6392
J. Phys. Chem. C 2007, 111, 6392-6397
Electrical and Structural Characterization of Biphenylethanethiol SAMs B. Lu1 ssem,† L. Mu1 ller-Meskamp,† S. Kartha1 user,*,† M. Homberger,‡ U. Simon,‡ and R. Waser†,§ Institute for Solid State Research and CNI, Research Centre Ju¨lich GmbH, 52425 Ju¨lich, Germany, Institute of Inorganic Chemistry, RWTH Aachen, Landoltweg 1, 52074 Aachen, Germany, and Institut fu¨r Werkstoffe der Elektrotechnik 2, RWTH Aachen, Sommerfeldstrasse 24, 52074 Aachen, Germany ReceiVed: NoVember 10, 2006; In Final Form: February 28, 2007
The electron-transfer properties of a full-coverage self-assembled monolayer of biphenylethanethiol (CH3C6H4-C6H4-(CH2)2-SH, BP2) is studied by current versus distance spectroscopy. With this, the electrical properties of different functional groups in one molecule are vertically resolved. The current versus distance plot is shown to consist of three different parts, whose logarithmic slope is interpreted as the decay length of the vacuum, the biphenyl group, and the alkane chain. This finding is described in a trilayer tunnel junction model and the decay constant, βPh, of the tunneling current through the biphenyl group is estimated. Besides characterizing the biphenylalkanethiol monolayers electrically, the structure of these layers is determined by UHV-STM. Two new structures with rectangular (4 × 6x3) lattices are identified for 4′-methyl-1,1′-biphenyl4-ethanethiol self-assembled monolayers on (111)-oriented gold surfaces.
1. Introduction SAMs built of aromatic thiols display various properties, which make them interesting for future applications. For example, aromatic SAMs employed for lithography quasipolymerize and thereby form a negative resist, whereas aliphatic SAMs act as a positive resist under electron radiation (i.e., the SAM is destroyed).1 However, for molecular electronics it is of great importance that the tunneling conductance of an ordered self-assembled monolayer can be varied due to the electrontransfer properties of the molecular building blocks. For example, Bumm et al.2 showed that an aromatic molecule inserted into an alkanethiol SAM behaves like a molecular wire; that is, the embedded molecules show an increased conduction compared to the surrounding alkanethiol SAM. Self-assembled monolayers of different aromatic molecules like arenethiols,3 phenylthiols,4 biphenylthiols,5,6 or terphenylthiols7 have been studied. Tao et al.8 have shown that the order of a biphenylthiol SAM is greatly increased by the inclusion of a CH2 group between the sulfur and biphenyl groups. This observation triggered several studies concerning the structure of SAMs built by molecules of the homologous series of 4′-methyl-1,1′-biphenyl-4-alkanethiols (CH3-C6H4C6H4-(CH2)n-SH, BPn, n being the number of CH2 units included between the biphenyl and sulfur groups. This kind of SAMs was studied using IR spectroscopy and ellipsometry,8,9 near-edge X-ray adsorption fine structure (NEXAFS), Fourier transform absorption spectroscopy (FTIRRAS), X-ray photoelectron spectroscopy (XPS),10 high-resolution XPS,11 cyclic voltammetry12 and scanning tunneling microscopy (STM).11,13-15 SAMs of BPn can be divided into two classes: one class in which an even number of CH2 groups is included between the sulfur group and the biphenyl unit (BPn, n ) even) and another * Corresponding author. E-mail:
[email protected]. † Institute for Solid State Research and CNI, Research Centre Ju ¨ lich GmbH. ‡ Institute for Inorganic Chemistry, RWTH Aachen. § Institut fu ¨ r Werkstoffe der Elektrotechnik 2, RWTH Aachen.
