HK40025702B - Electrodes with silicon oxide active materials for lithium ion cells achieving high capacity, high energy density and long cycle life performance - Google Patents

Description

Electrodes with silicon oxide active materials for achieving high capacity, high energy density, and long cycle life performance in lithium-ion batteries

Cross-references to related applications

This application claims priority to U.S. Patent Application 15/948,160, filed April 9, 2018, entitled “Electrode with silicon oxide active material for achieving high capacity, high energy density and long cycle life performance of lithium-ion batteries,” and U.S. Provisional Patent Application 62/609,930, filed December 22, 2017, entitled “Electrode with silicon oxide active material for achieving high capacity and long cycle life performance of lithium-ion batteries and resulting high energy density battery,” both of which are incorporated herein by reference.

U.S. federally funded research and development

This invention was completed with the support of the U.S. government under the United States Advanced Battery Consortium (Project No. 14-2141-ABC), grant number DE-EE0006250, granted by the U.S. Department of Energy. The U.S. government holds certain rights to this invention.

Invention Field

This invention relates to the formation of a negative electrode incorporating a high-capacity silicon oxide active material while achieving good cycle performance. The invention also relates to a battery obtained by assembling these high-capacity, long-life negative electrodes into a battery with high energy density.

Background of the Invention

Lithium-ion batteries are widely used in consumer electronics due to their high energy density. For some current commercial batteries, the anode material can be graphite, and the cathode material can include lithium cobalt oxide ( _LiCoO₂ ), lithium manganese oxide _{( LiMn₂O₄} ), lithium iron phosphate ( _LiFePO₄ ), lithium nickel oxide ( _LiNiO₂ ), lithium nickel cobalt oxide ( _LiNiCoO₂ ), lithium nickel cobalt manganese oxide ( _LiNiMnCoO₂ ), lithium nickel cobalt aluminum oxide ( _LiNiCoAlO₂ ), etc. For the anode, lithium titanate is a good alternative to graphite with good cycling performance, but it has a lower energy density. Other graphite alternatives such as tin oxide and silicon have the potential to provide improved energy density. However, some high-capacity anode materials have been found to be commercially unsuitable due to high irreversible capacity loss and poor discharge and recharge cycling associated with structural changes and unusually large volume expansion, especially for silicon, where the structural changes and unusually large volume expansion are related to lithium intercalation/alloying. Structural changes and large volume changes can compromise the structural integrity of the electrodes, thereby reducing cycle efficiency.

Invention Overview

In a first aspect, the present invention relates to a lithium-ion battery comprising a negative electrode, a positive electrode, a separator between the negative and positive electrodes; an electrolyte comprising a lithium salt and a non-aqueous solvent; and a container encapsulating other battery components. The negative electrode may comprise about 75% to about 92% by weight of an active material, about 1% to about 7% by weight of nanoscale conductive carbon, and about 6% to about 20% by weight of a polymer binder, wherein the active material may comprise about 40% to about 95% by weight of a silicon oxide-based material and about 5% to about 60% by weight of graphite. In some embodiments, the positive electrode comprises a nickel-rich lithium nickel cobalt manganese oxide, conductive carbon, and a polymer binder, wherein the nickel-rich lithium nickel cobalt manganese oxide is approximately represented by the formula LiNi _{_x}_{Mn_y}Co_{_z}O_₂, where x + y + z ≈ 1, 0.45 ≤ x, 0.025 ≤ y ≤ 0.35, and 0.025 ≤ z ≤ 0.35.

In another aspect, the present invention relates to a negative electrode for a lithium-ion battery, the negative electrode comprising about 78 wt% to about 92 wt% of an active material, about 1 wt% to about 7 wt% of nanoscale conductive carbon, and about 6 wt% to about 20 wt% of a polymer binder, wherein the polymer binder comprises at least about 50 wt% of polyimide and at least about 5 wt% of a different second polymer binder with an elastic modulus not greater than about 2.4 GPa. In some embodiments, the different second polymer binder has an elongation of at least about 35%.

In another aspect, the present invention relates to a lithium-ion battery comprising: a negative electrode comprising silicon oxide, graphite, nanoscale conductive carbon, and a polymer binder; a positive electrode comprising lithium nickel cobalt manganese oxide, conductive carbon, and a polymer binder; a separator between the negative and positive electrodes; an electrolyte comprising a lithium salt and a non-aqueous solvent; and a container encapsulating other battery components. In some embodiments, the lithium-ion battery has an energy density of at least 235 Wh/kg when discharged at 30°C at a rate of C/3 from a selected charging voltage of at least about 4.25 V to 2.5 V, and when cycled at a rate of C/3 between 2.3 V and a selected charging voltage from the 5th cycle to the 450th cycle, the capacity of the lithium-ion battery at 450 cycles is at least about 80% of the capacity at the 5th cycle.

In other aspects, the present invention relates to a lithium-ion battery comprising: a negative electrode comprising silicon oxide, graphite, nanoscale conductive carbon, and a polymer binder; a positive electrode comprising lithium nickel cobalt manganese oxide, conductive carbon, and a polymer binder; a separator between the negative and positive electrodes; an electrolyte comprising a lithium salt and a non-aqueous solvent; and a container encapsulating other battery components. In some embodiments, the lithium-ion battery has an energy density of at least 235 Wh/kg when discharged from 4.35 V to 2.3 V at a rate of C/3 at 30 °C, and when cycled from the 5th cycle to the 450th cycle at a rate of C/3 between 2.3 V and 4.35 V, the capacity of the lithium-ion battery at 450 cycles is at least about 80% of the capacity at the 5th cycle.

Brief description of the attached diagram

Figure 1 is an unfolded diagram of a pouch battery with battery cells that are separate from the two parts of the pouch casing.

Figure 2 is a perspective lower surface view of the assembled pouch cell of Figure 1.

Figure 3 is a bottom view of the pouch battery in Figure 2.

Figure 4 is an illustration of one embodiment of a battery cell including an electrode stack.

Figure 5 is a graph of the specific capacity (as a function of cycle number) of a set of half-cells (lithium foil electrodes) with a set of electrodes having active materials having SiO _x /Si/C composite material and optional selected blend amounts of graphite.

Figure 6 is a graph of the specific capacity (as a function of cycle number) of an all-button cell having a blend of lithium metal oxides for the positive electrode active material and a set of negative electrodes having a SiO _x /Si/C composite material and optional selected blend amounts of graphite.

Figure 7 is a graph of the standardized specific capacity (as a function of the cycle) based on the specific capacity graph in Figure 6.

Figure 8 is a graph showing the specific capacity (as a function of cycles) of a group of half-cells with blends of active materials and different nanoscale carbon conductive additives.

Figure 9 is a graph of the specific capacity (as a function of cycles) of a silicon-based electrode with a blend of lithium metal oxides for the positive electrode active material, as shown in Figure 8.

Figure 10 is a graph of the standardized specific capacity (as a function of the cycle) of the specific capacity in Figure 9.

Figure 11 is a graph of the specific capacity (as a function of cycles) of a half-cell with electrodes containing different polymer binder formulations.

Figure 12 is a graph showing the specific capacity (as a function of cycles) of a full cell with the corresponding negative electrode and positive electrode containing a lithium metal oxide blend used in the half cell of Figure 11.

Figure 13 is a graph of the normalized capacity (as a function of cycles) of the full cell in Figure 12.

Figure 14 is a graph of the specific capacity of a half-cell with alternative active material formulations relative to the batteries in Figures 11-13, wherein the electrodes contain five different formulations with four different binder compositions.

Figure 15 is a graph showing the specific capacity (as a function of cycles) of a full cell with the corresponding negative electrode and positive electrode containing a lithium metal oxide blend used in the half cell of Figure 14.

Figure 16 is a graph of the normalized capacity (as a function of cycles) of the full cell in Figure 15.

Figure 17 is a set of graphs showing the battery voltage (as a function of specific capacity) of a coin cell with a negative and a positive electrode cyclically within the voltage window indicated in the figure, wherein the negative electrode has a blend of silicon oxide/carbon composite particles and graphite, and the positive electrode has LiNi _0.6 Mn _0.2 Co _0.2 O ₂ (NMC622) or LiNi _0.8 Mn _0.1 Co _0.1 O ₂ (NMC811).

Figure 18 is a set of graphs showing the discharge specific capacity (as a function of the number of cycles) of the three button cell battery types indicated in Figure 17.

Figure 19 is a set of graphs showing the standardized discharge specific capacity (as a function of cycle number) of three coin cells used to produce the results depicted in Figure 18.

Figure 20 is a graph of the battery voltage (as a function of specific capacity) during the first cycle of a coin cell with an NMC811 series positive electrode and a silicon oxide-graphite blend negative electrode at a charge and discharge rate of C/20.

Figure 21 is a graph of the voltage (as a function of specific capacity) of the coin cell used to generate Figure 20 during the second charge/discharge cycle with a charge rate of C/6 and a discharge rate of C/10.

Figure 22 is a graph of the discharge specific capacity (as a function of cycles) of two equivalent coin cells similar to those used to generate the graph in Figure 20.

Figure 23 is a graph of the normalized capacity (as a function of cycles) used to generate the button cell of Figure 22.

Figure 24 is a front view of a pouch cell designed to operate at approximately 21 Ah.

Figure 25 is a side view of the pouch battery in Figure 24.

Figure 26 is a front view of a pouch cell designed to operate at approximately 11 Ah.

Figure 27 is a side view of the pouch battery in Figure 26.

Figure 28 is a graph of the normalized capacity (as a function of cycles) of the pouch cell shown in Figures 24-27.

Figure 29 is a graph of the specific energy (as a function of the number of cycles) of a pouch cell using a graphite-blended silicon oxide/carbon and NMC622 cathode within a voltage window from 4.3V to 2.3V.

Figure 30 is a graph of the normalized specific energy (as a function of cycle number) of the two pouch cells used to produce the results in Figure 29.

Figure 31 is a graph of the discharge capacity (as a function of cycle number) of the two pouch cells used to produce the results in Figure 29.

Figure 32 is a graph of the normalized discharge capacity (as a function of cycle number) of the two pouch cells used to produce the results in Figure 29.

Figure 33 is a graph of the specific energy (as a function of cycles) of two pouch cells with a “11Ah design” using a positive electrode with NMC811 active material and an improved silicon oxide anode as described herein.

Figure 34 is a graph of the normalized specific energy (as a function of cycle) of the battery used to generate the graph in Figure 33.

Figure 35 is a graph of the battery capacity (as a function of the cycle) used to generate the graph in Figure 33.

Figure 36 is a graph of the normalized capacity of the battery (as a function of cycles) used to generate the graph in Figure 33.

