1288
Langmuir 2002, 18, 1288-1293
Self-Assembled Monolayers on Gold of Ferrocene-Terminated Thiols and Hydroxyalkanethiols Tommaso Auletta, Frank C. J. M. van Veggel,* and David N. Reinhoudt* Supramolecular Chemistry and Technology and MESA+ Research Institute, University of Twente, P.O. Box 217, 7500 AE Enschede, The Netherlands Received September 25, 2001. In Final Form: November 15, 2001 In this paper, a study on the adsorption of mixed self-assembled monolayers (SAMs) for two different combinations of thiols (Fc(CH2)6SH/HO(CH2)2SH and Fc(CH2)16SH/HO(CH2)11SH (Fc ) ferrocene)) is presented, to obtain surfaces with single isolated ferrocenylalkanethiols embedded in shorter hydroxyalkanethiols. These hydrophilic substrates with very low surface concentration of ferrocene moieties are required to perform force spectroscopy experiments on host-guest supramolecular complexes, by using SAMs of β-cyclodextrin heptathioether adsorbed on gold-coated AFM tips. Several SAMs have been prepared on polycrystalline gold electrodes from 1 mM thiol solutions, changing the ferrocenylalkanethiols/ hydroxyalkanethiols ratio. The amount of electroactive component immobilized on the electrode was determined by cyclic voltammetry, and it has been related to the solution composition. The general trend is that the longer chain component is preferentially adsorbed, suggesting a thermodynamic control of the adsorption. However, relevant differences in the layer formation and composition can be observed for the two systems on the basis of a different balance of the driving forces that govern the adsorption process. Fc(CH2)16SH is adsorbed to a larger extent, compared to the Fc(CH2)6SH for the same solution composition, as shown by higher charge densities values. Furthermore for the Fc(CH2)16SH/HO(CH2)11SH system upon increase of the percentage of ferrocenylalkanethiols in solution, a shift in the redox peak position to more positive values is observed, indicating that phase segregation occurs. The differences between the two systems can be related to chain length effects from which arise favorable enthalpic contributions to the adsorption of the longer chain component.
Introduction Self-assembled monolayer of sulfur-containing molecules on clean gold substrates provide an easy and fast way to form homogeneous monomolecular films with welldefined surface properties, both from solution and from vacuum.1-5 By using a mixture of adsorbates, it is therefore possible to introduce different chemical functionalities on a surface;6,7 the amount of each component adsorbed on the surface and their reciprocal distribution are functions of many different factors such as solution composition, chain length, and temperature. A general trend observed when thiols with different chain lengths are used to form monolayers is that the longer chain component is preferentially adsorbed, with respect to the solution composi* To whom correspondence should be addressed. Fax: +31 53 4894645. Phone: +31 53 4892980. E-mail:
[email protected]. (1) Chailapakul O.; Sun, L.; Xu, C.; Crooks, R. M. J. Am. Chem. Soc. 1993, 115, 12459. (2) Ulman A. An Introduction to Ultrathin Organic Films; From Langmuir-Blodgett to Self-Assembly; Academic Press: New York, 1991. (3) (a) Dubois, L. H.; Nuzzo, R. G. Annu. Rev. Phys. Chem. 1992, 43, 437. (b) Allara, D. L. Biosens. Bioelectron. 1995, 10, 771. (c) Bishop, A. R.; Nuzzo, R. G. Curr. Opin. Colloid Interface Sci. 1996, 1, 127. (4) (a) Ulman, A. Chem. Rev. 1996, 96, 1533. (b) Ulman, A. Characterization of Organic Thin Films; Butterworth-Heinemann: Boston, MA, 1995. (c) Rubinstein, I.; Steinberg, S.; Tor, Y.; Shanzer, A.; Sagiv, J. Nature 1988, 332, 426. (5) Nuzzo, R. G.; Allara, D. L. J. Am. Chem. Soc. 1983, 105, 4481. (6) (a) Bain, C. D.; Whitesides, G. M. J. Am. Chem. Soc. 1988, 110, 6560. (b) Bain, C. D.; Whitesides, G. M. J. Am. Chem. Soc. 1988, 110, 3665. (c) Whitesides, G. M.; Laibinis, P. E. Langmuir 1990, 6, 87. (d) Pale-Grosdemange, C.; Simon, E. S.; Prime, K. L.; Whitesides, G. M. J. Am. Chem. Soc. 1991, 113, 12. (e) Laibinis, P. E.; Fox, M. A.; Folkers, J. P.; Whitesides, G. M.; Deutch, J. Langmuir 1991, 7, 3167. (f) Bertilsson, L.; Liedberg, B. Langmuir 1993, 9, 141. (g) Folkers, J. P.; Laibinis, P. E.; Whitesides, G. M.; Deutch, J. J. Phys. Chem. 1994, 98, 563. (7) Kakiuchi, T.; Iida, M.; Gon, N.; Hobara, D.; Imabayashi, S.-I.; Niki, K. Langmuir 2001, 17, 1599.
