BORON NITRIDE-IONIC LIQUID COMPOSITES AND THEIR USE FOR ENERGY

STORAGE DEVICES

CROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application claims priority to U.S. Provisional Patent Application No. 62/128,353, filed on March 4, 2015. The entirety of the aforementioned application is incorporated herein by reference.

STATEMENT REGARDING FEDERALLY SPONSORED RESEARCH [0002] Not applicable.

BACKGROUND

[0003] Current components of energy storage devices have numerous limitations, including limited stability at high temperatures. As such, a need exists for the development of more stable energy storage device components.

SUMMARY

[0004] In some embodiments, the present disclosure pertains to compositions that include a boron nitride, an ionic liquid, and a lithium salt. In some embodiments, the ionic liquid in the composition includes a piperidinium-based ionic liquid, such as 1 -methyl- 1 -prop ylpiperidinium bis(trifluoromethylsulfonyl)imide. In some embodiments, the lithium salt is lithium bis(trifluoromethylsulfonyl)imide. In some embodiments, the boron nitride in the composition is hexagonal boron nitride. [0005] In some embodiments, the compositions of the present disclosure are in the form of a composite. In some embodiments, the compositions of the present disclosure are utilized as an electrolyte, a separator, or an electrolyte- separator combination in an energy storage device.

[0006] In some embodiments, the present disclosure pertains to methods of making the compositions of the present disclosure. In some embodiments, the methods of the present disclosure include a step of contacting a boron nitride, an ionic liquid, and a lithium salt. In some embodiments, the contacting occurs by mixing the boron nitride, the ionic liquid, and the lithium salt. In some embodiments, the boron nitride is in solid form while the ionic liquid and the lithium salt are in a solution. In some embodiments, the mixing occurs at a 1:2 boron nitride to solution wt./wt. ratio.

[0007] In some embodiments, the methods of the present disclosure also include a step of incorporating the compositions of the present disclosure into an energy storage device. In some embodiments, the present disclosure pertains to energy storage devices that contain the compositions of the present disclosure. In some embodiments, the compositions of the present disclosure are utilized as electrolytes, separators or electrolyte-separator combinations in the energy storage device. In some embodiments, the energy storage device is a battery, such as a lithium-ion battery. In some embodiments, the energy storage device is capable of being operated at temperatures of more than about 100°C.

DESCRIPTION OF THE FIGURES

[0008] FIGURE 1 provides schemes for making boron nitride-based composite materials (FIGS. 1A-B), and an illustration of a battery that contains the composite materials (FIG. 1C).

[0009] FIGURE 2 shows the measurement of transport, structural and electrochemical properties of various electrolyte systems, including the boron nitride-based composite materials. FIG. 2A provides an Arrhenius plot of ionic conductivity for the 1 mol L ^"1 solution of lithium bis(trifluoromethylsulfonyl)imide ("LiTFSI") in 1 -methyl- 1 -prop ylpiperidinium bis(trifluorosulfonyl)imide ("PP13"), and for the BN- and clay-based composites. FIG. 2B provides the evolution of Tu ₊ with temperature for a hexagonal boron nitride ("h-BN") composite and for the LiTFSI/PP13 solution. FIG. 2C provides the Raman spectra of the BN system in the region of TFSI ^" full anion vibration. The spectra for LiTFSI solutions in room temperature ionic liquids ("RTILs") with different concentrations is also provided for comparison. The dots are experimental points, and the red line is the fit obtained when combining the two band components. FIG. 2D provides anodic (top) and cathodic (bottom) linear voltammetry scans at 120°C for the neat ionic liquid, the 1 mol L ^"1 LiTFSI/PP13 solution and the h-BN-based electrolyte. The legends apply for both plots.

[0010] FIGURE 3 shows the performance of lithium titanate ("LTO") half-cells containing h- BN composites as electrolyte. FIG. 3A provides cyclic voltammograms taken at 120°C with a scan rate of 0.1 mV s ^"1 , showing symmetric and sharp peaks. FIG. 3B provides cyclic stability for a cell operated for 32 days at a C/8 rate at 120°C, showing stable capacity of 158 mAh g ^"1 . FIG. 3C provides comparison between the charge/discharge profiles of a half-cell containing the h-BN composite (cycled at 120°C) and a half-cell containing 1 mol L ^"1 lithium hexafluorophosphate ("LiPF6") in ethylene carbonate ("EC") and dimethyl carbonate ("DMC") (EC:DMC 1: 1 (v/v), cycled at 23°C), obtained at a C/8 rate. The performance is very similar, with both systems showing low polarization. [0011] FIGURE 4 shows perspectives for h-BN composite electrolytes. FIG. 4A provides cyclic voltammetry and FIG. 4B provides cyclic stability of a LTO half-cell containing the h-BN composite tested at 150°C. Although there are changes in the electrode reaction kinetics over time, the electrolyte is electrochemically stable and the capacity fade is negligible.

[0012] FIGURE 5 shows compatibility of the h-BN electrolyte with cathode materials. FIG. 5A provides cyclic voltammogram of a Lii _+x Mn ₂ 0 ₄ (LMO) half-cell tested at 120°C. FIG. 5B provides the charge-discharge profile of the half-cell. The results show that, at high temperatures, the h-BN electrolyte allows performance similar to conventional electrolytes tested at room temperature.

[0013] FIGURE 6 shows modeling of the ionic conductivities of the electrolytes. FIG. 6A provides a Vogel-Tammann-Fulcher ("VTF") plot for the 1 mol L ^"1 solution of LiTFSI in PP13. FIG. 6B provides a VTF plot for the h-BN composite. FIG. 6C provides a VTF plot for the clay-based composite.

[0014] FIGURE 7 shows additional transport properties for the h-BN electrolytes. FIGS. 7A and 7B provide an example of results obtained for determining T _L i ₊ for the h-BN composite at 23 °C and 60°C, respectively, showing reduced resistance of the passivation film and an accelerated achievement of the steady state at higher temperatures. FIG. 7C provides typical impedance behavior of symmetric Li/electrolyte/Li cells obtained for the electrolytes (mostly 1 mol L ^"1 LiTFSI/PP13) at 100°C and 120°C, before and after the potentiostatic polarization employed to measure Tu ₊ . The Nyquist plot was purposely represented with asymmetric axis for clarity. FIGS. 7D-E provide dependence of ionic conductivity and Tu ₊ , respectively, on LiTFSI concentration in PP13 solutions.

[0015] FIGURE 8 shows scanning electron microscopy ("SEM") images. FIG. 8A provides an SEM image of pure BN. FIG. 8B provides an SEM image of fresh BN-RTIL electrolytes. FIG. 8C provides an SEM image of BN-based electrolyte after 30 cycles in a cathode half-cell. A 10 nm layer of gold was sputtered on top of the samples to minimize charging. [0016] FIGURE 9 shows additional electrochemical data for the electrolytes. FIG. 9A provides linear sweep voltammetry showing the electrochemical stability of the electrolytes at room temperature. The anodic scan is on the top, and the cathodic scan is on the bottom. FIGS. 9B-C provide cyclic voltammograms of the BN electrolyte in a 3-electrode setup (Ni as working electrode and Li metal as both reference and counter) showing a reversible plating/stripping behavior at room temperature and at 120°C, respectively.

