HIGH-ENTROPY OXIDES

Abstract

Disclosed are high-entropy oxides, and methods of their preparation. The high-entropy oxide is characterised by a sub-micron particle size and rod-like particle shape. The method of its preparation includes a co-precipitation step, preferably using an oxalate compound as a precipitating agent. Also disclosed are an electrode, e.g. an anode, a catalyst and an electrochemical cell comprising the high-entropy oxide.

Claims

1. A method of preparing a high-entropy oxide, the method comprising: (a) mixing a solution comprising at least four elementally different metal cations in a solvent, each metal cation making up at least 5% of the total number of the four or more elementally different metal cations, with a precipitating agent to obtain a solid material comprising the at least four metal cations; (b) thermally treating the solid material to obtain a high-entropy oxide; wherein the precipitating agent comprises an organic anion.

2. The method of claim 1, wherein the thermal treatment includes a calcining process to produce a high-entropy oxide intermediate.

3. The method of claim 2, wherein the thermal treatment includes annealing the high-entropy oxide intermediate to obtain the high-entropy oxide.

4. The method of claim 3, wherein the high-entropy oxide intermediate is mixed with a solid-state dispersant before annealing.

5. The method of any one of claims 1 to 4, wherein the thermal treatment includes the use of a controlled atmosphere.

6. The method any one of claims 1 to 5, wherein the solution comprises at least five elementally different metal cations.

7. The method of any one of claims 1 to 6, wherein each metal cation is independently selected from the group consisting of cations of Be, Mg, Ca, Sr, Ba, Mn, Fe, Co, Ni, Cu, Zn, Gd, Pb and Pt.

8. The method of any one of claims 1 to 7, wherein the metal cations are independently selected from the group consisting of cations of Mg, Co, Ni, Cu and Zn.

9. The method of any one of claims 1 to 8, wherein the metal cations are independently selected from the group consisting of cations of Mg, Mn, Fe, Co and Ni.

10. The method of any one of claims 1 to 9, wherein the precipitating agent is an oxalate compound.

11. The method of any one of claims 1 to 10, wherein the solvent comprises water and ethylene glycol.

12. The method of claim 10 or 11 wherein the oxalate compound is ammonium oxalate.

13. A method of preparing a high-entropy oxide, the method comprising: (a) mixing a solution comprising at least four elementally different metal cations in a solvent, each metal cation making up at least 5% of the total number of the four or more elementally different metal cations, with a precipitating agent to obtain a solid material comprising the at least four metal cations; (b) thermally treating the solid material to obtain a high-entropy oxide intermediate; (c) mixing the high-entropy oxide intermediate with a solid-state dispersant and annealing the high-entropy oxide intermediate to form the high-entropy oxide.

14. An oxalate salt comprising four or more elementally different metal cations, each metal cation making up at least 5% of the total number of metal cations.

15. The oxalate salt of claim 14, wherein each metal cation makes up between 5% and 30% of the total number of metal cations.

16. The oxalate salt of claim 14 or 15, in the form of particles comprising the four or more elementally different metal cations.

17. The oxalate salt of any one of claims 14 to 16, wherein each metal cation is independently selected from the group consisting of cations of Be, Mg, Ca, Sr, Ba, Mn, Fe, Co, Ni, Cu, Zn, Gd, Pb and Pt.

18. The oxalate salt of any one of claims 14 to 17, comprising a rod-like particle shape.

19. The oxalate salt of claim 18, wherein the length:width ratio of the particles is between about 1:1.5 to about 1:3.5.

20. The oxalate salt of any one of claims 14 to 19, represented by the formula (A.sub.vB.sub.wC.sub.xD.sub.yE.sub.z)C.sub.2O.sub.4, wherein v, w, x, y and z are each independently about 0.05 to about 0.30, and wherein A, B, C, D, and E are each independently selected from the group consisting of cations of Be, Mg, Ca, Sr, Ba, Mn, Fe, Co, Ni, Cu, Zn, Gd, Pb and Pt.

Description

BRIEF DESCRIPTION OF THE SEVERAL VIEWS OF THE DRAWINGS

[0119] A number of example embodiments will now be described by way of example with reference to the accompanying drawings, in which:

[0120] FIG. 1 shows (a) a SEM image of as-precipitated oxalate precursor described in Example 1, and (b) a TEM bright-field image of a few oxalate precursor bundles described in Example 1.

[0121] FIG. 2 shows a) a high-angle annular dark-field scanning transmission electron microscopic (HAADF-STEM) image of an oxalate bundle described in Example 1 with a scale bar of 200 nm, and (b)-(f) show EDS elemental mappings of Cu, Co, Ni, Zn, and Mg, respectively.

[0122] FIG. 3 shows a thermogravimetric analysis (TGA), and differential scanning calorimetry (DSC) and derivative thermogravimetry (DTG, dotted line) plots of an oxalate precursor described in Example 1.

[0123] FIG. 4 shows XRD patterns of the high-entropy oxides annealed from room temperature to different terminal temperatures and then held at each temperature for 3 hours, as described in Example 2. From bottom to top: 700 C., 800 C., 900 C. and 1000 C. Square, asterisk, triangle, and hash indicate rocksalt, tenorite (CuO), spinel (CO.sub.3O.sub.4), wurtzite (ZnO) phases, respectively.

