Minerals Engineering 24 (2011) 449–454

Contents lists available at ScienceDirect

Minerals Engineering

journal homepage: www.elsevier .com/locate /mineng

The effect of acrylamide-co-vinylpyrrolidinone copolymer on the depressionof talc in mixed nickel mineral flotation

Andy Leung, James Wiltshire, Anton Blencowe, Qiang Fu, David H. Solomon, Greg G. Qiao ⇑Polymer Science Group, Department of Chemical and Biomolecular Engineering, The University of Melbourne, Parkville, Victoria 3010, Australia

a r t i c l e i n f o

Article history:Received 18 October 2010Accepted 15 December 2010Available online 8 January 2011

Keywords:Sulphide oresFlotation depressantsFroth flotationIndustrial mineralsMineral processing

0892-6875/$ - see front matter � 2011 Elsevier Ltd. Adoi:10.1016/j.mineng.2010.12.010

⇑ Corresponding author. Tel.: +61 3 83448665.E-mail address: gregghq@unimelb.edu.au (G.G. Qia

a b s t r a c t

Copolymers of acrylamide and vinylpyrrolidinone with varying compositions have been synthesised andemployed to depress talc in a model flotation system with process plant operation conditions. Adsorptionisotherms indicated that the hydrophilic acrylamide homopolymer has a very low affinity for the hydro-phobic talc surface, whereas vinylpyrrolidinone homopolymer strongly adsorbs onto the talc surface.Micro-flotation experiments revealed that the copolymer system can induce stronger talc depressionthan the homopolymer variants, with the most effective copolymer depressant having 25–30% vinyl-pyrrolidinone incorporation. The copolymer system is observed to have inherited the strong talc affinityof vinylpyrrolidinone polymer and the strong hydrophilic property from polyacrylamide. This combinedeffect facilitates the desired strong talc depression in single mineral flotation. However, this copolymersystem has similar adsorption affinity on both the talc and pentlandite, hence depressing both talc andpentlandite in the mixed mineral flotation system. This research shows that a sufficient hydrophobic bal-ance on the polymer is necessary for the adsorption and subsequent depression for talc. However, poly-mer with high adsorption selectivity is required to be a successful synthetic talc depressant for mixedmineral system.

� 2011 Elsevier Ltd. All rights reserved.

1. Introduction

Talc is a common gangue mineral found closely associated withmany complex valuable mineral ores. The natural hydrophobicityof talc means that it tends to be collected with the valuable miner-als during the froth flotation process. As talc is essentially a mag-nesium silicate mineral (Deer et al., 1992), high talc content inflotation concentrates causes complications during the down-stream smelting process and imposes financial penalties uponthe concentrate seller. The depression of talc in flotation processeshas therefore been the focus of a great deal of research.

Guar gum is one of the most widely used industrial depressantsfor hydrophobic talceous gangue minerals in flotation of nickelbearing ores. However, guar is a natural product and is subject toseasonal variation. Thus, there is a lack of control over its micro-structural properties such as the galactose:mannose ratio and dis-tribution, a problem which can lead to fluctuations in depressantperformance.

Other polymers have been used to depress talc in flotation. Thebasic concept is to employ polymers with particular functionalgroups to render the gangue mineral hydrophilic and hence prevent

ll rights reserved.

o).

the flotation of the gangue mineral (Chen et al., 2003). Previous re-searches in the field have focused on the use of polysaccharidessuch as dextrin (Liu et al., 2000; Rath et al., 1997; Sedeva et al.,2010) and carboxymethyl cellulose (Burdukova et al., 2008;Cuba-Chiem et al., 2008; Khraisheha et al., 2005; Shortridge et al.,2000), as well as polyacrylamides (Beattie et al., 2006a,b; Chiemet al., 2006; Morris et al., 2002) as synthetic depressants for talcwith limited success.

