DEVICE FOR THE ENERGY-EFFICIENT, LOW TEMPERATURE IN-SITU RESOURCE UTILIZATION(ISRU) PRODUCTION OF METHANE, HYDROCARBONS AND OXYGEN ON MARS

Description

RELATED APPLICATIONS

This application claims priority to U.S. Provisional application 63/616,488 filed Dec. 29, 2023 which is incorporated herein by reference in its entirety.

STATEMENT REGARDING FEDERALLY FUNDED RESEARCH

This invention was made with government support under 80NSSC22K1766 and HR0011-25-3-0310 awarded by the NASA and DARPA respectively. The government has certain rights in the invention.

FIELD

The present disclosure relates generally to the field of chemical engineering. More particularly, it concerns devices and methods/processes for methane and ethanol production.

BACKGROUND

The current goal of the United States National Aeronautics and Space Administration (NASA)'s is to land humans on Mars in the 2030's. The exploration of Mars is also a priority for several national and private space entities with lander missions from the US and China, and orbiters from the US, China, Europe, Russia, India, and the United Arab Emirates (UAE) presently active there (1). The duration of any mission to Mars is circumscribed by the constraints imposed by the mass that can be sent to Mars from Earth. For example, SpaceX's Falcon Heavy (weighing 1420 tons fully loaded) is designed to deliver a 16.8-ton payload to Mars (1.8% of total weight). The immense energetic cost of moving material out of Earth's gravity well is illustrated by the following example—completely provisioning a Mars mission from Earth, including 35 metric tons of propellant needed for the return journey, is estimated to require ca 400 metric tons of propellant (fuel and oxidant) on 4-5 heavy lift launch vehicles (20). Thus, any economical long-term (weeks to months) mission necessitates the exploitation of resources present on Mars for life-support and energy (2).

Key inputs for in-situ resource utilization (ISRU) have been identified in the Martian atmosphere and surface. The Martian atmosphere significantly differs from that of Earth's with its predominant constituent being CO₂ with the atmospheric pressure on Mars (6.36 mbar) being significantly lower than that of Earth (1013 mbar) (detailed in Table 1). Critically, the relatively low diurnal temperature range on Mars also suggests a heating energy penalty for any ISRU process carried out there.

In addition to abundant CO₂ in the Martian atmosphere, the comprehensive elucidation of Martian regolithic geochemistry by a succession of robotic landers and orbiter missions has led to the identification of deliquescent perchlorate salts (9,10,12,13). This includes the notable discovery of significant quantities of perchlorate and sulfate salts of sodium and magnesium by NASA's Phoenix lander (through its wet chemistry instrument (WCI)) and optically observed sublimation of water ice discovered under a few inches of regolith (11). Multiple pathways allow for the existence of water on (or under) the surface of present-day Mars through premelting (4), by adsorption of atmospheric vapor at grain-ice boundaries(5) and by greenhouse melting(6)). Perchlorate salts play an important role in any extant hydrological cycle on Mars by depressing the freezing point of water ca −70° C. (9,18) and allowing for the existence of liquid water—a key ISRU feedstock (16,17). Contemporary liquid water flows are also indicated by the geological observations of the Mars Reconnaissance Orbiter (MRO). NASA's Spirit and Opportunity rovers found historic evidence of super-cold local acidic brines through observation of acidic aqueous activity, evaporation, and desiccation of the Martian regolith (8). Furthermore, 2×10¹⁸ kg of water on Mars is present in the form of polar ice caps as evidenced by the Mars Odyssey Gamma Ray Spectrometer (GRS) and Mars Express spacecrafts. The Mars Advanced Radar for Subsurface and Ionosphere Sounding (MARSIS) instrument on-board the Mars Express spacecraft has also detected multiple sub-glacial water bodies underneath the Martian south pole at Ultimi Scopuli. NASA's “Mars Oxygen ISRU Experiment” (MOXIE) on the Perseverance rover is the most advanced ISRU demonstration to date. This high temperature (1073-1273K) electrolyzer testbed produces O₂ and CO from atmospheric CO₂ through the reaction shown below:

2 ⁢ CO 2 ( g ) = 2 ⁢ CO ( g ) + O 2 ( g ) ( 1 )

Presently, the MOXIE's proven power normalized production of O₂ is 0.8 grams W⁻¹ day⁻¹ (at 1037K) (21). This system could potentially be sent to the surface 26 months before a manned mission in order to produce the 30 metric tons of oxygen needed to support a human mission to Mars. Proposals exist for a larger 1 metric ton MOXIE-like electrolyzer to produce 25-30 metric tons of oxygen for life support and oxidant (19). Despite MOXIE's successful production of O₂, the need for downstream processing to remove CO and the inherent risks of CO toxicity implies increased balance of plant (BOP) requirements and reduced overall energy efficiency. Furthermore, high operational temperatures introduce issues of thermal waste and safety.

A critical design issue is the nature of engines powering the descent stage landing on Mars. Methalox engines, utilizing cryogenic methane as fuel and cryogenic oxygen as oxidant, are the preferred power source (a listing representative of global efforts towards developing methalox engines is presented in Table 2). An issue for future missions is that MOXIE only produces O₂ used in methalox engines, leaving the need for the CH₄ component unfulfilled.

Thus, there remains a need for additional processes, systems, and devices for ISRU.

SUMMARY

In addressing problems associated with low temperature chemical processes, the present disclosure describes an integrated ultra-low temperature electrolyzer that produces both fuel and oxidant utilizing Martian atmospheric CO₂ in conjunction with liquid brines present on Mars. Some embodiments of the disclosed system(s) produce an oxidant with the added benefit of producing methane (CH₄) all at a low energy consumption. In some embodiments, the disclosed electrolyzer operates at ultra-low (230K and 500 k) temperatures. In other embodiments, the disclosed electrolyzer may use Earth atmospheric concentration of CO₂ at Earth pressure along with a liquid brine.

In some aspects, the invention is directed towards an electrochemical device and associated up-stream and downstream processes for the harvesting and subsequent conversion of gaseous carbon dioxide to value added hydrocarbons such as CO, HCHO, CH₃OH, CH₄, C₂H₄, and C₂H₅OH. In certain configurations the electrochemical device can also produce pure oxygen as a valuable by-product. This device is targeted for use in both the Martian atmosphere and with sources of CO₂ on Earth. Furthermore, the use of highly concentrated brine electrolytes (or seawater or brackish water) makes this device a much more attractive proposition compared to existing systems that need a highly purified water stream to function.

Certain embodiments are directed to a low temperature electrochemical process comprising: introducing feedstock into an electrolysis chamber having cation exchange membrane (CEM) positioned between an anode and a cathode forming an anode compartment and a cathode compartment, the cathode comprising a copper electrocatalyst, wherein a brine feedstock is introduced on the anode side of the CEM and the CO₂ containing feedstock is introduced on the cathode side of the CEM; and applying electrochemical potential across the CEM generating oxygen and hydrocarbon products, the process being performed at a temperature of 90 degrees to −60 degrees Celsius. The low temperature electrochemical process Can be performed at a temperature between 15 degrees to −15 degrees Celsius. The brine feedstock can include perchlorate salts, chloride salts, carbonate salts, bicarbonate salts, sulfate salts, bisulfate salts or combinations thereof. The copper catalyst can be a sulfonated tetrafluoroethylene fluoropolymer-copolymer-coated copper foam or a copper oxide deposited on copper foam with the copper oxides consisting of copper (I) oxide and copper (II) oxide in ratios of 10:90 to 0:100. In certain aspects the low temperature electrochemical process can be performed at a pH of 5 to 9. the low temperature electrochemical process can be performed under atmospheric pressure. The low temperature electrochemical process can further include separately collecting the hydrocarbon product and O₂ product.

Certain embodiments are directed to an electrochemical cell including a cathode composed of a material effective for the reduction of CO₂ to hydrocarbons, an anode composed of a material effective for the oxidation of H₂O to O₂, an electrolyte comprising an aqueous solution chloride salts, carbonate salts, bicarbonate salts, sulfate salts, bisulfate salts or combinations thereof, an ion exchange membrane separating the cathode and the anode; an ion-conductive electrolyte capable of facilitating the transport of ions between the cathode and anode; and a power supply connected to the cathode and the anode for driving the electrolysis process, the electrochemical cell configured to operate at temperatures below 0 degrees Celsius. The cathode can include a catalyst material selected from the group consisting of copper, nickel, zinc, tin, and their oxides and/or alloys. The anode can include a metal oxide such as iridium oxide or another suitable oxide catalyst.

Other embodiments of the invention are discussed throughout this application. Any embodiment discussed with respect to one aspect of the invention applies to other aspects of the invention as well and vice versa. Each embodiment described herein is understood to be embodiments of the invention that are applicable to all aspects of the invention. It is contemplated that any embodiment discussed herein can be implemented with respect to any method or composition of the invention, and vice versa. Furthermore, compositions, devices, systems and kits of the invention can be used to achieve methods/processes of the invention.

The use of the word “a” or “an” when used in conjunction with the term “comprising” in the claims and/or the specification may mean “one,” but it is also consistent with the meaning of “one or more,” “at least one,” and “one or more than one.”

Throughout this application, the term “about” is used to indicate that a value includes the standard deviation of error for the device or method being employed to determine the value.

The use of the term “or” in the claims is used to mean “and/or” unless explicitly indicated to refer to alternatives only or the alternatives are mutually exclusive, although the disclosure supports a definition that refers to only alternatives and “and/or.”

As used in this specification and claim(s), the words “comprising” (and any form of comprising, such as “comprise” and “comprises”), “having” (and any form of having, such as “have” and “has”), “including” (and any form of including, such as “includes” and “include”) or “containing” (and any form of containing, such as “contains” and “contain”) are inclusive or open-ended and do not exclude additional, unrecited elements or method steps.

As used herein, the terms “comprises,” “comprising,” “includes,” “including,” “has,” “having,” “contains”, “containing,” “characterized by” or any other variation thereof, are intended to encompass a non-exclusive inclusion, subject to any limitation explicitly indicated otherwise, of the recited components. For example, a chemical composition and/or method that “comprises” a list of elements (e.g., components or features or steps) is not necessarily limited to only those elements (or components or features or steps) but may include other elements (or components or features or steps) not expressly listed or inherent to the chemical composition and/or method.

As used herein, the transitional phrases “consists of” and “consisting of” exclude any element, step, or component not specified. For example, “consists of” or “consisting of” used in a claim would limit the claim to the components, materials or steps specifically recited in the claim except for impurities ordinarily associated therewith (i.e., impurities within a given component). When the phrase “consists of” or “consisting of” appears in a clause of the body of a claim, rather than immediately following the preamble, the phrase “consists of” or “consisting of” limits only the elements (or components or steps) set forth in that clause; other elements (or components) are not excluded from the claim as a whole.

As used herein, the transitional phrases “consists essentially of” and “consisting essentially of” are used to define a chemical composition and/or method that includes materials, steps, features, components, or elements, in addition to those literally disclosed, provided that these additional materials, steps, features, components, or elements do not materially affect the basic and novel characteristic(s) of the claimed invention. The term “consisting essentially of” occupies a middle ground between “comprising” and “consisting of”.

Other objects, features and advantages of the present invention will become apparent from the following detailed description. It should be understood, however, that the detailed description and the specific examples, while indicating specific embodiments of the invention, are given by way of illustration only, since various changes and modifications within the spirit and scope of the invention will become apparent to those skilled in the art from this detailed description.

DESCRIPTION OF THE DRAWINGS

The following drawings form part of the present specification and are included to further demonstrate certain aspects of the present invention. The invention may be better understood by reference to one or more of these drawings in combination with the detailed description of the specification embodiments presented herein.

FIG. 1. Schematic representation of a CO2-brine electrolyzer for operation on Mars.

FIG. 2A-2B. Thermodynamic properties for (a) brine electrolysis and (b) CO2—brine electrolysis between 230K and 500 k.

FIG. 3A-3B. Nernst potentials at 636 Pa and 101325 Pa for (a) brine electrolysis and (b) CO2-brine electrolysis.

FIG. 4A-4D. Nernst voltage for carrying out electrolysis at (a) 236K-636 Pa, (b) 298K-636 Pa, (c) 236K-101325 Pa and (d) 298K-636 Pa.

FIG. 5. Performance of the brine Electrolyzer operating with Pt—C cathode and Pb₂Ru₂O₇ anode at 237K and 101325 Pa.

FIG. 6. Performance of the brine Electrolyzer operating with Pt—C cathode and RuO₂ anode at 237K and 101325 Pa.

FIG. 7. Performance of the CO₂-brine electrolyzer operating with Cu(111) cathode and RuO₂ anode at 255K and 101325 Pa.

FIG. 8. Volume of CH₄ and O₂ produced and power requirements for a stack of CO₂-brine electrolyzer operating with 10 cells of 100 cm² each at 255K and 101325 Pa.

FIG. 9. Schematic representation of a zero-gap electrolyzer.

FIG. 10A-10B. (A and B) C1 product reaction pathways.

FIG. 11A-11B. (A and B) |E_emf| and E_{thermoneutral} (V) vs. Temperature (K)

FIG. 12A-12B. Cathode potentials (V) vs temperature (K) in acidic (a) and basic conditions (b).

FIG. 13A-13F. 3-D Nernst models for each hypothetical Martian CO₂ Electrolyzer (A-F), E_Nernst (V) vs. conversion (reaction extant e) vs. temperature (K)

FIG. 14. Overall performance (E_Cell (V)) vs. current density for proposed Martian electrolyzers (excluding HCHO)

FIG. 15. Theoretical production rates at various faradaic efficiencies and current densities.

FIG. 16. ASPEN simulation set up.

FIG. 17. Energy requirements for separation.

FIG. 18. Normalized power requirement.

FIG. 20A-20D. Polarization curves: (A, C) Nafion coated copper electrode. (B, D) Cu₂O coated copper electrode.

FIG. 21A-21H. Characterization of the CO₂ reduction electrode. (a) XRD of as-prepared Nafion coated Cu foam; (b) SEM imaged of the Nafion coated Cu foam and (c-h) elemental mapping of the Nafion coated Cu foam.

FIG. 22. Top: Raw intensity data recorded by the detector for incident laser light (I₀) and transmitted laser light (I_t). Bottom: Simulated and measured spectrally-resolved absorbance showing distinct absorption peaks corresponding to methane. Left column: 10.6 Torr, 298 K, and 1% CH₄/1% CO/98% CO₂ Right column: 4.5 Torr, 298 K, and 99% CH₄/1% CO₂.

FIG. 23. Left: Raw intensity data recorded by the detector for incident laser light (I₀, with empty optical fiber cell) and transmitted laser light (I_t, with the cell containing process gases from electrolyzer) during operation of the catalytic electrolyzer. Right: Calculated spectrally-resolved absorbance showing distinct absorption peaks corresponding to methane at 9 Torr and 296 K.

FIG. 24A-24E. (A) XRD data confirmed the presence of different phases of copper oxide. (B-D) FESEM images clearly revealed a well-defined crystal structure, notably in the form of truncated octahedrons. The composition of the deposition bath, along with deposition time and temperature, was found to significantly influence the Cu-oxide phase and the ratio between CuO and Cu₂O. Further work is needed to identify the optimal Cu-oxide catalyst composition for maximizing ethanol (EtOH) production through CO₂ reduction.

FIG. 25A-25B. CO₂RR experiments in an undivided cell using a two-electrode setup in 2.8 M magnesium perchlorate for 30 minutes at different potentials. (A) Calibration curve for ethanol concentrations ranging from 1000 ppm to 6 lakh ppm. (B) Faradaic efficiency for ethanol production at different voltages: 2.5 V: 86.51%,3.0 V: 13.27%,3.5 V: 6.36%.

FIG. 26. NMR analysis of a system test at room temperature using a copper oxide electrode in an undivided cell with 0.5M KHCO₃ and a three-electrode setup at −1.1 V vs RHE for 30 minutes. The results showed the presence of ethanol.

FIG. 27. NMR analysis of a system test at room temperature using electrodeposition of Cu₂O catalyst in an undivided Cell with 2.8 M Mg(ClO₄)₂ using a 3 electrode set up and an eCO2RR applied potential: −1.05V vs RHE for 30 mins.

FIG. 28. NMR analysis of catalyst 2: Electrodeposition of Cu₂O in undivided cell, electrolyte: 2.8 M Mg(ClO₄)₂ electrochemical setup with 3 electrodes eCO₂RR applied potential of −1.05V vs RHE with an electrolysis time of 30 mins at −20° C., analyzed by NMR.

FIG. 29. NMR analysis of same experimental setup, potential, and electrolysis time as FIG. 28 at −40° C.