[0124] FIG. 5 shows an XRD pattern of the high-entropy oxide (Mg.sub.0.2Co.sub.0.2Ni.sub.0.2Cu.sub.0.2Zn.sub.0.2)O prepared via solid-state dispersant assisted annealing. The projections of different crystal planes are exhibited next to their corresponding diffraction peaks.

[0125] FIG. 6 shows a Rietveld refinement of corresponding XRD patterns.

[0126] FIG. 7 shows (a) a SEM image of high-entropy oxide rods, (b) a bright-field TEM image of high-entropy oxide rods, and (c) high-resolution TEM image of a high-entropy oxide rod showing rocksalt lattice fringes of (111) planes parallel to the longitudinal axis of the rod, while the top-left corner inset is the corresponding FFT patterns.

[0127] FIG. 8 shows (a-c) a HAADF-STEM image, an ABF image, and a contrast-reversed image of the ABF image taken along [100] orientation (metal cations and oxygen anions are depicted as large and small spheres, respectively), and (d) ABF image taken along [110] orientation (drifts of 0 atomic columns are demonstrated as arrows, and the direction of an arrow indicates the drifting orientation).

[0128] FIG. 9 shows (a) the CV curves in the first 5 cycles, (b) the CV curves at crescent scan rates from 0.3 mV/s to 1.0 mV/s, and (c) the voltage profiles in different cycles at a current rate of 0.2 /g.

[0129] FIG. 10 shows the cycling performance of a high-entropy oxide anode in 470 cycles at 0.2 A/g.

[0130] FIG. 11 shows (a) the rate performance, (b) the voltage profiles at different current rates, and (c) the cycling performance of conventionally annealed (CA) high-entropy oxide anode at 0.1-A/g.

[0131] FIG. 12 shows the first discharge profiles of the high-entropy oxide anode prepared by conventional annealing (dashed) and dispersant assisted annealing (solid) at a current rate of 0.1 A/g, and inset image is the differential capacity plots (dQ/dV) of two discharge profiles.

[0132] FIG. 13 shows XRD patterns of (a) a product annealed without external oxygen, (b) a product annealed with excessive oxygen, (c) a product annealed with external oxygen and insufficient annealing time, (d) a product of sufficient annealing and external oxygen, according to Example 6.

[0133] FIG. 14 shows an SEM image of as-precipitated oxalate precursor described in Example 5.

[0134] FIG. 15 shows oxidation states of Mn, Fe, Co and Ni in the high-entropy oxide of Example 6, investigated by X-ray absorption near edge structure (XANES).

[0135] FIG. 16 illustrates an aspect of the subject matter in accordance with one embodiment.

[0136] FIG. 17 shows a) a high-angle annular dark-field scanning transmission electron microscopic (HAADF-STEM) image (scale bar of 100 nm), and (b)-(f) show EDS elemental mappings of Mg (b), Mn (c), Fe (d), Co (e) and Ni (f), for the high-entropy oxide particle described in Example 6.

[0137] FIG. 18 shows a schematic of a half cell described in Example 3.

[0138] FIG. 19 shows a schematic of a coin cell described in Example 4.

DETAILED DESCRIPTION

Examples

Characterization Methods

[0139] The composition of the samples was determined by ICP emission spectroscopy (ICP-OES Agilent 5110). About 10 mg of respective samples were dissolved in 10 ml aqua regia at 200 C. in a PTFE beaker. Analysis was undertaken using five different calibration solutions. The structures and phase purity of different samples were characterized by X-ray diffraction (X'Pert Pro MPD with Cu K radiation). The thermal analysis was performed on a simultaneous thermal analyzer (Netzsch STA 449 F.3 Jupiter) in a Pt crucible with a ramp rate of 5 C./min in flowing air. X-ray photoelectron spectroscopy measurements were performed on a Thermo Escalab 250 XI with a monochromatic Al K source and a spot size of 400 m. All spectra were calibrated with the C 1s peak of adventitious hydrocarbons at 284.8 eV before fitting.

[0140] SEM images were taken on a field-emission scanning electron microscope (Thermo Fisher Scientific Apreo S) operating at 30 kV and 0.4 nA. Bright-field TEM images and EDS linear scanning results were obtained on a transmission electron microscope (FEI Tecnai F30) equipped with an XFlash 6T-60 EDS detector (Bruker). The atomic-scale characterizations of individual high-entropy oxide nanorods were conducted on an aberration-corrected S/TEM (FEI Titan Cubed Themis G2 300, FEI) at an accelerating voltage of 300 kV with a convergence semi-angle of 25 mrad. The scanning/TEM was equipped with a monochromator, an EDS detector (Bruker), a Gatan imaging filter (GIF Quantum ER/965, Gatan) of high-resolution electron energy loss spectrometer, and a high-speed K2 camera (Gatan). The multiple-inelastic-scattering background in the core-loss region was removed by Fourier ratio deconvolution of the low energy-loss signal. Line profiles were collected from EDS mappings of each element cation using Gatan DigitalMicrograph GMS3 software. The profiles were converted into text files via the script (Export Profile as Tabbed Text) created by Dave Mitchell, which is available on the website: www.dmscripting.com.

