Langmuir 2008, 24, 5249-5252
5249
Temporal Evolution of the Composition of Mixed Monolayers on TiO2 Surfaces: Evidence for a Dimerization-Induced Chelate Effect Gregory R. Soja, Jonathan R. Mann, and David F. Watson* Department of Chemistry, UniVersity at Buffalo, The State UniVersity of New York, Buffalo, New York 14260-3000 ReceiVed March 7, 2008 Mixed monolayers of octanoic acid (OA) and 16-mercaptohexadecanoic acid (MHDA) were adsorbed to nanocrystalline TiO2 films from mixed solutions in tetrahydrofuran. For a range of solution compositions, the mole fraction of MHDA within the mixed monolayers (χMHDA,surf) exceeded that of the coadsorption solution. In addition, χMHDA,surf increased with time, while the sum of the surface coverages of MHDA and OA remained constant. To account for these effects, we propose a mechanism involving disulfide formation between the terminal thiol groups of surface-adsorbed MHDA molecules. Disulfide formation leads to an increase in the surface adduct formation constant (Kad) of dimeric MHDA, causing the gradual displacement of OA from the surface. The mechanism is supported by spectroscopic evidence and desorption kinetics. These are the first examples of mixed monolayers that undergo time-dependent compositional changes as a result of covalent bond formation between surfactants. Our findings illustrate that dimerization and other intermolecular interactions between surfactants may dramatically influence the composition and terminal functionalization of a wide range of mixed monolayer systems.
Introduction The preparation of mixed monolayers on surfaces has received significant attention over the past two decades. Binary mixed monolayers have served as model systems for evaluating the influence of surfactant functionalization on the physical properties, chemical reactivity, and biological response of surfaces and interfaces.1–8 Most investigations have focused on binary mixed monolayers of functionalized alkanethiols on gold surfaces.1,2,5,7 Mixed monolayers of alkylcarboxylates and alkylphosphonates on metal oxide surfaces have also been reported.9–12 In principle, the properties and reactivity of mixed monolayers can be tuned by varying the structure, functionalization, relative abundance, and spatial distribution of surfactants. Therefore, mixedmonolayer-functionalized surfaces are promising for applications in sensing, biomolecular recognition, and molecular electronics.5,13–16 In addition, mixed-monolayer-functionalized surfaces and mixed-monolayer-protected nanoparticles have been used as substrates and building blocks for materials assembly.12,17 * Corresponding author. E-mail:
[email protected]. (1) Bain, C. D.; Evall, J.; Whitesides, G. M. J. Am. Chem. Soc. 1989, 111, 7155–7164. (2) Bain, C. D.; Whitesides, G. M. J. Am. Chem. Soc. 1989, 111, 7164–7175. (3) Prime, K. L.; Whitesides, G. M. J. Am. Chem. Soc. 1993, 115, 10714– 10721. (4) Atre, S. V.; Liedberg, B.; Allara, D. Langmuir 1995, 11, 3882–3893. (5) Ostuni, E.; Yan, L.; Whitesides, G. M. Colloids Surf., B 1999, 15, 3–30. (6) Chapman, R. G.; Ostuni, E.; Takayama, S.; Holmlin, R. E.; Yan, L.; Whitesides, G. M. J. Am. Chem. Soc. 2000, 122, 8303–8304. (7) Love, J. C.; Estroff, L. A.; Kriebel, J. K.; Nuzzo, R. G.; Whitesides, G. M. Chem. ReV. 2005, 105, 1103–1169. (8) Arima, Y.; Iwata, H. Biomaterials 2007, 28, 3074–3082. (9) Tosatti, S.; Michel, R.; Textor, M.; Spencer, N. D. Langmuir 2002, 18, 3537–3548. (10) Zwahlen, M.; Tosatti, S.; Textor, M.; Ha¨hner, G. Langmuir 2002, 18, 3957–3962. (11) Sasahara, A.; Uetsuka, H.; Onishi, H. Langmuir 2003, 19, 7474–7477. (12) Mann, J. R.; Watson, D. F. Langmuir 2007, 23, 10924–10928. (13) Drechsler, U.; Erdogan, B.; Rotello, V. M. Chem.sEur. J. 2004, 10, 5570–5579. (14) Yasutomi, S.; Morita, T.; Imanishi, Y.; Kimura, S. Science 2004, 304, 1944–1948. (15) Luderer, F.; Walschus, U. Top. Curr. Chem. 2005, 260, 37–56. (16) Heimel, G.; Romaner, L.; Zojer, E.; Bre´das, J.-L. Nano Lett. 2007, 7, 932–940.
