STABLE LITHIUM FLUORIDE-BASED CATHODES FOR METAL AND METAL-ION BATTERIES
20200373580 ยท 2020-11-26
Inventors
- Gleb Yushin (Atlanta, GA)
- Eugene Berdichevsky (Oakland, CA)
- Bogdan Zdyrko (Clemson, SC, US)
- Alexander Jacobs (Oakland, CA, US)
- Daniel Gordon (Oakland, CA, US)
- Nicholas Ingle (Berkeley, CA, US)
- Laura Gerber (Oakland, CA, US)
Cpc classification
H01M4/13
ELECTRICITY
Y02E60/10
GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
H01M10/0525
ELECTRICITY
International classification
H01M4/62
ELECTRICITY
H01M4/13
ELECTRICITY
H01M4/36
ELECTRICITY
H01M4/583
ELECTRICITY
Abstract
A battery electrode composition is provided that comprises composite particles. Each composite particle may comprise, for example, active lithium fluoride/metal nanocomposite material optionally embedded into a nanoporous, electrically-conductive skeleton matrix material particle(s), where each of these composite particles is further encased in a Li-ion permeable, chemically and mechanically robust, protective outer shell that is impermeable to electrolyte solvent molecules. The active lithium fluoride/metal nanocomposite material is provided to store and release Li ions during battery operation.
Claims
1. A Li or Li-ion battery electrode composition comprising composite particles, each composite particle comprising: a mixture of metal and lithium fluoride (LiF) materials provided to store and release Li ions during battery operation; a skeleton matrix material into which the mixture is embedded to form an active material core; and a Li-ion permeable shell at least partially encasing the active material core and protecting the metal and LiF materials from interaction with a battery electrolyte.
Description
BRIEF DESCRIPTION OF THE DRAWINGS
[0025] The accompanying drawings are presented to aid in the description of embodiments of the invention and are provided solely for illustration of the embodiments and not limitation thereof.
[0026]
[0027]
[0028]
[0029]
[0030]
[0031]
[0032]
[0033]
DETAILED DESCRIPTION
[0034] Aspects of the present invention are disclosed in the following description and related drawings directed to specific embodiments of the invention. The term embodiments of the invention does not require that all embodiments of the invention include the discussed feature, advantage, process, or mode of operation, and alternate embodiments may be devised without departing from the scope of the invention. Additionally, well-known elements of the invention may not be described in detail or may be omitted so as not to obscure other, more relevant details.
[0035] While the description below may describe certain examples in the context of Li and Li-ion batteries (for brevity and convenience, and because of the current popularity of Li technology), it will be appreciated that various aspects may be applicable to other rechargeable and primary, metal and metal-ion batteries (such as Na-ion, Mg-ion, K-ion, Ca-ion, Al-ion, and others). Further, while the description below may also describe certain examples of the material formulations in a Li-free (e.g., charged) state, it will be appreciated that various aspects may be applicable to Li-containing electrodes (e.g., in either a partially or fully discharged state).
[0036] Similarly, while the description below may describe certain examples in the context of LiF chemistry, it will be appreciated that various aspects may be applicable to other lithium halide chemistries (such as LiCl, for example).
[0037]
[0038] Both liquid and solid electrolytes may be used for the designs herein. Conventional liquid electrolytes for Li- or Na-based batteries of this type are generally composed of a single Li or Na salt (such as LiPF.sub.6 for Li-ion batteries and NaPF.sub.6 or NaClO.sub.4 salts for Na-ion batteries) in a mixture of solvents (such as a mixture of carbonates). The most common salt used in a Li-ion battery electrolyte, for example, is LiPF.sub.6, while less common salts include lithium tetrafluoroborate (LiBF.sub.4), lithium perchlorate (LiClO.sub.4), lithium bis(oxalato)borate (LiB(C.sub.2O.sub.4).sub.2, lithium difluoro(oxalate)borate (LiBF.sub.2(C.sub.2O.sub.4)), various lithium imides (such as SO.sub.2FN.sup.(Li.sup.+)SO.sub.2F, CF.sub.3SO.sub.2N.sup.(Li.sup.+)SO.sub.2CF.sub.3, CF.sub.3CF.sub.2SO.sub.2N.sup.(Li.sup.+)SO.sub.2CF.sub.3, CF.sub.3CF.sub.2SO.sub.2N.sup.(Li.sup.+)SO.sub.2CF.sub.2CF.sub.3, CF.sub.3SO.sub.2N.sup.(Li.sup.+)SO.sub.2CF.sub.2OCF.sub.3, CF.sub.3OCF.sub.2SO.sub.2N.sup.(Li.sup.+)SO.sub.2CF.sub.2OCF.sub.3, C.sub.6F.sub.5SO.sub.2N.sup.(Li.sup.+)SO.sub.2CF.sub.3, C.sub.6F.sub.5SO.sub.2N.sup.(Li.sup.+)SO.sub.2C.sub.6F.sub.5 or CF.sub.3SO.sub.2N.sup.(Li.sup.+)SO.sub.2PhCF.sub.3, and others) and others. Electrolytes for Mg-ion, K-ion, Ca-ion, and Al-ion batteries are often more exotic as these batteries are in earlier stages of development. They may comprise different salts and solvents (in some cases, ionic liquids may replace organic solvents for certain applications).
[0039] In addition, solid electrolytes may provide some advantages for fluoride-based cathodes, such as stability against oxidation at high cathode potentials, reduced undesirable side reactions between the cathode and electrolyte, as well as enhanced safety. Examples of the solid ceramic electrolytes include sulfide-based electrolytes (such as Li.sub.2SP.sub.2S.sub.5, Li.sub.2SGa.sub.2S.sub.3GeS.sub.2, Li.sub.2SSiS.sub.2, etc.), halide-based electrolytes, lithium-oxy-halide and lithium-metal-oxy-halide based electrolytes (e.g., Li.sub.3OCl, Li.sub.3OCl.sub.0.5I.sub.0.5, Li.sub.2HOCl.sub.0.5Br.sub.0.5, Li.sub.1.5H.sub.0.5Al.sub.0.3OCl.sub.0.5Br.sub.0.5, etc.), oxide-based electrolytes (such as LiLaTiO garnet, LiLaTaO garnet, LiSiO glass, LiGeO glass, Li.sub.9SiAlO.sub.8, etc.), mixed sulfide-oxide electrolytes (such as Li.sub.2SSiS.sub.2Li.sub.4SiO.sub.4, LiILa.sub.2O.sub.2SLa.sub.2O.sub.2S.sub.2, etc.), and many others. The use of solid electrolytes with conversion-based cathodes has been hindered by the inability of ceramics to accommodate the volume changes that take place during charge and discharge cycling. Their use with fluoride-based cathodes has been particularly difficult because fluoride-based active cathode particles exhibit large volume changes.
[0040] Conventional cathode materials utilized in metal-ion batteries are of an intercalation-type. Metal ions are intercalated into and occupy the interstitial positions of such materials during the discharge of a battery. However, such cathodes exhibit small gravimetric and more importantly small volumetric capacities: typically less than around 220 mAh/g active material and less than around 700 mAh/cm.sup.3 at the electrode level, respectively. This low capacity of intercalation-type cathodes limits the energy density and specific energy of metal-ion batteries.
[0041] Fluoride-based cathodes may offer outstanding technological potential due to their very high capacities, in some cases exceeding 300 mAh/g (and greater than 1200 mAh/cm.sup.3 at the electrode level). For example, FeF.sub.3 offers a theoretical specific capacity of 712 mAh/g; FeF.sub.2 offers a theoretical specific capacity of 571 mAh/g; MnF.sub.3 offers a theoretical specific capacity of 719 mAh/g; CuF.sub.2 offers a theoretical specific capacity of 528 mAh/g; NiF.sub.2 offers a theoretical specific capacity of 554 mAh/g; PbF.sub.2 offers a theoretical specific capacity of 219 mAh/g; BiF.sub.3 offers a theoretical specific capacity of 302 mAh/g; BiF.sub.5 offers a theoretical specific capacity of 441 mAh/g; SnF.sub.2 offers a theoretical specific capacity of 342 mAh/g; SnF.sub.4 offers a theoretical specific capacity of 551 mAh/g; SbF.sub.3 offers a theoretical specific capacity of 450 mAh/g; SbF.sub.5 offers a theoretical specific capacity of 618 mAh/g; CdF.sub.2 offers a theoretical specific capacity of 356 mAh/g; and ZnF.sub.2 offers a theoretical specific capacity of 519 mAh/g.
