LUNAR IN SITU RESOURCE UTILIZATION

Abstract

An exemplary embodiment of the present disclosure provides a process for extracting metals and/or metalloids from extraterrestrial regolith. The process can comprise providing a sample of extraterrestrial regolith, wherein the sample can comprise an oxide of one or more metals and/or metalloids, and chemically reacting the oxide of the one or more metals and/or metalloids with carbon via carbothermal reduction to produce one or more metals and/or metalloids.

Description

CROSS-REFERENCE TO RELATED APPLICATIONS

Statement Regarding Federally Sponsored Research or Development

Not Applicable

THE NAMES OF THE PARTIES TO A JOINT RESEARCH AGREEMENT

Not Applicable

SEQUENCE LISTING

Not Applicable

STATEMENT REGARDING PRIOR DISCLOSURES BY THE INVENTOR OR A JOINT INVENTOR

Not Applicable

BACKGROUND OF THE DISCLOSURE

1. Field of the Invention

The various embodiments of the present disclosure relate generally to lunar in situ resource utilization, and, in particular, to systems and methods for extracting metals and/or metalloids from lunar regolith.

2. Description of Related Art

Human space exploration is predicated upon establishing a permanent presence on the Moon. The harvesting of existing lunar resources is essential for moving forward with a strong focus on in-situ resource utilization (ISRU). The extraction of valuable metals and metalloids contained in lunar regolith is of particular interest, including Si, Al, Mg, Ca, Fe, Ti, and Na, which are integral for the development of perovskites for optical glass and other industrial and scientific infrastructures. Accordingly, there is a need for improved systems and methods for harvesting metals and metalloids in a lunar environment. The present disclosure provides such systems and methods.

BRIEF SUMMARY OF THE INVENTION

An exemplary embodiment of the present disclosure provides a process for extracting metals and/or metalloids from extraterrestrial regolith. The process can comprise providing a sample of extraterrestrial regolith, wherein the sample can comprise an oxide of one or more metals and/or metalloids, and chemically reacting the oxide of the one or more metals and/or metalloids with carbon via carbothermal reduction to produce one or more metals and/or metalloids.

In any of the embodiments disclosed herein, chemically reacting the oxide comprises mixing the sample with a carbon-containing reducing agent to form a mixture, and directing concentrated solar irradiation to the mixture as process heat.

Another exemplary embodiment of the present disclosure provides a process for extracting metals and/or metalloids from a material. The process can comprise providing a sample of a material, wherein the sample can contain an oxide of one or more metals and/or metalloids; and chemically reacting the oxide of the one or more metals and/or metalloids with carbon via carbothermal reduction to produce one or more metals and/or metalloids, wherein chemically reacting the oxide can comprise contacting the sample with a carbon-containing reducing agent to form a mixture and directing concentrated solar irradiation to the mixture.

In any of the embodiments disclosed herein, chemically reacting the oxide can be carried out at a pressure between 3×10⁻¹⁵ bar and 1×10⁻¹² bar.

In any of the embodiments disclosed herein, chemically reacting the oxide can be carried out at a temperature between 750° C. and 2000° C.

In any of the embodiments disclosed herein, the extraterrestrial regolith can be lunar regolith.

In any of the embodiments disclosed herein, the carbon-containing reducing agent can be derived from a fixed carbon source.

In any of the embodiments disclosed herein, the carbon-containing reducing agent can be derived from a carbon source selected from the group consisting of human feces and subsurface carbon-bearing polar ices.

In any of the embodiments disclosed herein, the carbon-containing reducing agent can be selected from the group consisting of activated carbon, CH₄, CO, C₂H₄, CO₂, and CH₃OH.

In any of the embodiments disclosed herein, the concentrated solar irradiation can be provided by a solar concentrating thermal technology.

In any of the embodiments disclosed herein, the solar concentrating thermal technology can provide concentrated sunlight at solar concentration ratios of up to 5000 suns.

In any of the embodiments disclosed herein, the solar concentrating thermal technology can provide a temperature between 1000° C. and 2000° C.

In any of the embodiments disclosed herein, the metals and/or metalloids can be selected from the group consisting of aluminum, silicon, iron, calcium, magnesium, sodium, and titanium.

BRIEF DESCRIPTION OF THE DRAWINGS

The following detailed description of specific embodiments of the disclosure will be better understood when read in conjunction with the appended drawings. For the purpose of illustrating the disclosure, specific embodiments are shown in the drawings. It should be understood, however, that the disclosure is not limited to the precise arrangements and instrumentalities of the embodiments shown in the drawings.

FIG. 1 provides schematics outlining the two different pathways for producing metals and metalloids from lunar regolith using concentrated solar irradiation for direct thermal reduction and carbothermal reduction, in accordance with some embodiments of the present disclosure.

FIG. 2 provides a schematic depiction of a high flux solar simulator.

FIG. 3 provides schematics outlining the setup of thermogravimetry experiments to carbothermally reduce JSC-1A+C in 100% Ar environment, in accordance with some embodiments of the present disclosure. The Ar flow is controlled by a flow controller (FC) and passes through the O₂ trap before entering the thermogravimetric analyzer (TGA). The gas chromatography (GC) and mass spectrometry (MS) are connected to the gas exhaust line before the fume hood.

FIG. 4 provides schematics outlining the setup of ultra-high vacuum experiments to carbothermally reduce JSC-1A+C in ~10⁻¹² bar environment, in accordance with some embodiments of the present disclosure. The ultra-high vacuum (UHV) chamber is equipped with a turbomolecular pump (TP) and a mechanical vacuum pump (MVP) to maintain vacuum, and a quadrupole mass spectrometry (QMS) to measure gas evolution. The Cu foil is mounted above the sample to collect volatiles leaving the sample.

FIGS. 5A-5B provide chemical equilibrium compositions in terms of normalized moles of Si-species (FIG. 5A) and Fe-species (FIG. 5B) as a function of temperature for isobaric processes at 10⁻⁸ bar at C to JSC-1A mass ratio of 0.331, in accordance with some embodiments of the present disclosure.

FIGS. 6A-6B provide chemical equilibrium compositions in terms of normalized moles of SiO₂ (solid line) and Si(g) (dashed line) (FIG. 6A), and SiO(g) (solid line) and SiC (dashed line) (FIG. 6B) as a function of temperature and pressure at C to JSC-1A mass ratio of 0.331, in accordance with some embodiments of the present disclosure.

FIG. 7 provides chemical equilibrium compositions in terms of normalized moles of Fe (solid line) and Fe(g) (dashed line) as a function of temperature and pressure at C to JSC-1A mass ratio of 0.331, in accordance with some embodiments of the present disclosure.

FIG. 8 provides normalized mass change (dotted line), molar concentration of CO (solid line), and temperature (dashed line) versus time for thermogravimetry performed to examine the carbothermal reduction of JSC-1A in 100 mL_N/min of Ar, in accordance with some embodiments of the present disclosure.

FIG. 9 provides cumulated volatile mass loss (shaded light gray region) and CO evolution from sample (shaded dark gray region) versus experimental time for thermogravimetry performed to examine the carbothermal reduction of JSC-1A in 100 mL_N/min of Ar, in accordance with some embodiments of the present disclosure.

FIGS. 10A-10B provide temperature (dashed line) and MS signal (amu 28, solid line) versus time of ultra-high vacuum experiments at highest isotherms of 1000° C. (Experiment 1) (FIG. 10A) and 1250° C. (Experiment 2) (FIG. 10B) to examine the carbothermal reduction of JSC-1A in ultra-high vacuum environment, in accordance with some embodiments of the present disclosure.

FIGS. 11A-11C provide temperature (dashed line) and MS signal (amu 28, solid line) versus time of ultra-high vacuum experiments at highest isotherms of 1500° C. (Experiment 3) (FIG. 11A), 1750° C. (Experiment 4) (FIG. 11B), and (c) 2000° C. (Experiment 5) (FIG. 11C) to examine the carbothermal reduction of JSC-1A in ultra-high vacuum environment, in accordance with some embodiments of the present disclosure.

FIGS. 12A-12E provide SEM-captured sample image (FIG. 12A) and elemental scan by EDS/X of O (FIG. 12B), Al (FIG. 12C), Si (FIG. 12D), and Mg (FIG. 12E) of JSC-1A, in accordance with some embodiments of the present disclosure.

FIGS. 13A-13G provide SEM-captured sample image (FIG. 13A) and elemental scan by EDS/X of O (FIG. 13B), Al (FIG. 13C), Mg (FIG. 13D), Si (FIG. 13E), Fe (FIG. 13F), and Ti (FIG. 13G) of sample after TGA, in accordance with some embodiments of the present disclosure.

FIGS. 14A-14B provide X-ray diffraction normalized intensity patterns versus 2 h of loose sample plotted against reference SiC normalized intensity patterns (diamond markers) (FIG. 14A) and melted sample plotted against reference Mg₂SiO₄ normalized intensity patterns (square markers) (FIG. 14B), in accordance with some embodiments of the present disclosure.

FIGS. 15A-15C provide TEM images of a small fragment of the loose sample after TGA at 2 μm (FIG. 15A), 50 nm (FIG. 15B), and 10 nm (FIG. 15C) magnification, in accordance with some embodiments of the present disclosure.

FIGS. 16A-16E provide EDS/X-captured sample image (FIG. 16A) and elemental scan by EDS/X of Si (FIG. 16B), C (FIG. 16C), Fe (FIG. 16D), and O (FIG. 16E) of the same small fragment of loose sample after TGA as studied by the TEM, in accordance with some embodiments of the present disclosure.

FIG. 17 provides a schematic depicting the tube furnace experimental setup to reduce JSC-1A+C, LMS-1+C, and LHS-1+C in 100% Ar. The Ar flow is controlled by a flow controller (FC) and passes through an O₂ trap before entering the furnace. A thin circular Mo foil with ≤3 mm holes is mounted at the exhaust end of the Al₂O₃ tube to collect vapor deposits evolving from the sample during experiments. The gas chromatography (GC) and mass spectrometry (MS) are connected to the gas exhaust line before the fume hood to measure temporal gas evolution.

FIGS. 18A-18B provide chemical equilibrium compositions in terms of normalized moles of SiO(g) [solid lines], SiC (dashed lines), and Si(g) [dotted lines] for LMS-1+C (FIG. 18A) and LHS-1+C (FIG. 18B) systems as a function of temperature and pressure.

FIGS. 19A-19B provide chemical equilibrium compositions in terms of normalized moles of Al₂O(g) [solid lines] and Al(g) [dotted lines] for LMS-1+C (FIG. 19A) and LHS-1+C (FIG. 19B) systems as a function of temperature and pressure.

FIGS. 20A-20B provide chemical equilibrium compositions in terms of normalized moles of Mg(g) for LMS-1+C (FIG. 20A) and Ca(g) for LHS-1+C (FIG. 20B) systems as a function of temperature and pressure.

FIGS. 21A-21C provide molar concentration of CO for JSC-1A+C (solid lines) (FIG. 21A), LMS-1+C (solid lines) (FIG. 21B), and LHS-1+C (solid lines) (FIG. 21C), and temperature (dotted line) versus time for tube furnace experiments to examine the carbothermal reduction of JSC-1A, LMS-1, and LHS-1, respectively, in 100 mL_N/min of Ar.

FIG. 22 provides normalized mass of post-experimental sample, CO evolution, and all other vapor evolution from sample during tube furnace experiments for JSC-1A+C (solid line), LMS-1+C (dashed line), and LHS-1+C (dotted line).

FIGS. 23A-23C provide images of post-experimental samples melted onto the Mo foils for JSC-1A+C (FIG. 23A), LMS-1+C (FIG. 23B), and LHS-1+C (FIG. 23C) tube furnace experiments annotated with black squares of varying line styles marking regions used for material characterization work.

FIG. 24 provides x-ray diffraction normalized intensity patterns versus 2θ for dashed square region of post JSC-1A+C tube furnace experiment sample plotted against reference Fe_0.4Al_0.6 (diamond markers), Al_0.5Si_0.75O_2.25 (triangle markers), and Mo (circle markers) normalized intensity patterns.

FIGS. 25A-25D provide x-ray diffraction normalized intensity patterns versus 2θ for dashed and dotted-dashed square regions of post LMS-1+C tube furnace experiment sample plotted against reference Mo (diamond markers), Si (circle markers) (FIG. 25A); Ca₃Al₂P₂Si₂O₁₅ (triangle markers) (FIG. 25B), Ca₂Fe₉O₁₃ (square markers) (FIG. 25C), and Mo₃Si (hexagon markers) (FIG. 25D) normalized intensity patterns.

FIG. 26 provides x-ray diffraction normalized intensity patterns versus 2θ for dotted square region of post LMS-1+C tube furnace experiment sample plotted against reference Fe_0.5Al_0.5 (diamond markers) normalized intensity patterns.

FIGS. 27A-27D provide x-ray diffraction normalized intensity patterns versus 2θ for dashed square region of post LHS-1+C tube furnace experiment sample plotted against reference MoSi₂ (diamond markers) (FIG. 27A), CaAl₂Si₂O₈ (circle markers) (FIG. 27B), SiC (triangle markers) (FIG. 27C), and AlMoTi₂ (square markers) (FIG. 27D) normalized intensity patterns.

FIGS. 28A-28B provide x-ray diffraction normalized intensity patterns versus 2θ for solid square region of post LHS-1+C tube furnace experiment sample plotted against reference Fe (diamond markers) (FIG. 28A), Mo (circle markers) (FIG. 28B), and SiO₂ (triangle markers) (FIG. 28B) normalized intensity patterns.

DETAILED DESCRIPTION OF THE INVENTION

To facilitate an understanding of the principles and features of the present disclosure, various illustrative embodiments are explained below. The components, steps, and materials described hereinafter as making up various elements of the embodiments disclosed herein are intended to be illustrative and not restrictive. Many suitable components, steps, and materials that would perform the same or similar functions as the components, steps, and materials described herein are intended to be embraced within the scope of the disclosure. Such other components, steps, and materials not described herein can include, but are not limited to, similar components or steps that are developed after development of the embodiments disclosed herein. As used in the specification and the appended claims, the singular forms “a,” “an” and “the” include plural referents unless the context clearly dictates otherwise.

Also, in describing the preferred exemplary embodiments, terminology will be resorted to for the sake of clarity. It is intended that each term contemplates its broadest meaning as understood by those skilled in the art and includes all technical equivalents which operate in a similar manner to accomplish a similar purpose.

Ranges can be expressed herein as from “about” or “approximately” one particular value and/or to “about” or “approximately” another particular value. When such a range is expressed, another exemplary embodiment includes from the one particular value and/or to the other particular value.

By “comprising” or “containing” or “including” is meant that at least the named compound, member, particle, or method step is present in the composition or article or method, but does not exclude the presence of other compounds, materials, particles, method steps, even if the other such compounds, material, particles, method steps have the same function as what is named.

Mention of one or more method steps does not preclude the presence of additional method steps or intervening method steps between those steps expressly identified. Similarly, it is also to be understood that the mention of one or more components in a device or system does not preclude the presence of additional components or intervening components between those components expressly identified.

The materials described as making up the various members of the invention are intended to be illustrative and not restrictive. Many suitable materials that would perform the same or a similar function as the materials described herein are intended to be embraced within the scope of the invention. Such other materials not described herein can include, but are not limited to, for example, materials that are developed after the time of the development of the invention.

As discussed above, lunar regolith contains many valuable metals, including Si, Fe, Ti, K, and Ca (Crawford, I. A., “Lunar Resources: A Review,” Progress in Physical Geography: Earth and Environment, 39:137-167 (2015); Cannon, “Planetary simulant database,” Free resource for regolith simulant information [Online] Available: https://simulantdb. com/simulants/jsc1.php [Accessed 14 Jun. 2022]; Papike et al., “The lunar regolith: Chemistry, mineralogy, and petrology,” Rev. Geophys., 20:761-826 (1982)). These metals and metalloids are bound with oxygen as oxides, which are contained in Plagioclase, Olivine, and other minerals (Papike et al., “The lunar regolith: Chemistry, mineralogy, and petrology,” Rev. Geophys., 20:761-826 (1982); Cannon, “Planetary simulant database,” Free resource for regolith simulant information [Online] Available: https://simulantdb.com/simulants/jsc1.php [Accessed 14 Jun. 2022]; Schreiner et al., “Thermophysical property models for lunar regolith,” Adv. Space Res., 57:1209-1222 (2016)), necessitating chemical reductions that require energy and, potentially, reducing agents. Sunlight is a viable heat source that is capable of driving such thermochemical processes, as the Moon is ideally situated for harnessing sunlight due to little atmospheric attenuation and long lunar day (Palos et al., “Lunar ISRU energy storage and electricity generation,” Acta Astronaut, 170:412-420 (2020)). The lunar declination angle of 1.5° compared to Earth's 23° (Burke, J. D., “Perpetual Sunshine, Moderate Temperatures and Perpetual Cold as Lunar Polar Resources,” In: Badescu, V. (Ed.), Moon: Prospective Energy and Material Resources, Heidelberg, Germany: Springer Berlin (2012)), and a lunar day that is equivalent to ~ 14 Earth days (Palos et al., “Lunar ISRU energy storage and electricity generation,” Acta Astronaut, 170:412-420 (2020)) result in an average direct-normal solar irradiation on the Moon of ~1361 W/m² (Williams, 2021, Moon Fact Sheet [Online]. NASA. Available: https://nssdc.gsfc.nasa.gov/planetary/factsheet/moonfact.html [Accessed 14 Jun. 2022]), significantly higher than on Earth due to no atmospheric attenuation. Pairing the process to solar concentrating thermal technologies, which further concentrate sunlight (e.g., heliostats fields) to solar concentration ratios of up to 5000 suns (Gill et al., Characterization of a 6 kW high-flux solar simulator with an array of xenon arc lamps capable of concentrations of nearly 5000 suns,” Rev. Sci. Instrum., 86: 125107 (2015)), facilitates operating temperatures between 100° and 2000° C., where the reduced gravity on the Moon compared to Earth allows for lighter infrastructures (Palos et al., “Lunar ISRU energy storage and electricity generation,” Acta Astronaut, 170:412-420 (2020)). This enables solar thermochemical processes specifically designed to extract the metalloids and metals contained in lunar regolith which further benefit from very low lunar surface pressures (fugacity) of 3×10⁻¹⁵ bar (Williams, 2021, Moon Fact Sheet [Online]. NASA. Available: https://nssdc.gsfc.nasa.gov/planetary/factsheet/moonfact.html [Accessed 14 Jun. 2022]), according to Le Chatelier's principle. The coupling of concentrating solar irradiation to ultra-low pressures to drive thermochemical processes enables elevated temperature processes under thermodynamically favorable conditions for producing pure metals and metalloids.