Precisely controlling the composition of mixed monolayers, while necessary for tuning their properties and reactivity, is not straightforward. Mixed monolayers are typically prepared by coadsorption from mixed solutions.1,2,7 Because of differences in the solubility and surface-binding affinity of surfactants, the compositions of mixed monolayers often differ from those of the corresponding mixed coadsorption solutions.1,2,18–20 Noncovalent interactions between surfactants have also been shown to influence mixed-monolayer composition.2,21–24 Given the myriad potential applications of mixed monolayers, understanding the influence of surfactant structure and functionalization on mixed-monolayer composition represents an important and ongoing challenge in surface chemistry. In this letter, we describe a mixed monolayer system that undergoes unprecedented time-dependent compositional changes after the initial coadsorption reaction. The mixed monolayers consist of n-octanoic acid (OA) and 16-mercaptohexadecanoic acid (MHDA) adsorbed to the surfaces of nanocrystalline TiO2 films. The composition of the mixed monolayers was found to depend on both the composition of the coadsorption solution and the reaction time for surfactant adsorption. To account for these observations, we present a mechanism in which the formation of disulfide bonds between neighboring MHDA thiol groups leads to increased coverage of MHDA on the TiO2 surface. Our findings illustrate that covalent interactions between surfactant functional groups can dramatically influence the composition and adsorption–desorption equilibria of mixed monolayers. (17) Shenhar, R.; Rotello, V. M. Acc. Chem. Res. 2003, 36, 549–561. (18) Folkers, J. P.; Laibinis, P. E.; Whitesides, G. M.; Deutch, J. J. Phys. Chem. 1994, 98, 563–571. (19) Chen, S.; Li, L.; Boozer, C. L.; Jiang, S. J. Phys. Chem. B 2001, 105, 2975–2980. (20) Choo, H.; Cutler, E.; Shon, Y.-S. Langmuir 2003, 19, 8555–8559. (21) Bain, C. D.; Whitesides, G. M. J. Am. Chem. Soc. 1988, 110, 3665–3666. (22) Laibinis, P. E.; Fox, M. A.; Folkers, J. P.; Whitesides, G. M. Langmuir 1991, 7, 3167–3173. (23) Kang, J. F.; Liao, S.; Jordan, R.; Ulman, A. J. Am. Chem. Soc. 1998, 120, 9662–9667. (24) Carot, M. L.; Macagno, V. A.; Paredes-Olivera, P.; Patrito, E. M. J. Phys. Chem. C 2007, 111, 4294–4304.
