Bias Voltage Induced Surface-Enhanced Raman ... - ACS Publications

Subscriber access provided by ECU Libraries

C: Surfaces, Interfaces, Porous Materials, and Catalysis

Bias Voltage Induced SERS Enhancement on the Single Molecule Junction Satoshi Kaneko, Koji Yasuraoka, and Manabu Kiguchi J. Phys. Chem. C, Just Accepted Manuscript • DOI: 10.1021/acs.jpcc.8b11595 • Publication Date (Web): 01 Mar 2019 Downloaded from http://pubs.acs.org on March 7, 2019

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 22 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Bias Voltage Induced SERS Enhancement on the Single Molecule Junction Satoshi Kaneko, Koji Yasuraoka, Manabu Kiguchi*

Department of Chemistry, Tokyo Institute of Technology, 2-12-1 W4-10 Ookayama, Meguroku, 152-8551 Tokyo, Japan

1 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 22

ABSTRACT

We have studied the surface enhanced Raman scattering (SERS) from the aminobenzenethiol (ABT) and benezenedithiol (BDT) single molecules bridging Au electrodes (single molecule junction) at different bias voltages. The SERS intensity of the ABT single molecule junction increased with the bias voltage, and the non-totally symmetric b2 mode appeared at the high bias voltage. Meanwhile, the SERS intensity did not change with the bias voltage in the case of the BDT single molecule junction. The bias voltage induced SERS intensity and appearance of the b2 mode for the ABT single molecule junction can be explained by the resonance effect. The energy difference between the metal occupied state and the lowest unoccupied molecular orbital (LUMO) of ABT deceased with an increase in the bias voltage. The charge transfer resonance taking place between the metal occupied state and the LUMO was, thus, allowed at higher bias voltages, which caused the enhancement of SERS intensity and appearance of the b2 mode.


Page 3 of 22 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1. Introduction Since the discovery of surface enhanced Raman scattering (SERS) in 1970s, SERS becomes a powerful technique to study the low concentration analytes or even single molecules.1-2 The SERS remarkable sensitivity and non-destructive nature have found its way into a number of fields including analytical chemistry, surface science, biochemistry and medicine.3-6 The SERS remarkable sensitivity has its origin in the dramatic increase of Raman scattering that takes place for analytes in the vicinity of metal nano structures. Typical enhancement factors for SERS are in the range of 104 – 1014 depending on experimental conditions. Raman enhancement is the product of two contributions: electromagnetic (EM) and chemical (CM) contributions.7-8 The EM contribution takes into account the enhancement of electromagnetic fields on metallic surfaces, while the CM accounts for resonance Raman scattering resulting from the charge transfer resonance taking place between metal and the molecular orbital. The EM contribution can be controlled by the shape and size of the metal nano structures, and distance between the structures.9-10 While the EM contribution has been intensively studied with well-defined metal nano structures and understood, there are still unclear points in the CM contribution. It is because the CM contribution is sensitive to the atomic structure of the metal-molecule interface, and the precise determination and control of the metal-molecule interface structure are still difficult for samples with detectable SERS. In this study, we have tried to control the CM contribution by the bias voltage applied 3 ACS Paragon Plus Environment


Page 4 of 22

to the single molecule bridging metal electrodes (single molecule junction). The single molecule junction has attracted wide attention because of its potential application in ultra small electronic devices, such as diode, memory, switch, transistor.11-13 The single molecule junction is also an ideal test bed for investigating the fundamentals of electron transport at the nanoscale.14 The tunnelling electron transport, hopping electron transport through single molecules, and inelastic electron tunnelling effect have been investigated for the single molecule junctions. Furthermore, the single molecule junction is a suitable system for SERS, because the gaps of a few nanometers between metal electrodes have been identified as hot spots with particularly strong Raman enhancement.15-16 The SERS study on the single molecule junction has been reported for benzenedithiol (BDT), bipyridine (BPY) and several other small molecules.17-18 The presence of a molecule at the junction is confirmed by the SERS measurement, which cannot be identified by the solo measurement of electrical conductance. However, the CM contribution has not be investigated on the single molecule junction, due to the low stability of the single molecule junction at room temperature. We prepared highly stable nano gap electrodes using the lithographic technique, which enabled us to do detail studies on SERS of the single molecule junction. Using the highly stable nano gap electrode, we revealed the effect of the bias voltage on the SERS spectra for the aminobenzenethiol (ABT) and BDT single molecule junction. In the case of the ABT single molecule junction, the fraction of samples with detectable SERS (signal intensity is more than 4 ACS Paragon Plus Environment

Page 5 of 22 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


10CPS) and SERS intensity increased with the bias voltage. Together with the increase in SERS intensity, the non-totally symmetric b2 mode appeared at high bias voltages. These experimental results can be explained by the charge transfer resonance taking place between metal occupied state and the lowest unoccupied molecular orbital (LUMO) of ABT.

