FLOTATION AND WASHRNG OF SLWY GELAHHOUS FRECEPITATES OF OXYGENATED METALS Thesis for the Dogm 91‘ M. S. MWHFGAN STATE UNWEafiTY Donald Rober? parker 1958 “4rfi‘5 This is to certify that the thesis entitled Flotation and Washing of Slimy Gelatinous Precipitates of Oxygenated Metals presented by Donald Robert Parker has been accepted towards fulfillment of the requirements for Master of Science degree in Chemical Engineering 0-169 FLOTATION AND WASHING OF SLINY GELATINOUS PRECIPITATES OF OXYGENATED METALS by DONALD ROBERT PARKER AN ABSTRACT Submitted to the School for Advanced Graduate Studies of Michigan State University of Agriculture and Applied Science in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Chemical Engineering 1958 ,/’122;;f7wéégg? i Approved: <:///' éVV~L/‘f '"M ) ABSTRACT This is a report of an investigation to examine the feasibility of using flotation as a method for washing slimy gelatinous precipitates. The precipitates studied were the hydroxides of magnesium, nickel, copper, cadmium, and zinc, and the hydrous oxides of aluminum, iron, manganese, chromium, and cobalt. Particular emphasis was placed upon the flota- tion and washing preperties of gelatinous magnesium and aluminum. Flotation of the precipitates was accomplished in a machine designed to cause flotation by release of dissolved air. The machine was similar in principle to the Sveen- Pedersen Saveall used in the paper industry for white water clarification. Oleic acid was found to be effective as a collector for all the precipitates studied with the exception of cupric hydroxide. Cupric hydroxide was floated using eosin red dye as an activator and emulsified light machine oil as a collector. Data are presented to show the relative efficiency of several organic acids as collector reagents. The influence of temperature upon collector efficiency and adherence during washing was also studied. Results indicate that at temperatures exceeding 120° F the gelatinous precipitates of aluminum and magnesium may be floated and washed three times with no appreciable loss of precipitate. FLOTATION AND WASHING OF SLIMY GELATINOUS PRECIPITATES OF OXYGENATED METALS by DONALD ROBERT PARKER A THESIS Submitted to the School for Advanced Graduate Studies of Michigan State University of Agriculture and Applied Science in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Chemical Engineering 1958 :4, Vs '. e ' ' 1‘ , l "J '! I I Z ,4 ,z" . '1: Ln“ (.5 *3 :3 ACECII OU’.’LEDGEF$EIIT The author wishes to eXpress his appreciation to Dr. Clyde C. Dewitt for he guidance and encouragement which he gave throughout the course of this investigation. Thanks are also due to the Dura-Metallic Corporation of Kalamazoo, Michigan, for their generous donation of a Dura- Scal air tight bearing for this work. TABLE OF CONTENTS CHAPTER PAGE ACKNOWLEDGEMENT . . . . . . . . . . . . . . . . . . ii I. III'I‘RODUCTION O O O O O C O O C I O O O C O O C l Flotation Applied to Problem of Washing Slimy PPeCipitates O O O O O O O O O O O O O O 0 Scope of the Investigation . . . . . . . . . O‘UIF' II. EXPERIMENTAL EQUIPMENT . . . . . . . . . . . . III. PRELIMINARY EXPERIMENTAL WORK . . . . . . . . 11 Survey of Metallic Hydroxides and Hydrated 0x1des O O O O O O C I O O C O I O O O O O 11 Flotation EXperiments with Digested Bauxite 11 Experimental Variables . . . . . . . . . . . 13 IV . 12R OCEDURE O O O O O O C O O O O O O O C O O C 15 Usual procedure Followed in Making Individual Trial Flotation Tests on Gelatinous Aluminum and Magnesium . . . . . . . . . . . . . . 15 v. EXPERIMENTAL SECTION . . . . . . . . . . . . . 18 ‘Reagents Tested as Collectors . . . . . . . 18 Water Emulsions of Oleic and Linoleic Acid . 18 Turbidity Determination to Measure Flotation Results 0 O O O O O O O O O O O I O O C . 2O Flotation of Magnesium Hydroxide at 1150 F Using an Oleic Acid Emulsion . . . . . . . 28 Effect of Temperature on Efficiency of COlleCtor O O O O O O O O O O O O O O O O 39 Repeated Flotation and Washing of the Aluminum Oxide and Magnesium Hydroxide Precipitates ho VI. DISUCSSION . . . . . . . . . . . . . . . . . SS Bubble Attachment . . . . . . . . . . . . . 55 iii CHAPTER PAC“ Relative Efficiency of Some Organic Acids as ’ Collectors . . . . . . . . . . . . . . . . . 56 Frothing . . . . . . . . . . . . . . . . . . . 60 pH as a Variable . . . . . . . . . . . . . . . 62 Air Flow Through the Chamber . . . . . . . . . 63 Air Pressure in the Chamber . . . . . . . . . 63 Release of Air Pressure . . . . . . . . . . . 6S Agitation . . . . . . . . . . . . . . . . . . 6S Washing the Precipitates . . . . . . . . . . . 66 Hardness of Water and Its Effect upon Results 67 VII. CONCLUSIONS . . . . . . . . . . . . . e e . e 0 7O BIBLIOGRAPIIY O O O O O O O O O O O O O 0 O O O O O O O 71 iv FIGURE 1. ll. 12. 15. 16. 17. 18. 19h- 20. 21. LIST OF FIGURES PAGE The flotation machine and auxilary equipment . . 7 The flotation machine and auxilary equipment . . 7 Close-up of lower section of flotation machine . 8 Sketch of the air pressure relief valve . . . . 9 Transmission curve for gelatinous aluminum . . . 2h Transmission curve for gelatinous magnesium . . 26 Mg(OH)2 flotation with an oleic acid emulsion . 29 Mg(OH)2 flotation with an oleic acid emulsion . 3O Mg(OH)9 flotation with an oleic acid soap SOlugion . . C 0 O . O O O O C O O O O O O C . 33 Mg(OH) flotation with an oleic acid soap SOluéion O O O C . O O C I O O C O O O O O O 0 3h A120 .3H 0 flotation with a lauric acid soap SO uti n O O O O O O O O O O O O O C O 0 O O O 36 A120 .3H20 flotation with a lauric acid soap SO ution o e e o e e e e e e e e e o e .0 e e O 37 Mg(OH) flotation with a lauric acid soap 301.11%1 on Q C C C O O O O C O 0 0 C O O O O C O 1+0 Mg(OH)a flotation with a lauric acid soap solueion . . . . . . . . . . ... . . . . . . . hl Effect of temperature on flotation of Mg(OH)2 . h3 Effect of temperature on flotation of Mg(OH)2 . hh Effect of temperature on flotation of A1203.3H20 h? Adherence of various collector films on gelatinous aluminum 0 O O O O O O O O 0‘ O O O O O O O O C 51 Adherence of linoleic and lauric acid films on gelatinous magnesium . . . . . . . . . . . . . S2 Bubble attachment comparison . . . . . . . . . . S7 Flotation efficiency as a function of air pres- sure with the Sveen-Pedersen Saveall . . . . . 