. and Anderson, J. S ., Modern Aspects of Inorganic Chemistry, D. Van Nostrand Co. , New York, 1947. 39. Ephriam, 40. Ephriam, F ritz, Inorganic Chemistry, Interscience Pub­ lishers, Inc., New York,- 1948 (ed. Thorne, P. C. L«. and Roberts, E. R. ). 41. Erickson, 8, 1942). 42. Everson, C arrie J . , United States Patent 348, 157 (Aug. 24, 1886). 43. Feigl, F ., 44. Feigl, F. 45. F e rris , 1930). 46. Galloway, C. 16, 1940). A. P. , J. In Chem. E. J. , S. E.-, Ber. J. Min. 63, J. 126, Am. Chem. 1310-19 1008 (1928). 1928 (1930). United States Patent 2, 295, 459 (Sept. Ber. _58, and Bondi, L. W. , von., Russ. Phys. Chem. 46, Abstr. 9, 1903 (1914). Eng. F ritz, F. 2294 (1925). A ., Ber. 64, 2819 (1931). United States Patent 1, 784, 280 (Dec. 9, M. , United States Patent 2, 201, 727 (July 119 47. Gaudin, A. M ., Principles of Mineral Dressing, 1st e d ., McGraw-Hill Book Company, New York, 1939. 48. Gaudin, A. 11, 1941). 49. Gaudin, A. M. , Glover, H. , Hansen, M. S. and O rr, C. W. , Foundations of Flotation: Technical Paper Univ. Utah 1928, No. 1. 50. Gaudin, A. M. and Sorenson, P. M ., Foundations of Flotation P a rt H: Technical Paper Univ. Utahl 1928, No. 4. 51. Gaudin, A. M. May, 1946. 52. Geyer, B. P ., and Smith, Soc. 64, 1649-51 (1942). 53. Gilman, Henry, Organic Chemistry, 2nd ed. , pp. 18681883 (vol. II), John Wiley and Sons, New York, 1943. 54. Green, L,.,McCarthy, J. , Chem. 128, 447 (1939). 55. Guile, Ralph, The Condensation of Some T ertia ry Di­ methyl Amyl Carbinols with Phenol in Presence of Alum­ inum Chloride, unpublished Ph. D. thesis, Michigan State College, 1938. 46 numbered pages. 56. Gurewitsch, 5 7. Gutzeit, G., 1946). 58. Gutzeit, G. , Am. Inst, No. 2077, Nov. 1946. 1946. 59. Haendler, H. M. and Geyer, Soc. 60, 2813-14 (1938). M. , United States Patent 2, 231, 265 (Feb. A. , and P re lle r, Ber. G. G. S ., Min. M cP ., J. and King, 32, C. Tech. Am. G ., 10, Chem. J. Biol. 2424 (1899). United States Patent 2, 395, 866 (Mar. Min. Met. Engrs. Tech. In Min. Tech. 10, p. 3, B. P. , J. Am. 5, Pub. Nov., Chem. A 120 60. Haendler, H. M. and Smith, Soc. 61, 2624-6 (1939). 61. Henrich, 62. Henrich, Ferd. and Eisenach, [2] 70, 332 (1904).-/ H. , 63. Huang-Minlon, Soc. 64. Hedrick, G. W ., The Condensation of the T ertiary Hep_ tyl Alcohols with Phenol in Presence of Aluminum Chlor­ ide as a Catalyst, unpublished Ph. D. thesis, Michigan State College, 1937. 46 numbered pages. Ferd. , J . J. Prakt. Am. G. M cP ., Chem. Chem. J. [2] 70, J. Am. 315 (1904). Prakt. 68, Chem. Chem. 2488 (1946). « 65. Huston, R. C. , 18, 1936). United States Patent 2, 051, 300 (Aug. 66. Huston, R. C. and Bailey, 68, 1382 (1946).' 67. Huston, R. C. 69-71 (1939). 68. Huston, R. C. , Guile, R ., Bailey, D. L. , Curtis, R. , and Esterdahl, 'M ., J. Am. Chem. Soc. 67, 899-901 (1945). 69. Huston, R. C. and Hedrick, 59, 2001-3 (1937). 70. Huston, R. C. and Hsieh, 58, 439 (1936). T. 71. Huston, R. C. 2655 (1942). G. , 72. Ilinsky, 73. Ingold, C. K. 173 (1924). and Guile, D. L«. , J. R ., and Meloy, M. and von Knorre, and Piggott, J. G. Am. W. , Y. , H. A. , J. Am. Am. Am. Ber. Chem. Chem. J. J. J. G. , Am. Soc. 61, Chem. Chem. Chem. 18, Soc. Soc. Soc. Soc. 64, 69 (1885). Chem. Soc. 125, A 121 74. Jonassen, H. B. and Dexter, 71, 1553-6 (1949). 75. Kamm, O. and Marvel, C. S. , Organic Syntheses Coll. Vol. I, 2nd e d ., p. 30. John Wiley and Sons, New York, 1941. 76. Keefer, R. M. , Andrews, L. J. and Kepner, J. Am. Chem. Soc. _71, 2381-3 (1949). 77. Kharman, Emil, Ghatyas, L. W. and Shternov, J. Am. Chem. Soc. 53, 3397-407 (1931). 78. Kietabl, 79. Knowles, 80. Koenigs, W ., 81. Krohnke, F. , Ber. 82. Kubota, M. ,J. Chem. Soc. Japan 61, 1176-81, (1940); in Chemical Abstr. 37, 4049 (1940). 83. Lange, N. A. (ed. ) Handbook of Chemistry, 7th e d ., p. 132. Handbook Publishers, Inc. , Sandusky, Ohio, 1949. 