Mechanisms of ethyl(hydroxyethyl) cellulose–solid interaction: Influence of hydrophobic modification

Journal of Colloid and Interface Science 293 (2006) 322–332 www.elsevier.com/locate/jcis

Mechanisms of ethyl(hydroxyethyl) cellulose–solid interaction: Influence of hydrophobic modification Jing Wang, P. Somasundaran ∗ NSF Industry/University Cooperative Research Center for Advanced Studies in Novel Surfactants, Columbia University, New York, NY 10027, USA Received 5 May 2005; accepted 20 June 2005 Available online 2 August 2005

Abstract Hydroxyethyl cellulose and its hydrophobically modified derivatives are widely used in many industrial areas such as pharmaceuticals, cosmetics, textiles, paint and mineral industries. However, the interaction mechanisms of these biopolymers and solids have not been established. In this work, the interaction mechanism and conformation of hydrophobically modified ethyl(hydroxyethyl) cellulose (C14 -EHEC) have been investigated using spectroscopic, AFM and allied techniques. Comparison was made with corresponding unmodified analogue in order to investigate the effects of the hydrophobic modification. Electrokinetic studies showed that polysaccharides adsorption decreased the negative zeta potential of talc but did not reverse the charge. EHEC adsorption on talc was not found to be affected significantly by changes in solution conditions such as pH and ionic strength, ruling out electrostatic force as the controlling factor. However, HM-EHEC adsorption was found to increase markedly with increase in ionic strength from 0.1 to 1 suggesting a role for the hydrophobic force in this adsorption process. Fluorescence spectroscopic studies conducted to investigate the role of hydrophobic bonding using pyrene probe showed no evidence of the formation of hydrophobic domains at talc–aqueous interface. Urea, a hydrogen bond breaker, reduced the adsorption of HM-EHEC on talc markedly. In FTIR study, the changes in the infrared bands, associated with the C–O stretch coupled to the C–C stretch and O–H deformation, were significant and therefore support strong hydrogen bonding of HM-EHEC on the solid surface. Moreover, Langmuir modeling of the adsorption isotherms suggests hydrogen bonding to be a major force for the adsorption of EHEC and C14 -EHEC on solid since the adsorption free energies of these polymers were close to that for hydrogen bond formation. All of the above results suggest that the main driving force for EHEC adsorption on talc is hydrogen bonding rather than electrostatic interaction or hydrophobic force. For hydrophobically modified C14 -EHEC, hydrophobic force plays a synergetic role in adsorption along with hydrogen bonding. From computer modeling and AFM imaging, it is proposed that C0 -EHEC and C14 -EHEC adsorb flat on talc with ethylene oxide side chains and hydrophobic groups protruding out from the surface into bulk water phase.  2005 Elsevier Inc. All rights reserved. Keywords: Talc; EHEC; Hydrophobic modification; Polymer adsorption; Fluorescence; FTIR; AFM; Model; Zeta potential

1. Introduction The mineral talc is a hydrous magnesium silicate with the chemical formula Mg3 Si4 O10 (OH)2 [1]. Its elementary sheet is composed of a layer of magnesium-oxygen/hydroxyl octahedra sandwiched between two layers of silicon-oxygen tetrahedra. The main surfaces of this sheet do not contain hydroxyl groups or active ions and hence the hydrophobicity. * Corresponding author.

E-mail address: [email protected] (P. Somasundaran). 0021-9797/$ – see front matter  2005 Elsevier Inc. All rights reserved. doi:10.1016/j.jcis.2005.06.072

The edges of the sheets are, however, composed of hydroxyl, silicon, magnesium and oxygen ions and this contributes to its hydrophilicity. The nonionic polymers used in the present investigation, ethyl(hydroxyethyl) cellulose (EHEC) and hydrophobically modified ethyl(hydroxyethyl) cellulose (HM-EHEC) are important associative thickeners. The repeating units of EHEC and HM-EHEC are shown in Fig. 1. Previous studies have dealt with the rheology and phase behavior of HM-EHEC in an aqueous solution [2–4], with less emphasis on the study of interactions between these polymers and solids, which is

J. Wang, P. Somasundaran / Journal of Colloid and Interface Science 293 (2006) 322–332

323

(a)

(b) Fig. 1. The monomeric structure of (a) EHEC, (b) HM-EHEC.

