《高等选矿学》课程教学资源（文献资料）Fine Particle Processing - Separation methods in mineral processing. Fine particles in mineral processing. Flotation Entrainment

1 Fine Particle Processing Separation methods in mineral processing. Fine particles in mineral processing. Flotation Entrainment. SEPARATION METHODS IN MINERAL PROCESSING Gravity Magnetic Electric Phys1cochem1cal Concentration Separatfon Separation Separation in Dense in Water in A1r Media Figure 1. Classification of separation methods. Flotation is the most important separation method in the group of the physicochemical methods. It is based on differences in surface properties of treated minerals while the other methods are based on differences in bulk properties such as density, magnetic susceptibility, etc. Figure 2. Illustration of the flotation separation process in which hydrophobic particles(the particles which attach to bubbles)are separated from the hydrophilic particles(the particles which do not attach to bubbles)

1 Fine Particle Particle Particle Particle Processing Processing Processing Processing Separation Separation Separation Separation methods methods methods methodsin mineral mineral mineral mineral processing. processing. processing. processing. Fine particles particles particles particlesin mineral mineral mineral mineral processing. processing. processing. processing. Flotation Flotation Flotation Flotation. Entrainment Entrainment Entrainment Entrainment. Figure 1. Classification of separation methods. Flotation is the most important separation method in the group of the physicochemical methods. It is based on differences in surface properties of treated minerals while the other methods are based on differences in bulk properties such as density, magnetic susceptibility, etc. Figure 2. Illustration of the flotation separation process in which hydrophobic particles (the particles which attach to bubbles) are separated from the hydrophilic particles (the particles which do not attach to bubbles)

2The separation results strongly depend on particle size and all methods can handle onlyparticles ofagiven sizerange.Figure 3. Optimal particle size ranges for different operationsUnit operations in the mineral processing plant can broadly be divided into four distinctgroups:comminution (that also includes classification),separation (beneficiation, concentration),dewatering of the beneficiation products, and water clarification (Fig. 4).COMMINUTIONCONCENTRATIONWATERCLARIFICATIONPRODUCTDEWATERINGFig.4.Unit operations in mineral processingplant circuits

2 The separation results strongly depend on particle size and all methods can handle only particles of a given size range. Figure 3. Optimal particle size ranges for different operations. Unit operations in the mineral processing plant can broadly be divided into four distinct groups: comminution (that also includes classification), separation (beneficiation, concentration), dewatering of the beneficiation products, and water clarification (Fig. 4). Fig. 4. Unit operations in mineral processing plant circuits

3The separation processescanonlyyield satisfactoryresults if mineral particles in thefeedareliberated.This isachievedbysizereductionoftherun-of-minerockdowntotheliberationsize. In the lowgrade ores -which are now mined -the valuable minerals are disseminated in theform of very fine grains and thus to achieve liberation the rock must by crushed and grounddown to this liberation size (Fig. 5).Figure 5. Liberation size of the grains of valuable mineral.Flotation,however,cannotefficientlydeal withveryfineparticles.10090hei80700040ChalcociteTime1minFrotherPPG400O-KEXOKE0m01000AverageParticleSize(um)Figure 6.Response ofchalcociteto collector addition (potassium ethyl xanthate)(afterTrahar,1981)

3 The separation processes can only yield satisfactory results if mineral particles in the feed are liberated. This is achieved by size reduction of the run-of-mine rock down to the liberation size. In the low grade ores - which are now mined - the valuable minerals are disseminated in the form of very fine grains and thus to achieve liberation the rock must by crushed and ground down to this liberation size (Fig. 5). Figure 5. Liberation size of the grains of valuable mineral. Flotation, however, cannot efficiently deal with very fine particles. Figure 6. Response of chalcocite to collector addition (potassium ethyl xanthate) (after Trahar, 1981)

4esrr604020C51020501005002Averageparticlesize(um)Figure 7. Recovery of siderite as a function of particle size. Recovery in the presence of frotherbutwithoutcollector()isassumedtobebyentrainmentonlywhereasrecoverywithcollectorand forther () is by entrainment and true flotation."The difference curve (dotted line)estimates recovery due to the “"true flotation" only. (after Warren and Trahar, 1981)Whileparticles in the-100+20um sizerange floatmostlybya trueflotationmechanism, fine particles can also be transported to a froth product by entrainment. Therefore,flotation of very finely ground flotation feeds is not only difficult and slow it is also veryunselective(1)P:P=probability of flotation,P,=probability of particle-to-bubble collisionPa=probability of particle-to-bubble attachment, P,=probability of the formation of stableparticle-bubble aggregate.The first factor, the probability of collision, is purely hydrodynamic,it depends onparticle and bubble sizes and hydrodynamic conditions prevailing in the flotation cell. For veryfine particles the probability of collision is very small.The Ps, is entirely determined by the wettability of the mineral particles and also by their densityIt is, therefore, proportional to the contact angle, the larger is hydrophobicity (the larger is thecontact angle) the more likely it is that the formed particle-bubble aggregate will be stableenough to withstand shearing forces in the flotation cell

