Electrochemical synthesis of direct reduced metal and metal-derived compounds

Abstract

An electrowinning process for reduction of metal from a metal-containing ore in an electrochemical cell comprising oppositely disposed electrodes immersed in a water-in-salt electrolyte can include dissolving the metal-containing ore in the water-in-salt electrolyte under acidic conditions to form a solution comprising one or more metal cations from the metal-containing order; and applying a potential to selectively deposit one or more of the metal cations from the solution onto one of the electrodes, thereby growing a high-purity metal plate on the electrode.

Description

Docket No. Docket No.30275 / 59668 UM 2024-153-02 ELECTROCHEMICAL SYNTHESIS OF DIRECT REDUCED METAL AND METAL- DERIVED COMPOUNDS CROSS-REFERENCE TO RELATED APPLICATION

[0001] The benefit of priority of U.S. Provisional Patent Application No.63 / 543,011 filed October 6, 2023, is hereby claimed and the disclosure is incorporated herein by reference in its entirety. FIELD

[0002] The disclosure relates to electrochemical method for reduction of dissolved metal ore to a component metal. BACKGROUND

[0003] Direct reduced iron (DRI) is a type of iron produced using a process that involves the reduction of iron ore in the presence of a reducing agent such as natural gas or coal. DRI is conventionally produced using solid-state reduction processes that do not require coke or other carbon-based fuels. The process generally starts with the reduction of iron ore to a solid sponge-like form, which is then compacted into briquettes or pellets that can be easily transported and stored. Carbothermic ore reduction in a blast furnace is a common process used. This process results in released CO2.

[0004] Steel production is a significant source of greenhouse gas emissions globally. According to the International Energy Agency (IEA), the iron and steel sector accounted for approximately 7% of total global CO2emissions in 2020. Iron makes up about 95% of steel, and iron-creation contributes to 90% of emissions associated with steel production. Addressing the decarbonization of steel productions requires a sustainable way to produce iron from iron ore. SUMMARY

[0005] Electrowinning processes known in the art have not been successfully applied to the production of directed reduced iron. Efficiencies of acidic electrowinning are poor due to parasitic hydrogen evolution reaction (HER), limiting the usefulness of electrowinning to alkaline electrowinning.

[0006] Acidic conditions in an electrowinning process would allow for electrodeposition of iron at smaller potentials, leading to higher energy efficiency.

[0007] Methods of the disclosure provide a direct electrochemical process to reduce dissolved metal-containing ore to a metal through a combination of electrode potential control and suppression of hydrogen evolution using high-concentrated electrolytes and additives. In particular, methods of the disclosure utilize water-in-salt electrolytes toDocket No. Docket No.30275 / 59668 UM 2024-153-02 suppress HER through kinetically locking in water molecules within the solvation shell. Methods of the disclosure, thereby allow for acidic electrowinning to be performed with significant improvement in efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] Figure 1: illustrates a workflow for a method in accordance with the disclosure, characterization of the resulting reduced metal, and considerations for scale-up of the method of the disclosure to an industrial scale of > 1kg / hr for 100 hours.

[0009] Figure 2: a) Two surface configurations of H at 1 / 2 ML coverage (top: 2×1; bottom (2×2)-2H) b) Configurations of O at 1 / 6 ML coverage using an orthogonal (top) and non- orthogonal (bottom) unit cell c) Configurations of OH at 1 / 3 ML using an orthogonal (top) and non-orthogonal (bottom) unit cell.

[0010] Figure 3: Mixed OH-H phases formed via chemical dissociation of water on the surface. Two mixed phases at 1 / 2 ML total coverage are shown in a) 2×1 and b) (2×2)-2H configurations. c) mixed phase at 1 ML total coverage.

[0011] Figure 4: Surface Pourbaix slices for pH values of a) 0.0, b) 7.0, and c) 14.0. A sampling of the most important surface phases is shown for clarity. Within the acidic and neutral regimes, the 1 H phase and dissolution are dominant. Under alkaline conditions O phases emerge that can prevent dissolution.

[0012] Figure 5: Full surface Pourbaix diagram.

[0013] Figure 6: Surface phases for adsorbates at coverages 1 / 4 ML, 1 ML and 1 / 2 ML.

[0014] Figure 7: Surface phases for adsorbates at coverages 1 / 6 ML, 1 / 3 ML as well as mixed OH + H phases.

[0015] Figure 8: Free energy diagram for sequential electrodeposition of iron atoms on an Fe110 terrace. Light grey atoms in the atomic images highlight the iron atoms deposited so far. The supercells have been repeated to visualize the overarching patterns.

[0016] Figure 9: Free energy diagram for sequential electrodeposition of iron atoms at an Fe210 step. Light grey atoms in the atomic images highlight the iron atoms deposited so far. The supercells have been repeated to visualize the overarching patterns.

[0017] Figure 10: Schematics of the absorption structures with hydrogen placed a) at terrace surface hollow site (top and side views), b) below terrace short bridge site (top and side views), c) hollow site near step edge, d) hollow site adjacent to step, e) below short bridge site near step edge. Outermost surface iron atoms are darkest grey, subsurface ironDocket No. Docket No.30275 / 59668 UM 2024-153-02 atoms are grey, and hydrogen atoms have a thick outline. f) Summary of calculated absorption enthalpies for all paths considered, shaded-coded by mechanism.

[0018] Figure 11: Free energy diagrams for HER on the terrace sites (1 / 4 ML and 1 ML coverage), beside the step, near the step edge, and on a hydrogenated step. DETAILED DESCRIPTION

[0019] An electrowinning process for reduction of metal from a metal-containing ore in accordance with the disclosure can be performed in an electrochemical cell comprising oppositely disposed electrodes immersed in a water-in-salt electrolyte. The process (Fig.1) includes dissolving the metal-containing ore in the water-in-salt electrolyte under acidic conditions to form a solution comprising one or more metal cations from the metal-containing ore. It has been observed that the use of water-in-salt electrolytes is advantageous for suppressing hydrogen evolution reaction (HER). The process then includes applying a potential to selectively deposit one or more of the metal cations from the solution onto one of the electrodes, thereby growing a high-purity metal plate on the electrode. The growth electrode can be, for example, the cathode. A highly pure metal plate can be grown on the cathode, while oxygen gas evolution reaction (OER) occurs at the anode.

[0020] The acidic conditions can include a pH of the electrolyte being -2 to 7, for example, -2 to 5.

[0021] Methods of the disclosure can provide for extraction of pure metal from a metal- containing ore in a room temperature aqueous acidic electrowinning process, thereby avoiding the problem of prior known processes in requiring high temperature and coke resulting in CO2formation. In the methods of the disclosure, it was observed that the use of water-in-salt electrolyte suppressed hydrogen evolution reaction (HER) and enhanced faradaic efficiency, and high exchange current densities were needed to achieve sufficient metal-salt dissolution.

[0022] The metal to be extracted from the ore can be any one or more of iron, tin, nickel, cobalt, cadmium, chromium, zinc, manganese, and titanium. Selective deposition of metal species from the dissolved ore can be accomplished through tuning of the potential applied. The potential can be selected from the Pourbaix diagram of the metal by a region where a pure metal phase exists within the acidic pH (i.e., pH <7) where HER is suppressed, as indicated by the HER line in the Pourbaix diagram. For example, the potential can be applied in a range of -0.1 V vs standard hydrogen electrode (SHE) to -1.0 V vs SHE. Such potential can be useful when the metal is iron, for example. For reduction of Ti from the feedstock, the potential can be -1.4 V vs SHE to -2.3 V vs SHE and pH can be in the range -Docket No. Docket No.30275 / 59668 UM 2024-153-02

[0023] The electrode onto which the metal is electrodeposited can include or can be entirely formed of the same metal to be deposited. For example, for deposition of iron from an iron-containing ore, the electrode can be formed of iron. The electrode onto which the metal is electrodeposited can be surface modified. For example, surface modification with iron-alloys (e.g., Fe-Ga, Fe-Si, Fe-Al, etc) can be used to allow for favoring of electrodeposition over hydrogen evolution or metal oxidation. The electrode for electrodeposition can be selected such that it has good nucleation characteristics or metal electrodeposition along with a low surface diffusion barrier. Some examples are Mn, Co, and other bimetallic alloys of ferrous metals.