[0017] FIGURE 10 shows additional cyclic stability data at 120°C for LTO half-cells prepared using the h-BN composite (FIGS. lOA-C). The respective C-rate is indicated in the plot. FIG. 10D provides the average coulombic efficiency as a function of the current density for LTO half- cells cycled at 120°C. The values are representative of all cells tested with the same current density. FIG. 10E provides rate capability studies at 120°C. FIG. 10F provides cycling data for the same cells at different temperatures at a C/3 rate.

[0018] FIGURE 11 shows electrochemical behavior of LTO half-cells operating at 24°C. FIG. 11A provides cyclic voltammograms obtained at a 0.1 mVs scan rate. FIG. 11B provides comparison of charge/discharge profiles between cells cycled at a C/8 rate at room temperature and 120°C. The elevated ionic conductivity and accelerated reaction kinetics at higher temperatures leads to a lower polarization and to large specific capacities. FIG. 11C provides a Nyquist plot from impedance measurements of an uncycled cell. FIG. 11D provides cycling behavior of the cell at a C/8 rate. The scattering in the first few points are due to an initially noisy cycling, sometimes observed at room temperature.

[0019] FIGURE 12 provides data relating to the characterization of h-BN-based composite electrolytes aged at 120°C before being used in a half-cell. FIG. 12A is a photograph of the electrolyte paste, showing no visible changes even after 20 days of exposure to high temperature. Also shown are Nyquist plots obtained at 120°C for uncycled half-cells assembled using electrolytes aged for 10 days (FIG. 12B) and 20 days (FIG. 12C). The insets, with units in kohm, show the high frequency region. FIG. 12D shows voltammograms for the first five cycles of the half-cell prepared using the h-BN composite aged for 20 days.

[0020] FIGURE 13 shows electrode aging experiments. FIG. 13A provides delithiation capacities at 25 °C for LTO half-cells assembled using fresh and aged electrodes, with 1 mol L ^"1 LiPF ₆ in EC:DMC 1: 1 (v/v) as electrolyte. FIGS. 13B-D provide electrochemical impedance spectra for cells prepared using a fresh electrode and electrodes aged at 120°C for 10 days and 20 days, respectively. Insets show an amplified view of the high frequency region. All spectra were collected in the delithiated state.

[0021] FIGURE 14 shows impedance spectra at 120 °C for LTO half-cells containing the h-BN composites at different cycling stages at high temperatures. Shown are fresh cells (FIG. 14A), cells after 5 cycles (FIG. 14B), and cells after 50 cycles (FIG. 14C). A C/8 rate was employed in the experiments. All spectra were collected in the delithiated state.

[0022] FIGURE 15 shows self-discharge (SD) measurements for a LTO half-cell at 120 °C. FIG. 15A shows charge-discharge profile for the sample allowed to self-discharge for 24 hours. The galvanostatic portion (C/8 rate) is shown as blue, while the SD curve is colored red. FIG. 15B shows the delithiation capacity for each cycle. The 4 ^th cycle, pointed by the red arrow, is the capacity retrieved after 24 hours of SD.

[0023] FIGURE 16 shows electrochemical characterization for the LMO-LTO full-cell. FIG. 16A provides cyclic voltammetry for the full-cell using conventional organic electrolytes at room temperature. FIGS. 16B-D provide cyclic voltammetry, charge discharge profile, and cyclic stability, respectively, at 120°C using h-BN composite electrolytes.

DETAILED DESCRIPTION

[0024] It is to be understood that both the foregoing general description and the following detailed description are illustrative and explanatory, and are not restrictive of the subject matter, as claimed. In this application, the use of the singular includes the plural, the word "a" or "an" means "at least one", and the use of "or" means "and/or", unless specifically stated otherwise. Furthermore, the use of the term "including", as well as other forms, such as "includes" and "included", is not limiting. Also, terms such as "element" or "component" encompass both elements or components comprising one unit and elements or components that comprise more than one unit unless specifically stated otherwise.

[0025] The section headings used herein are for organizational purposes and are not to be construed as limiting the subject matter described. All documents, or portions of documents, cited in this application, including, but not limited to, patents, patent applications, articles, books, and treatises, are hereby expressly incorporated herein by reference in their entirety for any purpose.

[0026] Many energy storage devices have performance limitations, including limited stability at high temperatures. For instance, lithium- ion batteries ("LIBs") present a combination of extended cycle life and high energy density, which make them the standard power source for portable devices. However, LIBs have limited performance even at mildly elevated temperatures, such as temperatures ranging from 60 °C to 80 °C.

[0027] Problems with energy storage device performance at high temperatures likely arise due to the failure of separators and electrolytes. In particular, separators are usually composed of a thin polymeric film, which can soften or shrink upon heating, thereby resulting in electrical short circuit.

[0028] Likewise, commercial electrolytes are based on organic solvents with low boiling points. Therefore, an increase in the temperature of an energy storage device can elevate the internal pressure of a cell, thereby potentially damaging the electrodes and the packaging.

[0029] The aforementioned limitations with separators and electrolytes present significant safety concerns. For instance, the aforementioned limitations can provoke battery explosions. [0030] Moreover, energy storage device operation at high rates may lead to thermal dissipations that increase the internal temperature of a cell, thereby facilitating runaway and failure. As such, the development of temperature-friendly technologies is desirable to meet the safety expectations for energy storage devices.

[0031] Efforts have been made to develop safe and reliable energy storage devices capable of working at temperatures over 100 °C. For instance, lithium-containing solutions in ionic liquids as the electrolyte have been utilized, particularly due to the high thermal stability and negligible vapor pressure of the composition.

[0032] Likewise, to address issues associated with the mechanical stability of separators, polymer electrolytes have been utilized. However, polymer electrolytes lack mechanical robustness for prolonged exposure to temperatures higher than 100 °C. As such, a need exists for the development of more stable energy storage device components.

[0033] Compositions

[0034] In some embodiments, the present disclosure pertains to compositions that include a boron nitride, an ionic liquid, and a lithium salt. The compositions of the present disclosure can have various types of boron nitrides, ionic liquids and lithium salts at various ratios.

[0035] In some embodiments, the boron nitrides in the compositions of the present disclosure include a boron nitride in a hexagonal phase, such as hexagonal boron nitride. In some embodiments, the hexagonal boron nitrides in the compositions of the present disclosure include, without limitation, particulate hexagonal boron nitrides, exfoliated hexagonal boron nitrides (e.g., exfoliated hexagonal boron nitrides where the layers of the structure have been separated into individual flakes), and combinations thereof.

