sensors Review

Surface-Enhanced Raman Scattering in Molecular Junctions Madoka Iwane, Shintaro Fujii * and Manabu Kiguchi * Department of Chemistry, Tokyo Institute of Technology, 2-12-1 Ookayama, Meguro-ku, Tokyo 152-8551, Japan; [email protected] * Correspondence: [email protected] (S.F.); [email protected] (M.K.); Tel.: +81-3-5734-2071 (S.F & M.K.) Received: 12 July 2017; Accepted: 16 August 2017; Published: 18 August 2017

Abstract: Surface-enhanced Raman scattering (SERS) is a surface-sensitive vibrational spectroscopy that allows Raman spectroscopy on a single molecular scale. Here, we present a review of SERS from molecular junctions, in which a single molecule or molecules are made to have contact from the top to the bottom of metal surfaces. The molecular junctions are nice platforms for SERS as well as transport measurement. Electronic characterization based on the transport measurements of molecular junctions has been extensively studied for the development of miniaturized electronic devices. Simultaneous SERS and transport measurement of the molecular junctions allow both structural (geometrical) and electronic information on the single molecule scale. The improvement of SERS measurement on molecular junctions open the door toward new nanoscience and nanotechnology in molecular electronics. Keywords: surface-enhanced Raman scattering; molecular electronics; single-molecular junction; electron transport

1. Introduction Molecular junctions, where a small number of molecules bridge metal electrodes, have been envisioned as components for miniaturized electronic circuits since the 1970s. Aviram et al. first proposed theoretically that electrical rectification was possible with a molecular junction where a donor π system is bound to an acceptor π system via a σ bonded tunneling bridge [1]. In 1997, Metzger et al. experimentally showed the electrical rectification in the hexadecylquinolinium tricyanoquinodimethanide (C16 H33 Q-3CNQ) molecular junction [2]. At present, various functionalities including diode, transistor, and switch have been reported for single-molecular junctions [3–10]. Song et al. reported the transistor characteristics of alkanedithiol and benzene dithiol (BDT) single-molecule junctions [3]. The molecular currents were modulated by more than two orders of magnitude along with controlling the gate voltage between ±3 V. A single molecular diode was reported by Díez-Pérez et al. in 2009 [6]. They prepared a single molecular diode with the diblock dipyrimidinyldiphenyl molecule using the scanning tunneling microscope-break junction (STM-BJ) technique. The orientation of the asymmetric molecule was controlled through a selective deprotection strategy. The average rectification ratio at a 1.5 V bias was approximately five to one from positive to negative bias polarities. Kiguchi et al. reported single molecular resistive switch with the covered oligothiophene (QT) molecules [4]. The QT molecule has two same anchors (i.e., thiophene rings) at both the termini. The QT single-molecule junction showed three distinct conductance states depending on the gap size. The conductance was tuned by controlling the anchoring positions in the same molecule with the gap distance. Although various functionalities have been reported for molecular junctions, molecular devices are still far away from practical application, due to large variability of the device performance. Sensors 2017, 17, 1901; doi:10.3390/s17081901

www.mdpi.com/journal/sensors

Sensors 2017, 17, 1901

2 of 14

For example, several different values (0.1 G0 (G0 = 2 e2 /h), 0.01 G0 , 4 × 10−4 G0 ) were reported as conductance of a benzenedithiol (BDT) single-molecular junction [3,11–14]. One of the origins of the variability of the charge transport property is due to the lack of the direct structural information of the single-molecular junction. Theoretical study showed that the conductance of the BDT single-molecular junction depended on the adsorption geometries on Au electrodes. The calculated conductance was 0.078 G0 , 0.040 G0 , and 0.004 G0 for bridge, hollow, atop site, respectively [13]. Generally, only electrical conductivity of the single-molecular junctions has been discussed without structural and electronical characterization of the molecular junctions. Vibrational spectroscopy is the most straightforward method to determine the atomic and electronic structure of the molecular junction because it provides a molecular fingerprint that can be used to identify the bridging molecule and molecular adsorption site. In 2002, point-contact spectroscopy (PCS) was first exercised to the hydrogen single-molecular junction [15]. PCS observed a small conductance change caused by the electron—phonon interaction of the single-molecular junction. Peaks in d2 I/dV 2 -V curves provide the vibrational energy of the single-molecular junction, which unambiguously clarifies the existence of the molecule between metal electrodes. The PCS and inelastic electron tunneling spectroscopy (IETS) have been employed for numerous molecular junctions and were established as standard methods to identify the molecular junctions [16–18]. However, there is a drawback in these spectroscopies. The peak width in the spectrum increases with the temperature, and we can measure PCS and IETS only at low temperature (~4 K). For practical application, it is essential to measure the vibrational spectroscopy at room temperature. Optical spectroscopic techniques, such as IR and Raman spectroscopy, are promising for vibration spectroscopy of the molecular junctions at room temperature. However, it seems difficult to use optical spectroscopes to observe the molecular junction. First, it is not easy to focus light to single-molecule size; second, the signal from a single or few molecules is too weak to detect. Fortunately, surface-enhanced Raman scattering (SERS) can overcome these difficulties thanks to the enhanced field formed between metal electrodes [19–25]. When light is irradiated onto metal nano electrodes whose size is smaller than the light wavelength, localized surface plasmon is excited in the metal nano electrodes. As the metal nano electrodes approach each other, the localized plasmon modes hybridize, resulting in a strong electric field between metal nano electrodes. This enhanced field increases the Raman signal. Since Raman scattering is a second-order optical process, the metallic nanostructure acts as an amplifier for both the incoming and scattered wave fields. The SERS enhancement factor can be as large as 1015 in the metal nanogap, which is sufficient to study Raman scattering of a single molecule [26–29]. SERS measurement of a single molecule was first demonstrated using random aggregates of colloidal Ag nanoparticles with crystal violet and rhodamine 6 G molecules. In the nano particle system, the molecules bridge metal nano structures, so this system can regarded as molecular junctions. The SERS of the molecule in the molecular junction could be detected thanks to the similar enhancement mechanism. In SERS measurement of molecular junctions, simultaneous SERS and conductance measurement are crucial. The conductance of the junction provides the information about the molecular junction; gap size, the number of the bridging molecules in the junction, and so on. The complementary information is obtained by SERS and conductance measurements. In this review article, we discuss the SERS studies on the molecular junctions. 2. SERS Measurement on the Molecular Junction In 2006, the first SERS study of the molecular junction was reported by Tian et al. [30]. The BDT molecular junction was prepared by mechanically controllable break junction (MCBJ). In the MCBJ technique, a metal wire is attached on a flexible substrate. Target molecules are adsorbed on the surface of metal by immersing the wire in the solution containing the molecules. The sample is mounted in a three-point bending configuration. By bending the substrate, the metal wire is mechanically broken, leaving two atomically sharp electrodes separated by a nanogap. Molecules diffuse on the metal wire, and finally molecules are trapped in the nanoscale gap, forming molecular junctions [31,32].

