Local Protonation Control using Plasmonic Activation

All SERS and TERS measurements are performed with an experimental setup described previously.1 Laser radiation is focused using an oil immersion microscope objective (40X, 1.35NA, Olympus). The scattered signal is collected with the same objective and passes through a dichroic mirror and notch filter before it enters the spectrometer (Acton Advanced SP2750 A, SI GmbH, Germany). The control experiments under continuous flow of argon and temperature dependent studies are performed in a Biocell (JPK, Germany).

In the present study 4-mercaptopyridine (4-MPY) was chosen as a model compound for plasmon induced protonation.The adsorption of 4-MPY on metal surfaces can occur via three potential sites (1) the sulphur function, (2) via lone pair electrons of the nitrogen and (3) via aromatic p-electrons of the ring.9][20][21] Furthermore, pH dependent surface enhanced Raman scattering (SERS) studies of 4-MPY show that at pH 4 12 the unprotonated (UP) compound is observed.This is indicated by a single Raman peak at 1575 cm À1 (ring stretching mode with unprotonated nitrogen).3][24] Protonation on SERS substrates has been studied on various molecular systems 22,[25][26][27] using different experimental techniques and all these studies show a pH dependency.Generally, the protonation is reversible and depending on the pH conditions, protonation and deprotonation can be observed.
For the detailed investigation of such surface dependent reactions, a site specific detection at single molecule level would be extremely valuable.Tip-enhanced Raman scattering (TERS) provides a tool to investigate structural information of surfaces with nanometre resolution using a plasmonic hot spot at a scanning probe tip to probe the sample interface.9][30][31][32] Recently it has been shown that surface plasmon induced chemical reactions can be monitored using TERS under ambient conditions and also under ultrahigh vacuum conditions. 12,13In the present study we use a combined SERS and TERS approach to investigate the protonation of 4-MPY.
Initially SERS spectra of 4-MPY adsorbed on silver island films using a laser wavelength of 532 nm at 125 mW were measured over several minutes.Fig. 1a shows 10 selected spectra measured at different time-points.Initially Raman band at 1575 cm À1 (indicating the unprotonated compound) with high intensity and a very small protonated band at 1608 cm À1 is visible.Gradually the Raman band at 1608 cm À1 increases indicating the protonation of 4-MPY.Thus a conversion from unprotonated to protonated 4-MPY is monitored.Fig. 1b clarifies this conversion in the form of a colour coded intensity plot.The results suggest that a proton source must be present, as no further reaction of the 4-MPY (e.g.decomposition) could be detected.Different hydrogen containing components in air could be responsible for the observations.As the dissociation of H 2 8 and H 2 O 9-11 under surface plasmon conditions were already shown, this further strengthens our assumption of protonation of 4-MPY using surface plasmon under ambient conditions.Fig. 1c shows the intensity of unprotonated and protonated peak as time progresses.One can see that the intensity of the protonated peak increases initially and reaches to a maximum.We also noticed that the intensity of different peaks shown in Fig. 1a decrease as time progresses.This could be attributed to an oxidation of the silver island film during the measurement.This effect will not be discussed further as it will not affect the main conclusion of this work.
Considering H 2 O or H 2 under ambient as a possible hydrogen sources required for the surface plasmon induced protonation reaction, we performed time dependent measurements under a continuous flow of argon.Fig. 2a shows the time development of 10 spectra at different times-points.The data clearly indicate that no protonation takes place.Fig. 2b shows all time dependent SERS spectra as a colour coded intensity plot.
4][35] Consequently, employing  the same incident laser power (50 mW) and 632 nm laser excitation (Fig. 3b) compared to 532 nm (Fig. 3a), no protonation could be detected, thus the intensity of the surface plasmon with 632 nm excitation is not sufficient to start the reaction.Low power (18 mW) of 532 nm laser excitation (see Fig. 3c) shows also no protonation signature.Thus a minimum intensity of surface plasmon is the decisive factor to initiate the protonation which can be controlled using either incident laser power or excitation wavelength.
The results of different excitation wavelengths at 50 mW further imply that the protonation reaction is not induced due to a temperature increase in the laser focus since under same laser power the temperature should be approximately the same.To further the temperature dependence we performed temperature dependent experiments.A low incident power (12 mW @ 532 nm) was used such that no protonation could be observed at room temperature, increasing the temperature of the substrate up to 60 1C (data not shown), did not yield the protonation either.