essential for coating, dispersion, flotation and other applications. Specific interaction mechanisms between polysaccharides and solid surfaces are generally not well established. While Balajee [5], Afenya [6] and Jucker [7] have attributed the adsorption of polysaccharides to hydrogen bonding, Liu and Laskowski [8] observed that the surface metallic sites were responsible for polysaccharide adsorption. Other workers suggested polysaccharide adsorption to involve physical interaction via hydrophobic bonding [9–11]. Steenberg and Harris [12] have proposed a combination of hydrophobic and hydrophilic interaction, with polysaccharides adsorbing firstly on the faces of talc via hydrophobic force, and then on the hydrophilic edges via hydrogen bonding. If their conclusion is correct, then the hydrophobic modification of polysaccharide would strengthen the interaction between the polymer and the solid and improve polymer adsorption significantly. However, we have observed recently that hydrogen bonding rather than hydrophobic force is the governing mechanism for polysaccharide such as guar gum on talc [13]. Therefore, to study the role of hydrophobic force on polysaccharide adsorption, EHEC was hydrophobically modified and two polymers were comparatively studied in this work. The objective of this work was to clarify the mechanistic aspects of the interactions between EHEC/C14 -EHEC and talc using a combination of spectroscopic, microscopic, electrostatic adsorption techniques and molecular modeling.

2. Experimental 2.1. Materials Ethyl(hydroxyethyl) cellulose (EHEC) and hydrophobically modified ethyl(hydroxyethyl) cellulose (HM-EHEC) were obtained from Akzo-Nobel Inc. They have approximately the same weight average molecular weight of 100,000. The hydrophobic moiety in the HM-EHEC is tetradecyl (C14 ). All EHEC and HM-EHEC stock solutions were prepared by adding powder into vigorously stirred water and furthur stirring for 30 min. Talc sample of 40 µm size and BET surface area of 19.75 m2 /g was obtained from Cytec Industries. Mica in the form of flat plate purchased from Ted Pella Inc. was used in AFM studies. pH of the solutions and suspensions was adjusted using Fisher standard hydrochloric acid or sodium hydroxide solutions. Reagent grade potassium chloride from APACHE Chemicals Inc. was used to adjust the ionic strength of solutions. 1-pyrene butyric acid and 1,3-dicyclohexyl carbodiimide (DCC) bought from Aldrich were used in the fluorescence probe labeling of the polymers. Urea (Fisher Chemical), phenol (EM Science) and 98% sulfuric acid (Amend Drug & Chemical Co. Inc.) were used for colorimetric experiments. The water used was triply distilled.

324

J. Wang, P. Somasundaran / Journal of Colloid and Interface Science 293 (2006) 322–332

2.2. Experiments Experiments done along with the adsorption measurements to investigate the adsorption mechanisms include FTIR for exploring the role of hydrogen bonding in adsorption, fluorescence spectroscopy to study the role of hydrophobic force, AFM and molecular modeling to explore the conformation of the adsorbed polysaccharides.

Stock solution of pyrene-labeled polymer was then added to the talc suspension. The vials containing the suspensions were quickly wrapped with aluminum foil to keep the light away and left to condition overnight. Finally, the hydrophobicity of the supernatant and the solid were analyzed separately using a LS-1 fluorescence spectrophotometer (Photon Technology International). 2.8. AFM analysis

2.3. Adsorption measurements The suspensions of solids, with ionic strength adjusted using KCl and pH adjusted using HCl or KOH at 10% solid loading, were ultrasonicated for 30 min and then stirred magnetically for 2 h. Polymer stock solution was then added to the solid suspension and left for conditioning overnight. The suspensions were centrifuged and the supernatants pipetted out for determination of guar concentration by a Total Organic Carbon Analyzer. The adsorption density of polymer on solid was calculated from the data for initial and residual polymer concentrations. 2.4. Colorimetric method A colorimetric method described by Dubois et al. [14] was used in the experiments. 0.05 ml 80% phenol and 5 ml 98% sulfuric acid were added to 2 ml of supernatant obtained after centrifugation, and after 4 h of color development under warm conditions, absorbance was measured at a wavelength of 487.5 nm using SHIMADZU UV-1201 UV– vis spectrophotometer. Adsorption density of polymer on talc was calculated from the difference in absorbance.

