- Email: [email protected]
Reducing quartz gangue entrainment in sulphide ore flotation by high molecular weight polyethylene oxide
International Journal of Mineral Processing 97 (2010) 44–51
Contents lists available at ScienceDirect
International Journal of Mineral Processing j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / i j m i n p r o
Reducing quartz gangue entrainment in sulphide ore flotation by high molecular weight polyethylene oxide J. Gong a, Y. Peng b, A. Bouajila c, M. Ourriban c, A. Yeung a, Qi Liu a,⁎ a b c
Department of Chemical & Materials Engineering, University of Alberta, Edmonton, Canada School of Chemical Engineering, University of Queensland, Brisbane, Australia COREM, 1180 rue de la Minéralogie, Quebec City, Quebec, Canada
a r t i c l e
i n f o
Article history: Received 15 October 2009 Received in revised form 12 July 2010 Accepted 30 July 2010 Available online 5 August 2010 Keywords: Polyethylene oxide Flotation Entrainment Homo-aggregation Hetero-aggregation Photometric dispersion analysis
a b s t r a c t Polyethylene oxide (PEO) was tested to flocculate and depress fine quartz particles in the batch flotation of artificial mixtures of chalcopyrite–quartz as well as a commercial Au–Cu sulphide ore sample. The aggregation/ dispersion behaviors of quartz, chalcopyrite and their mixtures in the presence of PEO with and without potassium amyl xanthate (KAX) were studied by photometric dispersion analysis (PDA), scanning electron microscopy (SEM) and zeta potential measurements. Batch flotation results indicated that the addition of low dosages of PEO improved value mineral recovery and concentrate grade during the flotation of both the artificial mixtures of chalcopyrite–quartz and the Au–Cu sulphide ore sample. Aggregation/dispersion test results revealed that the PEO caused non-selective flocculation of quartz and chalcopyrite, forming large hetero-aggregates. However, the addition of KAX caused the chalcopyrite particles to break away from the hetero-aggregates, leading to separate homo-aggregates of quartz and chalcopyrite. The flotation of the fine chalcopyrite and the depression of the fine quartz were thus both improved. © 2010 Elsevier B.V. All rights reserved.
1. Introduction Flotation separation of fine and ultrafine mineral particles is a major challenge faced by the mineral processing industry today. The fine and ultrafine particles cause two problems both of which lower separation efficiency. One problem is the loss of value hydrophobic mineral particles due to the inefficient collection of the particles by gas bubbles, resulting in low value recovery (the recovery problem), and the other is the deterioration in concentrate grade caused by mechanical and hydraulic entrainment of hydrophilic gangue particles (the entrainment problem). Research effort over the past several decades has been focused on understanding the flotation behaviors of fine and ultrafine particles and developing processes to improve their recovery, and much less research has been focused on reducing the entrainment of the fine and ultrafine gangue minerals. A lack of effective methods to reduce fine gangue particle entrainment may be one of the reasons for the scarcity of commercial application of new techniques to improve fine particle recovery, although many such techniques have been proposed (see, for example, the summaries by Fuerstenau, 1980; Mathur et al., 2000; and Rubio, 2003).
⁎ Corresponding author. Fax: +1 780 492 2881. E-mail address: [email protected] (Q. Liu). 0301-7516/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.minpro.2010.07.009
Studies reported in the literature (e.g., Johnson et al., 1974; Trahar, 1981; Warren, 1984; Kirjavainen, 1996; Neethling and Cilliers, 2001; Zheng et al., 2006) provided good models of gangue entrainment, an understanding of the mechanisms of hydraulic and mechanical entrainment, and the effects of water recovery, aeration rate, froth thickness and pulp density on such entrainment. Parallel to this fundamental research, some studies have been carried out to lower the entrainment in froth flotation. To encourage particle drainage through the froth zone, water spray to the froth layer, vibration of the froth zone, or the use of a centrifugal force field flotation cell, have been tested (Kaya, 1989; Cheng et al., 1999; Banisi et al., 2003). Some researchers tried to reduce water recovery (thus entrainment) by increasing the flotation rate of hydrophobic particles so that value minerals could be recovered in as short a time period as possible (Akdemir et al., 2005). Mulleneers et al. (2002) modified mechanical flotation cells by adding a counter current sedimentation zone which is expected to prevent entrainment. Cao and Liu (2006) and Liu et al. (2006) showed that the entrainment of fine and ultrafine hydrophilic particles can be reduced by enlarging their particle sizes using either inorganic depressants (such as ZnSO4 which coagulated and depressed fine sphalerite (Cao and Liu, 2006)), or high molecular weight polymers (Liu et al., 2006). In this paper, we report results of a study to extend the particle sizeenlargement concept to bench mechanical flotation of artificial mixtures of chalcopyrite and quartz, as well as a commercial Au–Cu ore sample. Polyethylene oxide (PEO) was chosen to flocculate quartz particles
J. Gong et al. / International Journal of Mineral Processing 97 (2010) 44–51
because it has shown a high affinity towards quartz (Rubio and Kitchener, 1976). Although PEO is not a commonly used flocculant in mineral industry, it has been widely used in pulp and paper (van de Ven et al., 2004). Several recent studies have also shown that PEO is potentially useful in dewatering mineral and oil sands tailings (Scheiner and Smelley, 1984; Sworska et al., 2000; Mpofu et al., 2003; Wu et al., 2006). 2. Experimental 2.1. Test samples and reagents A quartz single mineral sample was obtained from Iron Ore Company of Canada. X-ray fluorescence (XRF) analysis indicated that it contains 99.9% SiO2. A chalcopyrite single mineral sample, originated from Durango, Mexico, was purchased from Ward's Natural Sciences. Inductively Coupled Plasma Mass Spectrometric (ICP-MS) analysis showed that it contains 31.8% Cu, 27.9% Fe and 39.4% S, representing a purity of 91.8% CuFeS2. The quartz and chalcopyrite single mineral samples were separately crushed to collect the −3.2 + 0.6 mm particles to be used in batch grind-flotation tests. A small portion of the −3.2 + 0.6 mm particles were dry ground in a mechanized agate mortar and pestle grinder manufactured by Fritsch GmbH. The ground mineral particles were screened and the −20 μm size fraction was collected and used in flocculation/dispersion, zeta-potential and scanning electron microscopic measurements. A commercial Au–Cu sulphide ore sample was taken from one of the member mining companies of COREM, Quebec, Canada, from the SAG mill discharge point. The sample has a prevailing particle size of about 1 mm, and contains chalcopyrite as the value mineral. The Au was associated with the chalcopyrite. The ore sample assayed 0.37 g/t Au, 0.07% Cu, 4.82% Fe and 32.2% Si. The collected sample was air-dried, riffled into 1 kg batches and stored in a freezer in individual sealed plastic bags. Two polyethylene oxide (PEO) reagents, with molecular weights of 1 million and 8 million, respectively, were obtained from Polysciences, Inc. Fresh stock solutions of PEO were prepared at a concentration of 1 g/L daily. Both PEOs were able to induce flocculation of the mineral suspensions, and the PEO with a higher molecular weight was used in some tests to facilitate experimental observations (such as in the photometric dispersion analyzer) due to the stronger flocculation it caused. Potassium amyl xanthate (KAX) and sodium isopropyl xanthate (NaIPX) were used as collectors for the sulphide minerals and were obtained from Charles Tennant & Company Ltd and Cytec Industries, respectively. Both xanthate reagents were purified by following the procedure described by DeWitt and Roper (1932). Dowfroth 250 (DF250) was obtained from Charles Tennant & Company and used as a frother. Other common chemical reagents used in the experiment, such as sodium hydroxide and hydrochloric acid, etc., were purchased from Fisher Scientific. They were used directly without further purification. 2.2. Flotation tests Batch flotation tests on quartz single mineral and synthetic mixtures of chalcopyrite and quartz were conducted in COREM labs, using a Denver laboratory flotation machine in a 1.25-L cell. Prior to flotation, 100 g of quartz or a mixture of 100 g quartz and 20 g chalcopyrite were ground with 150 mL of demineralized water in a zircon pebble mill to 80% passing 20 μm. Mill discharge was transferred to the 1.25-L flotation cell and conditioned first with PEO (molecular weight= 1 million, various dosages), then NaIPX (60 g/t) and DF250 (150 g/t) for 2 min with each reagent addition at 1200 rpm of agitation. Lime (CaO) was used to adjust pulp pH to 9 in the grinding mill and the flotation cell. The
45
flotation froth was scraped every 10 s and four rougher concentrates were collected at 0.5, 2, 4 and 8 min, and analyzed separately. Batch flotation tests on the Au–Cu sulphide ore sample were conducted in a laboratory Wemco flotation machine with a 2-L flotation cell at the University of Alberta. Prior to flotation, 1 kg of the ore sample was ground with 700 mL of Edmonton tap water in a mild steel ball mill to 90% passing 75 μm. The ground slurry was transferred to the 2-L flotation cell, conditioned for 3 min with lime addition to a pH of 10, another 3 min with PEO (molecular weight = 8 million) (if used), and finally with KAX for 3 min. This was followed by two stages of rougher flotation (2 min for the first stage and 3 min for the second stage). The combined rougher concentrate was reground for cleaner flotation, which was conducted for 6 min with and without the addition of PEO. 2.3. Flocculation/dispersion measurements The degree of flocculation or dispersion of mineral suspension was measured by a Photometric Dispersion Analyzer (PDA 2000) manufactured by Rank Brothers Inc. Photometric dispersion analysis is a sensitive technique developed to monitor the aggregation state of particle suspensions by Gregory and Nelson (1986). Compared to the traditional settling tests, it has the advantage of providing information about the change in the aggregation or dispersion state of the suspended particles, such as initial aggregation, floc growth, floc breakage and re-aggregation, etc., in real time. It has been widely used by many researchers (Gregory and Carlson, 2003; Hopkins and Ducoste, 2003; Jin et al., 2007). Its principle has been described in detail by Gregory and Nelson (1986) and Ching et al. (1994). So only a brief description will be given here. The PDA measures the intensity of light transmitted through a flowing suspension. For a well-agitated suspension, there is a small fluctuation in the intensity of the transmitted light due to the random variation in the particulate composition of the slurry flowing through the path of the light beam. A photodiode is used to monitor the transmitted light intensity. The output of the light intensity is divided into two components, a large mean value (DC) and a small fluctuating value (AC). The DC value corresponds to the average transmitted light intensity and depends on the turbidity of the flowing suspension. Compared to the DC value, the AC value is very small and can be amplified by a desired factor. The root mean square (RMS) of the AC value is then derived and divided by the DC value. The result is reported as a “Ratio” reading. The Ratio value is very sensitive to the concentration and size of the suspended particles. When aggregation occurs, the Ratio value increases. Conversely, when the suspension becomes more dispersed, the Ratio reading decreases. Therefore, the relative change in “Ratio” value is a sensitive indicator of the aggregation or dispersion state of the suspension. In each test, certain amount of minerals (−20 μm) and water were agitated using a magnetic stirrer in a 250-mL beaker and circulated through the flow cell of the PDA 2000 using a peristaltic pump. The magnetic stirring bar was 51 mm long and the stirring speed was fixed at 330 rpm. The Ratio output as a function of time was recorded by a computer connected to the PDA 2000. In such a way, the flocculation/ dispersion state of the suspension was monitored in real time following the addition of any reagents to the circulating suspension. Sworska et al. (2000) observed that hydrodynamic conditions are important factors in the flocculation of oil sands tailings by PEO. They observed that when the PEO was added to the tailings slurry under vigorous agitation (100–200 rpm in a baffled 500-mL beaker), the tailings slurry could be well flocculated. However, when the PEO solution was added to the tailings slurry in a 500-mL graduated cylinder followed by moving a plunge up and down five times, no flocculation of the tailings slurry was observed (Sworska et al., 2000). In our work, both the batch mechanical flotation machine and the magnetic stirring/ peristaltic pumping seemed to have provided sufficient agitation, and flocculation was observed in both cases. However, the hydrodynamics in the batch mechanical flotation machine was almost certainly different
46
J. Gong et al. / International Journal of Mineral Processing 97 (2010) 44–51
from those in the magnetically stirred beakers. The effect of such differences will be commented later in the “Discussion” section. 2.4. Zeta potential measurements Zeta potential measurement was performed using a ZetaPALS zeta potential analyzer manufactured by Brookhaven Instruments. All solutions and reagents were prepared with stock solutions of 10−2 mol/L KCl to maintain a constant ionic strength. To prepare the stock suspension, 1 g of the mineral (−20 μm) was added to a 1-L volumetric flask containing the 10−2 mol/L KCl solution. The flask was then sealed and inverted several times to keep the mineral particles well dispersed. For each zeta potential measurement, 10 mL of the stock mineral slurry was withdrawn and diluted with 90 mL of 10−2 mol/L KCl solution, adjusted to appropriate pH, and then treated by desired reagents. A very small amount (about 2 mL) of this conditioned particle suspension was transferred to the plastic sample cell of the ZetaPALS for zeta potential measurement.