[0042] In addition, in cases where the fluoride-forming element is inexpensive, fluoride-based cathodes offer a low cost potential as well. The 5-year averaged wholesale commodity cost of many fluoride-forming elements is reasonably low. For example, in 2013, the cost of Fe was only around $0.2/kg; the cost of Cu was only around $4-9/kg; the cost of Zn was only around $1-2/kg; the cost of Cd was only around $1/kg; the cost of Pb was only around $1-2/kg; and the cost of Sb was only around $6-15/kg.
[0043] However, many fluorides with high theoretical capacity and high theoretical energy density (such as CuF.sub.2, NiF.sub.2, PbF.sub.2, BiF.sub.3, BiF.sub.5, SnF.sub.2, SnF.sub.4, SbF.sub.3, CdF.sub.2, ZnF.sub.2, and others) have been believed not to be practical for use in rechargeable Li-ion batteries due to the previously observed lack of stability and very large polarizations experimentally observed when they were used in conventional cathode configurations, where metal fluorides were mechanically mixed with carbon additives or deposited on the outer surface of carbon particles.
[0044] One advantage of some of these so-called impractical fluorides (such as CuF.sub.2, PbF.sub.2, SnF.sub.2, CdF.sub.2, ZnF.sub.2, and others) over more generally used (and still not very practical) FeF.sub.2 and FeF.sub.3 is a more flat discharge curve and often (e.g. in case of CuF.sub.2) higher energy density.
[0045] In contrast to the small structural, chemical, and volumetric differences observed during insertion/extraction of Li ions into/out of so-called intercalation cathode compounds (where Li is inserted/intercalated into the interstitials of the intercalation crystals), fluorides exhibit dramatic structural changes and significant volume changes accompanying cell cycling. During electrochemical Li insertion into a metal fluoride-based cathode, a displacement/conversion process takes place, where Li displaces solid fluoride-forming element(s) (such as metals or semimetals, or in some cases semiconductors), leading to the formation of solid LiF and clusters of the fluoride-forming element(s), typically only 2-10 nanometer in size. Theoretically, the Li capacity of fluorides is determined by their stoichiometry and the density of the fluoride-forming metal according to the following reaction (which assumes fully reversible electrochemical transformation during charge and discharge):
xLi.sup.++xe.sup.+MF.sub.x.Math.xLiF+M(Eq. 1)
where M is a fluoride-forming element.
[0046] Mechanistically, it is believed that initial insertion of Li into some of the metal fluorides with metals possessing a higher oxidation state (such as FeF.sub.3) takes place as intercalation. For example, during electrochemical reaction of Li with FeF.sub.3, Li first intercalates into the structure forming:
Li.sup.++e.sup.+FeF.sub.3.fwdarw.LiFeF.sub.3(Eq. 2)
Only after additional Li insertion, a conversion reaction transforms the reaction products to LiF and interconnected Fe nanoparticles according to:
2Li.sup.++2e.sup.+LiFeF.sub.3.fwdarw.3LiF+Fe(Eq. 3)
[0047] As discussed in the background above, conventional fluoride cathodes may suffer from limitations, such as (i) low electrical conductivity; (ii) low ionic conductivity, and (iii) volume expansion during electrochemical lithiation and formation of LiF and metal clusters. Other limitations include (iv) gas generation during fluoride reactions with electrolytes (particularly at high potentials), which may cause battery degradation; (v) formation of surface species during surface reactions with the electrolyte, which may increase resistance and reduce reversibility of electrochemical reactions; (vi) oxidation of metals and dissolution of the metal fluorides during cycling, which may increase resistance, damage the solid electrolyte interphase (SEI) layer on the anode, and reduce both the power performance and cycle stability of battery cells; (vii) irreversible changes within their structure during battery operation (such as irreversible growth of the LiF and metal clusters/nanoparticles), which may also lead to irreversible resistance growth capacity losses; and (viii) the need to couple a metal fluoride-comprising cathode with a reactive and difficult to handle Li-comprising anode to form a functional cell. The present disclosure allows one to overcome some of the above-discussed above challenges and produce stable, high capacity, high energy density fluoride-based cathodes.
[0048] Formation of suitable protective shells around the electrode or around the individual particles comprising metal fluoride material may overcome at least some of the above-discussed limitations, but often not all of them (for example, a core-shell material may still require the use of Li-containing anodes). Furthermore, formation of shells is often challenging, expensive, insufficiently reliable (for achieving stability during cycling or storage, particularly at a partially or fully discharged state) and potentially dangerous. For example, formation of various shells around CuF.sub.2-comprising material (particularly at temperatures above 60 C. in an HF-free or F-free environment) generally results in undesirable reactions, such as conversion of CuF.sub.2 into a Cu metal, HF vapors, F.sub.2 gas, or various other fluorinated compounds (depending on the particular chemical synthesis route utilized for the shell deposition). The use of an HF- or F-gaseous environment may help to prevent CuF.sub.2 decomposition or conversion reactions in some cases, but it significantly increases the fabrication cost and potential hazard during this operation. Even once formed, the produced protective shells may become broken during cell operation if the expansion (induced by electrochemical lithiation of the electrodes) induces stresses sufficiently large to initiate cracks and fractures in the shell during the initial or the subsequent cycles (fatigue). Once broken, the shells do not provide the requisite stabilization against side reactions. The present disclosure provides synthesis routes to overcome such challenges and deposit suitable coatings that may better withstand stresses during battery cycling.
[0049] Embedding metal fluoride material into a suitable skeleton matrix material and thus forming fluoride-comprising composites provides complimentary advantages. This approach and particle architecture may reduce volume changes in particles and electrodes during cell operation. This approach and particle architecture may significantly reduce irreversible changes in the electrode during cycling (for example, by preventing irreversible growth of metal clusters) and thus reduce cell degradation and resistance growth. This approach and particle architecture may significantly reduce ionic and/or electrical resistance of the cathode particles. However, for some of the metal fluoride materials (for example, for CuF.sub.2) formation of such particle architecture may be challenging, expensive, or dangerous. The present disclosure provides not only the advantageous architecture of the composites, but also describes advanced approaches for suitable formation of such composite particles, where fluorides are imbedded into the skeleton matrix materials.
[0050] The present disclosure provides for advanced composite materials for battery electrodes comprising fluorides as active material(s). Instead of the traditionally used transition metal fluorides (such as FeF.sub.2, FeF.sub.3, CoF.sub.2, and NiF.sub.2), the disclosed composite materials may comprise lithium fluoride (LiF) intermixed with clusters or nanoparticles of metal(s) or metal alloy(s). This particle architecture may provide the following advantages: (i) the material does not expand significantly since Li is already present in these cathodes; (ii) the cathodes comprising such composite particles may be matched with Li-free anodes (such as graphite anodes, Si-comprising anodes, Sn-comprising anodes, etc.); and (iii) formation of protective shells around such particles is significantly more straightforward and cheaper because LiF is significantly more stable and less reactive than many other metal fluorides (in particularly than CuF).
[0051] Since the formation of lithium fluoride (LiF) intermixed with clusters or nanoparticles of metal(s) in such a way as to avoid oxidation of metal clusters/nanoparticles (with the formation of metal oxides) may be challenging, the present disclosure provides examples of suitable methods that minimize or eliminate the undesirable formation of metal oxides instead of the desired pure metal(s).
[0052] The present disclosure provides favorable compositions of the metal in the corresponding metal clusters or nanoparticles of various shapes or layers in the above-discussed composite materials. This improved metal composition allows for improved cathode properties, such as improved mechanical and electrochemical stability, reduced energy losses during cycling, and improved rate performance, to name a few.
[0053] The present disclosure also provides advantageous architectures of the composites comprising lithium fluoride (LiF) (or other halides, such as LiCl) intermixed with clusters or nanoparticles of metal(s) of various shapes or layers and further embedded into the skeleton material matrix (for example, into the skeleton matrix particles), which further improves stability, reduces volume changes during cycling, and improves other properties of the cathodes based on such composites (such as rate performance).