The atomic force microscopy (AFM) system used was a Nanoscope III from Digital instruments. The measurements were performed in air or underwater in tapping mode using a V-shaped Si3 N4 cantilever covered with gold on the back for laser beam reflection. Two types of tips (MESP-10 and NP20) were used in air and underwater respectively. All images were collected in the height mode, which keeps the force constant. 2.9. Molecular modeling Molecular modeling (Macromodel) is a powerful tool for exploring the conformation of polysaccharide in solvents. Macromodel was applied to study the conformation of these polymers. Accelrys Material Studio was also used in helping to understand the conformation of adsorbed polysaccharides on solid.

3. Results and discussion 3.1. Molecular modeling

2.5. Electrokinetic measurements Small amounts of talc were added to 10−3

M KNO3 solution and ultrasonicated for 30 min, magnetically stirred for 2 h and the pH adjusted to the desired value using HCl or KOH. Finally, the polymer stock solution was added and left for conditioning for 16 h. The zeta potential was then measured using a Zeta meter. 2.6. FTIR study The infrared spectra of talc, polysaccharides and polymer adsorbed talc were recorded using a Model FTS7000 series Fourier transform infrared spectrometer from Digilab operating in the range of 4000–350 cm−1 . Approximately 2 mg of the desired powder sample was thoroughly mixed with 200 mg of spectroscopic grade KBr and pressed into pellets for recording the spectra. 2.7. Fluorescence spectroscopy Suspensions of talc at 10% solid loading were ultrasonicated for 30 min and mixed for 2 h with a magnetic stirrer.

Fig. 2 shows the computer simulated molecular space filling structure of EHEC and C14 -EHEC in both vacuum and aqueous environments. Each of them contains 5 repeat units. It is seen that polymer chains tend to be in a more stretched helical form in water. The complex helical structure shown in Fig. 3 is a piece of polymer chain containing 50 repeat units. It is clear that the OH groups tend to stick out of the helix. For HM(C14 )-EHEC, from the longitudinal view of structure, it appears that the C14 hydrophobic tails stay in the center of the helical structure (Fig. 3). According to the simulation, there is no change in the simulated structure when the number of repeat units is increased further from 50 to 100. From computer simulation by Accelrys Material Studio (Fig. 4), it was found that both EHEC and C14 -EHEC molecules chose relatively flat conformation to adsorb on talc surface. Compared to Fig. 3, it is clear to see the significant change of the conformation of C14 -EHEC after adsorption on solid surface. Therefore, it can be proposed that C0 EHEC and C14 -EHEC adsorb on talc with ethylene oxide side chains and hydrophobic groups protruding out from the surface into bulk water to make the surface hydrophobic.

J. Wang, P. Somasundaran / Journal of Colloid and Interface Science 293 (2006) 322–332

325

Fig. 2. Computer simulated molecular CPK (space filling) structure of: (a) EHEC, (b) C14 -EHEC with and without aqueous environment (each contains five repeat units).

3.2. Electrokinetic studies It is evident from the electrokinetic data shown in Fig. 5 that the surface of talc is negatively charged and the isoelectric point (i.e.p.) is located around pH 2.5. Above pH 2.5, the electronegative character increases with increase in pH up to 4 and thereafter remains almost constant. The adsorption of EHEC and C14 -EHEC decreased the negative zeta potential of talc (without any significant shift of i.e.p.) but did not reverse the charge, which indicates that the adsorption of these polymers only masks the surface charge on talc since the shift of shear plane due to adsorbed polymer layer is the reason for the change in zeta potential upon polymer adsorption. 3.3. Adsorption studies 3.3.1. Influence of pH Figs. 6 and 7 show adsorption isotherms of EHEC and C14 -EHEC on talc at pH 3 and pH 9. It can be seen that the adsorption density does not vary measurably between pH 3 and 9. If electrostatic interaction plays an important role in the adsorption process, the adsorption density of polymer on talc should have been affected significantly since the surface charge of talc does change with change in pH from 3 to 9. Such is not the case here, and it can be concluded that

