- Email: [email protected]
Measurement of Interaction Forces between Silica and α-Alumina by Atomic Force Microscopy
JOURNAL OF COLLOID AND INTERFACE SCIENCE ARTICLE NO.
184, 594–600 (1996)
0656
Measurement of Interaction Forces between Silica and a-Alumina by Atomic Force Microscopy S. VEERAMASUNENI, M. R. YALAMANCHILI,1
AND
J. D. MILLER 2
Department of Metallurgical Engineering, 412 William C. Browning Building, College of Mines and Earth Sciences, University of Utah, Salt Lake City, Utah 84112 Received March 26, 1996; accepted July 25, 1996
Interaction forces between a silica sphere and an a-alumina substrate at various pH values were measured by atomic force microscopy (AFM). As expected, at pH values of 10.8 and 10.2 when the surfaces are similarly charged, a repulsive force was observed. On the other hand, at pH values of 5.5, 6.4, and 8.6 when the surfaces are oppositely charged, attractive forces were observed. Experimental force vs separation distance curves were found to be in good agreement with theoretical predictions based on electrostatic and van der Waals interactions. Interestingly, when the force/radius values at a particular separation distance were plotted against pH, the transition from an attractive to a repulsive force occurred at pH 9.3, which is very close to the point of zero charge (pzc) of a-alumina as determined from electrophoresis experiments. These results suggest that AFM force measurements can be used to estimate the pzc of materials. This method may be of particular significance for soluble salt minerals where conventional electrophoretic measurements are not possible at high ionic strengths. Finally, results from transmittance studies further confirmed the interaction between silica and alpha alumina particles in suspensions at various pH values as would be expected based on the results from atomic force microscopy measurements. q 1996 Academic Press, Inc.
Key Words: electrostatic forces; aggregation; dispersion; point of zero charge; atomic force microscopy; silica; alumina.
6). The most widely used method for direct force measurements has been the surface force apparatus (2) in which forces between two macroscopic surfaces (usually mica) are measured. In recent years, atomic force microscopy (AFM) has gained popularity as a technique for force measurements (7). Forces between a sharp tip and a flat surface have been measured both in air (8, 9) and in aqueous media (10, 11, 12). However, these measured forces were not compared to theoretical predictions because of the ill-defined geometry of the tip. Recent advances in force measurements by atomic force microscopy now allow direct force measurements between a sphere and a macroscopic surface (13–15) or a sphere and a sphere (16). In the present work, forces between silica and a-alumina surfaces have been measured by AFM to determine the interactions between these materials at different pH values. The experimental force curves were then compared to the theoretical forces calculated based on electrostatic and van der Waals interactions between charged surfaces. In addition, the results from AFM force measurements were used to determine the point of zero charge of a-alumina. Electrokinetic measurements were conducted for both silica and alumina as a function of pH to further substantiate the results from experimental force measurements.
INTRODUCTION
EXPERIMENTAL
Surface forces play a dominant role in many scientific and technological processes including adsorption, film stability, wetting, emulsions and suspensions, bubble particle interactions, flotation, etc. Fundamental understanding of the nature and origin of surface forces will allow greater control of these technological processes. During the past 25 years, several methods (direct and indirect) have been developed for the measurement of surface and intermolecular forces (1– 1 Present address: MEMC Electronic Materials, Inc., 501 Pearl Dr., P.O. Box 8, St. Peters, MO 63376. 2 To whom correspondence should be addressed.
JCIS 4521
/
6g19$$$221
Sapphire ( a-alumina) single-crystal windows (13 1 1 mm) were purchased from Harrick Scientific Corp. Silica microspheres with a mean diameter of 4.8 mm were purchased in dry condition from Bangs Laboratories, Inc. Tipless cantilever probes (V-shaped) were purchased from Digital Instruments for the AFM experiments. a-Alumina (Alfa Aesar, 99.99%, Ç1 mm) and silica (Sigma, 99.9%, 1–5 mm) in powder form were purchased from Sigma Chemicals for transmittance studies. The chemicals used in the present study include reagent-
594
0021-9797/96 $18.00 Copyright q 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.
AID
Materials
11-20-96 10:44:54
coidas
AP: Colloid
INTERACTION FORCES BETWEEN DISSIMILAR SURFACES MEASURED BY AFM
FIG. 1. SEM photograph of a 4.8-mm hydrophilic silica particle glued to the V-shaped AFM cantilever.
grade potassium chloride (KCl) with purity greater than 99% (Mallinckrodt, Inc.), reagent-grade acetone (Fisher Scientific), reagent-grade methanol (EM Science), reagent-grade sodium hydroxide (NaOH, Mallinckrodt, Inc.), and reagentgrade hydrochloric acid (HCl, EM Science). A Milli-Q water system (Millipore) provided high-purity water (18 MV ) with a surface tension of 72 { 0.2 mN/m at 227C. The pH of the high-purity water stabilized at pH 5.8 { 0.1 after equilibration with the atmosphere. Cleaning Procedures The optical quality sapphire single-crystal windows used in this study were rinsed with 0.1 M KOH/water/methanol/ water/acetone/water followed by drying and plasma cleaning. Plasma cleaning was done in a Tegal plasma chemistry reactor, and the sapphire windows were subjected to oxygen plasma for 30 min on each face to remove any residual organic contaminants. FTIR transmission experiments were conducted to ensure that the sapphire windows were free of any residual organic contaminants. Silica spheres were cleaned by washing with water/methanol/water/acetone/ water followed by drying and plasma cleaning with argon for 40 min. Glassware was soaked in chromic acid, rinsed with large amounts of water, soaked in 7 M KOH, rinsed with water, and dried just prior to use. The liquid cell and the tubing used in AFM experiments were cleaned by rinsing with water/methanol/water/acetone/water followed by vacuum drying.