Invention Details

Improvements in electrode design have enabled longer-term cycling performance while leveraging the high specific capacity of silicon-based active materials such as silicon oxides. To achieve cycle stability, specific features of the anode (negative electrode) design have been found to be appropriately considered to achieve unexpectedly improved cycle performance. In particular, the active material can be designed to include blends of silicon-based active materials but with a significant graphite component. Furthermore, binder properties have been found to contribute significantly to cycle stability in some embodiments, and advantageous blends of high-mechanical-strength polyimides with more elastic polymers are described herein. Although graphite is conductive, nanoscale carbon can also be used appropriately to provide conductivity to the electrode, providing unexpected improvements in cycle performance. These improvements can then be used individually or in combination to form a negative electrode with desired capacity while achieving significant cycle stability. In some electrode embodiments, batteries achieving greater than 600 charge/discharge cycles while retaining at least 80% of the initial (post-formation) discharge specific capacity can be formed. These negative electrode improvements have also been found to be compatible with the applicant's battery design, thereby achieving correspondingly high energy densities for the corresponding batteries using appropriate positive electrode designs. A high-energy-density battery with good cycling performance is described, featuring a cathode material incorporating a nickel-rich lithium nickel cobalt manganese oxide cathode active material. Therefore, batteries with long-term cycle stability can be formed in larger battery forms suitable for automotive applications and other suitable applications, while achieving an initial post-formation energy density of at least approximately 235 Wh/kg.

Lithium has been used in both primary and secondary batteries. A striking characteristic of lithium used in batteries or battery packs is its light weight and the fact that it is the most electropositive metal, and these characteristics can also be advantageously incorporated into lithium-based batteries. Certain forms of metals, metal oxides, and carbon materials are known to incorporate lithium ions from the electrolyte into their structure through intercalation, alloying, or similar mechanisms. The positive electrode of a lithium-based battery typically contains an active material that is reversibly intercalated with lithium or alloyed with lithium. Lithium-ion batteries generally refer to batteries in which the negative electrode active material is also a lithium-intercalated/alloyed material. Unless explicitly distinguished as used herein and for convenience, the terms cell and battery pack, and their variants, are used interchangeably.

The batteries described herein are lithium-ion batteries using a non-aqueous electrolyte solution containing lithium cations and suitable anions. For rechargeable lithium-ion batteries, oxidation occurs at the cathode (positive electrode) during charging, where lithium ions are extracted and electrons are released. During discharge, reduction occurs at the cathode, where lithium ions are inserted and electrons are consumed. Similarly, during charging, reduction occurs at the anode (negative electrode), where lithium ions are absorbed and electrons are consumed, and during discharge, oxidation occurs at the anode, where lithium ions and electrons are released. Unless otherwise specified, the performance values mentioned herein are at room temperature, approximately 23 ± 2 °C. As described below, some tests of silicon-based active materials are performed in lithium-ion batteries with lithium metal electrodes (referred to as half-cells) or in lithium-ion batteries with a positive electrode containing lithium metal oxide (referred to as full-cells). In half-cells with silicon-based electrodes, the lithium electrode acts as the negative electrode, and the silicon electrode acts as the positive electrode, which is the opposite of its usual role as the negative electrode in lithium-ion batteries.

The term "element" in this text is used in its conventional way to refer to a member of the periodic table, wherein if an element is in a composition, it has a suitable oxidation state, and wherein when stated as being in elemental form, it is in its elemental form M ^₀ . Therefore, a metallic element is generally simply a metallic state in its elemental form, or a suitable alloy of metals in their elemental form. In other words, apart from metal alloys, metal oxides or other metal compositions are generally nonmetals.

When lithium-ion batteries are in use, the absorption and release of lithium from the positive and negative electrodes cause structural changes in the electroactive materials. As long as these changes are inherently reversible, the material's capacity does not change with cycling. However, the capacity of the active material has been observed to decrease to varying degrees with cycling. Therefore, after several cycles, the battery performance degrades below acceptable levels, necessitating battery replacement. Furthermore, irreversible capacity loss typically occurs during the first cycle of a battery, which is significantly greater than the capacity loss in each subsequent cycle. Irreversible capacity loss (IRCL) is the difference between the initial charge capacity and the first discharge capacity of a new battery. Lithium metal oxide-based cathodes may exhibit some IRCL, leading to some compensation for the negative electrode in terms of available lithium for cycling. Irreversible capacity loss can result in a corresponding decrease in the battery's capacity, energy, and power due to changes in the battery materials during the initial cycling period.

Elemental silicon and other silicon-based active materials have attracted considerable attention as potential anode materials due to silicon's very high specific capacity in terms of lithium uptake and release. Elemental silicon forms alloys with lithium, which theoretically can have lithium contents corresponding to more than 4 lithium atoms/silicon atoms (e.g., _Li₄₄Si ). Therefore, the theoretical specific capacity of silicon is in the range of 4000 to 4400 mAh/g, significantly greater than the theoretical capacity of graphite (approximately 370 mAh/g). Graphite is believed to have lithium intercalation reaching a level of approximately 1 lithium atom for every 6 carbon atoms ( _LiC₆ ). Furthermore, elemental silicon, silicon alloys, and silicon composites can exhibit low potentials relative to lithium metal, similar to graphite. However, silicon undergoes very large volume changes when alloyed with lithium. Large volume expansions of two to three times or more of the original volume have been observed, and these large volume changes are associated with a significant decrease in the cycle stability of batteries with silicon-based anodes. It has also been found that silicon suboxides (i.e., SiO _{_x}, x<2) are ideal active materials for lithium-ion batteries, exhibiting high specific capacity relative to lithium alloys in some embodiments. The mention of silicon suboxides acknowledges silicon dioxide as a fully oxidized form of silicon. Unless specifically stated otherwise, for convenience, silicon suboxides may generally be referred to as silicon oxides, and are not limited to silicon monoxide (SiO).

In the embodiments of particular interest, the silicon-based active material may comprise elemental silicon and/or low-valent silicon oxides as the primary active material. Low-valent silicon oxides have been found to be particularly effective in achieving longer cycle stability. To stabilize the silicon-based active material and to improve conductivity, carbon can be incorporated into the composite active material. Long-term cycle stability remains challenging for carbon composites containing nanoscale elemental silicon and/or silicon oxides, but the applicant has achieved moderate cycle stability for consumer electronics applications, as described above. Longer cycle stability is described herein based on mixtures of electroactive graphite and silicon-based composites, along with other electrode design improvements. As discussed in detail below, stable silicon-based electrodes may also incorporate additional conductive materials, such as nanoscale carbon, and binder blends that also significantly contribute to improved cycle stability.

The active materials used in lithium-ion secondary batteries as described herein typically include, for example, positive electrode (i.e., cathode) active materials with a moderately high average voltage relative to lithium and silicon-based active materials for the negative electrode (i.e., anode). Generally, a variety of cathode materials can be used. For example, commercially available cathode active materials can be used with existing commercial production availability. Such cathode active materials include, for example, lithium cobalt oxide ( _LiCoO₂ ), LiNi¹ _/ ³Mn¹ _{/³Co¹ / ³O₂} (L333 or NMC111), _LiNiCoAlO₂ (NCA), other lithium nickel manganese cobalt oxides (NMC), _LiMn₂O₄ (lithium manganese oxide _spinel ), their modified forms, or mixtures thereof.

Nickel-rich lithium nickel cobalt manganese oxide (LiNi _{_x}Mn_{_y}Co _{_z} O _₂ , 0.45≤x, 0.05≤y, z≤0.35) is of interest due to its lower cost and lower flammability risk compared to lithium cobalt oxide, as well as its ability to cycle at higher voltages. The results show that nickel-rich lithium nickel manganese cobalt oxide active materials paired with improved silicon-based anodes form batteries with good cycle stability and high energy density. Furthermore, for example, U.S. Patent 8,389,160 (hereinafter referred to as '160 Patent) entitled "Cathode Material for Lithium-ion Batteries with High Discharge Specific Capacity and Method for Synthesizing These Materials" by Venkatachalam et al., and U.S. Patent 8,465,873 (hereinafter referred to as '873 Patent) entitled "Cathode Material for High Discharge Capacity Lithium-ion Batteries" by Lopez et al., describe recently developed materials with high specific capacity, having a layered crystal structure and being lithium-rich relative to the _LiMO2 (M = non-lithium metal) reference composition, both of which are incorporated herein by reference. As discussed further below, blends of lithium-rich + manganese-rich NMC and nickel-rich NMC cathode active compositions have been found to provide particularly desirable battery performance and good cycle stability.

Specifically, desirable cycling results can be obtained from nickel-rich lithium nickel manganese cobalt oxides (N-NMC), which can be represented by the formula LiNi _{_x}Mn _{_y} Co _{_z} O_₂, where x ≥ 0.45 and x + y + z ≈ 1. Commercially available formulations of these compounds include, for example, LiNi _{_0.5} Mn_0.3Co _{_0.2 O ₂} (BASF), LiNi _{_0.6}Mn _{_0.2} Co_0.2O_₂ (Targray, Canada), and LiNi _{_0.8}Mn _{_0.1} Co _{_0.1} O_₂ (Targray, Canada and LG Chemical). In the industry, NCM and NMC are used interchangeably when cobalt and manganese are listed in the corresponding order, and the statements are equivalent and based solely on personal preference. Improved cycling stability can be achieved using blends of nickel-rich NMC and lithium-rich + manganese-rich lithium nickel manganese cobalt oxides (which can be referred to as high-capacity manganese-rich compositions). Typically, electrodes with a combination of electroactive compositions contain significant amounts of both materials, usually at least about 5% by weight of each relative to the total active material of the electrode. Battery performance indicates that blends of these active materials cycle better than individual N-NMC materials at higher charging voltages, while the N-NMC materials offer other desirable battery performance characteristics.

Positive electrode active materials can have stabilizing coatings. Stabilizing nanocoatings for positive electrode active materials are further described in U.S. Patent Application 2011/0111298, entitled "Coated Positive Electrode Material for Lithium-ion Batteries," by Lopez et al.; U.S. Patent 8,535,832, entitled "Metal Oxide Coated Positive Electrode Material for Lithium-based Batteries," by Karthikeyan et al.; and U.S. Patent 8,663,849, entitled "Metal Halide Coating on Positive Electrode Material for Lithium-ion Batteries and Corresponding Battery," by Venkatachalam et al., all of which are incorporated herein by reference. Specifically, these stabilizing coatings can be used to significantly improve the cycling performance of lithium-rich + manganese-rich nickel-manganese-cobalt oxide electrodes.

As stated above, silicon-based electrodes present challenges in achieving cycling performance suitable for commercial applications. For consumer electronics applications, a reasonable cycling target might be approximately 250-450 cycles without unacceptable performance loss, but for transportation and similar higher-capacity applications, greater cycle stability is required. The applicant has achieved battery designs suitable for consumer electronics applications that achieve suitable performance using silicon-based anodes. These batteries are described in U.S. Patent Application 2015/0050535 (hereinafter referred to as '535 application'), entitled "Lithium-ion Battery with High-Capacity Anode Material and Good Cycling Performance for Consumer Electronics," published by Amiruddin et al., which is incorporated herein by reference. The novel battery design described herein provides cycling performance exceeding the target cycle stability for consumer electronics while simultaneously achieving performance suitable for transportation and other high-capacity applications.