6h V TABLE II. III. IV. V. VI. VII. VIII. IX. XI. XII. XIII. XIV. LIST OF TABLES PAGE Metallic hydroxides and hydrated oxides floated and reagent used . . . . . . . . . . . . . . 12 Reagents tested as collectors . . . . . . . . 19 Hydrated aluminum oxide transmittance data . . 25 Magnesium hydroxide transmittance data . . . . 2? Magnesium hydroxide floation with an oleic acid emulsion containing approximately 1.0 ml. per 200 mls. solution . . . . . . . . . 31 Magnesium hydroxide flotation with an oleic acid soap containing 1.0 ml. oleic acid per 200 mls. SOlution e o e e e e e e e e I 35 Hydrated aluminum oxide flotation using a lauric acid soap containing 1.0 gram lauric acid per 200 mls. of solution . . . . . . . . . . 38 Magnesium hydroxide flotation using a lauric acid soap solution containing 1.0 gram of lauric acid per 200 mls. of solution . . . . t2 Effect of temperature on flotation of Mg(0H) using 5.0 mls. of an oleic acid emulsion 2 containing approximately 1.0 ml. oleic acid per 200 mls . . . . . . . . . . . . . . . . ES Effect of temperature on flotation of gelatinous aluminum oxide using 0.6 mls. of oleic acid soap containing one part per 200 parts by VOlumo e e e e e e e e e e e e e e e e e e 0 us Effect of temperature on flotation of gelatinous oxide using 0.6 mls. of linoleic acid soap containing one part per 200 parts by volume . E9 Repeated washing and flotation of gelatinous aluminm O O O O O O O O 0 O 0 I O O O O 0 0 SB Repeated washing and flotation of gelatinous magnesium . . . . . . . . . . . . . . . . . Sh Analysis of service water supplied to buildings on the north campus of Michigan State Univer— sity. This is a typical analysis . . . . . 69 vi CHAPTER I INTRODUCTION The separation, grading, and purification of materials, is a problemwwifih'which engineers and in particular mining and chemical engineers are frequently confronted. To solve these problems vast sums are invested annually by many in- dustries for milling and screening plants, filters, frac- tionation columns, settling basins, and many other types of separation equipment. The separation and purification of materials is a primary justification for the employment of thousands of chemical engineers. There is a continual search for more effective and econdmical means for the separ- ation and purification of materials. The principal of flotation provides in some industries, and in particular the mining industry, an effective, and in many cases, a very efficient means for the separation of materials. In the mining industry, where this principle is currently finding its primary application, it is being used, each year, for the concentration of 150,000,000 tons of non-ferrous ore in the Western Hemisphere alone.(h) The application of this principle is no longer limited to the minerals industry however. The number of materials separated or concentrated by flotation has, in recent years, grown rapidly. Such diverse materials as coal, diamonds, Sulfur, collodial oil, peas, seeds, paper mill white water wastes, dyes, certain bacteria, viruses, many types of industrial wastes, and sewage are being separated by flo- tation.(7) It appears that the method will come into even greater use as the horizons of technology and imagination are extended. All materials which are not chemically inert may theoretically be concentrated by air flotation from liquids in which the finely divided materials are suspended. How- ever, as is often the case, the breach between that which is theoretically possible, and that which is at present tech- nologically feasible, is great. The primary problem usually associated with flotation systems is a problem involving surface chemistry. In order to separate a desired particle by flotation, it is necessary to form on at least a portion of that particle, a surface which is sufficiently hydrophobic as to cause the particle to be held at a gas-liquid interface. It is often necessary that similar hydrOphobic surfaces be absent from such foreign particles that are present as impurities. The technique for bringing about such a condition in the suSpended material being separated is, in essence, the science of flotation. The flotation technique, now known as gross flotation, was first demonstrated by the Norwegian engineer, Nils Pedersen, and patented in the United States in 1921(11). Pedersen's machine was also patented in Norway in 192h and in Germany in 1925. Pedersen's machine was designed to recover white water waste from the Fourdrinier machines of the paper industry. This white water contained suSpended fibers, rosin size, glue, casein, clay, and mineral pig- ments. In Pedersen's machine,air at 30-h5 p.s.i.g. was dissolved in the white water suSpension. After a short time the pres- sure was released and the air coming out of solution as myriads of submicroscopic bubbles adsorbed on the surface of the white water particles,floating them, in a bubble-like blanket, to the surface of the liquid from whence they were removed mechanically. Pedersen's development was not immediately successful. Pedersen's invention did not prosper until the intro- duction by Karl Sveen, another Scandinavian engineer, of an agglomerating material which came to be known as Sveen's glue. This glue was made from animal glue, rosin size, alum, and formaldehyde; the formaldehyde serving to prevent decomposition of the mixture. Sveen glue, when used in the Pedersen machine, materially aided the flotation process and the Sveen-Pedersen Saveall, as it became known, mmssoon adOpted by the paper industry throughout much of Europe. The Sveen-Pedersen Saveall was introduced into the United States in 1937. Two years later, August 17, 1939, (11) presented a very complete description Ward D. Harrison of the commercial Operation of the Sveen-Pedersen Saveall before an assembly of the Technical Association of the Pulp and Paper Industries. He reported reduction in white water 1+ solids from 9 pounds to 0.3 pounds per thousand gallons in some cases. The utilization of the Sveen-Pedersen machine is now common in the United States paper industry. Its