当前位置：文档库 › Effect of mechanical and chemical clay removals by hydrocyclone

Effect of mechanical and chemical clay removals by hydrocyclone

Effect of mechanical and chemical clay removals by

hydrocyclone and dispersants on coal ?otation

William J.Oats,Orhan Ozdemir,Anh V.Nguyen *

School of Chemical Engineering,The University of Queensland Brisbane,Queensland 4072,Australia

a r t i c l e i n f o Article history:

Received 26July 2009

Accepted 9December 2009

Available online 13January 2010Keywords:Coal ?otation Slime Clay

Dispersants Hydrocyclone Colloid stability DLVO forces

a b s t r a c t

Fine minerals,mostly clays,are known to have a detrimental effect on coal ?otation.This paper focuses on the effect of mechanical and chemical removals of ?ne minerals by hydrocyclone and dispersants on coal ?otation.The experimental results showed that the ?otation recovery slightly increased from med-ium acidic to medium alkaline ranges.The ?otation experiments carried out with dispersants at different dosages showed that the dispersants did not enhance the ?otation recovery signi?cantly.However,the removal of the ?ne fraction from the feed using a hydrocyclone signi?cantly increased the ?otation recovery.The bubble–particle attachment tests also indicated that the attachment time between an air bubble and the coal particles increased in the presence of clay particles.These attachment time results clearly showed that the clay particles adversely affected the ?otation of coal particles by covering the coal surfaces which reduced the ef?ciency of bubble–coal attachment.An analysis based on the colloid stabil-ity theory showed that the clay coating was governed by the van der Waals attraction and that the dou-ble-layer interaction played a secondary role.It was also concluded that the best way to increase the ?otation recovery in the presence of clays was to remove these ?ne minerals by mechanical means such as hydrocylones.

1.Introduction

Fine particles less than 10l m coat the surface of many valuable minerals,making the particles hydrophilic and preventing the adsorption of ?otation collectors onto the mineral surfaces,which hinders mineral separation.These ?ne particles also consume ?o-tation reagents and increase the operational costs.

There have been many studies conducted to investigate the negative effects caused by ?ne particles (Arnold and Aplan,1986a;Burdon et al.,1976;Celik et al.,1998,2002;Jowett et al.,1956;Mishra,1978;Quast et al.,2008;Xu et al.,2003).These stud-ies showed that the slime coating of the mineral surface occurs due to electrostatic attraction between opposite charged coarse and ?ne particles.For example,while there was no attraction between the same charged ?ne and coarse galena particles,there was a sig-ni?cant loss on ?otation recovery when oppositely charged ?nes were present in the system (Gaudin et al.,1960).It is explained that the ?ne clay particles are electrostatically attracted to the sur-face of coal particles and the attachment is heavily dependent on the magnitude and sign of the zeta potentials of the coal particles and the ?nes (Arnold and Aplan,1986a;Xu et al.,2003).The effect of clay minerals on boron ?otation was also speci?cally studied

(Celik et al.,2002).The studies showed that while boron minerals (colemanite,ulexite,etc.)?oated with both anionic and cationic surfactants in the absence of clay,even as little as 1%of clay addi-tion reduced the boron ?otation recovery considerably (Celik et al.,2002).The effect of ?ne clay particles on several coal samples from more hydrophobic to moderate hydrophobic was also the subject of many studies (Jowett et al.,1956;Burdon et al.,1976;Mishra,1978;Arnold and Aplan,1986a,b;Quast et al.,2008).The results con?rmed that the presence of ?ne particles decreased the recov-ery and rate of coal ?otation.