Fig. 1. Quartz recovery as a function of water recovery in the batch flotation of pure quartz at several PEO dosages. Molecular weight of PEO is 1 million.
2.5. SEM/EDS observations A JEOL 6301F SEM scanning electron microscope (SEM) coupled with an energy dispersive X-ray spectrometer (EDS) was used to examine the quartz–chalcopyrite flocs. This machine is equipped with a PGT (Princeton Gamma-Tech) IMIX digital imaging system, a solid state silicon backscattered electron detector and a liquid nitrogen cooled silicon (lithium) detector for energy dispersive X-ray analysis. Since all the constituent elements in chalcopyrite (Cu, Fe and S) have higher atomic mass than those in quartz (Si and O), chalcopyrite particles can scatter more electrons and hence appear much brighter than quartz particles in SEM images, especially back-scattered electron images. EDS also showed that the brighter spots are mostly comprised of Cu, Fe and S, indicating that they are CuFeS2 while the majority of the gray spots are Si and O, i.e., SiO2. To prepare a SEM sample, chalcopyrite and quartz were mixed in 250-mL beakers into a 10% solid suspension, at a ratio of 10% chalcopyrite and 90% quartz. The suspension was agitated using a magnetic stirrer and the pH was adjusted to 9. Two parallel tests were conducted: one with only PEO addition, and the other with KAX addition 3 min after PEO. Small slurry samples were taken from the beakers at 1 min and 15 min after PEO addition (the 15 minute samples may have also been treated by KAX). All slurry samples were diluted 20 times with distilled water and small amount of the diluted suspension (two or three drops) was then deposited on the SEM sample stand such that the flocs would be well separated and would not overlap with each other after being dried in air. The dried samples were coated with chromium to render them conductive before being placed into the SEM sample chamber.
remained but the degree of entrainment was significantly reduced, from 0.73 without any PEO, to 0.34 at 80 g/t PEO (6.4 mg/L). The reduction of quartz entrainment was likely due to the bridging flocculation of quartz by the PEO (more on this in the next section). When the PEO flocs of quartz form, they could potentially trap sulphide mineral particles and affect their flotation. In fact, one of the primary concerns with the use of high molecular weight polymer depressants in froth flotation is that the flocs can trap value minerals and lower their recovery. That is why most of the polymeric depressants used in froth flotation are low molecular weight polymers, typically up to a few thousand grams per mole. In order to examine if the possible PEO flocs of quartz would trap sulphide mineral particles, batch flotation tests were performed on synthetic quartz–chalcopyrite mineral mixtures, which were ground together to 80% passing 20 μm. Flotation of the ground slurry was carried out at about 6 wt.% solids, using different dosages of PEO (molecular weight= 1 million), and 60 g/t NaIPX which serves as a collector for the chalcopyrite. Fig. 2 shows a recovery–grade plot of the batch flotation results. It can be seen that the recovery–grade curves moved upward as PEO dosage was increased from 0 to 100 g/t (9.6 mg/L), indicating that the use of PEO improved the separation of chalcopyrite from the quartz– chalcopyrite mixtures. PEO was then tested as a selective depressant in the flotation of a commercial Au–Cu sulphide ore sample in which quartz was the major gangue mineral. This ore sample was ground to 90% passing 75 μm. Two
3. Results 3.1. Batch flotation Batch flotation tests were first carried out using pure quartz. The quartz particles were ground to 80% passing 20 μm in a laboratory zircon pebble mill and floated in a 1.25-L flotation cell with the addition of 60 g/ t NaIPX, 150 g/t of DF250 and varying dosages of PEO (molecular weight= 1 million) at 6 wt.% solids. Since no collector for quartz was used, quartz recovery in these tests was considered purely a result of entrainment. Fig. 1 shows the relationship between quartz recovery and water recovery at different PEO dosages. As can be seen, when no PEO was used, quartz recovery increases linearly with water recovery, consistent with the observations of other researchers (e.g., Trahar, 1981). The slope of the line, variously called degree of entrainment or entrainment factor (Smith and Warren, 1989), can be used to indicate the severity of entrainment. After adding PEO, the linear relationship
Fig. 2. Recovery–grade relationship of the batch flotation of synthetic chalcopyrite– quartz mixtures. Molecular weight of PEO is 1 million.
J. Gong et al. / International Journal of Mineral Processing 97 (2010) 44–51
Fig. 3. Gold and copper recovery as a function of Si content in rougher and cleaner concentrates in the batch flotation of a commercial Au–Cu sulphide ore. Molecular weight of PEO is 8 million.
sets of batch flotation tests were carried out, both with rougher flotation followed by cleaner flotation. In one set of tests, PEO (molecular weight= 8 million) was used in both rougher and cleaner flotation, while in the other, no PEO was used in either rougher or cleaner. The flotation was conducted at about 30 wt.% solids and a pulp pH of 10.3, with 160 g/t KAX (potassium amyl xanthate) as a collector for the sulphide minerals and gold. In the tests in which PEO was used, 10 g/t was added in rougher and 5 g/t was added in the cleaner. Fig. 3 shows the flotation results, with Cu and Au recovery plotted as a function of Si content in the rougher and cleaner concentrates. As can be seen, higher Cu and Au recoveries could be achieved at the same Si content in both the rougher and cleaner concentrates when PEO was used. This means that at the same Cu and Au recovery, the Si contents in the rougher and cleaner concentrates were much lower when PEO was used. Therefore, the addition of PEO had the benefit of improving the concentrate grade while maintaining or increasing copper and gold recovery during the flotation of the Au–Cu ore sample. 3.2. Flocculation measurement with PDA The batch flotation results show that the addition of PEO can reduce quartz entrainment and improve concentrate grade in chalcopyrite and sulphide ore flotation. In order to confirm that the lowering of quartz entrainment is due to the bridging flocculation by PEO, a series of photometric dispersion analyzer (PDA) measurements were performed. 3.2.1. Flocculation of 1:1 quartz–chalcopyrite mixtures The first set of PDA measurements was carried out on suspensions of chalcopyrite and quartz mixtures (50 wt.% quartz and 50 wt.% chalcopyrite). The suspension was prepared at 1 wt.% solids and pH 9, and was agitated with a magnetic stir bar (51 mm long) at 330 rpm while being circulated through the PDA 2000 photometric dispersion analyzer. Three tests were conducted. In the first test, only PEO (molecular weight= 8 million) was added (to a final concentration of 10 mg/L). In the next two tests, KAX (300 mg/L) was added 3 min before or after the addition of PEO while the Ratio output was continuously monitored. The results are presented in Fig. 4, where the points of PEO and KAX additions are noted. As can be seen from Fig. 4, in all tests, the addition of PEO caused an immediate increase in the Ratio value by about 5 times, indicating that the particles were flocculated by PEO. A typical SEM image of the suspension, shown in Fig. 5, confirms the presence of large flocs and it also shows that chalcopyrite and quartz particles formed hetero-flocs, i.e., they were flocculated together.