This paper presents an approach in employing a copolymer ofacrylamide (Am) and vinylpyrrolidinone (VPD) as a syntheticdepressant in froth flotation system. The concept of this syntheticdepressant is to develop a polymer with balanced hydrophobic/hydrophilic nature to overcome the natural hydrophobicity of talcand hence to prevent it from attaching to the air bubbles in thefroth flotation process. However, the acrylamide homopolymerdoes not adsorb strongly onto talc and hence has a weak talcdepression because of the opposing hydrophobicities of the poly-mer (hydrophilic) and talc (hydrophobic). This paper investigatesa solution to this problem by incorporation of less hydrophilicmonomer into the acrylamide polymer, thus providing anchorpoints for the hydrophilic copolymer to adsorb onto the hydropho-bic surface of talc and to reduce its floatability. Micro-flotation andlaboratory-scale flotation devices, as well as isotherm adsorptionexperiments, have been employed for the study of flotation andadsorption selectivity respectively in this research.

Table 1Characteristics of the synthetic depressant.

Polymer % Ama % VPDa Mw (Da)b PDIb

P(Am) 100 0 306 500 3.14P(Am-co-VPD5%) 95 5 423 100 3.10P(Am-co-VPD8%) 92 8 507 700 2.29P(Am-co-VPD11%) 89 11 423 300 1.57P(Am-co-VPD25%) 75 25 624 900 2.58P(Am-co-VPD37%) 63 37 316 200 2.45P(Am-co-VPD50%) 50 50 328 800 1.94P(Am-co-VPD63%) 37 63 496 500 1.67P(Am-co-VPD75%) 25 75 241 700 1.36P(Am-co-VPD90%) 10 90 353 000 1.35P(VPD) 0 100 247 000 1.11

a Am and VPD composition measured by 1H-NMR.b Molecular weight (Mw) and polydispersity (PDI) measured by GPC.

450 A. Leung et al. / Minerals Engineering 24 (2011) 449–454

2. Experimental

2.1. Materials

Talc sample for single mineral flotation was sourced from theMt. Fitton mine site in South Australia with XRF analysis showingmore than 99% purity. The sample was milled and wet-screened,before being collected for use in the flotation study. The volumeweighted mean diameter, D[4,3], was 234 lm as determined usinga Malvern Mastersizer and the BET surface area was measured as2.29 m2/g. Brucite sample used in the single mineral flotationwas sourced from CSIRO with a volume weighted mean diameter,D[4,3], of 81 lm and a BET surface area of 2.94 m2/g.

Talc sample used in the adsorption studies was purchased fromUnimin Australia (>95% purity) with a volume weighted meandiameter, D[4,3], of 30 lm and a BET surface area of 3.69 m2/g. Thistalc sample was not use in flotation experimentation since itsparticle size is too fine for froth flotation.

Pentlandite sample used in the adsorption studies was sourcedfrom CSIRO with a volume weighted mean diameter, D[4,3], of107 lm and a BET surface area of 1.06 m2/g.

For the mixed mineral experiments, high grade sample of talcwas obtained from Steetley Minerals and was crushed to pass asieve mesh size of 1.65 mm and screened to reject the materialpassing 200 lm. The pentlandite was concentrated from a highgrade sample from Kambalda, WA. Non-sulphides were first re-jected by hand sorting and the upgraded portion was then crushedto pass a sieve mesh size of 700 lm. The �300 lm fraction, whichwas low grade, was rejected and the +300 lm fraction was splitinto closely sized fractions. These size fractions were upgradedindividually in a laboratory magnetic separator to reject pyrrhotite.

Polyfroth H57 (Huntsman), acrylamide (Am) (Merck, 98%), 1-vi-nyl-2-pyrrolidone (VPD) (Sigma–Aldrich, P99%), 2,20-azobis(2-methylpropionamidine) dihydrochloride (AMPA) (Sigma–Aldrich,97%), NaCl (Chem-Supply, 99%), CaCl2 (Merck, P99%), KCl (AjaxChemicals, 99%), NaHCO3 (Sigma–Aldrich, P99%), MgSO4�7H2O(Sigma–Aldrich, P99%), guar (Sigma–Aldrich) and sodium ethylxanthate (SEX) (CSIRO) were used as received.

2.2. General procedure for synthesis of P(Am-co-VPD) copolymers

Copolymers of Am and VPD (see Fig. 1) were synthesized bycopolymerizing Am and VPD in water at 100 �C under argon for2 h using AMPA as initiator. The composition of the final copolymerwas controlled by varying the initial Am to VPD monomer ratio (seeTable 1). The resulting polymers were precipitated into acetone,collected via filtration and dried under vacuum and stored in a des-iccator at room temperature. The corresponding VPD compositionof each copolymer was evaluated by 1H-NMR spectroscopy.