[0054] For many applications, it may be advantageous for the skeleton matrix material to be in the form of individual particles (powders). For significantly improved structural and chemical stability, the skeleton matrix for each composite particle may preferably be in the form of a single monolithic particle (a single-bodied particle). For many applications, it may be advantageous for the skeleton matrix material particles to be of substantially spherical shape (e.g., in order to enhance mechanical properties or the stability of individual particles). For many applications, it may be advantageous for the skeleton matrix material particles to be uniform in size (e.g., with a difference Between a so-called D90 parameter and a so-called D10 parameter being less than a so-called D50 or a median particle size; more preferably less than 50% of an average size; even more preferably less than 20% of an average size). In this case the particle architecture may be optimized for enhanced stability and sufficiently high rate, and the electrode-level ionic resistance may be minimized by forming straight channels for electrolyte ion transport within the electrode. As will be further described, it may also be advantageous for the skeleton matrix material to be electrically conductive (e.g., exhibit electrical conductivity above about 0.000001 S/m; preferably above 0.001 S/m; more preferably above 0.1 S/m; and even more preferably above around 10 S/m). For many applications, it may be advantageous for the electrode comprising the skeleton matrix material particles or composite particles (such as particles comprising a porous skeleton matrix material filled with lithium fluoride intermixed with clusters or nanoparticles of metal(s)) to comprise conductive additives (e.g., carbon black, carbon nanotubes, carbon fibers, graphite flakes, etc.) that are chemically bonded (or sintered) to the surface of such particles. In this case the fluoride-based composite particles may exhibit better stability during electrochemical cycling in cells due to significantly slower electrode degradation. It may further be advantageous for conductive additives bonded to neighboring composite particles to additionally form chemical bonds with each other.
[0055] The present disclosure also provides an improved architecture of such composites, where each composite particle comprises a shell at least partially encasing the active material (such as a lithium fluoride material intermixed with metal clusters or metal nanoparticles), the shell being substantially permeable to the Li ions stored and released by the active fluoride material and being substantially impermeable to electrolyte solvent molecules and (preferably) to ions of the metals in the above-discussed metal clusters/nanoparticles during cell operation. If the active material (such as a lithium fluoride material intermixed with metal clusters or metal nanoparticles) is embedded into a skeleton matrix material, the shell may at least partially encase the composite comprised of the skeleton matrix material with the embedded active material (such as a lithium fluoride material intermixed with metal clusters or metal nanoparticles). The presence of the protective shell(s) significantly reduces the undesirable side reactions of metal fluorides with the electrolyte and thus improves cell cycle stability and overall performance. Examples of a suitable structure of the Li-ion permeable protective shell(s), a suitable composition of the Li-ion permeable protective shell material, and suitable methods of shell fabrication are also provided.
[0056] The present disclosure also provides suitable methods for fabrication of the above-discussed composites.
[0057] As mentioned briefly above, instead of using transition metal fluorides, this disclosure provides an intermixed lithium fluoride (LiF)/metal matrix nanocomposite material. In one example, this material may be viewed as a metal fluoride cathode material in a fully lithiated (discharged) state. However, this material offers significantly more flexibility in the design and optimization of the cathode properties than simply a lithiated transition metal fluoride and thus allows for the formation of significantly better cathodes. In one example, this material may have an excess of LiF (which may be beneficial if some of the LiF is lost during subsequent processing) or an excess of metal nanoparticles (which may be beneficial for improvement of the rate capability of the cathode due to retaining electrically conductive material within an electrically insulative metal fluoride matrix and also due to the remaining metal serving as heterogeneous nucleation sites for the formation of metal clusters/nanoparticles during the conversion reaction). Another advantage of this material is its high chemical and thermal resistance. For example, if CuF.sub.2 is heated in the presence of H.sub.2, C, or other reducing agents at very moderate temperatures (sometimes as low as about 80 to about 200 C.), the Cu.sup.2+ ions of CuF.sub.2 may reduce to Cu.sup.0 metal (note that since this occurs without the simultaneous formation of LiF, an electrochemical reaction back to CuF.sub.2 is prevented from occurring if a battery is assembled with this material.). In contrast, when a mixed LiF/Cu matrix is heated at similar conditions, it will remain largely unaffected (because LiF is more stable than CuF.sub.2 and because Cu metal is already in the reduced state). Therefore, the formation of protective outer shells (shells that would protect this material from undesirable interactions with the electrolyte) and/or electrically conductive shells (shells that would provide electrical connectivity between individual particles and the current collector) around the LiF/Cu composite matrix particles may be performed at elevated temperatures and in a reducing environment (e.g., by using a CVD process that may involve H.sub.2 generation or by using a thermal decomposition of a precursor material, etc.) and thus be of significantly higher quality (better mechanical stability, better integrity, better uniformity, etc.). Furthermore, since this material already has Li in the structure, further expansion is prevented from taking place during cell operation, and thus, fracturing of the electrode, the composite particles, or the outer protective shells may be significantly reduced or minimized.
[0058]
[0059]
[0060]
[0061] Alternatively, the pores within the porous LiF 201 may be interconnected. In this case, the conductive M 202 may form an interconnected network (as described above with reference to
[0062]
[0063] The use of another electrically conductive material (e.g., another metal, conductive carbon, conductive polymer, conductive oxide, etc.) within various types of LiF/M nanocomposite particles may be advantageous for enhancing performance characteristics (e.g., stability, rate, etc.) of the cathodes.
[0064] In some applications it may be advantageous for the LiF 201 to contain dopants to enhance its electrical conductivity.
[0065]
[0066]
[0067] This intermixing of the LiF 301 and the metal(s) or metal alloy(s) M 302 material may have architectures similar to those previously described in
[0068]
[0069] The shape of the particles in
[0070] In some designs, the metal clusters or metal nanoparticles M 202/M 302 of the above-discussed lithium fluoride-comprising composites may comprise more than one type of metal atoms to exhibit more favorable electrochemical behavior in cells (such as better rate performance, smaller hysteresis, better cycle stability, better low temperature performance, better high temperature performance, better stability, easier manufacturability, etc., to name a few). The origin of such performance improvements is believed to be related to the favorable bonding between individual metal atoms that (i) prevents the growth of metal clusters during the repeated conversion reactions during battery operation; (ii) minimizes the size of metal and LiF clusters in the composite; (iii) provides lower energy pathways for the conversion reaction; (iv) prevents leaching some of the metal ions (e.g., Cu.sup.2+ ions) out of the composite particles; and (v) minimizes side reactions; or may be related to other known or unknown phenomena. In some cases, the use of mixed metals in the formation of composites allows improved control of the resulting particle morphology and architecture. In some designs, the metal clusters or metal nanoparticles of different composition may be present in a single composite particle. In some designs, metal clusters or metal nanoparticles are composed of several metals to offer improved performance. In some designs, the bonding between at least some of the metal atoms is partially ionic (due to some difference in the electronegativity of different metal atoms). In some designs, the metals of the mixed metal clusters or nanoparticles exhibit the same crystal structure (e.g., all face centered cubic, FCC, or all base centered cubic, BCC) at room temperature at equilibrium conditions. In some designs, metal clusters or metal nanoparticles may comprise a solid solution of metals. In some designs, metal clusters or metal nanoparticles may comprise intermetallic compounds. In some designs, metal clusters or metal nanoparticles may exhibit a core-shell structure, where the composition of the metal cluster/nanoparticle changes from the center to the perimeter of such particles. In some designs, the average composition of the metal cluster/nanoparticle may change from the center to the perimeter of the metal-LiF-skeleton matrix material composite particles.