Sensors 2017, 17, 1901

3 of 13

Sensors 2017, 17, 1901 3 of 14 diffuse on the metal wire, and finally molecules are trapped in the nanoscale gap, forming molecular junctions [31,32]. Using the experimental setup shown in Figure 1a, the authors observed SERS enhancement only at the gap. The SERS signals from the smooth surface and edge of electrodes were Using the experimental setup shown in Figure 1a, the authors observed SERS enhancement only at the weak. The polarization dependence of incident light on SERS supported the molecular junction and gap. The SERS signals from the smooth surface and edge of electrodes were weak. The polarization gave significant SERS signals. Strong SERS signals were detected when the polarization of the dependence of incident light on SERS supported the molecular junction and gave significant SERS incident laser was parallel to the junction axis. When the polarization of the incident laser was signals. Strong SERS signals were detected when the polarization of the incident laser was parallel perpendicular to the junction axis, the SERS intensity decreased greatly. In the SERS of BDT to the junction axis. When the polarization of the incident laser was perpendicular to the junction molecular junction, strong peaks were observed at 1068 cm−1 and 1568 cm−1, which were assigned to axis, the SERS intensity decreased greatly. In the SERS of BDT molecular junction, strong peaks a ring breathing mode (ν1)−and C=C stretching mode (ν8a) of the BDT molecule, respectively. Other were observed at 1068 cm 1 and 1568 cm−1 , which were assigned to a ring breathing mode (ν1 ) and peaks were also assigned to the vibrational modes of the BDT molecule, which clearly showed the C=C stretching mode (ν8a ) of the BDT molecule, respectively. Other peaks were also assigned to the existence of the BDT molecules in the gap between the metal electrodes. The authors studied the vibrational modes of the BDT molecule, which clearly showed the existence of the BDT molecules in the dependence of SERS intensity as a function of the gap width. The gap width was evaluated by gap between the metal electrodes. The authors studied the dependence of SERS intensity as a function measuring the tunneling current across the gap. Figure 1b shows three SERS spectra of BDT of the gap width. The gap width was evaluated by measuring the tunneling current across the gap. molecules with different gap widths. The SERS intensity increased considerably when the gap width Figure 1b shows three SERS spectra of BDT molecules with different gap widths. The SERS intensity was decreased from 0.8 to 0.4 nm. This enhancement in SERS intensity was explained by the increase increased considerably when the gap width was decreased from 0.8 to 0.4 nm. This enhancement in in electromagnetic field as one reduced the gap. SERS intensity was explained by the increase in electromagnetic field as one reduced the gap.

Figure Figure 1. 1. (a) (a) The The experimental experimental setup setup for for surface-enhanced surface-enhanced Raman Raman scattering scattering (SERS) (SERS) of of molecular molecular junctions fabricated with a mechanically controllable break junction (MCBJ); (b) SERS of 1,4junctions fabricated with a mechanically controllable break junction (MCBJ); (b) SERS of benzenedithiol (BDT) molecular junctions with different gap widths. The gap width was (A) 0.8 nm, 1,4-benzenedithiol (BDT) molecular junctions with different gap widths. The gap width was (A) 632.8 [30].nm) [30]. (B) (C) 0.4 0.8 0.6 nm,nm, (B) and 0.6 nm, andnm. (C) (λ 0.4ex:nm. (λnm) : 632.8 ex

The first SERS of the single-molecular junction was reported by Liu et al. in 2011 [33]. They The first SERS of the single-molecular junction was reported by Liu et al. in 2011 [33]. They measured measured SERS of the single-molecular junction with the fishing-mode tip-enhanced Raman SERS of the single-molecular junction with the fishing-mode tip-enhanced Raman scattering (TERS). scattering (TERS). The fishing-mode TERS (FM-TERS) is achieved by combining ‘fishing-mode’ STM The fishing-mode TERS (FM-TERS) is achieved by combining ‘fishing-mode’ STM (FM-STM) with TERS. (FM-STM) with TERS. In the FM-STM mode (Figure 2a), the proportional gain and the integral gain In the FM-STM mode (Figure 2a), the proportional gain and the integral gain are decreased in order are decreased in order to decrease the response of the STM feedback. The current (conductance) to decrease the response of the STM feedback. The current (conductance) though the STM junction is though the STM junction is continuously monitored. When a molecular junction is formed, the continuously monitored. When a molecular junction is formed, the electric current increases. The STM tip electric current increases. The STM tip is retracted to decrease the current under the low-feedback is retracted to decrease the current under the low-feedback condition. As the STM tip is moved away condition. As the STM tip is moved away from the surface, the molecular junction breaks and the from the surface, the molecular junction breaks and the electric current decreases. The low-feedback electric current decreases. The low-feedback system forces the tip to approach to the surface to regain system forces the tip to approach to the surface to regain the tunneling current. Figure 2b shows the the tunneling current. Figure 2b shows the time sequence conductance curve for 4,4′-bipyridine (BPY) time sequence conductance curve for 4,40 -bipyridine (BPY) in the junction between a Au STM tip and in the junction between a Au STM tip and a Au substrate. Current jumps were observed and the time a Au substrate. Current jumps were observed and the time scale of each jump was a fraction of a scale of each jump was a fraction of a millisecond. These current jumps were attributed to the millisecond. These current jumps were attributed to the formation (ON) and breaking (OFF) of the formation (ON) and breaking (OFF) of the BPY single-molecular junction. The TERS signal fluctuated BPY single-molecular junction. The TERS signal fluctuated when the electric conductance switched when the electric conductance switched between the single-molecular junction (ON) state and the between the single-molecular junction (ON) state and the breaking (OFF) state, indicating that the change breaking (OFF) state, indicating that the change in the TERS signal originated from the single in the TERS signal originated from the single molecular events. The TERS signal from the ON state molecular events. The TERS signal from the ON state showed the increase in intensities and peak showed the increase in intensities and peak widths, and shifting of peak positions (Figure 2c). Clear C—C widths, and shifting of peak positions (Figure−2c). Clear C—C stretching mode was observed around stretching mode was observed around 1620 cm 1 (C—N/C—C dephasing stretching mode; ν ) in SERS, 1620 cm−1 (C—N/C—C dephasing stretching mode; ν8a) in SERS, which confirmed that8a the BPY which confirmed that the BPY molecule bridged Au electrodes. molecule bridged Au electrodes.

Sensors 2017, 17, 1901 Sensors 2017, 17, 1901

4 of 14 4 of 13

Figure Figure 2. 2. (a) (a) Schematic Schematic view view of of the the fishing-mode fishing-mode tip-enhanced tip-enhanced Raman Raman scattering scattering (FM-TERS) (FM-TERS) for for simultaneous measurement of conductance and TERS for single-molecular junctions; (b) Time course simultaneous measurement of conductance and TERS for single-molecular junctions; (b) Time course of 0 -bipyridine/Au (111) (111) of conductance Au scanning tunneling microscopy tip/4,4′-bipyridine/Au system; (c) conductance for for the the Au scanning tunneling microscopy tip/4,4 system; (c) TERS TERS spectra corresponding the ONOFF andstates. OFF states. (λex: 632.8 nm, intensity: laser intensity: 5 mW) spectra corresponding to the to ON and (λex : 632.8 nm, laser 5 mW) [33]. [33].