47
Fig. 4 shows that after reaching the maximum value, the Ratio reading in all three cases started to drop quickly. This means that the flocs were not strong and they were easy to break by the mechanical forces such as magnetic stirring and the peristaltic pumping. The rate at which the Ratio value drops can be taken as an indication of the breakage rate of the flocs. Since the initial part of the dropping segment of the curve appeared to be linear, the slope of that segment was calculated to indicate the dropping rate of the Ratio reading. For the case where only PEO was added, the slope in the initial dropping segment was found to be −0.49 min–1. When KAX was added 3 min after PEO, the slope increased to −0.66 min–1, indicating a faster floc breakage with the addition of KAX. When KAX was added 3 min before PEO, the Ratio reading increased from 1 to about 2.2 immediately following the KAX addition. This increase was most likely caused by the aggregation of the chalcopyrite due to its surface hydrophobicity induced by the adsorbed KAX. Subsequent addition of PEO 3 min later resulted in a further increase of the Ratio reading from about 2.2 to about 5, followed by a sharp drop. In fact, the slope of the dropping segment of the Ratio curve in this case was the highest, at −0.95 min–1. The above PDA measurement results reveal that PEO caused unselective hetero-aggregation of the quartz and chalcopyrite particles. The flocs were not strong and tend to break as a result of magnetic stirring and peristaltic pumping. The addition of KAX accelerated the breakage of these hetero-aggregates, especially when KAX was added ahead of PEO. 3.2.2. Flocculation of quartz and chalcopyrite single minerals To examine the roles that KAX plays in the quartz–chalcopyrite mixture system, its effect on quartz and chalcopyrite single mineral flocculation was studied. The test procedures were the same as above except that only quartz or chalcopyrite was used in each test. Fig. 6 shows the results of pure quartz particle suspension at about 5 wt.% solids (i.e., about 5 times more concentrated than the quartz– chalcopyrite mixture tests) at pH 9. As can be seen, when only KAX was added, there was no change in the Ratio reading. This is as expected since xanthate is known to have no interactions with quartz. However, when PEO was added to a concentration of 10 mg/L, there was an abrupt increase in the Ratio reading from nearly zero to close to 13 (the much higher Ratio value was probably due to the higher particle concentration). The quartz was flocculated by the PEO. These flocs were not stable, and the Ratio reading dropped quickly after reaching the maximum. Adding KAX after PEO did not seem to have caused any difference in the Ratio reading. Therefore, it can be concluded that KAX has no effect on quartz particle flocculation by PEO.
Fig. 4. Effect of KAX and PEO on the flocculation/dispersion of 1:1 quartz–chalcopyrite mixtures. Molecular weight of PEO is 8 million. The point of PEO or KAX addition is noted by a vertical arrow.
48
J. Gong et al. / International Journal of Mineral Processing 97 (2010) 44–51
Fig. 6. Effect of KAX and PEO on the flocculation/dispersion of quartz. Molecular weight of PEO is 8 million. The point of PEO or KAX addition is noted by a vertical arrow.
seemed to have been returned to the initial dispersed state when only PEO was used. When KAX was added 3 min after PEO, the breakage rate (slope) increased to a much higher value at −5.46 min–1, indicating that KAX sped up the breakage of the chalcopyrite–PEO flocs. However, it is interesting to note that after this initial sudden drop following KAX addition, the Ratio reading then stabilized at about 3.0 rather than dropping continuously to the original value of 1.0. It is speculated that KAX caused the breakage of the PEO flocs of chalcopyrite, but some of the chalcopyrite particles may have been re-flocculated by KAX, for the formed aggregates seem to be resistant to breakage, similar to the case when KAX was added alone. When KAX was added first, followed by PEO 3 min later, there was an initial small and gradual increase in the Ratio reading from 1 to about 2.7, again possibly by the flocculation of the hydrophobized chalcopyrite. This was followed by a sudden increase of the Ratio reading to above 5 after adding the PEO. The Ratio reading quickly dropped, with a slope of −1.35 min–1, and finally stabilized at about 2.5.
3.2.3. Flocculation of quartz–chalcopyrite mixtures with varying amounts of chalcopyrite The foregoing description of the PDA test results seem to point to the same trend, that the addition of KAX promotes the formation of breakage-resistant chalcopyrite–KAX flocs, and breaks chalcopyrite– PEO flocs. However, the addition of KAX does not affect quartz nor the quartz–PEO flocs. Therefore, it seems that in the hetero-aggregates of Fig. 5. (a) SEM image of a typical PEO floc of quartz–chalcopyrite. (b) EDS spectrum of point (1): chalcopyrite, and (c) EDS spectrum of point (2): quartz. Molecular weight of PEO is 8 million.
However, as can be seen from Fig. 7, the phenomenon was very different in the chalcopyrite single mineral system. With the addition of KAX to a concentration of 300 mg/L, there was an initial sudden increase of the Ratio reading of the 0.5 wt.% chalcopyrite suspension from 1 to about 2, followed by a steady increase to about 5. The Ratio reading never decreased in the 30-min recording period, as have been observed in the tests with quartz–chalcopyrite mixtures or the quartz single mineral. Clearly, the KAX caused the aggregation of the chalcopyrite particles and the aggregates were more resistant to breakage than the PEO flocs. Fig. 7 also shows that when PEO was added, chalcopyrite particles were flocculated and the Ratio reading increased from 1 to over 5, but the flocs were not stable and broke up gradually. The breakage rate, i.e., the slope of the dropping segment of the Ratio curve, was calculated to be −0.84 min–1. The Ratio reading continued the dropping trend, reaching the initial value of about 1.0 in 30 min, i.e., the chalcopyrite particles
Fig. 7. Effect of KAX and PEO on the flocculation/dispersion of chalcopyrite. Molecular weight of PEO is 8 million. The point of PEO or KAX addition is noted by a vertical arrow.
J. Gong et al. / International Journal of Mineral Processing 97 (2010) 44–51
quartz and chalcopyrite formed by the addition of PEO, the chalcopyrite particles are the “weak spots” once KAX was added. The xanthate anions are known to adsorb on the chalcopyrite particles through chemical and/or electrochemical reactions (Fuerstenau, 1982; Mielczarski et al., 1996; Woods, 1984). Such a strong chemisorption may have caused the desorption of the PEO from the chalcopyrite surfaces, leading to the disintegration of the quartz–chalcopyrite hetero-aggregates. The removed chalcopyrite particles, however, could form hydrophobic chalcopyrite–KAX flocs due to the surface hydrophobicity induced by KAX. This hypothesis seems to be borne out by the results shown in Fig. 8. This figure shows PDA measurement results on quartz– chalcopyrite mixtures with different chalcopyrite contents. As can be seen, the slopes of the decreasing segment of the Ratio curves increased with increasing chalcopyrite content after adding KAX, at −0.15 min–1 for 10 wt.%, −0.7 min–1 for 30 wt.%, and −5.46 min–1 for 100 wt.% chalcopyrite. Also, the final Ratio readings were at about 1.0, 1.5 and 2.5 for the mixtures with 10 wt.%, 30 wt.% and 100 wt.% chalcopyrite, respectively. In other words, after adding KAX, the more chalcopyrite the quartz–chalcopyrite mixture contains, the faster the PEO flocs break, and the higher the final Ratio reading. 4. Discussion To understand the observed batch flotation and PDA flocculation/ dispersion test results, it is envisaged that the following may be what happened during the flotation process: when PEO was added to the quartz–chalcopyrite mixture, it flocculated both minerals together to form hetero-aggregates. With continuous stirring, especially in a batch mechanical flotation cell, some of the flocs were broken. When KAX was added after PEO addition, it adsorbed on chalcopyrite surfaces and partially or completely (depending on KAX dosages) replaced the PEO, which further weakens the hetero-aggregates. As a result, chalcopyrite particles disassociated from the hetero-aggregates, causing further breakup of the flocs. This postulate is consistent with the two generally accepted flocs breakage mechanisms, i.e., large-scale fragmentation, and surface erosion (Pandya and Spielman, 1982; Jarvis et al., 2005). Fragmentation is thought to be caused by tensile stress across the entire floc, while surface erosion is attributed to the shearing stress on the floc surface. In this case, the desorption of PEO by KAX renders chalcopyrite particles on the surface less resistant to the shearing stress so chalcopyrite particles are stripped off from the flocs more easily (note that the chalcopyrite particles in the hetero-flocs were scattered (see Fig. 5) and hydrophobic forces will not come into play even after the KAX has adsorbed on the chalcopyrite particles). The flocs become
Fig. 8. Effect of PEO and KAX on the flocculation/dispersion of quartz–chalcopyrite mixtures with different chalcopyrite content. Molecular weight of PEO is 8 million. The point of PEO or KAX addition is noted by a vertical arrow.
49
weaker and the fragmentation worsens, which in turn exposes more chalcopyrite particles to shearing stress, eventually leading to the disintegration of the large PEO flocs, and forming smaller flocs that contain quartz mostly. The individual chalcopyrite particles then associate to form hydrophobic aggregates under agitation, due to the surface hydrophobicity caused by the adsorption of KAX. Fig. 9 shows two SEM images that demonstrate such an effect of KAX. Both images show samples that were taken from the quartz–chalcopyrite mixture suspensions 15 min after PEO was added, but in the sample shown in Fig. 9b, KAX was added 3 min after PEO. The quartz– chalcopyrite mixtures contained 10 wt.% chalcopyrite. Fig. 9a and b shows typical field of view of these two samples. The most obvious difference between these two images is that large hetero-aggregates existed in the image shown in Fig. 9a where only PEO was added. These large flocs were not observed in Fig. 9b where both PEO and KAX were used. Another difference is the way by which chalcopyrite particles are incorporated in the flocs. In Fig. 9a (with no KAX), chalcopyrite particles are mostly incorporated in large and medium size flocs. In Fig. 9b (with KAX), the small flocs are concentrated in either chalcopyrite or quartz. In other words, the large hetero-aggregates formed by the addition of PEO were broken as a result of subsequent KAX addition, to form smaller homo-aggregates of quartz and chalcopyrite. Even though the quartz flocs were broken under agitation and the action of KAX, their sizes are still larger than the single dispersed quartz particles. Therefore, the quartz flocs are less likely to be hydraulically carried over to the concentrate, and some of them may be large enough to overcome the fluid drag and drain
Fig. 9. SEM images of suspension samples taken 15 min after the addition of PEO, (a) without KAX; (b) KAX added 3 min after PEO. Bright particles: chalcopyrite; gray particles: quartz. Molecular weight of PEO is 8 million.
