Layered Sodium Metal Oxides For Na-ion Batteries

Abstract

A composition having the general formula: Na.sub.aMn.sub.bFe.sub.cTi.sub.dM.sub.eO.sub.2, wherein: M comprises one or more elements selected from the group consisting of aluminium, magnesium, zinc, copper, silicon, and zirconium; and wherein: 0.5<a1; 0.1b0.7; 0.1c0.7; 0<d0.3; and 0<e0.5, and wherein the composition is a layered sodium metal oxide material having at least a first phase and a second phase, wherein each phase is different and independently comprises one or more P2-type structures, one or more O3-type structures or one or more P3-type structures. Also described are methods of synthesizing layered sodium metal oxide materials as well as electrodes and energy storage devices including such compositions.

Claims

1. A composition having the general formula:
Na.sub.aMn.sub.bFe.sub.cTi.sub.dM.sub.eO.sub.2, wherein: M comprises one or more elements selected from the group consisting of aluminium, magnesium, zinc, copper, silicon, and zirconium; and wherein: 0.5<a1; 0.1b0.7; 0.1c0.7; 0<d0.3; and 0<e0.5, wherein: the composition is a layered sodium metal oxide material having at least a first phase and a second phase, wherein each phase is different and independently comprises one or more P2-type structures, one or more O3-type structures, or one or more P3-type structures.

2. The composition of claim 1, wherein: (i) the first phase comprises one or more P2-type structures and the second phase comprises one or more O3-type structures; (ii) the first phase comprises one or more P2-type structures and the second phase comprises one or more P3-type structures; or (iii) the first phase comprises one or more P3-type structures and the second phase comprises one or more O3-type structures.

3. The composition of claim 1, wherein: (i) the first phase comprises one or more P2-type structures and the second phase comprises one or more O3-type structures; or (ii) the first phase comprises one or more P2-type structures and the second phase comprises one or more P3-type structures.

4. The composition of claim 1, wherein the first phase comprises one or more P2-type structures and the second phase comprises one or more O3-type structures.

5. The composition of claim 1, wherein the composition consists of: (i) a first phase comprising one or more P2-type structures and a second phase comprising one or more O3-type structures; (ii) a first phase comprising one or more P3-type structures and a second phase comprising one or more O3-type structures; (iii) a first phase comprising one or more P2-type structures, a second phase comprising one or more O3-type structures, and a third phase comprising one or more P3-type structures; or (iv) a first phase comprising one or more P2-type structures and a second phase comprising one or more P3-type structures.

6. The composition of claim 1, wherein: 0.6a0.9; and/or 0.2b0.5; and/or 0.2c0.5; and/or b=c; and/or d=e.

7. The composition of claim 1, wherein M comprises any one or more elements selected from the group consisting of aluminium, copper, magnesium, and zirconium.

8. The composition of claim 1, wherein M comprises one or more elements selected from the group consisting of magnesium, zinc, copper, aluminium, silicon, and zirconium.

9. The composition of claim 1, wherein M comprises aluminium and copper.

10. The composition of claim 1, having the general formula:
Na.sub.aMn.sub.bFe.sub.cTi.sub.dAl.sub.mM.sub.nO.sub.2, wherein: M comprises one or more elements selected from the group consisting of magnesium, zinc, copper, aluminium, silicon, and zirconium; and wherein: 0<m0.2; and 0<<0.2.

11. The composition of claim 1, wherein the layered sodium metal oxide material comprises from 0.1 to 99.9 wt % of the first phase and from 0.1 to 99.9 wt % of the second phase.

12. An electrode comprising the layered sodium metal oxide material of claim 1.

13. An energy storage device comprising the layered sodium metal oxide material of claim 1, wherein the energy storage device is a sodium-ion battery.

14. A method of forming a layered sodium metal oxide material as defined in claim 1 via a sol-gel route, the method comprising: (a) providing a metal salt solution, the metal salts including salts of Na, Mn, Fe, and M; (b) mixing a Ti source with the metal salt solution; (c) mixing a gelator with the metal salt solution to form a sol-gel solution; (d) increasing the pH of the sol-gel solution; (e) heating the sol-gel solution to form a gel; and (f) subjecting the gel to calcination to obtain the layered sodium metal oxide material; wherein M comprises one or more elements selected from the group consisting of aluminium, magnesium, zinc, copper, silicon, and zirconium.

15. The method of claim 14, wherein the gelator is a carboxylic acid.

16. The method of claim 14, wherein the stoichiometric ratio of gelator to metal salts is 1:1.

17. The method of claim 14, wherein step (d) includes increasing the pH of the sol-gel solution to a pH of 6 to 10.

18. The method of claim 14, wherein step (e) includes heating the sol-gel solution to a temperature from 60 to 100 C.

19. The method of claim 14, wherein step (f) includes subjecting the gel to calcination in an oxidising atmosphere.

20. The method of claim 14, wherein step (f) includes: (g) calcining the gel at a first temperature of 400 to 600 C., then (h) calcining the gel at a second temperature of 600 to 1200 C., and, where the layered sodium metal oxide material comprises one or more P3-type structures, (i) calcining the gel at a third temperature of 400 to 600 C.