[0024] The water-in-salt electrolyte comprises a salt concentration of at least 4 M. The water-in-salt electrolyte can include, for example, a polar solvent with a high donor number (e.g., a solvent with a Gutman donor number >15). For example, the water-in-salt electrolyte can include iron bis(trifluoromethanesulfonyl)imide (FeTFSI) in water. For example, the water-in-salt electrolyte comprises a eutectic salt. For example, the eutectic salt can include FeTFSI and one or more cations, such as tetramethyl ammonium and tetramethylethyl ammonium. For example, the eutectic salt comprises a single cation and multiple anions. In such a eutectic salt, for example, the cation can be iron and the multiple anions comprise TFSI and nitrate. Suitable anions can include anions with small donor number (e.g., anions with a Gutman donor number <40).

[0025] It was shown that greater than 1 M of iron sulfate is soluble in a water-in-salt formulation having greater than 5 M non-iron salt concentration. This indicates that high iron-salt solubility can be maintained in the process of the disclosure. The water-in-salt electrolyte kinetically blocks protons / water molecules from reaching the cathode / anode, thereby suppressing both HER and OER. It has been further demonstrated using a robotic test stand, Otto, that OER overpotentials can be turned while maintaining HER overpotentials.

[0026] Methods of the disclosure can be used with various grades of ore feedstock, thereby expanding the direct reduction of metals from ore to lower-grade feedstocks. For example, for iron-based ore, taconite can be used as a feedstock, in addition to higher grade feedstocks used in conventional processes such as hematite and magnetite. Typical contaminants that may still be present after salt extraction possess more negative reduction potentials, indicating they will not be plated (Fe2+ / Fe: -0.45 V, Al3+: -1.68 V, Mg2+: -2.37 V, Ca2+: -2.87 V, K+: -2.93 V) during the electrodeposition steps. This is also true for the salt cations present in water-in-salt electrolyte (Na+:-2.71, Li+:-3.04 etc).Docket No. Docket No.30275 / 59668 UM 2024-153-02

[0027] The Nernst equation can be used to estimate energy intensity of electrowinning step. Using n=2 electron process even with a cell voltage of ~3V, the process of the disclosure is estimated to be capable of achieving 10.08 GJ / t of Fe produced. Process and product scalability potential can be obtained also through electrowinning of metals such as Cu and Zn.

[0028] A thermodynamic analysis of the Fe110 aqueous electrochemical interface was conducted and used to develop an acidic electrowinning scheme for use in the processes of the disclosure. Surface Pourbaix diagrams were used to investigate the thermodynamically most favorable surface phases of O*, OH*, and H*on Fe110 as a function of pH and USHE. Interestingly, the results showed that there does not exist a region in Pourbaix space where Fe110 is thermodynamically predicted to be both stable towards dissolution and free from any adsorbates. As HER suppression is required to access acidic conditions with selective electrodeposition, this analysis was extended to calculate a surface Pourbaix diagram without hydrogenation phases formed via a Volmer process. Through this suppression, a prevalence for low-coverage surface phases emerged, which could be preferable for electroplating. In particular, a low-coverage oxygen phase was identified in a region of both weak acidity and small applied potential. This analysis paves the pathway for potential routes to electrowinning under acidic conditions.

[0029] To facilitate the thermodynamic analysis presented here, spin-polarized density functional theory (DFT) calculations were conducted using GPAW via the Atomic Simulation Environment package. Projector augmented wave potentials were used along with the RPBE exchange-correlation functional. AutoCat was used to generate all initial structures. Surface coverage was incorporated via a combination of periodic supercell dimensions and explicit inclusion of the adsorbates.

[0030] Modeling this aqueous environment, the following water oxidation reactions can occur on the surface forming either O*or OH*: H2O(l) + * → OH*+ H+(aq) + e- (1) H2O(l)+ * → O*+ 2H+(aq)+ 2e- (2)

[0031] where * indicates an unoccupied surface site and X*is an X molecule adsorbed on the surface.

[0032] Additionally, the formation of H*species on the surface can occur through a Volmer process: H+(aq)+ e- + * → H∗(3)Docket No. Docket No.30275 / 59668 UM 2024-153-02

[0033] In predicting the most thermodynamically favorable surface phases of Fe110, the energetics of the above processes are key. Calculation of these free energies were done by invoking the computational hydrogen electrode (CHE). This approach sets the reference potential to be that for which the reaction 1 / 2H2(g) ⇀↽ H+(aq) + e- is in equilibrium. From this choice, surface phase formation energies as a function of pH and USHE can be extracted, taking into account the number of consumed or produced electrons and protons. For example, the free energy change upon formation of an O* phase formed via Eq.2 can be expressed as: ∆GO∗(pH, USHE) = ∆G0O∗ + 2kBT ln 10 pH − 2eUSHE(4) where kB is the Boltzmann constant, e fundamental charge, and ∆G0O∗ is the energy upon formation at 0 V (vs SHE) and pH = 0. ∆G0O∗ is obtained directly from DFT, and similar energy expressions can be derived for surface phases involving H* and OH*. Moreover, to allow for comparisons across monolayer (ML) coverages, these calculated energies are subsequently normalized to be per surface atom.

[0034] These energies are not only governed by the strength of adsorbate interactions with Fe110 relative to their reference phases, but also adsorbate coverage and patterns on the surface. At higher coverages steric effects can become more pronounced, and at a given coverage there may exist multiple configurations.

[0035] In addition to the surface phases, stability of the Fe110 surface is also important and the following iron dissolution reaction was considered: Fe(s) → Fe2+(aq)+ 2e- (5)

[0036] Following a similar logic as for the adsorbate phases, the energy change upon dissolution at aFe2+= 1 can be expressed as:

[0037] where is the redox potential for this process at standard conditions (-0.447 V vs SHE). It is worth noting that as this process does not produce or consume H+, it is independent of the pH of the environment. Dissolution of the iron surface forming HFeO2–is also possible, but we assume this will have a considerable kinetic barrier and thus exclude it from our analysis.Docket No. Docket No.30275 / 59668 UM 2024-153-02 Surface Phases of Fe110

[0038] Identifying the most favorable adsorption sites for O*, H*, and OH*, adsorption energies were calculated at 1 / 4 ML coverage by relaxing the adsorbates at all unique surface sites on the 110 facet: hollow, long bridge, short bridge, and on top. These calculated adsorption energy values are given in Table 1. It was observed that for H* at this coverage, the hollow site yielded the most negative adsorption energy. Much of the previous literature, both computational and experimental, identifies either the hollow or long bridge as the most favorable site. However, we highlight the small distance distinguishing these two sites. Similarly, OH* and O* were also observed to prefer the hollow site, and for all other coverages we assume this site will be most favorable. Table 1: Adsorption energies of H, O, and OH at 1 / 4 ML coverage on Fe110 ∆G0(eV) adsorbate hollow on top long bridge short bridge H -0.60 0.40 -0.29 -0.56 O -0.99 0.82 -0.67 -0.52 OH-0.590.31 -0.30 -0.39

[0039] To develop as complete a picture of the surface as possible, several different surface coverages and phases were modeled.

