Using Surface-Enhanced Raman Spectroscopy to Probe Surface-Localized Nonthermal Plasma Activation

Low-temperature, nonthermal plasmas generate a complex environment even when operated in nonreactive gases. Plasma-produced species impinge on exposed surfaces, and their thermalization is highly localized at the surface. Here we present a Raman thermometry approach to quantifying the resulting degree of surface heating. A nanostructured silver substrate is used to enhance the Raman signal and make it easily distinguishable from the background radiation from the plasma. Phenyl phosphonic acid is used as a molecular probe. Even under moderate plasma power and density, we measure a significant degree of vibrational excitation for the phenyl group, corresponding to an increase in surface temperature of ∼80 °C at a plasma density of 2 × 1010 cm–3. This work confirms that surface-localized thermal effects can be quantified in low-temperature plasma processes. Their characterization is needed to improve our understanding of the plasma-induced activation of surface reactions, which is highly relevant for a broad range of plasma-driven processes.

Raman thermometry with argon plasma exposure.Figure 1a shows a schematic of the optical system to measure Raman scattering signal from PPA adsorbed on the lab-made SERS substrates.
A continuous wave laser at λ=532 nm was irradiated to the sample inside the chamber.Power of laser was controlled by the density filters (Supplementary Figure 3).Both Raman and Rayleigh scattering were collimated by a lens (focal distance = 5 cm) and focused to a monochromator (Acton series, Princeton Instruments).A double notch filter was positioned ahead of the monochromator slit to minimize the Rayleigh scattering signal from the sample.Spectra, taken with a 1 second of acquisition time, were averaged (10 spectra in total) using a software (LightField, Princeton Instruments).This method is well described in our previous study. 1gon plasma exposure was performed using a commercial environmental chamber (HVC-DRP-5 Harrick Scientific).The chamber was modified by replacing one of the KBr windows with

S3
Pyrex tubular reactor (See Figure 1b), allowing the irradiation of plasmas onto the sample.A copper electrode wrapped around the reactor was connected to an RF power supply (RFPP RF-5S, Advanced Energy) and a matching network (MFJ-989D, MFJ) to ignite plasmas.A flow of 20 sccm of argon gas was flowed into the system, controlled by mass flow controllers (MKS 1179C, MKS Instruments).The pressure was maintained at 3.7 torr.The temperature of the system was regulated using a temperature controller (ATK-024-3, Harrick Scientific).
The actual power to sustain plasmas was obtained by measuring the RF voltage, current, and phase difference during the discharge using an oscilloscope (TES2024C, Tektronix). 2 It is observed that approximately 10% of the power supplied by the RF power source is effectively coupled to the discharge, consistent with our previous experiences with similar small plasma reactors at low power regime.

Plasma density measurements.
We performed capacitive probe measurements to obtain the plasma density (i.e., ion density) of argon plasma as a function of RF input power.A copper wire tip with a diameter of 2.5 mm and a length of 0.5 mm was inserted into the chamber by replacing one of the KBr windows.The probe tip was then connected to a 200 pF capacitor and an RF power supply (RFPP RF-5S, Advanced Energy) capable of generating square wave functions for pulsed operation.The capacitor voltage was measured using an oscilloscope (TES2024C, Tektronix).The capacitor was charged negatively through pulses generated by the RF power supply, then discharged by the ion flux from the argon plasma toward the probe. 3nal of Physical Chemistry Letters S4 It was assumed that Ar ions penetrate the sheath edge at the Bohm velocity.It is necessary to know the electron temperature of argon plasma for calculating the Bohm velocity.The electron temperature was estimated by balancing the rate of ionization and ion wall losses. 4This approach is well summarized in our previous report. 5We obtained a value of 5.5 eV for the electron temperature and used that value to calculate the plasma density.Supplementary Figure 4 shows the measured ion density as a function of RF input power.
FTIR measurements.The same setup used for Raman measurements has been employed for the FTIR measurements.The reaction chamber (HVC-DRP-5 Harrick Scientific) was mounted on a Praying Mantis diffuse reflectance adapter (DRP-XXX, Harrick Scientific).The adapter was then equipped in a FTIR spectrometer (iS50, Thermo Fisher Scientific).All spectra were obtained through liquid-nitrogen-cooled HgCdTe detector and averaged via 16 scans at a resolution of 4 cm -1 .The experimental conditions for plasma exposure are also exactly the same as those used for the Raman measurements for the consistency.We set the ramp rate at 10 ℃/min through the temperature controller (ATK-024-3, Harrick Scientific).