Concentrated solar irradiation is a novel lunar regolith processing technique that can provide the necessary radiative heat fluxes to drive the volatilization and thermochemical processes necessary to extract materials from the lunar regolith. There are numerous advantages to using solar irradiation to drive ISRU processes on the moon compared to Earth: The solar irradiation is not attenuated by an atmosphere; there is no wind and ⅙th gravity loads allow for lighter support infrastructure; the lunar tilt from the celestial equator (lunar declination angle) is 1.5°, compared to the Earth's 23°, resulting in little seasonal variation in the solar irradiation (Burke, “Perpetual Sunshine, Moderate Temperatures and Perpetual Cold as Lunar Polar Resources,” in Badescu, ed., Moon: Prospective Energy and Material Resources, Heidelberg, Germany: Springer, pp. 335-345 (2012)).

The direct thermal reduction of SiO₂ to Si and Fe₂O₃ or FeO to Fe are favorable at elevated temperatures with favorable co-production of Fe(g) and Si(g) predicted.

Recombination of the gaseous products with O₂ is likely upon cooling (Schlüter and Cowley, “Review of techniques for in-situ oxygen extraction on the moon,” Planetary Space Sci., 181:104753 (2020)), requiring rapid quenching. A proposed system of rapid quenching involves cooled metal plates to condense metals before recombination, and O₂ pumped into a compressed storage (Schlüter and Cowley, “Review of techniques for in-situ oxygen extraction on the moon,” Planetary Space Sci., 181:104753 (2020)). Circumventing recombination is also possible by adding reducing agents (e.g., carbon) available on the Moon with the added advantage of decreasing the reaction onset temperatures. Two different reducing pathways using concentrated solar thermal (CST) technologies are outlined in FIG. 1: (1) direct thermal reduction and (2) carbothermal reduction. The direct thermal and carbothermal reduction of lunar simulants for O₂ production has been extensively studied (Balasubramaniam et al., “The Reduction of Lunar Regolith by Carbothermal Processing Using Methane,” Int. J. Miner. Process., 96:54-61 (2010); Gustafson et al., “Carbon reduction of lunar regolith for oxygen production,” AIP Conf. Proc., 746:1224-1228 (2005); Troisi et al., “Oxygen extraction from lunar dry regolith: Thermodynamic numerical characterization of the carbothermal reduction,” Acta Astronaut, 199:113-124 (2022)) whereas interest in metals and metalloids production is increasing (Samouhos et al., “In-situ resource utilization: ferrosilicon and SiC production from BP-1 lunar regolith simulant via carbothermal reduction,” Planet. Space Sci., 212:105414 (2022); Shaw et al., “Thermodynamic modelling of ultra-high vacuum thermal decomposition for lunar resource processing,” Planet. Space Sci., 204:105272 (2021); Shaw et al., “High Vacuum Solar Thermal Dissociation for Metal and Oxide Extraction,” In: Reddy, eds., New Directions in Mineral Processing, Extractive Metallurgy, Recycling and Waste Minimization, Cham, Switzerland: Springer Nature Switzerland, pp. 77-86 (2023)). The carbothermal reduction pathway is often explored for extraterrestrial [e.g., Mars (Nababan et al., “Metals extraction on Mars through carbothermic reduction,” Acta Astronaut., 198:564-576 (2022))] oxygen and metal productions. The direct thermal reduction pathway for producing metals and metalloids from lunar regolith or simulants has been explored at ultra-high vacuum conditions to mimic lunar surface conditions (Shaw et al., “Thermodynamic modelling of ultra-high vacuum thermal decomposition for lunar resource processing,” Planet. Space Sci., 204:105272 (2021); Shaw et al., “High Vacuum Solar Thermal Dissociation for Metal and Oxide Extraction,” In: Reddy, eds., New Directions in Mineral Processing, Extractive Metallurgy, Recycling and Waste Minimization, Cham, Switzerland: Springer Nature Switzerland, pp. 77-86 (2023)). However, the carbothermal reduction pathway with the focus of metals and metalloids production at inert and ultra-high vacuum conditions is underexplored.

As described herein, the carbothermal reduction pathway for producing metals and metalloids from JSC-1A lunar regolith simulant was explored. CH₄, readily found on the Moon (Cadogan et al., “Carbon Chemistry of the Lunar Surface,” Nature, 231:29-31 (1971); Hodges et al., “Methane in the lunar exosphere: Implications for solar wind carbon escape,” Geophys. Res. Lett., 43:6742-6748 (2016)), is popularly studied as the reducing agent in the carbothermal reduction pathway of lunar simulants (Prinetto et al., “Terrestrial demonstrator for a low-temperature carbothermal reduction process on lunar regolith simulant: Design and AIV activities,” Planet. Space Sci., 225:105618 (2023); Troisi et al., “Oxygen extraction from lunar dry regolith: Thermodynamic numerical characterization of the carbothermal reduction,” Acta Astronaut, 199:113-124 (2022)) where the initial process often involves cracking CH₄ into C and H₂, where C is used as a reducing agent for regolith analogues and H₂ is repurposed or recycled into CH₄ (Balasubramaniam et al., “The Reduction of Lunar Regolith by Carbothermal Processing Using Methane,” Int. J. Miner. Process., 96:54-61 (2010); Lee et al., “A simulation study on the direct carbothermal reduction of SiO₂ for Si metal,” Curr. Appl Phys., 10: S218-S221 (2010); Prinetto et al., “Terrestrial demonstrator for a low-temperature carbothermal reduction process on lunar regolith simulant: Design and AIV activities,” Planet. Space Sci., 225:105618 (2023)). Due to the low surface pressure on the Moon, CH₄ cracking occurs at temperatures lower than the reduction temperatures of oxides/minerals present in lunar regolith, making CH₄ a viable source of C for carbothermal reduction of lunar regolith [CH₄→C+2H₂, T^eq<25° C. (Roine, HSC Chemistry 7.11.7.11 ed.: Metso Outotec (2009))]. Activated carbon was used in this work as the reducing agent to eliminate the secondary process of H₂ and extend the analysis validity to different sources of carbon to prepare for other fixed carbonaceous sources [e.g., human feces (Bittencourt et al., “Thermodynamic Assessment of Human Feces Gasification: An Experimental-based Approach,” SN Applied Sci., 1:1077 (2019))] becoming available on Moon in the future as a result of a human presence.

Thermodynamic modeling of carbothermal reduction of lunar regolith with CH₄ and C has been conducted with the focus on potential oxygen yield (Troisi et al., “Oxygen extraction from lunar dry regolith: Thermodynamic numerical characterization of the carbothermal reduction,” Acta Astronaut, 199:113-124 (2022)). However, the production of different species and intermediaries, such as metals, metalloids, reduced metal oxides, and reduced minerals of lunar simulants, has not been investigated. Herein, thermodynamic analysis was conducted to predict equilibrium chemical compositions for the carbothermal reduction of JSC-1A lunar regolith simulant. Thermogravimetry coupled to gas analytics was then used to assess reaction extents, rates, and products with the aim to successfully demonstrate the carbothermal reduction of JSC-1A with activated carbon and quantify volatiles production. Solid-state surface material characterization was performed to examine particle morphologies and elemental distributions of samples before and after thermogravimetry, along with chemical compositions of sample after thermogravimetry to assess the transpired reduction reactions. Ultra-high vacuum experiments were conducted in combination with solid-state surface material characterization to mimic the low lunar surface pressure and corroborate the thermogravimetry as well as identify volatiles leaving the sample at varying temperatures. The thermodynamic and experimental analyses serve to deepen the understanding of carbothermal reduction of JSC-1A lunar simulants for metal or metalloid production in inert and at high vacuum environment, and they provide guidance for operating conditions and subsequent solar-driven reactor designs for carbothermal reduction processes of lunar regolith on Moon.

A generic process to produce these valuable products includes prospecting, excavating, beneficiating and processing lunar regolith before storage and distribution. The various elemental compositions (e.g., Fe, Al, Ca) across lunar surface are extensively identified and measured via remote sensing (Pieters et al., “The Moon Mineralogy Mapper (M3) on Chandrayaan-1,” Current Science, 96 (4): 500-505 (2009); Prettyman, T. H., Chapter 54—Remote Sensing of Chemical Elements Using Nuclear Spectroscopy, in Encyclopedia of the Solar System (Third Edition), T. Spohn, D. Breuer, and T. V. Johnson, Eds, 2014, Elsevier: Boston. p. 1161-1183; Ouyang et al., “Primary scientific results of Chang'E-1 lunar mission,” Science China Earth Sciences, 53 (11): 1565-1581 (2010)) as part of the prospecting and surface exploration efforts. The resource mappings (Prettyman, T. H., Chapter 54—Remote Sensing of Chemical Elements Using Nuclear Spectroscopy, in Encyclopedia of the Solar System (Third Edition), T. Spohn, D. Breuer, and T. V. Johnson, Eds, 2014, Elsevier: Boston. p. 1161-1183; Dhingra, D., “The New Moon: Major Advances in Lunar Science Enabled by Compositional Remote Sensing from Recent Missions,” Geosciences, 8 (12): 498 (2018); Andrea Marinoni, P. G., MINERAL MAPPING FROM CHANGE 1 DATA. 10th EARSeL SIG Imaging Spectroscopy Workshop (2017)) are useful for informing decisions on landing sites and potential habitation sites. Site preparation is also needed before excavation and collection efforts are conducted. Dust is an important factor to consider as the lunar regolith contains large amounts of fine dust that adhere to any and every surface (Cannon et al., “Working with lunar surface materials: Review and analysis of dust mitigation and regolith conveyance technologies,” Acta Astronautica, 196:259-274 (2022)). Clearing the excavation sites from fine top layers of regolith and leveling potential inclined surfaces help with dust control and site preparation for excavation and lunar regolith collection. Various regolith moving systems to excavate and move regolith have been investigated and proposed in the field, which typically involves transportation of regolith from excavation site to ISRU sites, storage of lunar regolith before processing, and initial beneficiation work (e.g., size separation/sieving, magnetic separation) of lunar regolith (Cannon et al., “Working with lunar surface materials: Review and analysis of dust mitigation and regolith conveyance technologies,” Acta Astronautica, 196:259-274 (2022); Mueller et al., Evolution of Regilith Feed Systems for Lunar ISRU O2 Production Plants, in 48th AIAA Aerospace Sciences Meeting Including the New Horizons Forum and Aerospace Exposition (2010) American Institute of Aeronautics and Astronautics; Regolith Extraction, Storage, and Transfer under Micro-Gravity, in Earth and Space 2016. p. 584-597; Guerrero-Gonzalez and Zabel, “System analysis of an ISRU production plant: Extraction of metals and oxygen from lunar regolith,” Acta Astronautica, 203:187-201 (2023)). Upon transportation to ISRU sites and initial beneficiation process, regolith is reacted or processed to yield useful products. Lunar regolith processing pathways that are being investigated in the field include direct thermal heating (Shaw et al., “Thermodynamic modelling of ultra-high vacuum thermal decomposition for lunar resource processing,” Planetary and Space Science, 204:105272 (2021); Šeško et al., “Oxygen production by solar vapor-phase pyrolysis of lunar regolith simulant,” Acta Astronautica, 224:215-225 (2024)), H₂ reduction (Reiss et al., “Thermogravimetric analysis of chemical reduction processes to produce oxygen from lunar regolith,” Planetary and Space Science, 181:104795 (2020); Sargeant et al., “Hydrogen reduction of ilmenite: Towards an in situ resource utilization demonstration on the surface of the Moon,” Planetary and Space Science, 180:104751 (2020)), carbothermal reduction (Sargeant et al., “Hydrogen reduction of ilmenite: Towards an in situ resource utilization demonstration on the surface of the Moon,” Planetary and Space Science, 180:104751 (2020); Peng et al., “Influencing factors for the preparation of FeO in lunar soil simulant using high-temperature carbothermic reduction,” Advances in Space Research, 70 (10): 3220-3230 (2022); Samouhos et al., “In-situ resource utilization: ferrosilicon and SiC production from BP-1 lunar regolith simulant via carbothermal reduction,” Planetary and Space Science, 212:105414 (2022); Kaur et al., “Thermodynamic and experimental study on carbothermal reduction of JSC-1A lunar regolith simulant for metal and metalloid production,” Advances in Space Research, 73 (8): 4024-4039 (2024)), and molten regolith electrolysis (Burke et al., “Modeling electrolysis in reduced gravity: producing oxygen from in-situ resources at the moon and beyond,” Frontiers in Space Technologies, 5 (2024); Schreiner et al., Development of a Molten Regolith Electrolysis Reactor Model for Lunar In-Situ Resource Utilization?, in 8th Symposium on Space Resource Utilization (2015), American Institute of Aeronautics and Astronautics). The products of these reactions, dependent on reaction mechanisms, are ideally collected and stored separately. The metals or metalloids (e.g., Fe, Si, Al, SiC) and gases or vapors (e.g., O₂, H₂O, CO) are then distributed to potential construction, habitation, and research sites, or spacecraft fuel stations for utilization. Slugs formed during primary processing may possibly be stored and used for secondary processing or other beneficiation processes.

Processing lunar regolith as an O₂ source for life support, propellant for space systems, and human habitation has been extensively studied (Šeško et al., “Oxygen production by solar vapor-phase pyrolysis of lunar regolith simulant,” Acta Astronautica, 224:215-225 (2024); Reiss et al., “Thermogravimetric analysis of chemical reduction processes to produce oxygen from lunar regolith,” Planetary and Space Science, 181:104795 (2020); Lu and Reddy, “Extraction of Metals and Oxygen from Lunar Soil,” High Temperature Materials and Processes, 27 (4): 223-234 (2008); Schlüter and Cowley, “Review of techniques for In-Situ oxygen extraction on the moon,” Planetary and Space Science, 181:104753 (2020)) and continues to be a priority with conceptualization of ISRU facilities (Mueller et al., Evolution of Regilith Feed Systems for Lunar ISRU O₂ Production Plants, in 48th AIAA Aerospace Sciences Meeting Including the New Horizons Forum and Aerospace Exposition (2010) American Institute of Aeronautics and Astronautics; Dottori et al., “Lunar pilot plant payload design toward in situ demonstration of oxygen extraction by carbothermal reduction,” Aerospace Europe Conference 2023, 2023; Kosuke et al., “Hybrid lunar ISRU plant: a comparative analysis with carbothermal reduction and water extraction,” Acta Astronautica, 2024)). Producing and using metals/metalloids for construction materials (Toklu and Akpinar, “Lunar soils, simulants and lunar construction materials: An overview,” Advances in Space Research, 70 (3): 762-779 (2022); Farries et al., “Sintered or melted regolith for lunar construction: state-of-the-art review and future research directions,” Construction and Building Materials, 296:123627 (2021)) to fabricate scientific, industrial, and habitation infrastructures on the Moon or to commercially supply resources back to Earth has been underexplored with increasing research interest in recent years (Lu and Reddy, “Extraction of Metals and Oxygen from Lunar Soil,” High Temperature Materials and Processes, 27 (4): 223-234 (2008); Peng et al., “Influencing factors for the preparation of FeO in lunar soil simulant using high-temperature carbothermic reduction,” Advances in Space Research, 70 (10): 3220-3230 (2022); Samouhos et al., “In-situ resource utilization: ferrosilicon and SiC production from BP-1 lunar regolith simulant via carbothermal reduction,” Planetary and Space Science, 212:105414 (2022); Kaur et al., “Thermodynamic and experimental study on carbothermal reduction of JSC-1A lunar regolith simulant for metal and metalloid production,” Advances in Space Research, 73 (8): 4024-4039 (2024)). While various pathways have been postulated, significant knowledge gaps are still present, especially for carbothermal reduction aimed at metal and metalloid production. Thermodynamic and experimental investigations on operating conditions to target and produce different compounds are needed to realize ISRU goals. Hence, carbothermal reduction of lunar regolith simulants is explored in this work thermodynamically and experimentally. C is used as the reducing agent in carbothermal reductions (Balasubramaniam et al., “The reduction of lunar regolith by carbothermal processing using methane,” International Journal of Mineral Processing, 96 (1): 54-61 (2010) to reduce the reaction onset temperatures compared to direct thermal heating and avoid problematic recombination upon cooling (Schlüter and Cowley, “Review of techniques for In-Situ oxygen extraction on the moon,” Planetary and Space Science, 181:104753 (2020)). C is potentially available on Moon as CH₄ in the exosphere (Hodges, “Methane in the lunar exosphere: Implications for solar wind carbon escape,” Geophysical Research Letters, 43 (13): 6742-6748 (2016)), subsurface carbon-bearing polar ices (Cannon, K. M., “Accessible Carbon on the Moon,” Earth and Planetary Astrophysics (2021)), or from human waste upon presence in the future (Bittencourt et al., “Thermodynamic assessment of human feces gasification: an experimental-based approach,” SN Applied Sciences, 1 (9): 1077 (2019); Duan et al., “Human waste anaerobic digestion as a promising low-carbon strategy: Operating performance, microbial dynamics and environmental footprint,” Journal of Cleaner Production, 256:120414 (2020)). Concentrated solar thermal (CST) technologies provide an ideal process heat source due to the lack of atmosphere on Moon (Palos et al., “Lunar ISRU energy storage and electricity generation,” Acta Astronautica, 170:412-420 (2020)) to attenuate solar irradiation and long lunar day (~14 Earth days (Palos et al., “Lunar ISRU energy storage and electricity generation,” Acta Astronautica, 170:412-420 (2020))), enabling processes at elevated temperatures required for carbothermal reduction. The low surface pressure of 3×10⁻¹⁵ bar on the Moon (Williams, D. R. Moon Fact Sheet. 2021 [cited 2022, 14 Jun. 2022]; Available from: https://nssdc.gsfc.nasa.gov/planetary/factsheet/moonfact.html) is ideal for driving reactions which become spontaneous at lower temperatures according to Le Chatelier's principle.