FIG. 30A-30D. (A) XRD data. FESEM images showing (B) electrodeposited Cu₂O crystal structure, (C) electrodeposited CuO crystal structure, and (D) electrodeposited bilayer CuO on Cu₂O crystal structure.

FIG. 31A-31D. (A) Schematic process flow diagram for a Martian CO₂-regolithic brine electrolyzer and associated vapor-liquid equilibrium diagrams for the water-ethanol system at (B) 101325 Pa, (C) 5098 Pa, (D) 636 Pa.

FIG. 32. E_emf, E_{thermoneutral} and half-cell potentials as a function of temperature.

FIG. 33. Nernst potential for CO2RR to ethanol at (a) 101325 Pa, (b) 5098 Pa and (c) 636 Pa.

FIG. 34A-34B. Performance of the ethanol producing CO2-perchlorate brine electrolyzer operating with CuO cathode and RuO₂ anode at various internal cell pressures and at temperatures of (A) 255K and (B) 298K.

FIG. 35A-35C. Predicted power normalized production rates for ethanol with varying overall system efficiencies (listed in the figure legend) in an CO₂—H₂O electrolyzer operating with a stack of 10 cells of 100 cm2 each at 255K and various internal stack pressures.

FIG. 36A-36C. Predicted power normalized production rates for ethanol with varying overall system efficiencies (listed in the figure legend) in an CO₂—H₂O electrolyzer operating with a stack of 10 cells of 100 cm² each at 298K and various internal stack pressures.

FIG. 37A-37H. XRD of pristine Cu foam, and Nafion® coated Cu foam before and after electrolysis. (B) SEM imaged of the Nafion® coated Cu foam and (C—H) elemental mapping of the Nafion® coated Cu foam.

FIG. 38A-38B. (A) H-cell setup used in our experiments; (B) HPLC trace showing presence of ethanol in the electrolyte.

DESCRIPTION

The following discussion is directed to various embodiments of the invention. The term “invention” is not intended to refer to any particular embodiment or otherwise limit the scope of the disclosure. Although one or more of these embodiments may be preferred, the embodiments disclosed should not be interpreted, or otherwise used, as limiting the scope of the disclosure, including the claims. In addition, one of skill in the art will understand that the following description has broad application, and the discussion of any embodiment is meant only to be an example of that embodiment and not intended to imply that the scope of the disclosure, including the claims, is limited to that embodiment.

Provided herein are an integrated ultra-low temperature electrolyzers that produces both fuel and oxidant. In certain aspects the electrolyzer operates under conditions representative of a Martian atmosphere, e.g., using CO₂ concentrations representative of a Martian atmosphere in conjunction with liquid brines that can be found on Mars. The ultra-low temperature CO₂-brine electrolyzer (FIG. 1) can produce methane as fuel (in addition to O₂) through the following full cell reaction:

CO 2 ( g ) + 2 ⁢ H 2 ⁢ O ( l ) = CH 4 ( g ) + 2 ⁢ O 2 ( g ) ( 2 )

One objective of this invention is to utilize a liquid acidic brine feed stock and atmospheric CO₂ to operate an electrolyzer. The Nernst potential, the choice of anode/cathode and overpotentials for the electrocatalyst at the anode/cathode were correspondingly calculated through experiments to model the performance of the electrolyzers under Martian conditions. The power requirement capacity and CH₄ generation capacity of CO₂-brine electrolyzer containing a stack of 10 cells with 100 cm² is also described.

I. THE ADVANTAGES OF MARTIAN CONDITIONS

Because of the low temperatures on Mars, it was previously thought that due to Arrhenius kinetics that the lower temperature made electrochemical CO₂ reduction reaction (CO₂RR) an unfavorable process for CH₄ production. However, recently it was shown that at these lower temperatures, Cu electrocatalysts exhibit anti-Arrhenius kinetics and, counterintuitively, improved CH₄ selectivity going from 293K to 255K (23). This is due to the solubility of CO₂ increasing and the H₂-evolution reaction kinetics decreasing as temperature decreases. At these lower temperatures, it was demonstrated that gas hydrate crystalline structures called clathrates restrict rotation and translation of gas molecules (24) and that may influence the CO₂RR's selectivity. Furthermore, at these lower temperatures, the competing HER pathway is shifted towards negative overpotentials, and is directly inhibited by adsorbed CO from the CO₂RR, which leads to a much higher selectivity and faradaic efficiency for CO₂ reduction on copper electrodes (25). Ultimately, some of the conditions that make Mars uninhabitable for humans are anticipated to be beneficial for the synthesis of methalox engine propellant using a liquid brine and CO₂ electrolysis system.

II. PARAMETERS TO DESIGN AND OPERATE ELECTROLYZER

To estimate the operating cell voltage of the electrolyzer (E_cell) on Martian conditions, information on parameters like electromotive force (E_emf) and Nernst potential (E_Nernst) for electrolysis, activation overpotential for anode

( η act anode )

and cathode

( η act cathode )

along with Ohmic losses (E_IR) are required to be calculated as shown below,

E cell = E Nernst + ❘ "\[LeftBracketingBar]" η act anode ❘ "\[RightBracketingBar]" + ❘ "\[LeftBracketingBar]" η act cathode ❘ "\[RightBracketingBar]" + E IR ( 3 )

E_Nernst is estimated from E_emf and the reaction stoichiometry. This is shown as an example through water electrolysis as,

E Nernst = ❘ "\[LeftBracketingBar]" E emf ❘ "\[RightBracketingBar]" - RT nF ⁢ ln ⁡ ( [ Products ] [ Reactants ] ) ( 4 ) H 2 ⁢ O ( l ) = H 2 ( g ) + 1 2 ⁢ O 2 ( g ) ( 5 )

E Nernst = ❘ "\[LeftBracketingBar]" E emf ❘ "\[RightBracketingBar]" - RT 2 ⁢ F ⁢ ln ( [ H 2 ] [ O 2 ] 1 2 [ H 2 ⁢ O ] ) ( 6 )

For example, in the case of water electrolysis, the |E_emf| is 1.229V at 298K and 101325 Pa. But the E_Nernst is 1.278V assuming 90% conversion at 298K and 101,325 Pa as estimated through chemical reaction equilibria. Activation overpotential for anode

( η act anode )

and cathode

( η act cathode )

are estimated from the differences between the onset potential from linear sweep voltammetry for anode

( E onset anode )

or cathode

( E onset cathode )

with the Nernst potential for the anode

( E ner anode )

or cathode

( E ner cathode )

for half-cell reactions at corresponding pressure and temperature.

H act anode = E onset anode - E ner anode η act cathode = E onset cathode - E ner cathode ( 7 )

For example, in the case of water electrolysis, the activation overpotential for oxygen evolution reaction (OER) at anode

( η act anode )

and hydrogen evolution reaction (HER) at cathode

( η act cathode )

are calculated from their half-cell Nernst potentials as,

H 2 ⁢ O = 1 2 ⁢ O 2 + 2 ⁢ H + + 2 ⁢ e - ( 8 ) 2 ⁢ H + + 2 ⁢ e - = H 2 ( 9 )

at standard temperature and pressure (STP) of 298K, 1 atm and pH=0 is,

η act anode = E onset anode - 1 . 2 ⁢ 78 η act cathode = E onset cathode - 0 ( 10 )

The Ohmic loss in the electrolyzer is estimated from the resistance of the membrane and is estimated using Ohms law as,

E IR = IR mem ( 11 )

where, I is the current density (mA cm⁻²) and R_mem is the resistance of the membrane (Ω cm²).

To design the operation of an electrolyzer working at Martian conditions, for both, the electrolysis of brine (water containing Mg(ClO₄)₂) and electrolysis of brine with CO₂, the thermodynamic properties of liquid water at ultracold conditions needs to be estimated. Care is taken for careful estimation of Nernst potentials for the corresponding half-cell reactions instead of equilibrium potentials for half-cell reactions at anode and cathode for all cases.

In some embodiments the thermodynamic properties, the electromotive force, and the corresponding equilibrium potentials for half-cell reactions at the anode and cathode for the following reactions are required for designing the operation of an electrolyzer working at Martian conditions.

H 2 ⁢ O ⁢ ( l ) = H 2 ⁢ ( g ) + 1 2 ⁢ O 2 ⁢ ( g ) ( Brine - electrolysis ) ( 12 ) CO 2 ( g ) + 2 ⁢ H 2 ⁢ O ( l ) = CH 4 ⁢ ( g ) + 2 ⁢ O 2 ( g ) ( ⁢ CO 2 - Brine ⁢ electrolysis ) ( 13 )

III. THERMODYNAMIC PROPERTIES AT MARTIAN CONDITIONS

Thermodynamic properties like change in entropy (dS) and change in enthalpy (dH) are required to calculate the free energy change (dG) as shown below,

dG = dH - T ⁢ dS ( 14 )

The estimated dG is correlated to |E_emf| using the equation,

Δ ⁢ G = - nF ⁢ ❘ "\[LeftBracketingBar]" E emf ❘ "\[RightBracketingBar]" ( 15 )

where the number of electrons (n) is either 2 or 8 depending on whether electrolysis or electrolysis is carried out by the electrolyzer, and F corresponds to the number of electrons and Faraday constant (˜96485 C mol⁻¹).

For brine electrolysis, the thermodynamic properties of H₂O in liquid phase are between 230K to 270K. The thermodynamic properties (dS and dH) are calculated from the specific heat capacity (C_p, J mol⁻¹ K⁻¹) expression of H₂O reported in literature²⁶ for temperatures between 230K to 270K with T_ref at 298K as given below,

C p = 0 . 4 ⁢ 4 * ( T - 2 ⁢ 2 ⁢ 2 2 ⁢ 2 ⁢ 2 ) - 2 . 5 + 74.3 ( for - H 2 ⁢ O ) ( 16 ) dS = C p * ln ( T T ref ) ( 17 ) dH = C p * ( T - T ref ) ( 18 )

An expression correlating E_emf with the temperature (T) for temperatures between 230K to 1300K is also estimated and the thermoneutral voltage (E_{thermoneutral}) is calculated from the corresponding dH values as given below,

E emf = ( 2 ⁢ e - 7 * T 2 ) - ( 6 ⁢ e - 4 * T ) + 1 .4251 ( 19 ) E thermoneutral = dH 2 ⁢ F ( 20 )

For CO₂-Brine electrolysis, in addition to the thermodynamic properties of H₂O in liquid phase, thermodynamic properties of CO₂ and CH₄ are required between 230K to 270K. The thermodynamic properties (dS and dH) are calculated from the specific heat capacity (C_p, J mol⁻¹ K⁻¹) expression of CO₂ and CH₄ as reported in literature (27,28) for temperatures between 220-300K and 120-500K with T_ref at 298 K as given below,

C p = 0.039 * T + 25.743 ( for - CO 2 ) ( 21 ) C p = 0 . 0 ⁢ 0 ⁢ 0 ⁢ 1 * T 2 - 0 . 0 ⁢ 3 ⁢ 8 * T + 36.839 ( for - CH 4 ) ( 22 ) dS = C p * ln ⁡ ( T T ref ) ( 23 ) dH = C p * ( T - T ref ) ( 24 )

An expression correlating E_emf with the temperature (T) for temperatures between 230K to 1300K is also estimated and the thermoneutral voltage (E_{thermoneutral}) is calculated from the corresponding dH values as given below,

E emf = ( 1 ⁢ e - 7 * T 2 ) - ( 2 ⁢ e - 4 * T ) + 1 .1238 ( 25 ) E thermoneutral = d ⁢ H 8 ⁢ F ( 26 )

The thermodynamic properties like dG, dH along with E_emf and E_{thermoneutral} for brine electrolysis and CO2-brine electrolysis are shown in FIG. 2.

IV. NERNST POTENTIALS FOR ELECTROLYSIS

The Nernst potentials at 101325 Pa for brine electrolysis is estimated using the expression,

E Nernst = ❘ "\[LeftBracketingBar]" E emf ❘ "\[RightBracketingBar]" - R ⁢ T 2 ⁢ F ⁢ ln ⁡ ( [ y H 2 ] [ y O 2 ] 1 2 [ y H 2 ⁢ O ] ) ( 27 )

where the mole fractions of H₂, O₂ and H₂O are written as y_H2, y_O2 and y_H2O and y_H2O is ˜1 as it is in liquid phase and the reaction reduces to,

E Nernst = ❘ "\[LeftBracketingBar]" E emf ❘ "\[RightBracketingBar]" - R ⁢ T 2 ⁢ F ⁢ ln ⁡ ( [ y H 2 ] [ y O 2 ] 1 2 ) ( 28 )

The Nernst potentials at 636 Pa for brine electrolysis is estimated using the expression,

E Nernst = ❘ "\[LeftBracketingBar]" E emf ❘ "\[RightBracketingBar]" - R ⁢ T 2 ⁢ F ⁢ ln ⁡ ( [ y H 2 ] [ y O 2 ] 1 2 * [ P Mars P Std ] 3 2 ) ( 29 )

The additional pressure ratio term,

P Mars P Std ,

allows the Nernst Potential to be adjusted for the pressure correction and the pressure at Mars is 636 Pa and standard Earth pressure is 101325 Pa. Similarly, the Nernst potentials at 101325 Pa for CO₂-brine electrolysis is estimated using the expression,

E Nernst = ❘ "\[LeftBracketingBar]" E emf ❘ "\[RightBracketingBar]" - R ⁢ T 8 ⁢ F ⁢ ln ⁡ ( [ y CH 4 ] [ y O 2 ] 2 [ y CO 2 ] ) ( 30 )

where the mole fractions of CH₄, O₂ and CO₂ are written as y_CH4, y_O2 and y_CO2 and y_H2O is ˜1 as it is in liquid phase.

The Nernst potentials at 636 Pa for CO₂-brine electrolysis is estimated after incorporating the corrections for pressure is (29-30),

E Nernst = ❘ "\[LeftBracketingBar]" E emf ❘ "\[RightBracketingBar]" - R ⁢ T 8 ⁢ F ⁢ ln ⁡ ( [ y CH 4 ] [ y O 2 ] 2 [ y CO 2 ] * [ P Mars P Std ] 2 ) ( 31 )

3Dimensional (3D) surface plots for the Nernst potentials at various reaction extent (conversion, F) at temperatures ranging from 230K to 300K are estimated as shown in FIG. 3 for brine electrolysis (FIG. 3A) and CO₂-brine electrolysis (FIG. 3B) at 636 Pa and 101325 Pa. For better visualization, the 2Dimensional (2D) plots of the Nernst potential at various F values at 236K or 298K and 636 Pa or 101326 Pa are shown in FIG. 4 for brine electrolysis and CO₂-brine electrolysis. CO₂ electrolysis to carbon monoxide (CO) and O₂, MOXIE at low temperature was also studied for comparison. The results clearly show the advantages of operating a CO₂-brine electrolyzer over a brine electrolyzer and MOXIE at low temperatures with lesser values of Nernst potential on Martian conditions and Earth conditions.

V. BRINE ELECTROLYSIS

Some assumptions can be considered before modeling the operation of brine electrolyzer as stated below, (1) The brine electrolyzer operates at 237K and 101325 Pa with a conversion efficiency of 90%. (2) The brine solution is made up of 2.8M Mg(ClO₄)₂ dissolved in water and is assumed to be in liquid phase at 237K and 101325 Pa. The brine solution is acidic with pH=3 as measured at standard atmospheric conditions.

Case 1: Oxygen evolution reaction (OER) is carried out at the anode by lead ruthenate (Pb₂Ru₂O₇) and hydrogen evolution reaction (HER) is carried out at the cathode by a composite of platinum-carbon black (Pt—C) and the activation overpotentials are calculated from the onset potentials of experiments with the brine at 237K and 101325 Pa and compared with experimental data.

Case 2: OER is carried out at the anode by ruthenium oxide (RuO₂) and HER is carried out at the cathode by a composite of platinum-carbon black (Pt—C) and the activation overpotentials are calculated from the onset potentials of experiments with brine at 237K and 101325 Pa.