[0141] All chemicals were purchased from Sigma Aldrich and were used without further purification.

Example 1a: Co-Precipitation of an Oxalate Precursor of (Mg.SUB.0.2.Co.SUB.0.2.Ni.SUB.0.2.Cu.SUB.0.2.Zn.SUB.0.2.)O High-Entropy Oxide

[0142] The preferred method for production of a precursor for forming a high-entropy oxide ((Mg.sub.0.2Co.sub.0.2Ni.sub.0.2Cu.sub.0.2Zn.sub.0.2)O) involved the use of oxalate anions to achieve near-equimolar deposition of elementally different cations by forming corresponding complexes in solution. Oxalate anions were found to form precipitating complexes with Mg, Co, Ni, Cu and Zn cations in polar and protic solutions. Oxalate anions were particularly preferred because they were found to have similar rates of precipitate formation for each metal cation, such that each precipitating particle comprised proportions of metal cations that were substantially equivalent with the proportion of metal cations in the reaction solution.

[0143] MgCl.sub.2.Math.6H.sub.2O (99%, 0.55 mmol), CuCl.sub.2.Math.6H.sub.2O (99%, 0.5 mmol), CoCl.sub.2.Math.6H.sub.2O (99%, 0.5 mmol), NiCl.sub.2.Math.6H.sub.2O (99.9%, 0.5 mmol), and Zn(NO.sub.3).sub.2.Math.6H.sub.2O (98%, 0.5 mmol) were dissolved in a mixed solution of 10 ml deionized water and 20 ml ethylene glycol, marked as solution A. Zinc nitrate was preferred as a zinc source. Zinc chloride is less desirable as a zinc source because it forms insoluble zinc oxychloride in an aqueous solution. An excess of magnesium chloride was added because the magnesium oxalate precipitate is slightly soluble. Then, ammonium oxalate monohydrate ((NH.sub.4).sub.2C.sub.2O.sub.4.Math.H.sub.2O, 99%, 2.55 mmol) was dissolved into another mixed solution of 10 ml deionized water and 20 ml ethylene glycol at 50 C., marked as solution B. Ammonium oxalate was the preferred precipitating agent because it is soluble in the water:ethylene glycol solvent.

[0144] The precipitation of the metal cations was less uniform when oxalic acid was used compared to oxalate salts (without being bound by theory, this is likely due to the acidity of oxalic acid affecting the relative solubility of the metal cations in solution). Group I counter ions such as sodium oxalate are insoluble in ethylene glycol and the oxalic acid may give rise to a dissolution of as-precipitated oxalates. Both solutions A and B were heated up to 50 C. under stirring. After that, solution B (oxalate ions) was rapidly poured into solution A (metal ions) under vigorous stirring. The suspension was further stirred at 50 C. for 8 hours, followed by separating the oxalate precursor from the reaction solution by centrifugation. The precursor was washed with water and absolute ethanol several times before drying at 70 C. overnight.

[0145] Without wishing to be bound by theory, the inventors believe the complex was formed in a step-wise polymerisation. The copper-oxalate complex was first formed, followed by forming the complexes of other three transition metals ions (that is, Ni.sup.2+, Zn.sup.2+, and Co.sup.2+). Due to a low value of the critical stability constant (log K), Mg.sup.2+ cations are most difficult to coordinate with oxalate ions. Therefore, the chains of the magnesium-oxalate complexes were formed last. This process provided a precipitated oxalate precursor having a rod-like shape and hierarchical bundle structure.

[0146] FIG. 1(a) illustrates a scanning electron microscopic (SEM) image of the as-prepared oxalate precursor. The precursor precipitated according to the process above showed bundle-like structure. The transmission electron microscopic (TEM) image in FIG. 1(b) demonstrates that the oxalate bundles were monodispersed. Each bundle was 500 nm to 1 m long and with an average cross-section of 180180 nm.sup.2. The TEM image also revealed that the as-precipitated precursor was dense and solid.

[0147] Energy-dispersive X-ray spectroscopic (EDS) analysis was conducted on one of these bundles. The signal of Cu was more intense at the centre. In contrast, the Mg signal was relatively stronger on the periphery of the bundle. The signals of Ni, Zn, Co were almost uniform across the entire oxalate bundle. Hence, the results of the EDS analysis corroborated the above hypothesis on the hierarchical structure.

[0148] FIG. 2a shows a high-angle annular dark-field scanning transmission electron microscopic (HAADF-STEM) image of an oxalate bundle. In FIG. 2(b)-(f), the elemental mappings of Cu, Co, Ni, Zn, and Mg are displayed separately. At a glance, the overall distribution of 5 metal cations across the bundle suggests the successful co-precipitation of a multi-component system on a sub-micron scale. Additionally, a close-up observation indicates that the Mg concentration was more intensive on the periphery while those of Cu were denser at the core (see FIG. 2(b) and FIG. 2(f)).

