Adhesion of Single Polyelectrolyte Molecules on Silica, Mica, and

To understand the role of HPAM, single-molecule force spectroscopy was employed for the first time to measure the desorption/adhesion forces of single...
0 downloads 0 Views 260KB Size
Download PDF
1652

Langmuir 2006, 22, 1652-1659

Adhesion of Single Polyelectrolyte Molecules on Silica, Mica, and Bitumen Surfaces Jun Long, Zhenghe Xu,* and Jacob H. Masliyah Department of Chemical and Materials Engineering, UniVersity of Alberta, Edmonton, Alberta, Canada T6G 2G6 ReceiVed October 12, 2005. In Final Form: NoVember 24, 2005 In a recent study (Energy Fuels 2005, 19, 936), a partially hydrolyzed polyacrylamide (HPAM) was used as a process aid to recover bitumen from oil sand ores. It was found that HPAM addition at the bitumen extraction step not only improved bitumen recovery but also enhanced fine solids settling in the tailings stream. To understand the role of HPAM, single-molecule force spectroscopy was employed for the first time to measure the desorption/adhesion forces of single HPAM molecules on silica, mica, and bitumen surfaces using an atomic force microscope (AFM). Silicon wafers with an oxidized surface layer and newly cleaved mica were used, respectively, to represent sand grains and clays in oil sands. The force measurements were carried out in deionized water and in commercial plant process water under equilibrium conditions. The desorption/adhesion forces of HPAM obtained on mica, silica, and bitumen surfaces were ∼200, 40, and 80 pN in deionized water and ∼100, 50, and 40 pN in the plant process water, respectively. The measured adhesion forces together with the zeta potential values of these surfaces indicate that the polymer would preferentially adsorb onto clay surfaces rather than onto bitumen surfaces. It is the selective adsorption of HPAM that benefits both bitumen recovery and tailings settling when the polymer was added directly to the bitumen extraction process at an appropriate dosage.

1. Introduction The adsorption of polyelectrolytes at solid-liquid or liquidliquid interfaces plays a critical role in many biological systems and industrial processes. In biological systems, for example, DNA and RNA molecules, proteins, and many polysaccharides are polyelectrolytes.2 Among polysaccharides, hyaluronic acid has an important function in the lubrication of joints, and the complex glucoprotein mucin is the main building block of the protective coating on many internal surfaces in the body. In many industrial processes, polyelectrolytes are also widely used, for instance, for rheology control, as wet and dry strength additives, as flocculating or dispersing agents, and for surface conditioning.3 In a recent study,1 a partially hydrolyzed polyacrylamide (HPAM) was used for the first time as a process aid to recover bitumen from oil sand ores using a water-based bitumen extraction process. It was found that HPAM addition at the bitumen extraction step improved not only bitumen recovery but also fine solids settling in the tailings stream. It is evident that, whether in biological systems or industrial applications, the binding of polyelectrolytes at solid-liquid interfaces plays a controlling role in determining the functionality of the polyelectrolytes. Understanding the binding forces can provide fundamental insights into the behavior of the polyelectrolytes. The present study focuses on understanding the role of the aforementioned HPAM in bitumen extraction by directly measuring the adhesion of HPAM molecules on the surfaces of various components present in oil sands. In a typical water-based bitumen extraction process,4 oil sand lumps are mixed with water and process aids such as sodium * To whom correspondence [email protected].

should

be

addressed.

E-mail:

(1) Li, H.; Long, J.; Xu, Z.; Masliyah, J. H. Energy Fuels 2005, 19, 936. (2) Claesson, P. M.; Poptoshev, E.; Blomberg, E.; Dedinaite, A. AdV. Colloid Interface Sci. 2005, 114-115, 173. (3) Tripathy, S. K., Kumar, J., Nalwa, H. S., Eds. Handbook of Polyelectrolytes and Their Applications; American Scientific Publishers: Stevenson Ranch, CA, 2002; Vol. 3.

