Synthetic methods for porous bimetallic electrocatalysts

Abstract

Facile methods for the synthesis metal oxide electrocatalysts are described herein. In some embodiments, a method of making a doped cobalt oxide comprises providing a powder mixture comprising cobalt salt, dopant transition metal (Me) salt, and water soluble metal salt (WS), and milling the powder composition. The milled powder is calcined to provide a Me x Co1 - x O y -WS matrix, and the Me x Co1 - x O y -WS matrix is subsequently washed with water or aqueous-based solution to remove WS, leaving behind porous Me x Co1 - x O y .

Description

[0001] SYNTHETIC METHODS FOR POROUS BIMETALLIC ELECTROCATALYSTS RELATED APPLICATION DATA The present application claims priority pursuant to Article 8 of the Patent Cooperation Treaty to United States Provisional Patent Application Serial Number 63 / 633,342 filed April 12, 2024 which is incorporated herein by reference in its entirety. FIELD The present invention relates to metal oxide catalysts and, in particular, to porous bimetallic electrocatalysts for the oxygen evolution reaction. BACKGROUND Decarbonizing the energy sector by exploring renewable energy sources (solar, wind, hydro etc.) has evolved as an attractive strategy to achieve sustainable energy supply while ensuring environmental preservation. For example, H2O, CO2, N2 which are abundant in the atmosphere could be converted into important fuels or value-added chemical such as hydrogen, hydrocarbons, oxygenates and ammonia via electro(photo)chemical processes. Since these reactions are thermodynamically unfavorable, energy input in form of light or electrical bias is required to drive such reactions, making catalyst highly indispensable in the decarbonization agenda. For instance, hydrogen production via electro(photo)chemical water splitting have emerged as the more desired cleaner alternative to steam reforming. Fundamentally, electrochemical water splitting proceeds through two distinct reactions; The anodic oxygen evolution reaction (OER) that proceed through a four-electron transfer process and the cathodic hydrogen evolution reaction (HER) that proceeds through a two-electron transfer. The multi-step OER is more thermodynamically costly and therefore controls the efficiency of the water-splitting process. Unfortunately, benchmark catalyst for the OER is the very scarce and expensive (< 3% of all metals) platinum group metals such as iridium (Ir) and ruthenium (Ru) thus impeding the practical deployment of technologies that rely on such reaction. This speaks to the large interest in developing non-noble metal electrocatalyst for the OER with desirable efficiencies. To achieve electrocatalyst with high efficiencies, the number of active sites and the intrinsic activity of each site must be strategically maximized. For instance, hydrothermally synthesized LaCoO3 perovskite oxide with porous and hollow structures exhibited 4-6 times higher OER activity compared to bulk LaCoO3. Apparently, achieving amorphous and porous surface morphologies lead to enhanced active site availability. Similarly, hydrothermally synthesized spinel NiCo2O4 ordered into a KIT-6 network with subsequent template removal shows enhanced OER performance in alkaline media, with a lower overpotential of 350mV compared to 385mV exhibited by non- ordered commercial NiCo2O4. The enhanced OER performance and stability of the KIT-6 ordered structure was attributed to increased ECSA and superior charge transfer capacity. Furthermore, low temperature phosphating strategy was adopted to create CoPx -CoOy with a porous interconnected structure. The amorphous nature and porosity alongside the presence of P atoms lead to enhanced OER performance, achieving an overpotential of 322mV and a tafel slope of 102mV / dec. On the other hand, doping has evolved as a viable means to not only tune site availability but also modifies the electronic structure of the catalysts, leading to materials with tunable electrocatalytic activity. For example, doping of Sr into LaCoO3 increases the oxygen vacancy concentration that enhances its performance and stability for the OER. Similarly, doping of nickel into cobalt oxide reduces the charge transfer resistance leading to enhanced OER performance in alkaline media. This strategy has become extremely popular and is manifest in the large number of multi-elements electrocatalyst such as the recently emerging high entropy materials. To achieve metal oxides with well-defined porosity and interconnected network, soft or hard templating technique is readily utilized. Soft templating technique involves the coordination programming of inorganic metals into self-assembled aggregated surfactants. Although effective, the soft templates strategy is faced with drawbacks such as long processing times, sacrificial nature of templates and excess use of organic solvents. On the other hand, the hard template method explores