Viscosity and glass formation behavior of lunar regolith simulants

Melt viscosity is central to choices of lunar processing technologies, including additive manufacturing, molten regolith electrolysis, and sintering processes, all of which may be central to future lunar exploration and in-situ resource utilization (ISRU). Here, we investigate the viscosity of a diverse set of lunar regolith simulants, specifically LX-M100, LX-T100, LMS-1, LHS-1, JSC-1A, EAC-1A, OPRL2N, OPRL2NT, OPRH2N, and OPRH3N. High-temperature concentric cylinder viscometry together with differential scanning calorimetry (DSC) / ‘Shift Factor’-derived viscosities were combined to describe viscosity across a broad temperature range, from superliquidus conditions to temperatures just above the glass transition ( T g ). Selected fitted VFT-curves were compared with data measured by fast differential scanning calorimetry (FDSC). Post-viscometry glasses, obtained from reheated and then quenched samples, were analyzed using electron probe microanalysis (EPMA) and rock magnetic measurements. The results of these indicate post-viscometry heterogeneities (nano/microlites / liquid immiscibility) for some of the samples leading to their exclusion from calorimetric study but not impacting the interpretation of high-temperature viscosity measurements. Empirical and machine-learning-based viscosity models were assessed against experimental data, with the configurational-entropy model of Sehlke and Whittington (2016) performing robustly across the full range, while the recent gpvisc framework ( Le Losq et al., 2025 ) yielded the lowest errors at high temperatures and near T g These findings emphasize the importance of melt homogeneity and careful characterization to ensure accurate T g -based viscosity predictions, vital for diverse lunar ISRU technologies especially under well-characterized redox conditions.