[0040] At 1 / 2 ML coverage for all three adsorbates two phases were considered, as described by Jiang and Carter: (2×1) and (2×2)-2H. The former has a rectangular pattern on the surface whereas the latter forms a hexagonal pattern, as shown in Fig.2A H* shows a slight preference towards the hexagonal configuration at 1 / 2 ML coverage, in agreement with previous work. OH* and O* also show a slight preference towards the hexagonal phase at this coverage. However, on a per-atom basis, this preference is at most a difference of 30 meV for the three adsorbates.

[0041] At 1 / 6 ML and 1 / 3 ML coverages, two different configurations were considered for each. One is formed by using an orthogonal cell where the other uses a conventional non- orthogonal cell in the xy plane (Fig.2B-2C). H* shows a preference towards the orthogonal configuration at both 1 / 6 ML and 1 / 3 ML coverage, but by only ∼5 meV / surface atom. The differing behavior of H* may be rationalized by decreased steric effects, asDocket No. Docket No.30275 / 59668 UM 2024-153-02 evidenced by the vertical OH* orientations on the surface. Both OH* and O* prefer the pattern formed via the non-orthogonal cell at these coverages.

[0042] In addition to these single adsorbate phases, mixed OH* + H* phases were also considered, as they can be formed via the chemical dissociation of water. At a total of 1 / 2 ML, we model both patterns in the mixed phase at this coverage with 1 / 4 ML OH* + 1 / 4 ML H* (Fig.3A, 3B). Again, minimal energetic difference was observed, with a marginal preference towards the hexagonal phase of ∼5 meV / surface atom. The higher coverage mixed phase 1 / 2 ML OH* + 1 / 2 ML H* was also modeled (Fig.3C). Interestingly this phase was predicted to be thermodynamically stable, unlike the 1 ML O* and 1 ML OH* phases, possibly through the OH-H hydrogen bonding as evidenced from the induced tilt of the OH moieties. Visualizations of all surface phases considered are aggregated in Fig.6, 7 with their corresponding energies provided in Table 2 below:Docket No. Docket No.30275 / 59668 UM 2024-153-02 Table 2: Energies of all surface phases in this work. ∆G0is the total energy of the phase, ne is the number of electrons produced upon formation, nH+is the number of protons produced upon formation, area is the number of surface atoms for that phase. phase∆G0(eV)nenH+area Fe110 0.0 0 0 Fe2+-0.89 2 0 1 / 4 H -0.60 -1 -1 4 1 / 4 O -0.99 2 2 4 1 / 4 OH -0.59 1 1 4 1 H -0.28 -1 -1 1 1 O 0.52 2 2 1 1 OH 0.55 1 1 1 1 / 2 H (2×1) -1.24 -1 -1 4 1 / 2 H (2×2)-2H -1.25 -1 -1 4 1 / 2 O (2×1) -0.81 2 2 4 1 / 2 O (2×2)-2H -0.93 2 2 4 1 / 2 OH (2×1) -0.38 1 1 4 1 / 2 OH (2×2)-2H -0.48 1 1 4 1 / 6 H (orthogonal) -0.33 -1 -1 6 1 / 6 H -0.33 -1 -1 6 1 / 6 O (orthogonal) -0.57 2 2 6 1 / 6 O -0.68 2 2 6 1 / 6 OH (orthogonal) -0.24 1 1 6 1 / 6 OH -0.31 1 1 6 1 / 3 H (orthogonal) -0.67 -1 -1 6 1 / 3 H -0.63 -1 -1 6 1 / 3 O (orthogonal) -1.03 2 2 6 1 / 3 O -1.18 2 2 6 1 / 3 OH (orthogonal) -0.29 1 1 6 1 / 3 OH -0.50 1 1 6 1 / 4 OH + 1 / 4 H (2×1) -0.77 0 0 4 1 / 4 OH + 1 / 4 H (2×2)-2H -0.79 0 0 4 1 / 2 OH + 1 / 2 H -0.28 0 0 4 Mapping surface Pourbaix space

[0043] Aggregating the above calculations and corresponding pH and USHE dependencies, ∆G was visualized as a function of USHEfor a selection of phases in Fig.4. In the acidic and neutral regimes (Fig.4A and Fig.4B), it was observed that for negative applied potentials the surface was predicted to be fully hydrogenated. Increasing the potential was predicted to lead to dissolution of the Fe110 surface beyond approximately -0.15 V (vs SHE) at pH = 0.0 and -0.30 V (vs SHE) at pH = 7.0.Docket No. Docket No.30275 / 59668 UM 2024-153-02

[0044] Moving into the alkaline regime with a pH of 14.0 (Fig.4c), it was predicted that oxygenated surface phases will emerge that prevent dissolution. Previous targeted surface science works have observed the formation of oxygen covered phases on Fe110, thereby demonstrating a general predisposition for the formation of these phases.

[0045] Continuing this analysis and treating pH as a continuous variable, the surface Pourbaix diagram was plotted in Fig.4a. As this approach does not incorporate changes in the bulk, the pH and USHErange is selected to mostly coincide with stable Fe from bulk Pourbaix diagram. Throughout this space there is no region where a clean Fe110 surface was thermodynamically preferred. For most negative potentials, 1 ML H* was the favored phase. At potentials that were near 0 V (vs SHE) or were positive, either O* phases were predicted to form, or dissolution of the surface is favored depending on the pH. With the aim of promoting pure iron growth, and identifying promising environmental conditions for this process, optimizing for low coverage surface phases may be beneficial for ease of deposition due to increased exposed surface area. Thus, this diagram aligns with previous motivation towards exploring alkaline conditions.

[0046] However, since iron ore such as hematite (Fe2O3) is most soluble in acidic conditions, a preference for pH < 7 arises. This motivates a reevaluation of avenues towards facilitating acidic electrowinning. But in contrast to alkaline conditions, the challenge of selectivity against HER becomes much more severe. Referring to Fig.5, most potentials required to avoid surface dissolution under acidic conditions will activate HER, thereby limiting faradaic efficiency of deposition. To make this approach viable, strategies towards HER suppression must be adopted.

[0047] Interestingly, HER suppression is also an optimization goal within the aqueous Li- ion battery community as this reaction limits the width of the voltage operating window. One strategy applied within this context is water-in-salt electrolytes (Water-in-salt electrolyte), where salt is added in very high concentrations. This has the effect of forming new hydration structures that introduce a desolvation barrier, thereby kinetically blocking HER. While this can also impact oxygen evolution reaction (OER) kinetics, the anodic reaction in an electrowinning cell, appropriate tuning of the water-in-salt electrolyte can decouple HER and OER suppression.

[0048] There are several design parameters that can be used for tuning the salt. For example, pairing a sodium cation with a weakly interacting anion could be effective. Moreover, eutectic salts open the door towards optimization through mixing multiple anions for improved solubility.Docket No. Docket No.30275 / 59668 UM 2024-153-02

[0049] As a by-product of this HER suppression, the surface phase thermodynamics may also be influenced. To probe this impact further, another surface Pourbaix diagram was constructed, except now excluding the formation of any H* phases via Volmer processes (the first step of HER, Eq.3). Implicitly it was also assumed that HER suppression is achieved without a passivation film formed from the salt, which has been demonstrated as possible in the lithium context.

