Raman Scattering Enhancements Due to Super- and Subradiant Collective Plasmon Modes on Large-Area 2D-Au Arrays.

Ordered metal nanoparticle (MNP) arrays with ultrasmall interparticle gaps S exhibit strong enhancement of the electromagnetic (EM) near-field, known as hotspots, for surface-enhanced Raman spectroscopy (SERS) sensing. These arrays, with uniform gap sizes, are also essential for studying nonlinear Raman scattering effects and surface selection rules. Optical characterization of the fabricated large-area Au arrays with S ≪ r, where r is the MNP radius, revealed the excitation of hybridized-collective plasmon modes with giant EM near-field enhancement. We found that the SERS enhancement associated with a subradiant plasmon mode depends primarily on the interparticle gap distance S, rather than on the ordering of MNPs into arrays. However, arranging MNPs in the form of arrays influenced the far-field scattering of the super-radiant mode excited at a longer wavelength, resulting in lower but highly anisotropic SERS enhancements that depend on the far-field excitation polarization angle σ. This study on ordered MNP arrays with ultrasmall interparticle gaps S ≪ r highlights the roles of S and MNP ordering in SERS enhancement of an analyte. This understanding is pivotal for designing SERS substrates with very small interparticle gaps, as they generate a large number of intense and well-distributed SERS hotspots. Furthermore, an anisotropy-induced SERS dichroism effect was observed. Polarization-dependent SERS intensities varied based on the excitation wavelength λ(exc) and its corresponding Stokes wavelength positions related to the excited plasmon mode. As a result, the SERS dichroism of lower-frequency Stokes-shifted peaks exhibited a cos(2)(σ) dependence, whereas higher-frequency Stokes-shifted peaks exhibited a sin(2)(σ) dependence. This observation validates the EM near-field mechanism of SERS. The fabricated large-area 2D-Au arrays meet most of the essential requirements for efficient, robust, and reliable large-area SERS sensing.