Stable molecular junction could be fabricated with the fixed electrode on the flat substrate. The Stable molecular junction could be fabricated with the fixed electrode on the flat substrate. The fixed fixed electrodes could be prepared using electromigration (EM) technique [34]. The EM process electrodes could be prepared using electromigration (EM) technique [34]. The EM process originates from originates from the momentum transfer from electrons to metal atoms, so called electron wind force, the momentum transfer from electrons to metal atoms, so called electron wind force, and/or Joule heating. and/or Joule heating. When a metal wire is electrically heated, the metal atoms become mobile. When a metal wire is electrically heated, the metal atoms become mobile. Because the momentum transfers Because the momentum transfers from electrons to metal atoms, the atoms move in the opposite from electrons to metal atoms, the atoms move in the opposite direction of current flow. The EM process direction of current flow. The EM process is accomplished by ramping a voltage across the metal wire is accomplished by ramping a voltage across the metal wire while monitoring the current. As the current while monitoring the current. As the current increases, the conductance starts to change. Upon increases, the conductance starts to change. Upon further increase of the current, the conductance drops further increase of the current, the conductance drops suddenly to almost zero, and the metal wire suddenly to almost zero, and the metal wire breaks, forming the nanoscale gap. Although the number breaks, forming the nanoscale gap. Although the number of the bridging molecules between the fixed of the bridging molecules between the fixed electrodes cannot be controlled, mechanical stability of the electrodes cannot be controlled, mechanical stability of the electrodes is much higher than those electrodes is much higher than those fabricated by MCBJ, FM-STM, and other break junction techniques. fabricated by MCBJ, FM-STM, and other break junction techniques. Figure 3a is the scanning electron Figure 3a is the scanning electron (SEM) image of the fixed electrodes fabricated on a Si substrate, and (SEM) image of the fixed electrodes fabricated on a Si substrate, and on which p-mercaptoaniline on which p-mercaptoaniline (pMA) was deposited from solutions [34]. Figure 3b is the spatial map of (pMA) was deposited from solutions [34]. Figure 3b is the spatial map of the Si 520 cm−1 peak in the the Si 520 cm−1 peak in the same region shown in Figure 3a. The Si peak was not observed on the Au same region shown in Figure 3a. The Si peak was not observed on the Au electrodes. Figure 3c shows electrodes. Figure 3c shows the distribution of the pMA-SERS signal of 1590 cm−1 (a1 symmetry mode). the distribution of the pMA-SERS signal of 1590 cm−1 (a1 symmetry mode). The SERS signal was only The SERS signal was only observed at the nanogap. Figure 3d shows the example of the simultaneous observed at the nanogap. Figure 3d shows the example of the simultaneous conductance and the conductance and the SERS measurements of a pMA molecular junction (a pMA molecule trapped in SERS measurements of a pMA molecular junction (a pMA molecule trapped in the Au electrodes). the Au electrodes). The rapid changes in the SERS signals correlated with conductance changes but the The rapid changes in the SERS signals correlated with conductance changes but the relationship was relationship was complicated. In some periods, the increases in SERS intensity correlated with increase in complicated. In some periods, the increases in SERS intensity correlated with increase in conductance, conductance, while increases in SERS intensity correlated with decrease in conductance in other period. while increases in SERS intensity correlated with decrease in conductance in other period. The The observed changes in conductance and SERS of molecular junction can be explained by the change in observed changes in conductance and SERS of molecular junction can be explained by the change in the atomic configuration of the molecular junction. the atomic configuration of the molecular junction. Using a highly stable molecular junction with fixed electrodes, the temperatures of the charge-transporting molecular junctions were investigated by Ward et al. and Ioffe et al. [35,36]. ef f They evaluated the effective temperature (Tν ) of the molecular junction for each vibrational mode by ef f the Stokes (S) and the anti-Stokes (AS) ratio [35,36]. Here, Tν is represented by   IνAS ( v L + v ν )4 ef f = A exp −} v /k T ν ν B ν IνS ( v L − v ν )4

(1)

where IνS and IνAS are the intensity of the Stokes and anti-Stokes Raman mode, v L and vν are the frequency of the incident laser, wave number of the vibrational mode, and Aν is a correction factor, respectively [35]. Figure 4a shows the bias voltage dependence of SERS for the three-ring oligophenylene vinylene (OPV3) molecular junction. The intensity of the anti-Stokes SERS increased ef f ef f with the voltage. Figure 4b shows the Tν for different modes as a function of bias voltage. The Tν Figure 3. (a) Scanning electron microscope (SEM) image of the sample; (b) Map of thewith substrate Si 520 linearly increased with the bias voltage. Although the effective temperature varied the vibrational cm−1 peak; (c) Map of the p-mercaptoaniline (pMA) SERS signal of 1590 cm−1 (a1 symmetry mode); (d) Waterfall plot of SERS and conductance for a pMA molecular junction (λex: 785 nm, laser intensity: 0.5 mW) [34].

same region shown in Figure 3a. The Si peak was not observed on the Au electrodes. Figure 3c shows the distribution of the pMA-SERS signal of 1590 cm−1 (a1 symmetry mode). The SERS signal was only observed at the nanogap. Figure 3d shows the example of the simultaneous conductance and the SERS measurements of a pMA molecular junction (a pMA molecule trapped in the Au electrodes). The rapid changes in the SERS signals correlated with conductance changes but the relationship was Sensors 2017, 17, 1901 5 of 14 complicated. In some periods, the increases in SERS intensity correlated with increase in conductance, while increases in SERS intensity correlated with decrease in conductance in other period. The modes, generally the effective temperature can increase several hundred by by the application of observed changesspeaking in conductance and SERS of molecular junctionbycan be explained change in theatomic bias voltages with a few hundred millivolts. the configuration of the molecular junction. Sensors 2017, 17, 1901

5 of 13

Using a highly stable molecular junction with fixed electrodes, the temperatures of the chargetransporting molecular junctions were investigated by Ward et al. and Ioffe et al. [35,36]. They evaluated the effective temperature (Tνeff) of the molecular junction for each vibrational mode by the Stokes (S) and the anti-Stokes (AS) ratio [35,36]. Here, Tνeff is represented by exp

/

(1)

and are the intensity of the Stokes and anti-Stokes Raman mode, and are the where frequency of the incident laser, wave number of the vibrational mode, and is a correction factor, respectively [35]. Figure 4a shows the bias voltage dependence of SERS for the three-ring oligophenylene vinylene (OPV3) molecular junction. The intensity of the anti-Stokes SERS increased withFigure the voltage. Figure 4belectron shows the Tνeff for different modes as a function ofthe bias voltage. The Tνeff microscope (SEM) image of the sample; (b) Map of Si 520 Figure3.3.(a) (a)Scanning Scanningelectron microscope (SEM) image of the sample; (b) Map of substrate the substrate Si −1 −1 linearly increased with the bias voltage. (pMA) Although the effective temperature varied with −1 peak; cm peak; (c) Map of the SERS signal of 1590 cm (a−1 1symmetry mode); (d) the 520 cm (c) Map ofp-mercaptoaniline the p-mercaptoaniline (pMA) SERS signal of 1590 cm (a1 symmetry mode); vibrational modes, generally speaking the effective temperature can increase by several hundred Waterfall plot of SERS and conductance for a pMA molecular junction (λ ex : 785 nm, laser intensity: 0.5 by (d) Waterfall plot of SERS and conductance for a pMA molecular junction (λex : 785 nm, laser intensity: application of the bias voltages with a few hundred millivolts. mW) [34].[34]. 0.5 mW)

Figure Figure 4. 4. (a) (a) Sample Sample Raman Raman spectra spectra of of the the three-ring three-ring oligophenylene oligophenylene vinylene vinylene (OPV3) (OPV3) molecular molecular junction. The full scale of the anti-Stokes and Stokes signal are 235 counts and junction. The full scale of the anti-Stokes and Stokes signal are 235 counts and 10,000 10,000 counts, counts, −1 respectively; (b) Effective Effectivetemperature temperatureofof molecular junction a function of voltage: bias voltage: 1317 respectively; (b) thethe molecular junction as aas function of bias 1317 cm −1 (red) and 1625 −1 (blue) [35]. − 1 cm cm (red) and 1625 cm (blue) [35].

3. 3. Correlation Correlation between between SERS SERS and and Atomic Atomic Structure Structure of of the the Molecular Molecular Junction Junction Raman Raman shift shift provides provides us us the the information information about about the the atomic atomic structure structure of of the the molecular molecular junction. junction. Konishi et al. observed the structural change of the BPY single-molecular junction as a change Konishi et al. observed the structural change of the BPY single-molecular junction as a change in in the the SERS SERS spectra spectra [37]. [37]. First, First, they they confirmed confirmed that that the the SERS SERS signal signal originated originated from from the the single-molecular single-molecular junction junction by by the the correlation correlation between between SERS SERS intensity intensity and and electric electric conductance. conductance. Figure Figure 55 shows shows the the SERS SERS intensity as a function of conductance of the BPY molecular junctions fabricated with nano MCBJ intensity as a function of conductance of the BPY molecular junctions fabricated with nano MCBJ electrodes. 0.01 G G0.. Previous research electrodes. A A significant significant increase increase in in SERS SERS intensity intensity was was observed observed around around 0.01 0 Previous research −2 G0 [38], indicating the revealed that the conductance of the BPY single-molecular junction was 10 revealed that the conductance of the BPY single-molecular junction was 10−2 G0 [38], indicating the single single molecular molecular origin origin of of the the SERS SERS spectrum. spectrum.