Thermodynamic analyses with the focus on O₂ or metals and metalloids yield have been conducted (Shaw et al., “Thermodynamic modelling of ultra-high vacuum thermal decomposition for lunar resource processing,” Planetary and Space Science, 204:105272 (2021); Šeško et al., “Oxygen production by solar vapor-phase pyrolysis of lunar regolith simulant,” Acta Astronautica, 224:215-225 (2024); Lu and Reddy, “Extraction of Metals and Oxygen from Lunar Soil,” High Temperature Materials and Processes, 27 (4): 223-234 (2008); Kaur et al., “Thermodynamic and experimental study on carbothermal reduction of JSC-1A lunar regolith simulant for metal and metalloid production,” Advances in Space Research, 73 (8): 4024-4039 (2024); Schlüter and Cowley, “Review of techniques for In-Situ oxygen extraction on the moon,” Planetary and Space Science, 181:104753 (2020); Troisi et al., “Oxygen extraction from lunar dry regolith: Thermodynamic numerical characterization of the carbothermal reduction,” Acta Astronautica, 199:113-124 (2022)). However, they are deficient for generic highland and mare simulants, which possibly expands the scope of relevancy for future ISRU work in terms of regolith compositions. Thermodynamic analyses were conducted in this work to forecast equilibrium compositions for the carbothermal reduction of (1) lunar highland simulant (LHS-1) and (2) lunar mare simulant (LMS-1) as functions of equilibrium total pressures and temperatures. The results were used to inform and guide further experimental investigations in a tube furnace to study reaction mechanisms of carbothermal reduction of different simulant types. Solid-state material characterization was also conducted to characterize the solid and gaseous products. Results provide insights into the feasibility of metal or metalloid production in an inert environment, laying the foundation for further analysis on processing lunar regolith carbothermally. While ground truth is essential to realize lunar ISRU, utilizing lunar regolith simulants in this work provides valuable insights with comparable reactions with actual lunar regolith, as these simulants are carefully curated to chemically behave analogously to lunar regolith (Toklu and Akpinar, “Lunar soils, simulants and lunar construction materials: An overview,” Advances in Space Research, 70 (3): 762-77 (2022); Clendenen et al., “Temperature programmed desorption comparison of lunar regolith to lunar regolith simulants LMS-1 and LHS-1,” Earth and Planetary Science Letters, 592:117632 (2022)).

An exemplary embodiment of the present disclosure provides a process for extracting metals and/or metalloids from extraterrestrial regolith. The process comprises providing a sample of extraterrestrial regolith, wherein the sample comprises an oxide of one or more metals and/or metalloids, and chemically reacting the oxide of the one or more metals and/or metalloids with carbon via carbothermal reduction to produce one or more metals and/or metalloids.

In any of the embodiments disclosed herein, chemically reacting the oxide can comprise mixing the sample with a carbon-containing reducing agent to form a mixture, and directing concentrated solar irradiation to the mixture as process heat.

Another exemplary embodiment of the present disclosure provides a process for extracting metals and/or metalloids from a material. The process comprises providing a sample of a material, wherein the sample contains an oxide of one or more metals and/or metalloids; and chemically reacting the oxide of the one or more metals and/or metalloids with carbon via carbothermal reduction to produce one or more metals and/or metalloids, wherein chemically reacting the oxide comprises contacting the sample with a carbon-containing reducing agent to form a mixture and directing concentrated solar irradiation to the mixture.

In any of the embodiments disclosed herein, chemically reacting the oxide can be carried out at a pressure between 3×10⁻¹⁵ bar and 1×10⁻¹² bar. For example, in some embodiments, chemically reacting the oxide can be carried out at a pressure between 3×10⁻¹⁵ bar and 1×10⁻¹⁴ bar, between 3×10⁻¹⁵ bar and 5×10⁻¹⁴ bar, between 3×10⁻¹⁵ bar and 1×10⁻¹³ bar, between 3×10⁻¹⁵ bar and 5×10⁻¹³ bar, between 3×10⁻¹⁵ bar and 1×10⁻¹² bar, between 1×10⁻¹⁴ bar and 5×10⁻¹⁴ bar, between 1×10⁻¹⁴ bar and 1×10⁻¹³ bar, between 1×10⁻¹⁴ bar and 5×10⁻¹³ bar, between 1×10⁻¹⁴ bar and 1×10⁻¹² bar, between 5×10⁻¹⁴ bar and 1×10⁻¹³ bar, between 5×10⁻¹⁴ bar and 5×10⁻¹³ bar, between 5×10⁻¹⁴ bar and 1×10⁻¹² bar, between 1×10⁻¹³ bar and 5×10⁻¹³ bar, between 1×10⁻¹³ bar and 1×10⁻¹² bar, or between 5×10⁻¹³ bar and 1×10⁻¹² bar.

In some embodiments, chemically reacting the oxide is carried out in an inert environment at a pressure of 1 bar to simulate the lunar surface's low O₂ environment.

In any of the embodiments disclosed herein, chemically reacting the oxide can be carried out at a temperature between 750° C. and 2000° C. For example, in some embodiments, chemically reacting the oxide can be carried out at a temperature between 750° C. and 1000° C., between 750° C. and 1250° C., between 750° C. and 1500° C., between 750° C. and 1750° C., between 1000° C. and 1250° C., between 1000° C. and 1500° C., between 1000° C. and 1750° C., between 1000° C. and 2000° C., between 1250° C. and 1500° C., between 1250° C. and 1750° C., between 1250° C. and 2000° C., between 1500° C. and 1750° C., between 1500° C. and 2000° C., or between 1750° C. and 2000° C.

In any of the embodiments disclosed herein, the extraterrestrial regolith can be lunar regolith. Lunar regolith refers to the layer of loose, heterogeneous material covering solid rock on the surface of the Moon. This material can be composed of a mixture of fine dust, small rocks, and broken fragments of larger rocks. Lunar regolith can vary in depth, typically ranging from about 4 to 5 meters in the maria (the dark, basaltic plains) to 10 to 15 meters in the highlands. It plays a crucial role in lunar exploration and potential future lunar habitation, as it can be used for in-situ resource utilization (ISRU) to support human activities on the Moon. In some embodiments lunar regolith simulants can be used. Lunar regolith simulants can be synthesized on Earth to simulate the mineralogical, particle-size distribution, and chemical behavior of lunar regolith for research and technology development purposes due to current scarcity of lunar regolith.

In any of the embodiments disclosed herein, the carbon-containing reducing agent is derived from a fixed carbon source. For example, in some embodiments the carbon-containing reducing agent can be derived from a carbon source selected from the group consisting of human feces and subsurface carbon-bearing polar ices. In some embodiments, the carbon-containing reducing agent can be activated carbon, CH₄, CO, C₂H₄, CO₂, and CH₃OH.

In any of the embodiments disclosed herein, the concentrated solar irradiation is provided by a solar concentrating thermal technology. Solar concentrators can use optical media, such as lenses and mirrors, to focus incident (solar) light. Concentrated light can then be focused to produce thermal or electrical energy via photovoltaic (PV) cells. Solar concentrators have been applied in industry, consumer products, and numerous advanced technologies. This concentrated sunlight can be used to heat up materials, such as lunar regolith. Concentrator technologies have applications across the electromagnetic (EM) spectrum; paraboloid shaped antennae dishes used for satellite communications use similar geometries for radio waves that mirror-based concentrators use for visible light. Parabolic trough technologies are applied for cutting-edge terrestrial solar power plants and in space via a Stretched-Lens Array (SLA) Fresnel concentrator design. Suitable solar concentrating thermal technologies include without limitation Cassegrain reflectors, offset parabolic concentrators, compound parabolic concentrators, trough (line focus) concentrators, stretched lens array, and Fresnel reflectors. In some embodiments, secondary concentrators can be used to further concentrate focused light after being directed by an initial concentrator, or to move focused light to a desired location.

In some embodiments a high-flux solar simulator (FIG. 2) can be coupled to an upward flow reactor to mimic carbothermal reduction conditions on the lunar surface.

In any of the embodiments disclosed herein, the solar concentrating thermal technology can provide concentrated sunlight at solar concentration ratios of up to 5000 suns. For example, in some embodiments, the solar concentrating thermal technology can provide concentrated sunlight at solar concentration ratios of up to 500 suns, up to 1000 suns, up to 1500 suns, up to 2000 suns, up to 2500 suns, up to 3000 suns, up to 3500 suns, up to 4000 suns, or up to 4500 suns.

In any of the embodiments disclosed herein, the solar concentrating thermal technology can provide a temperature between 1000° C. and 2000° C. For example, in some embodiments, the solar concentrating thermal technology can provide a temperature between 1000° C. and 1200° C., between 1000° C. and 1400° C., between 1000° C. and 1600° C., between 1000° C. and 1800° C., between 1200° C. and 1400° C., between 1200° C. and 1600° C., between 1200° C. and 1800° C., between 1200° C. and 2000° C., between 1400° C. and 1600° C., between 1400° C. and 1800° C., between 1400° C. and 2000° C., between 1600° C. and 1800° C., between 1600° C. and 2000° C., or between 1800° C. and 2000° C.

In some embodiments, the different metal and/or metalloid oxides will be separated prior to carbothermal reduction to avoid problematic separation after reduction and potentially reduce product impurities. In some embodiments, different separation methodologies can be utilized for each of the different oxides by targeting temperatures just above the melting points for the different oxides to realize separation.

In any embodiment disclosed herein, the metals and/or metalloids can be aluminum, silicon, iron, calcium, magnesium, sodium, and titanium.

It is to be understood that the embodiments and claims disclosed herein are not limited in their application to the details of construction and arrangement of the components set forth in the description and illustrated in the drawings. Rather, the description and the drawings provide examples of the embodiments envisioned. The embodiments and claims disclosed herein are further capable of other embodiments and of being practiced and carried out in various ways. Also, it is to be understood that the phraseology and terminology employed herein are for the purposes of description and should not be regarded as limiting the claims.

Accordingly, those skilled in the art will appreciate that the conception upon which the application and claims are based may be readily utilized as a basis for the design of other structures, methods, and systems for carrying out the several purposes of the embodiments and claims presented in this application. It is important, therefore, that the claims be regarded as including such equivalent constructions.

Furthermore, the purpose of the foregoing Abstract is to enable the United States Patent and Trademark Office and the public generally, and especially including the practitioners in the art who are not familiar with patent and legal terms or phraseology, to determine quickly from a cursory inspection the nature and essence of the technical disclosure of the application. The Abstract is neither intended to define the claims of the application, nor is it intended to be limiting to the scope of the claims in any way.

EXAMPLES

The following Examples are presented to illustrate various aspects of the present application, but are not intended to limit the scope of the claimed application.

Example 1—Thermodynamic and Experimental Study on Carbothermal Reduction of JSC-1A Lunar Regolith Simulant for Metal and Metalloid Production

The carbothermal reduction of JSC-1A lunar simulant with activated carbon was investigated for producing metals and metalloids. Chemical equilibrium compositions were examined using Gibb's free energy minimization to identify favorable operating temperatures at 0.1, 10⁻⁸, and 3×10⁻¹⁵ bar for Fe and Si production. Complete conversion of Fe₂O₃ to Fe(g) was predicted at temperatures above 850° C. and about 90% conversion of SiO₂ to Si(g) was predicted at temperatures above 1000° C. for lunar surface pressure of 3×10⁻¹⁵ bar. Thermogravimetry was used to examine the reactions with mixtures of JSC-1A and stoichiometric activated carbon in 100% Ar up to 1500° C., and comparisons between temporal mass losses and CO evolution were used to estimate volatile production. The results were compared to ultra-high vacuum experiments at ~10-12 bar with temporal CO measured. Solid-state surface material characterization was performed using scanning electron microscopy, energy-dispersive X-ray spectroscopy, X-ray photoelectron spectroscopy, X-ray diffractometry, and transmission electron microscopy to identify elemental distributions and particle morphology of sample surface before and after experiment as well as chemical compositions present after experiment. Changes in elemental distributions and morphologies of Si—, Fe—, and Ti— in the sample before and after the experiment provided strong evidence of reactions of Si—, Fe—, and Ti— compounds in JSC-1A. Characterization of the loose sample after TGA confirmed the presence of crystalline SiC, Si, and Fe. Deposits collected on Cu foils mounted above the sample in the ultra-high vacuum experiments were found to contain Fe—, Al—, Mg—, and Si— volatiles.

Materials

Chemical and Mineral Compositions of JSC-1A

JSC-1A is a low-Ti mare simulant with a chemical composition simulating Apollo sample 14163 (Cannon, “Planetary simulant database,” Free resource for regolith simulant information [Online]. Available: https://simulantdb.com/simulants/jsc1.php [Accessed 14 Jun. 2022]). The mass fractions of different oxides in JSC-1A are given in Table 1 for baseline comparison with experimental results. Even though Fe is more commonly present as FeO in lunar regolith (Papike et al., “The lunar regolith: Chemistry, mineralogy, and petrology,” Rev. Geophys., 20:761-826 (1982)) due to the highly reduced state of lunar surface (Li et al., “Widespread hematite at high latitudes of the moon,” Sci. Adv., 6 (36): eaba1940 (2020)), Fe₂O₃ is more common in terrestrial minerals (Schreiner et al., “Thermophysical property models for lunar regolith,” Adv. Space Res., 57:1209-1222 (2016)) and is found in lunar regolith due to oxygen-containing Earth wind, presence of ice near the lunar poles, and heat from interplanetary dust impacts (Li et al., “Widespread hematite at high latitudes of the moon,” Sci. Adv., 6 (36): eaba1940 (2020)). Coupling with the higher Fe content as compared to other types of simulants (Crawford, I. A., “Lunar Resources: A Review,” Progress in Physical Geography: Earth and Environment, 39:137-167 (2015)), JSC-1A is an attractive lunar simulant to study since Fe is particularly useful for construction purposes on Moon (Crawford, I. A., “Lunar Resources: A Review,” Progress in Physical Geography: Earth and Environment, 39:137-167 (2015)). Substantial amounts of SiO₂ and Fe₂O₃ (47.40% and 11.40%, respectively) are present for ISRU production.

The most abundant minerals in JSC-1A are natural basaltic glass (49.3%), plagioclase (37.1%), and olivine (9.0%), with <1.5% abundances of other minerals (Cannon, “Planetary simulant database,” Free resource for regolith simulant information [Online]. Available: https://simulantdb.com/simulants/jsc1.php [Accessed 14 Jun. 2022]). The natural basaltic glass in JSC-1A is a mineral with 45-52% SiO₂ by mass and other oxides (e.g., Al₂O₃, Fe₂O₃, CaO) (Lima et al., “Basaltic glass-ceramic: A short review,” Boleti'n De La Sociedad Espan ola De Cera mica y Vidrio, 61:2-12 (2022)). The plagioclase contained in JSC-1A is represented with two end members: (1) NaAlSi₃O₈ (albite) and (2) CaAl₂Si₂O₈ (anorthite) (Mineralogy, H. I. O. 2023. Plagioclase [Online].mindat.org: Hudson Institute of Mineralogy. Available: https://www.mindat.org/min-9264.html [Accessed Jul. 27, 2023 2023]) and is present in JSC-1A as Ca_0.7Na_0.3—Al_1.7Si_2.3O₈ (Owens, “Characterization summary of JSC-1AF lunar Mare regolith simulant,” NASA (2006)), whereas olivine contained in JSC-1A is represented with two end members: (1) Mg₂SiO₄ (forsterite) and (2) Fe₂SiO₄ (fayalite) (Geologyscience (2018) Olivine [Online]. geologyscience.com: GeologyScience. Available: https://geologyscience.com/minerals/olivine/[Accessed Jul. 27, 2023]) and is present in JSC-1A as Mg_1.46Fe_0.54SiO₄ (Owens, “Characterization summary of JSC-1AF lunar Mare regolith simulant,” NASA (2006)). The compositions of plagioclase and olivine are also expressed as oxides (Owens, “Characterization summary of JSC-1AF lunar Mare regolith simulant,” NASA (2006)).

TABLE 1
Oxide species of JSC-1A by mass fraction (Cannon,
“Planetary simulant database,” Free resource for
regolith simulant information [Online].
Available: https://simulantdb. com/simulants/
jsc1.php [Accessed Jun. 14, 2022])
Oxide Species Weight %
SiO2          47.40
Al2O3         16.10
Fe2O3         11.40
CaO           10.50
MgO           7.72
Na2O          2.94
TiO2          1.56
K2O           0.80
P2O5          0.59
MnO           0.18
Cr2O3         0.03
Total         99.22

Methodologies

The carbothermal reduction of JSC-1A was examined both thermodynamically and experimentally to examine different reaction conditions coupled to solid-state materials characterization.

Thermodynamic Analysis

The carbothermal reduction reactions of all oxides pre-sent in the simulant along with the equilibrium temperatures (ΔG=0) above which the reaction becomes spontaneous (ΔG<0) are given in Table 2 for lunar surface pressure of 3×10⁻¹⁵ bar using available thermodynamic properties (Roine, HSC Chemistry 7.11.7.11 ed.: Metso Outotec (2009)). These results provided guidance for favorable operating conditions and associated solar concentrating infrastructure. However, they do not account for changes in activity coefficients and other reaction pathways at lower reduction temperatures or the formation of more favorable intermediaries (e.g., SiC). Increases in pressure are also expected as the reactions occur and gaseous products are produced including volatiles, metals/metalloids, and oxides. The addition of C significantly reduces the reduction temperature to enable the direct harvesting of Si(g) or Si from SiO₂ and Fe or Fe(g) from Fe₂O₃. Reactions with Fe₂O₃ were predicted at the lowest temperature with respect to other reactions, and CaO, MgO, and Al₂O₃ reactions were favorable at higher temperatures. Na₂O, MnO, and Cr₂O₃ reactions as well as SiO₂ and TiO₂ reactions were favorable at similar temperatures.