hydroxide to form a slurry. Bitumen is released or “liberated” from the sand grains in the slurry, and the liberated bitumen is aerated with air bubbles. The aerated bitumen is then separated and recovered as a bitumen froth by flotation. Such a process includes two essential microsubprocesses: bitumen liberation and aeration.5 These two microsubprocesses determine bitumen recovery. Normally, the water-based processes can achieve high bitumen recovery (90% or higher) for good processing ores. However, for poor processing ores containing a relatively large amount of fine solids, it is difficult to obtain high bitumen recovery and acceptable bitumen froth quality.4 The poor bitumen recovery in such systems has been attributed to the so-called slime coating of bitumen droplet surfaces (i.e., bitumen surfaces coated with a layer of fine particles). The presence of the slime coating not only reduces the bitumen flotation rate and recovery by setting up a steric barrier that retards bitumen drops to contact air bubbles but also deteriorates the froth quality by carrying the attached fine solids to the froth product, which bears significant implications for subsequent bitumen froth cleaning. To avoid the slime coating and thus to improve bitumen recovery for poor processing oil sand ores, a preliminary attempt was made recently by Li et al.1 They added a partially hydrolyzed polyacrylamide (HPAM) directly to the bitumen extraction process of a poor processing ore and found that both bitumen recovery and tailings (fine solids) settling were significantly improved. HPAM is an anionic polyelectrolyte and has been identified as an effective flocculant in oil sand tailings treatment.6-8 Because it was used for the first time as a process aid to recover bitumen from oil sands, its role in the recovery (4) Masliyah, J. H.; Zhou, Z.; Xu, Z.; Czarnecki, J.; Hamza, H. Can. J. Chem. Eng. 2004, 82, 628. (5) Liu, J.; Xu, Z.; Masliyah, J. H. AIChE J. 2004, 50, 1927. (6) Cymerman, G.; Kwong, T.; Lord, E.; Hamza, H.; Xu, Y. In Polymers in Mineral Processing; Laskowski, J. S., Ed.; Canadian Institute of Mining, Metallurgy and Petroleum: Montreal, 1999. (7) Sworska, A.; Laskowski, J. S.; Cymerman, G. Int. J. Miner. Process. 2000, 60, 143. (8) Long, J.; Li, H.; Xu, Z.; Masliyah, J. H. AIChE J. 2005, 25, 371.

10.1021/la052757f CCC: $33.50 © 2006 American Chemical Society Published on Web 12/22/2005

Adhesion on Silica, Mica, and Bitumen Surfaces

process and how it improves bitumen recovery remain to be investigated. Li et al.1 speculated that the polymer could enhance the flocculation of fines while reducing the heterocoagulation between fines and bitumen. To have such an effect, the polymer must be able to adsorb selectively on the surface of fines and flocculate them so that they have less chance to slime-coat bitumen droplets. Therefore, it becomes essential to investigate the adsorption selectivity of HPAM on the surfaces of various components present in oil sands, including sand, bitumen, and clays. Measuring interaction forces between HPAM molecules and these surfaces is a direct and very powerful method of accomplishing this. With the development of an AFM-based technique, singlemolecule force spectroscopy (SMFS), it has become possible to measure inter- and intramolecular forces precisely on the scale of a single molecule.9 This technique has been successfully used to study the stretching of single biological macromolecules and synthetic polymer chains. Recently, it was applied to investigate the desorption of single polymer chains preadsorbed on solid substrates. The forces needed to pull single polymer molecules off a solid substrate were measured for a number of systems.10-20 The measured forces are denoted as “desorption forces” to make it clear that these forces are measured when detaching a polymer from a surface. The force spectroscopy measurements may be performed under equilibrium or nonequilibrium conditions.16,21-23 Nonequilibrium conditions are typically reached if the force loading time is comparable to or shorter than the natural lifetime of the investigated bonds. The rupture of the bonds is an irreversible process, and the rupture forces depend on the force loading rates (force/time). In contrast, if the dissociation and reassociation of the binding partners occur on a much shorter time scale, the polymer spacer connecting two neighboring bonds will not be stretched and therefore no rupture events is observed. In this case, the force measurement represents a process of continually peeling off the polymer chain from the surface and thus takes place under thermal equilibrium conditions. The measured forces under such conditions are force loading rate independent, and the investigated process is reversible. Thus, the measured desorption forces are identical to the equilibrium adsorption forces or the adhesion forces. In the present study, the SMFS technique was employed for the first time to investigate the adsorption of single HPAM molecules on sand, clay, and bitumen surfaces. The desorption/ adhesion forces were measured on these surfaces under equilibrium conditions. The goal is to provide new insights into the (9) Janshoff, M. N.; Oberdo¨rfer, Y.; Fuchs, H. Angew. Chem. 2000, 112, 3346. (10) Hugel, T.; Grosholz, M.; Clausen-Schaumann, H.; Pfau, A.; Gaub, H.; Seitz, M. Macromolecules 2001, 34, 1039. (11) Conti, M.; Bustanji, Y.; Falini, G.; Ferruti, P.; Stefoni, S.; Samorı`, B. ChemPhysChem 2001, 2, 610. (12) Zhang, W.; Cui, S.; Fu, Y.; Zhang, X. J. Phys. Chem. B 2002, 106, 12705. (13) Seitz, M.; Friedsam, C.; Jostl, W.; Hugel, T.; Gaub, H. E. ChemPhysChem 2003, 4, 986. (14) Cui, S.; Zhang, W.; Xu, Q.; Wang, C.; Zhang, X. Macromol. Symp. 2003, 195, 109. (15) Cui, S. Liu, C.; Zhang, X. Nano Lett. 2003, 3, 245. (16) Friedsam, C.; Seitz, M.; Gaub, H. E. J. Phys.: Condens. Matter 2004, 16, S2369. (17) Friedsam, C.; Becares, A. D. C.; Jonas, U.; Gaub, H. E.; Seitz, M. ChemPhysChem 2004, 5, 388. (18) Friedsam, C.; Becares, A. D. C.; Jonas, U.; Seitz, M.; Gaub, H. E. New J. Phys. 2004, 9, 1. (19) Cui, S.; Liu, C.; Wang, Z.; Zhang, X.; Strandman, S.; Tenhu, H. Macromolecules 2004, 37, 946. (20) Shi, W.; Wang, Z.; Cui, S.; Zhang, X.; Bo, Z. Macromolecules 2005, 38, 861. (21) Evans, E.; Ritchie, K. Biophys. J. 1997, 72, 1541. (22) Merkel, R.; Nassoy, P.; Leung, A.; Ritchie, K.; Evans, E. Nature 1999, 397, 50. (23) Liphardt, J.; Dumont, S.; Smith, S. B.; Tinoco, I., Jr.; Bustamante, C. Science 2002, 296, 1832.

