Thermophysical property evolution during molten regolith electrolysis. (15th September 2022)

• Record Type: Journal Article
• Authors: Humbert, Matthew S.; Brooks, Geoffrey A.; Duffy, Alan R.; Hargrave, Chad; Rhamdhani, M. Akbar
• Is Part Of: Planetary and space science. Volume 219(2022)
• Publication Date: 2022-09-15
• Subjects: Molten oxide electrolysis -- ISRU; Space sciences -- Periodicals; Atmosphere, Upper -- Periodicals; Sciences spatiales -- Périodiques; Haute atmosphère -- Périodiques; 523
• DOI: 10.1016/j.pss.2022.105527
• Languages: English
• ISSNs: 0032-0633

Abstract

Abstract: The electrolysis of lunar regolith for oxygen generation has been under investigation since before the Apollo missions. One critical and overlooked aspect in work to date has been how the composition and physical properties of the molten regolith change during electrolysis process as oxides are reduced. The regolith composition also varies across the surface of the moon, and thus the molten properties will also vary. In this work we aggregate previous empirical models for physical properties of mixtures of molten oxides spanning the possible composition ranges of the electrolyte. The physical properties most important to reactor operation are liquidus temperature, electrical conductivity, and viscosity. In this work we show that these properties vary dramatically for compositions present on the lunar surface and during electrolysis. From this model review we illustrate that the liquidus temperature and heat capacity of the melt rises during electrolysis while viscosity, electrical resistivity, and thermal conductivity increase with the removal of iron oxide and decrease with the removal of silicon oxide. This new framework provides Molten Regolith Electrolysis (MRE) system designers a more accurate method for evaluating system performance during operation. In addition to lunar applications, the models reviewed herein are applicable to those seeking to develop Molten Oxide Electrolysis (MOE) for terrestrial, carbon-free metal production. The code developed during this project has been released on github file exchange; a link is provided as supplementary material. Highlights: Physical property models disagree with one another. Temperature and composition have large effect on properties. Uncertainty quantification has not been performed. Composition and properties change during processing.