10.1021/la800731p CCC: $40.75 2008 American Chemical Society Published on Web 04/10/2008
5250 Langmuir, Vol. 24, No. 10, 2008
Letters
Experimental Section Materials and Instrumentation. Titanium(IV) tetraisopropoxide was obtained from Alfa Aesar. Sodium borohydride was obtained from Fisher Scientific. MHDA, n-hexadecanoic acid (HDA), OA, and polyethylene glycol were obtained from Aldrich. Tetrahydrofuran (THF) and nitric acid were obtained from a variety of sources. Reagents were used without further purification. FTIR spectra were obtained using a Perkin-Elmer 1760-X IR spectrometer with a resolution of 2 cm-1. TiO2 Synthesis. Nanocrystalline TiO2 films were prepared following the method of Heimer et al.25 Titanium(IV) tetraisopropoxide (50 mL) was slowly added to rapidly stirring dilute nitric acid (300 mL, 0.5% v/v). The mixture was boiled until the total volume was reduced to 90 mL. This solution was then held at 200 °C in a sealed vessel for 15 h, after which polyethylene glycol (5.4 g) was added and the solution was stirred for an additional 8 h. The films were prepared by spreading a thin layer of this mixture over glass slides and then annealing in air (430 °C, 30 min). SEM and XRD measurements have shown that the films are 4.1 ( 0.9 µm thick and consist of anatase TiO2 with average particle diameters of 20-30 nm.12,26 Preparation and Characterization of Monolayers. TiO2 films were immersed in freshly prepared THF solutions of OA, HDA, and/or MHDA. The sum of the concentrations of alkanoic and mercaptoalkanoic acids in solution was always 2.0 mM, which is sufficient to yield saturation surface coverages.12 Five minutes of immersion was found to be sufficient to reach saturation surface coverages of pure monolayers of OA, HDA, and MHDA adsorbed as carboxylates. After immersion, the films were rinsed with THF and allowed to dry. Transmission-mode IR spectra from 2600 to 3200 cm-1 were obtained to characterize the composition of monolayers. The TiO2-coated glass slides absorbed strongly below 2300 cm-1 but were transparent throughout the C-H and S-H stretching regions. Approximately 5 mg/mL sodium borohydride was added to solutions to prepare films under reducing conditions. Surface Coverage Calculations. All IR spectra were baseline corrected. A Gaussian and a constant were subtracted from each spectrum to minimize the baseline absorbances in the regions of 3200-3000 cm-1 and 2800-2600 cm-1. The average of the spectra from six films soaked in 2.0 mM MHDA was then used as a second baseline to determine the contribution from the asymmetric CH3 stretching (νa(CH3)) band of OA to the total absorbance at 2961 cm-1. This value, together with the extinction coefficient of OA OA at 2961 cm-1 (2961 ) (1.40 × 105 cm2 mol-1 as calculated from Beer-Lambert plots of OA solutions in CCl4), was used to calculate the surface coverage of OA (ΓOA) on each film (1). The calculated ΓOA value, together with the extinction coefficient of the νa(CH2) OA band of OA at 2928 cm-1 (2928 ) (3.51 × 105 cm2 mol-1), was used to calculate the absorbance of OA at 2928 cm-1. This value was subtracted from the total measured absorbance at 2928 cm-1 to yield the contribution from MHDA. The surface coverage of MHDA (ΓMHDA) was calculated from the net absorbance and the extinction MDHA coefficient of MHDA at 2928 cm-1 (2928 ) (8.92 × 105 cm2 mol-1) (2). The mole fraction of MHDA on the surface (χMHDA,surf) was calculated as ΓMHDA divided by the sum of ΓOA and ΓMHDA (3).
A(νa(CH3)OA) OA 2961
) ΓOA
A(νa(CH2)total) - (ΓOAOA 2928) MHDA 2928
) ΓMHDA
ΓMHDA ) χMHDA,surf ΓMHDA + ΓOA
(1)
(2)
Figure 1. IR spectra of TiO2 films as a function of immersion time in THF solutions containing 0.2 mM MHDA and 1.8 mM OA.