Regarding silicon, oxygen-deficient silicon oxides, such as silicon oxide SiO_{_x}, 0.1 ≤ x ≤ 1.9, can be intercalated with lithium/alloyed with lithium to enable oxygen-deficient silicon oxides to function as active materials in lithium-ion batteries. Silicon oxides can be doped with a relatively large amount of lithium to allow the material to exhibit a large specific capacity. However, the capacity of silicon oxides is generally observed to decay rapidly with battery cycling. Commercial silicon-based materials containing SiO_x that can be combined with carbon and silicon nanocrystals are available from several suppliers, including Alfa Aesar (USA), Sigma-Aldrich (USA), Shin-Etsu (Japan), Osaka Titanium Corporation (Japan), and Nanostructured and Amorphous Materials Corp. (USA). Further specific suitable formulations of silicon-based compositions are described below. The applicant has achieved cycle stability of silicon oxide-based active materials using the electrode formulations described herein. In some embodiments, a negative electrode comprising a combination of graphitic carbon active materials and silicon-based active materials can ideally extend cycle life with an acceptable decrease in specific capacity, and the excellent cycle performance described herein utilizes such active material blends.

To achieve the results described herein, various design improvements were examined independently or in combination to provide improved cycle performance, and for at least some embodiments, combinations of specific electrode features can provide unexpectedly synergistic performance improvements for longer cycle stability. Specifically, the negative electrode can be designed to contain a composite binder with high tensile strength while introducing some elongation. Nanoscale conductive carbon, such as carbon nanotubes, carbon black, carbon nanofibers, or combinations thereof, has been found to improve the cycle performance of negative electrodes with silicon-based active materials as conductive electrode additives. Electrolytes with excellent cycle performance at higher voltages >4.3V can be provided. These features can be combined with designs for electrode loading and density that provide good energy density-based performance for resulting consumer electronics battery designs. Cycling can be further improved by adding supplemental lithium to the battery and/or by adjusting the balance of active materials in each electrode.

The graphite component of the active material blend in the negative electrode can provide conductivity. However, it has been found that an appropriate amount of nanoscale carbon can further stabilize the negative electrode in terms of cycling. Nanoscale carbon can take the form of carbon nanotubes, carbon nanofibers, or carbon nanoparticles such as carbon black. The usefulness of nanoscale conductive carbon for the cycling stability of silicon-based negative electrodes has previously been found, as described in U.S. Patent 9,190,694B2, entitled "High-Capacity Anode Material for Lithium-Ion Batteries," by Lopez et al., and U.S. Patent 9,780,358B2, entitled "Battery Design Using High-Capacity Anode and Cathode Materials," by Masarapu et al., both of which are incorporated herein by reference. Typically, the electrode contains at least about 1% by weight of nanoscale conductive carbon to achieve stable cycling.

In lithium-ion batteries, reactive lithium for cycling is typically provided in the positive electrode active material, which is transferred to the negative electrode during the initial charge of the battery and then available for discharge. Silicon-based negative electrodes often exhibit large irreversible capacity losses during the first charge of the battery. This capacity loss is often associated with irreversible changes corresponding to the material during the first charge. For example, a solid electrolyte interface (SEI) layer forms along with the negative electrode active material as a result of reactions with the typical electrolyte used in the battery. If a stable SEI layer forms, it can stabilize the battery during cycling. Other irreversible changes can be presumed with respect to the silicon-based active composition. The irreversible capacity loss in the first cycle is typically significantly greater than any per-cycle capacity loss associated with subsequent cycles, although the second, third, and subsequent cycles may also have large per-cycle capacity losses because the initial changes are carried over to the earlier cycles rather than being completed entirely in the first cycle. Large irreversible capacity losses (IRCL) can reduce cycle capacity and the battery's energy and power output during cycling. In larger battery configurations, capacity may increase at lower cycle counts due to practical effects, such as improved electrolyte permeation through the electrode stack.

To reduce energy and power output losses in batteries due to irreversible capacity loss, supplemental lithium can be added to provide additional lithium to the battery. The introduction of supplemental lithium can reduce the introduction of cathode active materials that do not cycle due to active lithium capacity loss associated with IRCL. Supplemental lithium refers to active lithium, different from the cathode active material, introduced directly or indirectly into the battery to replace lithium lost in the irreversible process and to provide other beneficial effects. The applicant has found that supplemental lithium provided in an amount greater than that corresponding to compensating for irreversible capacity loss can further stabilize the cycle. In the case of a lithium-rich + manganese-rich nickel-manganese-cobalt oxide cathode active material, this cycle stabilization is described in U.S. Patent 9,166,222 (hereinafter referred to as '222 Patent') entitled "Lithium-ion Battery Using Supplemental Lithium" by Amiruddin et al., which is incorporated herein by reference.

Various methods can be used to introduce supplemental lithium, including, for example, adding a lithium active material (e.g., lithium metal powder or foil) to the negative electrode, adding a sacrificial lithium source to the positive electrode, incorporating a sacrificial lithium electrode into the battery structure, and electrochemical pre-lithiation of the negative electrode. These methods are further described in U.S. Patent Application 2011/0111294 (hereinafter referred to as '294 application), entitled "High-Capacity Anode Material for Lithium-Ion Batteries" by Lopez et al., both of which are incorporated herein by reference. In some embodiments, it has been found that electrochemical methods can be convenient, such as those described in PCT Application WO 2013/082330, entitled "Method for Alkalinizing the Anode" by Grant et al., which is also incorporated herein by reference. Typically, supplemental lithium can be introduced in a certain amount to compensate for a portion, almost all, or a greater amount than the irreversible capacity loss, but it is typically no more than 30% higher than the capacity of the negative electrode active material than the irreversible capacity loss. In some implementations, supplemental lithium can compensate for approximately 90% to approximately 200% of the irreversible capacity loss during the first cycle of the anode.

The applicant's prior work has found that the use of a high tensile strength polymer binder significantly improves the cycling performance of silicon anodes, and the high tensile strength polymer binder can be achieved by using a suitable polyimide. Specifically, the tensile strength of the polymer binder can be at least about 60 MPa. To extend cycling stability even further, polymer binder blends have been found to provide further improved cycling performance. One component of the polymer binder blend can be a high tensile strength polymer, such as a polyimide, and the second polymer can have a lower Young's modulus (elastic modulus) value to provide a more elastic polymer, such as polyvinylidene fluoride, carboxymethyl cellulose, styrene-butadiene rubber, lithium-ionized polyacrylic acid, or mixtures thereof. In addition to providing tensile strength, the polymer binder should also provide good adhesion so that the electrode remains laminated with the current collector. The desired blend can contain at least about 50 wt% of a high tensile strength polymer and at least about 5 wt% of a polymer binder, the polymer binder having a Young's modulus not exceeding about 2.4 GPa and, for some embodiments, an elongation of at least about 35%.

Anode design typically involves balancing multiple factors to achieve target energy and power densities while still providing reasonable cycle life. As seen in the results of the following examples, batteries with silicon-based anode active materials can cycle over 600 times while retaining over 80% of their capacity. Meanwhile, realistic anode designs can be matched with reasonable cathode designs to achieve good cycle life and high energy density values. The battery design and the balance of design features used to achieve these achievements are described in detail below.

Typically, the battery designs described herein are suitable for cylindrical batteries or more rectangular or prismatic shapes. Cylindrical batteries typically have wound electrode structures, while prismatic batteries can have wound or stacked electrodes. Generally, to achieve the desired performance using an electrode design with appropriate electrode loading and density, a battery can include multiple electrodes of various polarities, which can be stacked together with a separator material between the battery electrodes. Wound electrodes can provide similar effects with reasonable internal resistance due to electronic conductivity and ion migration, as well as good stacking of the electrodes within a suitable container. Battery size typically affects the total capacity and energy output of the battery. The designs described herein are based on achieving ideal high energy density while providing ideal cycling performance for batteries based on silicon-based active materials.

Electrode structure

Battery electrodes comprise active materials, binders, and conductive additives. The electrodes are formed into sheets, dried, and pressed to achieve the desired density and porosity. Electrode sheets are typically formed directly onto a metallic current collector, such as a metal foil or thin metal mesh. For various battery structures, electrode layers are formed on both sides of the current collector to provide ideal performance in the assembled battery or battery pack. Electrode layers on each side of the current collector can be considered elements of the same electrode structure because they are at the same potential in the battery; however, the current collector itself, although part of the electrode structure, is generally not considered part of the electrode because it is electrochemically inert. Therefore, when referring to the physical aspects of an electrode, it usually refers to the single electrode layer within the electrode structure. Conductive current collectors facilitate the flow of electrons between the electrode and the external circuitry.

In some implementations, when high load levels are used on the positive or negative electrode, the electrode density can be reduced to provide good electrode cycle stability. Electrode density is a function of compression stress within a reasonable range. Generally, electrode density cannot be arbitrarily increased without sacrificing performance with respect to load levels while simultaneously achieving the desired cycle performance and capacity at higher discharge rates. The characteristics of specific negative and positive electrode layers are provided in the following sections.

In some embodiments, the current collector may be formed of nickel, aluminum, stainless steel, copper, etc. Electrode material may be cast as a thin film onto the current collector. The electrode material with the current collector is then dried, for example, in an oven, to remove solvent from the electrode. In some embodiments, the dried electrode material in contact with the current collector foil or other structure may be subjected to a pressure of about 2 to about 10 kg/ ^cm² . The thickness of the current collector for the positive electrode may be about 5 micrometers to about 30 micrometers, in other embodiments about 10 micrometers to about 25 micrometers, and in further embodiments about 14 micrometers to about 20 micrometers. In one embodiment, an aluminum foil current collector is used for the positive electrode. The thickness of the current collector for the negative electrode may be about 2 micrometers to about 20 micrometers, in other embodiments about 4 micrometers to about 14 micrometers, and in further embodiments about 6 micrometers to about 10 micrometers. In one embodiment, copper foil is used as the current collector for the negative electrode. Those skilled in the art will recognize that other ranges of current collector thickness within the explicitly defined ranges described above are to be considered and are within the scope of this disclosure.

negative electrode

The basic electrode design comprises a blend of an active composition, a polymer binder, and a conductive diluent. As described above, in some embodiments, the improved electrode design may involve a blend of a polymer binder blend and an active composition, along with a nanoscale conductive carbon additive. The active material blend may contain a majority of a silicon-based active material, such as a silicon oxide composite, and at least 10% by weight of distinct graphite. Furthermore, it has been found that stabilization of electrode cycling using silicon-based active materials can be achieved by using a blend of polyimide, which provides high mechanical strength, with a portion of a more deformable polymer that still provides good electrode performance in the synergistic binder blend. Although graphite can provide conductivity to the electrode, it has also been found that, in some embodiments, a certain amount of distinct nanoscale conductive carbon can be important for the ability to generate a long-cycle negative electrode. Typically, nanoscale conductive carbon is not considered electrochemically active, while graphite is. These improved design aspects are then incorporated into electrodes with further improvements to the previously discovered silicon-based electrode design.

Significant attention has been directed toward high-capacity silicon-based anode active materials. For batteries containing significant amounts of silicon, silicon-based active materials have generally not achieved cycle stability suitable for automotive applications. '535 applications have demonstrated successful cycling for applications such as consumer electronics, where cycling up to approximately 200 to 300 cycles is achieved with at least 80% of the initial capacity. The applicant has been particularly successful in achieving cycle stability using materials primarily based on silicon oxide composites. This work presents an electrode that can smoothly cycle for over 600 cycles over a wide voltage range at reasonable rates without capacity degradation below 80%. Therefore, this work relates to extending cycle stability to a range suitable for automotive applications.