Dispersants are widely used to modify the colloidal interactions between particles by creating electrostatic and/or steric repulsion which counterbalances the contribution of the van der Waals attraction to the total net force.These dispersants are mostly anio-nic polymers and adsorb on mineral surfaces,making them more negative (Huynh et al.,2000).Classi?cation methods such as cy-clones are more economical and ef?cient for separating ?ne parti-cles from the coarse minerals (Salter and Childs,1984;Greet and Smart,1997).Removing ?ne particles from coal samples by sieving or desliming with a hyrocyclone improved the coal ?otation per-formance (Quast et al.,2008).The ultrasonic treatment to disperse the clay particles from boron minerals also increased the ?otation recovery of the minerals and reduced the required collector dosage (Celik et al.,1998).In addition,some studies showed that tabling should also be used for removing the ?ne particles (Burdon et al.,1976).

*Corresponding author.

E-mail address:Anh.Nguyen@https://www.wendangku.net/doc/c7964046.html,.au (A.V.Nguyen).

Minerals Engineering 23(2010)413–419

Contents lists available at ScienceDirect

Minerals Engineering

journal homepage:www.else v i e r.c o m /l o ca t e /m i n e n g

In the coal industry,clay particles have also a signi?cant effect on ?ltration (Jowett et al.,1956;Burdon et al.,1976;Mishra,1978;Arnold and Aplan,1986a,b;Quast et al.,2008).The clay minerals such as kaolinite,illite,and montmorillonite can have different charge distributions and signs on the surfaces and edges,and can cover the surface of coal particles in different ways,depending on pH and coal type and liberation.As a result,Arnold and Aplan showed that while one type of clay mineral such as bentonite sig-ni?cantly depresses the coal ?otation,clays such as kaolinite or il-lite do not show any effect on the coal ?otation (Arnold and Aplan,1986a ).

The aim of this study was to investigate the effect of slime par-ticles on the ?otation of coal samples from BHP Billiton Mitsubishi Alliance (BMA).Speci?cally,the research focused on comparing the ef?ciency of slime removal using hydrocyclone and slime disper-sion by chemicals on the coal ?otation.The combined effect of pH and dispersants on the coal ?otation was also studied.Finally,the effect of ?ne clay particles on the bubble–particle attachment was investigated using the attachment time measurements to identify the physics underlying the effect of clays on coal ?otation.

2.Materials and methods 2.1.Materials

The coking coal samples,moderately hydrophobic,used for the ?otation tests were obtained from a BHP Billiton Mitsubishi Alli-ance mine in Central Queensland,Australia.The ?otation experi-ments were carried out with à0.5mm size fraction.Fig.1shows the particle size distribution of the ?otation feed.In addition,the

ash content of the sample with respect to particle size was also ob-tained (Fig.1).As shown in Fig.1,the ash content of the sample slightly increases with decreasing particle size.In particular,the ash content sharply increases for the à150l m fractions.It is important to note that the ash content of à38l m fraction (mostly clay minerals)is very high,about 53%.The removal of the ?ne fraction from the feed signi?cantly reduces the ash content of the feed.

Diesel and methylisobutylcarbinol (MIBC)were used as the col-lector and frother,respectively.The diesel dosage was 100g/t and MIBC dosage was 15ppm for all experiments.Reagent grade so-dium silicate (Sigma–Aldrich,Australia)and sodium hexameta-phosphate (Fisher Scienti?c,Australia)were also used as the dispersants.Analytical grade NaOH and HCl were used to control pH.

2.2.Flotation experiments

The ?otation experiments were carried out with à0.5mm size fraction using a small laboratory Agitair ?otation cell (2.5L).For each ?otation experiment,250g of a coal sample was added into the cell and mixed with tap water at 900rpm for 5min without re-agent addition.During this conditioning step,the acid or base was added to adjust the solution pH.After 5min conditioning,the de-sired amount of diesel and MIBC was added and mixed for 3min.Finally,air was introduced at a ?ow rate of 10L/min into the cell and froth was collected at 1,3,5,and 8min.The impeller speed was kept at 600rpm.Feed solid concentration was 10%by weight.In the case of ?otation experiments with dispersants,the only dif-ference from the previous procedure was that the dispersants were added into the suspension after the initial 5min conditioning,and the suspension was mixed for an additional 5min.All experiments with dispersants were carried out at natural pH ($8).Concentrates and tailing were ?ltered,dried at 80°C,and weighed for further processing and analysis.For the ash analysis,an about 10g of dried sample from each product was ?rst ground using a mortar and pes-tle.Then,approximately 2g of a ground sample from each product was burned in an oven at 815°C for 2h.The ash left over was weighed to calculate the ash content.2.3.Hydrocyclone desliming

