Biohybrid systems consisting of redox enzymes and colloidal semiconductor nanocrystals offer an exciting platform for driving complex chemical reactions with light. As Nature’s catalysts redox enzymes catalyze chemical transformations with an efficiency and selectivity rarely achieved synthetically. However, enzyme catalysis is often limited by the transport of electrons from a second protein. Nanocrystals can replace this transport protein by transferring photoexcitation electrons to the enzyme. This allows the catalysis to be driven by light and facilitates tuning of the rate of transfer. The interplay between the nanocrystal and enzyme have complex kinetics which are difficult to model analytically. Thus, we turn to kinetic Monte Carlo (KMC), a technique which simulates the time evolution of a system based on a kinetic model. To highlight the value of these methods we examine two systems: CdS nanorods with a hydrogenase and CdS quantum dots with a nitrogenase.

In the nanorod–hydrogenase system we first investigated the factors which limit the rate of electron transfer. By simulating a range of excitation frequencies, nanorod/enzyme ratios, and hole transfer rates we were able to predict which regimes are limited by excitation rate as opposed to hole transfer efficiency, or other factors such as the turnover of the enzyme itself. However, transferring electrons to the enzyme is only the first step in catalysis, so we then expanded our studies to explore the often neglected back-reactions in this system: back transfer of an electron from the enzyme to the nanorod, and oxidation of H2 back into protons. In particular, the experimental data shows a counterintuitive trend in which more catalyst can decrease the catalytic yield. By simulating both back electron transfer and H2 oxidation we were able to demonstrate that significant H2 oxidation must occur to accurately model experimental trends. Overall, the trends identified in this work provide a framework for identifying and controlling the limiting reactions under various experimental conditions in nanocrystal-enzyme photochemistry.

We then explored the role of nanocrystal surface chemistry in a system composed of CdS quantum dots and a nitrogenase enzyme. We synthesized quantum dots capped with either 3-mercaptopropionic acid (MPA) or 4-mercaptobenzoic acid (MBA). It was spectroscopically determined that the MBA-capped dots had slower electron-hole recombination rates as well as faster hole transfer rates. This would lead one to expect that the MBA-capped dots would result in more efficient electron transfer and more product formation, however the experimental results are more nuanced. The nitrogenase enzyme produces both NH3 and H2, and while the system containing MBA-capped dots did produce more NH3, it produced less hydrogen. The result was that the total conversion yield of electrons to chemical products was nearly identical between the two quantum dot surfaces. The MBA capped dots did not make more overall product but did create a higher selectivity for NH3 over H2. KMC simulations were used to demonstrate that this could be explained by increased back electron transfer from the nitrogenase to MBA over MPA capped dots. The simulation shows that this not only explains the similar total yields, as fewer transferred electrons from MBA are converted into product, but also is able to explain the improved selectivity. The faster forward and back electron transfer results in the nitrogenase enzyme spending less time in hydrogen producing states, while still reaching the NH3 production state.

Ultimately these projects demonstrate how KMC modeling can be used to understand the complex interplay of processes in nanocrystal–enzyme biohybrids. By developing a clear mechanistic understanding of such systems, we can improve our design of nanocrystal–enzyme complexes with a focus on tuning the most relevant handles