templates such as silica, KIT-6, SBA-15, MCM-41 that requires corrosive reagents such as hydrofluoric acid (HF) or concentrated NaOH to remove. SUMMARY In view of the foregoing disadvantages, facile methods for the synthesis metal oxide electrocatalysts are described herein. In some embodiments, a method of making a doped cobalt oxide comprises providing a powder mixture comprising cobalt salt, dopant transition metal (Me) salt, and water soluble metal salt (WS), and milling the powder composition. Milling the powder composition can intimately mix the various salt components in preparation for insertion of cobalt ions and Me ions into the template of the water soluble metal salt (WS). The milled powder is calcined to provide a MexCo1-xOy-WS matrix, and the MexCo1-xOy-WS matrix is subsequently washed with water or aqueous-based solution to remove WS, leaving behind porous MexCo1-xOy. In some embodiments, an oxygen donor component is added to the powder composition prior to calcining. The oxygen donor component, for example, can be milled with the mixture of salts prior to calcining. The oxygen donor component, in some embodiments, comprises one or more powder metal hydroxides, such as sodium hydroxide (NaOH). In some embodiments, the cobalt salt and dopant transition metal salt are halides, such as chlorides. Moreover, the WS can also be a halide, such as an alkali metal halide. For example, the WS can be NaCl, in some embodiments. The MexCo1-xOyproduced by methods described herein can exhibit interconnected porosity. Morphology of the MexCo1-xOycan include a network of needles and / or cubes forming the interconnected porosity. Additionally, the porous MexCo1-xOycan be crystalline or a mixture of crystalline and amorphous phases. In some embodiments, the MexCo1-xOy has an interplanar spacing of 0.5 nm to 2.5 nm. The transition metal dopant, Me, can be any transition metal consistent with the technical objectives described herein. In some embodiments, the transition metal dopant is a first row transition metal. For example, Me can be selected from the group consisting of manganese, iron, nickel, and copper. Me can be present in the porous MexCo1-xOyin any desired amount. In some embodiments, for example, x ranges from 0.01 to 0.25 or 0.05 to 0.2. Methods described herein, in some embodiments, further comprise forming the porous MexCo1-xOy into an electrode and performing the oxygen evolution reaction (OER) with the electrode. In some embodiments, the electrode comprising MexCo1-xOy exhibits an overpotential less than 400 mV when participating in OER. The MexCo1-xOy electrode, for example, can exhibit an overpotential of 300-390 mV for OER. Additionally, the MexCo1-xOyelectrode, in some embodiments, can exhibit a charge transfer resistance less than 300 Ω, such as 90-250 Ω. These and other embodiments are further described in the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS FIG.1 illustrates a synthetic pathway for porous MexCo1-xOy, according to some embodiments. FIG.2(a) provided the XRD patterns of various metal oxides described herein. FIG.2(b) is a schematic illustration of spinel Co3O4, and its doped analogues. FIGS.3(a)-3(e) are scanning electron microscopy (SEM) images detailing morphologies of the synthesized MexCo1-xOy catalyst. FIGS.4(a)-4(e), 4(g)-(k), and 4(l)-4(p) are high resolution transmission electron microscopy (TEM) images of various metal oxides described herein. FIGS.5(a)-5(p) provide SEM elements maps of various metal oxides described herein. FIGS.6(a)-6(e) are XPS spectra of various metal oxides described herein. FIG.7(a) provides linear sweep voltammograms (LSV) showing current densities of metal oxides described herein normalized to the electrode surface area. FIG.7(b) shows linear fit of (ja-jc) / 2. FIG.7(c) provides LSV normalized to ECSA. FIG.7(d) provides overpotentials of various metal oxides described herein in OER. FIG.7(e) are Nynguist plots of electrodes comprising metal oxides described herein recorded at 0.2V vs Ag / AgCl. FIG.7(f) provides Tafel slopes of various metal oxides employed in electrodes described herein. DETIALED DESCRIPTION Embodiments described herein can be understood more readily by reference to the following detailed description and examples and their previous and following descriptions. Elements, apparatus and methods described herein, however, are not limited to the specific embodiments presented in the detailed description and examples. It should be recognized that these embodiments are merely illustrative of the principles of the present invention. Numerous modifications and adaptations will be readily apparent to those of skill in the art without departing from the spirit and scope of the invention. EXAMPLE 1 – Synthesis of MexCo1-xOyand associated electrode production FIG.1 illustrates a synthetic pathway for porous MexCo1-xOy, according to some embodiments. For the synthesis of porous Co3O4, CoCl2 (2mmol) and NaCl (0.5g) were weighed and charged into a 50ml yttria-stabilized milling jar and milled with 30 zirconia balls (3mm diameter) for 30 minutes using a planetary ball mill (model) operated on pulse mode (11 minutes on 1 minute off) to induce the insertion of cobalt ions into the NaCl crystalline to form MClx-NaCl. After the initial milling cycle (30 minutes), NaOH (4mmol) was weighed an added into the milling jar as an oxygen donor to facilitate the formation of oxides and subjected to further milling for one (1) hour to form MClx-NaCl-NaOH. To fully convert samples to oxides, increase crystallization and encourage maximum mixing of ions with the NaCl template, milled samples were place in a 50ml porcelain crucibles and calcined in a programmable muffle furnace (model F-47920-80, thermos scientific) at 4000C (10C / min up to appointed temperature) under air for 5 hours to form the MOx-NaCl matrix. Finally, MOx-NaCl was cooled down to room temperature and washed multiple times with Deionized water to remove the NaCl template. Washed samples were dried under vacuum (Fisherbrand, model 281A vacuum oven) at 850oC for 12 hours to final honeycomb structures. To synthesize various M-Co3O4analogues, requisite quantities of Ni, Mn, Fe, Cu required to achieve a mixture with M0.1Co0.9 (1:9) with total mixture of 2mmol. All systems were synthesized via similar process as Co3O4 discussed above. FIG.2(a) shows the XRD patterns of various metal oxides investigated in this study. All the samples show characteristic patterns corresponding to the (220), (311), (222), (400), (331), (511), (440) of Co3O4with other impurities corresponding to CoO phases. However, Fe-Co3O4and Cu- Co3O4shows higher proportion of the CoO phases. similarly, Fe and Cu doped Co3O4demonstrates higher crystallinity, evident in their peak narrowing and increased sharpness. careful inspection of all the XRD peaks shows that incorporation of Ni, Mn, Fe and Cu into the Co3O4 modified the crystal growth process with modification of facet preferences. As observed, the intensity of the (311) planes of the undoped Co3O4 was slightly lower than the (222) plane. However, upon doping, the crystal growth in the (222) planes were inhibited relative to the (311) planes. This reveals that Co-synthesis of Cobalt oxide with other transition metals could significantly affect face orientation and crystallinity. This phenomenon is very important as strong relationship between face orientation with catalytic activities of Co3O4 exists. The morphologies of the synthesized catalyst are shown in FIGS.3(a)-3(e). All catalyst evolves a honeycomb-like architecture with varying degrees of interconnection. Particles in FIGS. 3(c-e) which corresponds to Ni-Co3O4, Mn-Co3O4and Co3O4shows higher interconnection relative to Cu-Co3O4and Fe-Co3O4[3(a) and 3(b) respectively). The relatively higher interconnection in Ni- Co3O4, Mn-Co3O4and Co3O4may reflect lower degree of crystallinity suggesting a more porous and amorphous morphology. This morphological evolution closely agrees with the XRD results. To gain deeper insight into the nature of interconnection and particle morphologies. The morphologies of all samples were further examined via transmission electron microscope as shown in FIGS.4(a)-4(e). As seen from FIGS.4(a)-4(e), TEM results show varying morphologies and interconnecting levels in the various catalysts. Ni-Co3O4and Mn-Co3O4evolves highly connected network of needles and cubes with seemingly amorphous morphologies. Similarly, pure Co3O4[FIG.4(e)] evolves a crystalline microstructure with a cubic morphology. However, Cu-Co3O4 and Fe-Co3O4 (FIGS.4(c) and 4(d) respectively) exhibits lower levels of interconnection with absence of rod-like morphologies. In FIGS.4(g)-4(k), HR-TEM images show the lattice arrangement in the studied system. Presence of well-defined lattice fringes is ascribed to high degrees of crystallinity. However, careful inspection of the interplanar spacing reveals that Ni-Co3O4[FIG.4(l)] shows very broad spacing (d =2.1nm) typical of semi-crystalline or amorphous structure. Although all catalysts were synthesized under same conditions, different morphologies, crystallinity, and structures are obtained. FIGS.5(b)-5(p) shows the SEM-EDS spectra of Ni-Co3O4, Fe-Co3O4, Ni-Co3O4, Mn-Co3O4, Co3O4. SEM-EDS elemental mapping confirms the homogenous distribution of all metals and oxygen on the catalyst surface. As shown in FIGS.6(a)-6(e), the oxidation states of surface Mn, Fe, Ni, Cu and Co were studied by XPS. From FIG.5(a), the high-resolution spectra of Cobalt in all samples shows the Co 2p3 / 2 and Co 2p1 / 2 centered around 779.5eV and 794.5 eV respectively. Although the high spin Co(II) and low-spin Co(III) may sometime have overlapping binding energies, the spin orbital energy difference in all systems studied here is approximately 15eV, thus indicating that cobalt exists mainly as the low-spin Co(III)