TABLE 2
Carbothermal reduction reaction equations of all oxide
species present in JSC-1A simulant along with equilibrium
temperatures (ΔG = 0) above which these reactions become
spontaneous (ΔG < 0) at lunar surface pressure of 3 × 10−15 bar
	Oxide Species	Reaction	Teq, ° C.
	SiO2	SiO2 + 2C → Si + 2CO(g)	485.3
	Al2O3	Al2O3 + 3C → 2Al + 3CO(g)	712.3
	Fe2O3	Fe2O3 + 3C → 2Fe + 3CO(g)	83.0
	CaO	CaO + C → Ca + CO(g)	861.3
	MgO	MgO + C → Mg + CO(g)	712.2
	Na2O	Na2O + C → 2Na + CO(g)	294.6
	TiO2	TiO2 + 2C → Ti + 2CO(g)	506.8
	K2O	K2O + C → 2K + CO(g)	225.6
	P2O5	P2O5 + 5C → 2P + 5CO(g)	136.0
	MnO	MnO + C → Mn + CO(g)	331.5
	Cr2O3	Cr2O3 + 3C → 2Cr + 3CO(g)	332.7

Thermodynamic analysis based on the oxide compositions of JSC-1A is useful because the minerals in JSC-1A are broken down to their oxide compositions thermodynamically to produce similar equilibrium compositions since equilibrium compositions are only functions of elemental composition, activity coefficient, total pressure, and temperature. A thermodynamic analysis examining the carbothermal reduction of JSC-1A was performed to predict equilibrium chemical compositions as a function of temperature and pressure at a C to JSC-1A mass ratio of 0.331. The mass ratio represents the exact stoichiometry needed to carbothermally reduce all available oxides to metals/metalloids. Multiphase equilibrium calculations via Gibb's Free Energy minimization for isobaric processes were conducted for JSC-1A+C compositions (Roine, HSC Chemistry 7.11.7.11 ed.: Metso Outotec (2009)). All possible species were considered and classified into two phases: (1) gases, (2) solids and liquids (e.g., oxides, carbides, elements). A closed, isobaric system and an ideal solid solution of JSC-1A+C were assumed.

The equilibrium temperature and pressure ranges studied were 25≤T^eq≤2000° C. and 3×10⁻¹⁵≤p^eq≤0.1 bar, respectively. The molar fraction of SiO₂ and the conversions of SiO₂ to Si(g), SiO(g), and SiC are given, respectively, as:

$\begin{array}{cc}{X}_{{\mathrm{SiO}}_2}=\frac{n_{{\mathrm{SiO}}_2}^{\mathrm{eq}}}{n_{{\mathrm{SiO}}_2,i}},X_{\mathrm{Si}\left(g\right)}=\frac{n_{\mathrm{Si}\left(g\right)}^{\mathrm{eq}}}{n_{{\mathrm{SiO}}_2,i}},& \left(1\right)\end{array}$ $X_{\mathrm{SiO}\left(g\right)}=\frac{n_{\mathrm{SiO}\left(g\right)}^{\mathrm{eq}}}{n_{{\mathrm{SiO}}_2,i}},\mathrm{and}{X}_{\mathrm{SiC}}=\frac{n_{\mathrm{SiC}}^{\mathrm{eq}}}{n_{{\mathrm{SiO}}_2,i}}$

where

$n_{Si{O}_2}^{eq},n_{Si\left(g\right)}^{eq},n_{SiO\left(g\right)}^{eq},\mathrm{and}{n}_{SiC}^{eq}$

are molar amounts of SiO₂, Si(g), SiO(g), and SiC, respectively, at equilibrium and

$n_{{\mathrm{SiO}}_{2,i}}^{eq}$

is the initial molar amount of SiO₂. SiO(g) was considered due to its stability at elevated temperatures coupled to disproportionation at lower temperatures to Si and SiO₂. SiC has been previously observed as an intermediate at lower temperatures in Si production ((Loutzenhiser et al., 2010). Si was considered but not presented among the products in this work due to its low amounts and less favorability compared to Si(g). The conversions of Fe₂O₃ to Fe and Fe(g) are given, respectively, as:

$\begin{array}{cc}{X}_{\mathrm{Fe}}=\frac{n_{\mathrm{Fe}}^{\mathrm{eq}}}{2n_{{\mathrm{Fe}}_2{O}_3,i}}\mathrm{and}{X}_{\mathrm{Fe}\left(g\right)}=\frac{n_{\mathrm{Fe}\left(g\right)}^{\mathrm{eq}}}{2n_{{\mathrm{Fe}}_2{O}_3,i}}& \left(2\right)\end{array}$

where

$n_{Fe}^{eq}\mathrm{and}{n}_{Fe\left(g\right)}^{eq}$

are molar amounts of Fe and Fe(g) at chemical equilibrium, respectively, and n_Fe2O3,i is the initial molar amount of Fe₂O₃.

Thermogravimetry

The carbothermal reduction of JSC-1A was examined in a thermogravimetric analyzer (TGA, NETZSCH STA 449 F3 Jupiter, mass resolution of 1 μg). The schematic of the experiment setup is shown in FIG. 3. Activated carbon was used as the reducing agent because the C content is primarily responsible for the reduction reactions via CH₄ (Balasubramaniam et al., “The Reduction of Lunar Regolith by Carbothermal Processing Using Methane,” Int. J. Miner. Process., 96:54-61 (2010)). JSC-1A regolith simulant [mean particle size: 81-105 μm (Cannon, “Planetary simulant database,” Free resource for regolith simulant information [Online]. Available: https://simulantdb.com/simulants/jsc1.php [Accessed 14 Jun. 2022]), ORBITEC] was mechanically mixed with activated carbon (powdered, 100-400 mesh, Sigma-Aldrich) at a 0.331 mass ratio of activated carbon to JSC-1A to examine carbothermal reduction with stoichiometric C. Mixed samples of 67.6 mg were thinly spread on a flat Al₂O₃crucible in direct contact with an S-type thermocouple to measure the sample temperature over time. An Ar gas flow (99.999% Ultra High Purity) of 100 mL_N/min (where LN denotes liters at normal conditions; mass flow rates are calculated at 273 K and 1 atm) was introduced from the bottom of the TGA to create a low O₂ concentration analogous to lunar conditions. Temporal product gas evolutions were measured with gas chromatography (GC, Agilent, 490 Micro, 134-1 Hz) and mass spectrometry (MS, Omnistar GSD320 Gas Analysis System, 3-1 Hz). Due to the higher sampling rate of the MS, measurements from the GC were used to calibrate MS measurements from ionic current to molar concentrations. The samples were heated to 500° C. at 20 K/min, held isothermally for 2 h to remove moisture, further heated to 1500° C. at 10 K/min, held isothermally for 20 min, and cooled down to ambient temperature at a rate of 5 K/min. The sample mass was measured before and after the experiments (Mettler-Toledo ML54, mass resolution 0.1 mg) to verify mass loss measured by the TGA. Additional blank runs were performed on the TGA to correct for mass drift due to gas dynamics.

Ultra-High Vacuum Experiments

Ultra-high vacuum experiments with rapid heating rates were conducted to analyze volatiles under ultra-high vacuum (UHV, ~10⁻¹² bar) conditions. The schematic of the experiment setup is shown in FIG. 4. The sample holder (Ta strip) was contained in the UHV chamber equipped with a Quadrupole Mass Spectrometer (QMS, Pfeiffer Vacuum Prisma Plus QMG 220C-SEM) to measure evolving gases. The chamber was continuously pumped down to maintain the ultra-high vacuum condition throughout the experiments. A JSC-1A+C sample of ~3 mg with the same mass ratio as for TGA was mounted on a Ta strip as a dense layer of ~10 mm² area via H₂O suspension (dried in air at 300 K), and then heated to the desired high temperatures by resistive heating at a rate of 50 K/s. The same sample was used across all experiments to investigate the specific volatile that dissipates from the sample at various final temperatures. The sample temperature was monitored by a high-performance infrared pyrometer (Raytek, measurement range: 540-3000° C.). Cu foils were mounted 3 mm above the sample holder in the setup of each experiment to collect metal and metalloid volatiles for subsequent solid-state characterization. Five experiments were performed at different final temperatures using the ultra-high vacuum apparatus. The sample was heated up to the highest temperature in multiple isothermal steps before being cooled down to room temperature. The following schedules were used for each experiment:

1. The sample was heated in the first experiment to 750° C. and held isothermally for 3.5 min, and then heated to 1000° C. and held isothermally for 60 min.

2. The sample was reheated in the second experiment to 750° C. and held isothermally for 3.5 min, heated to 1000° C. and held isothermally for 7.5 min, and then heated to 1250° C. and held isothermally for 60 min.

3. The sample was reheated in the third experiment to 750° C. and held isothermally for 3.4 min, heated to 1000° C. and held isothermally for 7 min, heated to 1250° C. and held isothermally for 11 min, and then heated to 1500° C. and held isothermally for 60 min.

4. The sample was reheated in the fourth experiment to 750° C. and held isothermally for 3.5 min, heated to 1000° C. and held isothermally for 7 min, heated to 1250° C. and held isothermally for 11 min, heated to 1500° C. and held isothermally for 16 min, and then heated to 1750° C. and held isothermally for 30 min.

5. The sample was reheated in the fifth experiment to 750° C. and held isothermally for 3.4 min, heated to 1000° C. and held isothermally for 7 min, heated to 1250° C. and held isothermally for 11 min, heated to 1500° C. and held isothermally for 16 min, heated to 1750° C. and held isothermally for 15 min, and then heated to 2000° C. and held isothermally for 10 min.

The duration of the isotherms at higher temperatures was reduced to avoid melting of the Cu foils or the sample holder. Throughout the experiments, a slight increase in pressure was observed in steps corresponding to the different isotherms. The chamber pressure was at ~10⁻¹² bar at room temperature and increased up to ~10⁻⁹ bar at 2000° C. This increase in pressure was presumably due to radiative heating from the Ta strip desorbing water from the chamber walls. The work was used to corroborate observations from the TGA and extend the analysis to lower pressures.

Solid-State Surface Material Characterization

Solid-state surface material characterization was conducted via scanning electron microscopy (SEM, Zeiss Ultra60 FE, Oberkochen, Germany) and energy-dispersive X-ray spectroscopy (EDS/X, Zeiss Ultra60 FE, Oberkochen, Germany) to examine particle morphology and elemental distribution of samples before and after TGA. Further comparisons using X-ray photoelectron spectroscopy (XPS, ThermoFisher Scientific, Al Kα radiation) and X-ray diffractometry (XRD, Malvern PANalytical Alpha-1, Cu Kal radiation) were pursued to verify the elemental compositions at the surface and assess chemical compositions and conversions of the sample surface and assess chemical compositions and conversions of the sample surface. Transmission electron microscopy (TEM, FEI Tecnai G2 F30) was used to assess the atomic structure of the sample after TGA. EDS/X was used to characterize elemental distribution of the copper cooling plate for the ultra-high vacuum experiments.

Results and Discussion

The results from thermodynamic analysis, TGA, and ultra-high vacuum analysis, are discussed along with solid-state surface material characterization.

Thermodynamic Analysis

A total pressure of p^eq=10⁻⁸ bar was assumed for isobaric conditions to account for pressure increases due to evolving gases from the sample, and an ideal solid solution was used to estimate the solid fugacity. X for the carbothermal reduction of JSC-1A are given in FIGS. 5A-5B for Si-species (FIG. 5A) and Fe-species (FIG. 5B). SiO₂ dissociated from its mineral compounds as T^eq increased, and rapidly decreased at T^eq>600° C. as SiC became more favorable. Significant SiC was predicted at 600≤T^eq≤900° C., which rapidly decreased as SiO(g) became more favorable at T^eq>900° C. Very little Si was predicted at 900≤T^eq-≤1300° C. with significant Si(g) predicted at T^eq>1100° C., corresponding to rapidly decreasing SiC and SiO(g). No changes in equilibrium compositions were predicted at 1400<T^eq<2000° C., which corresponded to X_Si(g)~0.9, indicating that full conversion from SiO₂ to Si or Si(g) was not favorable over the given temperature range. A small amount of FeO was predicted at 25≤T^eq-≤300° C. before large amounts of Fe were predicted at T^eq>300° C. Fe rapidly decreased at 700° C. as Fe(g) became more thermodynamically favorable with X_Fe(g)~1 at T^eq>1200° C. Fe₃O₄ was present at chemical equilibrium for lower temperatures.

The effects of the total pressure were also investigated at 3×10⁻¹⁵≤p^eq≤0.1 bar for Si-species shown in FIGS. 6A-6B for SiO₂ and Si(g) (FIG. 6A), and SiO and SiC (FIG. 6B). SiO₂ rapidly decreased at lower T^eq with lower p^eq and the predicted Si-species shifted towards higher T^eq with increasing p^eq in accordance with Le Chatelier's principle. Si(g) was only predicted at p^eq≤10⁻⁸ bar for all T^eq examined. Si was predicted in significantly smaller amounts for all investigated p^eq, indicating that Si(g) was expected to be more favorable compared to Si for ISRU.

The effects of the total pressure for Fe-species at 3×10⁻¹⁵≤p^eq≤0.1 bar were examined, and the results are shown in FIG. 7 for Fe and Fe(g). The predicted Fe-species also shifted towards higher T^eq as p^eq increased in accordance with Le Chatelier's principle. Similar trends to Si(g) were observed for Fe and Fe(g): Larger amounts of Fe were predicted at elevated p^eq and significant amounts of Fe(g) were predicted for p^eq≤10⁻⁸ bar, which provides strong evidence for the viability of producing Fe for ISRU.

Conclusively, the analysis shows that X_Fe(g)→1 was forecast at T^eq≥1300° C. for p^eq=10⁻⁸ bar and at T^eq→850° C. for for p^eq=3×10⁻¹⁵ bar. X_Si(g)~90% was forecast at T^eq≥1450° C. for p^eq=10⁻⁸ bar and at T^eq≥1000° C. for p^eq=3×10⁻¹⁵ bar. The results also indicate favorable T^eq and low lunar surface pressures are advantageous for the solar-driven carbothermal reduction of the JSC-1A lunar regolith.

Thermogravimetry

The results from the TGA are shown in FIG. 8 with normalized mass change Δm/m_i×100%, (dotted line), CO molar concentration y_CO×100%, (solid) line with the MS signal (amu 28) calibrated by GC signal of y_CO, and temperature T (dashed line) versus time. An initial decrease in Δm/m_i×100% was observed, likely due to activated carbon pyrolysis, evidenced by the small peak of CO at t=20 min. Increasing y_CO was observed while ramping up to 1500° C. with small initial peaks that presumably represented the reduction reactions of Fe₂O₃ to Fe₃O₄ and subsequent reactions to FeO and Fe or Fe(g), followed by a significantly larger peak forecasting the conversions of SiO₂ to SiO(g) or SiC and a subsequent reaction to Si or Si(g). Smaller CO peaks were expected for reactions of Fe-species due to smaller initial amounts of Fe₂O₃ in the simulant compared to SiO₂ (Table 1). These reactions were also expected to occur at lower temperatures than reactions of Si-species based on thermodynamics. A significant decrease in Δm/m_i×100%~50% was observed that coincided with larger amounts of CO measured in the product gases as C reacted with oxides/minerals to produce CO. The results also showed that the sample continued to lose mass after CO evolved from the sample, indicative of volatile Si(g), SiO(g), and Fe(g) evolving from the samples. The mass loss reflects C conversions to CO and volatiles leaving the sample as predicted from the thermodynamics, noting that this process was not in chemical equilibrium as gaseous products were continually removed. A significant amount of the sample from the carbothermal reduction as melted onto the Al₂O₃ crucible as the melting points of Si and FeO were exceeded. Only about a quarter of the post-experiment sample was not completely melted and was spread evenly above the melted sample.

The measured initial and final sample masses via an analytical scale are provided in Table 3 along with the computed mass of CO evolved from the sample. Measured final sample mass confirmed the ~50% mass loss during experiment as measured by TGA. The results from Table 3 also show that CO evolution only accounted for ~45% of sample mass loss during the experiment. The cumulated volatile loss (shaded light gray region) and evolution of CO (shaded dark gray region) from sample are given in FIG. 9. The mass balance shown in FIG. 9 confirmed that 45% sample mass loss was due to CO evolution, and the rest was due to volatiles leaving the sample, presumably SiO(g), Fe(g), and Si(g), based on thermodynamics. The results provided a strong foundation for further analyses to optimize the C content and produce Si— and Fe— compounds from JSC-1A in a solar-driven process. Further analyses of the sample were conducted using solid-state material characterization tools for better understanding of the transpired reactions.

TABLE 3
Sample mass measured before and after the carbothermal reduction
experiment and computed mass of CO evolved from the sample
Species                          m, mg
Initial sample mass (JSC-1A + C) 67.6
Final sample mass (loose + melted sample) 33.8
CO evolved from sample           15.2
Balance (moisture loss + evolved volatiles) 18.6

Ultra-High Vacuum Experiments

The heating profiles (dashed line) of the five experiments and the MS signals (amu 28, solid line) corresponding to CO evolving from the sample during experiments are shown in FIGS. 10A-10B for the highest isotherms of 1000° C. (FIG. 10A) and 1250° C. (FIG. 10B) and in FIGS. 11A-11C for the highest isotherms of 1500° C. (FIG. 11A), 1750° C. (FIG. 11B), and 2000° C. (FIG. 11C). CO peaks were observed at the beginning of every isotherm in all experiments except at 750° C. in 1750° C. (FIG. 11B) and 2000° C. (FIG. 11C) experiments. The CO peaks were indicative of carbothermal reduction with larger conversions occurring at higher isotherms in each experiment. The reduction increased between each experiment up to the 1500° C. experiment (FIG. 11A). The results were supported by thermodynamics forecasting of more oxides/minerals reducing carbothermally, and volatile formation becoming more favorable with increasing temperature. During the 1750° C. experiment, the reduction significantly decreased, likely a result of previous volatile evolution at lower temperatures from the sample. During the 2000° C. experiment, increased CO was observed, likely due to other reactions becoming favorable [e.g., Si(g)]. CO peaks at a given isotherm reduced in consecutive experiments with higher reductions and volatile formation occurring in previous experiments. The absence of CO peaks in the 1750 and 2000° C. experiments at 750° C. were likely due to this.

