Creating a Redox Materials Database for Solar-Thermochemical Air Separation and Fuels Production

Josua Vieten*, Dorottya Guban, Martin Roeb, Christian Sattler, Institute of Solar Research, DLR (German Aerospace Center), Germany; Patrick Huck, Matthew Horton, Kristin Persson, Lawrence Berkeley National Laboratory, USA; Brendan Bulfin, ETH Zurich, Switzerland

15th Annual NH3 Fuel Conference, Pittsburgh, PA, October 31, 2018
NH3 Energy+ Topical Conference at the AIChE Annual Meeting


Converting heat from renewable sources into other forms of energy is considered an essential factor in the reduction of greenhouse gas emissions. For instance, high temperatures can be reached using concentrated solar power (CSP), and the thus-captured energy can be converted into so-called solar fuels via thermochemical processes. These consist of the partial reduction of a redox material, usually a metal oxide, at high temperatures following the exothermic re-oxidation of this material at a lower temperature level using steam or CO2, which are thus converted into hydrogen or carbon monoxide, respectively. These two gases can be combined to generate syngas for the production of hydrocarbons (see Fig. 1). Through the same process, a stream of mostly inert gas can be produced by re-oxidation with air, allowing air separation using renewable energy sources. Hydrogen production and air separation can also provide the feedstock for ammonia production through the Haber-Bosch process, as the achieved oxygen partial pressures can be kept low enough to avoid catalyst poisoning. [2] Ammonia produced through this method can be used for fertilizer production, or as a fuel for energy storage.

Achieving efficient air separation and fuels production through solar-thermochemical processes is challenging but possible. Finding suitable redox materials depending on the respective process conditions through evaluation of the materials thermodynamics is a key point in reaching high process efficiencies. [1, 3-5] Within a materials screening for these applications, we prepared perovskite solid solutions with the general composition AxB1-xMyN1-yO3-¦Ä with A, B = Ca, Sr and M, N = Ti, Mn, Fe, Co, Cu using a modified Pechini method. [5] Their redox enthalpies and entropies as a function of the non-stoichiometry ¦Ä can be tuned by adjusting their composition. We obtained experimental data gathered via equilibrium oxygen uptake and release measurements using thermogravimetric analysis, and theoretical data gathered via density functional theory (DFT) calculations. The experimental data, i.e., redox enthalpies and entropies, are fit using a novel empirical model, in order to generate interactive isotherms, isobars, as well as graphs at constant non-stoichiometry which are referred to as isoredox plots (see Fig. 2). In a joint effort between the German Aerospace Center and the Lawrence Berkeley National Laboratory, the data is used to create a search engine for redox materials data based upon the infrastructure of The Materials Project. [6] The data is included into MPContribs [7], which is the framework for external contributors to publicly share their data on the Materials Project website. Many of the functions included in this contribution are based on the databases included in The Materials Project.

Moreover, theoretical data is collected for a large set of perovskite redox materials, including solid solutions in a large compositional range. This data is generated by pre-evaluating possible stable candidate materials using the Goldschmidt tolerance factor and subjecting this data to Density Functional Theory (DFT) based calculations. The resulting structural data contains the energies of all atoms in the structure and allows to calculate redox enthalpies, and, via the elastic tensors of the materials and their composition, redox entropies. Thus, a full model of thermodynamic properties is generated. Both experimental and theoretical datasets are used to create a redox materials database. By defining specific target process conditions, such as the reduction and oxidation temperatures and oxygen partial pressures, it is possible to find the most efficient redox material for each specific application using the novel perovskite search engine.

These factors, in summary, accelerate the finding of new materials by replacing large sets of experiments by a computer-based pre-selection step. By finding many new redox materials and selecting the best candidates for further studies, we allow a major leap towards more efficient renewable energy conversion and storage, including ammonia production and solar fuels generation.

[1] J. Vieten, B. Bulfin, F. Call, M. Lange, M. Schmuecker, A. Francke, M. Roeb, C. Sattler, Journal of Materials Chemistry A, 4, 13652-13659 (2016)
[2] B. Bulfin, J. Lapp, D. Guban, J. Vieten, S. Richter, S. Brendelberger, M. Roeb, C. Sattler, submitted, (2018)
[3] J. Vieten, B. Bulfin, M. Senholdt, M. Roeb, C. Sattler, M. Schmuecker, Solid State Ionics 308, 149-155 (2017)
[4] B. Bulfin, J. Vieten, C. Agrafiotis, M. Roeb, C. Sattler, Journal of Materials Chemistry A, 5, 18951-18966 (2017)
[5] J. Vieten, B. Bulfin, M. Roeb, C. Sattler, Solid State Ionics 315, 92-97 (2018)
[6] A. Jain, S.P. Ong, G. Hautier, W. Chen, W.D. Richards, S. Dacek, S. Cholia, D. Gunter, D. Skinner, G. Ceder, K.A. Persson, APL Materials, 011002 (2013)
[7], RedoxThermoCSP contribution under construction.

Read the abstract at the AIChE website.


Download this presentation [PDF].


Institute of Solar Research, DLR
Lawrence Berkeley National Laboratory
Professorship of Renewable Energy Carriers, ETH Zurich
Learn more about the 2018 NH3 Fuel Conference

Leave a Reply