- Email: [email protected]
Poly (N-isopropylacrylamide) (PNIPAM) as a flotation collector: Effect of temperature and molecular weight
Minerals Engineering 23 (2010) 921–927
Contents lists available at ScienceDirect
Minerals Engineering journal homepage: www.elsevier.com/locate/mineng
Poly (N-isopropylacrylamide) (PNIPAM) as a flotation collector: Effect of temperature and molecular weight Elizaveta Burdukova a, Haihong Li b, Dee J. Bradshaw c, George V. Franks a,* a
Australia Mineral Science Research Institute, Department of Chemical and Biomolecular Engineering, The University of Melbourne, Victoria 3010, Australia Alberta Research Council, Alberta Innovates, 250 Karl Clark Road, Edmonton, Alberta, AB, Canada c Julius Kruttschnitt Mineral Research Centre, University of Queensland, Queensland 4072, Australia b
a r t i c l e
i n f o
Article history: Received 15 December 2009 Accepted 6 March 2010 Available online 31 March 2010 Keywords: Mineral processing Flotation reagents Flotation collectors Fine particle processing
a b s t r a c t Poly (N-isopropylacrylamide) (PNIPAM), a temperature responsive polymer, was tested for its potential use as a collector in a quartz flotation system. The effect of PNIPAM on the surface characteristics of quartz particles were studied using induction time, contact angle and zeta potential measurement and analysed in terms of the probability of bubble/particle attachment and the probability of formation of stable bubble/particle aggregates. It was found that probability of bubble/particle attachment of quartz significantly increases in the presence of PNIPAM, particularly at temperatures above the lower critical solution temperature (LCST) of 32 °C. Furthermore, the probability of bubble/particle attachment increases with increasing PNIPAM molecular weight. This was attributed to the increased hydrophobicity of the quartz surface as well as the decrease in the double layer repulsion between bubbles and particles. This leads to the conclusion that PNIPAM could act as an effective collector in a flotation system. Ó 2010 Elsevier Ltd. All rights reserved.
1. Introduction Poly (N-isopropylacrylamide) (PNIPAM) is a temperature responsive polymer. At temperature below its lower critical solution temperature (LCST) of 32 °C, PNIPAM is soluble in water and hydrophilic. The polymer macromolecules are hydrated and fully extended, with extensive intermolecular hydrogen bonding between water molecules and polymer chains (Saunders et al., 1999; Sakohara et al., 2002; Sun et al., 2004). At temperatures above the LCST, the hydrogen bonds between water molecules and polymer chains are broken and instead, intramolecular and intermolecular hydrogen bonds between the C@O and N–H groups are formed. The formation of intramolecular bonds causes the molecule to coil up, exposing its hydrophobic core. This renders the polymers hydrophobic and insoluble in water (Saunders et al., 1999; Sakohara et al., 2002; Sun et al., 2004). The studies of PNIPAM in the context of mineral processing applications, originate in its potential use as a temperature responsive flocculant at temperatures above the LCST. It can be used as a reversibly responsive flocculant and dispersant (Guillet et al., 1985; Deng et al., 1996; Sakohara et al., 2002; Sakohara and Nishikawa, 2004; Franks, 2005; Li et al., 2007, 2009). The reversibility of the state of aggregation and dispersion can be * Corresponding author. E-mail address: [email protected] (G.V. Franks). 0892-6875/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.mineng.2010.03.003
used to produce both rapid sedimentation (at temperatures above the LCST) and enhance sediment consolidation and dewatering (at temperatures below the LCST). At high temperature, the inter-particle forces in a mineral suspension are changed from repulsive to attractive; causing the formation of hydrophobic particle aggregates (flocs). After the flocs settle, the temperature can be reduced, which produces repulsion between particles as the polymer becomes hydrophilic. The repulsion allows for additional consolidation of the mineral sediment leading to improved water recovery. As well as being an effective flocculant/dispersant, preliminary investigations demonstrated that PNIPAM can be an effective collector in the flotation of oxide minerals (Franks et al., 2009). In current flotation practice, flotation of oxide minerals is generally achieved using short chain surfactant collectors (e.g. dodecyl amine and sodium dodecyl sulphate) (Quast, 2000; Araujo et al., 2005). Preliminary testing in a simple flotation system showed that at temperatures above the LCST, the presence of PNIPAM induced flotation of both silica and kaolinite, achieving recoveries significantly greater than those in the presence of dodecyl amine (Franks et al., 2009). The ability of PNIPAM to induce both flocculation and flotation of previously dispersed hydrophilic particles (at temperatures above the LSCT) potentially makes it a highly valuable reagent in the flotation of ultrafine mineral particles. Furthermore, PNIPAM is a non-toxic polymer, which can potentially provide an environmentally friendly substitute to industrial collectors which
922
E. Burdukova et al. / Minerals Engineering 23 (2010) 921–927
are often environmentally harmful (Davis et al., 1976). However, there is still insufficient evidence to determine whether or not PNIPAM would present a commercially viable substitute for collectors currently used in industry. The likelihood that a mineral particle is floatable is generally expressed in terms of probabilities, as shown in Eq. (1), where P is the overall flotation probability, PC is the probability of bubble/particle collision, PA is the probability of bubble/particle attachment and PS is the probability of formation of a stable bubble/particle aggregate (Laskowski, 1986; Ralston, 1992).
P ¼ PC PA PS
ð1Þ
The probability of collision (PC) is strongly related to hydrodynamic factors such as cell turbulence, pulp viscosity and particle size (Schulze, 1984). As such, collision probability is unlikely to be influenced by the presence of collector in the flotation system, and is not considered further. The probability of formation of stable bubble/particle aggregates (PS) is a very strong function of the magnitude of the three phase contact angle between the air bubble, mineral particle and aqueous medium, which directly determines the strength of the bubble/particle bond (Laskowski, 1986). The increased strength of this bond determines how likely a particle is to detach from a bubble. For this reason, in some publications this term (PS) is replaced by the probability of bubble/particle detachment (PD), where PS is inversely related to PD (Laskowski, 1986; Ralston, 1992). The probability of bubble/particle attachment (PA) is largely dependent on the surface forces existing between a bubble and a particle. It has been shown that in an aqueous medium, the interactions between a negatively charged mineral particle and a bubble are similar to those between two negatively charged mineral particles (Derjaguin, 1954; Laskowski, 1986). Therefore, the probability of bubble particle attachment is subject to double layer forces, as described by the modified DLVO theory, mainly: van der Waals forces, electrical double layer repulsion and hydrophobic attraction. It follows that the probability of bubble particle attachment will decrease with increasing double layer repulsion between mineral particles and bubbles and increase with increasing particle hydrophobicity. Induction time measurements are one way of characterising the probability of particle/bubble attachment (Nguyen et al., 1997). Induction time is defined as the time for the thinning of the intervening liquid film between an air bubble and a hydrophobic particle to a critical thickness at which the film will rupture spontaneously (Laskowski, 1974). Induction time is measured by bringing captive bubbles into contact with a bed of particles for a controlled time interval (typically between 10 and 1500 ms), to determine whether or not mineral particles attach to bubbles such that they can be lifted up from a particle bed. The particle/bubble contacts are repeated numerous times, estimating the probability of bubble/particle pickup at each contact time. The contact time corresponding to 50% probability of particle pickup is termed the induction time. This technique allows for a simple and reliable estimate of the strength of particle/bubble attachment (Eigeles and Volova, 1968; Yordan and Yoon, 1986, 1988; Holuszko et al., 2008; Burdukova and Laskowski, 2009). The purpose of this work is to study the effect of temperature sensitive PNIPAM on the probability of attachment of quartz particles to air bubbles, using induction time measurements. These measurements are then interpreted by characterising the particle surfaces in terms of their hydrophobicity (as indicated by contact angle measurements) and electrical double layer repulsion (as indicated by zeta potential measurements).
