FI ELD OF THE I NVENTION

The present invention relates to novel doped nickelate-containing compositions and their method of preparation, to electrodes containing the novel doped nickelate-containing compositions, and to the use of these electrodes, for example in energy storage devices.

BACKGROUND OF THE INVENTION

Sodium-ion batteries are analogous in many ways to the lithium-ion batteries that are in common use today; they are both reusable secondary batteries that comprise an anode (negative electrode), a cathode (positive electrode) and an electrolyte material, both are capable of storing energy, and they both charge and discharge via a similar reaction mechanism. When a sodium-ion (or lithium-ion battery) is charging, Na ⁺ (or Li ⁺ ) ions de- intercalate from the cathode and insert into the anode. Meanwhile charge balancing electrons pass from the cathode through the external circuit containing the charger and into the anode of the battery. During discharge the same process occurs but in the opposite direction.

Lithium-ion battery technology has enjoyed a lot of attention in recent years and provides the preferred portable battery for most electronic devices in use today; however lithium is not a cheap metal to source and is considered too expensive for use in large scale applications. By contrast sodium-ion battery technology is still in its relative infancy but is seen as advantageous; sodium is much more abundant than lithium and some researchers predict this will provide a cheaper and more durable way to store energy into the future, particularly for large scale applications such as storing energy on the electrical grid. Nevertheless a lot of work has yet to be done before sodium-ion batteries are a commercial reality.

NaNio.5Mno.5O2 is a known Na-ion material in which the nickel is present as Ni ²⁺ while the manganese is present as Mn ⁴⁺ . The material is ordered with the Na and Ni atoms residing in discrete sites within the structure. The nickel ions (Ni ²⁺ ) are a redox element which contributes to the reversible specific capacity and the manganese ions (Mn ⁴⁺ ) play the role of a structure stabilizer. Compound NaNio.5Tio.5O2 is analogous to NaNio.5Mno.5O2 in that the Ni ²⁺ ions provide the active redox centre and the Ti ⁴⁺ ions are present for structure stabilization. There is plenty of literature describing the preparation of NaNio.5Mno.5O2 (and to a lesser extent NaNio.5Tio.5O2) as the precursor for making LiNio.5Mno.5O2 and LiNio.5Tio.5O2 by Na→ Li ion exchange for Li-ion applications. A direct synthesis method to make these Li materials yields undesirable disordered materials, for example, as a result of the lithium and nickel atoms sharing structural sites. However, recent electrochemical studies reported by Komaba et al Adv. Funct. Mater. 2011 , 21 , 3859 describe the sodium insertion performance of hard-carbon and layered NaNio.5Mno.5O2 electrodes in propylene carbonate electrolyte solutions. The results obtained show that although NaNio.5Mno.5O2 exhibits some reversible charging and discharging ability, the capacity of the material fades by 25% or more, after only 40 cycles.

It is typically possible to predict the maximum charge capacity for sodium and lithium nickelate compounds based on the Ni ²⁺ to Ni ⁴⁺ redox process, however as Zhonghua Lu and J. R. Dahn, J. Electrochemical Society, 149 (7) A815-A822 (2002) explain, the electrochemical behaviour of cells made using lithium containing compounds, e.g. Li[NixLi ₍₁ 3_2x/3 ₎ Mn(2/3-x/3 ₎ ]02, where x=1/6, 1/4, 1/3 and 5/12 do not always follow this conventional wisdom. These workers report that when cells containing lithium-nickelate materials are charged at voltages up to 4.45V, this causes lithium to be removed until the Mn oxidation state reaches 4+; thus giving an expected charge capacity of 2x. However, when lithium cells where x < ½ are charged to higher voltages, e.g. between 4.5 and 4.7V, they exhibit a long plateau approximately corresponding to 1-2x and subsequent to this plateau, these materials reversibly cycle at capacities over 225mAh/g. Put simply, lithium-containing compounds of the formula Li[NixLi ₍₁ 3_2x/3 ₎ Mn(2/3-x/3 ₎ ]02 where the amount of nickel is less than 0.5, exhibit a significantly higher charge capacity than would be expected from conventional theoretical calculation. By contrast, Lu and Dahn note that a similar increase in charge capacity is not observed when x = 0.5, i.e. LiNio.5Mno.5O2, as there is sufficient nickel present to remove all of the lithium. To explain the higher than expected charge capacity phenomenon, Lu and Dahn demonstrated that their lithium materials undergo a nonreversible loss of oxygen when they are charged to higher voltages, and these oxygen deficient materials then react reversibly with lithium. Notwithstanding this, although this increased charge capacity is a potentially interesting discovery, the commercial utility of such compounds is hindered by the high cost of lithium, as discussed above.

US2007/0218361 teaches a sodium ion secondary battery which comprises a positive electrode active material which includes a sodium-containing transition metal oxide Na _a Li M _x 02±a- The M includes at least two of manganese, iron, cobalt and nickel. In such materials it is apparently extremely important that the amount of sodium is neither too large (otherwise excess sodium oxide or sodium hydride is produced which causes the positive electrode to become highly moisture absorptive) nor too small. In the latter case the amount of sodium ions that can be intercalated and de-intercalated is said to be reduced and this results in a high discharge capacity not being able to be obtained. This prior art describes that the optimum amount of sodium, i.e. the value of a, is preferably 0.6 to 1.1 , further preferably 0.9 to 1.1 and more preferably 1. Meanwhile the amount of oxygen is also described as critical to performance. Too much oxygen is said to occupy transition metal and/or alkaline metal sites, and presumably this will hinder re-intercalation of the sodium ions during charge/discharge. Too little oxygen is said to produce material with a crystal structure with lots of defects. The optimum range for a is from 0 to 0.1. Another feature of the specific examples described in US2007/0218361 is that they all contain manganese in oxidation state +3. This is to allow sodium extraction by a manganese oxidation process (Mn ⁺³ →Mn ⁴⁺ ).

The present invention aims to provide a cost effective electrode that contains an active material that is capable of achieving a considerably higher specific charge capacity than would be expected from conventional theoretical calculations. Further, it is desirable for such active materials to be straightforward to manufacture and easy to handle and store. Further still, the present invention aims to provide an electrode which is able to be recharged multiple times without significant loss in charge capacity. In particular the present invention will provide an energy storage device that utilises an electrode of the present invention for use in a sodium-ion cell or a sodium metal cell.

Therefore in a first aspect, the present invention provides an electrode containing one or more active materials comprising:

A _a M ¹ v M ² w M ³ x M ⁴ _y M ⁵ z 0 _2-c (Formula 1)

wherein

A comprises either sodium or a mixed alkali metal in which sodium is the major constituent;

M ¹ is nickel in oxidation state less than or equal to 4+,

M ² comprises a metal in oxidation state less than or equal to 4+,

M ³ comprises a metal in oxidation state 2+,

M ⁴ comprises a metal in oxidation state less than or equal to 4+, and

M ⁵ comprises a metal in oxidation state 3+

wherein

0≤ a≤ 1 , preferably 0≤ a < 0.5

v > 0, preferably 0 < v < 0.5

at least one of w and y is > 0

x≥ 0, preferably x > 0

z≥ 0 c > 0.1 , preferably 0.1 < c≤ 0.5

where (a, v, w, x, y, z and c) are chosen to maintain electroneutrality.

