Photoexcited Hot Electron Catalysis in Plasmon-Resonant Grating Structures with Platinum, Nickel, and Ruthenium Coatings.

We report the electrochemical potential dependence of photocatalysis produced by hot electrons in plasmon-resonant grating structures. Here, corrugated metal surfaces with a period of 520 nm are illuminated with 785 nm wavelength laser light swept as a function of incident angle. At incident angles corresponding to plasmon-resonant excitation, we observe sharp peaks in the electrochemical photocurrent and dips in the photoreflectance consistent with the conditions under which there is wavevector matching between the incident light and the spacing between the lines in the grating. In addition to the bare plasmonic metal surface (i.e., Au), which is catalytically inert, we have measured grating structures with a thin layer of Pt, Ru, and Ni catalyst coatings. For the bare Au grating, we observe that the plasmon-resonant photocurrent remains relatively featureless over the applied potential range from -0.8 to +1.2 V vs NHE. For the Pt-coated grating, we observe a sharp peak around -0.3 V vs NHE, three times larger than the bare Au grating, and near complete suppression of the oxidation half-reaction, reflecting the reducing nature of Pt as a good hydrogen evolution reaction catalyst. The photocurrent associated with the Pt-coated grating is less noisy and produces higher photocurrents than the bare Au grating due to the faster kinetics (i.e., charge transfer) associated with the Pt-coated surface. The plasmon-resonant grating structures enable us to compare plasmon-resonant excitation with that of bulk metal interband absorption simply by rotating the polarization of the light while leaving all other parameters of the experiment fixed (i.e., wavelength, potential, electrochemical solution, sample surface, etc.). A 64X plasmon-resonant enhancement (i.e., p-to-s polarized photocurrent ratio) is observed for the Pt-coated grating compared to 28X for the bare grating. The nickel-coated grating shows an increase in the hot-electron photocurrent enhancement in both oxidation and reduction half-reactions. Similarly, Ru-coated gratings show an increase in hot-electron photocurrents in the oxidation half-reaction compared to the bare Au grating. Plasmon-resonant enhancement factors of 36X and 15X are observed in the p-to-s polarized photocurrent ratio for the Ni and Ru gratings, respectively.

Voltage-Induced Inversion of Band Bending and Photovoltages at Semiconductor/Liquid Interfaces.

At semiconductor/liquid interfaces, the surface potential and photovoltages are produced by a combination of band bending and quasi-Fermi-level splitting at the semiconductor surface, which are usually treated in a qualitative fashion. As such, it is important to develop quantitative metrics for the band energies and photovoltaics at these interfaces. Here, we present a spectroscopic method for monitoring the photovoltages produced at semiconductor/liquid junctions. The surface reporter molecule mercaptobenzonitrile (MBN) is functionalized on the photoelectrode surface (p-type silicon) and is measured using in situ surface-enhanced Raman scattering (SERS) spectroscopy with a water immersion lens under electrochemical working conditions. In particular, the vibrational frequency of the C≡N stretch mode (ωCN) around 2225 cm-1 is sensitive to the local electric field in solution at the electrode/electrolyte interface via the vibrational Stark effect. Over the applied potential range from -0.8 to 0.6 V vs Ag/AgCl, we observe ωCN to increase from 2220 to 2229 cm-1 (at low laser power). As the incident laser power is increased from 83.5 µW to 13.3 mW, we observe additional shifts of ΔωCN = ±1 cm-1, corresponding to photovoltages produced at the semiconductor/liquid interface ΔV = ±0.2 V. Based on Mott-Schottky measurements, the flat band potential (FBP) occurs at -0.39 V vs Ag/AgCl. For applied potentials above the FBP, we observe ΔωCN > 0 (i.e., blue-shifts ∼1 cm-1) corresponding to positive photovoltages, whereas for applied potentials below the flat band potential, we observe ΔωCN < 0 (i.e., red-shifts ∼1 cm-1) corresponding to negative photovoltages. These spectroscopic observations reveal voltage-induced changes in the band bending at the semiconductor/liquid junction that, thus far, have been difficult to measure.

Measuring Local pKa and pH Using Surface Enhanced Raman Spectroscopy of 4-Mercaptobenzoic Acid.