2. Experimental details 2.1. Materials 2.1.1. Temperature sensitive polymers A temperature sensitive, non-ionic polymer – poly (N-isopropylacrylamide) (PNIPAM), was used in this work. Three molecular weights of this polymer were used, termed: low, medium and high. The low molecular weight (0.23 MDa) PNIPAM was purchased from Sigma–Aldrich, Australia. The PNIPAMs with the medium molecular weight (1.32 MDa) and the high molecular weight (3.6 and 4.5 MDa) were synthesized in our laboratories as described in detail elsewhere (O’Shea et al., 2007). The synthesis process of these high molecular weight polymers is such that relatively small quantities of the polymer can be produced per batch and it is not easy to obtain consistent molecular weights of polymers between batches. For this reason the high molecular weight polymer used in the different sections of the work varies in molecular weight between 3.6 and 4.5 MDa. In the case of contact angle measurements, the PNIPAM with medium molecular weight (2.0 MDa) was purchased from Polymer Source Inc., Canada. 2.1.2. Quartz Quartz powder (Silica 400G) was obtained from UNIMIN Australia Limited for use in zeta potential measurements. The particles had a size distribution of d50 10 lm, and a BET surface area of 1.7 m2 g1. The powder contained 99.6% SiO2 and traces of alumina, ferric oxide, titania and lime. For the use in induction time measurements, quartz particles (Silica 100 WQ) were obtained from UNIMIN Australia Limited. The particles were dry sieved to obtain a particle size distribution of 90 lm < Dp < 150 lm. The particles contained 99.5% SiO2 and traces of alumina, ferric oxide, titania and lime. 3. Methods 3.1. Contact angle measurement The sessile drop technique was used for the measurement of advancing contact angle, using the OCA Dataphysics measurement system, provided by the Department of Chemistry, University of Melbourne. Silica glass surfaces were cleaned by first washing them in surfactant solution followed by dilute NaOH, then finally rinsing with deionised water. For each molecular weight PNIPAM, three clean silica glass slides (26 76 mm) were coated with polymer solutions so as to produce the surface concentration of 0.1 g of polymer per m2 of glass surface. The slides were dried in an oven at 80 °C for about 30 min. The temperature of the glass slides during contact angle measurements was controlled by placing the slides on the top surface of an aluminium stage which has its temperature controlled by a water bath in order to achieve the temperature of either 22 °C (below the LCST) or 50 °C (above the LCST). Five minutes were allowed to make sure that the slide had reached the desired temperature. A drop of deionised water at pH 6, with 3 lL volume, was introduced onto the slide through a computer controlled micro-syringe. The measurement was taken 10 s after the liquid drop was put on the glass slide. The measurements were conducted in triplicate. 3.2. Zeta potential measurements Zeta potential measurements were performed using the Zeta Acustosizer, manufactured by Colloidal Dynamics Inc., Sydney, Australia. The instrument is equipped with a recirculating water bath, which allowed for the measurement being performed at both 22 °C and 50 °C. Suspensions of quartz (5 wt.%) were prepared using deionised water containing 40 ppm of PNIPAM (corresponding to
E. Burdukova et al. / Minerals Engineering 23 (2010) 921–927
923
400 g ton1 of quartz) at a variety of molecular weights. The suspensions were allowed to stand for 24 h prior to measurement. The measurements were performed at pH 8, using 1 M NaOH solution for pH adjustment, with the background electrolyte of 102 KCl. 3.3. Induction time measurements Quartz particles were treated with PNIPAMs with a variety of molecular weights, prior to induction time measurement. Ten grams of quartz were placed in a beaker containing 100 mL of distilled water adjusted to pH 8 using 1 M NaOH solution for pH adjustment, with the background electrolyte of 102 KCl. The beaker was immersed in a water bath controlled to either 22 °C or 50 °C. Each suspension was dosed with 40 ppm of PNIPAM, which corresponded to 400 g of polymer per ton of quartz. The suspensions were allowed to equilibrate for 1 h, while being continuously agitated with a magnetic stirrer. After 1 h, the solids were filtered out and gently washed with distilled water adjusted to the appropriate temperature. The PNIPAM coated solids were placed in the oven to dry at 70 °C. Induction time measurements were performed with the MCT 100 Induction Time Meter, provided by the Julius Kruttschnitt Mineral Research Centre, University of Queensland. The temperature of the measurement was kept controlled by a small water jacket, fitted around a measurement vessel. Water from an adjacent recirculating water bath was pumped though the water jacket, maintaining a constant temperature of either 22 °C (below the LCST) or 50 °C (above the LCST). The photograph of the test apparatus is shown in Fig. 1. Approximately 2 g of solids were placed in the measurement vessel (enough to cover the vessel bottom) and covered with distilled water adjusted to the desired temperature and pH 8. The bubble was brought into contact with the particle bed at set contact times ranging between 10 and 1200 ms. For each contact time, the probability of particle pickup was estimated by counting the number of times particles were picked up by a bubble out of 40– 80 contacts. The induction time was estimated as the contact time at which the pickup probability was 50%. The error in the induction time value was estimated by calculating the 95% confidence interval of the linear regression of contact times as a function of pickup probability as shown in Fig. 2 (Weisberg, 2005). 4. Results and discussion 4.1. Effect of PNIPAM temperature on particle/bubble attachment The effect of temperature on particle/bubble attachment in the presence of PNIPAM is illustrated by measuring the induction time
Fig. 2. Example of the calculation of the error in the induction time measurement, for quartz in the presence of 400 g ton1 of 1.32 MDa PNIPAM below the LCST.
of quartz at temperatures above and below the LCST, both in the presence and absence of midrange molecular weight (1.32 MDa) polymer. The pickup probability of quartz particles by an air bubble as a function of contact time is presented in Fig. 3. The obtained curves were analysed to obtain an estimate of induction time, corresponding to the particle pickup probability of 50%, in a manner described in Fig. 2. It is important to note that in the case of quartz in the absence of polymer, the pickup probability of 50% was not reached within the operational range of the instrument. In these cases the induction time value had to be extrapolated, which carries with it a high degree of uncertainty, as indicted by the high error bars in Fig. 4. Here and elsewhere in the paper, the error bars are indicative of the 95% confidence interval. Figs. 3 and 4 demonstrate that in the absence of polymer, quartz particles can be picked up by an air bubble, which is indicative of a very mild natural hydrophobicity. However, the estimated induction times of these particles range between 1.5 and 2.5 s, which indicates that in a turbulent environment of a real flotation system, bubble/particle attachment is unlikely to occur. The induction time of quartz in the absence of polymer remained unchanged as a function of temperature. Figs. 3 and 4 also clearly show that when 1.32 MDa PNIPAM is present on the quartz surfaces, the induction time of quartz decreases dramatically. The particle pickup probability curves undergo an increase towards higher probabilities and become steeper. The induction times at both high and low temperatures undergo
Fig. 1. Induction time measurement apparatus.
924
E. Burdukova et al. / Minerals Engineering 23 (2010) 921–927
Fig. 3. Probability of quartz particle pickup as a function of bubble/particle contact time in the presence and absence of 400 g ton1 of PNIPAM at temperatures above and below the LCST.
Fig. 4. Induction time of quartz particles in the presence and absence of PNIPAM at temperatures above and below the LCST.
a significant decrease, indicating an elevated degree of particle/ bubble attachment. At a temperature above the LCST (50 °C), the particle pickup probabilities are significantly higher than those at 22 °C (below the LCST). Although significantly higher than in the absence of polymer, the particle pickup probabilities at 22 °C are still very low, not reaching 50% within the instrument operational range. As a result, the induction time at a higher temperature is significantly lower than that at 22 °C, corresponding to a much higher probability of bubble/particle attachment. The effect of PNIPAM on bubble/particle attachment can be explained by characterising its effect on the particle surfaces. As discussed earlier, the probability of bubble/particle attachment is strongly dependent on the magnitude of electrical double layer repulsion between bubbles and mineral particles. The changes in degree of double layer repulsion can be gauged by using zeta potential measurements. Fig. 5 demonstrates the effect of PNIPAM at temperatures above and below the LCST on the zeta potential of quartz particles. The figure shows that at room temperature, in the absence of PNIPAM, quartz surfaces carry a very strong negative charge, which is typical of this mineral at pH 8 (Brien and Kar, 1968). However, the magnitude of the zeta potential of quartz particles appears to be slightly diminished at higher temperature. The graph also clearly
Fig. 5. Zeta potential measurement of quartz particles in the presence and absence of 400 g ton1 PNIPAM at temperatures above and below the LCST, at pH 8.
demonstrates that the addition of PNIPAM causes a significant decrease in zeta potential at temperatures both above and below the LCST. However, at 50 °C the resultant zeta potential is lower than that at 22 °C. Another component of the forces between bubbles and particles which affects attachment probability is hydrophobic attraction, determined by the degree of hydrophobicity of the particle surfaces. This was estimated by measuring the three phase contact angles between air bubbles and PNIPAM coated silica glass surfaces in aqueous solution at both low and high temperatures. The results of the measurement are presented Fig. 6. The graph shows that in the absence of polymer, silica glass slides exhibit a very small contact angle, indicating that they are hydrophilic. This is not consistent with induction time measurements which showed that there existed a low probability of particle pickup by air bubbles in the absence of polymer. However, the quarts particles are likely to have a slightly higher contact angle than the clean silica glass slides. Since the quartz particles were not cleaned they are likely to have a small amount of contamination from the environment which causes their contact angle to be larger than 5°. Therefore the quartz particles without the polymer have a low probability of being picked up by a bubble, as seen in Figs. 3 and 4. When PNIPAM is added to the system, the contact angles undergo a significant increase. At lower temperatures, the contact angles
Fig. 6. Contact angles of silica glass surfaces in the presence and absence of PNIPAM at temperatures above and below the LCST (adapted from Burdukova et al. (2010)).