tion.8 Phase segregation phenomena are related to the difference in chain length of the two components (stability scale: long-long > short-short > long-short)9-11 or can be induced by the presence of specific functional groups as a hydrogen bond donor/acceptor.12 By increase of the temperature at which the layers are formed, the phase segregation phenomena are suppressed or drastically reduced, depending on the chain length difference of the two components, probably due to a lower number of defects present in the SAMs.13 Thus, to achieve control over the layer composition with respect to the solution from which the layer itself is formed both the kinetics and thermodynamic of the adsorption process have to be taken in consideration. The interactions that lead to the formation of a stable monomolecular film can in general be separated in two main contributions: the binding force between the headgroup (sulfur) and gold (S-Au bond strength energy 45 kcal‚mol-1)14 and the interchain interactions (van der Waals attractions on the order of a few kcal‚mol-1/CH2)3a (Figure 1). (8) (a) Bain, C. D.; Whitesides, G. M. J. Am. Chem. Soc. 1989, 111, 7164. (b) Labinis, P. E.; Nuzzo, R. G.; Whitesides, G. M. J. Phys. Chem. 1992, 96, 5097. (9) Folkers, J. P.; Laibinis, P. E.; Whitesides, G. M. Langmuir 1992, 8, 1330. (10) Offord, D. A.; John, C. M.; Linford, M. R.; Griffin, J. H. Langmuir 1994, 10, 883. (11) Hobara, D.; Ota, M.; Imabayashi, S.-I.; Niki, K.; Kakiuchi, T. J. Electroanal. Chem. 1998, 444, 113. (12) (a) Chen, S.; Li, L.; Boozer, C. L.; Jiang, S. J. Phys. Chem. B 2001, 105, 2975. (b) Smith, R. K.; Reed S. M.; Lewis, P. A.; Monnell, J. D.; Clegg, R. S.; Kelly, K. F.; Bumm, L. A.; Hutchinson, J. E.; Weiss, P. S. J. Phys. Chem. B 2001, 105, 1119. (c) Stranick, S. J.; Parikh, A. N.; Tao, Y.-T.; Allara, D. L.; Weiss, P. S. J. Phys. Chem. 1994, 98, 7636. (d) Stranick, S. J.; Atre, S. V.; Parikh, A. N.; Wood, M. C.; Allara, D. L.; Winograd, N.; Weiss, P. S. Nanotechnology 1996, 7, 438. (13) Chen, S.; Li, L.; Boozer, C. L.; Jiang, S. Langmuir 2000, 16, 9287.
10.1021/la011474u CCC: $22.00 © 2002 American Chemical Society Published on Web 01/26/2002
Self-Assembled Monolayers on Gold
Figure 1. Schematic representation of stabilizing interactions in SAMs (adapted from ref 2).
