Hydrometallurgical treatment of tailings with high zinc content

Hydrometallurgical treatment of tailings with high zinc contentShirin Espiari a,Fereshteh Rashchi a,⁎,S.K.Sadrnezhaad ba Department of Metallurgy and Materials Engineering,University of Tehran,PO Box11365-4563,Tehran,Iranb Department of Materials Science and Engineering,Sharif University of Technology,PO Box11365-9466,Tehran,IranReceived24September2005;received in revised form15January2006;accepted23January2006Available online20March2006AbstractZinc exists as smithsonite and hemimorphite in the lead flotation tailings from the Dandi mineral processing plant in north western Iran.In this research,zinc-rich tailings produced in the Dandi plant were characterized mineralogically and a leaching study was carried out to assess the effect of several parameters on the kinetics of zinc dissolution.Parameters studied included: sulfuric acid concentration,reaction time,temperature and slurry density.It was found that leaching is controlled by a single rate-controlling step with an activation energy of23.5kJ/mol.To overcome some of the filtration problems associated with polymerization of silicic acid,lime was added as a coagulant.The optimum pH,holding time and temperature required to maximize the filtration rate were determined.©2006Elsevier B.V.All rights reserved.Keywords:Lead–zinc oxide ores;Leaching;Kinetics;Silica coagulation1.IntroductionCurrently,zinc is produced mostly from zinc sulfideores because sulfides are easy to separate from gangueand to concentrate by conventional flotation techniques.As new sphalerite mines are becoming more difficult tofind,new processes to produce zinc metal from oxidizedzinc ores are being developed.Oxidized ores containzinc in various carbonate and silicates minerals such assmithsonite(ZnCO3),hydrozincite(2ZnCO3·3Zn(OH)2),zincite(ZnO),willemite(ZnSiO4)and hemi-morphite(Zn2SiO3·H2O).Recently,Sole et al.(2005)reviewed the firstcommercial application of zinc solvent extraction for aprimary zinc ore containing∼10.6%Zn in Skorpion,Namibia.In the process an atmospheric leach in sulfuricacid is applied to remove iron,aluminum and silica fromsolution by precipitation.Zinc is then selectivelyextracted by solvent extraction.Extensive work has been carried out on the treatmentof zinc oxide ores by hydrometallurgical and pyromet-allurgical methods.In particular,there are severalprocesses for the treatment of zinc silicate ores.Leachingis the first step in the hydrometallurgical route andleaching kinetics is important from the economic point ofview.Thomas and Fray(1981)studied leaching of oxidezinc materials with chlorine and chlorine hydrate.Theyfound that the rate of leaching of the Adrar Turkish orecould be described using a shrinking core diffusionmodel,and that the rate of leaching was controlled bysurface reaction.In all cases studied,lead was alsoleached with zinc.However,iron oxides remainedvirtually undissolved.Abdel-Aal(2000)investigatedthe kinetics of sulfuric acid leaching of low-grade zincsilicate ore.Abdel-Aal found that leaching of about94%Hydrometallurgy82(2006)54–62⁎Corresponding author.E-mail address: Rashchi@ut.ac.ir (F. Rashchi).0304-386X/$-see front matter©2006Elsevier B.V.All rights reserved.doi:10.1016/j.hydromet.2006.01.005of zinc was achieved after180min of reaction with10% sulfuric acid at70°C.Diffusion through the product layer was found the controlling step during the leaching reaction.In contrast,Zhao and Stanforth(2000)studied the production of zinc powder by alkaline treatment of smithsonite ores.The best leaching results were obtained in presence of5M NaOH,at90–95°C and a reaction time of90min.Frenay(1985)studied leaching of oxidized zinc ores in various solution media and obtained the best leaching results with sulfuric acid and caustic soda.Mineralogical studies showed that smith-sonite can be completely leached but hemimorphite is relatively refractory to leaching.Matthew and Elsner(1977)developed a continuous process in which silica was dissolved first and then coagulated by controlling the pH.The ore was leached with spent electrolyte.The pH of the leach solution was then raised to pH4to5.5using a neutralizing agent to precipitate and coagulate the colloidal silica.Dufresne (1976)used a quick leach method to treat zinc silicate ores.The method was based on the chemistry of a water-starved system and effectively rejected silica from many silicate materials.Alternatively,Kumar and Biswas (1986)used