Hydrogen as an energy carrier has by far the highest gravimetric energy density, which, when produced by a renewable pathway would go a long in achieving the goal of carbon neutral and responsible economy. Photocatalytic water splitting is a promising approach for the production of solar hydrogen. The technological potential of this approach is unclear and it is not known how the choice of the components (semiconducting particles, electrolyte, mediator, etc.), the design of the system, and the operating conditions affect the performance of the system.Firstly, we present a steady 1-dimensional model of a single photocatalyst particle which couples semiconductor physics to account for the charge carrier generation and transport with the electrochemical kinetics governed by the butler-volmer equation. We account for the solar absorption and the charge carrier generation with Lambert-Beer’s law. We then solve for the carrier transport and conservation with Poisson’s equations and the hole and electron transport by drift diffusion currents. We assume the recombination in the particle to be a Shockley-Reed-Hall trap recombination. The semiconductor electrolyte interface is treated as an ideal Schottky contact. We show the effect of intrinsic parameters like doping concentration, electron affinity (band positions), relative permittivity on the operating current density. We also present the effect of particle size on the operating current density.Secondly, we present a transient 0-dimensional model, which accounts for ideal solar absorption, ideal charge generation and transport, and the kinetic characteristics at the catalytic sites for two sets of photocatalyst semiconductor particles. We solve for species transport (mediator species and protons) and conservation in the one or two reactor compartments separated by a semipermeable membrane. We evaluate this 0-dimensional model on a reactor level as well, comparing three different reactor types, these three types include: i) a single photocatalyst reactor, ii) a membrane-separated dual compartment reactor and iii) a membrane-separated dual compartment reactor with wire and two auxiliary electrodes. We put forth a theoretically limiting efficiency which considers no overpotentials and a realistic case which accounts for losses for each type of reactor. For these reactor designs, we establish a link between the thermodynamic limits and materials requirements, changes in material requirements when more realistic operation and losses are considered, and compare the three reactor designs.Conclusively, the steady 1-dimensional particle model allows us to evaluate the effect of intrinsic material properties of a single photocatalyst on the operating current density, and the transient 0-dimensional model giving us a broader evaluation on a reactor design level with respect to the employed photocatalysts and redox mediators.