Photocatalysis with plasmonic nanostructures has lately emerged as a transformative paradigm to drive and alter chemical reactions using light. At the surface of metallic nanoparticles, photoexcitation results in strong near fields, short-lived high-energy "hot" carriers, and light-induced heating, thus creating a local environment where reactions can occur with enhanced efficiencies. In this context, it is critical to understand how to manipulate the nonequilibrium processes triggered by light, as their ultrafast (femto- to picoseconds) relaxation dynamics compete with the process of energy transfer toward the reactants. Accurate predictions of the plasmon photocatalytic activity can lead to optimized nanophotonic architectures with enhanced selectivity and rates, operating beyond the intrinsic limitations of the steady state. Here, we report on an original modeling approach to quantify, with space, time, and energy resolution, the ultrafast energy exchange from plasmonic hot carriers (HCs) to molecular systems adsorbed on the metal nanoparticle surface while consistently accounting for photothermal bond activation. Our analysis, illustrated for a few typical cases, reveals that the most energetic nonequilibrium carriers (i.e., with energies well far from the Fermi level) may introduce a wavelength-dependence of the reaction rates, and it elucidates on the role of the carriers closer to the Fermi energy and the photothermally heated lattice, suggesting ways to enhance and optimize each contribution. We show that the overall reaction rates can benefit strongly from using pulsed illumination with the optimal pulse width determined by the properties of the system. Taken together, these results contribute to the rational design of nanoreactors for pulsed catalysis, which calls for predictive modeling of the ultrafast HC-hot adsorbate energy transfer.
Read full abstract