[0149] The chemical composition of oxalate precursor was further determined by inductively coupled plasma optical emission spectrometry (ICP-OES). The results indicated that molar percentages of each metal cation are relatively close to 20%, i.e., they are close to equimolar.

Example 1b: Alternative Anions

[0150] Several alternative anions were investigated for the preparation of high-entropy oxide precursor materials. These included hydroxide anions, formate anions, acetate anions, and citrate anions. The methodology of Example 1a was followed, except that oxalate was substituted for stoichiometric equivalents of sodium hydroxide, ammonium hydroxide, hexamethylenetetramine, formate, acetate, and citrate. Remarks on the suitability of these anions are provided in Table 1 below.

TABLE-US-00001 TABLE 1 Precipitating Functional agent species Remarks NaOH OH.sup. Use of this precipitating agent resulted in bulky hydroxide precipitates of metal cations NH.sub.3H.sub.2O NH.sub.4OH Use of these precipitating agents resulted in incomplete precipitation of Mg cations and/or Hexamethylene- redissolution of Cu hydroxidate tetramine (HMTA) precipitate by forming tetra-ammine copper complex Ammonium formate HCOO Use of these precipitating agents Ammonium acetate CH.sub.3COO resulted in the formation of Lithium acetate CH.sub.3COO agglomerated floccules Citric acid C.sub.6H.sub.6O.sub.7.sup.2 Ammonium citrate C.sub.6H.sub.6O.sub.7.sup.2

Example 2: Calcining and High Temperature Annealing of (Mg.SUB.0.2.Co.SUB.0.2.Ni.SUB.0.2.Cu.SUB.0.2.Zn.SUB.0.2.)O from Oxalate Precursor

[0151] To prepare an oxide intermediate, the dried oxalate precursor was calcined in a muffle furnace at 400 C. for 3 hours with a ramp rate of 10 C./min.

[0152] Two different annealing approaches were used to treat the calcined oxide intermediate to form a high-entropy oxide.

[0153] In a conventional approach, the oxide intermediate (as calcined, appearing as a black powder) was annealed in a muffle furnace at different temperatures for 3 hours. The temperatures used were 700 C., 800 C., 900 C. and 1000 C. The ramp rates of all processes were 10 C./min.

[0154] The second approach was an annealing process assisted by solid-state dispersants. 0.2 g oxide intermediate was re-dispersed in 40 ml deionized water with 2 g K.sub.2SO.sub.4 under stirring for 30 minutes. After being sonicated for 20 minutes, the suspension was put into an oven and heated up to 120 C. After the water was completely removed, the solid product was finely ground in a mortar. The powder was subsequently placed in ceramic crucibles and annealed at 1000 C. for 3 hours with the same ramp rate used previously. After annealing, K.sub.2SO.sub.4 was dissolved in water, while the solid product was separated out via vacuum filtration. Particles of the high-entropy oxide having a rod-like shape were obtained after washing the product with adequate deionized water and drying the sample in the oven.

[0155] The high-entropy oxide may be quenched following annealing. The high-entropy oxide may be quenched by rapidly cooling the high-entropy oxide immediately following the annealing process. Rapid cooling may be achieved by removing the high-entropy oxide from the oven and allowing to cool by exposure to ambient (e.g., room temperature) air. Rapid cooling may also be achieved by contacting the high-entropy oxide with cooled fluids such as cooled air or liquid nitrogen.

[0156] Superior particle size distributions and homogeneity, and superior electrochemical performance were obtained by an annealing process of at least 5 hours.

[0157] Thermogravimetric (TG) analysis of the high-entropy oxide in air from 30 C. to 1000 C. at a ramp rate of 5 C./min showed two notable weight-loss stages between 100 C. and 400 C. (FIG. 3). The weight loss at a lower temperature, represented by the inflexion point at 169 C. in derivative thermogravimetry (DTG), corresponds to water loss. The weight loss at a higher temperature corresponds to the decomposition of oxalate precursor, which is represented by the inflexion point at about 326 C. in DTG. At 1000 C., around 38% of the mass remains as the high-entropy product. Differential scanning calorimetry (DSC) further confirms that the water loss process is endothermic while the decomposition of oxalates is exothermic. Additionally, an enormous endothermic peak positioned at 740 C. is observed. This peak is indicative of the entropy-driven solid solution process, including the incorporation of Zn.sup.2+ into rocksalt structure and the conversion of spinel Co.sub.3O.sub.4 into CoO. It is believed that the mixing of various components on a sub-micron or even a nanometre scale in oxalate precursor facilitates the solid-solution process at a lower annealing temperature relative to known process. Furthermore, according to the X-ray diffraction (XRD) patterns in FIG. 4, the Bragg peaks indexed to tenorite CuO can be observed after heating the precursor at 800 C. for 3 hours. Those peaks disappear after further escalating the temperature to above 900 C., as tenorite CuO is gradually incorporated into the rocksalt structure, coinciding with the subtle endothermic peak in the DSC curve centred at 830 C.