Preferred electrodes of the present invention contain one or more active materials comprising:

A _a M ¹ v M ² w M ³ x M ⁴ _y M ⁵ z 0 _2-c (Formula 1)

wherein

A comprises either sodium or a mixed alkali metal in which sodium is the major constituent;

M ¹ is nickel in oxidation state 4+,

M ² comprises a metal in oxidation state 4+,

M ³ comprises a metal in oxidation state 2+,

M ⁴ comprises a metal in oxidation state 4+, and

M ⁵ comprises a metal in oxidation state 3+

wherein

0≤ a < 1 , preferably 0≤ a < 0.5

v > 0, preferably 0 < v < 0.5

at least one of w and y is > 0

x≥ 0, preferably x > 0

z≥ 0

c > 0.1 , preferably 0.1 < c≤ 0.5

where (a, v, w, x, y, z and c) are chosen to maintain electroneutrality.

Particularly preferred electrodes contain one or more active materials comprising:

A _a M ¹ v M ² w M ³ x M ⁴ _y M ⁵ z 0 _2-c (Formula 1)

wherein

A comprises either sodium or a mixed alkali metal in which sodium is the major constituent;

M ¹ is nickel in oxidation state less than 4+,

M ² comprises a metal in oxidation state less than or equal to 4+,

M ³ comprises a metal in oxidation state 2+,

M ⁴ comprises a metal in oxidation state less than or equal to 4+, and

M ⁵ comprises a metal in oxidation state 3+

wherein

0≤ a≤ 1 , preferably 0≤ a < 0.5

v > 0, preferably 0 < v < 0.5 at least one of w and y is > 0

x≥ 0, preferably x > 0

z≥ 0

c > 0.1 , preferably 0.1 < c≤ 0.5

where (a, v, w, x, y, z and c) are chosen to maintain electroneutrality.

Metals M ² and M ⁴ may be the same or a different metal in oxidation state less than or equal to 4+. Moreover M ² and M ⁴ are interchangeable with each other. When M ₂ = M ⁴ then Formula 1 may be written either as:

A _a M ¹ v M ² w M ³ x M\ M ⁵ z 0 _2-c ,

or

A _a M ¹ v M ² _W+Y M ³ x M ⁵ z 0 _2-c ,

or

A _a M ¹ v M ³ x M ⁴ Y _+W M ⁵ z 0 _2-c , and all of these forms of the equation are to be regarded as equivalent.

Compounds having sodium alone as the chosen alkali metal are especially preferred.

Preferably, M ² comprises a metal selected from one or more of manganese, titanium and zirconium; M ³ comprises a metal selected from one or more of magnesium, calcium, copper, zinc and cobalt; M ⁴ comprises a metal selected from one or more of manganese, titanium and zirconium; and M ⁵ comprises a metal selected from one or more of aluminium, iron, cobalt, molybdenum, chromium, vanadium, scandium and yttrium.

As discussed below, the active materials of Formula 1 are surprisingly capable of providing a specific charge capacity that is considerably higher than that predicted from conventional theoretical calculations.

Especially preferred electrodes contain one or more active materials comprising:

A _a M ¹ v M ² w M ³ x M ⁴ _y M ⁵ z 0 _2-c (Formula 1)

wherein

0≤ a≤ 1 ; preferably 0≤ a < 0.5; further preferably 0≤ a≤0.4; and highly preferably 0.1≤ a≤ 0.4;

v > 0, preferably 0 < v <0.5; 0 < w < 0.5;

0 < x < 0.5; preferably 0 < x < 0.5;

0 < y < 0.5;

z≥ 0; and

c≥ 0.11 ; preferably c≥ 0.13 and highly preferably c≥ 0.15, preferably the upper limit for c is ≤0.5,

where a, v, w, x, y, z and c are chosen to maintain electroneutrality.

In a highly preferred active material of the present invention, (v+w) < 0.8, especially when M ² = Mn and/or Ti.

Advantageous electrodes of the present invention contain one or more active materials comprising: io.25 io.25Mgo.25 no.25O1.75;

N io.33 n ₀ .33Mg ₀ .i 67T10.167Ο1.83;

Nio.33Mno.33Cuo.i67Tio.i 67O1.833;

N io.33Mno.33Zno.i 6 T10.167Ο1.833;

Nio.33Mno.33Cao.i67Tio. 67O1.833 ^' ,

Nio.45Mno.45Mgo.05Tio.05O1. ₉₅ ;

Nio.35Mno.35Mgo.15Tio.15O1.85;

Nio.3Mno.3Mgo.2Tio.2Oi 8;

Nio.4Mno.4Mgo.1Tio.1O1 9;

Nio.4Mno.4Mgo.05Tio.05Alo.1 O1 9;

Nio.35Mno.35Mgo.05Tio.05Alo.2O1.85;

Nlo.33Mno.33Mgo.11TI0.11 Alo.1101.83;

Nio.3Mno.3Mgo.05Tio.05Alo.3O1 ₈ ;

Nio.35Mno.35Mgo.1Tio.1Alo.1 O1.85;

Nio.3Mno.3Mgo.1Tio.1Alo.2Oi 8;

Ni0.33Mn0.33AI0.33O1.83;

Nio.35Mgo.15Mno.5O1.85;

Nio.333Mgo.i67Mno.5O1.333;

Nio.35Mgo.15Mno.5O1.85;

Nio.3Mgo.2Mno.5Oi 8; Nao.2Nio.25Tio.25Mgo.25Mno.25O1.85 ^' ,

Nao.2Nio.333Mno.333Mgo.i 67Tio.i 67O1.933;

Nao.2Nio.333Mno.333CUo.i67Tio.i 67O1.933;

Nao.2Nio.333Mno.333Zno.i6 Tio.i 67O1.933;

Nao.2Nio.333Mno.333Cao.i 67Tio.i 67O1.933;

Nao.05Nio.45Mno.45Mgo.05Tio.05O1.975;

Nao.2Nio.35Mno.35Mgo.15Tio.15O1.95;

Nao.2Nio.3Mno.3Mgo.2Tio.2O1 9;

Nao.1 Nio.4Mno.4Mgo.1Tio.1 O1.95;

Nao.1 Nio.4Mno.4Mgo.05Tio.05Alo.1 O1.95;

Na0.2Ni0.35Mn0.35Mg0.05Ti0.05AI0.2O1.95;

Nao.2Nio.333Mno.333Mgo.i 11T10.111 AI0.1101.933;