MHDA. Alkanoic acids and mercaptoalkanoic acids adsorb to metal oxides as carboxylates, primarily by monodentate coordination of oxygen to the surface.26,27 For the preparation of mixed monolayers, the concentrations of OA and MHDA were varied while maintaining a total carboxylic acid concentration of 2.0 mM. Solution MHDA mole fractions (χMHDA,soln) of 0.10, 0.25, 0.50, and 0.75 were used. The compositions of the mixed monolayers on TiO2 were determined by an analysis of the C-H stretching region of the IR spectra. In a typical experiment, TiO2-coated glass slides were immersed in mixed THF solutions of OA and MHDA for varying amounts of time. The films were rinsed thoroughly with THF before IR spectra were obtained. Spectra of TiO2 films that had been immersed in THF solutions containing 1.8 mM OA and 0.2 mM MHDA (χMHDA,soln ) 0.1) are shown in Figure 1. Each spectrum represents the average of four TiO2 films. Bands were assigned on the basis of previously reported spectra of alkanoic acids adsorbed to metal oxide surfaces.27,28 The spectra changed dramatically with immersion time. The absorbance of the 2961 cm-1 νa(CH3) band decreased with immersion time, whereas the absorbances of the 2928 cm-1 νa(CH2) band and the 2857 cm-1 νs(CH2) band increased. (The absorbance of the 2878 cm-1 νs(CH3) band appeared to decrease with time, but the band was weak and poorly resolved.) The observed spectral changes were not caused by a photochemical reaction because identical results were obtained for films prepared and analyzed in the dark and under laboratory lighting. In a control experiment, mixed monolayers of HDA and OA on TiO2 surfaces were prepared. The IR spectra of mixed HDA-OA monolayers were invariant with time. The changes in the IR spectra of mixed monolayers of MHDA and OA are consistent with a decrease in the surface coverage of OA and an increase in the surface coverage of MHDA with increasing immersion time. Because OA contains a terminal methyl group and MHDA does not and because of the different number of CH2 groups, the composition of mixed monolayers could be calculated from the measured absorbances at 2961 and 2928 cm-1, as described in the Experimental Section. The IR data reveal four interesting aspects of the mixedmonolayer formation process. First, for each composition of the
(3)
Results and Discussion The mixed monolayers were prepared by immersing TiO2coated glass slides in THF solutions containing both OA and
(25) Heimer, T. A.; D’Arcangelis, S. T.; Farzad, F.; Stipkala, J. M.; Meyer, G. J. Inorg. Chem. 1996, 35, 5319–5324. (26) Dibbell, R. S.; Soja, G. R.; Hoth, R. M.; Watson, D. F. Langmuir 2007, 23, 3432–3439. (27) Allara, D. L.; Nuzzo, R. G. Langmuir 1985, 1, 52–66. (28) Arnold, R.; Azzam, W.; Terfort, A.; Wo¨ll, C. Langmuir 2002, 18, 3980– 3992.
Letters
Langmuir, Vol. 24, No. 10, 2008 5251 Scheme 1. Dimerization-Induced Chelate Effect
Figure 2. Surface coverages of MHDA (ΓMHDA) and OA (ΓOA) and total surface coverage (Γtotal) as a function of immersion time in coadsorption solutions with MHDA mole fractions (χMHDA,soln) of 0.25 (shaded black symbols) and 0.50 (open red symbols). Each data point represents the average of four films.
Figure 3. MHDA mole fractions within mixed monolayers (χMHDA,surf) as a function of immersion time and MHDA mole fraction in the adsorption solution (χMHDA,soln). Each data point represents the average of four films.