As described herein, improved cycling results were obtained using an active composition blended with silicon-based active materials and graphite carbon. Typically, the total capacity of the negative electrode blended active material can be at least about 750 mAh/g, in further embodiments at least about 900 mAh/g, in other embodiments at least about 1000 mAh/g, and in still other embodiments at least about 1100 mAh/g when cycling relative to lithium metal at a C/3 rate from 5 mV to 1.5 V. The blended active material can contain at least about 40 wt% silicon-based active material, in further embodiments at least about 50 wt% silicon-based active material, in other embodiments about 55 wt% to about 95 wt% silicon-based active material, and in still other embodiments about 60 wt% to about 90 wt% silicon-based active material. Accordingly, the blended active material may comprise about 5% by weight to about 60% by weight of graphite, in a further embodiment about 7% by weight to about 50% by weight of graphite, in another embodiment about 8% by weight to about 45% by weight of graphite, and in other embodiments about 10% by weight to about 40% by weight of graphite. Those skilled in the art will recognize that further ranges of discharge specific capacity and concentration of the silicon-based active material within the aforementioned defined ranges are to be considered and are within the scope of this disclosure.

As described above and in the detailed description below, suitable silicon-based active materials can include composite materials having a carbon component. Silicon-based active materials are discussed in detail in the following sections. Unlike blends involving mixtures held together by polymer binders, composite materials refer to granular materials whose components are tightly bonded together at a suitable scale to form a monolithic material with effective homogeneity. Composite material components can include, for example, silicon, oxygen, carbon, etc. While not wishing to be theoretically limited, it is generally considered that the carbon component of silicon-based composite materials is electrochemically active and is typically not graphite, although activity is an abstract concept given the tight bonding in composite materials, and the crystal structure can be extremely complex and difficult to evaluate. In any case, it will be readily understood by those skilled in the art that the carbon component of a composite material is distinguishable from the distinct graphite in a composite material not included in an active material blend. The following examples are based on commercial composite material compositions believed to contain predominantly low-valence silicon oxides with some amounts of elemental silicon crystals and elemental carbon in the combined composite granular material.

Graphite is commercially available in both natural and synthetic forms, and suitable graphite includes both natural and synthetic graphite. Graphite is a crystalline form of carbon having sheet-like covalently bonded carbon. As used herein, graphite refers to graphitic carbon that does not require perfect crystallinity, and some natural graphite materials may have some crystalline impurities. However, graphite generally refers to materials whose structure is dominated by graphite, as will be recognized in the art. Graphite is conductive along the planes of the covalently bonded carbon sheets stacked in the crystal. Crystalline carbon in graphite form can intercalate lithium, making it a proven electrochemically active material for lithium-ion batteries.

The average particle size of the graphite particles can be from about 1 micrometer to about 30 micrometers, in further embodiments from about 1.5 micrometers to about 25 micrometers, and in other embodiments from about 2 micrometers to about 20 micrometers. Generally, it is ideal for graphite to not contain particles larger than the electrode thickness to avoid an uneven electrode surface, and graphite particles significantly smaller than micrometers are likely to be less crystalline. In some embodiments, the D50 (mass median diameter) of the graphite carbon can be from about 5 micrometers to about 50 micrometers, in further embodiments from about 7 micrometers to about 45 micrometers, and in other embodiments from about 10 micrometers to about 8 micrometers to about 40 micrometers. Additionally, in some embodiments, the BET surface area of the graphite carbon active material (which can be evaluated according to ISO 4652) can be from about 1 ^m² /g to about 100 ^m² /g, in further embodiments from about 5 ^m² /g to about 85 ^m² /g, and in other embodiments from about 7.5 ^m² /g to about 60 ^m² /g. Those skilled in the art will recognize that other ranges of particle size and surface area of the graphite carbon active material are to be considered and are within the scope of this disclosure. In contrast, conductive carbon black and the like (which are called paracrystalline) typically have a surface area of at least about 40 ^m² /g to 1000 ^m² /g or greater.

Regarding polymer binders, the applicant has achieved reasonable cycling performance for silicon-based batteries using high tensile strength binders such as polyimide binders. See U.S. Patent 9,601,228 (hereinafter referred to as '228 Patent) entitled "Silicon Oxide-Based High-Capacity Anode Material for Lithium-Ion Batteries" by Deng et al., which is incorporated herein by reference. In some embodiments where longer cycling stability is achieved, polymer binder blends have been unexpectedly found to further stabilize the cycles. In particular, a second polymer or combination of polymers providing a lower elastic modulus (corresponding to greater elasticity) can be blended with a high tensile strength polyimide. The binder blend typically contains at least about 50% by weight of polyimide, at least about 55% by weight in further embodiments, and about 60% to about 95% by weight in other embodiments. Similarly, binder blends typically contain at least about 5% by weight of a polymer having a lower elastic modulus, at least about 10% by weight in further embodiments, and about 12% to about 40% by weight in other embodiments, as further described below. Those skilled in the art will recognize that additional ranges of polymer amounts within the clearly defined ranges described above are to be considered and are within the scope of this disclosure. The polymers in the blends may be selected to be soluble in the same solvent.

Polyimides are polymers based on repeating units of the imide monomer structure. Polyimide polymer chains can be aliphatic, but for high tensile strength applications, the polymer backbone is typically aromatic, with the backbone extending along the nitrogen atoms of the polyimide structure. For silicon-based anodes exhibiting significant morphological changes during cycling, thermosetting polyimide polymers have been found ideal for high-capacity anodes, likely due to their high mechanical strength. The table below provides suppliers of high-tensile-strength polyimide polymers and their corresponding names.

The tensile strength of the polyimide polymer can be at least about 60 MPa, at least about 100 MPa in further embodiments, and at least about 125 MPa in other embodiments. Some commercial polyimides with high tensile strength may also have high elongation values, which are the amount of elongation the polymer withstands before tearing. In some embodiments, the elongation of the polyimide can be at least about 40%, at least about 50% in further embodiments, and at least about 55% in other embodiments. Tensile strength and elongation values can be measured according to procedures in ASTM D638-10 standard test method for tensile properties of plastics or ASTM D882-91 standard test method for tensile properties of thin plastic sheets, both of which are incorporated herein by reference. Based on values reported by commercial suppliers, results for polyimides from these alternative ASTM schemes appear similar to each other. Those skilled in the art will recognize that additional ranges of polymer properties within the clearly defined ranges described above are to be considered and are within the scope of this disclosure.

Suitable, more flexible polymer components can be selected that are electrochemically inert in the battery and compatible with processing using polyimide. In particular, suitable more flexible polymer components include, for example, polyvinylidene fluoride (PVdF), carboxymethyl cellulose (CMC), styrene-butadiene rubber (SBR), lithium-modified polyacrylic acid (LiPAA), or mixtures thereof. Regarding polymer properties, some important properties for high-capacity anode applications are outlined in the following table.

adhesive elongation Tensile strength (MPa) Elastic modulus (GPa) PVDF 5-50% 30-45 1.0-2.5 polyimide 30-100% 60-300 2.5-7 CMC 30-40% 10-15 1-5 SBR 400-600% 1-25 0.01-0.1 LiPAA 1-6% 90 1-4

PVDF refers to polyvinylidene fluoride, CMC refers to sodium carboxymethyl cellulose, SBR refers to styrene-butadiene rubber, and LiPAA refers to lithium-modified polyacrylic acid. PVDF, CMC, and SBR are commercially available from various sources. LiPAA can be made from LiOH and commercial polyacrylic acid (PAA). For example, a stoichiometric amount of LiOH can be added to a solution of PAA at a rate of one mole of LiOH per monomer unit of PAA. The formation and use of LiPAA are further described in Li et al., “Lithium polyacrylate as a binder for tin-cobalt-carbon negative electrodes in lithium-ion batteries,” Electrochemica Acta 55 (2010) 2991-2995, which is incorporated herein by reference.

Elongation refers to the percentage of elongation before the polymer tears. Typically, to accommodate silicon-based materials, an elongation of at least about 30% is required, at least about 50% in some embodiments, and at least about 70% in further embodiments. For polymer binder blends, for more elastic polymer binder components, an elastic modulus (alternatively referred to as Young's modulus or tensile modulus) of no more than about 2.4 GPa, no more than about 2.25 GPa in further embodiments, no more than about 2 GPa in other embodiments, and no more than about 1.8 GPa in yet another embodiment may be required. Those skilled in the art will recognize that further ranges of properties of the more elastic polymer components within the aforementioned defined ranges are to be considered and are within the scope of this disclosure.

To form electrodes, powders can be blended with polymers in a suitable liquid, such as a solvent used to dissolve the polymer. Polyimides and PVdF are typically treated in N-methylpyrrolidone (NMP), but other suitable organic solvents can be used. Water-treatable polyimides are commercially available and suitable for blending with a variety of other polymers. The granular components of the electrode, i.e., the active material and nanoscale conductive carbon, can be blended with a polymer binder in a solvent to form a paste. The resulting paste can then be pressed into electrode structures.

The loading of active material in the binder can be substantial. In some embodiments, the negative electrode has about 75% to about 92% by weight of negative electrode active material, in other embodiments about 77% to about 90% by weight of negative electrode active material, and in further embodiments about 78% to about 88% by weight of negative electrode active material. In some embodiments, the negative electrode has about 6% to about 20% by weight of polymer binder, in other embodiments about 7% to 19% by weight of polymer binder, and in further embodiments about 8% to 18% by weight of polymer binder. Additionally, in some embodiments, the negative electrode contains about 1% to about 7% by weight of nanoscale conductive carbon, in further embodiments about 1.5% to about 6.5% by weight, and in other embodiments about 2% to about 6% by weight of nanoscale conductive carbon. Those skilled in the art will recognize that further ranges of polymer loading within the aforementioned explicit ranges are to be considered and are within the scope of this disclosure.

For improved cycling performance of the negative electrode, nanoscale carbon additives or combinations thereof have been found to be particularly desirable. Nanoscale conductive carbon generally refers to at least two submicron-sized high-surface-area elemental carbon particles. Suitable nanoscale conductive carbons include, for example, carbon black, carbon nanotubes, and carbon nanofibers. In some embodiments, the nanoscale conductive carbon additive used in the negative electrode may include carbon nanotubes, carbon nanofibers, carbon nanoparticles (e.g., carbon black), or combinations thereof. In some embodiments, to achieve improved performance, the conductive additive may have a conductivity of at least about 40 S/cm, in some embodiments at least about 50 S/cm, and in further embodiments at least about 60 S/cm. Those skilled in the art will recognize that further ranges of particle loading and conductivity within the clearly defined ranges described above are to be considered and are within the scope of this disclosure.

Conductivity, the reciprocal of resistivity, can be reported by the distributor and is typically measured using specific techniques developed by the distributor. For example, the measurement of carbon black resistance is performed between two copper electrodes using Super P ^™ carbon black; see Timcal Graphite & Carbon, A Synopsis of Analytical Procedures, 2008, www.timcal.com. Suitable supplemental conductive additives can also be added to improve long-term cycling stability. Alternatively, some suppliers describe conductive carbon concentrations that achieve the conductive percolation threshold.