single spherical particle of silica was mounted on the AFM cantilever tip using a speed bonder and an activator (Loctite Corporation) by means of a micromanipulator and a CCD camera/monitor system. Extreme care was taken to prevent spreading of the glue on the cantilever tip. Optical microscopy was used to ensure that the glue used to mount the silica sphere did not spread on the cantilever. Figure 1 shows a SEM micrograph of a silica sphere (4.8 mm in diameter) mounted on a V-shaped cantilever. Interaction forces were measured in 1 1 10 03 M KCl between the silica microsphere glued to the AFM cantilever and the flat sapphire substrate. The sphere/plate geometry facilitated the application of Derjaguin’s approximation to determine the interaction forces involved (17). All the force measurements were conducted in an aqueous environment (1 1 10 03 M KCl) as a function of pH. The deflection versus Z (sample displacement in z direction) curves were obtained by moving the sapphire sample mounted on a piezo crystal along the z direction toward the silica sphere glued to the cantilever while monitoring the cantilever deflection. The initial nonlinear portion of the deflection vs Z curve represents the changes in attractive or repulsive forces with distance. After contact is established between the sample surface and the sphere, the force curve again becomes linear since the piezo and the sphere glued to the AFM cantilever move together. The actual force versus separation distance plots were obtained by using appropriate values for the spring constant of the cantilever and the radius of sphere glued to the cantilever. The spring constants for the cantilevers used were calibrated from the dimensions of the cantilever and the Young’s modulus (E) of the cantilever material. The cantilever dimensions were determined by scanning electron microscopy. For the cantilevers used in the present work, the spring constants, k values, were calculated from (18) D* Å kÅ
Et 3 12
[1]
6D *b 3 d , L 3 (4d 3 / b 3 )
[2]
where L, b, d, and t are as shown in Fig. 2. Since the
Experimental Procedures Force measurements were conducted using a Nanoscope E atomic force microscope (Digital Instruments, Inc.). A
AID
JCIS 4521
/
6g19$$$222
11-20-96 10:44:54
595
FIG. 2. Characteristic dimensions of the AFM cantilever.
coidas
AP: Colloid
596
VEERAMASUNENI, YALAMANCHILI, AND MILLER
TABLE 1 Spring Constants Calculated for the Cantilevers Used
transmittance measurements. pH was measured again just prior to the transmittance measurements in order to verify whether there was any drift in the pH measured.
Spring constant, k Cantilever No.
Calculated from Eq. [2] (N/m)
Supplied by Digital Instruments (N/m)
1 2 3
0.126 0.101 0.121
0.12 0.12 0.12
cantilevers are made of silicon, a Young’s modulus value of E Å 1.5 1 10 11 N/m 2 was used. There may be some error associated in estimating the k values by this method due to the microscopic measurements of the cantilever dimensions, particularly the thickness, d. Similarly, taking the value of E from a handbook may also introduce some error since E is sensitive to changes in the sample purity, the crystallinity, and the method of sample preparation. In any case, a reasonable agreement between the k values calculated by this method and those supplied by Digital Instruments was obtained as shown in Table 1. It should be mentioned here that excellent reproducibility was observed in these force measurements depending on the consistency achieved in attaching the silica sphere at the same location on the cantilever tip. The practice (adopted by many researchers in the past) of taking the average of several force cycles with the same sphere at the same or a different location on the substrate is believed to introduce large errors in the force measurements due to changes occurring in the physical and chemical properties of the colloidal probe during repeated contacts of the probe with the surface during these force measurements (19). Because of these reasons, force measurements were repeated at least three times with a new silica sphere attached for each experiment at a particular pH. Electrokinetic measurements for a-alumina and silica (Bangs Laboratories) were performed as a function of pH by laser–Doppler electrophoresis using the Zetasizer 3 (Malvern Instruments, London). The samples at the desired level were conditioned for 30 min after pH adjustment and were injected into the cell for electrophoresis measurements. Other details of the instrument and the experimental procedure are provided elsewhere (20). Finally, transmittance measurements were conducted using a Bausch & Lomb Spectronic 20 spectrophotometer with the wavelength set at 580 nm. Carbon black and 1 1 10 03 M KCl solutions were used to calibrate the instrument. Silica and a-alumina colloidal particles (1%) were added simultaneously to 1 1 10 03 M KCl solution in a 50-cc beaker at a particular pH value. The suspension then was conditioned for 10 min and was allowed to settle for 30 min before
AID
JCIS 4521
/
6g19$$$222
11-20-96 10:44:54
RESULTS AND DISCUSSION
Force Measurements The force / radius vs separation curves shown in Fig. 3 were obtained from the raw data by using an AFM analysis program ( 21 ) . To obtain a force / radius vs separation curve, it is necessary to define zeros for both force and separation distance and to convert the diode signal to cantilever deflection and force. The above program allows one to select both the zero of force and the zero of separation. The zero of force was chosen where the deflection was constant ( where the plate and sphere were far apart ) and the zero of separation was chosen when the cantilever deflection became linear with respect to sample displacement at high force. Then the actual force vs separation distance plots were obtained by using appropriate values for the spring constant of the cantilever and the radius of silica sphere glued to the cantilever. Figure 3 presents the force / radius vs separation distance diagrams for the silica / alumina system at different pH values ranging from pH 5.5 to 10.8. It can be seen from this figure that the measured interaction forces vary with pH. The changes in these forces with pH can be
FIG. 3. Force/radius vs separation distance curves obtained at different pH values between a silica sphere and an a-alumina substrate.
coidas
AP: Colloid
INTERACTION FORCES BETWEEN DISSIMILAR SURFACES MEASURED BY AFM
597
FIG. 4. Comparison of experimental force/radius vs separation distance curves with the theoretical values calculated based on electrostatic and van der Waals forces at different pH values. Solid lines represent the theory.