aluminum sulfate for coagulation of silica gel when recovering zinc from Zawar ancient siliceous slag in sulfuric acid.They found that neutralization, coagulation and precipitation of silicic acid with Al3+ ions distinctly improved zinc recovery.Bodas(1996) developed a process for leaching zinc silicate ores from Thailand using sulfuric acid and Magnafloc156as coagulant.He found that a maximum of95%zinc leaching was obtained at a4.5M acid concentration after 3h of leaching at70°C and a1:5solid:liquid weight ratio.Bodas found that0.5g Magnafloc156was required to coagulate the silicic acid produced in a slurry containing1Kg ore.Willemite and hemimorphite are readily soluble in dilute sulfuric acid.Reactive silica dissolves according to the following reaction(Iler,1955):SiO2ðsÞþ2H2O¼SiðOHÞ4ðaqÞð1ÞMonosilicic acid(Si(OH)4)polymerizes with time to produce polysilicic acid.Polymerization involves con-densation of silanol(SiOH)groups to form siloxane (Si–O–Si)bonds.The overall equation for the poly-merization reaction may be written as follows:ðSiO2Þnþ2n H2O¼n SiðOHÞ4ð2ÞWith continued polymerization,polymers attain a range of colloidal sizes(0.001to0.1μm)depending on factors such as temperature,pH and impurities.The particles of colloidal silica may or may not be stable. Unstable particles of colloidal silica may aggregate to produce an open network structure(a gel)(Matthew and Elsner,1977).The challenge is to optimize precipitation of colloidal silica without forming an unfilterable gel and without loosing a large amount of zinc sulfate, which is present in the filtrate.Coagulation of colloidal silica helps to increase the filtration rate.Monhemius and Terry(1983)investigated the influence of temperature,pH,acid type and surface area on the kinetics of the acid dissolution of natural and synthetic willemites and natural hemimorphite.They found that dissolution of hemimorphite was diffusion-ally controlled and that dissolution of willemite was chemically controlled.The objective of this research was to study extraction of zinc from tailings produced in the Dandi plant using sulfuric acid.Mineralogical studies were carried out to identify the main phases present in the mineral and in the gangue and to identify the liberation of these phases. In the leaching experiments,the effect of acid concentration,holding time,agitation rate and solid/ liquid ratio on zinc extraction was investigated.The effect of addition of lime on the coagulation of silicic acid was also studied.2.Experimental procedure2.1.Materials/analysisThe sample was provided by the Dandi mineral processing plant in north western Iran.The feed for the Dandi mineral processing plant is currently from the Angouran mine,a large lead–zinc mine in northwestern Iran.In the Dandi plant,low grade ore is divided in two size fractions0–2and2–15mm.In the2–15mm size fraction,gangue is separated using heavy media separation techniques.After grinding,the0–2mm size fraction enters the lead flotation circuit,which produces a lead concentrate and zinc-rich oxide-silicate tailings. The sample used in this study was taken directly from the0–2mm tailings containing ca.22%zinc.These tailings are currently stockpiled.The sample was sieved and about80%of the tailings were smaller than300μm (50mesh).Mineralogical analysis of the sample was performed by X-ray diffraction(XRD),X-ray Fluorescence(XRF), polished sections and thin sections microscopy.The XRD analysis suggests that the major mineral phases in the tailing are smithsonite(ZnCO3),quartz(SiO2), followed by hemimorphite[Zn4Si2O7(OH)2·H2O],55S.Espiari et al./Hydrometallurgy82(2006)54–62calcite [CaCO 3],cerussite [PbCO 3],kaolinite [Al 2Si 2O 5(OH)4]and jarosite [KFe 3(SO 4)2(OH)6]as minor phases.The XRF results showed that the sample contained 37%ZnO,23.7%SiO 2,6.2%Al 2O 3,3.85%Fe 2O 3,2.95%CaO,3.75%PbO,0.49%SO 3,0.68%MgO,0.06%P 2O 5,and 0.81%K 2O.Sieve analysis showed that about 80%of the tailing was smaller than 300μm (50mesh).2.2.Mineralogical examinationThe tailing was divided into homogeneous,small size fractions using the riffle separator.The samples were wet screened to produce a 38–300μm size fraction (+50–400mesh).Thin sections were prepared by cementing a thin slice of the tailing to glass.A cover slip was cemented on top of tailing (Gribble and Hall,1993).Transmitted light microscopy was used to examine minerals that transmitted light in the thinsections.These minerals included most of the gangue or metallic-free minerals.A