Carbon black refers to synthetic carbon materials and may alternatively be called acetylene black, furnace black, thermal black, or other names indicating the method of synthesis. Carbon black generally refers to amorphous carbon, but there are indications that at least some forms of carbon black contain small structural domains with short or medium-range sequences corresponding to the crystal structure of graphite or diamond; however, for practical applications, the material can be considered amorphous. Under ISO technical specification 80004-1 (2010), carbon black is a nanostructured material. Primary particles of carbon black can be on the order of tens of nanometers, but primary particles are typically hard-fused into chains or other aggregates, and the smallest dispersible unit can be considered to be about 80 nm to 800 nm, which is still submicron. Carbon black is commercially available, and it has been synthesized to provide the desired levels of conductivity, such as Timcal, Akzo Nobel, Shawinigan (Chevron-Phillips), and Black Pearls (Cabot).

Carbon nanofibers are high aspect ratio fibers that typically consist of sheets, cones, or other forms of graphene layers, while carbon nanotubes consist of graphene sheets folded into tubes. Carbon nanofibers can have diameters below 250 nm and are commercially available, for example, from Pyrograf Products, Inc., or from American Elements, Inc. Carbon nanotubes have been found to be ideal conductive additives, improving the cycling performance of both positive and negative electrodes. Single-walled or multi-walled carbon nanotubes are also available from American Elements, Inc. (CA, USA), Canano Technologies (China), Fuji, Inc. (Japan), Alfa Aesar (MA, USA), or NanoLabs (MA, USA).

The positive and negative electrodes used in the batteries described herein can have high active material loading levels and reasonably high electrode densities. For a given active material loading level, density is negatively correlated with thickness, such that electrodes with higher densities are thinner than electrodes with lower densities. The loading level is equal to the density multiplied by the thickness. In some embodiments, the negative electrode active material loading level of the battery negative electrode is at least about 1.5 mg/ ^cm² , in other embodiments about 2 mg/ ^cm² to about 8 mg/ ^cm² , in still other embodiments about 2.5 mg/ ^cm² to about 6 mg/ ^cm² , and in other embodiments about 3 mg/ ^cm² to about 4.5 mg/ ^cm² . In some embodiments, the active material density of the battery negative electrode is about 0.5 g/cc (cc = cubic centimeters ( ^cm³ )) to about 2 g/cc, in other embodiments about 0.6 g/cc to about 1.5 g/cc, and in still other embodiments about 0.7 g/cc to about 1.3 g/cc. Similarly, the average dry thickness of the silicon oxide electrode can be at least about 15 micrometers, at least about 20 micrometers in a further embodiment, and from about 25 micrometers to about 75 micrometers in another embodiment. The resulting silicon oxide electrode can exhibit a capacity/unit area of at least about 3.5 mAh/ ^cm² , at least about 4.5 mAh/ ^cm² in a further embodiment, and at least about 6 mAh/ ^cm² in another embodiment. Those skilled in the art will recognize that further ranges of active material loading levels and electrode densities within the aforementioned defined ranges are to be considered and are within the scope of this disclosure.

High-capacity silicon-based anode materials

Typically, the battery designs described herein are based on high-capacity anode active materials. Specifically, when cycled at a rate of C/10 relative to lithium metal from 0.005V to 1.5V, the specific capacity of the anode active material is typically at least about 800 mAh/g, at least about 900 mAh/g in further embodiments, at least about 1000 mAh/g in other embodiments, at least about 1150 mAh/g in some embodiments, and at least about 1400 mAh/g in other embodiments. As implicitly, the specific capacity of the anode active material can be evaluated in batteries with a lithium metal counter electrode. However, in the batteries described herein, the anode can exhibit reasonably comparable specific capacity when cycled relative to a high-capacity lithium metal oxide cathode active material. In batteries with non-lithium metal electrodes, the specific capacity of each electrode can be evaluated by dividing the battery capacity by the weight of each active material. As described herein, ideal cycling results can be obtained using a combination of silicon-based and graphitic carbon active materials, and good capacity has been observed.

Elemental silicon, silicon alloys, and silicon composites can possess low potentials relative to lithium metal, similar to graphite. However, elemental silicon typically undergoes very large volume changes when alloyed with lithium. Large volume expansions, ranging from two to four times the original volume or greater, have been observed, and these large volume changes are associated with a significant decrease in the cycle stability of batteries with silicon-based anodes.

In the batteries described herein, commercially available low-valent silicon oxides, composites of elemental silicon and carbon can be used. Additionally, other formulations of silicon-based anode active materials with high capacity and reasonable cycle performance have been developed. Several silicon-based compositions are described below, offering potential and promising alternatives to commercially available SiO-based compositions.

Furthermore, high-capacity silicon-based materials in the negative electrode of lithium-ion batteries may exhibit large irreversible capacity loss (IRCL) in some formulations during the first charge/discharge cycle. High IRCL in silicon-based anodes can consume a significant portion of the capacity available for the battery's energy output. Since the cathode (i.e., the positive electrode) supplies all lithium in conventional lithium-ion batteries, high IRCL in the anode (i.e., the negative electrode) can lead to low-energy batteries. To compensate for large anode IRCL, supplementary lithium can be added directly or indirectly to the negative electrode material to offset the IRCL. The use of supplementary lithium to improve the performance of silicon-based electrodes is also described in '294 application and '228 patent, both of which are cited above and incorporated herein by reference. The use of supplementary lithium in improved battery designs is further described below.

The anode of the battery described herein can utilize a nanostructured active silicon-based material to better accommodate volume expansion and thereby maintain the mechanical electrode stability and cycle life of the battery. Nanostructured silicon-based anode compositions are disclosed in U.S. Patent 9,139,441 ('441 Patent), entitled "Porous Silicon-Based Anode Material Formed Using Metal Reduction" by Angushamy et al. (these are incorporated herein by reference). Suitable nanostructured silicon can include, for example, nanoporous silicon and nanoparticle silicon. Additionally, nanostructured silicon can be formed into composites with carbon and/or alloys with other metal elements. The goal of designing improved silicon-based materials is to further stabilize the anode material during cycling, while maintaining high specific capacity and reducing irreversible capacity loss in the first charge-discharge cycle in some embodiments. Furthermore, pyrolytic carbon coatings have been observed to stabilize silicon-based materials in terms of battery performance.

Ideal high-capacity anode active materials can include porous silicon (pSi) based materials and/or composites of porous silicon based materials. Typically, pSi based materials comprise highly porous crystalline silicon, which offers a high surface area and/or high pore volume relative to bulk silicon. Although nanostructured porous silicon can be formed by various methods such as electrochemical etching of silicon wafers, nanostructured porous silicon obtained by metal reduction of silicon oxide powder has achieved particularly good battery performance. In particular, this material exhibits exceptionally good cycle performance while maintaining a high specific capacity. The formation of composites of pSi based materials with carbon-based materials or metals can further stabilize the anode mechanically to improve cycle performance. Further descriptions of pSi based materials from silicon oxide reduction can be found in the above-cited '441 patent.

Regarding composite materials, a nanostructured silicon component can be combined with, for example, carbon nanoparticles and/or carbon nanofibers within a tightly coupled composite material. For example, the components can be milled to form a composite material in which the materials are tightly associated. Typically, association is considered to have mechanical characteristics, such as softer silicon coated on or mechanically attached to a harder carbon material. In another or alternative embodiment, silicon can be milled together with metal powder to form an alloy, which may have a corresponding nanostructure. A carbon component can be combined with a silicon-metal alloy to form a multi-component composite material.

Additionally, carbon coatings can be applied to silicon-based materials to improve conductivity, and these coatings have been shown to stabilize silicon-based materials by improving cycling performance and reducing irreversible capacity loss. Ideally, carbon coatings can be formed by pyrolyzing an organic composition. The organic composition can be pyrolyzed at relatively high temperatures, such as about 800°C to about 900°C, to form a hard, amorphous coating. In some embodiments, the desired organic composition can be dissolved in a suitable solvent, such as water and/or a volatile organic solvent, for use in combination with the silicon-based component. The dispersion can be thoroughly mixed with the silicon-based composition. After drying the mixture to remove the solvent, the dried mixture of silicon-based material with the carbon precursor can be heated in an oxygen-free atmosphere to pyrolyze the organic composition, such as an organic polymer, some lower molecular weight solid organic compositions, etc., thereby forming a carbon coating.

Regarding silicon, oxygen-deficient silicon oxides, such as silicon oxide SiO_{_x}, 0.1 ≤ x ≤ 1.9, can be intercalated with lithium/alloyed with lithium to enable them to function as active materials in lithium-ion batteries. These oxygen-deficient silicon oxide materials are generally referred to as silicon oxide-based materials, and in some embodiments may contain varying amounts of silicon, silicon oxide, and silicon dioxide. Oxygen-deficient silicon oxides can be doped with a relatively large amount of lithium to allow the material to exhibit a high specific capacity. However, it has been observed that silicon oxides typically exhibit a rapid decrease in capacity with battery cycling, as observed with elemental silicon.

Silicon oxide compositions have been formulated into composite materials with high capacity and excellent cycling performance, as described in the aforementioned '228 patent. In particular, oxygen-deficient silicon oxides can be combined with conductive materials such as conductive carbon or metal powders to form composite materials, which unexpectedly and significantly improves cycling performance while providing high specific capacity values. Furthermore, milling silicon oxides into smaller particles, such as submicron structures, can further improve the material's properties.

Typically, a variety of composite materials can be used, and these can include silicon oxides, carbon components such as graphite particles (Gr), inert metal powders (M), elemental silicon (Si), particularly nanoparticles, pyrolytic carbon coatings (HC), carbon nanofibers (CNF), or combinations thereof. The component structure may or may not correspond to the structure of the components within the composite material. Therefore, the general composition of a composite material can be expressed as αSiO-βGr-χHC-δM-εCNF-φSi, where α, β, χ, δ, ε, and φ are relative weights and can be chosen such that α+β+χ+δ+ε+φ＝1. Typically, 0.35<α<1, 0≤β<0.6, 0≤χ<0.65, 0≤δ<0.65, 0≤ε<0.65, and 0≤φ<0.65. Certain subgroups within these composite material ranges are of particular interest. In some embodiments, a composite material having SiO and one or more carbon-based components is desirable, which can be represented by the formula αSiO-βGr-χHC-εCNF, where 0.35<α<0.9, 0≤β<0.6, 0≤χ<0.65, and 0≤ε<0.65 (δ＝0 and φ＝0), in a further embodiment 0.35<α<0.8, 0.1≤β<0.6, 0.0≤χ<0.55, and 0≤ε<0.55, in some embodiments 0.35<α<0.8, 0≤β<0.45, 0.0≤χ<0.55, and 0.1≤ε<0.65, and in another embodiment 0.35<α<0.8, 0≤β<0.55, 0.1≤χ<0.65, and 0≤ε<0.55. In another or alternative embodiment, a composite material having SiO, inert metal powder and optionally one or more conductive carbon components can be formed, which can be represented by the formula αSiO-βGr-χHC-δM-εCNF, where 0.35<α<1, 0≤β<0.55, 0≤χ<0.55, 0.1≤δ<0.65, and 0≤ε<0.55. In further, additional, or alternative embodiments, a composite material of SiO with elemental silicon and optionally one or more conductive carbon components can be formed, which can be represented by the formula αSiO-βGr-χHC-εCNF-φSi, where 0.35 < α < 1, 0 ≤ β < 0.55, 0 ≤ χ < 0.55, 0 ≤ ε < 0.55, and 0.1 ≤ φ < 0.65, and in a further embodiment 0.35 < α < 1, 0 ≤ β < 0.45, 0.1 ≤ χ < 0.55, 0 ≤ ε < 0.45, and 0.1 ≤ φ < 0.55. Those skilled in the art will recognize that additional scopes beyond the explicit scope described above are to be considered and are within the scope of this disclosure. As used herein, reference to composite materials implies the application of significant bonding forces, such as by HEMM milling, to achieve tight association of the materials, which differs from simple blending, which is not considered to form a composite material.