coadsorption solution, the sum of the surface coverages of OA and MHDA remained essentially constant (∼10-7 mol/cm2) for immersion times from 30 min to 6 h (Figure 2). These coverages are consistent with the saturation surface coverages of pure MHDA, OA, and HDA monolayers on nanocrystalline TiO2.12 Second, for each composition of the coadsorption solution and at all immersion times, χMHDA,surf was greater than χMHDA,soln (Figure 3). Third, for each composition of the coadsorption solution, χMHDA,surf increased with immersion time (Figures 2 and 3). The composition of mixed monolayers changed more rapidly as χMHDA,soln increased. Fourth, for a given immersion time, χMHDA,surf increased with χMHDA,soln. Coadsorption solutions with χMHDA,soln ) 0.10 yielded monolayers with χMHDA,surf ) 0.72 ( 0.09 after 6 h of immersion. For χMHDA,soln ) 0.75, the monolayers essentially consisted entirely of MHDA for all immersion times. Importantly, for all coadsorption solutions, sufficient volumes were used to ensure that the amounts of MHDA and OA in solution exceeded the amounts required for saturation surface coverages of pure monolayers. Therefore, neither surfactant was a limiting reagent in the adsorption process. Taken together, the data imply that the surfactants of mixed monolayers are in dynamic equilibrium with solution and that MHDA displaces OA after saturation surface coverages have been established. Our earlier equilibrium binding experiments
have shown that the surface adduct formation constants (Kad) of MHDA and alkanoic acids on nanocrystalline TiO2 films are nearly identical.12 Thus, the preferential adsorption of MHDA cannot simply be attributed to differences in the relative solubilities or surface affinities of MHDA and OA. Furthermore, differential solvation alone would not account for the time dependence of mixed-monolayer composition. Instead, we propose a mechanism (Scheme 1) in which the time-dependent increase in the MHDA surface coverage is caused by disulfide formation between the thiol groups of surfaceadsorbed MHDA molecules. Solvated and surface-adsorbed thiols are readily oxidized to disulfides by weak oxidizing agents, including molecular oxygen.29,30 Therefore, disulfide bonds are likely to form between adjacent MHDA molecules on the TiO2 surface. Disulfide-bridged MHDA dimers are bidentate surfactants with two carboxylate groups for surface attachment, whereas OA and MHDA monomers are monodentate. The binding constants of bidentate ligands are typically several orders of magnitude greater than those of monodentate ligands.31,32 This well-established chelate effect can result from entropic contributions, an enthalpic component from the formation of metallocycles, and an increased rate of attachment of the second ligating atom of bidentate ligands as a result of its high local concentration. Within mixed monolayers of MHDA and OA, the formation of disulfide-bridged MHDA dimers causes an increase in Kad relative to OA or monomeric MHDA. (Alternatively, the dissociation constant (Kd) of dimeric MHDA is less than that of OA or monomeric MHDA.) Disulfide formation causes k1/k-1 to be greater than k1′/k-1′ (Scheme 1). In addition, the high concentration of the dissociated carboxylic acid group of dimeric MHDA at the TiO2 surface promotes rapid readsorption. As a result of these factors, the dissociation of OA from the TiO2 surface should be more favorable than the dissociation of disulfide-bridged MHDA dimers. The proposed mechanism suggests that dynamic equilibrium favors an increase in χMHDA,surf after the initial establishment of saturation surface coverage, consistent with the measured changes in monolayer composition. Given that the measured Kad values of mercaptoalkanoic acids and alkanoic acids are nearly identical,12 disulfide formation (and the resultant increase in Kad) must occur on the TiO2 surface rather than in solution. If disulfides formed in solution, then the Kad values of mercaptoalkanoic acids would be significantly greater than those of alkanoic acids. Three experiments were performed to confirm the influence of disulfide formation on mixed monolayer composition. First, we obtained IR spectra of MHDA-functionalized TiO2 films prepared in the presence of NaBH4 (Figure 4a). Borohydride is (29) Wallace, T. J.; Schriesheim, A. J. Org. Chem. 1962, 27, 1514–1516. (30) Yiannios, C. N.; Karabinos, J. V. J. Org. Chem. 1963, 28, 3246–3248. (31) Basolo, F.; Pearson, R. G. Mechanisms of Inorganic Reactions: A Study of Metal Complexes in Solution; John Wiley and Sons: New York, 1967. (32) Martell, A. E. AdV. Chem. Ser. 1967, 62, 272–294.