4 Figure 7. Recovery of siderite as a function of particle size. Recovery in the presence of frother but without collector (●) is assumed to be by entrainment only whereas recovery with collector and forther (■) is by entrainment and “true flotation.” The difference curve (dotted line) estimates recovery due to the “true flotation” only. (after Warren and Trahar, 1981). While particles in the -100 + 20 μm size range float mostly by a true flotation mechanism, fine particles can also be transported to a froth product by entrainment. Therefore, flotation of very finely ground flotation feeds is not only difficult and slow it is also very unselective. P = (1) P = probability of flotation, Pc = probability of particle-to-bubble collision, Pa = probability of particle-to-bubble attachment, Ps = probability of the formation of stable particle-bubble aggregate. The first factor, the probability of collision, is purely hydrodynamic, it depends on particle and bubble sizes and hydrodynamic conditions prevailing in the flotation cell. For very fine particles the probability of collision is very small. The Ps, is entirely determined by the wettability of the mineral particles and also by their density. It is, therefore, proportional to the contact angle, the larger is hydrophobicity (the larger is the contact angle) the more likely it is that the formed particle-bubble aggregate will be stable enough to withstand shearing forces in the flotation cell

ROTHHASEOCCPHASESLURRYHydrophobicParticles图HydrophilicParticlesFigure 8.Simplified description of flotation kinetics:(1)transfer from the slurry to the froth ofhydrophobic particles (true flotation); (2) transfer from the froth cell over the cell lip; (3) drop-back from the froth to the slurry and (4) entrainment of gangue particles (after Laplante et al.,1989),Fig.9.Schematicillustrationoffrothdrainage.Filledsymbolsrepresenthydrophobic particles, open symbolshydrophilicparticles.It is well established that the stability of the froth critically affects the grade of the frothproduct. If the froth is too stable there is no additional upgrading in the froth.Since entrainment results from the water which is carried on to the forth layer along with all fineparticle (also gangue), the upgrading must be related to the froth stability and drainage whichremoves hydrophilic particles back to the pulp

5 Figure 8. Simplified description of flotation kinetics: (1) transfer from the slurry to the froth of hydrophobic particles (true flotation); (2) transfer from the froth cell over the cell lip ; (3) drop￾back from the froth to the slurry and (4) entrainment of gangue particles (after Laplante et al., 1989). It is well established that the stability of the froth critically affects the grade of the froth product. If the froth is too stable there is no additional upgrading in the froth. Since entrainment results from the water which is carried on to the forth layer along with all fine particle (also gangue), the upgrading must be related to the froth stability and drainage which removes hydrophilic particles back to the pulp. Fig.9. Schematic illustration of froth drainage. Filled symbols represent hydrophobic particles, open symbols hydrophilic particles

6Fig. 10. Additional upgrading in the flotationTNFROTHfroth.βistheincreasedmetalcontentSresultingfromtheupgradingprocesstakingtplroplaceinthefroth(possibleonlyinahealthyfroth,thatisthefrothwhichisnottoostable)Metal conteutPure liquids do not foam. For a liquid to foam, it must be able to form a shell around thegas bubble that opposes the thinning of the lamellae. Foaming does not occur in pure liquidsbecause there exist no such mechanism (Kitchener and Cooper,1959).Since the foamsexemplified by those generated in the flotation frother solutions areunstable and practically canexistonlyduringbubbling,thesearedynamicnon-equilibriumsystems.Thestabilityof suchasystem results from a dynamic balance between destructive and stabilizing forces.The difficulties inherent in giving a comprehensive scientific analysis of flotation frotherswereindepthanalyzedbyWrobel60yearsago/Wrobel,19531.Thesituation60yearslaterisnot that different and the terms “powerful" and “"selective"are still commonly used to describethe properties of these flotation agents. The frothers that are purchased for commercial useusually come along with the information exemplified by Table 1.Table 1.Flotation forther characteristics as provided by manufacturersPropertyFrother1Frother2Frother3200250400Molecularweight71222Viscosity,cP0.9700.9800.988Density,g/cm3below-50below-50below-50Freez point, 325250285Flashpoint,oFWhile the information provided in Table 1 is important for handling these products itdoes not say anything about theirflotation properties. Some manufacturers,therefore,provide some additional qualitative information in which these products may befurther characterized as “selective" or “powerful". So, what we - who have to use these products-do?Well,we develop a research program and screen the acquired products following somegeneral guidelines which may vary depending on the school.Inthefundamental studiesonflotationfotherstherearemanyunknowns,andoneknownfact. It is well accepted that pure liquids do not foam. When surface active molecules are present,however,theiradsorptionatthegas/liquidinterfaceservestoretardthelossofliquidfromthelamellae and to produce a moremechanically stable system.This directly leads to a simpleconclusionthatrelatesfrother activityto its surfacetension.However,intheconcentration