electrostatic interaction is not the dominant force for the adsorption of EHEC and C14 -EHEC on talc. 3.3.2. Influence of ionic strength Adsorption isotherms of EHEC and C14 -EHEC on talc at pH 9 at different ionic strengths are shown in Figs. 8 and 9. If hydrophobic force makes a significant contribution to adsorption of a polymer on a mineral, the adsorption density would have increased with the addition of salt due to “salting-out” effect [15]. For EHEC, there is no significant change in the adsorption density as salt concentration is increased from 0 to 0.1 and 1 M. Thus it would appear that the hydrophobic force is not the main driving force for the adsorption of EHEC on talc. In the case of C14 -EHEC, almost no effect is observed with the increase of salt concentration from 0 to 0.1. However, as salt concentration increased further from 0.1 to 1, adsorption density increased markedly. Hydrophobic force may play a role in the case of adsorption of HM-EHEC on talc. 3.3.3. Adsorption on talc with and without urea To explore the role of hydrogen bonding, adsorption tests were carried out in the presence of urea, a hydrogen bond breaker. Urea is a strong hydrogen bonding acceptor and it can be expected to affect the hydrogen bonding between the solid and the polymer in solution by preferential formation of hydrogen bonds between themselves and the polysaccha-

326

J. Wang, P. Somasundaran / Journal of Colloid and Interface Science 293 (2006) 322–332

Fig. 3. Computer simulated (a) latitudinal view and (b) longitudinal view of molecular CPK (space filling) structure of EHEC and C14 -EHEC in aqueous environment (each contains 50 repeat units).

Fig. 4. Computer simulated structure of (a) EHEC and (b) C14 -EHEC molecule on talc surface.

J. Wang, P. Somasundaran / Journal of Colloid and Interface Science 293 (2006) 322–332

327

Fig. 5. Zeta potential of talc as a function of pH in the presence and absence of EHEC and HM-EHEC, 0.001 M KCl.

Fig. 6. Adsorption isotherm of EHEC on talc at different pH (no salt).

Fig. 8. Adsorption isotherm of EHEC on talc at different ionic strength (pH 9).

Fig. 7. Adsorption isotherm of HM-EHEC on talc at different pH (no salt).

Fig. 9. Adsorption isotherm of HM-EHEC on talc at different ionic strength (pH 9).

rides and water. Adsorption isotherms for C14 -EHEC on talc in the presence and absence of urea at pH 9 are shown in Fig. 10. It is evident that urea, the hydrogen bond breaker, re-

duces the adsorption of C14 -EHEC on talc markedly. Therefore, it is proposed that hydrogen bonding plays a important role in the adsorption of C14 -EHEC on talc.

328

J. Wang, P. Somasundaran / Journal of Colloid and Interface Science 293 (2006) 322–332

3.3.4. Modeling of polysaccharide adsorption on solid surface To obtain a better understanding of the adsorption mechanism of C14 -EHEC and EHEC on solids, the following parameters are extracted from the adsorption isotherms as shown in Table 1: standard free energy of adsorption
G0ads , maximum amount of polymer adsorbed per mass of solid (x/m)max and effective substrate area occupied per polymer chain σ 0 at the talc/solution interface.

The adsorption isotherms of EHEC and HM-EHEC on solids in this study exhibited a pseudo-Langmuirian behavior. According to the Langmuir adsorption model applied to the polymer systems [16], surface coverage θ = KC/(1 + KC) = (x/m)/(x/m)max .

(1)

Equation (1) can be expressed as
 m/x = 1/ K(x/m)max Ceq + 1/(x/m)max ,

(2)

where Ceq is the equilibrium solution concentration of polymer, x is the amount of polymer adsorbed, m is the mass of solid substrate, K is the Langmuir adsorption equilibrium constant and (x/m)max is the maximum amount of polymer adsorbed per mass of solid. A plot of m/x against 1/Ceq should yield a linear relationship. The values of K and (x/m)max can be determined from the intercept and slope of such a plot as shown in Figs. 11 and 12. Table 1 The calculated values of (x/m)max ,
G0ads and σ 0 for the adsorption of EHEC and HMC on talc at T = 25 ◦ C

Fig. 10. Adsorption isotherms of HM-EHEC on talc with and without urea.

Polymer

(x/m)max (mg/m2 )


G0ads (kJ/mol)

σ0 (nm2 )

EHEC HM-EHEC

1.31 0.93

−23.0262 −28.3803

126.3 178.65

Fig. 11. Langmuir plot for adsorption of EHEC on talc at pH 9, T = 25 ◦ C.

Fig. 12. Langmuir plot for adsorption of HM-EHEC on talc at pH 9, T = 25 ◦ C.

J. Wang, P. Somasundaran / Journal of Colloid and Interface Science 293 (2006) 322–332

329

Fig. 13. FTIR spectrum of C14 -EHEC.