class in which this number is odd (BPn, n ) odd). These two classes differ, for example, in the resistance to electron beam modification1 in the desorption potentials and in the resistance against replacement by alkanethiols.11,12 In all of these studies, BPn with n being even is shown to be more unstable. This oddeven effect is explained by the different molecular conformations of the SAMs, that is, different Au-S-C bonding angles and thus different orientations of the biphenyl group of odd- and even-numbered BPn. For an odd number of CH2 units, the adaption of the conformationally preferred C-S-Au angle of 104° 9 causes the C-phenyl bond and thus the biphenyl unit to be directed rather upright. By IR spectroscopy and NEXAFS, it is shown that the biphenyl unit in SAMs of BPn with n being odd is inclined only by 23° out of the surface normal. Furthermore, STM experiments showed that the molecule adopts a (2x3×3) lattice with intermolecular distances of 8.66 and 5 Å, which are in analogy to well-known alkanethiol lattices on (111)oriented gold surfaces.16,17 Therefore, for an odd number of CH2 units, the molecule can adopt a low-energy conformation, in terms of both sulfur bonding to the gold lattice and orientation of the biphenyl unit. For an even number of CH2 groups, due to the direction of the C-biphenyl bond, the tilt of the biphenyl group is significantly increased compared to odd-numbered BPn. To relax this unfavorable conformation of the biphenyl unit to some extent, the C-S-Au bonding angle is increased so that the alkane chain adopts a rather upright position and the tilt angle of the biphenyl group is reduced. The biphenyl unit is tilted by 45° out of the surface normal, and the C-S-Au bonding angle is increased to above 130°.9 Both the increased tilt of the biphenyl group, which increases the distance between neighboring biphenyl groups, and the increase in C-S-Au bonding angle are energetically costly. Consequently, energetically unfavorable structures and therefore less-stable SAMs result. However, because of the very complex interplay between the preferred orientation of the biphenyl unit, the preferred Au-
10.1021/jp067459l CCC: $37.00 © 2007 American Chemical Society Published on Web 04/11/2007
Biphenylethanethiol SAMs
Figure 1. STM image of a BP2 SAM (300 × 300 nm2,Vt ) 0.2 V, It ) 500 pA).
S-C bonding angle and the binding site, more than one surface structure has been found by STM for BPn, n ) even. The probably best-characterized structure is a (3 × 5x3) structure.14,15 The structure consists of eight molecules with an intermolecular distance of 8.66 Å along the 〈011〉 and 6.25 Å along the 〈211〉 direction. The distance of 6.25 Å implies that the molecules have to bind at different sites on the gold lattice14 and that the density of molecules on the gold surface is decreased by approximately 25% compared to odd-numbered BPn. All studies so far have focused on the structural characterization of SAMs of biphenylalkanethiols. However, because of the combination of aromatic and aliphatic groups in one molecule, SAMs built of BPn should show interesting electronic properties. Furthermore, because of the large tilt of the biphenyl group in the biphenylethanethiol self-assembled monolayer, this system is an ideal test system for current-transport measurements. Therefore, in this work the electronic properties of SAMs of BP2 are studied by current versus distance spectroscopy. 2. Experimental Details Gold films are prepared in a two-step deposition process described previously.18 Biphenylthiols are synthesized according to Rong et al.9 via Grignard C-C coupling reactions of the corresponding phenyl- or alkylbromides, and finally the conversion of the bromide into the thiol via thiourea. The gold films are kept in a 0.1 mM ethanolic solution of the molecules overnight. The films are rinsed thoroughly before they are transferred to the UHV-STM. Following deposition, the films are heated in vacuum (1 × 10-9 mbar) by an indirect resistive heating element at 64 °C for 1-2 h to desorb physisorbed molecules, before the SAMs are characterized by a UHV-STM (JEOL JSPM 4500, base pressure 1 × 10-9 mbar). 3. Structural Characterization Before characterizing the SAM electrically, the structure is determined by STM. In Figure 1, a 300 × 300 nm2 STM scan of BP2 is shown. The film is grown from solution at room temperature and annealed at 64 °C for 2 h in vacuum. The film shows gold vacancy islands, characteristic of SAMs
J. Phys. Chem. C, Vol. 111, No. 17, 2007 6393 based on gold/sulfur bonding. Between individual domains of BP2 molecules, the domain boundaries are separated by trenches. Two structures can be resolved in biphenylethanethiol SAMs grown at room temperature, shown in Figure 2. The first structure is shown in the STM scan in Figure 2a. It can be described by a quadratic lattice with a lattice constant of ∼0.6 nm (0.61 × 0.62 nm2). Assuming commensurability, this structure can be mapped to the gold lattice as a (4 × 6x3) structure as shown in Figure 2d. The structure consists of 10 molecules. Along the 〈211〉 direction, the molecules adopt different bonding sites and after five molecules the same bonding site as that for the first one is reached again. It should be noted that because of the variety of bonding sites, like already observed for other structures of BPn with n being even (see Table 