50
J. Gong et al. / International Journal of Mineral Processing 97 (2010) 44–51
back to the pulp from the froth layer, i.e., the entrainment of quartz is reduced. In fact, the zeta potential measurement results reveal the different competitive adsorption behaviors of PEO and KAX on the mineral surface. As can be seen from Fig. 10, PEO was able to adsorb on both quartz and chalcopyrite surfaces, driving the zeta potentials of both minerals to near zero. As PEO is a nonionic polymer, its adsorption could stretch the shear plane of the electrical double layer on the mineral surface far away from the surface, lowering the magnitude of the zeta potential. On the other hand, the addition of KAX only altered the zeta potentials of chalcopyrite, making it more negatively charged as more KAX was added, and it did not affect the zeta potential of quartz. In fact, it is known that xanthate anions can chemisorb on chalcopyrite surfaces through the formation of surface metal xanthate compounds, giving rise to a negative surface charge. In this context, it is KAX that has lent selectivity to this system. Since KAX (xanthate) is a specific collector for all sulphide minerals, the combined use of xanthate and PEO should benefit any sulphide ore flotation with rejection of quartz and silicate gangue by lowering its mechanical entrainment. It is realized that the hydrodynamic conditions in a batch mechanical flotation machine are different from a suspension that undergoes magnetic-bar stirring and peristaltic pumping. Whether the same aggregation and dispersion behavior of the quartz–chalcopyrite mixtures with the addition of PEO and xanthate as observed in the PDA tests occurred in the batch mechanical flotation machine is yet to be verified, especially in view of the “fragile” nature of the PEO flocs. However, the improved flotation results as shown by the grade– recovery relationship after the addition of PEO serve as an indication that similar aggregation or dispersion behavior may have happened in the mechanical flotation machine. In fact, our tests using a 3-L baffled container agitated at several hundred rpm showed similar aggregation behavior of xanthate and PEO on quartz and galena. In this context, it is important to note that Sworska et al. (2000) reported that a relatively strong agitation (hydrodynamic conditions) was required when the PEO solution was introduced to the oil sands tailings to cause the flocculation of the latter, and Ding and Laskowski (2007) showed that a strong agitation broke the large polyacrylamide flocs and caused the formation of selectively flocculated smaller flocs of coal and gangue particles. Clearly, the pulp hydrodynamic condition is an important operating parameter in controlling the flocculation behavior of PEO and the selectivity of the flocculation. This work described the chemical effect of xanthate on breaking PEO flocs of chalcopyrite and quartz. The effect could be further enhanced by different hydrodynamics. In fact, hydrodynamic condition is an important factor that will be systematically studied. This is because, unlike tailings dewatering, large flocs are not necessarily desirable when polymers are used in a froth flotation machine with the intention to lower hydraulic entrainment.
Fig. 10. Effect of KAX and PEO on the zeta potentials of chalcopyrite and quartz. Molecular weight of PEO is 8 million.
5. Conclusions Polyethylene oxides (PEO), with molecular weights of 1 million or 8 million, were found to be able to lower the mechanical entrainment of quartz in batch flotation. When PEO was used in the batch flotation of a commercial Au–Cu sulphide ore sample, it also reduced quartz entrainment and improved copper and gold recovery. As a result, rougher and cleaner concentrates with higher grade and higher Cu and Au recovery were obtained after the addition of PEO. Aggregation/ dispersion studies using a photometric dispersion analyzer (PDA) and a scanning electron microscope (SEM), coupled with zeta potential measurements, showed that during the PEO-assisted flotation, PEO flocculated quartz and chalcopyrite together, forming hetero-aggregates. Subsequent addition of a xanthate collector caused the chalcopyrite particles to break away from the PEO hetero-aggregates, due to the competitive adsorption of xanthate which probably removed PEO from the chalcopyrite surfaces. The removed chalcopyrite particles were able to form homo-aggregates, possibly due to the surface hydrophobicity induced by the adsorbed xanthate. The chalcopyrite–xanthate homo-aggregates were more resistant to breakage than the chalcopyrite–PEO flocs. The combined use of PEO and xanthate could therefore potentially benefit the froth flotation of polymetallic sulphide ores to reject fine and ultrafine quartz and silicate gangue. Acknowledgements This project is supported by the Natural Sciences and Engineering Research Council of Canada (NSERC) through a discovery grant, and by COREM, Quebec and NSERC through a collaborative research and development (CRD) project grant (2005–2007). References Akdemir, U., Guler, T., Yildiztekin, G., 2005. Flotation and entrainment behavior of minerals in talc–calcite separation. Scand. J. Metall. 34, 241–244. Banisi, S., Sarvi, M., Hamidi, D., Fazeli, A., 2003. Flotation circuit improvements at the Sarcheshmeh copper mine. Trans. Inst. Min. Metall. C. 112, C198–C205. Cao, M., Liu, Q., 2006. Reexamining the functions of zinc sulfate as a selective depressant in differential sulfide flotation—the role of coagulation. J. Colloid Interface Sci. 301, 523–531. Cheng, H., Cai, C., Zhang, X., Zhang, J., 1999. A highly selective flotation cell with oscillating separator. In: Xie, H., Golosinski, T.S. (Eds.), Mining Science and Technology '99. A.A. Balkema, Rotterdam/Brookfield, pp. 551–553. Ching, H.W., Tanaka, T.S., Elimelech, M., 1994. Dynamics of coagulation of kaolin particles with ferric chloride. Water Res. 28, 559–569. DeWitt, C.C., Roper, E.E., 1932. The surface relations of potassium ethyl xanthate and pine oil. I. J. Am. Chem. Soc. 54, 444–455. Ding, K.J., Laskowski, J.S., 2007. Effect of conditioning on selective flocculation with polyacrylamide in the coal reverse flotation. Trans. Inst. Min. Metall. C. 116, C108–C114. Fuerstenau, D.W., 1980. Fine particle flotation. In: Somasundaran, P. (Ed.), Fine Particle Processing. AIME, New York, pp. 669–705. Fuerstenau, M.C., 1982. Adsorption of sulphydryl collectors. In: King, P. (Ed.), Principle of Flotation: South African Institute of Mining and Metallurgy Monograph Series, 3, pp. 91–108. Johannesburg. Gregory, D., Carlson, K., 2003. Relationship of pH and floc formation kinetics to granular media filtration performance. Environ. Sci. Technol. 37, 1398–1403. Gregory, J., Nelson, D.W., 1986. Monitoring of aggregates in flowing suspensions. Colloids Surf. 18, 175–188. Hopkins, D.C., Ducoste, J.J., 2003. Characterizing flocculation under heterogeneous turbulence. J. Colloid Interface Sci. 264, 184–194. Jarvis, P., Jefferson, B., Gregory, J., Parsons, S.A., 2005. A review of floc strength and breakage. Water Res. 39, 3121–3137. Jin, P.K.K., Wang, X.C.C., Chai, H.X., 2007. Evaluation of floc strength by morphological analysis and PDA online monitoring. Water Sci. Technol. 56, 117–124. Johnson, N.W., McKee, D.J., Lynch, A.J., 1974. Flotation rates of nonsulfide minerals in chalcopyrite flotation process. Trans. AIME 256, 204–226. Kaya, M., 1989. Froth washing in mechanical flotation cells. Ph. D. Thesis, McGill University, Montreal. Kirjavainen, V.M., 1996. Review and analysis of factors controlling the mechanical flotation of gangue minerals. Int. J. Miner. Process. 46, 21–34. Liu, Q., Wannas, D., Peng, Y., 2006. Exploiting the dual functions of polymer depressants in fine particle flotation. Int. J. Miner. Process. 80, 244–254. Mathur, S., Singh, P., Moudgil, B.M., 2000. Advances in selective flocculation technology for solid–solid separations. Int. J. Miner. Process. 58, 201–222.
J. Gong et al. / International Journal of Mineral Processing 97 (2010) 44–51 Mielczarski, J.A., Mielczarski, E., Cases, J.M., 1996. Interaction of amyl xanthate with chalcopyrite, tetrahedrite, and tennantite at controlled potentials. Simulation and spectroelectrochemical results for two-component adsorption layers. Langmuir 12, 6521–6529. Mpofu, P., Addai-Mensah, J., Ralston, J., 2003. Investigation of the effect of polymer structure type on flocculation, rheology and dewatering behaviour of kaolinite dispersions. Int. J. Miner. Process. 71, 247–268. Mulleneers, H.A.E., Koopal, L.K., Bruning, H., Rulkens, W.H., 2002. Selective separation of fine particles by a new flotation approach. Sep. Sci. Technol. 37, 2097–2112. Neethling, S.J., Cilliers, J.J., 2001. Simulation of the effect of froth washing on flotation performance. Chem. Eng. Sci. 56, 6303–6311. Pandya, J.D., Spielman, L.A., 1982. Floc breakage in agitated suspensions — theory and data-processing strategy. J. Colloid Interface Sci. 90, 517–531. Rubio, J., 2003. Unconventional flocculation and flotation techniques. In: Ralston, J., Miller, J., Rubio, J. (Eds.), Flotation and Flocculation: From Fundamentals to Applications. Proceedings from Strategic Conference and Workshop, Hawaii, pp. 17–32. Rubio, J., Kitchener, J.A., 1976. Mechanism of adsorption of poly(ethylene oxide) flocculant on silica. J. Colloid Interface Sci. 57, 132–142. Scheiner, B.J., Smelley, A.G., 1984. Dewatering of fine particle mining wastes using polyethylene oxide flocculant. Miner. Metall. Process. 1, 71–75.
51
Smith, P.G., Warren, L.J., 1989. Entrainment of particles into flotation froths. In: Laskowski, J.S. (Ed.), Frothing in Flotation. Gorgan and Breach, Glasgow, pp. 123–145. Sworska, A., Laskowski, J.S., Cymerman, G., 2000. Flocculation of the Syncrude fine tailings part II. Effect of hydrodynamic conditions. Int. J. Miner. Process. 60, 153–161. Trahar, W.J., 1981. A rational interpretation of the role of particle-size in flotation. Int. J. Miner. Process. 8, 289–327. van de Ven, T.G.M., Qasaimeh, M.A., Paris, J., 2004. PEO-induced flocculation of fines: effects of PEO dissolution conditions and shear history. Colloids Surf. A. 248, 151–156. Warren, L.J., 1984. Ultrafine particles in flotation. In: Jones, M.H., Woodcock, J.T. (Eds.), Principles of Mineral Flotation. Australian IMM, Melbourne, pp. 185–214. Woods, R., 1984. Electrochemistry of sulfide flotation. In: Jones, M.H., Woodcock, J.T. (Eds.), Principles of Mineral Flotation, The Wark Symposium: Australian Institute of Mining and Metallurgy, Symposia Series No. 40, Victoria, Australia, pp. 91–115. Wu, M.R., Paris, J., Van De Ven, T.G.M., 2006. Polyethylene oxide induced fines flocculation and retention: from bench top experiments to paper machine performance. Nord. Pulp Pap. Res. J. 21, 646–652. Zheng, X., Johnson, N.W., Franzidis, J.P., 2006. Modelling of entrainment in industrial flotation cells: water recovery and degree of entrainment. Miner. Eng. 19, 1191–1203.