[0050] By applying this suppression, much of the 1 ML H* phase in the original surface Pourbaix diagram was substituted for the 1 / 4 ML OH* + 1 / 4 ML H* phase. This phase stems from water dissociation forming OH* and H* on the surface. As no H+was involved with this process, and it was purely chemical, neither pH nor applied potential will influence the energy of this phase. While there is the possibility that this phase can be kinetically blocked in a fully aqueous environment, from a thermodynamic view its formation is always preferable over a clean surface, even with HER suppression This change in surface behavior halves the total surface coverage from 1 ML to 1 / 2 ML and reinforces the value of HER suppression. Moreover, upon closer inspection it was noted that the 1 / 4 ML O* phase extended into the weak acidic region at relatively low potentials. This has the benefit of decreasing the total coverage further to 1 / 4 ML while also being in the preferred pH for ore dissolution. With this observation, this region of phase space was identified as particularly promising for electrochemical growth of pure iron.

[0051] Several surface phases on Fe110 ranging in coverage from 1 / 6 ML to 1 ML and including mixed adsorbate phases were modeled. Using the associated calculated energies of these phases, surface Pourbaix diagrams were generated, and it was observed that there was a thermodynamic preference for adsorbates in all regions of surface stability. Considering strategies towards efficient electrowinning that operates under acidic conditions, it was observed that water-in-salt electrolyte may be useful in eliminating or at least significantly reducing the competing HER. Building on this discussion and suppressing hy- drogenation phases that are formed via Volmer processes, the surface Pourbaix diagram was generated under these conditions and a prevalence of lower coverage phases was found. Through this finding, promising regions of the acidic phase space were identified to explore via HER suppression, optimizing for low coverage. Specifically, from HER suppression a 1 / 4 ML O phase that emerged in the weak acid and low potential regime was identified. Computational Details

[0052] For all DFT calculations a target grid spacing of 0.16 Å was applied with Fermi- Dirac smearing of 0.05 eV to help improve self-consistent field convergence. As adsorbatesDocket No. Docket No.30275 / 59668 UM 2024-153-02 were placed on a single side of the slab, a dipole correction was applied in the z-direction to correct for the asymmetry. Every slab consisted of four layers, with the bottom two layers fixed to their bulk positions. All structures were relaxed until the maximum force was less than 0.05 eV / Å.

[0053] To model an assortment of adsorbate coverages and patterns, periodicity in the xy- plane, varying dimensions of supercells, and number of explicit adsorbates were leveraged. Accordingly, k-meshes were chosen for each slab size (Table 3). We note that for the 3×2 slab, we considered both orthogonal and non-orthogonal slabs to model their slightly different patterns. Table 3: k-mesh used for each supercells considered in this study. supercell dimensions k-mesh 1×1 12×12×1 2×2 6×6×1 3×2 (orthogonal)4×8×13×2 (non-orthogonal)4×6×1

[0054] To calculate the Gibbs Free energies for each phase we set the reference electrode to be the computational hydrogen electrode (CHE). This is defined as the potential for which the reaction ^^+ ^^is in equilibrium. This choice is convenient as it avoids ^ the need to calculate µ^^^^^^since it sets µ^^ ^^^^^^= ^ µ^^^^^at pH = 0. Furthermore, the chemical potential for the electron can then be expressed as µe_ =−eUSHEwhere USHEis the applied potential relative to the standard hydrogen electrode (SHE).

[0055] Additionally, it was assumed that the water is at its liquid-vapor equilibrium (p = 0.035 bar and T = 298 K) which sets µH2O(l)= µH2O(g). The electric field was neglected.

[0056] With the above choices and approximations, it was possible to calculate adsorption free energies for the formation of surface phases involving H*, O*, and OH*. The energy changes for the reactions described in Eq.1-3 above at 0 V vs SHE and pH = 0 can be expressed as below:Docket No. Docket No.30275 / 59668 UM 2024-153-02

[0057] where GX* is the energy of the modeled slab with 1 X adsorbed (with X∈{H, O, OH}) and G∗is the energy of the clean slab. For systems where multiple adsorbates are explicitly modeled on the surface, the coefficients for the reference states (e.g., µH2(g)) are adjusted accordingly (see ne–and nH+columns in Table 2). Furthermore, this same logic is used for considering mixed phases. For example, the energy for forming a mixed OH* and H* phase can be expressed by combining the reactions described in Eq.1,3:

[0058] These Gibbs free energies are calculated by applying vibrational corrections to the DFT total energies as follows:

[0059] where EDFT is the DFT total energy, ZPE is the zero-point energy and TS is the entropic term. For adsorbates the ZPE was calculated in the harmonic approximation at 1 / 4 ML coverages and assumed this will be relatively independent of coverage. It was also assumed that configurational entropy is negligible on the surface. On the other-hand for the gas phase references, the ZPEs were calculated in the ideal gas limit and entropies taken from previously established experimental values at the liquid-vapor equilibrium. The corrections are summarized in Table 4.Docket No. Docket No.30275 / 59668 UM 2024-153-02 Table 4: Vibrational corrections to adsorbates in eV TS T∆S ZPE ∆ZPE ∆ZPE - T∆S H2O0.67 0 0.57 0 0 H2O → OH* + ½ H20.21 -0.47 0.46 -0.11 0.36H2O → O* + H2 0.41 -0.26 0.33 -0.24 0.02 H2 0.41 0.27 H* 0 0.16O* 0 0.06 OH* 0 0.32

[0060] Using the chemical potential of the electron and a pH correction, we can use these adsorption free energies to then calculate the free energy as a function of applied potential and pH as follows:

[0061] Taking these calculated values and normalizing by the surface area of their corresponding surface phase, we can then compare values to predict the most thermodynamically stable phase as a function of pH and USHE. Surface Phases of Fe210

[0062] Focusing on the acidic to neutral region, a surface Pourbaix diagram was calculated for the Fe210 step, modeling 1 / 5 ML, 3 / 5 ML, and 1 ML coverage of the steps by O and H. Since the terrace surface Pourbaix diagram predicted a drive toward 1 ML H coverage, the terrace region of this step model was covered with hydrogens. Herein it was found that for all regions where the surface is predicted to be stable, there will be a thermodynamic drive toward keeping the step free from adsorbates.

[0063] This analysis did not include solvation in calculating the thermodynamics of these surface phases. Through hydrogen-bonding, the water molecules from the aqueous environment can stabilize the surface adsorbates. For example, the water layer structure on metallic interfaces is an area of extensive study. Of particular relevance here, OH has beenDocket No. Docket No.30275 / 59668 UM 2024-153-02 observed to incorporate itself into the water layer network on Pt which can stabilize it on the surface. On the other hand, solvation has a negligible effect on atomic O adsorbates.

[0064] To gauge how this effect may change the physical conclusions of the model, focus was placed on the neutral to acidic region and applied potentials leading to a stable surface. Here it was investigated the stabilization required to displace the 1 ML H phase as most favored. Without solvation, the most stable phase of pure OH is the 1 / 4 ML OH phase, where the most stable mixed phase involving OH is the 1 / 4 ML OH + 1 / 4 ML H phase. The difference between the formation energy of these phases and the 1 ML H phase as a function of pH and applied potential was calculated. At pH = 0, the OH molecules would need to be stabilized by 10.51 eV for the entire 1 ML H region to be replaced by 1 / 4 OH. The difference at this pH is slightly smaller for the mixed 1 / 4 ML OH + 1 / 4 ML H phase, 8.32 eV, but remains substantial. For the OH phases to replace part of the 1 ML H phase (i.e., up until -0.5 V), the differences are 4.32 eV and 2.32 eV for the pure and mixed OH phases, respectively. Moving into the neutral regime, the difference is further decreased but there remains a large gap to overcome. At pH = 7 and for the OH phases to become more thermodynamically favorable until -0.5 V, the mixed phase is 0.65 eV less stable than 1 ML H. For context, OH is stabilized by 0.33 eV on Pt by the water layer, implying solvation alone is unlikely to significantly alter the predicted surface Pourbaix diagram for the acidic to neutral regimes.