Solid-State Surface Material Characterization

XPS was used to measure the elemental atomic percentages of the sample surfaces before and after experimentation with the results listed in Table 4. The JSC-1A sample was scanned prior to carbothermal reduction to measure the simulant elemental compositions, and the residual sample from TGA that melted onto the Al₂O₃ crucible was also scanned to identify reactions and elements present. The sample mixture of JSC-1A and activated carbon sample was scanned as well, and the results indicated smaller amounts of JSC-1A were present due to higher volume of activated carbon overwhelming the sample mixture (activated carbon has lower density than JSC-1A). Three different spots for each of the sample surfaces were scanned with circular spots of 400 μm in diameter. The scanned atomic percentages were compared to the weight percentages reported in Table 1 and converted to atomic percentages.

Cr, K, P, and Mn were below the detection limit of the XPS. Smaller amounts of Ca, Mg, and O and more Al, Fe, Na, and Ti were detected in the JSC-1A scans compared to the reported JSC-1A composition with similar amounts of Si. Residual C was also present in the surface JSC-1A scans due to atmospheric exposure prior to scanning (adventitious carbon contamination). The presence of C resulted in the different atomic percentages in the simulant compared to Table 1, presumably because the surface compositions were quantified, and the reported JSC-1A composition was bulk compositions.

TABLE 4
The atomic percentages of elements present in JSC-1A, and at three locations on
the surface of JSC-1A and samples before and after the carbothermal reduction
experiment measured by XPS
	Atomic %
				Post-experiment melted
Element	JSC-1A	JSC-1A	JSC-1A + C	sample
scan	(Cannon)	Spot 1	Spot 2	Spot 3	Spot 1	Spot 2	Spot 3	Spot 1	Spot 2	Spot 3
Si2p	16.09	16.20	17.53	16.71	3.56	1.60	2.03	14.34	12.46	12.50
Ca2p	3.82	—	2.17	2.45	—	—	—	3.36	3.16	2.92
Ti2p	0.40	—	0.52	0.63	—	—	—	0.79	0.87	2.32
O1s	69.84	57.17	52.81	52.83	22.88	22.64	19.15	58.21	60.61	54.46
C1s	—	9.34	8.98	8.95	69.14	73.59	76.81	8.80	9.97	18.96

The sample variance of Si and Ti scans significantly increased between before and after TGA, evidencing the increased variation of elemental compositions between scan locations after TGA, indicative of carbothermal reactions of Si— and Ti— compounds. The sample variance of Ca scans did not change significantly between before and after TGA, presumably due to Ca— compounds not carbothermally reducing (Table 2 indicate that reactions with Ca— compounds occur at elevated temperatures). Fe, Mg, and Na were not detectable after the TGA, likely due to volatile transport or relative compositional changes.

The Al scans after the experiment did not yield useful information, as the samples were melted onto the Al₂O₃ crucible, introducing possible sample contamination. Significant amounts of oxygen observed from the O scans after experimentation suggested that oxides were still present, potentially from unreacted oxides/minerals, reduced oxides/minerals upon reactions, and Al₂O₃ crucible contributions. The O scans of the surface sample before and after experiment also accounted for oxide layer formations on the sample due to atmospheric exposure. The C scans of spot 1 and 2 after TGA were comparable with the C scans of JSC-1A, indicating that most, if not all, of the activated carbon was reacted. The higher amount of C in spot 3 scan likely indicated carbide formations.

XPS was also used to measure peak locations as a function of binding energy for elemental presence in JSC-1A samples before and after TGA. The scans were performed on the same spots as the atomic percentage scans, and the peak locations were identified from each scan and are listed in Table 5.

TABLE 5
Peak locations of elements as a function of binding energy at three locations on
the surface of JSC-1A and samples before and after the carbothermal reduction
experiment measured by XPS
		Peak Binding Energy
	Electron							Post-experiment
Element	Energy	JSC-1A	JSC-1A + C	melted sample
Scan	Level	Spot 1	Spot 2	Spot 3	Spot 1	Spot 2	Spot 3	Spot 1	Spot 2	Spot 3
Si2p	2p3	103.84	103.90	103.23	106.00	106.72	106.65	103.47	104.65	104.40
	2p1	104.57	104.43	103.96	106.54	107.45	107.18	104.20	105.38	105.13
	2p3	—	—	—	—	—	—	107.38	107.24	106.83
	2p1	—	—	—	—	—	—	107.91	107.77	107.37
Ca2p	2p3	—	349.16	348.40	—	—	—	349.17	348.54	349.95
	2p1	—	352.76	352.00	—	—	—	352.53	351.85	353.55
	2p3	—	352.24	351.35	—	—	—	352.50	352.00	352.83
	2p1	—	355.54	354.65	—	—	—	355.99	355.58	356.37
Ti2p	2p3	—	459.86	459.47	—	—	—	462.24	462.21	462.46
	2p1	—	465.86	465.47	—	—	—	468.24	468.21	468.46
O1s	1 s	531.95	532.58	532.24	531.62	532.98	532.52	532.91	534.10	534.07
		533.11	534.03	533.94	534.27	535.49	534.53	534.84	536.67	536.40
		537.37	537.91	537.53	537.45	538.18	536.50			—
C1s	1 s	285.84	286.29	286.05	285.24	284.97	285.93	286.00	288.07	286.97
		287.72	290.39	287.78	286.92	286.64	288.67	289.88	290.33	289.19
		290.13	294.98	290.30	288.86	289.17	290.95	—	292.52	291.48
		294.44		294.69	292.24	292.20				—

The absence of Fe, Mg, and Na was consistent with the elemental analysis after TGA. Slight peak shifts between the JSC-1A and the mixed sample for O and C scans were observed, suggesting variations in peak locations between scanning sessions. Significant peak shifts in Si scans were observed between the JSC-1A and the mixed sample, suggesting room temperature reactions with Si-during mechanical mixing of JSC-1A and activated carbon. The Si and Ti peaks after TGA showed significant shifting compared to the JSC-1A peak locations, indicative of reactions occurring with Si— and Ti— compounds in the sample during the experiment. Additional Si peaks were also observed, indicating formation of new Si— compounds due to reduction. The peak locations of Ca in the sample after TGA did not differ significantly from the peak locations of Ca for JSC-1A. This was evidence of no reaction with Ca— compounds. A reduction in the number of peaks was observed for O before and after TGA samples. The third O peak in the mixed sample was not observed after TGA, indicating that reduction reactions occurred during the experiment: O was bound with C to form CO as supported by the TGA results. A similar reduction in the number of peaks was also observed for the C scans between the before and after TGA samples. The first C peaks observed in the mixed sample were likely subsequently reacted during the experiment to reduce the oxides/minerals in the sample and produce CO as the byproduct. The other O and C peaks were attributed to oxide layer formation and adventitious carbon contamination, respectively, due to atmospheric exposure.

The SEM and EDS/X were used to examine particle morpholog and elemental distributions, respectively. The capture and scans are shown in FIGS. 12A-12E for JSC-1A with sample image (FIG. 12A), O scan (FIG. 12B), Al scan (FIG. 12C), Si scan (FIG. 12D), and Mg scan (FIG. 12E). The particle morphology of JSC-1A was examined with the SEM and featured irregular particle structures with many jagged edges and polydisperse particles (Schieber et al., “Characterization of H₂O transport through Johnson Space Center number 1A lunar regolith simulant at low pressure for in-situ resource utilization,” Phys. Fluids, 33:037117 (2021)). All elements of the oxides/minerals present in JSC-1A were detected by the EDS/X, except for Cr, Mn, and P for the JSC-1A surface scan due to their low presence. O, Si, Al, and Mg scans coincided with one another, indicating that the JSC-1A sample was homogenous. The Mg scan revealed higher concentration spots consistently throughout the material. Scans of all other elements were consistent with the O, Si, and Al scans.

SEM with EDS/X scans are given in FIGS. 13A-13G for sample image (FIG. 13A), O scan (FIG. 13B), Al scan (FIG. 13C), Mg scan (FIG. 13D), Si scan (FIG. 13E), Fe scan (FIG. 13F), and Ti scan (FIG. 13G) for sample after TGA. Circular-like depositions were observed on the surface and attributed to Si, Fe, and Ti, indicative of carbothermal reduction of the related oxides/minerals during TGA, resulting in inhomogeneity in sample morphology. Si, Fe, and Ti scans overlapped for some of the depositions, suggesting elemental binding. The Mg scan of the after TGA sample revealed Mg— compounds were still present and unreacted. Fe and Mg were detected by the EDS/X and the SEM in contrast to XPS suggesting that Fe and Mg were present in the subsurface of the sample after TGA. Ca was also detected without significant changes from the JSC-1A scan. The Al and O scans of the sample after TGA were consistent, confirming the presence of Al₂O₃ from the crucible.

XRD was performed on the sample after TGA to determine prominent chemical compositions present in the sample. The XRD normalized intensity patterns are shown in FIGS. 14A-14B for loose portion of the sample which was not melted onto the Al₂O₃ crucible (FIG. 14A) and portion of the sample which was melted onto the Al₂O₃ crucible (FIG. 14B).

The patterns were analyzed using an XRD analysis software (HighScore, Malvern PANalytical) via phase identification that match candidate constituents to results based on a large reference intensity patterns database (PDF-4+database, ICDD). SiC was present in the loose sample with reference intensity patterns (010751541, diamond markers) shown in FIG. 14A. The result supported the thermodynamic equilibrium analysis forecasting SiC as a stable intermediate product of carbothermal reduction of SiO₂. Analysis via HighScore suggested the presence of Mg₂SiO₄ in the sample that was melted onto the Al₂O₃ crucible with the reference intensity patterns (010778396, square markers) shown in FIG. 14B, indicating that the olivine, specifically Mg₂SiO₄, present in JSC-1A was unreacted and melted onto the Al₂O₃ crucible at elevated temperatures (Table 2 indicates that reaction with Mg— compound occurs at elevated temperature). Significant amounts of other crystal peaks were not identified because of the low matching probability with the intensity patterns database, likely due to their low compositions. However, they do show that both loose and melted samples were polycrystalline. The XRD patterns of JSC-1A (before experiments) are found elsewhere (Ray et al., “JSC-1A lunar soil simulant: Characterization, glass formation, and selected glass properties,” J. Non Cryst. Solids, 356:2369-2374 (2010); Wilkerson et al., “Outgassing behavior and heat treatment optimization of JSC-1A lunar regolith simulant,” Icarus, 400:115577 (2023)) showing the presence of major mineral phases (e.g., olivine, plagioclase) in the simulant.

The loose portion of the sample was also studied with TEM to verify the XRD results of SiC with crystalline structures. A very small fragment of the sample as seen by the TEM is given in FIGS. 15A-15C for magnifications of 2 μm (FIG. 15A), 50 nm (FIG. 15B), and 10 nm (FIG. 15C). Results showed nanowires with lattice fringes evidencing crystalline structures of the sample. The presence of an amorphous oxide layer on the surface of the sample (FIGS. 15B-15C) was also observed, confirming the O scans with XPS and EDS/X that indicated atmospheric exposure.

An EDS/X couple to TEM was used to identify elements present in the same small fragment of the sample and the results are shown in FIGS. 16A-16E for the sample image (FIG. 16A), and elemental scans of Si (FIG. 16B), C (FIG. 16C), Fe (FIG. 16D), and O (FIG. 16E). Si and C scans significantly overlapped, confirming the presence of SiC as detected by XRD. The Si scan also showed Si in locations where C or O elements were not present, confirming the presence of Si. The Fe scan not intersecting with other elemental scans confirms the presence of Fe in the sample, but in a very small quantity, providing insights into the unidentified crystal peaks in the XRD analysis of the loose sample and the low Fe signal in EDS/X analysis. The EDS/X analysis, supplemented by TEM, validated the strong presence of crystalline SiC and a low quantity presence of crystalline Si and Fe in the loose sample after TGA, and the formation of amorphous oxide layers on the surface of these crystals.

Ultra-High Vacuum Experiments

EDS/X was used on the five Cu collecting foils to examine the elemental presence of volatiles evolving from the sample. The elements detected for each Cu collecting foil are listed in Table 6, arranged in decreasing weight percentage.

TABLE 6
Elemental presence of volatiles on the Cu collecting
foils arranged by decreasing weight percentage for
the ultra-high vacuum experiments as detected by EDS/X
Highest Isothermal Step of Experiment Elements Detected by EDS/X
1000° C.                         Cu, C, O
1250° C.                         Cu, C, O, Si, Fe, Mg
1500° C.                         Cu, C, O, Fe, Al, Si
1750° C.                         Cu, C, O, Si
2000° C.                         Cu, C, Si, O

The volatiles were thinly deposited onto the Cu foils, therefore, elements representing the Cu foils, namely Cu, C, and O, were detected on the higher end of the elemental distribution. No volatiles were collected during the 1000° C. experiment, despite the presence of CO peaks, indicating that carbothermal reduction occurred without volatile evolution. Mg evolved at 1250° C., Al evolved at 1500° C., and Fe evolved at 1250 and 1500° C. Due to the low presence of Mg, Al, and Fe in the simulant, the volatiles were not observed in subsequent experiments. Si evolved at ≥1250° C. However, Si decreased during the 1500° C. experiment, before increasing in subsequent experiments as evidenced by the weight percentage (Table 6) coupled with the CO peaks (FIGS. 11A-11C). SiO(g) likely evolved at 1250° C. and, potentially, at 1500° C. in the subsequent experiment. The reduction in Si during the 1500° C. experiment suggested the depletion of SiO(g). Increases in Si in subsequent experiments were likely from evolving Si(g). All volatiles, except Si, evolved from the sample at ≤1500° C. Ca was not detected on the Cu foils, which was consistent with TGA results where Ca was not observed to reduce.

Conclusions

The carbothermal reduction of JSC-1A via activated carbon was investigated in this work by conducting thermodynamic equilibrium analysis, thermogravimetry, ultra-high vacuum thermal heating experiments, and solid-state surface material characterization. Thermodynamic equilibrium analysis showed complete conversions of Fe₂O₃ to Fe or Fe(g) and up to 90% conversion of SiO₂ to Si or Si(g) at lunar surface pressure and relatively low temperatures as described by Le Chatelier's principle. This analysis showed the potential advantage of the low-pressure lunar surface by reducing the onset reaction temperatures for ISRU, specifically for carbothermal reductions of lunar regolith. Activated carbon was mixed with JSC-1A regolith to act as the reducing agent to produce useful metals or metalloids. Thermogravimetry showed ~50% mass loss that coincided with CO, indicative of carbothermal reduction. Mass loss continued to occur after CO evolution, which was indicative of volatiles, such as SiO(g), Si(g), and Fe(g), being released from sample at elevated temperatures. Results from the surface material characterization strongly indicated that reduction reactions occurred with Si—, Fe—, and Ti— compounds causing the particle morphology, elemental compositions, and chemical compositions to significantly change. The strong presence of crystalline SiC and low quantities of crystalline Si and Fe were confirmed to be formed upon reduction via TGA with a thin amorphous oxide layer forming around the crystals due to atmospheric exposure. Results also supported the evolution of volatiles from the sample and suggest those volatiles to include Fe(g) and Mg(g). Elemental Cr, K, P, and Mn were not identified by any of the surface characterization tools, as the concentrations of the related oxides/minerals were low. The ultra-high vacuum experiments showed Fe—, Mg—, and Al— volatiles evolving from the sample mixture at relatively low temperatures whereas Si-volatiles evolved continuously over the different temperatures examined.

The work lays the foundation and pathway to further investigate the carbothermal reduction of lunar regolith on the Moon surface and subsequently design and optimize reactors. Further work is necessary to investigate the dependence of reaction kinetics on other variables, such as the amounts of C mixed with JSC-1A and heating/cooling rates. Further investigation of carbothermal reduction of other types of simulants is also necessary to understand the feasibility of producing different metals and metalloids due to the different oxide and mineral compositions. Simulants with Fe present as FeO should be investigated for carbothermal reduction, as FeO is more commonly found in lunar regolith compared to Fe₂O₃ (Papike et al., “The lunar regolith: Chemistry, mineralogy, and petrology,” Rev. Geophys., 20:761-826 (1982)). This work will provide references for comparison with guidance on operating conditions (e.g., temperature and pressure) for future experiments with concentrating solar irradiation as the heat source to carbothermally reduce lunar simulants.