21. The method of claim 20, wherein step (g) includes calcining the gel at the first temperature for 2 to 6 hours and step (h) includes calcining the gel at the second temperature for 0.5 to 20 hours.

22. A method of forming a layered sodium metal oxide material as defined in claim 1 via a solid-state route, the method comprising: a) providing a sodium source, b) providing Mn.sub.3O.sub.4, Fe.sub.2O.sub.3, TiO.sub.2, c) providing an M oxide, wherein M comprises one or more elements selected from the group consisting of aluminium, magnesium, zinc, copper, silicon, and zirconium; d) milling the compounds of steps a), b), and c) together; e) pelletising the mixture from step d); f) calcining the pelletised mixture from step e).

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0120] FIG. 1 is a diagram of the crystal structures of P2-type, O3-type and P3-type layered sodium metal oxides.

[0121] FIGS. 2a-b are powder X-ray diffractograms showing the presence of O3 and/or P2 phases in each material.

[0122] FIGS. 3a-d show charge-discharge load curves for each material at 25 mA g.sup.1.

[0123] FIG. 4 shows discharge capacities for each material across 100 cycles at 25 mA g.sup.1.

[0124] FIG. 5 shows discharge capacities for each material at specific currents of 25, 50, 100, 250 and 500 mA g.sup.1.

[0125] FIGS. 6a-d show cyclic voltammograms for each of the materials measured for five cycles in a potential window of 2.5-4.2 V vs. Na.sup.+/Na at a scan rate of 0.030 mV s.sup.1.

[0126] FIGS. 7a-c are ex situ X-ray diffraction patterns collected in the pristine state, after charging to 4.0 V or 4.2 V, and after discharging to 2.5 V.

[0127] FIG. 8a shows a voltage profile for the P2 material cycled between 2.5-4.3 V.

[0128] FIG. 8b shows a voltage profile for the O3/P2 material cycled between 2.2-4.2 V.

[0129] FIG. 9a shows specific discharge capacities and electrode discharge energy densities for the P2 material cycled between 2.5-4.3 V.

[0130] FIG. 9b shows specific discharge capacities and electrode discharge energy densities for the O3/P2 material cycled between 2.2-4.2 V.

[0131] FIGS. 10a-b are powder X-ray diffractograms showing the presence of O3 and/or P2 phases in the series Na.sub.0.75Mn.sub.0.4Fe.sub.0.4Ti.sub.0.1M.sub.0.1O.sub.2.

[0132] FIG. 11 shows discharge capacities for materials in the series Na.sub.0.75Mn.sub.0.4Fe.sub.0.4Ti.sub.0.1M.sub.0.1O.sub.2.

[0133] FIG. 12 shows powder X-ray diffraction patterns for solid-state and sol-gel synthesised materials.

[0134] FIG. 13 shows a selected area electron diffraction (SAED) pattern for a single particle of Na.sub.0.75Mn.sub.0.35Fe.sub.0.35Ti.sub.0.1Al.sub.0.1Cu.sub.0.1O.sub.2 material.

[0135] FIG. 14 shows charge-discharge load curves of the solid-state synthesised material cycled between 2.5-4.2 V at 25 mA g.sup.1.

[0136] FIG. 15 shows the discharge capacities of the solid-state and sol-gel synthesised materials cycled between 2.5-4.2 V at 25 mA g.sup.1.

[0137] FIG. 16 shows powder X-ray diffraction pattern confirming the P2/P3 nature of the bi-phasic material Na.sub.0.80Mn.sub.0.4Fe.sub.0.3Ti.sub.0.15Cu.sub.0.15O.sub.2.

[0138] FIG. 17 shows powder X-ray diffraction pattern confirming the O3/P2 and P2/O3 nature of the bi-phasic materials shown.

[0139] FIG. 18 shows the discharge capacities of the materials shown, cycled between 2.5-4.2 V at 25 mA g.sup.1.

[0140] FIG. 19 shows powder X-ray diffraction pattern confirming the O3/P2 nature of the bi-phasic materials shown.

[0141] FIG. 20 shows the discharge capacities of the materials shown, cycled between 2.5-4.2 V at 25 mA g.sup.1.

[0142] FIG. 21 shows powder X-ray diffraction pattern confirming the P2/O3 nature of the bi-phasic materials shown.

[0143] FIG. 22 shows the discharge capacities of the materials shown, cycled between 2.5-4.2 V at 25 mA g.sup.1.

[0144] FIG. 23 shows the discharge capacity of the P2/P3 Na.sub.0.80Mn.sub.0.4Fe.sub.0.3Ti.sub.0.15Cu.sub.0.15O.sub.2.