Another problem to be faced dealing with mixed monolayers, especially containing low amounts of one of the two adsorbates, is the analysis and quantification of the adsorbed components. Electrochemistry provides a straightforward way to quantify the amount of electroactive centers adsorbed on a gold electrode and thus allows an indirect evaluation of the layer composition. In particular, mixed SAMs of ferrocene-containing thiols and n-alkanethiols have widely been investigated.15-19 The redox properties of ferrocenylalkanethiols coadsorbed with n-alkanethiols have been studied as a function of the chain length of the cocomponent and of the presence of polar groups (e.g. amide). The electron-transfer rates between the ferrocene moiety and the gold electrode have been correlated with the nature of different linking moieties between the redox center and the alkyl chain, and the double layer effects on the interfacial redox reaction have been investigated. The final aim of this work is the preparation of surfaces with only a few, isolated ferrocenylalkanethiols/µm2, embedded in a layer of a hydroxyl-terminated thiols, to produce surfaces on which complexation of the ferrocene moieties by β-cyclodextrins moieties can still occur. These substrates are required to perform force spectroscopy measurements of single host-guest complexes, by using SAMs of a β-cyclodextrin heptathioether adsorbed on a gold-coated AFM tip.20 The ferrocenylalkanethiols have to be longer than the cocomponent and homogeneously spaced on the surface, while the hydroxyl moieties are required to reduce the aspecific interactions between the SAM on the substrates and the one on the AFM tip. Moreover, the electrochemical properties of the ferrocene (14) Nuzzo, R. G.; Zegarski, B. R.; Dubois, L. H. J. Am. Chem. Soc. 1987, 109, 733. (15) Chidsey, E. D.; Bertozzi, C. R.; Putwinski, T. M.; Mujsce, A. M. J. Am. Chem. Soc. 1990, 112, 4301. (16) Smalley, J. F.; Feldberg, S. W.; Chidsey, C. E. D.; Linford, M. R.; Newton, M. D.; Liu, Y.-P.J. Phys. Chem. 1995, 99, 13141. (17) Sabapathy, R. C.; Bhattacharyya, S.; Leavy, M. C.; Cleland, W. E.; Hussey, C. L. Langmuir 1998, 14, 124. (18) Sek. S.; Misicka, A.; Bilewicz, R. J. Phys. Chem. B 2000, 104, 5399. (19) (a) Creager, S. E.; Rowe, G. K. Anal. Chim. Acta 1991, 246, 233. (b) Creager, S. E.; Rowe, G. K. J. Electroanal. Chem. 1994, 370, 203. (c) Rowe, G. K.; Creager, S. E. Langmuir 1991, 7, 2307. (d) Weber, K.; Hockett, L.; Creager, S. J. Phys. Chem. B 1997, 101, 8286-8291. (e) Sumner, J. J.; Weber, K. S.; Hockett, L A. Creager; S. E. J. Phys. Chem. B 2000, 104, 7449. (20) (a) Scho¨nherr, H.; Beulen, M. W. J.; Bu¨gler, J.; Huskens, J.; van Veggel, F. C. J. M.; Reinhoudt, D. N.; Vancso, G. J. J. Am. Chem. Soc. 2000, 122, 4963. (b) Bu¨gler, J.; Beulen, M. W. J.; Lammerink, B.; Geurts, F. A. J.; Biemond, E. M. E. F.; Van Leerdam, K. G. C.; Van Veggel, F. C. J. M.; Engberts, J. F. J.; Reinhoudt, D. N. Langmuir 1998, 14, 6424. (c) Beulen, M. W. J.; Bu¨gler, J.; De Jong, M. R.; Lammerink, B.; Huskens, J.; Scho¨nherr, H.; Vancso, G. J.; Boukamp, B. A.; Wieder, H.; Offenhauser, A.; Knoll, W.; Van Veggel, F. C. J. M.; Reinhoudt, D. N. Chem. Eur. J. 2000, 6, 1176.