For example; Am (12.0 g, 168.8 mmol), VPD (6.0 mL, 56.3 mmol)and AMPA (2.0 g, 7.5 mmol) were dissolved in MilliQ water(270 mL, 18 MX cm) and heated at 100 �C under an argon atmo-sphere for 2 h. After cooling to room temperature the solutionwas diluted with water (130 mL) and precipitated into acetone

Fig. 1. Structure of the P(Am-co-VPD) copolymer.

(4 L) with rapid stirring. The resulting precipitate was collectedby filtration and dried in vacuo for 48 h to afford a fluffy white so-lid, 18.7 g (92%); 1H-NMR spectroscopic analysis provided a VPDcomposition of 25%; GPC: Mw = 624 900, PDI = 2.58.

2.3. Polymer molecular weight and composition measurements

The molecular weight characteristics of the polymers weredetermined via size exclusion chromatography performed on aShimadzu liquid chromatography system fitted with a Wyattmini-DAWN TREOS laser light scattering detector (690 nm,30 mW) and a Shimadzu RID-10A differential refractometer(690 nm), and using three phenogel columns in series (500, 104

and 106 Å porosity, 5 lm bead size). An aqueous solution (50 mMNaNO3, 0.02% w/v of NaN3) was used as the eluent at a flowrateof 1 mL/min and with the column temperature set at 50 �C.

The corresponding composition of each polymer was analysedusing proton nuclear magnetic resonance (1H-NMR) spectroscopyon a Varian Unity + 400 in D2O at 400 MHz.

2.4. Hallimond tube flotation tests

Micro-flotation tests were performed using a modified Halli-mond tube. Hyper saline solution used was 0.825 M NaCl,0.011 M CaCl2, 0.028 M KCl, 0.003 M NaHCO3 and 0.099 M MgSO4

in distilled water. 20 mL of the hyper saline solution was addedto 2 g of the Mt. Fitton talc in a 28 mL polypropylene container.The pH was adjusted to pH 4.5 with the addition of 80 lL of0.25% v/v H2SO4. Then, 0.75 mL of 2 mg/mL depressant preparedin distilled water was added into the talc sample to make a finaldepressant concentration of 750 g of depressant per tonne of min-eral (750 g/t). The talc sample was conditioned by mixing on arotator for 10 min. For the flotation experiment, the talc samplewas then transferred to the Hallimond tube followed by additionof the top-up solution, which consisted of 0.001% v/v of H57 frotherin 200 mL of hyper saline solution at pH 4.5. The talc sample wasagitated by a custom made mechanical impeller at a stirring rateof 500 rpm and high purity nitrogen gas was delivered at the rateof 10 mL per minute. After a set flotation time of 3, 5 and 8 min, theconcentrate and the tailing were separately collected, filtered byWhatman Grade 541 filter paper, dried and weighed. The recover-ies are expressed on a weight basis. The experimental error wasmeasured to be ±2.5%.

2.5. Isotherm adsorption studies

Preliminary time-dependent adsorption experiments indicatedthat maximum adsorption was attained within 15 min. For alladsorption tests shown in this study, the time of equilibration

A. Leung et al. / Minerals Engineering 24 (2011) 449–454 451

was fixed at 60 min. Five grams of mineral samples were weighedseparately into 50 mL centrifuge tubes and made up to 15 mL byadding polymer solution with known concentration (ranging from0 to 2000 ppm) prepared in distilled water. The mineral suspen-sions were then agitated for 60 min at 1400 rpm using a magneticstirrer at room temperature and the samples were monitoredcontinuously to ensure no sedimentation had occurred due toinadequate mixing. After equilibration the suspensions were cen-trifuged for 1 min at 4400 rpm using an Eppendorf 5702 centrifuge.The supernatant liquid of each sample was filtered by WhatmanGrade 541 filter paper and the residual polymer concentration inthe supernatant was determined via a Shimadzu TOC-VCSH totalorganic carbon analyser. It was assumed that the amount of poly-mer depleted from solution is adsorbed onto the talc surface.