[0036] The compositions of the present disclosure can also have various ionic liquids. For instance, in some embodiments, the ionic liquids in the compositions of the present disclosure can include, without limitation, imidazolium-based ionic liquids, piperidinium-based ionic liquids, phosphonium-based ionic liquids, ammonium-based ionic liquids and combinations thereof. In some embodiments, the ionic liquid is a piperidinium-based ionic liquid, such as 1- methyl- 1-propylpiperidinium bis(trifluoromethylsulfonyl)imide.

[0037] The compositions of the present disclosure can also have various lithium salts. For instance, in some embodiments, the lithium salts in the compositions of the present disclosure can include, without limitation, lithium hexafluorophosphate ("LiPFc" or "LiPF ₆ "), lithium bis(trifluoromethylsulfonyl)imide ("LiTFSI"), lithium bis(fluorosulfonyl)imide ("LiFSI"), lithium bis(oxalato)borate ("LiBOB"), and combinations thereof. In some embodiments, the lithium salt is lithium bis(trifluoromethylsulfonyl)imide.

[0038] In some embodiments, the compositions of the present disclosure include a mixture of hexagonal boron nitride, a piperidinium-based ionic liquid, and a lithium salt. The compositions of the present disclosure can also be in various forms. For instance, in some embodiments, the compositions of the present disclosure are in the form of a composite material. In some embodiments, the compositions of the present disclosure have a consistency of a paste. In some embodiments, the compositions of the present disclosure are in the form of freestanding membranes, such as freestanding films.

[0039] In some embodiments, the compositions of the present disclosure also include a polymer binder. In some embodiments, the polymer binder maintains the compositions of the present disclosure in the form of a free standing membrane. In some embodiments, the polymer binder includes one or more thermoplastic polymers. In some embodiments, the thermoplastic polymers include, without limitation, poly(methyl methacrylate), polymethacrylic acid, polyvinylidene fluoride, acrylonitrile butadiene styrene, polylactic acid, polybenzimidazole, polycarbonate, polyether sulfone, poly ether ether ketone, polyetherimide, polyethylene, polypropylene, polystyrene, polyvinyl chloride, polyimide, polyacrylonitrile, polyurethane, and combinations thereof. In some embodiments, the polymer binder includes one or more thermosetting polymers. In some embodiments, the thermosetting polymers include, without limitation, polyimide, polyurethane, epoxy, polyester, and combinations thereof.

[0040] The compositions of the present disclosure can have various advantageous properties. In some embodiments, the compositions of the present disclosure are thermally stable and nonvolatile. In some embodiments, the compositions of the present disclosure are operable and safe at high temperatures. In some embodiments, the compositions of the present disclosure have ionic conductivities that range from about 0.1 mS/cm to about 5 mS/cm. In some embodiments, the compositions of the present disclosure have ionic conductivities of about 3 mS/cm.

[0041] In some embodiments, the compositions of the present disclosure have electrochemical stability windows that range from about 1 V to about 6 V. In some embodiments, the compositions of the present disclosure have electrochemical stabilities of about 5 V at 120 °C.

[0042] In some embodiments, the compositions of the present disclosure can demonstrate extended thermal stability. For instance, when used in combination with conventional electrodes, the compositions of the present disclosure can demonstrate thermal stability at temperatures above 100 °C for over 100 cycles with a total capacity fade of less than 5%. In some embodiments, the compositions of the present disclosure can demonstrate thermal and electrochemical stability for over 600 cycles at 120 °C, with a total capacity fade of less than 3%. In some embodiments, the compositions of the present disclosure can demonstrate thermal and electrochemical stability for over 50 cycles at 150 °C, with a total capacity fade of less than 2%.

[0043] In some embodiments, the compositions of the present disclosure can demonstrate thermal and electrochemical stability for cycling periods of over 32 days. In some embodiments, the compositions of the present disclosure can demonstrate a self-discharge of less than 3% of capacity while stored at 120 °C for 24 hours in the fully charged state. In some embodiments, the compositions of the present disclosure can demonstrate thermal and electrochemical stability for storage times over 20 days at 120 °C. [0044] Methods of Making Compositions

[0045] In some embodiments, the present disclosure pertains to methods of making the compositions of the present disclosure. In some embodiments, the methods of the present disclosure involve contacting a boron nitride, an ionic liquid and a lithium salt to form the composition.

[0046] The contacting of the different components of the compositions of the present disclosure can occur by various methods. For instance, in some embodiments, the contacting occurs by heating boron nitrides, ionic liquids, and lithium salts. In some embodiments, the contacting occurs by incubating boron nitrides, ionic liquids, and lithium salts. In some embodiments, the contacting occurs by mixing boron nitrides, ionic liquids, and lithium salts. In some embodiments, the mixing occurs by various methods, such as stirring, sonication, physical agitation, and combinations of such steps.

[0047] The components of the compositions of the present disclosure can be in various states during the contacting step. For instance, in some embodiments, the boron nitride is in solid form, and the ionic liquid and the lithium salt are in a solution.

[0048] The components of the compositions of the present disclosure can be in various ratios during the contacting step. For instance, in some embodiments where the boron nitride is in solid form and the ionic liquid and the lithium salt are in a solution, the contacting occurs at a 1:2 boron nitride to solution wt./wt. ratio. In some embodiments where the boron nitride is in solid form and the ionic liquid and the lithium salt are in a solution, the contacting occurs at a 1:2.5 boron nitride to solution wt./wt. ratio. In some embodiments where the boron nitride is in solid form and the ionic liquid and the lithium salt are in a solution, the contacting occurs at a 1:3 boron nitride to solution wt./wt. ratio. In some embodiments, equimolar ratios of boron nitrides, ionic liquids, and lithium salts are utilized during the contacting step. [0049] In some embodiments, the contacting of lithium salts and ionic liquids occurs in a solvent, such as an organic solvent. In some embodiments, the contacting results in the dissolution of the lithium salts with the ionic liquids in the solvent to form a solution. Thereafter, the boron nitride is combined with the formed solution by various methods described previously, such as by physical agitation.

[0050] In some embodiments, the properties of the formed compositions may be tailored by adjusting the amounts of boron nitrides, ionic liquids and lithium salts that are mixed together. In some embodiments, the choice of the ionic liquids and the lithium salts can define the electrochemical properties of the formed composition while the choice of boron nitride can provide mechanical stability to the composition.

[0051] In some embodiments, the methods of the present disclosure also include a step of incorporating the formed composition into an energy storage device. In some embodiments, the formed composition is utilized as an electrolyte, a separator, or an electrolyte- separator combination in the energy storage device.

[0052] In some embodiments illustrated in FIG. 1A, the compositions of the present disclosure are formed by mixing a lithium salt with an ionic liquid (step 10). Thereafter, appropriate amounts of a boron nitride (e.g., hexagonal boron nitride) are mixed with the ionic liquid/lithium salt solution (step 12), thereby forming a composite (step 14). The formed composite is then incorporated into an energy storage device (step 16).

[0053] In some embodiments, the mixing of the ionic liquid/lithium salt solution with the boron nitride occurs by adding the boron nitride to the ionic liquid/lithium salt solution. In some embodiments, the mixing of the ionic liquid/lithium salt solution with the boron nitride occurs by adding the ionic liquid/lithium salt solution to the boron nitride.

