Force Curve Measurements between n-Decanethiol Self-Assembled Monolayers in Inert Solvent and in Electrochemical Environment∗

It was revealed that electrostatic forces between electrodes under the applied bias are almost completely suppressed by an electrochemical potential control. The force operating between n-decanethiol self-assembled monolayers (SAMs) on gold was examined by atomic force microscopy in liquid n-dodecane and in an aqueous solution of HClO4. In inert n-dodecane, the bias voltage was simply applied between the tip and the sample, and the force curve exhibited a long-range attractive force in accordance with the simple capacitance model. Meanwhile under the independent control of the tip and sample potentials in HClO4, the force curve did not depend on the applied potentials due to the small double-layer capacitance. [DOI: 10.1380/ejssnt.2009.731]


I. INTRODUCTION
Self-assembled monolayers (SAMs) of n-alkanethiols (HS(CH 2 ) n CH 3 ) on gold provide a promising strategy for constructing stable, well-defined monolayers on electrode surfaces [1].SAMs are of crystalline order as long as the alkyl chain is sufficiently long (n ≥ 9).These layers strongly block the electrochemical oxidation of the underlying gold and also electron transfer with redox couples in solution [2].Many groups have attempted to verify basic assumptions about the applicability of fundamental electrical double-layer models in electrochemistry.The differential capacitance of SAMs can be determined by conventional electrochemical methods, such as cyclic voltammetry and impedance spectroscopy, and compared with that obtained theoretically [2].
It is well known that atomic force microscopy (AFM) can be used to study the force between the tip and the sample as well as surface topography [3].In a force measurement, the force is monitored as the tip approaches and contacts the sample, and is then withdrawn from the sample.This method, the so-called force curve measurement, has also been applied to potential-controlled SAMs [3][4][5][6][7].In general, an insulating silicon nitride or silicon tip, where the tip surface is terminated by a silanol group, is used without potential control.Bard and colleagues have systematically studied the diffuse double layer of SAMs as a function of electrode potential [8][9][10][11].These direct force measurements have provided a significant insight into double-layer properties, such as surface potential and charge [3].
Although the double-layer properties are now reasonably well understood in the case of the above-mentioned systems, the force measurement between two potentialcontrolled electrodes has been very limited [12][13][14][15][16].Such knowledge is valuable for analyzing the image of electrochemical scanning tunneling microscopy (EC-STM).It has been known that EC-STM is one of the most important techniques for studying the electrode surface on an atomic scale [17,18].In this technique, the tunneling current is monitored by applying different tip and sample potentials (E tip = E sample ), where the potential difference (E tip − E sample ) defines the bias voltage.Unlike the typical STM measurements in vacuum, air, and inert solvents, the potential profile created by mixing the tip and sample double layers might change the sample states during EC-STM imaging [19].Recently, the electron transfer between potential-controlled electrodes through electroactive molecules has attracted considerable attention in the field of molecular electronics, because electron transfer rates can be controlled by simply applying appropriate potentials (electrochemical gate effect) [20][21][22][23][24].The knowledge of the potential profile may help to understand the electron transfer mechanism and design more sophisticated systems.
In this report, we present force curve measurements between n-decanethiol (HS(CH 2 ) 9 CH 3 ; C 10 H 21 SH) SAMs in n-dodecane and in a 0.05 M HClO 4 aqueous solution under potential control (Fig. 1).While in the case of n-dodecane, a bias voltage was simply applied between the tip and the sample, as in typical STM measurements (Fig. 1(a)), in the case of an aqueous HClO 4 solution, the tip and sample potentials were independently controlled using a bipotentiostat, as in EC-STM measurements (Fig. 1(b)).The potential and/or bias dependences of the forces are interpreted with respect to the potential profile between the tip and the sample.

II. EXPERIMENTAL
An electrolyte solution was prepared using ultrapuregrade HClO 4 (Cica-Merck) and Milli-Q water (Nihon Millipore).n-decanethiol was purchased from Kanto Chemical.All other chemicals were of reagent grade or better and used without further purification.A gold-coated silicon nitride AFM tip (OMCL-TR800PB, Olympus) was cleaned by following the procedure reported by Fujihira et al. [25].Briefly, the tip was exposed to UV light in air for 30 min, and then immersed in ethanol for 30 min at 70 • C.An evaporated gold film on mica with a (111)oriented surface was annealed with a butane flame and used as the sample substrate [26,27].The tip and sample substrates were immersed in a 1.0 mM ethanol solution of C 10 H 21 SH overnight, rinsed, and dried with N 2 gas.Freshly prepared tip and sample were used for each set of experiments.
Force curve measurements were carried out using a commercial AFM apparatus (MFP-3D, Asylum Research) equipped with a homemade PTFE cell.To avoid the damage of the tip surface, the applied force during tip sample contact was kept to less than 500 pN.A bipotentiostat (HR-101B, Hokuto Denko) was used to control the tip and sample potentials independently.AuO x and Pt wire electrodes were used as reference and counter electrodes, respectively [26,27].All electrochemical potentials are presented with respect to the Au/AuO x reference electrode.For each AFM tip, its spring constant was calibrated by the thermal noise method reported by Hutter and Bechhoefer [28].The potential dependence of the spring constant was negligible in the potential range used in this study.Cyclic voltammetry was performed using a potentiostat (BAS 100B, Bioanalytical Systems).

