Synthesis of partially disordered phase in rocksalt materials via electrochemical pulsing

Abstract

A method of processing a rocksalt material for δ phase transformation includes obtaining a synthesized disordered rocksalt (DRX) material; heating the DRX material to above a predetermined temperature; and electrochemically pulsing the DRX material over one or more predetermined voltage windows, and / or at one or more predetermined current density rates, for a predetermined duration.

Description

[0001] SYNTHESIS OF PARTIALLY DISORDERED PHASE IN ROCKSALT MATERIALS VIA ELECTROCHEMICAL PULSING

[0002] FIELD OF THE INVENTION

[0003] The present invention relates to rocksalt materials having partially disordered spinel-like “8 phase” structures, and methods for promoting accelerated transformation of the 8 phase structures in these materials to reduce manufacturing times of the same for use in electrode engineering.

[0004] BACKGROUND OF THE INVENTION

[0005] Demands in the automotive industry, including industry-wide goals of reducing greenhouse gas emissions, have promoted a transition to electric vehicles which has led to rapid growth in the lithium-ion battery industry.

[0006] Modem lithium-ion batteries are typically made with cathode materials having a general composition of LiNii-xMxCh, where the metal M is commonly chosen from Co, Mn, Al and combinations thereof. Conventionally, cathode materials have been made with a layered structure that provides superior energy density and prolonged power emission, which is critical for vehicle applications. However, these conventional cathode materials present cost and affordability issues in vehicle manufacture applications as they require rare metals, such as nickel or cobalt, which are of high cost and limited availability. The limited availability of these rare metals also raises concerns that these materials may not be a sustainable option for long-term vehicle manufacturing needs.

[0007] Efforts to identify alternative cathode material compositions of more abundance have led to the identification of lithium iron phosphate (LiFePCU) as a viable and sustainable alternative. However, lithium iron phosphate has a significant energy density disadvantage that is undesirable for vehicle applications. As such, there remains a pressing need for a cathode material that is both earth abundant and characterized by a high energy density, and which would be commercially feasible for high-throughput manufacturing.

[0008] SUMMARY OF THE INVENTION

[0009] The present invention is inclusive of methods of processing a rocksalt material for transformation of a spinel-like nano-domain structure, term 8 phase structures. The methods generally include: obtaining a synthesized disordered rocksalt (DRX) material; processing the DRX material into an electrode by mixing it with carbon and a binder; assembling a lithium-ion cell using the processed electrode; heating the lithium-ion cell above a predetermined temperature; and electrochemically pulsing the DRX material over one or more predetermined voltage windows and / or at one or more predetermined current density rates for a predetermined duration.

[0010] The DRX material is a prior-synthesized DRX material, preferably with a Li-excess and a general formula of Lii+xTMi-xCL-yFy. It is further preferred that the transition metal (TM) comprise Mn, optionally with one or more additional metals, with the DRX material being Mn-rich. The method may be used with either small-particle DRX materials having particles with a minor axis of less than about 500 nanometers (nm), or large-particle DRX materials having particles with a minor axis greater than about 500 nm. Without being bound by theory, it is expected that large-particle DRX materials used in the inventive methods will most commonly have particles with a minor axis in a range of about 500 nm to about 5 microns, though some DRX materials with particle sizes greater than about 5 microns may also be suitable.

[0011] The predetermined temperature above which the DRX material is heated is approximately that temperature needed for the DRX material to tolerate the elevated currents used during the formation cycling. When the DRX material is an Li -excess material, the predetermined temperature is an elevated temperature that increases Li diffusivity of the DRX material. The predetermined temperature may be in a range of about 25°C to about 50°C, and in some instances may be higher than 50°C.

[0012] The voltage window may range from about 1.5 V to about 4.8 V, preferably from about 2.5 V to about 4.5 V, and more preferably from about 3.0 V to about 4.5 V. The current density rate may range from about 100 mA / g to about 1000 mA / g, in some instances from about 100 mA / g to about 500 mA / g, and in some instances from about 250 mA / g to about 500 mA / g (normalized by the mass of the electrochemically active material).

[0013] The predetermined duration may be a predetermined period of time and may range from about 1 day to about 5 days, in some instances from about 1 day to about 2 days, and in some instances may be more than 5 days. Alternatively, the predetermined duration may be a predetermined number of pulses, and may range from 10 to 100 pulses, in some instances from 20 to 40 pulses, and in some instances from 60 to 80 pulses.

[0014] Methods according to the present invention may further include, after heating the DRX material above the predetermined temperature, and prior to electrochemically pulsing the DRX material, charging the DRX material to an initial standard charge. The initial standard charge may exceed the maximum voltage of the voltage window used for subsequent electrochemical pulsing and may be approximately 4.8 V. Methods according to the present invention may further include, upon cessation of the predetermined duration for the electrochemical pulsing, executing at least one galvanostatic charge cycle of the DRX material. The one or more galvanostatic charge cycles executed at this time are test cycles, for example, to verify that the material is transformed, or for any other testing apart from cell utilization. A test cycle may include charging the DRX material to a voltage exceeding a maximum of the voltage window used for electrochemical pulsing and discharging the DRX material to a charge below a minimum of the voltage window used for electrochemical pulsing. A test cycle may include charging the DRX material to approximately 4.8 V and discharging the DRX material to approximately 2.0 V. This test cycling may also serve the dual function of forming the anode’s solid electrolyte interphase (SEI) for which the electrochemical pulsing targeting the cathode may or may not be well suited.

[0015] The 5 phase structure is characterized by partially disordered spinel-like domains with short coherence lengths, measuring less than 10 nm in size, with an average size of approximately 4 nm. The extent of 16c / 16d site disorder obtained from a single-phase, partially disordered spinel refinement model with periodic interruptions to the spinel-like order, but not the rocksalt parent lattice, as determined by synchrotron X-ray diffraction, may be approximately 5%. These interruptions in spinel-like order result from anti-phase boundaries formed between different spinel-like domains that nucleate during transformation and impinge on each other as they grow. These boundaries consist of roughly 1 nm thick disordered regions, maintained because the domains are different variants of spinel-like order formed on the higher symmetry rocksalt lattice. The different domains that meet at anti-phase boundaries represent the eight spinel translational and rotational variants resulting from the symmetry reduction from the cubic rocksalt lattice (Fm3m) to that of spinel (Fd3m). There are no significant regions of disorder, though there may be some regions of disorder. A fully transformed 5 phase structure may be characterized by a partially disordered nano-domain structure (approximately 5% 16c / 16d disorder and 4nm domains), thin disordered anti-phase boundaries between different spinel-like variants, with substantial homogeneity and minimal disorder.

[0016] The present invention is inclusive of DRX materials (and methods of processing the same) with large single-crystalline particles, with particles sizes greater than 1 pm (e.g., 2-3 pm, on average), that cycle with greater than 600 Wh / kg. The 8 phase structure in these DRX materials have minimal remnant disorder, with a high degree of order within domains, and anti-phase boundaries between adjacent domains. The 6 phase domains of these Mn-rich DRX materials contain at least four and up to eight spinel variants, produced in large, unmilled particles and transformed by electrochemical pulsing to a stable, fully transformed state. These 5 phase DRX materials may be obtained via synthesis of a disordered-rocksalt material, for example, by solid-state or molten salt methods, and subsequently transformed using an electrochemical pulse sequencing.

[0017] The present invention is inclusive of a cathode active material, comprising an Li-TM-0 compound comprising a particle distribution with particle sizes (P) in a range of 100 nm < P < 10 pm, where TM is one or more transition metals; wherein particles in the particle distribution have a crystal structure comprising a disordered rocksalt (DRX) lattice with a plurality of variant 8-phase domains of spinel -like lattice formed therein, each 8-phase domain differing from one another, with adjacent 8-phase domains separated by an anti-phase boundary. The crystal structure may comprise at least four variant 8-phase domains, optionally eight variant 8-phase domains, with anti-phase boundaries therebetween measuring approximately 1 nm.

[0018] Each 8-phase domain measures in a range from approximately 3 nm to approximately 10 nm, with the 8-phase domains differing from one another based on at least one difference chosen from: lattice orientation; occupancy of the 16c / 16d Wyckoff sites; and coherence length. Differences in lattice orientation may include a translational difference in lattice orientation, a rotational difference in lattice orientation, and a combination thereof. Each 8-phase domain has a 16c / 16d site disorder percentage of approximately 10.2% to approximately 3.6%; optionally 10.2% to approximately 3.6%. Each 8-phase domain has a coherence length (L) of L < 10 nm; optionally L < 5 nm; optionally L < 4 nm; and optionally in a range of 2.4 nm < L < 3.9 nm.

[0019] The cathode active material is formed from a transformed DRX material that was previously processed to form the 8-phase domains through electrochemical pulse-sequencing, the pre-processed, as-synthesized DRX material having been obtained via a prior-synthesis process comprising one of: a solid-state synthesis process and a molten-state synthesis process. The as-synthesized DRX material having been, optionally, subjected to particle-size-reduction processing prior to electrochemical pulsesequencing for transformation of the 8-phase domains; or subjected to electrochemical pulsesequencing for transformation of the 8-phase domains without prior particle-size-reduction processing.

[0020] The particle distribution comprises particle sizes (P) of 100 nm < P < 10 pm; optionally, 100 nm < P < 500 nm; optionally, 500 nm < P < 1 pm; optionally, 1 pm < P < 10 pm; optionally, 1 pm < P < 5 pm; optionally, 2 pm < P < 3 pm. The transition metal (TM) may comprise Mn, alone or in combination with one or more other transition metals, with Mn in an amount of 0.6 < Mn < 0.7.

[0021] Both the foregoing general description and the following detailed description are exemplary and explanatory only and are intended to provide further explanation of the invention as claimed. The accompanying drawings are included to provide a further understanding of the invention; are incorporated in and constitute part of this specification; illustrate embodiments of the invention; and, together with the description, serve to explain the principles of the invention.

[0022] BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Further features and advantages of the invention can be ascertained from the following detailed description that is provided in connection with the drawings described below:

[0024] FIG. 1 provides a schematic representation of the inventive process, illustrating an as- synthesized DRX material, a pulse sequencing processing of the DRX material, and a 8 phase DRX material (8-DRX) produced from the pulse sequencing process.

[0025] FIG. 2 shows an x-ray diffraction (XRD) scan for an as-synthesized DRX material having a formula Li1.1Mn0.8Ti0.1O1.9F01.

[0026] FIGS. 3A-3B show energy dispersive X-ray spectroscopy (EDS) images of Mn, Ti, O and F, in a DRX material, in which: FIG. 3A shows EDS images of an as-synthesized solid-state DRX material; and FIG. 3B shows EDS images of the DRX material after ball-milling.

[0027] FIGS. 4A-4B show scanning electron microscopic (SEM) images of the DRX material, in which: FIG. 4A shows an SEM image of the as-synthesized solid-state DRX material; and FIG. 4B shows an SEM image of the DRX material after ball-milling.

[0028] FIGS. 5A-5B show examples of the inventive electrochemical pulse-cycling process for the rapid transformation of 8 phase structures within a prior-synthesized DRX material, in which: FIG. 5A provides a schematic illustration of atypical pulse-cycling process; and FIG. 5B provides a further detailed, non-limiting example of a pulse-cycling process.

[0029] FIG. 6 shows the results of temperature differences during pulse sequencing on the transformation of 8 phase.

[0030] FIGS. 7A-7D show results of galvanostatic charge cycles of various DRX materials processed via pulse-cycling with variations in voltage windows, in which: FIG. 7A shows a first collection of galvanostatic charge cycles of DRX materials across various voltage windows; FIG. 7B shows a second collection of galvanostatic charge cycles of DRX materials across various voltage windows; FIG. 7C shows differential capacity (dQ / dV) plots for the galvanostatic charge cycles in FIG. 7B; and FIG. 7D shows a third collection of galvanostatic charge cycles of DRX materials across various voltage windows.