[0054] In some embodiments illustrated in FIG. IB, the compositions of the present disclosure are formed by mixing lithium bis(trifluoromethylsulfonyl)imide ("LiTFSI") with 1 -methyl- 1- propylpiperidinium bis(trifluoromethylsulfonyl)imide (a room temperature ionic liquid labeled as "RTIL") to form a solution (step 20). Thereafter, appropriate amounts of a hexagonal boron nitride are mixed with the solution (step 22) to form a composition that resembles a conductive paste.

[0055] Energy Storage Devices

[0056] In some embodiments, the present disclosure pertains to energy storage devices that contain the compositions of the present disclosure. In some embodiments, the energy storage devices include, without limitation, capacitors (e.g., electrochemical double-layer capacitors), super capacitors, micro supercapacitors, pseudo capacitors, batteries, photovoltaic devices, photovoltaic cells, and combinations thereof.

[0057] In some embodiments, the energy storage devices of the present disclosure include a battery. In some embodiments, the battery includes, without limitation, micro batteries, lithium- ion batteries, lithium- sulfur batteries, sodium-ion batteries, magnesium-ion batteries, aluminum- ion batteries, and combinations thereof. In some embodiments, the battery is a lithium- ion battery.

[0058] The compositions of the present disclosure may be utilized as various components of an energy storage device. For instance, in some embodiments, the compositions of the present disclosure are utilized as an electrolyte in the energy storage device. In some embodiments, the compositions of the present disclosure are utilized as a separator in the energy storage device. In some embodiments, the compositions of the present disclosure are utilized as an electrolyte and a separator in the energy storage device (i.e., an electrolyte-separator combination).

[0059] An example of an energy storage device that contains a composition of the present disclosure is shown in FIG. 1C. In FIG. 1C, the energy storage device is lithium- ion battery 30. Lithium-ion battery 30 contains anode 32, electrolyte component 34, and cathode 36. Electrolyte component 34 in this example contains a composition of the present disclosure. In some embodiments, electrolyte component 34 can be an electrolyte composite, a separator, or an electrolyte- separator combination.

[0060] In some embodiments, the present disclosure pertains to methods of making the energy storage devices of the present disclosure. In some embodiments, the methods include incorporating the compositions of the present disclosure as components of the energy storage device. For instance, in some embodiments, energy storage devices are fabricated by combining the compositions of the present disclosure with electrodes (e.g., combining electrolyte component 34 with anode 32 and cathode 36, as illustrated in FIG. 1C).

[0061] The energy storage devices of the present disclosure can have various advantageous properties. For instance, in some embodiments, the energy storage devices of the present disclosure are capable of being operated at temperatures that range from about 20 °C to about 200 °C. In some embodiments, the energy storage devices of the present disclosure are capable of being operated at temperatures that range from about 20 °C to about 150 °C. In some embodiments, the energy storage devices of the present disclosure are capable of being operated at temperatures that range from about 20 °C to about 120 °C. In some embodiments, the energy storage devices of the present disclosure are capable of being operated at temperatures of more than about 80 °C. In some embodiments, the energy storage devices of the present disclosure are capable of being operated at temperatures of more than about 100 °C. In some embodiments, the energy storage devices of the present disclosure are capable of being operated at temperatures of more than about 120 °C. In some embodiments, the energy storage devices of the present disclosure are capable of being operated at temperatures of more than about 150 °C.

[0062] The energy storage devices of the present disclosure can also be operational at temperatures that are lower than the aforementioned operational temperatures. For instance, in some embodiments where an energy storage device is capable of being operated at temperatures of more than about 100 °C, the energy storage device can also be operational at temperatures that are below 100 °C. [0063] In some embodiments, the energy storage devices of the present disclosure have a theoretical capacity of more than about 100 mAh/g at 120 °C. In some embodiments, the energy storage devices of the present disclosure have a theoretical capacity of more than about 160 mAh/g at 120 °C. In some embodiments, the energy storage devices of the present disclosure have a theoretical capacity of more than about 175 mAh/g at 120 °C.

[0064] In some embodiments, the energy storage devices of the present disclosure have a stable capacity retention over 100 cycles. For instance, in some embodiments, the energy storage devices of the present disclosure have a stable capacity retention over 600 cycles of recharging at high current rates of 3C. In some embodiments, the energy storage devices of the present disclosure have a stable capacity retention over 100 cycles of recharging at high current rates of 1C. In some embodiments, the energy storage devices of the present disclosure have a stable capacity retention over 100 cycles of recharging at current rates of C/3. In some embodiments, the energy storage devices of the present disclosure have a stable capacity retention over 100 cycles of recharging at current rates of C/8.

[0065] Reference will now be made to more specific embodiments of the present disclosure and experimental results that provide support for such embodiments. However, Applicants note that the disclosure below is for illustrative purposes only and is not intended to limit the scope of the claimed subject matter in any way.

[0066] Example 1. Hexagonal Boron Nitride-Based Electrolyte Composite for Li-Ion Battery Operation from Room Temperature to 150 °C

[0067] In this Example, Applicants report a composite electrolyte/separator that can extend the capability of Li-ion batteries to high temperatures. A stoichiometric mixture of hexagonal boron nitride, piperidinium-based ionic liquid and a lithium salt has been formulated, with ionic conductivity reaching 3 mS/cm, electrochemical stability higher than 5 V vs Li/Li ⁺ and extended thermal stability. The composite was used in combination with conventional electrodes and demonstrated to be stable for over 600 cycles at 120 °C, with a total capacity fade of less than 3%.

[0068] Electrolytes were prepared by mixing bentonite clay or hexagonal boron nitride ("h-BN") with an appropriate weight of a 1 mol L ^"1 solution of lithium bis(trifluoromethylsulfonyl)imide ("LiTFSI") in the ionic liquid 1 -methyl- 1 -prop ylpiperidinium bis(trifluorosulfonyl)imide ("PP13"). Although bentonite and boron nitride have a similar flake size of ~1 μιτι, h-BN was found to absorb twice as much room temperature ionic liquid ("RTIL") solution as clay before it reached the same paste-like consistence. Without being bound by theory, such observations are likely due to a higher specific volume of the h-BN powder.

[0069] The ionic conductivities of RTIL solution, clay-RTIL and h-BN-RTIL electrolytes were recorded from room temperature to 150 °C, as shown in FIG. 2A. The addition of ceramic particles was found to reduce ion mobility, with a more marked effect at lower temperatures. At 24 °C, the conductivity for the h-BN composite is -0.2 mS cm ^"1 , above the benchmark value of 0.1 mS cm ^"1 for battery electrolytes. The clay-RTIL system lies below that value.

[0070] The ionic conductivity dependence on temperature followed a Vogel-Tammann-Fulcher ("VTF") behavior (FIG. 6). The fitted parameters are presented in Table 1.