III. RESULTS AND DISCUSSION
Figure 2(a) shows typical approach force curves between C 10 H 21 SH SAMs in n-dodecane.The bias voltage applied to the tip was changed from 0 to 1.0 V. Zero force was selected as the force when the tip was far away from the sample surface (> 300 nm), which can be considered to be constant for different biases since interactions become negligible at such a distance.The force was not zero in a long range (> 50 nm).The force depends on the tip sample separation and applied bias.Because long-range forces decreased with applied bias, the electrostatic nature of these forces was implicated.According to Tivanski et al., the electrostatic force, the capacitive force between the tip and the sample, can be approximated as where ε 0 and ε represent the permittivity of free space and the effective dielectric constant of the SAM solvent system, respectively, V is the applied bias voltage, R is the effective tip radius, and d is the tip sample separation [29].This equation is derived from the assumption that the tip shape can be modeled as a truncated cone ended by a spherical apex.ously indicated by Thomas et al., we consider that small differences in surface packing and coverage between the tip and the sample contribute to the contact potential difference [30].It is known that the alkanethiols adsorbed onto gold surfaces possess dipole moment pointing to their methyl group (from negative to positive) [31].Asadi et al. have reported that the lower the coverage of alkanethiol SAMs, the higher the work function [32].Thus, the positive contact potential difference observed in this study indicates the lower coverage of the tip surface, as expected from the large curvature of the underlying gold.Considering the contact potential difference V c , we modify Eq. ( 1) as The solid lines in Fig. 2(a) represent fits of the force curves obtained using Eq. ( 2), where ε = 2.5 [29] and R = 30 nm.The experimental data can be well reproduced by the capacitive force theory, except for very short tip sample separations at which nonelectrostatic contributions exist.Hence, the potential profile between the tip and the sample in an inert solvent can be described using a simple capacitance model.We note that the retraction force curves were not reproducible, irrespective of the applied bias voltage, presumably because a small amount of water dissolved in n-dodecane accumulated near the tip and sample contact.For a long-distance region, Eq. ( 2) represents the force curve and is not affected by the presence of a small amount of water.
To examine the force between the potential-controlled tip and sample in an aqueous solution, we first performed electrochemical measurements.Figure 3 shows typical cyclic voltammograms of C 10 H 21 SH SAM on Au(111) in a 0.05 M HClO 4 solution.The current was proportional to the scan rate and was almost potential-independent, characteristic behavior of thin layers with a low dielectric constant [2].The calculated capacitance (∼2 µF/cm 2 ) is in agreement with the previously reported value [2,33,34].To prevent the reductive and oxidative desorptions of C 10 H 21 SH molecules, the potentials applied to the tip and the sample were restricted between −0.8 and −0.4 V vs Au/AuO x (vertical dotted lines) [24,26,27] AFM tip showed similar features, but the plateau current was not well defined, owing to the undefined electrode area.
The typical force curve between C 10 H 21 SH SAMs in the 0.05 M HClO 4 solution is shown in Fig. 4. The tip and sample potentials were set at −0.8 and −0.4 V vs Au/AuO x , respectively.The sample-to-tip distance was first decreased (red) and then increased from the contact region (blue).The hysteresis in the force curve (i.e., approach versus retraction) corresponds to the adhesion between the tip and sample surfaces.The adhesion force required to separate the tip from the sample was 8±1 nN and irrespective of the tip and sample potentials.This indicates that the electrostatic force does not contribute to the adhesion force.It is known that the Johnson Kendall Roberts (JKR) theory of adhesion mechanics adequately describes adhesion forces measured between a SAM-modified tip and a SAM-modified sample [35][36][37][38][39][40].The JKR model predicts that the adhesion force F ad is given by where W ad is the work of adhesion to pull the tip off the sample in medium, and γ is the interfacial energy [41].By using the literature value of 52 mJ/m 2 for γ [42], the tip radius is calculated to be ∼15 nm.This value is onehalf that obtained from the capacitance model in the inert solvent.Possible reasons for this discrepancy are (1) the lower γ value of the tip, as indicated by the lower coverage in the previous section, and/or (2) difference in effective tip radius due to different interaction ranges [43].
Figure 5 shows approach force curves between C 10 H 21 SH SAMs taken in the 0.05 M HClO 4 solution at various tip and sample potentials.Each curve represents an average of 10 consecutive measurements.In contrast to Fig. 2(a), it is obvious that the applied potentials do not affect the force between the tip and the sample, and the force curves show no long-range force.Although it is known that, in some cases, the force measurements between hydrophobic SAMs show a long-range (> 30 nm) attractive force due to preexisting nanobubbles [44,45], we never observed such a force.The atomically flat surface used in this study might have prevented the dissolved air from forming the nanobubbles.The attractive force was apparently observed at about 7 nm.Now, we consider the potential profile in an electrochemical environment.Figure 6 shows schematics of the potential profile between the tip and the sample (a) in an inert solvent and (b) in an electrochemical environment.In the inert solvent, a simple capacitance is created between the tip and the sample, as discussed above.On the other hand, the double-layer capacitances between the SAM-modified electrodes and the electrolyte solution are created in the electrochemical environment.The potential of zero charge (pzc) of n-alkanethiol SAMs is known to be about −1.5 V vs Au/AuO x [46,47].Thus, both the tip and sample electrodes are positively charged, as schematically shown in Fig. 6(b).
Although potential-dependent repulsive forces between the electrodes charged with the same sign are expected, we observed potential-independent net attractive forces.We attribute the absence of the net repulsive forces to (1) the small surface charge density due to the small capacitance (∼2 µF/cm 2 ) and (2) the relatively small Debye length (the screening length of charge; 1.4 nm for 0.05 M HClO 4 ).The surface charge density (0.09-0.14 e/nm 2 ) calculated from the capacitance is much less than the surface density of C 10 H 21 SH SAMs (4.6 molecules/nm 2 ).Preliminary force measurements between positively charged NH 3+ -terminated SAMs (∼3 molecules/nm 2 [48]) under the same experimental conditions showed the net repulsive forces.This indicates that the small surface charge density is one of the sources of the net attractive forces observed in this study.The origin of the attractive forces is the hydrophobic force as well as van der Waals force, which is beyond the scope of this study.Finally, we discuss previous experiments reported by Kwon and Gewirth.They performed force measurements between potential-controlled n-alkanethiol SAMs in an alkaline solution (0.1 M NaOH) and studied the reductive desorption of SAMs [16].They found a potentialindependent repulsive force between the n-alkanethiol SAMs, irrespective of alkyl chain length (C 2 H 5 SH and C 16 H 33 SH).The magnitude of the repulsive force reached at least about 100 pN (corrected for our tip radius), which was sufficiently above our detection limit.As mentioned in their report, the origin of the repulsive force in the alkaline solution is unknown at this stage, and the potentialindependent attractive force obtained in our study is more http://www.sssj.org/ejssnt(J-Stage: http://www.jstage.jst.go.jp/browse/ejssnt/) e-Journal of Surface Science and Nanotechnology Volume 7 (2009) easily interpreted.Systematic studies are required to reveal the pH dependence of the force.