6 Pure liquids do not foam. For a liquid to foam, it must be able to form a shell around the gas bubble that opposes the thinning of the lamellae. Foaming does not occur in pure liquids because there exist no such mechanism (Kitchener and Cooper, 1959). Since the foams exemplified by those generated in the flotation frother solutions are unstable and practically can exist only during bubbling, these are dynamic non-equilibrium systems. The stability of such a system results from a dynamic balance between destructive and stabilizing forces. The difficulties inherent in giving a comprehensive scientific analysis of flotation frothers were in depth analyzed by Wrobel 60 years ago [Wrobel, 1953]. The situation 60 years later is not that different and the terms “powerful” and “selective” are still commonly used to describe the properties of these flotation agents. The frothers that are purchased for commercial use usually come along with the information exemplified by Table 1. Table 1. Flotation Flotation Flotation Flotation forther forther forther forther characteristics characteristics characteristics characteristics as provided provided provided provided by manufacturers manufacturers manufacturers manufacturers Property Frother 1 Frother 2 Frother 3 Molecular weight Viscosity, cP Density, g/cm3 Freez point, oC Flashpoint, oF 200 7 0.970 below –50 250 250 12 0.980 below –50 285 400 22 0.988 below -50 325 While the information provided in Table 1 is important for handling these products it does not say anything about their flotation properties. Some manufacturers, therefore, provide some additional qualitative information in which these products may be further characterized as “selective” or “powerful”. So, what we - who have to use these products - do? Well, we develop a research program and screen the acquired products following some general guidelines which may vary depending on the school. In the fundamental studies on flotation fothers there are many unknowns, and one known fact. It is well accepted that pure liquids do not foam. When surface active molecules are present, however, their adsorption at the gas/liquid interface serves to retard the loss of liquid from the lamellae and to produce a more mechanically stable system. This directly leads to a simple conclusion that relates frother activity to its surface tension. However, in the concentration Fig. 10. Additional upgrading in the flotation froth. ∆β is the increased metal content resulting from the upgradingprocess taking place in the froth (possible only in a healthy froth, that isthe froth which is nottoo stable)

7ranges in which frothers affect foaming and bubble size the water surface tension is practicallythat of water (Sweet et al., 1997)2.020sroanaarinonHexanol(rt)1.515Hexanol(size)1.010Haxano(S.T.)MIBC(size)0.5MIBCMIBC(S.T.)(rt)o0110410310210810510"3Concentration,moldmFig.11.Normalizedretentiontime(foamability),bubblediameterand surfacetensionof n-hexanol and MIBC aqueous solutions [C. Sweet, J. van Googstraten, M. Harris andJ.S. Laskowski, 1997]As Fig. 11 demonstrates, the surface tension is the least sensitive to the frotherconcentration, both bubble size and retention time are affected by the frother at theconcentrationsatwhichthesurfacetensionvaluesarestill identical tothatof distilledwaterRecent results prove without any doubt that frothers control the size of bubbles by decreasingbubblecoalescence(Cho&Laskowski,Int.J.Min.Proc.,vol.64(2002),andthatthebubblecoalescence can be entirely prevented at frother concentrations exceeding the critical coalescenceconcentration (CCC)Figure 12 shows surface tension isotherms for a few selected frothers. As this figuresdemonstrates the two common frothers, MIBC and DF- 2oo, affect surface tensions of aqueoussolutions very little