With the estimation of E_Nernst for brine electrolysis at various temperatures (˜230-500K) and pressures (636 Pa and 101325 Pa) using the thermodynamic data and the Nernst equation, the activation overpotential at the anode and cathode needs to be further estimated 13 at a specific operating conditions of a brine electrolyzer. The E_Nernst for a brine electrolyzer operating at 237K and 101325 Pa with a conversion efficiency of 90% is

E Nernst = 1.3404 V ( 32 )

Case 1: The activation overpotential for OER at Pb₂Ru₂O₇ anode and HER at Pt—C cathode are calculated from half-cell Nernst potentials,

H 2 ⁢ O = 1 2 ⁢ O 2 + 2 ⁢ H + + 2 ⁢ e - ⁢ ( E ner anode , 237 ⁢ k , 101325 ⁢ Pa = - 1.304 ⁢ V ) ( 33 ) 2 ⁢ H + + 2 ⁢ e - = H 2 ⁢ ( E ner cathode , 237 ⁢ k , 101325 ⁢ Pa = 0 ⁢ V ) ( 34 )

at electrolyzer operating conditions of 237 K, 101325 Pa and pH=3 is,

H 2 ⁢ O = 1 2 ⁢ O 2 + 2 ⁢ H + + 2 ⁢ e - ⁢ ( E ner anode , 237 ⁢ k , 101325 ⁢ Pa = - 1.162905 ⁢ V ) ( 35 ) 2 ⁢ H + + 2 ⁢ e - = H 2 ⁢ ( E ner cathode , 237 ⁢ k , 101325 ⁢ Pa = - 0.14109 ⁢ V ) ( 36 ) ❘ "\[LeftBracketingBar]" η act Pb 2 ⁢ Ru 2 ⁢ O 7 ❘ "\[RightBracketingBar]" = E onset Pb 2 ⁢ Ru 2 ⁢ O 7 - 1.162905 ❘ "\[LeftBracketingBar]" η act Pt - C ❘ "\[RightBracketingBar]" = E onset Pt - C - ( - 0.14109 ) ( 37 )

E onset Pb 2 ⁢ Ru 2 ⁢ O 7 ⁢ and ⁢ E onset Pt - C

for the anode and cathode are estimated to be 1.3489V and −0.5511V from the experiments performed on brine at 237K and 101325 Pa. The calculated activation overpotential for the electrodes are,

❘ "\[LeftBracketingBar]" η act Pb 2 ⁢ Ru 2 ⁢ O 7 ❘ "\[RightBracketingBar]" = 0.185885 V ⁢ and ⁢ ❘ "\[LeftBracketingBar]" η onset Pt - C ❘ "\[RightBracketingBar]" = 0.41001 V ( 38 )

The total activation overpotential,

η act = ❘ "\[LeftBracketingBar]" η act Pb 2 ⁢ Ru 2 ⁢ O 7 ❘ "\[RightBracketingBar]" + ❘ "\[LeftBracketingBar]" η act Pt - C ❘ "\[RightBracketingBar]" = 0.599 V ( 39 )

The Ohmic loss is estimated from the R_mem at varying I values. The R_mem is reported to vary between 0.2-0.7 Ω cm² and the maximum value of 0.7 Ω cm² is used for estimation of E_Ω. A model representing the performance of the brine electrolyzer (solid red line) at 237K and 101325 Pa with Pb₂Ru₂O₇ anode and Pt—C cathode is shown in FIG. 5. The experimental values (brown squares) are compared with the performance of the model as shown in FIG. 5. The experimental values lie above the 70% efficiency (dotted red line) of the estimated values of the model performance.

Case 2: The activation overpotential for oxygen evolution reaction (OER) at RuO₂ anode and hydrogen evolution reaction (HER) at Pt—C cathode are calculated from half-cell Nernst potentials,

H 2 ⁢ O = 1 2 ⁢ O 2 + 2 ⁢ H + + 2 ⁢ e - ⁢ ( E ner anode , 237 ⁢ k , 101325 ⁢ Pa = - 1.304 ⁢ V ) ( 40 ) 2 ⁢ H + + 2 ⁢ e - = H 2 ⁢ ( E ner cathode , 237 ⁢ k , 101325 ⁢ Pa = 0 ⁢ V ) ( 41 )

at electrolyzer operating conditions of 237K, 101325 Pa and pH=3,

H 2 ⁢ O = 1 2 ⁢ O 2 + 2 ⁢ H + + 2 ⁢ e - ⁢ ( E ner anode , 237 ⁢ k , 101325 ⁢ Pa = - 1.162905 ⁢ V ) ( 42 ) 2 ⁢ H + + 2 ⁢ e - = H 2 ⁢ ( E ner cathode , 237 ⁢ k , 101325 ⁢ Pa = - 0.14109 ⁢ V ) ( 43 ) ❘ "\[LeftBracketingBar]" η act RuO 2 ❘ "\[RightBracketingBar]" = E onset RuO 2 - 1.162905 ❘ "\[LeftBracketingBar]" η act Pt - C ❘ "\[RightBracketingBar]" = E onset Pt - C - ( - 0.14109 ) ( 44 )

E onset RuO 2 ⁢ and ⁢ E onset Pt - C

for the anode and cathode are estimated to be 1.5489V and −0.5511V from the experiments performed with brine at 237K and 101325 Pa. The calculated activation overpotential for the electrodes are,

❘ "\[LeftBracketingBar]" η act RuO 2 ❘ "\[RightBracketingBar]" = 0.385995 V ⁢ ❘ "\[LeftBracketingBar]" η act Pt - C ❘ "\[RightBracketingBar]" = 0.41001 V ( 45 )

The total activation overpotential,

η act = ❘ "\[LeftBracketingBar]" η act RuO 2 ❘ "\[RightBracketingBar]" + ❘ "\[LeftBracketingBar]" η act Pt - C ❘ "\[RightBracketingBar]" = 0.796 V ( 46 )

The Ohmic loss is estimated from the R_mem at varying I values. The R_mem is reported to vary between 0.2-0.7 Ωcm² and the maximum value of 0.7 Ωcm² is used for estimation of E_IR. A model representing the performance of the brine electrolyzer (red line) at 237K and 101325 Pa with RuO₂ anode and Pt—C cathode is shown in FIG. 6.

VI. ELECTROLYSIS UNDER MARTIAN CONDITIONS

Some assumptions are considered before modeling the operation of the CO₂-brine electrolyzer as stated below, (1) The CO₂-brine electrolyzer operates at 255K and 101325 Pa with a conversion efficiency of 90%. (2) The brine solution is made up of 2.8 M Mg(ClO₄)₂ dissolved in water and is assumed to be in liquid phase at 255K and 101325 Pa. (3) The brine solution is acidic with pH=3 as measured at standard atmospheric conditions. (4) OER is carried out at the anode by ruthenium oxide (RuO₂) and carbon-dioxide reduction reaction (CO₂ RR) is carried out at the cathode by copper (Cu(111)) and the 17 activation overpotentials are calculated from the onset potentials of experiments with brine at 255K and 101325 Pa.

With the estimation of E_Nernst for the CO₂-brine electrolysis at various temperatures (˜230-500K) and pressures (636 Pa and 101325 Pa) using the thermodynamic data and the Nernst equation, the E_Nernst for the CO₂-brine electrolyzer operating at 255K and 101325 Pa with a conversion efficiency of 90% is,

E Nernst = 1.07904 V ( 47 )

The activation overpotential for oxygen evolution reaction (OER) at RuO₂ anode and carbon-dioxide reduction reaction (CO₂RR) at Cu(111) cathode are calculated from half-cell Nernst potentials,

4 ⁢ H 2 ⁢ O = 2 ⁢ O 2 + 8 ⁢ H + + 8 ⁢ e - ⁢ ( E ner anode , 237 ⁢ k , 101325 ⁢ Pa = - 1.2968 ⁢ V ) ( 48 ) CO 2 + 8 ⁢ H + + 8 ⁢ e - = CH 4 + 2 ⁢ H 2 ⁢ O ⁢ ( E ner cathode , 237 ⁢ k , 101325 ⁢ Pa = 0.21773 V ) ( 49 )

at electrolyzer operating conditions of 237 K, 101325 Pa and pH=3 is,

4 ⁢ H 2 ⁢ O = 2 ⁢ O 2 + 8 ⁢ H + + 8 ⁢ e - ⁢ ( E ner anode , 237 ⁢ k , 101325 ⁢ Pa = - 1.14496 ⁢ V ) ( 50 ) CO 2 + 8 ⁢ H + + 8 ⁢ e - = CH 4 + 2 ⁢ H 2 ⁢ O ⁢ ( E ner cathode , 237 ⁢ k , 101325 ⁢ Pa = 0.058189 V ) ( 51 ) ❘ "\[LeftBracketingBar]" η act RuO 2 ❘ "\[RightBracketingBar]" = E onset RuO 2 - 1.14496 ❘ "\[LeftBracketingBar]" η act Cu ⁡ ( 111 ) ❘ "\[RightBracketingBar]" = E onset Cu ⁡ ( 111 ) - ( 0.058189 ) ( 52 )

E onset RuO 2 ⁢ and ⁢ E onset Cu ⁡ ( 111 )

for the anode and cathode are estimated to be 1.48819V and −0.25V from the experiments performed with brine at 255K and 101325 Pa. The calculated activation overpotential for the electrodes are,

❘ "\[LeftBracketingBar]" η act RuO 2 ❘ "\[RightBracketingBar]" = 0.3432 V ⁢ ❘ "\[LeftBracketingBar]" η act Pt - C ❘ "\[RightBracketingBar]" = 0.3159 V ( 53 )

The total activation overpotential,

η act = ❘ "\[LeftBracketingBar]" η act RuO 2 ❘ "\[RightBracketingBar]" + ❘ "\[LeftBracketingBar]" η act Pt - C ❘ "\[RightBracketingBar]" = 0.65915 V ( 54 )

Since the only experimental data is available for CO₂RR by Cu(111) cathode and OER by RuO₂ anode, the model of a CO₂-brine electrolyzer requires information on the Ohmic loss by the membrane to display the performance. R_mem is assumed to be ˜1.22 Ωcm² i.e., assuming the conductivity of a membrane for the CO₂-brine electrolyzer to be equal to the conductivity of Nafion 117 at 303K. A model representing the performance of the CO₂-brine electrolyzer (red line) at 255K and 101325 Pa with RuO₂ anode and Cu(111) cathode is shown in FIG. 7.

Described herein is a low temperature electrolyzer operating on brine, with and without CO₂ at 236K and 255K. It was found that the Nernst potential varies slightly with change in temperature between 236K and 298K. However, there are no substantial changes in the Nernst potential with changes in pressure between 636 Pa and 101325 Pa. Hence, the experimental study or modeling of electrolyzers on earth at 101325 Pa and at temperatures between 236 K to 298 K would resemble the performance of electrolyzers on Martian atmosphere. The brine electrolysis is carried out on 2.8 M Mg(ClO₄)₂ containing solution which simulates the brines existing in Martian conditions for generation of O₂ and the CO₂-brine electrolysis is carried out with CO₂ and brine for generation of both O₂ and CH₄. The performance of our brine electrolyzer was studied with two different catalysts at anode for OER at 236K. The Pb₂Ru₂O₇ anode outperforms RuO₂ anode as the former has low overpotentials for OER and the performance of the brine electrolyzer was compared with experiments conducted at 236K and 101325 Pa. RuO₂ shows more stability as compared to Pb₂Ru₂O₇ as the brine is operated at acidic conditions. The Pb₂Ru₂O₇ anode (for OER) is used along with Cu(111) cathode (for CO₂RR) in CO₂-brine electrolysis and the performance was studied at 255K and 101325 Pa.

As shown in our model (FIG. 8), a CO₂-brine electrolyzer could produce the necessary methane fuel that a larger MOXIE design cannot produce for future manned Martian missions. An electrolyzer system operating with 2V applied at each cell has a theoretical power normalized production of 406 LW⁻¹day⁻¹ of CH₄ and 2.08 LW⁻¹day⁻¹ of O₂, assuming 90% O₂ and 35% CH₄ catalyst faradaic efficiency. The CO₂-brine electrolyzer produces both components of methalox engine propellant, with the additional life-support O₂ applicability. The theoretical oxygen production is x times more power efficient than MOXIE. This theoretical value will likely change as our group, and the MOXIE team further model the cryogenic process of storing the propellant and further balance of plant (BOP). Additionally, as we research and develop more efficient electrocatalysts, the electrolyzer's CH₄ production would increase if CO₂RR electrocatalysts have the efficiencies reported in Table 3. Furthermore, the ultra-cold temperatures and liquid brines present on the Martian surface have been shown to increase the methane selectivity of copper CO₂RR electrocatalysts. Further developing a CO₂RR electrocatalyst with these conditions in mind is a part of the developmental roadmap of our research.

An important consensus is that a scaled-up MOXIE system is imperative for the first manned mission to Mars, and not directly competing with our CO₂-brine electrolyzer. Our proposed Martian CO₂-brine electrolyzer would currently require astronauts to operate initially, due to the need for a source of water/brine feed stock. Hypothetically, astronauts could either mine brine ice or harvest sub-surface brine-water by creating a well. Beyond that, with further analogous system development, future missions could be supported by both systems remotely.

To further expand this technique considering the abundance of Martian brine and the technology developed to harvest it, we presumed that reliable Martian ISRU feedstocks are liquid brine and atmospheric CO₂ for electro-chemical synthesis of key products to ensure sustainable exploration on Mars. The electro-chemical synthesis of CO, HCHO, CH₃OH, CH₄, C₂H₄, and C₂H₅OH as key products is via the aforementioned Martian CO₂ and liquid brine as electrolyzer feedstocks for CO₂ reduction. Table 4 is a summary of the applications of key products and their uses as industrial precursors and vital rocket propellant components.

B. Electrolyzer Configuration

CO₂ electrolysis has advanced significantly since their early integration for renewable energy syngas pilot plants (35) in the 1980's. Literature report the advantages of aqueous ionic and non-ionic liquid additives, in significantly aiding the selectivity of the reaction towards C₂₊ hydrocarbon products (36). In recent years, the CO₂ electrolysis was largely carried out using a cation exchange membrane due to the high cationic conductivity. But, the acidic environment negatively influences the selectivity for CO₂RR (37). However, further studies of alkaline CO₂ electrolyzers remain prospective due to the advancements in anion exchange membrane technology (38) and their selectivity towards C₂ hydrocarbons, which serve as feedstock for synthesizing other petrochemical products. The modeling efforts herein are conducted under the assumption that the CO₂—brine electrolyzer would utilize—CO₂ from the Martian atmosphere at the cathode, brine solution containing magnesium perchlorate at the anode with a cation exchange membrane, to synthesize key products. A 2-D schematic diagram of the electrolyzer is shown in FIG. 9. The working configuration is a zero-gap with a catalyst coated porous electrodes to allow for brine permeation and CO₂ gas diffusion.

Catalyst Selectivity. Throughout the field of CO₂ electrolysis, copper-based catalysts have exhibited a majority of the higher selectivity towards C₁ and C₂₊ hydrocarbon products(39,40). Highly efficient Martian key resource production is possible through tailoring these copper catalysts to selectively produce to and increase the faradaic efficiency of CH₃OH, CH₄, C₂H₄, and C₂H₅OH synthesis to above 80%, additionally, utilizing Ag and Au catalysts to efficiently produce CO with faradaic efficiencies above 95%, and lastly, exploiting the unique case of boron doped diamond efficiently producing HCHO with a faradaic efficiency of ˜75% as seen in Table 5. Our modelling incorporates the reported faradaic efficiencies and activity with the published activation overpotentials of these cutting-edge catalysts in order to model electrolyzer performance on Mars and theoretically determine the production rates of CO, CH₃OH, CH₄, C₂H₄, and C₂H₅OH.

Overall CO₂ Electrolyzer Reactions. Researchers have exploited catalyst descriptors in conjunction with density functional theory (DFT) calculations to design these highly selective and high performing electrocatalysts that favor key products from CO₂ reduction. Table 6 can be used as an overall reference for the reactions happening within the proposed Martian electrolyzers.

Reaction Pathways. Each mechanism of the acidic reactions from Table 6 have been theoretically determined from DFT and experimental analysis of these high performing CO₂ electroreduction catalysts. Current theoretical mapping of these proposed reaction pathways and their intermediates can be seen in FIGS. 10A and 10B.

Key Product Mechanism Explorations

1. CO (Carbon Monoxide) CO₂RR Cathode Mechanism

Electroreduction of CO₂ into CO is a promising pathway for an intermediate in fuel generation, as seen in the Fischer-Tropsch process (63), in fact carbon monoxide and oxygen are already being produced by MOXIE on the Martian rover Perseverance (5). Thus, carbon monoxide is a potential key product for Martian ISRU-based propellant plant technology. There are multiple reduction reaction mechanisms that can be carried out on the electrode's surface, which is dependent on whether the electrolyzer is in acidic or alkaline conditions, or in MOXIE's case a Solid Oxide Electrolyzer (SOXE). The possible acidic and alkaline reaction schemes present in this theoretical electrolyzer can be seen in Table 6.