[0071] In some designs, the metal clusters or metal nanoparticles of the M 202/302 may comprise Cu atoms. In some designs, the atomic fraction of Cu atoms in the clusters/nanoparticles may range from about 10 at. % to about 100 at. % (as a portion of the total metal atoms in the cluster/nanoparticle). In some designs, the metal clusters or metal nanoparticles may comprise Fe atoms. In some designs, the average atomic fraction of Fe may range from about 3 at. % to about 100 at. %. In some designs, the metal clusters or metal nanoparticles may comprise Ni atoms. In some designs, the average atomic fraction of Ni may range from about 0.05 at. % to about 100 at. %. In some designs, the metal clusters or metal nanoparticles may comprise Co atoms. In some designs, the average atomic fraction of Co may range from about 0.05 at. % to about 100 at. %. In some designs, the metal clusters or metal nanoparticles may comprise Ti atoms. In some designs, the average atomic fraction of Ti may range from about 0.01 at. % to about 21 at. %. In some designs, the metal clusters or metal nanoparticles may comprise Zn atoms. In some designs, the average atomic fraction of Zn may range from about 0.05 at. % to about 46 at. %. In some designs, the metal clusters or metal nanoparticles may comprise Bi atoms. In some designs, the average atomic fraction of Bi may range from about 0.01 at. % to about 100 at. %. In some designs, the metal clusters or metal nanoparticles may comprise Pb atoms. In some designs, the average atomic fraction of Pb may range from about 0.01 at. % to about 5 at. %. In some designs, the metal clusters or metal nanoparticles may comprise Sb atoms. In some designs, the average atomic fraction of Sb may range from about 0.01 at. % to about 35 at. %. In some designs, the metal clusters or metal nanoparticles may comprise Sn atoms. In some designs, the average atomic fraction of Sn may range from about 0.05 at. % to about 26 at. %. In some designs, the metal clusters or metal nanoparticles may comprise Cd atoms. In some designs, the average atomic fraction of Cd may range from about 0.01 at. % to about 50 at. %. In some designs, the metal clusters or metal nanoparticles may comprise Cr atoms. In some designs, the average atomic fraction of Cr may range from about 0.001 at. % to about 2 at. %. In some designs, the metal clusters or metal nanoparticles may comprise Zr atoms. In some designs, the average atomic fraction of Zr may range from about 0.1 at. % to about 28 at. %. In some designs, the metal clusters or metal nanoparticles may comprise Nb atoms. In some designs, the average atomic fraction of Nb may range from about 0.1 at. % to about 40 at. %. In some designs, the metal clusters or metal nanoparticles may comprise Mo atoms. In some designs, the average atomic fraction of Mo may range from about 0.01 at. % to about 20 at. %. In some designs, the metal clusters or metal nanoparticles may comprise Hf atoms. In some designs, the average atomic fraction of Hf may range from about 0.01 at. % to about 41 at. %. In some designs, the metal clusters or metal nanoparticles may comprise Ta atoms. In some designs, the average atomic fraction of Ta may range from about 0.01 at. % to about 35 at. %. In some designs, the metal clusters or metal nanoparticles may comprise Si atoms. In some designs, the average atomic fraction of Si may range from about 0.01 at. % to about 1 at. %. Larger quantities of Si may lead to the formation of gaseous SiF.sub.4, which may leak out during material processing or cell operation. In some designs, the metal clusters or metal nanoparticles may comprise La atoms. In some designs, the average atomic fraction of La may range from about 0.01 at. % to about 20 at. %. In some designs, the metal clusters or metal nanoparticles may comprise Ce atoms. In some designs, the average atomic fraction of Ce may range from about 0.01 at. % to about 20 at. %.
[0072] In some designs, the metal clusters, metal nanoparticles, or porous metal particles (such as the M 202 in
[0073] When the skeleton matrix material 305 is utilized for the cathode composite construction, the metal clusters/nanoparticles may be infiltrated into the pores of the skeleton matrix material 305 or deposited onto the surface of the skeleton matrix material 305 (such as skeleton matrix material particles). The infiltration temperature may be sufficiently low to prevent excessive growth of metal particles (their minimum characteristic dimensions may preferably stay within 50 nm, or more preferably within 5 nm). Since it may be important to prevent formation of oxide on the metal surface, the concentration of oxygen containing reactive species (such as O.sub.2, CO.sub.2, H.sub.2O, and others) in the deposition chamber may preferably be kept to a minimum (so that the total amount of O atoms entering the chamber during the deposition may preferably be significantly below the total number of metal atoms deposited, preferably 10 at. % or below).
[0074] In some designs, the metal clusters or metal nanoparticles may be produced or deposited by electroless deposition or electrodeposition on skeleton matrix material substrates (e.g., into the pores of porous skeleton matrix particles).
[0075] In some designs, the metal clusters, metal nanoparticles, or porous metal(s) may also be produced via a wet synthesis route, either by homogeneous nucleation in the bulk of a solution or by heterogeneous nucleation on the surface of the skeleton matrix material.
[0076] In some designs, the metal clusters or metal nanoparticles (e.g., the M 202 in
[0077] In some designs, the LiF clusters, LiF layers, LiF nanoparticles, or LiF porous particles (the LiF 201 of
[0078] When a skeleton matrix material is utilized for the cathode composite construction (e.g., as in
[0079] As an alternative to the subsequent formation or deposition of metal (M) and LiF clusters/nanoparticles/porous particles, metal fluoride (MF.sub.x) (nano)particles may be first formed (or deposited or infiltrated into a skeleton matrix material) and then converted into the LiF/M nanocomposite by using one of the chemical lithiation procedures (preferably in an anhydrous environment or other environments free from those solvents that may induce undesirable interactions either with LiF or with M). These chemical lithiation processes may reduce a metal (M) from the MF.sub.x into its metallic state and simultaneously form LiF.
[0080]
[0081] Subsequently, the MF.sub.x compositions may be placed in a temperature-controlled environment of the reactor (block 402A). Controlling temperature may be useful because lithiation temperature affects the size and morphology of structural features in the obtained M/LiF composites. At some temperatures (and compositions), M and LiF may form interpenetrating networks during lithiation of MF.sub.x particles. At other temperatures (and compositions), one may form porous M with LiF filling its pores during lithiation of MF.sub.x. At yet other temperatures (and compositions), one may form porous LiF with M filling its pores. In some conditions, the obtained M/LiF composite particles largely retain the initial shape of the MF.sub.x particles. Such a situation may be preferable in many cases (e.g., when control of the composite morphology is important). At too high temperatures of the lithiation reactions, the size of M clusters or LiF clusters may be undesirably large and morphology of the M/LiF composite may deviate significantly from the morphology and size of the initial MF.sub.x particle. Unfortunately, chemical lithiation may be a highly exothermic process, where heat is released upon the chemical lithiation reaction. If such a reaction proceeds too fast and the heat extraction from the reactor is too slow, the control of the reaction temperature may be lost and the desired M/LiF composite morphology may not be obtained. The third step may involve lithiation of the MF.sub.x compositions (block 403A). Excess of metal (M) may be introduced into the M/LiF composite by, for example, post-deposition (e.g., by CVD, ALD, or wet chemistry routessuch as precipitation from the solution with subsequent reduction). Alternatively, a metal precursor may be added to the MF.sub.x particles prior to chemical lithiation. In this case the lithiation procedure may simultaneously reduce the metal precursor to form an additional metal component of the composite. Excess LiF may be introduced into the M/LiF composite by, for example, post-deposition (e.g., by CVD, ALD, or wet chemistry routes). Alternatively, fluorinating a composite matrix material may provide extra fluorine needed for the additional LiF formation during the chemical lithiation. By-products of the chemical reactions may be removed after synthesis.
[0082] In some designs, lithiation (block 403A of
[0083] Many lithiation/reducing agents are reactive to air and moisture and may need to be handled under a dry, inert atmosphere. Ideally, lithiating/reducing agents should produce LiF and MF.sub.x without side reactions and with reaction byproducts that are easily removable. Byproducts may be removed by extraction with organic solvents or evaporation/sublimation at reduced pressures or elevated temperatures. Alkyllithium reagents produce short-chain hydrocarbon byproducts that may be either extracted with organic solvents or removed by evaporation under reduced pressure or elevated temperature. A drawback of alkyllithium reagents is that they may produce alkylate metal salts (e.g., producing M(alkyl).sub.x), which is an undesired reaction in this case. The byproducts of organic radical anions (including ketyl radical anions or polycyclic aromatic hydrocarbon radical anions) may be uncharged organic small molecules that may be extracted with organic solvents or removed under reduced pressure or elevated temperature by evaporation or sublimation. The byproducts of alkyllithium magnesates, alkyl magnesium compounds, and hydrocarbons may also be easily removable by extraction in organic solvents or be removed by evaporation (e.g., under reduced pressure or elevated temperatures). Lithium trialkylborohydrides produce hydrogen gas and trialkylboranes, which may be easily removed by evaporation, but because trialkylboranes are pyrophoric this reagent may be less desirable for safety and ease of handling.
[0084] Reactions with chemical reducing agents may be conducted in organic solvents, specifically including ethereal solvents such as tetrahydrofuran, dimethoxyethane, diethylether, or dioxolane or hydrocarbon solvents such as C5-C9 alkyl chains. Reactions may be conducted under inert atmosphere to prevent oxidation by O.sub.2 or oxidation or hydration by water. The temperature of lithation reactions may typically range from about 78 C. to about 600 C. (too low temperature restricts reaction kinetics too much, while too high temperature makes it less controllable). Pressure of the reaction may range from about 0.05 atm to about 50 atm (near-atmospheric pressure may typically be suitable). Concentration of the soluble reducing agents may range from about 0.001 M to about 10 M (or more, in some cases) in the reaction solvent. Molar ratios of lithium cations Li.sup.+ relative to metal cations M.sup.n+ may range from about 0.5n to about 100n. The byproducts and solvents from reaction may be removed by filtration, centrifugation, evaporation, sublimation, carbonization, a combination thereof, or other methods.