In order to remove the ?ne fraction from the feed,the hydrocy-clone separation was conducted using an AKW hydrocyclone (AKW Apparate +Verfahren GmbH,Germany).As known from the stan-dard literature,there are many factors affecting the cyclone ef?-ciency such as the apex diameter,vortex diameter,feed percent solids,feed rate etc.However,since the project aimed to remove the ?ne particles as much as possible after easy adjustments,we only carried out some cyclone tests at different vortex and apex diameters at constant feed percent solids and feed rate.The testing was conducted at a solids percentage of 10%and a ?ow rate of 3L/s to keep the results consistent.The combinations investigated are shown in Table 1.As seen from the table,four different combina-tions of vortex ?nder and apex were used to ?nd the optimum conditions.

Under?ow and over?ow products for the separations were col-lected at the same time during the tests.After the separation,the particle size distributions of the under?ow and over?ow products were determined using a Master Sizer 2000(Malvern,UK).The data obtained from the results are presented in Table 2.The aim of this separation was to maximize the rejection of à38l m sample to the over?ow stream.Based on these results,the 15mm diame-ter spigot and the 29mm vortex ?nder were chosen to produce the sample for ?otation tests due to obtained particle cut size.2.4.Bubble–particle attachment experiments

The bubble–particle attachment tests were carried out with coal samples containing ?ne fraction,mostly clays,in order to see the effect of these particles on the bubble–particle attachment time.The coal sample (212–150l m)suspension was prepared for the

Table 1

Hydrocyclone test combinations.

Test #Vortex ?nder diameter (mm)Apex diameter (mm)1291522910344104

414W.J.Oats et al./Minerals Engineering 23(2010)413–419

tests(1g in100mL distilled water).In addition,the clay suspen-sions were prepared from the suspension containingà38l m frac-tions.The solid percentage was20%by weight.After mixing the sample for30min,the clay suspension was kept for a while in or-der to allow the coarse particles to settle down.Then,a small

amount of the suspension was added into the coal suspension to obtain the required clay concentration.The suspensions were ini-tially very turbid and were not clear enough for optically visualiz-ing the particle attachment in the bubble–particle attachment tests.Therefore,the attachment experiments were carried out after the sedimentation of clay minerals overnight using the attachment timer device,which was described in a previous paper(Ozdemir et al.,2009).

3.Results and discussions

3.1.Effect of pH on?otation

pH is one of the most important factors in controlling slime coating due to electrostatic attraction between the clays and coal particles.For this reason,in the?rst set of experiments,the?ota-tion of coal was carried out as a function of pH.Shown in Fig.2are the recovery results plotted against suspension pH.The coal recov-ery was lower than70%and increased with increasing suspension pH.

3.2.Effect of dispersants on?otation

Prior to the?otation tests with hydrocyclone-sized samples,the effect of the polymeric dispersants including sodium hexameta-phosphate(HMP)and sodium silicate(SS)on the coal?otation was investigated.The results are shown in Fig.3.As seen from the?gure,the use of the dispersants slightly increased the?otation recovery.The effect of sodium silicate as dispersant on coal?ota-tion recovery was a slightly better than sodium hexametaphos-phate.Overall,the dispersants could not increase the?otation recovery signi?cantly.

3.3.Effect of hydrocyclone desliming

The results for coal?otation carried out with coal sample des-limed using the hydrocyclone are shown in Fig.4.The results show an increased recovery of94.1%with an ash content of14.02%.Evi-dently,the removal of?nes from the feed signi?cantly increased the?otation recovery.Included in Fig.4are the previous?otation results obtained with the dispersants.The comparison between the ?otation recoveries obtained by the mechanical and chemical rem-ovals clearly shows that the best way to eliminate the detrimental effect of?ne clay particles on the coal?otation is to deslime the feed using a hydrocyclone.