[0065] The analysis was based on thermodynamics and did not incorporate kinetics. Thus Fig.5 represents phases the surface will be driven towards. While not shown in the diagram (Fig.5), intermediate coverages or combinations of these surface phases may coexist on the surface. Theoretical overpotentials for iron electrodeposition

[0066] To understand the most relevant region of the surface Pourbaix for iron electrodeposition (the reverse reaction of Eq.5), the theoretical limiting potentials and overpotentials associated with plating were calculated. The theoretical limiting potential UL is the least negative applied potential for which all steps are downhill in a free energy landscape for a given reaction mechanism. Taking the difference between this limiting potential and the equilibrium potentialyields the theoretical overpotential.

[0067] For iron electrowinning, the reaction of interest is electrodeposition. The energetics of this process govern the efficiency of the growth of pure iron plates. Herein, two main growth mechanisms were considered: (i) island nucleation on a Fe110 terrace and (ii) kink formation at a Fe210 step. Modeling both of these mechanisms was performed byDocket No. Docket No.30275 / 59668 UM 2024-153-02 sequentially placing one iron atom at a time until the terrace or step was reproduced. As a simplification, both of these surface models were modeled without the presence of H*, O*, and OH* adsorbates. Thus, these calculated potentials can be interpreted as a scenario where adsorbate formation is blocked either kinetically or through electrolyte design. Adsorbates present on the surface are expected to generally have a detrimental impact on these overpotential values and possibly plate composition.

[0068] The growth mechanism stemming from island nucleation on a terrace (Fig.8) was first modeled. In this case, iron atoms are placed until another full layer of iron is deposited on the surface. At the standard reduction potential for iron, the first two iron deposition steps are endothermic, and the latter two steps are exothermic. Ideally, placement of the final iron should return the energy to 0 eV as the terrace has been reproduced. However, the observed deviance of 0.07 eV is close to the applied convergence criteria for the bulk energy, 0.05 eV. Moreover, this deviation may be a consequence of differing coordination values of the reference bulk and surface iron atoms, as observed in other metals. Applying limiting potential analysis to this free energy diagram, the most uphill step in this growth mechanism is placement of the first iron atom with a value of 0.50 eV. The limiting potential is calculated byreferenced to SHE. Since the theoretical overpotential is the difference between the standard reduction potential and limiting potential, it is calculated by. Therefore, this initial island nucleation step is the potential determining step, yielding a limiting potential UL of -0.70 V vs SHE. Propagating this quantity to obtain the theoretical overpotential, this value was calculated to be 0.25 V for deposition on a terrace.

[0069] Further investigating the terrace growth mechanism, it was found that for placing the second iron atom on the surface there are two symmetrically unique configurations. Their energies are comparable, with values of 0.66 eV and 0.71 eV respectively. Since this second step is not potential determining, both deposition paths are thermodynamically accessible when the potential applied is at least UL.

[0070] Growth via kink formation was further investigated using the Fe210 surface facet as model. In alignment with previous low-energy electron diffraction experiments, significant relaxation from the bulk with comparable interlayer spacings was observed. Applying this model for the step growth mechanism at the standard reduction potential (Fig.9), the first step is energetically uphill to nucleate the kink with subsequent steps relatively flat in energy. Thus, as in growth at the terrace, The PDS was observed to be the first deposition step. In contrast to growth on the terrace, however, the magnitude of this PDS was observed to be relativelyDocket No. Docket No.30275 / 59668 UM 2024-153-02 smaller (0.26 eV) leading to a ULof -0.58 V vs SHE. The theoretical overpotential for growth at the step site was calculated to be 0.13 V.

[0071] Comparing the theoretical overpotentials between growth at the terrace and step sites, it was predicted through this idealized model for growth to mainly occur via kink formation. This is in alignment with previous conclusions drawn for electroplating of other metals. Without intending to be bound by theory, it is believed that solvation effects through adsorbate formation as well as diffusion kinetics, not included in this model, could influence these growth mechanisms.

[0072] The size restricted model herein focuses on the overall growth energetics without considering the grow of island clusters as intermediate growth stage. Previous epitaxial growth investigations have observed island formation followed by coalescence with increasing coverage. Specifically for electrodeposition, layer-by-layer growth with relatively flat deposits after island formation has also been reported using this method.

[0073] Additionally, the calculated limiting potentials of both terrace and step site mechanisms both lie within the 1 ML H region of the surface Pourbaix diagram (Fig.5). This observation implies that potentials required to electroplate the iron cations from solution will also thermodynamically drive hydrogen adsorbate formation. In other words, iron deposition may proceed on Fe110 in the presence of hydrogen adsorbates thereby introducing the possibility for more complicated interactions and growth mechanisms. A coupled analysis further investigating the interplay of hydrogen as the layer grows is necessary to gauge the likelihood of more intertwined deposition mechanisms. Hydrogen evolution and absorption

[0074] As the surface phases associated with hydrogen formation are most thermodynamically favorable under the required conditions for iron electrodeposition, iron- hydrogen interactions were further investigated. Experimental investigations into the interplay of iron and hydrogen have proposed that the hydrogen adsorption state can lead to two further interaction modes: hydrogen absorption into the bulk and HER. Investigating each of these processes, both geometries terrace and steps were considered to gauge impact stemming from different coordination states of the surface iron atoms. Previous work has highlighted increased reactivity of iron steps with both nitrogen and water, emphasizing the interplay between surface morphology and surface-adsorbate interactions.

[0075] For absorption, the hydrogen will first adsorb onto the surface at the hollow site via a Volmer process (Fig.10a). On the terrace, to allow for greater reconstruction of the slab to accommodate hydrogen entry into the subsurface, the number of layers in the model was increased from 4 to 5 with the top 3 layers free to move. In the x-y dimensions the terraceDocket No. Docket No.30275 / 59668 UM 2024-153-02 supercell model is kept at 2×2. With this thicker slab the hydrogen was relaxed on the surface at the hollow site, and the calculated adsorption enthalpy was -0.91 eV. To identify the most stable hydrogen site upon entry into the slab, the hydrogen was relaxed in the subsurface underneath the top site, hollow site, long bridge, and short bridge. For initial placement in the subsurface below the hollow and long bridge sites, the hydrogen atoms were spontaneously ejected to the surface. Calculating the enthalpies referenced to gas phase for placement under the short bridge and top site, values of 0.03 eV and 0.11 eV, respectively were obtained. This thermodynamic preference towards hydrogen absorption beneath the short bridge is in qualitative agreement with earlier reports, and the configuration is illustrated in Fig.10b. From these values, the enthalpy for terrace absorption was calculated to be 0.94 eV. This finding is in good agreement with previous reports of 0.98 eV60 and 1.00 eV. Moreover, the enthalpy of absorption was calculated to below the short bridge for 1 hydrogen from the fully hydrogenated Fe110 terrace. This process was calculated to be slightly less endothermic (0.83 eV) than for the clean surface indicating that neighboring hydrogens may help encourage this absorption process.

[0076] Extending this analysis to include entry at step sites, two scenarios were evaluated for the clean 210 step: (i) absorption from near the step edge (Fig.10c) and (ii) absorption from adjacent to the step (Fig.10d). In both mechanisms, the hydrogen diffused to below the short bridge near the step edge since it was most favorable in the terrace (Fig.10e). Starting from a hollow site near the step edge, the enthalpy of absorption was calculated to be 1.06 eV. Alternatively, if absorption starts from a hollow site directly adjacent to the step, the calculated enthalpy of absorption was 0.75 eV. While this latter case exhibits a less uphill enthalpy for absorption, calculation of the kinetic barriers is required to determine whether it is predicted to be the more likely absorption mode. It was additionally considered, where the step is hydrogenated on the terrace and hydrogen enters from the step. This process is less endothermic than the mechanism near the step edge but is comparable to entry from beside the step (0.78 eV). Similar to the step edge, this indicates that nearby hydrogens may help hydrogen entry in some configurations.