All polymers exhibited relatively high affinity isotherms, andhence can be analysed using the Langmuir expression for adsorp-tion at the solid–water interface (Hunter, 2001):

h ¼ Cads

Cmads

¼ K:Ceq

55:5þ K:Ceq¼ b:Ceq

1þ b:Ceq

where Cads is the adsorbed amount (mg/m2), Cmads is the plateau ad-

sorbed amount (mg/m2), Ceq is the polymer equilibrium solutionconcentration (ppm), K is the adsorption equilibrium constant andb is the Langmuir affinity constant (ppm�1). This equation can berearranged to allow a straight line fit to data plotted as Ceq/Cads ver-sus Ceq:

Ceq

Cads¼ 1

Cmads:b

þ Ceq

Cmads

The plateau adsorbed amount and the Langmuir affinity con-stant can be determined in this expression. The affinity constantcan be used to determine the equilibrium constant, K,(K = b � 55.5, where 55.5 mol/L is the concentration of water for di-lute solutions). All parameters derived from the Langmuir analysisare quoted with error bars as determined from the curve fitting er-ror analysis (which takes into account errors in the individual datapoints) and propagation of errors.

2.6. Mixed mineral flotation

Mixed mineral flotation experiments of talc, quartz and pent-landite were conducted on a modified 3 L Denver style cell (Guy,1992; Senior et al., 2009). For all mixed mineral flotation tests,50 g of equal amounts of talc and pentlandite and 450 g of quartzwere mixed with synthetic saline water and ground for 20 min ina ceramic mill with ceramic balls at the natural pH at 67% solidsby weight.

In the flotation cell, the impeller was driven from below to al-low the whole surface of the froth to be scraped with a paddle atconstant depth and time intervals. The cell was fitted with a rubberdiaphragm, sight tube and electronic sensor for automatic detec-tion and control of pulp level. With this cell design no air is en-trained during conditioning, unlike most commercial cells, andthe flotation gas only enters the pulp when the gas on/off controlis activated. The gas flow rate was 8 L/min. The pH was monitoredcontinuously during testing and was measured with a glass/calo-mel electrode.

For all mixed mineral tests, the ball-mill ground slurry wastransferred to the flotation cell, the water level was raised, andthe pH adjusted to the test value. The SEX collector and talc depres-sant were added sequentially and the pulp was conditioned for7 min. For both conditioning and flotation the impeller speedwas 1200 rpm. Automatic titrators were used to deliver dilutesolutions of sulphuric acid to maintain the pH at the test value.Frother was added continually at the rate of 6 mg/min starting

1 min before flotation. Concentrates were collected at 0.5, 1, 2, 4and 8 min. The scraping rate for flotation was once every 2 s forthe first minute and once every 5 s thereafter. The concentrateand the tailing were separately collected, filtered, dried andweighed. The magnesium (Mg) and nickel (Ni) content from talcand pentlandite respectively were analyzed by inductively coupledplasma (ICP) mass spectrometry. The experimental error was mea-sured to be ±5%.

3. Results and discussions

3.1. Flotation studies

The effectiveness of the P(Am-co-VPD) copolymers was evalu-ated by the modified Hallimond tube, which is a widely used smallscale flotation device for the study of interaction between additivesand minerals in a flotation system.

Initial studies on the modified Hallimond tube had shown thatthe flotation characteristic of mineral particles in the absence ofdepressant was different in the hyper saline water system (withfrother and pH adjustment) comparing to a pure water system.In the hyper saline water system, both the naturally hydrophilic(brucite) and hydrophobic (talc) mineral particles transported tothe froth phase much faster under the same operation conditions(identical air flowrate and impeller stirring speed) in the modifiedHallimond tube than with the original column (Fig. 2). Talc recov-ery with the hyper saline water system reached a maximumplateau (�90%) in 2 min, whereas the recovery in distilled watertook � 8 min to reach the same level. Also, the maximum recoveryof brucite jumped from a value of less then 20% in distilled water in5 min to 50% in the hyper saline water. Introduction of the hypersaline water and frother in the flotation system significantly in-creased the recovery of the hydrophilic mineral in the concentrateof the modified Hallimond tube system. This recovered hydrophilicmineral is caused by entrainment from the raising air bubble and isnot true flotation. This high level of entrainment was reduced byincreasing the column height of the modified Hallimond tube by8 cm, which significantly reduced the maximum carryover of thebrucite down to 20% while still having a fast recovery of talc(�90% in 2 min). All subsequent flotation experiments with thehomopolymers and P(Am-co-VPD) copolymers were performedon the height extended Hallimond tube system.