explained by electrostatic and van der Waals forces ( 12 ) . The van der Waals attraction is independent of pH, but the electrostatic force changes with pH as the surface charge of silica and alumina are pH dependent. The pzc of silica is about pH 2.0. Therefore, above pH 2.0 silica is negatively charged. Consequently at the pH values where the experiments were conducted, the nature of the
AID
JCIS 4521
/
6g19$$$222
11-20-96 10:44:54
interaction force depends mainly on the surface charge of the alumina. A repulsive electrostatic force is expected when alumina is negatively charged and an attractive force is expected when alumina is positively charged under these conditions. From Fig. 3, it can be concluded that at the pH values of 5.5, 6.4, and 8.6, alumina is positively charged since attractive forces were observed. Similarly,
coidas
AP: Colloid
598
VEERAMASUNENI, YALAMANCHILI, AND MILLER
TABLE 2 Comparison of the Best Fit Values of the Surface Potential (from AFM Force Curves) with the Measured Zeta Potential Values for a-Alumina
pH
Best fit values, c0 (mV),a calculated from AFM curves
10.8 10.2 8.6 6.4 5.5
034 032 11 24 31
a
Zeta potential (mV) from electrophoresis measurements 039 037 12 26 28
{ { { { {
1.8 1.9 1.8 1.8 1.9
Average of 3 measurements.
at pH values of 10.2 and 10.8 alumina is negatively charged since repulsive forces were observed. The attractive component observed at the pH values of 10.2 and 10.8 is due to van der Waals forces ( 12 ) . In the following section the experimental force measurements are compared to the theoretical forces calculated based on electrostatic and the van der Waals interactions. Theoretical Calculations In the present study, the following equation, based on the linearized Poisson-Boltzmann equation, the Derjaguin approximation, and van der Waals interactions between a charged sphere and a charged plane was used to calculate the theoretical force (F/R) between silica and sapphire surfaces (22): F 4ps1s2 0 kD A Å e 0 . ee0k R 6D 2
[3]
In Eq. [3], s1 and s2 are the charge densities of silica and sapphire, respectively, e is the dielectric constant of the medium, e0 is the permittivity of free space, k 01 is the Debye length, A is the Hamaker constant, and D is the separation distance. This result is based on the assumption that the surface charge remains constant as the two surfaces approach one another. Previous numerical solutions showed that the constant surface charge assumption is better than the other simple alternative of holding the surface potential constant because there are large changes of surface potential as the two double layers interpenetrate (16). Further, the surface charge densities ( s1 and s2 ) were calculated using the Grahame equation (22), s Å 0.117[KCl] 1 / 2 sin h( c0 /51.4),
JCIS 4521
/
6g19$$$222
The Point of Zero Charge (PZC) for a-Alumina The point of zero charge is an important parameter that describes the nature of surfaces in aqueous solutions (24, 25). Basically, the pzc can be used as an index of the propensity of the surface to become either positively or negatively charged in aqueous solutions. There are several conventional methods available for the determination of the pzc (26– 28). However, these methods cannot be used in the case of semisoluble and soluble minerals because of the experimental difficulties involved in dealing with the high ionic strength solutions (20). In such cases, AFM can be a very useful technique for the pzc determination. In this regard, an attempt has been made to examine the use of AFM force measurements (as a function of pH) for estimating the pzc. As mentioned above, the results from force measurements between silica and a-alumina have been used to estimate
[4]
where s is in Coulombs/m 2 and c0 is the surface potential in millivolts. The Hamaker constant (A) for silica and sapphire interacting across water was calculated to be 1.41 1 10 020 J on the basis of the Lifshitz theory (22). The Debye length,
AID
k 01 , was calculated to be 9.6 nm. Figure 4 compares the theoretically fitted and experimentally measured F/R values as a function of the separation distance at different pH values. Equation [3] was fitted using a least-squares procedure (solver) available in Excel spreadsheet software. During the above fitting procedure the surface charge density of silica, calculated from the measured zeta potential values, was used. Then from the best fit value for surface charge density of a-alumina in Eq. [3], c0 was back calculated by using Eq. [4] and these results were compared to the measured zeta potential values as shown Table 2. It can be noted from Table 2 that there is a good agreement between the fitted and the experimentally measured potential values, indicating that the experimental force measurements by AFM are reasonably accurate. In these calculations, surface potential was approximated by the measured zeta potential because of the low ionic strengths involved (23).
11-20-96 10:44:54
FIG. 5. The point of zero charge of a-alumina as determined from AFM force measurements and electrokinetic measurements.
coidas
AP: Colloid
INTERACTION FORCES BETWEEN DISSIMILAR SURFACES MEASURED BY AFM
the point of zero charge for a-alumina. The force/radius values at a separation distance of 5 nm (obtained from Fig. 3) were plotted against the pH values as shown in Fig. 5. It can be noted from Fig. 5 that attractive forces between silica and a-alumina in the acidic range changed to repulsive forces at about a pH of 9.3. This pH of 9.3 should yield an estimate of the point of zero charge of a-alumina. Because of the existence of both van der Waals and repulsive electrostatic forces (between the neutral a-alumina surface and the charged silica surface) at the point of zero charge there is a small difference between the pzc estimated from force measurements and the pzc values determined from electrokinetic measurements for a-alumina. Figure 5 also shows the measured zeta potential of aalumina as a function of pH. The results indicate that the pzc is pH 9.1. The pzc for a-alumina as determined from electrokinetic measurements in this work is in good agreement with the pzc values reported in the literature (24). It should be noted that the pzc for a-alumina determined from the electrokinetic measurements is very close to the pH at which attractive forces change to repulsive forces as shown in Fig. 5. It has been pointed out that the pzc of a-alumina is particularly sensitive to adsorption of silica from solution (29). However, the good agreement between the literature values and experimental values in the present study suggests that there is no substantial adsorption of silica from solution. Further, it is evident from the data presented in Fig. 5 that AFM force measurements can be used to determine the pzc of a mineral surface. Similar AFM experiments were conducted with a silica sphere and a colemanite substrate and the pzc for colemanite, a boron mineral, was found to be pH 10.1, which agrees well with previously reported values (30). It should be mentioned that AFM will be a useful method for the determination of the pzc for soluble salt minerals where conventional methods are not possible to use because of the high ionic strengths involved. To further substantiate the results obtained from the AFM force measurements, the stability of silica and a-alumina suspensions was examined as a function of pH by transmittance studies. Figure 6 shows the results of transmittance studies carried out for colloidal size silica and a-alumina suspensions in 1 1 10 03 M KCl solution at various pH values. As expected, large transmittance values ( Ç80%) were obtained at low pH values where silica and a-alumina particles are oppositely charged and the suspensions are unstable. As the pH approaches the pzc, transmittance decreases and at pH values greater than the pzc where the particles are similarly charged almost zero transmittance was observed because under these conditions the suspensions are stabilized. These results further confirm the results from AFM force measurements, which explain the aggregation/ dispersion behavior of silica and a-alumina suspensions as a function of pH.