[0158] Conventional annealing treatments at 1000 C. may lead to severe aggregation of particles. Therefore, to circumvent the formation of large aggregates, a solid-state dispersant was used during high-temperature annealing. Without wishing to be bound by theory, it is believed that the solid-state dispersant suppresses the aggregation and crystallite growth. Considering that the annealing temperature could be as high as 1000 C., potassium sulfate was selected as the dispersant since its melting point is 1069 C. Specifically, the oxalate precursor was first annealed at 400 C. for 3 hours to covert oxalates into a mixed-oxides intermediate. The phases in the intermediate are confirmed by XRD. Despite the poor crystallinity, the phases in the intermediate could be indexed to rocksalt NiO, tenorite CuO, rocksalt MgO, spinel Co.sub.3O.sub.4, and wurtzite ZnO. SEM and TEM images revealed that the bundle-like structure is preserved after moderate-temperature annealing. Interestingly, the intermediate rods have a mesoporous structure. This is because the thermal decomposition of oxalate precursor leaves substantial inner voids within these rods. The as-annealed intermediate was finely dispersed in K.sub.2SO.sub.4, followed by further annealing the mixture at 1000 C. for 3 hours.

[0159] The XRD pattern in FIG. 5 demonstrates that the (Mg.sub.0.2Co.sub.0.2Ni.sub.0.2Cu.sub.0.2Zn.sub.0.2)O high-entropy oxide prepared via a solid-state dispersant assisted annealing is a single-phase compound without any impure phases. Peak positions are shown in Table 2, below. The Rietveld refinement results in FIG. 6 show good convergence and low R-factors, validating that the (Mg.sub.0.2Co.sub.0.2Ni.sub.0.2Cu.sub.0.2Zn.sub.0.2)O high-entropy oxide prepared via solid-state dispersant assisted annealing has an fcc cubic crystal structure with the Fm3m space group. The refined lattice parameters are a=4.2395(3) , b=4.2395(3) , c=4.2395(3) , ===90, V=76.199(8) . In such a structure, oxygen anions occupy the 4a sites, whereas the octahedral 4b sites are randomly co-occupied by Co, Cu, Ni, Zn, and Mg ions with a coordination number of 6. This is a particular example of a high-entropy oxide that is an entropy stabilised oxide.

TABLE-US-00002 TABLE 2 Peak positions (2) (111) (200) (220) (311) (222) (Mg.sub.0.2Co.sub.0.2Ni.sub.0.2Cu.sub.0.2Zn.sub.0.2)O 36.8 42.7 62.0 74.3 78.2

[0160] FIG. 7(a) displays an SEM image of the as-annealed high-entropy oxide rods, validating that the uniformity and bundle-like structure of the oxalate precursor are preserved to a large extent after annealing. Unlike the porous intermediate, the high-entropy oxide nanorods were fairly dense, evidenced in the bright-field TEM image (FIG. 7(b)). As shown in FIG. 7(c), the high-resolution TEM image exhibits the (111) lattice fringes of the high-entropy oxide with an interplanar spacing of 0.246 nm. The axial (longitudinal) direction of the nanosized rod is along [111]. The corresponding Fast Fourier Transform (FFT) patterns are represented in the top-left corner. The EDS linear scanning across the width of a high-entropy oxide rod showed the distribution of 5 cations in high-entropy oxide become more uniform across the rod when compared with the linear scanning results for the oxalate precursor.

[0161] FIG. 8(a) shows the atomic-resolution HAADF-STEM image of a high-entropy oxide nanorod projected along the [100] orientation. The image explicitly reveals that the high-entropy oxide has an fcc sublattice of metal cations with oxygen anions residing at the octahedral holes, indicated by four large spheres (Me) and 12 small spheres (O). FIG. 8(b) shows an annular bright-field (ABF) image, indicating that the atomic columns of metals are axis-aligned with O atomic columns along [100] orientation. The ABF image along [110] direction (FIG. 8(d)) shows that some O anions exhibit a subtle drift from their perfect octahedral sites. This drift is believed to arise from the anion sublattice distortion caused by the Jahn-Teller effects on tetrahedrally coordinated Cu.sup.2+ in an octahedral configuration.

Example 3: Electrochemical Performance of (Mg.SUB.0.2.Co.SUB.0.2.Ni.SUB.0.2.Cu.SUB.0.2.Zn.SUB.0.2.)O High-Entropy Oxide

[0162] The obtained high-entropy oxides were used to assemble electrochemical half cells, where lithium discs were employed as both counter and reference electrodes.

[0163] FIG. 18 shows a schematic of a half cell in the form of a coin cell 1816, comprising a top cap 1802, high-entropy oxide working electrode (anode layer 1804), a polymer separator 1806, a counter electrode (lithium disc 1808), a stainless steel spacer 1810, an o-ring 1812 and a bottom cap 1814.