polished surface of the tailing sample was also prepared and reflected light microscopy was used to examine minerals that did not transmit light in a thin section,but reflected light to varying degrees when polished (Sarabi,2000).Microscopic analysis of the polished and thin sections confirmed the existence of the phases detected by XRD.This analysis provided useful information concerning the liberation of the minerals,especially quartz as a gangue mineral,which could complicate leaching of the tailing.Fig.1shows liberated grains of quartz and smithsonite in all size fractions of the tailing and Fig.2shows hemimorphite present as a minor phase.To identify calcite grains,thin sections were colored with red alizarin (figure is not shown).Microscopic analysis (polished and thin sections)was used to calculate the relative amounts of the components of the tailing according to the ASTM C 295norm.Smithsonite (56.4%)was the dominant phase but there was also 28.6%quartz (Table 1).2.3.LeachingIn the leaching and kinetic experiments,sulfuric acid from Merck with a purity of 95–98%and a density of 1.82g/ml was used.Calcium hydroxide [Ca(OH)2]from Merck was used as neutralizing agent.Leaching tests were carried out in a 1-l beaker.The reaction mixture was agitated and heated on a hot plate.The target temperature was controlled within ±5°C.A Buchner funnel was used for the filtration step.In the kinetic tests,70g of ore were added to 1400mL of an agitated 2M sulfuric acid solution held at a fixed temperature.The range of temperatures studied was 20to 60°C.The ore size was −200to +270mesh (53–75μm),the solid :liquid ratio was kept constant at 1:20.The optimum leaching conditions were determined as follows:100g of tailing were leached with 400mL sulfuric acid of different concentrations and stirred with a magnetic stirrer rotating at 480rpm.To determine the effect of conditioning time,400g of the sample was leached in a 2M sulfuric acid solution at 60°C for different intervals at a solid :liquid ratio of 1:4.The optimum solid :liquid ratio was determined using different amounts of the solid sample in 2MsulfuricFig.1.Liberated grains of quartz (grey)and smithsonite (white).Polished section.Reflected light.Table 1Microscopic analysis of the components of the tailing (Wt.%)Phase ZnCO 3SiO 2Wt.%56.428.6Fig.2.Grains of hemimorphite (blue).Thin section.Transmitted light.56S.Espiari et al./Hydrometallurgy 82(2006)54–62acid at 60°C with a stirring rate of 480rpm and 120min conditioning time.Finally the stirring rate was varied,while other parameters were fixed.Unless stated otherwise,these optimum conditions were used for the data presented in the figures below.From these initial experiments,the optimum leaching conditions were determined to be:2M sulfuric acid,2h leaching at 60°C,1:4solid :liquid ratio and 480rpm agitation.After leaching at the optimum conditions,a fixed quantity of lime was added to the hot slurry (50°C)and the effect of pH,time and temperature on the filtration rate (volume of filtrate per m 2h)was investigated.Finally,a thin section from the filter cake was prepared and studied under the microscope.3.Results and discussionSeveral batch leach tests were conducted to deter-mine the optimum conditions for extraction of zinc.Fig.3shows the correlation between recovery of zinc and acid concentration after leaching for 1h at 60°C.Zinc recovery increases up to 88%with acid concen-tration up to 2M and remains almost constant at higher acid concentrations.While the rate of leaching increases by increasing the concentration of the acid,this can also increase dissolution of undesirable minerals and reagent consumption (Habashi,1993).Dissolution of silica also depends on the acidity of the slurry.At pH b 2,formation of silicic acid is greater than at pH N 2(Bodas,1996).Fig.4shows the variation in recovery of zinc as a function of the leaching time.Nearly 98%Zn was extracted in 120min at 60°C in 2M acid concentration.Dissolution of zinc is a solid –liquid reaction.There-fore,the reaction time has a significant effect on the progress of the reaction.However,the extraction of zinc depends on the iron content of the sphalerite(Bodas,1996).The higher the iron content,the lower the zinc extraction in weakly acidic solutions.This is due to a coating of the smithsonite grains with iron hydroxides (Frenay,1985).The rate of leaching increases by decreasing the solid:liquid ratio and hence the excess of acid present.Fig.5shows the variation