Solution-based methods for synthesizing various Si-SiO _x -CM (M = metal) composite materials are described in U.S. Patent Application 2014/0308585, entitled "Silicon-based Active Materials for Lithium-ion Batteries and Synthesis Using Solution Processing," published by Han et al., which is incorporated herein by reference. Silicon-based carbon composite materials having graphene sheets are described in U.S. Patent Application 2014/0370387, entitled "Silicon-Silicon Oxide-Carbon Composite Materials for Lithium-ion Battery Electrodes and Methods for Forming the Composite Material," published by Angushamy et al., which is incorporated herein by reference. Commercially available materials believed to contain SiO _x -Si-C or SiO _x -Si composite materials are used in the batteries described in the examples.

The capacity of the anode significantly affects the energy density of the battery. In batteries with the same output, a higher specific capacity of the anode material results in a lower anode weight. When the negative electrode is made of silicon-based materials, the specific capacity of the electrode at a discharge rate of C/3 relative to lithium metal from 1.5V to 5mV can be about 800 mAh/g to 2500 mAh/g, in further embodiments about 900 mAh/g to about 2300 mAh/g, and in other embodiments about 950 mAh/g to about 2200 mAh/g. Those skilled in the art will recognize that other ranges of discharge specific capacity within the above-defined ranges are to be considered and are within the scope of this disclosure.

positive electrode

The improved negative electrode described above allows for the effective introduction of various positive electrode chemistry. The selected composition can be co-blended with suitable binders and conductive materials into the positive electrode. This section focuses on positive electrode active materials that are particularly ideal for high-voltage cycling and moderately high capacity. Furthermore, this section describes the overall electrode composition and properties.

To some extent, the desired application of the final battery can influence the choice of cathode composition. From this perspective, a wide range of compositions are described below. For automotive applications and similar applications, specific cathode chemistry properties have been found to be ideal for achieving high energy density and cycling to over 600 cycles while retaining at least 80% of capacity, although some materials offer promising results with slightly lower cycle stability. Specifically, blends of nickel-rich lithium nickel cobalt manganese oxide and (lithium + manganese)-rich lithium nickel cobalt manganese oxide provide ideal cathode performance within a voltage range suitable for providing high energy density and long-term cycle stability. Furthermore, when paired with the silicon-based anodes described herein, the nickel-rich lithium nickel cobalt manganese oxide alone, as an active material, can provide ideally high energy density due to its good average discharge voltage during cycling. Blending with some (lithium + manganese)-rich lithium nickel cobalt manganese oxide can improve cycle stability, but results in some loss of energy density due to a slight decrease in average voltage. Examples of active material blends and two separate nickel-rich lithium nickel cobalt manganese oxides are provided below.

Nickel-rich lithium nickel manganese cobalt oxide (N-NMC) can provide ideal cycle and capacity performance for the lithium-ion batteries described herein. Specifically, nickel-rich lithium can be approximately represented by the formula LiNi _{_x}Mn_{_y}Co_{_z}O_₂, where x + y + z ≈ 1, 0.45 ≤ x, 0.025 ≤ y, z ≤ 0.35, and in a further embodiment, 0.50 ≤ x, 0.03 ≤ y, z ≤ 0.325, and 0.55 ≤ x, 0.04 ≤ y, z ≤ 0.3. The amount of nickel can influence the selected charging voltage, thereby balancing cycle stability and discharge energy density. For x values in the range of 0.525 ≤ x ≤ 0.7, the selected charging voltage can be from 4.25 V to 4.375 V. For x values in the range of 0.7 ≤ x ≤ 0.9, the selected charging voltage can be from 4.05 V to 4.325 V. Those skilled in the art will recognize that additional ranges of compositions and selected charging voltages within the aforementioned specific ranges should be considered and are within the scope of this disclosure. These compositions have been found to provide relatively stable high-voltage cycling, good capacity, and desirable impedance. N-NMC powders can be synthesized using coprecipitation techniques such as those further described below, and they are commercially available, for example, from BASF (Germany), TODA (Japan), L&F Materials Corp. (Korea), Unicore (Belgium), and JinheMaterials Corp. (China).

For N-NMC compositions, the average voltage tends to be slightly higher with increasing nickel content, but the charging voltage for stable cycling tends to be slightly lower with increasing nickel content. Therefore, trade-offs can exist regarding the selection of active materials, although N-NMC active materials can provide good cycling performance and reasonably high capacity and energy density. For some embodiments, it has been found that the cycling stability of active materials containing N-NMC compositions can be improved by forming physical blends with lithium-rich and manganese-rich NMC compositions (LM-NMC), the composition of which is described in more detail below. When combined with the silicon-based anodes described herein, the blends can introduce desirable battery performance.

As described above, ideal blends may comprise N-NMC and (lithium-rich + manganese-rich) lithium nickel manganese cobalt oxides (LM-NMC or). These compositions can be approximately represented by the formula Li _1+b Ni _α Mn _β Co _γ A _δ O _2-z F _z , where b+α+β+γ+δ≈1, b is in the range of about 0.04 to about 0.3, α is in the range of 0 to about 0.4, β is in the range of about 0.2 to about 0.65, γ is in the range of 0 to about 0.46, δ is in the range of about 0 to about 0.15, and z is in the range of 0 to 0.2, provided that α and γ are not simultaneously 0, and where A is a metal different from lithium, manganese, nickel, and cobalt. In some embodiments, A may be Mg, Sr, Ba, Cd, Zn, Al, Ga, B, Zr, Ti, Ca, Ce, Y, Nb, Cr, Fe, V, or combinations thereof. Furthermore, in other or alternative embodiments, Li _{_1+b} _Ni _αMn_{_β}Co _γA_{_δ}O_₂, wherein 0.05≤b≤0.125, 0.225≤α≤0.35, 0.35≤β≤0.45, 0.15≤γ≤0.3, 0≤δ≤0.05 and having up to 5 mol% oxygen can be replaced with a fluorine dopant. Those skilled in the art will recognize that further ranges of composition within the explicitly defined ranges described above are to be considered and are within the scope of this disclosure.

LM-NMC cathode materials can be advantageously synthesized via co-precipitation and sol-gel processes detailed in the '160 and '873 patents. In some embodiments, the cathode material is synthesized by precipitating a mixed metal hydroxide or carbonate composition from a solution containing +2 cations, wherein the hydroxide or carbonate composition has a selected composition. The metal hydroxide or carbonate precipitate is then subjected to one or more heat treatments to form a crystalline layered lithium metal oxide composition. The carbonate co-precipitation process described in the '873 patent yields the desired lithium-rich metal oxide material, which contains cobalt in its composition and exhibits high specific capacity performance and excellent tap density. These patents also describe the effective use of metal fluoride coatings to improve performance and cycling.

The synthesis methods for high-capacity cathode active materials outlined above have shown suitability for forming materials with high tap density. This is further described in the '873 patent cited above. Due to the high tap density and excellent cycling performance, batteries can exhibit high total capacity when the active material is incorporated into the cathode. Generally, a higher tap density can be advantageously used to obtain high electrode density without sacrificing material performance if the high-tap-density material has desirable properties.

It has been found that coatings on LM-NMC cathode active materials can improve the performance of the corresponding batteries. Suitable coating materials, generally considered electrochemically inert during battery cycling, can include metal fluorides, metal oxides, or metal non-fluorinated halides. The results for LM-NMC in the following examples were obtained using LM-NMC materials coated with metal fluorides.

For example, PCT application WO 2006/109930A, entitled "Cathode Active Material Coated with Fluorine Compound for Lithium Secondary Batteries and Preparation Method Thereof," by Sun et al., describes the general use of metal fluoride compositions as coatings for cathode active materials (specifically _LiCoO₂ and _{LiMn₂O₄ )} , which is incorporated herein by reference. An improved metal fluoride coating with a suitable designed thickness is described in U.S. Patent Application 2011/0111298 ('298 application), entitled "Coated Cathode Material for Lithium-ion Batteries," by Lopez et al., which is incorporated herein by reference. Suitable metal oxide coatings are further described, for example, in U.S. Patent 8,535,832B2, entitled "Metal Oxide Coated Cathode Material for Lithium-based Batteries," by Karthikeyan et al., which is incorporated herein by reference. The discovery of nonfluorinated metal halides as ideal coatings for cathode active materials is described in U.S. Patent 8,663,849B2, entitled "Metal Halide Coatings on Cathode Materials for Lithium-ion Batteries and Corresponding Batteries" by Venkatachalam et al., which is incorporated herein by reference. The synthesis methods, together with the coating, provide excellent performance in terms of capacity and cycle performance. The ideal properties of the active material, along with the use of the ideal anode material described herein, provide improved battery performance.

For these active materials, long-term cycle stability has been achieved at higher cycle voltages, as described in U.S. Patent 8,928,286, entitled "Very Long Cycles of Lithium-ion Batteries with Lithium-Rich Cathode Materials" by Amiruddin et al., which is incorporated herein by reference. Among LM-NMC compositions in this range, certain compositions have been found to exhibit particularly desirable performance. See, for example, U.S. Patent 8,394,534B2, entitled "Layer-Layer Lithium-Rich Complex Metal Oxides with High Specific Capacity and Excellent Cycles" by Lopez et al., which is incorporated herein by reference. Some LM-NMC compositions have been found to exhibit low DC resistance while maintaining high capacity and excellent cycle performance, as described in U.S. Patent 9,552,901B2 (hereinafter referred to as the '901 Patent), entitled "Lithium-ion Batteries with High Energy Density, Excellent Cycle Performance, and Low Internal Impedance" by Amiruddin et al., which is incorporated herein by reference.

Regarding the active material blend for the positive electrode, the active material may comprise about 3% to about 85% LM-NMC, in a further embodiment about 5% to about 75% LM-NMC, in another embodiment about 6% to about 70% LM-NMC, and in still other embodiments about 7% to about 65% LM-NMC. Similarly, in the positive electrode active material blend, the active material may comprise about 15% to about 97% N-NMC, in a further embodiment about 25% to about 95% LM-NMC, in another embodiment about 30% to about 94% LM-NMC, and in still other embodiments about 35% to about 93% N-NMC. The positive electrode active material may optionally contain 0 to 25% by weight of other active materials, such as lithium cobalt oxide, LiNi _{0.33Mn 0.33Co 0.33O 2} (NMC111), LiNi _{0.8Co 0.15Al 0.05O 2} (NCA), mixtures thereof, etc. Those skilled in the art will recognize that further ranges of composition blends within the clearly defined scope above are to be considered and are within the scope of this disclosure.