[0031] FIGS. 8A-8C show results of galvanostatic charge cycles of various DRX materials, in which: FIG. 8A shows results from a first full galvanostatic charge of the DRX materials processed via pulsecycling using different numbers of pulses; FIG. 8B shows differential capacity (dQ / dV) plots for the galvanostatic charge cycles in FIG. 8A; and FIG. 8C shows the initial cycling stability of the DRX materials from FIG. 8A.

[0032] FIGS. 9A-9C show results of galvanostatic charge cycles of various DRX materials, in which: FIG. 9A shows results from a first full galvanostatic charge of the DRX materials processed via pulsecycling using different current density rates; FIG. 9B shows differential capacity (dQ / dV) plots for the galvanostatic charge cycles in FIG. 9A; and FIG. 9C shows the initial cycling stability of the DRX materials from FIG. 9A.

[0033] FIGS. 10A-10B show time efficacy of pulsing at different rates, in which FIG. 10A shows capacity delivered vs. pulse number; and FIG. 10B shows capacity delivered vs. elapsed time during pulsing of ball-milled solid-state material 80 times at 50°C, from 3.0 V to 4.5 V at different rates.

[0034] FIG. 11 shows XRD scans illustrating the evolution of spinel-like structure peaks in DRX materials following various pulse-cycling.

[0035] FIGS. 12A-12H show synchrotron XRD images of DRX materials subjected to different numbers of pulses, in which FIG. 12A shows XRD images of a pristine material as a powder, and in which the remaining figures show XRD images for electrodes measured ex situ, in which DRX materials were processed with a different number of pulse cycles, in which: FIG. 12B shows results after 5 pulses; FIG. 12C shows results after 10 pulses; FIG. 12D shows results after 20 pulses; FIG. 12E shows results after 40 pulses; FIG. 12F shows results after 60 pulses; FIG. 12G shows results after 80 pulses; and FIG. 12H shows results after 100 pulses.

[0036] FIG. 13 shows scanning electron nano-diffraction (SEND) mapping performed on pulse sequenced solid-state synthesized material.

[0037] FIGS. 14A-14F illustrate correlations between electrochemical change, domain growth, and ordering during 5 phase transformation, in which: FIG. 14A shows peak differential capacity of 3V plateau after pulsing vs. estimated domain size; FIG. 14B shows peak differential capacity of 3V plateau after pulsing vs. estimated 16c / 16d site disorder; FIG. 14C shows length of 4V plateau after pulsing vs. estimated domain size; FIG. 14D shows length of 4V plateau after pulsing vs. estimated 16c / 16d site disorder; FIG. 14E shows length of 4V plateau during pulsing vs. estimated domain size; and FIG. 14F shows length of 4V plateau during pulsing vs. estimated 16c / 16d site disorder.

[0038] FIG. 15 shows evolution of the 4V plateau feature in the voltage curve over the course of pulse sequencing from 5 pulse cycles to 100 pulse cycles. FIG. 16 shows differential capacity (dQ / dV) of the 3 V plateau feature as a function of total time in pulse sequencing from 5 pulse cycles to 100 pulse cycles at 100 mA / g and cycling at 20 mA / g.

[0039] FIG. 17 shows a comparison of DRX materials subjected to various numbers of pulse cycles and charge cycles, with equivalent number of the two processes.

[0040] FIG. 18 shows galvanostatic charge cycles of various DRX materials that were processed via pulse-cycling, both with and without additional post-transformation pulse-cycling at ineffective voltage windows above and below the ideal range of approximately 3.0 V - 4.5 V.

[0041] FIG. 19 shows an SEM image of Li1.1Mn0.8Ti0.1O1.9F01 synthesized by the molten salt method.

[0042] FIG. 20 shows galvanostatic charge cycles of various DRX materials processed via pulsecycling using different (constant) current density rates.

[0043] FIG. 21 shows galvanostatic charge cycles of various DRX materials that were processed via pulse-cycling, with constant voltage holds at 3.0 V / 4.5 V down to 10 mA / g rather than constant current galvanostatic pulsing as in FIG. 20, with the results achieved at the 1stand 10thoperational cycles at varying cycling temperatures.

[0044] FIG. 22 shows comparative discharge capacity results of various DRX materials processed according to inventive methods and a comparative control DRX material processed according to a conventional method.

[0045] FIG. 23 shows SEM micrograph imaging of molten salt synthesized DRX material showing single crystals with a homogenous particle size of 2-3 pm.

[0046] FIG. 24 shows HAADF-STEM and EDS imaging performed on the molten salt synthesized DRX material showing a uniform distribution of Mn, Ti, O, and F.

[0047] FIG. 25 shows XRD imaging of the molten salt synthesized DRX material confirming that the product is a phase-pure DRX.

[0048] FIG. 26 shows first full cycles of constant-current pulsed molten salt synthesized DRX material, pulsed at 50°C and low constant-current rates, along with a pristine material sample.

[0049] FIGS. 27A-27D show test results of constant-current pulsed molten salt synthesized DRX material, in which: FIG. 27A shows a first charge of the molten salt synthesized DRX material electrode at 50°C and 10 mA / g, with a decrease in voltage at roughly 3.5 V attributed to large overpotential at low state of charge in large particle DRX; FIG. 27B shows voltage and current profile of constant voltage cycling from 10 A / g down to 10 mA / g; FIG. 27C shows duration of each pulse cycle over the course of 80 pulse cycles in a pulse sequence; FIG. 27D shows specific capacity of each charge and discharge pulse cycle. FIGS. 28A-28C show further test results of molten salt synthesized DRX material, in which: FIG. 28A shows cycling performance of the molten salt synthesized DRX material at 25°C after 80 pulses with constant voltage holds; FIG. 28B shows cycling performance of the molten salt synthesized DRX material at 50°C after 80 pulses with constant voltage holds; and FIG. 28C shows extended cycling of pulsed molten salt synthesized DRX material at 25°C and at 50°C, along with a control cell which has not been pulsed prior to cycling.

[0050] FIG. 29 shows synchrotron XRD imaging of a 5 phase transformed molten salt synthesized DRX material, measured ex situ after pulse sequencing.

[0051] FIGS. 3OA-3OF show multimodal structural characterizations of 8 phase in a molten salt synthesized DRX material, in which: FIG. 30A shows a low magnification HAADF-STEM image of the pulse cycled molten salt synthesized DRX material; FIG. 30B shows mean SEND patterns from the region marked by the solid line box in FIG. 30A; FIG. 30C show spatial distribution of the 8 phase obtained by virtual imaging the diffraction spots unique to the Fd3m space group (scale bar: 5 nm); FIG. 30D shows atomic resolution HAADF-STEM micrograph collected along the

[0110] zone axis from the region marked in the solid-line box in FIG. 30A; FIG. 30E shows the 8 phase lattice obtained by Fourier filtering of frequency Fl in FIG. 30D; and FIG. 30F shows the 8 phase lattice obtained by Fourier filtering of frequency F2 in FIG. 30D.

[0052] FIG. 31 shows diffraction patterns from select points indexed based on the structural information obtained from the synchrotron XRD refinement of a 8 phase transformed molten salt synthesized DRX material.

[0053] FIG. 32 shows virtual imaging of marked scattering range for 8 phase structure, with scattering angles peaks unique to the spinel space group Fd3m marked as annular circles with peak vectors of (0.175, 0.250) A ' and (0.59, 0.67) A"1.

[0054] FIG. 33 provides a raw atomic resolution HAADF-STEM image of the 8 phase, without filtering.

[0055] DETAILED DESCRIPTION OF THE INVENTION

[0056] The following disclosure discusses the present invention with reference to the examples shown in the accompanying drawings, though does not limit the invention to those examples.

[0057] The use of any and all examples, or exemplary language ( .g., “such as”) provided herein is intended merely to better illuminate the invention and does not pose a limitation on the scope of the invention unless otherwise claimed. No language in the specification should be construed as indicating any non-claimed element as essential or otherwise critical to the practice of the invention, unless otherwise made clear in context.

[0058] As used herein, the singular forms “a,” “an,” and “the” include plural referents unless the context clearly dictates otherwise. Unless indicated otherwise by context, the term “or” is to be understood as an inclusive “or.” Terms such as “first”, “second”, “third”, etc. when used to describe multiple devices or elements, are so used only to convey the relative actions, positioning and / or functions of the separate devices, and do not necessitate either a specific order for such devices or elements, or any specific quantity or ranking of such devices or elements.

[0059] The word “substantially”, as used herein with respect to any property or circumstance, refers to a degree of deviation that is sufficiently small so as to not appreciably detract from the identified property or circumstance. The exact degree of deviation allowable in a given circumstance will depend on the specific context, as would be understood by one having ordinary skill in the art.

[0060] Use of the terms “about” or “approximately” are intended to describe values above and / or below a stated value or range, as would be understood by one having ordinary skill in the art in the respective context. In some instances, this may encompass values in a range of approx. + / — 10%; in other instances, there may be encompassed values in a range of approx. -+7—5%; in yet other instances values in a range of approx. + / -2% may be encompassed; and in yet further instances, this may encompass values in a range of approx. + / -1%.

[0061] It will be understood that the terms “comprises” and / or “comprising,” when used in this specification, specify the presence of stated features, integers, steps, operations, elements, and / or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof, unless otherwise indicated or clearly contradicted by context.

[0062] The term “5 phase” (delta phase) as used herein is intended to describe the specific partially disordered crystal structure formed during cycling of Mn-rich rocksalt materials. This structure is characterized by a coherent parent rocksalt lattice on which partial spinel-like ordering occurs. This ordering means that in the 5 phase Li is mostly found on sites similar to the 8a / 16c site of a spinel crystal, while Mn and other transition metals occupy mostly the 16d-like site. However, in both the case of Li and the transition metals, there is significant occupancy of the alternate sites, constituting the partial disorder of the material. This structure is distinct from the ordered spinel phase obtained at high temperature, such as the material LiMn2O4, in which Li and Mn are found essentially entirely on 8a and 16d sites, respectively. The terms “disordered rocksalt” and “DRX” as used herein are intended to describe the occupancy of multiple different crystallographic sites on the parent rocksalt lattice by Li and transition metals. These terms are further intended to encompass rocksalt materials with any degree of disorder, including “fully disordered rocksalts” and “partially disordered rocksalts”, unless otherwise indicated or clearly contradicted by context. In the case of an as-synthesized fully disordered rocksalt, the high degree of disorder means there is no long-range order at all, and there is only one inequivalent cation site on which all metals are distributed. In the case of the 5 phase which forms during pulse-cycling, the term “disorder” is used to indicate that while there is some long-range order and distinct 16c / 16d sites, there is still a probability that Li or transition metals will occur on any given site. For example, a given site may appear in a spinel -like environment as if it were a 16c site on which Li would typically be found in a discharged material, but instead is occupied by a Mn ion. The term “disorder” may also indicate that the local definition of which sites are the 16c sites and which are the 16d sites is different at different locations in the material (in addition to local disorder).

[0063] The term “transition metals” as used herein is intended to describe metals in the d-block (Groups 3-12) of the periodic table and other metals which perform an equivalent role in lithium (transition) metal oxide cathode materials. These elements may play either an active role in electrochemical processes by oxidation or reduction or remain in a fixed oxidation state. Some representative examples include manganese (Mn being oxidized from a 3+ state to 4+), and titanium (Ti as only 4+). Other metals which are not “transition metals” in the traditional context such as, but not limited to, aluminum (Al) are often referred to under this term in the battery field literature due to serving a similar purpose in the material as electrochemically inert metals such as Ti. Therefore, metals outside of the d-block may be included in the DRX formulation and referred to as “transition metals” in some instances.

[0064] The terms “pulsing”, “pulse sequencing”, and “pulse-cycling” as used herein are intended to refer to the inventive process of formation cycling or electrochemical pulsing via the application of high rate, restricted voltages. This will be understood as separate and distinct from traditional galvanostatic cycling via the wider voltage ranges traditionally used in research settings, which may also be referred to herein as “charge-cycling”.

