Pathways to Boost Solar-to-Hydrogen Efficiencies for Z-Scheme Photocatalytic Reactors: Learnings from Equivalent Circuit and Continuum Multiphysics Models

Abstract

Z-scheme semiconductor particles present in an aqueous solution with soluble redox shuttles provides a promising two-step approach to split water to produce hydrogen and oxygen using sunglight. This approach facilitates efficient utilization of the incident solar spectrum with tandem light-absorbers and promotes safe reactor operation with hydrogen and oxygen produced in physically distinct locations. Thermodynamically, these systems are projected to achieve solar-to-hydrogen (STH) efficiencies that are comparable to the electrode architectures, which are as large as 34% with idealized light absorbers and perfect reaction selectivity1. Up to 4% STH efficiencies were theoretically projected for state-of-the-art light absorbers and cocatalysts with diffusion-driven redox shuttle transport between reaction sites2. However, practical demonstrations are limited to efficiencies that are significantly smaller (less than 1%)3. Two main factors impeding actual performance in these systems include: (a) undesired reactions of any redox species, including hydrogen, oxygen, or the redox shuttle species, on the photocatalyst site; and (b) mass-transfer limitations of reactant/product species between oxygen and hydrogen evolution reaction sites. In this work, I will present materials- and device-level insights obtained from physics-based models of varying complexities. An expanded, zero-dimensional (0-D), equivalent-circuit model is developed to probe the effectiveness of semipermeable oxide coatings to achieve reaction selectivity. This model introduces powerful innovations to model parallel/competing redox reactions at the valence and conduction bands, while also accounting for mass-transfer limitations. Moreover, our framework provides a tractable approach to quantify the multidimensional dependencies of coating transport and kinetic parameters, concentration of light absorbers, and species mass-transfer rates on the STH efficiencies. At the reactor-scale, we assess the viability of natural convective currents to enhance species mixing. Temperature and concentration profiles are predicted from a two-dimensional (2-D), coupled flow, heat, and mass-transfer model of two reaction compartments separated by a membrane. Modeling predictions are compared to and validated against experimental measurements on dual-compartment reactor prototypes. Enhanced rate of species transport is predicted in the presence of thermal convection in addition to diffusion. Improved species mixing is observed for the horizontal, as compared to the vertically stacked, arrangement of reaction compartments, and is attributed to attenuation of incident sunlight by the membrane present along the optical path. With 2-cm tall reaction compartments, the minimum redox shuttle concentration needed to sustain a 10% STH efficiency decreases by 80% in the presence of natural convection as compared to diffusion. Overall, these results provide new insights and strategies to realize highly effective materials-to-device scale designs for Z-scheme photocatalytic solar fuel reactors with soluble redox shuttles. References (1) Keene, S.; Bala Chandran, R.; Ardo, S. Calculations of Theoretical Efficiencies for Electrochemically-Mediated Tandem Solar Water Splitting as a Function of Bandgap Energies and Redox Shuttle Potential. Energy Environ. Sci. 2019, 12 (1), 261–272. https://doi.org/10.1039/C8EE01828F.(2) Bala Chandran, R.; Breen, S.; Shao, Y.; Ardo, S.; Weber, A. Z. Evaluating Particle-Suspension Reactor Designs for Z-Scheme Solar Water Splitting via Transport and Kinetic Modeling. Energy Environ. Sci. 2018, 11 (1), 115–135. https://doi.org/10.1039/C7EE01360D.(3) Wang, Z.; Hisatomi, T.; Li, R.; Sayama, K.; Liu, G.; Domen, K.; Li, C.; Wang, L. Efficiency Accreditation and Testing Protocols for Particulate Photocatalysts toward Solar Fuel Production. Joule 2021, 5 (2), 344–359. https://doi.org/10.1016/j.joule.2021.01.001.