Regarding performance, the positive electrode active material blends can provide improved cycling performance at high specific capacity and high energy density, enabling cycling to greater than 600 cycles, and can be combined with high-capacity silicon-based active materials. The performance of these blends is further described below, and specific embodiments are described in the examples. Ideal cycling performance has been demonstrated for both N-NMC alone and blends of N-NMC and LM-NMC.

As described above, the positive electrode typically comprises an active material and a conductive material within a binder. The active material loading in the electrode can be substantial. In some embodiments, the positive electrode comprises about 85% to about 99% of the positive electrode active material, in other embodiments about 90% to about 98% of the positive electrode active material, and in further embodiments about 95% to about 97.5% of the positive electrode active material. In some embodiments, the positive electrode has about 0.75% to about 10% of the polymer binder, in other embodiments about 0.8% to about 7.5% of the polymer binder, and in further embodiments about 0.9% to about 5% of the polymer binder. The positive electrode composition may also typically contain conductive additives different from those in the electroactive composition. In some embodiments, the positive electrode may have about 0.4% to about 12% of the conductive additive, in further embodiments about 0.45% to about 7% of the conductive additive, and in other embodiments about 0.5% to about 5% of the conductive additive. Those skilled in the art will recognize that other ranges of particle loading within the clearly defined ranges described above are to be considered and are within the scope of this disclosure. The positive electrode active material has been described above. Suitable polymer binders for the positive electrode include, for example, polyvinylidene fluoride (PVDF), polyethylene oxide (PE), polyimide, polyethylene, polypropylene, polytetrafluoroethylene (PTFE), polyacrylates, rubbers such as ethylene-propylene-diene monomer (EPDM) rubber or styrene-butadiene rubber (SBR), copolymers thereof, or mixtures thereof. For the positive electrode, PVDF can be used with good results, and the positive electrode in the examples uses a PVDF binder. Conductive additives for the negative electrode are described in detail, and nanoscale conductive carbon can be effectively used for the positive electrode.

For a given loading level, the electrode density (of the active material) is negatively correlated with thickness, such that electrodes with higher density are thinner than those with lower density. The loading is equal to density multiplied by thickness. In some embodiments, the loading level of the positive electrode active material in the battery cathode is about 10 to about 40 mg/ ^cm² , in other embodiments about 12 to about 37.5 mg/ ^cm² , in still other embodiments about 13 to about 35 mg/ ^cm² , and in other embodiments about 20 to about 32.5 mg/ ^cm² . In some embodiments, the density of the active material in the battery cathode is about 2.5 g/cc to about 4.6 g/cc, in other embodiments about 3.0 g/cc to 4.4 g/cc, and in still other embodiments about 3.25 g/cc to about 4.3 g/cc. In a further embodiment, the thickness of the cathode on each side of the current collector after compression and drying of the positive electrode active material can be about 45 micrometers to about 300 micrometers, in some embodiments about 80 micrometers to about 275 micrometers, and in still other embodiments about 90 micrometers to about 250 micrometers. Those skilled in the art will recognize that other ranges of active material loading levels, electrode thicknesses, and electrode densities within the aforementioned specific ranges should be considered and are within the scope of this disclosure.

Lithium supplement

The improved high-energy battery designs described herein typically include supplemental lithium, and this section relates to methods for incorporating supplemental lithium in appropriate embodiments. Generally, the inclusion of supplemental lithium is ideal for batteries with silicon-based anode active materials because the materials exhibit high irreversible capacity loss during the initial charging period of the battery. Additionally, supplemental lithium unexpectedly stabilizes the cycling performance of LM-NMC. Various methods can be employed to introduce supplemental lithium into the battery, although after the corresponding initial reaction and/or charging, the anode becomes associated with excess lithium from the supplemental lithium for cycling. Regarding the anode in a battery with supplemental lithium, the structure and/or composition of the anode can change relative to its initial structure and composition after the first cycle and after subsequent cycles.

Depending on the method used to introduce supplemental lithium, the positive electrode may initially contain a source of supplemental lithium, and/or a sacrificial electrode containing supplemental lithium may be introduced. Alternatively or additionally, supplemental lithium may associate with the negative electrode. In some embodiments, unlike purely chemical or mechanical methods, electrochemical methods may be used to introduce supplemental lithium into the negative electrode. If the supplemental lithium is initially located in the positive electrode or a separate electrode, the negative electrode may remain unchanged in the absence of lithium until the battery is charged, or at least until the circuit between the negative electrode and the electrode containing supplemental lithium is closed in the presence of an electrolyte and a separator. For example, the positive electrode or supplemental electrode may contain elemental lithium, lithium alloys, and/or other sacrificial lithium sources, among other electrode components.

If sacrificial lithium is included in the positive electrode, lithium from the sacrificial lithium source is loaded into the negative electrode during the charging reaction. The voltage during charging based on the sacrificial lithium source can be significantly different from the voltage when charging based on the positive electrode active material. For example, elemental lithium in the positive electrode can charge the negative electrode active material without applying an external voltage, because the oxidation of elemental lithium can drive the reaction as long as the circuit is closed. For some sacrificial lithium source materials, an external voltage is applied to oxidize the sacrificial lithium source in the positive electrode and drive lithium into the negative electrode active material. Charging can typically be performed using constant current, step-constant voltage charging, or other convenient charging schemes. However, at the end of the charging process, the battery should be charged to the desired voltage, thus also involving the extraction of lithium from the positive electrode active material (e.g., deintercalation or dealloying).

In a further embodiment, at least a portion of the supplemental lithium is initially associated with the negative electrode. For example, the supplemental lithium can be in the form of metallic lithium, lithium alloys, or other lithium sources that are more electronegative than the negative electrode active material. Elemental lithium can be in the form of thin films, such as those formed by evaporation, sputtering, or ablation, lithium or lithium alloy foils, and/or powders. Elemental lithium, particularly in powder form, can be coated to stabilize the lithium for processing purposes, and commercial lithium powders, such as those from FMC Corporation, are sold as special coatings for stability. The coating typically does not alter the properties of the lithium powder used for electrochemical applications. After the negative electrode comes into contact with the electrolyte, a reaction can occur, and the supplemental lithium is transferred to the negative electrode active material. Because the electrode is internally conductive, a closed circuit is not required to provide the electron flow resulting from the reaction. A solid electrolyte interface (SEI) layer can also be formed during this process. Therefore, the supplemental lithium loaded into the negative electrode active material is typically consumed in the formation of the SEI layer. Excess lithium released from the lithium-rich positive electrode active material can also be deposited into the negative electrode active material during the final charging of the battery. The supplementary lithium in the negative electrode should be more electronegative than the active material in the negative electrode because there is no way to make the supplementary lithium source react with the active material in the same electrode by applying voltage.

In some embodiments, supplementary lithium associated with the negative electrode can be incorporated into the negative electrode as a powder. Specifically, the negative electrode may comprise an active negative electrode composition and a supplementary lithium source within a polymer binder matrix, as well as any conductive powder (if present). In another or alternative embodiment, the supplementary lithium is disposed along the surface of the electrode. For example, the negative electrode may comprise an active layer having an active negative electrode composition and a supplementary lithium source layer on the surface of the active layer. The supplementary lithium source layer may comprise a foil of lithium or a lithium alloy, supplementary lithium powder within a polymer binder, and/or particles of supplementary lithium source material disposed on the surface of the active layer. In an alternative configuration, the supplementary lithium source layer is located between the active layer and the current collector. Additionally, in some embodiments, the negative electrode may comprise supplementary lithium source layers on both surfaces of the active layer.

An arrangement for electrochemical preloading of lithium may include an electrode having a silicon-based active material formed on a current collector, placed in a container containing an electrolyte and a sheet of lithium source material for contact electrodes. The lithium source material sheet may include lithium foil, lithium alloy foil, or optionally lithium source material in a polymer binder along with conductive powder, which is in direct contact with the negative electrode to be preloaded with lithium, allowing electrons to flow between the materials to maintain charge neutrality while various reactions occur. In the subsequent reaction, lithium is loaded into the silicon-based active material through intercalation, alloying, etc. In an alternative or further embodiment, the negative electrode active material may be mixed into the electrolyte and the lithium source material for incorporating supplemental lithium before being formed into an electrode with the polymer binder, allowing the materials to react spontaneously in the electrolyte.

In some implementations, a lithium source within the electrodes can be assembled into a battery together with electrodes preloaded with lithium to be used. A separator can be placed between the electrodes. Current can flow between the electrodes to provide controlled electrochemical pre-lithiation. Depending on the composition of the lithium source, a voltage may or may not be required to drive lithium deposition within the silicon-based active material. The apparatus for performing this lithiation process can include a container housing an electrolyte and a battery comprising an electrode to be used as the negative electrode in the final battery, a current collector, a separator, and a sacrificial electrode containing a lithium source, such as lithium metal foil, wherein the separator is between the sacrificial electrode and the electrode having a silicon-based active material. A convenient sacrificial electrode can comprise lithium foil, lithium powder embedded in a polymer, or a lithium alloy, but any electrode with extractable lithium can be used. The container for the lithiated battery can include a conventional battery case, a beaker, or any other convenient structure. This configuration provides the advantage of being able to measure the current to quantify the degree of lithiation of the negative electrode. Furthermore, the negative electrode can be cycled once or more, wherein the negative electrode active material is loaded with lithium to near full load. In this way, the SEI layer can be formed to the desired degree of control during lithium preloading of the negative electrode active material. Then, the negative electrode is fully formed during the preparation of the negative electrode with selected lithium preloading.

Typically, for embodiments in which supplemental lithium is used, the amount of supplemental lithium preloaded or available for loading into the active composition can be at least about 2.5% of the capacity, in further embodiments about 3% to about 55% of the capacity, in other embodiments about 5% to about 52.5% of the capacity, and in some embodiments about 5% to about 50% of the capacity of the negative electrode active material. The supplemental lithium can be chosen to substantially balance the IRCL of the negative electrode, but other amounts of supplemental lithium may be used as needed. In some embodiments, supplemental lithium is added in an amount having an oxidation capacity corresponding to 60% to 180% of the first-cycle IRCL of the negative electrode, in further embodiments it is 80% to 165%, and in other embodiments it is 90% to 155%. Those skilled in the art will recognize that additional ranges of percentages within the aforementioned specific ranges are to be considered and are within the scope of this disclosure. Therefore, the contribution to the IRCL of the negative electrode can be effectively reduced or eliminated by the addition of supplemental lithium, such that the measured IRCL of the battery represents, in part or in majority, the contribution from the IRCL of the positive electrode, which is not reduced by the presence of supplemental lithium. Those skilled in the art will recognize that additional scope of IRCL, within the expressly defined scope above, is to be considered and is within the scope of this disclosure.