3.4.Bubble–particle attachment tests

The ef?ciency of?otation critically depends on the ability of air bubbles to collect hydrophobic particles from the suspension(Ngu-yen and Schulze,2004;Binks and Horozov,2006;Wilson,2007). The slime coating of the coal surfaces considerably in?uences the coal?otation performance as seen from the previous results.In or-der to provide a better understanding of the slime effect on coal

Table2

Results of the hydrocyclone experiments.

Vortex/apex

diameter

Over?ow Under?ow

d10 (l m)d50

(l m)

d90

(l m)

d10

(l m)

d50

(l m)

d90

(l m)

29/10 1.515.276.498.5290.0566.4 33/10 1.614.071.6102.2268.7535.5 44/10 1.819.3102.4126.8318.1598.5 29/15 1.610.241.262.2246.9536.9

W.J.Oats et al./Minerals Engineering23(2010)413–419415

?otation,the bubble–particle attachment tests were carried out with coal particles in the absence and presence of the ?nes.The re-sults are presented in Fig.5.As seen from Fig.5,the contact time for coal particles was found to be shorter than 10ms (100%attach-ment)in the absence of the ?nes.However,the experiment done with the coal particles containing 0.1%?ne fraction showed that the attachment percentage of the coal on the bubble at 10ms was about 80%.

Fig.6shows the distinctive difference in attachment of coal par-ticles onto the bubble surface in the absence and presence of ?ne clay particles which normally adhere to the coal particle surface.The experimental results strongly suggest that the slime coating signi?cantly prevents the attachment of the coal particles to the bubble,and hence reduces the coal ?otation recovery.3.5.Analysis of clay coating using forces of colloidal interaction In order to gain a better understanding of clay coating on coal particles,an analysis of the coating using the Derjaguin–Landau–

Verwey–Overbeek (DLVO)theory of the aggregate stability of col-loidal particles was completed using forces of colloidal interaction (Deryagin and Landau,1941;Verwey and Overbeek,1948;Adam-son and Gast,1997).The DLVO theory includes the van der Waals and electrical double-layer forces.These forces are brie?y de-scribed below.

The van der Waals force,F v dW ,is determined using the Lifshitz approach based on quantum physics.The effects of electrolyte-screening and electromagnetic retardation are accounted for by the advanced theory,which gives (Nguyen and Schulze,2004)

F v dW

?à

1d àA eh ;j T ?1àA h

t1dA

e1T

where h is the shortest separation distance between the surfaces.In

Eq.(1),the effective radius for the particles with radii,R 1and R 2,is de?ned as R ?R 1R 2=eR 1tR 2T.The Hamaker–Lifshitz function,A ,is de?ned as

A eh ;j T?e1t2j h Te à2j h A 0tA 1eh T

e2T

where j is the Debye constant.For water at 25°C,j =3.288I ,where j is measured in nm à1and the ionic strength,I ,is measured in mol/L.The zero-frequency term of the Hamaker–Lifshitz function,A 0,is a function of the static dielectric constants of the particles and water,and can be described as

A 0?3k

B T X 1

m ?1m

à380àe 1e 180àe 2

e 2

m e3T

where k B is the Boltzmann constant and T is the absolute tempera-ture,e 1is the static dielectric constant of coal ($2.5)and e 2is the

static dielectric constant of clay ($2for layered silicates).The in?-nite sum is equal to 0.89.The non-zero-frequency term,A 1,which accounts for the electromagnetic retardation effect that follows from the limited velocity of propagation of the electromagnetic waves,is described as

A 1eh T?3 h x 8???