The Langmuir adsorption equilibrium constant, K, can be considered to represent the affinity of a polymer for a particular surface. It can be related to the standard free energy of adsorption,
G0ads , from the expression given in Eq. (3):
G0ads = −RT ln K,

(3)

where R is the general gas constant (R = 8.314 J/(mol K)) and T is the absolute temperature. From the maximum adsorption density, the effective area occupied on the substrate surface per polymer chain, σ 0 , can be calculated using the following equation: σ0 = θ

A

, ns2 NA

(4)

where NA is Avogadro’s number. θ is the fraction of the surface which is covered by polymer, A is the substrate surface area, ns2 is the number of moles of polymer adsorbed. The number of moles of polymer adsorbed per unit area of solid, ns2 /A, is given by Eq. (5). ns2 (x/m)max , = A Mw

(5)

where Mw is the molecular weight of the polymer. Hence, the effective substrate area occupied per polymer molecule, σ 0 , can be calculated using Eq. (6). σ0 = θ

Mw . (x/m)max NA

(6)

From Table 1, it can be seen that
G0ads , calculated for EHEC and HM-EHEC indicates that adsorption of these polymers on talc is highly favored, i.e.
G0ads < 0. It is apparent that the values of
G0ads are around −25 kJ/mol, close to the free energy of hydrogen bond formation. Moreover, if hydrophobic interaction is the major force for adsorption, then a value of about −50 kJ/mol will be expected for
G0ads [17], which is much greater than the calculated value. These results thus further confirm the major role of hydrogen bonding rather than hydrophobic force in controlling CMC adsorption. In addition,
G0ads was determined to be more negative for the adsorption of HM-EHEC than the

other ones because of the synergetic effect of hydrophobic interaction in adsorption. For a particular polymer, the value of σ 0 will be determined by the conformation that the polymer adopts upon adsorption. Clearly, σ 0 will be much greater if the polymer is adsorbed mainly in “trains” (a conformation in which the majority of the polymer is in contact with the surface) than if it adsorbs in the form of a high degree of “loops” and “tails”. As shown in Table 1, the σ 0 of polysaccharides are very high, which means that most segments of these polymers adsorb as trains on talc. 3.4. FTIR spectroscopy For pure C14 -EHEC (Fig. 13), the band at 2924 cm−1 is due to C–H stretching of the –CH2 groups. The band due to ring stretching of glucose appears at around 1611 cm−1 . In addition, the bands in the region 1350–1450 cm−1 are due to symmetrical deformations of CH2 and COH groups. The bands due to primary alcoholic –CH2 OH stretching mode and CH2 twisting vibrations appear at 1078 and 1021 cm−1 , respectively. The weak bands around 770 cm−1 are due to ring stretching and ring deformation of α-D-(1–4) and α-D(1–6) linkages. Fig. 14 shows the spectra of talc, C14 -EHEC and C14 EHEC adsorbed talc. The FTIR spectrum of talc is shown in Fig. 14a. According to the characteristic IR frequencies of talc reported by other researchers [18], the band at around 1039 cm−1 can be assigned to the out-of-plane symmetric Si–O–Si mode. The Si–O bending vibration for talc has been observed at 432 cm−1 in the spectrum. The other bands in the spectrum around 654, 604, 550, 481 and 450 cm−1 are probably associated with various Mg–OH modes. The band at around 384 cm−1 appears to involve mixed vibrations of the Si–O network, the octahedral cations and the hydroxy groups. The band at 374 cm−1 is associated with the symmetric Mg–OH vibration. Talc also shows a single Mg–OH stretching band around 3677 cm−1 due to the centrosymmetric relationship between the hydroxy groups on both sides of the octahedral layers. The FTIR spectrum of

330

J. Wang, P. Somasundaran / Journal of Colloid and Interface Science 293 (2006) 322–332

Fig. 14. FTIR spectra of talc, C14 -EHEC and C14 -EHEC adsorbed talc.