[0077] As all absorption mechanisms investigated are endothermic (Fig.10f), there is an initial barrier for hydrogen entry into the pristine iron surface. However, defects and vacancies have been identified as key towards hydrogen embrittlement through the introduction of traps, and after initial absorption into the bulk, the hydrogen atoms are anticipated to exhibit increased mobility.Docket No. Docket No.30275 / 59668 UM 2024-153-02

[0078] In addition to entering the iron, adsorbed hydrogen on the surface may alternatively further react to form H2via HER. After the initial Volmer step, HER may proceed via a Heyrovsky process as follows:

[0079] For HER, has generally been used as a descriptor towards catalytic activity. The adsorption energy of hydrogen on the terrace and several step configurations was calculated and the results summarized in Fig.11. All subsequent values are calculated for pH = 0. Starting with the terrace, our calculated adsorption energies used in constructing the surface Pourbaix are -0.60 eV / H atom and -0.28 eV / H atom for 1 / 4 ML and 1 ML, respectively. Of these two phases, we predict full coverage to be generally more dominant by the surface Pourbaix (Fig.5). In comparison to the calculated limiting potential for iron growth on the terrace, -0.70 V, the -0.42 V limiting potential for HER is predicted to be competing with iron plating, as expected.

[0080] HER at the Fe210 step was investigated by considering multiple potential active sites in the vicinity. Initially placing the hydrogen between two step iron atoms, the system relaxed to one of two sites: (i) beside the step above a hollow site (Fig.10c) or (ii) near the step edge (Fig.10d). For each of these configurations the ZPE was calculated to obtain more accurate vibrational corrections. In addition to these two sites, the system was relaxed starting with hydrogen singly coordinated to a step iron, and the system converged to a configuration beside the step but over a top site. Performing vibrational analysis on this configuration, however, revealed an imaginary mode from oscillating parallel to the step edge indicating it is metastable. Further analyses were focused on the two stable sites.

[0081] The corresponding adsorption energies were calculated to be -0.09 eV / H atom beside the step and -0.34 eV / H atom near the step edge. With these values closer to zero than on the terrace, the viewpoint that the step sites may be more reactive than the terrace is reinforced. Comparing these adsorption energies with the step growth limiting potential, -0.58 V, competition between HER and iron deposition are predicted.

[0082] Under neutral to acidic conditions, the surface will be driven towards forming a hydrogen overlayer on the terrace. Calculations of the HER landscape on the 210 step were performed where the terrace portion is covered by hydrogen (Fig 11). In this scenario, the hydrogen is stabilized on the step edge at the bridge between two step iron atoms. With the introduction of the nearby hydrogen overlayer in the model, the hydrogen binding at the step is weakened further to 0.02 eV / H atom. Therefore, hydrogen coverage on the terrace sites increases the reactivity of the steps toward HER even further.Docket No. Docket No.30275 / 59668 UM 2024-153-02

[0083] As an alternative to the Heyrovsky step, two adsorbed hydrogens may chemicallyrecombine via the Tafel process, Previous experimental studies into the mechanism have observed this Tafel step to be more prevalent at lower overpotentials and Heyrovsky at moderate to higher overpotentials. Without intending to be bound by theory it is believed that this thermodynamic analysis using the Heyrovsky step may be interpreted as an upper bound on HER performance. As the Tafel step is a largely kinetic process, a kinetic analysis is required for a more comprehensive view of HER on iron. Further, the results here emphasize HER as a challenge towards the efficient electrowinning of iron and requires electrolytes that can suppress it (e.g., water in salt electrolytes).

[0084] The foregoing description is given for clearness of understanding only, and no unnecessary limitations should be understood therefrom, as modifications within the scope of the disclosure may be apparent to those having ordinary skill in the art.

[0085] All patents, patent applications, government publications, government regulations, and literature references cited in this specification are hereby incorporated herein by reference in their entirety. In the case of conflict, the present description, including definitions, will control.