p eB 1àB 3TeB 2àB 3T12I 2eh T????????????????B 2tB 3p àI 1eh T

????????????????B 1tB 3p

e4T

where h is the Planck constant (divided by 2p ),x is the absorption

frequency in the UV ($2?1016rad/s for water).B 1and B 2are equal to square of the refractive index of coal (1.78)particles (Foster and Howarth,1968;Goodarzi and Murchison,1973;Im and Ahluwalia,1993)and clay (1.38)particles (Zhang et al.,2005;Friedrich et al.,2008;Ravi Kumar et al.,2008).For water,B 3?1:887.The functions

80100

A t t a c h m e n t p e r c e n t a

g e (%)

Bubble-particle contact time (ms)

Coal without fine clay particles

Coal with fine clay particles (0.1% by weight)

Fig.5.Bubble–particle attachment versus contact time for coal particles in the absence and presence of ?ne clay particles as measured by the attachment timer device.

Fig.6.The attachment of the coal particles onto air bubbles in the absence (left picture)and presence of (0.1%)?ne particles (clay)particles (right picture)under the same conditions.The left bubble surface is covered by the attached coal particles,while the presence of ?ne clay particles make the suspension on the right picture very turbid and cover the coal surface,making the particles dif?cult to attach to the bubble surface.

416W.J.Oats et al./Minerals Engineering 23(2010)413–419

I1and I2in Eq.(4)which account for the electromagnetic retarda-tion are given as

I iehT?f1teh=k iTq gà1=qe5Twhere q=1.185.The characteristic wavelengths,k i,are measured in units of length as

k j?

p2x

????????????????????????

B3eB jtB3T

;e6T

where v is the speed of light in vacuum.

The electrical double-layer force can be predicted from the solu-tion of the Poisson–Boltzmann equation with the boundary condi-tion at constant surface potential or constant surface charge (Nguyen and Schulze,2004).The double-layer force,F w,obtained with the boundary conditions at constant surface potential is de-scribed by

F w?2p ee0j

expej hTàew1T2àew2T2

expe2j hTà1

e7T

where e0is the permittivity of vacuum,and e is the relative permit-tivity(the dielectric constant)of the solution(e=80for water).w1 and w

are the surface(zeta)potentials of the particles.The dou-ble-layer force,F r,obtained with the boundary conditions at con-stant surface charge is described by

F r?2p ee0j

expej hTtew1T2tew2T2

expe2j hTà1

e8T

The surface(zeta)potential of the particles were measured by the micro electrophoresis using an ZetaSizer Nano-ZS(Malvern, UK).The zeta potential of the bubbles was determined using a zeta meter(Rank Brothers,UK)using a procedure available in the liter-ature(Jameson and Kubota,1993).The electrolyte concentration of about1mM was similar to the laboratory?otation conditions.The results of the zeta potentials are shown in Fig.7.

The DLVO theory predicts the interaction force by the sum of the van der Waals and double-layer forces described by Eqs.(1) and(7)for the interaction at constant surface potential or Eqs.

(1)and(8)for the interaction at constant surface charge.Shown in Fig.8are the results obtained for the colloidal forces under the normal conditions of unadjusted pH=8of the suspensions. As can be seen from Fig.8,the double-layer forces for the two sur-face charging mechanisms are repulsive and hence the double-layer interaction under the surface charge regulation occurring be-tween the two extreme cases is also repulsive.On the contrary,the van der Waals force is attractive at all separation distances.In par-ticular,at short separation distance,the van der Waals attraction is so strong that the double-layer repulsion is overcome,resulting in strong net DLVO attraction between the particles at close contact. It is concluded that the van der Waals attraction is the driving force for the clay coating.The analysis is carried out for spherical parti-cles for simplicity and can be extended to the real coal and clay particles which are rough and non-spherical.The extension will re-sult in correction factors but the principal conclusion will not be changed.