use effects a very considerable economy in material losses and avoids resultant stream pollution. New uses are rapidly being found for the Sveen-Pedersen Saveall. Among these new usages are: the removal of oil from brines and water, recovery of fines from synthetic rubber polymerizing processes, concentration of citrus pulp, recovery of soap spills, recovery of fish pro- tein, hydration of fruit skins, seeds, and stems; re- moval of emulsified oil from railroad round house waste water, removal of carbon from airplane engine cleaning water, removal of grease from meat packing wastes, re- moval of solids from beet sugar waste waters, removal of starch from water, removal of stillage solids from winery waste water, removal of fines from fermentation leps, removal of albuminoids from beet and cane eager solutions, and removal of fine solids from sewage. Flotatiqg Applied 32 Problem of Washing Slimy Precipitates It is necessary in certain processes to filter and wash the slimy gelatinous precipitates of metallic oxides or hydroxides. This operation is a very difficult and time consuming task. Filtration of such precipitates is nec- . essarily costly. Flotation of these precipitates would allow simultaneous separation from salution and washing to be combined in one operation. The demonstration of such a flotation procedure based on known flotation principles was a part of the main objective of the present investigation. Scope 22 the Investigation The purpose of this investigation was to examine the possibility of using the gross flotation technique for the washing of slimy precipitates of metallic hydroxides and hydrated metallic oxides. This work involved a study of the kind and amount of effective collector reagents for the efficient flotation of such precipitates in a machine de- signed to effect the release of air dissolved in water sus- pensions of such precipitates at pressures higher than atmospheric. The gelatinous precipitates of magnesium hydroxide and hydrated aluminum oxide were chosen for a detailed study; these materials were considered to be representative of oxygenated slimy precipitates. Other hydrated metaflic oxide precipitates were shown to be amenable to washing by the gross flotation technique. CHAPTER II 6 EXPERIMENTAL EQUIPMENT The primary item of equipment used in this investigation was a flotation machine especially designed for the project. (See Figures 1, 2, and 3). This machine consisted of a pressure chamber built of standard pyrex glass pipe; this chamber was provided with an internal agitator. Air, under pressure, was applied to the base of the machine and bubbled up through the agitated suspension. The air at 80 p.s.i.g. was brought to the machine through c0pper tubing. An air filter preceded the valve regulating the air supply to the machine. This filter prevented the introduction of entrained oil into the pressure chamber of the machine. Air pressure in the machine was regulated by means of a specially built relief valve. (See Figure h). This valve consisted of a steel ball bearing which rode on a brass seat; it was weighted by means of a lever arm resting upon it. A sliding weight on the lever arm, held in place by a set screw, served as a pressure adjustment. During Opera- tion of the machine there was a continual flow of air through the chamber and contents, since the air supplied to the machine was at 80 p.s.i.g., whereas pressure in the chamber was held constant throughout all of the experimental work at 35 p.s.i.g. The flotation machine and auxilary equipment. Figure 1. Figure 2. Close-up of lower section of flotation machine. ’! Figure 3. Notice clarity of water beneath floated layer. The c0pper tubing air connection and thermocouple wires are shown here. The thermos flask containing ice and used as a reference junction with the potentiometer is also evident. av Nth at uni aw kQQQOSn. teas. \ '\ \ \ \ P-‘ I \ . " wzxmfiwfl I fl~'-'—“’.-"'.'-—‘-"-"‘----—-------—--—------—-------. \ awash e235, .\ \\ 3% £Y\ «a S was: ea c \ N\ g KNxVNQ N W3 WQMQQ WSW his}; 41 \\ 10 The internal agitator assembly consisted of a a" stain- less steel shaft fitted with four stainless steel impellers equi-spaced along the shaft. The shaft was supported at the bottom end by a nylon bearing, and at the t0p end by a special pressure sealed bearing(5) through which it extended to the outside of the pressure chamber. Outside, and immed- iately above the chamber top, the agitator shaft was support- ed by a pillow block bearing which prevented the application of a lateral stress upon the sealed bearing. The agitator shaft was driven through a V-belt connection by means of a hydraulic speed reducer, which was in turn connected by a V-belt drive to a % h.p. electric motor. The agitator was driven at approximately 300 R.P.M. Materials entered the machine through a %” stainless steel pipe fitted with a i" stainless steel valve. A funnel served to direct liquids into the entrance pipe. Liquids were removed by gravity through a valved %" stainless steel pipe at the bottom of the pressure chamber. Both the top and the bottom of the pressure chamber were sealed by stain- less steel plates, fitted with rubber gaskets. These plates held the pipes and the agitator shaft bearings. The pressur- ized air connection was attached to the bottom plate. The Pyrex pipe aluminum flanges were standard size. CHAPTER III 11 PRELIMINARY EXPERIMENTAL WORK Survey gf Metallic Hydroxides and Hydrated Oxides The first work undertaken was a survey to determine the floatability of many of the metals forming hydroxides and hydrated oxides.(17) It was found (see Table I) that the precipitated hydroxides of aluminum, magnesium, manganese, chromium, cobalt, cadmium, nickel, zinc, and iron could be floated using oleic acid as a collector. In the case of iron, it was found necessary to add a small quantity of activated silica, in order to make the oleic acid an effec- - tive collector. The hydrous hydroxide of bi-valent copper was floated with an emulsion of light maChine oil in methanol after first activating the precipitate with eosin red dye which it adsorbs.