2. Experimental The Au nano gap electrodes were prepared through a series of standard nanofabrication techniques.17 An insulating SiO2 film (thickness: 1 μm) was deposited on the polished phosphor bronze substrate (thickness: 0.5 mm) with sputtering. The SiO2 layer provides electrical insulation and limits the intensity of the Raman background scattering. On the SiO2 layer, a polyimide film was deposited with spin-coating. The nanosized Au junction (the narrowest constriction: 300 nm×150 nm) was prepared on the polyimide-coated substrate by means of electron beam lithography and lift-off processing. The Cr and Au films (3 nm/130 nm) were thermally deposited on the substrate. Subsequently, the polyimide underneath the Au junctions was removed by isotropic reactive ion etching using O2 plasma (80 W). A freestanding Au nano bridge (length: 2 μm) was fabricated on the substrate through these processes (Fig. 1c, Fig. S1). The substrate was mounted on a custom made mechanically controllable break-junction (MCBJ) system, consisting of a stacked piezo-element (NEC tokin) and two fixed counter supports (Fig. 1ab, Fig. S2). The ABT or BDT molecules were adsorbed on the 5 ACS Paragon Plus Environment


Page 6 of 22

unbroken Au nano bridge using self-assembly from 1mM ABT or BDT ethanol solution. The Au nano bridge was stretched and eventually broken by gradually bending the substrate using a piezoelectrically controlled push-rod. All measurements were performed at room temperature. The electrical measurements were performed with a Keithley 428 programmable amplifier. The SERS signals were collected using a NanoFinder30 Raman microprobe (Tokyo Instruments) with a near-infrared laser (λex = 785 nm) as an excitation light. The laser beam was focused onto the junctions using an objective lens with 50× magnification and 0.50 numerical aperture. The laser spot diameter was ~1 μm. The incident light was polarized parallel to the junction axis. The conductance and SERS of the junction were continuously measured during the breaking process of the junction. The duration time of the Raman spectra was 960 ms. After this SERS measurement, the bias voltage was swept from +1 V to -1V within 5 ms to get the I-V curve.

Figure 1: (a) Photo image of the experimental setup for the SERS measurement of the single 6 ACS Paragon Plus Environment

Page 7 of 22 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


molecule junction. (b) Schematic view of the experimental setup, (c) SEM image of the Au nano gap electrode, (d) Molecular structure of Amino benzenethiol (ABT) and benzenedithiol (BDT) (e) Typical conductance trace of the Au junction covered with the BDT molecules during the breaking the junction.

3. Result and Discussion The formation of the single molecule junction was confirmed by the electrical conductance of the junction. Figure 1e shows the typical conductance trace of the Au junction covered with BDT molecules. The conductance traces showed conductance step around 0.01 G0 (2e2/h) after showing the 1 G0 step, which corresponded to the Au atomic junction. The previously reported study on the ABT and BDT revealed that the conductance of single molecule junction was 0.01 G0.17, 19 Therefore, the 0.01 G0 step appeared in the conductance trace (Fig. 1(e)) corresponds to the formation of the BDT single molecule junction. The SERS signal was strongly enhanced when the conductance of the junction was 0.01 G0 (Fig. S3), indicating that the SERS signal came from the single molecule bridging metal electrodes. Figure 2(a,b) show the SERS spectra of the ABT and BDT single molecule junctions. The SERS spectrum of the ABT single molecule junction showed two distinct Raman bands, which were assigned to a C-S stretching mode (~1070 cm-1, ν7a) and a C-C stretching mode (~1580 cm-1, ν8a).20 The C-S stretching mode (ν1) and the C-C stretching mode (ν8a) appeared 7 ACS Paragon Plus Environment