[0065] Recitations of value ranges herein, unless indicated otherwise, serve as shorthand for referring individually to each separate value falling within the respective ranges, including the endpoints of the range, each separate value within the range, and all intermediate ranges subsumed by the overall range, with each incorporated into the specification as if individually recited herein. Unless indicated otherwise, or clearly contradicted by context, methods described herein can be performed with the individual steps executed in any suitable order, including: the precise order disclosed, without any intermediate steps or with one or more further steps interposed between the disclosed steps; with the disclosed steps performed in an order other than the exact order disclosed; with one or more steps performed simultaneously; and with one or more disclosed steps omitted.

[0066] The present invention addresses lithium-ion batteries having cathode materials with disordered rocksalt structures (DRX) that have a lithium-excess[1]. These DRX materials are defined by a nearly random arrangement of cations, including Li and one or more transition metals, in a rocksalt oxide or oxyfluoride.

[0067] DRX materials display impressive energy density when synthesized with an excess of Li, as opposed to a balanced 1: 1 ratio of Li to transition metal(s) as otherwise provided in a typical LiMCL formulation of an ordered rocksalt structure. This is because a lithium-excess enhances lattice site connectivity, enabling Li percolation through low-energy pathways between lattice sites. These DRX materials present an advantage in that they may be synthesized with only earth abundant metals such as Mn and Ti, as opposed to the rare metals in typical cathode materials.

[0068] Unfortunately, DRX materials generally display lower Li diffusivities than materials having ordered structures, as characterized for example by slower Li transport in the DRX materials. To counter this, these DRX materials are ball-milled to reduce the as-synthesized powders to a sufficiently small particle size that enables higher energy densities. However, there remains a reduced electrodelevel energy density with these DRX materials due to the necessity of a large carbon content in the electrodes to provide electrical contact with the increased surface are of the cathode particles that results from the ball-milling of the DRX materials, with further potential of a diminished cycle life, neither of which is ideal nor practical for vehicle applications.

[0069] Greater success has been found with variants of Mn-rich DRX materials that have low inactive dopant content, such as Lii+xMni.x-yMyO2-zFz, where x < 0.2, y < 0.2, z < 0.2. Mn-rich DRX materials have been observed to undergo a partial structural transformation during electrochemical cycling upon which regions of the DRX material transform to the 8 phase, as indicated, among other characteristics, by the presence of 3 V and 4 V plateaus characteristic of spinel-like order[2,3]. This partially disordered structure has been found to increase the energy density and rate performance of the DRX material due to improved Li transport through low-energy 3D diffusion pathways in the transformed crystal structure. Additionally, the improved performance of these partially transformed materials is retained for a longer period of time, yielding an improved cycle life relative to DRX materials without such transformation.

[0070] The stable cycling of these transformed 8 phase materials is remarkable when compared to the rapid degradation of the archetypal ordered spinel structure LiMn2O4 (Lio.sMnCL) when Li is inserted to the rocksalt composition of Li2Mn2O4 (LiMnCh). This degradation is attributed to a two-phase reaction which occurs simultaneously with an anisotropic crystal volume increase from the collective Jahn-Teller distortion of the Mn3+ions, which yields an inhomogeneous volume change that results in disintegration of the cathode particles, leading LiMn2O4 to only be usable over the 4 V plateau where Li is extracted only from tetrahedral sites. The variant DRX materials which transform to the 8 phase are of great interest due to their ability to cycle more Li, thereby yielding a higher energy density without the poor cycle life of an ordered spinel structure. Hereafter, the spinel -like structures that result upon partial transformation of the variant DRX materials may be referred to simply as 8 phase structures as has been adopted in the scientific literature on these materials.

[0071] While the DRX materials with 8 phase structures provide excellent performance characteristics, complete transformation of a DRX material to yield 8 phase structures typically requires at least 20-30 charge / discharge cycles between 2.0 V and 4.8 V. Due to the relatively low currents that DRX materials sustain immediately after synthesis and ball-milling in initial cycles (e.g., approximately 20 mA / g), this charge / discharge cycling (i.e., a charge-cycling) may take on average 2-4 weeks for complete transformation of the 8 phase structures. If seeking to employ DRX materials with larger particle sizes e.g., without ball-milling), for improved cycle life and electrode energy density, then it is likely a complete transformation could not be achieved at all due to the poor diffusivity of the initial DRX material. Given the needs for high product throughput in the commercial manufacturing of batteries, it is expected that partially transformed DRX materials with 8 phase structures will not be adopted commercially if the transformation process continues to require 2-4 weeks of processing, as this is not likely to be deemed commercially viable for the necessary manufacturing throughput.

[0072] DRX materials with 8 phase structures are a promising class of earth-abundant materials that possess adequate energy densities sufficient for use in lithium-ion battery cathodes in vehicle applications. The main drawback of these transformed DRX materials is the long processing time needed for forming the 8 phase structures under conventional methods.

[0073] The present invention provides methods for accelerating the transformation of DRX materials to more rapidly yield 8 phase structures, while achieving comparable properties as those yielded by conventional processing methods. These inventive methods are expected to greatly improve commercial feasibility of 6 phase materials as an alternative chemistry for use in lithium-ion battery cathodes for vehicle applications, reducing reliance on materials that otherwise require more expensive and less readily available metals such and nickel and cobalt.

[0074] Methods according to the present invention proceed, generally, as follows:

[0075] (1) Synthesize or otherwise obtain a prior-synthesized DRX material;

[0076] (2) Assemble a typical lithium-ion battery cell using the DRX material as the active material in a cathode comprising DRX, carbon, and a plastic binder;

[0077] (3) Elevate the temperature of the DRX containing cell above a predetermined threshold; and

[0078] (4) Subject the DRX material to repeat electrochemical pulse-cycling over one or more predetermined voltage ranges, and / or at one or more predetermined current density rates, for a predetermined period (e.g., number of pulses or duration of time).

[0079] The present invention is further inclusive of the compounds, compositions and materials produced by the foregoing methods, as well as products produced therefrom, including, without limitation, cathode active materials with a Li-TM-0 (e.g. , Li-Mn-O) oxide or oxyfluoride composition, and Li-ion rechargeable batteries incorporating such cathode active materials.

[0080] FIG. 1 provides a schematic representation of the inventive process, illustrating the prior- synthesized, parent DRX material in the Fm3m space group; a charge / discharge state of the DRX material during a pulse cycle in the pulse sequencing process; and a resulting 5 phase DRX material (8-DRX) in the Fd3m space group, produced by pulse sequencing process.

[0081] The synthesized DRX material may be of any given compositional formula, though preferably has a general formula of Lii+xTMi-x02-yFywhere x < 0.2, y < 0.2. In preferred examples, the transition metal TM will comprise Mn, optionally with one or more additional metals, such that the DRX material has a general formula of either Lii+xMni-xO2-yFyor Lii+xMni-x-yTMyO2-zFz, where TM is chosen from Ti and / or one or more other metals, such as Al, Cr, Ni, and other such like metals. Preferably, the DRX material is Mn-rich, with Mn being the most prevalent transition metal in the composition, preferably with Mn > 0.6, and more preferably Mn > 0.7. The as-synthesized DRX material may be one that was previously subjected to particle-size-reduction processing (e.g., ballmilling) prior to the inventive pulse-cycling or may be one that was not previously subjected to particle-size-reduction processing. A DRX material that was subjected to particle-size-reduction (i.e., a small-particle DRX material) may comprise particles with a typical minor axis of on average less than 500 nanometers (nm) and a DRX material that was not subjected to particle-size-reduction (z.e., a large-particle DRX material) may comprise particles with a typical minor axis of greater than about 500 nm. Particles in this context will include single crystal particles having the corresponding diameter (as with the molten salt synthesized material described here) and secondary particles comprised of multiple smaller primary particles that together have the corresponding diameter.

[0082] The DRX material is to be heated to above a predetermined elevated temperature that is approximately the practical limit for industrial formation cycling of the respective DRX material given the constraints of the pulsing apparatus, the thermal stability of the electrolyte, anode or other cell components, and the constraints of the cell casing. As used here, the language “elevated temperature” refers to a temperature exceeding room temperature and which improves the kinetics of high-rate pulse charges and discharges to yield an improvement in Li diffusivity during pulse-cycling. For example, for Mn-rich DRX materials, the predetermined elevated temperature may be approximately 50°C, which was observed as approximately the practical limit for formation cycling thereof, as set by a halfcell apparatus used during development of the present invention. Elevated temperatures that may be used in the inventive methods may range from about 25°C to about 50°C, including all individual temperatures and temperature ranges therebetween. However, examples according to the present invention may use elevated temperatures exceeding 50°C depending on the specific compositions in question. Elevating the temperature of the DRX material increases Li diffusivity, facilitating removal of Li from tetrahedral sites of the crystal structure to create interstitial vacancies that promote more facile migration of the transition metal ions (e. , Mn) in place thereof. The elevated temperature also enables the DRX material to tolerate an increased current while keeping Li content substantially homogeneous between the surface and bulk of the cathode material.

[0083] The cell containing the DRX material is heated to above an elevated temperature and subjected to a series of electrochemical pulses for a predetermined duration during which the material is repeatedly charged and discharged across one or more predetermined voltage window(s) and / or at one or more predetermined current density rate(s). This is done with the intention of promoting the transformation to the 8 phase prior to standard battery use. Rapid, repeat pulse cycling promotes full transformation of a DRX material with minimal degradation, with 3 V and 4 V plateau features growing in early pulse cycles and slowing in later pulse cycles.

[0084] The predetermined duration may be a predetermined period of time or a predetermined number of pulses. When the predetermined duration is a period of time, the period of time may be less than about 1 to about 2 days in instances where quicker processing of the DRX material is favored or may be approximately 5 days or greater in instances where increased maximum capacity of the DRX material is favored. When the predetermined duration is a number of pulses, the number of pulses may be any number of pules that has been determined in advance to yield a targeted processing time, a targeted material property, or a combination of the two. For example, 20-40 pulses may be used in instances where quicker processing time is favored, and 60-80 pulses (or more) may be used in instances where maximum capacity of the DRX material is favored. Without being bound by theory, while pulse sequencing beyond 80 pulses may further increase a maximum capacity of the DRX material, there is expected diminishing returns to the increase in capacity at pulses beyond approximately 80 in total such that gains to maximum capacity are not likely to justify the trade-off in additional processing time.

[0085] The voltage window may range from about 1.5 V to about 4.8 V; preferably from about 2.5 V to about 4.5 V; and more preferably from about 3.0 V to about 4.5 V. Current density rates may range from about 100 mA / g to about 1000 mA / g; preferably from about 100 mA / g to about 500 mA / g; and more preferably about 250 mA / g to about 500 mA / g. The current density rate may fluctuate during pulse-cycling, and in some examples pulse-cycling may be performed with current density rates across multiple of the foregoing ranges.

[0086] A 5 phase structure is achieved through the migration of transition metal ions between interstitial sites in the crystal structure. In examples of Mn-rich DRX materials, migration of Mn ions begins with movement to adjacent, vacant interstitial tetrahedral sites, mainly those which do not faceshare with any other ions. This requires a large vacancy concentration with most Li ions removed at a higher voltage, as the Mn ions are unlikely to persist at the tetrahedral sites when Li is present. Upon Li insertion, during discharge, Mn ions potentially move to a new octahedral site due to unfavorable interaction of tetrahedral Mn ions with face sharing Li ions, thus requiring that Li be cycled rather than held at a fixed voltage4. Mn ions mainly migrate when at low oxidation states of either Mn2+or Mn3+due to Mn4+being immobile on octahedral sites, thus requiring a low state of charge and a low- to-moderate voltage3.