Sensors 2017, 17, 1901 Sensors 2017, 17, 1901

of 13 14 66 of 6 of 13

Figure 5. SERS intensity as a function of the conductance of the BPY molecular junction. The arrow Figure 5. SERS intensity as a function of the conductancejunction of the BPY molecular junction. The arrow indicates conductance regime of BPY [37].molecular The hollow and solid squares Figure 5. the SERS intensity as a function ofsingle-molecular the conductance of the BPY junction. The arrow indicates the conductance regime of BPY single-molecular junction [37]. The hollow and solid squares correspond the molecular junctions the totally symmetric a mode and thesolid non-totally indicates thetoconductance regime of BPYshowing single-molecular junction [37]. The hollow and squares correspondb to the in molecular junctions showing the totally symmetric a mode and the non-totally symmetric mode SERS, respectively. correspond to the molecular junctions showing the totally symmetric a mode and the non-totally symmetric b mode in SERS, respectively. symmetric b mode in SERS, respectively.

Typically, two types of SERS spectra were observed for the BPY single-molecular junction Typically, two types of998, SERS spectra were the BPY junction −1 (bottomfor (Figure 6a). The bands 840, 1026 and cmobserved of Figure 6a) single-molecular were assigned the nonTypically, two typesat of SERS spectra were1205 observed for the BPY single-molecular junctionto (Figure 6a). −1 (bottom (Figure 6a). The bands at 840, 998, 1026 and 1205 cm of Figure 6a) were assigned to the non−1 (top of Figure 6a) was assigned to the non-totally totally symmetric b 1 mode. The band at 975 cm − 1 The bands at 840, b998, 1026 and 1205 cm (bottom of Figure 6a) were assigned to the non-totally totally symmetric 1 mode. The band at 975 cm−1 (top of Figure 6a) was assigned to the non-totally symmetric b2 mode. modes. The theoretical showed the b2 mode when the BPY −1 (top of Figure symmetric b The band at 975 cmcalculation 6a) that was to theappeared non-totally symmetric b2 1 symmetric b2 modes. The theoretical calculation showed thatassigned the b2 mode appeared when the BPY molecule vertically bridged the gap, and the b 1 mode appeared when the molecule became tilted. The modes. The theoretical calculation showed that the b mode appeared when the BPY molecule vertically 2 appeared when the molecule became tilted. The molecule vertically bridged thesingle-molecular gap, and the b1 mode conformational change of the junction wasbecame observed as aThe change in SERS spectra. bridged the gap, and the b mode appeared when the molecule tilted. conformational change 1 conformational change of the single-molecular junction was observed as a change in SERS spectra. Figure 6b depicts time course of the conductance and the intensity of b 1 and b2 mode in the SERS of the single-molecular junction was observed as a change in SERS spectra. Figure 6b depicts time Figure of 6b the depicts time course of the conductance and the intensity of b1 and b2 mode in thecourse SERS spectra BPY single-molecular junction. The b 1 mode was detected until 3 s, and the conductance of the conductance and the intensity of b and b mode in the SERS spectra of the BPY single-molecular 1 2 spectra of thethan BPY0.01 single-molecular junction. The disappeared b1 mode was detected until 3 the s, and the conductance was higher G0. detected At 4 s, until the b31 s,mode with rapid decrease junction. The than b1 mode was and the conductancetogether was higher than 0.01 G0 . At 4 s, the in bin1 was higher 0.01 G 0. At 4 s, the b1 mode disappeared together with the rapid decrease conductance below 0.01 G 0. At 7 s, the b2 mode appeared and conductance recovered to the initial mode disappeared together with the rapid decrease in conductance below 0.01 G . At 7 s, the b mode 0 conductance below 0.01 G0. spectra At 7 s, the b2 modethat appeared andmolecule conductance recovered to the the2 metal initial value. Theand observed SERS indicated the The BPY initially bridged appeared conductance recovered to the initial value. observed SERS spectra indicated that the value. Thewith observed SERS spectra indicated thattothe BPY molecule initially bridged the metal electrodes its molecular long axis inclining the junction axis, afterinclining which the molecular BPY molecule initially bridged the metal electrodes with its molecular long axis to the junction electrodes with(3its inclining to the axis,with after the molecular junction broke s),molecular and finallylong the axis molecular bridges thejunction electrodes itswhich molecular long axis axis, afterbroke which(3 thes),molecular junction broke (3 s),bridges and finally the molecular bridges the electrodes junction and finally the molecular the electrodes with its molecular long with axis parallel to the junction axis. its molecular long axis parallel to the junction axis. parallel to the junction axis.

Figure of SERS SERS spectra spectra of of the the BPY BPY single-molecular single-molecular junctions junctions (λ (λex ex: 785 nm, laser Figure 6. 6. (a) (a) Two Two types types of : 785 nm, laser Figure 6. (a) Two types of SERS spectra of the BPY single-molecular junctions (λ : 785 nm, laser and bex 2 mode, and the intensity: Raman intensity intensity of the bb1 mode intensity: 0.5 0.5 mW); mW); (b) (b) Time Time evolution evolution of of the the Raman of the mode and b 1 2 mode, and the 1 mode and b2 mode, and the intensity: 0.5 mW); (b) Time evolution of the Raman intensity of the b conductance conductance of of the the BPY BPY single-molecular single-molecular junction; junction; (c) (c) Time Time course course of of the the Raman Raman shift shift of of the the ring ring conductance of around the BPY1050 single-molecular junction; (c) of Time course of the Raman junction shift of the ring −1 − 1 breathing mode cm and the conductance the BPY single-molecular breathing mode around 1050 cm −1 and the conductance of the BPY single-molecular junction[37]. [37]. breathing mode around 1050 cm and the conductance of the BPY single-molecular junction [37].

Sensors 2017, 17, 1901

7 of 14

As well as the dramatic change in the atomic configuration of the junction, a slight molecular orientation change was detected as a change in the Raman shift and conductance of the BPY single-molecular junction. Figure 6c shows the time course of electric conductance and the wavenumber of the ring breathing mode around 1050 cm−1 in the single-molecular junction regime. The wavenumber of the ring breathing-mode increased (decreased) when the conductance decreased (increased). The observed anticorrelation between wavenumber and electric conductance could be explained based on the metal—molecule interaction. When a molecule adsorbs on the metal surface, the highest occupied molecular orbital (HOMO) of BPY hybridizes with the metal unoccupied states, and the lowest unoccupied molecular orbital (LUMO) hybridizes with the metal occupied states. These two interactions lead to the strong molecule—metal bond. Through these interactions, electrons are removed from the HOMO (bonding) and are injected into the LUMO (anti-bonding). Both electron transfers weakens the bonds of molecule itself. The degree of decrease in the molecular bond strength, corresponding to the wavenumber, depends on the strength of the interaction between metal and molecule. An increase in the metal–molecule interaction, thus, leads to a decrease in the energy of the wavenumber of the molecule. Meanwhile, the electric conductance of the single-molecular junction is described as: 2e2 4Γ2 , (2) G= h ∆2 + 4Γ2 in the single-level tunneling model. Here Γ and ∆ are the electric coupling between the metal and molecular orbitals and the energy difference between the metal and molecular orbitals, respectively. The electric conductance increases with the electric coupling, that is, the metal—molecule interaction. The increase in the metal—molecule interaction, thus, causes the increase in the electric conductance of the single-molecular junction, and decrease in the wavenumber, as experimentally observed in Figure 6c. The fluctuation of the wavenumber synchronized with the change in electric conductance directly showed the dynamic motion of the single-molecular junction. In the above studies, the atomic structure of the single-molecular junction is argued by considering the electric conductance and vibrational energy in SERS. The current—voltage (I—V) response provides much more information than conductance (e.g., electronic structure, metal—molecule interface structure of the molecular junction). Assuming that electron transport through a single channel, I—V response is represented as ( ) 1 1 2e −1 2 eV − ε 0 −1 2 eV + ε 0 I (V ) = Γ tan ( ) + tan ( ) h Γ Γ

(3)

where ε0 is the energy of the conduction orbital [39–41]. The parameters, ε0 and Γ, are sensitive to the metal—molecule interface structure of the single-molecular junction. By fitting the I–V response to Equation (3), ε0 and Γ can be obtained. Figure 7a shows the distribution of I–V responses for BDT single-molecular junctions. Three statistically high-probable nonlinear curves were clearly observed (H, M and L). The distribution of strength Γ also indicates probable three Γ’s in the Γ histogram (Figure 7b) [42]. By comparing the experimentally obtained ε0 , Γ, and conductance with the calculated ones, the experimentally observed H, M and L states were assigned to the bridge, hollow, and top adsorption site geometry, respectively. Then, the SERS of the BDT single-molecular junction is discussed based on the results obtained by I—V responses. Orange counts in the Γ histogram (Figure 7b) corresponds to SERS active samples. The counts were found to concentrate in the H state (bridge). In other words, we can conclude that the BDT molecule occupies the bridge site when its SERS signal is detected. Moreover, the SERS intensity was found to increase with Γ. Figure 7c shows the relationship between the intensity of the SERS signal (Is ) and Γ on a log—log plot. The observed distribution clearly corresponded to a power law relationship, Is ∝ Γ α . This is the first in-situ study of the correlation between the optical and electronic properties in single-molecular junctions [42].

