Atomistic QM/Classical Modeling of Surface-Enhanced Infrared Absorption

TL;DR

This work develops a fully atomistic multiscale QM/$ω$FQ($μ$) framework to model surface-enhanced infrared absorption (SEIRA) spectra of molecules on plasmonic substrates, coupling a DFT description of the molecule with frequency-dependent, atomistic plasmonic models for metals and graphene. The method computes IR intensities from the coupled QM/MM system and applies it to adenine adsorbed on gold nanoparticles and graphene disks, benchmarking against SERS data and available experiments. Results show that adsorption orientation and substrate geometry critically shape SEIRA and SERS enhancements, with average SEIRA enhancements around $AEF\approx5$ and maximum enhancements ($MEF$) up to ~30, while SERS can reach higher factors. The framework provides a predictive tool for designing plasmonic SEIRA substrates and demonstrates graphene’s tunable IR response via size and Fermi energy control, although solvent effects and configurational sampling remain important future considerations.

Abstract

We present a multiscale quantum mechanics/classical (QM/MM) approach for modeling surface-enhanced infrared absorption (SEIRA) spectra of molecules adsorbed on plasmonic nanostructures. The molecular subsystem is described at the density functional theory (DFT) level, while the plasmonic material is represented using fully atomistic, frequency-dependent Fluctuating Charges ($ω$FQ) and Fluctuating Charges and Dipoles ($ω$FQF$μ$) models. These schemes enable an accurate and computationally efficient description of the plasmonic response of both graphene-based materials and noble metal nanostructures, achieving accuracy comparable to ab initio methods. The proposed methodology is applied to the calculation of SEIRA spectra of adenine adsorbed on gold nanoparticles and graphene sheets. The quality and robustness of the approach are assessed through comparison with surface-enhanced Raman scattering (SERS) spectra and available experimental data. The results demonstrate that the proposed framework provides a reliable route to simulate vibrational responses of plasmon-molecule hybrid systems.

Figures (8)

• Figure 1: Molecular structure and atom labelling of adenine.
• Figure 2: Molecular configurations of ADE adsorbed on the vertex (a) and face (b) positions of Au$_{10179}$ Ih NP.
• Figure 3: Normalized SEIRA (a)) and SERS (b)) spectra of the six configurations of ADE adsorbed on the vertex and face of Au$_{10179}$ Ih. SERS spectra are calculated at the PRF of the gold substrate (2.21 eV/560 nm). Stars denote $i$-MEF. MEF values are also reported.
• Figure 4: Normalized SEIRA (left panel) and SERS (right panel) spectra, calculated by averaging N3/N9, N7/N10, N1/N10, and all (Calc) configurations of ADE on Au$_{10179}$ Ih, together with the experimental results.kundu2009adenine. The characteristic regions of the SEIRA and SERS spectra are highlighted.
• Figure 5: Graphical depiction of ADE adsorbed on a graphene disk.