Sam le A / S cm ^"1 K ^"05 E _a / kJ mol ^"1 To / K R

BN + 1 mol L ^"1 0.77 3.95 216 0.99945

Clav mol I . 0.3 ii in O.W636 the coefficient of determination for the fitting Table 1. Fitting parameters obtained by applying the VTF model to the conductivity data for the studied electrolytes.

[0071] The pre-exponential factor A is proportional to the number of charge carriers in the electrolyte. Although factor A evidently dropped with addition of ceramic components, Applicants note that the ratio between A _h -BN and A _c i _ay is ~2. Assuming that both composites have comparable densities, this factor is larger than expected, likely due to the presence of -30% more ionic liquid solution in the h-BN system. The results suggest that the actual nature of ceramic within the electrolyte impacts the ion distribution and the effective free volume.

[0072] Applicants also observed that the presence of ceramic particles resulted in a decrease in the pseudo-activation energy, possibly due to a local partial screening of coulombic interactions in the ionic liquid, with a more significant effect observed with the presence of the charged platelets of clay. The ideal glass transition temperature is higher in the composites than in the RTIL solution, indicating stronger solute- solvent interactions.

[0073] Although the plots in FIG. 2A depict the overall ionic motion in the investigated electrolytes, it does not represent the Li ⁺ conductivity, which is a metric for LIB operation. The ionic nature of RTILs suggest the existence of several charged species other than Li ⁺ migrating in the cell, eventually accumulating in the electrode vicinity, building up overpotential and thereby compromising device operation, especially at high rates.

[0074] The lithium ion transference number ("Τυ ₊ ") was measured following an ac/dc polarization method (Evans et al., Polymer, 1987, 28, 2324). The technique is based on the assumption that, in a symmetric cell containing electrodes which are non-blocking only towards lithium ions, the application of a constant bias over a long time leads to a steady state, in which the overall current flow corresponds to the flux of lithium ions only. A typical polarization measurement is shown in FIG. 7A, with the impedance plots before and after applying the bias shown as inset. The resistance of the passivation layer formed on top of the lithium metal electrodes were in the range of hundreds of ohms at room temperature, quickly dropping upon heating the cell (FIG. 7B).

[0075] The T _Li+ at 23 °C was found to be 0.093+0.029, 0.076+0.009 and 0.060+ 0.001 for the 1 mol L ^"1 solution of LiTFSI in PP13, the h-BN composite, and the clay composite, respectively. The values were in the same range of other reports for ionic liquids in the literature.

[0076] Without being bound by theory, changes in the transference number may occur as a consequence of adding extra ions to the electrolyte or by uneven changes in the diffusion constants of charge carriers. Hexagonal boron nitride is expected to weakly interact with ionic liquids, even though it is not clear how its presence can affect Tu ₊ .

[0077] Additionally, the negatively charged clay platelets may influence the ion mobility in the electrolyte. Ceramic fillers at small concentrations have been widely reported to improve transport properties in polymer electrolytes by promoting salt dissociation and avoiding chain recrystallization. Nevertheless, at higher concentrations, the particles tend to aggregate, reducing the effective contact area with the electrolyte and negatively affecting charge transport.

[0078] Since the clay composite has a high solid content, the actual role of the ceramic particles might be reduced due to phase segregation. Another factor to consider is that ion exchange might be happening at a small extent between the liquid phase and the clay, thus depleting the concentration of Li ⁺ in the electrolyte and reducing the transference number. Due to the improved ionic conductivity and lithium ion mobility, the h-BN-based composite was further characterized for use in LIBs operating at a wide temperature range.

[0079] The h-BN composite has a uniform distribution of boron nitride particles and ionic liquid, as observed using scanning electron microscopy in FIG. 8. The specific interactions between the two phases are very weak and no major changes were observed in the Raman spectrum of the composite, as compared to the 1M solution of LiTFSI in ionic liquid (full data not shown here). [0080] However, the solvation of lithium ions by TFSI ^" can be monitored by analyzing the band at -745 cm ^"1 assigned to the full anion vibration, as shown in FIG. 2C. This band can be resolved into two components, originated from the vibration of free anion (-742 cm ^"1 ) and Li ⁺ - bound TFSI ^" (748 cm ^"1 ). As shown in the plots, addition of lithium salt increases the fraction of anions involved in strong solvation, as indicated by the enhanced intensity of the high frequency component. The analysis of the spectra also show that addition of boron nitride increases the relative area of the bound TFSI ^" component by -5%, showing that Li+ is actually more intensely solvated by the anions in the presence of the ceramic.

[0081] Since the h-BN does not show strong interactions with the other components, this effect might be a consequence of a confinement of the ionic liquid solution by the h-BN particles, reducing the average separation between the ions in the system. This is also in agreement with the higher T ₀ values found after fitting the ionic conductivity data using the VTF model (FIG. 6).

[0082] Temperature effects on lithium ion transference numbers were also measured for both the RTIL solution and the h-BN electrolyte (FIG. 2B). The values present only a slight increase above 80 °C, indicating that the Li ⁺ solvation remains nearly unchanged within the temperature range, and that the temperature dependence of the diffusion coefficients of all species in the liquid phase is about the same. At higher temperatures, the resistance of the passivation layer formed on top of the lithium electrodes in the symmetric cells drops drastically, and the increased ionic mobility favors the achievement of a steady state much more quickly than at room temperature (FIG. 7B).

[0083] The boron nitride-RTIL composite presented adequate conductivity and stable Li ⁺ transport properties in the range of 23 °C to at least 120 °C. However, the successful application in LIBs requires compatibility with an appropriate potential range, without the presence of unwanted reactions.

[0084] The electrochemical stability of the electrolyte was evaluated through 3 -electrode measurements, using lithium metal as both the counter and reference electrodes, with a stainless steel working electrode. Linear voltammetry scans taken at 120 °C are shown in FIG. 2D for the RTIL, the 1M solution of LiTFSI in PP13, and for the h-BN composite (data at 23 °C is provided in FIG. 9A).

[0085] FIG. 2D demonstrates that the anodic stability of the RTIL at high temperature increases after addition of LiTFSI, with an even larger enhancement attained when h-BN is mixed with the RTIL-based solution. The electrochemical window was found to be broader than 4.7 V (from 5 V down to < 0.3 V vs Li/Li ⁺ ) at 120 °C, thus enabling its use with high energy density cathodes and anodes. Since the anodic stability might be strongly related to the reactivity of TFST, the intensification of Li-TFSI interactions with addition of h-BN, as observed by Raman spectroscopy, might explain the observation of a wider electrochemical window. The electrolyte was also found to be compatible with lithium metal electrodes, showing a reversible plating/stripping behavior both at room and high temperatures (FIGS. 9B-C).

[0086] Although there are several reports on effects of cell overheating and abuse of lithium secondary batteries, little is known about actual devices operating at temperatures above 80 °C. Thus, a need exists to understand thermal effects on fundamental cell processes.

[0087] High energy density anode materials generally rely on the formation of a stable passivation layer to achieve stable performance. Moreover, it is rather unpredictable how the complex and heterogeneous structure of a passivation layer would behave if exposed to such conditions.

