Model-Based Design of an Indirectly Irradiated Thermochemical Hydrogen Production Reactor Capable of Radiant Heat Recovery

Abstract

Renewable/green hydrogen is of great interest as an alternative fuel for decarbonizing sectors such as shipping, aviation, chemicals, and heavy industry. The high cost of green hydrogen through electrolysis, an off-the-shelf mature technology, has led researchers to explore alternative water-splitting methods including thermochemistry, which can also be used for cosplitting of H2O and CO2 to produce syngas that can be converted to liquid fuels. Moreover, the process can operate on stored high temperature heat, making 24/7 operation possible. This thesis focuses specifically on the two-step thermochemical redox cycle using non-stoichiometric metal oxides. While the process has been demonstrated at the lab and pilot scales, efficiencies have so far been limited by the large temperature swing between the reduction and oxidation conditions, resulting in high sensible heat losses. In our previous work, we have introduced the Reactor Train System (RTS), a concept that features multiple and identical individually sealed, indirectly irradiated, metal oxide-containing reactors which move between a hot reduction zone and a cooler oxidation zone, engaging in counterflow radiative heat recovery in between. Prior modeling of the RTS, which revealed promise for high efficiency and heat recovery effectiveness, used either zero- or one-dimensional models of the RTS reactors and assumed a basic reactor design that featured a sapphire window for radiative heat transfer between the source and the redox material. A detailed conceptual design and higher-fidelity modeling of the RTS reactors is the focus of this thesis. This thesis is a comprehensive documentation of the model-based iterative design process of a novel thermochemical hydrogen reactor with highly unique and challenging functional requirements, from initial concept to early prototyping. The primary engineering challenge is that the structural pressure vessel also acts as the heat transfer interface, and must serve both purposes while undergoing extreme thermal cycling. The original windowed reactor concept is first investigated using a radiative heat transfer model, with findings of unfavorable heat losses and concerns regarding practicality guiding us towards a reactor design using a fully ceramic vessel acting also as a heat transfer interface. A more advanced thermomechanical model was then used to select a geometry which we call the Multi-Tubular Radiative Recovery Reactor (MiTR3 ) instead of one larger ceramic vessel, and to study the design parameters of the MiTR3 such as tube wall thickness with critical insight into the stress and failure probability of the ceramic tubes. Besides its mechanical strength and favorable thermal properties, this design is scalable and adaptable to different operating conditions and redox materials. Moreover, it utilizes easy to assemble off-the-shelf components. We then further augmented our modeling capabilities with multidimensional, time-dependent thermo-fluid and chemical reaction physics, incorporating both reduction and oxidation kinetics into the conservation equations for full-cycle simulations using ceria as the metal oxide. This enabled further study of the impact of important parameters, especially operational parameters such as redox material loading and form factor, gas flow rates, etc., and a deeper understanding of realistic system level efficiencies and productivities that take into consideration the impact of auxiliary components such as vacuum pumping and gas separation technologies on both. Finally, our ongoing experimental work with a benchtop-scale, single-tube reactor prototype aimed at derisking components and validating modeling results is presented, alongside plans for future prototyping efforts.