Each of these reactions have a specific mechanism and intermediates that form at the interface of the electrode where CO₂ is being reduced. In the case of MOXIE, atmospheric CO₂ is fed into the SOXE and reduced into CO and O₂. This SOXE technology utilizes a high performing Ni-ceria Cermet cathode, a doped lanthanum cobalt ferrite anode, and scandia stabilized Zirconia (ScSZ) as the solid electrolyte (64). The CO₂ binds to the Ni-ceria cermet interface where an electric field is applied to reduce CO₂, this uniquely forms O²⁻ ions that are carried through the ScSZ solid electrolyte to the anode. The ScSZ has physical vacancies in the lattice structure that carry the O²⁻ ions to the anode where O₂ forms and desorbs from the surface of the doped lanthanum cobalt ferrite. SOXE and solid oxide fuel cell (SOFC) technology can further undergo the Boudouard reaction where CO is couples with another CO molecule and forms C+CO₂, which creates carbon coking in the reactor (65). Avoiding this carbon coking was a critical design consideration that led to our initial modelling of a Martian ISRU aqueous low temperature electrolyzer (6). Thus, we have focused on the 2e⁻ reduction mechanism seen in FIG. 11A.

In the case of an aqueous ISRU electrolysis approach, atmospheric CO₂ and the brine are the feedstocks. The pH of the catholyte can be acidic, alkaline (basic), or neutral, which changes the mechanism of how CO₂ reacts at the surface interface of the electrode and subsequently changes the reaction intermediates and the electric double layer surrounding the electrode interface. For an example, take the highest performing CO synthesizing catalyst in Table 5, an electrolyzer that utilizes a neutral 0.5M KHCO₃ electrolyte (pH=7.5), a Ag nanocrystal cathode, and a Nafion® 212 cation exchange membrane (CEM) (41). Due to this system exchanging H⁺ cations across the cation exchange membrane (CEM) with the neutral KHCO₃ electrolyte, the acidic reaction takes place. Our modelling efforts incorporate theoretical acidic reaction conditions across a cation exchange membrane, as mentioned in the electrolyzer configuration. This reaction mechanism is a 2-electron transfer, starting with the dissolved CO₂ adsorbing to the surface of the proposed electrolyzer cathode:

2. HCHO (Formaldehyde) CO₂RR Cathode Mechanism

Electrochemically reducing CO₂ into formaldehyde is one of the more difficult reactions to tailor a catalyst to favor. Nakata et al., reported that boron doped diamond can achieve up to a 74% FEHCHO in an acidic reaction scheme with a Nafion® membrane in 0.1M methanol electrolyte (46). This was the only study to report a highly selective formaldehyde product from CO₂ reduction. This can largely be attributed to formaldehyde being an intermediate during CO₂ reduction into methanol (66). Additionally, it is worth noting that formaldehyde has been experimentally shown to not be an intermediate in the pathway towards reduction into methane, this can be further explained by the work of Bagger et al. where their group determined that formaldehyde is a key intermediate towards methanol production when utilizing weak binding metal catalysts that have weaker carbon binding that form a formaldehyde intermediate as a result (67). Production of formaldehyde on boron doped diamond (BDD) electrodes in methanol electrolyte takes advantage of this mechanism with the COOH* carboxylic acid intermediate forming at the surface of the electrode. Where in, the acidic reaction is carried out from the Nafion® membrane H⁺ cation transport, with further protonation at the surface of the BDD electrode yielding a HCHO formaldehyde product. Nakata's group experimentally determined that this mechanism is likely due to the presence of sp³ carbon in BDD, where there is an initial reduction of CO₂ into a formic acid intermediate, and a final reduction into formaldehyde at the sites of sp³ carbon on the BDD electrode interface with HER as the main competing reaction (46). The alkaline cathode reaction mechanism is assumed to be unfavorable, considering that formaldehyde production relies on synthesizing carboxylic acid formic acid intermediates at the BDD sp³ carbon sites and protonation from the Nafion® exchange membrane.

3. CH₃OH (Methanol) CO₂RR Cathode Mechanism

In electrochemical systems that selectively produce methanol from CO₂ reduction there are predominantly Nafion® membranes with the acidic reaction taking place in the electrolyzers. As reviewed previously, the intermediate that leads to higher selectivity towards methanol is formaldehyde (HCHO*). The initial step for CO₂ reduction into methanol is the formation of the initial carboxylic acid intermediate COOH* and is further reduced in the following reaction pathway (68): CO₂→COOH*→CO*→COH*→CHOH*→CH₂OH*→CH₃OH. The key to tailoring catalysts to follow this pathway is to utilize electrode catalysts that have a CO* endergonic step (positive free energy) which will continue to reduce the intermediates on the surface into methanol, rather than desorb as a CO product (47,69).

4. CH₄ (Methane) CO₂RR Cathode Mechanism

CO₂ reduction into methane mechanism follows the same initial intermediate steps that produce methanol. Catalysts that are tailored to produce methane rather than methanol have active sites that bind a C* intermediate over a CHOH* intermediate. The reaction pathway thus becomes (47): CO₂→COOH*→CO*→COH*→C*→CH*→

CH 2 * → CH 3 *

→CH₄. Through utilizing catalysts, like copper, you can promote the protonation of CO* to COH* and tailor the catalyst to form C* and CH* through alloying, ligand stabilization, and tethering, or by using promoters (70). The highest performing CH₄ electrocatalyst is copper with a drop casted Nafion® surface modification, increasing the FE to 88% as seen in Table 5.
5. C₂H₄(Ethylene) CO₂RR Cathode Mechanism

In order to produce any C₂₊ products from CO₂ reduction the CO* intermediate must go through dimerization, where a C—C bond is formed when a hydrogenated COH* intermediate binds with a CO* intermediate to form OCCOH* intermediates at the surface of the cathode, additionally, CO*—CO* dimerization is also a possible pathway, however protonation from OCCO*—OCCOH is instantaneous and the same reaction pathways will proceed. This dimerization is the most favorable pathway for the formation of C₂₊ products, and thus, utilizing catalysts that reduce the activation barrier for C—C dimerization and further C₂H₄ desorption will avoid other C₂₊ products (54). The favorable C₂H₄ dimerization pathway is as follows: CO₂→COOH*→CO*+COH*→OCCOH*→OCC*+H₂O→OCCH*→OCCH₂→OCCH*→

OCCH 2 * → OCCH 3 * →

C₂H₄. The initial intermediates have a carbon bond at the surface, however, when the dimerization binding forms OCCOH* and is further hydrogenated, the water desorbs leaving a OCC*intermediate that has only one carbon bond with the surface and an oxygen bond with the cathode surface. Once the OCCH* intermediate is formed, the carbon is no longer bound to the catalyst surface, leaving only an oxygen bond at the surface, which is lost through hydrogenation and the final desorption of C₂H₄. The higher selectivity could be attributed to the dimerization pathway proceeding with the oxygen—surface bond rather than following a pathway with only carbon—surface bonds (54). Ethylene can form from a carbon—surface pathway as follows (60): CO₂→COOH*→CO*+COH*→OCCOH*→HOCCOH*→HOCC*+H₂O→HOCHC*→HCC*+H₂O→H₂CC*→H₂CCH*→C₂H₄. Additionally, in the formation of methane, there is a side reaction of the CH* intermediate going through dimerization and forming C₂H₄(71). However, these pathways are side reactions that were determined to be less thermodynamically favored through DFT calculations and experiments (60,71).

Multiple groups have shown that reducing CO₂ into ethylene through the overall alkaline reaction has a positive effect on the dimerization step (56,57). Sargent et al. was able to determine that using a hydroxide mediated surface created Cu(OH)₂ molecules that stabilized the OCCO* intermediates through strong dipole interactions (58). This interaction lowers the activation energy barrier for the dimerization step in the reaction pathway and proceeds to form C₂H₄ and other C₂₊ products under alkaline conditions. It is worth pointing out that the most efficient ethylene producing CO₂ electrocatalyst reported in literature is CuO nanoplates with 84.5% FE_C2H4 performed in acidic conditions (54). However, with further development of alkaline condition catalysts that utilize the dipole interactions that increase the dimerization mechanism of the C₂₊ pathway, an alkaline reaction scheme could be favored for ethylene synthesis.

6. C₂H₅OH (Ethanol) CO₂RR Cathode Mechanism

C₂H₅OH formation relies on the same initial C₂H₄ dimerization of CO* and COH* intermediates, however, the highest performing reaction pathway relies on the carbon—surface bond pathway rather than the oxygen—surface bond pathway, effectively switching the main and side reaction pathways compared to C₂H₄. This reaction pathway is as follows (60): CO₂—COOH*→CO*+COH*→OCCOH*→HOCCOH*→HOCC*+H₂O→HOCCH*→HOCHCH*→

HOCHCH 2 * → HOCHCH 3 *

→C₂H₅OH. A majority of the high performing ethanol producing CO₂RR electrocatalysts in literature utilized an acidic reaction and cation exchange membrane, compared to the dual acidic—alkaline C₂H₄ electrolyzers (59-61).

VII. CO₂ ELECTROLYZER MODEL DEVELOPMENT

Thermodynamic Properties. The first step in modelling these electrolyzers is to gather the thermodynamic properties (dS, dH and dG between 230K to 298K at 101.325 KPa) for each reaction depicted in Table 6. Modelling the electro-reduction of carbon-dioxide (CO₂) to the various key products carbon monoxide (CO), formaldehyde (HCHO), methanol (CH₃OH), methane (CH₄), ethanol (C₂H₅OH) and ethylene (C₂H₄) depends on the number of electrons utilized for reduction of CO₂. Enthalpy, entropy, and gibbs free energy (dS, dH and dG) are specific for each reaction, thus each model has different electrochemical values based on the number of electrons in the reaction and the initial thermodynamic values. However, the operation of the electrolyzer can be carried out under both, acidic and basic conditions, and the overall thermodynamic potential, i.e., electromotive force (emf) for the reaction, remains unchanged for each product as shown in Table 6.

The change in enthalpy (dH) is calculated using the following relationship—

dH = H 298 ⁢ K + c p × ( T - T ref ) ( 1 )

The change in entropy (dS) is calculated using the following relation with the specific heat capacity (c_p)—

dS = S 298 ⁢ K + c p × ln ⁡ ( T T ref ) ( 2 )

Having obtained dH and dS for the given reaction, the free energy change (dG) is calculated as shown below—

dG = dH - T ⁢ dS ( 3 )

dG is in turn used to calculate the reversible electromotive force (E_emf) i.e., the minimum thermodynamic potential required for the reactions to occur, using the following equation—

Δ ⁢ G = - nF ⁢ ❘ "\[LeftBracketingBar]" E emf ❘ "\[RightBracketingBar]" ( 4 )

where the number of electrons (n) is 2 to 12 depending on whether brine electrolysis or CO₂-brine electrolysis is carried out by the electrolyzer, and F is Faraday's constant (96,485 C mol⁻¹).

In the case of the anode, where the half-cell reaction of brine oxidation is carried out, the thermodynamic properties (dS and dH) are calculated from the specific heat capacity (C_p, J mol⁻¹ K⁻¹) expression,

C p H 2 ⁢ O = 0.44 × ( T - 222 2 ⁢ 2 ⁢ 2 ) - 2 . 5 + 7 ⁢ 4 . 3 ( 5 )

Similarly, for the cathode, where the half-cell reaction of CO₂ reduction is carried out to form desired key products, the thermodynamic properties (dS and dH) are calculated from the specific heat capacity (C_p, J mol⁻¹ K⁻¹) data⁷² as a function of temperature (T) as the following equations,

C p CO 2 = 0 . 0 ⁢ 3 ⁢ 9 × T + 2 ⁢ 5 .74 ( 6 ) C p CO = 2 ⁢ 9 . 0 ⁢ 9 ⁢ 9 + ( - 1 × 1 ⁢ 0 - 4 × T ) + ( 7 × 1 ⁢ 0 - 7 × T 2 ) ( 7 ) C p HCHO = 2 ⁢ 4 . 0 ⁢ 7 ⁢ 9 ⁢ 2 + ( 0 . 0 ⁢ 3 ⁢ 9 ⁢ 7 × T ) ( 8 ) C p C ⁢ H 3 ⁢ OH = 4 ⁢ 6 . 5 ⁢ 6 ⁢ 4 + ( 0 . 1 ⁢ 1 ⁢ 5 ⁢ 3 × T ) ( 9 ) C p C ⁢ H 4 = ( 0 . 0 ⁢ 0 ⁢ 0 ⁢ 1 × T 2 ) - ( 0 . 0 ⁢ 3 ⁢ 8 × T ) + 36.84 ( 10 ) C p C 2 ⁢ H 5 ⁢ OH = ( 0.6102 × T ) + 61. 98 ( 11 ) C p C 2 ⁢ H 4 = ( 0 . 0 ⁢ 8 ⁢ 3 ⁢ 4 × T ) + 1 ⁢ 8 . 3 ⁢ 7 ⁢ 4 ( 12 )

Having obtained the thermodynamic values as detailed above, we calculated |E_emf| and E_{thermoneutral} across the temperature range (230K to 298K) for each reaction in Table 6. We also obtained the following general empirical correlation between |E_emf| vs. temperature (T) for various key products, from the calculated values—

❘ "\[LeftBracketingBar]" E emf CO ❘ "\[RightBracketingBar]" = ( 5 ⁢ e - 4 × T ) - 1 .47 ( 13 ) ❘ "\[LeftBracketingBar]" E emf HCHO ❘ "\[RightBracketingBar]" = ( 9 ⁢ e - 5 × T ) - 1 . 3 ⁢ 9 ( 14 ) ❘ "\[LeftBracketingBar]" E emf CH 3 ⁢ OH ❘ "\[RightBracketingBar]" = ( 2 ⁢ e - 4 × T ) - 1 . 2 ⁢ 8 ( 15 ) ❘ "\[LeftBracketingBar]" E emf CH 4 ❘ "\[RightBracketingBar]" = ( 5 ⁢ e - 5 × T ) - 1 . 0 ⁢ 8 ( 16 ) ❘ "\[LeftBracketingBar]" E emf C 2 ⁢ H 5 ⁢ OH ❘ "\[RightBracketingBar]" = ( 3 ⁢ e - 4 × T ) - 1 . 2 ⁢ 3 ( 17 ) ❘ "\[LeftBracketingBar]" E emf C 2 ⁢ H 4 ❘ "\[RightBracketingBar]" = ( 5 ⁢ e - 4 × T ) - 1 . 3 ⁢ 0 ( 18 )

The thermoneutral voltage (E_{thermoneutral}) is calculated from the corresponding dH values using eqn. (19).

E thermoneutral = dH nF ( 19 )

The value of n varies between 2 to 12 for the corresponding key product. The |E_emf| and E_{thermoneutral} values for reactions in Table 6 from 230K to 298K are depicted in FIG. 11. As you can see in FIG. 11, the |E_emf| and E_{thermoneutral} increased inversely to the temperature, meaning that the reactions will require more energy to proceed on Mars, which is expected due to the C_p values and Arrhenius kinetics. The effect of these anomalous C_p values at <273K was found to be more significant than any changes due to phase transitions.

It is worth noting that the cathode half-cell potentials can vary significantly between acidic and basic operating conditions of the electrolyzer while the overall cell potentials remain unaltered. The estimation of cathode half-cell potentials for each key product under acidic conditions is straightforward, as the anodic half-cell potentials for water oxidation remains same for anodic half-cell potentials for water electrolysis. However, the estimation of cathode half-cell potentials for each key product under basic conditions is quite complicated as the anodic half-cell potentials for water oxidation is not the same for anodic half-cell potentials for water electrolysis. Hence, the change of anodic half-cell potentials for water oxidation under basic conditions is determined using the below expression (73).

E T = E 298.15 + ( T - 298.15 ) ⁢ ( dE o dT ) ( 20 )

For example, the half-cell potential for water oxidation under alkaline conditions at 298.15 K is −401 mV and the value of

( dE o dT )

is −1.6816. Hence, the half-cell potential for water oxidation at 230 K is,

E 230 = E 298.15 + ( 230 - 298.15 ) ⁢ ( - 1.6816 ) = - 286.4 ⁢ mV

Similarly, the anodic-half cell potentials for water oxidation were determined between 230K to 298K and the corresponding cathodic half-cell potentials for each key product under basic conditions were calculated. The effect of temperature on cathode half-cell potentials under acidic and basic conditions are shown in FIG. 12.

Nernst Potentials for Electrolysis. The Nernst equation provides information about the equilibrium potential of the electrolyzer under open circuit conditions. The equilibrium potential is the sum of contributions from reversible electromotive force (minimum thermodynamic potential) and concentration (activity) of electroactive species in a reaction.