[0088] In order to simplify the system on a device level, cell testing was performed using half- cells based on Lithium Titanate (Li ₄ TisOi ₂ , "LTO"), which presents a high lithiation potential of -1.5 V and the absence of a thick, permanent SEI layer. LTO half-cells containing the h-BN- based electrolyte were tested at 120 °C. The electrochemical performance is shown in FIGS. 3 and 10. [0089] The cyclic voltammograms (FIG. 3A) show sharp and symmetric peaks corresponding to the lithium insertion and extraction in the electrodes, with the complete absence of additional reactions. Cells were tested by constant-current charge-discharge at different C-rates (FIGS. 3B and lOA-C), showing unprecedented cyclic stability, without significant capacity fade, even after extended testing times.

[0090] As an example, FIG. 3B shows the cyclic stability for a cell cycled at a -C/8 rate, where the capacity is nearly constant at 158 mAh g ^"1 for more than 50 cycles and a total test time of over a month. The lithiation kinetics was found to be so favored at high temperatures that changing the C-rate had only minor effects on the delivered capacity (FIG. 10E). The absence of thermal runway on the cells and the non-flammability and negligible vapor pressure of the electrolytes show that the h-BN composite allows reliable and safe operation for Li-ion batteries.

[0091] In contrast to previous reports in the literature of Li-ion cells operating within this same temperature, the half-cells prepared using this composite electrolyte delivered exceptional coulombic efficiency values at 120 °C for a broad range of current densities, even after several charge-discharge cycles, as presented in FIG. 10D. Efficiencies over 98% were observed, showing the high reversibility of the electrode reactions at such extreme conditions and posing a good indication of the stability of the electrolyte under a long term exposure to harsh environments while involved in electrochemical processes. The high capacity values and low polarization observed for the cells at 120 °C were actually comparable to the performance of devices prepared with a conventional organic electrolyte and operated at room temperature (FIG. 3C).

[0092] Since the electrolyte presented reasonable ionic conductivity and T _L i ⁺ at 24 °C, it was also operational at room temperature. The sluggish transport properties at reduced temperatures accounted for a larger overpotential (FIGS. 11A-B) and high total resistance (FIG. 11C). However, stable capacities up to -90 mAh g ^"1 were achieved, over 50% of that at 120 °C. Moreover, full capacity could be achieved at temperatures as low as 60 °C, as shown in FIG. 10F.

[0093] A big challenge in successfully operating an electrochemical device at 120 °C is to assure that all components are stable over long periods under such conditions. Although most of the literature discusses the thermal stability of ionic liquids based on thermogravimetric analysis ("TGA"), this might not be a realistic figure of merit for applications that require long term exposition to high temperatures. Chemical transformations in the ionic liquid at lower temperatures may form non- volatile electrochemically active species that can strongly affect cell performance. For instance, it has been found that alkylimidazoles were being formed after heating certain imidazolium-based RTILs for 24 hours at temperatures above 120 °C.

[0094] To probe the long term stability of the electrolyte, aging tests were conducted by exposing a sealed sample of the h-BN composite to high temperature for different periods and then using it to prepare LTO half-cells. Exposure of the electrolyte paste to 120 °C for a period of 20 days did not lead to any visible changes (FIG. 12A). High temperature electrochemical performance of the cells prepared using such aged electrolyte was similar to the one presented by half-cells containing fresh electrolyte. Impedance spectra for cells using the composite aged for 10 and 20 days are presented in FIGS. 12B-C, respectively. The only changes are a small increase in the ESR and the existence of a semicircle at high frequency.

[0095] Small changes in the electrolyte conductivity can happen upon electrolyte aging, while the development of a semicircle can be related to a small increase in the charge transfer resistance or due to the passivation of the lithium metal electrode. From the symmetric cells used to determine Li ⁺ transference number, it is known that any passivation film would have a very low resistance at 120 °C (typically < 1 Ω). Therefore, charge transfer may account for most of the change.

[0096] Additionally, the cells were tested by cyclic voltammetry. The observed profiles for aged samples were found to closely resemble the one for cells using fresh composite (compare FIG. 12D to FIG. 3A). This further supports the optimal stability of the electrolyte towards high temperature environments and limited effects of aging in the overall performance. Scanning electron microscopy images (FIGS. 8A-C) also support that there are no major changes in the state of boron nitride particles in the electrolyte, even after extensive cycling.

[0097] Although the electrolyte presents noticeable performance at 120 °C, high temperatures could possibly allow thermally-activated processes to happen, triggering additional self- discharge mechanisms. For instance, it has been reported that some RTILs are incompatible with electrodes in the charged, high energy state, presenting an onset for thermal runway at temperatures lower than for conventional EC:DEC -based electrolytes. An alternative way of probing the stability of the h-BN composite electrolyte toward the charged electrode at high temperatures is through the analysis of self-discharge behavior and further stability of the cell upon recharge.

[0098] The profiles of the galvanostatic charge/discharge cycles before and after the self- discharge, along with the self-discharge curve itself, are shown in FIG. 15A. During the self- discharge (red portion of the plot), the cell potential quickly rises to a value slightly lower than the plateau voltage for LTO. The capacity retrieved from the cell after 24 hours of open circuit at 120 °C corresponds to 97% of the cell capacity. Furthermore, no permanent capacity loss was observed after the self-discharge (FIG. 15B).

[0099] As a comparison, larger capacity losses than the one reported here were observed for other conventional systems, even at 60 °C for the same period. The high extent of capacity retention and the absence of irreversible losses after self-discharge provide an independent evidence of the electrolyte stability in harsh environments, even in the presence of a charged electrode.

[00100] In additional experiments, Applicants increased the temperature of operation of the half- cell to 150 °C. FIG. 4A shows a cyclic voltammetry at 150 °C for a LTO half-cell, showing that the electrochemical stability of the electrolyte holds up at higher temperatures and that the mechanical properties of the h-BN system is still large enough to avoid short-circuits. Although the peaks display some extent of broadening after the first few cycles, the cell is able to deliver a stable capacity with a high coulombic efficiency of 97% (FIG. 4B). A change in the shape of the peaks was also observed in CV measurements with slower scan rates and appeared to be time-dependent. Without being bound by theory, such patterns may be due to changes in the electrode that affect the overall kinetics of the device, possibly involving redistribution of the polyvinylidene fluoride ("PVDF") binder. To the best of Applicants' knowledge, the results in FIG. 4 represent the highest temperature of operation ever reported for a Li-ion battery that can also perform at room temperature.

[00101] The electrochemical properties of extra-lithiated manganese oxide (Lii _+x Mn ₂ _ _x 0 ₄ , "LMO") as electrode were also studied. As demonstrated from the charge-discharge profile and cyclic voltammograms for a LMO half-cell at 120 °C (FIGS. 5A-B, respectively), this material presents two lithiation plateaus, with onsets at ~3 V and ~4 V, with a total practical delithiation capacity of 146 mAh g ^"1 . The high extent of reversibility of the electrode reactions observed in the CV plot is a good indication that LMO is suitable for high temperature applications and is proof that the composite electrolyte can be utilized as cathode components.