E. Burdukova et al. / Minerals Engineering 23 (2010) 921–927
fail to reach 30°, which is still indicative of very low levels of hydrophobicity. However, at 50 °C the contact angles increase beyond 70°, within the range typically exhibited by highly hydrophobic particles. These results clearly indicate that the presence of PNIPAM on quartz surfaces (at temperatures both above and below the LCST), significantly decreases their induction time, increases their hydrophobicity and diminishes their zeta potential. This indicates an increased probability of particle/bubble attachment due to diminished electrical double layer repulsion and increased hydrophobic attraction between bubbles and particles (Fuersteanau et al., 1983; Laskowski, 1986). Furthermore, the increased three phase contact angle of quartz particles increases the stability of the bubble/particle aggregates, further contributing to the increased probability of flotation (Eq. (1)). However, this effect is significantly more pronounced at 50 °C (above the LCST) than at 22 °C (below the LCST). It is important to note, that while the induction time and zeta potential measurements were performed at pH 8, the contact angle measurements were obtained at pH 6. However, the difference in the zeta potential of quartz at these two pH levels is practically negligible (Brien and Kar, 1968) making the measurements directly comparable. The dramatic decrease in quartz induction times in the presence of PNIPAM at temperatures above the LCST is not unexpected. Preliminary studies have shown that PNIPAM is capable of inducing floatability in previously hydrophilic material at high temperatures (Franks et al., 2008, 2009). This is largely due to the fact that at temperatures above the LCST, the polymer conformation is such that the molecule is coiled up, with its hydrophobic core exposed. This renders the macromolecules highly hydrophobic and insoluble in water (Sun et al., 2004). These hydrophobic molecules have also been shown to have a very high affinity for quartz surfaces in aqueous medium. It has also been demonstrated that at temperatures above the LCST, PNIIPAM readily adsorbs onto quartz surfaces, resulting in very high adsorption densities at relatively low concentrations (Li et al., 2009; O’Shea et al., submitted for publication). The dense adsorbed layer of highly hydrophobic, non-ionic polymer is likely to result in the increased hydrophobicity and decreased zeta potential of the mineral surfaces, which leads to enhanced particle/bubble attachment. The slight decrease in the induction time of quartz (indicative of enhanced particle/bubble attachment) in the presence of PNIPAM at temperatures below the LCST is more surprising. At lower temperatures PNIPAM molecules are fully extended and hydrated and as such are hydrophilic (Sun et al., 2004). Furthermore, at lower temperatures PNIPAM has a relatively low affinity for quartz surfaces as indicated by very low adsorption densities of this polymer onto quartz (Li et al., 2009; O’Shea et al., submitted for publication). One can speculate that at 22 °C, the adsorption of PNIPAM onto the mineral substrate takes place via the hydrogen bonding of the carbonyl and amine groups of the polymer to the glass surface. This type of bonding would then expose the more hydrophobic core of the macromolecules to the aqueous solution, rendering the mineral slightly hydrophobic. The enhanced hydrophobicity as well as the decreased zeta potential (stemming from the shift of the shear plane away from the particle surface by a thick layer of adsorbed polymer (Dukhin and Derjaguin, 1976)), can be responsible for the marginal increase in attachment probability.
925
of PNIPAM was estimated using induction time measurements. All the tests were performed at temperatures above the LCST (50 °C), and at a PNIPAM dosage of 400 g ton1. The pickup probability of quartz particles by an air bubble as a function of contact time is presented in Fig. 7, while the corresponding induction times (pickup probability of 50%) are shown in Fig. 8. The pickup probabilities also increase significantly with the increasing molecular weight of PNIPAM molecules. As a result, the induction times of quartz particles decrease dramatically with increasing polymer molecular weight as shown in Fig. 8 (note the logarithmic scale on the induction time axis). As in previous figures, the error bars correspond to 95% confidence intervals. The induction time measurements can once again be interpreted in terms of the effect of increasing molecular weight of PNIPAM on the surface properties of quartz and the effect they have on particle/bubble interactions. The probability of particle/bubble attachment is likely to increase due to a reduction in electrical double layer repulsion, as the zeta potential of the particles becomes more neutral. The attachment probability is also likely to increase due to an increase in hydrophobic attraction between particles and bubbles with increasing three phase contact angle (Burdukova et al., 2010).
Fig. 7. Probability of quartz particle pickup as a function of bubble/particle contact time in the presence of 400 g ton1 of PNIPAM at 50 °C, at a variety of molecular weights.
4.2. Effect of PNIPAM molecular weight on particle/bubble attachment Having established that PNIPAM is capable of inducing particle/ bubble attachment of previously hydrophilic solids at temperatures above the LCST, we can now examine the effect of increasing PNIPAM molecular weight. The probability of particle/bubble attachment of quartz as a function of increasing molecular weight
Fig. 8. Induction time of quartz particles in the presence of 400 g ton1 of PNIPAM at 50 °C, at a variety of molecular weights.
926
E. Burdukova et al. / Minerals Engineering 23 (2010) 921–927
The zeta potential of quartz particles (indicative of the degree of double layer repulsion between particles and bubbles) was again measured, with the results presented in Fig. 9. The data clearly shows that the quartz zeta potential undergoes a significant decrease in magnitude with increasing molecular weight of PNIPAM. This results in significant decrease in the electrical double layer repulsion between bubbles and quartz particles with increasing PNIPAM molecular weight. The probability of particle/bubble attachment is likely to increase due to a reduction in electrical double layer repulsion, as the zeta potential of the particles becomes more neutral. The hydrophobicity of quartz surfaces as a function of increasing PNIPAM molecular weight was estimated using contact angle measurements at 50 °C with the results presented in Fig. 10. The figure demonstrates that the contact angle (and hence hydrophobicity) of the PNIPAM coated silica surfaces increases significantly with increasing molecular weight of PNIPAM. This in turn is indicative of the increase in the degree of hydrophobic attraction between bubbles and quartz particles (Burdukova et al., 2010). The progressive decrease in zeta potential with the corresponding increase in hydrophobicity of the quartz particles leads to the increase in the probability of bubble/particle attachment. This is clearly reflected in the decrease in the induction time of quartz
particles with increasing PNIPAM molecular weight, as shown in Figs. 7 and 8. Furthermore, the increasing contact angle between quartz surfaces and air bubbles serves to increase the probability of the formation of the stable bubble/particle aggregate, which further increases the overall probability of flotation (Eq. (1)). At temperatures above the LCST, PNIPAM molecules of all molecular weights exist as highly hydrophobic, water insoluble colloids. However, the above results strongly demonstrate that the ability of these colloids to render a quartz particle floatable is a strong function of polymer molecular weight. It has been previously demonstrated (Li et al., 2009; O’Shea et al., submitted for publication) that at temperatures above LCST, higher molecular weight PNIPAM polymers have a significantly greater affinity for quartz surfaces than those of a lower molecular weight. The adsorption density of PNIPAM at similar solution concentrations tended to significantly increase with molecular weight. The increased thickness of the adsorbed layer of neutral polymer is likely to shift the shear plane further away from the particle surface, thus diminishing its zeta potential (Dukhin and Derjaguin, 1976). Furthermore, the denser adsorbed layer of hydrophobic polymer colloids on the particle surfaces effectively renders them more hydrophobic. Both of these factors then contribute to the increased probability of bubble/particle attachment, while the increased particle hydrophobicity also contributes to the probability of formation of stable bubble/particle aggregates through higher three phase contact angles.
5. Conclusions From the results presented and discussed above the following conclusions can be drawn:
Fig. 9. Zeta potential of quartz particles in the presence of 400 g ton1 of PNIPAM at 50 °C, at a variety of molecular weights.