For example, in the case of a CO₂ electrolyzer operating at Earth's atmospheric pressure of 101325 Pa in Martian temperatures producing CH₃OH, the E_Nernst can be defined as,

E Nernst CH 3 ⁢ O ⁢ H = ❘ "\[LeftBracketingBar]" E emf CH 3 ⁢ O ⁢ H ❘ "\[RightBracketingBar]" - R ⁢ T 6 ⁢ F ⁢ ln ⁢ ( [ y CH 3 ⁢ OH ] [ y O 2 ] 3 2 [ y CO 2 ] [ y H 2 ⁢ O ] 2 ) ( 21 )

where the mole fractions of CH₃OH, O₂, CO₂ and H₂O are written as y_CH3OH, y_O2, y_CO2 and y_H2O. Since the y_CH3OH and y_H2O are ˜1 in liquid phase and the electrolyzer is operated at 1 atm, the expression is reduced to,

E Nernst CH 3 ⁢ O ⁢ H = ❘ "\[LeftBracketingBar]" E emf CH 3 ⁢ O ⁢ H ❘ "\[RightBracketingBar]" - R ⁢ T 6 ⁢ F ⁢ ln ⁢ ( 1 [ y CO 2 ] ) ( 22 )

Hence, the

E Nernst CH 3 ⁢ O ⁢ H

for a CO₂—brine electrolyzer that can synthesize CH₃OH as a key product, is determined using the thermodynamic parameters (E_emf) and the contributions from concentration (activity) of electroactive species, which depends on the extent (conversion) of the reaction. The Nernst potential expressed in terms of conversion (8) for CH₃OH production as,

E Nernst CH 3 ⁢ O ⁢ H = ❘ "\[LeftBracketingBar]" E emf CH 3 ⁢ O ⁢ H ❘ "\[RightBracketingBar]" - R ⁢ T 6 ⁢ F ⁢ ln ⁢ ( [ 3 - 0.5 ε ] [ 1 - ε ] ) ( 23 )

Similarly, the Nernst potentials expressed for CO, HCHO, CH₄, C₂H₅OH and C₂H₄ production are,

E Nernst CO = ❘ "\[LeftBracketingBar]" E emf CO ❘ "\[RightBracketingBar]" - R ⁢ T 2 ⁢ F ⁢ ln ⁢ ( [ ε ] [ 1 - ε ] ) ( 24 ) E Nernst HCHO = ❘ "\[LeftBracketingBar]" E emf HCHO ❘ "\[RightBracketingBar]" - R ⁢ T 4 ⁢ F ⁢ ln ⁢ ( [ ε ] [ 1 - ε ] ) ( 25 ) E Nernst CH 4 = ❘ "\[LeftBracketingBar]" E emf CH 4 ❘ "\[RightBracketingBar]" - R ⁢ T 8 ⁢ F ⁢ ln ⁢ ( [ ε ] [ 1 - ε ] ) ( 26 ) E Nernst C 3 ⁢ H 5 ⁢ OH = ❘ "\[LeftBracketingBar]" E emf C 2 ⁢ H 5 ⁢ O ⁢ H ❘ "\[RightBracketingBar]" - R ⁢ T 12 ⁢ F ⁢ ln ⁢ ( [ 5 - ε ] [ 2 - 2 ⁢ ε ] ) ( 27 ) E Nernst C 2 ⁢ H 4 = ❘ "\[LeftBracketingBar]" E emf C 2 ⁢ H 4 ❘ "\[RightBracketingBar]" - R ⁢ T 12 ⁢ F ⁢ ln ⁢ ( [ ε ] [ 2 - 2 ⁢ ε ] ) ( 28 )

3D plots of the Nernst potentials for various key products at different reaction extents (conversion, e) and temperatures ranging from 230K to 298K were calculated and plotted as surface plots as shown in FIG. 13 for CO₂—brine electrolysis.

Polarization models. These polarization models incorporate real-world validated operational voltages of the electrolyzers (8) (E_cell) working under Martian conditions (255K, 1 atm), and combine the thermodynamics (electromotive force (E_emf)) and the open circuit voltage known as the Nernst potential (E_Nernst), along with activation overpotentials for electrodes (anode

( η act anode ) ,

cathode

( η act cathode ) )

and transport losses ((Ohmic losses (E_IR) due to electronic and ionic resistances) to calculate more realistic theoretical production rates of key products. The polarization model is thus the sum of all these contributions as shown below—

E c ⁢ e ⁢ l ⁢ l = E Nernst + ❘ "\[LeftBracketingBar]" η act a ⁢ n ⁢ o ⁢ d ⁢ e ❘ "\[RightBracketingBar]" + ❘ "\[LeftBracketingBar]" η act cathode ❘ "\[RightBracketingBar]" + E IR ( 29 )

E_Nernst at 255 k and 1 atm is calculated from E_emf and the reaction stoichiometry using the Nernst equation (74) as detailed in the previous section.

The activation overpotentials for the electrodes are required for calculating the operating voltage (E_cell) of the electrolyzer. The activation overpotential for the anode

( η act a ⁢ n ⁢ o ⁢ d ⁢ e ) ,

an RuO₂ anode, was calculated at 255K and 1 atm in our groups previous work⁸. The activation overpotentials for the cathode

( η act cathode )

at 255K and 1 atm are calculated using the overpotential values reported in literature at room temperature for electrodes corresponding to each key product respectively. The rate constant (k) is corelated to activation energy (E_a) as (75),

k = k o ⁢ exp ⁡ ( - E a R ⁢ T ) ( 30 )

The rate constants are correlated to current as,

I = nF ⁢ AkC ⁢ and ⁢ I o = nF ⁢ Ak o ⁢ C ( 31 )

Hence, the current is related to activation energy as,

I = I o ⁢ exp ⁡ ( - E a R ⁢ T ) ( 32 )

It is very clear from the literature that, the overpotential is correlated to current (I) and exchange current density (I_o) using the Tafel expression as,

I = I o ⁢ exp ⁡ ( α ⁢ nF ⁢ η R ⁢ T ) ( 33 )

Hence, the overpotential is correlated to the activation energy (E_a) as,

E a = α ⁢ nF ⁢ η ( 34 )

Where α, n, F, η are the transfer coefficient (˜0.5), number of electrons (2-12), Faraday constant and overpotential values, respectively. Once the E_a values are calculated (for each corresponding key product) from the overpotential values reported in the literature at room temperature or vice-versa, the

I I o

values are calculated, and the corresponding E_a values are calculated for 255K which is used to estimate the corresponding overpotential values at 255K.

Having obtained the thermodynamic contributions and the overpotential losses to overall electrolyzer polarization, we turned to the Ohmic losses in the electrolyzers. The most resistive component (and hence the major contributor to Ohmic loses) is the resistance of the membrane, which is calculated using Ohm's law-E_IR=jASR_mem where, j is the current density (mA cm⁻²) and ASR_mem is the area specific resistance (ASR) of the membrane (Ω cm²). The Ohmic loss contributions are parameterized in our model of the CO₂-brine electrolyzers by assuming the conductivity of the separator membranes for the CO₂-brine electrolyzers to be equal to the conductivity of Nafion 117 at 303K. Thus, ASR_mem is estimated⁷⁶ to be ˜1.22 Ω cm². Nafion® 117 was selected for this polarization model due to the availability of data at lower temperatures and compatibility with the acidic conditions present in these electrolyzers.

Electrolyzer Performances and Production rates. These polarization models are used to express the performance of theoretical CO₂-brine electrolyzers operating at Martian conditions to synthesize various key products through ISRU to enable sustained crew travel to and from Mars. The polarization models applied to the Martian brine electrolyzer account for inefficiencies at several levels to provide realistic predictions of performances and production rates. First, a reactant conversion of 90% is assumed in the Nernst equation, reducing the operating potential of the electrolyzer.

For example, as discussed above, the Nernst potential for CH₃OH production is expressed in terms of conversion (8) as,

E N ⁢ e ⁢ rnst CH 3 ⁢ O ⁢ H = ❘ "\[LeftBracketingBar]" E e ⁢ m ⁢ f CH 3 ⁢ O ⁢ H ❘ "\[RightBracketingBar]" - R ⁢ T 6 ⁢ F ⁢ ln ⁢ ( [ 3 - 0 . 5 ⁢ ε ] [ 1 - ε ] )

The potential of

❘ "\[LeftBracketingBar]" E e ⁢ m ⁢ f C ⁢ H 3 ⁢ O ⁢ H ❘ "\[RightBracketingBar]"

is estimated using the thermodynamic parameters at 255 K is calculated to be 1.22 V. Accounting for 90% conversion, ε=0.9, the Nernst potential becomes,

E N ⁢ e ⁢ rnst C ⁢ H 3 ⁢ O ⁢ H = 1 . 2 ⁢ 2 - 8 . 3 ⁢ 1 ⁢ 4 * 2 ⁢ 5 ⁢ 5 6 * 9 ⁢ 6 ⁢ 4 ⁢ 8 ⁢ 5 ⁢ ln ⁢ ( [ 3 - ( 0 . 5 * 0 . 9 ) ] [ 1 - 0 . 9 ] ) E N ⁢ e ⁢ rnst C ⁢ H 3 ⁢ O ⁢ H = 1.21 V

Secondly, the practical open circuit potential (OCP) (which should ideally equal the Nernst potential) is assumed to be a mixed potential with a value equal to 70% of the calculated Nernst potential to account for potential contributions from unwanted side-reactions (77). Hence, the overall cell potential becomes,

E c ⁢ e ⁢ l ⁢ l = 0 . 7 ⁢ E N ⁢ e ⁢ rnst + ❘ "\[LeftBracketingBar]" η act a ⁢ n ⁢ o ⁢ d ⁢ e ❘ "\[RightBracketingBar]" + ❘ "\[LeftBracketingBar]" η act cathode ❘ "\[RightBracketingBar]" + E IR

The individual overpotential values of the anodes and cathodes in addition to the overall activation overpotentials

( η act A ⁢ n ⁢ o ⁢ d ⁢ e + η act Cathode )

for each key product is shown in Table 7. The performances of the CO₂-brine electrolyzers for various key products are shown using operating lines in FIG. 14.

The cell potential requirements for operating these electrolyzers at varying current densities (100-1500 mA cm⁻²) is captured using the operating line which accounts for all the losses of the electrolyzer operation. The production of CO consumes the maximum cell potential and the production of C₂H₄ consumes the least cell potential, with all other key products lying between CO and C₂H₄.

Formaldehyde (HCHO) is an intermediate product in the CO₂ electroreduction into CH₃OH pathway as seen in FIG. 10a due to the theory that formaldehyde desorbs off the surface of BDD due to the presence of sp³ carbon (46). There is no HCHO operating line in FIG. 14 due to the lack of activation overpotential data in their work. However, we theorize the operating line would for HCHO would reside between CO and CH₃OH because HCHO is an intermediate within the CH₃OH reaction. The production rates of the key products at various Faradaic efficiencies (25%-100%) are plotted in FIG. 15A-15E). The production rates of various key products from literature in Table 8 are plotted over the corresponding products in FIG. 15A-15E). However, formaldehyde's importance as an industrial precursor seen in Table 4 warrants further experimental analysis and catalyst tailoring and can eventually be fully incorporated into our group's modelling and analysis efforts.

Building off our groups previously validated Martian ISRU electrolyzer (6) we were able to determine the theoretical production rates of CO, CH₃OH, C₂H₄, and C₂H₅OH, and incorporate the CH₄ results from our previous Martian ISRU electrolyzer model. Additionally, we modelled the Nernst potential of HCHO due to its importance as industrial precursor and relevance as an intermediate in CO₂ electroreduction, therefore only requiring the experimental determination of the electrode overpotentials to finish modelling theoretical Martian production rates. As seen in Table 6 the production rates of the highest performing CO₂ electrocatalysts in literature can theoretically produce 0.052-1.13 g per cm² per day of CO, 0.0004-0.086 g per cm² per day of CH₃OH, 0.001-0.048 g per cm² per day of CH₄, 0.0164-0.171 g per cm² per day of C₂H₄, 0.008-0.0125 g per cm² per day of C₂H₅OH. These production numbers can increase by the time crewed Martian missions start with further improvements in catalyst activity, and selectivity towards these desired key products. Currently, the cutting-edge CO₂ electroreduction catalysts have impressive selectivity, but require an increase in activity for industrial scale terrestrial production. However, Martian ISRU electrolyzers have the advantage of direct atmospheric capture of CO₂ and a niche use case scenario due to the immense cost of transporting materials to the Martian surface. Humanity can begin to achieve sustainable crewed travel to and from Mars by exploiting the Martian environment and the cutting edge electrolyzer technology.

References The following references, to the extent that they provide exemplary procedural or other details supplementary to those set forth herein, are specifically incorporated herein by reference.

• 1. Every Mission to Mars, Ever. The Planetary Society. 5/13, 2022.
• 2. Falcon Heavy The World's Most Powerful Rocket. May 8, 2022, 2022.
• 3. Williams, Mars Fact Sheet. Accessed May 10, 2022.
• 4. Dash et al., Reviews of Modern Physics. 2006; 78(3):695-741.
• 5. Möhlmann, International Journal of Astrobiology. 2009, 9(1):45-49.
• 6. Möhlmann, Icarus. 2010/05/01/2010, 207(1):140-148.
• 7. Möhlmann and Thomsen, Icarus. 2011/03/01/2011, 212(1):123-130.
• 8. Fairén et al., Planetary and Space Science. 2009/03/01/2009, 57(3):276-287.
• 9. Gayen et al., PNAS USA 2020, 117(50):31685-31689.
• 10. Duri et al., Frontiers in Astronomy and Space Sciences 2022
• 11. Kounaves et al., Geophysical Research Letters. 2010, 37(9):1-5.
• 12. Ramirez et al., International Journal of Astrobiology 2017, 18(1):18-24.
• 13. Eichler et al., Icarus. 2021, 354
• 14. Oze et al., Soil Systems 2021, 5(3)
• 15. Davila et al., International Journal of Astrobiology 2013, 12(4):321-325.
• 16. Rennó et al., Journal of Geophysical Research. 2009, 114
• 17. Zorzano et al., Geophysical Research Letters 2009, 36(20)
• 18. Nair and Unnikrishnan, ACS Omega Apr. 28 2020, 5(16):9391-97.
• 19. Michael et al., The Mars Oxygen ISRU Experiment (MOXIE) Abstract. 2014.
• 20. Drake, Human Exploration of Mars Design Reference Architecture 5.0. Vol. 5. 2009.
• 21. Hecht, MOXIE. NASA.
• 22. Kato et al., EUCASS. 2017, 7th:12.
• 23. Sargeant et al., ACS Catalysis. 2020, 10(14):7464-74.
• 24. Liang et al., The Journal of Physical Chemistry C. 2016, 120(30):16298-304.
• 25. Ooka et al., Langmuir. Sep. 19 2017, 33(37):9307-13.
• 26. Tombari and Salvetti, Chemical physics letters 1999, (300):749-751.
• 27. Goodwin, Thermophysical properties of methane, from 90 to 500 K at pressures to 700 bar. 1974
• 28. Tables of thermal properties of gases; comprising tables of thermodynamic and transport properties of air, argon, carbon dioxide, carbon monoxide, hydrogen, nitrogen, oxygen, and steam (U.S. Dept. of Commerce, National Bureau of Standards) (1955).
• 29. O'Brien, Large Scale Hydrogen Production From Nuclear Energy Using High Temperature Electrolysis. 2010:287-308.
• 30. Jayasayee et al., Chapter 4 In: Arai, Garche, Colmenares, eds. Electrochemical Power Sources: Fundamentals, Systems, and Applications. Elsevier; 2021:47-79.
• 31. Yadav et al., Chapter 3 In: Arai, Garche, Colmenares, eds. Electrochemical Power Sources: Fundamentals, Systems, and Applications. Elsevier; 2021:23-45.
• 32. Lin et al., Angew Chem Int Ed Engl. Dec. 7 2020, 59(50):22408-413.
• 33. Pan and Barile, Energy & Environmental Science 2020, 13(10):3567-78.
• 34. Umeda et al., ACS Applied Energy Materials 2019, 3(1):1119-27.
• 35. Sedighian Rasouli et al., Chem Catalysis. 2022; 2(4):908-16.
• 36. Han et al., J Am Chem Soc. Jul. 22 2020, 142(29):12563-567.
• 37. Wang et al., J Phys Chem Lett. Sep. 3 2020, 11(17):7261-66.
• 38. Liu et al., Journal of Materials Chemistry A. 2020, 8(7):3541-62.
• 39. Kuhl et al., Energy & Environmental Science. 2012, 5(5)
• 40. Zhao et al., Journal of Materials Chemistry A. 2017; 5(38):20239-243.
• 41. Weng et al., Angewandte Chemie International Edition. 2017, 56(42):13135-139.
• 42. Wang et al., ACS Catalysis. 2018, 8(8):7113-19.
• 43. Jeon et al., Journal of the American Chemical Society. 2019, 141(50):19879-887.
• 44. Sargeant et al., ACS Catalysis, 10(14), 7464-74.