[00102] Next, Applicants made a preliminary attempt to test the operability of the electrolyte in a full-cell configuration. Using LTO and Lii _+x Mn ₂ _ _x 0 ₄ as electrodes, cells were tested with the h- BN composite electrolyte and also with conventional organic electrolytes. The cell configuration was balanced by considering only the high voltage plateau of the cathode. Cyclic voltammetry tests performed at 120 °C for the h-BN resulted in comparable profile as that of the organic electrolyte at room temperature (FIGS. 16A-B). The lack of spurious peaks at high temperatures indicates the compatibility of the electrolyte with the potential range of a full-cell operation.

[00103] The cell with h-BN electrolyte tested at 120 °C delivered a stable voltage profile with distinctive plateaus for both anodic and cathodic potentials, resulting in a reasonable cyclic stability (FIGS. 16C-D). These initial results build a solid starting point for developing stable battery systems with high temperature stability.

[00104] In summary, Applicants present an electrolyte composite based on hexagonal boron nitride and ionic liquid that allows Li-ion battery operation with high efficiency over an unprecedented temperature range, spanning at least from 24 °C to 150 °C. The presence of a ceramic component in the electrolyte had little overall effect on the transport properties, although contributing to an increase of the electrochemical window. The composite was also shown to be compatible towards a cathode material, enabling full-cell testing at high temperatures.

[00105] Example 1.1. Electrode preparation, electrolyte composition and cell assembly

[00106] Electrodes were fabricated by coating slurries containing 80% active material, 10% graphite and 10% poly(vinilydene difluoride) binder onto current collectors. The h-BN composite was prepared by mixing boron nitride powder and a 1 mol L ^"1 solution of LiTFSI in the RTIL 1 -methyl- 1 -prop ylpiperidinium bis(trifluorosulfonyl)imide, in a 1:2 wt./wt. ratio. More detailed protocols are provided herein.

[00107] The electrodes were prepared by grinding the active materials for the anode (Lithium Titanate, "LTO", < 200 nm particle size, spinel, Sigma- Aldrich) or the cathode (extra-lithiated manganese oxide, Lii _+x Mn ₂ - _x 0 ₄ , "LMO", synthesis described below) with ultra-fine graphite and poly(vinilydene difluoride) ("PVDF", Sigma- Aldrich), in a proportion of 80: 10: 10 and adding enough N-methyl-2- pyrrolidinone ("NMP", Sigma- Aldrich) to form a viscous slurry. The slurry was then cast onto copper current collectors (1.21 cm ), by either manually coating or spray coating. Both techniques provided comparable results. The electrodes were then dried for at least 24 hours at 85 °C under vacuum. Typical loadings were in the range of 2-5 mg for anodes and 5-7 mg for cathodes, depending on the technique employed for casting the slurry. Typical thicknesses are in the order of 30-40 μιη for both electrodes. [00108] The electrolyte was prepared by thoroughly mixing appropriate amounts of bulk boron nitride powder (~ 1 pm flakes, Sigma- Aldrich) or bentonite clay (~1 pm flakes, Southern Clay Products) and a 1 mol L ^"1 solution of bis(trifluoromethylsulfonyl)imide lithium salt ("LiTFSI") in the room temperature ionic liquid ("RTIL") 1 -methyl- 1 -prop ylpiperidinium bis(trifluoromethylsulfonyl)imide ("PP13"). The concentration of the RTIL solution was selected based on ionic conductivity and Li-ion transference number measurements (FIGS. 7D-E).

[00109] The paste-like mixture had a 1:2 BN- and 1: 1 clay-to-solution ratio. The typical total weight of the electrolyte was -150 mg per cell. After homogenization, the paste was manually cast onto the electrode until the whole area was covered, forming a layer with typically 250-300 μιη of thickness. Both coin cells and Swagelok-type cells were tested and provided similar results. Electrolyte preparation and cell assembly were performed in an Argon-filled glovebox with moisture and oxygen levels inferior to 0.1 ppm. Several results for cells operating at high temperature were compared to conventional cells tested at room temperature for validation, in which a 1 mol L ^"1 solution of LiPF6 in 1: 1 v/v ethylene carbonate/dimethyl carbonate was used as electrolyte.

[00110] Example 1.2. Electrochemical characterization

[00111] Galvanostatic Charge-Discharge ("CD") tests were performed on a BT-2000 battery cycler (Arbin Instruments), while Cyclic Voltammetry ("CV") and Electrochemical Impedance Spectroscopy ("EIS") were performed on an Autolab PGSTAT 302N potentiostat. CVs were taken at 0.1 mV s ^"1 and EIS was performed at open circuit potential, in the range 1 MHz - 100 Hz with an amplitude of 10 mV.

[00112] Raman measurements were conducted by sealing the samples in glass tubes in a glove box. The data was collected in a backscattering geometry using a Renishaw spectrometer with a 514-nm laser and a 5X objective. [00113] Ionic conductivities were measured by EIS, with the electrolyte placed between two stainless steel current collectors in a Swagelok cell without a spring, in a way to fix the cell constant at all temperatures. The electrolyte thickness was taken after the measurements were performed. For the RTIL solutions, conductivity measurements were taken using filter paper as separator.

[00114] All aging experiments were performed by sealing the cell component (electrode or electrolyte) in a coin cell case and storing it under the appropriate condition. The assembly and disassembly of cases were always performed in a glovebox. Lithium ion transference numbers (Τϋ+) were measured using symmetric Li/electrolyte/Li cells.

[00115] Example 1.3. Cathode active material synthesis

[00116] Spinel Lii _+x Mn ₂ _ _x 0 ₄ compound was prepared by a sol-gel method from a stoichiometric composition of lithium acetate (CH ₃ COOL1 2H ₂ 0, Sigma Aldrich) and manganese acetate (CH ₃ COO) ₂ Mn4H ₂ 0, Sigma Aldrich), with citric acid as a chelating agent. Primarily, the reactants were dissolved in Millipore water at 70 °C with an effective stirring. An optimized proportion of reactants to citric acid of 3:1 was employed to achieve homogenous solution. The process of heating and stirring persisted until a transparent gel was obtained, which was dried and kept in an oven at 300 °C for 12 hours. The resultant powder was ground and sintered at 850 °C for 12 hours under air atmosphere to attain desired compound.

[00117] Example 1.4. Ionic conductivity modeling

[00118] The experimental data was fitted using the following Vogel-Tammann-Fuchel equation (Equation 1). [00119] In Equation 1, o is the ionic conductivity, A is a pre-exponential factor, T is the absolute temperature, T ₀ is the ideal glass transition temperature, and E _a is a pseudo-activation energy. VTF plots for all samples are shown in FIG. 6 and the fitted parameters are listed in Table 1.

[00120] Lithium ion transference numbers were calculated using the equation as modified by Choe et al. (Chem. Mater., Vol. 9, pp. 369-379, 1997), accounting for variations in the electrolyte resistance (Equation 2).