Fig. 15. FTIR spectra of talc, C14 -EHEC and C14 -EHEC adsorbed talc (850–1200 cm−1 ).

talc after interaction with C14 -EHEC is depicted in Fig. 14b. It is apparent that several bands appear on the spectrum in the region 1000–1050 cm−1 (Fig. 15b) which is attributed to the hydrogen bond formation between the O of –CH2 – CH2 –O–CH2 –CH2 – and Si–O–H of talc edge after adsorption [19]. The changes in the infrared bands in the region 1000–1080 cm−1 , associated with the C–O stretch coupled to the C–C stretch and O–H deformation, were significant and support strong hydrogen bonding of C14 -EHEC to the solid surface, since hydrogen bonding would be expected to affect the C–O stretching.

emission spectrum will increase as the environment becomes more hydrophobic and is termed polarity parameter. Fluorescence data for the pyrene labeled EHEC and HM-EHEC adsorbed talc-solution interfaces are shown in Figs. 16 and 17. Pyrene fluorescence gave no indication for the formation of hydrophobic domains at the talc-aqueous interface because the polarity parameter decreased with the increase in concentration of both polymers, suggesting instead the formation of hydrophilic domains for the adsorption of EHEC and HM-EHEC on talc. 3.6. AFM analysis

3.5. Fluorescence spectroscopy The microstructure of the polymer adsorbed layer was probed using fluorescence spectroscopy in order to elucidate the nature of HM-EHEC adsorption. Pyrene fluorescence is sensitive to the medium in which it resides [20]. The ratio of the intensities of the first to the third peak (I3 /I1 ) on a pyrene

From AFM imaging [21–24] of polymer adsorbed mica surface (Fig. 18a), HM-EHEC is found to form isolated clusters evenly on the solid. As the concentration was increased from 200 to 1000 ppm, the solid surface was almost fully covered by these clusters (Fig. 18b), however the size of each cluster remained the same with increase in polymer

J. Wang, P. Somasundaran / Journal of Colloid and Interface Science 293 (2006) 322–332

Fig. 16. Pyrene fluorescence for HM-EHEC and talc interface (pH 8.5, no salt).

concentration with the topography of C0 -EHEC adsorbed mica very different (Fig. 18c). C0 -EHEC forms a smooth layer with small uncovered spots uniformly distributed on mica. With the increase of polymer concentration from 200 to 1000 ppm, the number of the uncovered sites decreased with an increase in their size (Fig. 18d). The thickness of adsorbed layer is around 3 nm, which further proved the flat adsorbed conformation of these polymers on solid surface as is suggested in molecular modeling. Therefore, it is proposed that C0 -EHEC adsorbs flat on solid surfaces by hy-

331

Fig. 17. Pyrene fluorescence for EHEC and talc interface (pH 8.5, no salt).

drogen bonding with ethylene oxide side chains with large aggregates forming due to intramolecular and intermolecular hydrogen-bonding interactions [25,26]. But for C14 -EHEC, only small clusters form because of the presence of hydrophobic chains on the polymer.

4. Conclusions (1) Electrokinetic studies of adsorption of EHEC and HMEHEC on talc in the pH range of 2–11 showed that

Fig. 18. AFM images of C14 -EHEC and C0 -EHEC adsorbed on mica at (a) C14 -EHEC, 200 ppm, (b) C14 -EHEC, 1000 ppm, (c) C0 -EHEC, 200 ppm, (d) C0 -EHEC, 1000 ppm.

332

(2)

(3)

(4)

(5)

(6)

(7)

(8)

J. Wang, P. Somasundaran / Journal of Colloid and Interface Science 293 (2006) 322–332

polymer decreased the negative zeta potential of talc but did not reverse it. Adsorption studies also showed that the adsorption of EHEC and HM-EHEC on talc is not affected by changes in pH. These results suggest that the electrostatic force is not the dominant force for the adsorption process. The adsorption of HM-EHEC but not EHEC on talc was affected significantly by the ionic strength level. Hydrophobic force plays a role in adsorption of HM-EHEC on talc. The changes of the infrared bands in the region 1000– 1080 cm−1 , associated with the C–O stretch coupled to the C–C stretch and O–H deformation, were significant and therefore support strong hydrogen bonding of these polysaccharides to the solid surfaces. Pyrene fluorescence gave no indication for hydrophobic domain formation for the adsorption of EHEC and HMEHEC at talc–aqueous interfaces. Urea, a hydrogen bond breaker, reduces significantly the adsorption of HM-EHEC on talc. This result also supports a mechanism involving hydrogen bonding rather than a hydrophobic one. Polysaccharides were also found from AFM imaging to adsorb in a very flat conformation according to the effective area per polymer chain on the substrate surface. From molecular modeling, helical structures were observed. The polymers were found to adsorb flat on solid surface to let more of their OH groups in contact with the surface. Langmuir modeling of adsorption isotherm further suggested hydrogen bonding to be the dominant force for polysaccharide adsorption since the adsorption free energy of these polymers close to that of hydrogen bond formation.