(18) The flotation was difficult in this case. Flotation Experiments with Digested Bauxitg Attenpts were also made to float the residual red mud from the digestion of bauxite with caustic. A 50 pound sample of Suriname bauxite was generously donated for this work by the Aluminum Company of America. The digestion was performed in an autoclave according to directions outlined by Dr. J. W. Newsome,(1u) Chief of the Alumina and Chemicals Division, of the Aluminum Company of America. Several reagents were tried as collec- tors for the red mud but none of those tried was found to be effective. macs cacao w modmz opfixoaehm ++QN 9e msosemm 1 case cacao m mowz commences s+ee msoaphm Hosanna: Hosanna: 2a use cos chaos as see semen m meaz oeaxoees: +sso powuamassm msoaphm eao< oaoao m moaz oeaxoeesm ++az eschew: pwo< cacao m momz epfixo ++oo msoaphm efio< cacao w mozmz mam oeawo +++AU escapmm 32 208 o. moaaz was 02513:: msoaemm woeafim poum>fiuo< ewes cacao w wodmz mam opaxo +++om escaphm gem cacao Ha modmz “mam opaxoaemm I m2 escapmm pace cacao m.o miudmz mam opfimo popsaehm.4eeam macaehm “soue>fimw

\1.\V\ at N RXN Q\_mm\ Kins an). WN\\ Q Q\ QM. \ Q J C Q 2 wt Q W L f\\ Re . n/ \HV. \ 7 0 Q a \M ...f\‘ x e r \V \ N w sneak 3.0%.... awake Q \a .._ uox ZEM seams QCux VG Nb Rm... hkxk Ewirkefiq exte\%§\ O .. 3 bQ\ x QMRVQ 4k fitQNxex 3s...wt\ .740 000%. named». .3}... :0. - oxamux ktexm 1.4 q N \ 9% 9.2 .1 9.2 9.3 mi g: 3.. m we a N w 10l 4 £4 - - 7-3: [:1 -;I....--.l- .74. r- 3---- -.-...4.i..-l1-1 « 1 4 . l4 1 “rue . ex p... H .w. «MVWK EQQK NANNQ m L 6% 1 an F - l. :1 _ .09 >30 ........§w Saw 2...... to sea... SS... kegs... s Q stag 31 H.0H 0.00 00.0 m.00 00H 0.00 ea :.mH 0.00 00.0 m.00 mad 0.00 0 0.0a 0.00 00.0 00 00H 0.00 0 0.0a 0.00 00.0 00 00H «.00 0 0.00 0.00 00.0 00 0HH 0.00 0 o.HH m.00 00.0 00 NHH 0.mm 0 0.0 0.00 00.0 00 0H0 :.Hm m 0.0 0.00 00.0 00 0H0 a.0a : 0.0 0.00 m0.H om mHH 0.0a m 0.0 0.0m om.H on sea 0.0 m 0.0 0.00 m0.o ma 00H 0.: H €0.02... .00. r100... .000. 1.10.0. 0...... it... 00000 00000; e mammqmm fleece acco.mmm HeaeaeH 0060 cacao pcoo pom pCon>Hsdm mowaa I oaSpwacQEop owmao>< .aoauaaom .mHE oom non .HE o.H haopeefixoaaae wcfizfiwpcoo scamasso twee cacao am an“; :000000H0 ocfixoachn Edaeocmwfi .> mqm8: 5 so made Qua cm. 6 set E E Rut sue Q unisex -..__ ‘5‘ QN §v cm 0% QQ\ .37 J/Vz’v'l J //4/_S‘ N631 04.. uh. 3 Q\ NWSWVKK ‘3 x «3%: $ch Ema Sun EMS 3258a \ stud klewmé 9mm, mum Sow QdN 9.3 3\ en, . i J a J _ 41 M.“ 9% 4” .NN. Wuhan rxskk QKWQ hm. W n: .0/ PM cm H w w“ 0 -3.» .1 - . a? :xfikxxuuw .3? any rank QC? .356 ii 3‘53th \vak b 3% QRAQV) 99% bkuVQ >R\ TK\\\ >\§\ka& Q N\ ”win? 3S H.mm m.00 00.H 0.00 00H 0.00 om.H m.am m.00 00.H m.00 0aa 0.00 00.H m.Hm m.00 00.H m.00 maa 0.0: 00.H m.ea m.00 00.H m.00 maa H.00 00.0 0.ma 0.00 00.H ~.~0 mad 0.mm 00.0 0.0a m.00 e0.a ~.m0 mad m.am 0m.0 0.0 m.e0 m0.H 0.m0 sad H.ea 00.0 5.0 0.m0 00.H 00 :HH 0.NH 0m.0 0.m 0.m0 :0.H m0 maa p.0a mm.0 mwm m.0e mm.a 00 :HH m.0 0m.0 0300 0030; flflwemdm mg g 0000 Q % MWmemmwm m odaa I endprOQEop owwpo>< .GOapzaom .mHE com pom vfiow cacao .HE o.H mcficfiwucoo meow pace cacao cw 29H: :oapmpoam opaxonohn ssfimocwmz .H> mqm< .coauzaom mo .mHE oom pom cwom oo.m m~.m 0m.m 00.m 00.0 00.m m~.H mmmm hmehoy vfio< ownswq vCon>Hnwm 0.0 m.m 0.m 0.: 0.: 0.: m.m $0. camsaa Ewyw o.a maficfimpcoo mwom cwow capswa a mafia: zofipwpoah ooaxo Ezcfiesam Uoudpomm .HH> mqm¢8 39 adding 2.0 grams of sodium hydroxide, and adding water slowly with mixing until all of the lauric acid had been sanonified and a final volume of MOO mls. was reached. This lauric acid sodium soap solution therefore contained 1.0 gram of lauric acid per 200 mls. The curve of Figure 12 is somewhat risleading and re- quires explanation. The curve might be interpreted to indicate that flotation improved as the ratio of lauric.acid to Mg(OH)2 decreased. All the evidence available however, strongly suggests that the ratio of lauric acid to M3{QH)2 floated is high when the total amount of lauric acid present is low due to the relatively large solubility of lauric acid in the water solution, and that after the solubility requirement of lauric acid in water has been satisifed the behavior is analogous to that of oleic acid as shown in Figure 1C. In both cases a maximum of flotation is indicated at definite reagent concentrations. Figures 13 and 1h show the flotation results obtained in floating Mg(OH)2 with the same lauric acid soap used in floating the A1203.mH20 of Figures ll and 12. Effect of Temperature 2g Efficiency 22 Collector Figure 15 shows the effect of varying temperature upon the flotation of Mg(OH)2 using the same emulsion used pre- viously in obtaining the curves of Figures 7 and 8 for the flotation of Mg(OH)2. The effectiveness of 5.0 mls. of the m)\ I NwlSw>ux 06 5 2.“.ka «mm, osmium; QQQ .wxmxokx V we“? 0. b &V QM, GM. 0 ...\ 0 9m. 40 L OQ\ Wk‘ “k Si? ‘ m “L u“ a u ‘U\U \ t . “I! r“ u J u 0 . .0 . a- . . . «W» \.\J "u . 2st 3 do Swan abs. o is»... u \\ {it} \éxkfie Q wise? 04v JJ/V JJ/N§Nt/é/J $\ nearest e? Ease Q $wa $33 Qua Sea... has. q no): we. - steam a I 0 GE 1 I..+ e3 03 e: 03 are me 1 if K. QNQ \ Wkekhkm‘oxtwk Nmnwfiux‘xs :4? 00m. wank QC§ waxes. V. 235% o\ Qnusxxkxkob \wak er 23.3 see deem, Sea Sit... N ex 323 >§h§e wk «is? on d 3.0. 1 l -06 QB cm 0% 00\ (7711/0 7.7% yaw/y .é L2 ”0.0 m~.o 05.0 NN.H moan Namo~w2\efiaswq odpmm pause; 0.m0 m.0e 0.00 m.00 eopmoflm Hence Illmelll uaoo hem so.H em.a 00.H ~0.H mooma .coapsaom no .mas oom pom pace caused me Seam o.a maficfiwpcoo scapsaom doom pace canswa w wcfimo coapmuoau opaxopphn Esfimocwwz “mswpwv eopwoam movmz Hmpoe m.so 0.Hm m.so be oocwpuHE Imnwhe 9:00 pom ®HH HNH HNH QHH chaps IsomEoe HewpficH I chapwhomfiou owmho>< .HHH> mqm<9 0.H 0.H m.H m.m moan headnov cfio¢ canned Amdaw>findm 0.0 0.m o.m 0.m .MHE pcomeom In.\ MnKo w\h\ k \NIEIutkmnk 9%anme Qe\ 92 Si 9? wa II3 Q\\ 03 em 4 W a . . NH HRS 20$ shad m .32 3%. ~31 QC? Em «0 .Ex xvhksxgeeias $§§r§§u 2e}. 43th QC? 2% we .kkwex mot Are? to zqksfi s Q 20 Nessie an? E ks a Putt. l 60 on l QQ\ Jame/14w mew 0&7 \\ humxxo n\,\I\ k \ lequika ON NU. when! NWSK QWQQRKNK M QW\ 9h.\ QV\ QM:\ QN.