1), this structure has only a very loose relationship to the gold lattice. The second structure is shown in Figure 2e and the cross sections in Figure 2f and g, respectively. It can be described by the same (4 × 6x3) unit cell as the previous structure (measured unit cell size: 1.27 × 3.08 nm2). The only difference lies in the position of the molecules in the center of the unit cell. Compared to the structure shown in Figure 2d, the molecules building the middle line in the unit cell move to the left (see Figure 2g), resulting in a centered structure. The structure shown in Figure 2a can be compared to the (3 × 5x3) structure described by Azzam et al.14,15 (see Table 1). Compared to this structure, the unit cell shown in Figure 2d is expanded along the 〈011〉 direction (from 0.86 to 1.2 nm) and the spacing of the molecules along the 〈211〉 direction is decreased slightly from 0.625 to 0.6 nm. This decreased spacing causes the length of the unit cell to increase to 6x3 compared to 5x3 because only after five molecules the same bonding site on the gold lattice as for the first molecule can be reached again. Although this difference in intermolecular spacing along the 〈211〉 direction is very small and probably lies within the error margin of the STM, the expansion along the 〈011〉 direction is too large to be explained by measurement artifacts. This small variations in the packing motif most likely stem from differences in the preparation procedures (see Table 1) while the observed polymorphism is in accordance with recently found polymorphism for BP4 and BP6.11 The deposition of BP2 at room temperature has resulted in structures that deviate slightly from the unit cell described in literature. By deposition of BP2 at higher temperatures, it is possible to give further explanations to this observation. In Figure 3 a SAM of BP2 that is deposited at 90 °C for 7 h from solution is shown. The film deviates in several characteristics from the room-temperature sample. Similar to annealing experiments of SAMs of alkanethiols,19 the number of vacancy islands decreases and their diameter increases. Furthermore, the molecules show crystalline packing and the area of the domains of the high-temperature sample is limited only by the terrace size of the (111) gold surface. These domains with explicitly increased size display clearly a long-range periodic height variation. The period and orientation of this long-range pattern varies slightly on the sample (3-7 nm). This observation is in accordance with the studies of Cyganik et al. on SAMs of BP2 deposited at 70 °C.13 They explained this pattern by a mismatch between the lattice constant favored by BP2 and the Au lattice. The stress due to this mismatch is relaxed by forming small regions in which the SAM is commensurate with the gold lattice. These small regions are separated by so-called solitons or domain walls in which the stress is released due to molecules
6394 J. Phys. Chem. C, Vol. 111, No. 17, 2007
Lu¨ssem et al.
Figure 2. High-resolution UHV-STM scans of a biphenylethanethiol SAM showing two structures that can be described by a (4 × 6x3) unit cell; (a) quadratic lattice (scan area 12 × 12 nm2); (b and c) cross sections along A and B; (d) schematic of the quadratic lattice; (e) centered lattice (scan area 14 × 14 nm2); (f and g) cross sections along lines C and D; (h) schematic of the centered lattice.
TABLE 1: Structures of Even-Numbered BPn SAMs deposition temperature
reference
BP2
(3 × 5x3) (4 × 6x3) quadratic (4 × 6x3) centered
structure
RT, 333-343 K 330 K 330 K
14,15 this work this work
BP4
(3 × 5x3) (6x3 × 2x3) (2x3 x x13)
RT, 333-343 K, 360 K 373-423 K 383 K
14,15 11 11
BP6
(3 × 5x3) (6x3 × 2x3)
RT, 343 K 373-423 K
14 11
adopting different adsorption sites, resulting in a low registry with the gold lattice. This model proposes regions of perfect registry with the gold lattice and regions in which the registry is broken. The mismatch between the SAM and the gold lattice is relaxed only in the domain walls. Similarly, one can also postulate that the stress due to the mismatch is not only relaxed in the domain walls but that the mismatch is also homogeneously distributed over the whole SAM. Such a model should be taken into account if
the chain-to-chain interaction between the biphenylethanethiol molecules would at least be on the order of the S-Au coupling. This is likely due to the strongly enlarged C-S-Au angle of about 130°, which destabilizes the S-Au bond, and small differences in adsorption energies among different adsorption configurations of alkanethiolates on (111) Au surfaces.20 Furthermore, a homogeneous change of adsorption sites is supported by the observation that the height pattern is smooth and continuous (see Figure 3) and does not show sharp transitions, as one would expect for the model, in which the misfit is relaxed only in the domain walls. Either of these explanations shows that BP2 exhibits a high structural misfit with the gold lattice. As already discussed in the introduction, BPn with n being even can only bind to the gold lattice by increasing the Au-S-C bond angle and the tilt of the biphenyl unit. Both of these deviations from the energetically lowest configuration generate stress in the SAM. Therefore, SAMs of BPn with n being even are less stable than those for n being odd and have only a low registry with the gold lattice.