We report spectroscopic measurements of the local pH and pKa at an electrode/electrolyte interface using surface enhanced Raman scattering (SERS) spectroscopy of 4-mercaptobenzoic acid (4-MBA). In acidic and basic solutions, the protonated and deprotonated carboxyl functional groups at the electrode surface exist in the solution as -COOH and -COO-, which have different Raman active vibrational features at around 1697 and 1414 cm-1, respectively. In pH neutral water, as the applied electrochemical potential is varied from negative to positive, the acidic form of the 4-MBA (i.e., -COOH) decreases in Raman intensity and the basic form (i.e., -COO-) increases in Raman intensity. The change in local ion concentration is due to the application of electrochemical potentials and the accumulation of ions near the electrode surface. Under various applied potentials, the ratio of 1697 and 1587 cm-1 (pH-independent) peak areas spans the range between 0.7 and 0, and the ratio of the 1414 and 1587 cm-1 peak areas ranges from 0 to 0.3. By fitting these data to a normalized sigmoid function, we obtain the percentage of surface protonation/deprotonation, which can be related to the pKa and pH of the system. Thus, we can measure the local pKa at the electrode surface using the surface enhanced Raman signal of the 4-MBA.

Plasma-enhanced electrostatic precipitation of diesel exhaust particulates using nanosecond high voltage pulse discharge for mobile source emission control.

This study reports enhancement in the electrostatic precipitation (ESP) of diesel engine exhaust particulates using high voltage nanosecond pulse discharge in conjunction with a negative direct current (DC) bias voltage. The high voltage (20 kV) nanosecond pulses produce ion densities that are several orders of magnitude higher than those in the corona produced by a standard DC-only ESP. This plasma-enhanced electrostatic precipitator (PE-ESP) demonstrated 95 % remediation of PM and consumes less than 1 % of the engine power (i.e., 37 kW diesel engine at 75 % load). While the DC-only ESP remediation increases linearly with applied voltage, the plasma-enhanced ESP remains approximately constant over the applied range of negative DC biases. Numerical simulations of the PE-ESP process agree with the DC-only experimental results and enable us to verify the charge-based mechanism of enhancement provided by the nanosecond high voltage pulse plasma. Two different reactor configurations with different flow rates yielded the same remediation values despite one having half the flow rate of the other. This indicates that the reactor can be made even smaller without sacrificing performance. Here, this study finds that the plasma enhancement enables high remediation values at low DC voltages and smaller ESP reactors to be made with high remediation.

In Situ Investigation of Ultrafast Dynamics of Hot Electron-Driven Photocatalysis in Plasmon-Resonant Grating Structures.

Understanding the relaxation and injection dynamics of hot electrons is crucial to utilizing them in photocatalytic applications. While most studies have focused on hot carrier dynamics at metal/semiconductor interfaces, we study the in situ dynamics of direct hot electron injection from metal to adsorbates. Here, we report a hot electron-driven hydrogen evolution reaction (HER) by exciting the localized surface plasmon resonance (LSPR) in Au grating photoelectrodes. In situ ultrafast transient absorption (TA) measurements show a depletion peak resulting from hot electrons. When the sample is immersed in solution under -1 V applied potential, the extracted electron-phonon interaction time decreases from 0.94 to 0.67 ps because of additional energy dissipation channels. The LSPR TA signal is redshifted with delay time because of charge transfer and subsequent change in the dielectric constant of nearby solution. Plateau-like photocurrent peaks appear when exciting a 266 nm linewidth grating with p-polarized (on resonance) light, accompanied by a similar profile in the measured absorptance. Double peaks in the photocurrent measurement are observed when irradiating a 300 nm linewidth grating. The enhancement factor (i.e., reaction rate) is 15.6× between p-polarized and s-polarized light for the 300 nm linewidth grating and 4.4× for the 266 nm linewidth grating. Finite-difference time domain (FDTD) simulations show two resonant modes for both grating structures, corresponding to dipolar LSPR modes at the metal/fused silica and metal/water interfaces. To our knowledge, this is the first work in which LSPR-induced hot electron-driven photochemistry and in situ photoexcited carrier dynamics are studied on the same plasmon resonance structure with and without adsorbates.

Asymmetric response of interfacial water to applied electric fields.