5252 Langmuir, Vol. 24, No. 10, 2008
Letters
Figure 4. (a) ν(S-H) region of the IR spectra of MHDA-functionalized TiO2 films prepared from Ar-purged MHDA solutions with NaBH4 (–) and air-exposed MHDA solutions without NaBH4 (- - -). (b) Equilibrium MHDA mole fractions of mixed monolayers (χMHDA,surf) as a function of MHDA mole fractions in adsorption solutions (χMHDA,soln) with (0) and without (9) NaBH4. The dashed line has a slope of 1 and corresponds to χMHDA,surf equaling χMHDA,soln. Each data point represents the average of four films.
known to reduce disulfides to thiols.33,34 The spectra of TiO2 films that had been immersed in Ar-purged, NaBH4-saturated solutions of MHDA contained a weak band centered at 2545 cm-1 corresponding to the S-H stretching (ν(S-H)) mode.35 The spectra of TiO2 films that had been immersed in air-exposed solutions of MHDA without NaBH4 contained no ν(S-H) band. These spectral differences clearly indicate that the thiol groups of MHDA are readily oxidized to disulfides by oxygen in the absence of a reducing agent. Second, we prepared mixed monolayers from THF solutions containing OA, MHDA, and saturated NaBH4. For a range of solution compositions, the relative equilibrium surface coverages of OA and MHDA in the presence of NaBH4 were nearly equal to their relative concentrations in the coadsorption solution (Figure 4b). (χMHDA,surf was slightly lower than χMHDA,soln, which we attribute to differences in the solubility of MHDA and OA in THF.) The lack of enrichment of the MHDA surface coverage in the presence of NaBH4 provides compelling evidence that disulfide formation is responsible for this effect in the absence of a reducing agent. Importantly, these data also reveal that the equilibrium composition of mixed monolayers of OA and MHDA can be precisely controlled by preparing the monolayers in the presence of a reducing agent, thereby minimizing disulfide formation. Finally, we measured the relative desorption kinetics of pure monolayers of OA and MHDA (Figure 5). OA- and MHDAfunctionalized TiO2 films were immersed in air-exposed THF containing 0.1 mM HCl. (HCl was added to THF to simulate the proton concentration in mixed coadsorption solutions, which are in equilibrium with the surface-adsorbed carboxylates of OA and MHDA.) The desorption of OA occurred significantly more rapidly than that of MHDA. After 4 h of immersion in acidified THF, the surface coverages of MHDA and OA decreased to (45 ( 2)% and (19 ( 1)% of their original values, respectively. The greater stability of MHDA compared to that of OA on the TiO2 surface provides further evidence that disulfide formation gives rise to an increase in Kadthrough a chelate effect. (33) Jocelyn, P. C. Methods Enzymol. 1987, 143, 246–256. (34) Hansen, R. E.; Osteraard, H.; Norgaard, P.; Winther, J. R. Anal. Biochem. 2007, 363, 77–82. (35) Young, A. G.; Green, D. P.; McQuillan, A. J. Langmuir 2007, 23, 12923– 12931.
Figure 5. Relative surface coverages (Γ) of MHDA (9) and OA (0) on nanocrystalline TiO2 films as a function of immersion time in THF with 0.1 mM HCl. Each data point represents the average of four films.
Conclusions We have shown that disulfide-induced dimerization causes a time-dependent increase in the surface coverage of thiolterminated surfactants within mixed monolayers on TiO2 surfaces. These represent the first mixed monolayers that undergo timedependent compositional changes through covalent bond formation after initially reaching saturation surface coverage. Our findings may be broadly applicable because they highlight the influence of intermolecular interactions on the composition and terminal functionalization of mixed monolayers and, therefore, the properties and reactivity of mixed-monolayer-functionalized surfaces. Acknowledgment. This work was funded, in part, by the National Science Foundation (CHE-0645678) and the State University of New York. In addition, acknowledgment is made to the Donors of the American Chemical Society Petroleum Research Fund for partial support of this research. LA800731P