\ Q\\ QQ\ « q q 4 a a 9% w w W MN wank“ Eskx this . es w w. \ m. \\ QQ\ « .Wéx QQN Who. QQx u P36. “9: xnwkatkoknkxh wlxxxxuxkeau ESQ §>xuu QC‘ not. so -khwewwefi so»? he 2st Es E 5 mesh redeem us. Chopin TABLE IX. Effect of temperature on flotation of Mg(OH) using 5.0 mls. of an oleic acid emulsion containing approximately 1.0 ml. oleic acid per 200 mls. Initial Temper- Per Cent Total Mg(c§12 Per Cent 2; ature Trans- Pleated Total mittance (Grams; Floated 116 77 1.77 88.5 128 90.5 1.92 96.0 1,411» 9605 1'97 9805 151+ 99.6 1.99 99.5 1.6 emulsion was determined at several different temperatures and as Figures 17 and 18 indicate the variation of collector efficiency with temperature is great. A correSponding temperature effect is shown in Figure 17 in floating hydrated aluminum oxide using the same sodium soap solution of oleic acid as was used in obtaining the curves of Figures 9 and 10 for Mg(OH)2. The preparation of this oleic acid soap solution has been previously de- scribed. Shown also in Figure 17 is a similar curve obtained with a linoleic acid sodium soap solution prepared in exactly the same manner as the previously described oleic acid soap solution was prepared. The results are shown on the same plot for comparative purposes._ Repeated Flotation and Washing 2f the Aluminum Oxide and Magnesium Hydroxide Frecipitates Washing of the aluminum and magnesium precipitates was accomplished by the following stepwise procedure: 1. The suspended precipitate was first floated in the usual manner. 2. One-half of the total volume, or 850 mls. of the water layer beneath the floated material was then drained out of the machine. 3. An equal quantity (850 mls.) of fresh water was then added to the machine. (In the case of Mg(OH) an additional 20 mls. of 28% ammonia was also 2 adled to maintain pH above lO.h). h. The material was then refloated in the usual manner using a ten minute agitation and a ten minute settling period to complete the cycle. in m \ m... sol EMIZ mQstK bmmtwa Imthkmxtmmtwk 9: Q3 QNx Q\\ 09 Q% Q% on Go m. a by N Vme Xsmkk %Km.,n... Q\.\N\ ,w\uu NV .1 Q\VKV VxN NQ>\\ N 0s». e .v. to St and wk 2% Mgfisémmtmk ks u onto. 9m. 4 Qt L o3 oé JJNb’JJ/N S‘Nb’b’l h8 TABLE X. Effect of temperature on flotation of.gelatinous aluminum oxide using 0.6 mls. of oleic acid soap containing one part per 200 parts by volume. 1331‘ mIEEEEEo TEEEEET FTSEEZd 7h 62 1.90 57.6 '75 57 1.70 51.5 95 71 2.23 67.6 95 7h 2.35 71.2 102 70 2.20 66.7 10h 67 2.07 62.8 116 92 3.00 90.8 116 95 3.12 9h.5 121 100 3.30 100.0 126 102 3.30 100.0 #9 TABLE XI. Effect of temperature on flotation of gelatinous aluminum oxide using 0.6 mls. of linoleic acid soap containing one part per 200 parts by volume. Initial 4—203:- -:-_-_::- Edi-$123333“? ~30 -‘-’-‘- Tb’fl" mittance m FTSZI‘E‘Jd 72 72 2.25 68.2 91 81 2.60 78.7 93 7h 2.35 71.2 at 91 .2 .98 9o . 3 102 91 2.98 90.3 113 90 2.95 89.u 118 92 3.00 ‘90.8 123 96 3.15 95.0 126 97 3.17 96.0 127 98 3.23 97.8 1&1 3.28 99.h 99.5 50 The curves of Figures 18 and 19 were obtained by sampling the lower water layer which was removed in Step 2 of the previously listed procedure. Test conditions for each curve, as marked, are given in Tables XII and XIII. Notice the nearly parallel behavior of oleic acid as compared with linoleic acid and also how much better the collector film adheres to the magnesium precipitate as compared to the aluminum precipitate. Notice also the poor adherence of the lauric acid collector in both aluminum and magnesium flotation and that best results were obtained by heating the solutions before flotation. By running at higher temperatures, excellent flotation was assured with a near minimum of collector. The dip indicated in the curves shown in Figures 18 and 19 suggest that possibly the first washing caused particles with collector deficiencies to drOp out. 3 was 00.. 33th.»: .0. gases? 0. m _ M 51 *0 ‘0 Mg Numwg EQQK webmxme ill I .III QC§ bkxbmxu I...T.... QEEES -IIQII: mean 103 III- I I obs 00.30 0 . NEE 33 .1 $00 {31 wa§ b\ .323 w .XMK 20 320 on $03; do >6 02 0E 0.0.x ow u use QStss so wozwrfixfia EJ/Vl/Jl/NS‘Né/é/Z oé m \ was ex..\ .n.%>\\}\miw\\£ \Q wxwmaxtbé % h e In .0 m, N \ 52 i. . , ”Sta Em 06>: 0. Ti. QC.» Swami / / rsxmzwzbbt 33.2..an «Maw 2Q 93 NW... Q\b\\ U\w\\.&\ V Q>\\\ 9\NVQ>\\ V LxQ NbQNQLV§Q§ ON QV Qm JQNb’lL/NS‘A/é/é/J Qé 53 O O O HJO‘O Infiivdo 0 OH "\N “INF-a) e .d-c-Iom O 00.0 m0.m 00.H 00.0 0 o . aw. OO.H oo.H 0 MO HNK" on d’ can 0 on: wmlwlg Hduoa w Haves O O 0 COOH 00am Jiflhfifll CUFKOCD Hmoo OHHM O canons Imwofi voz no.0 canes: Imwoe poz Ammmmwp 0Q amen... 00H 00 00H .Es Infidmr 9:00 900 HOH haa mma N000 IwHIwfiI laHO EOE HaHpacH OHWM OHNM Egl 0am new Dawn: 0 .000; 02 02 new no» new .lILFII UOHHom .EscfiEus msocfipaaow ho ceapmpoah cad madness woueoaom .HHN Handy ofisaop hp 00N\H .9000 v.0 vuo< cacaonaq oEdao> kn 00N\H .0000 ~.0 cuo< cacao .nHz 00m nod seam a .0000 o.m cao< caused oESHo> hp 00N\H .0000 0.0 Udo< cacao osaHo> hp 00N\H .ane 0.0 cae< oaoaoafiq .n anomeom ado «om comb mm 0% m% N% a% 9008 Sh om.H 00.H HH H0 m0.0 00 mod H0.0 00 0HH oases: 00H :mH undo:— 902 00.0 00 00 00.0 mm H0 00.0 mm mm 00.0 mm 00.0 0 H0.0 00 «ca OHDwfifl Ieeoa poz HOH 00H no.0 00 odd canes: Hod 00H lmwOE p 02 nmfiaawv oondpp E ensue 900A Incembl “lg pmm«oe 0:00 000 HaapacH fl Iii-“Ia I OHN MJMONQ OHNM 0) (U 3 .mHE 00m pom Edam H .0000 00% 0.m vno< owhseq Nfi 055H0> hp 00N\H .0000 new 0.0 pao¢ oaoHocHA H% M .0H2 0000 ooaaom pcomeom comb sfidamocmwe msocfipwflow mo aoapdeHm use magnum; poudomom .HHHN mam<fi O ‘ I a *‘J [I 31‘ H U1 \R Bubble Attachment The method of gross flotation as used in this work was different from the usual mineral froth flotation in that bubble attachment was the result, almost exclusively, of gas precipitation upon the surface of the floated particles. In the usual froth flotation of mineral particles, bubble attachment is the result primarily of collision between the bubble and mineral particles in t1 -‘t,t d war is (8) r- u 0 1 10 1e agi.a,e Sicpens on. In such a system it can be shown that as particle size is reduced the chances for successful bubble-particle encounter become much less. is a consequence, systems containing very small particles are floated with difficulty. Surprising as it might seem, very small particles are also difficult to float using the gas precipitation method, see N periments. For