A capacitive energy storage device

Abstract

The invention provides a capacitive energy storage device comprising: a lithium metal anode; a capacitive cathode comprising an ion-adsorption electrode material; an aprotic electrolyte comprising lithium salt; a nitrogen-based component selected from (i) a nitrogen oxide anion, present in the aprotic electrolyte, (ii) a reaction product of a nitrogen oxide anion, present in a solid electrolyte interphase layer between the lithium metal anode and the aprotic electrolyte, and (iii) combinations thereof; and a sulfur-based component selected from (i) a polysulfide anion, present in the aprotic electrolyte, (ii) a polysulfide anion or reaction product thereof, present in a solid electrolyte interphase layer between the lithium metal anode and the aprotic electrolyte, and (iii) combinations thereof.

Description

A capacitive energy storage deviceTechnical Field

[0001] The invention relates to a capacitive energy storage device comprising a lithium metal anode, a capacitive cathode, an aprotic electrolyte comprising lithium salt, a nitrogen-based component selected from a nitrogen oxide anion or reaction product thereof and a sulfur-based component selected from a polysulfide anion or reaction product thereof. The invention further relates to a method of conditioning a capacitive energy storage device, and to a method of storing electrical energy in a capacitive energy storage device.Background of Invention[2] Lithium-ion capacitors (LICs) are hybrid energy storage devices with anodes as used in lithium-ion batteries and cathodes as used in supercapacitors. The cathode typically comprises a high surface area conductive carbon material, such as activated carbon, with excellent charge storage capacity in an electric double layer that develops at the interface between the electrode and the electrolyte. The anode is typically graphite. Battery-type electrolytes compatible with the graphite anode are used, for example carbonate-based electrolytes containing LiPFe as the lithium salt.[3] Since charging and discharging the LIC does not rely on the kinetically limiting faradaic processes of a battery cathode, LICs provide a power density similar to a supercapacitor but with an energy density in an intermediate range between typical batteries and supercapacitors. The improved energy density relative to supercapacitors is achieved due to the extended voltage window (up to about 3.8V vs only 2.7V for supercapacitors) and energy storage in the anode by bulk battery-type mechanisms. The cycle life of LICs is significantly improved compared to Li ion batteries but cannot approach the nearly unlimited cycle life of supercapacitors due to degradation processes at the anode.[4] As such, LICs are viewed as a separate technology from batteries and supercapacitors, with specific applications that require medium energy, high power, and good durability. Some applications include backup power sources in servers andstorage devices for integrated circuits, processors and memory, photovoltaic power generation and energy recovery systems in industrial machinery, electric and hybrid vehicles and short-range transportation systems.[5] A substantial portion of the lithium is typically introduced to a lithium-ion battery via the active cathode material, with the anode substantially free of lithium when the battery is fabricated. However, the active cathode material of an LIC contains no lithium. Therefore, for the device to function, it is essential to add lithium to the cell in a pre-lithiation stage. The technological complications and high cost of this procedure present a significant hurdle to the mass adoption of LICs.[6] Pre-lithiation may be achieved by incorporating sacrificial lithium metal which is partially or completely dissolved upon the first charge of the LIC device. Metallic lithium, electrically connected to the graphite electrode, forms a local battery at the time of immersion into the electrolytic solution, initiating doping of lithium ions into the graphite. The dissolution of the sacrificial lithium metal results in a volume change inside the cell, which needs to be compensated carefully to maintain good contact between the components, adding to manufacturing complexity. Alternatively, pre- lithiation of the anode can be done in a separate cell, adding to the manufacturing complexity. Other reported approaches to pre-lithiation are also unsatisfactory. Relying on excess lithium in the electrolyte depletes the electrolyte of lithium as the anode is lithiated, causing low ionic conductivity and poor cycle life. Introducing a sacrificial lithium source (e.g. lithiated transition metal oxide or an organic lithium salt) in the cathode formulation requires an initial lithium extraction step at an undesirably high voltage and introduces extraneous residual components to the cell.[7] Over long-term cycling, LIC devices also suffer from instability issues related to the anode. The electrolyte solvent can co-insert into the graphitic structure during the intercalation of lithium ions. Gas generated when the solvent is reduced upon subsequent cycling can exfoliate the graphite sheets. In addition, the deintercalation and intercalation of lithium ions result in a volume change of the graphite sheets and mechanical damage to the electrode. As a result of these processes, the solid electrolyte interphase (SEI) is constantly disrupted and regenerated during cycling, leading to electrolyte consumption.[8] For lithium batteries, lithium metal anodes are theoretically favoured over the alternatives due to the high theoretical capacity (3860 mAh g-1) and low electrochemical potential. However, despite considerable research effort, practical applications for lithium metal anodes in batteries remain elusive due to safety concerns and the stability issues encountered over extended cycling.[9] The use of a lithium metal anode, instead of the typical graphite anode, in a LIC-type device (here termed a lithium metal capacitor, LMC), would in principle eliminate the need for a pre-lithiation step, thus simplifying the manufacturability of the device. Moreover, an LMC device should in principle have a higher energy density than a conventional LIC due to the reduction of anode weight. However, capacitive energy storage devices should preferably provide improved cycle life performance relative to batteries, presenting a significant challenge to the development of a suitably stable LMC considering the difficulties evident from battery research. To date, LMCs having satisfactory cycling stability, particularly when cycled at discharge rates representative of practical applications of capacitive energy storage devices, have not been developed.

[0010] There is therefore an ongoing need for capacitive energy storage devices which at least partially address one or more of the above-mentioned short-comings, or provide a useful alternative.

[0011] A reference herein to a patent document or other matter which is given as prior art is not to be taken as an admission that the document or matter was known or that the information it contains was part of the common general knowledge as at the priority date of any of the claims.Summary of Invention

[0012] In accordance with a first aspect the invention provides a capacitive energy storage device comprising: a lithium metal anode; a capacitive cathode comprising an ion-adsorption electrode material; an aprotic electrolyte comprising lithium salt; a nitrogen-based component selected from (i) a nitrogen oxide anion, present in the aprotic electrolyte, (ii) a reaction product of a nitrogen oxide anion, present in a solid electrolyte interphase layer between the lithium metal anode and the aprotic electrolyte, and (iii) combinations thereof; and a sulfur-based component selected from (i) apolysulfide anion, present in the aprotic electrolyte, (ii) a polysulfide anion or reaction product thereof, present in a solid electrolyte interphase layer between the lithium metal anode and the aprotic electrolyte, and (iii) combinations thereof.

[0013] The inventors have surprisingly found that such a capacitive energy storage device, or LMC, is capable of stable, long-term cycling under charge and discharge current densities corresponding to high power density applications of capacitive energy storage devices such as supercapacitors and LIC’s. There is no requirement to engineer the lithium-electrolyte interface during device fabrication. Instead, a nitrogen oxide anion and a polysulfide anion may be added to the initial aprotic electrolyte composition. During the early charge-discharge cycles of the device, which are preferably conducted under well-controlled and mild cycling conditions as a conditioning stage, a complex solid electrolyte interphase (SEI) layer is formed at the interface between electrolyte and anode. This SEI layer, which is believed to comprise reaction products of the polysulfide anion, the nitrogen oxide anion and other electrolyte components, modulates lithium electrodeposition and stripping during charging and discharging, protecting the electrolyte from rapid degradation and prolonging cycle life. Significant enhancements in long-term device performance can thus be obtained, even under highly demanding cycling conditions and despite the initial presence of the redoxactive polysulfide anion in the aprotic electrolyte and the subsequent depletion of that polysulfide anion from the aprotic electrolyte. Surprisingly, excellent results were obtained with only small initial loadings of polysulfide anion, thus avoiding interference with the cathode performance or persistent parasitic redox reactions during extended cycling.

[0014] In some embodiments of the first aspect, the sulfur-based component is present in an amount of less than 1 mg cm-2, or less than 0.5 mg cm-2, or less than 0.4 mg cm2, or less than about 0.3 mg cm2, based on the mass of sulfur in the sulfurbased component relative to the lithium metal anode surface area. In some embodiments, the sulfur-based component is present in an amount of 0.01 to 0.5 mg cm2, such as 0.04 to 0.4 mg cm2, based on the mass of sulfur in the sulfur-based component relative to the lithium metal anode surface area. In some embodiments, the concentration of polysulfide in the aprotic electrolyte, prior to cycling of thecapacitive energy storage device, is less than 0.15 M, or less than 0.1 M, such as less than 0.05 M.

[0015] In some embodiments, the nitrogen oxide anion is nitrate.

[0016] In some embodiments, the ion-adsorption electrode material is a carbonbased electrode material. The carbon-based electrode material may be selected from the group consisting of activated carbon, porous carbon, graphene, reduced graphene oxide, expanded graphite, exfoliated graphite, carbon nanotubes, carbon aerogel and combinations thereof. For example, the carbon-based electrode material may be selected from activated carbon and reduced graphene oxide.

[0017] In some embodiments, the ion-adsorption electrode material has a BET surface area of between about 100 and 3000 m2 / g, such as between about 1000 and 3000 m2 / g.

[0018] In some embodiments, the lithium salt comprises a fluorinated anion. In some embodiments, the lithium salt comprises a fluorinated anion selected from the group consisting of fluorinated alkyl sulfonyl imide, a fluorinated alkyl sulfonyl methide, a fluorinated alkyl sulfonate, a fluorinated alkyl carbonate, a fluorinated aryl borate, a fluorinated alkyl phosphate and combinations thereof. For example, the lithium salt may comprise bis(trifluoromethanesulfonyl)imide.

[0019] In some embodiments, the aprotic electrolyte comprises a molecular liquid carrier comprising one or more ethers. The molecular liquid carrier may comprise at least 70 wt.%, such as at least 90 wt%, of the one or more ethers based on the total weight of the liquid carrier.

[0020] In some embodiments, the molecular liquid carrier comprises 1 ,3-dioxalane (DOL) and / or dimethoxyether (DME). The molecular liquid carrier may comprises DOL and DME in a ratio of 1 :10 to 10:1 (v / v), such as in a ratio of 1 :2 to 2:1 (v / v).

[0021] In some embodiments, the aprotic electrolyte comprises lithium cations in a concentration of 0.1 and 4 M, or between 0.2 and 3 M, such as between 1 and 2 M.

[0022] In some embodiments, the aprotic electrolyte comprises the nitrogen oxide anion. The aprotic electrolyte may comprise the nitrogen oxide anion in a concentration of from 0.05 to 1 .5 M, such as from 0.1 to 1 .0 M, for example from 0.25 to 0.75 M.

[0023] In some embodiments, the capacitive energy storage device comprises a reaction product of a polysulfide anion, present in a solid electrolyte interphase layer between the lithium metal anode and the aprotic electrolyte.

[0024] In some embodiments, the aprotic electrolyte is substantially free of soluble polysulfide anion. In particular, the aprotic electrolyte may be substantially free of soluble polysulfide anion after the capacitive energy storage device has been subjected to one or more charge-discharge cycles, such as a plurality of conditioning chargedischarge cycles.

[0025] In some embodiments, the reaction product is derived from a soluble polysulfide anion initially present in the aprotic electrolyte by charging and / or discharging the capacitive energy storage device.

[0026] In some embodiments, the sulfur-based component comprises U2S2 and / or U2S, present in the solid electrolyte interphase layer.

[0027] In some embodiments, the solid electrolyte interphase layer comprises a reaction product of the nitrogen oxide anion.

[0028] In some embodiments, the capacitive energy storage device retains at least 85%, such as at least 90%, of an initial capacitance when subjected to at least 1000 subsequent cycles at 1 A g1, based on the mass of ion-adsorption electrode material, between a lower cut-off voltage of 2.2V vs Li / Li+and an upper cut-off voltage of 3.8V vs Li / Li+.

[0029] In some embodiments, the capacitive energy storage device has a specific energy of at least 8 Wh.kg1, such as at least 10 Wh.kg1, based on the total mass of the capacitive energy storage device excluding packaging when cycling the capacitive energy storage device at 1 A g1, based on the mass of ion-adsorption electrode material, between a lower cut-off voltage of 2.2V vs Li / Li+and an upper cut-off voltage of 3.8V vs Li / Li+.