Example 2—Metal and Metalloid Production from Lunar Regolith Simulants Via Carbothermal Reduction: Thermodynamic and Experimental Analyses

Metal and metalloid production from lunar regolith simulants via carbothermal reduction was investigated using activated carbon under inert conditions. JSC-1A, LMS-1, and LHS-1 lunar regolith simulants were thermodynamically and experimentally studied over a range of conditions to assess metal and metalloid production as function of oxide and mineral composition. Thermodynamic analyses were conducted for the carbothermal reduction of LMS-1 and LHS-1 mixed with stoichiometric C to identify equilibrium compositions and favorable reactions. The temperature and pressure ranges studied were between 25 and 2000° C., and 3×10⁻¹⁵ and 1 bar, respectively. Complete conversions of SiO₂ to Si(g) and Al₂O₃ to Al(g) were forecast for all simulants at total pressures of 10⁻⁸ and 3×10⁻¹⁵ bar. Similarly, complete conversion of CaO to Ca(g) in LHS-1 was predicted at total pressures of 10⁻⁸ and 3×10⁻¹⁵ bar whereas complete conversion of MgO to Mg(g) in LMS-1 was predicted at all studied pressures. Tube furnace experiments using the three simulants mixed with stoichiometric activated carbon were conducted to examine the carbothermal reactions up to 1600° C. in 100% Ar. Mass balance analyses estimated CO evolution and vapor deposits for each experiment. Solid-state surface material characterization was conducted via energy-dispersive x-ray spectroscopy, x-ray photoelectron spectroscopy and x-ray diffractometry to identify elemental presence and their chemical compositions of post-experimental samples and collected metal vapors. Oxides were strongly present in post-experimental samples indicating that samples are not fully reduced. SiC, silicates, and elements (e.g., Si, Al, Fe) were also identified in significant amounts that are potentially useful for lunar in-situ resource utilization. Si compounds were prominently observed on all vapor collectors with small amounts of Ca compounds observed for JSC-1A+C and LHS-1+C experiments.

Methodologies

The thermodynamics of carbothermal reduction of LMS-1 and LHS-1 were modeled to predict equilibrium chemical compositions as a function of temperature and pressure. The carbothermal reduction of these simulants along with Johnson Space Center (JSC-1A) simulant was also examined experimentally with coupling to solid-state surface material characterization to investigate reaction conditions and characterize reactions.

Chemical and Mineral Compositions of Lunar Regolith Simulants

Three different types of lunar regolith simulants were investigated in this work representing the different regions of the Moon to study the reaction mechanism and feasibility of metal or metalloid production as a function of varying regolith compositions: (1) JSC-1A; (2) LMS-1; and (3) LHS-1 (Table 7).

TABLE 7
Oxide compositions of JSC-1A, LMS-1, and LHS-1 by mass fraction.
	Wt. %
Oxide Species	JSC-1A	LMS-1	LHS-1
SiO2	47.40	46.9	51.2
Al2O3	16.10	12.4	26.6
FeO	—	8.6	2.7
Fe2O3	11.40	—	—
CaO	10.50	7.0	12.8
MgO	7.72	16.8	1.6
Na2O	2.94	1.7	2.9
TiO2	1.56	3.6	0.6
K2O	0.80	0.7	0.5
P2O5	0.59	0.2	0.1
MnO	0.18	0.2	0.1
Cr2O3	0.03	—	—
Total	99.22	98.1	99.0

JSC-1A is a low-Ti mare simulant with bulk chemistry simulating Apollo 14 sample 14163 (Mckay et al., “JSC-1: A NEW LUNAR SOIL SIMULANT,” Engineering, Construction, and Operations in Space IV, p. 857-86638 (1994). Substantial amounts of SiO₂, Al₂O₃, and Fe₂O₃ (47.40%, 16.10%, and 11.40%, respectively) are present for ISRU production (Table 7). The most abundant minerals in JSC-1A are natural basaltic glass (49.3%), plagioclase (37.1%), and olivine (9.0%), with <1.5% abundances of other minerals (Taylor et al., “JSC-1 as the Lunar Soil Simulant of Choice,” Meteoritics and Planetary Science Supplement, 40:5180 (2005). The plagioclase contained in JSC-1A is present as Ca_0.7Na_0.3Al_1.7Si_2.3O₈ (Owens, “Characterization Summary of JSC-1AF Lunar Mare Regolith Simulant,” NASA. p. 3-5 (2006)), whereas olivine is present in JSC-1A as Mg_1.46Fe_0.54SiO₄ (Owens, “Characterization Summary of JSC-1AF Lunar Mare Regolith Simulant,” NASA. p. 3-5 (2006)). The compositions of plagioclase and olivine are also expressed as oxides (Owens, “Characterization Summary of JSC-1AF Lunar Mare Regolith Simulant,” NASA. pp. 3-5 (2006)).

LMS-1 is a low- to moderate-Ti simulant representing the average or generic composition of mare soils found in darker cratered regions of the Moon (Tech, “Lunar Mare (LMS-1) High-Fidelity Moon Dust Simulant,” [cited 2024, 27 Feb.]; Available from: https://spaceresourcetech.com/collections/lunar-simulants/products/lms-1-lunar-mare-simulant). Substantial amounts of SiO₂, MgO, and Al₂O₃ (46.9%, 16.8%, and 12.4%, respectively) are present for ISRU production. The minerals in LMS-1 by descending weight percentage are pyroxene (32.8%), natural basaltic glass (32.0%), anorthosite (19.8%), olivine (11.1%) and ilmenite (4.3%) (Tech, “Lunar Mare (LMS-1) High-Fidelity Moon Dust Simulant,” [cited 2024, 27 Feb.]; Available from: https://spaceresourcetech. com/collections/lunar-simulants/products/lms-1-lunar-mare-simulant).

Conversely, LHS-1 is a simulant representing the average or generic composition of highland soils found in the lighter cratered regions of the Moon, including the south pole (Tech, “Lunar Highlands (LHS-1) High-Fidelity Moon Dust Simulant,” [cited 2024, 27 Feb.]; Available from: https://spaceresourcetech.com/collections/lunar-simulants/products/lhs-1-lunar-highlands-simulant). Substantial amounts of SiO₂, Al₂O₃, and CaO (51.2%, 26.6%, and 12.8%, respectively) are present for ISRU production. The most abundant minerals in LHS-1 by weight percentage are plagioclase (74.4%) and natural basaltic glass (24.7%) with <0.5% of other minerals (Tech, “Lunar Highlands (LHS-1) High-Fidelity Moon Dust Simulant,” [cited 2024, 27 Feb.]; Available from: https://spaceresourcetech.com/collections/lunar-simulants/products/lhs-1-lunar-highlands-simulant). The plagioclase mineral in LHS-1 is present as CaAl₂Si₂O₈ (anorthite) (Tech, “Lunar Highlands (LHS-1) High-Fidelity Moon Dust Simulant,” [cited 27 Feb. 2024]; Available from: https://spaceresourcetech. com/collections/lunar-simulants/products/lhs-1-lunar-highlands-simulant). The natural basaltic glass contained in all simulants is a mineral with 45-52% SiO₂ by mass and other oxides (e.g., Al₂O₃, Fe₂O₃, CaO) (Lima, et al., “Basaltic glass-ceramic: A short review,” Boletín de la Sociedad Española de Cerámica y Vidrio, 61 (1): 2-12 (2022)).

Fe is more commonly present as FeO in lunar regolith (Papike et al., “The lunar regolith: Chemistry, mineralogy, and petrology,” Reviews of Geophysics, 20 (4): 761-826 (1982)) and simulants (e.g., LMS-1, LHS-1) due to the highly reduced state of lunar surface (Li, et al., “Widespread hematite at high latitudes of the Moon,” Science Advances, 6 (36): eaba1940 (2020)). However, Fe in JSC-1A is present as Fe₂O₃ which is more common in terrestrial minerals (Schreiner et al., “Thermophysical property models for lunar regolith,” Advances in Space Research, 57 (5): 1209-1222 (2016)) and is potentially found in lunar regolith due to oxygen-containing Earth wind, presence of ice near the lunar poles, and heat from interplanetary dust impacts (Li, et al., “Widespread hematite at high latitudes of the Moon,” Science Advances, 6 (36): eaba1940 (2020)). Coupling with the higher Fe content as compared to other types of simulants, including LMS-1 and LHS-1 (Tech, “Lunar Mare (LMS-1) High-Fidelity Moon Dust Simulant,” [cited 27 Feb. 2024]; Available from: https://spaceresourcetech.com/collections/lunar-simulants/products/lms-1-lunar-mare-simulant; Tech, “Lunar Highlands (LHS-1) High-Fidelity Moon Dust Simulant,” [cited 2024, 27 Feb.]; Available from: https://spaceresourcetech.com/collections/lunar-simulants/products/lhs-1-lunar-highlands-simulant; Farries et al., “Sintered or melted regolith for lunar construction: state-of-the-art review and future research directions,” Construction and Building Materials, 296:123627 (2021)), JSC-1A is an attractive lunar simulant to study since Fe is particularly useful for construction purposes on Moon (Farries et al., “Sintered or melted regolith for lunar construction: state-of-the-art review and future research directions,” Construction and Building Materials, 296:123627 (2021)).

The lunar regolith simulants were mechanically mixed with activated carbon (powdered, 100-400 mesh, Sigma-Aldrich) at mass ratios representing the exact stoichiometric amount of C needed to reduce all available oxides in a simulant to metals/metalloids. The mass ratios for C to JSC-1A, C to LMS-1, and C to LHS-1 are 0.332, 0.319, and 0.344 respectively. Activated carbon was used as the reducing agent because the C content is primarily responsible for the reduction reactions via CH₄ (Balasubramaniam et al., “The reduction of lunar regolith by carbothermal processing using methane,” International Journal of Mineral Processing, 96 (1): 54-61 (2010)).

Thermodynamic Analyses

Carbothermal reduction reactions for all oxides present in JSC-1A along with their respective equilibrium temperatures of T^eq at ΔG=0, above which the reactions becomes spontaneous (ΔG<0) are given for lunar surface pressure of 3×10⁻¹⁵ bar using available thermodynamic properties as (Kaur et al., “Thermodynamic and experimental study on carbothermal reduction of JSC-1A lunar regolith simulant for metal and metalloid production,” Advances in Space Research, 73 (8): 4024-4039 (2024)) and are applicable for simulants (e.g., LMS-1, LHS-1) containing similar oxides. The carbothermal reaction equation for FeO is represented as:

$\begin{array}{cc}\mathrm{FeO}+C\to \mathrm{Fe}+\mathrm{CO}\left(g\right)& \left(3\right)\end{array}$

for which T^eq=131.6° C. (ΔG=0) above which reaction becomes spontaneous for lunar surface pressure using available thermodynamic properties (Roine, HSC Chemistry 7.11. 2009, Metso Outotec). The carbothermal reduction of FeO is favorable at a slightly higher temperature than for Fe₂O₃, however, the reaction favorability of Fe₂O₃ and FeO are forecast at the lowest temperatures compared to other carbothermal oxide reductions. The results provide guidance for favorable reactions and operating conditions, but they do not account for activity coefficient changes and other reaction pathways that produce more favorable intermediaries [e.g., SiO(g), Al₂O(g)].

Multiphase equilibrium calculations via Gibb's free energy minimization for isobaric processes were conducted for the LMS-1+C and LHS-1+C systems. Similar thermodynamic analyses for the JSC-1A+C system have been previously performed (Kaur et al., “Thermodynamic and experimental study on carbothermal reduction of JSC-1A lunar regolith simulant for metal and metalloid production,” Advances in Space Research, 73 (8): 4024-4039 (2024)). The analyses assumed a closed, isobaric system and all possible compounds were considered and grouped into two phases: (1) gases, (2) solids and liquids (e.g., oxides, elements, carbides). LMS-1+C and LHS-1+C were assumed as ideal solid solutions.

The conversions of SiO₂ to Si(g), SiO(g), and SiC in LMS-1 and LHS-1 are given, respectively, as:

$\begin{array}{cc}{X}_{\mathrm{Si}\left(g\right)}=\frac{n_{\mathrm{Si}\left(g\right)}^{\mathrm{eq}}}{n_{{\mathrm{SiO}}_2,i}},X_{\mathrm{SiO}\left(g\right)}=\frac{n_{\mathrm{SiO}\left(g\right)}^{\mathrm{eq}}}{n_{{\mathrm{SiO}}_2,i}},\mathrm{and}{X}_{\mathrm{SiC}}=\frac{n_{\mathrm{SiC}}^{\mathrm{eq}}}{n_{{\mathrm{SiO}}_2,i}}& \left(4\right)\end{array}$

$n_{Si\left(g\right)}^{eq},n_{SiO\left(g\right)}^{eq},\mathrm{and}{n}_{SiC}^{eq}$

Where are molar amounts of Si(g), SiO(g), and SiC, respectively, at equilibrium; and n_SiO2,i is the initial molar amount of SiO₂, SiO(g) and SiC were considered as they have been previously observed as favorable intermediaries in Si production via carbothermal reduction (Loutzenhiser and Steinfeld, “Production of Si by vacuum carbothermal reduction of SiO₂ using concentrated solar energy,” JOM, 62 (9): 49-54 (2010)). The conversions of Al₂O₃ to Al₂O(g) and Al(g) in LMS-1 and LHS-1 are given, respectively, as:

$\begin{array}{cc}{X}_{{\mathrm{Al}}_{2}O\left(g\right)}=\frac{n_{{\mathrm{Al}}_{2}O\left(g\right)}^{\mathrm{eq}}}{2n_{{\mathrm{Al}}_2{O}_3,i}},\mathrm{and}{X}_{\mathrm{Al}\left(g\right)}=\frac{n_{\mathrm{Al}\left(g\right)}^{\mathrm{eq}}}{2n_{{\mathrm{Al}}_2{O}_3,i}}& \left(5\right)\end{array}$

where

$n_{A{l}_{2}O\left(g\right)}^{eq},n_{Al\left(g\right)}^{eq}$

are molar amounts of Al₂O(g) and Al(g) at chemical equilibrium, respectively; and n_Al2O3,i is the initial molar amount of Al₂O₃. Al₂O(g) was considered due to favorability as an intermediary in carbothermal reduction of Al₂O₃ to Al (Halmann and Steinfeld, “Carbothermal reduction of alumina: Thermochemical equilibrium calculations and experimental investigation,” Energy, 32 (12): 2420-2427 (2007)). The conversion of MgO to Mg(g) in LMS-1 is given as:

$\begin{array}{cc}{X}_{\mathrm{Mg}\left(g\right)}=\frac{n_{\mathrm{Mg}\left(g\right)}^{\mathrm{eq}}}{n_{\mathrm{MgO},i}}& \left(6\right)\end{array}$

Where

$n_{Mg\left(g\right)}^{eq}$

is the molar amount of Mg(g) at chemical equilibrium; and n_MgO,i is the initial molar amount of MgO. The conversion of CaO to Ca(g) in LHS-1 is given as:

$\begin{array}{cc}{X}_{\mathrm{Ca}\left(g\right)}=\frac{n_{\mathrm{Ca}\left(g\right)}^{\mathrm{eq}}}{n_{\mathrm{CaO},i}}& \left(7\right)\end{array}$

where

$n_{Ca\left(g\right)}^{eq}$

is the molar amount of Ca(g) at chemical equilibrium; and n_CaO,i is the initial molar amount of CaO. Si, Al, Ca, and Mg were considered but not presented among the products in this work due to less favorability compared to Si(g), Al(g), Ca(g), and Mg(g) at elevated temperatures.

Tube Furnace Experiments

The carbothermal reduction of JSC-1A, LMS-1, and LHS-1 was examined in a tube furnace (GSL-1700X, MTI Corporation) as depicted in FIG. 17. The Zr₂O₃ crucible (3 ml, flat bottom boat, Almath Crucibles) lined with Mo foil (≥99.9%, 0.025 mm thickness, Sigma Aldrich) was filled with ~150 mg of mixed samples and placed in the center of the Al₂O₃tube (1″ O. D., 0.75″ I. D., 42″ L, International Ceramic Engineering). An Ar gas flow (99.999% ultra-high purity, Airgas) of 100 mL_N/min (where LN denotes liters at normal conditions; mass flow rates are calculated at 273 K and 1 atm) was pretreated with an O₂ scrubber and introduced through the tube to create minimal O₂ concentrations analogous to lunar conditions. A thin circular Mo foil with ≤3 mm holes to allow gas flows was mounted at the exhaust to condense and collect vapor deposits during experimentation for further analysis. Molar concentrations of temporal product gas evolutions were measured via gas chromatography (GC, 490 Micro, 134-1 Hz, Agilent) and mass spectrometry (MS, Omnistar GSD320 Gas Analysis System, 3-1 Hz). The furnace temperature was monitored by a B-type thermocouple fitted within the heating chamber of the furnace.

The mixed samples were heated to 500° C. and held isothermally for 1 h to remove moisture, further heated to 750° C. and held isothermally for 2 h, further heated to 1000° C. and held isothermally for 2 h, further heated to 1250° C. and held isothermally for 2 h, further heated to 1500° C. and held isothermally for 2 h, and further heated to 1600° C. and held isothermally for 20 min before being cooled down to ambient temperature at a heating/cooling rate of 2 K/min. The slow heating/cooling rate and shorter isotherm at 1600° C. was to minimize the deterioration of the Al₂O₃ tube. The sample masses were measured before and after each experiment (Mettler-Toledo ML54, mass resolution 0.1 mg) to estimate mass loss and vapor production.

Solid-State Surface Material Characterization

The post-experimental samples were characterized via energy dispersive X-ray spectroscopy (EDS, Zeiss Ultra60 FE, Oberkochen, Germany) to examine elemental presence and distributions. Further characterizations via X-ray photoelectron spectroscopy (XPS, ThermoFisher Scientific, Al Kα radiation) and X-ray diffractometry (XRD, powder, Rigaku MiniFlex, Cu Kal radiation) were conducted to identify chemical states and compositions of the post-experimental sample surface. The Mo vapor collectors from each experiment were characterized via EDS and XPS to identify the elemental distribution and chemical state, respectively, of the collected vapor deposits, giving insights into the mass loss during experiments.

Results and Discussion

The results from thermodynamic analysis and tube furnace experiments are provided along with solid-state surface material characterization of the post-experimental samples and vapor deposit collectors.