[0087] Without being bound by theory, it is expected that during cycling of a DRX material the transition metal (e.g. Mn) likely migrates during a moderate-to-high voltage during charge, with complete migration to a new octahedral site promoted during discharge, and that heating the DRX material to the predetermined elevated temperature may promote faster processing of the charge / discharge cycles via the inventive pulse-cycling as opposed to conventional charge-cycling. Herebelow a discussion is provided on a series of tests conducted on coin cells with Li metal anodes in which Li1.1Mn0.8Ti0.1O1.9F01 was chosen as an Mn-rich DRX material for partial transformation of 8 phase structures. Though the following discussion addresses working examples using this particular material, it will be understood that the process is not limited to this single example, and may be applied to any chosen composition, preferably Mn-rich materials. Though the following examples discuss DRX materials prepared under specific synthesizing processes, it will be understood that the present invention is not limited to these synthesizing processes, and that the invention may be used with DRX (or other Mn-rich) materials synthesized via any number of various methods.

[0088] FIG. 2 shows x-ray diffraction data for an as-synthesized DRX material having the formula Li1.1Mn0.8Ti0.1O1.9F01. This DRX material was synthesized via a solid-state method at 1100 °C and ball-milled with carbon to yield a reduced particle size. Energy Dispersive X-ray Spectroscopy (EDS) Mn, Ti, O and F particles in the as-synthesized solid-state material (FIG. 3A) and the ball-milled material (FIG. 3B) confirmed the material was single-phase, with uniformly distributed constituent elements. SEM imaging showed the initial as-synthesized solid-state material to have a relatively wide particle size distribution inclusive of particles measuring less than 0.5 pm to greater than 10 pm (FIG. 4A); with the subsequently ball-milled material having a more homogenous particle size distribution ranging from approximately 100 nm to approximately 500 nm, inclusive of secondary particle agglomerates (FIG. 4B). Particle size distribution (PSD) is measured by laser diffraction particle size analysis using a Malvern Mastersizer 3000 with an Aero S unit.

[0089] FIG. 5A presents a schematic for the inventive pulse-cycling process for transformation of 8 phase structures within an as-synthesized DRX material. As illustrated in the schematic, typical pulsecycling includes a resting period of the synthesized DRX material. Preferably during or prior to this resting period, the DRX material is heated to an elevated temperature to yield improved Li diffusivity in the DRX material. The DRX material remains at an elevated temperature throughout the entirety of the pulse-cycling process. Pulse-cycling begins with an initial charge to a standard voltage, at a predetermined current density rate, followed by a discharge. The schematic in FIG. 5A shows an example in which the standard voltage was 4.8 V, though any voltage may be used, including a voltage corresponding with the upper voltage of the pulse sequencing. The DRX material is then subjected to pulse sequencing during which the material is subjected to repeat pulse-cycling within a predetermined voltage window for a predetermined duration, with the duration being either a predetermined period of time or a predetermined number of pulses. Upon cessation of the predetermined duration (e.g., the time or number of pulses), the pulse sequencing transitions to a conventional galvanostatic cycle with a charge cycle up to a maximum charge value (e.g., 4.8V) followed by a discharge cycle down to a minimum charge value (e.g., 2.0V) at a predetermined current density rate (e.g, 20 mA / g). Though the illustrated schematic shows the pulsing sequencing being performed in a voltage window of 3.0- 4.5 V, it will be understood that this process may be performed with different voltage windows, at different temperatures, at different rates, and for different durations than those illustrated in the schematic.

[0090] FIG. 5B presents a further detailed non-limiting example of a pulse-cycling process according to the present invention. In this example, the process was performed with an initial resting period of approximately 6 hours (hours 0 to 6) with the DRX material heated to an elevated temperature of approximately 50°C, followed by an initial charge and discharge in which the DRX material is charged to approximately 4.8 V and discharged to approximately 3.0 V over an approximate two hours (hours 6 to 8). Pulse sequencing is then performed over approximately fifty-four hours (hours 8 to 62) while the DRX material remains at the elevated temperature. During pulse sequencing, the DRX material is repeatedly subjected to multiple pulse-cycles, with each pulse-cycle having a charge phase in which the charge is increased from approximately 3.0 V to approximately 4.5 V over twenty minutes followed by a discharge phase in which the charge is decreased from approximately 4.5 V to approximately 3.0 V over twenty minutes, such that each individual pulse-cycle has an approximate forty minute period, resulting in eighty-one pulse-cycles over the pulse sequencing window. Upon completion of pulse sequencing, the DRX material rests for approximately five hours (hours 62 to 67) at a steady voltage of approximately 3.25 V followed by a conventional galvanostatic cycle. During the galvanostatic cycle the DRX material is charged from approximately 3.25 V to approximately 4.8 V over approximately six hours (hours 67 to 73), discharged from approximately 4.8 V to approximately 2.0 V over approximately eleven hours (hours 73 to 84), and again charged from approximately 2.0 V to approximately 3.0 V over approximately four hours (hours 84 to 88).

[0091] Use of the elevated temperature (e.g, 50°C) is beneficial for promoting a higher rate of 5 phase transformation. This was observed from testing on two DRX materials that were prepared in accord with the inventive process, though with one material subjected to pulse sequencing at 25°C and the other subjected to pulse sequencing at 50°C. As seen in FIG. 6, when these two materials were subsequently cycled for twenty -four hours at 100 mA / g, the material processed under the elevated temperature was found to display greater specific capacity.

[0092] FIG. 7A shows results from a first full galvanostatic charge cycle of various DRX materials with the formulation Li1.1Mn0.8Ti0.1O19F01, with charge cycling of the material performed with the material at a temperature of 25°C at a rate of 20 mA / g (per gram of active material) in a voltage range of 2.0 V to 4.8 V. Each material was processed via pulse-cycling, though each with pulse sequencing in a different voltage window. Each sample was subjected to pulse sequencing for a duration of 5 days at a moderate rate of 100 mA / g (per gram of active material), and at an elevated temperature of 50°C. Each sample was processed with a pulse sequencing in different 1 V voltage windows, from 1.5 V to 4.8 V, in increments of 0.5 V; though one sample was tested in a truncated voltage window of 4.0-4.8 V due to electrolyte oxidation. These tests show that transformation of the 8 phase structure occurs at a moderate voltage, as only the pulse-cycling performed in voltage windows between 2.5-4.5 V were observed to yield notable 3 V and 4 V plateau features characteristic of 8 phase structures.

[0093] FIG. 7B shows additional results from a first full galvanostatic charge cycle of various DRX materials each again processed via pulse-cycling, along with a pristine (control) sample, with each sample cycled while at a temperature of 25°C, at a rate of 20 mA / g, in a voltage range of 2.0 V to 4.8 V. Again, each inventive sample was previously formed via pulse sequencing for a duration of 5 days at a moderate rate of 100 mA / g (per gram of active material), and at an elevated temperature of 50°C, though with pulse sequencing in different voltage windows. As seen in FIG. 7B, while pulse sequencing in all voltage ranges were found to promote transformation of 8 phase structures, voltages above 3 V were found to promote transformation to a relatively greater extent, as evidenced by the formation of the characteristic 3 V and 4 V plateau features characteristic of 8 phase structures. It was further observed that voltage windows of 3.0 V to 4.5 V and 3.0 - 4.8 V performed almost equivalently, suggesting there is little Mn migration in the range of 4.5 - 4.8 V. FIG. 7C shows corresponding differential capacity (dQ / dV) plots for these further samples, showing different extents of the development of the 3 V and 4 V plateau features.

[0094] FIG. 7D shows additional results from a first full galvanostatic charge cycle of various DRX materials that were each again processed via pulse-cycling with the same conditions as those from the tests in FIG. 4A, though in these tests three samples were processed with pulse sequencing in different 1 V voltage windows within the preferred voltage window of 2.5-4.5 V, at 0.5 V increments, and three samples were processed with pulse sequencing in expanded voltage windows of 2.5-4.0 V, 3.0-4.5 V and 3.0-4.8 V, respectively. With this further testing, the sample processed with pulse sequencing in a voltage window of 3.0-4.5 V was observed to yield results most closely conforming to those expected from complete transformation of the 8 phase structure, again based on the development of notable 3 V and 4 V plateau features. FIG. 8A shows tests results from a first full galvanostatic charge cycle of various DRX materials that were each processed via pulse-cycling, with each sample subjected to pulse sequencing at a rate of 100 mA / g in a voltage window of 3.0 - 4.5 V, and an elevated temperature of 50°C. A pristine material (control) sample is also included. The materials were charged from 2.0 to 4.8 V at a temperature of 25°C. However, each sample was subjected to a pulse sequencing for a different duration, in this instance based on a predetermined number of pulses, with separate samples processed for 5 pulses; 10 pulses; 20 pulses; 40 pulses; 60 pulses; 80 pulses and 100 pulses. It can be seen that the capacity of the material rises rapidly, reaching a value of approximately 195 mAh / g in only 5 pulses, requiring less than 1 day of pulsing. At this point, the 3 V and 4 V plateau features typical of spinel-like ordering are already visible, suggesting the formation of spinel-like environments has begun with only a few pulses. As the number of pulses is increased, these features become more pronounced, with the plateau features becoming both flatter and longer, which can be more clearly observed in the increased height and area of the peaks in the differential capacity curve in FIG. 8B. By 60 - 100 pulses, the changes in the voltage curve have slowed dramatically, with 80 and 100 pulses being almost identical in terms of capacity growth and plateau formation, with it being seen that the specific capacity results obtained from the 80-pulse sample and the 100-pulse sample were so close in character that the data curves for these results substantially overlap in FIG. 8A. This supports the theory that pulse durations of greater than 80 pulses are likely to yield insignificant differences. The 80 pulses at 100 mA / g enables transformation of the 8 phase in approximately 7 days, as compared to the greater than 20 days necessary for transformation via conventional charge cycling at the typical rate of 20 mA / g.

[0095] The long-term cycling of these cells after pulsing, as well as of the control, is shown in FIG. 8C. Surprisingly, it was found that pulse sequencing these materials leads to an increase in capacity, without a corresponding negative effect on subsequent cycle life. The cell pulsed 80 times, for example, possesses a higher capacity than the control throughout the entire duration of 100 cycles. It was found that 60-80 pulses were sufficient to produce fully transformed material. Though samples subjected to only 20-40 pulses were observed to yield incomplete transformations, these samples nonetheless achieved maximum capacities comparable to or surpassing the control material that was processed via conventional galvanostatic cycling while also displaying peaks indicative of 8 phase structure in ex-situ XRD scans (See FIG. 11). Thus, while pulse sequencing of 60-80 pulses may be desired for achieving greater maximum capacities, pulse sequencing of 20-40 pulses may be desired if there is deemed an acceptable trade-off of lesser maximum capacities in favor of reduced processing times.

[0096] FIG. 9A shows tests results from a first full galvanostatic charge cycle of various DRX materials that were each processed via pulse-cycling, with each sample subjected to a pulse sequencing of 80 pulses in a voltage window of 3.0-4.5 V, at an elevated temperature of 50°C. However, each sample was subjected to pulse sequencing at different rates - with separate samples processed at 100 mA / g; 250 mA / g; 500 mA / g and 1 A / g. A control sample was also provided. The materials were cycled in a range from 2.0 to 4.8 V, at a rate of 20 mA / g and a temperature of 25°C. As the rate during pulsing is increased from 100 mA / g to 200 mA / g, 500 mA / g and 1 A / g, the extent of transformation and the delivered capacity decreases in the first full cycle. The extent to which the 4 V feature forms, and the flattening of the 3 V plateau feature area both decrease when higher rates are employed, as more clearly seen in the dQ / dV profiles displayed in 9B. FIG. 9C shows the initial cycling stability of the same samples. These results indicate that pulse sequencing at higher currents lead to similarly incomplete transformations as pulse sequencing with a lesser number of pulses. For example, a sample processed with a pulse sequencing having 10 pulses at 100 mA / g (FIG. 8A) was observed to yield a maximum capacity and plateau features similar to those observed from a sample processed with a pulse sequencing having 80 pulses at 500 mA / g (FIG. 9A).