Langmuir, Vol. 22, No. 4, 2006 1653 Table 1. Some Properties of the Plant Process Watera K+

Na+ Mg2+ Ca2+ Cl- NO3- SO42- HCO3- pH surface tension, mJ/m2 14.3 503 19 48 431 1.48 63.1 647 8.4 68.5 a

Ion concentrations are in ppm.

selective adsorption of HPAM on these surfaces and consequently to understand clearly the role of HPAM in bitumen recovery. 2. Experimental Section Materials. The polyelectrolyte used is a partially hydrolyzed polyacrylamide (HPAM) obtained from Ciba Specialty Chemicals. It is a copolymer of sodium acrylate (CH2CHCOONa) and acrylamide (CH2CHCONH2) with a molecular weight of ∼17 500 000 and 22% anionicity.24 This water-soluble polyelectrolyte is commercially referred to as Percol 727. Because organic carboxylic acid has a pKa value around pH 4.5,25,26 the polymer will carry negative charges in a solution at pH above 4.5. In the current study, the pH values of the deionized water and plant process water are ∼7 and ∼8.2, respectively. At such high pH values, most of the carboxylic groups in HPAM molecules are negatively charged. To represent the sand grain surface, silicon wafers with an oxidized surface layer of ∼0.6 µm obtained from NANOFAB (University of Alberta, Canada) were used. The wafers were cut into small pieces of 10 × 10 mm2. Prior to the deposition of HPAM onto the surface for force measurements, these small wafers were thoroughly rinsed with deionized water and ethanol, followed by blow drying with ultrapure-grade nitrogen. They were then exposed to an ultraviolet light for more than 5 h to remove any possible organic contaminants. Newly cleaved mica sheets were used to represent the clay surface. The bitumen surface was prepared by coating a thin layer (∼100 nm) of bitumen onto the aforementioned 10 × 10 mm2 silicon wafers using a spin coater. Vacuum-distillation-feed bitumen, containing 17 wt % asphaltene, provided by Syncrude Canada Ltd. was used in this study. A detailed description on the preparation of the bitumen surface and the characteristics of the prepared bitumen surface can be found elsewhere.27 In the present study, both deionized water and industrial process water were used as the aqueous medium for the desorption force measurements. Deionized water with a resistivity of 18.2 MΩ cm was prepared with an Elix 5 followed by a Millipore-UV Plus water purification system (Millipore Inc., Canada). As in commercial oil sand operations, bitumen extraction is always carried out using plant recycle process water. To better understand the role of HPAM in bitumen extraction, industrial process water was also used for the force measurements. In this case, recycle plant process pond water from the Aurora commercial plant of Syncrude Canada Ltd. was used. Some relevant properties of the plant process water are given in Table 1. HPAM Deposition on Substrates. To pick up a single-molecule chain in the force measurements, a thin layer of HPAM had to be deposited onto the substrates, and an appropriate concentration of HPAM solutions had to be used for the deposition.28 In preliminary experiments, we tried a series of HPAM solution concentrations from 1 to 500 ppm. It was found that a concentration of 50 ppm gave the highest probability of picking up a single polymer chain on a silica surface. In this case, about 100 to 200 force curves from 1000 force curves show characteristic plateaus that indicate single (or multiple) chain desorption. In the current study, a 50 ppm HPAM concentration was chosen. The deposition procedure was as follows. (24) Alonos, E. A.; Laskowski, J. S. In Polymers in Mineral Processing; Laskowski, J. S., Ed.; Canadian Institute of Mining, Metallurgy and Petroleum: Montreal, 1999. (25) Kanatharana, J.; Sukpisan, J.; Sirivat, A.; Wang, S. Q. Polym. Eng Sci. 1996, 36, 2986. (26) Graveling, G. J.; Ragnarsdottiur, K. V.; Allen, G. C.; Eastman, J.; Brady, P. V.; Balsley, S. D.; Skuse, D. A. Geochim. Cosmochim. Acta 1997, 61, 3515. (27) Liu, J.; Xu, Z.; Masliyah, J. H. Langmuir 2003, 19, 3911. (28) Zhang, W.; Zou, S.; Wang, C.; Zhang, X. J. Phys. Chem. B 2000, 104, 10258.