[0145] FIG. 24 shows powder X-ray diffraction pattern confirming the O3/P3 nature of the bi-phasic materials O3/P3-Na.sub.0.77Mn.sub.0.4Fe.sub.0.4Ti.sub.0.1Cu.sub.0.1O.sub.2 and P3/O3-Na.sub.0.85Mn.sub.0.4Fe.sub.0.4Ti.sub.0.1Cu.sub.0.1O.sub.2.

[0146] FIG. 25 shows the discharge capacity of O3P3-Na.sub.0.77Mn.sub.0.4Fe.sub.0.4Ti.sub.0.1Cu.sub.0.1O.sub.2 and P3O3-Na.sub.0.85Mn.sub.0.4Fe.sub.0.4Ti.sub.0.1Cu.sub.0.1O.sub.2 relative to P3 and O3 single-phase materials.

[0147] FIG. 26 shows powder X-ray diffraction pattern confirming the O3P3P2 nature of the tri-phasic material O3P3P2 Na.sub.0.85Mn.sub.0.4Fe.sub.0.3Ti.sub.0.15Cu.sub.0.15O.sub.2.

[0148] FIG. 27 shows the discharge capacity of the O3P3P2 Na.sub.0.85Mn.sub.0.4Fe.sub.0.3Ti.sub.0.15Cu.sub.0.15O.sub.2 material.

EXAMPLE 1

Synthesis of Layered Sodium Metal Oxide Material

[0149] Four materials based on the chemistry Na.sub.0.72-0.75Mn.sub.0.4Fe.sub.0.4Ti.sub.0.1Al.sub.0.1O.sub.2, with different O3:P2 mass ratios of 1:0, 0.71:0.29, 0.69:0.31 and 0:1 (calculated by Rietveld Refinement), were synthesised using a citric acid sol-gel method. The target composition of the materials is detailed in Table 1.

[0150] Stoichiometric amounts of sodium nitrate, manganese nitrate, iron nitrate and aluminum nitrate were dissolved in di-ionised (DI) water and stirred for 10 mins. A 2 wt % excess of sodium nitrate was used. A sodium content of 0.72 was targeted for the pure phase P2 material, 0.73 for the majority P2 phase P2/O3 material and 0.75 was targeted for the majority O3 phase O3/P2 and pure phase O3 materials. Stoichiometric TiO.sub.2 nanopowder was then added to the solution under stirring, and left to homogenise under stirring for a further 10 mins. Citric acid was dissolved in a separate beaker (1:1 citric acid to metal ratio) and then added to the nitrate solution. After stirring for 2 hours, ammonium nitrate solution was added to adjust the pH from 1 to 8. The solution was then left to stir for a further 2 hours, before heating to 80 C. overnight for gel formation. The gel was then dried at 130 C. for 6 hours, before being ground in a pestle and mortar and calcined under air for 4 hours at 500 C., followed by 12 hours at 900 C. using a heating/cooling rate of 5 C./min. For the O3 material, the high temperature calcination was carried out at 800 C. for 1 hour. Once cooled to 250 C., the samples were removed and ground in a dry room before transferring to an argon-filled glovebox.

TABLE-US-00001 TABLE 1 Phase Sample composition Name Composition (O3:P2 ratio) P2 Na.sub.0.72Mn.sub.0.4Fe.sub.0.4Ti.sub.0.1Al.sub.0.1O.sub.2 0:1 P2/O3 Na.sub.0.73Mn.sub.0.4Fe.sub.0.4Ti.sub.0.1Al.sub.0.1O.sub.2 0.69:0.31 O3/P2 Na.sub.0.75Mn.sub.0.4Fe.sub.0.4Ti.sub.0.1Al.sub.0.1O.sub.2 0.71:0.29 O3 Na.sub.0.75Mn.sub.0.4Fe.sub.0.4Ti.sub.0.1Al.sub.0.1O.sub.2 1:0

Material Characterisation

[0151] Powder x-ray diffraction (XRD) patterns were obtained using a PANalytical Empyrean diffractometer in Bragg-Brentano geometry with Cu K.sub.1 radiation (=1.5406 ). Structures were refined by the Rietveld method using GSAS-II. Scanning electron microscopy (SEM) images of as-synthesised materials coupled with Energy Dispersive x-ray spectroscopy (EDS) were recorded on a JEOL JSM-6700F.

[0152] The results are shown in FIGS. 2a-b, where FIG. 2a shows the full range collected from 10-80 degrees 2, while FIG. 2b shows an expanded version of FIG. 2a focusing on the (001) peaks. In FIG. 2a, the left dashed line highlights a key peak for identifying the P2 phase, while the right dashed line highlights a key peak for identifying the O3 phase.