[0030] In some embodiments, the capacitive energy storage device comprises: a lithium metal anode; a capacitive cathode comprising an ion-adsorption electrode material; an aprotic electrolyte comprising lithium salt and a nitrogen oxide anion; and a sulfur-based component selected from (i) a polysulfide anion, present in the aprotic electrolyte, (ii) a polysulfide anion or reaction product thereof, present in a solid electrolyte interphase layer between the lithium metal anode and the aprotic electrolyte, and (iii) combinations thereof.

[0031] In accordance with a second aspect the invention provides a method of conditioning a capacitive energy storage device, the method comprising: providing a capacitive energy storage device comprising: a lithium metal anode; a capacitive cathode comprising an ion-adsorption electrode material; and an aprotic electrolyte comprising lithium salt, a nitrogen oxide anion and a polysulfide anion; and cycling the capacitive energy storage device through one or more conditioning charge-discharge cycles to form a solid electrolyte interphase layer between the lithium metal anode and the aprotic electrolyte, wherein the solid electrolyte interphase layer comprises a reaction product of the polysulfide anion.

[0032] In some embodiments of the second aspect, the capacitive energy storage device is cycled across a voltage range having a lower cut-off voltage of at least 2V vs Li / Li+, such as a lower cut-off voltage in the range of between 2.0V and 2.2V vs Li / Li+.

[0033] In some embodiments, the capacitive energy storage device is cycled across a voltage range having an upper cut-off voltage of no more than 4.2V vs Li / Li+, or not more than 4.0V vs Li / Li+, for example an upper cut-off voltage in the range of between 3.8V and 4.0V vs Li / Li+.

[0034] In some embodiments, cycling the capacitive energy storage device through the one or more conditioning charge-discharge cycles comprises charging and discharging the capacitive energy storage device at a maximum rate of less than 2 mA.cnr2, or less than 1 .5 mA.cnr2, such as less than 1 mA.cnr2.

[0035] In some embodiments, the capacitive energy storage device is cycled through at least 20, such as at least 40, conditioning charge-discharge cycles.

[0036] In some embodiments, the capacitive energy storage device is cycled through a plurality of conditioning charge-discharge cycles until the aprotic electrolyte is substantially free of soluble polysulfide anion.

[0037] In some embodiments, the polysulfide anion is present, prior to cycling the capacitive energy storage device, in an amount of less than 1 mg cm-2, or less than 0.5 mg cm-2, or less than 0.4 mg cm-2, or less than 0.3 mg cm-2, based on the mass of sulfur in the polysulfide anion relative to the lithium metal anode surface area. In some embodiments, the polysulfide anion is present in an amount of 0.01 to 0.5 mg cm-2, such as 0.04 to 0.4 mg cm-2, based on the mass of sulfur in the polysulfide anion relative to the lithium metal anode surface area. In some embodiments, the concentration of polysulfide in the aprotic electrolyte, prior to cycling the capacitive energy storage device, is less than 0.15 M, or less than 0.1 M, such as less than 0.05 M.

[0038] In some embodiments, the nitrogen oxide anion is nitrate.

[0039] In some embodiments, the ion-adsorption electrode material is a carbonbased electrode material.

[0040] In some embodiments, the aprotic electrolyte comprises a molecular liquid carrier comprising one or more ethers, preferably wherein the molecular liquid carrier comprises at least 70 wt.%, such as at least 90 wt%, of the one or more ethers based on the total weight of the liquid carrier.

[0041] In some embodiments, the electrolyte comprises the nitrogen oxide anion in a concentration of from 0.05 to 1 .5 M, or from 0.1 to 1 .0 M, such as from 0.25 to 0.75 M.

[0042] In some embodiments, the reaction product of the polysulfide anion comprises U2S2 and / or U2S.

[0043] In some embodiments, the solid electrolyte interphase layer further comprises a reaction product of the nitrogen oxide anion.

[0044] Other embodiments of the second aspect may have features as disclosed herein in the context of the first aspect.

[0045] In accordance with a third aspect the invention provides a method of storing electrical energy, comprising cycling (i) a capacitive energy storage device according to any embodiment of the first aspect or (ii) a capacitive energy storage device conditioned by a method according to any embodiment of the second aspect through a plurality of charge-discharge cycles.

[0046] In some embodiments of the third aspect, the capacitive energy storage device is cycled through at least 1000, such as at least 10000, charge-discharge cycles.

[0047] In some embodiments, the capacitive energy storage device provides a Coulombic efficiency of at least 98%, such as at least 99%, in a charge-discharge cycle following the at least 1000, or at least 10000, charge-discharge cycles.

[0048] In some embodiments, the capacitive energy storage device is cycled between a lower cut-off voltage of between 1 ,5V and 2.5V vs Li+and an upper cut-off voltage of between 3.5V and 4.2V vs Li+in one or more of the charge-discharge cycles. In some embodiments, the capacitive energy storage device is cycled between a lower cut-off voltage of between 2.0V and 2.3 V vs Li / Li+and an upper cut-off voltage of between 3.8V and 4.0 V vs Li / Li+in one or more of the charge-discharge cycles.

[0049] In some embodiments, wherein the capacitive energy storage device is discharged at a rate of at least 5 mA. cm2, preferably at least 7.5 mA. cm2, more preferably at least 10 mA. cm2, in one or more of the charge-discharge cycles.

[0050] In some embodiments, the capacitive energy storage device is discharged at a rate of above 2C, preferably above 5C, more preferably above 10C, most preferably above 20C, in one or more of the charge-discharge cycles.

[0051] In some embodiments, the capacitive energy storage device retains at least 80%, or at least 85%, such as at least 90%, of an initial capacitance corresponding to an initial charge-discharge cycle when subjected to at least 1000, or at least 10,000, subsequent charge-discharge cycles.

[0052] In some embodiments, the capacitive energy storage device discharges at least 8 Wh.kg1, such as at least 10 Wh.kg1, based on the total mass of the capacitive energy storage device excluding packaging in one or more of the charge-discharge cycles.

[0053] Where the terms “comprise”, “comprises” and “comprising” are used in the specification (including the claims) they are to be interpreted as specifying the stated features, integers, steps or components, but not precluding the presence of one or more other features, integers, steps or components, or group thereof.

[0054] Further aspects of the invention appear below in the detailed description of the invention.Brief Description of Drawings

[0055] Embodiments of the invention will herein be illustrated by way of example only with reference to the accompanying drawings in which:

[0056] Figure 1 schematically depicts a capacitive energy storage device according to some embodiments of the invention.

[0057] Figure 2 shows the pore size distribution of activated carbon used in the fabrication of capacitive cathodes in Example 1 .

[0058] Figure 3 is a cyclic voltammogram showing the current response during conditioning charge-discharge cycles (constant scan rate of 2 mVs’1) of a coin cell lithium metal capacitor, as performed in Example 4.

[0059] Figure shows the voltage response during conditioning charge-discharge cycles (constant current rate of 1 mA cm’2) of a coin cell lithium metal capacitor, as performed in Example 4.

[0060] Figure 5 is a graph of specific capacity during extended cycling current at densities between 3 and 12 mA cm’2with coin cell lithium metal capacitors having different electrolytes compositions, as performed in Example 5.

[0061] Figure 6 is a cyclic voltammogram showing the current response, during charge-discharge cycles (constant scan rate of 2 mVs’1) to different upper cut-off voltages, of a coin cell lithium metal capacitor, as performed in Example 6.

[0062] Figure 7 is a graph of specific capacity and Coulombic efficiency during extended cycling at current density of 3 mA cm’2in a voltage range of 2.0 to 4.2 V vs Li / Li+with a coin cell lithium metal capacitor, as performed in Example 7.

[0063] Figure 8 is a graph of specific capacity and Coulombic efficiency during extended cycling at current density of 6 mA cm-2in a voltage range of 2.2 to 3.8 V vs Li / Li+with a coin cell lithium metal capacitor, as performed in Example 7.

[0064] Figure 9 is a graph comparing the specific capacity and Coulombic efficiency during extended cycling at current density of 6 mA cm-2with a coin cell lithium metal capacitor and a symmetrical supercapacitor, as performed in Example 8.

[0065] Figure 10 is a graph of gravimetric capacity and arial capacity during extended cycling at current densities in the range of 0.33 to 1 A g’1with a coin cell lithium metal capacitor having elevated loading of activated carbon on the cathode, as performed in Example 9.

[0066] Figure 11 is a graph comparing the gravimetric capacity and arial capacity (cycled at a current density of 1 A g-1in the range of 2.0 to 4.0 V) with a coin cell lithium metal capacitor having different loadings of activated carbon on the cathode, as performed in Example 9.

[0067] Figure 12 is a graph of specific capacity and Coulombic efficiency during extended cycling (current density of 1 A g-1) with a coin cell lithium metal capacitor with a reduced graphene oxide cathode, as performed in Example 10.

[0068] Figure 13 is a graph of energy density and Coulombic efficiency during extended cycling (current density of 1 A g’1) with a pouch cell lithium metal capacitor, as performed in Example 12.

[0069] Figure 14 is a graph of energy density during extended cycling (current density of between 0.1 and 1 A g’1) with a pouch cell lithium metal capacitor having reduced electrolyte volume, as performed in Example 12.

[0070] Figure 15 is a graph comparing the properties of a pouch cell lithium metal capacitor against a commercial lithium-ion capacitor, as performed in Example 13.

[0071] Figure 16 shows scanning electron microscopy images of lithium metal anodes after cycling in different electrolytes, as performed in Example 14.

[0072] Figure 17 shows X-ray photoelectron spectroscopy S2p spectra for lithium metal electrodes after cycling in different electrolytes, as performed in Example 14.

[0073] Figure 18 is a graph of specific capacity during extended cycling (current density of 0.5 A g’1) of pouch cell lithium metal capacitors with different concentrations of LiNOa in the aprotic electrolyte, as performed in Example 16.

[0074] Figure 19 is a graph of specific capacity during extended cycling (current density of 0.5 A g’1) of pouch cell lithium metal capacitors with different concentrations of lithium polysulfide loaded in the cell, as performed in Example 17.

[0075] Figure 20 is a graph of specific capacity during extended cycling (current density of 0.5 A g-1) of pouch cell lithium metal capacitors after preconditioning cycles at different charge-discharge rates, as performed in Example 18.

[0076] Figure 21 shows the impedance resistance of pouch cell lithium metal capacitors with applied stack pressure, and no applied stack pressure, as performed in Example 19.Detailed DescriptionCapacitive energy storage device

[0077] The present invention relates to a capacitive energy storage device. The capacitive energy storage device comprises a lithium metal anode, a capacitive cathode comprising an ion-adsorption electrode material, an aprotic electrolyte comprising lithium salt; a nitrogen-based component, and a sulfur-based component. The nitrogen-based component is selected from (i) a nitrogen oxide anion, present in the aprotic electrolyte, (ii) a reaction product of a nitrogen oxide anion, present in a solid electrolyte interphase layer between the lithium metal anode and the aprotic electrolyte, and (iii) combinations thereof. The sulfur-based component is selected from (i) a polysulfide anion which is present in the aprotic electrolyte, (ii) a polysulfide anion or reaction product thereof which is present in a solid electrolyte interphase layer between the lithium metal anode and the aprotic electrolyte, and (iii) combinations thereof.

[0078] As used herein, a capacitive energy storage device is an electrochemical cell which stores electrical energy by capacitive charge storage mechanisms, includingelectrostatic double-layer capacitance and electrochemical pseudocapacitance. The capacitive energy storage devices of the present disclosure preferably store charge primarily by electrostatic double-layer capacitance. A capacitive energy storage device is thus distinguished from a rechargeable battery where charge is stored in the electrode materials of both electrodes by bulk electrochemical (faradaic) processes, such as lithium intercalation in the case of lithium battery cathodes.Lithium metal anode

[0079] The capacitive energy storage device disclosed herein includes a lithium metal anode (also known as a negative electrode) which comprises, and may consist of, lithium metal. Metallic lithium is theoretically an ideal anode material for high energy storage devices due to its high specific capacity (3,860 mAh.g-1) and very low electrochemical potential. However, lithium metal anodes are susceptible to dendritic lithium growth over repeated charge-discharge cycles, repeatedly disrupting the SEI and causing poor Coulombic efficiencies and capacity fading due to reactions of the high surface area lithium metal with the electrolyte, or even complete device failure due to short circuiting. These issues are exacerbated when cycling the device at high current densities, and it is therefore surprising that a cell configured for and used in high power density application, such as a LMC, can provide acceptable long-term term cycling stability despite the use of a lithium metal anode.