VIII. EXAMPLES

The following examples as well as the figures are included to demonstrate preferred embodiments of the invention. It should be appreciated by those of skill in the art that the techniques disclosed in the examples or figures represent techniques discovered by the inventors to function well in the practice of the invention, and thus can be considered to constitute preferred modes for its practice. However, those of skill in the art should, in light of the present disclosure, appreciate that many changes can be made in the specific embodiments which are disclosed and still obtain a like or similar result without departing from the spirit and scope of the invention.

Example 1

Simulation and Thermodynamic Modeling

The aim of in-situ resource utilization is to reduce cargo weight during takeoff while ensuring astronauts have essential resources throughout the mission. Leaving Earth's atmosphere requires considerable energy, with mass directly impacting this energy requirement. However, for long trips like those to Mars, maintaining a sufficient oxygen supply is crucial. Thus, electrolysis is used to produce oxygen and minimize spacecraft mass. Sankarasubramanian and colleagues demonstrated successful electrolysis of Martian brines under Martian conditions, aiming to enhance NASA's Mars Oxygen in situ Resource Utilization (MOXIE) while addressing safety concerns. However, the cathode's faradaic efficiency limits pure methane production, necessitating purification of the effluent stream.

To separate cathode products, cryogenic distillation columns are simulated due to their low boiling points. Distillation utilizes differences in boiling points to obtain pure fuel sources for methalox engines. Efficient column design is crucial for feed mixtures with similar boiling points and adjusting pressure can alter phase behavior. Temperature and pressure affect vapor-liquid equilibrium, where more volatile components concentrate in the vapor phase and less volatile ones in the liquid phase. Mass and heat transfer between phases are crucial for purification. As the more volatile component evaporates, it transfers from liquid to vapor phases on each tray, with heat transfer occurring similarly. Less volatile components transfer from vapor to liquid phases, absorbing enthalpy of vaporization. The liquid phase descends until the reboiler heats it, further removing volatile components. These fundamentals enable the separation of the cathodic effluent stream to purify methane for fuel.

Expanding on prior research operating a carbon dioxide-brine electrolyzer at −18° C. and 636 Pa, this study aims to assess product separation feasibility under varied pressures via simulation. The primary goal is to optimize energy efficiency while maintaining product purity.

A. Methodology

1. Simulation and Thermodynamic Modeling

A simulation was conducted using Aspen Plus to refine methane and separate other cathode products. Aspen Plus is a widely used software tool in chemical engineering for designing, optimizing, and analyzing chemical processes. It provides a simulation environment for downstream distillation columns, allowing for the modeling of steady-state columns and predicting feed behavior under different conditions. Aspen Plus offers a range of thermodynamic models to describe the behavior of chemical components and mixtures accurately. For this simulation, the Peng-Robinson equation of state model with the NRTL activity coefficient model was employed. The Peng-Robinson model is effective for non-ideal systems containing polar compounds, hydrogen bonding, and asymmetrical molecules, accurately depicting phase equilibria calculations in chemical processes, including separation and purification steps.

Initially, a downstream process was established, but after integrating process intensification techniques, energy fluctuations were minimized. Aspen Plus simulations enable engineers to intensify processes based on parameters such as economics, energy requirements, and efficiency. In the developed simulation, the downstream purification process was specifically designed to maximize tray efficiency and minimize the heat duties of the condenser and reboiler for each change in operating conditions. This facilitated the optimal design of a series of distillation columns to compare changes in operating conditions accurately.

B. Simulation Set-Up

A unified simulation design was adopted to maintain purity standards while varying the operating pressure from Martian atmospheric conditions to Earth's atmospheric pressure. This design utilized two separate cryogenic distillation columns in series to obtain pure methane. Three scenarios were tested by altering the operating pressure exiting the electrolyzer, 636 Pa, 5600 Pa, and 101325 Pa. For each of the three scenarios, the feed entered from the effluent of the electrolyzer into the first cryogenic distillation column. In FIG. 18, this is titled “FEED”. Due to nature of the boiling points, syn gas (carbon monoxide and hydrogen) remain as a vapor and separated off the top of the first column. At the bottom of the first column, the stream labeled “BOT” are the liquid bottoms that are the feed for the second cryogenic distillation column. Here methane is purified as a gas at the top of the second column and ethylene remains a liquid. In all three scenarios, this separation remains similar due to the boiling point differences remaining as an inherent property of the compounds in solution.

C. Results and Discussion

At lower pressures, to maintain the vapor-liquid equilibrium, all of the products need lower temperatures to remain a liquid. This is because the relationship between temperature and pressure. At lower pressures, lower temperatures are generally needed to maintain components in the liquid phase. For the cryogenic distillations, this dictates the amount of energy required for separation. As the operating pressure was raised to Earth's atmosphere, the energy demand for product separation decreased across both distillation columns. This trend suggests that higher pressures facilitate the condensation of the gaseous cathode products with reduced energy input. Although, the normalized power requirement initially indicates the minimal energy needed for equivalent pure methane production, a closer examination reveals potential complexities.

Referring to FIG. 19, at the lowest pressure (636 Pa) the heat duty on the condenser and reboilers for each distillation column was the highest of the three scenarios. While at the highest pressure (101325 Pa), the duty on the condensers and reboilers was the lowest. As noted, this generally occurs because the higher pressure causes the compounds to have increased liquid phase equilibrium concentration at the same temperature. This thermodynamic property dictated the optimal separation conditions.

Due to the thermodynamic constraints, increasing pressure would require less energy for the products to separate. This would allow for less energy used per kilogram of Methane produced. This is demonstrated in FIG. 20, where increasing the pressure of the cryogenic distillation columns nearly removes the need for about 500 kW of energy per kilogram of Methane. This would make it evident that operating at higher pressures during product separation would be the best course of action to reduce energy usage. However, this would affect the product composition of the feed at the cathode. Different operating conditions in the cathode would need to be tested to see how that affects the rate of Methane produced. This would change the normalized power requirement and give a clearer picture for the operating conditions of the system that would minimize the overall energy required and maximize the amount of Methane.

Variations in feed operating pressure may necessitate the electrolyzer to function at higher pressures, potentially reducing methane output at the cathode and consequently elevating the normalized power requirement. Thus, it becomes imperative to investigate the percentage of methane produced at varying operating pressures and leverage simulation to ascertain the most optimal conditions for both the electrolyzer and the separation process.

Example 2

In-Situ Resource Utilization (ISRU) Through Electrolytic Reduction of CO₂ Enables Future Crewed Missions to Mars

This study is undertaken to support long-term Mars colonization by developing ISRU technologies to produce multi-carbon products like methane, ethanol using Martian CO₂, brines, and the hydrological cycle at Average Martian temperatures (about −40° C.). Metallic copper, a unique heterogeneous catalyst, is used for converting CO₂ into hydrocarbons and oxygenates due to its ability to facilitate C—C coupling via *CO adsorption. Studies reveal that polycrystalline copper can produce over 16 CO₂ reduction products, including 12 multi-carbon (C2, C3) compounds. Copper oxides' enriched oxygen vacancy promotes CO₂ activation for improved ECR selectively C2+ products. This research focuses on synthesizing copper-based catalysts for CO₂RR products in concentrated Brine. Compare CO₂RR results with baseline KHCO₃ solution.

Objectives include the investigation of the effectiveness of Nafion®-coated copper catalysts for methane production, identification of the optimal composition of copper oxide catalysts for multi-carbon (C2+) products, focusing on ethanol production, and comparing Martian Brine CO₂RR result with baseline KHCO₃ solution.

A. Methods

In our methodology, we prepared copper-based catalysts and ran CO₂RR in Martian Brine. The first catalyst, a Nafion-coated copper foam, was optimized for methane production. Electropolishing with 70% H3PO4 and dipped with 5 wt % nafion solution and dried. The ionomer coating enhances the selectivity for C1 products while suppressing the hydrogen evolution reaction (HER). The second catalyst, electrodeposited Cu₂O on copper foam, was synthesized using a Tartaric bath to improve oxygen vacancies, which favor the formation of C2 products. This process resulted in a distinct red-colored Cu₂O layer.

Preparation of catalyst comprising Nafion® coated on copper foam. Copper was electropolished with 70% Orthophosphoric Acid (WE: Copper Foam (3*3 cm), CE:Copper Plate (2.5*3 cm), RE: Ag/AgCl, Potential: 2V, Time:10 sec). After electropolishing the copper was washed with MilliQ water and showed with airflow followed by dipping in Nafion solution (5 wt %) for 15 minutes. The dipped copper was then dried at 80° C. for 12 hours (hrs).

Preparation of electrodeposited Cu₂O on copper foam. Copper was electrodeposited using the following conditions: Cu₂O (0.3 mol L⁻¹ of Cu²⁺, 0.3 mol L⁻¹ of Tartaric acid, 1.5 mol L⁻¹ NaOH) (WE: Copper Foam (1.5*2.5 cm), CE: Pt Coil (RE: Ag/AgCl, Potential: −0.45V, pH: 13, Time:20 mins phases). The product of the electrodeposition was air dried at room temperature.

Preparation of catalyst comprising Electrodeposited bilayer CuO/Cu₂O on copper foam. Electrodeposition of Cu₂O (0.3 mol L⁻¹ of Cu²⁺, 0.3 mol L⁻¹ of Tartaric acid, 1.5 mol L⁻¹ NaOH) (WE:Copper Foam (1.5*2.5 cm),CE:Pt Coil (RE:Ag/AgCl, Potential: −0.45V, pH:13, Time:20 mins both phases).

Experimental details. Two experimental setups were employed: an undivided cell and an H-Cell, and two kind electrode setups used for CO₂ reduction and products were analyzed using techniques such as laser adsorption spectroscopy mainly for methane, HPLC, and NMR, enabling for liquid products.

B. Results

Polarization curves (FIG. 19A-19D). The polarization curves show that H-Cell response were dominated by significant ohmic loses (resistances in the order of 20-60 Ω recorded). The undivided cells show clear activation region up to 2.2V followed by transition to ohmic polarization these experiments were run in a 2 electrode system.

Initial studies. In the initial studies methane detection was performed using a Nafion-coated copper catalyst. By introducing organic compounds on the Cu surface properties such as background charge, counterions, and ion-exchange capacity were tailored to optimize the chemical microenvironment near the catalyst. This approach enhanced C1 product formation we ran CO₂RR for 30 mins by DC power supply source, increasing the voltages by 1V up to 5V. After that we characterized catalyst to see how Nafion coating is affected and see uniform coating on electrodes based on EDAX mapping. Minimal catalyst degradation was observed after CO₂RR.

To detect methane, a high-dynamic range laser absorption spectroscopy (LAS) technique was developed based on the Capillary Absorption Spectrometer concept (FIG. 20). This method uses a low-volume (1 mL) gas cell within a hollow optical fiber with a reflective inner coating, guiding a tunable laser beam to a detector. By varying the optical path-length, the technique achieved high sensitivity for small gas quantities. Measurements at low pressures (<50 Torr), relevant to the Martian atmosphere, successfully detected methane concentrations of 400 ppm. The method demonstrated robustness against interference from water vapor and CO₂, with results showing >98% methane selectivity over other reactants and intermediates. (FIG. 21A-21H) This makes it a highly sensitive and reliable technique for gas detection in bench-scale processes, ideal for ISRU applications on Mars.

Measurements. In calibration and experimental measurements, two barometrically-prepared methane (CH₄) mixtures were tested using the Capillary Absorption Spectrometer (CAS) system. The first mixture, with a low CH₄ concentration (1%), was measured at 10.6 Torr and 298 K, where four absorption features were observed (FIG. 22C). Two prominent peaks corresponded to ¹²CH₄ (abundant isotope of carbon.), while two smaller peaks were attributed to the natural less abundance isotopes of ¹³CH₄. The system demonstrated strong peak absorbance (αv≈2.1), highlighting a detection limit of ˜50 ppm with a noise floor near α≈0.01, showcasing the system's high sensitivity at low concentrations.

For the high CH₄ concentration (99%) mixture measured at 4.5 Torr and 298 K, six absorption features were identified (FIG. 22D). However, due to optically thick conditions, the peak absorbance was much lower (αv≈0.018) as the incident laser light was nearly fully absorbed along the optical path, resulting in insufficient transmitted light (It). This led to a lower signal-to-noise ratio (SNR), which must be considered in uncertainty analyses. To address this, the Interband Cascade Laser ICL modulation settings were adjusted to access a different range of spectral frequencies. Overall, the CAS system effectively detected methane across varying concentrations, demonstrating high sensitivity for low CH₄ levels and highlighting the need for optimization at higher concentrations. Successfully methane is experimentally produced in Martian brine and detected by LAS (FIG. 23A-23B).

Electrodeposited Cu₂O Catalyst. Electrodeposition was used to form a copper oxide on copper foam catalyst. The XRD data confirmed the presence of different phases of copper oxide (FIG. 24A). FESEM images clearly revealed a well-defined crystal structure, notably in the form of truncated octahedrons. The composition of the deposition bath, along with deposition time and temperature, was found to significantly influence the Cu-oxide phase and the ratio between CuO and Cu₂O (FIG. 24B-24E).

CO₂RR experiments were conducted in an undivided cell using a two-electrode setup in 2.8 M magnesium perchlorate for 30 minutes at different potentials. Due to the high concentration of the brine solution to prevent column, we employed a liquid-liquid extraction method with DCM for HPLC analysis. A calibration curve was established for ethanol concentrations ranging from 1000 ppm to 6 lakh ppm. Ethanol was detected, but the peaks overlapped with other components, making accurate quantification challenging.

Calibration Samples

HPLC Column Inputs

The Faradaic efficiency for ethanol production was calculated at different voltages: 2.5 V: 86.51%, 3.0 V: 13.27%, 3.5 V: 6.36%. The observed decrease in ethanol Faradaic efficiency with increasing voltage aligns with thermodynamic model predictions, which suggest increased production of other products at higher voltages. Given the limitations of HPLC for high-concentration brine solutions, we decided to shift to NMR analysis for more precise product detection and quantification.

Before proceeding to low-temperature experiments, we first tested the system at room temperature using a copper oxide electrode in an undivided cell with a three-electrode setup at −1.1 V vs RHE for 30 minutes. The results showed the presence of ethanol, but the ethanol triplet peak was not clearly observed, likely due to NMR noise range issues. This experiment serves as a critical first step to validate the system and optimize conditions before moving forward to low-temperature runs. Sample was prepared by mixing 2 ml of CO₂RR electrolyte with 100 μL Internal Standards (25 mM of Phenol+5 mM of DMSO) from this 250 μL and 350 μL of D₂O water, totally 600 μL. A wet suppression detection method using H₂O (90%)+D₂O (10%) with salt was used over 64 scans (FIG. 26).

In the next step, we followed existing literature and conducted CO₂RR at −18° C. in 4.4 mM magnesium perchlorate using the same potential for 30 minutes (FIG. 27). During the experiment, we observed that the current was not stable throughout, and upon analysis, the electrolyte did not form a viscous solution as expected. The NMR results confirmed the presence of ethanol with a clear triplet peak. Surprisingly, we also detected ethylene glycol and methanol, both showing singlet peaks. These promising results encouraged us to proceed further to −40° C., where 2.8M magnesium perchlorate will remain in liquid form and not freeze, allowing us to explore CO₂RR performance under even lower temperatures. Sample preparation included mixing 2 ml of CO₂RR electrolyte with 100 μL Internal Standards (25 mM of Phenol+5 mM of DMSO) from this 250 μL and 350 μL of D₂O water, total 600 μL. Detection method was wet suppression method (H₂O (90%)+D₂O (10%) with salt.

We followed the same experimental setup, potential, and electrolysis time at −40° C. (FIG. 28). We observed, the current was not stable, and the electrolyte formed a viscous solution at this temperature. Upon analyzing the CO₂RR samples, we observed that the ethanol triplet peak intensity increased fourfold compared to the results at −20° C. Additionally, the production of ethylene glycol and methanol also increased significantly. These findings highlight that at low temperatures, increased CO₂ solubility and decreased HER kinetics favor the formation of hydrocarbons, making electrochemical CO₂ reduction more efficient. Sample preparation included mixing 2 ml of CO₂RR electrolyte with 100 μL Internal Standards (25 mM of Phenol+5 mM of DMSO) from this 250 μL and 350 μL of D₂O water, totally 600 μL. Detection method was wet suppression method (H₂O (90%)+D₂O (10%) with salt).

