A computational scheme to predict the thermodynamic ability of photocatalysts to drive watersplitting is applied to a number of polymeric systems in order to explain their apparent inability to oxidise water.

We study the thermodynamic ability of carbon nitride materials to act as water splitting photocatalysts using a computational approach that involves a combination of density functional theory (DFT) and time-dependent DFT (TD-DFT) calculations on cluster models of both triazine-and heptazine-based structures. We first use TD-DFT to calculate the absorption spectra of the different cluster models and compare these spectra to those measured experimentally and then calculate using DFT and TD-DFT the reduction potentials of the free electron, free hole, and exciton in these models. We predict that all classes of carbon nitride structures considered should thermodynamically be able to reduce protons and oxidize water. We further provide evidence for the hypothesis that the experimental lack of overall water splitting activity for pure carbon nitride arises from the fact that water oxidation is a four-hole reaction and hence very susceptible to competition with electron−hole recombination. Finally, we propose that the recently reported overall water splitting activity of carbon nitride loaded with polypyrrole nanoparticles arises from a junction formed at the interface of both materials, which assists in keeping electrons and holes apart.

Based on insights from computational chemistry calculations, the ability of polymers to act as water splitting photocatalysts for the production of renewable hydrogen from water and sunlight is discussed. Specifically, the important role of exciton dissociation in these materials is highlighted, as well as the possible microscopic origins of the experimentally observed changes in the photocatalytic activity of a polymer with increasing chain length or changing chemical composition. The reason why water oxidation, with polymeric photocatalysts, is difficult, and which polymer properties to target when developing new polymers for water splitting photocatalysis are, finally, also discussed.

ESI-1 TP 3 stacked structuresFig. S1 Calculated spectra of TP3 crystal structure (CS1), a stack of two and three of these molecules (CS2 and CS3 respectively) and a laterally arranged CS2 stacks