[0080] The anode may comprise a layer of metallic lithium having at least a thickness sufficient for the required charge capacity of the cell. For example, the anode may comprise a layer of lithium metal having a thickness of between 5 pm and 500 pm, for example between 50 pm and 200 pm. It will be appreciated that the lithium metal anode in any particular implementation should preferably contain sufficient lithium metal to exceed the capacity of the cathode, so that is not fully depleted on complete discharge of the device. The metallic lithium anode may be present on a current collector, for example a metallic current collector such as copper foil. Alternatively, the lithium metal may act as its own current collector. This may advantageously reduce the weight of the device and thus increase the energy and power density.Capacitive cathode

[0081] The capacitive energy storage device disclosed herein includes a capacitive cathode (positive electrode) which comprises an ion-adsorption electrode material. A capacitive cathode stores electrical charge primarily at the interface between the electrolyte and the electrode material, in contrast to battery cathodes which store charge primarily in the bulk of the electrode material by electrochemical (faradaic) processes, such as lithium intercalation in the case of lithium batteries. As such, capacitive cathodes are capable of rapid charge and discharge in comparison to batteries where the rates are subjected to kinetic limitations associated with the bulk faradaic processes. Capacitive cathodes are used in supercapacitors, and it is expected that capacitive cathodes developed for supercapacitor applications, and particularly for EDLC supercapacitors, will generally be suitable for use in the presently disclosed device.

[0082] Preferably, the capacitive cathode stores charge primarily by electrostatic capacitance. In this mode, charge is believed to be stored by charge separation of electrolyte ions in a Helmholtz double layer at the interface between the electrolyte and the surface of the electrode material. Such electrodes are used in electric double-layer capacitors (EDLC supercapacitors). However, it is not excluded that the capacitive cathode may also store a portion of the charge by pseudocapacitive mechanisms, i.e. by rapidly reversible redox or intercalation processes involving electrolyte ions that occur at the surface of a pseudocapacitive electrode material. In some embodiments, at least 70%, or at least 80%, or at least 90% of the charge stored by the cathode in use is stored as electrostatic double layer capacitance.

[0083] Consistent with the type of charge storage required at the cathode, the cathode comprises an ion-adsorption electrode material. Suitable ion-absorption materials are conductive, high surface area materials which facilitate charge storage by adsorption of electrolyte ions at the surface, typically in an electric double layer. In some embodiments, the ion-adsorption electrode material has a surface area of between about 100 and 3000 m2 / g, such as between about 1000 and 3000 m2 / g, or between 2000 and 3000 m2 / g. The surface area may be measured by Brunauer- Emmett-Teller (BET) surface area analysis methods known to those of skill in the art.

[0084] In some embodiments, the ion-adsorption electrode material comprises, or consists of, a carbon-based electrode material. Carbon-based electrode material, suchas activated carbon, are generally porous materials, thus providing a high surface area for charge storage. Suitable carbon-based electrode material may include activated carbon, porous carbon, graphene, reduced graphene oxide, expanded or exfoliated graphite, carbon nanotubes, and carbon aerogels. The inventors have demonstrated good performance with activated carbon and reduced graphene oxide. Activated carbon is favoured in many commercial supercapacitors, while reduced graphene oxide is also of particular interest in such applications due to the possibility for functionalisation and tailoring of the pore size to specific electrolytes, and the high volumetric energy density.

[0085] The cathode may comprise a porous cathode formulation comprising the ion-absorption electrode material and other particulate additives. In some embodiments, the porous cathode formulation comprises a pseudocapacitive electrode material, such as a conductive polymer, transition metal oxide, or metal nanoparticles. Typically, any pseudocapacitive electrode material would be in addition to an electrostatic ion-absorption electrode material such as a carbon-based electrode material. In some embodiments, the porous cathode formulation comprises a conductive additive, such as carbon black. In some embodiments, the porous cathode formulation comprises a polymeric binder. Suitable polymeric binders for electrode formulations are known to those of skill in the art. One suitable example is carboxymethylcellulose (CMC).

[0086] Consistent with the nature of a capacitive cathode, the cathode typically does not include a substantial component of bulk battery-type cathode material to store charge by bulk faradaic mechanisms. In some embodiments, the cathode therefore does not include, and is thus substantially free, of lithium metal oxides and lithium metal phosphates. In some embodiments, the cathode is substantially free of non-carbon lithium-intercalation materials. In some embodiments, where the cathode is configured to store charge primarily by electrostatic capacitance, the cathode is substantially free of lithium-intercalation materials and pseudocapacitive materials.

[0087] The cathode may comprise a current collector, for example a metallic foil such as an aluminium foil. The ion-absorption electrode material may be coated onto the current collector together with any other particulate additives, typically in a solvent-borne electrode slurry which is then dried to produce the porous electrode formulation on the current collector.

[0088] The cathode includes a sufficient loading of the ion-absorption electrode material, and any pseudocapacitive electrode materials present, to provide the required capacitance of the capacitive energy storage device. In some embodiments, the capacitive cathode comprises the ion-absorption electrode material (such as a carbonbased electrode material) in an amount of between 0.5 and 15 mg / cnr2, or between 1 and 12 mg / cnr2, such as between 2 and 10 mg / cnr2.Aprotic electrolyte

[0089] The capacitive energy storage device disclosed herein includes an aprotic electrolyte which comprises lithium salt. The electrolyte is in contact with both the cathode and anode in the device, and thus facilitates lithium ion transport between anode and cathode during discharge and charging of the device. Typically, the anode and cathode are separated by a porous separator, and at least a portion of the electrolyte is infiltrated into the separator.

[0090] The aprotic electrolyte comprises sufficient lithium salt to provide the required lithium ion conductivity during cycling of the device. In some embodiments, the aprotic electrolyte comprises lithium cations in an amount of between 0.1 and 4 M, such as between 0.2 and 3 M, or between 1 and 2 M.

[0091] The lithium salt comprises a suitable counterion (anion). The anion should be sufficiently electrochemically stable to allow prolonged cycling of the device and should facilitate solubility of the lithium salt in the electrolyte. Without wishing to be limited by any theory, the choice of anion is also believed to affect one, or both of, the capacitance and long-term cycling stability of the device. The inventors observed that a higher initial capacitance was obtained, in an otherwise identical lithium metal capacitor, when using an ether-based electrolyte containing lithium bis(trifluoromethanesulfonyl)imide (LiTFSI) than when using a carbonate-based electrolyte containing lithium hexafluorophosphate (LIPFe) as typically used in a lithium ion capacitor. It is proposed that the larger solvated radius of TFSI anions, compared to PF6anions, facilitates improved charge storage in the electrical double layer on carbon-based electrode materials such as activated carbon. Furthermore, it isproposed that degradation of a small fraction of the anion during conditioning of the device may play a role in establishing a favourable SEI.

[0092] In some embodiments, the lithium salt comprises a fluorinated anion, for example a perfluorinated anion. Such anions are commonly sufficiently electrochemically stable for energy storage applications. Moreover, decomposition products of the fluorinated anion, such as lithium fluoride, may advantageously be incorporated into the SEI layer on the lithium metal anode.

[0093] In some embodiments, the lithium salt comprises an anion selected from the group consisting of a fluorinated alkyl sulfonyl imide (i.e. [RF-SO2N-R’]_where RFis a fluoroalkyl or perfluoroalkyl, and R’ is an organyl group, preferably an electron withdrawing organyl group, and more preferably -SO2RF), a fluorinated alkyl sulfonyl methide (i.e. [RF-SO2C-R’2] where RFis a fluoroalkyl, optionally perfluoroalkyl, and each R’ is an organyl group, preferably an electron withdrawing organyl group, and more preferably -SO2RF), a fluorinated alkyl sulfonate (i.e. [RFSO3]’ where RFis a fluoroalkyl, optionally perfluoroalkyl), a fluorinated alkyl carbonate (i.e. [RFCO2]’ where RFis a fluoroalkyl, optionally perfluoroalkyl), a fluorinated aryl borate (i.e. [BArF4]’ where ArFis a fluorinated, optionally perfluorinated aryl group) and a fluorinated alkyl phosphate (i.e. [P(RF)nF6-n]’ where RFis a fluoroalkyl or perfluoroalkyl and n is from 1 -6) and combinations thereof. In other embodiments, the lithium salt comprises an anion selected from the group consisting of PFe’, BF , bis(oxalato)borate (BOB, B[C2O4]2’), and combinations thereof.

[0094] In some embodiments, the lithium salt comprises an anion selected from the group consisting of bis(trifluoromethanesulfonyl)imide (TFSI), bis(fluorosulfonyl)imide (FSI), (fluorosulfonyl)(trifluoromethanesulfonyl)imide (FTFSI), tris(trifluoromethanesulfonyl)methide, tetrakis(3,5-bis(trifluoromethyl)-2,4,6-trifluoro- phenyl)borate, tetrakis(3,5-bis(trifluoromethyl)phenyl)borate, tetrakis(pentafluorophenyl)borate, C1-10 perfluoroalkylsulfonate, C1-10 perfluoroalkylcarbonate. In some preferred embodiments, the anion of the lithium salt comprises, or consists of TFSI.

[0095] The aprotic electrolyte may be a liquid. The aprotic electrolyte will thus comprise a liquid carrier in which the functional ionic components (lithium salt, nitrogenoxide anion, soluble polysulfide, etc) are dissolved. In some embodiments, the aprotic electrolyte comprises a molecular liquid carrier. The molecular liquid carrier may comprise one or more ether molecular solvents. Suitably, the molecular liquid carrier may comprise one or more ethers as the primary molecular liquid component, for example at least 70 wt.%, or at least 90 wt%, or 100 wt.%, based on the total weight of the liquid carrier. Ether-based molecular liquid carriers are considered more compatible with lithium metal anodes than some electrolytes commonly used together with graphite anodes, such as carbonate-based electrodes. Ethers are considered less susceptible to reaction with lithium metal, while the more limited degradation that does occur may advantageously improve the SEI due to reduction and polymerisation of the ether molecules which imparts elasticity to the SEI layer.

[0096] Non-limiting examples of suitable ethers include 1 ,3-dioxalane (DOL) and / or dimethoxyether (DME). DOL in particular may polymerise to form an elastic barrier component of the SEI, thus accommodating the volume increase and decrease of the lithium metal anode during electroplating and stripping of the lithium metal when cycling the device. In some embodiments, the molecular liquid carrier comprises DOL and DME in a ratio of 1 :10 to 10:1 (v / v), such a 1 :2 to 2:1 (v / v), or about 1 :1 (v / v).

[0097] It is also envisaged that the aprotic electrolyte may be a gel or polymeric electrolyte. For example, a gelled or polymeric aprotic electrolyte may comprise poly(vinylidene difluoride) (PVDF), optionally cross-linked for example with hexafluoropropylene. A PVDF-based electrolyte membrane may comprise molecular liquid components, such as ethers, together with the required ionic components.

[0098] As used herein, an aprotic electrolyte refers to an electrolyte which lacks protic solvent components. The aprotic electrolyte therefore does not include, and is generally substantially free of, protic solvents such as water, alcohols, acids and the like.

[0099] The capacitive energy storage device contains sufficient aprotic electrolyte to facilitate stable cycling of the device. A sufficient volume should be included to wet and penetrate the electrodes and infiltrate the separator, thus providing good ionic conductivity. However, the aprotic electrolyte contributes a substantial portion of the weight of the device, and it is desirable to minimise the amount of electrolyte whilemaintaining acceptable functionality, so as to increase the energy and power density of the device. One approach to doing so is to increase the concentration of lithium cations in the electrolyte.Nitrogen-based component

[0100] The capacitive energy storage device disclosed herein includes a nitrogenbased component selected from (i) a nitrogen oxide anion, present in the aprotic electrolyte, (ii) a reaction product of a nitrogen oxide anion, present in a solid electrolyte interphase layer between the lithium metal anode and the aprotic electrolyte, and (iii) combinations thereof.

[0101] In some embodiments, the aprotic electrolyte comprises the nitrogen oxide anion, as a solubilised species. Preferably, the nitrogen oxide anion is added to the electrolyte as a lithium salt, thus adding to the lithium concentration provided by the primary lithium salt. However, it is not excluded that a different salt of the nitrogen oxide anion is used provided that the cation is suitably compatible with the cell operation.

[0102] In some embodiments, the aprotic electrolyte comprises the nitrogen oxide anion in a concentration of from 0.05 to 1 .5 M (or to the solubility limit in the electrolyte at room temperature, if lower), such as from 0.1 to 1.0 M, or from 0.175 to 0.75 M, for example from 0.25 to 0.75 M. In some embodiments, the aprotic electrolyte comprises the nitrogen oxide anion in a concentration of from 0.175 M to the solubility limit of the nitrogen oxide anion in the electrolyte at room temperature. The nitrogen oxide anion is typically present in the aprotic electrolyte, for example in the above-mentioned concentrations, at least initially (i.e. prior to electrochemical cycling of the device), with a portion of the nitrogen oxide anion then being consumed during cycling by incorporation into the SEI. In some embodiments, at least a portion of the nitrogen oxide anion remains present in the aprotic electrolyte throughout the cycle lifetime of the device. Without wishing to be limited by any theory, nitrogen oxide anion solubilised in the aprotic electrolyte may play a helpful ongoing role in modulating lithium electrodeposited morphology on the anode during extended cycling of the device.

[0103] However, it is also envisaged that the nitrogen oxide anion initially present may be substantially depleted from the aprotic electrolyte by incorporation into the SEI, with the resultant SEI modulating subsequent cycling of the device despite thefunctional absence of soluble nitrogen oxide anion in the electrolyte. Accordingly, the nitrogen-based component may consist of a reaction product of a nitrogen oxide anion, present in a solid electrolyte interphase layer between the lithium metal anode and the aprotic electrolyte. The reaction product may be a reaction product of the nitrogen oxygen anion and lithium metal. The reaction products of the nitrogen oxide anion may include insoluble species such as LiaN and / or LiNxOy (where x and y represent relative stoichiometries of nitrogen and oxygen in the partially reduced species).

[0104] In some embodiments, the nitrogen oxide anion is nitrate (NOa'). Lithium nitrate has been successfully used in lithium sulfur batteries to improve cycling stability, and it is understood in that context that a portion of the nitrate anion is incorporated into the SEI, together with sulfur-based species derived from soluble polysulfides, by reaction at the surface of the lithium metal anode to form insoluble species such as Li Na and LiNxOy. Nitrite (NOa-), another nitrogen oxide anion, is reported to be an intermediate species in this process. Without limitation by theory, it is proposed that the nitrogen oxide anion plays a similar role in the present system, albeit in a synergistic effect together with a polysulfide anion additive.Sulfur-based component

[0105] The capacitive energy storage device disclosed herein includes a sulfurbased component selected from (i) a polysulfide anion which is present in the aprotic electrolyte, (ii) a polysulfide anion or reaction product thereof, present in a solid electrolyte interphase layer between the lithium metal anode and the aprotic electrolyte, and (iii) combinations thereof.

[0106] As used herein, a polysulfide anion has the formula Sn2’ where n is an integer from 1 to 8. A range of lithium polysulfide species, having the formula Li2Sn where n is from 2 to 8, are reported intermediates in the cathodic half-reaction of lithium sulfur batteries (S + 2Li++ 2 e- - U2S). The less-reduced intermediates, Li2Sn where n is from 4-8, are highly soluble in the battery electrolyte, allowing them to migrate to the lithium metal anode where they undergo parasitic reactions leading to capacity fading and poor Coulombic efficiencies. The elimination or mitigation of this polysulfide shuttling is a key goal in developing commercially viable lithium-sulfur batteries. One approach previously employed to this end is the use of lithium nitrate additive in theelectrolyte. It is presently believed that the presence of both nitrate and soluble polysulfide in the battery electrolyte leads to the formation of a stabilised SEI, incorporating derivatives of both species, which prolongs to some extent the cycle lifetime of the cell. Unfortunately, the SEI in lithium sulfur batteries remains susceptible to cracking due to constant attack of the highly corrosive polysulfide, ion fluctuation, and volume change-induced stress evolution in the cell, leaving freshly formed lithium surfaces in dynamic exchange with the electrolyte. The continuous reformation of the SEI is accompanied by the continuous consumption of the electrolyte and LiNOa, causing the cell to fail.

[0107] Despite the cycle stability problems caused by soluble polysulfides in lithium battery chemistry, and the limited mitigation provided by a nitrate additive in that context (and uncertainties as to the compatibility of redox-active polysulfides on the performance of a high surface area capacitive cathode), the inventors recognised that an analogous approach to SEI formation may be more effective in a lithium metal capacitor since the polysulfide could be introduced, in small and controllable amounts, as an additive.

[0108] The capacitive energy storage device may thus include a polysulfide anion, such as Li2Sn where n is from 4-8, which is present in the aprotic electrolyte as an additive. Indeed, the aprotic electrolyte typically does contain a soluble polysulfide anion in the device as fabricated and during at least the initial charge-discharge cycles of the device. For example, the aprotic electrolyte before commencement of cycling may include a soluble polysulfide anion in an amount of between 0.001 M and 0.1 M, such as between 0.005 M and 0.05 M. However, the polysulfide anion initially present will deplete and may in some embodiments be fully consumed by incorporation into the SEI layer, so that the aprotic electrolyte is ultimately substantially free of soluble polysulfide anion during extended cycling of the device. At that stage, the sulfur-based component of the capacitive energy storage device includes only a polysulfide anion or reaction product thereof which is present, as a solid, in a solid electrolyte interphase layer between the lithium metal anode and the aprotic electrolyte. The reaction product may be a reaction product of the polysulfide anion and lithium metal.

[0109] Since the incorporation of the polysulfide anion into the SEI may be critical to the performance of the cell, the areal loading of polysulfide relative to the lithiummetal anode working area, is proposed as a relevant parameter. In some embodiments, the sulfur-based component is present in an amount of less than 1 mg cm-2, or less than 0.75 mg cm-2, or less than 0.5 mg cm-2, or less than 0.4 mg cm-2, or less than about 0.3 mg cm-2. Excessively higher concentrations of polysulfide anion in the electrolyte may lead to parasitic reactions during cycling (of the type observed in lithium sulfur batteries), capacity loss of the cathode and impractical conditioning requirements, e.g. a high number of cycles to deplete polysulfides from the aprotic electrolyte and thus obtain rectangular capacitive behaviour. In some embodiments, the sulfur-based component is present in an amount of 0.01 to 0.5 mg cm-2, such as 0.04 to 0.4 mg cm-2, based on the mass of sulfur in the sulfur-based component relative to the lithium metal anode surface area. This can be achieved by loading a corresponding amount of polysulfide anion into the cell, based on the volume of electrolyte and its polysulfide anion concentration. In some embodiments, the concentration of polysulfide in the aprotic electrolyte prior to cycling (i.e. prior to electrochemical deposition of the polysulfide in the SEI), is less than 0.15 M, or less than 0.1 M, such as less than 0.05 M.

[0110] The SEI may comprise U2S2 and / or U2S. The inventors have identified these components in the SEI layer by X-ray photoelectron spectroscopy. It is proposed that these products are formed by reduction of soluble polysulfide, i.e. Li2Sn where n is from 4-8, with electroplated lithium metal formed on the lithium metal anode during cycling of the device.

[0111] The SEI produced upon cycling the capacitive energy device has a complex composition comprising, in addition to the polysulfide anion or reaction product thereof, a variety of other electrolyte degradation products. These are believed to include reaction products of the nitrogen oxide anion, such as LiaN and / or LiNxOy, reaction products of the lithium salt anion, such as Li F, and reaction products of the liquid carrier, such as reduced ether derivatives. As will be discussed hereafter, the SEI is preferably formed in an initial conditioning phase where the electrochemical energy storage device is cycled under mild conditions, such as at low current density.

[0112] The inventors have demonstrated that a capacitive energy storage device comprising both a nitrogen oxide anion in the aprotic electrolyte and a sulfur-based component as disclosed herein (initially present in the electrolyte, and then at leastpartially incorporated into the SEI) provides a substantial improvement in extended cycling performance. This is attributed to the improved, low surface area morphology of lithium electrodeposited on the lithium metal anode surface, as observed by scanning electron microscopy. In the absence of polysulfide, the lithium nitrate additive provides limited control over the lithium electrodeposition, leading to dendritic lithium formation and a high rate of electrolyte consumption. Without wishing to be limited by any theory, it is proposed that the SEI incorporating derivatives of both the nitrogen oxide anion and a polysulfide anion has synergistically enhanced properties which modulate lithium electrodeposition and stripping during cycling of the device. Surprisingly, the improvement is sufficient to allow extended cycling through thousands of charge discharge cycles of the capacitive energy storage device, despite the high current densities associated with the use of such devices.Device configuration

[0113] The cathode and anode of the capacitive energy storage device are electrically isolated from each other. The capacitive energy storage device may thus include a separator interposed between the capacitive cathode and the lithium metal anode. The separator may be a porous film, such as a microporous polymeric film. For example, the porous separator may comprise polyolefin (polyethylene, polypropylene and the like), fluorine resin (polytetrafluoroethylene and the like), polyaramid or polyimide. Alternatively, the porous separator may comprise paper or non-woven fabrics comprising resin fibre or glass fibre.

[0114] In some embodiments, the capacitive energy storage device may be pressurized, meaning that it is subjected to an externally applied pressure. Such pressurization may advantageously reduce the equivalent series resistance (ESR) of the device, thus enhancing cycle life.

[0115] Depicted in Figure 1 is a capacitive energy storage device 100 according to some embodiments of the invention. Device 100 comprises lithium metal anode 102 and capacitive cathode 104, separated by porous separator 106. Lithium metal anode 102 comprises lithium metal layer 108 and optional copper foil current collector 1 10. Capacitive cathode 104 comprises porous cathode formulation 112 including an ionadsorption electrode material, such as a carbon-based electrode material, together withoptional additives such as a conductive additive and a polymeric binder. Device 100 further comprises aprotic electrolyte 1 16, which is infiltrated through the porosity of separator 106 and which wets the surfaces of lithium metal anode 102 and capacitive cathode 103, infiltrating the porosity of porous cathode formulation 1 12 and the ionadsorption electrode material therein. Aprotic electrolyte 1 16 thus provides ionic conductivity between the cathode and anode. Aprotic electrolyte 1 16 comprises lithium salt, such as LiTFSI, a nitrogen oxide anion, preferably as LiNOa, and a soluble lithium polysulfide, such as one or more of LiaSn where n is from 4 to 8. These components are carried in a molecular liquid carrier, such as an ether or mixture of ethers.

[0116] In use, device 100 is subjected to repeated charge-discharge cycles via an external circuit (not shown), as represented by arrow A. In each discharge cycle, lithium metal is oxidised and stripped from the surface of lithium metal layer 108. The resultant lithium cations migrate through the aprotic electrolyte to capacitive cathode 104, while electrons pass through the external circuit to capacitive cathode 104. The charges are stored in an electric double layer which forms at the interface between the aprotic electrolyte 1 16 and the surface of the high surface-area ion-adsorption electrode material. The electric current, corresponding to the release of electrical energy from the device, may be used to do work in applications requiring high power density. In each charge cycle, corresponding to storage of electrical energy in the device, lithium cations migrate back through the aprotic electrolyte to lithium metal anode 102 when the device is polarised, where they are reduced and electrodeposited as lithium metal on the surface of lithium metal layer 108.

[0117] During the initial cycles of device 100, solid electrolyte interphase (SEI) layer 1 18 forms at the interface between the aprotic electrolyte 1 16 and lithium metal layer 108. SEI layer 1 18 has a complex composition, comprising one or more reaction products of the soluble lithium polysulfide with the electrodeposited lithium metal (such as I 2S2 and U2S), together with other SEI components derived from electrolyte components such as the nitrogen oxide anion, the anion of the lithium salt, and the molecular liquid carrier. Over multiple cycles, the soluble lithium polysulfide may be depleted from the aprotic electrolyte by incorporation into SEI layer 1 18, so that the aprotic electrolyte 1 16a is substantially free of soluble polysulfide anion.

[0118] The resultant SEI layer 118 has been found to modulate the electroplating and stripping of lithium metal on lithium metal anode 102, causing the formation of low surface area plate-like deposits of electrodeposited lithium metal in contrast to a mossy I dendritic high surface morphology formed in the absence of the polysulfide additive. The robust SEI layer, and the favourable lithium metal morphology, limits the ongoing degradation of electrolyte components by reaction with freshly deposited lithium metal, thus extending the cycle life of device 100. This effect has been found to persist even after the soluble lithium polysulfide has been fully depleted from aprotic electrolyte 1 16.Device performance characteristics

[0119] As a result of the benefits disclosed herein, embodiments of the capacitive energy storage device may have one or more of the following properties:• retention of at least 85%, or at least 90%, of an initial capacitance when subjected to at least 1000, or at least 5000, or at least 10,000 subsequent cycles at 1 A g-1(based on the mass of ion-adsorption electrode material) when cycling the device between a lower cut-off voltage of 2.2V vs Li / Li+and an upper cut-off voltage of 3.8V vs Li / Li+. The initial capacitance may be measured in any nominated charge-discharge cycle, for example after one or more conditioning cycles.• a specific energy of at least 8 Wh.kg1, preferably at least 10 Wh.kg1, based on the total mass of the capacitive energy storage device (excluding packaging) when cycling the capacitive energy storage device at 1 A g-1(based on the mass of ion-adsorption electrode material) between a lower cut-off voltage of 2.2V vs Li / Li+and an upper cut-off voltage of 3.8V vs Li / Li+.

[0120] The characterisation of a given capacitive energy storage device according to these parameters may be by measured by methods known to those of skill in the art, for example by using a potentiostat, such as a Biologic potentiostat.Method of conditioning a capacitive energy storage device

[0121] The present invention also relates to a method of conditioning a capacitive energy storage device. The method comprises providing a capacitive energy storagedevice comprising: a lithium metal anode, a capacitive cathode comprising an ionadsorption electrode material, and an aprotic electrolyte comprising lithium salt, a nitrogen oxide anion and a polysulfide anion. This capacitive energy storage device is then cycled through one or more conditioning charge-discharge cycles so as to form a solid electrolyte interphase layer between the lithium metal anode and the aprotic electrolyte. The solid electrolyte interphase layer comprises a reaction product of the polysulfide anion.

[0122] The capacitive energy storage device, as provided, is generally as previously disclosed herein except for the requirement that the electrolyte initially comprises a nitrogen oxide anion and a polysulfide anion.

[0123] The conditioning cycles may provide the conditioned capacitive energy storage device with improved properties in comparison to a corresponding unconditioned device which is subjected immediately to cycling in capacitive energy storage applications. The improved properties may include one or more of: improved long-term cycling stability, improved capacitance, and improved Coulombic efficiency. As previously disclosed herein, such benefits are ascribed to the formation of a durable SEI capable of modulating the morphology of lithium metal electrodeposition on the anode. Therefore, the conditioning cycles may be conducted under cycling conditions designed to produce an effective SEI layer. In particular, the conditioning cycles may be conducted under mild and well-controlled cycling conditions, for example at lower current density and / or across a narrower voltage window than the device may ultimately be subjected to in use. Optionally, the conditioning cycles may be applied as part of the device manufacture.

[0124] In some embodiments, the capacitive energy storage device is cycled across a voltage range having a lower cut-off voltage of at least 2V vs Li / Li+, such as in the range of between 2.0V and 2.2V vs Li / Li+, during the conditioning cycles. It is considered important to remain above 2.0V while soluble polysulfide anion remains dissolved in the electrolyte to avoid over-reduction of the polysulfide on the cathode. The deposition of insoluble species such U2S2 and U2S on the cathode may block the porosity of the cathode leading to capacity loss.

[0125] In some embodiments, the capacitive energy storage device is cycled across a voltage range having an upper cut-off voltage of no more than 4.2V vs Li / Li+, such as not more than 4.0V vs Li / Li+, for example in the range of between 3.8V and 4.0V vs Li / Li+, during the conditioning cycles. Remaining below 4.2V, and preferably below 4.0V, may avoid or acceptably limit oxidative degradation of the electrolyte during the conditioning.

[0126] In some embodiments, the capacitive energy storage device is charged and discharged at a maximum rate of less than 2 mA.cnr2, such as less than 1 .5 mA.cnr2, for example less than 1 mA.cnr2, based on the working area of the lithium metal anode. In some embodiments, the capacitive energy storage device is charged and discharged at a maximum rate of less than 1 A.g1, such as less than 0.5 A.g1, for example less than 0.33 A.g1, based on the mass of the ion-adsorption electrode material. Maintaining low current densities in these ranges during the conditioning of the device is considered helpful to facilitate the formation of a favourable SEI composition and morphology.

[0127] The number of conditioning cycles may be selected to provide desirable device properties in subsequent use. In some embodiments, the capacitive energy storage device is cycled through at least 20 conditioning charge-discharge cycles, such as between 20 and 100 cycles. Optionally, the conditioning cycles are continued until most or all of the polysulfide anion is depleted from the aprotic electrolyte. In some embodiments, the capacitive energy storage device is cycled through a plurality of charge-discharge cycles, preferably each at low current densities as disclosed herein, until the aprotic electrolyte is substantially free of soluble polysulfide anion. This may be determined by monitoring the current-potential response, as soluble polysulfides will be oxidised and reduced during the conditioning phase until fully incorporated into an SEI layer. Thus, an aprotic electrolyte is substantially free of soluble polysulfide anion when peaks corresponding to oxidation or reduction of soluble polysulfides are no longer evident when cycling the device.

[0128] Referring again to Figure 1 , a method according to some embodiments of the invention comprises providing a capacitive energy storage device 100 as described herein, including aprotic electrolyte 1 16 which comprises lithium salt, such as LiTFSI, a nitrogen oxide anion, preferably as LiNOa, and a soluble lithium polysulfide, such asone or more of Li2Sn where n is from 4 to 8. These components are carried in a molecular liquid carrier, such as an ether or mixture of ethers. Device 100 is cycled through a series of conditioning charge-discharge cycles via an external circuit (not shown), as represented by arrow A. Preferably, the conditioning cycles are conducted across a voltage range having a lower cut-off voltage of at least 2V vs Li / Li+, and at a maximum charge and discharge rate of less than 1.5 mA. cm-2, preferably less than 1 mA.cnr2, based on the working area of the lithium metal anode.

[0129] Solid electrolyte interphase (SEI) layer 1 18 thus forms at the interface between the aprotic electrolyte 1 16 and lithium metal layer 108. SEI layer 1 18 has a complex composition, comprising one or more reaction products of the soluble lithium polysulfide, such as U2S2 and l_i2S, together with other SEI components derived from electrolyte components such as the nitrogen oxide anion, the anion of the lithium salt, and the molecular liquid carrier. Over multiple conditioning cycles, the soluble lithium polysulfide may be depleted from the aprotic electrolyte by incorporation into SEI layer 1 18, so that the aprotic electrolyte 1 16a is substantially free of soluble polysulfide anion. The resultant SEI layer 1 18 forming during the conditioning cycles has been found to facilitate the long-term cycling of the device, as disclosed herein.Method of storing electrical energy

[0130] The present invention also relates to a method of storing electrical energy. The method comprises cycling (i) a capacitive energy storage device as disclosed herein, or (ii) a capacitive energy storage device conditioned by a method as disclosed herein, through a plurality of charge-discharge cycles to successively store and release electrical energy.

[0131] The capacitive energy storage devices according to embodiments of the invention may be acceptably stable over prolonged cycling under realistic use conditions. In some embodiments, the capacitive energy storage device is cycled through at least 1000, or at least 10,000, such as at least 50,000, or at least 100,000 charge-discharge cycles. Despite such prolonged cycling, the capacitive energy storage device may provide a Coulombic efficiency of at least 98%, or at least 99%, in the final charge-discharge cycles.

[0132] In some embodiments, the capacitive energy storage device is cycled through a plurality of charge-discharge cycles between a lower cut-off voltage of between 1 ,5V and 2.5V vs Li+and an upper cut-off voltage of between 3.5V and 4.2V vs Li+. Cycling to lower cut-off voltages than 2.0V may be possible, particularly when the device has been conditioned as disclosed herein, once soluble polysulfide anion has been irreversibly depleted from the aprotic electrolyte. In some embodiments, the capacitive energy storage device is cycled between a lower cut-off voltage of between 2.0V and 2.3 V vs Li / Li+and an upper cut-off voltage of between 3.8V and 4.0 V vs Li / Li+. Cycling in a narrower voltage range may provide extended cycle life, albeit at lower energy storage capacity in each cycle.

[0133] The capacitive energy storage devices according to embodiments of the invention are suitable for high power density applications, i.e. applications requiring more rapid charging and / or discharging than can be provided by a rechargeable battery. In some embodiments, therefore, the capacitive energy storage device is discharged and / or charged at a rate of at least 5 mA.cnr2, or at least 7.5 mA.cnr2, such as at least 10 mA.cnr2, in one or more of the charge-discharge cycles, or in each charge discharge cycle. In some embodiments, the capacitive energy storage device is discharged and / or charged at a rate of at least 10C, such as at least 20C, or at least 30C, or at least 50C. In contrast, batteries are typically cycled in the range of 1 to 2C.

[0134] The capacitive energy storage devices according to embodiments of the invention are highly stable over long term cycling. In some embodiments, therefore, the capacitive energy storage device retains at least 80%, such as at least 85%, for example at least 90%, of an initial capacitance corresponding to an initial chargedischarge cycle when subjected to at least 1000, preferably at least 10,000, subsequent charge-discharge cycles. The initial capacitance may be measured in any nominated charge-discharge cycle, for example after one or more conditioning cycles.

[0135] The capacitive energy storage devices according to embodiments of the invention provide excellent energy density, and in particular greater energy density than many commercially available supercapacitors. In some embodiments, the capacitive energy storage device discharges at least s Wh. kg1, such as at least 10 Wh.kg1, based on the total mass of the capacitive energy storage device (excluding packaging) in one or more of the charge-discharge cycles.EXAMPLES

[0136] The present invention is described with reference to the following examples. It is to be understood that the examples are illustrative of and not limiting to the invention described herein.Example 1. Preparation of cathodes

[0137] Activated carbon (AC) electrodes for coin cells were made by doctor blading a slurry of 80 wt% AC (ASAC-25, Adven), 10 wt% carbon black (Black Pearl 2000, CABOT) and 10 wt% carboxymethylcellulose (CMC) in water (c.a. 25% solids) onto a sheet of aluminium foil current collector, and drying to remove the water. Electrodes of 1.13 cm-2were cut out of the sheet.

[0138] The AC used has similar characteristics to that used in commercial supercapacitor electrodes. Surface area and pore textural characteristics were obtained from the N2 adsorption isotherms for pressures up to 1 bar measured by a volumetric method using a Micromeritics ASAP 2420 instrument at 77 K (liquid nitrogen bath). Samples were evacuated and activated at 200 °C under dynamic vacuum at 10-6 Torr for 12 h to remove any residual solvent and measure the sample mass precisely. Gas adsorption measurements were performed using ultrahigh-purity nitrogen. Brunauer-Emmett-Teller surface area and pore size distribution data were calculated from the N2 adsorption isotherms based on the density functional theory model in the software provided within the Micromeritics ASAP 2420 instrument. The pore size distribution of the AC is shown in Figure 2. The BET surface area of the AC is about 2000-2500 m2 / g.

[0139] Reduced graphene oxide (rGO) was prepared via a thermal expansion process of graphene oxide, and followed by a ball milling process as described by Dong et al (Advanced Functional Materials 2019, 29 (24), 1901 127). The BET surface area of the rGO is about 200-300 m2 / g. An rGO ink was produced by mixing all components (rGO and 5% CMC in water, c.a. 25% solids) in a vacuum mixer at 600 RPM for 1 hour. The rGO electrodes were made by doctor blading the slurry onto a sheet of aluminium foil current collector and drying to remove the water. Electrodes of 1 .13 cm-2were cut out of the sheet.

[0140] For both AC and rGO electrodes, the amount of active material on the electrodes was measured using a microbalance of a thermogravimetric instrument with a weighing precision of 0.1 pg. The areal loading of carbon was controlled in the range of 1 .9 - 6 mg cm-2.

[0141] Cathodes were produced as shown in T able 1 .Table 1.Example 2. Preparation of electrolyte

[0142] A Li2Se solution was prepared according to previously reported methods (Zhou et al, Journal of Materials Chemistry A 2016, 4 (15), 5406-5409). Thus, elemental sulfur and U2S powder were mixed in an DOL + DME solvent (DOL:DME = 1 :1 v / v) at 50 °C for 36 h under stirring in an argon glove box, with a molar ratio of 8:5 (8 U2S + 5Ss — 8Li2Se). The resultant product was centrifuged at 3610.7 x g for 10 min to abandon the particles inside, and the remaining red-brown solution is the I 2S6 solution.

[0143] Ether-based electrolytes (<0.003% water content) were prepared by dissolving bis(trifluoromethylsulfonyl)imide lithium (LiTFSI) and lithium nitrate (LiNOa) to the desired concentrations in DOL and DME (1 :1 , v / v) in an Ar-containing glovebox (<0.1 ppm H2O and <0.1 ppm O2). The U2S6 solution was then added to the electrolyte to give the desired concentration of Li PS. The addition of polysulfide was observed to turn the electrode a yellowish colour.

[0144] A comparative carbonate-based electrolyte was prepared by dissolving LiPFe in a 1 :1 (v / v) mixture of ethylene carbonate (EC) and dimethylcarbonate (DMC).

[0145] Electrolytes were thus produced as shown in Table 2.Table 2.tetraethylammonium tetrafluoroborate; electrolyte procured from Sigma AldrichExample 3. Fabrication of coin cell energy storage devices

[0146] Lithium metal capacitors, in the form of a standard coin cell (CR2032), were fabricated using porous carbon electrodes (as produced in Example 1 ) as cathode, a Celgard porous polyethylene membrane (Celgard 2730, 20 pm thickness, 16 mm diameter, 1 pm pore size, and 43% porosity) as separator, lithium chip anode (15.6 mm diameter x 0.2 mm, purchased from Shandong Gelon LIB Co. Ltd., China) and 40 pL of electrolyte (as produced in Example 2).

[0147] A coin cell comprising cathode C1 -1 (as produced in Example 1 ) and electrolyte E2-1 (as produced in Example 2) had the following features: electrolyte :carbon ratio = c.a. 13 pL.mg1; LiPS loading of 0.082 mg; LiPS areal loading of 0.043 mg cm2(relative to anode area).

[0148] A coin cell supercapacitor (CR2032) was fabricated using two porous carbon electrodes (C1 -1 ; Example 1 ), a Celgard membrane separator and 40 pL of electrolyte E2-5 (as produced in Example 2).

[0149] Coin cell (CR2032) symmetrical lithium cells were fabricated using two lithium chip electrodes (15.6 mm diameter x 0.2 mm), a Celgard membrane, and 40 pL of electrolyte (as produced in Example 2).Example 4. Conditioning of coin cell lithium metal capacitor

[0150] Lithium metal capacitors, as fabricated in Example 3 with the C1 -1 cathode and E2-1 electrolyte, were cycled using a BioLogic Potentiostat to evaluate electrochemical performance.

[0151] In an effort to form and stabilise an effective SEI layer, the lithium metal capacitors were initially cycled at low scan rates, of 2 mVs’1in cyclic voltammetry mode or at 1 mA cm2(0.33 A g1of AC) in constant current mode, in the range of 2.0 to 4.0 V vs Li / Li+.

[0152] The cyclic voltammogram in Figure 3 reveals an incomplete reduction peak between 2.3 to 2.0 V during the early cycles (cycles 1 and 25 shown), assigned to the reduction of U2S6 to U2S4 on the surface and probably within the bulk of the conducive and porous AC cathode. A subsequent oxidation peak between 2.3 and 2.5 V is assigned to the conversation of U2S4 to U2S6 and probably U2S8. All of these species are highly soluble in the electrolyte. SEI formation cycles were continued until the disappearance of the redox peaks, thus ensuring that substantially all the soluble lithium polysulfide species had been utilized in the formation of the SEI (and thus irreversibly depleted from the electrolyte). After that, the cell exhibited the rectangular capacitive behaviour of a typical LIC device configured with a lithiated graphite anode and an AC cathode (representative cycle 50 shown).

[0153] Similarly, the voltage-discharge capacity profile in Figure 4, for the constant current SEI formation mode, exhibited a short plateau at 2.3-2.0 V related to the conversion reaction of polysulfides within the higher-order range (Li2Sx, 4<x<8).

[0154] It is important to set the lower cut-off voltage above 2.0 V to prevent the uncontrolled reduction of U2S6 and U2S4 to the insoluble U2S2 and U2S during conditioning. These insoluble species are the final products of the reactions between lithium and sulfur. Their deposition on the surface of the AC cathode results in localized passivation of the cathode surface. The resultant blocking reduces the high-rate capability of a LMC device. It is also considered important to cycle the LMC cell at low current densities during conditioning (substantially lower current densities than in use for energy storage application), thus allowing a favourable SEI to be established under mild conditions.Example 5. Long-term cycling of coin cell lithium metal capacitors

[0155] Lithium metal capacitors, as fabricated in Example 3 with the C1 -1 cathode (3 mg / cm2AC) and different electrolytes, were cycled using a BioLogic Potentiostat to investigate the effect of the electrolyte composition on electrochemical performance. The cells were initially conditioned as described in Example 4 (c.a. 50 cycles). Figure 5 compares the subsequent cycling performance of cells with (1 ) ether-based electrolyte containing LiNOa and LiPS additives (E2-1 ), (2) ether-based electrolyte containing LiNOa additive only (E2-2) and (3) carbonate-based electrolyte without additives (E2-4) over 6000 cycles conducted at different cycling rates (in constant current mode). Cycles 1 -500 were conducted at 3 mA cm-2, cycles 501 -1000 at 6 mA cm2, cycles 1001 -5000 at 12 mA cm2, and cycles 5001 -6000 again at 3 mA cm2(same current densities for both charging and discharging steps).

[0156] The performance of a cell with ether-based electrolyte containing no additives (E2-3) was also investigated (not shown in Figure 5). In the absence of LiNOa additive, the cell failed after only 50 cycles. This indicates the important role that LiNOa plays in passivating the lithium metal anode and establishing a viable SEI during cycling with an ether-based electrolyte.

[0157] The capacitance of the cell with only LiNOa in DOL / DME (E2-2) was initially higher than the cell with both LiNOa and LiPS in DOL / DME (E2-1 ) and the cell with the commercial LiPFe in EC / DMC electrolyte (E2-4) at all the cycling rates. For example, in cycles 1 -500 at current density of 3 mA cm’2(equivalent to 1 A g’1), the AC cathode delivers 60 mAh g1with the ether-based electrolyte and 55 mAh g1with the carbonate electrolyte, a demonstration of the higher energy that can be achieved by using an ether-based electrolyte instead of the carbonate electrolyte conventionally used in lithium ion capacitor systems. Without wishing to be limited by theory, it is proposed that the larger capacitance achieved in the ether-based electrolyte arises from the larger solvated radius of TFSI anions compared to PFe. The distortion of solvation shells of the larger size TFSI anions in small pores (0.8-1.4 nm) of typical AC materials may allow a closer approach of the ion centre to the electrode surface, leading to improved capacitance.

[0158] While Li PS did not improve initial capacitance (cf. cycles 1 -1500 for electrolytes E2-1 vs E2-2), the advantage of the LiPS additive is evident from the impressive stability of capacitance in long-term cycling - particularly when the cell is subjected to cycling at high current density. All three cells shown in Figure 5 exhibited good stability over the first 1000 cycles at 3 and 6 mA cm-2. However, significant differences were observed during the 4000 cycles at 12 mA cm-2. When the cycling current was subsequently switched back to 3 mA cm’2, allowing a comparison against the initial performance during cycles 1 -500, it was observed that the cell with carbonate- based electrolyte (E2-4) had lost half of its initial capacity, with rapid capacity decay continuing afterwards. In stark contrast, the cell with ether-based electrolyte and both LiNOa and LiPS additives (E2-1 ) substantially reverted to its initial capacity, indicating minimal damage to the lithium metal anode and electrolyte composition despite the high-rate long-term cycling. The importance of the LiPS additive to complement the LiNOa is evident from the comparative failure of the cell with ether-based electrolyte but only LiNOa additive (E2-2), which rapidly degraded after only 1000 cycles at 12 mA cm’2and failed soon after 2000 cycles at 12 mA cm’2.

[0159] Without wishing to be limited by theory, the improved performance obtained when using both LiNOa and LiPS additives was attributed to the formation of a solid electrolyte interphase layer between the lithium metal anode and the electrolyte which is highly effective to modulate lithium electrodeposition on the lithium metal anode, even at the high charge / discharge current densities characteristic of a capacitive energy storage device.Example 6. Cycling of coin cell lithium metal capacitors to different upper cut-off voltages

[0160] A lithium metal capacitor, as fabricated in Example 3 with the C1 -1 cathode (3 mg / cm2AC) and the E2-1 electrolyte (ether-based with LiPS and LiNOa additives), was cycled using a BioLogic Potentiostat to investigate the effect of extended voltage ranges on electrochemical performance. The lithium metal capacitor was cycled through 50 cycles at a rate of 2 mVs’1with a lower cut-off voltage of 2 V vs Li / Li+and upper cut-off voltages of 4.0 V, 4.1 V, 4.2 V, 4.3 V, 4.4 V and 4.5 V (10 cycles at each condition). As can be seen in the cyclic voltammograms in Figure 6, which show the10th cycle at each condition, the rectangular behaviour starts to break down at 4.3 V, which suggests a limiting voltage of 4.2 V for this system.Example 7. Long-term cycling of coin cell lithium metal capacitors in different voltage ranges

[0161] Lithium metal capacitors, as fabricated in Example 3 with the C1 -1 cathode (3 mg / cm2AC) and the E2-1 electrolyte (ether-based with LiPS and LiNOa additives), were cycled using a BioLogic Potentiostat to investigate the electrochemical performance in long-term cycling. The cells were initially conditioned as described in Example 4 (c.a. 50 cycles).

[0162] Figure 7 demonstrates the long-term cycling behaviour of such a cell at an extended voltage window of 2 - 4.2 V and a current density of 3 mA cm-2. It is believed that the low Coulombic efficiency over the first 500 cycles is caused by high voltage solvent oxidation in an extended conditioning phase. The cell demonstrates an excellent specific capacity of 85 mAh g-1, and retains more than 80 % of the initial capacity for nearly 1400 cycles. After that, there is a relatively sudden decay in the performance of the cell, where the Coulombic efficiency drops from a stable 99.4 % to 98.5 % within only 40 cycles. The result suggests a shorter but nevertheless useful life of the device when operated at a higher voltage window (up to 4.2 V) to increase the capacity.

[0163] Commercially available LIC devices using AC cathodes are commonly rated for use in the range of 2.2-3.8 V. Figure 8 shows the cycle life of the new lithium metal capacitor operating in this voltage window at a cycle rate of 6 mA cm’2. The cell demonstrates capacity retention of 95 % over 8000 cycles, and near-quantitative Coulombic efficiency throughout.Example 8. Comparative performance of coin cell lithium metal capacitors against EDLC supercapacitor

[0164] The comparative performance of (i) a lithium metal capacitor, as fabricated in Example 3 with the C1 -1 cathode (3 mg / cm2AC) and the E2-1 electrolyte (ether- based with LiPS and LiNOa additives), and conditioned as described in Example 4 (c.a. 50 cycles), and (ii) a symmetrical supercapacitor with identical AC cathodes andNEt4BF4 in acetonitrile electrolyte (Example 3) was investigated by cycling the devices using a BioLogic Potentiostat. Figure 9 shows the cycling behaviour of the cells at a current density of 3 mA cm’2(1 A g’1of AC). The supercapacitor was cycled between 2.0 and 2.7 V (the maximum voltage of supercapacitors) while the lithium metal capacitor was cycled between 2.0 and 4.0 V. Since the amount of energy stored in a capacitor increases quadratically with voltage, the energy density of the lithium metal capacitor can be expected to be more than twice that of an equivalent EDLC supercapacitor. The measured capacity of the lithium metal capacitor was indeed about double that of the supercapacitor, with excellent capacity retention and Coulombic efficiency over 2000 cycles.Example 9. Cycling of coin cell lithium metal capacitors with increased cathode loading

[0165] A lithium metal capacitor, as fabricated in Example 3 with the C1 -3 cathode having elevated AC loading (6 mg / cm2AC) and the E2-1 electrolyte (ether-based with LiPS and LiNOa additives), and conditioned as described in Example 4 (c.a. 50 cycles), was cycled using a BioLogic Potentiostat to investigate the electrochemical performance. As seen in Figure 10, cycling of the cell at a current density of 1 A g1(6 mA cm’2) delivered a high areal capacity of about 0.3 mA cm’2with excellent capacity retention over 500 cycles. Figure 1 1 shows the capacity of lithium metal capacitors, as fabricated in Example 3 with a range of different AC loadings (C1 -1 , C1 -2 and C1 -3 cathodes with 3, 4 and 6 mg / cm2AC respectively) and the E2-1 electrolyte, and conditioned as described in Example 4. The cells were cycled at a current density of 1 A g’1in the range of 2.0 to 4.0 V. While the gravimetric capacity decreases with increased AC loading, the areal capacity and thus the total energy storage capacity of the device increases. The results indicate that the areal loading of AC can only be increased to a certain optimum before limitations on electrolyte transport within the cathode layer outweigh the improved energy storage capacity.Example 10. Long-term cycling of coin cell lithium metal capacitors with a reduced graphene oxide cathode

[0166] A lithium metal capacitor with reduced graphene oxide cathode, as fabricated in Example 3 with the C1 -4 cathode (1.9 mg / cm2rGO) and E2-1 electrolyte(ether-based electrolyte with LiNOa and LiPS additives), and initially conditioned as described in Example 4 (c.a. 50 cycles), was cycled using a BioLogic Potentiostat. Figure 12 shows the performance of the cell over 6000 cycles conducted at a cycling rate of 1 A g1(1 .9 mA cm2) in the voltage range of 2.8 V to 3.8 V. The cell retained more than 80% of the initial capacity over 30,000 cycles while maintaining a Columbic efficiency of > 99% throughout.Example 11. Fabrication of pouch cell lithium metal capacitors

[0167] A significant challenge for lithium-metal batteries is the transition from high- performing lab-scale coin cells to pouch cell prototypes. Lithium metal capacitors, in the form of a pouch cell, were thus fabricated to evaluate scalability of the newly developed capacitive device.

[0168] Double-sided capacitive electrodes (3 cm x 4.5 cm) were prepared by coating AC slurry (as described in Example 1 ) on both sides of an Al foil, yielding AC loading of about 4 mg cm-2. The total amount of AC on each electrode was thus 108 mg. An Al tab was welded onto the as-prepared electrode. After stacking the electrode against a Celgard separator, the stack was transferred into an Ar glovebox (<0.1 ppm H2O and <0.1 ppm O2) and a Li foil anode (100 pm thickness, cut to the same size, 3 cm x 4.5 cm, as the AC cathode) was stacked against the separator. A Ni tab was bonded to the lithium anode with a two-spot welder. The required amount of electrolyte (ether-based with LiPS and LiNOa additives; Example 2) was injected into the stack. Then, the pouch cell package was sealed under vacuum. The pouch cells had a capacitance of 10 Farads.

[0169] Pouch cells containing (1 ) 980 pL of electrolyte E2-1 and (2) 900 pL of electrolyte E2-6 were produced by this methodology. The first pouch cell had the following features: electrolyte:carbon ratio = c.a. 9 pL.mg1; LiPS loading of 2.01 mg; LiPS areal loading of 0.074 mg cm2(relative to anode area). The second pouch cell had the following features: electrolyte:carbon ratio = c.a. 8 pL.mg-1’ LiPS loading of 1 .85 mg; LiPS areal loading of 0.068 mg cm2(relative to anode area). Both cells were initially conditioned by 50 cycles at 1 mA cm-2(0.25 A g-1AC) in constant current mode, in the range of 2.2 to 3.8 V vs Li / Li+.Example 12. Cycling of pouch cell lithium metal capacitors

[0170] Figure 13 shows the cycling performance of the 10 F lithium metal capacitor pouch cell containing 980 pL of electrolyte E2-1 (produced in Example 1 1 ) over 100 cycles at 1 A g1, equivalent to a cycling rate of 30C, in the voltage range of 2.2 to 3.8 V vs Li / Li+. The capacitor was subject to an external pressure of 43 kPa applied in the normal direction to the stack by a kettlebell during cycling. The capacity retention of the pouch cell after 1000 cycles was 93.2%, despite the high cycling rate, with the specific capacity being about 10-20% lower than for an equivalent coin cell. The Coulombic efficiency of the cell was 100% through the cycling.

[0171] The energy density of the cell was calculated to be about 8.0 Wh kg’1(excluding the packaging) at 1 A g-1. The weight proportion of each component of the pouch cell (excluding packaging) is also shown in Figure 13. The electrolyte contributes 78% of the weight, with the ether solvents (DOL and DME) being 85% of this.

[0172] To increase the specific energy at the pouch level, the volume of the electrolyte in the cell was decreased while increasing the concentration of LiTFSI in the electrolyte from 0.75 M to 2 M. Figure 14 shows the cycling performance of the 10 F lithium metal capacitor pouch cell containing 900 pL of electrolyte E2-6 at a series of different cycle rates in the voltage range of 2.2 to 3.8 V vs Li / Li+. The capacitor was subject to an external pressure of 43 kPa applied in the normal direction to the stack by a kettlebell during cycling. This second cell, where the electrolyte contributes only 71 % of the weight, provided 22% higher specific energy at the same current density (1 .0 A g’1) as the first pouch cell. The second cell provided an energy density of 13 Wh kg’1at 0.1 A g’1, 1 1 .9 Wh kg’1at 0.5 A g’1, and 11 .0 Wh kg’1at 1 A g’1.Example 13. Comparison of pouch cell lithium metal capacitors against commercial lithium ion capacitors

[0173] The performance of the 10 Farad pouch cell lithium metal capacitor (reduced volume of electrolyte, as tested in Example 1 1 ) was compared against that of a commercial 14 Farad lithium ion capacitor (cylindrical format) produced by Jianghai (Nantong Jianghai Capacitor Co). The Jianghai device included a pre-lithiated graphite anode and cathode with about 7 mg cm’2loading of AC. The mass of the device components was determined by device disassembly and weighing.

[0174] The comparison in device characteristics, based on cycling the second pouch cell in Example 12 at 1 A g1in the voltage range of 2.2 to 3.8 V vs Li / L is shown in Figure 15. Despite using an AC electrode with 42% less areal loading of AC, the specific energy of the lithium metal capacitor is only 28% less than that of the commercial lithium ion capacitor. This suggests that a lithium metal capacitor may produce similar or even superior performance while being produced with a considerably more straightforward manufacturing process.Example 14. Characterisation of lithium metal morphology and SEI composition after cycling symmetrical lithium cells in different electrolytes

[0175] Symmetrical lithium cells, as fabricated in Example 3 with electrolytes E2-1 , E2-2 or E2-4, were cycled using a BioLogic Potentiostat to investigate the effect of electrolyte composition on the composition of the SEI layer and the evolution of lithium metal anode morphology. The cells were cycled through the following sequence: 100 cycles at 1 mA cm-2, 100 cycles at 4 mA cm-2, 100 cycles at 8 mA cm-2, and 100 cycles at 12 mA cm-2. Each cycle consisted of 6 minutes charge and 6 minutes discharge with no open circuit time between. The cycling steps at a charge density of 4 mA cm-2and above are considered representative of the operation of an LMC.

[0176] After cycling, the cells were disassembled to recover the lithium metal electrodes for analysis by scanning electron microscopy (SEM). Transport of the lithium anodes from glovebox to the SEM was performed using a remote-controlled atmosphere sensitive sample transfer shuttle. SEM was performed using a Zeiss Merlin field emission scanning electron microscope operated in the secondary electron. Energy dispersive X-ray spectroscopy (EDS) was used to identify elements present within the samples. The EDS system used was X-Max Xtreme, manufactured by Oxford Instruments Pty Ltd. An accelerating voltage > 5 kV was used for EDS mapping.

[0177] Figure 16 depicts top-view SEM images of (a) the lithium electrodes without cycling, (b) the lithium electrode cycled in carbonate-based electrolyte without additives (E2-4), (c) the lithium electrode cycled in ether-based electrolyte containing LiNOa additive only (E2-2) and (d) the lithium electrode cycled in ether-based electrolyte containing LiNOa and LiPS additives (E2-1 ).

[0178] The lithium anode cycled in electrolyte E2-4 was cracked (as seen in lower magnification SEM images) and the electrodeposited lithium on the anode surface exhibited a mossy, uneven and high surface area morphology (Figure 16b). A similar lithium morphology (albeit without anode cracking) was observed for the lithium anode cycled in electrolyte E2-2 (Figure 16c), indicating that LiNOa by itself does not satisfactorily mediate the lithium electrodeposition. Dendritic, high surface are lithium electrodeposition during charging of a LMC is expected to damage the SEI. As a result of the continuous damage and reformation of the SEI during cycling, the electrolyte is gradually consumed leading to early cell failure.

[0179] The lithium anode cycled in electrolyte E2-1 exhibited a very different, low surface area morphology (Figure 16d), with large plates of lithium formed on the surface in contrast to the dendritic morphology obtained in the other electrolytes. Such lithium electrodeposition during charging of a LMC is expected to reduce or avoid damage to the SEI and the resultant electrolyte consumption, thus extending the cell life.

[0180] The electrodes cycled in electrolyte E2-2 and electrolyte E2-1 were also analysed by out X-ray photoelectron spectroscopy (XPS), after first rinsing with DOL solvent to remove any salt residues. XPS was performed on a Thermo Scientific Nexsa Surface Analysis System calibrated to Au 4f and equipped with a hemispherical analyser. The incident radiation was monochromatic Al Ka X-rays (1486.6 eV) at 72 W (6 mA and 12 kV, 400 x 800 pm2 spot). Survey (wide) and high-resolution (narrow) scans were recorded at analyser pass energies of 150 and 50 eV and step sizes of 1 .0 eV and 0.1 eV, respectively. The base pressure in the analysis chamber was less than 5.0 x 10-9 mbar. A low-energy dual-beam (ion and electron) flood gun was used to compensate for surface charging. To prevent surface oxidation prior to analysis, the samples were transported from an argon glove box to the load chamber of the XPS instrument using a vacuum transfer module. Data processing was carried out using Avantage software version 5.9921 and the energy calibration was referenced to the main line of C 1 s at 284.8 eV. All peaks were modelled using the Smart background type and the sulphur 2p3 / 2 - 2p1 / 2 doublet separation set to 1.18eV and peaks constrained to a 2:1 area ratio (2p3 / 2: 2p1 / 2).

[0181] The significant change in the chemical composition of the SEI formed when LiPS was present in the electrolyte is apparent in the high-resolution S2p, C1 s, and F1 sspectra. The S2p spectra for lithium electrode cycled in ether-based electrolyte containing LiNOa additive only (E2-2) and ether-based electrolyte containing LiNOa and LiPS additives (E2-1 ) are shown in Figure 17 (a) and (b) respectively.

[0182] The S2p spectrum of the electrode cycled without LiPS additive is fitted with a pair of doublet peaks in binding energies of 165-173 eV, assigned to fresh / decomposed LiTFSI compounds (RSO2R, sulphates, and sulphites), and a very low-intensity doublet peak in binding energies of 158-165 eV, attributed to insoluble LiaS and U2S2 compounds. The large abundance of fresh and decomposed LiTFSI species is attributed to trapping of the electrolyte in the porous morphology of this lithium electrode. In contrast, the S2p spectrum of the lithium anode cycled in the presence of LiPS was mainly fitted with the LiaS / LiaSa doublet peak with only a very small contribution from the decomposed LiTFSI species (sulphates, and sulphites) and no contribution from fresh LiTFSI. The difference in the intensity of the LiaS and U2S2 peaks in the S2p spectrum of the two electrodes indicates that the soluble LiPS additive in the electrolyte is incorporated into the SEI, likely via reduction by lithium metal to form insoluble LiaS and U2S2. It is proposed that the LiPS-containing electrolyte thus facilitates the formation and stabilization of a SEI with desirable composition and properties for mediating lithium electrodeposition. The F1 s spectra indicated the presence of LiF, in similar amounts, as another significant component of the SEI on both electrodes.Example 15. Characterisation of lithium metal morphology and SEI composition after cycling a LMC in different electrolytes

[0183] Lithium metal capacitors, as fabricated in Example 3 with the C1 -1 cathode (3 mg / cm2AC) and either the E2-1 or E2-2 electrolyte were cycled using a BioLogic Potentiostat for 300 cycles at a moderate current density of 1 A / g. No performance degradation was thus yet evident in either cell. However, the post-cycling morphology of the lithium metal anode surface of the two cells, as determined by SEM analysis (as described in Example 12), was very different. The anode cycled in the ether-based electrolyte containing LiNOa additive only (E2-2) had a high surface area, mossy I dendritic morphology (similar to Figure 16c) while the anode cycled in the ether-based electrolyte containing LiNOa and LiPS additives (E2-1 ) had a low surface area, platelike morphology (similar to Figure 16d).

[0184] EDS analysis of the electrode surfaces showed (i) a larger concentration of sulfur on the anode cycled in electrolyte E2-1 , indicating that the LiPS additive is incorporated into the SEI, and (ii) a larger concentration of oxygen on the anode cycled in electrolyte E2-2, consistent with greater electrolyte decomposition on the high surface area lithium morphology.Example 16. Effect of Lithium nitrate (LiNOs) loading in pouch cell lithium metal capacitors

[0185] Pouch cell lithium metal capacitors with different loading of LiNOs in the electrolyte were prepared and conditioned by the method of Example 11 . The capacitive carbon cathode had an AC loading of 3.3 mg cm-2and the Li foil anode thickness was 100 pm. Electrolytes were prepared as shown in Table 3, and added to the pouch cell in an amount of 635 pL (maintaining a similar AC:electrolyte ratio as Example 1 1 ).Table 3.

[0186] Figure 18 shows the cycling performance of the lithium metal capacitor pouch cells with the three electrolytes over 1000 cycles at 0.5 A g1in the range of 2.2 to 3.8 V vs Li / Li+. The pouches were subject to an external pressure of 43 kPa applied in the normal direction to the stack by a kettlebell. At a low nitrate loading (E16-3), the cell was stable for approximately 200 cycles but deteriorated thereafter. Much longer cycle lifetimes were evident at a medium nitrate loading (E16-2) and a high nitrate loading (E16-1 ; 0.75 M is near the solubility limit in this system). The results further demonstrates the contribution of nitrate to the formation of a stable SEI, in cooperation with the polysulfide additive.Example 17. Effect of polysulfide (LiPS) loading in pouch cell lithium metal capacitors

[0187] Pouch cell lithium metal capacitors with different loading of LiPS in the electrolyte were prepared and conditioned by the method of Example 11. The capacitive carbon cathode had an AC loading of 3.2 mg cm-2and the Li foil anode thickness was 100 pm. Electrolytes were prepared as shown in Table 4, and added to the pouch cell in an amount of 1200 pL.Table 4.

[0188] Figure 19 shows the cycling performance of the lithium metal capacitor pouch cells with both electrolytes at 0.5 A g’1(corresponding to 15 C), in the range of 2.2 to 3.8 V vs Li / Li+. The pouches were subject to an external pressure of 43 kPa applied in the normal direction to the stack by a kettlebell. Both pouch cells performed well, with the higher LiPS loading demonstrating slightly better cycling stability and comparable specific capacity.Example 18. Effect of preconditioning on pouch cell lithium metal capacitors

[0189] Pouch cell lithium metal capacitors with electrolyte E16-1 (Table 3) were prepared by the method of Example 11 . The capacitive carbon cathode had an AC loading of 3.3 mg cm’2and the Li foil anode thickness was 100 pm. The pouches underwent different pre-conditioning steps in the form of pre-cycling at 2 mVs’1and 1 mVs'1. As seen in Figure 20, both pouch cells exhibited similar initial capacitance but the pouch preconditioned at the low rate showed better long-term stability.Example 19. Effect of stack pressure on pouch cell lithium metal capacitors

[0190] Pouch cell lithium metal capacitors with electrolyte E16-1 (Table 3) were prepared by the method of Example 1 1 . The capacitive carbon cathode had an AC loading of 3.3 mg cm2and the Li foil anode thickness was 100 pm. One pouch was subject to an external pressure of 43 kPa applied in the normal direction to the stack by a kettlebell, while no pressure was applied to the other pouch. Electrochemical impedance spectroscopy (EIS) was conducted on both pouches in a fully charged state. As seen in Figure 21 , the pouch cell experiencing no pressure develops higher internal resistance compared to the pouch cell experiencing pressure. The pouch cell experiencing no pressure also exhibited poor rate capability and unstable cycling performance. The result demonstrates the importance of external pressurisation to ensure that the pouch cells have a low internal resistance.

[0191] Those skilled in the art will appreciate that the invention described herein is susceptible to variations and modifications other than those specifically described. It is understood that the invention includes all such variations and modifications which fall within the spirit and scope of the present invention.

Claims

AMENDED CLAIMS received by the International Bureau on 24 October 2024 (24.10.2024) Claims1 . A capacitive energy storage device comprising: a lithium metal anode; a capacitive cathode comprising an ion-adsorption electrode material; an aprotic electrolyte comprising lithium salt; a nitrogen-based component selected from (i) a nitrogen oxide anion, present in the aprotic electrolyte, (ii) a reaction product of a nitrogen oxide anion, present in a solid electrolyte interphase layer between the lithium metal anode and the aprotic electrolyte, and (iii) combinations thereof; and a sulfur-based component selected from (i) a polysulfide anion, present in the aprotic electrolyte, (ii) a polysulfide anion or reaction product thereof, present in a solid electrolyte interphase layer between the lithium metal anode and the aprotic electrolyte, and (iii) combinations thereof.

2. The capacitive energy storage device according to claim 1 , wherein the sulfurbased component is present in an amount of less than 0.5 mg cm-2, based on the mass of sulfur in the sulfur-based component relative to the lithium metal anode surface area.

3. The capacitive energy storage device according to claim 1 or claim 2, wherein the nitrogen oxide anion is nitrate.

4. The capacitive energy storage device according to any one of claims 1 to 3, wherein the ion-adsorption electrode material is a carbon-based electrode material.

5. The capacitive energy storage device according to claim 4, wherein the carbonbased electrode material is selected from the group consisting of activated carbon, porous carbon, graphene, reduced graphene oxide, expanded graphite, exfoliated graphite, carbon nanotubes, carbon aerogel and combinations thereof.AMENDED SHEET (ARTICLE 19)6. The capacitive energy storage device according to any one of claims 1 to 5, wherein the ion-adsorption electrode material has a BET surface area of between about 100 and 3000 m2 / g.

7. The capacitive energy storage device according to any one of claims 1 to 6, wherein the lithium salt comprises a fluorinated anion selected from the group consisting of fluorinated alkyl sulfonyl imide, a fluorinated alkyl sulfonyl methide, a fluorinated alkyl sulfonate, a fluorinated alkyl carbonate, a fluorinated aryl borate, a fluorinated alkyl phosphate and combinations thereof.

8. The capacitive energy storage device according to any one of claims 1 to 7, wherein the aprotic electrolyte comprises a molecular liquid carrier comprising one or more ethers.

9. The capacitive energy storage device according to any one of claims 1 to 8, wherein the aprotic electrolyte comprises lithium cations in a concentration of between 0.2 and 3 M.

10. The capacitive energy storage device according to any one of claims 1 to 9, wherein the aprotic electrolyte comprises the nitrogen oxide anion in a concentration of from 0.25 to 0.75 M.11 .The capacitive energy storage device according to any one of claims 1 to 10, comprising a reaction product of a polysulfide anion, present in a solid electrolyte interphase layer between the lithium metal anode and the aprotic electrolyte.

12. The capacitive energy storage device according to claim 11 , wherein the aprotic electrolyte is substantially free of soluble polysulfide anion.

13. The capacitive energy storage device according to any one of claims 1 to 12, wherein the sulfur-based component comprises U2S2 and / or U2S, present in the solid electrolyte interphase layer.AMENDED SHEET (ARTICLE 19)14. The capacitive energy storage device according to any one of claims 1 to 13, wherein the solid electrolyte interphase layer comprises a reaction product of the nitrogen oxide anion.

15. The capacitive energy storage device according to any one of claims 1 to 14, which retains at least 85% of an initial capacitance when subjected to at least 1000 subsequent cycles at 1 A g1, based on the mass of ion-adsorption electrode material, between a lower cut-off voltage of 2.2V vs Li / Li+and an upper cut-off voltage of 3.8V vs Li / Li+.

16. The capacitive energy storage device according to any one of claims 1 to 15, having a specific energy of at least 8 Wh.kg1based on the total mass of the capacitive energy storage device excluding packaging when cycling the capacitive energy storage device at 1 A g1, based on the mass of ion-adsorption electrode material, between a lower cut-off voltage of 2.2V vs Li / Li+and an upper cut-off voltage of 3.8V vs Li / Li+.

17. A method of conditioning a capacitive energy storage device, the method comprising: providing a capacitive energy storage device comprising: a lithium metal anode; a capacitive cathode comprising an ion-adsorption electrode material; and an aprotic electrolyte comprising lithium salt, a nitrogen oxide anion and a polysulfide anion; and cycling the capacitive energy storage device through one or more conditioning charge-discharge cycles to form a solid electrolyte interphase layer between the lithium metal anode and the aprotic electrolyte, wherein the solid electrolyte interphase layer comprises a reaction product of the polysulfide anion.

18. The method according to claim 17, wherein the capacitive energy storage device is cycled across a voltage range having a lower cut-off voltage of at least 2V vs Li / Li+.AMENDED SHEET (ARTICLE 19)19. The method according to claim 17 or claim 18, wherein the capacitive energy storage device is cycled across a voltage range having an upper cut-off voltage of no more than 4.0V vs Li / Li+.

20. The method according to any one of claims 17 to 19, wherein cycling the capacitive energy storage device through the one or more conditioning chargedischarge cycles comprises charging and discharging the capacitive energy storage device at a maximum rate of less than 2 mA.cnr2.21 .The method according to any one of claims 17 to 20, wherein the capacitive energy storage device is cycled through a plurality of conditioning chargedischarge cycles until the aprotic electrolyte is substantially free of soluble polysulfide anion.

22. The method according to any one of claims 17 to 21 , wherein the polysulfide anion is present, prior to cycling the capacitive energy storage device, in an amount of less than 0.5 mg cm2based on the mass of sulfur in the polysulfide anion relative to the lithium metal anode surface area.

23. The method according to any one of claims 17 to 22, wherein the ion-adsorption electrode material is a carbon-based electrode material.

24. The method according to any one of claims 17 to 23, wherein the electrolyte comprises the nitrogen oxide anion in a concentration of from 0.25 to 0.75 M.

25. The method according to any one of claims 17 to 24, wherein the reaction product of the polysulfide anion comprises U2S2 and / or U2S.

26. A method of storing electrical energy, comprising cycling (i) a capacitive energy storage device according to any one of claims 1 to 16 or (ii) a capacitive energy storage device conditioned by a method according to any one of claims 17 to 25 through a plurality of charge-discharge cycles.AMENDED SHEET (ARTICLE 19)27. The method according to claim 26, wherein the capacitive energy storage device is cycled through at least 1000 charge-discharge cycles, and wherein the capacitive energy storage device provides a Coulombic efficiency of at least 98% in a charge-discharge cycle following the at least 1000 charge-discharge cycles.

28. The method according to claim 26 or claim 27, wherein the capacitive energy storage device is cycled between a lower cut-off voltage of between 2.0V and 2.3V vs Li / Li+and an upper cut-off voltage of between 3.8V and 4.0V vs Li / Li+in one or more of the charge-discharge cycles.

29. The method according to any one of claims 26 to 28, wherein the capacitive energy storage device is discharged at a rate of above 5C in one or more of the charge-discharge cycles.

30. The method according to any one of claims 26 to 29, wherein the capacitive energy storage device retains at least 85% of an initial capacitance corresponding to an initial charge-discharge cycle when subjected to at least 1000 subsequent charge-discharge cycles.AMENDED SHEET (ARTICLE 19)

Images