example, experiments CD as it was used in th with the flotation of CaCO3 and drigd A1203.nH20 using linoleic acid soap as collector showed that the waterial could not beczompletely floated even though an excess of collector was used and it was run at a high temperature. These materials, C8603 and A120,.nH 0 (dried), showed a marked di ference in their {Lynnbility as compared with the previous gelatinous precipitates. The best results which could be obtained resulted in a transmittance reading of 87% in the case of dri7d A1 13.nH20. Similar results 56 were obtained with CaCOB. It was observed in these cases that there was no appreciable flocculation and that many very fine particles remained in suSpension upon release of the air pressure. It thus appears that flocculation is very essential to the gross flotation method. Unfortunately the method of gross flotation is non- selective. All foreign material particles in the suspension which are not too heavy or dense are floated either by direct collector activity or indirectly by becoming trapped or adsorbed by the straining action of the floating floc.(2) Bubbles are attached to the relatively large floc particles in gross flotation, whereas in mineral flotation the particles are attached to relatively large bubbles. These differences are illustrated below in Figure 20. Relative Efficiency of Some Organic Acids as Collectors It appears, from the results given in Table II, that there are two primary factors to be considered in selecting a straight chain organic acid as a collector for the oxygenated metals; they are melting point and solubility. The melting point of the acid determines, to a large extent, its diapersibility in solution whereas its solubility determines the quantity of the acid required. A more soluble acid is not so effective probably because the film which it forms is not so hydrophobic in comparison to films formed by the less soluble acids, and in addition more of the acid is necessary to satisfy the solubility requirements within the body of the rigid. 503.625 / 7' 7 /‘7( HME N 7 COMPflR/SON .- M/A/f x7571. f2 0777 770M GROSS FLO 7777770N [761/95 20 57 58 Keck and Jasberg(12) found from their studies on the flotative properties of magnetite, that with the saturated fatty acids valeric, caproic, caprylic, and stearic, flo- tation increased with the number of carbon atoms. 'Oleic acid was found to be a better collector than saturated stearic acid. They suggest that the inferiority of stearic acid was due to its lack of diapersion.(13) The results of Table II are in agreement with this suggestion. In Table II lauric acid was the last saturated soil in the series of increasing number of carbon atoms which had a low enough melting point to be sufficiently dispersible at 1150 Fxto be effective as a collector at that temperature. However from the standpoint of quantity, lauric acid was only about one-tenth as effective as unsaturated oleic or unsaturated linoleic acid, due most probably to its greater solubility. This may be seen from a comparison of Figures 9 and 13. It is interesting to note that at temperatures above 1150 F oleic acid is equal to or superior to linoleic acid as a collector according to the results of Figure 17. The oleic acid film is probably more effective at higher temper- atures because it is more hydrophobic. At temperatures below 115° linoleic acid is more effective as compared to oleic acid at the same temperature; linoleic acid appears to be more readily dispersed than is oleic acid. This increased effectiveness of linoleic acid may be due to its chemical 59 constitution, or those physical properties which allow it to be more easily dispersed. According to Gaudin(9) In practice where a carboxylic acid or its soap is required for collection, unsaturated acids such as oleic acid are preferred. The preference for unsaturated fatty acids has led to the search for cheap sources of multiply unsaturated fatty acids such as those of corn oil, cottonseed oil and fish oil foots. It is believed that these fatty acids have greater solubility in water and_a lower melting point than saturated fatty acids with hydrocarbon chains of equivalent length. That unsaturated acids are more effective than saturated acids may well be some function of their solubility. However, on the basis of these findings it is evident that solubility is a detriment to collector efficiency. It therefore appears that the effectiveness of the unsaturated acids such as oleic or linoleic acids must be explained in some other manner. In some instances the multiply unsaturated acids are readily oxidized. To determine whether or not oxidation during flotation with oleic acid was responsible for its high efficiency as a collector, Gaudin and Cole(6) floated fluorite with highly refined oleic acid, extracted the floated mineral with solvents for oleic acid after inoxidative gentle acidification, and examined the extracts for the presence or absence of double bonds. They concluded that oleic acid was unchanged, even when the mineral which it coated was floated repeatedly, that linoleic acid which has two non- -conjugated double bonds, was not effected appreciably more than oleic acid, but that linolenic acid (three non-conjugated 60 bonds) was measurably altered, even after one single flotation step. They concluded that the extraordinary utility of these reagents is related to some property of the compounds other than the oxidizability of their double bonds. (10) The recent work of Hackerman and Makrides suggests and supports another explanation for the superior results obtained with unsaturated acids. The only real difference between a saturated acid such as stearic acid and unsaturated acids such as oleic and linoleic acids is their comparative hydrogen deficiency. From the chemisorption viewpoint an unsaturated double bond possesses an extra pair of sharable electrons. The sharing of these electrons with positively charged surface atoms might well explain the greater efficiency of the unsaturated organic acids as flotation reagents. Frothing As has been suggested previously gross flotation is somewhat different from the usual froth flotation in the manner of bubble attachment as illustrated in Figure 20. The presence of excess froth is completely unnecessary in this type of flotation. It was observed that the degree of frothing of the agitated suSpension with added collector was in no way directly related to the final flotation results obtained. The appearance of froth even with a collector such as oleic acid did not necessarily guarantee good final results, whereas good results were often obtained with a minimum of froth. 