[00121] In equation 2, Ri and R _f are the bulk resistances before and after the dc polarization, respectively. R _p is the resistance of the passivation layer on top of lithium metal. I ₀ is the initial current after applying the bias. I _s is the current in the steady state.

[00122] According to the model proposed by Bruce and Vincent (J. Electroanal. Chem., 225, pp. 1-17, 1987), the bias applied to the cell to determine the lithium ion transference number should be small enough to keep a low concentration gradient within the electrolyte. For systems where the bulk resistance is comparable to or even larger than the resistance of the passivation film, potentials in the order of 5 mV are commonly employed for the task. However, when the electrolyte layer is much more conductive than the passivation film, the application of small biases leads to a large ohmic drop through the interface, and the effective potential in the electrolyte is too small to result in consistent readings.

[00123] In this Example, the dc relaxation step required higher voltages (20-130 mV) only for measurements performed at room temperature. In all other cases, the actual iR drop in the electrolyte was typically kept in the range of 2-5 mV, low enough to comply with the low concentration gradient requirement. [00124] Above 100 °C, the impedance spectra for some cells (-50% of 1M PP13 and -10% of h- BN composite) presented features that were not observed at other temperatures, as shown in FIG. 7C. The existence of two semicircles in the spectra can be ascribed to either a charge transfer resistance or to another passivation film. Resistances arising from both processes are expected to decrease with an increase in temperature, but the additional semicircle always seemed to present larger Rp values than it would be expected.

[00125] Example 1.5. Electrolyte characterization

[00126] Scanning electron microscopy images were acquired in a FEI Quanta 400 ESEM FEG, using 30 kV as acceleration voltage. The samples were carefully deposited over carbon tape and stored in an Argon-filled glovebox until used. The electrochemical stability of the neat ionic liquid, the lithium salt solution and the h-BN-based composite were evaluated through 3- electrode experiments, using stainless steel as a working electrode and lithium metal strips as both reference and counter electrode. In order to evaluate the stability windows both at room temperature and at 120 °C of the three electrolytes, an open-cell experiment inside an argon- filled glovebox was designed.

[00127] All the samples presented good cathodic stability, as shown in FIGS. 2C and 8. A broad and low intensity peak was found in all three samples at all temperatures and is possibly related to the formation of a passivation layer from the reaction of TFST anions with the stainless steel in the counter electrode. The position and intensity of the peak changed with the sample and temperature of the experiment.

[00128] At room temperature, addition of a lithium salt was found to keep the overall stability limit of anodes almost unchanged, although upshifting a low intensity redox peak. However, the h-BN composite presented enhanced anodic stability. At 120 °C, the anodic stability showed drastic changes, rising from -4 V for the RTIL to -4.9 V for the solution and -5.4 V for the h- BN electrolyte. Although the origin of this difference in behavior is unknown, it can be related to changes in the solvation state of the ions within the electrolyte or to differences in the surface reactivity of lithium metal with the different systems.

[00129] Example 1.6. Room temperature performance of LTQ half-cells using the h-BN-RTIL composite electrolyte

[00130] LTO half-cells tend to present high resistance at room temperature, as visible from FIG. llC, as a consequence of both lower ionic conductivities from the electrolyte (-0.2 mS cm ^"1 ) and a less efficient electrolyte wetting onto the electrode surface, due to the elevated viscosity of the composite. The combination of these two factors leads to larger cell polarization (FIGS. 11A-B) and limits the cell capacity. Oftentimes, an oscillating cycling behavior is initially observed until the cell stabilizes, as visible in FIG. 11D. Although the capacities reported for high temperature experiments presented consistent values, experiments at room temperature provided capacities ranging from 54 mAh g ^"1 to 101 mAh g ^"1 , being highly dependent on the thickness of the electrolyte layer.

[00131] Example 1.7. Electrode aging at high temperature

[00132] It has been found that cycling a cell with organic electrolytes at high temperatures caused a diffusion of PVDF binder within the cathode, forming a thin polymer-rich layer on the electrode surface. Moreover, part of the binder was transferred through the separator to the anode. This process was found to increase cell resistance and contribute to the capacity fade. It is also important to note that 120 °C is close to the melting point for PVDF, and the reduced mechanical properties can affect its binding properties and contribute to a reduced coulombic efficiency.

[00133] In order to confirm if the LTO anodes were stable towards the electrolyte at high temperatures, electrodes were sealed and placed in an oven at 120 °C for 20 days in the presence of the RTIL or the 1 mol L ^"1 solution of LiTFSI. The electrodes were them unsealed inside an Argon-filled glovebox, thoroughly washed with dimethyl carbonate and used to prepare half- cells using conventional organic electrolyte. The cyclic stability remained unchanged for the aged electrodes, with even a slightly higher capacity observed in comparison to cells prepared using fresh ones (FIG. 13A). However, even for electrodes aged in the absence of ionic liquid, the cells tended to present coulombic efficiencies slightly larger than one, indicating that a permanent change had taken place. Further investigation showed a large increase in the charge transfer resistances for cells using aged electrodes (FIGS. 13C-D).

[00134] Example 1.8. Impedance evolution during cycling at 120 °C

[00135] Despite the stable cycling performance at room temperature observed after individually aging the electrode and the electrolyte, the small changes observed in the charge transfer resistance and coulombic efficiencies indicate that time dependent processes may occur on a device level upon long term cycling. Based on the information gathered for the separated components, the impedance evolution of the half-cells at 120 °C can be analyzed. A total absence of interfacial processes was observed for an uncycled cell, resulting in a purely diffusive behavior under impedance spectroscopy test (FIG. 14A). A high frequency semicircle is visible after a few cycles at a C/8 rate (FIGS. 14B-C), from a convolution of the low resistive passivation layer on top of the lithium metal and charge transfer processes with contribution from both electrode and electrolyte. An additional contribution for the semicircle may come from an unexpected formation of a thin SEI-like layer on top of the LTO electrode, explaining the irreversible capacity in the first cycle of charge-discharge (FIGS. 3B and 10).

[00136] Electrolyte aging effects seemed to be dominant at early stages, with larger resistances from the electrode aging taking over upon further cycling. On these same plots, it is possible to observe that the electrolyte resistance is nearly constant, even after extensive cycling, further indicating the stability of the composite.

[00137] Example 1.9. Self-discharge tests [00138] The cells were initially cycled at 120 °C at a C/8 rate for 3 cycles, as shown in the blue curve in the profiles of FIG. 15A. Thereafter, the LTO was lithiated (charged) at a C/8 rate and then allowed to self-discharge (SD) (still at 120 °C, for 24 hours). The SD curve was recorded in an Autolab potentiostat and is shown in the red portion of FIG. 15A. It is possible to see that the potential quickly rises to a value close to the plateau potential until it stabilizes. After the time was over, the LTO was galvanostatically delithiated (C/8 rate) to evaluate the remaining charge stored in the cell. The half-cell was then cycled for two more cycles at the same rate to investigate if there was any irreversible loss of capacity.