Publications

We combine large-scale atomistic amorphous carbon models generated with DynReaxMas and optical simulations based on the ωFQ method to uncover how disorder, curing, and local morphology shape plasmonic hot spots. The approach identifies four recurrent classes of field-enhancement sites—dangling bonds, stacked graphene-like sheets, carbon chains, and atomistic defects—and shows that cured structures can produce stronger and more localized enhancements, with values comparable to defective metallic nanojunctions.

We present plasmonX, an open-source code for simulating the plasmonic response of nanostructures. It combines high-performance implementations of the atomistic frequency-dependent fluctuating charges and dipoles approach (ωFQFμ) and the continuum Boundary Element Method (BEM), enabling accurate modeling of metallic and graphene-based systems and in-depth analysis of their optical properties.

We introduce a hybrid multiscale method (ωFQFμ-BEM) that models metal nanoparticles with an implicit continuum core and atomistic surface. Coupled with quantum mechanics, this framework reproduces optical properties and Surface-Enhanced Raman Scattering (SERS) with high accuracy at a fraction of the computational cost of fully atomistic approaches

We present the first fully atomistic multiscale method (QM/uFQFμ) to model surface-enhanced fluorescence (SEF) of molecules near plasmonic nanostructures. By coupling quantum mechanics with an atomistic electrodynamical model, the approach captures how nanoparticle morphology, defects, and atomistic features control fluorescence quenching and enhancement.

We analyze the time evolution of plasmonic excitations in pentagonal silver clusters (Ag25+, Ag43+) using time-dependent density functional theory (TDDFT) and real-time descriptors. Transition contribution maps and induced densities reveal distinct plasmon-like collective responses for longitudinal excitations and molecular-like transitions for transverse ones, with a characteristic delay of the plasmon signal relative to the driving pulse.

We extend the ωFQFμ atomistic model to describe bimetallic nanoparticles (Ag–Au alloys and core–shell systems). The method captures plasmon resonance shifts, intensity changes, and interband/tunneling effects with high accuracy, reproducing experimental trends where continuum or simplified models fail.

We present the ωFQFμ atomistic model, built on classical physics (Drude conduction, interband polarizability, and tunneling), can reproduce optical responses of Ag and Au nanoparticles usually attributed to quantum effects. It captures plasmon size shifts, induced charge distributions, and nonlocal effects with accuracy comparable to TDDFT, but at a fraction of its cost.

We develop a suite of time-resolved analysis tools—induced densities, differential PDOS, and transition contribution maps—combined with TDDFT and GW/BSE to track electron dynamics under ultrafast laser pulses. Applied to HBDI, DNQDI, LiCN, and Ag22, the approach reveals excited-state populations, charge transfer, and collective responses in real time, offering a robust framework to interpret ultrafast spectroscopy and electron dynamics

Using molecular dynamics, we screened zeolite frameworks for the selective separation of sucrose and 6-kestose in aqueous solution. Among 253 candidates, three extra-large pore zeolites (AET, DON, ETR) were studied in detail. Simulations show that while all three favor sucrose uptake, DON exhibits the highest selectivity and flux, making it a strong candidate for oligosaccharide purification.