61 Frothing is related to the concentration of a potential frother in the solution as compared with its concentration at the gas-liquid interface. This relation is expressed by the Gibbs equation given here, which relates the two concentration factors to the rate of change of surface tension with concentration in the bulk of the fluid. Gibbs equation: /“: :9. .95 RT'dc Whe re, F difference in concentration between bulk of liquid and concentration at gas liquid interface concentration in bulk of liquid surface tension C xi R gas constant T temperature (Note: for greater precision activities should be used in place of the concentration factors in the above equation.) If the rate of change of surface tension with respect to concentration is negative, then f“ is positive and surface tension is lowered with increasing concentration of material in the bulk of the fluid. Such a decrease in surface ten- sion makes a froth or foam more stable, thermodynamically Speaking, since less energy is required for an extension of the surface. Generally, organic substances which are only slightly soluble are positively or preferentially adsorbed at the surface whereas inorganic salts are negatively adsorbed. 62 Organic substances which have both a hydrophobic portion and a polar hydrophylic portion in their molecules are often good frothing agents. Examples of such materials are pine oils, cresylic acid, and certain synthetic alcohols. The organic acids used in this work also acted as frothing agents. When the minimum of the organic acid required for effective flotation was used, very little frothing was evi- dent during agitation. The floated material gave the appearance of the usual froth as observed in traditional mineral flotation and only closer examination would reveal any distinction between the two types of flotation. The distinction is merely one of structure of the froth or bubble attachment as shown in Figure 20. Gross flotation, although different from the standpoint of bubble attachment, is, strictly speaking, a froth flotation. pg 33 3 Variable Consideration has to be given to the effect of pH in almost every flotation system. In some systems it has been found that the difference between float and non-float is merely a matter of one pH unit whereas in other systems the range may be as great as six pH units.(16) pH was not studied in this investigation primarily due to the solubility sensitivity of the materials to pH changes. The solubility consideration presented an obstacle to such a study since the method of analysis depended upon the assumption of constant concentration of the material being 63 floated. To overcome this solubility obstacle different methods of analysis would have been required. It is quite likely, that for the flotation of the aluminum precipitate, that a pH of about 7.0 would have been found to be optimum. Deshpande and Bhat(3) found that a pH of seven was Optimum for the flotation of alumina and Harrison(ll) reported a pH of about seven as being optimum for the flotation of white water in which gelatinous aluminum floc was one of the principal constituents floated. Li}: m Throgh _t_h_e_ Chamber . No attempt was made to regulate air flow through the agitated suspension but it was kept as constant as slight variations of the building's air supply would allow by opening the air inlet valve completely during all tests. It is felt that the total air flow was probably only important to the extent that it affected agitation of the suSpension. Air Pressure ig the Chamber Air pressure in the chamber was kept constant at 35 p.s.i.g. in order that other previously discussed effects (11) gives the curve shown in could be studied. Harrison figure 21 for the relation between air pressure and per cent efficiency for the flotation of paper mill white water. It seems quite likely, in View of the fact that aluminum floc was one of the principal constitutents of the white water system, for which this curve is given, that this relation should also hold quite closely for the system of gelatinous 6h \N .uuth\\ \ m6 hem NWS he wQQ 95V Q5 QV Gm. QN S .r L 4 . fl fll‘ll|\ul|.\||\ 3x .\ Vanessa zmwmmemox ‘ Ewwxe wlk xxx} News 3me was be éetuzsx \\ wax CENSQKN Retake Q %% Qw QQ\ 19/12/9491:”; aé 65 aluminum used in these eXperiments. It would not be sur- prising to find that the same curve also predicted results for the flotation of gelatinous magnesium. Release 2: Air Pressure From observation of the suSpension upon release of air pressure from the machine, it appeared that the results were not appreciably affected by the manner of release pro- vided the release was not too violent or too slow. If the release was too rapid it appeared that much air escaped without effect due to violent agitation of the suspension after some of the gas had already precipitated upon the surface of particles, in which case the bubbles were broken loose and floated free. When the release of air pressure was too slow, a somewhat similar effect was evident, due to turbulence introduced by air escaping from the air tube entering the chamber at the bottom. Both of these diffi- culties were absent when air release was gradual but not too slow. Agitation Agitation was constant throughout these experiments. The flotation machine was equipped with a hydraulic variable speed reducer, but this was run at maximum Speed at all times. A variation of agitator speed may have had an effect upon flotation results, especially in view of the assumed importance of dispersion to collector efficiency. 