Plasmonic Au Array SERS Substrate with Optimized Thin Film Oxide Substrate Layer

Abstract

This work studies the effect of a plasmonic array structure coupled with thin film oxide substrate layers on optical surface enhancement using a finite element method. Previous results have shown that as the nanowire spacing increases in the sub-100 nm range, enhancement decreases; however, this work improves upon previous results by extending the range above 100 nm. It also averages optical enhancement across the entire device surface rather than localized regions, which gives a more practical estimate of the sensor response. A significant finding is that in higher ranges, optical enhancement does not always decrease but instead has additional plasmonic modes at greater nanowire and spacing dimensions resonant with the period of the structure and the incident light wavelength, making it possible to optimize enhancement in more accessibly fabricated nanowire array structures. This work also studies surface enhancement to optimize the geometries of plasmonic wires and oxide substrate thickness. Periodic oscillations of surface enhancement are observed at specific oxide thicknesses. These results will help improve future research by providing optimized geometries for SERS molecular sensors.

Keywords: SERS; array; computational electromagnetics; grating; nano-optics; nanowires; plasmonics; thin film.

Figure 1

Depiction of the modeled grating of Au plasmonic nanowires bonded to a SiO₂ thin film by a Ti adhesion layer atop a Si substrate. (a) Shows a schematic of the simulated structure under the presence of λ₀ = 785 nm light incident normal to the surface and polarized in the x-direction, perpendicular to the length of the nanowires. The parameters swept in this study were wire width (w), electrode spacing (s), and SiO₂ thickness (t_SiO2); (b) Optical enhancement hotspots in 2D cross-sections of structures modeled as infinite wires with t_SiO2 = 330 nm, s = 10 nm, and (i) optimized value at w = 10 nm, (ii) median value at w = 200 nm, and (iii) minimum value at w = 400 nm.

Figure 2

(a) Plot of line average enhancement versus w at s = 10 nm, 50 nm, 100 nm, and 500 nm; (b) plots the enhancement values from the peaks in (a) at different spacing values, w_peak; and plots the enhancement value for a constant width, w = 50 nm, for spacing values in (a). The vertical purple dashed line in (a) illustrates from where the data for w = 50 nm in (b) is obtained, and the orange arrows show points used for w_peak. (b) Further, plot includes an additional data point not shown in (a) at s = 200 nm.

Figure 3

(a) Color plot of line average enhancement versus s and w with t_SiO2 = 200 nm. The arrows point to (i) optimized geometry values w = 70 nm and s = 10 nm, (ii) secondary peak geometry values w = 150 nm and s = 320 nm, and (iii) standard fabrication geometries w = 200 nm and s = 200 nm. The white dotted line indicates a resonant mode corresponding to the period of the structure at 785 nm, the same as λ₀; (b) Sweep of line average enhancement versus t_SiO2 at (i), (ii), and (iii); (c) Second iteration of (a) with maximum t_SiO2 = 330 nm found from (b). The arrows point to (v) the optimized enhancement value occurring at w = 50 nm and s = 10 nm, (iv) secondary peak geometry values w = 130 nm and s = 360 nm, and (iii) standard fabrication geometries w = 200 nm and s = 200 nm; (d) Second sweep of line average enhancement versus t_SiO2 at (iii), (iv), and (v) to optimize t_SiO2. The geometry is optimized at t_SiO2 = 50, 330, or 590 nm, w = 50 nm and s = 10 nm; however, a second peak occurs at t_SiO2 = 630 nm, w = 130 nm and s = 360 nm, which is more easily fabricated.

Figure 4

Color plot of line average enhancement versus s and w at three t_SiO2 values. (a) Maximized enhancement at t_SiO2 = 330 nm, (b) median at t_SiO2 = 290 nm, and (c) minimized at t_SiO2 = 250 nm.