[0027] . High resolution XPS spectra of Mn in Mn-Co3O4 [FIG.5(b)] gave two distinct peaks at 642.2 eV and 653.7eV which corresponds to the Mn 2p3 / 2 and Mn 2p1 / 2 respectively. The spin-orbital energy difference between 2p3 / 2 and 2p1 / 2 is 11.5eV which is typical for Mn existing in the Mn3+ oxidation state, indicating partial substitution of Mn within the Oh sites of spinel oxide. As shown in FIG.5(b), the core level XPS spectra of Ni 2p in Ni-Co3O4 evolves two major peaks centered at 855.2eV and 872.8eV which can be assigned to Ni 2p3 / 2 and Ni 2p1 / 2 respectively. The peaks located at 861.3eV and 879.6eV corresponds to satellite peaks of Ni 2p3 / 2 and Ni 2p1 / 2 respectively. The spin orbital energy separation between Ni 2p3 / 2 and Ni 2p1 / 2 is 17.6eV, typical of Ni(OH)2. High resolution peak of Cu 2p in Cu-Co3O4 gives the Cu 2p3 / 2 and Cu 2p1 / 2 signal centered at 933.1eV and 953.4eV respectively having a spin orbital energy difference of 20.3eV, with two strong satellite peaks centered around at 941.5eV and 962.6eV that are indicative of Cu2+, which shows Cu to exist as CuO. as observed, the Cu 2p3 / 2 peak is split into two peaks at 932.1 eV and 933.9eV which could be ascribed to Cu existing in the Cu0 and Cu2+ oxidation states respectively. This confirms that Cu exist in more than one oxidation states. Fe 2p XPS spectra of Fe- Co3O4 gives only two peaks at 711.5eV and 722.3eV which corresponds to the 2p3 / 2 and 2p1 / 2 without any satellite peak. Absence of satellite peaks has been associated with Fe existing as Fe3O4. The obtained catalyst was tested for electrocatalytic OER performance using the standard rotating disk electrode system featuring a glassy carbon electrode (0.071 cm2), platinum counter electrode and a Ag / AgCl reference electrode in N2 purged 1M KOH electrolyte. The linear sweep voltammograms (LSV) in FIG.7(a) shows current densities of all the catalyst system normalized to the working electrode’s surface area. Fe-Co3O4, Ni-Co3O4 and Mn-Co3O4 shows similar current densities which outperforms Cu-Co3O4 while pure Co3O4 shows least performance. Prior to obtaining the final LSV curves, pre-conditioning step that involves taking cyclic voltammograms (CV) scans at 100mV / s for 30 cycles was conducted to stabilize pre-catalysts to stable testing conditions. During preconditioning, Cu-Co3O4 shows the smallest onset potential, lowest overpotential and highest current density. However, after only few cycles, performance significantly deteriorated. Such performance changes are observed in all catalyst with varying degrees. The performance changes follow the order Cu-Co3O4 > Fe-Co3O4 > Mn-Co3O4 > Co3O4 > Ni-Co3O4. These capacity changes may be attributed to chemical changes to active sites during oxidation-reduction cycles where pre- catalysts are converted to more stable chemical states. Results indicates that Ni-Co3O4 undergoes minimal changes during the precondition. As shown in figure (S2) consecutive CV scans obtained under normal testing condition of 20mV / s after preconditioning steps shows excellent reproducibility. For direct comparisons of the various catalyst, we sought to determine their intrinsic catalytic performance by normalizing their respective currents to electrochemical surface area (ECSA). The ECSA were evaluated from the double layer capacitance (Cdl). The recorded CVs taken at different scan rates for the various catalysts are shown in FIG.7(b) shows linear fit of (ja-jc) / 2. ECSA values estimated from the Cdl is shown in Table 1. As expected, Ni-Co3O4 shows highest ECSA values due to its