Contents lists available at ScienceDirect
International Journal of Mineral Processing j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / i j m i n p r o
Reducing quartz gangue entrainment in sulphide ore flotation by high molecular weight polyethylene oxide J. Gong a, Y. Peng b, A. Bouajila c, M. Ourriban c, A. Yeung a, Qi Liu a,⁎ a b c
Department of Chemical & Materials Engineering, University of Alberta, Edmonton, Canada School of Chemical Engineering, University of Queensland, Brisbane, Australia COREM, 1180 rue de la Minéralogie, Quebec City, Quebec, Canada
a r t i c l e
i n f o
Article history: Received 15 October 2009 Received in revised form 12 July 2010 Accepted 30 July 2010 Available online 5 August 2010 Keywords: Polyethylene oxide Flotation Entrainment Homo-aggregation Hetero-aggregation Photometric dispersion analysis
a b s t r a c t Polyethylene oxide (PEO) was tested to flocculate and depress fine quartz particles in the batch flotation of artificial mixtures of chalcopyrite–quartz as well as a commercial Au–Cu sulphide ore sample. The aggregation/ dispersion behaviors of quartz, chalcopyrite and their mixtures in the presence of PEO with and without potassium amyl xanthate (KAX) were studied by photometric dispersion analysis (PDA), scanning electron microscopy (SEM) and zeta potential measurements. Batch flotation results indicated that the addition of low dosages of PEO improved value mineral recovery and concentrate grade during the flotation of both the artificial mixtures of chalcopyrite–quartz and the Au–Cu sulphide ore sample. Aggregation/dispersion test results revealed that the PEO caused non-selective flocculation of quartz and chalcopyrite, forming large hetero-aggregates. However, the addition of KAX caused the chalcopyrite particles to break away from the hetero-aggregates, leading to separate homo-aggregates of quartz and chalcopyrite. The flotation of the fine chalcopyrite and the depression of the fine quartz were thus both improved. © 2010 Elsevier B.V. All rights reserved.
1. Introduction Flotation separation of fine and ultrafine mineral particles is a major challenge faced by the mineral processing industry today. The fine and ultrafine particles cause two problems both of which lower separation efficiency. One problem is the loss of value hydrophobic mineral particles due to the inefficient collection of the particles by gas bubbles, resulting in low value recovery (the recovery problem), and the other is the deterioration in concentrate grade caused by mechanical and hydraulic entrainment of hydrophilic gangue particles (the entrainment problem). Research effort over the past several decades has been focused on understanding the flotation behaviors of fine and ultrafine particles and developing processes to improve their recovery, and much less research has been focused on reducing the entrainment of the fine and ultrafine gangue minerals. A lack of effective methods to reduce fine gangue particle entrainment may be one of the reasons for the scarcity of commercial application of new techniques to improve fine particle recovery, although many such techniques have been proposed (see, for example, the summaries by Fuerstenau, 1980; Mathur et al., 2000; and Rubio, 2003).
⁎ Corresponding author. Fax: +1 780 492 2881. E-mail address: [email protected] (Q. Liu). 0301-7516/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.minpro.2010.07.009
Studies reported in the literature (e.g., Johnson et al., 1974; Trahar, 1981; Warren, 1984; Kirjavainen, 1996; Neethling and Cilliers, 2001; Zheng et al., 2006) provided good models of gangue entrainment, an understanding of the mechanisms of hydraulic and mechanical entrainment, and the effects of water recovery, aeration rate, froth thickness and pulp density on such entrainment. Parallel to this fundamental research, some studies have been carried out to lower the entrainment in froth flotation. To encourage particle drainage through the froth zone, water spray to the froth layer, vibration of the froth zone, or the use of a centrifugal force field flotation cell, have been tested (Kaya, 1989; Cheng et al., 1999; Banisi et al., 2003). Some researchers tried to reduce water recovery (thus entrainment) by increasing the flotation rate of hydrophobic particles so that value minerals could be recovered in as short a time period as possible (Akdemir et al., 2005). Mulleneers et al. (2002) modified mechanical flotation cells by adding a counter current sedimentation zone which is expected to prevent entrainment. Cao and Liu (2006) and Liu et al. (2006) showed that the entrainment of fine and ultrafine hydrophilic particles can be reduced by enlarging their particle sizes using either inorganic depressants (such as ZnSO4 which coagulated and depressed fine sphalerite (Cao and Liu, 2006)), or high molecular weight polymers (Liu et al., 2006). In this paper, we report results of a study to extend the particle sizeenlargement concept to bench mechanical flotation of artificial mixtures of chalcopyrite and quartz, as well as a commercial Au–Cu ore sample. Polyethylene oxide (PEO) was chosen to flocculate quartz particles
J. Gong et al. / International Journal of Mineral Processing 97 (2010) 44–51
because it has shown a high affinity towards quartz (Rubio and Kitchener, 1976). Although PEO is not a commonly used flocculant in mineral industry, it has been widely used in pulp and paper (van de Ven et al., 2004). Several recent studies have also shown that PEO is potentially useful in dewatering mineral and oil sands tailings (Scheiner and Smelley, 1984; Sworska et al., 2000; Mpofu et al., 2003; Wu et al., 2006). 2. Experimental 2.1. Test samples and reagents A quartz single mineral sample was obtained from Iron Ore Company of Canada. X-ray fluorescence (XRF) analysis indicated that it contains 99.9% SiO2. A chalcopyrite single mineral sample, originated from Durango, Mexico, was purchased from Ward's Natural Sciences. Inductively Coupled Plasma Mass Spectrometric (ICP-MS) analysis showed that it contains 31.8% Cu, 27.9% Fe and 39.4% S, representing a purity of 91.8% CuFeS2. The quartz and chalcopyrite single mineral samples were separately crushed to collect the −3.2 + 0.6 mm particles to be used in batch grind-flotation tests. A small portion of the −3.2 + 0.6 mm particles were dry ground in a mechanized agate mortar and pestle grinder manufactured by Fritsch GmbH. The ground mineral particles were screened and the −20 μm size fraction was collected and used in flocculation/dispersion, zeta-potential and scanning electron microscopic measurements. A commercial Au–Cu sulphide ore sample was taken from one of the member mining companies of COREM, Quebec, Canada, from the SAG mill discharge point. The sample has a prevailing particle size of about 1 mm, and contains chalcopyrite as the value mineral. The Au was associated with the chalcopyrite. The ore sample assayed 0.37 g/t Au, 0.07% Cu, 4.82% Fe and 32.2% Si. The collected sample was air-dried, riffled into 1 kg batches and stored in a freezer in individual sealed plastic bags. Two polyethylene oxide (PEO) reagents, with molecular weights of 1 million and 8 million, respectively, were obtained from Polysciences, Inc. Fresh stock solutions of PEO were prepared at a concentration of 1 g/L daily. Both PEOs were able to induce flocculation of the mineral suspensions, and the PEO with a higher molecular weight was used in some tests to facilitate experimental observations (such as in the photometric dispersion analyzer) due to the stronger flocculation it caused. Potassium amyl xanthate (KAX) and sodium isopropyl xanthate (NaIPX) were used as collectors for the sulphide minerals and were obtained from Charles Tennant & Company Ltd and Cytec Industries, respectively. Both xanthate reagents were purified by following the procedure described by DeWitt and Roper (1932). Dowfroth 250 (DF250) was obtained from Charles Tennant & Company and used as a frother. Other common chemical reagents used in the experiment, such as sodium hydroxide and hydrochloric acid, etc., were purchased from Fisher Scientific. They were used directly without further purification. 2.2. Flotation tests Batch flotation tests on quartz single mineral and synthetic mixtures of chalcopyrite and quartz were conducted in COREM labs, using a Denver laboratory flotation machine in a 1.25-L cell. Prior to flotation, 100 g of quartz or a mixture of 100 g quartz and 20 g chalcopyrite were ground with 150 mL of demineralized water in a zircon pebble mill to 80% passing 20 μm. Mill discharge was transferred to the 1.25-L flotation cell and conditioned first with PEO (molecular weight= 1 million, various dosages), then NaIPX (60 g/t) and DF250 (150 g/t) for 2 min with each reagent addition at 1200 rpm of agitation. Lime (CaO) was used to adjust pulp pH to 9 in the grinding mill and the flotation cell. The
45
flotation froth was scraped every 10 s and four rougher concentrates were collected at 0.5, 2, 4 and 8 min, and analyzed separately. Batch flotation tests on the Au–Cu sulphide ore sample were conducted in a laboratory Wemco flotation machine with a 2-L flotation cell at the University of Alberta. Prior to flotation, 1 kg of the ore sample was ground with 700 mL of Edmonton tap water in a mild steel ball mill to 90% passing 75 μm. The ground slurry was transferred to the 2-L flotation cell, conditioned for 3 min with lime addition to a pH of 10, another 3 min with PEO (molecular weight = 8 million) (if used), and finally with KAX for 3 min. This was followed by two stages of rougher flotation (2 min for the first stage and 3 min for the second stage). The combined rougher concentrate was reground for cleaner flotation, which was conducted for 6 min with and without the addition of PEO. 2.3. Flocculation/dispersion measurements The degree of flocculation or dispersion of mineral suspension was measured by a Photometric Dispersion Analyzer (PDA 2000) manufactured by Rank Brothers Inc. Photometric dispersion analysis is a sensitive technique developed to monitor the aggregation state of particle suspensions by Gregory and Nelson (1986). Compared to the traditional settling tests, it has the advantage of providing information about the change in the aggregation or dispersion state of the suspended particles, such as initial aggregation, floc growth, floc breakage and re-aggregation, etc., in real time. It has been widely used by many researchers (Gregory and Carlson, 2003; Hopkins and Ducoste, 2003; Jin et al., 2007). Its principle has been described in detail by Gregory and Nelson (1986) and Ching et al. (1994). So only a brief description will be given here. The PDA measures the intensity of light transmitted through a flowing suspension. For a well-agitated suspension, there is a small fluctuation in the intensity of the transmitted light due to the random variation in the particulate composition of the slurry flowing through the path of the light beam. A photodiode is used to monitor the transmitted light intensity. The output of the light intensity is divided into two components, a large mean value (DC) and a small fluctuating value (AC). The DC value corresponds to the average transmitted light intensity and depends on the turbidity of the flowing suspension. Compared to the DC value, the AC value is very small and can be amplified by a desired factor. The root mean square (RMS) of the AC value is then derived and divided by the DC value. The result is reported as a “Ratio” reading. The Ratio value is very sensitive to the concentration and size of the suspended particles. When aggregation occurs, the Ratio value increases. Conversely, when the suspension becomes more dispersed, the Ratio reading decreases. Therefore, the relative change in “Ratio” value is a sensitive indicator of the aggregation or dispersion state of the suspension. In each test, certain amount of minerals (−20 μm) and water were agitated using a magnetic stirrer in a 250-mL beaker and circulated through the flow cell of the PDA 2000 using a peristaltic pump. The magnetic stirring bar was 51 mm long and the stirring speed was fixed at 330 rpm. The Ratio output as a function of time was recorded by a computer connected to the PDA 2000. In such a way, the flocculation/ dispersion state of the suspension was monitored in real time following the addition of any reagents to the circulating suspension. Sworska et al. (2000) observed that hydrodynamic conditions are important factors in the flocculation of oil sands tailings by PEO. They observed that when the PEO was added to the tailings slurry under vigorous agitation (100–200 rpm in a baffled 500-mL beaker), the tailings slurry could be well flocculated. However, when the PEO solution was added to the tailings slurry in a 500-mL graduated cylinder followed by moving a plunge up and down five times, no flocculation of the tailings slurry was observed (Sworska et al., 2000). In our work, both the batch mechanical flotation machine and the magnetic stirring/ peristaltic pumping seemed to have provided sufficient agitation, and flocculation was observed in both cases. However, the hydrodynamics in the batch mechanical flotation machine was almost certainly different
46
J. Gong et al. / International Journal of Mineral Processing 97 (2010) 44–51
from those in the magnetically stirred beakers. The effect of such differences will be commented later in the “Discussion” section. 2.4. Zeta potential measurements Zeta potential measurement was performed using a ZetaPALS zeta potential analyzer manufactured by Brookhaven Instruments. All solutions and reagents were prepared with stock solutions of 10−2 mol/L KCl to maintain a constant ionic strength. To prepare the stock suspension, 1 g of the mineral (−20 μm) was added to a 1-L volumetric flask containing the 10−2 mol/L KCl solution. The flask was then sealed and inverted several times to keep the mineral particles well dispersed. For each zeta potential measurement, 10 mL of the stock mineral slurry was withdrawn and diluted with 90 mL of 10−2 mol/L KCl solution, adjusted to appropriate pH, and then treated by desired reagents. A very small amount (about 2 mL) of this conditioned particle suspension was transferred to the plastic sample cell of the ZetaPALS for zeta potential measurement.