Results of the depressant effect of the P(Am-co-VPD) copoly-mers and guar using the modified Hallimond tube are summarizedin Fig. 3. Typically talc recovery using guar was about 40% in 5 min.The P(Am) homopolymer was shown to have basically no depres-sant effect on talc as the mineral recovery reached the maximumof �90% in 3 min, similar to the talc recovery without depressant.On the other hand, the P(VPD) homopolymer was capable of reduc-ing the talc recovery to between 20% and 40% over the period of3–8 min. This depressant effect is slightly stronger than guar.

Interesting results were observed when applying the P(Am-co-VPD) copolymer to the talc flotation system. As shown in Fig. 3,P(Am-co-VPD5%) was shown to considerably reduce the rate of talcrecovery. Talc recovery in 3 min was reduced to around 40%, yetthe amount of talc recovery still reached the maximum (�90%)in 8 min. With higher VPD composition incorporated on the acryl-amide polymer, the talc recovery was drastically reduced. With11% VPD composition, the copolymer started to show a strongerreduction on talc recovery (�18% in 5 min) than guar (�40% in5 min). The copolymer attained the strongest depression of talc(7–15% between 3 and 8 min) at � 25–30% VPD composition. Fur-ther increment of the VDP component in the acrylamide copolymerreduced the effectiveness of the polymer depressant. Nevertheless,the P(Am-co-VPD) copolymer with VPD composition between 25%

Flotation Time (mins)0 2 4 6 8

Min

eral

Rec

over

y (%

)

0

20

40

60

80

100

[Original Column] Talc (Distilled Water)[Original Column] Brucite (Distilled Water)[Original Column] Talc (H. Saline)[Original Column] Brucite (H. Saline)[Extended Column] Talc (H. Saline)[Extended Column] Brucite (H. Saline)

Original Column- Distilled Water

Original Column- H. Saline

Extended Column- H. Saline

Fig. 2. Effect of synthetic process plant water and Hallimond tube column height on the flotation of talc and brucite in the absence of depressant.

VPD % in P(Am-VPD) copolymer0 20 40 60 80 100

Min

eral

Rec

over

y %

0

20

40

60

80

100

3 mins5 mins8 mins

Pure P(VPD)

Pure P(Am)

Guar 5 mins–

Fig. 3. Talc recovery in hyper saline water with different P(Am-co-VPD) copolymersand Guar.

Fig. 4. Adsorption isotherms of P(Am), P(VPD), different P(Am-co-VPD) copolymersand guar on talc in distilled water. Insert: Langmuir affinity constants (b) of P(Am),P(VPD) and different P(Am-co-VPD) copolymers on talc.

Table 2Langmuir affinity constants (b) and plateau adsorbed amounts (Cm

ads) obtained byfitting the adsorption isotherm of different P(Am-co-VPD) copolymers and guar ontalc to the Langmuir model.

Depressant type Talc

b (ppm�1) Cmads (mg/m2) R2

P(Am) 0.0045 ± 0.0002 1.2783 ± 0.0291 0.99P(Am-co-VPD05%) 0.0546 ± 0.0243 1.4385 ± 0.2252 0.87P(Am-co-VPD25%) 0.0474 ± 0.0274 1.0083 ± 0.1659 0.83P(Am-co-VPD50%) 0.0392 ± 0.0105 0.8191 ± 0.0495 0.97P(Am-co-VPD75%) 0.0378 ± 0.0175 0.8972 ± 0.1147 0.91P(Am-co-VPD90%) 0.2599 ± 0.0992 0.6879 ± 0.0476 0.95P(VPD) 0.2152 ± 0.1290 1.0997 ± 0.1897 0.82Guar 0.1134 ± 0.0233 0.3571 ± 0.0181 0.98

452 A. Leung et al. / Minerals Engineering 24 (2011) 449–454

and 90% all had a lower talc recovery than guar and the P(VPD)homopolymer.