Biphenylethanethiol SAMs
J. Phys. Chem. C, Vol. 111, No. 17, 2007 6395
Figure 3. STM scan of a BP2 SAM grown at 90 °C: (a) 100 × 100 nm2, Vt ) 1.5 V, It ) 100 pA; (b) 20 × 20 nm2; (c) cross sections along line A.
This low registry with the gold lattice of BP2 can explain the variety of possible structures. Consequently, the structure of BP2 on Au(111) is not as well defined as that for oddnumbered BPn. The BP2 structures reported in literature and observed in this work are summarized in Table 1. 4. Electrical Characterization of SAMs of BP2 After having examined the structure of biphenylethanethiol SAMs, the current transport through layers of BP2 is measured by current versus distance spectroscopy. For this kind of spectroscopy, the STM tip is positioned above the SAM. As for the current versus voltage spectroscopy, the distance between the tip and the SAM is controlled by the tunneling set-point. Once the feedback of the STM has adjusted the tunneling setpoint, it is turned off, the tip is moved toward the SAM, and the current is measured simultaneously. Figure 4 shows the result of an IS measurement on a BP2SAM, deposited for 16 h 30 from 0.1 mM solution. The distance axis is a relative axis; a tip displacement of 0 nm denotes the position of the tip at the tunneling set-point (2 V, 100 pA). Negative displacements represent an increase in tunneling distance, and positive displacements a decrease in tunneling distance. Figure 4 is a logarithmic plot of the current. Four domains can be identified: Below -0.1 nm, the tunneling current is lower than the noise level (∼1 pA) so that only noise is measured. Above -0.1 nm, all domains show a logarithmic dependency of the current I on the distance d and differ only in the slope of the function ln(I) ) f(d). Although below 0.07 nm the slope is steep, above 0.07 nm the slope of the plot decreases abruptly. The slope stays almost constant up to a distance of 0.75 nm and increases again beyond this point. The proportionality of ln(I) and d can be explained easily by the simple exponential law commonly used to describe the current/distance dependence of molecules21-23
I ) G exp(-βLz)V
(1)
Figure 4. Dependence of the tunneling current on the distance. The origin of the x axis corresponds to the tunneling set-point (Vt ) 2 V, It ) 100 pA).
with Lz being the length of the molecule. It follows for the logarithm of the current (d ) -Lz)
1n(I) ) 1n(GV) + βd
(2)
that is, the slope or derivative of the function ln(I) ) f(d) is a direct measure for the decay constant, β. However, we found in our STM measurement setup that the decay constant, β, is not constant over the entire length of the device. Three different parts of the device with different decay constants can be identified: the tunneling gap between the STM tip and the molecule (βVac), the biphenyl moiety (βPh), and the alkane group (βCH). Therefore, eq 1 can be refined:
I ) G exp(-(βVacdVac + βPhdPh + βCHdCH))V
(3)
It is tempting to identify the different slopes of the IS curve of Figure 4 with the decay constants βVac, βPh, and βCH of eq 3.
6396 J. Phys. Chem. C, Vol. 111, No. 17, 2007
Figure 5. Schematic illustration of biphenylethanethiol in a selfassembled monolayer.
For βVac, this interpretation is straightforward because around the tunneling set-point the tip moves through the vacuum gap and the slope of the function ln(I) ) f(d) is the decay constant of vacuum βVac. When the tip is pushed further toward the SAM, it touches the surface of the SAM at a certain point and the decay constant changes. Therefore, the first kink in Figure 4 at 0.07 nm can be interpreted as the point at which the STM tip touches the SAM surface. The methyl group bound to the biphenyl group cannot be detected by IS spectroscopy because of its small size. In analogy to the interpretation of the decay constant of vacuum, the slope of the function ln(I) ) f(d) in the range beyond the contact point is interpreted as the decay constant of the biphenyl group βPh. Pushing the STM tip further into the SAM after the contact point has been reached may perturbe the SAM and change the tunneling conditions to a certain extent. However, there are different observations that forebode that this perturbation is small. First of all, the slope of the logarithmic plot of the tunneling current, that is, the decay constant βPh, is 4.7 ( 0.8 nm-1. This experimental directly accessible value is right within the published decay constants for an unperturbed phenylen layer (∼4-6 nm-1 21,22,24,25). Furthermore, the IS plot as shown in Figure 4 can be acquired on the same position of the STM tip at least 10-20 times without changing the overall characteristic of the plot. The last hint, that the perturbation is small, is the value of βPh. This value stays constant over a distance of 0.605 ( 0.9 nm and rises again if the tip is pushed further into the layer. This distance is close to the thickness of the biphenyl group if the tilt of 45° is taken into account (0.608 nm), like that illustrated in Figure 5. Therefore, the additional rise in slope for distances above 0.75 nm can be interpreted as the tip pushing through the biphenyl layer and reaching the alkane spacer chain (Figure 5). The decay constant of the alkane group cannot be extracted because the current amplifier limits the tunneling current to 100 µA, a value that was reached above ∼0.9 nm. The slope of the function ln(I) ) f(d) can indeed be interpreted as a combination of the decay constants characteristic for the biphenyl group and the alkyl group of the biphenylethanethiol molecules. All of these observations point toward the conclusion that IS spectroscopy can Vertically probe the decay constants of the SAM, without significantly damaging the SAM. This interpretation is in accordance with the IS measurements of Yasutake et al.,26 who measured the current versus distance dependency of SAMs of alkanethiols. They also observed two slopes in the
Lu¨ssem et al. logarithmic plot of the tunneling current and interpreted the slope of the current at positions where the tip is penetrating the SAM as the decay constant of alkanes. By these measurements, they obtained a βCH of 11-12 nm-1, which is in close agreement with decay constants measured by Bumm et al.27 The electronic transport measurements performed in this work additionally give some input to the discussion about the current transport mechanism in organic molecules. According to ref 28, the current transport through self-assembled monolayers of organic molecules can in principle follow the bond overlaps along the molecular backbones (through bond ) TB) or may have a major component from electrode to electrode (through space ) TS). In the latter case, the molecule modifies the current transport through the gap. In the case of biphenylethanethiols, it is possible to differentiate between these two pathways by geometrical considerations. The measurement of dPh that we used to estimate the decay constant, βPh ) 4.7 ( 0.8 nm-1, was taken directly from the IS plot and thus along the vertical pathway of the STM tip toward the gold substrate (Figure 5). If there is a major component of the current following the molecular backbone, then the application of dBP , the length of the biphenyl group, should lead to the right βPh. Employing dBP ) 0.86 nm instead of dPh ) dBP × sin 45° results in a value of the decay constant of βPh ) 3.32 nm-1, which is out of the range of βPh values reported in literature.21,22,24 This result indicates that the current through the biphenyl group has a strong component from electrode to electrode predominately modified by the biphenylthiol layer, implying the current transport through interacting biphenyl groups, that is, π-π interactions of neighboring molecules. It should be noted that this result corresponds to a setup with a STM tip above a vacuum gap and a biphenylethanethiol layer on a (111) Au surface, which means a molecule only bonds covalently to one electrode. 5. Conclusions The structures and electrical properties of biphenylethanethiol SAMs are studied. It is shown that BP2 forms a 0.61 × 0.62 nm2 or 1.27 × 0.61 nm2 rectangular lattice on (111)oriented gold surfaces. If this structure is commensurate, then it can be described by a (4 × 6x3) lattice. It is shown that IS spectroscopy is a tool for measuring the decay constant of the biphenylethanethiol SAM vertically resolved. The IS curve is shown to be composed of three parts, each part with a different decay constant, β. These parts are interpreted as the vacuum barrier above the SAM, the biphenyl unit, and the alkane chain. By these measurements, the decay length, β, of the biphenyl unit is determined to be 4.7 ( 0.8 nm-1. Acknowledgment. The research described in this publication was supported by the Bundesministerium fu¨r Bildung and Forschung under Grant No. 13N8361 and by the Institute of Functional Materials for Information Technology (IFMIT). We thank Prof. K. Szot for his helpful discussions and H. Haselier for the gold substrates. Further, we thank J. Kiesgen and B. Hahn for the preparation of the compound. References and Notes (1) Zharnikov, M.; Grunze, M. J. Vac. Sci. Technol., B 2002, 20, 17931807. (2) Bumm, L. A.; Arnold, J. J.; Cygan, M. T.; Dunbar, T. D.; Burgin, T. P.; Jones, L., II; Allara, D. L.; Tour, J. M.; Weiss, P. S. Science 1996, 271, 1705-1707. (3) Yang, G.; Qian, Y.; Engtrakul, C.; Sita, L. R.; Liu, G.-Y. J. Phys. Chem. B 2000, 104, 9059-9062.
Biphenylethanethiol SAMs (4) Frey, S.; Stadler, V.; Heister, K.; Eck, W.; Zharnikov, M.; Grunze, M. Langmuir 2001, 17, 2408-2415. (5) Leung, T. Y. B.; Schwartz, P.; Scoles, G.; Schreiber, F.; Ulman, A. Surf. Sci. 2000, 458, 34-52. (6) Azzam, W.; Fuxen, C.; Birkner, A.; Rong, H. -T.; Buck, M.; Wo¨ll, C. H. Langmuir 2003, 19, 4958-4968. (7) Azzam, W.; Bashir, A.; Terfort, A.; Strunskus, T.; Wo¨ll, C. H. Langmuir 2006, 22, 3647-3655. (8) Tao, Y.-T.; Wu, C.-C.; Eu, J.-Y.; Lin, W.-L. Langmuir 1997, 13, 4018-4023. (9) Rong, H.-T.; Frey, S.; Yang, Y.-J.; Zharnikov, M.; Buck, M.; Wu¨hn, M.; Wo¨ll, C. H.; Helmchen, G. Langmuir 2001, 17, 1582-1593. (10) Zharnikov, M.; Frey, S.; Rong, H.; Yang, Y.-J.; Heister, K.; Buck, M.; Grunze, M. Phys. Chem. Chem. Phys. 2000, 2, 3359-3362. (11) Cyganik, P.; Buck, M.; Strunskus, T.; Shaporenko, A.; WiltonEly, J. D.; Zharnikow, M.; Wo¨ll, C. H. J. Am. Chem. Soc. 2006, 128, 13868-13878. (12) Felgenhauer, T.; Rong, H.-T.; Buck, M. J. Electroanal. Chem. 2003, 550-551, 309-319. (13) Cyganik, P.; Buck, M.; Wilton-Ely, J. D.; Wo¨ll, C. H. J. Phys. Chem. B 2005, 109, 10902-10908. (14) Cyganik, P.; Buck, M.; Azzam, W.; Wo¨ll, C. H. J. Phys. Chem. B 2004, 108, 4989-4996. (15) Azzam, W.; Cyganik, P.; Witte, G.; Buck, M.; Wo¨ll, C. H. Langmuir 2003, 19, 8262-8270.
J. Phys. Chem. C, Vol. 111, No. 17, 2007 6397 (16) Love, J. C.; Estroff, L. A.; Kriebel, J. K.; Nuzzo, R. G.; Whitesides, G. M. Chem. ReV. 2005, 105, 1103-1170. (17) Lu¨ssem, B.; Mu¨ller-Meskamp, L.; Kartha¨user, S.; Waser, R. Langmuir 2005, 21, 5256-5258. (18) Lu¨ssem, B.; Kartha¨user, S.; Haselier, H.; Waser, R. Appl. Surf. Sci. 2005, 249, 197-202. (19) Bumm, L. A.; Arnold, J. J.; Charles, L. F.; Dunbar, T. D.; Allara, D. L.; Weiss, P. S. J. Am. Chem. Soc. 1999, 121, 8017-8021. (20) Cao, Y.; Ge, Q.; Dyer, D.; Wang, L. J. Phys. Chem. B 2003, 107, 3803-3807. (21) Wakamatsu, S.; Fujii, S.; Akiba, U.; Fujihira, M. Ultramicroscopy 2003, 97, 19-26. (22) Wold, D. J.; Haag, R.; Rampi, M. A.; Frisbie, C. D. J. Phys. Chem. B 2002, 106, 2813-2816. (23) Wang, W.; Lee, T.; Reed, M. A. Phys. ReV. B 2003, 68, 035416. (24) Wakamatsu, S.; Fujii, S.; Akiba, U.; Fujihira, M. Jpn. J. Appl. Phys. 2006, 45, 2736-2742. (25) Lu¨ssem, B.; Mu¨ller-Meskamp, L.; Kartha¨user, S.; Waser, R.; Homberger, M.; Simon, U. Langmuir 2006, 22, 3021-3027. (26) Yasutake, Y.; Azuma, Y.; Nagano, K.; Majima, Y. Mater. Res. Soc. Symp. Proc. 2004, 782, 17-21. (27) Bumm, L. A.; Arnold, J. J.; Dunbar, T. D.; Allara, D. L.; Weiss, P. S. J. Phys. Chem. B 1999, 103, 8122-8127. (28) Salomon, A.; Cahen, D.; Lindsay, S.; Tomfohr, J.; Engelkes, V.; Frisbie, C. AdV. Mater. 2003, 22, 1881.