Fig. 1. Quartz recovery as a function of water recovery in the batch flotation of pure quartz at several PEO dosages. Molecular weight of PEO is 1 million.
2.5. SEM/EDS observations A JEOL 6301F SEM scanning electron microscope (SEM) coupled with an energy dispersive X-ray spectrometer (EDS) was used to examine the quartz–chalcopyrite flocs. This machine is equipped with a PGT (Princeton Gamma-Tech) IMIX digital imaging system, a solid state silicon backscattered electron detector and a liquid nitrogen cooled silicon (lithium) detector for energy dispersive X-ray analysis. Since all the constituent elements in chalcopyrite (Cu, Fe and S) have higher atomic mass than those in quartz (Si and O), chalcopyrite particles can scatter more electrons and hence appear much brighter than quartz particles in SEM images, especially back-scattered electron images. EDS also showed that the brighter spots are mostly comprised of Cu, Fe and S, indicating that they are CuFeS2 while the majority of the gray spots are Si and O, i.e., SiO2. To prepare a SEM sample, chalcopyrite and quartz were mixed in 250-mL beakers into a 10% solid suspension, at a ratio of 10% chalcopyrite and 90% quartz. The suspension was agitated using a magnetic stirrer and the pH was adjusted to 9. Two parallel tests were conducted: one with only PEO addition, and the other with KAX addition 3 min after PEO. Small slurry samples were taken from the beakers at 1 min and 15 min after PEO addition (the 15 minute samples may have also been treated by KAX). All slurry samples were diluted 20 times with distilled water and small amount of the diluted suspension (two or three drops) was then deposited on the SEM sample stand such that the flocs would be well separated and would not overlap with each other after being dried in air. The dried samples were coated with chromium to render them conductive before being placed into the SEM sample chamber.
remained but the degree of entrainment was significantly reduced, from 0.73 without any PEO, to 0.34 at 80 g/t PEO (6.4 mg/L). The reduction of quartz entrainment was likely due to the bridging flocculation of quartz by the PEO (more on this in the next section). When the PEO flocs of quartz form, they could potentially trap sulphide mineral particles and affect their flotation. In fact, one of the primary concerns with the use of high molecular weight polymer depressants in froth flotation is that the flocs can trap value minerals and lower their recovery. That is why most of the polymeric depressants used in froth flotation are low molecular weight polymers, typically up to a few thousand grams per mole. In order to examine if the possible PEO flocs of quartz would trap sulphide mineral particles, batch flotation tests were performed on synthetic quartz–chalcopyrite mineral mixtures, which were ground together to 80% passing 20 μm. Flotation of the ground slurry was carried out at about 6 wt.% solids, using different dosages of PEO (molecular weight= 1 million), and 60 g/t NaIPX which serves as a collector for the chalcopyrite. Fig. 2 shows a recovery–grade plot of the batch flotation results. It can be seen that the recovery–grade curves moved upward as PEO dosage was increased from 0 to 100 g/t (9.6 mg/L), indicating that the use of PEO improved the separation of chalcopyrite from the quartz– chalcopyrite mixtures. PEO was then tested as a selective depressant in the flotation of a commercial Au–Cu sulphide ore sample in which quartz was the major gangue mineral. This ore sample was ground to 90% passing 75 μm. Two
3. Results 3.1. Batch flotation Batch flotation tests were first carried out using pure quartz. The quartz particles were ground to 80% passing 20 μm in a laboratory zircon pebble mill and floated in a 1.25-L flotation cell with the addition of 60 g/ t NaIPX, 150 g/t of DF250 and varying dosages of PEO (molecular weight= 1 million) at 6 wt.% solids. Since no collector for quartz was used, quartz recovery in these tests was considered purely a result of entrainment. Fig. 1 shows the relationship between quartz recovery and water recovery at different PEO dosages. As can be seen, when no PEO was used, quartz recovery increases linearly with water recovery, consistent with the observations of other researchers (e.g., Trahar, 1981). The slope of the line, variously called degree of entrainment or entrainment factor (Smith and Warren, 1989), can be used to indicate the severity of entrainment. After adding PEO, the linear relationship
Fig. 2. Recovery–grade relationship of the batch flotation of synthetic chalcopyrite– quartz mixtures. Molecular weight of PEO is 1 million.
J. Gong et al. / International Journal of Mineral Processing 97 (2010) 44–51
Fig. 3. Gold and copper recovery as a function of Si content in rougher and cleaner concentrates in the batch flotation of a commercial Au–Cu sulphide ore. Molecular weight of PEO is 8 million.
sets of batch flotation tests were carried out, both with rougher flotation followed by cleaner flotation. In one set of tests, PEO (molecular weight= 8 million) was used in both rougher and cleaner flotation, while in the other, no PEO was used in either rougher or cleaner. The flotation was conducted at about 30 wt.% solids and a pulp pH of 10.3, with 160 g/t KAX (potassium amyl xanthate) as a collector for the sulphide minerals and gold. In the tests in which PEO was used, 10 g/t was added in rougher and 5 g/t was added in the cleaner. Fig. 3 shows the flotation results, with Cu and Au recovery plotted as a function of Si content in the rougher and cleaner concentrates. As can be seen, higher Cu and Au recoveries could be achieved at the same Si content in both the rougher and cleaner concentrates when PEO was used. This means that at the same Cu and Au recovery, the Si contents in the rougher and cleaner concentrates were much lower when PEO was used. Therefore, the addition of PEO had the benefit of improving the concentrate grade while maintaining or increasing copper and gold recovery during the flotation of the Au–Cu ore sample. 3.2. Flocculation measurement with PDA The batch flotation results show that the addition of PEO can reduce quartz entrainment and improve concentrate grade in chalcopyrite and sulphide ore flotation. In order to confirm that the lowering of quartz entrainment is due to the bridging flocculation by PEO, a series of photometric dispersion analyzer (PDA) measurements were performed. 3.2.1. Flocculation of 1:1 quartz–chalcopyrite mixtures The first set of PDA measurements was carried out on suspensions of chalcopyrite and quartz mixtures (50 wt.% quartz and 50 wt.% chalcopyrite). The suspension was prepared at 1 wt.% solids and pH 9, and was agitated with a magnetic stir bar (51 mm long) at 330 rpm while being circulated through the PDA 2000 photometric dispersion analyzer. Three tests were conducted. In the first test, only PEO (molecular weight= 8 million) was added (to a final concentration of 10 mg/L). In the next two tests, KAX (300 mg/L) was added 3 min before or after the addition of PEO while the Ratio output was continuously monitored. The results are presented in Fig. 4, where the points of PEO and KAX additions are noted. As can be seen from Fig. 4, in all tests, the addition of PEO caused an immediate increase in the Ratio value by about 5 times, indicating that the particles were flocculated by PEO. A typical SEM image of the suspension, shown in Fig. 5, confirms the presence of large flocs and it also shows that chalcopyrite and quartz particles formed hetero-flocs, i.e., they were flocculated together.