AID
JCIS 4521
/
6g19$$$222
11-20-96 10:44:54
599
FIG. 6. Percentage transmittance as a function of pH for silica and aalumina suspensions in 1 1 10 03 M KCl.
CONCLUSIONS
Interaction forces have been measured between a 4.8-mm silica sphere and an a-alumina crystal by AFM in 1 1 10 03 M KCl at various pH values. Repulsive forces were observed when the surfaces are similarly charged and, on the other hand, attractive forces were observed when the surfaces are oppositely charged. Experimental data obtained were found to be in reasonable agreement with theoretically calculated interaction forces based on electrostatic and van der Waals interactions. The force/radius values at a separation distance of 5 nm when plotted against pH yield a reasonable estimate for the point of zero charge of a-alumina. Finally, the transmittance results further confirm the aggregation/dispersion behavior of silica and a-alumina expected according to the interaction forces measured by AFM. ACKNOWLEDGMENT This work was supported by the U.S. Department of Energy, Grant DEFG03-94ER14315.
REFERENCES 1. Derjaguin, B. V., Rabinovich, Ya. I., and Churaev, N. V., Nature 272, 313 (1978). 2. Israelachvili, J. N., and Adams, G. E., J. Chem. Soc. Faraday Trans. 1 74, 975 (1978). 3. Israelachvili, J. N., Adv. Colloid Interface Sci. 16, 31 (1982). 4. Prieve, D. C., and Frej, N. A., Langmuir 6, 396 (1990). 5. van de Ven, T. G. M., Warszynski, P., Wu, X., and Dabros, T., Langmuir 10, 3046 (1994). 6. Ninham, B. W., Adv. Colloid Interface Sci. 16, 3 (1982). 7. Binning, G., Quate, C. F., and Gerber, Ch., Phys. Rev. Lett. 56, 930 (1986). 8. Martin, Y., Williams, C., and Wickramasinghe, H., J. Appl. Phys. 61, 4723 (1987).
coidas
AP: Colloid
600
VEERAMASUNENI, YALAMANCHILI, AND MILLER
9. Burnham, N. N., Dominguez, D. D., Mowery, R. L., and Colten, R. J., Phys. Rev. Lett. 64, 1931 (1990). 10. Weisenhorn, A. L., Hansma, P. K., Albrecht, T. R., and Quate, C. F., Appl. Phys. Lett. 54, 2651 (1989). 11. Ducker, W. A., and Cook, R. F., Appl. Phys. Lett. 56, 2408 (1990). 12. Butt, H.-J., Biophys. J. 60, 1438 (1991). 13. Ducker, W. A., Senden, T. J., and Pashley, R. M., Nature 353, 239 (1991). 14. Butt, H.-J., Biophys. J. 60, 777 (1991). 15. Meager, L., J. Colloid Interface Sci. 152, 293 (1992). 16. Li, Y. Q., Tao, N. J., Pan, J., Garcia, A. A., and Lindsay, S. M., Langmuir 9, 637 (1993). 17. Derjaguin, B. V., Kolloid Z. 69, 155 (1934). 18. Sader, J. E., and White, L., J. Appl. Phys. 74, 1 (1993). 19. Hlady, V., Bioengineering Department, University of Utah, Salt Lake City, UT, 1996, private communication. 20. Miller, J. D., Yalamanchili, M. R., and Kellar, J. J., Langmuir 8, 1464 (1992).
AID
JCIS 4521
/
6g19$$$222
11-20-96 10:44:54
21. Chan, D. Y., University of Melbourne, Victoria, Australia. 22. Israelachvili, J. N., ‘‘Intermolecular and Surface Force,’’ 2nd Ed. Academic Press, New York, 1992. 23. Fuerstenau, M. C., Miller, J. D., and Kuhn, M. C., ‘‘Chemistry of Flotation.’’ SME, New York, 1985. 24. Yopps, J. A., and Fuerstenau, D. W., J. Colloid Sci. 19, 61 (1964). 25. Somasundaran, P., and Agar, G. E., J. Colloid Interface Sci. 24, 433 (1967). 26. Honig, E. P., and Hengst, J. H. TH., J. Colloid Interface Sci. 29, 510 (1969). 27. Noh, S. J., and Schwarz, A. J., J. Colloid Interface Sci. 130, 157 (1989). 28. Lin, X.-Y., Farhi, E., Arribart, H., ‘‘Advances in Particle Adhesion.’’ OPA, Amsterdam, 1996. 29. Horn, R. G., Clarke, D. R., and Clarkson, M. T., J. Mater. Res. 3(3), 413 (1988). 30. Hancer, M., and Celik, M. S., Separation Sci. Technol. 28, 1703 (1993).