Sensors 2017, 17, 1901 Sensors 2017, 17, 1901

8 of 14 8 of 13

Figure 7.7.(a) (a) Bidimensional histogram summarizing the individual I—V response of Figure Bidimensional I—VI—V histogram summarizing the individual I—V response of singlesingle-molecule BDT junctions; (b) Statistical distribution of Orange Γ. Orange countscorrespond correspondto to νν8a -active molecule BDT junctions; (b) Statistical distribution of Γ. counts 8a-active samples; (c) (c) Correlation Correlation between between the the average average intensity intensity of of the the SERS SERS signal signal (ν (ν11 and and νν88 modes) modes) as as aa samples; function of Γ on a log—log plot; (d) Photo-induced charge transfer transition from HOMO to metal function of Γ on a log—log plot; (d) Photo-induced charge transfer transition from HOMO to metal unoccupiedstate. state.The Thediscrete discretemolecular molecularlevel levelisis broadened broadened by by ΓΓ[42]. [42]. unoccupied

The The site site selectivity selectivity and and power power law law relationship relationship were were explained explained by by the the SERS SERS enhancement enhancement mechanism mechanism (Figure (Figure 7d). 7d). The The SERS SERS signal signal gains gains intensity intensity from from two two contributions: contributions: electromagnetic electromagnetic (EMM) [21,22,43,44]. The (EMM) and and chemical chemical (CM) (CM) effects effects [21,22,43,44]. The EMM EMM effect effect originates originates from from local local field field enhancement. Although the EMM effect was the major contributing factor in SERS of the BDT singleenhancement. Although the EMM effect was the major contributing factor in SERS of the BDT molecular junction, the sitethe selectively and power law relationship could could be explained by the single-molecular junction, site selectively and power law relationship be explained byCM the effect. One of the main sources for the CM effect is charge transfer resonance taking place between CM effect. One of the main sources for the CM effect is charge transfer resonance taking place between metal metal states states near near the the Fermi Fermi level level and and molecular molecular electronic electronic stats stats (from (from HOMO HOMO to to metal metal unoccupied unoccupied state stateor ormetal metaloccupied occupiedstate stateto toLUMO) LUMO)[45]. [45].The Thecharge chargetransfer transferbetween betweenthe themetal metaland andthe themolecule molecule easily occurs when the metal—molecule interaction is strong. Therefore, the SERS intensity increases easily occurs when the metal—molecule interaction is strong. Therefore, the SERS intensity increases with calculationbased basedon ona asingle-level single-level Anderson model reproduced power with Γ. Γ. The The theoretical theoretical calculation Anderson model reproduced thethe power law law relationship. As for the site selectivity, electronic coupling was small for the hollow and atop, relationship. As for the site selectivity, electronic coupling was small for the hollow and atop, and and the SERS intensity corresponding to sites thesewere sitesunder were the under the detection limit (SERS thus,thus, the SERS intensity corresponding to these detection limit (SERS inactive). inactive). high coupling state (H) iswhich visible, which site selectivity. Only highOnly coupling state (H) is visible, causes sitecauses selectivity. 4. 4.Effect Effectof ofLight LightIrradiation Irradiationand andBias BiasVoltage Voltage Application Application on Molecular Junction In the above abovediscussion, discussion, assume thateffect the of effect laser irradiation and bias voltage In the wewe assume that the laserofirradiation and bias voltage application application on thejunction molecular junction is negligibly small. the bias voltage-induced heating on the molecular is negligibly small. Although theAlthough bias voltage-induced heating was observed was observed for OPV3 molecular [35], it that is still that theof atomic structurejunction of the for OPV3 molecular junction [35], it junction is still assumed theassumed atomic structure the molecular molecular junctionby was changed by of bias voltage. However, the light irradiation was not changed thenot application of the biasapplication voltage. However, the light irradiation and application and application of bias voltage can affectjunction. the molecular Lithe et al. reported Raman the bias-driven of bias voltage can affect the molecular Li et al.junction. reported bias-driven shift for Raman shift for the C60 molecular junctions using the fixed electrode, which were fabricated using EM technique [46]. Figure 8a shows the SERS of the C60 molecular junction. The sharp peak at 520

Sensors 2017, 17, 1901

9 of 14

Sensors 17, 1901 the C 2017, molecular

9 of 13 junctions using the fixed electrode, which were fabricated using EM technique [46]. − 1 Figure 8a shows the SERS of the C60 molecular junction. The sharp peak at 520 cm is from the cm−1 is from the underlying Si substrate, and the peaks−between 1000 cm−1 and 1600 cm−1 are from underlying Si substrate, and the peaks between 1000 cm 1 and 1600 cm−1 are from vibrational modes vibrational modes of C60 in the junction. Figure 8b shows the SERS of C60 molecular junction as a of C60 in the junction. Figure 8b shows the SERS of C60 molecular junction as a function of the bias function of the bias voltages in the bias regime from –0.6 to 0.6 V. Many of the vibrational modes voltages in the bias regime from –0.6 to 0.6 V. Many of the vibrational modes shifted toward low shifted toward low energies when the bias voltage increased. This bias-driven shift is apparent as a energies when the bias voltage increased. This bias-driven shift is apparent as a curvature of the curvature of the spectral features (Figure 8b). The bias driven shifts can be represented by δω~V2. This spectral features (Figure 8b). The bias driven shifts can be represented by δω~V 2 . This quadratic quadratic dependence on the bias voltage is clearly seen in Figure 8c, which shows the Raman shift dependence on the bias voltage is clearly seen in Figure 8c, which shows the Raman shift as a function as a function of the bias voltage for a particular mode−of 1258 cm−1. They discussed that the C60 charge of the bias voltage for a particular mode of 1258 cm 1 . They discussed that the C60 charge state was state was changed by the application of the bias voltage, and the change in the charge state caused changed by the application of the bias voltage, and the change in the charge state caused the vibrational the vibrational energy shifts of the C60 molecular junction. By applying the bias voltage across the energy shifts of the C60 molecular junction. By applying the bias voltage across the junction, electron(s) junction, electron(s) can be injected to the closer lying LUMO and the addition of an electron to the can be injected to the closer lying LUMO and the addition of an electron to the antibonding LUMO antibonding LUMO softens intramolecular bonds. softens intramolecular bonds. 60

Figure 8. (a) Example of SERS of the C60 molecular junction fabricated with the electron migration Figure 8. (a) Example of SERS of the C60 molecular junction fabricated with the electron migration technique. technique. Inset: Inset:SEM SEMimage imageofofthe the electrode. electrode. Surrounding Surroundingfigures figuresillustrate illustratevibrational vibrational modes; modes; (b) SERS of the C 60 molecular junction as a function of the bias voltage; (c) Raman shift as a function (b) SERS of the C60 molecular junction as a function of the bias voltage; (c) Raman shift as a function of of the bias voltage a particular mode: 1258 [46]. −1−1[46]. the bias voltage forfor a particular mode: 1258 cmcm