Poly (N-isopropylacrylamide) (PNIPAM) acts to enhance the probability of particle/bubble attachment at temperatures above the lower critical solution temperature (LSCT). The resultant probability of particle/bubble attachment of quartz significantly increases with increasing molecular weight of PNIPAM. The increase in the probability of particle/bubble attachment can be attributed to the decrease in the magnitude of electrical double layer repulsion and an increase in hydrophobic attraction between bubbles and particles. The increase in the three phase contact angle between the bubbles and particles in aqueous medium also leads to increase in the probability of formation of stable bubble/particle aggregates. The increase in both the probability of particle/bubble attachment as well as the probability of formation of stable bubble/ particle aggregates leads to an increase of probability of flotation of quartz particles in the presence of PNIPAM, particularly at temperature greater than the LCST. From all of the above it can be concluded that high molecular weight PNIPAM could act as an effective flotation collector at temperatures above the LCST. In the future, the potential use of PNIPAM as a flotation collector will be investigated in terms of its selectivity, by using charged polymers. The effect of PNIPAM on the floatability of mineral particles and particle aggregates will be investigated in a flotation system to determine its viability as a commercial collector.
Acknowledgements 2
Fig. 10. Contact angle measurements on silica glass surfaces coated with 0.1 g/m of PNIPAM of increasing molecular weight, at pH 6, at 50 °C (adapted from Burdukova et al. (2010)).
The authors would like to acknowledge the following for their helpful contribution to this work:
E. Burdukova et al. / Minerals Engineering 23 (2010) 921–927
The Australian Research Council, AMIRA International, BHP/ Billiton, Rio Tinto, Orica, Anglo Platinum, Xstrata, Freeport McMoran and Areva NC, though the Australian Minerals Science Research Institute (AMSRI) (LP0667828), for their financial support. Mr. Boris Albijanic of Julius Kruttschnitt Mineral Research Centre, for assisting with the induction time measurements. Mr. J.P. O’Shea and A/Prof. Greg Qiao, for their help with polymer synthesis and characterisation. Mr. Hengbao ‘‘Alex” Zhang, for performing the zeta potential measurements.
References Araujo, A.C., Viana, P.R.M., Peres, A.E.C., 2005. Reagents in iron ores flotation. Minerals Engineering 18, 219–224. Brien, F.B., Kar, G., 1968. An electrophoresis study of the flotation properties of talc minerals. Journal of University of Washington 20 (1), 8–12. Burdukova, E., Laskowski, J.S., 2009. Effect of insoluble amine on bubble surfaces on particle–bubble attachment in potash flotation. Canadian Journal of Chemical Engineering 89, 441–447. Burdukova, E., Li, H., Ishida, N., O’Shea, J.P., Franks, G.V., 2010. Temperature controlled surface hydrophobicity and interaction forces induced by poly (Nisopropylacrylamide). Journal of Colloid and Interface Science 342, 586–592. Davis, F.T., Hyatt, D.E., Cox, C.H., 1976. In: Fuerstenau, M.C. (Ed.), Flotation, American Institute of Mining. Metallurgical and Petroleum Engineers, New York, pp. 1307–1336. Deng, Y., Xiao, H., Pelton, R., 1996. Temperature sensitive flocculants based on poly (N-isopropylacrylamide-co-diallyldimenthylammonium chloride). Journal of Colloid and Interface Science 188, 190. Derjaguin, B.V., 1954. A theory of the heterocoagulation, interaction and adhesion of dissimilar particles in solutions of electrolytes. Discussions of the Faraday Society 18, 85–98. Dukhin, S.S., Derjaguin, B.V., 1976. Electrophoresis, Academy of Science, USSR, Russian Text, Moscow. Eigeles, M.A., Volova, M.L., 1968. Activation of air bubbles by the insoluble products of reaction in flotation. In: Proc. 7th Int. Minerals Processing Congress, Leningrad, USSR. Franks, G.V., 2005. Solid/liquid separation using stimulant sensitive flocculation and consolidation. Journal of Colloid and Interface Science 292, 598–603. Franks, G.V., Li, H., O’Shea, J.P., Qiao, G.G., 2008. Application of poly (N-isopropyl acrylamide) (PNIPAM) as a multifunction process aid in mineral flotation and solid/liquid separation. In: Wang, D.D. et al. (Eds.), 24th Int. Minerals Processing Congress. China Scientific Book Service, Beijing, China. Franks, G.V., Li, H., O’Shea, J.P., Qiao, G.G., 2009. Temperature sensitive polymers as multiple function reagents in mineral processing. Advanced Powder Technology 20, 273–279.
927
Fuersteanau, D.W., Rosenbaum, J.M., Laskowski, J.S., 1983. Effect of surface functional groups on the flotation of coal. Colloids and Surfaces 8, 153–174. Guillet, J.E., Henskins, M., Murray, D.G., 1985. Polymeric Flocculants, US Patent, Editor. Holuszko, M.E., Franzidis, J.P., Min, A., Manlapig, E.V., 2008. The induction time measurement on galena and sphalerite particles with different hydrophobicity characteristics. In: Wang, D.D. et al. (Eds.), 24th Int. Minerals Processing Congress. China Scientific Book Service, Beijing, China. Laskowski, J.S., 1974. Particle–bubble attachment in flotation. Mineral Science Engineering 6 (4), 223–235. Laskowski, J.S., 1986. The relationship between floatability and hydrophobicity. In: Somadundaran, P. (Ed.), Advances in Mineral Processing. SME, pp. 189–208. Li, H., Long, J., Xu, Z., Masliyah, J.H., 2007. Flocculation of kaolinite clay suspensions using a temperature sensitive polymer. AIChE Journal 53, 479–488. Li, H., O’Shea, J.P., Franks, G.V., 2009. Effect of molecular weight of poly (N-isopropyl acrylamide) (PNIPAM) temperature-sensitive flocculants on dewatering. AIChE Journal 55 (8), 2070–2080. Nguyen, A.V., Schulze, H.J., Ralston, J., 1997. Elementary steps in particle–bubble attachment. International Journal of Mineral Processing 51, 183–195. O’Shea, J.P., Qiao, G.G., Spiniello, M., Franks, G.V., 2007. Temperature Mediated Adsorption of Poly (N-isopropylacrylamide) onto a-Alumina, Chemeca Melbourne, Australia. O’Shea, J.P., Qiao, G.G., Franks, G.V., submitted for publication. Solid–liquid separations with a temperature-responsive polymeric flocculant: effect of temperature and molecular weight on polymer adsorption. Journal of Colloid and Interface Science. Quast, K.B., 2000. A review of hematite flotation using 12-carbon chain collectors. Minerals Engineering 13 (13), 1361–1376. Ralston, J., 1992. The influence of particle size and contact angle in flotation. In: Laskowski, J.S., Ralston, J. (Eds.), Colloid Chemistry in Minerals Processing. Elsevier, Amsterdam, pp. 203–224. Sakohara, S., Nishikawa, K., 2004. Compaction of TiO2 suspension utilizing hydrophilic/hydrophobic transition of cationic thermosensitive polymers. Journal of Colloid and Interface Science 278, 304–309. Sakohara, S., Kimura, T., Nishikawa, K., 2002. Flocculation mechanism of suspended particles using the hydrophilic/hydrophobic transition of thermosensitive polymer. KONA 20, 246–250. Saunders, B.R., Crowther, H.M., Morris, G.E., Mears, S.J., Cosgrove, T., Vincent, B., 1999. Factors affecting the swelling of poly (N-isopropylacrylamide) microgel particles: fundamental and commercial implications. Colloids and Surfaces 149, 57–64. Schulze, H.J., 1984. Physico-Chemical Elementary Processes in Flotation. Elsevier, Amsterdam. Sun, T., Wang, G., Feng, L., Liu, B., Ma, Y., Jang, L., Zhu, D., 2004. Reversible switching between reversible hydrophilicity and superhydrophobicity. Angewante Chemie International Edition 43 (3), 357–360. Weisberg, S., 2005. Applied Linear Regression. Wiley Series in Probability and Statistics. John Wiley & Sons Inc., New Jersey. Yordan, J.L., Yoon, R.H., 1986. Induction time measurement for the quartz amine flotation system. SME Annual Meeting 1986, New Orleans. Yordan, J.L., Yoon, R.H., 1988. Induction time measurement for coal flotation system. In: Attia, Y.A., Moudgil, B.M., Chander, S. (Eds.), Interfacial Phenomena in Biotechnology, Process Technology. Elsevier, New York.