[0085] Some of the above-described lithiation methods may be useful for lithiating other conversion-type electrode materials (e.g., S, Si, Sn, etc.) for use in rechargeable Li and Li-ion batteries.
[0086] In some designs, it may be advantageous to coat the lithiated particles by a protective surface layer (e.g., carbon or hydrocarbons) in order to enhance their stability prior to conducting subsequent material processing steps (e.g., formation of shells around the composite particles or formation of electrodes from the composites, etc.). In some designs, such a protective coating may be deposited by expositing the lithiated composite (e.g., M-LiF or M-LiF-skeleton matrix material) into a stream of a hydrocarbon gas (e.g., in a temperature range from about room temperature to about 700 C.). In some designs, it may be advantageous not to expose the lithiated particles (either coated or not coated by a protective surface layer) to air, O.sub.2, moisture, or other harmful environments in order to prevent their undesirable degradation (e.g., oxidation of metal M components of the composite) prior to conducting subsequent material processing steps (e.g., formation of shells around the composite particles or formation of electrodes from the composites, etc.).
[0087]
[0088] In order to introduce more than one type of metal atom into the M nanoparticles/nanoclusters, several (more than one) different metal fluorides (e.g., M1F.sub.x1, M2F.sub.x2, M3F.sub.x3, M4F.sub.x4, their various mixture(s) or solid solution, etc.) may be used for the nanocomposite formation (may be introduced simultaneously or one after another, etc.). Advantages of this approach may include the decrease in an undesirably high reactivity and undesirable diffusion and growth of the M nanoparticles/nanoclusters during subsequent processing steps. Another advantage is the superior electrochemical performance of the composites comprising different metal atoms as cathodes in battery cells.
[0089]
[0090] In many cases, controlled formation of the metal fluoride MF.sub.x particles of the desirable composition, shape, size, and morphology can be challenging. Furthermore, infiltration of certain metal fluoride compositions into the pores of the skeleton matrix material may be even more challenging.
[0091]
[0092] As an illustrative example, an aqueous solution of a salt of a fluorosilicic acid (e.g., FeSiF.sub.6.H.sub.2O) may be impregnated into skeleton matrix material pores, dried, and the produced salt may be transformed into a metal fluoride (e.g., an iron fluoride FeF.sub.2) by annealing (for examples, in an inert argon (Ar) gas). The annealing process decomposes FeSiF.sub.6.H.sub.2O into FeF.sub.2 (more generally FeFX), water vapors, and volatile SiF.sub.4 (the SiF.sub.4 boiling point=86 C.), with volatile compounds being removed.
[0093]
[0094]
[0095] In one example embodiment of such an approach, metal or metal alloy (nano)particles (e.g., Cu, CuFe, CuNi, CuFeNi, or other types of suitable metal particles, preferably with over 10 wt. % Cu) are synthesized and coated by a polymer material (or other suitable skeleton matrix material, such as carbon, including either porous or dense carbon), producing metal-polymer (or metal-carbon) composite particles. In some designs, such particles may be agglomerated so that multiple (nano)particles are embedded into a polymer material. In some designs, multiple individual metal particles may be embedded into a polymer material. In some designs, metal particles (e.g., agglomerated particles) of one metal composition may be at least partially coated with another metal composition (in some designs, another metal composition may be more resistive to fluorination reaction and thus may remain conductive during electrochemical cycling; for example, coating at least part of the surface of Cu or CuFe alloy particles with Ni, which is more resistive to fluorination and has a higher ignition temperature in fluorine than Cu and Fe). In some designs, such a metal coating process (e.g., reduction of metal salts) induces agglomeration of metal (nano)particles. The produced metal-polymer (or metal-carbon) composite particles may be heat-treated (e.g., to induce sintering of nanoparticles or to carbonize a polymer) and may be additionally ball-milled (e.g., to separate individual composite particles). The produced powder may be fluorinated. In some designs, a part of the metal composition (e.g., metal component that is more resistant to fluorination) may at least partially remain in metallic (electrically conductive) state. The next step may involve an optional removal of the reaction by-products or excess F (e.g., if excessive fluorination of the skeleton matrix material is undesirable), such as by heating in an inert environment. Synthesis conditions (e.g., temperature, pressure, concentration of the reactive species, etc.) in each step may be controlled to achieve the desired morphology of the composite particles. In some designs, the produced metal fluoride-skeleton matrix material composite may be additionally coated with Li-ion permeable shells (e.g., comprising carbon) that may serve to protect the MFx from undesirable interactions with environment during subsequent processing steps or during electrochemical cycling.
[0096] In the case of relatively volatile metal fluorides (e.g., SbF.sub.5, SbF.sub.3, BiF.sub.5, SnF.sub.2, and others), skeleton matrix material filling with the active material may be performed by capillary condensation of the fluoride vapor. Because of capillary action, vapor is preferentially condensed into skeleton matrix material capillaries, starting from the smallest pores, which exhibit the strongest interactions with the vapors.
[0097] Formation, infiltration or deposition of some of the MF.sub.x materials (particularly in HF-free and F.sub.2-free environments) may sometimes be challenging. For example, CuF.sub.2 is difficult to handle due to its reactivity and the tendency of the material to capture moisture (CuF.sub.2 is very hygroscopic) and form CuO. Processing at elevated temperatures of composite materials containing hydrated CuF.sub.2 may produce Cu(OH)F and/or copper oxide and release hydrofluoric acid vapors.
[0098] Several suitable methods for nanostructured CuF.sub.2 processing (formation/deposition) are disclosed below.
[0099] As an example, hydrated CuF.sub.2 (such as CuF.sub.2*2H.sub.2O) (a light blue-colored powder) is first dissolved in a suitable solvent (such as 2,2-Bipyridyl, N,N,N,N,N-pentamethyldiethylenetriamine or Tris[2-(dimethylamino)ethyl]amine or pyridine, or other suitable solvent) generally producing a blue-colored solution. If a CuF.sub.2/skeleton matrix nanocomposite material is desired, this solution may be infiltrated into the skeleton matrix material pores. A vacuum may be applied to help draw the liquid into the small (e.g., C) pores and evaporate the solvent. The process may be repeated additional times to increase mass loading of CuF.sub.2. Since the deposited CuF.sub.2 may still be hydrated, the CuF.sub.2 may be dehydrated. One method for dehydrating hydrated CuF.sub.2 may involve heating under an inert gas (such as Ar or He) stream or under vacuum at temperatures ranging from about 150 to about 500 C. Another method may involve dispersing hydrated CuF.sub.2 in a solvent having a high (above about 150 C. evaporation point (e.g., ionic liquids) and heat-treating the suspension under an inert gas stream or under vacuum. The exact temperature range of the suitable annealing/heat-treatment conditions depends on the chemistry of the skeleton matrix material and the size of the CuF.sub.2 particles. For example, if the skeleton matrix material is made of a reducing agent, such as conductive carbon, too high of a temperature (e.g., above around 400 C.) may induce reduction of the material to Cu metal. Too low of a temperature may be insufficient to drive all H.sub.2O (or other solvent strongly bound to CuF.sub.2) away. To assist in CuF.sub.2 formation (e.g., to avoid formation of Cu(OH)F), annealing (heating) may also instead be performed in the presence of F.sub.2, HF, or its precursors (such as NH.sub.4F).
[0100] As another example, anhydrous CuF.sub.2 may be first dissolved in a suitable solvent (such as anhydrous 2,2-Bipyridyl, N,N,N,N,N-pentamethyldiethylenetriamine or anhydrous Tris[2-(dimethylamino)ethyl]amine or anhydrous pyridine, or other suitable anhydrous solvent), generally producing a blue-colored solution. If a CuF.sub.2/skeleton matrix material nanocomposite material is desired, this solution may be infiltrated into the skeleton matrix material pores. A vacuum may be applied to help draw the liquid into the small (e.g., C) pores and evaporate the solvent. The process may be repeated one or several times to increase mass loading of CuF.sub.2. If the solvent has a significant affinity to CuF.sub.2, heating under an inert gas (such as Ar or He) stream or under vacuum at temperatures ranging from 150 to 500 C. may be used. Another method may involve dispersing CuF.sub.2 or CuF.sub.2/skeleton matrix material nanocomposite (particles) in yet another solvent having a high (e.g., above 150 C.) evaporation point (e.g., ionic liquids) and heat-treating the suspension under an inert gas stream or under vacuum. The exact temperature range of the suitable annealing/heat-treatment conditions depends on the chemistry of the skeleton matrix material and the size of the CuF.sub.2 particles. In some examples, annealing (heating) may instead be performed in the presence of F.sub.2, HF, or its precursors (such as NH.sub.4F).