7 ranges in which frothers affect foaming and bubble size the water surface tension is practically that of water (Sweet et al., 1997). Fig. 11. Normalized retention time (foamability), bubble diameter and surface tension of n-hexanol and MIBC aqueous solutions [C. Sweet, J. van Googstraten, M. Harris and J.S. Laskowski, 1997]. As Fig. 11 demonstrates, the surface tension is the least sensitive to the frother concentration; both bubble size and retention time are affected by the frother at the concentrations at which the surface tension values are still identical to that of distilled water. Recent results prove without any doubt that frothers control the size of bubbles by decreasing bubble coalescence (Cho & Laskowski, Int.J.Min.Proc Int.J.Min.Proc Int.J.Min.Proc Int.J.Min.Proc., vol. 64 (2002), and that the bubble coalescence can be entirely prevented at frother concentrations exceeding the critical coalescence concentration (CCC). Figure 12 shows surface tension isotherms for a few selected frothers. As this figures demonstrates the two common frothers, MIBC and DF- 200, affect surface tensions of aqueous solutions very little

81E-0050.00010.0010.01LLI117575LIL中中00000-80中中70 70?b福TTT中国田 6565电-LL-()saE 6060E5555F5050口DF-200eunsLα-Terpineol045E 45LIII间DF-1012E4040+DiacetoneAlcoholL1OMIBCE3535LI3030TTTTTT1E-0050.00010.0010.01Concentration (mol/dm3)Figure12.Surfacetension isothermsforafew selected frothersInourtestsontheeffectoffrothersonbubblesizeweusedaUCTBubbleSizeMeter(Fig. 13). In thee tests bubbles were generated using a single hole sparger, Figure 14 shows theeffect of frother concentration when bubbles were generated using a bronzeplate with a singleholeeitherof0.10mmor0.15mmdiameter

8 1E-005 0.0001 0.001 0.01 Concentration (m ol/dm3) 30 35 40 45 50 55 60 65 70 75 Surface Tension (m N/m) 30 35 40 45 50 55 60 65 70 75 1E-005 0.0001 0.001 0.01 DF-200 α−Terpineol DF- 1012 Diacetone Alcohol M IBC Figure 12. Surface tension isotherms for a few selected frothers. In our tests on the effect of frothers on bubble size we used a UCT Bubble Size Meter (Fig. 13). In thee tests bubbles were generated using a single hole sparger. Figure 14 shows the effect of frother concentration when bubbles were generated using a bronze plate with a single hole either of 0.10 mm or 0.15 mm diameter

9Detector UnitPhoto-OdetectoroUnitGasBuretteMPUCapillaryTimerBubbleSizer·?BubbleDe-TankbubblerCoalescenceFunnelAirSpargerDanpingPotnIBM-PCSyntron Lapping-PolishingMachinePeristalticWaterPumpReservoirFigure 13. UCT bubble size measuring set-up.F

9 Figure 13. UCT bubble size measuring set-up. F

102.4+HEX+MIBCBubbles generatedfrom+DEH*DEMPH2.20.15mmholesparger(uu)-MPDEX2.01.81.6e1.4Bubblesgeneratedfrom1.20.10mmholesparger1.00.000020.000040.000060.000080.00010.000120.000140Concentration(mol/L)Figure2.Effectoffrotherconcentrationonbubble sizeusingabronze single-hole sparger.+DF200-DF250-DF1012xMIBC)oCCCforDF2oo1.5forMIBC0.5CCcforCODF1012DF250o0.10.20.30.4Concentration(mmol/L)Figure 15.Effect of frother concentration on bubbe size measured in an open-top Leeds flotationcell (S. Cho and J.S. Laskowski, CanJ.Chem.Eng., 80, 299- 305 (2002).Asourresultsdemonstratewhileintheexperimentswithsinglebubbleswhichcannotcollide with other bubbles, at frother concentrations typical for the froth flotation systems, there

10 0 0.1 0.2 0.3 0.4 Concentration (mmol/L) 0 0.5 1 1.5 2 2.5 SauterDiameter(mm) DF200 DF250 DF1012 MIBC CCC for DF200 CCC for DF250 CCC for DF1012 CCC for MIBC Figure 15. Effect of frother concentration on bubbe size measured in an open-top Leeds flotation cell (S. Cho and J.S. Laskowski, Can.J.Chem.Eng., 80, 299- 305 (2002). As our results demonstrate while in the experiments with single bubbles which can not collide with other bubbles, at frother concentrations typical for the froth flotation systems, there