Thermodynamic Analyses

Thermodynamic analyses based on the oxide compositions of LMS-1 and LHS-1 were conducted to predict equilibrium compositions as a function of 25≤T^eq≤2000° C. at the defined stoichiometric mass ratios. Total pressures of p^eq=0.1, 10-8, and 3×10-15 bar were investigated for the carbothermal reduction of LMS-1 and LHS-1. X_SiO(g), X_SiC, and X_Si(g) as a function of T^eq and p^eq are given in FIGS. 18A-18B for LMS-1+C (FIG. 18A) and LHS-1+C (FIG. 18B) systems for 0.1, 10-8, and 3×10⁻¹⁵ bar. Similar trends were observed in both systems in which significant amounts of SiC predicted at relatively low T^eq before rapidly decreasing as SiO(g) became more favorable with increasing T^eq. Larger amounts of SiC were predicted in the LHS-1+C system compared to the LMS-1+C system. Dissociation of SiO(g) to Si(g) was favorable with increasing T^eq, causing SiO(g) to rapidly decrease. The predicted Si-species shifted towards higher T^eq with increasing p^eq in accordance with Le Chatelier's principle. X_Si(g)→1 was predicted for p^eq=10⁻⁸ and 3×10⁻¹⁵ bar indicating that complete conversion from SiO₂ to Si(g) was favorable for both systems within the given temperature range.

X_Al2O, and X_Al(g) are given in FIGS. 19A-19B for LMS-1 (FIG. 19A) and LHS-1 (FIG. 19B) systems for pressures of 0.1, 10-8, and 3×10⁻¹⁵ bar. Similar trends were observed in both systems with significant amounts of Al₂O(g) and Al(g) predicted at similar T^eq before Al₂O(g) rapidly decreased as Al(g) became more favorable at elevated T^eq. The predicted Al-species shifted towards higher T^eq with increasing p^eq in accordance with Le Chatelier's principle. XAl(g)→1 was predicted for p^eq=10⁻⁸ and 3×10⁻¹⁵ bar, indicating that complete conversion from Al₂O₃ to Al(g) was favorable for both systems within the given temperature range.

X_Mg(g) is given in FIGS. 20A-20B for the LMS-1+C system (FIG. 20A) and X_Ca(g) for the LHS-1+C system (FIG. 20B) for pressures of 0.1, 10⁻⁸, and 3×10⁻¹⁵ bar. XMg(g)→1 was predicted for all studied total pressures, indicating that complete conversion from MgO to Mg(g) was favorable within the given temperature range. Similarly, X_Ca(g)→1 was predicted for p^eq=10⁻⁸ and 3×10⁻¹⁵ bar, indicating that complete conversion from CaO to Ca(g) was favorable within the given temperature range. The predicted equilibrium of Mg(g) and Ca(g) also shifted towards higher T^eq as p^eq increased in accordance with Le Chatelier's principle.

Tube Furnace Experiments

The results from the tube furnace experiments are shown in FIGS. 21A-21C with CO molar fraction of y_CO for JSC-1A+C (solid lines) (FIG. 21A), LMS-1+C (solid lines) (FIG. 21B), and LHS-1+C (solid lines) (FIG. 21C), and temperature of TTF (dotted line) versus time. CO evolved continuously at T_TF>500° C. for all three simulants, indicating that T_TF>500° C. is required for carbothermal reactions under these conditions. CO evolution, indicative of ongoing reactions, increased with increasing temperatures with the highest evolution occurring close to the 1500° C. isotherm for all simulants, evidence that elevated temperatures are required to completely reduce all three mixed samples. Reactions with SiO₂ likely contributed to these results, as all studied simulants had significant amounts of SiO₂ with reactions predicted at higher temperatures from thermodynamics. Significantly smaller amounts of CO evolved from the mixed samples during the 1600° C. isotherm, likely due to the short isotherm and large reaction extents for TTF<1600° C. JSC-1A+C showed larger CO evolution at lower temperatures, followed by LHS-1+C and LMS-1+C, likely because of the large Fe content in JSC-1A and large Al content in LHS-1: Thermodynamics forecasted reactions with Fe— and initial reactions with Al— as favorable at lower temperatures. At higher temperatures, LHS-1+C had the most CO produced, followed by LMS-1+C and JSC-1A+C, potentially due to differences in MgO, CaO, and Al₂O₃ compositions. Consistent results between experiments for each sample mixture provided strong evidence for repeatable results.

Mass balances for the tube furnace experiments were conducted with the masses normalized by the initial sample mass. The normalized mass of the post-experimental samples along with normalized total mass of CO and vapor evolutions from sample upon experimentation are given in FIG. 22 for JSC-1A+C (solid line), LMS-1+C (dashed line), and LHS-1+C (dotted line). The CO evolved was larger for LHS-1+C compared to JSC-1A+C and LMS-1+C, indicative of more complete reactions occurred in LHS-1+C as corroborated by yCO (FIGS. 21A-21C). The low vapor and large CO evolutions in LHS-1+C suggested that although significant reaction extents were achieved, most reduced compounds remained in the solid form after cooling down. Conversely, the increased vapor evolutions from LMS-1+C suggested that numerous compounds evolved from the sample, either upon reduction or via phase change. The large post-experimental samples for JSC-1A+C were a consequence of the comparable amounts of CO evolution to LMS-1+C and low amounts of vapor evolution compared to LHS-1+C.

Solid-State Surface Material Characterization

Post-Experimental Samples

The images of the Mo foils for JSC-1A+C (FIG. 23A), LMS-1+C (FIG. 23B), and LHS-1+C (FIG. 23C) experiments are shown in FIGS. 23A-23C with the characterized regions marked with black squares of different line styles. Non-homogeneities of post-experimental samples were likely a result of temperature gradients caused by non-uniform heating in the tube furnace.

EDS was used on the post-experimental samples to qualitatively measure elemental distributions. Several rectangular spots with areas of 0.21×0.17 cm², representing regions annotated in FIGS. 23A-23C, were scanned. The wt. % of detected elements and their standard deviations are listed in Tables 8-10 for JSC-1A, LMS-1+C, and LHS-1+C, respectively.

TABLE 8
Wt. % and standard deviations of elements detected by EDS for post-
experimental sample of JSC-1A + C tube furnace experiment
	Wt. %
	Dashed square region	Solid square region
Element	Spot 1	σ	Spot 2	σ	Spot 1	σ	Spot 2	σ
O	38.3	0.3	33.8	0.3	46.5	0.3	46.0	0.4
Mo	15.9	0.3	31.6	0.4	0.6	0.2	9.8	0.3
Si	13.2	0.1	7.9	0.1	4.2	0.1	12.9	0.1
Al	12.0	0.1	7.7	0.1	35.9	0.3	12.3	0.1
Ca	9.5	0.1	8.2	0.2	4.1	0.1	9.6	0.2
Fe	5.0	0.4	6.3	0.4	3.7	0.4	4.4	0.4
C	2.1	0.3	1.0	0.3	1.7	0.3	2.5	0.4
Ti	1.6	0.1	1.5	0.1	1.7	0.1	0.4	0.1
Mg	1.1	0.0	—	—	0.3	0.0	0.3	0.0
P	0.4	0.1	2.1	0.1	1.1	0.1	1.7	0.1

Significant amounts of Si, Al, Ca, and Fe were detected for JSC-1A+C in all spots, especially in the region marked with the dashed square, indicative of reduced compounds that consisted of these elements. Ti, Mg, and P were only identified in small amounts, likely due to their low presence in JSC-1A. The significant differences in elemental compositions between spots evidenced sample inhomogeneity. The solid square region contained larger amounts of Al and O, indicative of oxides bonded to Al (e.g., Al₂O₃). Mo was detected in the background of the scans, confirming the non-interference of the Mo foils with reactions. The O and C scans also accounted for contributions from atmospheric exposure.

TABLE 9
Wt. % and standard deviations of elements detected by EDS for
post-experimental sample of LMS-1 + C tube furnace experiment.
	Wt. %
		Dotted-
	Dashed	dashed	Solid	Dotted
	square	square	square	square
	region	region	region	region
Element	Spot 1	σ	Spot 1	σ	Spot 1	σ	Spot 1	σ
O	39.5	0.3	25.2	0.3	39.4	0.4	48.6	0.4
Mo	18.7	0.3	54.9	0.4	21.4	0.4	5.7	0.3
Si	6.2	0.1	2.6	0.1	16.1	0.2	22.3	0.2
Al	17.2	0.1	7.2	0.1	6.7	0.1	7.9	0.1
Ca	10.6	0.1	4.8	0.1	6.5	0.2	5.5	0.1
Fe	—	—	2.2	0.3	0.6	0.4	1.0	0.3
C	1.8	0.3	0.9	0.4	0.7	0.4	3.6	0.4
Ti	5.6	0.2	1.8	0.1	0.5	0.1	1.6	0.2
Mg	—	—	—	—	8.0	0.1	3.8	0.1
P	0.3	0.1	0.3	0.1	—	—	—	—

For the LMS-1+C experiment, larger amounts of Al, Ca, and Ti were detected in the dashed square region, indicative of larger concentrations in reduced form. The circular white melt formed in the solid square region consisted of significantly larger amounts of Mg than other regions, indicative of Mg concentrations, with a significant presence of Si, Al, and Ca. The dotted square region suggested the presence of SiC with larger concentrations of Si and C observed. Significant amounts of Mg, Al, and Ca were also observed in this region. The dotted-dashed square region, however, consisted of smaller elemental presences of condensed vapors and larger amounts of Mo from the Mo foil, providing further evidence of the migration of compounds to form concentration spots causing sample inhomogeneity. The O and C scans also accounted for contributions from atmospheric exposure.

For the LHS-1+C experiment, significant amounts of Al, Ca, and Si were detected in the dashed square regions, and large amounts of Al and Si with significant amounts of Ca detected in the solid square region, indicative of reduced compounds consisting of these elements with variations in concentrations. More O was observed in the solid square region, indicative of more oxides. Coupled to more Si, it is likely that significant amounts of these oxides bonded with Si. Ti was also present in the dashed square region, and Fe was present in both regions in significantly smaller quantities. Mg was concentrated at a spot in the solid square region, but it was not detected in significant quantities elsewhere. C was also observed to be higher in the same region, indicative of carbide formations, with either Mg (e.g., C₂Mg) or Si (e.g., SiC). The vast difference in elemental concentrations between spots provided strong evidence of sample segregation and inhomogeneity. Mo was detected in the background from the Mo foils, and the O and C scans accounted for contributions from atmospheric exposure.

TABLE 10
Wt. % and standard deviations of elements detected by EDS for post-
experimental sample of LHS-1 + C tube furnace experiment
	Dashed square region	Solid square region
Element	Spot 1	σ	Spot 2	σ	Spot 1	σ	Spot 2	σ
O	39.6	0.4	39.0	0.4	50.7	0.3	44.8	0.4
Mo	18.4	0.4	17.0	0.4	2.6	0.2	17.8	0.4
Si	6.7	0.1	3.7	0.1	17.4	0.1	13.2	0.2
Al	20.1	0.2	26.9	0.2	14.7	0.1	14.4	0.2
Ca	12.3	0.2	11.8	0.2	6.3	0.1	9.1	0.2
Fe	0.7	0.4	—	—	0.7	0.2	—	—
C	1.0	0.4	—	—	4.9	0.3	0.6	0.4
Ti	0.8	0.2	1.0	0.2	0.4	0.1	0.2	0.1
Mg	—	—	—	—	2.1	0.1	—	—

Further characterization was conducted via XPS and an XPS analysis software (Avantage, Thermo Scientific Avantage data system, ThermoFisher Scientific) on the post-experimental samples to identify oxidation states and possible chemical compositions of the elements from EDS, and the results are listed in Table 11 for JSC-1A, in Table 12 for LMS-1+C, and in Table 13 for LHS-1+C. Oxides (e.g., Al₂O₃, SiO₂) were prominently present in all post-experimental samples, indicating that samples were not completely reduced. Some elements (e.g., Al, Si, Ca) were observed, suggesting some elemental yields. Silicate and SiC were also commonly found in several regions of all post-experimental samples, evidence of favorable intermediaries predicted by thermodynamics. CaCO₃ was detected in many regions, presumably due to Ca oxides reacting with atmospheric CO₂ to form carbonates. Mo detected in the background showed evidence of significant oxidation of the Mo foils. The C 1s and O 1s also accounted for exposure to atmospheric conditions. Surface effects and reactions were easily captured by XPS, as it is significantly more surface sensitive than EDS due to its much smaller depth of analysis. This also accounts for the smaller presence of Mo detected by XPS compared to EDS.

TABLE 11
Peak binding energy locations of detected elements by XPS and possible
chemical compositions of post-experimental samples for tube furnace experiments of
JSC-1A + C.
			Peak Binding Energy,	Possible Chemical
	Region	Element	eV	Compositions
JSC-1A + C	Solid Square	O 1s	530.46, 531.18, 532.13	Metal oxides, Al2O3,
				metal carbonates
		A1 2p	73.39, 74.04	Elemental Al, Al2O3
		C 1s	284.37, 285.08, 286.23,	Adventitious carbon
			288.58	contamination, carbonate
		Si 2p	98.23, 101.87	Elemental Si, SiO4
		Ca 2p	346.77, 347.73	CaO, CaCO3
		Fe 2p	711.00	Fe2O3
		Ti 2p	458.57	TiO2
	Dashed Square	C 1s	284.44, 284.89	Carbide, adventitious
				carbon contamination
		O 1s	529.75, 530.81, 531.84	Metal oxides, Al2O3,
				SiO2
		Fe 2p	710.81	Fe2O3
		Ca 2p	346.97	CaO
		Si 2p	101.58, 101.88, 103.04	SiC, SiO4, SiO2
		Mo 3d	227.95, 229.60, 232.25	Mo, MoO2, MoO3
		Al 2p	73.90	Al2O3

TABLE 12
Peak binding energy locations of detected elements by XPS and possible
chemical compositions of post-experimental samples for tube furnace experiments of
LMS-1 + C.
			Peak Binding Energy,	Possible Chemical
	Region	Element	eV	Compositions
LMS-1 + C	Dotted Square	C 1s	284.35, 284.77, 288.28	Carbide, adventitious
				carbon contamination,
				carbonate
		O 1s	529.74, 531.14, 532.23	Metal oxides, SiO2
		Mo 3d	232.01	MoO3
		Ca 2p	346.46	Elemental Ca, CaO
		Si 2p	99.60, 101.09,102.00,	Elemental Si, SiC, SiO4,
			102.83	SiO2
	Solid Square	O 1s	530.33, 531.10, 531.91,	Metal oxides, Al2O3, metal
			532.75	carbonates, SiO2
		Mg 1s	1304.14	MgO
		Si 2p	101.66, 102.21, 102.93	SiC, SiO4, SiO2
		C 1s	284.48, 286.00, 288.75	Carbide, adventitious
				carbon contamination,
				carbonate
		Ca 2p	347.08, 347.99	CaO, CaCO3
		Al 2p	74.10	Al2O3
	Dotted-Dashed	O 1s	530.11, 531.11, 532.00,	Metal oxides, Al2O3, metal
			532.76	carbonates, SiO2
	Square	C 1s	284.64, 285.78, 289.24	Carbide, adventitious
				carbon contamination,
				carbonate
		Mo 3d	229.78, 232.30	MoO2, MoO3
		Al 2p	74.43	Al2O3
		Ca 2p	347.42, 348.72	CaO, CaCO3
		Si 2p	101.94, 102.80	SiC, SiO2
		Ti 2p	458.90	TiO2
	Dashed Square	O 1s	529.54, 530.98, 532.10	Metal oxides, Al2O3, metal
				carbonates
		Al 2p	74.26	Al2O3
		Si 2p	101.78, 102.53	SiC, SiO4
		Ca 2p	346.85, 348.44	CaO, CaCO3
		Ti 2p	458.65	TiO2
		C 1s	284.57, 285.98, 288.95	Carbide, adventitious
				carbon contamination,
				carbonate
		Mo 3d	230.21, 232.36	MoO2, MoO3

TABLE 13
Peak binding energy locations of detected elements by XPS and possible
chemical compositions of post-experimental samples
for tube furnace experiments of LHS-1 + C.
			Peak Binding	Possible Chemical
	Region	Element	Energy, eV	Compositions
LHS-1 +	Dashed	O 1s	530.52, 531.56,	Metal oxides, Al2O3, metal
C	Square		532.89	carbonates
		Al 2p	73.98, 74.48	Elemental Al, Al2O3
		Ca 2p	347.46, 348.18	CaO, CaCO3
		C 1s	284.49, 285.78,	Carbide, adventitious
			288.75	carbon contamination,
				carbonate
		Si 2p	101.69, 102.40	SiC, SiO4
		Mo 3d	227.92, 229.10,	Mo, MoO2, MoO3
			232.38
	Solid	O 1s	530.38, 531.57,	Metal oxides, Al2O3, metal
	Square		532.40	carbonates
		Al 2p	73.74, 74.55	Elemental Al, Al2O3
		Mo 3d	229.65, 232.38	MoO2, MoO3
		Si 2p	102.18	SiO4
		Ca 2p	347.11, 348.43	CaO, CaCO3
		C 1s	284.65, 285.83,	Carbide, adventitious
			289.28	carbon contamination,
				carbonate
		Mg 1s	1304.19	MgO
		Ti 2p	459.22	TiO2

XRD was performed on the post-experimental samples to identify dominant chemical compositions. The obtained intensity patterns were analyzed using an XRD analysis software (SmartLab Studio II, Rigaku) via phase identification that matched candidate compounds to results based on a large reference intensity pattern database (PDF-4+database, ICDD) coupled with information from EDS. The normalized intensity patterns for the dashed square region of post JSC-1A+C experiment sample are shown in FIG. 24. Fe_0.4Al_0.6 (diamond markers, 040028918), Al_0.5Si_0.75O_2.25 (triangle markers, 000371460), and Mo (circle markers, 040147439) were identified, and the reference peaks are shown in FIG. 24. Identification of Al_0.5Si_0.75O_2.25 confirmed that the O scans from EDS included presence of oxides, and they were bonded to Si and Al. Mo was a contribution from the Mo foil and Fe_0.4Al_0.6 suggested elemental binding upon reduction.