Nao.2Nio.3Mno.3Mgo.05Tio.05Alo.3O1 9;

Nao.2Nio.35Mno.35Mgo.1Tio.1Alo.1 O1. ₉₅ ;

Nao.2Nio.3Mno.3Mgo.1Tio.1Alo.2O1 9;

Nao.2Nio.333Mno.333Alo.333O1.933;

Na ₀ .i Ni ₀ . ₄ Mg ₀ .i Mn _{0 5} Oi ₉₅ ;

Nao.2Nio.35Mgo.15Mno.5O1.95;

Nao.2Nio.333Mgo.i67Mno.5O1. ₉₃₃ ;

Na ₀ .2Nio.3Mgo. ₂ Mno.50i.9;

Nao.1 Nio.4Mgo.1 Mno.5O1 ₉₅ ;

Nao.2Nio.35Mgo.15Mno.5O1.95; and

Nao.2Nio.3Mgo.2Mno.5O1.9. with Nio.33Mn ₀ .33M ³ o.i67Tio.i670i .833 and Na ₀ .2Nio.33Mn ₀ .33M ³ o.i67Tio.i6 0i .933 being the most favourable materials.

The active materials used in the electrodes of the present invention may be prepared using any known method. However, a particularly convenient method involves effecting the net loss of Na ₂ 0 from one or more active cathode materials of Formula 2, where Formula 2 is defined as

A'a M ¹ v'M ² w'M ³ x'M ⁴ y'M ⁵ z'0 ₂ (Formula 2)

wherein

A' comprises either sodium or a mixed alkali metal in which sodium is the major constituent;

M ¹ is nickel in oxidation state 2+, M ² comprises a metal in oxidation state 4+,

M ³ comprises a metal in oxidation state 2+,

M ⁴ comprises a metal in oxidation state 4+, and

M ⁵ comprises a metal in oxidation state 3+

1≤ a' < 2;

0 < v' < 0.5;

0 < w'≤0.5;

0≤x' < 0.5;

0≤ y' < 0.5;

z' > 0;

and wherein v', w', x', y' and z' are all chosen to maintain electroneutrality.

A highly efficient process of producing compositions of Formula 1 comprises: charging an electrochemical cell containing one or more active cathode materials of Formula 2 beyond the conventional theoretical specific capacity as determined by the Ni ²⁺ /Ni ⁴⁺ redox couple; wherein Formula 2 is defined as:

A'a ^< M ¹ v'M ² w'M ³ x ^< M ⁴ _y .M ⁵ _z .02 (Formula 2)

wherein

A' comprises either sodium or a mixed alkali metal in which sodium is the major constituent;

M ¹ is nickel in oxidation state 2+,

M ² comprises a metal in oxidation state 4+,

M ³ comprises a metal in oxidation state 2+,

M ⁴ comprises a metal in oxidation state 4+, and

M ⁵ comprises a metal in oxidation state 3+

1≤ a' < 2;

0 < v' < 0.5;

0 < w'≤0.5;

0≤x' < 0.5;

0≤ y' < 0.5;

z'≥ 0;

and wherein v', w', x', y' and z' are all chosen to maintain electroneutrality.

The compounds of Formula 2 are disclosed in Applicant's patent applications GB1212263.6, GB1212268.5 and GB1212261.0 which documents are incorporated herein by reference. The process of overcharging involves charging the electrochemical cell containing the one or more active cathode materials to a specific capacity greater than 180mAh/g. Typically this means charging the Na-ion cell to at least 4.2V. Preferably the overcharging process is the first charge performed on the active materials although it may be possible to charge the active materials to their "normal" specific capacity first and follow this with a process in which the materials are overcharged.

Charging doped sodium nickelate compounds of Formula 2 up to their theoretical capacity causes the oxidation of Ni ²⁺ to Ni ⁴⁺ and the removal of a portion of the sodium ions from the lattice. By contrast, charging such compounds above their conventional theoretical capacity, i.e. "over-charging" them, produces materials of Formula 1 in which

M ¹ is nickel in oxidation state 4+,

M ² comprises a metal in oxidation state 4+,

M ⁴ comprises a metal in oxidation state 4+,

0≤ a <1 , preferably 0≤ a < 0.5 and

C > 0.1 , preferably 0.1 < c≤ 0.5.

The enhanced charge capacity (or anomalous charge capacity) observed when materials of Formula 2 are "over-charged" is due to several changes that occur as the over-charging is taking place. Broadly speaking this is a combination of the conventional (or "normal") charge process using just the oxidation of Ni ²⁺ to Ni ⁴⁺ , together with other structural changes that occur exclusively at the higher voltages, such as the removal of a higher than expected (from conventional theory) number of sodium ions and also an irreversible partial loss of oxygen to yield 0 ₂ . _c -containing active materials.

The present invention therefore provides a method of increasing the specific charge capacity of an oxide-containing cathode composition for use in a Na-ion cell comprising:

charging the oxide-containing cathode composition in an electrochemical cell to cause the loss of oxygen from the oxide-containing cathode composition and thereby form an oxygen deficient oxide-containing cathode composition.

Further the present invention provides an oxygen deficient oxide-containing cathode composition, for example made using the above method.

Advantageously the present invention provides an electrode comprising an oxygen deficient oxide-containing cathode composition, for example made using the above method. The mechanism by which oxygen is lost from the active cathode composition during the overcharging process operates most efficiently when the oxide-containing active cathode composition comprises a layered structure. Layered A _x M0 ₂ materials are known to adopt several, very similar structural types, the most well-known of which have been categorised by Delmas et al. Physica B+C, 91 , (1980), 81 , as "03, P2 and P3". These notations describe the stacking of the layers; the letter refers to the way in which the oxygen ions are coordinated to the alkali ions, with the letter O denoting an octahedral coordination and the letter P denoting a prismatic coordination. The number refers to the number of M0 ₂ layers contained in the unit cell. In the case of the uncharged active cathode materials used in the electrodes of present invention, it is preferred that they adopt a layered a-NaFe0 ₂ type structure (space group R-3m, no. 166), which has an 03 crystal structure. Following charging, these materials are thought to adopt 03, P2 or P3 or modified versions of these crystal structures, although it is also possible that other crystal structures may be adopted.

The Applicant has noted that when layered active cathode materials, for example of Formula 2, are over-charged in a Na-ion cell, they undergo a structural change, which produces a reduction in the volume of the unit cell as compared with both the pristine precursor material, e.g. materials of Formula 2 before charging, and the precursor material after it has been charged to its conventional theoretical capacity, e.g. based on Ni ²⁺ to Ni ⁴⁺ . It is believed that such unit cell volume reduction is consistent with the loss of alkali metal-ions and oxygen loss. This is discussed further below in the relation to the Specific Examples. Another characteristic of overcharged oxide-containing cathode compositions is that they appear to lose crystallinity (compared with to the uncharged material); this is shown by the FWHM of the 003 peak of the XRD pattern.