in zinc recovery as a function of pulp density under the same conditions.An increase in the liquid :solid ratio of up to 4:1produces a maximum 98%Zn recovery.It should be noted that if dilution of the slurry increases,the gelatinous silicic acid becomes hydrolyzed and crystalline SiO 2will precipi-tate (Bodas,1996).Increasing the agitation speed usually increases the rate of leaching due to mineral particles remaining suspended in the liquor and a decrease in the thickness of the mass transfer boundary layer on the surface of the particles.Hence,diffusion of liquor to the surface ofthe000.511.522.533.520406080100Molar Concentration of AcidZ i n c R e c o v e r y (%)Fig.3.Recovery of zinc as a function of the concentration of sulfuric acid after 1h at 60°C.50100150200Leaching Time (min.)Z i n c R e c o v e r y (%)Fig.4.Recovery of zinc as a function of the leaching time under standard optimum conditions.123456Liquid to Solid RatioZ i n c R e c o v e r y (%)Fig.5.Recovery of zinc as a function of liquid to solid ratio.57S.Espiari et al./Hydrometallurgy 82(2006)54–62particles increases (Vaghar,1998).Fig.6shows the recovery of zinc in 2M acid was optimum at 480rpm and there was no benefit from intense agitation with increased capital and operational expenses.It can be observed from thin section picture of the filter cake of the leach residue (not shown here)that most of the smithsonite grains dissolved during leaching,as well as hemimorphite.However,some undissolved quartz particles remained.The effect of temperature on the rate of zinc extraction in 2M acid is plotted in Fig.7.The tailing size was −200to +270mesh (53–75μm)and the solid:liquid ratio was kept constant at 1:20.Clearly,the reaction temperature has a significant impact on the rate,but about 98%Zn was extracted after 120min at 60°C.4.Kinetic modelingZinc carbonate reacts with sulfuric acid according to the following reactionZnCO 3ðs ÞþH 2SO 4ðaq Þ→ZnSO 4ðaq ÞþH 2O ðl ÞþCO 2ðg Þð3ÞTherefore,the size of the initial particle is gradually reduced as the reaction proceeds.Thus,the shrinking core model (SCM)with variable particle size can be used to describe this reaction (Smith,1981;Levenspiel,1999;Sadrnezhaad et al.,2003).An isometric spherical particle of changing size with a surface reaction or diffusion control regime can be used to model this process.4.1.Liquid film diffusion controlThe fraction of zinc reacted at any time,t ,in a film diffusion control situation can be calculated from thefollowing equation which applies to small particles moving in Stokes regime (Levenspiel,1999):ts¼1Àð1Àx Þ2=3ð4Þwhere x is the fraction of zinc reacted.The time for complete disappearance of a particle,s ,can be calculated from:s ¼q R 202bDCð5Þwhere ρis density of zinc ore,R 0is radius of the un-reacted particle,b is stoichiometric coefficient of the reaction,D is molecular diffusion coefficient of the zinc containing species and C is concentration of sulfuric acid.The above equations apply to mono-size particles.Thus,the average particle size of the solid particle can be used in themodel.406080100Agitation Rate (rpm)Z i n c R e c o v e r y (%)Fig.6.Recovery of zinc as a function of agitation rate.306090120150Time (min.)Z i n c R e c o v e r y (%)Fig.7.Fractional recovery of zinc with time at various temperatures.00.20.40.60.81306090120150Time (min.)1- (1-x )2/3Fig.8.Plot of 1−(1−x )2/3vs.time at various temperatures.58S.Espiari et al./Hydrometallurgy 82(2006)54–62Based on the experimental data shown in Fig.7,the right-hand side of Eq.(4)is plotted against time in Fig.8. The highest value of the correlation coefficient R2for linearity of the plot is obtained for the lowest temperature data(i.e.,20°C).At higher temperatures,the experi-mental data deviates further from a linear plot.None of the straight lines can be considered a close enough fit to the experimental data to suggest a film diffusion model for the leaching process.4.2.Solid product diffusion controlDiffusion of the reagent or dissolved species through a solid reaction product at any time,t,can be calculated from the following equation(Levenspiel,1999):ts¼1À3ð1ÀxÞ2=3þ2ð1ÀxÞð6ÞIn order to test the possibility of diffusion through a solid reaction product,e.g.,silica,the right-hand side of Eq.