3.6.General discussion

The available literature indicates that?ne clay particles attach to the negatively charged coal particles and increase the ash con-tent of the product(Laskowski and Par?tt,1989).The clay surfaces exhibit negative charges whereas the clay edges exhibit positive charge(Iwasaki1962).The coal particles have an overall positive surface charge at low pH,and a negative charge at high pH(Aplan, 1976).Since the increase of coal recovery with pH is weak as shown in Fig.2,it is reasonable to assume that the double-layer interaction between the coal and clay particles play a secondary role in the clay coating:The double-layer interaction only reduces the repulsion between the particles with decreasing pH but cannot change from the repulsive to attractive interactions which are responsible for clay removal from and coating onto the coal sur-face.Indeed the force analysis described in Section 3.5clearly shows that the van der Waals attraction between the coal and clay particles is responsible to the clay coating.Since the van der Waals attraction is almost independent of pH,it is not possible to change the van der Waals attraction to repulsion to remove the clay from the coal surface.

The study by Arnold and Aplan(1986b)also showed that coal ?otation in the absence of clay was affected little by pH.However, the coal?otation in the presence of clays such as kaolinite,illite and bentonite showed different effects on the coal?otation.For example,kaolinite and illite did not show any signi?cant effect on coal recovery.The presence of bentonite in the system resulted in a considerable decrease in recovery.It was also observed that although some coals were highly hydrophobic and?oated very well at pH=6–9even in the presence of bentonite,but at pH=3.5the recovery drastically decreased.It was concluded that the clay coating on coal surfaces depended on the suspension pH and the type of clay.A similar observation was also noticed re-cently(Quast et al.,2008).It is possible that kaolinite was domi-

-2

-1

0510152025

(

)

Separationdistance, h(nm)

van der Waals

EDL at const.surf

sur.charge

EDL at const.surf

sur.potential

DLVO at const.surf

sur f.charge

DLVO at const.surf

sur.potential

Fig.8.Colloidal(DLVO)forces between a coal particle and a clay particle versus

inter-surface separation distance,normalized by dividing by the effective radius,

R?R1R2=eR1tR2T,where R1and R2are the particle radii.

W.J.Oats et al./Minerals Engineering23(2010)413–419417

nantly present in the coal sample used in this study.Otherwise,the ?otation recovery of the original coal sample would be signi?-cantly in?uenced by pH.Indeed,the SEM (Scanning Electron Microscopy)and XDS (X-ray Dispersive Spectroscopy)analysis shows that the coal sample dominantly contains kaolinite (Fig.9).Polymers are mostly used to prevent the slime coating on the coal surface.It was thought that if the positive charge on the edge of clays was reduced or negative charges on the surface of clays in-creased,both the coal and clay surfaces would be negatively (aver-age)charged.Therefore,this negative charge for coal and clays would increase the repulsion force between the clay and coal par-ticles and would reduce the clay coating on the coal surface.How-ever,the results from the ?otation experiments with polymers showed that the ?otation could not be improved,re-con?rming that the double-layer interaction only play a secondary role in the clay coating.There are some studies showing that the anionic phosphates may be used to remove slime coatings from valuable minerals (Mishra,1978;Huynh et al.,2000).For example,Mishra et al.showed that desliming the feed at 53l m and conditioning with KH 2PO 4signi?cantly increased the ?otation recovery.On the other hand,Arnold and Aplan showed that the addition of so-

dium hexametaphosphate and lignin sulphonate and sodium sili-cate as dispersants depressed coal ?otation (Arnold and Aplan,1986a ).Evidently,in addition to the surface charge,the coal hydro-phobicity can play an important role in the clay coating and removal.

Our results show that the ?ne clay minerals can be eliminated from the coal sample using a hydrocyclone.The test work showed that the deslimed sample signi?cantly increased the ?otation recovery.On average,the total recovery was 25%higher than the recovery of the raw sample without the desliming.However,the ash content of the concentrate was still relatively high (Table 3

Fig.9.SEM back scattered electron image (left)of a coal particle associated with a ?ne clay particle (#2)and the element spectrum (right)of the clay particle (Section 2)analysed using XDS.The small C peak indicates a small amount of coal present at this section.The high peaks of oxygen,aluminium and silicon show the signi?cant presence of kaolinite clay.