Electrochemical Characterisation

[0153] To investigate the electrochemical performance of the materials, slurries were prepared using the active material synthesised by the method above, super C65 carbon and Solef 5130 binder (a modified polyvinylidene fluoride (PVDF)), in the mass ratio 80:10:10, in n-methyl-2-pyrrolidone (NMP). The slurry was cast onto aluminum foil using a doctor blade. After drying, 10 mm diameter electrode discs were punched and used to prepare CR2325 coin cells. All slurry processing, casting, drying, punching and coin cell assembly was carried out in an argon-filled glovebox (O.sub.2<0.1 ppm, H.sub.2O<0.1 ppm). Sodium metal was used as a counter/reference electrode, a glass fiber paper (Whatman, GF/F) was used as the separator and 1 M NaPF.sub.6 in EC/DEC was used as the electrolyte. Galvanostatic charge/discharge cycling and cyclic voltammetry were carried out at 30 C. using a Biologic BCS-805 battery cycler.

[0154] The resulting load curves are shown in FIGS. 3a-d. Two main regions can be observed, with a low voltage region corresponding to the Mn.sup.3+/Mn.sup.4+ redox couple, and a high voltage region (ca. >3.0 V) resulting from the Fe.sup.3+/Fe.sup.4+ redox couple. Although the shapes are broadly similar for all materials, the region relating to Mn redox appears to increase with increasing P2 content. For all materials, the load curves become more linear upon cycling, suggesting fewer phase changes occur in later cycles.

[0155] In each case, the initial charge capacity was significantly higher than the initial discharge capacity, confirming that the materials all contained sufficient Na for use in a full cell, and do not rely on additional Na from the metallic counter electrode to achieve high capacities in half cell format. The initial charge capacities were 137, 151, 164 and 128 mAh g.sup.1 for O3, O3/P2, P2/O3 and P2, respectively. The initial discharge capacities were higher for the bi-phasic materials compared to the pure phase materials, with the O3/P2 and P2/O3 materials having initial discharge capacities of 110 and 98 mAh g.sup.1, respectively, compared to 95 and 92 mAh g.sup.1 for the pure phase O3 and P2 materials, respectively. This suggests that the bi-phasic materials have higher initial electrochemical activity compared to the pure phase materials, and that the same result could not be achieved by simply physically combining the two pure phase materials.

[0156] As shown in FIG. 4, over subsequent cycles the P2-containing materials underwent slight increases in their capacities, with the P2/O3 material increasing from 98 to 99 mAh g.sup.1 and the O3/P2 material increasing from 110 to 111 mAh g.sup.1. This effect was greatest in the pure phase P2 material, which saw an increase in capacity from 93 to 102 mAh g.sup.1 over the first 12 cycles, before beginning to gradually fade. In contrast, no activation was observed in the pure phase O3 material, which gradually faded throughout the entire range of cycling. After 100 cycles, the P2 material showed the highest cycling stability, with 83% of the maximum capacity retained. Both bi-phasic materials showed similar capacity fade, with 82% capacity retention after 100 cycles, while the pure phase O3 material showed the most fade, retaining 78% of its maximum capacity. This showed that even when P2 was the minor phase in the material, it had a significant effect in stabilising the cycling performance, which is likely due to reduced overall volume changing occurring in the bi-phasic materials, due to complementary lattice changes in the respective phases during cycling. The O3/P2 material, which consisted of 69% O3 phase, only has 1% greater capacity loss compared to the fully P2 phase.

[0157] As shown in FIG. 5, rate capability testing (carried out at 25, 50, 100, 200 and 500 mA g.sup.1) revealed that the high rate performance increased with increasing P2 content. The O3/P2 material showed significantly enhanced capacity at 500 mA g.sup.1 compared to the pure phase O3 material (45 mAh g.sup.1 compared to 17 mA g.sup.1, respectively), again demonstrating that including a small quantity of P2 phase in the material can significantly enhance performance. Overall, the pure phase P2 material showed the best rate capability with capacities of 94, 86, 77, 66 and 55 mAh g.sup.1 at 25, 50, 100, 200 and 500 mA g.sup.1 respectively, compared to 101, 92, 80, 59 and 49 mAh g.sup.1 for the P2/O3 material, 105, 94, 81, 57 and 45 mAh g.sup.1 for the O3/P2 material and 95, 86, 70, 44 and 17 mAh g.sup.1 for the pure phase O3 material. These results confirmed that at low rates, the bi-phasic materials have higher capacities than the pure phase materials, but as the rate increases the P2 content becomes the crucial factor in determining performance. This reveals that tuning the P2/O3 ratio can be used to design materials with different characteristics, such as using bi-phasic materials for high energy applications, or P2-rich materials for high power applications, without altering the underlying chemistry.

[0158] Cyclic voltammetry was carried out in the same potential window (2.5-4.2 V) using a scan rate of 0.030 mV s.sup.1, to gain insight into the electrochemical reactions occurring during cycling. The results are shown in FIGS. 6a-d.