[0101] CuSiF.sub.6 is an additional example of a suitable precursor. A non-aqueous solution of CuSiF.sub.6 may be first produced and either produced as a powder (particles) or infiltrated into the pores of the skeleton matrix material (in the case of the MF.sub.x/skeleton matrix material composite). Upon heating (in an inert environment), CuSiF.sub.6 decomposes, producing CuF.sub.2 and SiF.sub.4 gas.
[0102] Several methods may be utilized to prepare an anhydrous CuSiF.sub.6 and anhydrous CuSiF.sub.6 solution(s). In one approach, CuO may be first dissolved in an aqueous H.sub.2SiF.sub.6 solution. The O and H.sup.+ combine to form water, resulting in an aqueous Cu(II)SiF.sub.6 solution. Drying this solution results in the formation of hydrated Cu(II)SiF.sub.6. In a slightly different approach, Cu metal electrodes may be first submerged in an H.sub.2SiF.sub.6 aqueous solution and then a current may be passed through the electrodes, such that Cu is oxidized to form Cu.sup.2+ ions at the positive electrode and H.sup.+ ions are reduced to form H.sub.2 gas at the other electrode, thus replacing H.sup.+ in the solution with Cu.sup.2+. A particular current and placement of the electrodes far apart, with the positive electrode at the bottom of the electrolyte container, promotes Cu.sup.2+ formation and H.sup.+ reduction instead of O.sub.2 evolution and Cu.sup.2+ reduction.
[0103] Once hydrated Cu(II)SiF.sub.6 is produced, the water of hydration may be removed by careful heating under controlled conditions. In order to prevent formation of CuO and produce anhydrous Cu(II)SiF.sub.6, the water may first be removed (evaporated) from the solution to form hydrated Cu(II)SiF.sub.6 crystals and then H.sub.2O may be carefully extracted further. In one example, these hydrated Cu(II)SiF.sub.6 crystals may then be dissolved in suitable anhydrous non-aqueous solvent(s). Examples of such a solvent(s) include, but are not limited to, dimethylformamide (DMF), NMP, methanol, or others. A substantial amount of solvent may be used to dilute the water content, or the solution may be repeatedly evaporated at room or elevated temperature (either at atmospheric pressure or under vacuum; in some cases with F.sub.2 or HF vapors present) and more anhydrous solvent added to dilute the water content. Water may also be extracted from the hydrated Cu(II)SiF.sub.6 by using a conversion (replacement) reaction or by other mechanisms. This non-aqueous salt solution may then be used for impregnating pores of the skeleton matrix material or producing the nanoparticles or nanocomposites. In another example water may be removed by azeotropic distillation of Cu(II)SiF.sub.6 solution in a solvent mixture, where one solvent forms an azeotrope with water. Aromatic hydrocarbons (toluene, benzene, etc.) are examples of suitable solvents for this purpose.
[0104] In another approach, a replacement (or double replacement) reaction may be utilized. In one example, anhydrous XSiF.sub.6 and anhydrous CuY (where X and Y are positively charged and negatively charged ions; each may be either a single ion of magnitude 2 elementary charge or two ions of magnitude 1 elementary charge), may be dissolved in an anhydrous solvent for which both XSiF.sub.6 and CuY are soluble but CuSiF.sub.6 or XY is insoluble. A precipitated anhydrous CuSiF.sub.6 or non-aqueous CuSiF.sub.6 solution may be retrieved by decanting. Once precipitated, CuSiF.sub.6 may be dissolved in a suitable non-aqueous solvent. A somewhat similar approach may also be used for anhydrous CuF.sub.2 formation.
[0105] In yet another approach, ion exchange may be performed for anhydrous CuSiF.sub.6 formation. In one example, a non-aqueous solvent may be initially used that dissolves XSiF.sub.6, CuY, and CuSiF.sub.6. (X and Y have the same meanings as described above.) That is, XSiF.sub.6 may be dissolved on one side of a cation exchange membrane, and CuY on the other side of the cation exchange membrane. Cu.sup.2+ will diffuse across the membrane to replace X, which will diffuse across the membrane in the opposite direction. The original CuY solution on the one side of the membrane may be exchanged with fresh CuY solution as the Cu.sup.2+ concentration depletes and the X concentration increases in this solution, until all X is replaced with Cu.sup.2+ on the other side of the membrane.
[0106] Unfortunately, the low solubility of CuF.sub.2 and CuSiF.sub.6 found in various solvents may make the methods described above slow, which may be undesirable. The inventors have identified that intermediate formation (or use) of CuF.sub.2 precursor particles (e.g., other Cu salts, Cu oxides, Cu carbides, Cu nitrides, Cu metals, etc.) that are later converted to CuF during fluorination reaction(s) may be advantageous in some designs. Such a process of producing metal fluoride-based composite particles (particularly CuF.sub.2-comprising particles) is not trivial and has never to their knowledge been used in the synthesis of the related CuF.sub.2 (or Cu/LiF) comprising cathodes. Similarly, other metal fluorides as well as mixed metal fluorides may benefit from such an approach.
[0107]
[0108] As shown in
[0109]
[0110] Suitable examples of metal fluoride (MF.sub.x) precursors include, but are not limited to: (i) metals (M) (including mixed metalse.g., as alloys); (ii) metal oxides (MO.sub.y) (including mixed metal oxides); (iii) metal nitrides (MN.sub.z) (including mixed metal nitrides); (iv) various other metal salts (including mixed metal salts); (v) various metal carbides (including mixed and ternary carbides); and (vi) various combinations of (i), (ii), (iii), (iv), and (v).
[0111] Suitable examples of the precursor salts may include, but are not limited to: metal chlorides (MCl.sub.x), metal bromides (MBr.sub.x), metal nitrates M(NO.sub.3).sub.x, metal fluoroborates (M(BF.sub.4).sub.x), metal sulfates (M(SO.sub.4).sub.x), metal trifluoromethanesulfonate (M(OSO.sub.2CF.sub.3).sub.x), metal iso-butyrate (M[CO.sub.2CH(CH.sub.3).sub.2].sub.x), metal acetate (M(O.sub.2CCH.sub.3).sub.x), or metal acetylacetonate, to name a few. As previously discussed, examples of the metals in these precursor metals, oxides, nitrides, carbides, and salts may include Cu, Fe, Ni, Co, Ti, Zn, Bi, Pb, Sb, Sn, Cd, Cr, Zr, Nb, Mo, Hf, Ta, Si, La, Ce, and their mixtures. In some designs, high fractions of Cu and Fe may be advantageous for achieving high fractions of CuF.sub.2 and FeF.sub.3 in the composite cathode particles, which, in turn, results in high energy density of the cells. Precursor salts may be homoleptic or heteroleptic.
[0112] As illustrated in the method of
[0113] Fluorination reagents may be solution-based (for example, using a non-aqueous solution of tetrabutylammonium fluoride or tetramethylammonium fluoride, to name a few examples) or gas-phase (for example, using such fluorinating agents as fluorine (F.sub.2) or nitrogen fluoride (NF.sub.3), to name a few examples).
[0114] Several methods may be utilized to generate appropriate reactive species for the effective gas fluorination of the infiltrated salt(s). For example, NF.sub.3 gas may be heated (e.g., above around 200 C.) to decompose into F and NF.sub.2 radicals. Such radicals may transform precursor salt(s) to metal fluorides. For example, transform metal chloride(s) (e.g., FeCl.sub.2 or CuCl.sub.2 or their mixtures, etc.) into metal fluorides (e.g., FeF.sub.2 or CuF.sub.2 or their mixtures, etc., respectively).
[0115] Fluorination via the decomposition of NF.sub.3 gas or by using other fluorinating gases may be performed in either a static reaction vessel (no fluorinating gas flow) or one that has continuous flow of a fluorinating (e.g., NF.sub.3) gas.
[0116] In some designs, particles may be agitated during the fluorination reaction to increase uniformity of the fluorination. In some designs, a rotary-type reactor (e.g., rotary kiln reactor) may be used for powder agitation purposes. In some other designs, a mixer may be utilized for powder agitation. In some designs, vibrations of the reactor may be utilized for powder agitation. In some designs, the flow of gas may be utilized for powder agitation (e.g., as in a fluidized bed reactor).