84. Liebmann, 85. Ludt, R. W. and DeWitt, C. C. , Am. Inst. Min. Met. Engrs. Tech.. Pub. No. 2535B: in Mining Transactions, p. 49, Feb. 1949. 86. Mayr, 87. McGreal, M. E. and Niederl, Soc. 57, 2625 (1935). 88. McDonald, 89. Mellan, Ibert, Organic Reagents in Inorganic Analysis, pp. 7-14, 153. Blakiston Co., Philadelphia, 1941. Carl, Monats. 1£^ H. B. , Bur. C. , Ber. Ad. , Z. W. H ., J. Am. Chem Soc. R. E ., V. A. , 53 7 (1898). Stds. J. Res. 9, 1 (1932). 23, 3145 (1890). 60, 527 (1927). Ber. Anal. T. , T. 14, 1842 (1881). Chem. Eng. 1182-7 98, Min. J. J. 402 (1934). B. , 126, J. Am. Chem. 679 (1928). 122 90. Mellor, J. W. , p. 8, Vol. VI, London, 1923. 91. Morgan, G. T. 1599 (1924). 92. Morgan, G. T. and Main Smith, C olorists 41, 233 (1925). 93. M orris, T. M ., Amer. Inst. Min. Met. Engrs. Tech. Pub. 2734B. in Min. Eng. 187, No. 1, 91^-5 (1950). 94. Page, H. • J. 3241 (1923). 95. Page, 96. Paranhpte, K. , Phalnikar, N. L. and Nargund, J. Univ. Bombay 11, pt. 3, 120-3 (1942). W. T reatise on Inorganic Chemistry, Vol. p. 6. Longmans Green and Company, and Glover, and Heasman, C ., En g . Min. T. , B. J. J. D ., R. , 127, Chem. Dyers Soc. 123, 520 (1929). S. , P arkes, G. D. Chemistry, p. York, 1939. 99. Patek, J . , Amer. Inst. Min. Met. Engrs. Tech. Pub. No. 564 (1934). In Trans. A. ' I. M. M. E. 112, 486, 508 (1934). 103. Irdian Chem. K. 98. 102. J. Soc. 125, Parekh, N. B.and Shah, 19, 339-42 (1942) 101. C. , J. J. Soc. 97. 100. R. J. Chem. Ill, Soc. and Mellor, J. W. , Modern Inorganic 685. Longmans Green and Co. , New Patek, J . , United States Patent 2, 331, 722 (Oct. 1944). 12, Pauling, Linus, The Nature of the Chemical Bond, Cornell University P r e s s , Ithaca, New York, 1945. Perkins, R. 4, 1934). P ., Prilaschajew , United States Patent 1, 972, 599 (Sept. Nikolaus, Ber. 42, 4814 (1909). A 123 104. Putnam, M L E ., Button, E . , and Perkins, R. United States Patent 2, 039, 344 (May 5, 1936). 105. Reihler, Hans, 106. Rex, J ., 107. H. Z. Anorg. EngL Min. J. Chem. 149, 257, P ., 340-8 (1948). 69 (Dec. 1948). Rogers, J . , Sutherland, K. L . , Wark, E. E. and Wark, I. W ., Amer. Inst. Min. Met. Engrs Tech. Pub. No. 2022 (1946); in Min. Tech. 10, Jul. 1946. 108. Rolla, ia 79, 109. San’dulesco, G. and Girard, [4] 47, 930, 1300 (1929). 110. Schuhmann, (1942). 111. Sidgwick, N.V. , The Electronic Theory of Valency, Oxford University P re ss, London, 1927. 112. Sidgwick, N.V., J. Chem. Ibid., 127, 907 (1935). 113. Sidgwick, N.V .,Organic Chemistry of Nitrogen, 46, Oxford University P re ss, London, 1937. p. 114. Sidgwick, N. V. 125, 522 (1924). Soc. 115. Spedden, H • Rush, and Hannan, Wm. S. jr. , Amer. Inst. Min. Met. Engrs. Tech. Pub. No. 2354 (1948). In Min. Tech. 12, M ar., 1948. 11.6. Taggart, A, F. (ed .), Handbook of Mineral Dressing, Vol. I, John Wiley and Sons, New York, 1945. 117. Taggart, A. F. and H assialis, M. Min. Met. Bigrs. Tech. Pub. No. Min. Tech. 10, Sept. 1946. M. and Marinazelli, A. M ., Gazz. Chim. Ital­ 164-75 (1949). In Chem. Abstr. 43, 7300 (1949) R. jr. , J. A ., Phys. and Callow, Bull. Chem. Soc. 125, R. K ., Soc. 46, Chim. 891-902 527 (1934); J. Chem. D ., Amer. Inst. 2078 (1946); in 124 118. 119. 120. 121. 122. 123. 124. Traub, W. and Biltz, A ., Ber. 37, 3130 (1904). Tzukervanik, L. P. and Tambovtsevna, Univ. Asie Centrale 22, 221-4 (1938). Abstr. 34, 4729 (1938). V. D ., Bull. in Chem. Umfitsev, V. N ., Doklady Akad. Nauk S. S. R. 50^ 253-5 (1945). in Chem. Abstr. 43, 4627 (1949) Wark, Ian W ., Principles of Flotation, 1st e d ., Aus­ tralasian Institute of Mining and Metallurgy, In c ., Melbourne, Australia, 1938. Wells, A. F . , Structural Inorganic Chemistry, Oxford University P re s s , London, 1945* Wieland, H. Wohl, A ., and Beyel, Ber. 27, F. , 1432, Ann. p. 439, 196 (1924). 1815 (1894). 473. 120 ME5H m emu "couEcm mmitiT wem. JO-W O MESH n c o Micmm. horn r collector mmm t o MFJH FEED imomHT wi * \L / ^ Xi-ilO mejh ma.