[0097] While the capacity delivered during pulsing at 100 mA / g decreases slowly over the course of pulsing, the capacity delivered in a pulse decreases with increasing pulsing rate, as shown in FIGS. 10A and 10B. This could suggest that when high pulsing rates are employed, the utilization of material close to the surface of these particles is greater than the bulk, as is often seen in electrode materials at high rates. This is also supported by the greater retention of the sloped DRX-like feature centered at approximately 3.2 V in the discharge of materials pulsed at higher rates (FIG. 9B), relative to the material pulsed fewer times at a low rate (FIG. 8B). These findings suggest that there is a trade-off between increasing the rate of pulsing to save time, while also ensuring that all the material is being accessed to an extent sufficient to promote the transformation. The long-term cycling data of these high-rate pulsed materials is displayed in FIG. 9C. As with the material pulsed a different number of times at 100 mA / g, the materials pulsed at high rates retain a comparable or higher capacity than the control throughout cycling. Even the materials pulsed at 500 mA / g and 1 A / g deliver capacities of approximately 215 and approximately 195 mAh / g in subsequent cycling. This constitutes less than 24 and 12 hours of pulsing, representing a significant improvement over the greater than 20 days needed for conventional charge cycling at 20 mA / g. While not being bound by theory, it is expected that samples processed with pulse sequencing at higher rates over a greater number of pulses will have greater degree of 5 phase structure transformation at the surface, though a lesser degree of such transformation throughout the bulk of the material, whereas samples processed with pulse sequencing at lower rates over a fewer number of pulses will have a more homogenous though less complete 8 phase structure transformation throughout the bulk of the material.

[0098] Favorably, as seen in FIGS. 8C and 8, DRX materials processed by the inventive pulse-cycling have been observed in subsequent cycling to yield capacity profiles that reach a relative stability far quicker than DRX materials processed by conventional galvanostatic cycling (z.e., the “control”), which is instead observed to have an initially low capacity followed by a noticeable capacity increase before stabilizing. This is true of the pulse-cycling processes with pulse sequencing having a fewer number of pulses (FIG. 8C) and with pulse sequencing having higher rates (FIG. 9C). Whereas the control samples were observed to take around three weeks (at 20 mA / g) to fully transform the 8 phase structure, as characterized by a steady capacity level of greater than approximately 220 mAh / g, DRX materials processed via pulse-cycling were observed to produce 8 phase structures with similar capacity in as little as 1-2 days. It is also noteworthy that there was not observed any stability tradeoff in the DRX materials processed by pulse-cycling, as those materials retained capacities similar to or even greater than those of the conventionally processed control samples during subsequent cycling.

[0099] FIG. 11 shows the ex-situ XRD pattern of the pristine, ball-milled DRX material, along with DRX material after 5, 20 and 80 pulses. After just 5 pulses, significant and broad spinel-like features are visible in the pattern, as marked in FIG. 11 with an “S”, in addition to the peaks which are common between the DRX and spinel structures, as marked in FIG. 11 with an “D”. The S-labelled peaks that are not present for the rocksalt structure result from cell doubling, since the spinel unit cell is simply a 2 x 2 x 2 superstructure of rocksalt with a specific cation order. This also explains why only the spinel peaks are broad, as the spinel-like order can have shortrange coherence, without interrupting the long-range rocksalt lattice. After pulsing 20 or 80 times, these features become both narrower and more intense. Combined, these changes indicate that the disorder of the spinel-like environments continues to decrease, while the coherence length increases. This can be interpreted as a growth and continued ordering of spinel-like “domains” within the material, which decrease the broadness and increase the integrated intensity of these new features, respectively.

[0100] FIGS. 12A-12H display details of Rietveld refinement of XRD patterns on samples collected after 0, 5, 10, 20, 40, 60, 80, and 100 pulse cycles, respectively. It is notable that significant diffraction features of spinel-like order appear after just 5 pulses in FIG. 12B, as compared to the pristine material without any pulsing in FIG. 12A, and how the changes decelerate thereafter. FIG. 12G displays the XRD pattern of a “fully” transformed electrode after 80 pulses at 100 mA / g. Compared to the pristine material in FIG. 12A, this material has strong, broad features indicative of significant spinel -like order. The following Table 1 shows refined structural parameters for the DRX materials in FIGS. 12A-12H.

[0101] Previously, no acceptable refinement could be obtained for a two-phase model that contains a spinel with a short coherence length. A good fit could only be obtained with a single-phase model containing only a partially disordered spinel of short coherence length. With a coherence length of roughly 4 nm for the spinel-like domains, the best fit gives 16c / 16d disorder values as low as 2-3%. Higher amounts of 16c / 16d disorder are incompatible with the intensity of the spinel-like diffraction features. Significant inversion of the spinel, arising from 8a occupancy of Mn, is not likely as this would significantly increase the intensity of the (220) peak at approximately 12°, which is already well captured with a model possessing only 16c / 16d disorder. Because 16c / 16d disorder by itself leads only to reduced intensity of the spinel-like peaks, not broadening, disorder alone cannot explain the low coherence length of spinel-like order in the 5 phase. Since no significant remnant DRX phase is apparent in diffraction, the broadness of the spinel-like features can only be explained by periodic interruptions to the spinel-like order, either from disordered “walls” between domains, or antiphase boundaries between spinel domains of different rotational or translational symmetry.

[0102] In the samples addressed by FIGS. 12A-12H, as the number of pulses increases from 5 to 100 pulses, the 16c / 16d disorder rapidly falls while the coherence length of the ordering continues to increase. Between pulses 60 and 100, the changes in domain size slow and halt, in agreement with the lack of change in the voltage curve at this point. The apparent rapid formation of spinel-like features, combined with the slow increase in coherence length, suggests that spinel domains form throughout the material in earlier pulse cycles and coalesce to form larger domains over latter pulse cycles in the range of 60-80 pulses before the process stops. FIG. 13 illustrates SEND mapping performed on pulse sequenced solid-state synthesized DRX material after 80 pulse cycles, showing spatial distribution of the 8 phase obtained by virtual imaging with diffraction spots unique to the Fd3m space group, with a high degree of order distributed throughout the material, and no remnant disordered regions, in agreement with the XRD refinement.

[0103] FIGS. 14A-14F illustrate correlations between electrochemical change, domain growth, and ordering during 6 phase transformation. FIG. 14A shows peak differential capacity of 3 V plateau after pulsing vs. estimated domain size; FIG. 14B shows peak differential capacity of 3V plateau after pulsing vs. estimated 16c / 16d site disorder; FIG. 14C shows length of 4V plateau after pulsing vs. estimated domain size; FIG. 14D shows length of 4V plateau after pulsing vs. estimated 16c / 16d site disorder; FIG. 14E shows length of 4V plateau during pulsing vs. estimated domain size; and FIG. 14F shows length of 4V plateau during pulsing vs. estimated 16c / 16d site disorder.

[0104] The findings illustrated in FIGS. 14A-14F suggest the clearest structural relationship with evolution of the 8 phase transformation is the correlation between the peak differential capacity (dQ / dV) of the 3 V plateau-like feature on charge and the coherence length (which is taken as an estimate of the domain size), as illustrated in FIG. 14A. As the estimated domain size in these materials increases, the 3 V plateau-like feature seen in cycling becomes flatter, leading to a larger peak. A similar relationship is seen in FIG. 14C between the coherence length and the capacity delivered above 3.5 V, providing a simpler proxy for the extent of transformation.

[0105] An equivalent metric can be taken from the capacity of the 4 V region within a pulse itself (although the capacity is slightly lower than in cycling), suggesting that the point at which the transformation stops could be determined during pulse sequencing, without need of a normal charge cycle. Evolution of the voltage curve during pulsing, including the initial growth of the 4 V feature and the cessation of this growth, is shown in FIG. 15. In contrast to the coherence length, the refined 16c / 16d disorder falls rapidly to a level below 5% and does not appear to correlate with the changes in electrochemistry after about 40 pulses. Details of these refinements and the estimated coherence length and extent of 16c / 16d disorder are presented in Table 1 above.

[0106] Taking the differential capacity peak of the 3 V plateau as a proxy for the extent of transformation, we can compare the evolution of the control and pulse sequenced materials may be compared by taking the differential capacity peak of the 3 V plateau as a proxy for the extent of transformation. FIG. 16 plots peak differential capacity of control and pulse sequenced materials, along with the cumulative time taken in either charge cycling at 20 mA / g or pulse sequencing at 100 mA / g. As seen in this figure, 2 - 3 days of the inventive pulse sequencing (40 pulses) can reach an equivalent extent of transformation that would otherwise require 20 days of conventional charge cycling (control), and 7 days of inventive pulse sequencing (80 pulses) can reach an equivalent extent of transformation that would otherwise require over 50 days of conventional charge cycling (control). As the pulse sequencing shown in FIG. 16 is at only 100 mA / g, it is expected that 5 phase transformation may be further accelerated with optimization of the pulse sequencing process.

[0107] FIG. 17 shows a comparison between the electrochemistry of DRX materials after pulse sequencing different numbers of times at 100 mA / g and the conventionally charge cycled control cell after the same number of cycles at 20 mA / g. It is apparent that formation of the 4 V plateau feature occurs to a similar extent after an equivalent number of pulse cycles / charge cycles, though the 4 V plateau feature in the control stops growing after approximately 40 charge cycles, whereas the plateau feature in the pulse sequenced material continues to grow until approximately 60 pulse cycles.

[0108] This suggests the effectiveness of a single pulse cycle is roughly equivalent to that of a full charge cycle, supporting the notion that not all parts of the voltage window contribute equally to 5 phase transformation. However, the evolution of the 3 V plateau is quite different in pulse sequencing versus charge cycling. In the pulse sequenced material, the 3 V plateau becomes flatter and longer, but does not otherwise change significantly after additional pulse cycles. In contrast, the 3 V feature of the charge cycled material becomes shorter with additional charge cycles. This shortening of the 3 V plateau feature during charge cycling, while the 4V plateau feature continues to increase in capacity, leads the capacity of material to peak after 20-30 charge cycles, despite the 8 phase transformation not being complete at that point. While not bound by theory, it is expected this might be explained by material degradation or impedance growth which occurs at low voltage, but not in the voltage range used in pulse sequencing (3.0 - 4.5 V).

[0109] FIG. 18 shows results from further testing undertaken to determine if additional cycling might yield further effects on partially transformed DRX materials. These results show a first full galvanostatic charge cycle of three separate DRX materials that were each processed via pulse-cycling, with each sample initially subjected to pulse sequencing at a rate of 500 mA / g, in a voltage window of 3.0-4.5 V, for a duration of 6 hours, at an elevated temperature of 50°C. This initial pulse-cycling resulted in the partial transformation of 8 phase structures in each of the samples. One of these samples was then tested without any further processing, whereas the two other samples were subjected to further processing. A first of the further processed samples was subjected to additional pulse sequencing in a voltage window of 2.0-3.25 V for an additional 24 hours, and a second of the further processed samples was subjected to pulse sequencing in a voltage window of 3.75-4.8 V for an additional 24 hours. Subsequent testing of these three samples yielded highly similar outputs, suggesting that 3.0-4.5 V is an optimal voltage window for pulse-cycling of the DRX materials, with little to no effect on 5 phase structure transformation from additional pulse sequencing in other voltage ranges.

[0110] Additional testing was performed to study the partial transformation of 5 phase structures in large-particle DRX materials. Li1 1Mn08Ti01O19F01 was again chosen as an Mn-rich DRX material, though for these tests the DRX material was synthesized via a molten salt (MS) method in potassium chloride (KC1)6, using the same precursor as the earlier tests, to obtain a more homogenous morphology than would otherwise be possible from solid-state synthesis. This DRX material was not subjected to any particle-size-reduction processing (e.g., no ball-milling), resulting in particles with a typical minor axis of 2-5 microns, as shown by a SEM image of the resulting particles in FIG. 19.

[0111] FIG. 20 shows test results from a first full galvanostatic charge cycle of the large-particle DRX materials after processing via pulse-cycling in which both samples were subjected to pulse sequencing of 80 pulses in a voltage window of 3.0-4.5 V, at an elevated temperature of 50°C. However, to compensate for the larger particle size, the pulse sequencing of these materials was performed with lower current density rates. A first of the tested samples was pulsed at a rate of 50 mA / g, and a second of the tested samples pulsed at a rate of 20 mA / g. The results from this testing show that there was not any transformation in the sample pulsed at 50 mA / g; whereas transformation was found in the sample pulsed at 20 mA / g, though processing of this transformed sample took around 4 weeks.