1654 Langmuir, Vol. 22, No. 4, 2006 HPAM was dissolved in either deionized water or the plant process water at a concentration of 50 ppm. About 2 to 3 drops of the prepared HPAM solution was deposited onto a precleaned substrate (silica, mica, or bitumen) in a dust-free environment. The substrate was then left undisturbed for several hours. After it was completely dry, the substrate was rinsed using deionized water to remove unattached molecules and then blown dry with pure nitrogen. The substrate was then immediately used for force measurements. Force Measurement. A Nanoscope E atomic force microscope (AFM) with a vendor-supplied fluid cell (Digital Instruments, Santa Barbara, CA) was used for the desorption force measurements. Silicon nitride probes, also from Digital Instruments, were used as received. Cantilevers of lever type 200 µm narrow with a spring constant of ∼0.06 N/m were employed to measure the forces. The real spring constant of the cantilevers was calculated from its geometry as determined from scanning electron micrographs.29 The nonspecific interactions between an AFM tip and a polymer chain could be up to 1 nN in magnitude,9,12 which enables the measurement of weak intermolecular interactions by SMFS. Force measurements were performed in an aqueous environment in a fluid cell. The liquids used in the present study included deionized water and the plant process water (without polymer addition). The water used in desorption force measurements was the same as that used in HPAM deposition. More specifically, if the HPAM deposition on the substrate was carried out in deionized water, then the force measurements were performed in deionized water. If the plant process water was used for the deposition, then it was correspondingly used in the force measurements. The procedure for measuring desorption forces is similar to the standard procedure of colloid force measurement.30,31 Briefly, a prepared substrate mounted on the AFM scanner was brought into contact with an AFM tip. Because of the nonspecific interactions between the polymer and the tip, one or more molecules would adsorb onto the tip. During the retraction process or the separation between the tip and the substrate, the molecule(s) was (were) pulled off of the substrate. The deflection of the cantilever as a function of the extension was recorded and then converted to a force-extension curve. To ensure sufficient time for the polymer to adsorb onto the tip, very slow piezo expansion/retraction speeds, 40-1000 nm/s, were used in the present study. Over this speed range, the measured desorption forces were independent of the speed used, indicating the measurement of equilibrium desorption forces. Comparing with the speeds used by other investigators for equilibrium force measurements (e.g., 70-3000 nm/s,20 60-2720 nm/s,19 and 1500 nm/s16), the speeds used in the present study were sufficiently small to ensure that the force measurements were performed under equilibrium conditions. As in standard contact mode, the AFM would immediately start to scan over the sample surface after tip engagement. To ensure that the scanner moved in the Z direction only, the scan size was set to zero. Under such conditions, the deposited molecules on the substrate were not disturbed by the tip scan in these two directions. All of the force measurements were conducted at room temperature (23.0 ( 1.5°C). For each test condition, the measurement was performed at a number of different surface locations, and 2 to 3 substrate-tip pairs were used. When we picked up molecule strands with AFM tips, the same maximum loading force of ∼4.5 nN was used. To obtain more representative results for single-molecule desorption, more than 1000 force curves were recorded for each test condition. Normally, only 10-20% of these force curves indicate single-molecule desorption.

3. Force Curve Analysis A typical desorption force curve with a long plateau is shown in Figure 1a. The sharp peak in the initial part of the force curve (near zero extension) corresponds to a very strong nonspecific adhesion between the bare tip and the uncovered regions of the (29) Veeramasuneni, S.; Yalamanchili, M. R.; Miller, J. D. J. Colloid Interface Sci. 1996, 184, 594. (30) Ducker, W. A.; Senden, T. J.; Pashley, R. M. Langmuir 1992, 8, 1831. (31) Long, J.; Xu, Z.; Masliyah, J. H. Energy Fuels 2005, 19, 1440.