3.2. Adsorption studies

Detailed adsorption experiments were carried out to study theadsorption of the synthetic polymers on talc. The adsorption iso-therms of the different P(Am-co-VPD) copolymers and homopoly-mers on talc are shown in Fig. 4. The corresponding Langmuiradsorption parameters determined from curve fitting are shownin Table 2.

P(Am) homopolymer had the lowest affinity constant on thetalc sample, 0.0045 ppm�1, out of all the synthetic depressants.This explains the result of high talc recovery in the flotation exper-iment since the hydrophilic polymer does not adsorb very well tothe hydrophobic basal plane surface of talc and hence is unableto alter its hydrophobicity. On the other hand, the less hydrophilicP(VPD) homopolymer had adsorbed strongly on talc, with an affin-ity constant almost 50 times stronger (0.2152 ppm�1) than P(Am).This is strong evidence for polymer absorption on the basal plane

of talc due to hydrophobic–hydrophobic interactions. This stronginteraction is reflected in the flotation test as the P(VPD) homopol-ymer exhibited a much stronger talc depression than the P(Am).The P(Am-co-VPD) copolymers had affinity constants somewhere

Pentlandite (Ni) Recovery %40 50 60 70 80 90 100

Tal

c (M

g) R

ecov

ery

%

0

10

20

30

40

50

60

70

80

Guar [1x], SEX [1x]P(Am-co-VPD25%) [1x], SEX [1x]P(Am-co-VPD25%) [1.5x], SEX [1x]P(Am-co-VPD25%) [1.5x], SEX [1.5x]

Fig. 5. Mixed mineral (Talc/Quartz/Pentlandite) flotation results at pH 4.5 showingoptimization of flotation conditions for (P(Am-co-VPD25%)) copolymer.

Fig. 6. Adsorption isotherms of the P(Am-co-VPD25%) copolymer and guar onpentlandite in distilled water.

A. Leung et al. / Minerals Engineering 24 (2011) 449–454 453

between the two homopolymers as expected, ranging in between0.0045–0.2152 ppm�1. Interestingly, guar had a relatively highaffinity constant on the talc sample (0.1134 ppm�1), which wasabout 25 times stronger than P(Am) and half the value of P(VPD),but had a much lower plateau adsorbed amount (0.3571 mg/m2)than the copolymer system.

Nevertheless, the results from the isotherm adsorption experi-ment clearly show that the P(Am-co-VPD) copolymers had inheritedthe property of stronger talc affinity from the VPD component (seeFig. 4). Only a small amount of VPD incorporation (5%) was requiredto facilitate the strong adsorption onto the talc surface. Subsequenceincrement of the VPD incorporation (up to 75%) did not increase theadsorption affinity and hence the depressant effectiveness of thecopolymer remained identical (See insert of Fig. 4). It was only whenthe VPD composition reached about 90% that the adsorption affinitybegan to reach the same level as P(VPD). It is theorized that Am ismuch more hydrophilic than VPD and hence it will have a strongdepressant effect should it adsorb on the hydrophobic talc. How-ever, due to their differences in hydrophobicity, the P(Am) homo-polymer does not adsorb strongly onto talc as shown in theisotherm adsorption experiment. The VPD component functions asa less hydrophilic anchor in the hydrophilic P(Am-co-VPD) copoly-mer and facilitates the desired attachment on the hydrophobic basalplane of the talc due to hydrophobic–hydrophobic interactions. Theresultant attachment of the hydrophilic Am segment alters thesurface hydrophobicity of talc to hydrophilic and hence depressesits floatability. This combined effect achieved the maximum balancewhen the VPD composition reached � 25–30%. Further addition ofthe VPD on the P(Am-co-VPD) copolymer system was shown to haveadverse effect on the depressant effectiveness, possibly caused bythe reduction in the amount of the hydrophilic Am and hence thedepressant effect on talc.

