Non-Uniform Porous Structures and Cycling Control for Optimized Fixed-Bed Solar Thermochemical Water Splitting

Abstract

Solar thermochemical redox cycles provide a sustainable pathway for solar fuel processing. If done in porous (ceria) structures, they can profit from faster reaction rates owed to the enhanced heat and mass transport characteristics. However, the exact porous structure and operating conditions significantly affect the performance. We present a transient volume-averaged fixed-bed model of a thermochemical redox reactor utilizing macroporous ceria. We studied the porosity-dependent (ɛ = 0.4–0.9) and operating condition-dependent (solar concentration ratio, ratio of oxygen partial pressure to total pressure, and gas flowrate) performance of the fixed-bed ceria redox cycle. Structures with large porosity (ɛ = 0.9) showed better performance than low-porosity structures, owning to the enhanced heat absorption and resulting higher temperatures. We show that the cycle duration requires optimization according to the porosity of the structure. Two hours of operation for a structure with ɛ = 0.75 resulted in the largest hydrogen production (115.78mLgceria−1) if the single cycle duration was 240 s (i.e., 30 cycles in 2 h), while nearly five times less was produced for a 15 times longer single cycle duration (i.e., two cycles in 2 h). We subsequently introduced porous structures with different types of non-uniform porosity distributions. For an average porosity of ɛ = 0.75, the most favorable non-uniform porosity media exhibited higher porosity at the boundaries and a denser core. The fuel production of the best non-uniform porous structure was six times larger compared to a uniform porous structure. Adjusting on top of this the cycling conditions, a 14.6 times production gain was achieved. This work suggests that under non-isothermal operation condition for macroporous ceria redox fixed-bed cycling, non-uniform porous structure with higher porosity boundaries and a dense core benefit fuel production and porosity-dependent cycle duration modulation can be used to increase performance.