[0112] FIG. 21 shows the results from additional testing conducted to assess whether the increasing rate capability of the DRX material during transformation of the 8 phase structure would enable the materials to efficiently equilibrate at 3.0 V or 4.5 V if using a pulse-cycling with constant voltages up or down to a certain current rate (as opposed to constant current pulses). These large-particle samples were tested with pulse-cycling using constant voltage pulses at 3.0V / 4.5 V, with holds to + / - 10 mA / g. These tests show that the materials were fully (or nearly fully) transformed, with specific capacities of approximately 225 mAh / g when cycling at 50°C, and 175 mAh / g when cycling at 25°C, after less than 2 weeks of processing.

[0113] FIG. 22 provides comparative results of three samples processed via pulse-cycling and one control (large particle) sample processed via conventional charge-cycling. A first of the pulse-cycled samples was prepared as a small-particle material (with ball-milling) and processed with a pulse sequencing of 80 pulses, in a voltage window of 3.0-4.5 V, at a rate of 100 mA / g and a temperature of 50°C; and two others of the pulse-cycled samples were prepared as large-particle samples (without ball-milling) and processed with pulse sequencing using constant voltage pulses at 3.0 / 4.5 V, with holds to + / - 10 mA / g, also at a temperature of 50°C.

[0114] The samples were then tested with subsequent cycling at 25°C, with the exception that one of the large-particle samples was cycled at 50°C. The control sample was observed to have a relatively low capacity, without ever sufficiently transforming to achieve a useful capacity. Meanwhile, the samples prepared according to the pulse-cycling processes displayed relatively high capacities with considerable stability. Surprisingly, the large-particle sample cycled at 50°C displayed substantially similar properties to the small-particle sample cycled at 25°C, which is atypical of battery cycling at elevated temperatures. This suggests that large-particle DRX / 8 phase materials may be made stable at the elevated temperatures when processed via pulse-cycling, which would be of considerable importance as the availability of large-particle materials would significantly simplify electrode engineering.

[0115] Further testing was performed to study 8 phase transformation in large-particle DRX materials by again synthesizing Li1.1Mn0.8Ti0.1O1.9F01 as an Mn-rich DRX material via a molten salt method in which the synthesis temperature was increased to 1100°C and the holding time was reduced to 20 minutes, using a relatively high ratio of 3: 1 KC1 to precursor, while using the same solid-state precursors (Li2CO3 and transition metal oxides) typically employed for the synthesis of DRX. This yielded spheroidal single crystal Li1.1Mn0.8Ti01O1.9F01 DRX particles of 2-3 pm diameters, as shown in the SEM image in FIG. 23. Almost no particles with a diameter of less than 1 micron are observed. FIG. 24 shows TEM-EDS imaging of these particles, confirming that Mn, Ti, and F are uniformly distributed in the material, and FIG. 25 shows XRD imaging confirming that the product was a phase- pure DRX.

[0116] With the longer diffusion length of these molten salt synthesized materials, at approximately greater than 10 times larger radius, it was found that the capacity of the material in each pulse is much lower than for the solid-state material if a current of 100 mA / g is employed. This indicates that not all the material can be accessed in 2-3 pm particles. When a lower current of 20 mA / g is applied, more capacity is delivered in each pulse, at the expense of the pulsing taking many weeks, and the material achieving a capacity of 184 mAh / g after pulsing, as seen in FIG. 26.

[0117] To decrease processing time of these larger particles, the pulse sequencing process was modified to use pulse cycles with constant voltage holds until a lower limit of current is reached. The modified pulse sequencing process began with an initial charge at 10 mA / g to 4.8 V and was followed by pulse sequencing over 80 pulse cycles with constant voltage holds at 3.0 V and 4.5 V to 10 mA / g. This modified pulse sequencing process allows the initial pulses to proceed more slowly, with additional time for cycling of Li+ in the initial DRX material. However, as the transformation proceeds and the diffusivity of the material improves, higher rates may be expected to persist through more of the constant voltage pulse, leading to high average current densities and an increased average current density in later cycles. FIG. 27 A shows a first charge of molten salt material electrode at 50°C and 10 mA / g, with a decrease in voltage at roughly 3.5 V attributed to a large overpotential at low charge in the large particle DRX. FIG. 27B shows voltage and current profile of constant voltage cycling from 10 A / g down to 10 mA / g. FIG. 27C shows the duration of each pulse cycle over the course of 80 pulse cycles. FIG. 27D shows specific capacity of each charge and discharge in each of the pulse cycles. Higher charge capacities attributed to electrolyte oxidative events, which temporarily approached or exceeded 10 mA / g during a voltage hold at 4.5 V and 50°C. The modified pulse sequencing process took approximately nine days, with atypical pulse cycle of charge and discharge voltage holds lasting for roughly 3 hours.

[0118] The DRX material produced from the modified pulse sequencing process using 80 pulse cycles with constant voltage holds was subjected to subsequent cycling performance testing. FIG. 28A shows the 1st, 50thand 100thfull charge and discharge cycles of these tests. As seen by the clearly formed 3 V and 4 V plateau features, the modified pulse sequencing process successfully transformed the DRX material to the 8 phase, providing an initial discharge capacity of 174 mAh / g. The resulting voltage curve of this material is quite similar to that of the smaller particle, ball-milled, solid-state material, with two important differences. First, in the molten state synthesized material, the 3 V and 4V plateau features appear sharper than the smooth features present in the solid-state material. Second, the lower capacity of the large particle material is mostly due to a lower capacity at the end of the 3 V plateau feature, while the 4 V plateau feature has similar capacity to that seen in the ball-milled material.

[0119] Stability of the large particle material was further tested by cycling performed on a DRX material produced from the modified pulse sequencing process using 80 pulse cycles with constant voltage holds while the material remained at 50°C after pulsing was complete. As shown in FIG. 28B, the material provided 225 mAh / g and 700 Wh / kg in the 1stcycle at 50°C, with slight decreased in the 50thand 100thcycles. This demonstrates that a large capacity remains accessible in the large particle material, even if some capacity is inaccessible at room temperature due to kinetic limitations.

[0120] FIG. 28C shows extended cycling of the large particle material at 25°C yielding some initial capacity gain, after which the capacity remains remarkably stable. Over the course of cycling, the capacity of the material increases from 174 mAh / g to over 190 mAh / g before stabilizing, yielding over 600 Wh / kg. From cycles 50 to 100, the capacity stays almost constant, with no significant voltage fade or other changes to the voltage curve. The capacity retention at 50°C is also quite good, with 94.7% and 90.3% of the initial capacity retained at the 50thand 100thcycles, respectively. To confirm that the 6 phase formed here is indeed equivalent to that formed in the ball-milled solid-state material, additional synchrotron XRD imaging was performed on molten salt material recovered after the constant voltage pulse sequencing process. The refinement of these XRD patterns, as seen in FIG. 29, confirms that this molten salt synthesized material, like the solid-state synthesized material, possesses a structure with an estimated average spinel-like domain size of approximately 5 nm and a high degree of order (e.g., less than 5% 16c / 16d disorder). The additional peak at roughly 10° in these XRD patterns is attributed to the diffuse diffraction of carbon in the electrode.

[0121] FIG. 30A shows a low magnification HAADF-STEM image of a typical large molten salt particle of approximately 2 pm after pulsing. The mean of the SEND diffraction patterns collected across the solid line box marked in FIG. 30A is shown in FIG. 30B, in which the sharp diffraction spots indicate a high degree of crystallinity of the particle post-pulse-cycling and suggest that this particle remains single-crystalline. Diffraction patterns from selected points are indexed based on the structural information obtained from the synchrotron XRD refinement of the same samples (FIG. 29) and are provided in FIG. 31. FIG. 30C maps the extent of 5 phase formation by integrating the diffraction peaks unique to the 5 phase from each of the SEND patterns and normalizing the obtained virtual image across the full acquired scattering range (from 0.175 A1to 1.2 A1) to account for the thickness. Radial integration ranges are shown in FIG. 32. It can be seen that there is a uniform transformation from the DRX to the 6-phase, with no substantial amount of retained disordered regions. An atomic-resolution HAADF-STEM micrograph along the

[0110] zone axis from the region marked by the solid line box in FIG. 30A was collected to investigate cation ordering in the 6-phase. FIG. 30D shows a low pass Butterworth filtered HAADF-STEM micrograph, with a raw image thereof provided in FIG. 33. The fast Fourier transform of this HAADF-STEM micrograph is shown in the inset image in the bottom right comer of FIG. 30D, with the frequency components (Fl) and (F2) belonging to the lower symmetry spinel-like (Fd3m) lattice marked. Filtering the marked frequency components (Fl) and (F2) obtains the lattice fringes shown in FIG. 30E and FIG. 30F, respectively. The lattice fringes from the parent DRX phase (Fm3m) are obtained by filtering the frequency components marked by the unlabeled circles in the inset of FIG. 30D. This parent lattice is shown as an underlaid diagonal pattern in FIGS. 30E and 30F. The spinel lattice fringes in FIG. 30E show multiple half-lattice vector translations, which are representative of antiphase boundaries. One such boundary is marked as by a dashed line box in FIGS. 30D-30F. These boundaries are formed when one variant of the spinel ordering impinges on another variant, which is distinct by one of the rotational or translational symmetry operations of the parent rocksalt that are lost in the spinel symmetry. The domain size obtained by counting the average number of the lattice fringes between the antiphase boundaries is estimated to be in the range of 3 to 10 nm.

[0122] The microstructure of a DRX 8 phase is derived from the manner by which it is formed from the parent DRX material. The common oxygen sublattice and higher symmetry of a rocksalt allows for multiple possible spinel variants, consistent with the norm of the quotient of the rocksalt and spinel space group. These variants differ by their occupancy of the 16c / 16d Wyckoff sites on the Fd3m lattice. It has been observed that the 8 phase consists of nanometer-sized domains of these variants separated by anti-phase boundaries, with the domains achieving a high degree of order, without any significant remnant disordered regions, as seen in FIG. 1. Without being bound by theory, it is considered likely that the formation of 8 phase starts in multiple different places in a rocksalt material, independent of neighboring nuclei, and that these nuclei grow to impinge on each other.

[0123] In the tests addressed herein, the DRX sample was observed to have completely transformed after pulse sequencing, with the resulting 8 phase microstructure and electrochemical voltage profile remaining remarkably stable. The almost complete extent of 8 phase formation after pulsing is evident in the SEND images seen in FIG. 13 data shown in Supplementary Fig. S7 for a solid-state synthesized DRX material and in FIG. 30C for a molten salt synthesized DRX material. The material possesses no obvious large length scale (> 1 nm) disordered regions, as seen by the predominance of lattice fringes corresponding to the spinel phase (Fd3m space group) in nearly all regions of the HAADF-STEM micrograph images in FIGS. 30E and 30F.

[0124] Some remnant disorder is seen in the boundaries of these well-ordered spinel domains, consistent with the small amount of 16c / 16d site disorder required for a good fit of the XRD patterns, but the individual domains appear to be well-ordered. While the mean diffraction pattern from the SEND image shows sharp peaks indicative that a high degree of crystallinity is retained after 8 phase has formed (FIG. 30B), the XRD peaks unique to the Fd3m phase (odd 1 index) show selective broadening, indicating that the ordering that introduces these reflections has a small domain size. Estimating the crystallite size from the broadness of these odd I-indexed peaks with the Scherrer equation gives an approximately 4 nm domain size, consistent with the spinel domains seen in the Fourier filtered atomic-resolution STEM-HAADF imaging as shown in FIGS. 30D and 30E. Due to the presence of domain boundaries and the overlap of different rotational or translational domains, it may prove difficult in discerning the extent of (dis)ordering in a 6 phase structure through a visual inspection of a STEM micrograph. However, if different phases are present, such as DRX, spinel, and layered, it may be possible to clearly identify these regions in the local diffraction patterns in SEND imaging. These phases may be distinguished from the overlapping spinel domains through tandem use of SEND and HAADF-STEM measurements. While not being bound by theory, it is expected that the 5 phase comprises highly ordered spinel domains separated by thin antiphase boundary regions measuring approximately 1 nm), with no evidence of a significant fraction of other phases present.