Langmuir, Vol. 18, No. 4, 2002 1289
units allow a direct comparison of the number of surfaceconfined guests and the number of complexes measured by force spectroscopy experiments.21 To the best of our knowledge there are no reports in the literature concerning self-assembled monolayers on gold electrodes adsorbed from solutions containing ferrocenylalkanethiols and polar cocomponents focusing on the low ferrocene percentage regime. In this paper a study on the formation of mixed monolayers of ferrocenylalkanethiols and hydroxylterminated ones differing in chain length is reported. Cyclic voltammetry and contact angle measurements have been carried out to determine the amount of electroactive component adsorbed on the electrode that has been related to the percentage of the two components in solution. The driving forces that lead to the formation of a stable monomolecular film and determine the ratio of the adsorbed components have been investigated, and the results show that the longer chain component is usually adsorbed preferentially. In addition not only the difference in chain length between the thiols determines the final composition but also the absolute chain length affects significantly the adsorption process, leading to significant deviation from statistical mixtures. Experimental Section Synthesis. 2-Mercaptoethanol and 11-mercapto-1-undecanol were purchased from Aldrich and used without any further purification. 6-Ferrocenylhexanethiol and 16-ferrocenylhexadecanethiol were synthesized according to literature procedures15,19b All solvents used in monolayer preparation were of pa grade. Gold Substrates. Gold substrates were prepared by evaporating 200 nm of gold on a glass slide of 25 mm diameter with a 2 nm titanium layer for adhesion. Just before use the substrates were cleaned by oxygen plasma for 5 min and the resulting oxide layer was removed by leaving the substrates for 10 min in EtOH.22 Monolayer Preparation. All glassware used to prepare monolayers was immersed in pirana solution (concentrated H2SO4 and 33% H2O2 in a 3:1 ratio). (Warning: pirana should be handled with caution; it has detonated unexpectedly.23) Next, the glassware was rinsed with large amounts of high-purity water (Millipore). Monolayers were prepared by immersing the gold substrates in a 1 mM solution (total concentration of both components) at ambient condition (20 ( 3 °C) for 16 h. The samples were then removed from the solution and rinsed with large amounts of dichloromethane, ethanol, and water to wash away any physisorbed material. Electrochemistry. Electrochemical measurements were performed with an AUTOLAB PGSTAT10, in a homemade electrochemical cell equipped with a platinum counter electrode, a mercury sulfate reference electrode (VMSE ) +0.61 VNHE), and a screw cap holding the gold working electrode (area exposed to the solution ) 0.44 cm2). The cyclic voltammograms were performed in NaClO4 at scan rates of 0.1, 0.2, and 0.5 V/s. The charge values for the electroactive species adsorbed on the electrode and the corresponding standard deviations were calculated by averaging the redox peaks at different scan rates. Contact Angle (CA). The advancing and receding contact angles with Q2 Millipore water were measured on a Kru¨ss G10 contact angle measuring instrument, equipped with a CCD camera. Measurements with a drop of water, whose volume was (21) On the basis of the results presented in this work, several substrates were prepared and subsequently tested by using ferrocene percentages in solution between 1% and 0.2%. It was possible to measure host-guest interactions with a rupture force for a single complex of 56 ( 10 pN for both systems. Moreover, direct comparison of the number of decomplexation events obtained by means of this technique and the number of ferrocene molecules immobilized on the tip, calculated from the electrochemical data, gave a good correlation. Some data for this study have already been published; see ref 20a. (22) Ron, H.; Rubinstein, I. Langmuir 1994, 10, 4566. (23) (a) Dobbs, D. A.; Bergman, R. G.; Theopold, K. H. Chem. Eng. News 1990, 68 (17), 2. (b) Wnuk, T. Chem. Eng. News 1990, 68 (26), 2. (c) Matlow, S. L. Chem. Eng. News 1990, 68 (30), 2.
1290
Langmuir, Vol. 18, No. 4, 2002
Figure 2. Cyclic Voltammograms of mixed SAMs of 2-mercaptoethanol and 6-ferrocenylhexane thiol (solution composition reported) (NaClO4, 0.1 M; scan rate ) 0.1 V/s; VMSE ) +0.61 VNHE). gradually increased (advancing CA) and then decreased (receding CA), were repeated on three different sites of the sample. For each adsorbate composition three samples were measured for a total of 12 drops, of which the average values and the corresponding errors are reported.