The shift of the SERS peaks was observed for the BPY single-molecular junction fabricated with The shift of the SERS peaks was observed for the BPY single-molecular junction fabricated with the FM-STM [33]. Figure 9 shows the bias voltage dependence of SERS of the BPY single-molecular the FM-STM [33]. Figure 9 shows the bias voltage dependence of SERS of the BPY single-molecular junction. When the bias voltage was increased from 10 to 800 mV, one peak around 1610 cm−−11 (ν8a) junction. When the bias voltage was increased from 10 to 800 mV, one peak around 1610 cm (ν8a ) changed to double peaks above 100 mV. When the bias voltage was reversed, the peak splitting changed to double peaks above 100 mV. When the bias voltage was reversed, the peak splitting disappeared, indicating that the peak splitting relates with the bias voltage. The BPY molecule disappeared, indicating that the peak splitting relates with the bias voltage. The BPY molecule consists consists of two pyridine rings. The theoretical calculation revealed that the increase in the bias voltage of two pyridine rings. The theoretical calculation revealed that the increase in the bias voltage lowers lowers the Fermi level and increases the density of electric charge on the Au tip. This led to an increase the Fermi level and increases the density of electric charge on the Au tip. This led to an increase in the in the strength of the chemical bond between the Au tip and the pyridine ring in contact with it. The strength of the chemical bond between the Au tip and the pyridine ring in contact with it. The peak peak split could be understood in terms of the different bonding interactions between the Au tip and split could be understood in terms of the different bonding interactions between the Au tip and BPY BPY and between BPY and the Au. and between BPY and the Au.

Sensors 2017, 17, 1901 Sensors 2017, 17, 1901

10 of 14 10 of 13

Figure 9. (a)9.The biasbias voltage dependence of BPY BPYmolecular molecular junction featuring ν8a band Figure (a) The voltage dependenceon onTERS TERS of junction featuring the νthe 8a band 1 ); −1 (1610(1610 cm−cm (b) The schematic junction a low bias-voltage ); (b) The schematicofofthe theBPY BPY single-molecular single-molecular junction at aatlow andand highhigh bias-voltage (λex : (λ 632.8 nm,nm, laser intensity: ex: 632.8 laser intensity:5 5mW) mW)[33]. [33].

As well as the bias voltage, the light irradiation can affect the electric conductance through the

As well as the bias voltage, the light irradiation can affect the electric conductance through the single-molecular junction. Vadai et al. reported the plasmon-induced conductance enhancement in single-molecular junction. Vadai[47]. et al.They reported the plasmon-induced conductance in the single-molecular junction fabricated single-molecular junctions with aenhancement squeezable the single-molecular junction [47]. They fabricated single-molecular junctions with a squeezable break break junction (SBJ) technique (Figure 10a). The SBJ consists of two Au electrodes evaporated on top junction (SBJ)thick technique (Figure 10a).between The SBJ twomechanically Au electrodes evaporated on top of of 1 mm glass slides. The gap Auconsists electrodesofwas controlled by applying 1 mma squeezing thick glass slides. The electrodes was mechanically controlled applying force against thegap top between slide. The Au wavelength of the light was 781 nm and its powerby was 10 mW. Figure 10b shows the histogram of 2,7-diaminofluorene molecular a squeezing force against the conductance top slide. The wavelength of the light was(DAF) 781 nm and itsjunction power was measured without and laser illumination. conductance of the single 10 mW. Figure 10b shows thewith conductance histogramThe of 2,7-diaminofluorene (DAF)DAF-molecular molecular junction −3 G0 without laser illumination. Under the irradiation of light, a new high junction was 1.9 × 10 measured without and with laser illumination. The conductance of the single DAF-molecular junction conductance peak appeared in the conductance histogram at 3.7 × 10−3 G0 in addition to the was 1.9 × 10−3 G0 without laser illumination. Under the irradiation of light, a new high conductance characteristic low conductance peak at 1.9 × 10−3 G0. The low peak (1.9 × 10−3 G0) did not −3conductance peak appeared in the conductance histogram at 3.7 × 10 G in addition to the characteristic low 0 entirely shift to a high conductance value (3.7 × 10−3 G0) and the both low and high conductance peaks −3 G . The low conductance peak (1.9 × 10−3 G ) did not entirely shift to conductance peak at 1.9 × 10 0 0 were observed under the light illumination. The authors explained this observation in the following −3 G ) and the both low and high conductance peaks were observed a high conductance value (3.7 × 0 way. In each measurement of a10conductance trace, local hot spots can be formed on the surface of Au under the lightwhere illumination. The authors explained thisThe observation the following In each electrodes the localized surface plasmon is excited. formation in of local hot spot is way. sensitive measurement of a conductance trace, local hotpropagation spots can belength formed the surface surface of Au electrodes to local roughness of the surface, and the (Lon p) of localized plasmon iswhere about a few 10 nm. The localized plasmon the conductance of spot a single-molecular only the localized surface plasmon is excited. Theaffects formation of local hot is sensitive tojunction local roughness the distance the molecule in the and hot spot isplasmon smaller than Lp. Because this is The of theifsurface, and between the propagation length (Lpjunction ) of localized surface is about a few 10 nm. not necessarily in the case of all conductance measurements, only a fraction of the measurements was localized plasmon affects the conductance of a single-molecular junction only if the distance between affected by the localized surface plasmon. They explained the increase in conductance of the DAF the molecule in the junction and hot spot is smaller than Lp . Because this is not necessarily in the case single-molecular junction by taking the plasmon field as an oscillating potential Vω at the plasmon of all conductance measurements, only a fraction of the measurements was affected by the localized frequency ω across the junction. The calculated transmission probability of the DAF single-molecular surface plasmon. They explained the increase conductance of the DAFinsingle-molecular junction was dominated by HOMO (highest in occupied molecular orbital) the energy range junction of EF − by taking the plasmon field as an oscillating potential V at the plasmon frequency ω across the junction. ħω < E < EF + ħω. The magnitude of gate voltage (Vω)ωwas evaluated to be 0.169 V. The calculated transmission probability of the DAF single-molecular junction was dominated by HOMO (highest occupied molecular orbital) in the energy range of EF − h¯ ω < E < EF + h¯ ω. The magnitude of gate voltage (Vω ) was evaluated to be 0.169 V.

Sensors 2017, 17, 1901

Sensors 2017, 17, 1901

11 of 14

11 of 13

Figure 10. (a) Schematic view of squeezable break junction setup for conductance measurement of the Figure 10. (a) Schematic view of squeezable break junction setup for conductance measurement of molecular junction; (b) Conductance histograms of the 2,7-diaminofluorene molecular junction the molecular junction; (b) Conductance histograms of the 2,7-diaminofluorene molecular junction measured without (black curve) and with (red colored area) continuous wave illumination measured without (black curve) and with (red colored area) continuous wave illumination (wavelength: (wavelength: 781 nm, power: 10 mW). Inset: Examples of the conductance traces of the 2,7781 nm, power: 10 mW). Inset: Examples of the conductance traces of the 2,7-diaminofluorene molecular diaminofluorene molecular junctions without (black) and with (red) laser illumination [47]. junctions without (black) and with (red) laser illumination [47].

5. Conclusions

5. Conclusions In conclusion, we have reviewed the SERS studies of the molecular junction. The molecular junctions have been as components forstudies miniaturized circuits. The SERS one In conclusion, weenvisioned have reviewed the SERS of theelectronic molecular junction. Theismolecular of the promising tools to characterize the molecular devices during device operation at room junctions have been envisioned as components for miniaturized electronic circuits. The SERS is one of temperature. In SERS measurement of the molecular junction, simultaneous SERS and electrical the promising tools to characterize the molecular devices during device operation at room temperature. measurement are crucial. The complementary information is obtained by SERS and electrical In SERS measurement of the molecular junction, simultaneous SERS and electrical measurement measurements, which are essential to fully characterize the molecular junction. The simultaneous are conductance crucial. The complementary information is obtained bymolecule SERS and electrical measurements, and SERS measurement clarify the bridging of target between metal electrodes, which are essential to fully characterize the molecular junction. The simultaneous conductance and effective temperature, dynamical motion of molecule in the gap, and so on. The simultaneous and I—V SERSresponse measurement clarify the bridging of target molecule metal electrodes, andThe effective and SERS can clarify the atomic configuration atbetween the molecule—metal interface. correlated dynamical SERS measurements show selectivity towards adsorption sites. Site-sensitivity temperature, motion of molecule in the gap, one andofsothe on. The simultaneous I—V response crucialthe step towardconfiguration the reliable integration of millions of molecular components into a SERS and represents SERS can aclarify atomic at the molecule—metal interface. The correlated working device. The development of the new measurement technique open the door to new science. measurements show selectivity towards one of the adsorption sites. Site-sensitivity represents a Thestep further development of the SERS measurement technique shouldcomponents reveal interesting crucial toward the reliable integration of millions of molecular into aphenomena working device. and useful insight in nanosciece and nanotechnology.