3.3. Mixed minerals selectivity study

From the single mineral talc flotation investigation above, the25% VPD copolymer was identified as the most effective talcdepressant out of all the copolymers. Therefore, the P(Am-co-VPD25%) copolymer was employed as the model synthetic depres-sant in the selectivity study on the mixed mineral (pentlandite/quartz/talc) system. The selectivity and depressant effectivenessof the P(Am-co-VPD) copolymers was evaluated by the Denver Flo-tation Cell system, which has been optimised for using guar as thetalc depressant. The cumulative recovery of talc and pentlandite inthe concentrates with different depressants and flotation condi-tions are summarized in Fig. 5.

The results showed that the application of guar in the systemachieved a high pentlandite recovery (maximum nickel recoveryof �95%) and strong depression of talc (maximum magnesiumrecovery of �50%). It was found that the use of P(Am-co-VPD25%) copolymer resulted in a significant reduction in talcdepression compared to guar when the same dosage (37.5 g/t)was used, with a maximum talc recovery of about 70%. These re-sults indicate that if P(Am-co-VPD25%) were to be employed asdepressant, multiple flotation steps will be needed to achieve highpentlandite recovery with low talc contamination in the final prod-uct. Increasing the concentration of the copolymer depressant(56.25 g/t) resulted in improved talc depression (maximum mag-nesium recovery of �40%) but also adversely affected the flotationof pentlandite, dropping the maximum nickel recovery to about80%. This reduction in nickel recovery could be recovered to a cer-tain extent by the addition of more SEX collector (90 g/t comparingto the original of 60 g/t). While these results were positive, thedegree of improvement brought about by optimization of theflotation conditions is still not enough to achieve the same flota-tion results as guar.

The difference in mineral selectivity of the two depressants canbe explained by their respective adsorption affinity on pentlanditeand talc. Adsorption isotherms of P(Am-co-VPD25%) copolymerand guar on pentlandite are shown in Fig. 6, and the correspondingLangmuir affinity constants (b) and plateau adsorbed amounts(Cm

ads) obtained from Langmiur model fitting are summarized inTable 3.

The adsorption selectivity for P(Am-co-VPD25%) copolymerbased on the Langmuir affinity constants is fairly minimal, withthe affinity constant for talc (0.0474 ppm�1; see Table 2) beingvery similar to that of pentlandite (0.0516 ppm�1; see Table 3).Comparing to the affinity constants of guar, it can be seen that guarhas much higher selectivity potential, where the affinity constantfor talc (0.1134 ppm�1) was almost 25 times higher than that ofthe pentlandite value (0.0046 ppm�1). Because of this high adsorp-tion selectivity, guar has a much higher tendency to adsorb specif-ically to the talc surface rather than pentlandite and hencefacilitated a strong talc depression while maintaining a high pent-landite recovery in the mixed mineral flotation. On the other hand,the P(Am-co-VPD25%) copolymer basically has no surface prefer-ence and hence adsorbs equally on both the talc and pentlandite,which resulted in the proportional increase of depressant effecton both minerals when more copolymer was added into the mixedmineral system. This research shows that the P(Am-co-VPD) sys-tem is a promising depressant for talceous gangue minerals, but

Table 3Langmuir affinity constants (b) and plateau adsorbed amounts (Cm

ads) obtained byfitting the adsorption isotherm of P(Am-co-VPD25%) and guar on pentlandite surfaceto the Langmuir model.

Depressant type Pentlandite

b (ppm�1) Cmads (mg/m2) R2

P(Am-co-VPD25%) 0.0516 ± 0.0201 0.3448 ± 0.0143 0.96Guar 0.0046 ± 0.0014 0.7668 ± 0.0631 0.97

454 A. Leung et al. / Minerals Engineering 24 (2011) 449–454

a higher mineral adsorption selectivity is required for the applica-tion in a nickel recovery flotation system.

4. Conclusion

This study has successfully shown that the hydrophobicitycharacteristic has a major role in the adsorption of polymer ontotalc. A less hydrophilic functional group can be employed to aidthe docking of hydrophilic polymer onto the hydrophobic surfaceof talc. Flotation experiments performed under realistic flotationconditions showed that the P(Am-co-VPD) copolymer system hada relative but equally strong affinity for both talc and pentlandite.It has been shown that the copolymer system was capable ofdepressing talc much more effectively than either of guar or theconstituent homopolymers in single mineral flotation system.However, the copolymer system showed a much lower mineralselectivity than guar in the mixed mineral flotation. The absorptionstudy indicated that this low level of selectivity is possibly due tothe small difference of Langmuir affinity constant (b) between thetalc and pentlandite mineral surface. Nevertheless, this copolymersystem may be applicable to other talc rich ores refining process.