Table 3

Summary of ?nal ?otation experimental results.Conditions

Ash (%)Recovery (%)Raw sample at pH =9.5

9.7268.9Raw sample with sodium hexametaphosphate (1000g/t)

10.0966.3Raw sample with sodium silicate (1000g/t)

10.5369.0Deslimed sample using a hydrocyclone (natural pH)

14.02

94.1

418W.J.Oats et al./Minerals Engineering 23(2010)413–419

probably due to the middling particles with the locked ash miner-als.The entrainment should be ruled out since the?otation time was kept constant for all?otation experiments.

4.Conclusions

The effect of?ne particles on coal?otation was investigated. The?otation experiments were carried out with the raw coal sam-ples and the deslimed coal samples to examine the effect of mechanical and chemical removals of?ne clay minerals on coal ?otation.The?otation experiments showed that the?otation recovery of the raw coal samples slightly increased with increasing pH and that sodium silicate and sodium hexametaphosphate used as the dispersants did not enhance the?otation recovery signi?-cantly.However,the removal of the?ne fraction from the feed using a hydrocyclone signi?cantly increased the?otation recovery. The bubble–particle attachment tests with coal particles also indi-cated that?ne clay particles increased the attachment time be-tween an air bubble and the coal particles and signi?cantly reduced the ef?ciency of bubble–coal attachment and the coal?o-tation recovery.The clay coating was analyzed based on the colloid stability theory which incorporated the van der Waals and double-layer interactions.Micro electrophoresis was used to obtain the coal and clay surface potentials required in calculating the dou-ble-layer interaction force.The van der Waals interaction force was calculated using the refractive index of the coal and clay par-ticles,and water.The theoretical analysis showed that the clay coating was governed by the van der Waals attraction and that the double-layer interaction played a secondary role.It was also concluded that the best way to increase the?otation recovery in the presence of clays is to remove the?ne clay minerals by a hydrocyclone.

Acknowledgements

The authors gratefully acknowledge the Australian Research Council for?nancial support through a Discovery grant and BHP Billiton Mitsubishi Alliance(BMA)for funding the BMA Chair of Minerals Processing at the University of Queensland and the coal samples.The authors would like to thank Dr.Marc A.Hampton for his assistance in obtaining the SEM images of the coal samples. References

Adamson,A.W.,Gast,A.P.,1997.Physical Chemistry of Surfaces.New York,John Wiley&Sons,Inc.

Aplan,F.F.,1976.Coal?otation.In:Fuerstenau,M.C.(Ed.),Flotation:M.C.Gaudin Memorial Volume.AIME,New York,pp.1235–1264.

Arnold,B.J.,Aplan,F.F.,1986a.The effect of clay slimes on coal?otation,part I:the nature of the clay.International Journal of Mineral Processing17(3–4),225–242.Arnold,B.J.,Aplan,F.F.,1986b.The effect of clay slimes on coal?otation,part II:the role of water quality.International Journal of Mineral Processing17(3–4),243–260.

Binks, B.P.,Horozov,T.S.(Eds.),2006.Colloidal Particles at Liquid Interfaces.

Cambridge University Press,Cambridge.

Burdon,R.G.,Booth,R.W.&Mishra,S.K.(1976).Factors in?uencing the selection of processes for the bene?ciation of?ne coal.In:Proceedings of the Seventh International Coal Preparation Congress25,Australia.

Celik,M.S.,Elma,I.,Hancer,M.,Miller,J.D.,1998.Effect of in-situ ultrasonic treatment on the?oatability of slime coated colemanite.Innovations Miner.

Coal Process.In:Proceedings of the International Mineral Processing7th Symposium,pp.153–157.

Celik,M.S.,Hancer,M.,Miller,J.D.,2002.Flotation chemistry of boron minerals.

Journal of Colloid and Interface Science256(1),121–131.

Deryagin,B.,Landau,L.,1941.Theory of the stability of strongly charged lyophobic sols and of the adhesion of strongly charged particles in solutions of electrolytes.Acta Physicochimica URSS14,633–662.