A determination of when a material is said to be "overcharged" is made by observing the voltage curve for the particular material. At capacities less than or close to the conventional theoretical maximum charge capacity (i.e. the Ni ²⁺ /Ni ⁴⁺ redox couple) a first voltage feature is seen. However the curve continues to rise to a second voltage feature upon further or "over" charging. The materials are said to be "over-charged" from the beginning of this second voltage feature.

Therefore the present invention provides a method of optimising the specific charge capacity of an oxide-containing cathode composition for use in a sodium-ion cell comprising charging the composition onto at least a portion of a voltage feature which is observed as a result of the oxide-containing cathode composition being charged above the conventional theoretical maximum charge capacity.

The present invention also provides a method of optimising the specific charge capacity of an oxide-containing cathode composition in a sodium-ion cell comprising:

a) charging the sodium-ion cell beyond the conventional theoretical capacity of the

cathode based on the Ni ²⁺ /Ni ⁴⁺ redox couple; and b) degassing the Na-ion cell to remove gasses formed during the charging process.

A doped sodium nickelate of Formula 2 charged to its theoretical specific capacity limit based on the Ni ²⁺ /Ni ⁴⁺ redox couple, will re-yield a compound with no oxygen loss when discharged; i.e. at this level, charging is seen to be reversible. However, discharging compounds of Formula 2 that have been "over-charged" results in producing materials of Formula 1 in which

M ¹ is nickel in oxidation state less than 4+,

at least one of M ² and M ⁴ comprises a metal in oxidation state less than 4+,

0 < a≤ 1

0 < v < 0.5 and

c > 0.1 , preferably 0.1 < c < 0.5.

Consequently, during the discharge of over-charged compounds of Formula 2 the nickel is reduced from oxidation state 4+ to oxidation state 2+, some or all of the sodium-ions are reinserted into the lattice and the lost oxygen atoms are not replaced. Moreover, although the metals M ² and/or M ⁴ do not contribute anything to the initial overcharging process, upon discharge of the overcharged material it is found that at least one of the metals M ² and/or M ⁴ is also reduced to an oxidation state less than 4+. This allows some or all of the sodium ions that were removed during the first "over-charge" to be reinserted into the material, and therefore produces a material with a higher discharge capacity. It is clear from this description that the role of the Mn ⁴⁺ (or Ti ⁴⁺ ) in the uncharged materials of the present invention is to allow for the reduction (Na insertion) process during cell discharge. This is in direct contrast to the role of the manganese in US2007/0218361 which is to allow Na extraction by the Mn oxidation process Mn ³⁺ → Mn ⁴⁺ during the first charge process. The role of the M ³ and M ⁵ metals is to reduce the amount of nickel in the doped nickelate materials of Formula 1 , thereby allowing the "over charge" mechanism to take place as oxygen is released from the structure. The presence of the M ³ and M ⁵ metals also reduces the relative formula mass (formula weight) of these materials, thereby further improving the cost effectiveness of the electrode materials. In order to achieve anomalous capacity, there must be over twice the amount of sodium as there is nickel in the pristine materials (e.g. Formula 2) so that, when all Ni ²⁺ has oxidised to Ni ⁴⁺ , there is still some Na left in the material, which can then be removed when the material is overcharged and releases oxygen. This explains why anomalous capacity is not observed for NaNio.5Mno.5O2; in this case there is enough Ni ²⁺ present to remove all of the Na. In each of the compounds of Formula 2 where a' = 1 , following the requirement above, there has to be less than 0.5 Ni and the overall material must be electroneutral. Satisfaction of these two conditions is facilitated by the presence of an M ³ metal, and/or an M ⁵ metal.

When oxide-containing cathode compositions have undergone the over-charge/discharge process described, it is found that they may be cycled in the Na-ion cell between normal voltage limits for the Ni ²⁺ /Ni ⁴⁺ redox couple, typically this might be expected to be between 1 and 4V, and no further unconventional voltage plateaux are formed. Moreover, cycling the over-charged/discharged compositions at normal voltage limits maintains a higher specific capacity than would be expected from conventional theoretical calculations, i.e. the overcharged/discharged oxide-containing cathode compositions perform better than the pristine compounds when cycled using the same normal voltage limits.

Thus the present invention provides a method of using an oxide-containing cathode composition according to Formula 1 in a Na-ion cell comprising:

a) charging a Na-ion cell comprising an oxide-containing cathode composition

according to Formula 2 to a cell voltage beyond the conventional theoretical capacity of the cathode material based on the Ni ²⁺ /Ni ⁴⁺ redox couple; b) degassing the Na-ion cell to remove gasses formed during the charging process; and c) cycling the resulting Na-ion cell, which now comprises an oxide-containing cathode composition according to Formula 1 , over a voltage range within the normal voltage limits of the Ni ²⁺ /Ni ⁴⁺ redox couple.

As mentioned above, it is envisaged that the optimized oxide-containing cathode composition of the present invention may be prepared by a method that does not rely on overcharging a compound of Formula 2. Thus the present invention includes an optimised oxide-containing cathode composition made by any route.

The electrodes according to the present invention are suitable for use in many different applications, for example energy storage devices, rechargeable batteries, electrochemical devices and electrochromic devices.

Advantageously, the electrodes according to the invention are used in conjunction with a counter electrode and one or more electrolyte materials. The electrolyte materials may be any conventional or known materials and may comprise either aqueous electrolyte(s) or nonaqueous electrolyte(s).

The present invention also provides an energy storage device that utilises an electrode comprising the active materials described above, and particularly an energy storage device for use as a sodium-ion cell and/or a cell in which sodium is the major mobile ion; a cell in which sodium is the major alkali metal-ion; either of which may be used in a non-aqueous or an aqueous electrolyte system.

It is also possible to convert sodium-ion derivatives into mixed lithium-ion/sodium-ion materials using an ion exchange process.

Typical ways to achieve Na to Li-ion exchange include:

1. Mixing the sodium-ion material with a lithium-containing material e.g. LiN0 ₃ , heating to above the melting point of LiN0 ₃ (264°C), cooling and then washing to remove the excess UNO ₃ and side-reaction product

2. Treating the Na-ion material with an aqueous solution of lithium salts, for example 1 M LiCI in water; and

3. Treating the Na-ion material with a non-aqueous solution of lithium salts, for example LiBr in one or more aliphatic alcohols such as hexanol, propanol etc.