[0164] The anode layer 1804 was prepared via a typical slurry method. The high-entropy oxide powder synthesised according to Example 2 was mixed with carbon black and PVDF binder to form a slurry with a mass ratio of 8:1:1 in N-methyl-2-pyrrolidinone (NMP). The obtained slurry was coated onto a copper foil. Then, the film was heated on a hotplate at 80 C. to evaporate the NMP, followed by completely drying the film under vacuum at 80 C. overnight. CR2032 coin cells were assembled in a glove box under a pure argon atmosphere. In the coin cell 2-electrode configuration shown in FIG. 18, lithium disc 1808 was separated from the anode layer 1804 by the polymer separator 1806. The electrolyte (not shown) was 1 M LiPF.sub.6 dissolved in a mixture of ethylene carbonate (EC) and diethyl carbonate (DEC) with a volume ratio of 1:1. The mass loading of the active materials on the anode layer 1804 was 0.924-1.052 mg/cm.sup.2.

[0165] FIG. 9(a) presents the cyclic voltammetry (CV) curves of the cell at a scan rate of 0.1 mV.Math.s.sup.1. The curves are similar to those of the previously reported high-entropy oxides (Sarkar et al.). The intensive cathodic peak at 0.52 V is reduced remarkably after first cycling, indicating solid electrolyte interphase (SEI) formation and initial reduction of transition metal oxides into metals and Li.sub.2O. A subtle cathodic peak at about 1.2 V in the first lithiation can be attributed to the Cu.sup.2+/Cu.sup.+ transformation. In the following cycles, four redox peaks are detectable from the CV results. One pair of intensive redox peaks centred at 1.2 V (cathodic) and 1.8 V (anodic) can be ascribed to the reduction of transition metal oxides and re-oxidation of metals. Additionally, a pair of minor redox peaks positioned at 0.375 V and 0.80 V may stem from a combined effect of the alloying and dealloying processes of Zn metal with Li and the spin-polarised surface capacitance of Co and Ni nanoparticles.

[0166] In the previous studies (Sarkar et al. and Ghigna, P.; Airoldi, L.; Fracchia, M.; Callegari, D.; Anselmi-Tamburini, U.; D'Angelo, P.; Pianta, N.; Ruffo, R.; Cibin, G.; de Souza, D. O.; Quartarone, E., Lithiation Mechanism in High-Entropy Oxides as Anode Materials for Li-Ion Batteries: An Operando XAS Study. ACS Applied Materials & Interfaces 2020, 12 (45), 50344-50354), multiple additional cathodic peaks were commonly observed in the first lithiation process, suggesting multiple individual lithiation reactions of the involved cations. The multiple cathodic peaks imply that known high-entropy oxide materials may still preserve a small amount of binary or ternary oxides at some microdomains. The CV curves for subsequent cycles are almost identical after 2 cycles, indicating a superior capacity retention ability of the high-entropy oxide electrodes. Moreover, the CV curves at increasing scan rates from 0.3 to 1 mV.Math.s.sup.1 show similar redox trends in FIG. 9(b), which demonstrates a good electrochemical response to different current rates.

[0167] FIG. 9(c) exhibits the galvanostatic charge/discharge profiles of some representative cycles of the half-cell. The high-entropy oxide electrode delivers a relatively high discharge capacity of 1639 mAh.Math.g.sup.1 at a current of 0.2 A.Math.g1 during the first cycle. The mechanism of lithium storage in high-entropy oxides is through conversion type reaction and, therefore, the initial Coulombic efficiency of the high-entropy oxide anode merely reaches 50.4%. The lithiation capacity then drops to 650 mAh.Math.g.sup.1 at the 30.sup.th cycle, before it increases to 1170 mAh.Math.g.sup.1 at the 400.sup.th cycle progressively. It is worth noting that the voltage profiles of 400.sup.th and 470.sup.th are nearly overlapped, indicating that the capacity of the high-entropy oxide anode is stabilised after 400 cycles. Likewise, FIG. 10 shows the impressive cyclability of the high-entropy oxide anode, demonstrating that the high-entropy oxide of the present invention is impressively stable. Furthermore, the high-entropy oxide anode displays excellent rate performance at increasing current rates and impressive capacity retention (FIG. 11 (a)). More specifically, the high-entropy oxide anode delivers high specific capacities of 545, 470, 407, and 308 mAh/g at 0.2, 0.5, 1, and 3 A/g, respectively. The capacity is stabilised at around 510 mAh/g in post cycles at 0.2 A/g, indicating superior structural stability of the high-entropy oxide anode under electrochemical operating conditions. FIG. 11 (b) demonstrates the charge/discharge profiles of the high-entropy oxide anode at different current rates. The discharge profile at 0.2 A.Math.g.sup.1 after high-rate cycles is superimposed in FIG. 11 (b). It appears to be overlapped with a discharge profile at 0.2 A/g before high-rate cycles (black dashed line) from 1.5 to 1 V. Similarly, the discharge profiles in the same range display minor changes in capacity upon cycling in FIG. 9(c), implying that this conversion reaction process is highly reversible.