Long et al.

Figure 1. Force curve analysis. (a) A typical desorption force curve with a single plateau obtained under equilibrium conditions. In this case, the AFM tip picked up a single polymer chain/strand as shown in the inset. (b) A typical desorption force curve with multiple plateaus. (c) A typical force curve obtained under nonequilibrium conditions or when the tip picked up multiple overlapped polymer strands as shown in the inset. (d) A typical force curve showing only the interaction between the bare tip and the substrate when no polymer strands were picked up as shown in the inset.

substrate.10 The AFM tip has a curvature of ∼20 nm at its apex. Comparing with the cross section of an extended polymer chain, the tip is much larger. In the case in which one polymer chain/ strand was captured by the tip as shown in the inset of Figure 1a, only a small fraction of the apex surface area was occupied by the polymer, and the strong interaction between the tip and the substrate was retained, inducing the sharp peak at the beginning of the force curve. The following plateau of the force curve suggests that there was a constant force to desorb the polymer chain from the substrate surface. Because of the electrostatic

Adhesion on Silica, Mica, and Bitumen Surfaces

repulsion among the negatively charged functional groups along HPAM chains, the polymer chains assume an extended conformation in solutions. Consequently, the polymer chains deposited onto the substrate surface would be expected to have a trainlike conformation. When the AFM tip picked up such an adsorbed chain, the tip retraction progressively unzipped the sequence of its binding sites with the surface, and the pulling-off force remained constant, thus resulting in a long plateau on the force curve.11,15,16,19 The length of the plateau directly reflects the length of the detached polymer chain from the substrate, whereas the height of the plateau corresponds to the desorption force required to detach the polymer chain from the substrate surface. The presence of such plateaus also indicates that the measurement was performed under equilibrium conditions.16 Figure 1b shows a typical desorption force curve with multiple plateaus. Three plateaus are observed on this curve. When the substrate surface was covered by a layer of polymer molecules, it was highly possible that the AFM tip could pick up multiple chains adsorbed on the substrate. If these chains do not overlap with each other and the adsorbed segments have different lengths, multiple plateaus would be expected to appear on the desorption force curve. Each time a polymer strand was completely desorbed, a step on the desorption force curve would be observed until the last strand was fully detached from the substrate. The length of the plateau directly reflects the length of an adsorbed polymer chain, whereas the height of a plateau corresponds to the desorption force that is required to desorb one or multiple polymer strands from the substrate surface. In Figure 1b, the desorption forces represented by plateaus 1, 2, and 3 are about 320, 160, and 80 pN, respectively. These results indicate that plateau 1 represents the desorption of four adsorbed polymer strands. When two of the four strands were completely pulled off of the substrate, plateau 2 was obtained. This plateau represents the desorption of the two remaining strands as shown in the inset. After one more strand was pulled off, the last plateau, which reflects the desorption of a single strand, was recorded. If the force measurements were performed under nonequilibrium conditions16 or the adsorbed chains picked up by the AFM tip were overlapped as shown in the inset of Figure 1c, the force curves obtained would look typically like the one shown in Figure 1c. In this case, no clear plateau is observed. Figure 2d shows a typical force curve that features only one sharp peak at the beginning of the curve. As mentioned earlier, the sharp peak represents the interaction between the bare AFM tip and the substrate. In this case, no polymer strand was picked up by the AFM tip as shown in the inset of Figure 1d. A detailed analysis of typical features of desorption force curves under equilibrium and nonequilibrium conditions is given by Friedsam et al.16 In this study, only force curves with one or more clear plateaus are presented.

4. Results and Discussion Desorption Forces in Deionized Water. The measured desorption forces of HPAM molecules on silica surface in deionized water are shown in Figure 2. Representative forceextension curves are presented in Figure 2a. In most cases, only one plateau on each force curve was observed. This suggests that the desorption of two or more adsorbed polymer strands in parallel was rarely observed. The sharp peaks at the beginning of the curves show that the adhesion forces between the silica surface and the bare silicon nitride tip in deionized water could reach 600 pN. From the height of the long plateaus, the desorption force was determined. A distribution of the desorption forces obtained is given in Figure 2b. From the force distribution, the

Langmuir, Vol. 22, No. 4, 2006 1655

Figure 2. Desorption force of HPAM molecules on a silica surface in deionized water. (a) Four typical force curves obtained. (b) Histogram of the desorption forces. The histogram is normal as shown by the smooth curve. (c) The effect of the force loading rates on the measured desorption forces.