The development of the new measurement technique open the door to new science. The further development of the SERS measurement technique should reveal interesting phenomena Acknowledgments: This work was financially supported by Grants-in-Aids for Scientific Researchand (No.useful insight in nanosciece 26102013, 17K04971) and from nanotechnology. the Ministry of Education, Culture, Sports, Science and Technology (MEXT), and Asahi Glass, Mitsubishi, Tokuyama, Kato, and the Precise Measurement Technology Promotion foundations.

Acknowledgments: This work was financially supported by Grants-in-Aids for Scientific Research (No. 26102013, Author Contributions: M.I., S.F. and M.K. wrote the paper. 17K04971) from the Ministry of Education, Culture, Sports, Science and Technology (MEXT), and Asahi Glass, Mitsubishi, Tokuyama, and the Precise Technology Promotion foundations. Conflicts of Interest:Kato, The authors declare no Measurement conflicts of interest. Author Contributions: M.I., S.F. and M.K. wrote the paper. References Conflicts of Interest: The authors declare no conflicts of interest. 1. 2.

Aviram, A.; Ratner, M.A. Molecular rectifiers. Chem. Phys. Lett. 1974, 29, 277–283. Metzger, R.M.; Chen, B.; Höpfner, U.; Lakshmikantham, M.V.; Vuillaume, D.; Kawai, T.; Wu, X.; Tachibana, References H.; Hughes, T.V.; Sakurai, H.; et al. Unimolecular Electrical Rectification in Hexadecylquinolinium Tricyanoquinodimethanide. J. Am. Chem. Soc. 1997, 119, 10455–10466. 1. Aviram, A.; Ratner, M.A. Molecular rectifiers. Chem. Phys. Lett. 1974, 29, 277–283. [CrossRef] Song, H.; Kim, Chen, Y.; Jang,B.; Y.H.; Jeong, H.;U.; Reed, M.A.; Lee, T. Observation molecular orbital gating. Nature 2. 3.Metzger, R.M.; Höpfner, Lakshmikantham, M.V.; ofVuillaume, D.; Kawai, T.; Wu, X.; 2009, 462, 1039–1043. Tachibana, H.; Hughes, T.V.; Sakurai, H.; et al. Unimolecular Electrical Rectification in Hexadecylquinolinium

3.

Tricyanoquinodimethanide. J. Am. Chem. Soc. 1997, 119, 10455–10466. [CrossRef] Song, H.; Kim, Y.; Jang, Y.H.; Jeong, H.; Reed, M.A.; Lee, T. Observation of molecular orbital gating. Nature 2009, 462, 1039–1043. [CrossRef] [PubMed]

Sensors 2017, 17, 1901

4.

5. 6.

7. 8.

9.

10.

11. 12. 13. 14.

15. 16. 17.

18.

19. 20. 21. 22. 23. 24.

25.

12 of 14

Kiguchi, M.; Ohto, T.; Fujii, S.; Sugiyasu, K.; Nakajima, S.; Takeuchi, M.; Nakamura, H. Single Molecular Resistive Switch Obtained via Sliding Multiple Anchoring Points and Varying Effective Wire Length. J. Am. Chem. Soc. 2014, 136, 7327–7332. [CrossRef] [PubMed] Su, T.A.; Li, H.; Steigerwald, M.L.; Venkataraman, L.; Nuckolls, C. Stereoelectronic switching in single-molecule junctions. Nat. Chem. 2015, 7, 215–220. [CrossRef] [PubMed] Diez-Perez, I.; Hihath, J.; Lee, Y.; Yu, L.; Adamska, L.; Kozhushner, M.A.; Oleynik, I.I.; Tao, N. Rectification and stability of a single molecular diode with controlled orientation. Nat. Chem. 2009, 1, 635–641. [CrossRef] [PubMed] Fujii, S.; Ziatdinov, M.; Higashibayashi, S.; Sakurai, H.; Kiguchi, M. Bowl Inversion and Electronic Switching of Buckybowls on Gold. J. Am. Chem. Soc. 2016, 138, 12142–12149. [CrossRef] [PubMed] Fujii, S.; Tada, T.; Komoto, Y.; Osuga, T.; Murase, T.; Fujita, M.; Kiguchi, M. Rectifying Electron-Transport Properties through Stacks of Aromatic Molecules Inserted into a Self-Assembled Cage. J. Am. Chem. Soc. 2015, 137, 5939–5947. [CrossRef] [PubMed] Kiguchi, M.; Nakashima, S.; Tada, T.; Watanabe, S.; Tsuda, S.; Tsuji, Y.; Terao, J. Single-Molecule Conductance of π-Conjugated Rotaxane: New Method for Measuring Stipulated Electric Conductance of π-Conjugated Molecular Wire Using STM Break Junction. Small 2012, 8, 726–730. [CrossRef] [PubMed] Fujii, S.; Marque’s-Gonza’lez, S.; Shin, J.; Shinokubo, H.; Masuda, T.; Nishino, T.; Arasu, N.P.; Va’zquez, H.; Kiguchi, M. Highly-conducting molecular circuits based on antiaromaticity. Nat. Commun. 2017, 8. [CrossRef] [PubMed] Reed, M.A.; Zhou, C.; Muller, C.J.; Burgin, T.P.; Tour, J.M. Conductance of a molecular junction. Science 1997, 278, 252–254. [CrossRef] Kim, Y.; Pietsch, T.; Erbe, A.; Belzig, W.; Scheer, E. Benzenedithiol: A Broad-Range Single-Channel Molecular Conductor. Nano Lett. 2011, 11, 3734–3738. [CrossRef] [PubMed] Komoto, Y.; Fujii, S.; Nakamura, H.; Tada, T.; Nishino, T.; Kiguchi, M. Resolving metal-molecule interfaces at single-molecule junctions. Sci. Rep. 2016, 6. [CrossRef] [PubMed] Chuang, P.; Ho, S.; Smith, L.W.; Sfigakis, F.; Pepper, M.; Chen, C.; Fan, J.; Griffiths, J.P.; Farrer, I.; Beere, H.E.; et al. All-electric all-semiconductor spin field-effect transistors. Nat. Nanotechnol. 2015, 10, 35–39. [CrossRef] [PubMed] Smit, R.H.M.; Noat, Y.; Untiedt, C.; Lang, N.D.; van Hemert, M.C.; van Ruitenbeek, J.M. Measurement of the conductance of a hydrogen molecule. Nature 2002, 419, 906–909. [CrossRef] [PubMed] Kiguchi, M.; Stadler, R.; Kristensen, I.S.; Djukic, D.; van Ruitenbeek, J.M. Evidence for a Single Hydrogen Molecule Connected by an Atomic Chain. Phys. Rev. Lett. 2007, 98. [CrossRef] [PubMed] Kiguchi, M.; Tal, O.; Wohlthat, S.; Pauly, F.; Krieger, M.; Djukic, D.; Cuevas, J.C.; van Ruitenbeek, J.M. Highly Conductive Molecular Junctions Based on Direct Binding of Benzene to Platinum Electrodes. Phys. Rev. Lett. 2008, 101. [CrossRef] [PubMed] Hihath, J.; Arroyo, C.R.; Rubio-Bollinger, G.; Tao, N.; Agraït, N. Study of Electron-Phonon Interactions in a Single Molecule Covalently Connected to Two Electrodes. Nano Lett. 2008, 8, 1673–1678. [CrossRef] [PubMed] Moskovits, M. Surface-enhanced spectroscopy. Rev. Mod. Phys. 1985, 57, 783–826. [CrossRef] Fleischmann, M.; Hendra, P.J.; McQuillan, A.J. Raman spectra of pyridine adsorbed at a silver electrode. Chem. Phys. Lett. 1974, 26, 163–166. [CrossRef] Jeanmaire, D.L.; Van Duyne, R.P. Surface raman spectroelectrochemistry. J. Electroanal. Chem. 1977, 84, 1–20. [CrossRef] Albrecht, M.G.; Creighton, J.A. Anomalously intense Raman spectra of pyridine at a silver electrode. J. Am. Chem. Soc. 1977, 99, 5215–5217. [CrossRef] McQuillan, A.J. The discovery of surface-enhanced Raman scattering. Notes Rec. R. Soc. 2009, 63, 105–109. [CrossRef] Van Duyne, R.P. Laser Excitation of Raman Scattering from Adsorbed Molecules on Electrode Surfaces. In Chemical and Biochemical Applications of Lasers; Moore, C.B., Ed.; Elsevier Inc.: Evanston, IL, USA, 1979; Volume 4, pp. 101–185. Haynes, C.L.; Yonzon, C.R.; Zhang, X.; Van Duyne, R.P. Surface-enhanced Raman sensors: Early history and the development of sensors for quantitative biowarfare agent and glucose detection. J. Raman Spectrosc. 2005, 36, 471–484. [CrossRef]