[0086] Throughout the specification, where the compounds, compositions, methods, and / or processes are described as including components, steps, or materials, it is contemplated that the compounds, compositions, methods, and / or processes can also comprise, consist essentially of, or consist of any combination of the recited components or materials, unless described otherwise. Component concentrations can be expressed in terms of weight concentrations, unless specifically indicated otherwise. Combinations of components are contemplated to include homogeneous and / or heterogeneous mixtures, as would be understood by a person of ordinary skill in the art in view of the foregoing disclosure. References A. Dave, J. Mitchell, K. Kandasamy, H. Wang, S. Burke, B. Paria, B. Póczos, J. Whitacre, V. Viswanathan Cell Reports Phys. Sci. 1, 12, 100264 (2020) [2] A. Dave, K. L. Gering, J. Mitchell, J. Whitacre and V. Viswanathan J. Electrochem. Soc. 167, 013514 (2019) [3] L. Chen, et. al. ACS Energy Lett.5, 3, 968–974 (2020) • World Steel Association, Climate change and the production of iron and steel. 2021, • Daehn, K.; Basuhi, R.; Gregory, J.; Berlinger, M.; Somjit, V.; Olivetti, E. A. Innovations to decarbonize materials industries. Nature Reviews Materials 2022, 7, 275–294. • Fan, Z.; Friedmann, S. J. Low-carbon production of iron and steel: Technology options, economic assessment, and policy. Joule 2021, 5, 829–862. • Raabe, D. The Materials Science behind Sustainable Metals and Alloys. Chemical Reviews 2023, 123, 2436–2608.Docket No. Docket No.30275 / 59668 UM 2024-153-02 • Mallapragada, D. S.; Dvorkin, Y.; Modestino, M. A.; Esposito, D. V.; Smith, W. A.; Hodge, B.-M.; Harold, M. P.; Donnelly, V. M.; Nuz, A.; Bloomquist, C. et al. Decar- bonization of the chemical industry through electrification: Barriers and opportunities. Joule 2023, 7, 23–41. • Ellis, L. D.; Badel, A. F.; Chiang, M. L.; Park, R. J.-Y.; Chiang, Y.-M. Toward electro- chemical synthesis of cement—An electrolyzer-based process for decarbonating CaCO3 while producing useful gas streams. Proceedings of the National Academy of Sciences 2020, 117, 12584–12591. • Houchins, G.; Viswanathan, V. Electrochemical Synthesis of Cementitious Compounds. 2022; US Patent App.17 / 771,197. • Allanore, A.; Yin, L.; Sadoway, D. R. A new anode material for oxygen evolution in molten oxide electrolysis. Nature 2013, 497, 353–356. • Kim, H.; Paramore, J.; Allanore, A.; Sadoway, D. R. Electrolysis of molten iron oxide with an iridium anode: the role of electrolyte basicity. Journal of The Electrochemical Society 2011, 158, E101. • Tang, D.; Zheng, K.; Yin, H.; Mao, X.; Sadoway, D. R.; Wang, D. Electrochemical growth of a corrosion-resistant multi-layer scale to enable an oxygen-evolution inert anode in molten carbonate. Electrochimica acta 2018, 279, 250–257. • Wang, D.; Gmitter, A. J.; Sadoway, D. R. Production of oxygen gas and liquid metal by electrochemical decomposition of molten iron oxide. Journal of the Electrochemical Society 2011, 158, E51. • Allanore, A. Features and challenges of molten oxide electrolytes for metal extraction. Journal of The Electrochemical Society 2014, 162, E13. • Xia, R.; Overa, S.; Jiao, F. Emerging electrochemical processes to decarbonize the chemical industry. JACS Au 2022, 2, 1054–1070. • Allanore, A.; Lavelaine, H.; Valentin, G.; Birat, J.; Lapicque, F. Electrodeposition of metal iron from dissolved species in alkaline media. Journal of the Electrochemical Society 2007, 154, E187. • Haarberg, G. M.; Kvalheim, E.; Rolseth, S.; Murakami, T.; Pietrzyk, S.; Wang, S. Electrodeposition of iron from molten mixed chloride / fluoride electrolytes. ECS Transactions 2007, 3, 341. • Boehme, C.; Starykevich, M.; Constantinescu, G.; Kovalevsky, A. V.; Lopes, D. V. Low- Temperature Electrowinning of Iron from Mixed Hematite-Magnetite Alkaline Suspen- sions. Journal of The Electrochemical Society 2023, 170, 052502. • Vehmaanperä, P.; Salmimies, R.; Häkkinen, A. Thermodynamic and kinetic studies of dissolution of hematite in mixtures of oxalic and sulfuric acid. Mining, Metallurgy & Exploration 2021, 38, 69–80. • McBriarty, M. E.; Stubbs, J. E.; Eng, P. J.; Rosso, K. M. Reductive dissolution mechanisms at the hematite-electrolyte interface probed by in situ X-ray scattering. The Journal of Physical Chemistry C 2018, 123, 8077–8085. • Zhan, G.; Fang, Y.; Zhang, M.; Cao, S.; Xu, T.; Ling, C.; Gu, H.; Liu, X.; Zhang, L. Oxalate promoted iron dissolution of hematite via proton coupled electron transfer. Environmental Science: Nano 2022, 9, 1770–1779. • Allanore, A.; Lavelaine, H.; Valentin, G.; Birat, J.; Lapicque, F. Iron metal production by bulk electrolysis of iron ore particles in aqueous media. Journal of the Electrochemical Society 2008, 155, E125. • Yuan, B.; Kongstein, O. E.; Haarberg, G. M. Electrowinning of iron in aqueous alkaline solution using a rotating cathode. Journal of The Electrochemical Society 2008, 156, D64. • Pourbaix, M. Atlas of Electrochemical Equilibria in Aqueous Solutions; NACE Inter- national: Houston, Texas, 1974.Docket No. Docket No.30275 / 59668 UM 2024-153-02 • Kavalsky, L.; Viswanathan, V. Robust active site design of single-atom catalysts for electrochemical ammonia synthesis. The Journal of Physical Chemistry C 2020, 124, 23164–23176. • Tsiverioti, L. M.; Kavalsky, L.; Viswanathan, V. Robust Analysis of 4e–Versus 6e– Reduction of Nitrogen on Metal Surfaces and Single-Atom Alloys. The Journal of Physical Chemistry C 2022, 126, 12994–13003. • Montoya, J. H.; Tsai, C.; Vojvodic, A.; Nørskov, J. K. The challenge of electrochemical ammonia synthesis: a new perspective on the role of nitrogen scaling relations. ChemSusChem 2015, 8, 2180–2186. • Dinh, C.-T.; Burdyny, T.; Kibria, M. G.; Seifitokaldani, A.; Gabardo, C. M.; Garcı´a de Arquer, F. P.; Kiani, A.; Edwards, J. P.; De Luna, P.; Bushuyev, O. S. et al. CO2 electroreduction to ethylene via hydroxide-mediated copper catalysis at an abrupt interface. Science 2018, 360, 783–787. • Gauthier, J. A.; Stenlid, J. H.; Abild-Pedersen, F.; Head-Gordon, M.; Bell, A. T. The role of roughening to enhance selectivity to C2+ products during CO2 electroreduction on copper. ACS Energy Letters 2021, 6, 3252–3260. • Meijer, K.; Denys, M.; Lasar, J.; Birat, J.-P.; Still, G.; Overmaat, B. ULCOS: ultra-low CO2 steelmaking. Ironmaking & Steelmaking 2009, 36, 249–251. • Lopes, D. V.; Lisenkov, A. D.; Sergiienko, S. A.; Constantinescu, G.; Sarabando, A.; Quina, M. J.; Frade, J. R.; Kovalevsky, A. V. Alkaline Electrochemical Reduction of a Magnesium Ferrospinel into Metallic Iron for the Valorisation of Magnetite-Based Metallurgical Waste. Journal of The Electrochemical Society 2021, 168, 073504. • Lopes, D. V.; Kovalevsky, A. V.; Quina, M. J.; Frade, J. R. Electrochemical deposition of zero-valent iron from alkaline ceramic suspensions of Fe2- xAlxO3 for iron valorisation. Journal of The Electrochemical Society 2020, 167, 102508. • Lopes, D. V.; Quina, M. J.; Frade, J. R.; Kovalevsky, A. V. Prospects and challenges of the electrochemical reduction of iron oxides in alkaline media for steel production. Frontiers in Materials 2022, 9, 1010156. • Xu, L.; Kirvassilis, D.; Bai, Y.; Mavrikakis, M. Atomic and molecular adsorption on Fe (110). Surface Science 2018, 667, 54–65. • Jiang, D.; Carter, E. A. Adsorption and diffusion energetics of hydrogen atoms on Fe (1 10) from first principles. Surface Science 2003, 547, 85–98. • Wight, A.; Condon, N.; Leibsle, F.; Worthy, G.; Hodgson, A. Initial stages of Fe (110) oxidation at 300 K: kinetics and structure. Surface science 1995, 331, 133–137. • Dwyer, D.; Kelemen, S.; Kaldor, A. The water dissociation reaction on clean and oxi- dized iron (110). The Journal of Chemical Physics 1982, 76, 1832–1837. • Yin, X.; Wang, H.; Han, E.-H. Effects of solvation and applied potential on the adsorp- tion behaviors of H, O, OH and H2O on Fe (110) surface. Surface Science 2020, 691, 121504. • Liu, S.; Tian, X.; Wang, T.; Wen, X.; Li, Y.-W.; Wang, J.; Jiao, H. Coverage dependent water dissociative adsorption on Fe (110) from DFT computation. Physical Chemistry Chemical Physics 2015, 17, 8811–8821. • Ossowski, T.; Da Silva, J. L.; Kiejna, A. Water adsorption on the stoichiometric and defected Fe (110) surfaces. Surface Science 2018, 668, 144–149. • Chohan, U. K.; Jimenez-Melero, E.; Koehler, S. P. Surface atomic relaxation and mag- netism on hydrogen-adsorbed Fe (110) surfaces from first principles. Applied Surface Science 2016, 387, 385–392. • White, J. J.; Hinsch, J. J.; Bennett, W. W.; Wang, Y. Theoretical understanding of water adsorption on stepped iron surfaces. Applied Surface Science 2022, 605, 154650. • Beverskog, B.; Puigdomenech, I. Revised pourbaix diagrams for iron at 25–300 C. Corrosion Science 1996, 38, 2121–2135.Docket No. Docket No.30275 / 59668 UM 2024-153-02 • Sripad, S.; Krishnamurthy, D.; Viswanathan, V. The Iron-Age of Storage Batteries: Techno-Economic Promises and Challenges.2021, • Wang, Z.; Guo, X.; Montoya, J.; Nørskov, J. K. Predicting aqueous stability of solid with computed Pourbaix diagram using SCAN functional. npj Computational Materials 2020, 6, 160. • Hansen, H. A.; Rossmeisl, J.; Nørskov, J. K. Surface Pourbaix diagrams and oxygen reduction activity of Pt, Ag and Ni (111) surfaces studied by DFT. Physical Chemistry Chemical Physics 2008, 10, 3722–3730. • Sumaria, V.; Krishnamurthy, D.; Viswanathan, V. Quantifying confidence in DFT pre- dicted surface pourbaix diagrams and associated reaction pathways for chlorine evolu- tion. ACS Catalysis 2018, 8, 9034–9042. • Vinogradova, O.; Krishnamurthy, D.; Pande, V.; Viswanathan, V. Quantifying confidence in DFT-predicted surface pourbaix diagrams of transition-metal electrode– electrolyte interfaces. Langmuir 2018, 34, 12259–12269. • Liu, H.; Jia, X.; Cao, A.; Wei, L.; D’agostino, C.; Li, H. The surface states of transition metal X-ides under electrocatalytic conditions. The Journal of chemical physics 2023, 158. • Ghanekar, P. G.; Deshpande, S.; Greeley, J. Adsorbate chemical environment-based machine learning framework for heterogeneous catalysis. Nature Communications 2022, 13, 5788. • Mortensen, J. J.; Hansen, L. B.; Jacobsen, K. W. Real-space grid implementation of the projector augmented wave method. Phys. Rev. B 2005, 71, 035109. • Enkovaara, J.; Rostgaard, C.; Mortensen, J. J.; Chen, J.; Du lak, M.; Ferrighi, L.; Gavnholt, J.; Glinsvad, C.; Haikola, V.; Hansen, H. A. et al. Electronic structure calculations with GPAW: a real-space implementation of the projector augmented- wave method. Journal of Physics: Condensed Matter 2010, 22, 253202. • Larsen, A. H.; Mortensen, J. J.; Blomqvist, J.; Castelli, I. E.; Christensen, R.; Du lak, M.; Friis, J.; Groves, M. N.; Hammer, B.; Hargus, C. et al. The atomic simulation environment—a Python library for working with atoms. Journal of Physics: Condensed Matter 2017, 29, 273002. • Hammer, B.; Hansen, L. B.; Nørskov, J. K. Improved adsorption energetics within density-functional theory using revised Perdew-Burke-Ernzerhof functionals. Physical review B 1999, 59, 7413. • Kavalsky, L.; Hegde, V.; Meredig, B.; Viswanathan, V. A Multiobjective Closed-loop Approach Towards Autonomous Discovery of Electrocatalysts for Nitrogen Reduction. 2023, • Nørskov, J. K.; Rossmeisl, J.; Logadottir, A.; Lindqvist, L.; Kitchin, J. R.; Bligaard, T.; Jonsson, H. Origin of the overpotential for oxygen reduction at a fuel-cell cathode. The Journal of Physical Chemistry B 2004, 108, 17886–17892. • Vanysek, P. Electrochemical series. CRC handbook of chemistry and physics 2000, 8, 8–33. • Dave, A.; Mitchell, J.; Kandasamy, K.; Wang, H.; Burke, S.; Paria, B.; Póczos, B.; Whitacre, J.; Viswanathan, V. Autonomous discovery of battery electrolytes with robotic experimentation and machine learning. Cell Reports Physical Science 2020, 1, 100264. • Li, W.; Dahn, J. R.; Wainwright, D. S. Rechargeable lithium batteries with aqueous electrolytes. Science 1994, 264, 1115–1118. • Suo, L.; Borodin, O.; Gao, T.; Olguin, M.; Ho, J.; Fan, X.; Luo, C.; Wang, C.; Xu, K. “Water-in-salt” electrolyte enables high-voltage aqueous lithium-ion chemistries. Science 2015, 350, 938–943.Docket No. Docket No.30275 / 59668 UM 2024-153-02 • Zheng, J.; Tan, G.; Shan, P.; Liu, T.; Hu, J.; Feng, Y.; Yang, L.; Zhang, M.; Chen, Z.; Lin, Y. et al. Understanding thermodynamic and kinetic contributions in expanding the stability window of aqueous electrolytes. Chem 2018, 4, 2872–2882. • Chen, L.; Zhang, J.; Li, Q.; Vatamanu, J.; Ji, X.; Pollard, T. P.; Cui, C.; Hou, S.; Chen, J.; Yang, C. et al. A 63 m superconcentrated aqueous electrolyte for high-energy Li-ion batteries. ACS Energy Letters 2020, 5, 968–974. • Meng, Y. S.; Srinivasan, V.; Xu, K. Designing better electrolytes. Science 2022, 378, eabq3750. • Nørskov, J. K.; Rossmeisl, J.; Logadottir, A.; Lindqvist, L.; Kitchin, J. R.; Bligaard, T.; Jonsson, H. Origin of the overpotential for oxygen reduction at a fuel-cell cathode. The Journal of Physical Chemistry B 2004, 108, 17886–17892.