Contents lists available at ScienceDirect
Minerals Engineering journal homepage: www.elsevier.com/locate/mineng
Poly (N-isopropylacrylamide) (PNIPAM) as a flotation collector: Effect of temperature and molecular weight Elizaveta Burdukova a, Haihong Li b, Dee J. Bradshaw c, George V. Franks a,* a
Australia Mineral Science Research Institute, Department of Chemical and Biomolecular Engineering, The University of Melbourne, Victoria 3010, Australia Alberta Research Council, Alberta Innovates, 250 Karl Clark Road, Edmonton, Alberta, AB, Canada c Julius Kruttschnitt Mineral Research Centre, University of Queensland, Queensland 4072, Australia b
a r t i c l e
i n f o
Article history: Received 15 December 2009 Accepted 6 March 2010 Available online 31 March 2010 Keywords: Mineral processing Flotation reagents Flotation collectors Fine particle processing
a b s t r a c t Poly (N-isopropylacrylamide) (PNIPAM), a temperature responsive polymer, was tested for its potential use as a collector in a quartz flotation system. The effect of PNIPAM on the surface characteristics of quartz particles were studied using induction time, contact angle and zeta potential measurement and analysed in terms of the probability of bubble/particle attachment and the probability of formation of stable bubble/particle aggregates. It was found that probability of bubble/particle attachment of quartz significantly increases in the presence of PNIPAM, particularly at temperatures above the lower critical solution temperature (LCST) of 32 °C. Furthermore, the probability of bubble/particle attachment increases with increasing PNIPAM molecular weight. This was attributed to the increased hydrophobicity of the quartz surface as well as the decrease in the double layer repulsion between bubbles and particles. This leads to the conclusion that PNIPAM could act as an effective collector in a flotation system. Ó 2010 Elsevier Ltd. All rights reserved.
1. Introduction Poly (N-isopropylacrylamide) (PNIPAM) is a temperature responsive polymer. At temperature below its lower critical solution temperature (LCST) of 32 °C, PNIPAM is soluble in water and hydrophilic. The polymer macromolecules are hydrated and fully extended, with extensive intermolecular hydrogen bonding between water molecules and polymer chains (Saunders et al., 1999; Sakohara et al., 2002; Sun et al., 2004). At temperatures above the LCST, the hydrogen bonds between water molecules and polymer chains are broken and instead, intramolecular and intermolecular hydrogen bonds between the C@O and N–H groups are formed. The formation of intramolecular bonds causes the molecule to coil up, exposing its hydrophobic core. This renders the polymers hydrophobic and insoluble in water (Saunders et al., 1999; Sakohara et al., 2002; Sun et al., 2004). The studies of PNIPAM in the context of mineral processing applications, originate in its potential use as a temperature responsive flocculant at temperatures above the LCST. It can be used as a reversibly responsive flocculant and dispersant (Guillet et al., 1985; Deng et al., 1996; Sakohara et al., 2002; Sakohara and Nishikawa, 2004; Franks, 2005; Li et al., 2007, 2009). The reversibility of the state of aggregation and dispersion can be * Corresponding author. E-mail address: [email protected] (G.V. Franks). 0892-6875/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.mineng.2010.03.003
used to produce both rapid sedimentation (at temperatures above the LCST) and enhance sediment consolidation and dewatering (at temperatures below the LCST). At high temperature, the inter-particle forces in a mineral suspension are changed from repulsive to attractive; causing the formation of hydrophobic particle aggregates (flocs). After the flocs settle, the temperature can be reduced, which produces repulsion between particles as the polymer becomes hydrophilic. The repulsion allows for additional consolidation of the mineral sediment leading to improved water recovery. As well as being an effective flocculant/dispersant, preliminary investigations demonstrated that PNIPAM can be an effective collector in the flotation of oxide minerals (Franks et al., 2009). In current flotation practice, flotation of oxide minerals is generally achieved using short chain surfactant collectors (e.g. dodecyl amine and sodium dodecyl sulphate) (Quast, 2000; Araujo et al., 2005). Preliminary testing in a simple flotation system showed that at temperatures above the LCST, the presence of PNIPAM induced flotation of both silica and kaolinite, achieving recoveries significantly greater than those in the presence of dodecyl amine (Franks et al., 2009). The ability of PNIPAM to induce both flocculation and flotation of previously dispersed hydrophilic particles (at temperatures above the LSCT) potentially makes it a highly valuable reagent in the flotation of ultrafine mineral particles. Furthermore, PNIPAM is a non-toxic polymer, which can potentially provide an environmentally friendly substitute to industrial collectors which
922
E. Burdukova et al. / Minerals Engineering 23 (2010) 921–927
are often environmentally harmful (Davis et al., 1976). However, there is still insufficient evidence to determine whether or not PNIPAM would present a commercially viable substitute for collectors currently used in industry. The likelihood that a mineral particle is floatable is generally expressed in terms of probabilities, as shown in Eq. (1), where P is the overall flotation probability, PC is the probability of bubble/particle collision, PA is the probability of bubble/particle attachment and PS is the probability of formation of a stable bubble/particle aggregate (Laskowski, 1986; Ralston, 1992).
P ¼ PC PA PS
ð1Þ
The probability of collision (PC) is strongly related to hydrodynamic factors such as cell turbulence, pulp viscosity and particle size (Schulze, 1984). As such, collision probability is unlikely to be influenced by the presence of collector in the flotation system, and is not considered further. The probability of formation of stable bubble/particle aggregates (PS) is a very strong function of the magnitude of the three phase contact angle between the air bubble, mineral particle and aqueous medium, which directly determines the strength of the bubble/particle bond (Laskowski, 1986). The increased strength of this bond determines how likely a particle is to detach from a bubble. For this reason, in some publications this term (PS) is replaced by the probability of bubble/particle detachment (PD), where PS is inversely related to PD (Laskowski, 1986; Ralston, 1992). The probability of bubble/particle attachment (PA) is largely dependent on the surface forces existing between a bubble and a particle. It has been shown that in an aqueous medium, the interactions between a negatively charged mineral particle and a bubble are similar to those between two negatively charged mineral particles (Derjaguin, 1954; Laskowski, 1986). Therefore, the probability of bubble particle attachment is subject to double layer forces, as described by the modified DLVO theory, mainly: van der Waals forces, electrical double layer repulsion and hydrophobic attraction. It follows that the probability of bubble particle attachment will decrease with increasing double layer repulsion between mineral particles and bubbles and increase with increasing particle hydrophobicity. Induction time measurements are one way of characterising the probability of particle/bubble attachment (Nguyen et al., 1997). Induction time is defined as the time for the thinning of the intervening liquid film between an air bubble and a hydrophobic particle to a critical thickness at which the film will rupture spontaneously (Laskowski, 1974). Induction time is measured by bringing captive bubbles into contact with a bed of particles for a controlled time interval (typically between 10 and 1500 ms), to determine whether or not mineral particles attach to bubbles such that they can be lifted up from a particle bed. The particle/bubble contacts are repeated numerous times, estimating the probability of bubble/particle pickup at each contact time. The contact time corresponding to 50% probability of particle pickup is termed the induction time. This technique allows for a simple and reliable estimate of the strength of particle/bubble attachment (Eigeles and Volova, 1968; Yordan and Yoon, 1986, 1988; Holuszko et al., 2008; Burdukova and Laskowski, 2009). The purpose of this work is to study the effect of temperature sensitive PNIPAM on the probability of attachment of quartz particles to air bubbles, using induction time measurements. These measurements are then interpreted by characterising the particle surfaces in terms of their hydrophobicity (as indicated by contact angle measurements) and electrical double layer repulsion (as indicated by zeta potential measurements).
