Abstract Hydrogen production by solar-driven 2-step reduction–oxidation cycles has been successfully demonstrated in recent years. While the demonstrated efficiencies were quite low (up to 5.25% solar-to-fuel efficiency), there is a large potential for much higher efficiencies. This paper presents a detailed parametric investigation of a large scale (>100 kW) volumetric solar receiver-reactor, performed using a finite volume method model. The reactor temperature profile and extent of reduction are investigated by evaluating the effects of heat recovery, sweep gas mass flow, radiation flux and porosity. The current reactor concept is shown to have a temperature gradient across the absorber of 500–700 K, leading to a mean ceria reduction extent of 0.004–0.012, thus limiting the amount of hydrogen that can be generated to the order of several milligrams per cycle or less. It is also shown that heat recovery can reduce the temperature gradient across the absorber to 350 K, thus increasing the hydrogen generation significantly, up to four orders of magnitude higher than with no heat recovery (up to 5.25 g/cycle). The solar heat flux on the receiver can also significantly increase the hydrogen production and reduce the reduction step duration. Thus, the optimization of both design and operation of large volumetric reactors can increase the hydrogen generation rate significantly.