66 ‘Time of agitation was kept conStant at ten minutes. Both the speed of the agitator and the time of agitation may have reasonably had some bearing on the size of the floc which resulted, as well as upon the degree of dispersion of the collector. It was observed that a finer floc floated more readily than a course floc. Washing the Precipitatgs The procedure for repeated flotation and washing of the precipitate has been given previously and will not be re- peated here. The previous data shown in Figures 18 and 19 and tabulated in Tables XII and XIII indicate, the best results were obtained when the materials were floated at higher temperatures. When higher temperatures were used the loss of precipitate was held at its lowest level. Unfortunately the cost of heat is high.and quite probably would preclude the use of heat in most commercial applications of this principle. Examination of Figures 7, 9, ll, 13, and 15 indicate that in some cases the highest efficiencies are not attained even with excess collector at temperatures lower than 120°. Adherence of the collector film is poor at these lower temperatures. The contamination caused by an adhering acid film such as oleic acid or linoleic acid would amount to approximately 0.09% in the case of gelatinous magnesium calculated as 67 Mg(OH)2 and approximately O.ll% in the case of gelatinous A1203.nH20 calculated as A1203.3H20. Such a slight impurity would very likely be tolerable in most instances. An adher- ing acid film could of course be destroyed by calcination. Even in cases where it was found desirable to use the sodium soap of the acids as collectors no appreciable sodium should interfere since it seems reasonable that only the collector anion is responsible for film formation. Gross flotation precludes the elimination of insoluble impurities by this washing procedure due to the non-Specific nature of the gross flotation method. Only soluble impurities would be eliminated by such a procedure of repeated flotation and washing. It may be found possible to eliminate some impurities by proper control of pH to render them soluble. This could be a subject for further investigation. Hardness of Hater and Its Effect Upon Results A typical analysis of the tap water used in these ex- periments is given in Table XIV. It is evident that there is an appreciable amount of hardness due to the presence of Ca“ Mg“. This total hardness amounted to 339.0 parts per million as CaCOB. The Mg‘+ and Ca“ ions were both precipitated almost completely under the conditions of these tests. A turbidity of 7h% was measured when the pH, of 1500 mls. of the tap water used in these eXperiments, was raised to 10.5 by the addition of 20 mls. of 28% ammonia. 68 Assuming all of the calcium and magnesium was pre- cipitated, a simple calculation showed that 339 parts per million of CaCO3 was equivalent to 0.3h5 grams per 1700 mls. as Mg(OH)2. Since it was known that C‘JOB and MgCO3 also floated under these test conditions it could only be concluded that interference by these materials, in this investigation, was significant. In spite of such interference it was decided to continue the use of tap water due to the difficulty of obtaining sufficient quantities of distilled water for these experiments. 69 TABLE XIV. Analysis of service water supplied to buildings on the north campus of Michigan State University This is a typical analysis.( 5 P.P.M., CaCO3 Equivalent Silica 9.6 Iron 0.3 Galoium 221.0 Magnesium 118.0 Sodium 18.8 Potassium 1.6 Manganese 0 Carbonate O Bicarbonate 38h.0 Sulfate 32.6 Chloride 8.6 Fluoride 0.2 Nitrate 1.h Total Hardness as CaCO3 339.0 CONCLUSIONS I. II. III. IV. V. CHAPTER VII 70 CONCLUSIONS A machine, similar in principle to the Sveen-Pedersen Saveall, was shown to be an efficient device for the flotation of slimy gelatinous precipitates. Oleic and linoleic acids are excellent collectors for many slimy gelatinous metallic precipitates. I The efficiency of the organic acids as collectors for these precipitates was shown to be temperature dependent. Repeated flotation and washing of the gelatinous pre- cipitates of aluminum and magnesium was possible without the addition of more reagent at each flotation step. It was found possible to float and wash the gelatinous precipitates of aluminum and magnesium three times without appreciable loss of the precipitates provided a temperature exceeding 1209 F was used in the initial flotation step. BIBLIOGRAPHY 10. ll. 12. 13. 1h. 15. BIBLIOGRAPHY 71 Campbell, J. M. Private Communication. {1958} Coleman Model 1h. Spectrgphotometer Igstruction Haggai. D'Arcy, N. A. American Petroleum I stitute, Tulsa, Okla. Deshpande, F. K., and G. h. Bhat. "Bubble pick-up Studies with Alumina," g. Ind. Inst. Science, 37 A, 1A1-e (1955). Dewitt, C. C. Private Communication. (1957). Burs-Metallic Corporation, Kalamazoo, Michigan Gaudin, A. M., and Richard E. Cole. "Double-bond Reactivity of Oleic Acid During Flotation," Trans. A.I.M.M.E., 196, Tech. Note lhh-B. (1953). Gaudin, A. M. Flotation. 2nd ed. New York: Ncfiraw-Hili Book Company, Inc. 529-‘3. (1957). Gaudin, A. i. Elotation. 2nd ed. New York: McGraw-Hill Book Company, Inc. Bho. (1957) Gaudin, A. M. Flotation. 2nd ed. New York: McGraw-Hill Book Company, Inc. 256. (1957) Hackerman, Norman and A. C. Makrides. "Action of IOlar Organic Inhibitors," Ind. Eng. Chem. h6. 523-527. (195A). Harrison, h. D. "The commercial Operation of the Sveen- Pedersen.Saveall," T.A.P/P.I., Paper Trade Journal 109. 67-7h. (August 17, 1939). Keck, w. C., and Paul Jasberg. "A Study of the Flotation Properties of Magnetite," Trans. A.I.W.k.;., Vol. 13M- 129-b5. (1937). Keck, U. C.,and Paul Jasberg. "A Study of the Flotation Properties of Magnetite," Trans._£LILfl.M.E., Vol. 13A. 132. (1937). Newsome, J. W. Private Communicatigg. (1956). Sutherland, K. L., and I. W. dark. Principles gf Flotg; tion. Melbourne: Australasian Inst. Min. and Net. Inc. 51. (1955). 72 16. Sutherland, K. L., and I. W. Wark. Principles 2; Flota- tion. Melbourne: Australasian In t. Min. and Met. Inc. 86. (1955)- 17. Weiser, H. B. Inorganic Colloid Chemistry. New York: John Wiley and Sons, Inc., Vol. II. (1935). 18. Weiser, H. B. Inorganic Colloid Chemistry. New York: John Wiley and Sons, Inc., Vol. II. 376. (1935). \. ,- ' i. UHWU 4.: vloi text; fiflf"-’il Um:- r! I! '7". K? II I I II III '- l I II II III I I I'll, l l I l I III II» I 3 3142 7507