[0125] The efficacy of pulse sequencing in a limited voltage regime to form 5 phase is consistent with the kinetics of Mn migration and thermodynamics of transformation of spinel in Li-TM-0 (Li-Mn-O) oxide and oxyfluoride systems. Mn migration is most favorable when Mn is at a low valence and adjacent to tetrahedral sites which do not share any face with octahedra occupied by another transition metal. Such a configuration has been referred to as a 1-TM (tetrahedral) site. In a Mn-rich DRX with compositions as addressed herein, a given tetrahedral site has a 26.6% chance of being adjacent to 1 Mn and 3 Li (a ”1-Mn” site), assuming a random cation configuration. Given this high chance of a tetrahedron being a 1-Mn site, and the presence of 8 tetrahedra around each Mn, a very large number of potential hops should be available for Mn during the first charge of the DRX structure.

[0126] This may explain why a significant degree of spinel-like order is already observed as early as the 5thpulse cycle in pulse sequencing of a DRX material. As the transformation proceeds, the number of Mn experiencing favorable conditions for migration would be expected to decrease, as the formation of spinel -like order locks Mn ions into a configuration without such 1-TM tetrahedron (ordered spinel LiMmCU has only 0, 2, and 4-TM sites). Given the decreasing number of opportunities to hop, it is expected that the growth of 8 phase will decelerate dramatically, with substantially complete transformation of the 8 phase after approximately 60-80 pulse cycles.

[0127] A preferred voltage window for effectively forming 8-DRX may be based on the combined need for low valent Mn and Li vacancies. To allow a low-energy hop for a Mn adjacent to a 1-Mn site, the three other octahedral sites which face share with this tetrahedron should be vacated of Li+ upon charging, with the Mn valence remaining below 4+ as Mn4+ shows very little mobility. Upon discharge, the octahedral sites are refilled with Li to destabilize the tetrahedral occupancy of the Mn and push it into other adjacent octahedral sites (or possibly back to the original), creating a spinel -like configuration of octahedral Mn. These conditions require the material to visit a range of Li contents that [1] vacate most octahedra; and [2] promote Li to re-occupy these octahedra (< 3.5 V).

[0128] These conditions may explain observations from the pulse sequencing process addressed herein. For example, the similarity of DRX materials pulsed from 3.0-4.5 V and 3.0-4.8 V makes sense given that the average Mn valence is already nearly 4+ at 4.5 V, so there are few Mn3+to hop into the vacancies created by the last stage of charge. This may likewise explain why discharge cutoffs at low voltage, when relatively few octahedral vacancies exist, do not lead to transformation. It may also be noted that the stoichiometry of the material in the effective voltage range is similar to that of a spinel Li1.1-xMn0.8Ti0.1O1.9F01, where 0.5 < x < 0.8. Thus, without being bound by theory, it is possible that this region is effective because it combines a strong driving force towards spinel with favorable kinetic conditions for Mn migration.

[0129] Material Synthesis. Solid-state materials used in tests addressed herein were synthesized using a traditional solid-state method. Li2COo (Sigma Aldrich, 99.9%), Mn20s (Alfa Aesar, 99%), TiO2 (Sigma Aldrich, 99.9%), and LiF (Alfa Aesar, 99.9%) were used as precursors. All precursors were mixed in stoichiometric portions with ethanol in a Retsch PM 200 planetary ball-mill at a rate of 250 rpm for 12h. A 10% excess of Li2COs was added to compensate for possible loss during synthesis at 1100°C. The precursors were dried in an oven at 70°C overnight before being pelletized. In the case of molten salt synthesis, the same precursor was mixed with KC1 (Sigma Aldrich, 99%) by hand for 10 minutes in a 1 :3 Precursor :KC1 weight ratio before being pelletized. In both the case of molten salt and solid-state synthesis, pressed pellets were heated to 1100°C at a rate of 10°C min1under argon flow in a tube furnace and held at this temperature for 20 minutes before natural cooling to room temperature. For the molten salt material, the heated pellet was washed directly with deionized water twice, and ethanol once, before drying at 70°C overnight under vacuum. For solid-state material, the produced pellets were ground by hand prior to ball-milling with carbon. All powders were transferred to and stored in an argon filled glovebox. SEM images of as-synthesized and ball-milled powders were collected using a Zeiss Gemini Ultra-55 Analytical Field Emission SEM. SEM images are analyzed to measure the dimensions of the particles. The “major axis” is defined as the longest dimension of the particle in the image, while the “minor axis” is the longest dimension perpendicular to the major axis. The particle size P as determined on SEM images is the area-equivalent diameter DA, also termed circular-equivalent diameter, which is the diameter of a sphere having the same projected area as the particle’s projection as seen on an SEM image. Electrochemistry. Cathode films used in tests addressed herein were composed of active materials, Super C65 (Timcal), and polytetrafluoroethylene (PTFE, DuPont, Teflon 8A) in a weight ratio of 70:20: 10. Prior to cathode fabrication, the solid-state material was ball-milled with the Super C65 for 1 hour under argon atmosphere using a SPEX 800M mixer / mill. The molten salt material was mixed with Super C65 by hand in a mortar and pestle for 20 minutes. Both the hand-mixed molten salt and ball-milled solid-state mixtures with carbon were mixed manually with PTFE in a mortar and pestle. The mixture was then rolled into thin films inside a glovebox. The active material loading density of all films was approximately 3-4 mg / cm2. Commercial 1 M LiPFe in ethylene carbonate (EC) and dimethyl carbonate (DMC) solution (1 : 1 volume ratio) was used as the electrolyte. A glass microfiber filter (Whatman) was used as the separator. FMC Li metal foil was used as the anode. Coin cells were assembled inside the glovebox and tested on an Arbin battery test instrument. The specific capacities were calculated based on the weight of active materials (70%) in the cathode films. Pulsing was conducted in multi-zone temperature chambers connected to the same Arbin test instruments, which were set to either 25°C or 50°C, and cycled by combining typical galvanostatic or potentiostatic program steps. Molten salt synthesized samples were pulsed using an initial constant current step at + / - 10 A / g until the cutoff of either 4.5 V or 3.0 V was reached, and then a constant voltage hold was employed until the current dropped to + / - 10 mA / g. After pulsing, cells were directly switched to cycling at 25°C. For the galvanostatic intermittent titration (GITT) measurements, each step in the voltage profiles corresponded to a galvanostatic charge / discharge of 10 mAh / g at a rate of 20 mA / g followed by a 6 h relaxation step.

[0130] Ex Situ Synchrotron X-ray diffraction (XRD) and X-ray absorption spectroscopy (XAS). Synchrotron XRD and XAS patterns were measured at Beamline 7-BM of the National Synchrotron Light Source II (NSLS-II). In order to avoid the diffuse diffraction feature of PTFE coincident with the (111) peak of spinel-like order, ex situ cathode films for XRD were prepared by mixing the active materials, Super C65 and polyvinylidene fluoride (PVDF) in at a weight ratio of 7:2: 1 in NMP and casting onto aluminum. The resulting electrode film was dried overnight before assembly into cells to perform pulsing / cycling. After pul sing / cy cling, cells used to produce ex situ samples were held at a desired voltage for 6 h. The cells were then disassembled and the cathode film was washed with diethyl carbonate (DEC). Cycled PVDF films for XRD were peeled off the aluminum to allow packing into Kapton capillaries. PTFE films used for XAS were sealed in Kapton tape for transport and measurement. All synchrotron XRD refinements were carried out using the TOPAS software package. A single-phase spinel structure model was used when refining the XRD patterns of cycled and pulsed electrodes. The reflections generated by the spinel -like order, resulting from those with an odd I-index, were selectively broadened as previously reported for 8 and y-LiFeCE type ordering. The broadening factor used on the spinel-like peaks was used in the Scherrer equation to estimate the coherence length of the domains. Ex situ XAS measurements at the Mn and Ti K-edge were performed in a transmission mode on films identical to those used in electrochemical measurements. The energy calibration was accomplished by simultaneously measuring the spectra of Mn and Ti metal foil. Data processing was carried out with the Athena software package.

[0131] Transmission electron microscopy sample preparation. To prepare ex situ samples for SEND and HAADF-STEM characterization, intact electrodes identical to those used for electrochemical tests were collected after pulsing and washed with diethyl carbonate (DEC) and transferred under argon. A thin lamella suitable for transmission electron microscopy (TEM) imaging was prepared using the dual-beam focused ion beam (FIB) method, utilizing the Helios G4 UX instrument, housed at the National Center for Electron Microscopy (NCEM), Berkeley. To protect the sample, 100 nm thick Pt was deposited on the surface using a 5 kV electron beam and a beam current of 1.6 nA before lift out. Subsequently, 1.5 pm Pt was deposited on the top of the electron beam deposited Pt using a Ga-ion beam at 30 kV and 0.26 nA. After a lamella was lifted out from the cathode film, it was welded to a FIB grid using Pt. The lamella was thinned down to approximately 1 pm using a Ga-ion beam with 30 kV accelerating voltage and 0.75 nA current. To prevent beam damage subsequent thinning to approximately 100 nm was carried out using a 5 kV accelerating voltage and 0.15 nA current. A final thinning step was carried out using a 2 kV ion beam. To remove the amorphous layer and thin further to achieve regions with thickness of approximately 50 nm, a low energy Ar-ion beam at 900 eV was used for 8 mins in a nano-mill.

[0132] Scanning electron nano diffraction (SEND). The SEND patterns were collected using a ThemlS 60-300 microscope equipped with a Gatan K2 detector housed at NCEM, Berkeley. The indicated semi-convergence angle was 0.62 mrad. The SEND patterns were analyzed using the py4DSTEM package version 0.14.8. To reconstruct a virtual image for the spatial extent of the spinel phase, only the peaks unique to the spinel phase were integrated. These radial ranges are marked in FIG. 32. The scattering angles for these peaks are (0.175, 0.250) A'1, and (0.59, 0.67) A’1. The integrated virtual image was normalized for thickness by integrating the region between the scattering angles from 0.175 A'1to 1.2 A’1.

[0133] Atomic resolved HAADF-STEM. The atomic resolution HAADF-STEM imaging was performed using the TEAM I microscope (double aberration-corrected Thermo Fisher Scientific Titan 80-300 kV) housed at NCEM, Berkeley. The microscope was operated at 300 kV with an indicated convergence angle of 30 mrad. The particle was tilted to

[0110] zone axis before atomic resolved imaging. The raw HAADF image is shown in FIG. 33. An adaptive local thresholding of 51 pixels is performed in MATLAB to remove any non-uniform illumination in the image. To remove high frequency noise a third order low pass Butterworth filter with a frequency cut-off of 160 pixels is applied in the frequency domain. The filtered HAADF-STEM image is shown in FIG. 30D. The Bragg filtering was done to identify the spinel lattice by filtering for the two spinel peaks in the FFT of the HAADF-STEM micrograph, shown in the inset of FIG. 30D. The HAADF-STEM image is first windowed using a Tuckey (or tapered cosine) window. The application of this window reduces the spectral leakage and edge artifacts thereby enhancing the accuracy of subsequent frequency based analysis. Following this, filtering masks are applied to the spinel frequency components in the Fourier space and inverse Fourier Transformed to obtain the spinel lattices in real space. Only half of the spinel peaks are considered for the filtering since the conjugate points contain the same information in a Fourier transform of a real-valued image through Hermitian symmetry. The spinel lattices are shown in FIGS. 30E and 30F based on the spinel peaks marked in FIG. 30D, with the filtering of the spinel peak marked as Fl shown in FIG. 30E and filtering of the spinel peak marked as F2 shown in FIG. 30F. Following a similar approach, the parent DRX lattice was overlaid on the obtained spinel lattice. The frequency components corresponding to the respective DRX lattice in the Fourier space are marked as unlabeled circles in FIG. 30D. These mathematical filtering operations were achieved using an in-house MATLAB script.