amorphous and highly interconnected nature that encourages good catalyst- electrolyte contact, thus enhancing active site availability. From FIG.7(b) ECSA values follows the trend Ni-Co3O4>Co3O4>Mn-Co3O4>Fe-Co3O4>Cu-Co3O4. These ECSA values closely correspond to crystallinity levels as observed on the XRD results. Subsequently, the LSV is normalized to ECSA as shown in FIG.7(c). The intrinsic catalytic activities follow the trend Fe-Co3O4 >Mn-Co3O4>Cu- Co3O4>Ni-Co3O4> Co3O4. Fe-Co3O4 shows highest current densities (both disk and ECSA) despite having a relatively lower ECSA. Overpotentials (ECSA) of all the catalysts shown in FIG. 7(d) follows the trend Cu-Co3O4(348mV)>Fe-Co3O4(367mV)>Mn-Co3O4 = Ni-Co3O4 (387mV) > Co3O4 (470mV) without IR compensation. Although Cu-Co3O4 shows faster kinetics at lower potentials, it becomes electrochemically unstable at higher potentials and cycles. Electrochemical impedance spectroscopy (EIS) was performed to gain insight into how charge transfer resistance affects the OER performance of studied electrocatalysts. As shown in FIG. 7(e), Nynguist plots were recorded at 0.2V vs Ag / AgCl. The order of the charge transfer resistance is Ni- Co3O4 (99.83^^^^) < Fe- Co3O4 (199.01^^^^) < Cu- Co3O4 (220^^^^) < Mn-Co3O4 (280 ^^^^) < Co3O4 (373.89 ^^^^). Ni-Co3O4 shows the least charge transfer resistance while pure Co3O4 shows the highest. The high electrical conductivity of Ni-Co3O4 may be attributed to its amorphous and highly interconnected morphology. The trend in charge transfer resistance does not directly coincides with the OER performance. This is because other factors such as ECSA, crystal phase purity, chemical environment of active site etc. could significantly affect the overall OER performance. However, for pure Co3O4, the electrical conductivity directly coincides with the OER performance. Tafel slope was used to understand the reaction kinetics on each catalyst surface. Tafel slope follows the order Fe- Co3O4(70.2 mV dec-1) >Mn-Co3O4(76.9mV dec-1) > Ni-Co3O4(96.6mV dec-1) > Co3O4(126.8mV dec-1) > Cu-Co3O4(224.5mV dec-1). As set forth in this example, a series of bimetallic cobalt-based electrocatalysts have been synthesized via solvent free NaCl-solid solution mechanochemical strategy. Four different first row transition metals (Ni, Mn, Fe, Cu) were doped into Cobalt oxide to yield the corresponding bimetallic compounds. Honeycomb architectures with obvious porosity and interconnection was obtained. The electrocatalytic activities of synthesized compounds towards the oxygen evolution reaction were carefully studied. Although the performance of each catalyst varied in different fundamental aspects pertinent to OER activity, we discover that overall, Fe, Ni and Mn doped systems shows better performance, kinetics, and stability under testing conditions when compared to pure Cu-Co3O4 and Co3O4. This study further establishes a facile and cheap strategy to obtaining highly active transition metal electrocatalyst for electrochemical water splitting and, screens for the most favorable active sites via a careful comparative framework. Further pushing research efforts in the development of non-noble electrocatalyst for the OER. Various embodiments of the invention have been described in fulfillment of the various objectives of the invention. It should be recognized that these embodiments are merely illustrative of the principles of the present invention. Numerous modifications and adaptations thereof will be readily apparent to those skilled in the art without departing from the spirit and scope of the invention.