BRIEF DESCRIPTION OF THE DRAWINGS

The present invention will now be described with reference to the following figures in which: FIGURE 1 (A) is a comparison of XRD profiles for a) the precursor starting material, NaNio.25Tio.25Mgo.25Mno.25O2, b) an electrode, originally containing NaNio.25Tio.25Mgo.25Mno.25O2, recovered from a cell after charging to 120mAh/g and c) an electrode, originally containing NaNio.25Tio.25Mgo.25Mno.25O2 recovered from a cell after charging to 230mAh/g;

FIGURE 1 (B) shows the First Charge Process (Na-ion cell Voltage [V] versus Cathode Specific Charge Capacity [mAh/g]) for a Hard Carbon//NaNio _.25 Tio _.25 Mg _0.25 Mn _0.25 0 ₂ cell using an electrolyte of 0.5M sodium perchlorate in propylene carbonate (PC), charging to 230mAh/g;

FIGURE 2(A) shows the Charge-Discharge Voltage Profile for the first 4 cycles (Na-ion Cell Voltage [V] versus Cumulative Cathode Specific Capacity [mAh/g]) for a Hard Carbon//NaNio.35Mn ₀ .35Mg ₀ .i5Tio.i502 cell, cycled between 1.5 to 4.2V;

FIGURE 2(B) shows the Cycle Life Performance (Cathode Specific Capacity [mAh/g] versus Cycle Number) for a Hard Carbon// NaNio.35Mno.35Mgo.15Tio.15O2 cell, cycled between 1.5 to 4.2V;

FIGURE 2(C) shows the Third Cycle Discharge Voltage Profile (Na-ion Cell Voltage [V] versus Cathode Specific Capacity [mAh/g]) for a Hard Carbon//NaNio. ₃ 5Mn ₀ . ₃ 5Mg ₀ .i5Tio.i502 cell, cycled between 1.5 to 4.2V;

FIGURE 2(D) shows the Third Cycle Differential Capacity Profiles (Differential Capacity [mAh/g/V]) versus Na-ion Cell voltage [V]) for a Hard Carbon//NaNio. ₃ 5Mn ₀ . ₃ 5Mg ₀ .i5Tio.i502 cell, cycled between 1.5 to 4.2V;

FIGURE 3(A) shows the Charge-Discharge Voltage Profiles for the first 4 cycles (Na-ion cell Voltage [V] versus Cumulative Cathode Specific Capacity [mAh/g]) for a Hard Carbon//NaNio.35Mn ₀ .35Mg ₀ .i5Tio.i502 cell, cycled between 1.5 to 4.4V;

FIGURE 3(B) shows the Cycle Life Performance (Cathode Specific Capacity [mAh/g] versus Cycle Number) for a Hard Carbon// NaNio.35Mno.35Mgo.15Tio.15O2 cell, cycled between 1.5 to 4.4V;

FIGURE 3(C) shows the Third Cycle Discharge Voltage Profile (Na-ion Cell Voltage [V] versus Cathode Specific Capacity [mAh/g]) for a Hard Carbon//NaNio. ₃ 5Mn ₀ . ₃ 5Mg ₀ .i5Tio.i502 cell, cycled between 1.5 to 4.4V; FIGURE 3(D) shows the Third Cycle Differential Capacity Profiles (Differential Capacity [mAh/g/V]) versus Na-ion Cell voltage [V]) for a Hard Carbon//NaNio.35Mn ₀ .35Mgo.i5Tio.i502 cell, cycled between 1.5 to 4.4V;

FIGURE 4(A) shows the changes observed in the a and c cell parameters for NaNi25Tio.25Mgo.25Mrio.25O2 as pristine material, material charged to 120mAh/g and material charged to 230 mAh/g; and

FIGURE 4(B) shows the changes in unit cell volume for NaNi25Tio.25Mgo.25Mno.25O2 as pristine material, material charged to 120mAh/g and material charged to 230 mAh/g.

DETAILED DESCRIPTION

Any convenient process may be used to make the precursor materials (Formula 2) described above. For example, the following general method may be used:

GENERAL METHOD:

1) Intimately mix together the starting materials in the correct stoichiometric ratio and press into a pellet.

2) Heat the resulting mixture in a furnace using either an ambient air atmosphere, or a flowing inert atmosphere (e.g. argon or nitrogen), at a furnace temperature of between 400°C and 1500°C until reaction product forms.

3) Allow the product to cool before grinding it to a powder.

Table 1 below lists the starting materials and heating conditions used to prepare example precursor materials 1 to 12 of Formula 2.

TABLE 1

PRECURSOR MATERIAL OF STARTING FURNACE FORMULA 2 MATERIALS CONDITIONS

NaNio.45Mno.45Mgo.05Tio.05O2 Na ₂ C0 ₃ 1) Air/800°C, dwell

N1CO3 time of 8 hours.

Mg(OH) ₂

MnO _z

2) Air/900°C, dwell

Ti0 ₂ time of 8 hours.

NaNio 4oMn ₀ .4oMg ₀ ioTio 10O2 Na ₂ C0 ₃ 1) Air/800"C, dwell

N1CO3 time of 8 hours.

Mg(OH) ₂

MnO _z

2) Air/900°C, dwell

Ti0 ₂

time of 8 hours.

NaNio 35Mn ₀ .35Mg ₀ .i5Tio 15Ο2 Na ₂ C0 ₃ 1) Air/900°C, dwell

N1CO3 time of 8 hours.

Mn0 ₂

Mg(OH) ₂

2) Air/900°C, dwell

Ti0 ₂

time of 8 hours. iv NaNio 3oMn ₀ . ₃ oMgo. ₂ oTio ₂ o0 ₂ Na ₂ C0 ₃ 1) Air/900°C, dwell

N1CO3 time of 8 hours

Mn0 ₂

Mg(OH) ₂

2) Air/900°C, dwell

Ti0 ₂

time of 8 hours.

V NaNi ₀ . ₄ oTio ₅ oMgo io0 ₂ Na ₂ C0 ₂ 1) Air/900°C, dwell

N1CO3 time of 8 hours Ti0 ₂

Mg(OH) ₂

2) /900°C, dwell time of 8 hours. vi NaNi ₀ . ₄ oTio. ₄ oMgo.ioMn ₀ 10O2 Na ₂ C0 ₃ 1) Air/900°C, dwell

N1CO3 time of 8 hours

Ti0 ₂ ,

Mg(OH) ₂

2) Air/900°C, dwell

Mn0 ₂

time of 8 hours. vii NaNio ₄₀ Mno. ₄₀ Mgo. ₀₅ Tio. ₀₅ Alo.1O2 Na ₂ C0 ₃ 1) Air/800°C, dwell

N1CO3 time of 8 hours

Mg(OH) ₂

MnO _z

2) Air/900°C, dwell

Ti0 ₂

AI(OH) ₃ time of 8 hours. viii NaNio .45Mno.45Cuo.05Tio.05O2 Na ₂ C0 ₃ 1) Air/900°C, dwell

N1CO3 time of 8 hours

Mn0 ₂

CuO

2) Air/900°C, dwell

Ti0 ₂

time of 8 hours.