Our understanding of the dielectric response of interfacial water, which underlies the solvation properties and reaction rates at aqueous interfaces, relies on the linear response approximation: an external electric field induces a linearly proportional polarization. This implies antisymmetry with respect to the sign of the field. Atomistic simulations have suggested, however, that the polarization of interfacial water may deviate considerably from the linear response. Here we present an experimental study addressing this issue. We measured vibrational sum-frequency generation spectra of heavy water (D2O) near a monolayer graphene electrode, to study its response to an external electric field under controlled electrochemical conditions. The spectra of the OD stretch show a pronounced asymmetry for positive versus negative electrode charge. At negative charge below 5 × 1012 electrons per square centimetre, a peak of the non-hydrogen-bonded OD groups pointing towards the graphene surface is observed at a frequency of 2,700 per centimetre. At neutral or positive electrode potentials, this 'free-OD' peak disappears abruptly, and the spectra display broad peaks of hydrogen-bonded OD species (at 2,300-2,650 per centimetre). Miller's rule1 connects the vibrational sum-frequency generation response to the dielectric constant. The observed deviation from the linear response for electric fields of about ±3 × 108 volts per metre calls into question the validity of treating interfacial water as a simple dielectric medium.

Monitoring Reaction Intermediates in Plasma-Driven SO2, NO, and NO2 Remediation Chemistry Using In Situ SERS Spectroscopy.

In situ surface-enhanced Raman scattering (SERS) spectroscopy is used to identify the key reaction intermediates during the plasma-based removal of NO and SO2 under dry and wet conditions on Ag nanoparticles. Density functional theory (DFT) calculations are used to confirm the experimental observations by calculating the vibrational modes of the surface-bound intermediate species. Here, we provide spectroscopic evidence that the wet plasma increases the SO2 and the NOx removal through the formation of highly reactive OH radicals, driving the reactions to H2SO4 and HNO3, respectively. We observed the formation of SO3 and SO4 species in the SO2 wet-plasma-driven remediation, while in the dry plasma, we only identified SO3 adsorbed on the Ag surface. During the removal of NO in the dry and wet plasma, both NO2 and NO3 species were observed on the Ag surface; however, the concentration of NO3 species was enhanced under wet-plasma conditions. By closing the loop between the experimental and DFT-calculated spectra, we identified not only the adsorbed species associated with each peak in the SERS spectra but also their orientation and adsorption site, providing a detailed atomistic picture of the chemical reaction pathway and surface interaction chemistry.

Measuring Local Electric Fields and Local Charge Densities at Electrode Surfaces Using Graphene-Enhanced Raman Spectroscopy (GERS)-Based Stark-Shifts.

We report spectroscopic measurements of the local electric fields and local charge densities at electrode surfaces using graphene-enhanced Raman spectroscopy (GERS) based on the Stark-shifts of surface-bound molecules and the G band frequency shift in graphene. Here, monolayer graphene is used as the working electrode in a three-terminal potentiostat while Raman spectra are collected in situ under applied electrochemical potentials using a water immersion lens. First, a thin layer (1 Å) of copper(II) phthalocyanine (CuPc) molecules are deposited on monolayer graphene by thermal evaporation. GERS spectra are then taken in an aqueous solution as a function of the applied electrochemical potential. The shifts in vibrational frequencies of the graphene G band and CuPc are obtained simultaneously and correlated. The upshifts in the G band Raman mode are used to determine the free carrier density in the graphene sheet under these applied potentials. Of the three dominant peaks in the Raman spectra of CuPc (i.e., 1531, 1450, and 1340 cm-1), only the 1531 cm-1 peak exhibits Stark-shifts and can, thus, be used to report the local electric field strength at the electrode surface under electrochemical working conditions. Between applied electrochemical potentials from -0.8 V to 0.8 V vs NHE, the free carrier density in the graphene electrode spans a range from -4 × 1012 cm-2 to 2 × 1012 cm-2. Corresponding Stark-shifts in the CuPc peak around 1531 cm-1 are observed up to 1.0 cm-1 over a range of electric field strengths between -3.78 × 106 and 1.85 × 106 V/cm. Slightly larger Stark-shifts are observed in a 1 M KCl solution, compared to those observed in DI water, as expected based on the higher ion concentration of the electrolyte. Based on our data, we determine the Stark shift tuning rate to be 0.178 cm-1/ (106 V/cm), which is relatively small due to the planar nature of the CuPc molecule, which largely lies perpendicular to the electric field at this electrode surface. Computational simulations using density functional theory (DFT) predict similar Stark shifts and provide a detailed atomistic picture of the electric field-induced perturbations to the surface-bound CuPc molecules.