Sensors 2017, 17, 1901

26. 27. 28. 29.

30.

31. 32. 33.

34.

35. 36.

37.

38. 39. 40. 41. 42.

43. 44. 45.

13 of 14

Nie, S.; Emory, S.R. Probing Single Molecules and Single Nanoparticles by Surface-Enhanced Raman Scattering. Science 1997, 275, 1102–1106. [CrossRef] [PubMed] Kneipp, K.; Wang, Y.; Kneipp, H.; Perelman, L.T.; Itzkan, I.; Dasari, R.R.; Feld, M.S. Single Molecule Detection Using Surface-Enhanced Raman Scattering (SERS). Phys. Rev. Lett. 1997, 78, 1667–1670. [CrossRef] Ru, E.C.L.; Meyer, M.; Etchegoin, P.G. Proof of Single-Molecule Sensitivity in Surface Enhanced Raman Scattering (SERS) by Means of a Two-Analyte Technique. J. Phys. Chem. B 2006, 110, 1944–1948. [PubMed] Dieringer, J.A.; Lettan, R.B., II; Scheidt, K.A.; Duyne, R.P.V. A Frequency Domain Existence Proof of Single-Molecule Surface-Enhanced Raman Spectroscopy. J. Am. Chem. Soc. 2007, 129, 16249–16256. [CrossRef] [PubMed] Tian, J.; Liu, B.; Li, X.; Yang, Z.; Ren, B.; Wu, S.; Tao, N.; Tian, Z. Study of Molecular Junctions with a Combined Surface-Enhanced Raman and Mechanically Controllable Break Junction Method. J. Am. Chem. Soc. 2006, 128, 14748–14749. [CrossRef] [PubMed] Kaneko, S.; Nakazumi, T.; Kiguchi, M. Fabrication of a Well-Defined Single Benzene Molecule Junction Using Ag Electrodes. J. Phys. Chem. Lett. 2010, 1, 3520–3523. [CrossRef] Kiguchi, M.; Murakoshi, K. Conductance of Single C60 Molecule Bridging Metal Electrodes. J. Phys. Chem. C 2008, 112, 8140–8143. [CrossRef] Liu, Z.; Ding, S.Y.; Chen, Z.B.; Wang, X.; Tian, J.H.; Anema, J.R.; Zhou, X.S.; Wu, D.Y.; Mao, B.W.; Xu, X.; et al. Revealing the molecular structure of single-molecule junctions in different conductance states by fishing-mode tip-enhanced Raman spectroscopy. Nat. Commun. 2011, 2, 305. [CrossRef] [PubMed] Ward, D.R.; Halas, N.J.; Ciszek, J.W.; Tour, J.M.; Wu, Y.; Peter Nordlander, P.; Natelson, D. Simultaneous Measurements of Electronic Conduction and Raman Response in Molecular Junctions. Nano Lett. 2008, 8, 919–924. [CrossRef] [PubMed] Ward, D.R.; Corley, D.A.; Tour, J.M.; Natelson, D. Vibrational and electronic heating in nanoscale junctions. Nat. Nanotechnol. 2011, 6, 33–38. [CrossRef] [PubMed] Ioffe, Z.; Shamai, T.; Ophir, A.; Noy, G.; Yutsis, I.; Kfir, K.; Cheshnovsky, O.; Selzer, Y. Detection of heating in current-carrying molecular junctions by Raman scattering. Nat. Nanotechnol. 2008, 3, 727–732. [CrossRef] [PubMed] Konishi, T.; Kiguchi, M.; Takase, M.; Nagasawa, F.; Nabika, H.; Ikeda, K.; Uosaki, K.; Ueno, K.; Misawa, H.; Murakoshi, K. Single Molecule Dynamics at a Mechanically Controllable Break Junction in Solution at Room Temperature. J. Am. Chem. Soc. 2013, 135, 1009–1014. [CrossRef] [PubMed] Xu, B.; Tao, N.J. Measurement of Single-Molecule Resistance by Repeated Formation of Molecular Junctions. Science 2003, 301, 1221–1223. [CrossRef] [PubMed] Matsuhita, R.; Horikawa, M.; Naitoh, Y.; Nakamura, H.; Kiguchi, M. Conductance and SERS Measurement of Benzenedithiol Molecules Bridging Between Au Electrodes. J. Phys. Chem. C 2013, 117, 1791–1795. [CrossRef] Cuevas, J.C.; Scheer, E. Molecular Electronics: An Introduction to Theory and Experiment; World Scientific: Singapore, 2010. Cuniberti, G.; Fagas, G.; Richter, K. Introducing Molecular Electronics; Springer: Berlin/Heidelberg, Germany, 2005. Kaneko, S.; Murai, D.; Marques-Gonzalez, S.; Nakamura, H.; Komoto, Y.; Fujii, S.; Nishino, T.; Ikeda, K.; Tsukagoshi, K.; Kiguchi, M. Site-Selection in Single-Molecule Junction for Highly Reproducible Molecular Electronics. J. Am. Chem. Soc. 2016, 138, 1294–1300. [CrossRef] [PubMed] Campion, A.; Ivanecky, J.E., III; Child, C.M.; Foster, M. On the Mechanism of Chemical Enhancement in Surface-Enhanced Raman Scattering. J. Am. Chem. Soc. 1995, 117, 11807–11808. [CrossRef] Chulhai, D.V.; Hu, Z.; Moore, J.E.; Chen, X.; Jensen, L. Theory of Linear and Nonlinear Surface-Enhanced Vibrational Spectroscopies. Annu. Rev. Phys. Chem. 2016, 67, 541–564. [CrossRef] [PubMed] Lombardi, J.R.; Birke, R.L.; Lu, T.; Xu, J. Charge-transfer theory of surface enhanced Raman spectroscopy: Herzberg–Teller contributions. J. Chem. Phys. 1986, 84, 4174–4180. [CrossRef]

Sensors 2017, 17, 1901

46. 47.

14 of 14

Li, Y.; Doak, P.; Kronik, L.; Neaton, J.B.; Natelson, D. Voltage tuning of vibrational mode energies in single-molecule junctions. Proc. Natl. Acad. Sci. USA 2014, 111, 1282–1287. [CrossRef] [PubMed] Vadai, M.; Nachman, N.; Ben-Zion, M.; Bürkle, M.; Pauly, F.; Cuevas, J.C.; Selzer, Y. Plasmon-Induced Conductance Enhancement in Single-Molecule Junctions. J. Phys. Chem. Lett. 2013, 4, 2811–2816. [CrossRef] © 2017 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).