mean desorption force (a) was 40 pN with a standard deviation (σ) of 6 pN. A normal distribution curve based on a ) 40 pN and σ ) 6 pN is also shown in Figure 2b. Comparing the distribution of the measured forces with the normal distribution, one can conclude that the desorption force distribution is normal. This indicates that the force measurements were carried out under equilibrium conditions.16 To further demonstrate that the measured desorption forces were obtained under thermal equilibrium conditions, the measurements were performed at a variety of force loading rates. The force loading rate (force/time) can be obtained from the product of tip-substrate separation speed and tip spring constant.21 Figure 2c shows the effect of the force loading rate on the measured desorption forces. Within the range of force loading rates used, from 2.4 to 60 nN/s, the desorption force is independent of the force loading rate. This implies that the dissociation time of the bonds between the HPAM chains and the substrate is much shorter than the time scale in the force measurements and that the desorption was carried out in an equilibrium state. Thus, the desorption forces measured between adsorbed HPAM chains and the silica substrate are equal to the binding forces of HPAM on the substrate. Similar loading rate independence was also found by other investigators.10-20 For example, Cui et al.19 reported that the desorption force of poly(2-acrylamido-2-methylpropanesulfonic acid) on amino-modified quartz in deionized water was independent of the force loading rate in the range of 2.28-116.1 nN/s.

1656 Langmuir, Vol. 22, No. 4, 2006

Figure 3. Desorption force of HPAM in deionized water from a mica surface. (a) Five typical force curves obtained. (b) Desorption force distribution.

Figure 4. Desorption force of HPAM molecules on a bitumen surface in deionized water. (a) Typical force curves obtained. Where applicable, Fi (i ) 1, 2, 3, ...) represents the desorption forces of i strands. The inset shows the relation of F1-F4. (b) Desorption force distribution. The solid curve is the normal distribution. A closer look of the distribution of F1 is given in the inset.

Figure 3 presents the results of desorption forces of HPAM molecules on a mica surface in deionized water. The force curves obtained (Figure 3a) are similar to those shown in Figure 2a. In most cases, there is also only one plateau on each force curve. The desorption force distribution is given in Figure 3b, and it is also normal. The average desorption force (a) is about 200 pN with a standard deviation (σ) of 9 pN. Typical results of the desorption force of HPAM molecules on a bitumen surface in deionized water are shown in Figure 4a. Different from the force curves obtained on silica and mica surfaces (Figures 2a and 3a), in this case, some of the force curves exhibit several plateaus. This feature of the force curves indicates that the desorption of multiple HPAM strands from

Long et al.

Figure 5. Desorption force of HPAM in the plant process water from a silica surface. (a) Typical force curves. (b) Desorption force distribution. The inset provides a closer look at the distribution of F1.

bitumen surfaces occurred in parallel (i.e., at the same time). In this Figure and where applicable in this article, Fi (i )1, 2, 3...) represents the desorption force of i adsorbed polymer strands in parallel. In general, Fi ) iF1 where F1 is the desorption force of a single strand. The average desorption forces for a single strand (F1) to four strands in parallel (F4) on bitumen surfaces are given in the inset of Figure 4a. The corresponding desorption force distributions are shown in Figure 4b. For a single-strand desorption, F1 ) 80.7 ( 6.7 pN where 80.7 is the average desorption force and 6.7 is the corresponding standard deviation. The solid curve represents the normal distribution. The inset provides a closer look of the distribution of F1. Although plateaus representing desorption forces of multiple strands at 160, 240, and 320 pN were observed, the total number of these plateaus is much smaller than the number of plateaus representing singlestrand desorption. It is evident that the probability of detecting the desorption force of a single strand is much higher than that of several strands in parallel. Desorption Force in the Plant Process Water. The results of desorption forces of HPAM on the silica surface measured in the plant process water are shown in Figure 5. Plateaus representing the desorption of multiple strands are observed from typical force curves in Figure 5a. This feature of force curves is different from the case in deionized water (Figure 2a) where only plateaus resulting from single-strand desorption are observed for the silica surface. In the plant process water, some force curves show the desorption of up to five strands in parallel. The desorption force distributions in Figure 5b show a higher probability of detecting the desorption of double or triple strands in parallel than that of a single strand. The desorption force of a single strand closely follows the normal distribution, as shown by the solid curve in Figure 5b, with a ) 50 pN and σ ) 6.2 pN. The inset of Figure 5b provides a closer look at the desorption force distribution of single strands with the normal distribution. The results of desorption forces of HPAM on the mica surface in the plant process water are shown in Figure 6. The typical force curves given in Figure 6a show that most of the force curves have one long plateau, representing the desorption of single-molecule strands. The average desorption forces are about

Adhesion on Silica, Mica, and Bitumen Surfaces

Figure 6. Desorption force of HPAM molecules on a mica surface in the plant process water. (a) Typical force curves. (b) Desorption force distribution. The inset provides a closer look at the distribution of F1.