Page 8 of 22

at 1065 cm-1 and 1560 cm-1 for the BDT single molecule junction.21-22 For comparison, we show the Raman spectra of the bulk ABT and BDT in Fig. 2a, b. The energy of the vibrational modes for the single molecule junction was smaller than those for the bulk. The red shift of the vibrational modes can be explained by the metal-molecule interaction at the interface. When a molecule adsorbs on the metal surface, a metal orbital and molecular orbital interact with each other, and the chemical bond is formed between the metal and molecule.21 In forming the chemical bond, the electron transfers from a bonding molecular orbital (e.g. highest occupied molecular orbital (HOMO)) to metal unoccupied states, and from metal occupied states to an antibonding molecular orbital (e.g. LUMO). Both electron transfers weaken intra-molecular bonds resulting in the observed red shift of the vibrational modes for the single molecule junctions. In the case of the ABT, the photo induced dimerization reactions have been reported on Au nano particles during the SERS measurements.23 However, the absence of 1380cm-1 and 1430cm-1 peaks corresponding to the ABT dimer indicated that dimerization reaction did not take place at the junction in the present experimental condition.23 Figure 2c, d show the typical SERS spectra of the ABT and BDT single molecule junctions at different bias voltages. The SERS intensity increased with the bias voltage, and an additional CH bending mode (~ 1142 cm-1, ν9b) appeared at higher bias voltages for the ABT single molecule junction. Meanwhile, the shape and intensity of SERS did not change with the bias voltage for the BDT single molecule junctions. The obtained results did not depend on the 8 ACS Paragon Plus Environment

Page 9 of 22 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


polarity of the applied bias voltage (Fig. S4).

Figure 2: Example of the SERS spectrum of the (a) ABT and (b) BDT single molecule junction (red), together with the bulk Raman (black). Examples of SERS spectra for the (c) ABT and (d) BDT single molecule junctions at different bias voltages (0V: black, 0.1 V: purple, 0.2 V: orange).

The SERS intensity and fraction of samples with detectable SERS were statistically analysed for the ABT and BDT single molecule junctions. Figure 3 shows the average SERS intensities of the ν8a mode and fraction of samples with detectable SERS for the ABT and BDT single molecule junction. Here, we defined the sample with detectable SERS as one for which the ν8a mode displayed over 10 counts per second. Both average SERS intensity and fraction of samples with detectable SERS increased with the bias voltage for the ABT single molecule junctions, while they did not change with the bias voltage for the BDT single molecule junction.



Page 10 of 22

Figure 3: (a, b) The average SERS intensity (8a) and (c, d) fraction of samples with detectable SERS as a function of the bias voltage, constructed from 4098(1355), 5602(3792) and 3177(2715) at 0, 0.1, and 0.2 V for ABT, while 5128(3085), 6867(2038) and 8386(3692) at 0, 0.1, and 0.2 V for BDT single molecule junctions, respectively. The number of samples with detectable SERS are shown in parentheses.


Page 11 of 22 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 4: (a) Example of SERS spectra of the ABT single molecule junction at different bias voltages (0 V: black, 0.1 V: purple, 0.2 V: orange). The intensity of SERS spectra is normalized by the intensity of the ν8a mode. (b) Average intensity of ν8a mode, fraction of sample with detectable ν8a mode (Pν8a) and ν9b mode (Pν9b), and the intensity of ν9b mode (A9b) as a function of the bias voltage. The intensity of ν9b signal is normalized by the intensity of ν8a mode.

The enhancement of the SERS intensity and appearance of the b2 mode (ν9b) for the ABT single molecule junction are discussed in the following. It should be noticed that the intensity of the b2 mode was significantly enhanced compared to other vibrational modes at the high bias voltage. Figure 4(a) shows the example of normalized SERS spectra of the ABT single molecule junction at different bias voltages. Here, the intensity of SERS spectra was normalized by the intensity of the normal totally symmetric ν8a mode. The relative intensity of the b2 mode increased with the bias voltage. Figure 4b shows the average normalized intensity 11 ACS Paragon Plus Environment


Page 12 of 22

of the ν9b mode, fraction of samples with detectable ν9b and ν8a modes, and average SERS intensity of the ν8a mode. It is clear that all four values correlated with each other, and they increased with the bias voltage. Since the ABT and BDT molecules have similar molecular backbone, the bias induced change in SERS spectra for ABT should originate from its characteristic electronic structure. It is well known that the charge transfer resonance taking place between metal occupied state and the LUMO contributes to the enhancement of SERS signal for the ABT molecules adsorbed on the Au substrates.24