2. Experimental details 2.1. Materials 2.1.1. Temperature sensitive polymers A temperature sensitive, non-ionic polymer – poly (N-isopropylacrylamide) (PNIPAM), was used in this work. Three molecular weights of this polymer were used, termed: low, medium and high. The low molecular weight (0.23 MDa) PNIPAM was purchased from Sigma–Aldrich, Australia. The PNIPAMs with the medium molecular weight (1.32 MDa) and the high molecular weight (3.6 and 4.5 MDa) were synthesized in our laboratories as described in detail elsewhere (O’Shea et al., 2007). The synthesis process of these high molecular weight polymers is such that relatively small quantities of the polymer can be produced per batch and it is not easy to obtain consistent molecular weights of polymers between batches. For this reason the high molecular weight polymer used in the different sections of the work varies in molecular weight between 3.6 and 4.5 MDa. In the case of contact angle measurements, the PNIPAM with medium molecular weight (2.0 MDa) was purchased from Polymer Source Inc., Canada. 2.1.2. Quartz Quartz powder (Silica 400G) was obtained from UNIMIN Australia Limited for use in zeta potential measurements. The particles had a size distribution of d50 10 lm, and a BET surface area of 1.7 m2 g1. The powder contained 99.6% SiO2 and traces of alumina, ferric oxide, titania and lime. For the use in induction time measurements, quartz particles (Silica 100 WQ) were obtained from UNIMIN Australia Limited. The particles were dry sieved to obtain a particle size distribution of 90 lm < Dp < 150 lm. The particles contained 99.5% SiO2 and traces of alumina, ferric oxide, titania and lime. 3. Methods 3.1. Contact angle measurement The sessile drop technique was used for the measurement of advancing contact angle, using the OCA Dataphysics measurement system, provided by the Department of Chemistry, University of Melbourne. Silica glass surfaces were cleaned by first washing them in surfactant solution followed by dilute NaOH, then finally rinsing with deionised water. For each molecular weight PNIPAM, three clean silica glass slides (26 76 mm) were coated with polymer solutions so as to produce the surface concentration of 0.1 g of polymer per m2 of glass surface. The slides were dried in an oven at 80 °C for about 30 min. The temperature of the glass slides during contact angle measurements was controlled by placing the slides on the top surface of an aluminium stage which has its temperature controlled by a water bath in order to achieve the temperature of either 22 °C (below the LCST) or 50 °C (above the LCST). Five minutes were allowed to make sure that the slide had reached the desired temperature. A drop of deionised water at pH 6, with 3 lL volume, was introduced onto the slide through a computer controlled micro-syringe. The measurement was taken 10 s after the liquid drop was put on the glass slide. The measurements were conducted in triplicate. 3.2. Zeta potential measurements Zeta potential measurements were performed using the Zeta Acustosizer, manufactured by Colloidal Dynamics Inc., Sydney, Australia. The instrument is equipped with a recirculating water bath, which allowed for the measurement being performed at both 22 °C and 50 °C. Suspensions of quartz (5 wt.%) were prepared using deionised water containing 40 ppm of PNIPAM (corresponding to
E. Burdukova et al. / Minerals Engineering 23 (2010) 921–927
923
400 g ton1 of quartz) at a variety of molecular weights. The suspensions were allowed to stand for 24 h prior to measurement. The measurements were performed at pH 8, using 1 M NaOH solution for pH adjustment, with the background electrolyte of 102 KCl. 3.3. Induction time measurements Quartz particles were treated with PNIPAMs with a variety of molecular weights, prior to induction time measurement. Ten grams of quartz were placed in a beaker containing 100 mL of distilled water adjusted to pH 8 using 1 M NaOH solution for pH adjustment, with the background electrolyte of 102 KCl. The beaker was immersed in a water bath controlled to either 22 °C or 50 °C. Each suspension was dosed with 40 ppm of PNIPAM, which corresponded to 400 g of polymer per ton of quartz. The suspensions were allowed to equilibrate for 1 h, while being continuously agitated with a magnetic stirrer. After 1 h, the solids were filtered out and gently washed with distilled water adjusted to the appropriate temperature. The PNIPAM coated solids were placed in the oven to dry at 70 °C. Induction time measurements were performed with the MCT 100 Induction Time Meter, provided by the Julius Kruttschnitt Mineral Research Centre, University of Queensland. The temperature of the measurement was kept controlled by a small water jacket, fitted around a measurement vessel. Water from an adjacent recirculating water bath was pumped though the water jacket, maintaining a constant temperature of either 22 °C (below the LCST) or 50 °C (above the LCST). The photograph of the test apparatus is shown in Fig. 1. Approximately 2 g of solids were placed in the measurement vessel (enough to cover the vessel bottom) and covered with distilled water adjusted to the desired temperature and pH 8. The bubble was brought into contact with the particle bed at set contact times ranging between 10 and 1200 ms. For each contact time, the probability of particle pickup was estimated by counting the number of times particles were picked up by a bubble out of 40– 80 contacts. The induction time was estimated as the contact time at which the pickup probability was 50%. The error in the induction time value was estimated by calculating the 95% confidence interval of the linear regression of contact times as a function of pickup probability as shown in Fig. 2 (Weisberg, 2005). 4. Results and discussion 4.1. Effect of PNIPAM temperature on particle/bubble attachment The effect of temperature on particle/bubble attachment in the presence of PNIPAM is illustrated by measuring the induction time
Fig. 2. Example of the calculation of the error in the induction time measurement, for quartz in the presence of 400 g ton1 of 1.32 MDa PNIPAM below the LCST.
of quartz at temperatures above and below the LCST, both in the presence and absence of midrange molecular weight (1.32 MDa) polymer. The pickup probability of quartz particles by an air bubble as a function of contact time is presented in Fig. 3. The obtained curves were analysed to obtain an estimate of induction time, corresponding to the particle pickup probability of 50%, in a manner described in Fig. 2. It is important to note that in the case of quartz in the absence of polymer, the pickup probability of 50% was not reached within the operational range of the instrument. In these cases the induction time value had to be extrapolated, which carries with it a high degree of uncertainty, as indicted by the high error bars in Fig. 4. Here and elsewhere in the paper, the error bars are indicative of the 95% confidence interval. Figs. 3 and 4 demonstrate that in the absence of polymer, quartz particles can be picked up by an air bubble, which is indicative of a very mild natural hydrophobicity. However, the estimated induction times of these particles range between 1.5 and 2.5 s, which indicates that in a turbulent environment of a real flotation system, bubble/particle attachment is unlikely to occur. The induction time of quartz in the absence of polymer remained unchanged as a function of temperature. Figs. 3 and 4 also clearly show that when 1.32 MDa PNIPAM is present on the quartz surfaces, the induction time of quartz decreases dramatically. The particle pickup probability curves undergo an increase towards higher probabilities and become steeper. The induction times at both high and low temperatures undergo
Fig. 1. Induction time measurement apparatus.
924
E. Burdukova et al. / Minerals Engineering 23 (2010) 921–927
Fig. 3. Probability of quartz particle pickup as a function of bubble/particle contact time in the presence and absence of 400 g ton1 of PNIPAM at temperatures above and below the LCST.
Fig. 4. Induction time of quartz particles in the presence and absence of PNIPAM at temperatures above and below the LCST.
a significant decrease, indicating an elevated degree of particle/ bubble attachment. At a temperature above the LCST (50 °C), the particle pickup probabilities are significantly higher than those at 22 °C (below the LCST). Although significantly higher than in the absence of polymer, the particle pickup probabilities at 22 °C are still very low, not reaching 50% within the instrument operational range. As a result, the induction time at a higher temperature is significantly lower than that at 22 °C, corresponding to a much higher probability of bubble/particle attachment. The effect of PNIPAM on bubble/particle attachment can be explained by characterising its effect on the particle surfaces. As discussed earlier, the probability of bubble/particle attachment is strongly dependent on the magnitude of electrical double layer repulsion between bubbles and mineral particles. The changes in degree of double layer repulsion can be gauged by using zeta potential measurements. Fig. 5 demonstrates the effect of PNIPAM at temperatures above and below the LCST on the zeta potential of quartz particles. The figure shows that at room temperature, in the absence of PNIPAM, quartz surfaces carry a very strong negative charge, which is typical of this mineral at pH 8 (Brien and Kar, 1968). However, the magnitude of the zeta potential of quartz particles appears to be slightly diminished at higher temperature. The graph also clearly
Fig. 5. Zeta potential measurement of quartz particles in the presence and absence of 400 g ton1 PNIPAM at temperatures above and below the LCST, at pH 8.
demonstrates that the addition of PNIPAM causes a significant decrease in zeta potential at temperatures both above and below the LCST. However, at 50 °C the resultant zeta potential is lower than that at 22 °C. Another component of the forces between bubbles and particles which affects attachment probability is hydrophobic attraction, determined by the degree of hydrophobicity of the particle surfaces. This was estimated by measuring the three phase contact angles between air bubbles and PNIPAM coated silica glass surfaces in aqueous solution at both low and high temperatures. The results of the measurement are presented Fig. 6. The graph shows that in the absence of polymer, silica glass slides exhibit a very small contact angle, indicating that they are hydrophilic. This is not consistent with induction time measurements which showed that there existed a low probability of particle pickup by air bubbles in the absence of polymer. However, the quarts particles are likely to have a slightly higher contact angle than the clean silica glass slides. Since the quartz particles were not cleaned they are likely to have a small amount of contamination from the environment which causes their contact angle to be larger than 5°. Therefore the quartz particles without the polymer have a low probability of being picked up by a bubble, as seen in Figs. 3 and 4. When PNIPAM is added to the system, the contact angles undergo a significant increase. At lower temperatures, the contact angles
Fig. 6. Contact angles of silica glass surfaces in the presence and absence of PNIPAM at temperatures above and below the LCST (adapted from Burdukova et al. (2010)).