Results and Discussion Well-defined electrochemical systems can be obtained by self-assembly of thiols containing an electroactive center in which all redox centers are at the same distance from the electrode and therefore the entire monolayer can be addressed in a single scan. The redox current shows a linear dependence with respect to the scan rates, and by integrating the current evolved in a redox cycle it is possible to calculate the surface coverage of the electroactive center and relate it to the concentration of the solution from which the layer is formed. The full width at half-maximum of the anodic voltammometric wave, ∆Efwhm, provides an indication on the interaction taking places between different centers. In the ideal situation where no interactions between the redox centers occur ∆Efwhm ) 3.53RT/nF (90.3/n mV at 24 °C)24 and deviations toward higher and smaller values have been attributed to interacting redox centers. 6-Ferrocenylhexanethiol/2-Mercaptoethanol (Fc(CH2)6SH/HO(CH2)2SH). Mixed monolayers of 6-ferrocenylhexanethiol and 2-mercaptoethanol have been adsorbed from solutions in which the total concentration was kept constant at 1 mM, and cyclic voltammetry measurements have been performed in the presence of NaClO4 as background electrolyte. The cyclic voltammograms recorded for different layers at a scan rate of 0.1 V/s are reported in Figure 2 and show a significant change in the shape and features of the curves with a higher percentage of ferrocene. In particular, the single, symmetric peak observed for ferrocene percentages in solution between 5 and 10% (∆Efwhm ) 120 mV) broadens, showing even multiple peaks above 20% (∆Efwhm ) 204 mV), with changes occurring also for the peak position that moves to more positive values from -0.234 VMSE up to -0.136 VHSE. In addition, at low ferrocene coverage no peak splitting between the anodic and cathodic sweep can be observed as a result of the fast enough electron-transfer kinetic, while an increase of the electroactive molecule in the layer leads to a peak splitting up to 30 mV. The shape of the CVs indicates that by increasing the concentration of the Fc(CH2)6SH in solution a gradual (24) Bard, A. J.; Faulkner, L. R. Electrochemical Methods: Fundamentals and Applications; John Wiley & Sons: New York, 1980; p 522.
Auletta et al.
Figure 3. Charge density of mixed SAMs of 6-ferrocenylhexanethiol and 2-mercaptoethanol (2) and theoretical plot (solid line).
change in the electrochemistry properties of the redox center occurs; below 10% each molecule experiences almost the same environment, and they can be treated as isolated species. Therefore, the peak position and the absence of peak splitting can be related to single, noninteracting isolated ferrocene molecules inserted in the 2-mercaptoethanol environment. Upon increase of the amount of ferrocenethiol on the electrode, the electroactive centers start interacting. However, due to the presence of the cocomponent they are embedded in different chemical surroundings and therefore behave as entities with different electrochemical properties: this shows up in the CVs as a broadening of the peak. The peak shift is also related to this phenomena, because upon oxidation of a ferrocene center a positive charge is formed and the oxidation process of any other ferrocene molecule in close proximity, with formation of yet another positive ferrocenium ion, will be more difficult, thus requiring a more anodic potential. Furthermore, the peak position is also affected by the creation of a less polar environment, due to the presence of the longer alkyl chains, unfavorable with respect to the creation of a positive charge. The values of charge density versus the layer composition are reported in Figure 3 together with a theoretical plot in which the ratio of the two components adsorbed on the electrode is the same as in solution (statistical mixture). This plot was built on the basis of geometrical assumptions, considering that an alkyl chain occupies around 20 Å2,25 while the ferrocene headgroup was treated as a hard sphere of 6.6 Å of diameter.15,26 In this latter case a straight line describes the adsorption process up to a critical concentration value that lies around a ferrocene percentage in solution of 60%. Above this value, due to sterical reasons, no more ferrocene can be fitted on the layer and the charge density reaches a plateau; 2-mercaptoethanol fills the space between the alkyl chain. The corresponding surface coverage for such a full ferrocene layer is 2.7 × 1014 cm-2.15 In the experimental plot the charge density increases upon increasing the percentage of the electroactive component in solution, and in this case the maximum surface coverage calculated for the ideal system is already reached at a ferrocene percentage in solution of 40% (2.74 × 1014 cm-2). The composition of the mixed SAMs adsorbed on polycrystalline gold electrode is a function of the percentage of the two components in solution. Below 10% ferrocene (25) (a) Chidsey, C. E. D.; Liu, G.-Y.; Rowntree, P.; Scoles, G. J. Chem. Phys. 1989, 91, 4421. (b) Camillone, Chidsey, C. E. D., III; Liu, G.-Y.; Putvinki, T. M.; Scoles, G. J. Chem. Phys. 1991, 94, 8493. (c) Fenter, P.; Eisenberger, P.; Liang, K. S. Phys. Rev. Lett. 1993, 70, 2447. (d) Widrig, C. A.; Alves, C. A.; Porter, M. D. J. Am. Chem. Soc. 1991, 113, 2805. (26) Seiler, P.; Dunitz, J. D. Acta Crystallogr. 1979, B35, 1068.