Ikeda et al. showed the

electrochemical potential dependence of SERS spectra of ABT on Au surface. They collected SERS spectrum of ABT SAMs on the Au smooth substrate by creating a sandwich-like structure using Au nano particles (Au nano particles-ABT molecules-Au film junction). They controlled the energy difference between the Fermi level of the Au electrode and the LUMO by the electrochemical potential.25 At the resonant condition, where the energy difference agreed with the energy of the incident light (1.96 eV), the SERS intensity of normal a1 modes increased together with an increase in the background and appearance of the b2 mode, which agreed with the present experimental results of the ABT single molecule junction at high bias voltages. The enhancement of the SERS intensity and appearance of the b2 mode could be explained by the resonance effect. We then discuss the reason why the resonance effect became significant for the ABT single molecule junction at high bias voltages. Under the application of the bias voltage on the 12 ACS Paragon Plus Environment

Page 13 of 22 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


single molecule junction, the molecular orbital shifts depending on the ratio of the electronic coupling to metal leads (Right and left electrodes).26-27 When a molecule strongly interacts with one of the electrodes, the molecular level follows the Fermi level of the metal electrode which strongly interacts with the molecule. Meanwhile, the Fermi level of the right and left electrodes shifts by same amount with different direction relative to the molecular orbital, when the bias voltage is applied to the single molecule junction where the electronic coupling is symmetric. The ratio of electronic coupling to the metal leads can be evaluated by the I-V curve of the single molecule junction.28-29 Figure 5 is an example of I-V curve of the ABT single molecule junction. In a single level tunnelling model, the transmission probability (τ(E)) is represented by 𝜏(𝐸) = (𝛤

4𝛤𝐿𝛤𝑅

2 2 𝐿 + 𝛤𝑅) + (E ― 𝜀0)

(1)

where ε0 and ΓL(R) are the energy of the conduction orbital and the electronic coupling energy between the molecule and the left (right) electrode, respectively. Integration of the transmission probability within an energy window given by the chemical potentials of the electrodes results in an analytical expression for the I-V curve of the single molecule junction given by 8𝑒

{

𝐼(𝑉) = ℎ 𝑠(1 ― 𝑠)𝛤 arctan

(

)} (2)

𝑠𝑒𝑉 ― 𝜀0

(1 ― 𝑠)𝑒𝑉 + 𝜀0

𝛤

𝛤

) + arctan (

where Γ = ΓL + ΓR and s = ΓL / (ΓL+ΓR). The ratio of the electronic coupling (ΓL/ΓR) is obtained by fitting the measured I-V curve to equation (2). In the case of the ABT single molecule junction, the average ratio was determined to be 1.0 (0.1). Therefore, the Fermi level of the 13 ACS Paragon Plus Environment


Page 14 of 22

right and left electrodes would shift by same amount with different direction relative to the molecular orbital, when the bias voltage is applied to the ABT single molecule junction. The energy difference between the Fermi level of the one of the electrodes and the LUMO would decrease with an increase in the bias voltage, and it could match with the energy of the incident light at the certain bias voltage. Under this resonant condition, SERS intensity would be significantly enhanced, which agreed with the present experimental results. Figure 6 shows the schematic image of the photo induced transition from the metal occupied state to the LUMO for the ABT single molecule junction at low and high bias voltages. In the case of the ABT single molecule junction, the energy difference between the LUMO and the Fermi level (~2eV) is larger than the energy of the incident light (1.58eV). 7, 25 Here it should be noticed that the LUMO are energetically broadened due to the hybridization between molecular orbital and metal orbitals. There are finite unoccupied molecule states below the centre of the LUMO states, although the density of states decreases with a decrease in energy. Therefore, the photo induced transition from the metal occupied state to the unoccupied molecular states can be possible at high bias voltages (0.1~0.2eV). Meanwhile, the LUMO level is much higher (>3 eV) for the BDT single molecule junction.30-31 Therefore, the photo induced transition from the metal occupied state to the unoccupied molecular states cannot be possible in the present experimental condition (bias voltage

Bias Voltage Induced Surface-Enhanced Raman ... - ACS Publications

Recommend Documents