E. Burdukova et al. / Minerals Engineering 23 (2010) 921–927
fail to reach 30°, which is still indicative of very low levels of hydrophobicity. However, at 50 °C the contact angles increase beyond 70°, within the range typically exhibited by highly hydrophobic particles. These results clearly indicate that the presence of PNIPAM on quartz surfaces (at temperatures both above and below the LCST), significantly decreases their induction time, increases their hydrophobicity and diminishes their zeta potential. This indicates an increased probability of particle/bubble attachment due to diminished electrical double layer repulsion and increased hydrophobic attraction between bubbles and particles (Fuersteanau et al., 1983; Laskowski, 1986). Furthermore, the increased three phase contact angle of quartz particles increases the stability of the bubble/particle aggregates, further contributing to the increased probability of flotation (Eq. (1)). However, this effect is significantly more pronounced at 50 °C (above the LCST) than at 22 °C (below the LCST). It is important to note, that while the induction time and zeta potential measurements were performed at pH 8, the contact angle measurements were obtained at pH 6. However, the difference in the zeta potential of quartz at these two pH levels is practically negligible (Brien and Kar, 1968) making the measurements directly comparable. The dramatic decrease in quartz induction times in the presence of PNIPAM at temperatures above the LCST is not unexpected. Preliminary studies have shown that PNIPAM is capable of inducing floatability in previously hydrophilic material at high temperatures (Franks et al., 2008, 2009). This is largely due to the fact that at temperatures above the LCST, the polymer conformation is such that the molecule is coiled up, with its hydrophobic core exposed. This renders the macromolecules highly hydrophobic and insoluble in water (Sun et al., 2004). These hydrophobic molecules have also been shown to have a very high affinity for quartz surfaces in aqueous medium. It has also been demonstrated that at temperatures above the LCST, PNIIPAM readily adsorbs onto quartz surfaces, resulting in very high adsorption densities at relatively low concentrations (Li et al., 2009; O’Shea et al., submitted for publication). The dense adsorbed layer of highly hydrophobic, non-ionic polymer is likely to result in the increased hydrophobicity and decreased zeta potential of the mineral surfaces, which leads to enhanced particle/bubble attachment. The slight decrease in the induction time of quartz (indicative of enhanced particle/bubble attachment) in the presence of PNIPAM at temperatures below the LCST is more surprising. At lower temperatures PNIPAM molecules are fully extended and hydrated and as such are hydrophilic (Sun et al., 2004). Furthermore, at lower temperatures PNIPAM has a relatively low affinity for quartz surfaces as indicated by very low adsorption densities of this polymer onto quartz (Li et al., 2009; O’Shea et al., submitted for publication). One can speculate that at 22 °C, the adsorption of PNIPAM onto the mineral substrate takes place via the hydrogen bonding of the carbonyl and amine groups of the polymer to the glass surface. This type of bonding would then expose the more hydrophobic core of the macromolecules to the aqueous solution, rendering the mineral slightly hydrophobic. The enhanced hydrophobicity as well as the decreased zeta potential (stemming from the shift of the shear plane away from the particle surface by a thick layer of adsorbed polymer (Dukhin and Derjaguin, 1976)), can be responsible for the marginal increase in attachment probability.
925
of PNIPAM was estimated using induction time measurements. All the tests were performed at temperatures above the LCST (50 °C), and at a PNIPAM dosage of 400 g ton1. The pickup probability of quartz particles by an air bubble as a function of contact time is presented in Fig. 7, while the corresponding induction times (pickup probability of 50%) are shown in Fig. 8. The pickup probabilities also increase significantly with the increasing molecular weight of PNIPAM molecules. As a result, the induction times of quartz particles decrease dramatically with increasing polymer molecular weight as shown in Fig. 8 (note the logarithmic scale on the induction time axis). As in previous figures, the error bars correspond to 95% confidence intervals. The induction time measurements can once again be interpreted in terms of the effect of increasing molecular weight of PNIPAM on the surface properties of quartz and the effect they have on particle/bubble interactions. The probability of particle/bubble attachment is likely to increase due to a reduction in electrical double layer repulsion, as the zeta potential of the particles becomes more neutral. The attachment probability is also likely to increase due to an increase in hydrophobic attraction between particles and bubbles with increasing three phase contact angle (Burdukova et al., 2010).
Fig. 7. Probability of quartz particle pickup as a function of bubble/particle contact time in the presence of 400 g ton1 of PNIPAM at 50 °C, at a variety of molecular weights.
4.2. Effect of PNIPAM molecular weight on particle/bubble attachment Having established that PNIPAM is capable of inducing particle/ bubble attachment of previously hydrophilic solids at temperatures above the LCST, we can now examine the effect of increasing PNIPAM molecular weight. The probability of particle/bubble attachment of quartz as a function of increasing molecular weight
Fig. 8. Induction time of quartz particles in the presence of 400 g ton1 of PNIPAM at 50 °C, at a variety of molecular weights.
926
E. Burdukova et al. / Minerals Engineering 23 (2010) 921–927
The zeta potential of quartz particles (indicative of the degree of double layer repulsion between particles and bubbles) was again measured, with the results presented in Fig. 9. The data clearly shows that the quartz zeta potential undergoes a significant decrease in magnitude with increasing molecular weight of PNIPAM. This results in significant decrease in the electrical double layer repulsion between bubbles and quartz particles with increasing PNIPAM molecular weight. The probability of particle/bubble attachment is likely to increase due to a reduction in electrical double layer repulsion, as the zeta potential of the particles becomes more neutral. The hydrophobicity of quartz surfaces as a function of increasing PNIPAM molecular weight was estimated using contact angle measurements at 50 °C with the results presented in Fig. 10. The figure demonstrates that the contact angle (and hence hydrophobicity) of the PNIPAM coated silica surfaces increases significantly with increasing molecular weight of PNIPAM. This in turn is indicative of the increase in the degree of hydrophobic attraction between bubbles and quartz particles (Burdukova et al., 2010). The progressive decrease in zeta potential with the corresponding increase in hydrophobicity of the quartz particles leads to the increase in the probability of bubble/particle attachment. This is clearly reflected in the decrease in the induction time of quartz
particles with increasing PNIPAM molecular weight, as shown in Figs. 7 and 8. Furthermore, the increasing contact angle between quartz surfaces and air bubbles serves to increase the probability of the formation of the stable bubble/particle aggregate, which further increases the overall probability of flotation (Eq. (1)). At temperatures above the LCST, PNIPAM molecules of all molecular weights exist as highly hydrophobic, water insoluble colloids. However, the above results strongly demonstrate that the ability of these colloids to render a quartz particle floatable is a strong function of polymer molecular weight. It has been previously demonstrated (Li et al., 2009; O’Shea et al., submitted for publication) that at temperatures above LCST, higher molecular weight PNIPAM polymers have a significantly greater affinity for quartz surfaces than those of a lower molecular weight. The adsorption density of PNIPAM at similar solution concentrations tended to significantly increase with molecular weight. The increased thickness of the adsorbed layer of neutral polymer is likely to shift the shear plane further away from the particle surface, thus diminishing its zeta potential (Dukhin and Derjaguin, 1976). Furthermore, the denser adsorbed layer of hydrophobic polymer colloids on the particle surfaces effectively renders them more hydrophobic. Both of these factors then contribute to the increased probability of bubble/particle attachment, while the increased particle hydrophobicity also contributes to the probability of formation of stable bubble/particle aggregates through higher three phase contact angles.
5. Conclusions From the results presented and discussed above the following conclusions can be drawn:
Fig. 9. Zeta potential of quartz particles in the presence of 400 g ton1 of PNIPAM at 50 °C, at a variety of molecular weights.