Self-Assembled Monolayers on Gold
in solution the experimental data are in good agreement with the theoretical calculation, indicating a statistical composition of the layer, with respect to the solution. However above this value the experimental curve that describes the adsorption process shows charge densities values higher than the theoretical one indicating that the longer chain component is preferentially adsorbed but external factor such as surface roughness, at most 1.2 for polycrystalline and vapor-deposited substrates,27 cannot explain this difference. The layer formation is then not only determined by the S-Au interaction, and other factors, mainly arising from interchain interactions, play a role in determining the ratio of the two components on the electrode. The balance of the different contributions is however apparently constant over the entire range of concentrations probably due to the fact that the interchain interactions are not strong enough to overcome the other contributions and predominate; thus, the charge densities values between 0% and 30% can be fitted to a first approximation linearly (R2 ) 0.982). Deviation from linearity, in particular at high ferrocene coverage, can be attributed to a saturation effect of the layer that cannot accommodate more electroactive moieties. These results are in agreement with the general features reported in previous studies where the coadsorption of electroactive molecules with alkanethiol monolayers has been investigated19c and the chemical and physical properties of the shorter chain component seem not to be relevant in determining the layer composition. Using as cocomponent a polar hydroxyl-terminated thiol instead of an alkanethiol, characterized by a different diffusion coefficient in the ethanolic solution from which the layer are formed, does apparently not modify the trend for the longer chain component to be preferentially adsorbed, indicating a prevalent thermodynamic control over the layer formation. This suggests that not only the difference in chain length but also the chain length itself has to play a decisive role in determining the driving forces for the layer formation, in particular above 10-12 carbon atom chains, due to enhanced interchain interactions. 16-Ferrocenylhexadecanethiol/11-Mercapto-1-undecanol (Fc(CH2)16SH/HO(CH2)11SH). On the basis of the above-reported results, it can be anticipated that the use of longer chain components can provide a more detailed picture of the layer formation process focusing in particular on additional factors arising from cooperativity effects. Several mixed SAMs have been prepared from solutions containing 11-mercapto-1-undecanol and 16-ferrocenylhexadecanethiol in different ratios, without changing the total thiol concentration of the solution. The cyclic voltammograms recorded for different layers at a scan rate of 0.1 V/s are plotted in Figure 4a, and upon increase of the amount of electroactive component, a remarkable broadening of the peak is present at more positive potential, up to 75% of ferrocene in solution for which, instead, a very sharp peak with a ∆Efwhm ) 62 mV appears. If we however focus in the region below 15% of ferrocene in solution (Figure 4b), the peaks are symmetric and show only a gradual change in the peak position (∆Efwhm ) 105 and 129 mV for the 1% and 10% ferrocenylalkanethiol, respectively). At low percentages of 16ferrocenylhexadecanethiol in solution the adsorbed ferrocene units can be treated as isolated species while upon increasing the molar fraction in solution they start interacting (∆Efwhm ) 212 mV for 20% ferrocene component (27) (a) Rodriguez, J. F., Mebrahtu, T.; Soriaga, M. P. J. Electroanal. Chem. 1987, 233, 283. (b) Schneider, T. W.; Buttry, D. A. J. Am. Chem. Soc. 1993, 115, 12391. (c) Walczak, M. M.; Alves, C. A.; Lamp, B. D.; Porter, M. D. J. Electroanal. Chem. 1995, 396, 103.
Langmuir, Vol. 18, No. 4, 2002 1291
Figure 4. Cyclic voltammograms of mixed SAMs of 11-mercapto-1-undecanol and 16-ferrocenylhexadecanethiol (solution composition reported) (NaClO4, 0.1 M; scan rate ) 0.1 V/s; VMSE ) +0.61 VNHE): (a) high 16-ferrocenylhexadecanethiol percentage; (b) low 16-ferrocenylhexadecanethiol percentage.