Acknowledgements

We would like to acknowledge the CRC for Polymers for thefunding of this research, as well as CSIRO Minerals for performingthe mixed mineral flotation experiments.

References

Beattie, D.A., Huynh, L., Kaggwa, G.B., Ralston, J., 2006a. Influence of adsorbedpolysaccharides and polyacrylamides on talc flotation. International Journal ofMineral Processing 78, 238–249.

Beattie, D.A., Huynh, L., Kaggwa, G.B.N., Ralston, J., 2006b. The effect ofpolysaccharides and polyacrylamides on the depression of talc and theflotation of sulphide minerals. Minerals Engineering 19, 598–608.

Burdukova, E., Van Leerdam, G.C., Prins, F.E., Smeink, R.G., Bradshaw, D.J., Laskowski,J.S., 2008. Effect of calcium ions on the adsorption of CMC onto the basal planesof New York talc – a ToF-SIMS study. Minerals Engineering 21, 1020–1025.

Chen, H.T., Ravishankar, S.A., Farinato, R.S., 2003. Rational polymer design for solid–liquid separations in mineral processing applications. International Journal ofMineral Processing 72, 75–86.

Chiem, L.T., Huynh, L., Ralston, J., Beattie, D.A., 2006. An in situ ATR–FTIR study ofpolyacrylamide adsorption at the talc surface. Journal of Colloid and InterfaceScience 297, 54–61.

Cuba-Chiem, L.T., Huynh, L., Ralston, J., Beattie, D.A., 2008. In situ particle film ATR-FTIR studies of CMC adsorption on talc: the effect of ionic strength andmultivalent metal ions. Minerals Engineering 21, 1013–1019.

Deer, W.A., Howie, R.A., Zussman, J., 1992. An Introduction to the Rock-formingMinerals. Wiley, New York.

Guy, P.J., 1992. The development of laboratory batch flotation equipment andpractice at CSIRO Australia. In: The AusIMM Annual Conference, Broken Hill.The Australasian Institute of Mining and Metallurgy, Melbourne, pp. 75–79.

Hunter, R.J., 2001. Foundations of Colloid Science. Oxford University Press, Oxford.Khraisheha, M., Hollandb, C., Creanyb, C., Harrisc, P., Parolisc, L., 2005. Effect of

molecular weight and concentration on the adsorption of CMC onto talc atdifferent ionic strengths. International Journal of Mineral Processing 75 (3-4),197–206.

Liu, Q., Zhanga, Y., Laskowskib, J.S., 2000. The adsorption of polysaccharides ontomineral surfaces: an acid/base interaction. International Journal of MineralProcessing 60 (3–4), 229–245.

Morris, G.E., Fornasiero, D., Ralston, J., 2002. Polymer depressants at the talc–waterinterface: adsorption isotherm, microflotation and electrokinetic studies.International Journal of Mineral Processing 67 (1–4), 211–227.

Rath, R.K., Subramanian, S., Laskowski, J.S., 1997. Adsorption of dextrin and guargum onto talc. A comparative study. Langmuir 13 (23), 6260–6266.

Sedeva, I.G., Fetzer, R., Fornasiero, D., Ralston, J., Beattie, D.A., 2010. Adsorption ofmodified dextrins to a hydrophobic surface. QCM-D studies, AFM imaging, anddynamic contact angle measurements. Journal of Colloid and Interface Science345, 417–426.

Senior, G.D., Smith, L.K., Silvester, E., Bruckard, W.J., 2009. The flotation ofgersdorffite in sulphide nickel systems – a single mineral study. InternationalJournal of Mineral Processing 93, 165–171.

Shortridge, P.G., Harris, P.J., Bradshaw, D.J., Koopal, L.K., 2000. The effect of chemicalcomposition and molecular weight of polysaccharide depressants on theflotation of talc. International Journal of Mineral Processing 59 (3), 215–224.