Figure 7. Desorption force of HPAM molecules on a bitumen surface in the plant process water. (a) Typical force curves. (b) Desorption force distribution.

100 pN with a standard deviation of 6.9 pN. Although there are plateaus showing the desorption of multiple strands, the number of these plateaus is much smaller, and the length of these plateaus is also limited to less than about 10 nm. Figure 6b shows the desorption force distribution. The inset of Figure 6b shows a close match of the desorption force distribution of single-molecule strands with the normal distribution. Figure 7 shows the results of desorption forces of HPAM on the bitumen surface in the plant process water. In this case, the desorption of single and double strands occurred, but the desorption of three or more strands was rarely observed as shown by the typical force curves in Figure 7a. The force distributions given in Figure 7b shows that the probability of single-strand desorption is similar to that of double-strand desorption. The desorption force of a single strand has an average value of 40 pN with a standard deviation of 4.8 pN, and the distribution of

Langmuir, Vol. 22, No. 4, 2006 1657

Figure 8. (a) Comparison of desorption forces of a single strand on silica, mica, and bitumen surfaces. (b) Probability that no polymer strand is picked up by the AFM tip under various conditions.

single-strand desorption forces follows the normal distribution as shown by the solid curve in Figure 7b. Comparison of Desorption Forces on Different Surfaces. To compare the adsorption of HPAM on sand grains, clays, and bitumen, the desorption/adhesion forces of single HPAM strands obtained on silica, mica, and bitumen surfaces in deionized water and the plant process water are plotted together in Figure 8a. The results are also summarized in Table 2. In deionized water, the average desorption/adhesion forces of single adsorbed strands are 40, 80, and 200 pN on silica, bitumen, and mica surfaces, respectively. These results clearly indicate that HPAM can adsorb onto the three surfaces in deionized water but the adsorption strength is different. The force obtained on the mica surface (∼200 pN) is more than twice that on the bitumen surface (∼80 pN). In the plant process water, the force on the silica surface (∼50 pN) is very close to the force on the bitumen surface (40 pN). The desorption force on mica is about 100 pN, which is nearly twice that on the silica or bitumen surface. Regardless of the type of water used in the force measurements, Figure 8a shows that the adsorption force of HPAM on the mica surface is much stronger than on silica and bitumen surfaces. Another important point that needs to be noted is that during the force measurements the probability of not picking up any polymer strands in the plant process water was much higher than in deionized water. If no adsorbed strand was picked up by the AFM tip, the obtained force curve would be similar to the one shown in Figure 1d. By counting the number of such curves in 1000 force curves obtained, the frequency of not picking up any polymer strands was obtained as shown in Figure 8b. In deionized water, the frequencies are 9.5, 13.5, and 23.5% on mica, silica, and bitumen surfaces, respectively. In the plant process water, these numbers increase correspondingly to 17.5, 22.5, and 64.5%. These results indicate that more HPAM molecules were adsorbed on the mica surface than on the bitumen surface although the same concentration of HPAM solution was used in the deposition of HPAM on these surfaces. In particular, the frequency that no adsorbed HPAM strand was picked up on the bitumen surface in the plant process water reaches a significantly high value of 64.5%, indicating that many fewer HPAM molecules were adsorbed on the bitumen surface than on mica and silica surfaces.

1658 Langmuir, Vol. 22, No. 4, 2006

Long et al.

Table 2. Summary of the Measured Forces and Zeta Potentials desorption force of single HPAM strands, pN

silica bitumen mica

zeta potential, mV

in deionized water (pH ∼7)

in the plant process water (pH ∼8.4)

in 1 mM KCl solutions (pH ∼8.2)

in the plant process water (pH ∼8.4)

40 ( 6.0 80 ( 6.7 200 ( 9.1

50 ( 6.3 40 ( 4.8 100 ( 6.9

-59a -76a -45b

-30c -45c -20d

a Data obtained by Liu et al.27 b The value is for clay fines from both good and poor oil sand ores.46 of clay fines from a transitional oil sand ore.1