Figure 5. Oxidation (9) and reduction (2) peak position versus solution composition of mixed SAMs of 16-ferrocenylhexadecanethiol and 11-mercapto-1-undecanol.
in solution), and for high ferrocene percentage domain formation is assumed. Strong coupled centers are present at 75% of ferrocene in solution as proven by the ∆Efwhm ) 62 mV characteristic of densely packed electroactive SAMs.28 The plot of the peak potential shift versus the solution composition (Figure 5) exhibits an S-shaped curve with an inflection between 15% and 20% of ferrocene in solution. With increase of the percentage of the ferrocene component in solution, the redox potential peak shifts toward more anodic values starting from a value of -0.234 VMSE (1% (28) Sato, Y.; Uosaki, K. Redox Mechanisms and Interfacial Properties of Molecules of Biological Importance; The Electrochemical Society: Pennington, NJ, 1993; p 299.
1292
Langmuir, Vol. 18, No. 4, 2002
Auletta et al.
Figure 6. Charge density of mixed SAMs of 16-ferrocenylhexadecanethiol and 11-mercapto-1-undecanol ([) and theoretical plot (solid line).
Figure 7. Advancing (b) and receding (9) contact angles and hysteresis (2) of mixed SAMs of 16-ferrocenylhexadecanethiol and 11-mercapto-1-undecanol.
Table 1. Oxidation Peak Position of Mixed SAMs of 6-Ferrocenylhexanethiol/2-Mercaptoethanol and 16-Ferrocenylhexadecanethiol/11-Mercapto-1-undecanol (VMSE ) +0.61 VNHE)
a linear relationship with respect to the solution composition, but they exhibit the same trend shown by the peak potential graph versus the solution composition, with an inflection point between 15 and 20% of ferrocene in solution. The advancing and the receding contact angle for monolayers adsorbed from solution with different composition are reported in Figure 7, together with the hysteresis value. The graph of the advancing contact angle shows an increase of the values with high percentages of the ferrocene in solution; also the hysteresis values increase with higher amounts of ferrocene immobilized on the surface. The increase of the contact angle indicates a change in the surface properties of the layer toward a more hydrophobic system, in accordance with the preferential adsorption of the longer chain component, with respect to the hydroxyl component. On the other hand, the hysteresis suggests that the preferential adsorption of the longer component introduces an higher degree of disorder in the surface. This behavior can be related to the domain formation of the ferrocene component separated each other by areas in which mainly the hydroxyl terminated thiol is present. The overall molecular ordering is then decreased, leading to a more disorganized outer surface and thus a higher hysteresis. Final Remarks. From the above-reported data for the system Fc(CH2)16SH/HO(CH2)11SH it is possible to define a critical ratio of the two components in solution that determines a change in the balance of the driving forces that rule the adsorption process, leading to a highly preferential adsorption of the longer component. On the basis of the plots of the charge density versus the solution composition and of the peak position shift, this critical composition lies between 10 and 15% of ferrocene in solution. As shown by the overlap with the theoretical plot, below 10% the layer composition is almost statistical, indicating that the layer formation is mainly ruled by the S-Au interaction, equal for both molecules. Above this composition threshold the longer component is preferentially adsorbed even in a more pronounced extent than Fc(CH2)6SH. The main difference between the two system is related to the absolute chain length of the alkyl spacer bearing the ferrocene moiety; therefore, the additional factor that rules the adsorption equilibrium is correlated to the interchain packing that now predominates, enhancing the tendency for the longer chain component to be adsorbed.30
ferrocene (%)
potential (VMSE) for Fc(CH2)16SH
potential (VMSE) for Fc(CH2)6SH
0.40 0.30 0.10 0.05
0.019 0.038 0.220 0.234
0.136 0.155 0.224 0.234