[0168] The electrochemical performance of high-entropy oxide materials prepared according to the present invention was compared with corresponding materials prepared using conventional annealing. The high-entropy oxide particles obtained after conventional annealing were also assembled into half-cells. FIG. 11 (c) displays the cycling performance of the conventionally annealed anode (CA-HEO) at a low current rate of 0.1 /g. Despite a low current rate, the CA-HEO anode exhibits a sudden capacity decay after 280 cycles. In addition, by comparing the first discharge profiles of both high-entropy oxide anodes, despite similar discharge capacities, the lithiation plateau observed in the high-entropy oxide rod anode was lower than that of the CA-HEO anode. There is a potential difference of around 46.8 mV between the plateaus of the two high-entropy oxide electrodes. Without wishing to be bound by theory, it is believed this lithiation potential difference may arise from the following two aspects. Firstly, the broad particle size distribution of CA-HEO may create uneven surface overpotentials and kinetics of lithiation/de-lithiation. Secondly, lithiation rates are dependent on the crystallographic directions. The high-entropy oxide rods synthesised in this study have an axial direction of <111>, and the sidewall planes of the rod are {110} and {112}. These non-closely packed planes are kinetically favourable for lithiation. Even though the fine size inevitably leads to a low initial Coulombic efficiency (that is, around 50%) caused by the enormous SEI formation, the short diffusion paths and stable 1D structure enable the high-entropy oxide rods anode a superior long-term cyclability and rate performance.

[0169] A comparison of the high-entropy oxide prepared according to the present invention with two known high-entropy oxide anodes is shown in Table 3. The high-entropy oxide anode according to the present invention delivers the most impressive electrochemical performance with the highest ratio of active material in anode. Furthermore, it is believed the low initial Coulombic efficiency can be effectively overcome by various pre-lithiation methods.

TABLE-US-00003 TABLE 3 High-entropy oxide anode Example 2 Qiu et al. Sarkar et al. Materials Mg.sub.0.2Co.sub.0.2Ni.sub.0.2Cu.sub.0.2Zn.sub.0.2O Structure Rock salt Synthesis Co-precipitation Solid-state Nebulized Spray method according to the synthesis Pyrolysis present invention Electrochemical 1188 mAh/g at a 920 mAh/g at a 590 mAh/g at a performance of current rate at current rate at current rate at active material 0.2 A/g after 470 0.1 A/g after 300 0.2 A/g after 500 (mAh/g) cycles; 308 mAh/g cycles; 490 mAh/g cycles; about 160 at a current rate at a current rate mAh/g at a current at 3 A/g. at 3 A/g. rate at 3 A/g. Proportion of 80% 70% 63% active material

Example 4: Electrochemical Cell

[0170] A coin cell 1902 comprising a high-entropy oxide anode was constructed with the high-entropy oxide of Example 2. A schematic of the constructed coin cell 1902 is shown in FIG. 19. The coin cell 1902 included a top cap 1904, an anode 1906, a separator 1908, cathode foil 1910, a stainless steel spacer 1912, a wave spring 1914, and a bottom cap 1916. Anode 1906 comprised the (Mg.sub.0.2Co.sub.0.2Ni.sub.0.2Cu.sub.0.2Zn.sub.0.2)O high-entropy oxide, and was prepared according to the slurry method, described above in Example 3. The cathode foil 1910 comprised LiFePO.sub.4 or LiNiCoMnO.sub.2.

Example 5: Co-Precipitation of an Oxalate Precursor of an (MgMnFeCoNi)O High-Entropy Oxide

[0171] An oxalate precursor of another high-entropy oxide ((MgMnFeCoNi)O, (Mg.sub.0.2Mn.sub.0.2Fe.sub.0.2Co.sub.0.2Ni.sub.0.2)O) was produced using a method similar to that of Example 1. However, some divalent cations involved (i.e., Fe.sup.2+ and Mn.sup.2+) are vulnerable to oxygen, as they are highly prone to be oxidized to their higher valence states. The fabrication of the precursor was therefore conducted in an inert atmosphere using a Schlenk line so that a pure Ar (Argon) environment could be used during the synthesis. A sample was made by first dissolving 0.1 mmol ascorbic acid (C.sub.6H.sub.8O.sub.6) in a mixed solution of 15 ml deionized H.sub.2O and 15 ml ethylene glycol. Ascorbic acid, which is a reducing agent, can effectively prevent the oxidation of Fe.sup.2+ and Mn.sup.2+ in the aqueous solution. Then, 1.1 mmol MgCl.sub.2 (98%), 1 mmol MnCl.sub.2.Math.4H.sub.2O (98%), 1 mmol FeCl.sub.2 (98%), 1 mmol CoCl.sub.2.Math.6H.sub.2O (99%), and 1 mmol NiCl.sub.2.Math.6H.sub.2O (100%) were dissolved into the above solution in a round bottom flask. After all metal chlorides were added to the solution, the flask was swiftly connected to the Schlenk line and purged with argon gas three times. This solution of metal ions was warmed up to 50 C. under stirring. In another mixed solution of deionized H.sub.2O and ethylene glycol (15 ml+15 ml), 5.1 mmol ammonium oxalate monohydrate (NH.sub.4).sub.2C.sub.2O.sub.4.Math.H.sub.2O was dissolved at 50 C. The solution was then deoxygenated using the Schlenk line before injecting it into the solution of chlorides under vigorous stirring. After reacting for 6 hours at 50 C., the oxalate precursor was washed and separated via centrifugation a couple of times, followed by drying the precursor at 50 C. overnight.

[0172] FIG. 14 shows an SEM image of a rod-shaped particle of the oxalate precursor, (MgMnFeCoNi)C.sub.2O.sub.4.