Foster,P.J.,Howarth, C.R.,1968.Optical constants of carbons and coals in the infrared.Carbon(Oxford)6(5),719–729.

Friedrich,F.,Steudel,A.,Weidler,P.G.,2008.Change of the refractive index of illite particles by reduction of the Fe content of the octahedral sheet.Clay and Clay Minerals56(5),505–510.

Gaudin,A.M.,Fuerstenau,D.W.,Miaw,H.L.,1960.Slime-coatings in galena?otation.

CIM Bulletin53,960–963.

Goodarzi,F.,Murchison,D.G.,1973.Oxidized vitrinites–their aromaticity,optical properties and possible detection.Fuel52,90–92.

Greet,C.J.,Smart,R.S.C.,1997.The effect of size separation by cyclosizing and sedimentation/decantation on mineral surfaces.Minerals Engineering10(9), 995–1011.

Huynh,L.,Feiler,A.,Michelmore,A.,Ralston,J.,Jenkins,P.,2000.Control of slime coatings by the use of anionic phosphates:a fundamental study.Minerals Engineering13(10–11),1059–1069.

Im,K.H.,Ahluwalia,R.K.,1993.Radiation properties of coal combustion products.

International Journal of Heat Mass Transfer36,293–302.

Jameson,G.J.,Kubota,K.,1993.A study of the electrophoretic mobility of a very small inert gas bubble suspended in aqueous inorganic electrolyte and cationic surfactant solutions.Journal of Chemical Engineering of Japan26(1), 7–12.

Jowett,A.,El-Sinbawy,H.,Smith,H.G.,1956.Slime coatings of coal in?otation pulps.Fuel35,303–309.

Laskowski,J.S.,Par?tt,G.D.,1989.Electrokinetics of coal–water suspensions.

Surfactant Sci.Ser.,32(Interfacial Phenom.Coal Technol.),pp.279–327. Mishra,S.K.(1978).The slime problem in Australian coal?otation.In:Australasian

I.M.M.,Mill Operators Conference.Mt Isa:pp.159–168.

Nguyen,A.V.,Schulze,H.J.,2004.Colloidal Science of Flotation.New York,Marcel Dekker.

Ozdemir,O.,Taran,E.,Hampton,M.A.,Karakashev,S.I.,Nguyen,A.V.,2009.Surface chemistry aspects of coal?otation in bore water.International Journal of Mineral Processing92(3–4),177–183.

Quast,K.,Ding,L.,Fornasiero,D.Ralston,J.,2008.Effect of slime clay particles on coal?otation.In:Proceedings of Chemeca2008,Newcastle,September28–October1,Australia.

Ravi Kumar,N.V.N.,Muralidhar,K.,Joshi,Y.M.,2008.On the refractive index of ageing dispersions of Laponite.Applied Clay Science42(1–2),326–330. Salter,P.,Childs,G.,1984.Application and operation of?ne cyclone circuits.In: Proceedings of the2nd International Conference on Hydrocyclones,Bath,BHRA. Verwey,E.J.W.,Overbeek,J.T.G.,1948.Theory of the Stability of Lyophobic Colloids.

Amsterdam,Elsevier.

Wilson,I.D.(Ed.),2007.Encyclopedia of Separation Science.Elsevier,Amsterdam. Xu,Z.,Liu,J.,Choung,J.W.,Zhou,Z.,2003.Electrokinetic study of clay interactions with coal in?otation.International Journal of Mineral Processing68(1–4),183–196.

Zhang,Y.-H.,Dang,Z.-M.,Xin,J.H.,Daoud,W.A.,Ji,J.-H.,Liu,Y.,Fei,B.,Li,Y.,Wu,J., Yang,S.,Li,L.-F.,2005.Dielectric properties of polyimide–mica hybrid?lms.

Macromolecular Rapid Communications26(18),1473–1477.

W.J.Oats et al./Minerals Engineering23(2010)413–419419