Claims

Docket No. Docket No.30275 / 59668 UM 2024-153-02 What is claimed is:

1. An electrowinning process for reduction of metal from a metal-containing ore in an electrochemical cell comprising oppositely disposed electrodes immersed in a water-in-salt electrolyte, comprising: dissolving the metal-containing ore in the water-in-salt electrolyte under acidic conditions to form a solution comprising one or more metal cations from the metal-containing ore; and applying a potential to selectively deposit one or more of the metal cations from the solution onto one of the electrodes as an electrode for electrodeposition, thereby growing a high-purity metal plate on the electrode for electrodeposition.

2. The process of claim 1, wherein the metal is iron, tin, nickel, cobalt, cadmium, chromium, zinc, manganese, or titanium.

3. The process of claim 1 or 2, wherein the electrode for electrodeposition comprises iron.

4. The process of any one of the preceding claims, wherein the electrode for electrodeposition is surface modified with alloys of iron, tin, cobalt, nickel for electrodeposition over hydrogen evolution or metal oxidation.

5. The process of any one of the preceding claims, wherein the electrode for electrodeposition comprises alloys of iron, tin, cobalt, and / or nickel and thereby achieves good nucleation characteristics or metal electrodeposition along with a low surface diffusion barrier.

6. The process of any one of the preceding claims, wherein the water-in-salt electrolyte comprises a salt concentration of at least 4 M.

7. The process of any one of the preceding claims, wherein the water-in-salt electrolyte comprises iron bis(trifluoromethanesulfonyl)imide (FeTFSI) in water.

8. The process of any one of the preceding claims, wherein the water-in-salt electrolyte comprises a eutectic salt comprising of two or more salts that enhance solubility beyond the solubility limit of the individual components.

9. The process of claim 8, wherein the eutectic salt comprises FeTFSI and a second salt containing one or more cations different from Fe.Docket No. Docket No.30275 / 59668 UM 2024-153-02 10. The process of claim 9, wherein the one or more cations include tetramethyl ammonium and tetramethylethyl ammonium.

11. The process of claim 8, wherein the eutectic salt comprises a single cation and multiple anions.

12. The process of claim 11, wherein the cation is iron and the multiple anions comprise TFSI and nitrate.

13. The process of any one of the preceding claims wherein the acidic conditions comprise the electrolyte having a pH of -2 to 7.

14. The process of any one of the preceding claims, wherein the potential applied for electrodeposition is -0.1 V vs SHE to -1.0 V vs SHE.

15. The process of any one of the preceding claims, wherein the metal is iron.

16. The process of any one of the preceding claims, wherein the metal is Ti.

17. The process of claim 16, wherein the potential applied for electrodeposition is -1.4 V vs SHE to -2.3 V vs SHE.

18. The process of claim 16 or 17, wherein the acidic conditions comprise the electrolyte having a pH of -2 to 5.5

Images