Resonant and Selective Excitation of Photocatalytically Active Defect Sites in TiO2.

It has been known for several decades that defects are largely responsible for the catalytically active sites on metal and semiconductor surfaces. However, it is difficult to directly probe these active sites because the defects associated with them are often relatively rare with respect to the stoichiometric crystalline surface. In the work presented here, we demonstrate a method to selectively probe defect-mediated photocatalysis through differential alternating current (ac) photocurrent (PC) measurements. In this approach, electrons are photoexcited from the valence band to a relatively narrow distribution of subband gap states in TiO2 and then subsequently to the ions in solution. Because of their limited number, these defect states fill up quickly, resulting in Pauli blocking, and are thereby undetectable under direct current or continuous wave excitation. In the method demonstrated here, the incident light is modulated with an optical chopper, whereas the PC is measured with a lock-in amplifier. Thin (5 nm) films of TiO2 deposited by atomic layer deposition on various metal films, including Au, Cu, and Al, exhibit the same wavelength-dependent PC spectra, with a broad peak centered around 2.0 eV corresponding to the band-to-defect transition associated with the hydrogen evolution reaction (HER). While the UV-vis absorption spectra of these films show no features at 2.0 eV, photoluminescence (PL) spectra of these photoelectrodes show a similar wavelength dependence with a peak of around 2.0 eV, corresponding to the subband gap emission associated with these defect sites. As a control, alumina (Al2O3) films exhibit no PL or PC over the visible wavelength range. The ac PC plotted as a function of electrode potential shows a peak of around -0.4 to -0.1 V versus normal hydrogen electrode, as the monoenergetic defect states are tuned through a resonance with the HER potential. This approach enables the direct photoexcitation of catalytically active defect sites to be studied selectively without the interference of the continuum interband transitions or the effects of Pauli blocking, which is limited by the slow turnover time of the catalytically active sites, typically on the order of 1 µs. We believe that this general approach provides an important new way to study the role of defects in catalysis in an area where selective spectroscopic studies of these are few.

Monitoring Local Electric Fields at Electrode Surfaces Using Surface Enhanced Raman Scattering-Based Stark-Shift Spectroscopy during Hydrogen Evolution Reactions.

We report the use of surface-enhanced Raman scattering (SERS) to measure the vibrational Stark shifts of surface-bound thiolated-benzonitrile molecules bound to an electrode surface during hydrogen evolution reactions (HERs). Here, the electrode surface consists of Au nanoislands deposited both with and without an underlying layer of monolayer graphene on a glass substrate. The Stark shifts observed in the nitrile (C-N) stretch frequency (around 2225 cm-1) are used to report the local electric field strength at the electrode surface under electrochemical working conditions. Under positive (i.e., oxidative) applied potentials [vs normal hydrogen electrode (NHE)], we observe blue shifts of up to 7.6 cm-1, which correspond to local electric fields of 22 mV/cm. Under negative applied potentials (vs NHE), the C-N stretch frequency is red-shifted by only about 1 cm-1. This corresponds to a regime in which the electrochemical current increases exponentially in the hydrogen evolution process. Under these finite electrochemical currents, we estimate the voltage drop across the solution ( V = IR). Correcting for this voltage drop results in a highly linear electric field versus applied electrochemical voltage relation. Here, the onset potential for the HER lies around 0.2 V versus NHE and the point of zero charge (PZC) occurs at 0.04 V versus NHE, based on the capacitance-voltage ( C- V) profile. The solution field is obtained by comparing the C-N stretch frequency in solution with that obtained in air. By evaluating the local electric field strength at the PZC and the onset potential, we can separate the solution field from the reaction field (i.e., electrode field), respectively. At the onset of HER, the solution field is -0.8 mV/cm and the electrode field is -1.2 mV/cm. At higher ion concentrations, we observe similar electric field strengths and more linear E-field versus applied potential behavior because of the relatively low resistance of the solution, which results in negligible voltage drops ( V = IR).