of Fc(CH2)16SH in solution), corresponding to isolated ferrocene species, similar to the Fc(CH2)6SH/HO(CH2)2SH system, up to a final value of +0.059 VMSE (100% of ferrocene in solution). From the analysis of the peak positions for the Fc(CH2)6SH and the Fc(CH2)16SH thiol mixed SAMs reported in Table 1 it is clear that for the same solution composition the peak shift occurs to a larger extent in Fc(CH2)16SH/ HO(CH2)11SH, supporting the conclusion of more strongly interacting redox centers. The peak splitting also shows a relationship with the solution composition, and below 15% it does not exceed 30 mV, while above this critical concentration it rises up to a maximum of 65 mV, for a percentage of ferrocene equal to 75%. The ferrocene component adsorbed on the gold electrode has been quantified by means of CV, and the charge densities have been related to the solution one (Figure 6). At low concentration (below 15% of Fc(CH2)16SH in solution) the data are in good agreement with the theoretical plot indicating a statistical composition and suggesting that other factors as for example surface roughness are not relevant in determining the amount of electroactive component. Above this “threshold” the longer chain component is preferentially adsorbed. The measured charge densities are even higher than the 6-ferrocenylhexanethiol coadsorbed with the 2-mercaptoethanol and increase since a final value of 7.4 × 10-5 C/cm2 corresponds to a coverage of 4.6 × 1014 cm-2, which means 1.7 times higher than a “full” covered layer.29 Chidsey and coworkers have already reported this phenomenon for longchain ferrocene-alkanethiols coadsorbed with n-alkanethiols with comparable chain length (15 carbon atoms) and attributed it to the possibility, for the ferrocene moieties, to back folding in the poorly packed alkyl chains.15 In our system the charge densities do not follow (29) A “full” layer is the one in which all the ferrocene moieties are well packed and no more can fit due to steric interactions. Even if the headgroups are well packed, a certain degree of disorder can be found in the underlying alkyl chains, due the difference in size between the ferrocene moieties and the alkyl chains, as shown in the calculation for the theoretical curve of the charge densities versus the solution composition.
(30) These results are not in contradiction with the work of Chidsey and co-workers,15 where the increase of chain length of both components (ferrocene thiol and unsubstituted thiol) does not affect the ferrocene surface coverage. In that case both components have comparable chain length; therefore, no discrimination could be expected on the basis of different lateral interaction.
Self-Assembled Monolayers on Gold
The adsorption of monolayers can be divided basically into two different steps: a first and fast step in which molecules are physisorbed on the wafer and lie with their axis almost parallel to the surface,31 described by a Langmuir adsorption isotherm, and a second slower step during which the molecules rearrange and organize their alkyl chains in an all-trans conformation with their axis close to the surface normal (θ ) 30°) and the sulfur headgroup binding to gold (chemisorption). The two-step formation mechanism of SAMs of alkanethiols on gold has been confirmed by using STM and AFM techniques.32-34 The realignment process starts from the boundaries of domains, and therefore, the process is driven mainly by lateral interactions. In the case of a mixed monolayer formation with components differing in chain length, during the second slower phase the longer chain component can reorganize in domains, due to the stronger interchain interactions and start forming island that can then grow by adsorption of molecules from the solution. This mechanism is of course more effective in the case of long-chain components as shown by the charge densities measured for the Fc(CH2)16SH, which is adsorbed in larger extent with respect to the Fc(CH2)6SH, even if the difference between the ferrocenylalkanethiol and the hydroxyl thiol is almost unchanged. (31) Camillone, N., III; Leung, T. Y. B.; Schwartz, P.; Eisenberger, P.; Scoles, G. Langmuir 1996, 12, 2737. (32) Poirier, G. E.; Pylant, E. D. Science 1996, 272, 1145. (33) Yamada, R.; Uosaki, K. Langmuir 1997, 13, 5218. (34) Tamada, K.; Hara, M.; Sasabe, H.; Knoll W. Langmuir 1997, 13, 1558.
Langmuir, Vol. 18, No. 4, 2002 1293
Conclusions In this paper the monolayer formation issue focusing on the chain length factor as a fundamental parameter of the adsorption process has been addressed. By means of cyclic voltammetry the amount of the electroactive component in mixed monolayers has been investigated and related to the solutions from which the layers are formed. From the contact angle no quantitative evaluation of the layer composition can be obtained, but, in combination with the electrochemical data, a general consistent picture of the adsorption process has been obtained. The use of a polar cocomponent in the layer formation does not affect the general trend for the longer chain component to be preferentially adsorbed indicating favorable thermodynamic contributions, related to enthalpy factors arising from van der Waals interactions between the alkyl chains. The preferential adsorption of the longer chain component is enhanced with even longer chain length. However, by using very low percentages (