(6)is plotted against time is shown in Fig.9.Again it is obvious that the data does not fit with a straight line and it is concluded,therefore,that the diffusion through a solid reaction product could not be rate determining in the leaching system used here.4.3.Reaction controlThe fraction of zinc reacted at any time,t,in a reaction control process can be calculated from the following equation which applies to small particles moving in Stokes regime(Levenspiel,1999):ts¼1Àð1ÀxÞ1=3ð7Þwheres¼q R0bk s Cð8Þand k s is rate constant of the reaction.Based on the experimental data plotted in Fig.7,the right-hand side of Eq.(7)is plotted against reaction time in Fig.10.When Figs.8,9and10are compared,it reveals that the R2values for Fig.10are closer to1than those of Figs.8and9,especially at higher temperatures. Therefore the data can be correlated to a chemical reaction model.The rates of most heterogeneous,low-temperature reactions are controlled by adsorption/desorption pro-cesses that are necessary for the chemical reactions to proceed.This is especially true when the reaction produces gas.In leaching systems,it is rare to find that diffusion control plays a significant role.On the other hand,in a number of leaching systems,investigators find it difficult to distinguish between the two reaction models of surface reaction control(Eq.(7))and liquid film diffusion control(Eq.(4)).This is due to the small difference between the two models as pointed out by Levenspiel(1999).To confirm the dominance of a single step regime, model calculations were carried out with two and three consecutive step regimes.The standard deviation criterion and the total conversion times evaluated by model calculations confirmed the dominance of a single step regime.Thus,it is concluded that a physico-chemical desorption process consistent with the obtained activation energy[Section 4.4]governs this system (Hashemi et al.,2004;Sadrnezhaad et al.,2003).4.4.Activation energyThe temperature dependence of the reaction rate constant can be calculated by the Arrhenius equation: k s¼A expÀQRTð9Þwhere A is frequency factor,Q is activation energy of the reaction,R is universal gas constant and T is absolute temperature.The values of k s at different temperatures can be calculated from the slope of the lines shown in Fig.10and Eq.(9).The Arrhenius plot(ln k s vs.1/T),shown in Fig.11,gives an activation energy of23.5kJ/mol for the reaction from0306090120150Time (min.)1-3(1-x)2/3+2(1-x)Fig.9.Plot of1−3(1−x)2/3+2(1−x)vs.time at various temperatures.59S.Espiari et al./Hydrometallurgy82(2006)54–62the slope of the line.The typical activation energy for a chemically controlled mechanism is N 40kJ/mol,whilst the activation energy for a physical adsorption/desorption mechanism is b 40kJ/mol (Sadrnezhaad,2004;Habashi,1969).As the calculated activation energy is lower than 40kJ/mol,it can be concluded that an adsorption/desorption process is the rate controlling step.In the adsorption process,the protons must reach the surface of ZnCO 3for the reaction to proceed (Eq.(3)).As there are three products generated forming two phases,desorption of products from the surface seem to cause a greater barrier to the progress of the reaction than adsorption of the reactant.This is consistent with formation of liquid and gas phase products during the leaching process.Thus,the physico-chemical desorp-tion process appears to create the greatest barrier to the leaching reaction and desorption can be considered to be the rate controlling step.5.Coagulation stageCoagulation of silica has a significant effect on the filtration rate of the slurry and has been considered to be due to several mechanisms.At instant collision of the particles,Van der Waals attraction draws two particles together and causes coagulation.However,the hydra-tion barrier layer or the electrostatic repulsion forces between the similarly charged particles can retard coagulation (Iler,1979).Allen and Matijevic (1969)proposed that with every cation adsorbed,the silica surface loses one silanol site for hydrogen bonding with water,and in that sense is “dehydrated ”.Increasing the “lyophobic ”character renders the solution more sensitive to coagulation by the electrolyte.Another mechanism is bridging by cations such as calcium ions.In this mechanism there could be a mutual matching of the positively charged calcium sites on one particle with negatively charged SiO-sites on a colliding particle and vice versa —with the calcium ions actingas2.933.13.2 3.3 3.4 3.510-3/T (K -1)l n KFig.11.Arrhenius plot for leaching of zinc oxidetailing.23456pHF i l t r a t i o n R a t e (m 3/m 2.h )Fig.12.Effect of pH of coagulation stage on filtration