Poly (N-isopropylacrylamide) (PNIPAM) acts to enhance the probability of particle/bubble attachment at temperatures above the lower critical solution temperature (LSCT). The resultant probability of particle/bubble attachment of quartz significantly increases with increasing molecular weight of PNIPAM. The increase in the probability of particle/bubble attachment can be attributed to the decrease in the magnitude of electrical double layer repulsion and an increase in hydrophobic attraction between bubbles and particles. The increase in the three phase contact angle between the bubbles and particles in aqueous medium also leads to increase in the probability of formation of stable bubble/particle aggregates. The increase in both the probability of particle/bubble attachment as well as the probability of formation of stable bubble/ particle aggregates leads to an increase of probability of flotation of quartz particles in the presence of PNIPAM, particularly at temperature greater than the LCST. From all of the above it can be concluded that high molecular weight PNIPAM could act as an effective flotation collector at temperatures above the LCST. In the future, the potential use of PNIPAM as a flotation collector will be investigated in terms of its selectivity, by using charged polymers. The effect of PNIPAM on the floatability of mineral particles and particle aggregates will be investigated in a flotation system to determine its viability as a commercial collector.
Acknowledgements 2
Fig. 10. Contact angle measurements on silica glass surfaces coated with 0.1 g/m of PNIPAM of increasing molecular weight, at pH 6, at 50 °C (adapted from Burdukova et al. (2010)).
The authors would like to acknowledge the following for their helpful contribution to this work:
E. Burdukova et al. / Minerals Engineering 23 (2010) 921–927
The Australian Research Council, AMIRA International, BHP/ Billiton, Rio Tinto, Orica, Anglo Platinum, Xstrata, Freeport McMoran and Areva NC, though the Australian Minerals Science Research Institute (AMSRI) (LP0667828), for their financial support. Mr. Boris Albijanic of Julius Kruttschnitt Mineral Research Centre, for assisting with the induction time measurements. Mr. J.P. O’Shea and A/Prof. Greg Qiao, for their help with polymer synthesis and characterisation. Mr. Hengbao ‘‘Alex” Zhang, for performing the zeta potential measurements.
References Araujo, A.C., Viana, P.R.M., Peres, A.E.C., 2005. Reagents in iron ores flotation. Minerals Engineering 18, 219–224. Brien, F.B., Kar, G., 1968. An electrophoresis study of the flotation properties of talc minerals. Journal of University of Washington 20 (1), 8–12. Burdukova, E., Laskowski, J.S., 2009. Effect of insoluble amine on bubble surfaces on particle–bubble attachment in potash flotation. Canadian Journal of Chemical Engineering 89, 441–447. Burdukova, E., Li, H., Ishida, N., O’Shea, J.P., Franks, G.V., 2010. Temperature controlled surface hydrophobicity and interaction forces induced by poly (Nisopropylacrylamide). Journal of Colloid and Interface Science 342, 586–592. Davis, F.T., Hyatt, D.E., Cox, C.H., 1976. In: Fuerstenau, M.C. (Ed.), Flotation, American Institute of Mining. Metallurgical and Petroleum Engineers, New York, pp. 1307–1336. Deng, Y., Xiao, H., Pelton, R., 1996. Temperature sensitive flocculants based on poly (N-isopropylacrylamide-co-diallyldimenthylammonium chloride). Journal of Colloid and Interface Science 188, 190. Derjaguin, B.V., 1954. A theory of the heterocoagulation, interaction and adhesion of dissimilar particles in solutions of electrolytes. Discussions of the Faraday Society 18, 85–98. Dukhin, S.S., Derjaguin, B.V., 1976. Electrophoresis, Academy of Science, USSR, Russian Text, Moscow. Eigeles, M.A., Volova, M.L., 1968. Activation of air bubbles by the insoluble products of reaction in flotation. In: Proc. 7th Int. Minerals Processing Congress, Leningrad, USSR. Franks, G.V., 2005. Solid/liquid separation using stimulant sensitive flocculation and consolidation. Journal of Colloid and Interface Science 292, 598–603. Franks, G.V., Li, H., O’Shea, J.P., Qiao, G.G., 2008. Application of poly (N-isopropyl acrylamide) (PNIPAM) as a multifunction process aid in mineral flotation and solid/liquid separation. In: Wang, D.D. et al. (Eds.), 24th Int. Minerals Processing Congress. China Scientific Book Service, Beijing, China. Franks, G.V., Li, H., O’Shea, J.P., Qiao, G.G., 2009. Temperature sensitive polymers as multiple function reagents in mineral processing. Advanced Powder Technology 20, 273–279.
927
Fuersteanau, D.W., Rosenbaum, J.M., Laskowski, J.S., 1983. Effect of surface functional groups on the flotation of coal. Colloids and Surfaces 8, 153–174. Guillet, J.E., Henskins, M., Murray, D.G., 1985. Polymeric Flocculants, US Patent, Editor. Holuszko, M.E., Franzidis, J.P., Min, A., Manlapig, E.V., 2008. The induction time measurement on galena and sphalerite particles with different hydrophobicity characteristics. In: Wang, D.D. et al. (Eds.), 24th Int. Minerals Processing Congress. China Scientific Book Service, Beijing, China. Laskowski, J.S., 1974. Particle–bubble attachment in flotation. Mineral Science Engineering 6 (4), 223–235. Laskowski, J.S., 1986. The relationship between floatability and hydrophobicity. In: Somadundaran, P. (Ed.), Advances in Mineral Processing. SME, pp. 189–208. Li, H., Long, J., Xu, Z., Masliyah, J.H., 2007. Flocculation of kaolinite clay suspensions using a temperature sensitive polymer. AIChE Journal 53, 479–488. Li, H., O’Shea, J.P., Franks, G.V., 2009. Effect of molecular weight of poly (N-isopropyl acrylamide) (PNIPAM) temperature-sensitive flocculants on dewatering. AIChE Journal 55 (8), 2070–2080. Nguyen, A.V., Schulze, H.J., Ralston, J., 1997. Elementary steps in particle–bubble attachment. International Journal of Mineral Processing 51, 183–195. O’Shea, J.P., Qiao, G.G., Spiniello, M., Franks, G.V., 2007. Temperature Mediated Adsorption of Poly (N-isopropylacrylamide) onto a-Alumina, Chemeca Melbourne, Australia. O’Shea, J.P., Qiao, G.G., Franks, G.V., submitted for publication. Solid–liquid separations with a temperature-responsive polymeric flocculant: effect of temperature and molecular weight on polymer adsorption. Journal of Colloid and Interface Science. Quast, K.B., 2000. A review of hematite flotation using 12-carbon chain collectors. Minerals Engineering 13 (13), 1361–1376. Ralston, J., 1992. The influence of particle size and contact angle in flotation. In: Laskowski, J.S., Ralston, J. (Eds.), Colloid Chemistry in Minerals Processing. Elsevier, Amsterdam, pp. 203–224. Sakohara, S., Nishikawa, K., 2004. Compaction of TiO2 suspension utilizing hydrophilic/hydrophobic transition of cationic thermosensitive polymers. Journal of Colloid and Interface Science 278, 304–309. Sakohara, S., Kimura, T., Nishikawa, K., 2002. Flocculation mechanism of suspended particles using the hydrophilic/hydrophobic transition of thermosensitive polymer. KONA 20, 246–250. Saunders, B.R., Crowther, H.M., Morris, G.E., Mears, S.J., Cosgrove, T., Vincent, B., 1999. Factors affecting the swelling of poly (N-isopropylacrylamide) microgel particles: fundamental and commercial implications. Colloids and Surfaces 149, 57–64. Schulze, H.J., 1984. Physico-Chemical Elementary Processes in Flotation. Elsevier, Amsterdam. Sun, T., Wang, G., Feng, L., Liu, B., Ma, Y., Jang, L., Zhu, D., 2004. Reversible switching between reversible hydrophilicity and superhydrophobicity. Angewante Chemie International Edition 43 (3), 357–360. Weisberg, S., 2005. Applied Linear Regression. Wiley Series in Probability and Statistics. John Wiley & Sons Inc., New Jersey. Yordan, J.L., Yoon, R.H., 1986. Induction time measurement for the quartz amine flotation system. SME Annual Meeting 1986, New Orleans. Yordan, J.L., Yoon, R.H., 1988. Induction time measurement for coal flotation system. In: Attia, Y.A., Moudgil, B.M., Chander, S. (Eds.), Interfacial Phenomena in Biotechnology, Process Technology. Elsevier, New York.