47
Fig. 4 shows that after reaching the maximum value, the Ratio reading in all three cases started to drop quickly. This means that the flocs were not strong and they were easy to break by the mechanical forces such as magnetic stirring and the peristaltic pumping. The rate at which the Ratio value drops can be taken as an indication of the breakage rate of the flocs. Since the initial part of the dropping segment of the curve appeared to be linear, the slope of that segment was calculated to indicate the dropping rate of the Ratio reading. For the case where only PEO was added, the slope in the initial dropping segment was found to be −0.49 min–1. When KAX was added 3 min after PEO, the slope increased to −0.66 min–1, indicating a faster floc breakage with the addition of KAX. When KAX was added 3 min before PEO, the Ratio reading increased from 1 to about 2.2 immediately following the KAX addition. This increase was most likely caused by the aggregation of the chalcopyrite due to its surface hydrophobicity induced by the adsorbed KAX. Subsequent addition of PEO 3 min later resulted in a further increase of the Ratio reading from about 2.2 to about 5, followed by a sharp drop. In fact, the slope of the dropping segment of the Ratio curve in this case was the highest, at −0.95 min–1. The above PDA measurement results reveal that PEO caused unselective hetero-aggregation of the quartz and chalcopyrite particles. The flocs were not strong and tend to break as a result of magnetic stirring and peristaltic pumping. The addition of KAX accelerated the breakage of these hetero-aggregates, especially when KAX was added ahead of PEO. 3.2.2. Flocculation of quartz and chalcopyrite single minerals To examine the roles that KAX plays in the quartz–chalcopyrite mixture system, its effect on quartz and chalcopyrite single mineral flocculation was studied. The test procedures were the same as above except that only quartz or chalcopyrite was used in each test. Fig. 6 shows the results of pure quartz particle suspension at about 5 wt.% solids (i.e., about 5 times more concentrated than the quartz– chalcopyrite mixture tests) at pH 9. As can be seen, when only KAX was added, there was no change in the Ratio reading. This is as expected since xanthate is known to have no interactions with quartz. However, when PEO was added to a concentration of 10 mg/L, there was an abrupt increase in the Ratio reading from nearly zero to close to 13 (the much higher Ratio value was probably due to the higher particle concentration). The quartz was flocculated by the PEO. These flocs were not stable, and the Ratio reading dropped quickly after reaching the maximum. Adding KAX after PEO did not seem to have caused any difference in the Ratio reading. Therefore, it can be concluded that KAX has no effect on quartz particle flocculation by PEO.
Fig. 4. Effect of KAX and PEO on the flocculation/dispersion of 1:1 quartz–chalcopyrite mixtures. Molecular weight of PEO is 8 million. The point of PEO or KAX addition is noted by a vertical arrow.
48
J. Gong et al. / International Journal of Mineral Processing 97 (2010) 44–51
Fig. 6. Effect of KAX and PEO on the flocculation/dispersion of quartz. Molecular weight of PEO is 8 million. The point of PEO or KAX addition is noted by a vertical arrow.
seemed to have been returned to the initial dispersed state when only PEO was used. When KAX was added 3 min after PEO, the breakage rate (slope) increased to a much higher value at −5.46 min–1, indicating that KAX sped up the breakage of the chalcopyrite–PEO flocs. However, it is interesting to note that after this initial sudden drop following KAX addition, the Ratio reading then stabilized at about 3.0 rather than dropping continuously to the original value of 1.0. It is speculated that KAX caused the breakage of the PEO flocs of chalcopyrite, but some of the chalcopyrite particles may have been re-flocculated by KAX, for the formed aggregates seem to be resistant to breakage, similar to the case when KAX was added alone. When KAX was added first, followed by PEO 3 min later, there was an initial small and gradual increase in the Ratio reading from 1 to about 2.7, again possibly by the flocculation of the hydrophobized chalcopyrite. This was followed by a sudden increase of the Ratio reading to above 5 after adding the PEO. The Ratio reading quickly dropped, with a slope of −1.35 min–1, and finally stabilized at about 2.5.
3.2.3. Flocculation of quartz–chalcopyrite mixtures with varying amounts of chalcopyrite The foregoing description of the PDA test results seem to point to the same trend, that the addition of KAX promotes the formation of breakage-resistant chalcopyrite–KAX flocs, and breaks chalcopyrite– PEO flocs. However, the addition of KAX does not affect quartz nor the quartz–PEO flocs. Therefore, it seems that in the hetero-aggregates of Fig. 5. (a) SEM image of a typical PEO floc of quartz–chalcopyrite. (b) EDS spectrum of point (1): chalcopyrite, and (c) EDS spectrum of point (2): quartz. Molecular weight of PEO is 8 million.
However, as can be seen from Fig. 7, the phenomenon was very different in the chalcopyrite single mineral system. With the addition of KAX to a concentration of 300 mg/L, there was an initial sudden increase of the Ratio reading of the 0.5 wt.% chalcopyrite suspension from 1 to about 2, followed by a steady increase to about 5. The Ratio reading never decreased in the 30-min recording period, as have been observed in the tests with quartz–chalcopyrite mixtures or the quartz single mineral. Clearly, the KAX caused the aggregation of the chalcopyrite particles and the aggregates were more resistant to breakage than the PEO flocs. Fig. 7 also shows that when PEO was added, chalcopyrite particles were flocculated and the Ratio reading increased from 1 to over 5, but the flocs were not stable and broke up gradually. The breakage rate, i.e., the slope of the dropping segment of the Ratio curve, was calculated to be −0.84 min–1. The Ratio reading continued the dropping trend, reaching the initial value of about 1.0 in 30 min, i.e., the chalcopyrite particles
Fig. 7. Effect of KAX and PEO on the flocculation/dispersion of chalcopyrite. Molecular weight of PEO is 8 million. The point of PEO or KAX addition is noted by a vertical arrow.
J. Gong et al. / International Journal of Mineral Processing 97 (2010) 44–51
quartz and chalcopyrite formed by the addition of PEO, the chalcopyrite particles are the “weak spots” once KAX was added. The xanthate anions are known to adsorb on the chalcopyrite particles through chemical and/or electrochemical reactions (Fuerstenau, 1982; Mielczarski et al., 1996; Woods, 1984). Such a strong chemisorption may have caused the desorption of the PEO from the chalcopyrite surfaces, leading to the disintegration of the quartz–chalcopyrite hetero-aggregates. The removed chalcopyrite particles, however, could form hydrophobic chalcopyrite–KAX flocs due to the surface hydrophobicity induced by KAX. This hypothesis seems to be borne out by the results shown in Fig. 8. This figure shows PDA measurement results on quartz– chalcopyrite mixtures with different chalcopyrite contents. As can be seen, the slopes of the decreasing segment of the Ratio curves increased with increasing chalcopyrite content after adding KAX, at −0.15 min–1 for 10 wt.%, −0.7 min–1 for 30 wt.%, and −5.46 min–1 for 100 wt.% chalcopyrite. Also, the final Ratio readings were at about 1.0, 1.5 and 2.5 for the mixtures with 10 wt.%, 30 wt.% and 100 wt.% chalcopyrite, respectively. In other words, after adding KAX, the more chalcopyrite the quartz–chalcopyrite mixture contains, the faster the PEO flocs break, and the higher the final Ratio reading. 4. Discussion To understand the observed batch flotation and PDA flocculation/ dispersion test results, it is envisaged that the following may be what happened during the flotation process: when PEO was added to the quartz–chalcopyrite mixture, it flocculated both minerals together to form hetero-aggregates. With continuous stirring, especially in a batch mechanical flotation cell, some of the flocs were broken. When KAX was added after PEO addition, it adsorbed on chalcopyrite surfaces and partially or completely (depending on KAX dosages) replaced the PEO, which further weakens the hetero-aggregates. As a result, chalcopyrite particles disassociated from the hetero-aggregates, causing further breakup of the flocs. This postulate is consistent with the two generally accepted flocs breakage mechanisms, i.e., large-scale fragmentation, and surface erosion (Pandya and Spielman, 1982; Jarvis et al., 2005). Fragmentation is thought to be caused by tensile stress across the entire floc, while surface erosion is attributed to the shearing stress on the floc surface. In this case, the desorption of PEO by KAX renders chalcopyrite particles on the surface less resistant to the shearing stress so chalcopyrite particles are stripped off from the flocs more easily (note that the chalcopyrite particles in the hetero-flocs were scattered (see Fig. 5) and hydrophobic forces will not come into play even after the KAX has adsorbed on the chalcopyrite particles). The flocs become
Fig. 8. Effect of PEO and KAX on the flocculation/dispersion of quartz–chalcopyrite mixtures with different chalcopyrite content. Molecular weight of PEO is 8 million. The point of PEO or KAX addition is noted by a vertical arrow.
49
weaker and the fragmentation worsens, which in turn exposes more chalcopyrite particles to shearing stress, eventually leading to the disintegration of the large PEO flocs, and forming smaller flocs that contain quartz mostly. The individual chalcopyrite particles then associate to form hydrophobic aggregates under agitation, due to the surface hydrophobicity caused by the adsorption of KAX. Fig. 9 shows two SEM images that demonstrate such an effect of KAX. Both images show samples that were taken from the quartz–chalcopyrite mixture suspensions 15 min after PEO was added, but in the sample shown in Fig. 9b, KAX was added 3 min after PEO. The quartz– chalcopyrite mixtures contained 10 wt.% chalcopyrite. Fig. 9a and b shows typical field of view of these two samples. The most obvious difference between these two images is that large hetero-aggregates existed in the image shown in Fig. 9a where only PEO was added. These large flocs were not observed in Fig. 9b where both PEO and KAX were used. Another difference is the way by which chalcopyrite particles are incorporated in the flocs. In Fig. 9a (with no KAX), chalcopyrite particles are mostly incorporated in large and medium size flocs. In Fig. 9b (with KAX), the small flocs are concentrated in either chalcopyrite or quartz. In other words, the large hetero-aggregates formed by the addition of PEO were broken as a result of subsequent KAX addition, to form smaller homo-aggregates of quartz and chalcopyrite. Even though the quartz flocs were broken under agitation and the action of KAX, their sizes are still larger than the single dispersed quartz particles. Therefore, the quartz flocs are less likely to be hydraulically carried over to the concentrate, and some of them may be large enough to overcome the fluid drag and drain
Fig. 9. SEM images of suspension samples taken 15 min after the addition of PEO, (a) without KAX; (b) KAX added 3 min after PEO. Bright particles: chalcopyrite; gray particles: quartz. Molecular weight of PEO is 8 million.
