Vector Beam-Enhanced Raman Decodes DNA Nucleobases

To overcome challenges in DNA nucleobase Raman analysis, such as strong backbone signals, spectral overlap, and background noise, we propose a surface-enhanced Raman scattering (SERS) method employing a k-space filter-assisted radial vector beam (RVB) coupled with gold nanospheres (AuNPs). A spatial frequency-modulated RVB is tightly focused to excite AuNPs, producing a localized plasmonic field with a strong longitudinal component that spatially overlaps with DNA segments located in the AuNP near-field region. This configuration achieves strong near-field enhancement while suppressing background noise. The orientation-dependent near-field selection is interpreted through the projection between the local plasmonic field and the vibrational polarizability derivatives of DNA modes. Under orientation-favorable configurations, backbone-related modes with weak field projection are relatively suppressed, whereas nucleobase-related modes with larger projection are preferentially enhanced. Experimental results show effective suppression of backbone signals (900-1250 cm-1) and clearer DFT-assisted assignment of nucleobase-related Raman bands (1350-1500 cm-1). Combined with density functional theory (DFT), overlapping peaks were accurately identified and decoupled. This approach provides a physical mechanism for selective enhancement and noise suppression, offering a label-free optical method for specific DNA nucleobase identification.