To better understand the measured forces, it is essential to understand the adsorption mechanisms of HPAM molecules on silica, mica, and bitumen surfaces. Several mechanisms have been suggested for the adsorption of HPAM on silica and mica surfaces. For example, some investigators32-34 considered that the driving force of the adsorption originates from attractive van der Waals forces between the polymer and the surfaces. Although there is an electrostatic repulsion between HPAM and the surfaces, several investigators still suggested that an electrostatic attraction can arise locally if the charge distribution on the surfaces is inhomogeneous (i.e., both negative and positive sites are present35-37). The adsorption can be caused by the spatial inhomogeneity of the surface charges, which creates attractive regions with charges different from the net charge of the surface. Others38-44 suggested the formation of hydrogen bonding as a possible adsorption mechanism. As for the adsorption of HPAM on the bitumen surface, no report has been found in the literature. To understand the adsorption mechanism, further investigation is needed. However, because the presence of surface oxygen on bitumen could provide sites to form hydrogen bonds with the hydrogen atoms of amide groups of HPAM molecules, it is possible that hydrogen bonding would contribute to the observed adsorption of HPAM on the bitumen surface. Because bitumen is a mixture, a possible surface charge inhomogeneity may also contribute to the adsorption of HPAM. In addition, surface charge inhomogeneity could also play an important role in the adsorption of HPAM in the plant process water. Because there are various electrolytes and surfactants present in the plant process water, the adsorption of these species on silica, mica, and bitumen surfaces could result in surface charge inhomogeneity and thus affect the adsorption of HPAM on these surfaces.

5. Role of HPAM in Bitumen Recovery Bitumen extraction is normally carried out under alkaline conditions. Under such conditions, the surfaces of various components present in the oil sand slurry (sand grains, clays, and bitumen) are negatively charged. Because HPAM molecules are also negatively charged, there is an electrostatic repulsion between (32) Holmberg, K.; Jonsson, B.; Kronberg, B.; Lindman, B. Surfactants and Polymers in Aqueous Solution; John Wiley & Sons: London, 2003; pp 403-435. (33) Dobrynin, A. V.; Rubinstein, M. J. Phys. Chem. B 2003, 107, 8260. (34) Zhao, F.; Dub, Y.-K.; Tang, J.; Li, X. C.; Yang, P. Colloids Surf., A 2005, 252, 153. (35) Samoshina, Y.; Diaz, A.; Becker, Y.; Nylander, T.; Lindman, B. Colloids Surf., A 2003, 231, 195. (36) Hoogeveen, N. G.; Stuart, M. A. C.; Fleer, G. J. J. Colloid Interface Sci. 1996, 182, 133. (37) Ellis, M.; Kong, C. Y.; Muthukumar, M. J. Chem. Phys. 2000, 112, 8723. (38) Graveling, G. J.; Ragnarsdottir, K. V.; Allen, G. J. C.; Eastman, J.; Brady, P. V.; Balsley, S. D.; Skuse, D. R. Geochim. Cosmochim. Acta 1997, 61, 3515. (39) GuCvellou, Y.; Noi’k, C.; Lecourtier, J.; Defives, D. Colloids Surf., A 1995, 100, 173. (40) Lee, L. T.; Somasundaranh, P. Langmuir 1989, 5, 854. (41) Peng, F. F.; Di, P. J. Colloid Interface Sci. 1994, 164, 229. (42) Yang, J. P.; Li, H. S. Chem. J. Chin. UniV. 1997, 18, 647. (43) Haschke, H.; Miles, M. J.; Sheppard, S. Single Mol. 2002, 2-3, 171. (44) Haschke, H.; Miles, M. J.; Koutsos, V. Macromolecules 2004, 37, 3799.

c

Data obtained by Zhao.45

d

The zeta potential

HPAM molecules and these surfaces. This repulsion would repel HPAM molecules from these surfaces. However, once the molecules attach to these surfaces, the adhesion forces would tend to hold them together. Therefore, the probability for HPAM molecules to attach to these surfaces is directly related to the magnitudes of both the electrostatic repulsion and adhesion force. For comparison, the measured zeta potentials of silica, bitumen, and clay fines in both the plant process water (pH ∼8.4) and 1 mM KCl solutions (pH ∼8.2) are listed in Table 2. Because the bitumen surface is the most negatively charged (-45 mV in the plant process water45 and -76 mV in a 1 mM KCl solution27) and the binding between HPAM and bitumen is the weakest (∼40 pN), the probability for HPAM molecules to attach to the bitumen surface is much less than that for attaching to the clay surface having the least negative charge (-20 mV in the plant process water1 and -45 mV in a 1 mM KCl solution46) and bears the strongest binding forces with HPAM (∼100 pN). In addition, in the present study, newly cleaved mica sheets were used to represent clays in oil sands. The desorption forces were measured only on the basal surfaces. However, in reality, HPAM molecules may adsorb on both the faces and edges of fine clay particles. It is well known that the edge and face of a clay lamella may carry opposite charges. Under certain conditions (e.g., pH