coidas
AP: Colloid
184, 594–600 (1996)
0656
Measurement of Interaction Forces between Silica and a-Alumina by Atomic Force Microscopy S. VEERAMASUNENI, M. R. YALAMANCHILI,1
AND
J. D. MILLER 2
Department of Metallurgical Engineering, 412 William C. Browning Building, College of Mines and Earth Sciences, University of Utah, Salt Lake City, Utah 84112 Received March 26, 1996; accepted July 25, 1996
Interaction forces between a silica sphere and an a-alumina substrate at various pH values were measured by atomic force microscopy (AFM). As expected, at pH values of 10.8 and 10.2 when the surfaces are similarly charged, a repulsive force was observed. On the other hand, at pH values of 5.5, 6.4, and 8.6 when the surfaces are oppositely charged, attractive forces were observed. Experimental force vs separation distance curves were found to be in good agreement with theoretical predictions based on electrostatic and van der Waals interactions. Interestingly, when the force/radius values at a particular separation distance were plotted against pH, the transition from an attractive to a repulsive force occurred at pH 9.3, which is very close to the point of zero charge (pzc) of a-alumina as determined from electrophoresis experiments. These results suggest that AFM force measurements can be used to estimate the pzc of materials. This method may be of particular significance for soluble salt minerals where conventional electrophoretic measurements are not possible at high ionic strengths. Finally, results from transmittance studies further confirmed the interaction between silica and alpha alumina particles in suspensions at various pH values as would be expected based on the results from atomic force microscopy measurements. q 1996 Academic Press, Inc.
Key Words: electrostatic forces; aggregation; dispersion; point of zero charge; atomic force microscopy; silica; alumina.
6). The most widely used method for direct force measurements has been the surface force apparatus (2) in which forces between two macroscopic surfaces (usually mica) are measured. In recent years, atomic force microscopy (AFM) has gained popularity as a technique for force measurements (7). Forces between a sharp tip and a flat surface have been measured both in air (8, 9) and in aqueous media (10, 11, 12). However, these measured forces were not compared to theoretical predictions because of the ill-defined geometry of the tip. Recent advances in force measurements by atomic force microscopy now allow direct force measurements between a sphere and a macroscopic surface (13–15) or a sphere and a sphere (16). In the present work, forces between silica and a-alumina surfaces have been measured by AFM to determine the interactions between these materials at different pH values. The experimental force curves were then compared to the theoretical forces calculated based on electrostatic and van der Waals interactions between charged surfaces. In addition, the results from AFM force measurements were used to determine the point of zero charge of a-alumina. Electrokinetic measurements were conducted for both silica and alumina as a function of pH to further substantiate the results from experimental force measurements.
INTRODUCTION
EXPERIMENTAL
Surface forces play a dominant role in many scientific and technological processes including adsorption, film stability, wetting, emulsions and suspensions, bubble particle interactions, flotation, etc. Fundamental understanding of the nature and origin of surface forces will allow greater control of these technological processes. During the past 25 years, several methods (direct and indirect) have been developed for the measurement of surface and intermolecular forces (1– 1 Present address: MEMC Electronic Materials, Inc., 501 Pearl Dr., P.O. Box 8, St. Peters, MO 63376. 2 To whom correspondence should be addressed.
JCIS 4521
/
6g19$$$221
Sapphire ( a-alumina) single-crystal windows (13 1 1 mm) were purchased from Harrick Scientific Corp. Silica microspheres with a mean diameter of 4.8 mm were purchased in dry condition from Bangs Laboratories, Inc. Tipless cantilever probes (V-shaped) were purchased from Digital Instruments for the AFM experiments. a-Alumina (Alfa Aesar, 99.99%, Ç1 mm) and silica (Sigma, 99.9%, 1–5 mm) in powder form were purchased from Sigma Chemicals for transmittance studies. The chemicals used in the present study include reagent-
594
0021-9797/96 $18.00 Copyright q 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.
AID
Materials
11-20-96 10:44:54
coidas
AP: Colloid
INTERACTION FORCES BETWEEN DISSIMILAR SURFACES MEASURED BY AFM
FIG. 1. SEM photograph of a 4.8-mm hydrophilic silica particle glued to the V-shaped AFM cantilever.
grade potassium chloride (KCl) with purity greater than 99% (Mallinckrodt, Inc.), reagent-grade acetone (Fisher Scientific), reagent-grade methanol (EM Science), reagent-grade sodium hydroxide (NaOH, Mallinckrodt, Inc.), and reagentgrade hydrochloric acid (HCl, EM Science). A Milli-Q water system (Millipore) provided high-purity water (18 MV ) with a surface tension of 72 { 0.2 mN/m at 227C. The pH of the high-purity water stabilized at pH 5.8 { 0.1 after equilibration with the atmosphere. Cleaning Procedures The optical quality sapphire single-crystal windows used in this study were rinsed with 0.1 M KOH/water/methanol/ water/acetone/water followed by drying and plasma cleaning. Plasma cleaning was done in a Tegal plasma chemistry reactor, and the sapphire windows were subjected to oxygen plasma for 30 min on each face to remove any residual organic contaminants. FTIR transmission experiments were conducted to ensure that the sapphire windows were free of any residual organic contaminants. Silica spheres were cleaned by washing with water/methanol/water/acetone/ water followed by drying and plasma cleaning with argon for 40 min. Glassware was soaked in chromic acid, rinsed with large amounts of water, soaked in 7 M KOH, rinsed with water, and dried just prior to use. The liquid cell and the tubing used in AFM experiments were cleaned by rinsing with water/methanol/water/acetone/water followed by vacuum drying.
single spherical particle of silica was mounted on the AFM cantilever tip using a speed bonder and an activator (Loctite Corporation) by means of a micromanipulator and a CCD camera/monitor system. Extreme care was taken to prevent spreading of the glue on the cantilever tip. Optical microscopy was used to ensure that the glue used to mount the silica sphere did not spread on the cantilever. Figure 1 shows a SEM micrograph of a silica sphere (4.8 mm in diameter) mounted on a V-shaped cantilever. Interaction forces were measured in 1 1 10 03 M KCl between the silica microsphere glued to the AFM cantilever and the flat sapphire substrate. The sphere/plate geometry facilitated the application of Derjaguin’s approximation to determine the interaction forces involved (17). All the force measurements were conducted in an aqueous environment (1 1 10 03 M KCl) as a function of pH. The deflection versus Z (sample displacement in z direction) curves were obtained by moving the sapphire sample mounted on a piezo crystal along the z direction toward the silica sphere glued to the cantilever while monitoring the cantilever deflection. The initial nonlinear portion of the deflection vs Z curve represents the changes in attractive or repulsive forces with distance. After contact is established between the sample surface and the sphere, the force curve again becomes linear since the piezo and the sphere glued to the AFM cantilever move together. The actual force versus separation distance plots were obtained by using appropriate values for the spring constant of the cantilever and the radius of sphere glued to the cantilever. The spring constants for the cantilevers used were calibrated from the dimensions of the cantilever and the Young’s modulus (E) of the cantilever material. The cantilever dimensions were determined by scanning electron microscopy. For the cantilevers used in the present work, the spring constants, k values, were calculated from (18) D* Å kÅ
Et 3 12
[1]
6D *b 3 d , L 3 (4d 3 / b 3 )
[2]
where L, b, d, and t are as shown in Fig. 2. Since the
Experimental Procedures Force measurements were conducted using a Nanoscope E atomic force microscope (Digital Instruments, Inc.). A
AID
JCIS 4521
/
6g19$$$222
11-20-96 10:44:54
595
FIG. 2. Characteristic dimensions of the AFM cantilever.
coidas
AP: Colloid
596
VEERAMASUNENI, YALAMANCHILI, AND MILLER
TABLE 1 Spring Constants Calculated for the Cantilevers Used
transmittance measurements. pH was measured again just prior to the transmittance measurements in order to verify whether there was any drift in the pH measured.
Spring constant, k Cantilever No.
Calculated from Eq. [2] (N/m)
Supplied by Digital Instruments (N/m)
1 2 3
0.126 0.101 0.121
0.12 0.12 0.12
cantilevers are made of silicon, a Young’s modulus value of E Å 1.5 1 10 11 N/m 2 was used. There may be some error associated in estimating the k values by this method due to the microscopic measurements of the cantilever dimensions, particularly the thickness, d. Similarly, taking the value of E from a handbook may also introduce some error since E is sensitive to changes in the sample purity, the crystallinity, and the method of sample preparation. In any case, a reasonable agreement between the k values calculated by this method and those supplied by Digital Instruments was obtained as shown in Table 1. It should be mentioned here that excellent reproducibility was observed in these force measurements depending on the consistency achieved in attaching the silica sphere at the same location on the cantilever tip. The practice (adopted by many researchers in the past) of taking the average of several force cycles with the same sphere at the same or a different location on the substrate is believed to introduce large errors in the force measurements due to changes occurring in the physical and chemical properties of the colloidal probe during repeated contacts of the probe with the surface during these force measurements (19). Because of these reasons, force measurements were repeated at least three times with a new silica sphere attached for each experiment at a particular pH. Electrokinetic measurements for a-alumina and silica (Bangs Laboratories) were performed as a function of pH by laser–Doppler electrophoresis using the Zetasizer 3 (Malvern Instruments, London). The samples at the desired level were conditioned for 30 min after pH adjustment and were injected into the cell for electrophoresis measurements. Other details of the instrument and the experimental procedure are provided elsewhere (20). Finally, transmittance measurements were conducted using a Bausch & Lomb Spectronic 20 spectrophotometer with the wavelength set at 580 nm. Carbon black and 1 1 10 03 M KCl solutions were used to calibrate the instrument. Silica and a-alumina colloidal particles (1%) were added simultaneously to 1 1 10 03 M KCl solution in a 50-cc beaker at a particular pH value. The suspension then was conditioned for 10 min and was allowed to settle for 30 min before
AID
JCIS 4521
/
6g19$$$222
11-20-96 10:44:54
RESULTS AND DISCUSSION
Force Measurements The force / radius vs separation curves shown in Fig. 3 were obtained from the raw data by using an AFM analysis program ( 21 ) . To obtain a force / radius vs separation curve, it is necessary to define zeros for both force and separation distance and to convert the diode signal to cantilever deflection and force. The above program allows one to select both the zero of force and the zero of separation. The zero of force was chosen where the deflection was constant ( where the plate and sphere were far apart ) and the zero of separation was chosen when the cantilever deflection became linear with respect to sample displacement at high force. Then the actual force vs separation distance plots were obtained by using appropriate values for the spring constant of the cantilever and the radius of silica sphere glued to the cantilever. Figure 3 presents the force / radius vs separation distance diagrams for the silica / alumina system at different pH values ranging from pH 5.5 to 10.8. It can be seen from this figure that the measured interaction forces vary with pH. The changes in these forces with pH can be
FIG. 3. Force/radius vs separation distance curves obtained at different pH values between a silica sphere and an a-alumina substrate.
coidas
AP: Colloid
INTERACTION FORCES BETWEEN DISSIMILAR SURFACES MEASURED BY AFM
597
FIG. 4. Comparison of experimental force/radius vs separation distance curves with the theoretical values calculated based on electrostatic and van der Waals forces at different pH values. Solid lines represent the theory.
explained by electrostatic and van der Waals forces ( 12 ) . The van der Waals attraction is independent of pH, but the electrostatic force changes with pH as the surface charge of silica and alumina are pH dependent. The pzc of silica is about pH 2.0. Therefore, above pH 2.0 silica is negatively charged. Consequently at the pH values where the experiments were conducted, the nature of the
AID
JCIS 4521
/
6g19$$$222
11-20-96 10:44:54
interaction force depends mainly on the surface charge of the alumina. A repulsive electrostatic force is expected when alumina is negatively charged and an attractive force is expected when alumina is positively charged under these conditions. From Fig. 3, it can be concluded that at the pH values of 5.5, 6.4, and 8.6, alumina is positively charged since attractive forces were observed. Similarly,
coidas
AP: Colloid
598
VEERAMASUNENI, YALAMANCHILI, AND MILLER
TABLE 2 Comparison of the Best Fit Values of the Surface Potential (from AFM Force Curves) with the Measured Zeta Potential Values for a-Alumina
pH
Best fit values, c0 (mV),a calculated from AFM curves
10.8 10.2 8.6 6.4 5.5
034 032 11 24 31
a
Zeta potential (mV) from electrophoresis measurements 039 037 12 26 28
{ { { { {
1.8 1.9 1.8 1.8 1.9
Average of 3 measurements.
at pH values of 10.2 and 10.8 alumina is negatively charged since repulsive forces were observed. The attractive component observed at the pH values of 10.2 and 10.8 is due to van der Waals forces ( 12 ) . In the following section the experimental force measurements are compared to the theoretical forces calculated based on electrostatic and the van der Waals interactions. Theoretical Calculations In the present study, the following equation, based on the linearized Poisson-Boltzmann equation, the Derjaguin approximation, and van der Waals interactions between a charged sphere and a charged plane was used to calculate the theoretical force (F/R) between silica and sapphire surfaces (22): F 4ps1s2 0 kD A Å e 0 . ee0k R 6D 2
[3]
In Eq. [3], s1 and s2 are the charge densities of silica and sapphire, respectively, e is the dielectric constant of the medium, e0 is the permittivity of free space, k 01 is the Debye length, A is the Hamaker constant, and D is the separation distance. This result is based on the assumption that the surface charge remains constant as the two surfaces approach one another. Previous numerical solutions showed that the constant surface charge assumption is better than the other simple alternative of holding the surface potential constant because there are large changes of surface potential as the two double layers interpenetrate (16). Further, the surface charge densities ( s1 and s2 ) were calculated using the Grahame equation (22), s Å 0.117[KCl] 1 / 2 sin h( c0 /51.4),
JCIS 4521
/
6g19$$$222
The Point of Zero Charge (PZC) for a-Alumina The point of zero charge is an important parameter that describes the nature of surfaces in aqueous solutions (24, 25). Basically, the pzc can be used as an index of the propensity of the surface to become either positively or negatively charged in aqueous solutions. There are several conventional methods available for the determination of the pzc (26– 28). However, these methods cannot be used in the case of semisoluble and soluble minerals because of the experimental difficulties involved in dealing with the high ionic strength solutions (20). In such cases, AFM can be a very useful technique for the pzc determination. In this regard, an attempt has been made to examine the use of AFM force measurements (as a function of pH) for estimating the pzc. As mentioned above, the results from force measurements between silica and a-alumina have been used to estimate
[4]
where s is in Coulombs/m 2 and c0 is the surface potential in millivolts. The Hamaker constant (A) for silica and sapphire interacting across water was calculated to be 1.41 1 10 020 J on the basis of the Lifshitz theory (22). The Debye length,
AID
k 01 , was calculated to be 9.6 nm. Figure 4 compares the theoretically fitted and experimentally measured F/R values as a function of the separation distance at different pH values. Equation [3] was fitted using a least-squares procedure (solver) available in Excel spreadsheet software. During the above fitting procedure the surface charge density of silica, calculated from the measured zeta potential values, was used. Then from the best fit value for surface charge density of a-alumina in Eq. [3], c0 was back calculated by using Eq. [4] and these results were compared to the measured zeta potential values as shown Table 2. It can be noted from Table 2 that there is a good agreement between the fitted and the experimentally measured potential values, indicating that the experimental force measurements by AFM are reasonably accurate. In these calculations, surface potential was approximated by the measured zeta potential because of the low ionic strengths involved (23).
11-20-96 10:44:54
FIG. 5. The point of zero charge of a-alumina as determined from AFM force measurements and electrokinetic measurements.
coidas
AP: Colloid
INTERACTION FORCES BETWEEN DISSIMILAR SURFACES MEASURED BY AFM
the point of zero charge for a-alumina. The force/radius values at a separation distance of 5 nm (obtained from Fig. 3) were plotted against the pH values as shown in Fig. 5. It can be noted from Fig. 5 that attractive forces between silica and a-alumina in the acidic range changed to repulsive forces at about a pH of 9.3. This pH of 9.3 should yield an estimate of the point of zero charge of a-alumina. Because of the existence of both van der Waals and repulsive electrostatic forces (between the neutral a-alumina surface and the charged silica surface) at the point of zero charge there is a small difference between the pzc estimated from force measurements and the pzc values determined from electrokinetic measurements for a-alumina. Figure 5 also shows the measured zeta potential of aalumina as a function of pH. The results indicate that the pzc is pH 9.1. The pzc for a-alumina as determined from electrokinetic measurements in this work is in good agreement with the pzc values reported in the literature (24). It should be noted that the pzc for a-alumina determined from the electrokinetic measurements is very close to the pH at which attractive forces change to repulsive forces as shown in Fig. 5. It has been pointed out that the pzc of a-alumina is particularly sensitive to adsorption of silica from solution (29). However, the good agreement between the literature values and experimental values in the present study suggests that there is no substantial adsorption of silica from solution. Further, it is evident from the data presented in Fig. 5 that AFM force measurements can be used to determine the pzc of a mineral surface. Similar AFM experiments were conducted with a silica sphere and a colemanite substrate and the pzc for colemanite, a boron mineral, was found to be pH 10.1, which agrees well with previously reported values (30). It should be mentioned that AFM will be a useful method for the determination of the pzc for soluble salt minerals where conventional methods are not possible to use because of the high ionic strengths involved. To further substantiate the results obtained from the AFM force measurements, the stability of silica and a-alumina suspensions was examined as a function of pH by transmittance studies. Figure 6 shows the results of transmittance studies carried out for colloidal size silica and a-alumina suspensions in 1 1 10 03 M KCl solution at various pH values. As expected, large transmittance values ( Ç80%) were obtained at low pH values where silica and a-alumina particles are oppositely charged and the suspensions are unstable. As the pH approaches the pzc, transmittance decreases and at pH values greater than the pzc where the particles are similarly charged almost zero transmittance was observed because under these conditions the suspensions are stabilized. These results further confirm the results from AFM force measurements, which explain the aggregation/ dispersion behavior of silica and a-alumina suspensions as a function of pH.
AID
JCIS 4521
/
6g19$$$222
11-20-96 10:44:54
599
FIG. 6. Percentage transmittance as a function of pH for silica and aalumina suspensions in 1 1 10 03 M KCl.
CONCLUSIONS
Interaction forces have been measured between a 4.8-mm silica sphere and an a-alumina crystal by AFM in 1 1 10 03 M KCl at various pH values. Repulsive forces were observed when the surfaces are similarly charged and, on the other hand, attractive forces were observed when the surfaces are oppositely charged. Experimental data obtained were found to be in reasonable agreement with theoretically calculated interaction forces based on electrostatic and van der Waals interactions. The force/radius values at a separation distance of 5 nm when plotted against pH yield a reasonable estimate for the point of zero charge of a-alumina. Finally, the transmittance results further confirm the aggregation/dispersion behavior of silica and a-alumina expected according to the interaction forces measured by AFM. ACKNOWLEDGMENT This work was supported by the U.S. Department of Energy, Grant DEFG03-94ER14315.
REFERENCES 1. Derjaguin, B. V., Rabinovich, Ya. I., and Churaev, N. V., Nature 272, 313 (1978). 2. Israelachvili, J. N., and Adams, G. E., J. Chem. Soc. Faraday Trans. 1 74, 975 (1978). 3. Israelachvili, J. N., Adv. Colloid Interface Sci. 16, 31 (1982). 4. Prieve, D. C., and Frej, N. A., Langmuir 6, 396 (1990). 5. van de Ven, T. G. M., Warszynski, P., Wu, X., and Dabros, T., Langmuir 10, 3046 (1994). 6. Ninham, B. W., Adv. Colloid Interface Sci. 16, 3 (1982). 7. Binning, G., Quate, C. F., and Gerber, Ch., Phys. Rev. Lett. 56, 930 (1986). 8. Martin, Y., Williams, C., and Wickramasinghe, H., J. Appl. Phys. 61, 4723 (1987).
coidas
AP: Colloid
600
VEERAMASUNENI, YALAMANCHILI, AND MILLER
9. Burnham, N. N., Dominguez, D. D., Mowery, R. L., and Colten, R. J., Phys. Rev. Lett. 64, 1931 (1990). 10. Weisenhorn, A. L., Hansma, P. K., Albrecht, T. R., and Quate, C. F., Appl. Phys. Lett. 54, 2651 (1989). 11. Ducker, W. A., and Cook, R. F., Appl. Phys. Lett. 56, 2408 (1990). 12. Butt, H.-J., Biophys. J. 60, 1438 (1991). 13. Ducker, W. A., Senden, T. J., and Pashley, R. M., Nature 353, 239 (1991). 14. Butt, H.-J., Biophys. J. 60, 777 (1991). 15. Meager, L., J. Colloid Interface Sci. 152, 293 (1992). 16. Li, Y. Q., Tao, N. J., Pan, J., Garcia, A. A., and Lindsay, S. M., Langmuir 9, 637 (1993). 17. Derjaguin, B. V., Kolloid Z. 69, 155 (1934). 18. Sader, J. E., and White, L., J. Appl. Phys. 74, 1 (1993). 19. Hlady, V., Bioengineering Department, University of Utah, Salt Lake City, UT, 1996, private communication. 20. Miller, J. D., Yalamanchili, M. R., and Kellar, J. J., Langmuir 8, 1464 (1992).
AID
JCIS 4521
/
6g19$$$222
11-20-96 10:44:54
21. Chan, D. Y., University of Melbourne, Victoria, Australia. 22. Israelachvili, J. N., ‘‘Intermolecular and Surface Force,’’ 2nd Ed. Academic Press, New York, 1992. 23. Fuerstenau, M. C., Miller, J. D., and Kuhn, M. C., ‘‘Chemistry of Flotation.’’ SME, New York, 1985. 24. Yopps, J. A., and Fuerstenau, D. W., J. Colloid Sci. 19, 61 (1964). 25. Somasundaran, P., and Agar, G. E., J. Colloid Interface Sci. 24, 433 (1967). 26. Honig, E. P., and Hengst, J. H. TH., J. Colloid Interface Sci. 29, 510 (1969). 27. Noh, S. J., and Schwarz, A. J., J. Colloid Interface Sci. 130, 157 (1989). 28. Lin, X.-Y., Farhi, E., Arribart, H., ‘‘Advances in Particle Adhesion.’’ OPA, Amsterdam, 1996. 29. Horn, R. G., Clarke, D. R., and Clarkson, M. T., J. Mater. Res. 3(3), 413 (1988). 30. Hancer, M., and Celik, M. S., Separation Sci. Technol. 28, 1703 (1993).
coidas
AP: Colloid