Claims

CLAIMS 1. A method of making a doped cobalt oxide comprising: providing a powder mixture comprising cobalt salt, dopant transition metal (Me) salt, and water soluble metal salt (WS); milling the powder composition; calcining the milled powder composition to provide a MexCo1-xOy-WS matrix; and washing the MexCo1-xOy-WS with water or aqueous-based solution to remove WS, leaving behind porous MexCo1-xOy.

2. The method of claim 1, wherein the cobalt salt comprises a cobalt halide, and the WS comprises alkali metal halide.

3. The method of claim 1, wherein the dopant transition metal (Me) salt comprises a halide.

4. The method of claim 1, wherein the cobalt salt comprises cobalt chloride, the dopant transition metal salt comprises chloride, and the WS comprises alkali chloride.

5. The method of claim 1 further comprising adding an oxygen donor component to the powder composition prior to calcining.

6. The method of claim 5, wherein the oxygen donor component comprises powder metal hydroxide.

7. The method of claim 1, wherein x ranges from 0.05 to 0.

2.

8. The method of claim 1, wherein porosity of the MexCo1-xOyis interconnected.

9. The method of claim 1, wherein the MexCo1-xOy is crystalline.

10. The method of claim 1, wherein the MexCo1-xOycomprises a mixture of crystalline and amorphous phases.

11. The method of claim 1, wherein Me is a first row transition metal.

12. The method of claim 11, wherein Me is selected from the group consisting of manganese, iron, nickel, and copper.

13. The method of claim 1, wherein the MexCo1-xOyhas an interplanar spacing of 0.5 nm to 2.5 nm.

14. The method of claim 1 further comprising forming the MexCo1-xOy into an electrode and performing the oxygen evolution reaction with the electrode.

15. The method of claim 14, wherein the electrode has an overpotential less than 400 mV.

16. The method of claim 14, wherein the electrode has an overpotential of 300-390 mV.

17. The method of claim 14, wherein the electrode has a charge transfer resistance less than 300 Ω.

18. The method of claim 14, wherein the electrode has a charge transfer resistance of 90-250 Ω.

Images