50
J. Gong et al. / International Journal of Mineral Processing 97 (2010) 44–51
back to the pulp from the froth layer, i.e., the entrainment of quartz is reduced. In fact, the zeta potential measurement results reveal the different competitive adsorption behaviors of PEO and KAX on the mineral surface. As can be seen from Fig. 10, PEO was able to adsorb on both quartz and chalcopyrite surfaces, driving the zeta potentials of both minerals to near zero. As PEO is a nonionic polymer, its adsorption could stretch the shear plane of the electrical double layer on the mineral surface far away from the surface, lowering the magnitude of the zeta potential. On the other hand, the addition of KAX only altered the zeta potentials of chalcopyrite, making it more negatively charged as more KAX was added, and it did not affect the zeta potential of quartz. In fact, it is known that xanthate anions can chemisorb on chalcopyrite surfaces through the formation of surface metal xanthate compounds, giving rise to a negative surface charge. In this context, it is KAX that has lent selectivity to this system. Since KAX (xanthate) is a specific collector for all sulphide minerals, the combined use of xanthate and PEO should benefit any sulphide ore flotation with rejection of quartz and silicate gangue by lowering its mechanical entrainment. It is realized that the hydrodynamic conditions in a batch mechanical flotation machine are different from a suspension that undergoes magnetic-bar stirring and peristaltic pumping. Whether the same aggregation and dispersion behavior of the quartz–chalcopyrite mixtures with the addition of PEO and xanthate as observed in the PDA tests occurred in the batch mechanical flotation machine is yet to be verified, especially in view of the “fragile” nature of the PEO flocs. However, the improved flotation results as shown by the grade– recovery relationship after the addition of PEO serve as an indication that similar aggregation or dispersion behavior may have happened in the mechanical flotation machine. In fact, our tests using a 3-L baffled container agitated at several hundred rpm showed similar aggregation behavior of xanthate and PEO on quartz and galena. In this context, it is important to note that Sworska et al. (2000) reported that a relatively strong agitation (hydrodynamic conditions) was required when the PEO solution was introduced to the oil sands tailings to cause the flocculation of the latter, and Ding and Laskowski (2007) showed that a strong agitation broke the large polyacrylamide flocs and caused the formation of selectively flocculated smaller flocs of coal and gangue particles. Clearly, the pulp hydrodynamic condition is an important operating parameter in controlling the flocculation behavior of PEO and the selectivity of the flocculation. This work described the chemical effect of xanthate on breaking PEO flocs of chalcopyrite and quartz. The effect could be further enhanced by different hydrodynamics. In fact, hydrodynamic condition is an important factor that will be systematically studied. This is because, unlike tailings dewatering, large flocs are not necessarily desirable when polymers are used in a froth flotation machine with the intention to lower hydraulic entrainment.
Fig. 10. Effect of KAX and PEO on the zeta potentials of chalcopyrite and quartz. Molecular weight of PEO is 8 million.
5. Conclusions Polyethylene oxides (PEO), with molecular weights of 1 million or 8 million, were found to be able to lower the mechanical entrainment of quartz in batch flotation. When PEO was used in the batch flotation of a commercial Au–Cu sulphide ore sample, it also reduced quartz entrainment and improved copper and gold recovery. As a result, rougher and cleaner concentrates with higher grade and higher Cu and Au recovery were obtained after the addition of PEO. Aggregation/ dispersion studies using a photometric dispersion analyzer (PDA) and a scanning electron microscope (SEM), coupled with zeta potential measurements, showed that during the PEO-assisted flotation, PEO flocculated quartz and chalcopyrite together, forming hetero-aggregates. Subsequent addition of a xanthate collector caused the chalcopyrite particles to break away from the PEO hetero-aggregates, due to the competitive adsorption of xanthate which probably removed PEO from the chalcopyrite surfaces. The removed chalcopyrite particles were able to form homo-aggregates, possibly due to the surface hydrophobicity induced by the adsorbed xanthate. The chalcopyrite–xanthate homo-aggregates were more resistant to breakage than the chalcopyrite–PEO flocs. The combined use of PEO and xanthate could therefore potentially benefit the froth flotation of polymetallic sulphide ores to reject fine and ultrafine quartz and silicate gangue. Acknowledgements This project is supported by the Natural Sciences and Engineering Research Council of Canada (NSERC) through a discovery grant, and by COREM, Quebec and NSERC through a collaborative research and development (CRD) project grant (2005–2007). References Akdemir, U., Guler, T., Yildiztekin, G., 2005. Flotation and entrainment behavior of minerals in talc–calcite separation. Scand. J. Metall. 34, 241–244. Banisi, S., Sarvi, M., Hamidi, D., Fazeli, A., 2003. Flotation circuit improvements at the Sarcheshmeh copper mine. Trans. Inst. Min. Metall. C. 112, C198–C205. Cao, M., Liu, Q., 2006. Reexamining the functions of zinc sulfate as a selective depressant in differential sulfide flotation—the role of coagulation. J. Colloid Interface Sci. 301, 523–531. Cheng, H., Cai, C., Zhang, X., Zhang, J., 1999. A highly selective flotation cell with oscillating separator. In: Xie, H., Golosinski, T.S. (Eds.), Mining Science and Technology '99. A.A. Balkema, Rotterdam/Brookfield, pp. 551–553. Ching, H.W., Tanaka, T.S., Elimelech, M., 1994. Dynamics of coagulation of kaolin particles with ferric chloride. Water Res. 28, 559–569. DeWitt, C.C., Roper, E.E., 1932. The surface relations of potassium ethyl xanthate and pine oil. I. J. Am. Chem. Soc. 54, 444–455. Ding, K.J., Laskowski, J.S., 2007. Effect of conditioning on selective flocculation with polyacrylamide in the coal reverse flotation. Trans. Inst. Min. Metall. C. 116, C108–C114. Fuerstenau, D.W., 1980. Fine particle flotation. In: Somasundaran, P. (Ed.), Fine Particle Processing. AIME, New York, pp. 669–705. Fuerstenau, M.C., 1982. Adsorption of sulphydryl collectors. In: King, P. (Ed.), Principle of Flotation: South African Institute of Mining and Metallurgy Monograph Series, 3, pp. 91–108. Johannesburg. Gregory, D., Carlson, K., 2003. Relationship of pH and floc formation kinetics to granular media filtration performance. Environ. Sci. Technol. 37, 1398–1403. Gregory, J., Nelson, D.W., 1986. Monitoring of aggregates in flowing suspensions. Colloids Surf. 18, 175–188. Hopkins, D.C., Ducoste, J.J., 2003. Characterizing flocculation under heterogeneous turbulence. J. Colloid Interface Sci. 264, 184–194. Jarvis, P., Jefferson, B., Gregory, J., Parsons, S.A., 2005. A review of floc strength and breakage. Water Res. 39, 3121–3137. Jin, P.K.K., Wang, X.C.C., Chai, H.X., 2007. Evaluation of floc strength by morphological analysis and PDA online monitoring. Water Sci. Technol. 56, 117–124. Johnson, N.W., McKee, D.J., Lynch, A.J., 1974. Flotation rates of nonsulfide minerals in chalcopyrite flotation process. Trans. AIME 256, 204–226. Kaya, M., 1989. Froth washing in mechanical flotation cells. Ph. D. Thesis, McGill University, Montreal. Kirjavainen, V.M., 1996. Review and analysis of factors controlling the mechanical flotation of gangue minerals. Int. J. Miner. Process. 46, 21–34. Liu, Q., Wannas, D., Peng, Y., 2006. Exploiting the dual functions of polymer depressants in fine particle flotation. Int. J. Miner. Process. 80, 244–254. Mathur, S., Singh, P., Moudgil, B.M., 2000. Advances in selective flocculation technology for solid–solid separations. Int. J. Miner. Process. 58, 201–222.
J. Gong et al. / International Journal of Mineral Processing 97 (2010) 44–51 Mielczarski, J.A., Mielczarski, E., Cases, J.M., 1996. Interaction of amyl xanthate with chalcopyrite, tetrahedrite, and tennantite at controlled potentials. Simulation and spectroelectrochemical results for two-component adsorption layers. Langmuir 12, 6521–6529. Mpofu, P., Addai-Mensah, J., Ralston, J., 2003. Investigation of the effect of polymer structure type on flocculation, rheology and dewatering behaviour of kaolinite dispersions. Int. J. Miner. Process. 71, 247–268. Mulleneers, H.A.E., Koopal, L.K., Bruning, H., Rulkens, W.H., 2002. Selective separation of fine particles by a new flotation approach. Sep. Sci. Technol. 37, 2097–2112. Neethling, S.J., Cilliers, J.J., 2001. Simulation of the effect of froth washing on flotation performance. Chem. Eng. Sci. 56, 6303–6311. Pandya, J.D., Spielman, L.A., 1982. Floc breakage in agitated suspensions — theory and data-processing strategy. J. Colloid Interface Sci. 90, 517–531. Rubio, J., 2003. Unconventional flocculation and flotation techniques. In: Ralston, J., Miller, J., Rubio, J. (Eds.), Flotation and Flocculation: From Fundamentals to Applications. Proceedings from Strategic Conference and Workshop, Hawaii, pp. 17–32. Rubio, J., Kitchener, J.A., 1976. Mechanism of adsorption of poly(ethylene oxide) flocculant on silica. J. Colloid Interface Sci. 57, 132–142. Scheiner, B.J., Smelley, A.G., 1984. Dewatering of fine particle mining wastes using polyethylene oxide flocculant. Miner. Metall. Process. 1, 71–75.
51
Smith, P.G., Warren, L.J., 1989. Entrainment of particles into flotation froths. In: Laskowski, J.S. (Ed.), Frothing in Flotation. Gorgan and Breach, Glasgow, pp. 123–145. Sworska, A., Laskowski, J.S., Cymerman, G., 2000. Flocculation of the Syncrude fine tailings part II. Effect of hydrodynamic conditions. Int. J. Miner. Process. 60, 153–161. Trahar, W.J., 1981. A rational interpretation of the role of particle-size in flotation. Int. J. Miner. Process. 8, 289–327. van de Ven, T.G.M., Qasaimeh, M.A., Paris, J., 2004. PEO-induced flocculation of fines: effects of PEO dissolution conditions and shear history. Colloids Surf. A. 248, 151–156. Warren, L.J., 1984. Ultrafine particles in flotation. In: Jones, M.H., Woodcock, J.T. (Eds.), Principles of Mineral Flotation. Australian IMM, Melbourne, pp. 185–214. Woods, R., 1984. Electrochemistry of sulfide flotation. In: Jones, M.H., Woodcock, J.T. (Eds.), Principles of Mineral Flotation, The Wark Symposium: Australian Institute of Mining and Metallurgy, Symposia Series No. 40, Victoria, Australia, pp. 91–115. Wu, M.R., Paris, J., Van De Ven, T.G.M., 2006. Polyethylene oxide induced fines flocculation and retention: from bench top experiments to paper machine performance. Nord. Pulp Pap. Res. J. 21, 646–652. Zheng, X., Johnson, N.W., Franzidis, J.P., 2006. Modelling of entrainment in industrial flotation cells: water recovery and degree of entrainment. Miner. Eng. 19, 1191–1203.