[0117] Alternatively to NF.sub.3, other gases (e.g., ClF.sub.3, SF.sub.6, and F.sub.2 gas) may also be used for the fluorination. An alternative method to using F.sub.2 is to generate fluorine or fluorine radicals in-situ (to minimize the safety hazard of F.sub.2), such as via the electrolysis of anhydrous hydrogen fluoride (HF), or an RF or microwave plasma source to break apart fluorine bearing gasses such as NF.sub.3, ClF.sub.3, or SF.sub.6. The use of plasma-assisted fluorination may be particularly attractive due to its high efficiency, high rate, and low temperature.
[0118] Fluorination of precursor salt compositions (including salts infiltrated into the porous skeleton matrix material) via the breakdown of NF.sub.3 by an RF remote plasma source was found to be a highly effective fluorination method that can be used over a wide range of temperatures (e.g., from below room temperature up to over 200 C.).
[0119]
[0120] In many designs, it may be highly advantageous to preserve the overall shape of the precursors during the fluorination or lithiation reactions. If the precursors are infiltrated into the pores of the skeleton matrix material, it may be particularly advantageous for the fluorinated or lithiated particles to remain within such pores because remaining within the pores enhances performance (rate, stability, etc.) of the metal fluoride-based battery electrodes. Unfortunately, fluorination, lithiation, or other chemical or physical processing steps may cause diffusion of the metal fluoride or its precursors from the interior of the porous matrix material (presumably to minimize the surface area of the nanoparticles; the energy for such undesirable mass transport may be provided by the heat supplied to the reaction vessel or generated during exothermic chemical reaction; in the case when the conversion process takes place in a solution, it is also possible that at least one intermediate species involved in such a conversion reaction may possess a sufficiently high solubility to migrate from the pores prior to precipitation on the outer surface of the particles). Processes may include measures to prevent diffusion of the metal fluoride or its precursors from the interior of the matrix material. An example of such a process is a conversion of an infiltrated precursor salt to another precursor with a higher melting point to lower mobility at elevated temperatures. Specifically, a metal nitrate salt may be infiltrated into the matrix material followed by heating in the presence of O.sub.2 to at least partially convert the metal nitrate to a metal oxide. This process may be conducted once or repeated multiple times to increase loading of a metal oxide in the skeleton matrix material before subsequent fluorination. Another example includes the co-infiltration of multiple metal salts together. It is hypothesized that the observed improved performance and stability of the mixed precursors during such transformation reactions may be related to changes in the thermodynamics or kinetics of the fluorination (or lithiation) reactions or other processes that are leading to reduced mobility (and the resulting reduced out-diffusion from the skeleton material pores).
[0121]
[0122] Direct contact of MF.sub.x, M, or LiF with the electrolyte may induce many undesirable side-reactions. Such reactions are particularly rapid with some of the MF.sub.x, such as CuF. For example, when Cu metal is exposed to the electrolyte, it may oxidize and dissolve as Cu.sup.1+ and Cu.sup.2+ species at 3.56V versus Li/Li.sup.+ and 3.38V vs. Li/Li.sup.+, respectively (cited are the standard oxidation potentials with 1M Cu.sup.+ or Cu.sup.2+ in solution). Unfortunately, these Cu oxidation potentials are very close to the redox potential for the conversion reaction of LiF and Cu metal to CuF.sub.2 of 3.55V vs. Li/Li.sup.+. Thus, the formation of Cu.sup.+ and Cu.sup.2+ may occur alongside the desired redox reaction, unless steps are taken to prevent Cu metal dissolution. Besides causing the loss of Cu in the electrode, the formation of Cu.sup.+ and Cu.sup.2+ in solution may also encourage the dissolution of LiF through the formation of [Cu(I)F.sub.4].sup.3 or [Cu(II)F.sub.6].sup.4 species and also damage the SEI on the anode. Other MF.sub.x may similarly suffer from such limitations. Formation of Li permeable shells around the LiF/M or LiF/M/skeleton matrix material nanocomposite (particles) may overcome these issues and prevent the dissolution of the active material.
[0123]
[0124]
[0125] The shell may comprise carbon, for example. The shell may be solely made of mostly graphitic (sp.sup.2-bonded) carbon or have an outer layer composed of graphitic (electrically conductive) carbon. The shell may comprise an active (e.g., Li storing) material layer, wherein the active material layer is formed from a different active material than in the core of the particles. The shell may comprise an intercalation-type active material, where accessible Li capacity, for example, of the intercalation material in the potential range of the cathode operation in a cell may range from about 0.1 mAh/g to about 300 mAh/g (where the low capacity may originate from Li intercalation in a different potential than that of the cathode operation). Examples of suitable lithium intercalation-type materials may include, but are not limited to various materials having a layered, spinel, olivine or tavorite crystal structure (such as lithium cobalt oxide, various lithium manganese oxides, various lithium nickel cobalt manganese oxides, various lithium cobalt aluminum oxides, various lithium polyanion compounds, such as various lithium phosphates, various lithium silicates, various other lithium compounds that comprise, for example, large (XO.sub.4, where X may be P, Si, As, Mo, W, S, etc.) anions, including those that comprise F in their crystal structure). The shell may also comprise a Li-free intercalation-type material, which is stable in the potential range of the core-shell cathode operation. Examples of suitable Li-free intercalation-type materials may include but are not limited to various oxides, nitrides and sulfides, to name a few. The active material of the active material layer may have a substantially lower capacity relative to the active fluoride material.
[0126]
[0127]
[0128]
[0129]
[0130]
[0131] In some designs, each composite particle may further comprise external channel pores extending from an outer surface of the skeleton matrix material towards the center of the skeleton matrix material, providing channels for diffusion of the ions into the active material disposed within the skeleton matrix material. At least some portion of the external channel pores may be filled with (i) a porous material having a different microstructure than the skeleton matrix material; (ii) an active material that does not undergo conversion reaction in the potential range of the composite cathode operation; and/or (iii) a solid electrolyte material. The surface of the external channel pores may be coated with a thin (from about 0.3 nm to about 50 nm) layer of (i) electrically conductive material; (ii) an active material that does not undergo conversion reaction in the potential range of the composite cathode operation; and/or (iii) a solid electrolyte material.
[0132] In some designs, shells may be deposited (at least, in part) by using one of the gas deposition methods, such as CVD or ALD. In some designs, the shell formation may involve multiple steps, where initially the shell precursor is first deposited conformably in a solution and then transformed (at least, in part) into the shell material via thermal decomposition and/or chemical reaction. In some designs, multiple approaches may be combined in order to produce conformal, essentially defect-free shells around individual particles. In some designs, shells may be deposited electrochemically (for example, after the electrodes comprising the M/LiF or M/LiF/skeleton matrix material-composite particles are assembled). In some designs, the shells may be deposited on electrodes comprising M/LiF or M/LiF/skeleton matrix material composite(s), in turn, produced by chemical or electrochemical lithiation of the corresponding MF.sub.x or MF.sub.x/skeleton matrix material composites.
[0133] In some designs, M/LiF or M/LiF/skeleton matrix material (nano)composite particles may exhibit a hierarchical architecture for improved stability or performance in cells.
[0134] In some designs, M/LiF or M/LiF/skeleton matrix material (nano)composite particles may exhibit a gradually changing composition for improved stability or performance in cells. In some examples, the volume fraction of the skeleton matrix material may increase (substantially, e.g., by 10% to 500%) near the surface of the (nano)composite particles or be reduced (substantially, e.g., by 10% to 500%) in the center of the (nano)composite particles. In some examples, the size of pores in the skeleton matrix material may reduce near the surface of the (nano)composite particles or be enlarged in the center of the (nano)composite particles. In some examples, one large central void may exist in the skeleton matrix material. Such a void may be partially (or completely) filled with M/LiF composite.
[0135] In some designs, an excess of M atoms (relative to LiF and the corresponding MF.sub.x formation) in M/LiF or M/LiF/skeleton matrix material composites may be advantageous, such as in applications requiring higher rate performance. This excess of M atoms may provide increased electrical conductivity and also serve as sites for heterogeneous nucleation and growth of M during the lithiation of MF.sub.x, thus reducing nucleation barrier and correspondingly reducing voltage hysteresis, and improving rate performance. In some designs, this excess of M atoms may be selected from those metals that have weaker bonding to F (more electropositive to the majority of M used in the electrochemical reaction) so that the excess M does not tend to get fluorinated.