Example 1 : The Preparation of Nin _? sTin _? sMg _{n ?} sMnn _? sO ₇ s

Active precursor doped nickelate material, NaNio.25Tio.25Mgo.25Mno.25O2 (compound xi in Table 1), prepared using the general method described above, was made into a hard carbon anode//NaNio _.25 Tio _.25 Mg _0.25 Mn _0.25 0 ₂ cathode cell using an electrolyte comprising 0.5M NaCI0 ₄ in propylene carbonate, see Figure 1 (A)a). Following a charge process to 230mAh/g, the cathode material was removed from the cell, washed several times in clean dimethyl carbonate and then dried at 70°C. Looking at Figure 1 (A)c), it is clear that the material obtained following the charge process to 230mAh/g is not the same as that shown in either Figure 1 (A)b) (charged to 120mAh/g) or that shown by the original active precursor doped nickelate material, NaNio.25Tio.25Mgo.25Mno.25O2 (as shown in Figure 1 (A) a). The proposed composition for the product obtained following the charge process to 230mAh/g is Nio.25Tio.25Mgo.25Mno.25O1.75, as determined by the following mass loss experiment.

Mass Loss Experiment to determine the composition of Nio ?sTio ?sMgo ?sMrio ?sOi 75- Cathode material = NaNio.25Tio.25Mgo.25Mno.25O2 Anode = Hard Carbon

Electrolyte = 0.5 M NaCI0 ₄ in Propylene Carbonate.

Active Mass of cathode in as-prepared cell = 21.5 mg

Following the charge process to 230 mAh/g, the cathode electrode disk was removed from the cell, washed several times in clean dimethyl carbonate to remove the electrolyte and then dried at 70°C.

The Active Mass of the washed cathode after first charge process shown in Figure 1 (B) = 16.1 mg

Thus the Active Mass loss = (21.5 mg - 16.1 mg) = 5.4 mg which equates to:

% mass loss = 25.1 %

If charge process was just by Na-ion extraction then mass loss should be:

Starting composition = NaNio.25Tio.25Mgo.25Mno.25O2

Molecular weight = 101.4 g/mol

Based on only the Ni ²⁺ to Ni ⁴⁺ redox process, on cell charge it is only possible to extract 0.5 Na-ion per formula unit. i.e.

NaNio.25Tio.25Mgo.25Mno.25O2 Nao.5Nio.25Tio.25Mgo.25Mno.25O2 + 0.5 Na ⁺ + 0.5 e ^" (1)

Therefore, the theoretical capacity based on this reaction (1) may be given by the following: Theoretical Specific Capacity = (0.5 x 96485)/(101.4 x 3.6) = 132 mAh/g

Thus the Expected %Mass loss for reaction (1 ) = (11.5/101.4) x 100 = 11.3 %

The charge process as shown in Figure 1 (B), corresponds to an actual cathode specific capacity of 230 mAh/g - i.e. far in excess of the expected theoretical specific capacity of 132 mAh/g.

Thus the following overall charge mechanism is:

NaNio.25Tio.25Mgo.25Mno.25O2 Nao.5Nio.25Tio.25Mgo.25Mno.25O2 + 0.5 Na ⁺ + 0.5 e ^" (1) followed by:

Nao.5Nio.25Tio.25Mgo.25Mno.25O2 Nio.25Tio.25Mgo.25Mno.25O1.75+ 0.5 Na ⁺ + 0.125 0 ₂ + 0.5 e ^" (2)

Looking at an overall process that relies on the complete extraction of Na:

NaNio.25Tio.25Mgo.25Mno.25O2 Ni ₀ .25Tio.25Mgo.25Mn ₀ .250i .75 ⁺ 1 -0 Na ⁺ + 0.125 0 ₂ + 1.0 e ^" (3) The theoretical capacity based on this reaction (3) may be given by:

Theoretical Specific Capacity = (1.0 x 96485)7(101.4 x 3.6) = 264 mAh/g. This compares well with the actual capacity achieved of 230 mAh/g.

Looking also at the Expected mass loss for reaction (3) = ((101.4-74.4)/101.4) x 100 % = 26.5 %. Again this percentage mass loss is very close to the 25 .1 % which is observed.

Thus on the basis that there is close correspondence between theoretical and actual results for both Specific Capacity and Expected Mass Loss, the Applicant has been able to determine with a high degree of certainty that Nio.25Tio.25Mgo.25Mno.25O1.75 is obtained when NaNio.25Tio.25Mgo.25Mno.25O2 is charged to the end of the second unconventional voltage plateau.

From reaction (3) above, it is proposed that the anomalous capacity arises as a result of the loss of Na ₂ 0, i.e. the production of active Na ⁺ -ions plus the liberation of 0 ₂ , and this produces a new layered oxide active material, Nio.25Tio.25Mgo.25Mno.25O1.75, as confirmed by comparing XRD Figure 1Aa) with that of Figure 1Ac). This is surprising because it is not the usual charging mechanism i.e. a simple Na ⁺ extraction from the cathode, but is a structural change that releases Na ⁺ and oxygen from the material to produce a new composition.

Example 2: The Preparation of Na _a Nio sMno sMgo isTio isO?-r.

The data shown in Figure 2A, 2B, 2C and 2D are derived from the constant current cycling data for a NaNio.35Mno.35Mgo.15Tio.15O2 (compound iii in Table 1) active precursor cathode material in a Na-ion cell where this cathode material is coupled with a Hard Carbon (Carbotron P/J) anode material. The electrolyte used is a 0.5 M solution of NaCI0 ₄ in propylene carbonate. The constant current data are collected at an approximate current density of 0.10 mA/cm ² between voltage limits of 1.50 and 4.20 V. To fully charge the Na-ion cell it, is potentiostatically held at 4.2 V at the end of the constant current charging process. The testing is carried out at room temperature. During the cell charging process, sodium- ions are extracted from the cathode active material, and inserted into the Hard Carbon anode. During the subsequent discharge process, sodium-ions are extracted from the Hard Carbon and re-inserted into the cathode active material.

Figure 2A shows the first four charge-discharge cycles (Na-ion Cell Voltage [V] versus Cumulative Cathode Specific Capacity [mAh/g]) for the Hard Carbon//NaNio. ₃ 5Mn ₀ . ₃ 5Mg ₀ .i5Tio.i502 cell. During the first charge process an anomalously high charge capacity is realized - a cathode specific capacity of 215 mAh/g is achieved - this figure is significantly larger than the theoretical specific capacity (which amounts to 178 mAh/g, based on the Ni ²⁺ to Ni ⁴⁺ redox couple) for the NaNio.35Mno.35Mgo.15Tio.15O2 active material. In particular, a two section voltage profile is clearly evident during this initial cell charge step. At cell voltages lower than about 4.0 V a sloping profile is evident, presumably reflecting the conventional Na extraction process from the NaNio.35Mno.35Mgo.15Tio.15O2 precursor material. At cell voltages greater than about 4.0 V, a more flat voltage region is evident which presumably reflects a new Na extraction process (i.e. not based on the Ni ²⁺ to Ni ⁴⁺ redox couple) occurring for the NaNio.35Mno.35Mgo.15Tio.15O2 precursor material. The active material that produces the anomalously high specific capacity is understood to be NaaNio.35Mno.35Mgo.15Tio.15O2.c- Importantly, as demonstrated below, this two section charge behaviour is not evident on subsequent cell charge profiles.

Figure 2(B) shows the cycle life performance (Cathode Specific Capacity [mAh/g] versus Cycle Number) for a Hard Carbon//NaNio. ₃ 5Mn ₀ . ₃ 5Mg ₀ .i5Tio.i502 cell. The cell shows good reversibility with the delivered cathode specific capacity reaching around 135 mAh/g after 4 cycles.

Figure 2(C) shows the third cycle discharge voltage profile (Na-ion Cell Voltage [V] versus Cathode Specific Capacity [mAh/g]) for the Hard Carbon//NaNio. ₃ 5Mn ₀ . ₃ 5Mg ₀ .i5Tio.i502 cell. The cathode specific capacity in this cycle corresponds to 130 mAh/g.

Figure 2(D) shows the third cycle differential capacity profiles (Differential Capacity [mAh/g/V] versus Na-ion Cell Voltage [V]) for the Hard Carbon//NaNio. ₃ 5Mn ₀ . ₃ 5Mg ₀ .i5Tio.i502 cell. These symmetrical data demonstrate the excellent reversibility of the ion-insertion reactions in this Na-ion cell.

Hard Carbon//NaNio.35Mno.35Mgo.i5Tio.i50 ₂ (Material =X0474A) Cell#204064

The data shown in Figures 3(A), 3(B), 3(C) and 3(D) are derived from the constant current cycling data for a NaNio.35Mno.35Mgo.15Tio.15O2 (compound iii in Table 1)) active precursor material in a Na-ion cell where this cathode material is coupled with a Hard Carbon (Carbotron P/J) anode material. The electrolyte used is a 0.5 M solution of NaCI0 ₄ in propylene carbonate. The constant current data are collected at an approximate current density of 0.10 mA/cm ² between voltage limits of 1.50 V and 4.40 V. To fully charge the cell the Na-ion cell is potentiostatically held at 4.4 V at the end of the constant current charging process. The testing is carried out at room temperature. During the cell charging process, sodium ions are extracted from the cathode active material, and inserted into the Hard Carbon anode. During the subsequent discharge process, sodium-ions are extracted from the Hard Carbon and re-inserted into the cathode active material. Figure 3(A) shows the first four charge-discharge cycles (Na-ion Cell Voltage [V] versus Cumulative Cathode Specific Capacity [mAh/g]) for the Hard Carbon//NaNio _.35 Mn ₀ .35Mgo.i5Tio _. i ₅ 0 ₂ cell (Cell#204064). During the first charge process an anomalously high charge capacity is realized - a cathode specific capacity of 226 mAh/g is achieved - a figure which is significantly larger than the theoretical specific capacity (which amounts to 178 mAh/g, based on the Ni ²⁺ to Ni ⁴⁺ redox couple) for the NaNio.35Mno.35Mgo.15Tio.15O2 active material. In particular, a two section voltage profile is clearly evident during this initial cell charge step. At cell voltages lower than about 4.0 V a sloping profile is evident, presumably reflecting the conventional Na extraction process from the NaNio.35Mno.35Mgo.15Tio.15O2 active material. At cell voltages greater than about 4.0 V, a more flat voltage region is evident which presumably reflects a new Na extraction process (i.e. not based on the Ni ²⁺ to Ni ⁴⁺ redox couple) occurring for the NaNio.35Mno.35Mgo.15Tio.15O2 active precursor material. Importantly, this two section charge behaviour is not evident on subsequent cell charge profiles.

Figure 3(B) shows the cycle life performance (Cathode Specific Capacity [mAh/g] versus Cycle Number) for the Hard Carbon//NaNio _.35 Mn _0.35 Mg _0. i ₅ Tio _. i ₅ 0 ₂ cell. The cell shows good reversibility with the delivered cathode specific capacity reaching around 130 mAh/g after 4 cycles.

Figure 3(C) shows the third cycle discharge voltage profile (Na-ion Cell Voltage [V] versus Cathode Specific Capacity [mAh/g]) for the Hard Carbon//NaNio _.35 Mn _0.35 Mg _0. i ₅ Tio _. i ₅ 0 ₂ cell. The cathode specific capacity in this cycle corresponds to 125 mAh/g.

Figure 3(D) shows the third cycle differential capacity profiles (Differential Capacity [mAh/g/V] versus Na-ion Cell Voltage [V]) for the Hard Carbon//NaNio _.35 Mn _0.35 Mg _0. i ₅ Tio _. i ₅ 0 ₂ cell. These symmetrical data demonstrate the excellent reversibility of the ion-insertion reactions in this Na-ion cell.

Other sodium doped nickelate electrode materials were also found to exhibit anomalous charge capacities:

TABLE 2

Precursor [Formula 2] Product [Formula 1] Theoretical Theoretical Actual

Charge Charge Charge

Capacity Capacity Capacity based on based on [mAh/g]

Ni ²⁺ /Ni ⁴⁺ removal of

only all Na

[mAh/g] [mAh/g]

Example 3. Nio.33Mno.33Mgo.i67 io.i67O1.833 169 256 239 NaNio.33Mno.33Mgo.i67Tio.i670 ₂

Example 4.

NaNio.33Mno.33CUo.167Tio.1670 ₂ io.33Mno.33Cuo.i67 io.i67O1.833 159 241 221

It is desirable for electrode materials to be safe during charge and discharge in an energy storage device. Li-ion batteries in common use today undergo safety/abuse testing, the results of such tests revealing that lithium oxide-based cathode materials are liable to liberate oxygen, which is a major contributing factor to an undesirable process known as thermal runaway. Prior to the present invention it might have been expected that a similar thermal runaway process would also be observed for sodium oxide-based materials, and that such sodium oxide-based materials would be unsafe and/or rendered completely useless by overcharging in rechargeable battery applications. However, the present invention has surprisingly demonstrated that when these sodium-based materials are overcharged, the 'labile' oxygen (which could contribute to thermal runaway) is caused to be removed from the structure; this yields the materials of the present invention which are highly thermodynamically stable and which are extremely effective and safe when used in reversible cathode materials.

Product Analysis using XRD

Analysis by X-ray diffraction techniques was conducted using a Siemens D5000 powder diffractometer to confirm that the desired target materials had been prepared, to establish the phase purity of the product material and to determine the types of impurities present. From this information it is possible to determine the unit cell lattice parameters.

The XRD operating conditions used to analyse the precursor electrode materials are as follows:

Slits sizes: 1 mm, 1 mm, 0.1 mm

Range: 2Θ = 5 ⁰ - 60 ⁰

X-ray Wavelength = 1.5418 A (Angstoms) (Cu Ka)

Speed: 1.0 seconds/step

Increment: 0.025 °

The XRD operating conditions used for ex-situ analysis of the electrodes are as follows: Slits sizes: 1 mm, 1 mm, 0.1 mm

Range: 2Θ = 10 ° - 60 °

X-ray Wavelength = 1.5418 A (Angstoms) (Cu Ka) Speed: 8.0 seconds/step

Increment: 0.015 °

Electrochemical Results

The target materials were tested either i) using a lithium metal anode test cell, or ii) using a Na-ion test cell using a hard carbon anode. It is also possible to test using a Li-ion cell with a graphite anode. Cells may be made using the following procedures:

Generic Procedure to Make a Lithium Metal Electrochemical Test Cell

The positive electrode is prepared by solvent-casting a slurry of the active material, conductive carbon, binder and solvent. The conductive carbon used is Super P (Timcal). PVdF co-polymer (e.g. Kynar Flex 2801 , Elf Atochem Inc.) is used as the binder, and acetone is employed as the solvent. The slurry is then cast onto glass and a free-standing electrode film is formed as the solvent evaporates. The electrode is then dried further at about 80°C. The electrode film contains the following components, expressed in percent by weight: 80% active material, 8% Super P carbon, and 12% Kynar 2801 binder. Optionally, an aluminium current collector may be used to contact the positive electrode. Metallic lithium on a copper current collector may be employed as the negative electrode. The electrolyte comprises one of the following: (i) a 1 M solution of LiPF ₆ in ethylene carbonate (EC) and dimethyl carbonate (DMC) in a weight ratio of 1 : 1 ; (ii) a 1 M solution of LiPF ₆ in ethylene carbonate (EC) and diethyl carbonate (DEC) in a weight ratio of 1 : 1 ; or (iii) a 1 M solution of LiPF ₆ in propylene carbonate (PC) A glass fibre separator (Whatman, GF/A) or a porous polypropylene separator (e.g. Celgard 2400) wetted by the electrolyte is interposed between the positive and negative electrodes.

A Na-ion electrochemical test cell containing the active material is constructed as follows: Generic Procedure to Make a Hard Carbon Na-ion Cell

The positive electrode is prepared by solvent-casting a slurry of the active material, conductive carbon, binder and solvent. The conductive carbon used is Super P (Timcal). PVdF co-polymer (e.g. Kynar Flex 2801 , Elf Atochem Inc.) is used as the binder, and acetone is employed as the solvent. The slurry is then cast onto glass and a free-standing electrode film is formed as the solvent evaporates. The electrode is then dried further at about 80°C. The electrode film contains the following components, expressed in percent by weight: 80% active material, 8% Super P carbon, and 12% Kynar 2801 binder. Optionally, an aluminium current collector may be used to contact the positive electrode.

The negative electrode is prepared by solvent-casting a slurry of the hard carbon active material (Carbotron P/J, supplied by Kureha), conductive carbon, binder and solvent. The conductive carbon used is Super P (Timcal). PVdF co-polymer (e.g. Kynar Flex 2801 , Elf Atochem Inc.) is used as the binder, and acetone is employed as the solvent. The slurry is then cast onto glass and a free-standing electrode film is formed as the solvent evaporates. The electrode is then dried further at about 80°C. The electrode film contains the following components, expressed in percent by weight: 84% active material, 4% Super P carbon, and 12% Kynar 2801 binder. Optionally, a copper current collector may be used to contact the negative electrode.

Generic Procedure to Make a Graphite Li-ion Cell

The positive electrode is prepared by solvent-casting a slurry of the active material, conductive carbon, binder and solvent. The conductive carbon used is Super P (Timcal). PVdF co-polymer (e.g. Kynar Flex 2801 , Elf Atochem Inc.) is used as the binder, and acetone is employed as the solvent. The slurry is then cast onto glass and a free-standing electrode film is formed as the solvent evaporates. The electrode is then dried further at about 80°C. The electrode film contains the following components, expressed in percent by weight: 80% active material, 8% Super P carbon, and 12% Kynar 2801 binder. Optionally, an aluminium current collector may be used to contact the positive electrode.

The negative electrode is prepared by solvent-casting a slurry of the graphite active material (Crystalline Graphite, supplied by Conoco Inc.), conductive carbon, binder and solvent. The conductive carbon used is Super P (Timcal). PVdF co-polymer (e.g. Kynar Flex 2801 , Elf Atochem Inc.) is used as the binder, and acetone is employed as the solvent. The slurry is then cast onto glass and a free-standing electrode film is formed as the solvent evaporates. The electrode is then dried further at about 80°C. The electrode film contains the following components, expressed in percent by weight: 92% active material, 2% Super P carbon, and 6% Kynar 2801 binder. Optionally, a copper current collector may be used to contact the negative electrode.

Cell Testing The cells are tested as follows, using Constant Current Cycling techniques.

The cell is cycled at a given current density between pre-set voltage limits. A commercial battery cycler from Maccor Inc. (Tulsa, OK, USA) is used. On charge, sodium (lithium)-ions are extracted from the cathode active material. During discharge, sodium (lithium)-ions are re-inserted into the cathode active material.

Cell Parameters at Various States of Charge of an Electrode Originally Containing NaNin _? sTin _? sMg _{n ?} sMnn _? sO? (Precursor material xiii)

Although the materials may change structure upon charge, for the purposes of calculating the cell parameters given in Table 2, an R-3m space group was used for the calculation at all states of charge.

Table 2: Calculated cell parameters for NaNio.2 ₅ Tio.2 ₅ Mgo.2 ₅ Mrio.2 ₅ O2 at different states of charge (standard deviation shown in brackets)

As the pristine material is charged to 120 mAh/g, nickel reduces in size as it is oxidised from Ni ²⁺ (IR for CN 6 = 0.69 A) to Ni ⁴⁺ (IR for CN 6 = 0.48 A), and this is reflected in the a- parameter. The c-parameter, however, increases upon charging to 120 mAh/g, and this is due to an increase in repulsion between the electronegative oxygen-ions of adjacent layers, as the sodium-ions are removed.

Upon further charging of this electrode, to 230 mAh/g, there is negligible further oxidation of the nickel, so the nickel oxidation state no longer contributes to the volume change of the unit cell. The Applicant does not wish to be rigidly bound to the following explanation but it is their current belief that the further application of an oxidative potential is capable of oxidising O ^2" in the lattice, thus releasing this from the structure as 0 ₂ , and this is accompanied by a removal of the remaining sodium-ions. The loss of both sodium and oxygen in this way appears to explain the observed reduction in both the a- and c-parameters, and the resulting overall reduction in unit cell volume.

These results are shown in Figures 4(A) and 4(B).