Electrodeposited Bilayer CuO/Cu₂O on Copper Foam. A third catalyst, a bilayer CuO/Cu₂O on copper foam, was developed in a tartaric acid bath, utilizing a solution-based chemical process with anodic and cathodic polarizations, eliminating the need for thermal oxidation to create CuO. The bilayer ratio was found to significantly influence CO₂ reduction reaction (CO₂RR) selectivity, offering a tunable approach to optimizing product yields (FIG. 29A-29D). We employed two experimental setups: an undivided cell and an H-Cell, CO₂ reduction products were analyzed using techniques such as laser adsorption spectroscopy mainly for methane, HPLC, and NMR, enabling for liquid products. The bilayer ratio was found to significantly influence CO₂ reduction reaction (CO₂RR) selectivity, offering a tunable approach to optimizing product yields.

The results of the study suggest: (1) that methane was successfully produced in Martian brine using Nafion coated copper catalysts, detected via laser adsorption spectroscopy. (2) Cu₂O catalysts synthesized with desired crystal structure produced ethanol in Martian brine at −20° C. and −40° C. Additional products like ethylene glycol and methanol were detected using NMR. (3) The effect of electrolytes and subzero temperatures on product selectivity showed significant variations in activity and product outcomes.

Example 3

Viability of Climate Enhancing Resource Utilization Through Electrolytic Carbon Dioxide Valorization to Ethanol on Mars

Sustainable, long-duration space exploration necessities the decoupling of operational consumable supplies from tenuous and expensive terrestrial supply chains. Herein, we examine the viability of utilizing the Martian brines (produced by soluble salts in the regolith depressing the freezing point of water) and the carbon dioxide rich atmosphere to produce ethanol through electrolysis. Ethanol is a valuable feedstock that exists as a liquid at typical Martian conditions. It can be used both as a fuel in combustion or electrochemical (fuel cell) systems and as a feedstock for further downstream processing into valuable chemicals such as polyethylene. The development of a high throughput CO₂-to-ethanol electrolyzer also holds commercial promise for repurposing and valorizing terrestrial atmospheric CO₂. Through the application of our electrolyzer polarization model (first detailed in AIChE Journal, 69(5):e18010 (2023); open access version at arXiv:2404.00800) to a hypothetical, zero-gap electrolyzer cell and a 10-cell stack, we examined the viable operational envelope of this electrolyzer across a range of operating temperatures, pressures (and backpressures) and inlet gas humidity. A range of catalyst candidates were considered and their effect on electrolyzer performance, product purity, side-reactions and overall throughput were determined to both identify promising CO₂-to-ethanol catalyst candidates and to delineate desirable electrocatalytic parameters (overpotential, activity) to guide future catalyst development. We found that ethanol is thermodynamically favored compared to any other CO₂ reduction produced except methane. The model indicates that CO₂-to-ethanol electrolyzers are indeed viable under Martian conditions and can potentially produce ethanol with both high selectivity (>50%) and low energy consumption (˜0.05 g ethanol per watt-hour). We also report the first (preliminary) experimental results demonstrating electrochemical CO₂-to-ethanol reduction from CO₂ saturated regolithic (perchlorate) brines.

ISRU design constraints are fundamentally different from terrestrial industrial technologies due to different operating conditions (varying pressure, temperature range, gravity etc.) and resource constraints. Crucially, most industrial technology on Earth is predicated on the ready availability of combustible fuels which in turn is used to produce steam that serves as the driving/working fluid in turbines and engines. But both combustible fuels (and oxidant) and steam will be constrained in space. On the other hand, abundant electricity from solar panels or various nuclear sources makes ISRU eminently amenable to electrochemical technologies which are currently uneconomical on Earth. This has been demonstrated by NASA's Mars in-situ resource utilization experiment (MOXIE) on-board the Perseverance rover which utilizes a solid oxide electrolyzer (SOE) to produce O₂ from CO₂ (4). The 10-cell MOXIE stack produced ˜12 g/h of O₂ on Mars using atmospheric CO₂. For context, humans need ca. 11 g/h of O₂ when sleeping to ca. 85 g/h of O₂ when running on a flat surface (5). The MOXIE stack consists of two 5-cell short stacks in series with electrode area=22.7 cm² and reportedly operates at 50% Faradaic efficiency (F.E). The 300 W power envelope consists of ca 115 W for operating the stack with the rest powering the balance of plant. The stack operation in turn consumes ca. 35 W for electrolysis of CO₂ (i=4A maximum at an 8.7 V stack operating voltage) and ca. 80 W for maintaining the stack temperature at 1037 K. CO can be produced as a side-product of this system and downstream processing would be required before use of the generated O₂ in life-support applications. Another, arguably more energy efficient, electrochemical ISRU approach has been the electrolysis of Martian regolithic/perchlorate brine under ambient Martian conditions of temperature and pressure (5). It has been shown that a lab-scale 25 cm² electrolyzer operating at −36° C. can produce pure O₂ and H₂ from (unfiltered/no purification) magnesium perchlorate brines at 70% faradaic efficiency and significantly lower energy consumption. Recently, we proposed a similar approach for the production of CH₄ and O₂ at Martian ambient conditions and showed it to be viable (6,16). Notably, that study develops a general model for electrolyzers accounting for both the thermodynamics and the polarization response. The model explicitly accounts for various potential catalyst and separator combinations in an electrolyzer and is broadly generalizable.

While these prior studies have focused on the production of fuel (H₂, CH₄) and O₂ (for use both as oxidant and for life support), longer term crewed exploration of Mars and beyond will benefit from ISRU approaches to produce commodity precursor chemicals. These precursor chemicals can in turn be processed into building materials or for other uses on-site. The reduction of Martian atmospheric CO₂ has the potential to produce a variety of hydrocarbon products including carbon monoxide (CO), methane (CH₄), formaldehyde (HCHO), methanol (CH₃OH), ethylene (C₂H₄) and ethanol (C₂H₅OH)(7). Nevertheless, the electrochemical production of more complex hydrocarbons is both energy intensive (a greater number of electrons transferred) and potentially less selective due to the branching reaction pathways. Ethanol, in addition to use as a beverage and in gasoline blends, is one of the important industrial raw materials and can serve as the precursor for the production of solvents, plastics, coolants, and fertilizers/pesticides(8). The present work seeks to examine the viability of producing ethanol through Martian ISRU. It is envisioned that a zero-gap electrolyzer stack will be fed with Martian perchlorate brine and humidified atmospheric CO₂. This electrolyzer will consist of corrosion resistant flow fields sandwiching (successively) a suitably selected CO₂ reduction reaction (CO₂RR) electrocatalyst (selective to ethanol) on the cathode side, an ionomeric membrane separator and an oxygen evolution reaction (OER) electrocatalyst on the anode side. Both catalysts will be bound to their respective electrodes using ionomeric binders chosen to match the polarity of the membrane separator. Material choices will be aided by learnings such as from the recent NASA report (15) on the use of metallic flow field on the anode (copper mesh) and cathode (nickel mesh) side for water electrolyzer operating on Martian brines with end plates made up of polycarbonate plastic filled with 20% glass fillers. FIG. 31 shows a representation of the system. We are designing the system to operate on a minimally purified Martian brine feed which is expected to consist of a concentrated solution that predominantly consists of dissolved perchlorate and sulfate salts (based on the Martian regolith composition). Thus, our preliminary CO₂ reduction experiments have been carried out in 2.8M Mg(ClO₄)₂ electrolytes. The cathodic feed is expected to consist of humidified Martian atmospheric CO₂, filtered to remove insoluble regolith dust. Back pressure valves are provided in the exit stream of the electrolyzer to passively pressurize the reactor and overcome the transport losses caused by the low pressure of CO₂, thereby enabling high current operation. We anticipate minimal undesired by products at the anode (explained below) and the desired O₂ product can be cooled in a heat exchanger and stored in a cryogenic state. The cathode product distribution will be a function of the catalyst selected and can be expected to be a mix of C1 and C2 products. At present, CO₂RR catalysts that can produce ethanol with 40% to 90% selectivity have been reported (10-12). We envision first using a separating tank to separate the gaseous and liquid products. The liquid products, expected to consist of unreacted water, ethanol, and perhaps methanol can then be distilled. As seen from the isobaric vapor liquid equilibrium (VLE) diagrams for the ethanol-water system at 101325 Pa, 5098 Pa and 636 Pa (FIG. 31B, 31C, 31D) the reduced atmospheric pressure on Mars is an advantage as it pushes the constant boiling azeotrope (vapor- and liquid fractions exhibit the same composition) from ca 85% to ca >96%, enabling the production of a purer ethanol fraction as the distillate. The boiling point of pure ethanol is also significantly reduced (from ca 78° C. to ca −6° C.) at Martian pressures, thereby reducing the heating duty of the distillation column.

The half-cell reactions in the electrolyzer can vary based on the operating pH and are as follows—

Desired Electrochemical Reaction:

Low pH operation—

Anode:6H2O(l)→3O2(g)+12H++12e−(−1.23V vs. SHE at STP)  (1)

Cathode:2CO2(g)+12H++12e−→C2H5OH(l)+3H2O(0.08V vs. SHE at STP)  (2)

High pH operation—

Anode:12OH−→6H2O+12e−+3O2(g)(−0.40 V vs. SHE at STP)  (3)

Cathode:2CO2(g)+9H2O+12e−→C2H5OH(l)+12OH−(−0.75V vs. SHE at STP)  (4)

In addition, the following undesired electrochemical/chemical reactions were considered based on the chemical species present in the regolithic brine solution.

Undesired Electrochemical Reaction(s):

ClO4−(aq)+2H++2e−→ClO3−(aq)+H2O(1.19 V vs.SHE at STP)

ClO3−(aq)+2H++2e−→ClO2(aq)+H2O(1.17 V vs.SHE at STP)

ClO3−(aq)+6H++6e−→Cl—(aq)+3H2O(1.45 V vs.SHE at STP)

ClO2+e−→ClO2−(aq)+H2O(1.04 V vs.SHE at STP)

Cl2+2e−→2Cl−(aq)(1.35 V vs.SHE at STP)

Undesired Chemical Reaction(s):

ClO3−(aq)+ClO2−(aq)+2H+→2ClO2(aq)+H2O

5ClO2−(aq)+4H+−→4C102+Cl−(aq)+2H2O

The reduction potential for the ethanol formation reaction is significantly lower than these reduction potentials and thus no side-reactions are anticipated with brine components at the cathode. While the oxidation potential (same value with the opposite sign as the reduction potentials) of some of these reactions are indeed close to the anodic half-cell potential, we anticipate having mostly ClO₄₋ (and perhaps Cl⁻) in the solution and thus these side reactions would not be an issue at anode due to absence of the required reactant species or the more negative oxidation potential.

Given the highly explosive nature of alkyl perchlorates, the formation of these compounds would be a safety hazard. The following summary of alkyl perchlorate chemistry is from Schumacher's monograph on perchlorates (17). Fortunately, such compounds are typically formed by the reaction of a perchlorate salt with an alkyl halide or by reacting anhydrous perchloric acid with the alcohol. No reports exist of alkyl perchlorates forming by reaction between the alcohol and a perchlorate salt. Furthermore, ethyl perchlorate is reported to be immiscible with water and is known to slowly undergo hydrolysis in contact with water, most likely producing ethanol and perchloric acid. Given that we are operating with a perchlorate solution rather than the pure salt and have low single pass conversion to ethanol, even if ethyl perchlorate is produced by a heretofore unreported reaction, it is expected to be in small quantities and is expected to be neutralized by hydrolysis.

Turning our attention to the desired product, the production of any C₂₊ products from CO₂RR requires that the CO* intermediate go through dimerization, where a C—C bond is formed when a hydrogenated COH* intermediate binds with a CO* intermediate to form OCCOH* intermediates at the surface of the cathode. This dimerization is the most favorable pathway for the formation of C₂₊ products, and thus, utilizing catalysts that reduce the activation barrier for C—C dimerization and further C₂H₄ desorption will avoid other C₂₊ products (9). The initial intermediates have a carbon bond at the surface, however, when the dimerization binding forms OCCOH* and is further hydrogenated, the water desorbs leaving a OCC*intermediate that has only one carbon bond with the surface and an oxygen bond with the cathode surface. The highest performing C₂H₅OH formation reaction pathway is as follows (10): CO2→COOH*→CO*+COH*→OCCOH*→HOCCOH*→HOCC*+H₂O→HOCCH*→HOCHCH*→HOCHCH₂*→HOCHCH₃*→C₂H₅OH. A majority of the high performing ethanol producing CO₂RR electrocatalysts in literature utilized an acidic reaction and cation exchange membrane, compared to the dual acidic—alkaline C₂H₄ electrolyzers (9-12). Thus, the present work examines the viability of a low-pH CO₂ to C₂H₅OH electrolyzer at Martian and terrestrial ambient temperatures and under various electrolyzer pressurization conditions.

Electrolyzer Cell Performance Modeling and Analysis. The model used in this study has been detailed and validated against brine electrolysis experimental data in our prior work (6,16). The validating experimental data was from an electrolyzer where we electrolyzed perchlorate brines (2.8 M Mg(ClO₄)₂) using Pt/C cathode and RuO₂ anode at Martian temperatures to synthesize H₂ and O₂ (5,6). This electrolyzer was operated at 237 K and 101327 Pa using a commercial Fumasep FAA-3-50 anion exchange membrane. The model used ideal gas theory for the calculation of thermodynamic parameters (dS, dH, and dG) which were used in turn to calculate the operating electromotive force (E_emf) of the electrolyzer. Ideal gas theory was found to be a good approximation for behavior under the very low Martian atmospheric pressure as seen from the excellent fit between model predictions and experimental data in our previous work (6,16). The E_emf is correlated to E_Nernst by accounting for the activity contribution and the single pass conversion of CO₂. The E_Nernst was used to obtain the operation of the electrolyzer E_cell by accounting for losses due to mixed potentials and energy efficiency of the electrolyzer. The model includes single pass conversion, mixed potential losses due to side reactions, activation overpotentials of RuO₂ anode and Pt/C cathode and ohmic losses due to the membrane. This study follows a similar approach to model the CO₂-brine electrolyzer. Since perchlorate brine is acidic in nature (pH-3), Nafion® membrane has been selected as the separator in the model. Specific changes made to examine the production of ethanol as the preferred product is detailed below.

1. Thermodynamic Model

The first step in modelling these electrolyzers was to gather the thermodynamic properties (dS, dH, and dG) at each of the temperature and pressures of interest for reactions (1) and (2). We chose to model the full cell reaction rather than separately considering each half-cell reaction given the difficulty in parameterizing ionic properties at the low temperatures and pressures of interest. The difference between the half-cell potentials will equate to the full cell potential and thus both approaches to modeling the cell voltage are appropriate, the focus was on the low pH reactions given the preceding discussion around the reaction pathway and the impact of the pH on this pathway. CO₂RR to ethanol (C₂H₅OH) proceeds through a 12-electron transfer pathway. Enthalpy, entropy, and Gibbs free energy (dS, dH, and dG) are specific to each reaction based on the number of electrons in the reaction and the initial thermodynamic values. The change in enthalpy (dH) was calculated using the following relationship—

dH=H_298K+c_p×(T−T_ref)  (5)

The change in entropy (dS) was calculated using the following relation with the specific heat capacity (cp)—

dS=S_298K+c_p×ln(T/T_ref)  (6)

Having obtained dH and dS for the given reaction, the free energy change (dG) was calculated as shown below—

dG=dH−T dS  (7)

dG was in turn used to calculate the reversible electromotive force (E_emf) i.e., the minimum thermodynamic potential required for the reactions to occur, using the following equation—

ΔG=−nF|E_emf|  (8)

where the number of electrons (n) is 12 in our case and F is Faraday's constant (96,485 C mol⁻¹).