In comparison to SERS where many nanoparticles are present in the laser focus and contribute to the overall Raman signal, in tip-enhanced Raman scattering (TERS) the signal is generated from a silver coated AFM tip of about 10 nm in radius and thus reducing the enhancing unit to a single nanoparticle.In a TERS experiment the silver coated AFM tip is positioned in the laser focus, thus only one specific plasmonic feature contributes to the signal.In order to warrant similar conditions a monolayer of 4-MPY was immobilized on a flat transparent gold nanoplate 36,37 (see ESI † for detail).
Fig. 4a shows 10 selected TERS spectra under ambient conditions and Fig. 4b shows the time-dependent TERS spectra plotted as a colour coded intensity plot.Interestingly under TERS conditions a much faster protonation of 4-MPY was observed compared to SERS.A control TERS measurement was also performed under inert atmosphere.Fig. 4c shows 10 selected spectra of TERS measurement under inert atmosphere at different timepoints and Fig. 4d shows the corresponding time dependent TERS spectra plotted as a colour coded intensity plot.The results nicely confirm the role of atmospheric conditions as seen in the SERS experiments.The difference between Au and Ag as the binding site has surprisingly little effect to the band positions of the protonated and unprotonated peaks.
With respect to the ''instantaneous'' protonation, previous theoretical and experimental studies showed that field confinement between two metal nanoparticles leads to an increased enhancement. 38,39This effect also occurs in a particle-on-metalsurface geometry (''gap-mode'') as in the case of the TERS experiment.Furthermore the tip-sample nanogap allows an efficient electron transfer from tip to sample. 40,41These effects can explain a faster protonation under TERS conditions with very defined and optimized geometry, whereas under SERS conditions many sites with different efficiencies contribute to the overall signal changes.Hence, in the case of the SERS experiments a gradual protonation is observed.To exclude any power related aspects a SERS experiment also using 650 mW (data not shown) was done and a comparable behavior regarding an increase in the reaction rate was observed which also  agrees with recent power dependent SERS measurement on 4-MPY. 24We also like to note a difference in the appearance of a broad peak around 980 cm À1 in TERS measurements.This peak is due to silicon tip and its nature changes from tip to tip.Since the TERS experiments with and without argon were performed with different tips, a direct comparison regarding the spectral background is difficult.
In conclusion, we report the experimental observation of protonation reaction under ambient condition using 4-MPY as a model system.While the actual mechanism of the protonation cannot be revealed with the presented experiments, it demonstrates the conditions under which the surface catalytic reaction can be controlled.The control experiment under a continuous flow of argon confirms that atmospheric H 2 O (H 2 cannot be fully excluded yet) acts as a potential proton source required for the reaction.The study further demonstrates that the intensity of the surface plasmon is a key factor to initiate the protonation reaction.The TERS experiments not only confirm the findings of the SERS, but also demonstrate a site specific protonation catalyst that can be located at specific sites and synchronously act as a structurally specific sensor.Interestingly, despite the lower number of plasmonic particles in the case of TERS the protonation happen much faster compared to SERS.This can be attributed to large electromagnetic fields produced in the metal-metal nanogap between tip and substrate and an efficient electron transfer.Consequently, the observed reaction depends more on the specific nanoparticle activity rather than the number of particles.This opens interesting perspective for site specific protonation with nanometre control.
Financial Support from the European Union and the state of Thuringia (FKZ: 2011 FE 9048; 2011 VF 0016) as well as through the Deutsche Forschungsgemeinschaft (FR 1348/19-1) is gratefully acknowledged.

Scheme 1
Scheme 1 Protonation of 4-MPY in the presence of metal nanoparticles and light.

Fig. 2
Fig. 2 (a) Time-dependent SERS spectra of 4-MPY under a continuous flow of argon during the experiment.Laser excitation at 532 nm/150 mW, integration time 1 s.(b) Colour coded intensity plot of same series in wavenumbers range from 1550-1630 cm À1 .No protonation is observed.

Fig. 3
Fig. 3 (a) Colour coded SERS intensity plot of 4-MPY under 532 nm laser radiation with laser power of 50 mW with an integration time of 2 s, clearly show protonation as a function of time.(b) SERS intensity plot of 4-MPY under 632 nm/50 mW laser radiation, integration time 4 s; no protonation is observed (c) SERS intensity plot of 4-MPY under 532 nm/18 mW laser radiation, integration time of 10 s; under low power conditions no protonation is observed either.