Abstract
Plasmon-driven chemical reactions are a subject that is currently capturing the attention of the research community and generates a fair amount of arguments about their origin. Taking into account that the lifetime of excited hot carriers in metals is very short, some mechanism is required to store carriers long enough and in sites that allow chemical reactions with the environment. One established mechanism is the injection of charges into either the valence or conduction band of a semiconductor, followed by a chemical reaction at the semiconductor surface. Here, we consider a somewhat less-explored pathway by which plasmon decay can cause a chemical reaction: the direct excitation of hybridized surface states by plasmons. Using a simple model, we evaluate theoretically the rate of direct excitation and find that it can be comparable and often exceed the rate of indirect excitation of surface states. Our findings correspond to prior experimental results. We also identify the conditions under which one can enhance the direct excitation efficiency and, thus, bring plasmon-driven photochemistry closer to practical applications.