In the case of the anode, where the half-cell reaction of perchlorate brine oxidation occurs, the thermodynamic properties (dS and dH) were calculated from the specific heat capacity (C_p, J mol⁻¹K¹) expression,

C p H 2 ⁢ O = 0.44 × ( T - 222 222 ) - 2.5 + 74.3 ( 9 )

Similarly, for the cathode, where the half-cell reaction of CO₂RR results in the formation of ethanol, the thermodynamic properties (dS and dH) were calculated from the specific heat capacity (C_p, J mol⁻¹ K⁻¹) data (13) as a function of temperature (T) using the following equation,

C p C 2 ⁢ H 2 ⁢ OH = ( 0.6102 × T ) - 61.098 ( 10 )

Having obtained the thermodynamic values as detailed above, we calculated |E_emf| and E_{thermoneutral} across the temperature range (230K to 298K). We also obtained the following general empirical correlation between |E_emf| vs. temperature (T) for ethanol from the calculated values—

❘ "\[LeftBracketingBar]" E emf C 2 ⁢ H 2 ⁢ OH ❘ "\[RightBracketingBar]" = ( 3 ⁢ e - 4 × T ) - 1.23 ( 11 )

The thermoneutral voltage (E_{thermoneutral}) was calculated from the corresponding dH values using the equation below—

E thermoneutral = dH / nF ( 12 )

Both |E_emf| and E_{thermoneutral} for the production of C₂H₅OH was found to lie between values for CH₄ production and water electrolysis (6). These parameters were in turn fed into the equations for the electrolyzer equilibrium potential as detailed below. FIG. 32 depicts the variation of E_emf and E_{thermoneutral} along with the half-cell potentials of anode and cathode as a function of temperature.

2. Nernst Potential

The Nernst equation provides information about the equilibrium potential of the electrolyzer under open circuit conditions. The equilibrium voltage is the sum of contributions from reversible electromotive force (minimum thermodynamic potential) and concentration (activity) of electroactive species in a reaction.

For example, in case of water electrolyzers operating at terrestrial atmospheric pressure of 101325 Pa, the E_Nernst can be defined as,

E Nernst = ❘ "\[LeftBracketingBar]" E emf ❘ "\[RightBracketingBar]" - RT 2 ⁢ F ⁢ ln ⁢ ( [ y H 2 ] [ y O 2 ] ? [ y H 2 ⁢ O ] ) ( 13 ) ? indicates text missing or illegible when filed

where the mole fractions of H₂, O₂ and H₂O are written as y_H2, y_O2 and y_H2O. Since the y_H2O is ˜1 in liquid phase, the expression reduced to,

E Nernst = ❘ "\[LeftBracketingBar]" E emf ❘ "\[RightBracketingBar]" - RT 2 ⁢ F ⁢ ln ⁢ ( [ y H 2 ] [ y O 2 ] ? ) ( 14 ) ? indicates text missing or illegible when filed

While temperature variations and effects are accounted for in calculating |E_emf| and also occurs in the prefactor to the second term on the right, the Nernst potentials incorporate pressure variation using another term. For example, for water electrolysis at Martian surface pressure (636 Pa), we would incorporate the additional term, P_Mars/P_Std, as shown below,

E Nernst = ❘ "\[LeftBracketingBar]" E emf ❘ "\[RightBracketingBar]" - RT 2 ⁢ F ? ( [ y H 2 ] [ y O 2 ] ? × [ P Mars P Std ] ? ) ( 15 ) ? indicates text missing or illegible when filed

Following the above discussion, the

E Nernst C 2 ⁢ H 5 ⁢ OH

for a CO₂— perchlorate brine electrolyzer that can synthesize C₂H₅OH as the cathode product, is determined using the thermodynamic parameters (E_emf) and the contributions from concentration (activity) of electroactive species, which in turn depends on the extent (single pass conversion) of the reaction.

E Nernst C 2 ⁢ H 5 ⁢ OH

was determined for electrolyzer operations at 255K and 298K and at pressures of 636 Pa, 5098 Pa and 101325 Pa.

The Nernst potential expressed in terms of single pass conversion (E) for C₂H₅OH production at 101325 Pa and 255K or 298K accounts for the liquid state of the ethanol product and is as follows—

E Nernst C 2 ⁢ H 5 ⁢ OH = ❘ "\[LeftBracketingBar]" E emf C 2 ⁢ H 5 ⁢ OH ❘ "\[RightBracketingBar]" - RT 12 ⁢ F ⁢ ln ⁢ ( 1 ? ? ) ( 16 ) ? indicates text missing or illegible when filed

Accounting for the vapor phase production of ethanol at pressures of 5098 Pa and 636 Pa (at temperatures of 255K or 298K), the Nernst potential is—

E Nernst C 2 ⁢ H 5 ⁢ OH = ❘ "\[LeftBracketingBar]" E emf C 2 ⁢ H 5 ⁢ OH ❘ "\[RightBracketingBar]" - RT 12 ⁢ F ⁢ ( ln ⁢ ( ? ? ? ? ) ? ( P Mars P Earth ) - 1 ) ( 17 ) ? indicates text missing or illegible when filed

FIG. 33 depicts surface plots of the Nernst potentials as a function of the single pass conversion (e) and temperature. Both pressure and temperature affect the Nernst potential, but the effect appears to be so small that we recommend choosing operating pressures and temperatures on the basis of economics and safety rather than electrolyzer performance.

3. Polarization Model

The polarization model combines the thermodynamic potentials calculated above with experimentally determined activation overpotentials for both electrodes (anode (η anode/act), cathode (η cathode/act)) and transport losses (Ohmic losses (E_IR) due to electronic and ionic resistances) to predict the polarization performance of a real electrolyzer at given conditions of temperature and pressure. The polarization model is thus the sum of all these contributions as shown below—

E cell = E Nernst + ❘ "\[LeftBracketingBar]" η act anode ❘ "\[RightBracketingBar]" + ❘ "\[LeftBracketingBar]" η act cathode ❘ "\[RightBracketingBar]" + E IR ( 18 )

The activation overpotentials for the electrodes are required for calculating the operating voltage (E_cell) of the CO₂— perchlorate brine electrolyzer. The activation overpotential for the anode (η anode/act), an RuO₂ anode, was obtained from our previous work (5). RuO₂ was used as anode for electrolyzer modeling due to its degradation resistance, compatibility under acidic conditions and due to availability of overpotential data at operating conditions of the electrolyzer, that is, 255 K and in a Mg(ClO₄)₂ electrolyte as discussed in previous study. The activation overpotentials for the cathode (η cathode/act) was obtained from the literature with the catalysts being selected on the basis of high activity and selectivity for ethanol.10-12 The rate constant (k) is corelated to activation energy (E_a) as follows—(14)

k = k o ⁢ exp ⁢ ( - E a RT ) ( 19 )

The rate constants are correlated to current by,

I = nFAkC ; I o = nFAk o ⁢ C ( 20 )

Hence, the current is related to activation energy by,

I = I o ⁢ exp ⁢ ( - E a RT ) ( 21 )

In turn, the overpotential is correlated to current (I) and exchange current density (Io) by the Tafel expression—

I = I o ⁢ exp ⁢ ( α ⁢ nF ⁢ η RT ) ( 22 )

Hence, the overpotential is correlated to the activation energy (Ea) by,

E a = α ⁢ nF ⁢ η ( 23 )

Where a, n, F, η are the transfer coefficient (˜0.5), number of electrons (12), Faraday constant and overpotential values respectively. Once the E_a values are calculated from the overpotential values reported in the literature at room temperature or vice-versa, the I/I_o values are calculated, and the corresponding E_a values are used to calculate the corresponding overpotential values at the temperature of interest. Having obtained the thermodynamic contributions and the overpotential losses to overall electrolyzer polarization, we turned to the Ohmic losses in the electrolyzers. The most resistive component (and hence the major contributor to Ohmic loses) is the resistance of the membrane, which was calculated using Ohm's law—

E IR = jASR mem ( 24 )

where, j is the current density (mA cm⁻²) and ASR_mem is the area specific resistance (ASR) of the membrane (Ω cm²). The Ohmic loss contributions are parameterized in our model of the CO₂-perchlorate brine electrolyzers by assuming the conductivity of the separator membranes for the CO2-perchlorate brine electrolyzers to be equal to the conductivity of Nafion® 117 at 303K. Thus, ASR_mem is estimated to be ˜1.22 Ω cm². Nafion® 117 was selected for this polarization model due to the availability of data at lower temperatures and compatibility with the acidic conditions present in these electrolyzers. The overall cell performance is shown in FIG. 32. Note that this cell voltage includes two inefficiency factors including an assumption of single pass conversion of 10% and a mixed potential at the cathode reducing the Nernst potential by 30%. Notably, changing the temperature from 255K to 298K resulted in a ca 400 mV reduction in electrolysis overpotential and is expected to result in a significantly more energy efficient electrolyzer. On the other hand, increasing the pressure from 636 Pa to 101325 Pa had negligible effect on the overpotential. Note that increasing the pressure is a passive process achieved by throttling the exit streams from the electrolyzer whereas operating at higher temperatures requires active heating.

Thus, the rate of heat dissipation is critical to determining the optimal operating conditions for the ethanol producing electrolyzer. The next section examines the performance of a realistic electrolyzer stack by extending the cell level model to a 10-cell stack and adding in operational inefficiencies.

Model Prediction of Electrolyzer Stack Performance. This section extends the single cell level model to a hypothetical 10-cell electrolyzer stack with 100 cm² electrochemically active area per cell. The polarization model detailed above applied to the Martian perchlorate brine electrolyzer account for inefficiencies at several levels to provide a realistic prediction of performance. First, a single pass conversion of 10% is assumed in the Nernst equation, reducing the operating potential of the electrolyzer.

Secondly, the practical open circuit potential (OCP) (which should ideally equal the Nernst potential) is assumed to be a mixed potential with a value equal to 70% of the calculated Nernst potential to account for potential contributions from unwanted side-reactions—

E cell = 0.7 E Nernst + ❘ "\[LeftBracketingBar]" η act anode ❘ "\[RightBracketingBar]" + ❘ "\[LeftBracketingBar]" η act cathode ❘ "\[RightBracketingBar]" + E IR ( 25 )

The current, voltage and this power were calculated similar to the case of the single cell but taking care to appropriately scale them to 10 cells in series.

FIG. 35 and FIG. 36 depict the ethanol production rate at 255K and 298K respectively. Ethanol production is depicted with variation in energy efficiency of the electrolyzers. The drop off in ethanol production shows that there is an optimal production rate that balances power consumption and output. The minimal effect of internal electrolyzer pressure changes is readily apparent in FIG. 34. Going from 636 Pa to 101325 Pa (a 160-fold increase) saw negligible changes (<0.05 g·W⁻¹·day¹) in the production rate. On the other hand, increasing the temperature to 298K saw the production rate increase by about 0.8 g·W⁻¹·day⁻¹ when operating at 101325 Pa, once again emphasizing the effect of temperature. Overall, a MOXIE like 300 W system can be conservatively expected to produce more than a liter of ethanol per day at 298K and 101625 Pa at 90% faradaic efficiency (note that copper oxide catalysts show >90% F.E).

Preliminary Experimental Results. We have demonstrated CO₂ reduction to ethanol in a perchlorate brine electrolyte in a H-cell configuration as detailed below. These tests were carried out at room temperature and efforts are ongoing to repeat them at −36° C. The electrochemical cell consisted of copper foam coated with Nafion® as cathode and graphite felt as anode in in a 2.8 M Mg(ClO₄)₂ perchlorate brine electrolyte. The two compartments of the H-cell were separated by a Nafion® 117 membrane separator.

The cathode was prepared by treating pure copper foam with 85 vol % orthophosphoric acid for 10 seconds at 2V (with copper plate acting as counter electrode) to get rid of the oxide coating. The acid treated copper foam was coated with Nafion® solution (5 wt. %) to prevent the Cu phases from oxidizing. FIG. 37A shows the x-ray diffraction (XRD) of pure copper foam (CF) along with Nafion® coated CF (before testing) and H-cell tested Nafion® coated CF (after testing). The XRD shows the stability of Nafion® coated CF with Cu and CuO phases being largely unchanged before and after testing in 2.8 M Mg(ClO₄)₂ perchlorate brine electrolyte. Scanning electron microscopy imaging of the Nafion® coated CF (before testing) is shown in FIG. 37B. The presence of Nafion® coating was confirmed through elemental mapping of individual elements in FIG. 37C-37H wherein the presence of carbon (C), fluorine (F), sulfur (S) indicated the presence of Nafion® with the phosphorous (P) coming from residual acid. Copper (Cu) and oxygen (O) were observed as expected from the base metal.

FIG. 38A shows the H-cell configuration used for synthesis of ethanol. The H-cell was filled with 2.8 M Mg(ClO₄)₂ perchlorate brine electrolyte on both sides, separated by a Nafion® membrane. CO₂ was purged for 20 minutes to attain equilibrium in the catholyte. A potential of 3V to 5V (accounting for significant ohmic losses) was applied for a time interval of 10 minutes each with CO₂ being continuously purged into the catholyte at room temperature. The active area of the electrodes (cathode and anode) are 32 cm² and a current of 30 mA and 62 mA were measured at the corresponding potentials. The product(s) synthesized in the 2.8 M Mg(ClO₄)₂ perchlorate brine electrolyte was separated using liquid-liquid extraction by exchanging with a suitable solvent i.e., dichloromethane. This exchange allows us to use a variety of analytical techniques for product quantification without the fear of damaging the instrument by sampling a concentrated brine solution. The product(s) exchanged dichloromethane was analyzed using high-performance liquid chromatography (HPLC) fitted with a Poroshell 120 EC-C18 column with a multi-wavelength detector (192 nm) and with water as the carrier. The results of the HPLC analysis are shown in FIG. 38B. The successful synthesis of ethanol was confirmed by comparing the peaks in the electrolyte sample with peaks from a pure ethanol sample. The partition coefficient of ethanol between the 2.8 M Mg(ClO₄)₂ perchlorate brine electrolyte and dichloromethane is being quantified and this will enable us to determine the exact amount ethanol produced for a given current input (i.e., Faradaic efficiency).

Conclusions. A comprehensive electrochemical model accounting for thermodynamic, kinetic, transport factors and operational inefficiencies has been presented and applied to the analysis of a CO₂-perchlorate brine electrolyzer intended to produce ethanol. Ethanol has been shown to be a viable electrolytic ISRU target despite requiring a 12-electron CO₂ reduction process and is found to be thermodynamically more favored than water electrolysis. The choice of operating conditions (passive pressurization vs. active heating) and the balance between production rate and power consumption are open questions that this model is well positioned to address. A first estimate of the performance of a modest 10-cell system indicates that ethanol production under Mars like conditions is viable even with catalysts available at present. We have also presented the first demonstration of ethanol production in perchlorate brines, validating our model prediction that it is viable to produce ethanol in these brine electrolytes.

REFERENCES

• 1 The Planetary Society. URL www.planetary.org/space-missions/every-mars-mission, 2022.
• 2 NASA, URL www.nasa.gov/sites/default/files/atoms/files/moon-investments-prepare-us-for-mars.pdf, 2022.
• 3 Reuters, URL www.reuters.com/business/aerospace-defense/china-plans-its-first-crewed-mission-mars-2033-2021-06-24/, 2021.
• 4 Hecht et al., Space Sci Rev, 217(9), 2021, pp. 1-76
• 5 Gayen et al., PNAS, 117(50), 2020, 31685-89
• 6 Shahid et al., AIChE Journal, 69(5), 2022, pp. e18010.
• 7 Wang et al., Adv. Mater., 33(46), 2021, 2005798.
• 8 Dagle et al., Ind. Eng. Chem. Res. 59(11), 2020, 4843-53.
• 9 Liu et al., Nature Communications, 13(1), 2022, 1877.
• 10 Shang et al., Small Methods, 6(2), 2022, 2101334.
• 11 Yuan et al., Catalysts, 7(7), 2017, 220.
• 12 Yadav et al., Energy & Fuels, 29(10), 2015, 6670-77.
• 13 Dillon and Penoncello, Inter. Journal of Thermophysics, 25(2), 2004, 321-35.
• 14 Bard and Faulkner, Electrochemical Methods: Fundamentals and Applications; Wiley, 2000.
• 15 H₂O Electrolysis with Impure Water Source Final—Report, NASA/TM-20230003630, 2023.
• 16 Shahid et al., arXiv:2404.00800 [physics.chem-ph](2024)
• 17 Schumacher, “Perchlorates”, ACS Monograph Series, Reinhold Publishing Corp, New York (1960)

All of the devices, compounds, material, compositions, and methods disclosed and claimed herein can be made and executed without undue experimentation in light of the present disclosure. While the disclosure may have focused on several embodiments or may have been described in terms of preferred embodiments, it will be apparent to those of skill in the art that variations and modifications may be applied to the compounds, compositions, and methods without departing from the spirit, scope, and concept of the invention. All variations and modifications apparent to those skilled in the art are deemed to be within the spirit, scope, and concept of the invention as defined by the appended claims.