IV. CONCLUSIONS
We compared the force curves between C 10 H 21 SH SAMs in n-dodecane and in 0.05 M HClO 4 under potential control.In 0.05 M HClO 4 , we found that the electrostatic force between the tip and the sample was apparently not observed due to the small capacitance and small Debye length used in this study.We consider that this result is an important starting point for more complex systems, such as SAMs containing charged functional groups and electroactive groups, because the effect of capacitive charging can be neglected when the thiol containing a long alkyl chain is used.On the other hand, the approach force curves taken in n-dodecane showed bias-dependent capacitive forces, demonstrating that the potential profiles in the inert solvent and electrochemical environment are different.Although it has been known that STM and EC-STM measurements of C 10 H 21 SH SAM give similar images [49,50], care must be taken to interpret the electron tunneling mechanisms due to the different potential profiles.Further experiments are required to reveal the potential profiles created at typical tunneling gaps.

FIG. 1 :
FIG. 1: Schematics of force curve measurements (a) in an inert solvent and (b) in an electrochemical environment.WE, RE, and CE indicate working, reference, and counter electrodes, respectively.

4 FIG. 2 :
FIG. 2: (a) Approach force curves between C10H21SH SAMs taken in n-dodecane.Different colors indicate different bias voltages applied to the tip.Black lines indicate the calculated electrostatic forces between the tip and the sample.(b) Force versus tip bias plots at tip sample separations indicated by vertical dotted lines in (a).

Figure 2 ( 3 FIG. 3 :
FIG. 3: Cyclic voltammograms of C10H21SH SAM on Au(111) taken in a 0.05 M HClO4 solution.Vertical dotted lines indicate the tip and sample potentials at which force curve measurements were performed.

FIG. 5 :FIG. 6 :
FIG. 5: Approach force curves between C10H21SH SAMs taken in a 0.05 M HClO4 solution at various tip and sample potentials.Each curve represents an average of 10 consecutive measurements.