Langmuir 1997, 13, 4939-4942
4939
Surface Chemical Reactions Probed with Scanning Force Microscopy Michel P. L. Werts, Eric W. van der Vegte, and Georges Hadziioannou* Department of Polymer Chemistry and Material Science Centre, University of Groningen, Nijenborgh 4, NL-9747 AG, The Netherlands Received April 9, 1997. In Final Form: July 18, 1997X In this letter we report the study of surface chemical reactions with scanning force microscopy (SFM) with chemical specificity. Using chemically modified SFM probes, we can determine the local surface reaction conversion during a chemical surface modification. The adhesion forces between a functionalized tip and a chemically modified surface were used to calculate local reaction conversions on a molecular level on model substrates. Moreover, the distribution of reacted (and unreacted) functional groups could be mapped by chemically specific lateral force imaging. This technique was applied to study surface modifications of polyethylene films upon etching with chromic acid.
Introduction Surface modifications are often used to improve the biocompatibility, adhesion, wetting, or frictional properties of materials.1 The change in the chemical nature of the surfaces can only be determined by macroscopic methods like contact angle, FT-IR, or XPS measurements. However, with scanning force microscopy (SFM) with chemical specificity,2 local chemical differences on a microscopic level can now be imaged directly and the functional group distribution can be determined. Well-defined self-assembled monolayers (SAMs) of alkanethiols3 with alcohol and acid end groups were used as a model system for surface modifications. The surface reaction conversions of the above-mentioned groups with butyl isocyanate were determined on a microscopic scale with SFM, and the reaction was mapped by friction force imaging. Many studies have addressed the chemical etching process of polymer surfaces, but all have made measurements on macroscopic scales (contact angle, XPS, FT-IR). To demonstrate the power of SFM with chemical specificity, we proved that we can monitor directly the chemical etching of a polymer surface in time on a microscopic level. pH-dependent adhesion force measurements on the etched surfaces identify the introduction of acidic functionalities. Experimental Section Materials. Dodecanethiol and n-heptane p.a. from Janssen Chimica and toluene p.a., ethanol p.a., and triethylamine p.a. X Abstract published in Advance ACS Abstracts, September 1, 1997.
(1) For example: (a) Holmes-Farley, S. R.; Whitesides, G. M. Langmuir 1987, 3, 62. (b) Hakiki, A.; Herz, J. E.; Beinert, G. Polymer 1992, 33, 4575. (c) Wilson, M. D.; Ferguson, G. S.; Whitesides, G. M. J. Am. Chem. Soc. 1990, 112, 1244. (d) Wu, Z.; Pittman, C. U., Jr.; Gardner, S. D. Carbon 1996, 34, 59. (e) Holmes-Farley, S. R.; Whitesides, G. M. Langmuir 1986, 2, 266. (2) (a) Frisbie, C. D.; Rozsnyai, L. F.; Noy, A.; Wrighton, M. S.; Lieber, C. M. Science 1994, 265, 2071. (b) Noy, A.; Frisbie, C. D.; Rozsnyai, L. F.; Wrighton, M. S.; Lieber, C. M. J. Am. Chem. Soc. 1995, 117, 7943. (c) Green, J.-B. D.; McDermott, M. T.; Porter, M. D.; Siperko, L. M. J. Phys. Chem. 1995, 99, 10960. (d) Thomas, R. C.; Houston, J. E.; Crooks, R. M.; Kim, T.; Michalske, T. A. J. Am. Chem. Soc. 1995, 117, 3830. (e) Sinniah, S. K.; Steel, A. B.; Miller, J. C.; Reutt-Robey, J. E. J. Am. Chem. Soc. 1996, 118, 8925. (f) Han, T.; Wiliams, J. M.; Beebe, T. P. Anal. Chim. Acta 1995, 307, 361. (g) van der Vegte, E. W.; Hadziioannou, G. Langmuir 1997, 13, 4357. (h) van der Vegte, E. W.; Hadziioannou, G. Submitted to J. Phys. Chem. (i) Vezenov, D. I.; Noy, A.; Rozsnyai, L. F.; Lieber, C. M. J. Am. Chem. Soc. 1997, 119, 2006. (3) For example: (a) Nuzzo, R. G.; Allara, D. L. J. Am. Chem. Soc. 1983, 105, 4481. (b) Porter, M. D.; Bright, T. B.; Allara, D. L.; Chidsey, C. E. D. J. Am. Chem. Soc. 1987, 109, 3559. (c) Bain, C. D.; Troughton, E. B.; Tao, Y.-T.; Evall, J.; Whitesides, G. M.; Nuzzo, R. G. J. Am. Chem. Soc. 1989, 11, 321. (d) Nuzzo, R. G.; Dubois, L. H.; Allara, D. L. J. Am. Chem. Soc. 1990, 112, 558. (e) Laibinis, P. E.; Fox, M. A.; Folkers, J. P.; Whitesides, G. M. Langmuir 1991, 7, 3167. (f) Delamarche, E.; Michel, B.; Kang, H.; Gerber, Ch. Langmuir 1994, 10, 4103.
S0743-7463(97)00364-8 CCC: $14.00
from Merck were used as received. Water used in experiments was deionized (18 MΩ.cm resistivity) with a Millipore Milli-Q filtration system. Gold-coated Si-cantilevers with an integrated Si3N4-tip (Topometrix GmbH) were calibrated, using a nondestructive method described by Hutter and Bechhoeffer,4 which is implemented in the data-acquisition software of Topometrix. The spring constant was determined to be 0.05 N/m. Thiols were available from previous studies.2g Butyl isocyanate was distilled at 67 °C and 200 mbar, prior to use. Polyethylene films were extracted with dichloromethane for 24 h and dried under vacuum at 40 °C for 2 h before use. Etching was done in chromic acid (H2SO4:H2O:K2Cr2O7 80:10:10) at 70 °C. Sample Preparation. Patterned model substrates were prepared on gold-coated Si-wafers via microcontact printing (µCP) with a polydimethylsilane (PDMS) stamp, exhibiting 20 × 20 µm2 squares, separated by 20 µm, as described by Whitesides et al.5 After they were rinsed twice sequentially with deionized water, alcohol, and n-heptane and finally with alcohol again, the stamps were dried under a prepurified nitrogen-stream and inked with 1 mM dodecanethiol in ethanol. Gold-coated Si-wafers were stamped immediately after gold-deposition, rinsed with either 1 mM 12-mercapto-1-undecanol or 1 mM 11-mercapto-1-undecanoic acid, washed with ethanol, and dried under a stream of prepurified nitrogen gas. Cantilevers were placed in a 1 mM solution of 11-mercaptoundecanamide in ethanol, immediately after gold-deposition for at least 12 h, washed with ethanol, and dried under a stream of prepurified nitrogen gas. Reaction Conditions. A solution of toluene (5.0 g), butyl isocyanate (5.0 g), and triethylamine (1.3 g) was prepared under inert conditions, and the patterned Si-wafers were placed in the solution at room temperature during the desired reaction times. Subsequently, the samples were rinsed with toluene and dried under a stream of prepurified nitrogen gas. Solutions without catalyst contained toluene (6.3 g) and butyl isocyanate (5.0 g). Instrumentation. Adhesion forces were determined from force-distance curves, measured in water with a Topometrix Explorer (TMX 1010) AFM with an open liquid cell. Lateral force images were obtained with a Topometrix Discoverer (TMX 2010) with a closed liquid cell. Scans of 75 × 75 µm2 were measured in ethanol with an amide-functionalized tip in contact mode. pH-dependent measurements on the polyethylene films were performed with a sulfate-functionalized tip6 in aqueous phosphate buffers.2h Conversion Determination. During the reaction of an OHfunctional surface with butyl isocyanate, the OH groups are converted into urethane links with a butyl as the end group of the SAM (Figure 1). The conversion (f) can be defined as the fraction of CH3 groups on the surface over the total number of end groups. (4) Hutter, J. L.; Bechhoeffer, J. Rev. Sci. Instrum. 1993, 64, 1868. (5) (a) Kumar, A.; Whitesides, G. M. Appl. Phys. Lett. 1993, 63, 2002. (b) Kumar, A.; Biebuyck, H. A.; Whitesides, G. M. Langmuir 1994, 10, 1498. (6) Bertilsson, L.; Liedberg, B. Langmuir 1993, 9, 141.
© 1997 American Chemical Society
4940 Langmuir, Vol. 13, No. 19, 1997
Letters
Figure 1. Reaction scheme of (a) an alcohol-functional surface and (b) an acid-functional surface with butyl isocyanate.
f)
[CH3] [CH3] + [OH]
(1)
We assume that the surface free energies γ12 (probe-surface interface) and γ23 (surface-medium interface) are only influenced by the change of the end groups during the reaction, that the energies are directly proportional to the fraction of CH3 and OH groups, and that γ13 (the probe-medium interface) is constant.1a To determine the local reaction conversion, we used the adhesion force calculated from the force-distance curves, measured with SFM.2g It was found that these systems can be described reasonably well with the Johnson-Kendall-Roberts theory of adhesion mechanics.7 According to the JKR theory, the measured adhesion force is proportional to the work of adhesion
3 Fadh ) - πRW12 2
(2)
The work of adhesion W12 is proportional to the surface free energies according to the Dupre´ equation and thus proportional to the reaction conversion (change in end groups):
f)
Fadh,t - Fadh,0 Fadh,∞ - Fadh,0
(3)
where Fadh,0 is the adhesion force before the reaction (OHfunctional surface), Fadh,∞ is the adhesion force at the end, and Fadh,t is the time-dependent adhesion force during the reaction.
Results Reaction Kinetics. To study the surface chemical reactions, a substrate with self-assembled monolayers was used with either OH or COOH terminal groups. During the reaction of butyl isocyanate with the OH-terminated SAM, the end groups are converted into a urethane link with the butyl as the end group of the SAM (Figure 1). The reaction conversion was studied with an amidefunctionalized SFM probe via force-distance profiles, from which the adhesion forces were evaluated. Since the amide (CONH2) groups on the SFM probe can form strong hydrogen bonds with an OH-functional surface, a large adhesion force between tip and sample is expected. During the reaction of butyl isocyanate with the OH-groups, the end groups of the SAMs will change from alcohol groups to methyl groups.8 The resulting CONH2-CH3 interac(7) Johnson, K. L.; Kendall, K.; Roberts, A. D. Proc. R. Soc. London A 1971, 324, 301. (8) Besides the fact that the chemical nature of the SAM changes upon reaction, also the chain length increases. This could induce shielding of the unreacted OH-groups, resulting in a larger distance between the OH-groups and the tip. Since the chain length increases only a few angstroms, the effect on the hydrogen bonding should be minimal, because these interactions are of relatively long range (F ∼ D-3).9 (9) Israelachvili, J. Intermolecular and Surface Forces, 2nd ed.; Academic Press: New York, 1992.
Figure 2. (a) Decrease of adhesion forces, measured with SFM with chemical specificity during the reaction of an OH-surface with catalyst (9), a COOH-surface with catalyst (b), and a COOH-surface without catalyst (2). (b) ‘Microscopic’ conversions calculated from these adhesion forces. Every data point represents an average over 50 force-distance measurements.
tions are much weaker (purely van der Waals forces) than the CONH2-OH interactions (van der Waals and hydrogen bonding), leading to a decrease in adhesion force (Figure 2a, 9). From the measured changes in the adhesion forces, the local chemical composition and the reaction conversion (eq 3) of a chemical surface modification can be determined on a microscopic scale (Figure 2b). If a COOH-functional surface is used, the reaction with butyl isocyanate forms an amide link and the surface will also become hydrophobic. The reaction of butyl isocyanate is found to be less efficient on a COOH-surface than on an OH-functional surface, since the slope of the conversion at the beginning of the reaction is less steep in the case of an acid surface (b). If no catalyst (triethylamine) is used, the slope of the conversion is even smaller, indicating a slower reaction rate (2). The higher reaction rates in the presence of the base catalyst have also been observed in the bulk, where presumably the initial formation of a base-isocyanate complex is involved.10 In general, the reaction rates decrease in the order OH with catalyst > COOH with catalyst > COOH without catalyst. A direct quantitative comparison between the alcohol and acid surfaces cannot be made, since different reaction mechanisms occur. The alcohol is converted into an urethane, while the acid is converted to an amide after the loss of carbon dioxide. Moreover, reaction constants reported in the literature10,11 apply only to reactions in the bulk performed under specific experimental conditions. Imaging. As shown in previous studies on structurally similar SAMs of alkanethiols,2a,b,g the magnitude of the adhesion force is directly proportional to the magnitude (10) Ephraim, S.; Woodward, A. E.; Mesrobian, R. B. J. Am. Chem. Soc. 1958, 80, 1326. (11) For example: (a) Hakiki, A.; Herz, J. E.; Beinert, G. Polymer 1992, 33, 4575. (b) Priola, A.; Bongiovanni, R.; Gozzelino, G. Eur. Polym. J. 1994, 30, 1047.
Letters
Langmuir, Vol. 13, No. 19, 1997 4941
Figure 3. Friction force images (70 × 70 µm2) of CH3/COOH patterns before and after the reaction with butyl isocyanate. Bright (dark) areas represent high (low) friction. The disappearance of contrast is observed when the acid surface changes into a methylterminated surface. (a) Before reaction and after (b) 30 min, (c) 90 min, and (d) 360 min.
of the friction force. In this way, a chemical map of the distribution of reacted (and unreacted) surface functional groups during a reaction can be obtained. This chemical mapping of the functional group distribution is demonstrated by the chemically specific lateral force imaging of a pattern of CH3/COOH regions of self-assembled monolayers which have reacted with butyl isocyanate in the presence of the catalyst. For the CH3/COOH pattern, a moderate frictional contrast can be seen before the reaction between the CH3 squares (dark, low friction) and the reactive COOH surroundings (bright, high friction) (Figure 3). During the reaction, the acid surfaces change into methyl-terminated surfaces, leading to a diminishing of the frictional contrast (following the decrease of the adhesion force2g). After 6 h the reaction is assumed to be complete, since the acid regions of the initial surface have approached the friction of the methyl regions and frictional contrast is no longer detectable (Figure 3d).12 In this way, differences in the surfaces functional group distribution can be imaged during a reaction. No differences in frictional force can be seen within the reacted acid regions
at the micron scale, indicating a homogeneous conversion across the surface. Etching of Polyethylene Films. As an example of the applications of this technique for measuring local changes in the surface chemical composition, we determined the time-dependent conversion of the etching process of polyethylene films with chromic acid. The (12) Recently it has been shown13 that friction and adhesion have a viscoelastic component if the length of the SAM chain exceeds eight carbons. However, in our experiments we start with a SAM of which both alkanethiols exhibit the same length, so the compliance should be equal for both components at that stage. Upon reaction, the acid areas are extended with four carbons, which should not alter the surface compliance significantly. Moreover, the intermolecular hydrogen bonding between the formed urethane links enhances the rigidity of the monolayer.14 A change in viscoelastic properties upon reaction should show differences in friction in the pattern even after the reaction is complete, which is not observed in Figure 3d. (13) (a) Barger, W.; Koleske, D.; Fledman, K.; Kruger, K.; Colton, R. Polym. Prepr. 1996, 37 (2), 606. (b) Bar, G.; Rubin, S.; Parikh, A. N.; Swanson, B. I.; Zawodzinski, T. A., Jr.; Whangbo, M.-H. Langmuir 1997, 13, 373. (14) Tam-Chang, S.-W.; Biebuyck, H. A.; Whitesides, G.; Jeon, N.; Nuzzo, G. Langmuir 1995, 11, 4371.
4942 Langmuir, Vol. 13, No. 19, 1997
Letters
first few minutes of the etching in which aldehydes and ketones are present and the strong increase at longer etching times because of the introduction of acid groups. To prove that acid groups are indeed formed, pHdependent measurements were performed on modified and unmodified polyethylene surfaces (Figure 4b).2h A sulfate SFM probe (SO3-) was used, since these groups are negatively charged over the whole pH-range; thus, negative charges on the surface can immediately be detected as a repulsive contribution to the total adhesion force. For the unmodified polyethylene surface, no pH dependence was found, since it has no acid groups, so the adhesion force is constant over the whole pH-range. For the film which was oxidized with chromic acid, a pH dependence can be seen. At low pH, the acid groups are protonated, giving strong hydrogen bonding. On increasing pH, the acid groups become deprotonated, which makes them negatively charged. The repulsive interactions between the negative probe and the surface cause a decrease in adhesion force above the pKa of the acid surface. A similar trend has been observed for COOH-terminated SAMs.2h,i
Figure 4. (a) Change in adhesion forces during the timedependent etching process of polyethylene films. (b) ‘Adhesion’ titration curves of polyethylene films: unmodified (9), etched with chromic acid for 45 min (b).
etching introduces alcohol, aldehyde, ketone and acid groups, which increases the hydrophilicity of the surface. The time-dependent change in hydrophilicity is measured as an increase in adhesion force between the surface and an amide-functionalized SFM probe (Figure 4a). During the first few minutes of the etching process, alcohol groups are formed which rapidly give scission products (aldehydes and ketones).15 These groups can finally be oxidized to acid groups which form much stronger hydrogen bonds with the amide tip than the aldehydes and ketones. This might explain the small increase in adhesion force in the (15) Blais, P.; Carlsson, D. J.; Csullog, G. W.; Wiles, D. M. J. Colloid Interface Sci. 1974, 47, 636.
Conclusions In this study we have shown that, using SFM with chemical specificity, one can determine the local functional group distribution and the reaction conversions of a surface modification reaction on a microscopic scale. Addition of the catalyst triethylamine to the reaction medium containing butyl isocyanate shows an expected increase in the conversion rate of COOH end groups. A further increase occurs when an OH-functional surface is used instead of a COOH-functional surface. Furthermore, with chemically specific lateral force imaging, we were able to map the distribution of reacted (and unreacted) functional groups on a microscopic scale during the reaction process. As an example of the characterization of surface modifications we have shown that the etching process of polyethylene films with chromic acid can be followed in time. pH-dependent measurements indicate that acid groups are present on the surface. Acknowledgment. This work was financially supported by the Netherlands Foundation for Chemical Research (SON). LA970364D