[0159] It can be seen that all four materials have broadly similar shapes, with peaks around 2.6-2.7 V corresponding to the Mn.sup.3+/Mn.sup.4+ redox couple and 3.3-3.5 V corresponding to the Fe.sup.3+/Fe.sup.4+ redox couple, which matches well with the load curves shown in FIGS. 3a-d. However, in the high voltage region (above 3.9 V), key differences can be observed depending on the phase ratio present in the materials. For the pure phase P2 material, only a single sharp peak is present in this region, with an onset of around 4.1 V. Likewise, only a single peak is present in this region for the pure phase O3 material, but this peak is broader and has an onset of around 3.9 V. Both these peaks are visible in the bi-phasic materials, with the relative intensity proportional to the content of P2 and O3 present in the sample. For the P2/O3 material, the peak present in the P2 material with an onset of 4.1 V has much greater intensity than the peak present in the O3 material with an onset of 3.9 V, while the opposite is true for the O3/P2 material.

[0160] This shows that both the O3 and P2 phases are electrochemically active in the bi-phasic materials, and therefore contribute to Na storage in the proportion that they were synthesised in. This is in contrast to some previous reports on O3/P2 bi-phasic materials, where only the major phase was active (DOI: 10.1039/c7ta11180k). While there was some gradual fade of these high voltage peaks across the first five cycles, overall there were limited changes to the features present, consistent with stable cycling performance. In particular, cycles 4 and 5 almost entirely overlap for all four materials.

Ex Situ Electrochemical Characterisation

[0161] To prepare materials for ex-situ characterisation, powder working electrodes were constructed by mixing the active material and super C65 carbon in the mass ratio 75:25 with no binder, using a swagelok-type cell. All other components were the same as used for the coin cells. The cells were charged to the desired state-of-charge, transferred to an argon-filled glovebox, disassembled, washed using dimethylcarbonate (DMC), and dried overnight under vacuum at room temperature. Glass capillaries were filled and sealed with vacuum grease, and X-ray diffraction patterns were collected in transmission mode (Debyey-Schrrer geometry) using Mo K radiation, =0.71 , on a PANalytical Empyrean diffractometer.

[0162] To investigate any changes in crystal structure during cycling, ex-situ XRD was carried out at selected states-of-charge for the P2, O3 and O3/P2 materials. The results are shown in FIGS. 7a-c.

[0163] Electrodes were extracted after charging to 4.0 V, 4.2 V and after discharging to 2.5 V, and compared to pristine materials. For the P2 material, no major changes in crystal structure could be seen at any state-of-charge, with the P2 structure being retained throughout the first cycle, showing that (de) sodiation occurs via a solid-solution pathway. Small shifts in the lattice parameters can be observed, with a slight expansion upon charging to 4.0 V, followed by a slight contraction after further charging to 4.2 V. After discharging (sodiation) to 2.5 V, the structure expands slightly, although it does not fully return to the original pristine state, showing that minor irreversible changes occur. This lack of the major phase change during the first cycle is consistent with the stable cycling performance shown by this material.

[0164] For the pure phase O3 material, no major structural change could be detected after charging to 4.0 V, although minor changes in the lattice parameters did occur. Upon further charging to 4.2 V, some significant changes could be seen. Specifically, all peaks broadened considerably, implying decrease in crystallinity and long-range order. As a result of this broadening, some peaks could no longer to be detected. Additionally, the (003) diffraction peak shifted from 7.4 to 7.9 degrees 2, consistent with a large contraction of the c-parameter and interlayer spacing. These changes were indexed to a new distorted O3 phase, showing that the CV peak between 3.9-4.1 V results from a O3.fwdarw.O3 transition. The large change in lattice parameters associated with the appearance of this new phase explains the poorer cycling stability observed in this material compared to the pure phase P2 material.

[0165] For the O3/P2 bi-phasic material, a small expansion of the c-parameter is observed upon charging to 4.0 V, with shifts of both the (002) and (003) diffraction peaks for the P2 and O3 phases, respectively. After further charging to 4.2 V, a loss of long-range order in the material was observed, with all remaining peaks broadening significantly and showing loss of intensity. Limited shift is seen for the P2 (002) diffraction peak, though there is a loss of intensity compared to what was seen for the pure phase P2 material at the same state-of-charge. For the O3 (003) diffraction peak, there is a shift to lower 20 values (contraction of the c-axis), as was observed for the formation of the O3 phase for the pure phase O3 material, although the contraction in the c-parameter is slightly smaller than was seen for the pure phase O3 material. That major changes occur in the diffraction peaks for both the P2 and O3 phase demonstrate clearly that both phases are electrochemically active in the bi-phasic material, and both contribute to capacity, allowing high energy densities to be obtained. This is in contrast to some previous reports of O3/P2 composites, where the minor phase has been reported to be inactive, which limits the available energy density. After discharging to 2.5 V, the original O3/P2 diffraction pattern reforms, showing that the phase changes seen during charging are reversible, which explains the stable long-term cycling stability shown by this material.

High Energy Density Testing

[0166] As discussed above, layered sodium metal oxide materials in accordance with the present invention displayed promising performance compared to other reported materials in the 2.5-4.2 V potential window. Further testing was carried out in wider potential windows to investigate the performance for high energy density cells. The results are shown in FIGS. 8a-b and 9a-b.

[0167] It was observed that the O3:P2 ratio had significant impact on the performance in different voltage windows, with the potential window needing optimising for each material studied. Nevertheless, high energy densities could be obtained from the pure phase P2 material when cycled in the potential window of 2.5-4.3 V, with an initial discharge capacity of 146 mAh g.sup.1. This corresponded to a cathode energy density of 480 W kg.sup.1, close to the energy density of the commercial Li-ion cathode material LiFePO.sub.4 (LFP). Although rapid fade occurred during the initial seven cycles, performance soon stabilised with a 7.sup.th cycle discharge capacity of 126 mAh g.sup.1 (energy density of 410 Wh kg.sup.1), of which 87% was retained by the 100th cycle. Crucially, 86% of the energy density was retained over the same cycle range, revealing that negligible voltage fade occurred, with the average discharge voltage remaining high at around 3.2 V throughout cycling. As well as high energy and stability, high energy efficiency (90%) was also shown, while the polarisation of 360 mV was relatively low for a material based on the Fe.sup.3+/Fe.sup.4+ redox couple. In addition, the initial coulombic efficiency of 91% is very suitable for full cell use, and the coulombic efficiency rapidly increased to over 99%, revealing that limited side-reactions occurred.

[0168] For the majority O3 bi-phasic material (O3/P2), a slightly lower potential window of 2.2-4.2 V was used for the high energy testing. An initial discharge capacity of 146 mAh g.sup.1 was achieved (430 W kg.sup.1, higher than the stable discharge energy obtained from the pure phase P2 material). Stable cycling was observed throughout the entire cycling range, with no initial fast fade. Overall, 86% of the initial discharge capacity was retained by the 50th cycle (126 mAh g.sup.1), demonstrating stable long-term cycling performance. In addition, the average voltage was high at just under 3 V, energy efficiency was ca. 89%, and the polarisation of 300 mV was even lower than the pure phase P2 material showed under high energy density testing.

[0169] The majority P2 phase bi-phasic material was also tested in both expanded potential windows. In the 2.5-4.3 V window, it displayed discharge capacities of 125 mAh g.sup.1 on the first cycle, 108 mAh g.sup.1 after 10 cycles, and 83 mAh g.sup.1 after 100 cycles, which corresponds to a 10.sup.th-100.sup.th capacity retention of 77%. This can be explained by higher polarisation in this material, which is likely linked to overcharging of the O3 part of the material up to 4.3 V, which may cause Fe migration and hence lower discharge capacities and cycling stability. In the 2.2-4.2 V potential window, an initial discharge capacity of 136 mAh g.sup.1 was returned, corresponding to an energy density of 407 W.Math.kg.sup.1, with 82% retention over 50 cycles. Whilst this is high, it is lower than observed for the O3-rich bi-phasic material O3/P2, which suggests that in this voltage window, high O3 content is key to achieving high capacities.

[0170] These results show that is it crucial to appropriately match the O3: P2 ratio to the desired potential window to achieve the required performance, and further demonstrate the versatility of being able to tune the O3: P2 ratio to match the desired parameters, without changing the underlying chemistry.

Example 2

[0171] A series of materials with different dopants was synthesised using a solid-state method and tested as positive electrodes for SIBs. The synthesized materials had the composition Na.sub.0.75Mn.sub.0.4Fe.sub.0.4Ti.sub.0.1M.sub.0.1O.sub.2, where M=Mg, Zn, Cu, Al, Si, or Zr. Other materials having a similar composition but not including M, or not including either Ti or M were also synthesized. In the solid state synthesis, stoichiometric amounts of sodium carbonate (2 wt % excess) was balled milled with Mn.sub.3O.sub.4, Fe.sub.2O.sub.3, TiO.sub.2, MgO, ZnO, Cu.sub.2O, Al.sub.2O.sub.3, SiO.sub.2 and ZrO.sub.2 (as appropriate depending on the desired composition) for 1 hour at 400 rpm. The resulting mixture was then pelletised and calcined under air for 12 h at 900 C. at a heating/cooling rate of 5 C./min. Once cooled to 250 C., the composition was removed and ground in a dry room before being transferred to an argon-filled glovebox.

[0172] It was discovered that changing the M element in the series Na.sub.0.75Mn.sub.0.4Fe.sub.0.4Ti.sub.0.1M.sub.0.1O.sub.2, where M=Mg.sup.2+, Zn.sup.2+, Cu.sup.2+, Al.sup.3+, Si.sup.4+ or Zr.sup.4+, could change the crystal structure from pure P2 phase to pure O3 with a range of P2/O3 bi-phasic materials in between, as shown in FIGS. 10a-b.

[0173] It was found that the average ionic radius of the elements in the transition metal layer could act as a rough predictor as to whether a chemical change would lead to a material having a greater/lower O3 or P2 content. A lower average ionic radius in the transition metal layers was associated with an increase in the P2 content, while a higher average ionic radius in the transition metal layers was associated with an increase in the O3 content.

[0174] The materials were tested as the positive electrode for SIBs in the voltage window of 2.5-4.2 V, as shown in FIG. 11. Several materials (where M=Cu, Mg, Zr and O) showed superior performance compared to a reference material MFCu-622, which is a P2-type material having the composition Na.sub.0.75Mn.sub.0.6Fe.sub.0.2Cu.sub.0.2O.sub.2, which was used as a benchmark due to being a high-performing, low-cost material.

Example 3

[0175] A material with the formula Na.sub.0.75Mn.sub.0.35Fe.sub.0.35Ti.sub.0.1Al.sub.0.1Cu.sub.0.1O.sub.2 was designed, to take advantage of the respective benefits observed from the use of Cu and Al dopants in Examples 1 and 2. Two materials with the same chemical composition, Na.sub.0.75Mn.sub.0.35Fe.sub.0.35Ti.sub.0.1Al.sub.0.1Cu.sub.0.102, were synthesised using the sol-gel and solid state synthetic routes used in Examples 1 and 2, respectively. The solid-state route led to a bi-phasic product with a significantly larger P2 content compared to the sol-gel route, showing the influence that the choice of synthetic route has on the O3: P2 phase ratio (as shown in FIG. 12).

[0176] To study the distribution of the O3 and P2 phases within the materials, i.e. to investigate whether the two phases (O3 and P2) exist as separate particles, or as intergrowths on the same particles, transmission electron microscopy (TEM) coupled with selected area electron diffraction (SAED) was used. FIG. 13 shows the resulting SAED pattern from one of the studied particles, which clearly showed the presence of both the O3 and P2 phases existing within one particle. This shows that the two phases exist as intergrowths on the same particles, which is likely to lead to additional benefits electrochemically due to reduced overall strain/volume changes during cycling, compared to a physical mixture of pure phase particles.

[0177] When cycled between 2.5-4.2 V at 25 mA g.sup.1, both sol-gel and solid-state materials showed promising performance, with initial discharge capacities of 114 and 109 mAh g.sup.1 and capacity retentions of 78 and 76% after 100 cycles, respectively (FIGS. 14-15, where FIG. 14 shows the load curves for the solid-state material).

Example 4

[0178] O3P2 and P2/O3 materials of compositions in which bc and/or de were synthesised and tested. Two materials comprising copper and aluminium dopants were synthesised with the following compositions: [0179] Na.sub.0.75Mn.sub.0.45Fe.sub.0.28Ti.sub.0.05Al.sub.0.05Cu.sub.0.17O.sub.2 (O3P2 material comprising 77% O3 phase and 23% P2 phase); and [0180] Na.sub.0.74Mn.sub.0.50Fe.sub.0.25Ti.sub.0.05Al.sub.0.05Cu.sub.0.15O.sub.2 (P2O3 material comprising 92% P2 phase and 8% O3 phase).

[0181] These materials were synthesised using the sol-gel synthetic routes used in Examples 1 and 2, respectively. FIG. 16 shows powder X-ray diffraction patterns for the synthesised materials, showing the presence of O3 and P2 phases.

[0182] When cycled between 2.5-4.2 V at 25 mA g.sup.1, both materials showed promising performance, with initial discharge capacities of about 100 and about 75 mAh g.sup.1 and all materials exhibiting a high capacity retention over at least 15 cycles (FIG. 17).

Example 5

[0183] O3P2 materials of compositions in which bc and/or de were synthesised and compared with material in which b=c and d=e. Five materials comprising copper dopants were synthesised with the following compositions: [0184] Na.sub.0.74Mn.sub.0.45Fe.sub.0.28Ti.sub.0.05Cu.sub.0.22O.sub.2 (O3P2 material comprising 85% O3 phase and 15% P2 phase); [0185] Na.sub.0.72Mn.sub.0.35Fe.sub.0.25Ti.sub.0.20Cu.sub.0.20O.sub.2 (O3P2 material comprising 78% O3 phase and 22% P2 phase); [0186] Na.sub.0.75Mn.sub.0.50Fe.sub.0.25Ti.sub.0.10Cu.sub.0.15O.sub.2 (O3P2 material comprising 60% O3 phase and 40% P2 phase); [0187] Na.sub.0.72Mn.sub.0.35Fe.sub.0.30Ti.sub.0.20Cu.sub.0.15O.sub.2 (O3P2 material comprising 67% O3 phase and 33% P2 phase); and [0188] Na.sub.0.80 Mn.sub.0.30Fe.sub.0.30Ti.sub.0.20Cu.sub.0.20O.sub.2 (O3P2 material comprising 88% O3 phase and 12% P2 phase).

[0189] These materials were synthesised using the sol-gel-synthetic routes used in Examples 1 and 2, respectively. FIG. 18 shows powder X-ray diffraction patterns for the synthesised materials, showing the presence of O3 and P2 phases.

[0190] When cycled between 2.5-4.2 V at 25 mA g.sup.1, all materials showed promising performance, with initial discharge capacities ranging from about 80 to about 97 mAh g.sup.1, and most materials exhibiting a high capacity retention over at least 40 cycles (FIG. 19).

Example 6

[0191] P2O3 materials of compositions in which bc and de were synthesised comprising copper dopants. Three materials were synthesised with the following compositions: [0192] Na.sub.0.72Mn.sub.0.45Fe.sub.0.25Ti.sub.0.10Cu.sub.0.20O.sub.2 (P2O3 material comprising 71% P2 phase and 29% O3 phase); [0193] Na.sub.0.73Mn.sub.0.50Fe.sub.0.25Ti.sub.0.10Cu.sub.0.15O.sub.2 (P2O3 material comprising 86% P2 phase and 14% O3 phase); and [0194] Na.sub.0.72Mn.sub.0.39Fe.sub.0.40Ti.sub.0.05Cu.sub.0.16O.sub.2 (P2O3 material comprising 84% P2 phase and 16% O3 phase).

[0195] These materials were synthesised using the sol-gel-synthetic routes used in Examples 1 and 2, respectively. FIG. 20 shows powder X-ray diffraction patterns for the synthesised materials, showing the presence of O3 and P2 phases.

[0196] When cycled between 2.5-4.2 V at 25 mA g.sup.1, all materials showed promising performance, with initial discharge capacities ranging from about 100 to about 105 mAh g.sup.1, and all materials exhibiting a high capacity retention over at least 15 cycles and in some cases up to 50 cycles (FIG. 21).

Example 4

[0197] A series of mixed phase materials, which include the P3 phase in addition to the O3 and/or P2 phases, were developed with phase compositions including P2P3, O3P3 and O3P2P3.

[0198] The P2P3 phase combination offers materials in which all sodium ions occupy prismatic sites, even at higher sodium contents such as Na.sub.0.80Mn.sub.0.4Fe.sub.0.3Ti.sub.0.15Cu.sub.0.15O.sub.2, (typical P3 or P2 sodium contents are around 0.67). Consequently, the P2P3 phase enables high voltage, good rate capability and good cycle life. The P2P3 Na.sub.0.80Mn.sub.0.4Fe.sub.0.3Ti.sub.0.15Cu.sub.0.15O.sub.2 material was synthesised using the sol-gel synthetic route and the same reagents as used in Example 1 (with the exception that ammonium nitrate was not used). Calcination was carried out under air for 5 hours at 500 C., followed by 12 hours at 1000 C., and then 6 hours at 500 C., using a heating/cooling rate of 5 C./min. The material exhibited a discharge voltage of up to 3.45 V (compared to 3.2-3.3 V for O3P2 materials) and 98% capacity retention up to 50 cycles (see FIG. 23).

[0199] Single phase O3-type and P3-type materials convert into one another during cycling (i.e. O3 single phase materials convert into P3, while P3 single phase materials convert into O3). These conversions are associated with volume changes and capacity fade. Consequently, the O3P3 phase combination offers materials exhibiting greater cycling stability by reducing the driving force for conversion between phases. Bi-phasic O3P3 materials (specifically O3P3-Na.sub.0.77Mn.sub.0.4Fe.sub.0.4Ti.sub.0.1Cu.sub.0.1O.sub.2 and P3O3-Na.sub.0.85Mn.sub.0.4Fe.sub.0.4Ti.sub.0.1Cu.sub.0.1O.sub.2) were synthesised using the sol-gel synthetic route and the same reagents as used in Example 1 (with the exception that ammonium nitrate was not used). For O3P3-Na.sub.0.77Mn.sub.0.4Fe.sub.0.4Ti.sub.0.1Cu.sub.0.1O.sub.2, calcination was carried out under air for 5 hours at 500 C., followed by 12 hours at 1000 C. using a heating/cooling rate of 5 C./min. For P3O3-Na.sub.0.85Mn.sub.0.4Fe.sub.0.4Ti.sub.0.1Cu.sub.0.1O.sub.2, calcination was carried out under air for 5 hours at 500 C., followed by 5 hours at 1000 C. using a heating/cooling rate of 5 C./min. The properties of the bi-phasic O3P3 materials were compared with O3 and P3 single phase materials. The mixed phase P3 and O3 materials delivered higher capacities than the O3 single phase material, with greater cycling stability than the P3 single phase material (see FIG. 25).

[0200] All three phases were combined into a single material: O3P3P2 Na.sub.0.85Mn.sub.0.4Fe.sub.0.3Ti.sub.0.15Cu.sub.0.15O.sub.2, which was synthesised using the sol-gel synthetic route and the same reagents as used in Example 1 (with the exception that ammonium nitrate was not used). Calcination was carried out under air for 5 hours at 500 C., followed by 12 hours at 1000 C., using a heating/cooling rate of 5 C./min. The tri-phasic material allowed for high sodium ion contents and long cycle life (see FIG. 27).