[0134] Energy Dispersive X-ray Spectroscopy. Energy-dispersive X-ray (EDX) spectroscopy elemental mappings were obtained using an FEI TitanX 60-300 transmission electron microscope housed at the NCEM facility, Berkeley. The spectral maps were collected at 300 kV accelerating voltage, Bruker-FEI Super-X four quadrant detector. The different elemental maps were drawn using Mn-ka, Ti-ka, O-ka, and F-ka peaks. The EDS data was analyzed using the Bruker esprit software.

[0135] The present invention provides methods of electrochemical pulse-cycling for in-situ formation of 8 phase structures in Mn-rich DRX. Whereas conventional processes for formation of a 6 phase in DRX materials are impractically long for commercial purposes, methods according to the present invention sufficiently accelerate transformation of the 8 phase to render these materials viable for commercial use.

[0136] Advantageously, the inventive process is also applicable for activating micron-sized single crystals, thereby obviating the need for ball-milling the DRX material with carbon. Such large particle, low surface area materials show near perfect cycling. During pulsing, no capacity or voltage fade is apparent, meaning that the transformation can be promoted without compromising later cycling. Once the 5 phase transformation is complete, minimal structural change occurs, in contrast to the voltage fade observed in Li and Mn-rich layered cathodes. The activation of large single crystals, and lack of degradation, render 8-DRX materials commercially viability as an earth-abundant and inexpensive cathode material.

[0137] The use of as-synthesized micron-sized materials, with a corresponding decrease in the amount of carbon (which is typically introduced during ball-milling of the material) may decrease the cost of composite cathode production for DRX materials. Improved cycling stability of these large particles may also considerably decrease the amount of specialized particle and electrolyte engineering required to attain long cycle life. This may also enable the use of voltage windows that forgo lower cutoff voltages below 2 V, which may eliminate a low voltage tail. These materials may further enable production of thermally safe and stable Mn-based DRX material cathodes with approximately 200 mAh / g and greater 600 Wh / kg at reasonable rates, with charge / discharging cycling beyond 100 cycles.

[0138] Although the present invention is described with reference to particular embodiments, it will be understood to those skilled in the art that the foregoing disclosure addresses exemplary embodiments only; that the scope of the invention is not limited to the disclosed embodiments; and that the scope of the invention may encompass any combination of the disclosed embodiments, in whole or in part, as well as additional embodiments embracing various changes and modifications relative to the examples disclosed herein without departing from the scope of the invention as defined in the appended claims and equivalents thereto.

[0139] To the extent necessary to understand or complete the disclosure of the present invention, all publications, patents, and patent applications mentioned herein are expressly incorporated by reference herein to the same extent as though each were individually so incorporated. No license, express or implied, is granted to any patent incorporated herein.

[0140] The present invention is not limited to the exemplary embodiments illustrated herein, but is instead characterized by the appended claims, which in no way limit the scope of the disclosure.

[0141] Citations

[0142] 1. Clement, R. J.; Lun, Z.; Ceder, G. Cation-Disordered Rocksalt Transition Metal Oxides and Oxyfluorides for High Energy Lithium-Ion Cathodes. Energ. Environ. Sci 2020, 13 (2), 345- 373. https: / / dot. org / 10.1039 / c9ee02803j. Li, L.; Ahn, J.; Yue, Y.; Tong, W.; Chen, G.; Wang, C. Fluorination-Enhanced Surface Stability of Disordered Rocksalt Cathodes. Adv. Mater. 2022, 34 (12), 2106256. https: / / dot. org 10.1002 / cidma.202106256. Ahn, J.; Ha, Y.; Satish, R.; Giovine, R.; Li, L.; Liu, J.; Wang, C.; Clement, R. J.; Kostecki, R.; Yang, W.; Chen, G. Exceptional Cycling Performance Enabled by Local Structural

[0143] Rearrangements in Disordered Rocksalt Cathodes. Adv Energy Mater 2022, 2200426. https : / / doi. org / 10.1002, cienm.202200426. Chen, T.; Yang, J.; Barroso-Luque, L.; Ceder, G. Removing the Two-Phase Transition in Spinel LiMn2O4 through Cation Disorder. Acs Energy Lett 2023, 8 (1), 314-319. https: / / doi. org / 10.1021, acsenergylett.2c02141. Reed, J.; Ceder, G. Role of Electronic Structure in the Susceptibility of Metastable Transition- Metal Oxide Structures to Transformation. Chem Rev 2004, 104 (10), 4513-4534. https: / / doi.org / ! 0.1021 / cr020733x. Chen, D.; Wu, J.; Papp, J. K.; McCloskey, B. D.; Yang, W .; Chen, G. Role of Redox-Inactive Transition-Metals in the Behavior of Cation-Disordered Rocksalt Cathodes. Small 2020, 16

[0144] (22), 2000656. https: / / doi.org / 10.1002 / smll.202000656.

Claims

CLAIMSWhat is claimed is:

1. A cathode active material, comprising: an Li-TM-0 compound comprising a particle distribution with particle sizes (P) in a range of 100 nm < P < 10 pm, where TM is one or more transition metals; wherein particles in the particle distribution have a crystal structure comprising a disordered rocksalt (DRX) lattice with a plurality of variant 5-phase domains of spinel-like lattice formed therein, each 5-phase domain differing from one another, with adjacent 5-phase domains separated by an anti-phase boundary.

2. The cathode active material according to claim 1, wherein the anti-phase boundaries measuring approximately 1 nm.

3. The cathode active material according to any one of claims 1-2, wherein each 5-phase domain measures in a range from approximately 3 nm to approximately 10 nm.

4. The cathode active material according to any one of claims 1-3, wherein the 5-phase domains differ from one another based on at least one difference chosen from: lattice orientation; occupancy of the 16c / 16d Wyckoff sites; and coherence length.

5. The cathode active material according to any one of claims 1-4, wherein the 5-phase domains differ from one another based on at least one difference in lattice orientation chosen from: a translational difference in lattice orientation, a rotational difference in lattice orientation, and a combination thereof.

6. The cathode active material according to any one of claims 1-5, wherein each 5-phase domain has a 16c / 16d site disorder percentage of approximately 10.2% to approximately 3.6%; optionally 10.2% to approximately 3.6%.

7. The cathode active material according to any one of claims 1-6, whereineach 8-phase domain has a coherence length (L) of L < 10 nm; optionally L < 5 nm; optionally L < 4 nm; and optionally in a range of 2.4 nm < L < 3.9 nm.

8. The cathode active material according to any one of claims 1-7, wherein the crystal structure comprises at least four variant 5-phase domains, optionally eight variant 8-phase domains.

9. The cathode active material according to any one of claims 1-8, wherein the particle distribution comprises particle sizes (P) of 100 nm < P < 1 pm; optionally, 100 nm < P < 500 nm.

10. The cathode active material according to any one of claims 1-8, wherein the particle distribution comprises particle sizes (P) of 1 pm < P < 10 pm; optionally, 1 pm < P < 5 pm; optionally, 2 pm < P < 3 pm.

11. The cathode active material according to any one of claims 1-10, wherein the transition metal (TM) comprises Mn, alone or in combination with one or more other transition metals.

12. The cathode active material according to any one of claims 1-11, wherein the transition metal (TM) comprises Mn an amount 0.6 < Mn < 0.7.

13. The cathode active material according to any one of claims 1-12, wherein the cathode active material is formed from a DRX material that was previously processed to form the 8-phase domains through electrochemical pulse-sequencing.

14. The cathode active material according to claim 13, wherein the DRX material is obtained through prior-synthesis comprising one of a solid-state synthesis process; a molten-state synthesis process.

15. The cathode active material according to claim 13, whereinthe as-synthesized DRX material is, optionally: subjected to particle-size-reduction processing prior to electrochemical pulse-sequencing; or transformed via electrochemical pulsesequencing without prior particle-size-reduction processing.

16. A method of processing a DRX material for use in a cathode active material according to any one of claims 1-15, comprising: obtaining a prior- synthesized DRX material; heating the DRX material to a predetermined temperature; and electrochemically pulsing the DRX material over a predetermined voltage window, at a predetermined current density rate, for a predetermined duration.

17. The method of claim 16, wherein the DRX material is a prior-synthesized DRX material.

18. The method according to claim 16 or 17, wherein the DRX material is a Li-excess material.

19. The method according to any one of claims 16-18, wherein the DRX material has a general formula Lii+xTMi-xCL-yFy, where TM comprises Mn.

20. The method according to any one of claims 16-19, wherein the DRX material is Mn-rich.

21. The method according to any one of claims 16-20, wherein the DRX material comprises particles having a minor axis of less than 500 nanometers.

22. The method according to any one of claims 16-20, wherein the DRX material comprises particles having a minor axis of in a range of about 500 nanometers to about 5 microns.

23. The method according to any one of claims 16-22, whereinthe predetermined temperature is approximately that temperature needed for formation cycling of the DRX material.

24. The method according to any one of claims 16-23, wherein the DRX material is a Li-excess material, and the predetermined temperature is an elevated temperature that increases Li diffusivity of the DRX material.

25. The method according to any one of claims 16-24, wherein the predetermined temperature is in a range of about 25°C to about 50°C.

26. The method according to any one of claims 16-25, wherein the predetermined temperature is approximately 50°C.

27. The method according to any one of claims 16-26, wherein the voltage window ranges from about 1.5 V to about 4.8 V.

28. The method according to any one of claims 16-27, wherein the voltage window ranges from about 2.5 V to about 4.5 V.

29. The method according to any one of claims 16-28, wherein the voltage window ranges from about 3.0 V to about 4.5 V.

30. The method according to any one of claims 16-29, wherein the current density rate is within a range from about 100 mA / g to about 1000 mA / g.

31. The method according to any one of claims 16-30, wherein the current density rate is within a range from about 100 mA / g to about 500 mA / g.

32. The method according to any one of claims 16-31, wherein the current density rate is within a range from about 250 mA / g to about 500 mA / g.

33. The method according to any one of claims 16-32, whereinthe predetermined duration is a predetermined period of time.

34. The method of claim 33, wherein the period of time is from about 1 day to about 5 days.

35. The method of claim 34, wherein the period of time is from about 1 day to about 2 days.

36. The method of claim 34, wherein the period of time is approximately 5 days.

37. The method according to any one of claims 16-32, wherein the predetermined duration is a predetermined number of pulses.

38. The method of claim 37, wherein the number of pulses is from 10 to 100.

39. The method of claim 38, wherein the number of pulses is from 20 to 40.

40. The method of claim 38, wherein the number of pulses is from 60 to 80.

41. The method according to any one of claims 16-40, further comprising: after heating the DRX material to the predetermined temperature, and prior to electrochemically pulsing the DRX material, charging the DRX material to an initial standard charge.

42. The method of claim 41, wherein the initial standard charge exceeds a maximum charge of the voltage window.

43. The method of claim 41, whereinthe initial standard charge is approximately 4.8 V.

44. The method according to any one of claims 16-43, further comprising: upon cessation of the predetermined duration for the electrochemical pulsing, executing at least one galvanostatic charge cycle of the DRX material.

45. The method of claim 44, wherein the galvanostatic charge cycle comprises charging the DRX material to a charge exceeding a maximum of the voltage window and discharging the DRX material to a charge below a minimum of the voltage window.

46. The method of claim 44, wherein the galvanostatic charge cycle comprises charging the DRX material to approximately 4.8 V and discharging the DRX material to approximately 2.0 V.

47. A DRX material processed according to the method of claim 16.

48. The DRX material of claim 47, wherein the DRX material comprises particles with a minor axis of less than about 500 nm.

49. The DRX material of claim 47, wherein the DRX material comprises particles with a minor axis in a range of about 500 nm to about 5 microns.

Images