rate.1306090120150Time (min.)1-(1-x )1/3Fig.10.Plotof 1−(1−x )1/3vs.time at various temperatures.050100150200Time (min.)F i l t r a t i o n R a t e (m 3/m 2.h )Fig.13.Effect of time in coagulation stage on filtration rate.60S.Espiari et al./Hydrometallurgy 82(2006)54–62the bridging units.There may be an appreciable area of inter-particle contact,with a number of Si/O −+Ca +−O/Si bridges formed between particles (Iler,1975).Addition of electrolytes lowers the zeta potential (Iler,1979),which reduces the charge of particles.If the charge reduces beyond a certain point,the particles come together,where the presumed Van der Waals attraction between them could overcome the ionic repulsion (Iler,1979).At the coagulation stage,the major factors affecting coagulation of the colloidal silica and pulp filtration rate are pH,temperature and holding time.The impact of pH on the pulp filtration rate is shown in Fig.12.By increasing the pH,the pulp filtration rate increases to a maximum of 2.3(m 3/m 2h)at pH 5.3,after 3h holding time at 50°C.Zinc hydroxide precipitates by increasing pH above 5.5(Jackson,1986).However,this depends on the temperature and concentration of zinc.Under the conditions used,zinc does not precipitate although some is adsorbed onto the residue.After leaching,the zinc concentration in solution was determined to be ca.1M.Calculations using the solubility product of zinc hydroxide (K sp =7×10−17)at pH 5.3,confirms that zinc does not precipitate as zinc hydroxide.Similarly,zinc does not precipitate as zinc carbonate at pH 5.3(K sp =1.2×10−10).At this pH,coagulation is presumably due to dehydration of the surface of silica and bridging by calcium ions.Since the point of zero charge (pzc)of silica is ca.2and the silica surface is negatively charged at pH N 2,Van der Waals attraction between the particles could not result in coagulation (Allen and Matijevic,1969).Coagulation involves a kinetic step,therefore,the rate of coagulation increases with time and temperature.The impact of holding time on the filtration rate is shown in Fig.13.The maximum filtration rate of 2.3(m 3/m 2h)wasobtained after 120min coagulation at a pH of 5.3.Increasing the holding time above 2h did not generally improve the filterability of the pulp.The effect of temperature on the pulp filtration rate was also investigated and the results are shown in Fig.14.Filtration rate increases by increasing the temperature up to 75°C.The maximum filtration rate measured was 4.1(m 3/m 2h),at 75°C and pH 5.3.However,it should be noted that the optimum coagula-tion temperature is a function of the reactivity of the neutralizing agent (Matthew and Elsner,1977).Never-theless,as can be seen from Figs.12,13and 14,pH has a more significant effect on coagulation of colloidal silica than either temperature or holding time.6.Conclusions1.A maximum zinc recovery of 98%was obtained by leaching with 2M sulfuric acid for 2h at 60°C with 1:4solid :liquid ratio agitated at 480rpm.2.Analysis of the experimental data indicates that leaching of zinc carbonate consists of a single rate controlling step.The rate controlling step used to describe this reaction is a physico-chemical desorp-tion process with activation energy of ca.23.5kJ/mol.3.For the silica coagulation stage,lime was added to raise the pH and a maximum filtration rate of ca.4.1(m 3/m 2h)was obtained at pH5.3,2h holding time and 75°C.AcknowledgementsThe authors would like to thank University of Tehran for supporting this work (Grant No.615/4/1051).ReferencesAbdel-Aal,E.A.,2000.Kinetics of sulfuric acid leaching of low-gradezinc silicate ore.Hydrometallurgy 55,247–254.Allen,L.H.,Matijevic,E.,1969.Colloid and Interface Science 31,287.Bodas,M.G.,1996.Hydrometallurgical treatment of zinc silicate orefrom Thailand.Hydrometallurgy 40,37–49.Dufresne,R.E.,1976.Quick leach of siliceous zinc ore.Journal ofMetals 28,8–12.Frenay,J.,1985.Leaching of oxidized zinc ore in various media.Hydrometallurgy 15,243–253.Gribble,C.D.,Hall,A.J.,1993.Optical Mineralogy:Principles andPractice.Chapman and Hall,London.Habashi,F.,1969.Principles of Extractive Metallurgy,vol.1.Gordonand Breach,New York.Habashi,F.,1993.A Textbook of val University,Quebec City,Canada.Hashemi,B.,Moghimi,Z.,Nemati,Z.A.,Sadrnezhaad,S.K.,2004.Determination of kinetic parameters of MgO –C.FourthIntl.Temperature (°C)F i l t r a t i o n R a t e (m 3/m 2.h )Fig.14.Effect of temperature of coagulation stage on filtration rate.61S.Espiari et al./Hydrometallurgy 82(2006)54–62。