[0136] In some designs, selected metal fluorides (such as MgF.sub.2 and NaF) may be added to LiF as dopants in order to improve rate performance (e.g., by serving as sites for heterogeneous nucleation of LiF during lithiation of (e.g., transition) metal fluorides). In one illustrative example, up to approximately 3% of MgF.sub.2 (relative to LiF) may be added to provide performance improvements. In another illustrative example, up to approximately 0.1% of NaF relative to LiF may be added.
[0137]
[0138]
[0139]
[0140]
[0141]
[0142]
[0143] In some designs, the skeleton matrix material may comprise 20-100 at. % carbon. In some designs, forming the skeleton matrix material may comprise, for example, forming a carbon-containing precursor, (optionally) oxidizing and carbonizing the carbon-containing precursor to form a carbonized particle, and activating the carbonized particle at elevated temperature to form the skeleton matrix material with a pore volume of greater than 50 vol. %.
[0144] In some designs, at least an outer portion of the shell may be deposited electrochemically during one or more initial battery cycles, during which electrochemical decomposition of at least some electrolyte components occurs.
[0145] In some designs, the protective layers or protective shells (or coating(s)) may be deposited from a vapor phase via vapor deposition methods. Examples of such methods include, but are not limited to, chemical vapor deposition (CVD), atomic layer deposition (ALD), plasma-enhanced ALD, plasma-enhanced CVD, vapor infiltration, and others. For some designs, the protective layer or protective shell material(s) may be deposited from a solution. Examples of suitable methods include sol-gel, layer-by-layer deposition, polymer adsorption, surface initiated polymerization, nanoparticles adsorption, spray drying, and others.
[0146] In some designs, the protective layer or protective shells (coating(s)) may comprise carbon. In some designs, the carbon content in the coating may range from about 10 wt. % to about 100 wt. %.
[0147] Several material compositions may be suitable for use in a skeleton matrix of the type described herein, including but not limited to: metals and semimetals, polymers (including polymers that are electrically conductive within the cathode operational potential range), oxides, carbides (including ternary and other mixed carbides), oxy-carbides (including partially fluorinated ones), oxy-fluorides, various intercalation-type cathode materials, metal nitrides, carbons (including partially or fully fluorinated ones, such as carbon fluorides), carbon oxy-fluorides, and others. In some designs, it may be advantageous for the skeleton matrix material to be in the form of individual, monolithic (single-bodied) particles. In some designs, the skeleton matrix material may comprise from about 0.05 at. % to about 50 at. % fluorine (F). In some designs (e.g., when chemical lithiation is used), at least a portion of this F may be converted to LiF during the lithiation step of the composite cathode synthesis. In some designs, it may be further advantageous (e.g., for maximizing energy density of the cathodes) for the volume fraction of the skeleton matrix material in the individual skeleton matrix materiallithium fluoridemetal composite particles not to exceed 35-40 vol. %, more preferably not to exceed 25-30 vol. %, and even more preferably not to exceed 15-20 vol. %. However, when its volume fraction becomes too small (e.g., smaller than about 3-5 vol. %) the functionality of the skeleton matrix material may typically be undesirably reduced. In some designs, it may be further advantageous for the skeleton matrix material particles to exhibit characteristic dimensions in the range of about 20 nm to about 20 microns. Larger particles are easier to handle during the slurry preparation, but too large particles (the size of which depends on the cell chemistry and operating conditions, but typically below about 20 microns) may reduce rate performance below the minimum (for a given application) value. In some designs, if porous skeleton matrix particles are used, the characteristic dimensions of the majority (e.g., above about 70 vol. %) of the skeleton matrix material pores (which may be completely or partially filled with active materials, such as a mixture of LiF and metals, metal fluorides, etc.) may preferably be in the range of about 1 nm to about 100 nm. In some designs, if porous skeleton matrix particles are used, the volume of pores within such particles may preferably be in the range of about 60 vol. % to about 95 vol. %. In some designs, at least some of the pores within the skeleton matrix material may be produced using sacrificial template materials (e.g., a sacrificial polymer, sacrificial metal, or sacrificial salt particles). The pores within the skeleton matrix material may be straight or curved and may be of various shapes (e.g., mostly slit shape, cylindrical shape, spherical shape, or have some other shape, including irregularly-shaped curved pores). If the pores are of elongated or slit shape, the orientation of such pores may depend on the skeleton particle preparation and may vary within a single particle. For example, in the center of the particles the pores may be oriented randomly or along the radius, while closer to the surface the pores may be oriented parallel to the surface of the particles. In this case one may achieve superior mechanical properties and, in some cases, simplify formation of the protective coatings or shells, which may serve to stabilize the metal or metal fluorides against irreversible changes and undesirable interactions with electrolyte. In some designs, porous skeleton matrix material particles may have smaller pores closer to the particle surface. In this case one may achieve superior mechanical properties and, in some cases, simplify formation of the protective coatings or shells. In some designs, porous skeleton matrix material particles may exhibit a smaller pore volume closer to the particle surface. In this case one may achieve superior mechanical properties and, in some cases, simplify formation of the protective coatings or shells. In some designs, when metal fluorides infiltrated in the pores of the porous skeleton matrix material undergo chemical lithiation, the skeleton matrix material may become irreversibly deformed without substantial changes in the mechanical integrity of the composite particles. In some designs (e.g., when either metal fluorides or a mixture of metal and LiF are infiltrated in the pores of the porous skeleton matrix material) the volume changes upon repeated electrochemical cycling of such composites may be accommodated mostly by reversible elastic deformations (e.g., by over 50% elastic deformation after cycle 5 and by over 90% elastic deformation after cycle 50); in this case high stability of cells may be achieved.
[0148] In some designs, skeleton matrix material particles may have conductive additives (e.g., carbon black particles, carbon nanotubes, graphene, exfoliated graphite, carbon fibers, etc.) strongly (e.g., chemically) bonded to its outer surface in order to enhance stability of the electrode during electrochemical cycling.
[0149] As discussed above, in one example, porous carbon may serve as an electrically conductive skeleton matrix material. Porous carbons having most of the pore walls being a monolayer-thick have been found to work particularly well, providing both high pore volume and sufficiently high conductivity, while minimizing the volume that inactive carbon atoms occupy. Porous carbons with experimentally measured Brunauer, Emmett and Teller (BET) specific surface area above 500 m.sup.2/g (more preferably above 1000 m.sup.2/g, even more preferably above 1500 m.sup.2/g) have been found to work well as a skeleton matrix material. An example of a high surface porous carbon is activated carbon. It may be produced by pyrolysis of carbon-containing organic precursors, followed by its either chemical or physical activation (partial oxidation of the resulting carbon to enhance its pore volume and specific surface area). Conventionally produced activated carbon particles often exhibit a majority of pores in the range from 0.3 nm to 100 nm. In some designs, those porous carbons that have a majority of pores (preferably more than 70% by volume) in the range from about 1 nm to about 50 nm may be particularly useful as a skeleton matrix material. In some designs, porous carbon skeleton matrix particles may be heated (in vacuum or inert environment) to temperatures of about 700 C. to about 2000 C. to enhance their electrical conductivity, their resistance to fluorination, or their mechanical strength.
[0150] In order to achieve a high energy cell (and often the lowest cost per unit energy in such cells and most favorable performance), when a battery is constructed with the cathode comprising mixed LiF and metal compositions and exhibiting a high volumetric capacity (e.g., above about 700 mAh/cc on the electrode level, more preferably above about 800 mAh/cc, even more preferably above about 900 mAh/cc, even more preferably above about 1000 mAh/cc), in some designs it may be preferred for the battery anode to also exhibit comparably high (or higher) volumetric capacity (e.g., above about 800 mAh/cc, even more preferably above about 900 mAh/cc, and in some designs more preferably above about 1000 mAh/cc or even above about 1100 mAh/cc). In some designs, it may be preferable for such high capacity anodes to comprise Si. In some designs, such high capacity anodes may comprise 30 at. % of Si or more. In some designs, such high capacity anodes may comprise 5 at. % of C or more in order to achieve better stability and higher rates. In some designs, it may be preferable for such high capacity anodes to comprise Li in order to further enhance cell energy density. In some designs, such high capacity anodes may comprise 3 at. % of Li or more.
[0151] The description is provided to enable any person skilled in the art to make or use embodiments of the present invention. It will be appreciated, however, that the present invention is not limited to the particular formulations, process steps, and materials disclosed herein, as various modifications to these embodiments will be readily apparent to those skilled in the art. That is, the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention.