pubs.acs.org/NanoLett
The Impact of E-Z Photo-Isomerization on Single Molecular Conductance Santiago Martin,*,†,‡ Wolfgang Haiss,† Simon J. Higgins,† and Richard J. Nichols*,† †
Centre for Nanoscale Science and Chemistry Department, University of Liverpool, Liverpool, Crown Street, L69 7ZD, United Kingdom, and ‡ Department of Organic and Physical Chemistry, University of Zaragoza, Campus Universitario, 50009, Zaragoza, Spain ABSTRACT The single molecule conductance of the E and Z isomers of 4,4′-(ethene-1,2-diyl)dibenzoic acid has been determined using two scanning tunneling microscopy (STM) methods for forming molecular break junctions [the I(s) (I ) current and s is distance) method and the in situ break junction technique]. Isomerization leads to significant changes in the electrical conductance of these molecules, with the Z isomer exhibiting a higher conductance than the E isomer. Isomerization is achieved directly on the gold surface through photoirradiation, and the STM is used to determine conductance before and after irradiation; reversible switching between the two isomers could be achieved through irradiation of the surface bound species at different wavelengths. In addition, three groups of molecular conductance values [A (“low”), B (“medium”), and C (“high”)] have been measured for these carboxylate-terminated molecules. The origin of these conductance groups as well as the increase of the conductance for the Z isomer have been analyzed by comparing the length of the molecules extended in the gap, derived from molecular modeling, with the experimentally observed break-off distance for both isomers. KEYWORDS Single molecule conductance, molecular electronics, STM, break junction, nanoelectronics, photoisomerization
I
n recent years the electrical conductance of a wide variety of compounds has been investigated down to the single molecule level using scanning tunneling microscopy (STM),1–3 conducting atomic force microscopy (AFM),4,5 and mechanically controllable break junctions.6–8 Using these methods, the relationship between molecular structure and electrical behavior of metal|molecule|metal junctions has been explored, and new paradigms for achieving switching in single molecule junctions have emerged. Switching of the electrical properties of single molecules has been achieved through imposition of an external stimulus, for instance, electrochemical potential,9 optical irradiation through an external gate electrode,10 or a covalent chemical modification.11 Optical switching has also been demonstrated using light-induced structural changes,12–15 in experiments employing a STM or a conducting AFM. Mativetsky et al.12 demonstrated that nanoscale domains of azobenzene derivatives could be photoisomerized between a lowconductance trans state and a high-conductance cis state. Kumar et al.13 also demonstrated photoisomerization of azobenzene derivatives by measuring changes in the STMdetermined apparent height of these molecules, isolated in alkanethiol self-assembled monolayers on gold. He et al.,14 on the other hand, demonstrated through single molecule measurements that photoinduced ring closure leads to increased conductance of dithienylethene chromophores. However, to date there has been no demonstration that * Corresponding authors. E-mail: liverpool.ac.uk (R.J.N.). Received for review: 12/23/2009 Published on Web: 05/25/2010
[email protected] © 2010 American Chemical Society
(S.M.),
simple photoinduced E-Z isomerization can be used to control the conductance of metal|molecule|metal junctions at the single molecule level. Here we show through single molecule conductance determination that such photoisomerization induced structural transitions result in a significant and reversible conductance change for molecular junctions. Photoisomerization is achieved directly on the gold surface, with STM single molecule conductance being determined before and after irradiation; photoisomerization can be reversibly achieved in either direction by choosing suitable wavelengths. STM-based methods have been used in air to determine the conductance of Au|molecule|Au junctions involving both E and Z isomers of 4,4′-(ethene-1,2-diyl)dibenzoic acid, and we show that these forms can be photochemically interconverted under irradiation. In addition, it is shown for the first time that molecules terminated with carboxylate end groups exhibit three different conductance groups, and we relate this to differing configurations of the end groups at the contact. The structures of the isomers are shown in Figure 1, and photoisomerization from the E to the Z form can be achieved by UV irradiation or vice versa upon visible irradiation. Driscoll et al. have shown by UV-vis spectroscopy that the compounds in Figure 1 can be reversibly photocycled be-
nichols@
FIGURE 1. Structure of the E and Z isomers of 4,4′-(ethene-1,2diyl)dibenzoic acid used to carry out the photoisomerization study. 2019
DOI: 10.1021/nl9042455 | Nano Lett. 2010, 10, 2019–2023
technique or the in situ BJ technique. Clear peaks at (0.22 ( 0.02) × 10-4 G0 (A), (0.70 ( 0.14) × 10-4 G0 (B), and (2.75 ( 0.50) × 10-4 G0 (C) were observed in the histograms, shown in Figure 2a, corresponding to the conductance of the E isomer, since these values are consistent with those presented in Table 1. The photoisomerization of the E isomer was then performed by UV irradiation25 as follows: once the single molecule conductance for the E isomer was determined, the STM tip was retracted, and the sample was irradiated with UV light (in situ) at 250-320 nm for at least 20 min. Then the tip was reapproached to the sample and new I(s) scans were recorded. After this irradiation, peaks at (0.32 ( 0.03) × 10-4 G0 (A), (1.27 ( 0.26) × 10-4 G0 (B), and (5.23 ( 0.46) × 10-4 G0 (C) were observed in the conduction histogram (presented in Figure 2b). All of these conductance values are consistent with those measured for the Z isomer (compare with Table 1). This confirms that E to Z isomerization occurred upon UV irradiation of the surface. Note here that when the irradiation time was less than 10 min, the histogram showed peaks corresponding to the conductance of both the E and Z isomers, according to the Table 1 (see Supporting Information). After these measurements, the sample was irradiated with visible light for 20 min following the same procedure as that mentioned above. Conductance histograms typical of the E isomer were seen for measurements made after this irradiation, consistent with photoisomerization back to the E isomer (see Supporting Information). The higher conductance of the Z isomer could be due either to a differing electronic structure (or orbital distribution) of the molecular bridge or to the Z isomer taking up a different configuration to its E counterpart; we refer to these two broad factors as “bridge electronic structure” and “junction configuration”, respectively, although we recognize in many instances that they may be intimately coupled. To elucidate the “bridge electronic structure” we have performed rudimentary density functional theory (DFT) calculations of both isomers connected to gold atoms through the carboxylate head groups. These indicate that there is no change in the energetic position of the first spanning molecular orbital (HOMO), which lies closest in energy to the Au Fermi energy upon isomerization from E to Z (see Supporting Information for details). Thus we move on to an analysis of the junction configuration on the molecular conductance. Schematic illustrations of the E and Z form placed between gold contacts are shown in Figure 3a. A tip-sample separation close to the length of the E isomer, as shown in Figure 3a, would imply that both carboxylate end groups can adsorb to the contacts with little direct interaction of the phenyl rings with the gold contacts. The geometry of the Z isomer in the junction, on the other hand, may result in a stronger interaction between the phenyl ring(s) and the gold contact(s). In this respect, we note that close proximity of gold contacts to phenyl rings has been previously shown to cause significant increases in junction
TABLE 1. Single Molecule Conductance Data for E and Z Isomersa conductance (× 10-4 G0) isomer
A (“low”)
E Z
0.18 ( 0.04 0.34 ( 0.09
B (“medium”) 0.69 ( 0.20 1.33 ( 0.24
C (“high”) 2.83 ( 0.31 5.37 ( 0.60
a Data for the three conductance groups (A, B, and C) are tabulated.
tween the E and Z isomers with high conversions.16 Dicarboxylic acids have been shown previously to form Au|molecule|Au junctions.17,18 On the basis of known Au/ RCOOH chemistry, it is likely that carboxylates are deprotonated on adsorption to Au. Two different methods for forming molecular break junctions were deployed; the I(s), I ) current and s is distance,9 and the in situ break junction (BJ)1 techniques. Both methods use an STM to form molecular junctions, but they differ in the way junctions are formed. In the I(s) technique, contact between the gold STM tip and the surface is avoided. In contrast, in the BJ technique, the tip is driven a certain distance into the substrate. The tip is then retracted until the metal contact cleaves, and molecular bridges can then form within this junction. These bridges then break upon further retraction of the tip, and molecular conductance can be determined through statistical analysis of many currents vs distance traces. Before carrying out the photoisomerization study, the single molecule conductance of both the E and Z isomers were individually determined (see Supporting Information for details). It has been previously shown for other compounds, such as alkanes,17,19–21 oligo(phenyleneethynylene)s [(-C6H4-CtC-)n, OPE],22 oligoynes [i.e., Py-(CtC)n-Py; Py ) 4-pyridyl]23 or bipyridines24 (i.e., with thiols, amines, or pyridines as functional groups to establish the contact between the molecule and the metals) that there is not one unique value for the single molecule conductance, with multiple single molecule conductance values being evident. Three groups of peaks, which have been attributed to differing contact morphologies between head groups and gold contacts,19,21 have been observed. Here we show that, for carboxylate-terminated molecules, three groups of molecular conductance values are also observed. The corresponding conduction values listed in Table 1 (A (“low”), B (“medium”), and C (“high”)) clearly demonstrate that the Z isomer exhibits a higher conductance than the E isomer for all three conductance groups. These conductance groups, as will be shown later in the text, can be assigned to the adsorption of the contacting carboxylate at either terrace or step sites, similar to other contact groups.21,23 Once we had determined the single molecule conductances for both pure compounds (Table 1), we carried out the photoisomerization study. A Au(111) substrate was immersed in a solution of the E isomer, and the single molecule conductance was determined using either the I(s) © 2010 American Chemical Society
2020
DOI: 10.1021/nl9042455 | Nano Lett. 2010, 10, 2019-–2023
FIGURE 2. Typical conductance traces together with the corresponding histograms for the E (a) and Z (b) isomers using the I(s) and the BJ techniques; conductance traces are shifted horizontally for clarity, and conductance histograms are built by adding together all the points of 215 conductance traces that showed discernible plateaus recorded before (a) and after (b) UV irradiation. Conductance data are presented in units of the conductance quantum G0 ) 2e2/h ) 77.4 µS. Utip ) 0.6 V. Solid lines are Gaussian fits. For histogram construction, the highconductance data were obtained by the BJ technique, and low-conductance data by the I(s) technique (see Supporting Information). The dashed line shows an exponential decay curve in the absence of molecular wire formation.
observed sbreak-off values are close to the values computed by molecular modeling for the E isomer completely extended between gold contacts. For the Z isomer, the sbreak-off values are slightly larger than the molecular modeling prediction. This poorer agreement of the break-off length for the Z isomer may be related to phenyl-ring rotation and interaction with the contacts during the junction stretching and eventual breaking process. However, Figure 3b clearly shows that the Z isomer breaks off at shorter tip-sample separations than the E isomer, consistent with the E isomer being the more extended form. Finally, we turn to the origin of the differing conduction groups, A, B, and C, observed for these molecules. We have analyzed average break-off distances for A and B events (end of the plateaus, marked by a rapid drop in conductance, observed in the conductance traces shown in Figure 2) and have added the initial tip-sample distance at the beginning of the I(s) retraction scan. The details of this break-off distance analysis are given in the Supporting Information. Figure 4 shows a two-dimensional histogram constructed from all traces with the break-off distance and a conductance for the E isomer being mapped. Two regions with a large number of counts, encircled by the purple (B events) and red (A events) dash lines, are visible. The measured difference between the centroid of both regions is 0.23 nm. A similar phenomenon is observed for the Z isomer (as marked
FIGURE 3. (a) Molecular structure of both isomers in the junction and (b) experimentally determined average break-off distance for group A (green, upper line) and group B (blue, lower line) events for E and Z isomers. The results of a molecular modeling calculation are shown as green (A events) and blue (B events) squares.
conductance by increasing transmission at the contacts.26 In this study it was shown that tilting of an aromatic molecular bridge brought phenyl groups into more intimate proximity to the gold contacts, resulting in large increases in junction conductance.26 In an analogous manner, we propose that rotation of the phenyl group in the Z isomer would also bring the ring into even closer proximity to the surface and would rationalize the larger conductance for the Z isomer, compared to the E isomer. The gap separation at which the molecular junction is cleaved can be estimated by careful calibration of the tipto-substrate distance (see Supporting Information).26 This experimentally measured break-off distance (sbreak-off) is compared to the length of the molecule in Figure 3b. The © 2010 American Chemical Society
2021
DOI: 10.1021/nl9042455 | Nano Lett. 2010, 10, 2019-–2023
been analyzed by comparing the length of the molecules extended in the gap derived from molecular modeling with the experimentally observed break-off distance for both isomers. Rotation of the phenyl groups in the Z form would be expected to bring the phenyl rings into close proximity with the gold contacts, which could offer an explanation for the increased transmission at the contacts. Acknowledgment. This work was supported by Engineering and Physical Sciences Research Council (EPSRC) under grant EP/C00678X/1 (Mechanisms of Single Molecule Conductance). S.M. acknowledges a postdoctoral fellowship and his Juan de la Cierva contract from the Ministerio de Ciencia e Innovacio´n of Spain.
FIGURE 4. Two-dimensional histogram constructed from 215 traces with a clear break-off distance for the E isomer. Two regions with a large number of counts are encircled by the purple (B events) and red (A events) dash lines. The high-conductance region (B events), around 6.5 × 10-5 G0, extends from 1.2 to 1.4 nm, and the lowconductance region (A events), around 1.75 × 10-5 G0, extends from 1.4 to 1.6 nm. Insets, model proposed to explain the origin for the differing conduction groups for both isomers. For the case of Group B events, one carboxylate group is adsorbed at a step or similar high coordination site, whereas Group A events involve adsorption of both contacting carboxylate groups on terraces. Each color refers to a count of conductance values versus tip-sample distance. The count rate increases in the sequence white-light blue-blue-orangeyellow. The color mapping is achieved by taking points every 0.01 nm (see Supporting Information for further details).
Supporting Information Available. UV-vis spectra, details of molecular adsorption, single molecule conductance for the E and Z isomer, details of the calibration of tip-substrate distance, reversible photoisomerization process, DFT calculations, references. This material is available free of charge via the Internet at http://pubs.acs.org. REFERENCES AND NOTES (1) (2) (3)
on the graph in Figure 3b). This separation is close to the height of a monatomic step on the Au(111) surface (0.235 nm), and such a separation between A and B events has been observed for other compounds, such as alkanedithiols21 and 4-pyridyl-terminated oligoynes.23 Given this spatial separation, close to one gold step height difference between group B and A, it is concluded that they are related by a structural transition involving a step height (or similar surface defect). In the model shown in the inset of Figure 4, for group B, one of the oxygen atoms (or both) of the carboxylate group is adsorbed at the base of a gold step edge. Conduction group C could then be explained by assuming that both carboxylate groups are adsorbed at step sites. It has been previously shown for molecules with thiol head groups that, as the junction is stretched during retraction of the STM tip, group B sites can be transformed into group A by pulling the head group up by one gold step height.21 This was also observed here for the case of the carboxylate head groups. In conclusion, we have demonstrated at the single molecule level that photoisomerization of E- and Z-4,4′-(ethene1,2-diyl)dibenzoic acid on a gold surface is a reversible process and is accompanied by significant change of the electrical conductance of these molecules. It has also been shown that, for the case of carboxylate head groups, three fundamental groups of conductance values can be observed [A (“low”), B (“medium”), and C (“high”)], similar to previous observations for thiol linking groups. The origin of these conductance groups as well as the increase of the conductance for the Z isomer, as compared to the E isomer, have © 2010 American Chemical Society
(4) (5) (6) (7) (8)
(9) (10) (11) (12) (13) (14) (15) (16) (17) (18)
2022
Xu, B. Q.; Tao, N. J. J. Science 2003, 301 (5637), 1221–1223. Haiss, W.; Nichols, R. J.; van Zalinge, H.; Higgins, S. J.; Bethell, D.; Schiffrin, D. J. Phys. Chem. Chem. Phys. 2004, 6 (17), 4330– 4337. Quek, S. Y.; Venkataraman, L.; Choi, H. J.; Loule, S. G.; Hybertsen, M. S.; Neaton, J. B. Nano Lett. 2007, 7, 3477–3482. Choi, S. H.; Kim, B.; Frisbie, C. D. Science 2008, 320 (5882), 1482– 1486. Cui, X. D.; Primak, A.; Zarate, X.; Tomfohr, J.; Sankey, O. F.; Moore, A. L.; Moore, T. A.; Gust, D.; Harris, G.; Lindsay, S. M. Science 2001, 294 (5542), 571–574. Reed, M. A.; Zhou, C.; Muller, C. J.; Burgin, T. P.; Tour, J. M. Science 1997, 278 (5336), 252–254. Weber, H. B.; Reichert, J.; Weigend, F.; Ochs, R.; Beckmann, D.; Mayor, M.; Ahlrichs, R.; von Lohneysen, H. Chem. Phys. 2002, 281 (2-3), 113–125. Huber, R.; Gonzalez, M. T.; Wu, S.; Langer, M.; Grunder, S.; Horhoiu, V.; Mayor, M.; Bryce, M. R.; Wang, C. S.; Jitchati, R.; Schonenberger, C.; Calame, M. J. Am. Chem. Soc. 2008, 130 (3), 1080–1084. Haiss, W.; van Zalinge, H.; Higgins, S. J.; Bethell, D.; Hobenreich, H.; Schiffrin, D. J.; Nichols, R. J. J. Am. Chem. Soc. 2003, 125 (50), 15294–15295. Kubatkin, S.; Danilov, A.; Hjort, M.; Cornil, J.; Bredas, J. L.; StuhrHansen, N.; Hedegard, P.; Bjornholm, T. Nature 2003, 425 (6959), 698–701. Jackel, F.; Watson, M. D.; Mullen, K.; Rabe, J. P. Phys. Rev. Lett. 2004, 92 (18), 188303. Mativetsky, J. M.; Pace, G.; Elbing, M.; Rampi, M. A.; Mayor, M.; Samori, P. J. Am. Chem. Soc. 2008, 130 (29), 9192–9193. Kumar, A. S.; Ye, T.; Takami, T.; Yu, B. C.; Flatt, A. K.; Tour, J. M.; Weiss, P. S. Nano Lett. 2008, 8 (6), 1644–1648. He, J.; Chen, F.; Liddell, P. A.; Andreasson, J.; Straight, S. D.; Gust, D.; Moore, T. A.; Moore, A. L.; Li, J.; Sankey, O. F.; Lindsay, S. M. Nanotechnology 2005, 16 (6), 695–702. Pace, G.; Ferri, V.; Grave, C.; Elbing, M.; von Hanisch, C.; Zharnikov, M.; Mayor, M.; Rampi, M. A.; Samori, P. Proc. Nat. Acad. Sci. U.S.A. 2007, 104 (24), 9937–9942. Driscoll, P. F.; Purohit, N.; Wanichacheva, N.; Lambert, C. R.; McGimpsey, W. G. Langmuir 2007, 23 (26), 13181–13187. Chen, F.; Li, X. L.; Hihath, J.; Huang, Z. F.; Tao, N. J. J. Am. Chem. Soc. 2006, 128 (49), 15874–15881. Martin, S.; Haiss, W.; Higgins, S.; Cea, P.; Lopez, M. C.; Nichols, R. J. J. Phys. Chem. C 2008, 112 (10), 3941–3948. DOI: 10.1021/nl9042455 | Nano Lett. 2010, 10, 2019-–2023
(19) Li, C.; Pobelov, I.; Wandlowski, T.; Bagrets, A.; Arnold, A.; Evers, F. J. Am. Chem. Soc. 2008, 130 (1), 318–326. (20) Li, X. L.; He, J.; Hihath, J.; Xu, B. Q.; Lindsay, S. M.; Tao, N. J. J. Am. Chem. Soc. 2006, 128 (6), 2135–2141. (21) Haiss, W.; Martin, S.; Leary, E.; van Zalinge, H.; Higgins, S. J.; Bouffier, L.; Nichols, R. J. J. Phys. Chem. C 2009, 113 (14), 5823– 5833. (22) Weibel, N.; Blaszczyk, A.; von Haenisch, C.; Mayor, M.; Pobelov, I.; Wandlowski, T.; Chen, F.; Tao, N. J. Eur. J. Org. Chem. 2008, (1), 136–149.
© 2010 American Chemical Society
(23) Wang, C. S.; Batsanov, A. S.; Bryce, M. R.; Martin, S.; Nichols, R. J.; Higgins, S. J.; Garcia-Suarez, V. M.; Lambert, C. J. J. Am. Chem. Soc. 2009, 131 (43), 15647–15654. (24) Quek, S. Y.; Kamenetska, M.; Steigerwald, M. L.; Choi, H. J.; Louie, S. G.; Hybertsen, M. S.; Neaton, J. B.; Venkataraman, L. Nat. Nanotechnol. 2009, 4 (4), 230–234. (25) Yamashita, S. Bull. Chem. Soc. Jpn. 1961, 34 (4), 490–493. (26) Haiss, W.; Wang, C. S.; Grace, I.; Batsanov, A. S.; Schiffrin, D. J.; Higgins, S. J.; Bryce, M. R.; Lambert, C. J.; Nichols, R. J. Nat. Mater. 2006, 5 (12), 995–1002.
2023
DOI: 10.1021/nl9042455 | Nano Lett. 2010, 10, 2019-–2023