Generally speaking, all the above results suggest the main driving force for EHEC and HM-EHEC adsorption on talc to be hydrogen bonding rather than electrostatic or hydrophobic force. Hydrophobic force plays a synergistic role in HM-EHEC adsorption process. In addition, conformational studies suggest a helical structure of EHEC and HM-EHEC in solution while they are found to adsorb flat on the solid surface.

Acknowledgments The authors acknowledge the support of the National Science Foundation (Grant #CTS-00-89530 dynamic ag-

gregation behavior of surfactant mixture in solutions at solid/liquid interface) and Akzo-Nobel Industries.

References [1] R.J. Piniazkiewicz, E.F. McCarthy, N.A. Genco, Talc. Industrial Minerals and Rocks, sixth ed., SME, Littleton, CO, 1990. [2] K. Thuresson, S. Nilsson, B. Lindman, Langmuir 12 (10) (1996) 2412– 2417. [3] S. Nilsson, K. Thuresson, B. Lindman, B. Nystroem, Macromolecules 33 (26) (2000) 9641–9649. [4] L. Karlson, PCT Int. Appl. (2000) 17. [5] S.R. Balajee, I. Iwasaki, Trans. AIME 244 (1969) 401–406. [6] P.M. Afenya, Int. J. Miner. Process. 9 (1982) 303–319. [7] B.A. Jucker, H. Harms, S.J. Hug, A.J.B. Zehnder, Colloids Surf. B 9 (1997) 331–343. [8] Q. Liu, J.S. Laskowski, Int. J. Miner. Process. 26 (1989) 297–316. [9] R.J. Pugh, Int. J. Miner. Process. 25 (1989) 101–130. [10] H.H. Haung, J.V. Calara, D.L. Bauer, J.D. Miller, in: N.N. Li (Ed.), Recent Developments in Separation Science, CRC Press, Chicago, IL, 1978, pp. 115–133. [11] S.R. Grano, L.F. Griffin, N.W. Johnson, R. St.C. Smart, J. Ralston, in: Proceedings of the Fourth Mill Operator’s Conference, The Australia Institute of Mining and Metallurgy, 1991, pp. 121–128. [12] E. Steenberg, J.P. Harris, S. Afr. J. Chem. 37 (1984) 85. [13] J. Wang, P. Somasundaran, D.R. Nagaraj, Min. Eng. 18 (1) (2005) 77– 81. [14] M. Dubois, K.A. Gilles, J.K. Hamilton, P.A. Rebers, F. Smith, Anal. Chem. 28 (1956) 350–356. [15] G. Rexwinkel, B.B.M. Heesink, W.P.M. van Swaaij, J. Chem. Eng. Data 44 (1999) 1139–1145. [16] D.J. Shaw, Introduction to Colloid and Surface Chemistry, Butterworth–Heinemann, Oxford, 1992. [17] H.-H. Haung, J.V. Calara, D.L. Bauer, J.D. Miller, Recent Dev. Sep. Sci. 4 (1978) 115–133. [18] J.T. Kloprogge, R.L. Frost, L. Rintoul, Phys. Chem. Chem. Phys. (1999) 2559–2564. [19] P.K. Weissenborn, L.J. Warren, J.G. Dunn, Colloids Surf. A 99 (1995) 11–27. [20] J.K. Thomas, The Chemistry of Excitation at Interfaces, Am. Chem. Soc., Washington, DC, 1984. [21] I. Capron, S. Alexandre, G. Müller, Polymer 39 (1998) 5725–5730. [22] T.M. McIntire, R.M. Penner, D.A. Brant, Macromolecules 28 (1995) 6375–6377. [23] A.P. Gunning, A.R. Kirby, M.J. Ridout, G.J. Brownsey, V.J. Morris, Macromolecules 29 (1996) 6791–6796. [24] K.J. Wilkinson, E. Balnois, G.G. Leppard, J. Buffle, Colloids Surf. A 155 (1999) 287–310. [25] A.P. Gunning, V.J. Morris, S. Al-Assaf, G.O. Phillips, Carbohydr. Polym. 30 (1996) 1–8. [26] M.K. Cowman, J. Liu, M. Li, D.M. Hittner, J.S. Kim, in: T.C. Laurent (Ed.), The Chemistry, Biology, and Medical Applications of Hyaluronan and its Derivatives, Portland Press, London, 1998, pp. 17–24.