Example 6: Preparation of the (MgMnFeCoNi)O High-Entropy Oxide

[0173] The (MgMnFeCoNi)O high-entropy oxide as a single-phase solid solution was formed directly by calcining the oxalate precursor at a high temperature. The precursor was calcined using a tube furnace at 1000 C. for 5 h in an Ar atmosphere within a lidded corundum ceramic boat. The precursor was placed in a quartz pan within the boat, and MnO.sub.2 as an oxygen generator was situated next to the quartz pan. The use of quartz pan allowed the MnO.sub.2 to be situated close to the precursor sample, while preventing contact and contamination. The quantity of precursor powder was 300 mg, and the quantity of MnO.sub.2 was 90 mg. Maintaining the temperature of 1000 C. for an extended duration of at least 5 hours was found to be an effective annealing process that resulted in a high-entropy oxide with good purity.

[0174] After formation, the high-entropy oxide was left to cool down naturally. The obtained dark brown powder was found to stable in room ambient conditions, as the phase purity showed no change after exposing the sample to air for several weeks.

[0175] It was found that if the calcining and annealing process was performed in a pure Ar atmosphere, without any oxygen source, the resulting material included a mixture of wstite (FeO) and Ni alloy (FIG. 13 (a)). The presence of the metallic phase can be ascribed to the generation of reductive by-products due to the decomposition of oxalate ligands (i.e., carbon and carbon monoxide) during the heat treatment.

[0176] MnO.sub.2 was used in the calcining process as an oxygen generator to mitigate the effects of a reductive environment. MnO.sub.2 decomposes progressively at high temperatures and releases a small amount of O.sub.2, either neutralizing reductive substances or slightly oxidizing the as-formed metallic products. The lidded boat provided sufficient containment of to maintain the oxidative environment. MnO.sub.2 undergoes a thermal decomposition as follows:

(400 C.800 C.) 2MnO.sub.2=Mn.sub.2O.sub.3+O.sub.2

(Above 800 C.) Mn.sub.2O.sub.3=2Mn.sub.3O.sub.4+O.sub.2

[0177] FIG. 13 (b)-(d) shows the XRD patterns of samples formed with the addition of MnO.sub.2 as an external oxygen source. An excess of oxygen from the decomposition of MnO.sub.2 produces spinel products (AB.sub.2O.sub.4, A=Mg, Mn, Fe, Co, Ni; BFe, Mn) as shown in FIG. 13 (b). The spinel ferrites are formed because Fe.sup.2+ and Mn.sup.2+ ions are proportionally oxidized into Fe.sup.3+ and Mn.sup.3+ by extra oxygen generated from MnO.sub.2 decomposition. A combination of trivalent Fe.sup.3+ and Mn.sup.3+ with remaining divalent cations (Mg.sup.2+, Mn.sup.2+, Fe.sup.2+, Co.sup.2+, and Ni.sup.2+) results in spinel products. These unwanted components can be avoided and/or minimised by utilising introducing into the annealing process an appropriate amount of oxygen. The amount can be readily determined by simple tests.

[0178] FIG. 13 (c) exhibits the co-existence of spinel and metallic phases in the oxide product that was calcined at 1000 C. with the controlled addition of MnO.sub.2 and annealed by maintaining at 1000 C. for one hour. Due to a limited oxygen diffusion rate within the fixed bed, insufficient annealing resulted in an overoxidized top layer and an under-oxidized bottom layer. This problem was addressed by prolonging the annealing duration. As shown in FIG. 13 (d), compared with the product in FIG. 13 (c), when the calcined oxide was annealed by maintaining the same temperature with the same amount of MnO.sub.2, the annealing time of 5 hours led to a well-crystallized single-phase solid solution having a rock salt crystal structure. Peak positions are shown in Table 4.

TABLE-US-00004 TABLE 4 Peak positions 2 (111) (200) (220) (311) (222) (MgMnFeCoNi)O 35.5 41.3 60.3 72.5 76.3

[0179] The oxidation states of each metal species were investigated by X-ray absorption near edge structure (XANES), and the results in FIG. 15 indicate that four transition metal elements unanimously have an average state close to +2. The slight increase in the pre-edge intensity for FeO reference material can be ascribed to the minor oxidization of octahedrally coordinated Fe.sup.2+ into Fe.sup.3+ with tetrahedral geometry.

[0180] FIG. 16 shows an SEM image of the as-annealed high-entropy oxide material. The SEM shows rod-shaped particles and some agglomeration between particles. The use of a solid-state dispersant, such as those described herein, would reduce agglomeration. FIG. 17 (a) shows the dark-field scanning transmission electron microscopy (DF-STEM) analysis and FIG. 17 (b)-(f) show energy dispersive spectroscopy (EDS) mapping of the high-entropy oxide particles. A uniform distribution of different metal species can be clearly observed within the representative (MgMnFeCoNi)O particles.

[0181] It is not the intention to limit the scope of the invention to the abovementioned examples only. As would be appreciated by a skilled person in the art, many variations are possible without departing from the scope of the invention as set out in the accompanying claims.