For the LMS-1+C experiment, the normalized intensity patterns for the dashed and dotted-dashed square regions are shown in FIGS. 25A-25D. Mo (diamond markers, 000040809), Si (circle markers, 000271402), Ca₃Al₂P₂Si₂O₁₅ (triangle markers, 000360108), Ca₂Fe₉O₁₃ (square markers, 000351277), and Mo₃Si (hexagon markers, 000510764) were identified, and the reference peaks are shown in FIGS. 25A-25D. The presence of Mo₃Si indicated that some binding of Si to the Mo foil likely accounted for the sticking of sample to the foil upon experimentation. Si was also identified, indicative of possible elemental yields. The identification of Ca₃Al₂P₂Si₂O₁₅ and Ca₂Fe₉O₁₃ indicated a strong oxide presence in sample with the possibility of further reduction. The normalized intensity patterns for dotted square region of post LMS-1+C experiment sample are shown in FIG. 26. Fe_0.5Al_0.5 (diamond markers, 040172997) was identified, and the reference peaks are shown in FIG. 26. The results provided evidence of elemental binding upon reduction in this region.

The normalized intensity patterns for the dashed square region for LHS-1+C experiment are shown in FIGS. 27A-27D. MoSi₂ (diamond markers, 010726181), CaAl₂Si₂O₈ (circle markers, 040132354), SiC (triangle markers, 040072139), and AlMoTi₂ (square markers, 000120074) were identified, with reference peaks shown in FIGS. 27A-27D. The presence of MoSi₂ and AlMoTi₂ indicated some binding of Si, Al, and Ti to the Mo foil likely accounting for the sticking of sample to the foil during experimentation. The identification of CaAl₂Si₂O₈ provided evidence of the oxide presence in sample with the possibility of further reduction. The normalized intensity patterns for solid square region of post LHS-1+C experiment sample are shown in FIGS. 28A-28B. Fe (diamond markers, 040188483), Mo (circle markers, 000040809), and SiO₂ (triangle markers, 040028513) were identified with reference peaks shown in FIGS. 28A-28B. Results yielded SiC, Fe, and SiO₂, which are likely useful for ISRU applications. Mo was also detected from the Mo foil.

Significant amounts of other crystal peaks in all post-experimental samples were unidentifiable because of the low matching probability with the intensity patterns database, likely due to low compositions. However, they do show that the reduced samples are polycrystalline. Other regions of the post-experimental samples were not analyzed by XRD due to small areas negating useful results.

Vapor Deposit Collectors

EDS was used to examine the Mo foils from each experiment for elemental analyses. Two different spots for each of the Mo foils were scanned with rectangular areas of 0.21×0.17 cm². The wt. % and standard deviations of the detected elements are listed in Table 14 for each experiment. Significant amounts of Si were detected on the collectors for all experiments, and small quantities of Al, Na, and Mg were detected in some. More Si was detected for LMS-1+C followed by JSC-1A+C and LHS-1+C with similar trends observed for the O scans, suggesting that O was bonded to Si in the form of SiO₂, likely due to disproportion of SiO(g) to Si and SiO₂ (Roine, A., HSC Chemistry 7.11. 2009, Metso Outotec) with decreasing temperatures at the exhaust end of the Al₂O₃ tube where Mo collectors were located. The opposite trends to Si scans were observed for the Mo scans with contributions from the Mo collectors in the background. The O scans also accounted for contributions from atmospheric exposure. Similarly, the C scans accounted for contributions from adventitious carbon contamination due to atmospheric exposure.

TABLE 14
The wt. % and standard deviations of elements detected by EDS on Mo vapor
deposit collectors for JSC-1A + C, LMS-1 + C, and LHS-1 + C tube furnace
experiments.
	Wt. %
	JSC-1A + C	LMS-1 + C	LHS-1 + C
	Spot		Spot		Spot		Spot		Spot		Spot
	1	σ	2	σ	1	σ	2	Σ	1	σ	2	σ
Mo	47.2	0.3	44.5	0.4	35.3	0.4	38.2	0.4	51.8	0.3	51.6	0.4
O	35.1	0.3	36.0	0.3	39.0	0.4	38.2	0.3	31.9	0.2	31.5	0.3
Si	16.6	0.1	18.7	0.2	24.9	0.2	22.1	0.2	13.8	0.1	13.7	0.1
C	0.7	0.3	0.9	0.4	0.7	0.4	1.3	0.4	1.8	0.2	2.5	0.3
Al					0.2	0.1	0.1	0.0	0.1	0.0	0.1	0.0
Na							0.1	0.0	0.1	0.0	0.2	0.0
Mg									0.3	0.0	0.3	0.0

Further analysis was conducted on the vapor deposit collectors via XPS and an XPS analysis software (Avantage, Thermo Scientific Avantage data system, ThermoFisher Scientific) to identify potential chemical compositions or oxidation states present on the collectors. Two different spots for each Mo collector were scanned with circular spots of 400 μm in diameter. The binding energies of the peaks for each detected element, along with their associated candidates of chemical compositions, are listed in Table 15 for all three experiments.

TABLE 15
Peak binding energy locations of detected elements by XPS and possible
chemical compositions present on Mo vapor deposit collectors for tube
furnace experiments of JSC-1A + C, LMS-1 + C, and LHS-1 + C.
			Peak Binding	Possible Chemical
	Element	Spot	Energy, e V	Compositions
JSC-1A +	O 1s	1	530.94, 532.66, 533.36	Metal oxides, metal
C		2	531.15, 532.42, 533.11	carbonates,
				adventitious
				contamination
	Si 2p	1	102.60, 103.48	SiC, SiO4, SiO2
		2	102.71, 103.50
	C 1s	1	284.06, 284.64,	Carbide, adventitious
			285.71, 288.68	contamination, metal
		2	284.28, 284.71,	carbonate
			286.34, 288.38
	Mo 3d	1	228.10, 229.01, 232.59	Mo, MoO2, MoO3
		2	228.07, 229.00, 232.60
	Ca 2p	1	347.05, 347.34	CaO, CaCO3
		2	347.02, 347.37
LMS-1 +	O 1s	1	532.35, 532.93	SiO2, adventitious
C		2	532.61, 533.07	contamination
	Si 2p	1	103.35
		2	103.41	SiO2
	C 1s	1	284.69, 286.05	Adventitious
		2	284.61, 285.56	contamination
	Mo 3d	1	228.00, 228.42, 232.70	Mo, MoO2, MoO3
		2	228.00, 228.36, 232.91
LHS-1 +	O 1s	1	530.93, 532.01, 532.75	Metal oxides, metal
C				carbonates,
		2	530.35, 531.34, 532.19	adventitious
				contamination
	Si 2p	1	102.55, 103.34	SiC, SiO4, SiO2
		2	102.02, 102.76
	C 1s	1	284.27, 284.86,	Carbide, adventitious
			286.39, 288.64	contamination, metal
		2	284.08, 284.56,	carbonate
			285.87, 288.29
	Na 1s	1	1071.94	Sodium oxide
		2	1071.85
	Ca 2p	1	347.07, 347.37	CaO, CaCO3
		2	346.61, 347.45
	Mg 1s	1	1303.64
		2	1303.49	MgO
	K 2p	1	293.66
		2	293.31	Elemental K
	Al 2p	1	74.75	Aluminosilicate,
		2	73.90	Al2O3

The O 1s peaks evidenced the presence of metal/metalloid oxides (e.g., SiO₂, SiO₄, CaO, MgO, Al₂O₃) supported by Si 2p, Ca 2p, Mg 1s, and Al 2p peaks. While oxides of Si were present in all collectors, CaO was only detected for JSC-1A+C and LHS-1+C, and MgO and Al₂O₃ were only detected for LHS-1+C. For JSC-1A+C and LHS-1+C, the O 1s and C 1s peaks suggested the presence of metal carbonates, specifically CaCO₃ as suggested by the Ca 2p peaks. The presence of SiC was strongly supported by the Si 2p and C 1s peaks. Evidence of some elemental K was observed in the LHS-1+C by the K 2p peaks. The O 1s and C 1s peaks accounted for atmospheric exposure effects. The Mo 3d peaks of all collectors revealed some oxidation of the Mo foils due to elevated temperatures. These findings generally corroborated EDS results. Ca compounds, significantly detected by XPS, were not detected by EDS, likely due to the quantities being below detection limits for EDS.

The finding from this work greatly expand upon previous thermodynamic analyses and experiments in inert conditions (Shaw et al., “Thermodynamic modelling of ultra-high vacuum thermal decomposition for lunar resource processing,” Planetary and Space Science, 204:105272 (2021); Šeško et al., “Oxygen production by solar vapor-phase pyrolysis of lunar regolith simulant,” Acta Astronautica, 224:215-225 (2024); Lu and Reddy, “Extraction of Metals and Oxygen from Lunar Soil,” High Temperature Materials and Processes, 27 (4): 223-234 (2008); Kaur et al., “Thermodynamic and experimental study on carbothermal reduction of JSC-1A lunar regolith simulant for metal and metalloid production,” Advances in Space Research, 73 (8): 4024-4039 (2024); Schlüter and Cowley, “Review of techniques for in-situ oxygen extraction on the moon,” Planetary Space Sci., 181:104753 (2020); Troisi et al., “Oxygen extraction from lunar dry regolith: Thermodynamic numerical characterization of the carbothermal reduction,” Acta Astronautica, 199:113-124 (2022)) by varying lunar regolith simulant compositions with the focus on metal or metalloid production. Thermodynamic analyses showed the potential advantage of the low lunar surface pressure and feasibility of ISRU, specifically for carbothermal reduction processes of lunar regolith. They also revealed that compound separation is possible by varying operating temperatures, evidenced by the solid-state surface material characterization of inhomogeneous post-experimental samples caused by uneven heating. Material characterization work also revealed that the compounds formed and evolved from the sample upon experimentation are dependent on simulant compositions. Experimental measurements highlighted the requirement for elevated temperatures to carbothermally reduce lunar regolith simulants, supported by thermodynamics. Mass balance analyses and comparison of gas product amounts suggested that two types of chemical reactions took place during experimentation: phase change reactions and carbothermal reduction.

The potential of thermodynamically and experimentally producing different useful products as a function of oxide and mineral compositions, temperature, and pressure were discussed and presented in this work. These results are valuable for informing and guiding further experimental work on carbothermal reduction of lunar regolith or simulants under varying pressures (ultra-high vacuum) or utilizing other heat inputs (concentrated solar irradiation). Further studies on optimizing operating conditions to extract targeted compounds from specific regolith types are needed for realizing ISRU objectives.

CONCLUSIONS

The carbothermal reductions of lunar mare simulant (LMS-1), lunar highland simulant (LHS-1) and Johnson Space Center (JSC-1A) simulant were investigated in this work by conducting thorough thermodynamic and experimental analyses. Chemical equilibrium compositions from the thermodynamic analyses of targeted compounds and favorable intermediaries were predicted as a function of temperature and pressure for LMS-1 and LHS-1, guiding operating conditions for experiments. Complete conversions of SiO₂ to Si(g), Al₂O₃ to Al(g), MgO to Mg(g), and CaO to Ca(g) were predicted at equilibrium pressures of ≤10⁻⁸ bar and equilibrium temperatures of <1500° C. via Le Chatelier's principle.

The carbothermal reduction of JSC-1A, LMS-1, and LHS-1 was then investigated in a tube furnace with activated carbon mixed with the different types of lunar simulants before experimentation. Results showed that temperatures of >500° C. were required for carbothermal reduction and elevated temperatures were required for complete reduction. Mass balance analyses were conducted to measure the evolved CO, the byproduct of carbothermal reactions, and evolved vapor from samples. LHS-1 was observed to have greater carbothermal reduction extents, and LMS-1 had the most phase change. JSC-1A had similar carbothermal reduction extents as LMS-1, which was expected as JSC-1A is also a type of mare simulant. Inhomogeneities in post-experimental samples evidenced capability of compound separation and production.

Solid-state surface material characterization tools were extensively used to characterize the post-experimental samples. Significant amounts of oxides were still present in post-experimental samples. However, elemental Si, Fe, and Al and favorable intermediaries of SiC and SiO₄ were identified, relevant for ISRU. Material characterization work was also conducted on the collected vapor deposits at the tube furnace exhaust. Results provided insights on vapor compounds for different lunar regolith simulants. Significant Si compounds were identified on the collectors of all systems with small presence of Ca and Al compounds in some.

Nomenclature
ΔG       change in Gibbs free energy, J
I        XRD intensity, counts
Imax     maximum XRD intensity, counts
m        mass, mg
mi       initial mass, mg
nAl2O3,i initial amount of Al2O3, moles
nCaO,i   initial amount of CaO, moles
nMgO,i   initial amount of MgO, moles
nSiO2,i  initial amount of SiO2, moles
n  A ⁢  l ⁡ ( g )    e ⁢ q   ⁢   equilibrium ⁢   amount ⁢   of ⁢    Al ( g )   , moles
n  A ⁢  l 2  ⁢  O ⁡ ( g )    e ⁢ q   ⁢   equilibrium ⁢   amount ⁢   of ⁢    Al 2  ⁢  O ( g )   , moles
n  C ⁢  a ⁡ ( g )    e ⁢ q   ⁢   equilibrium ⁢   amount ⁢   of ⁢    Ca ( g )   , moles
n  M ⁢  g ⁡ ( g )    e ⁢ q   ⁢   equilibrium ⁢   amount ⁢   of ⁢    Mg ( g )   , moles
n  S ⁢  i ⁡ ( g )    e ⁢ q   ⁢   equilibrium ⁢   amount ⁢   of ⁢    Si ( g )   , moles
n  S ⁢ i ⁢ C   e ⁢ q   ⁢   equilibrium ⁢   amount ⁢   of ⁢   SiC  , moles
n  S ⁢ i ⁢  O ⁡ ( g )    e ⁢ q   ⁢   equilibrium ⁢   amount ⁢   of ⁢    SiO ( g )   , moles
peq      experiment time, bar
t        experiment time, h
Teq      equilibrium temperature, ° C.
TTF      tube furnace temperature, ° C.
X        equilibrium molar fraction, -
XAl(g)   equilibrium molar fraction of Al(g), -
XAl2O(g) equilibrium molar fraction of Al2O(g), -
XCa(g)   equilibrium molar fraction of Ca(g), -
XMg(g)   equilibrium molar fraction of Mg(g), -
XSi(g)   equilibrium molar fraction of Si(g), -
XSiC     equilibrium molar fraction of SiC, -
XSiO(g)  equilibrium molar fraction of SiO(g), -
yCO      molar fraction of CO, -
□        x-ray diffraction angle, °
σ        standard deviation, weight %

Claims

1. A process for extracting metals and/or metalloids from extraterrestrial regolith comprising:

providing a sample of extraterrestrial regolith, wherein the sample comprises an oxide of one or more metals and/or metalloids;

chemically reacting the oxide of the one or more metals and/or metalloids with carbon via carbothermal reduction to produce one or more metals and/or metalloids.

2. The process according to claim 1, wherein chemically reacting the oxide comprises:

mixing the sample with a carbon-containing reducing agent to form a mixture; and

directing concentrated solar irradiation to the mixture as process heat.

3. The process according to claim 1, wherein chemically reacting the oxide is carried out at a pressure between 3×10−15 bar and 1×10−12 bar.

4. The process according to claim 1, wherein chemically reacting the oxide is carried out at a temperature between 750° C. and 2000° C.

5. The process according to claim 1, wherein the extraterrestrial regolith is lunar regolith.

6. The process according to claim 2, wherein the carbon-containing reducing agent is derived from a fixed carbon source.

7. The process according to claim 2, wherein the carbon-containing reducing agent is derived from a carbon source selected from the group consisting of human feces and subsurface carbon-bearing polar ices.

8. The process according to claim 2, wherein the carbon-containing reducing agent is selected from the group consisting of activated carbon, CH4, CO, C2H4, CO2, and CH3OH.

9. The process according to claim 14, wherein the concentrated solar irradiation is provided by a solar concentrating thermal technology.

10. The process according to claim 9, wherein the solar concentrating thermal technology provides concentrated sunlight at solar concentration ratios of up to 5000 suns.

11. (canceled)

12. The process according to claim 1, wherein the metals and/or metalloids are selected from the group consisting of aluminum, silicon, iron, calcium, magnesium, sodium, and titanium.

13. A process for extracting metals and/or metalloids from lunar regolith comprising:

chemically reacting an oxide of one or more metals and/or metalloids contained in a sample of the lunar regolith with a solid carbon-containing reducing agent derived from a fixed carbon source via carbothermal reduction performed in a vacuum environment having a pressure between 3×10−15 bar and 1×10−12 bar and at a temperature between 1250° C. and 2000° C. provided by concentrated solar irradiation to produce one or more metals and/or metalloids.

14. The process according to claim 13, wherein chemically reacting the oxide comprises:

contacting the sample with the solid carbon-containing reducing agent to form a mixture; and

exposing the mixture to the concentrated solar irradiation.

15. The process according to claim 14, wherein the concentrated solar irradiation provides concentrated sunlight at solar concentration ratios of at least 1000 suns.

16.-17. (canceled)

18. The process according to claim 14, wherein the solid carbon-containing reducing agent is derived from a carbon source selected from the group consisting of human feces and subsurface carbon-bearing polar ices.

19. (canceled)

20. The process according to claim 13, wherein the one or more metals and/or metalloids are selected from the group consisting of aluminum, silicon, calcium, magnesium, sodium, and titanium.

21. A process for extracting metals and/or metalloids from lunar regolith comprising:

chemically reacting an oxide of one or more metals and/or metalloids contained in a sample of the lunar regolith with a solid carbon-containing reducing agent derived from a fixed carbon source via carbothermal reduction performed in a vacuum environment having a pressure between 3×10−15 bar and 1×10−12 bar and at a temperature between 1250° C. and 2000° C. provided by concentrated solar irradiation to produce one or more metals and/or metalloids;

wherein: chemically reacting the oxide comprises: contacting the sample with the solid carbon-containing reducing agent to form a mixture; and exposing the mixture to the concentrated solar irradiation; the concentrated solar irradiation provides concentrated sunlight at solar concentration ratios of at least 5000 suns; the solid carbon-containing reducing agent is derived from a carbon source selected from the group consisting of human feces and subsurface carbon-bearing polar ices; and the one or more metals and/or metalloids are selected from the group consisting of aluminum, silicon, calcium, magnesium, sodium, and titanium.