Equilibrium of cathode and anode

It has been found that the overall performance of a battery depends on the capacity of both the negative and positive electrodes and their relative balance. Electrode balance has been found to be important for achieving particularly high battery energy density and good cycle performance. In some implementations, trade-offs can exist for achieving longer cycle stability and energy density. For longer cycle stability, balancing the battery to achieve a relatively lower energy density, but where the battery is suitable for stable long-term use under a wide range of operating parameters, can be ideal. High energy density can still be achieved using improved active materials and ideal electrode design, while obtaining more than 600 cycles with no more than 80% capacity degradation. Electrode balance can be evaluated in a variety of alternative ways, which can work effectively when a particular evaluation method is appropriately considered.

The capacity of the positive electrode active material can be estimated from the material's capacity, which can be measured by cycling the material relative to lithium metal foil. For example, for a given positive electrode, capacity can be evaluated by determining the insertion and extraction capabilities during the first charge/discharge cycle, where lithium is inserted or extracted from the positive electrode at a rate of C/20 to a voltage selected based on the material chemistry and the chosen battery design (typically 4.2V to 4.5V), and inserted or extracted back into the positive electrode to reach 2V, based on a slight adjustment based on the final anode voltage relative to lithium metal, for example, typically 0.1V, to reach a higher charging voltage relative to lithium metal. Similarly, for a given silicon-based electrode, insertion and extraction capabilities can be evaluated using a battery with a positive electrode containing silicon-based active material and a lithium foil negative electrode. Capacity is evaluated by determining the battery's insertion and extraction capabilities during the first charge/discharge cycle, where lithium is inserted/alloyed into the silicon electrode at a rate of C/20 to reach 5mV and extracted/dealloyed to reach 1.5V. In practical applications, the observed capacity may differ from the tested capacity due to various factors such as high-rate operation and voltage range variations. This could be due to battery design and the composition of the counter electrode, which is not lithium metal. For some evaluation methods, subsequent capacity measurements after the first cycle can be used to assess electrode balance, and higher discharge rates, such as C/3 or C/10, can be used if necessary. The use of balance measurements after a formation cycle or several formation cycles can be ideal, as balance is more based on conditions observed during battery use.

In most commercially available carbon-based batteries, an excess anode of approximately 7% to 10% relative to the cathode is used to prevent lithium plating. A significant problem with this excessive anode is that it increases the battery's weight, reducing its energy density. High-capacity silicon-based anodes can have an IRCL of approximately 10% to approximately 40%, compared to graphite with an IRCL of ~7% in the first charge-discharge cycle. After the first charge-discharge cycle, a large portion of the battery's capacity may become inactive, adding a significant amount of dead weight to the battery.

For high-capacity anode materials, the irreversible capacity loss of the negative electrode is typically greater than that of the positive electrode, resulting in additional lithium availability for the battery. If the negative electrode has a significantly higher irreversible capacity loss than the positive electrode, the initial charging of the negative electrode irreversibly consumes lithium, preventing it from supplying sufficient lithium to the positive electrode during subsequent discharges to meet its full lithium acceptance capacity. This leads to wasted positive electrode capacity, which correspondingly increases weight that does not contribute to cycling. As mentioned above, most or all of the lithium loss from the net IRCL (negative electrode IRCL minus positive electrode IRCL) can be compensated by supplemental lithium. The evaluation of electrode balance during the first formation cycle may or may not consider supplemental lithium. In subsequent cycles after the formation cycle or several cycles, any excess supplemental lithium not consumed for IRCL is typically alloyed into the anode material. Electrode balance can be evaluated in post-formation cycling stages, such as the fourth cycle at a selected rate, and these capacities can be estimated from electrode performance.

From the perspective of providing stable, longer-term cycling performance, it may be ideal to balance the electrodes to provide efficient use of both electrode capacities and to avoid lithium metal plating during cycling. Typically, electrode balancing is considered during electrode assembly, referring to the initial capacity of the electrode relative to the lithium metal.

Typically, battery life can be chosen to end when energy output decreases by approximately 20% from initial capacity at a constant discharge rate, but other values can be chosen as needed. For the materials described herein, capacity degradation at the negative electrode is generally greater than at the positive electrode during cycling, thus the avoidance of lithium metal deposition with cycling indicates a larger excess capacity at the negative electrode, further stabilizing the cycle. Generally, if the rate of negative electrode capacity decay is approximately twice that of the positive electrode, it is ideal to include at least 10% additional negative electrode capacity for cycling. In robust battery designs, at least approximately 10% additional negative electrode capacity may be required under various discharge conditions. Typically, a balance can be chosen such that the sum of the initial positive electrode charge capacity relative to the initial positive electrode charge capacity from open circuit voltage to the battery design charge voltage (typically 4.2V to 4.6V) at a C/20 rate, plus any supplemental lithium oxidation capacity, is approximately 110% to approximately 195% of the initial negative electrode charge capacity evaluated at a C/20 rate from open circuit voltage to 1.5V relative to lithium, in a further embodiment approximately 120% to approximately 185%, and in another embodiment approximately 130% to approximately 190%. Alternatively, the electrode balance can be evaluated at the fourth cycle at a discharge rate of C/10 or C/3, wherein the negative electrode capacity relative to the positive electrode capacity is approximately 110% to 195%, in a further embodiment approximately 120% to approximately 185%, and in another embodiment approximately 130% to approximately 190%. Those skilled in the art will recognize that additional ranges of balance within the above-defined ranges are to be considered and are within the scope of this disclosure. Such a balance is described in the following battery design.

General battery characteristics

The negative and positive electrode structures described above can be assembled into a suitable battery. As mentioned above, electrodes are typically formed in conjunction with current collectors to form the electrode structure. A separator is located between the positive and negative electrodes to form the battery. The separator is electrically insulating while providing at least selected ion conduction between the two electrodes. A variety of materials can be used as the separator. Some commercial separator materials can be formed from polymers such as polyethylene and/or polypropylene, which act as porous sheets providing ion conduction. Commercial polymer separators include, for example, a series of separator materials from Hoechst Celanese, Charlotte, N.C. Furthermore, ceramic-polymer composites have been developed for separator applications. These ceramic composite separators can be stable at higher temperatures, and the composites can reduce the risk of fire. Polymer-ceramic composites for lithium-ion battery separators are sold under the trademarks of Evonik Industries (Germany) and Tiejin Lielsort Korea Co., Ltd. Additionally, porous polymer sheets coated with gel-forming polymers can be used to form the separator. Such a separator design is further described in U.S. Patent 7,794,511B2, entitled "Battery Separator for Lithium Polymer Batteries," by Wensley et al., which is incorporated herein by reference. Suitable gel-forming polymers include, for example: polyvinylidene fluoride (PVDF), polyurethane, polyethylene oxide (PEO), propylene oxide (PPO), polyacrylonitrile, gelatin, polyacrylamide, polymethyl acrylate, polymethyl methacrylate, polyvinyl acetate, polyvinylpyrrolidone, polytetraethylene glycol diacrylate, copolymers thereof, and mixtures thereof.

Electrolytes provide ion transport between the anode and cathode of a battery during charging and discharging. A solution containing solvated ions is called an electrolyte, and an ionic composition that dissolves in a suitable liquid to form solvated ions is called an electrolyte salt. Electrolytes for lithium-ion batteries may contain one or more selected lithium salts. Suitable lithium salts typically have inert anions. Suitable lithium salts include, for example, lithium hexafluorophosphate, lithium hexafluoroarsenate, lithium bis(trifluoromethanesulfonylimide), lithium trifluoromethanesulfonate, lithium tri(trifluoromethanesulfonyl)methyl, lithium tetrafluoroborate, lithium perchlorate, lithium tetrachloroaluminate, lithium chloride, lithium difluorooxalatoborate, and combinations thereof. In some embodiments, the electrolyte contains a lithium salt with a concentration of 1M to 2M, but higher or lower concentrations may be used.

For lithium-ion batteries of interest, one or more lithium salts are typically dissolved in non-aqueous liquids. Solvents generally do not dissolve the electroactive materials. In some embodiments, suitable solvents may include, for example: propylene carbonate, dimethyl carbonate, diethyl carbonate, 2-methyltetrahydrofuran, dioxane, tetrahydrofuran, methyl ethyl carbonate, γ-butyrolactone, dimethyl sulfoxide, acetonitrile, formamide, dimethylformamide, triethylene glycol dimethyl ether (tris(ethylene glycol) dimethyl ether), diethylene glycol dimethyl ether (diethylene glycol dimethyl ether), DME (monoethylene glycol dimethyl ether or 1,2-dimethoxyethane or ethylene glycol dimethyl ether), nitromethane, and mixtures thereof. Particularly useful solvents for high-voltage lithium-ion batteries are further described in U.S. Patent 8,993,177, entitled “Lithium-ion Battery with High-Voltage Electrolyte and Additives” by Amiruddin et al., which is incorporated herein by reference.

For some implementations of batteries with silicon-based negative electrode active materials, electrolytes with fluorinated additives have shown to further improve battery performance. Fluorinated additives may include, for example: fluorinated ethylene carbonate, fluorinated ethylene carbonate, monochloroethylene carbonate, monobromoethylene carbonate, 4-(2,2,3,3-tetrafluoropropoxymethyl)-[1,3]dioxane-2-one, 4-(2,3,3,3-tetrafluoro-2-trifluoromethyl-propyl)-[1,3]dioxane-2-one, 4-trifluoromethyl-1,3-dioxane-2-one, bis(2,2,3,3-tetrafluoro-propyl) carbonate, bis(2,2,3,3,3-pentafluoro-propyl) carbonate, or mixtures thereof. In some embodiments, the electrolyte may contain about 1% to about 55% by weight of a halogenated carbonate, in further embodiments about 3% to about 50% by weight, and in other embodiments about 5% to about 45% by weight of the halogenated carbonate as part of the total electrolyte weight. Those skilled in the art will recognize that additional ranges of halogenated carbonate concentrations within the aforementioned defined ranges are to be considered and are within the scope of this disclosure. Furthermore, electrolytes containing fluoroethylene carbonate have been found to have excellent low-temperature performance, as described in U.S. Patent Application 2013/0157147 ('147 application) entitled "Low-Temperature Electrolyte for High-Capacity Lithium-Based Batteries" by Li et al., which is incorporated herein by reference. Additional fluorinated additives include, for example, fluorinated ethers, as described in U.S. Patent Application 2018/0062206 entitled "Fluorinated Ethers as Electrolyte Cosolvents for Lithium Metal Anodes" by Li et al. and WO 2018/051675 entitled "Lithium Secondary Batteries" by Takuya et al., both of which are incorporated herein by reference. Fluorinated electrolytes are available from Daikin America, Inc.

In some embodiments, electrolytes are formulated using commercial consumer electronics electrolytes comprising ethylene carbonate, diethyl carbonate, and fluorinated components, and excellent cycling results are demonstrated below. It has been found that suitable fluorinated electrolyte components such as fluorinated ethylene carbonate, fluorinated ethers, and/or fluorinated amines provide ideal stabilization for silicon-based electrodes compared to electrolytes that provide suitable commercial performance for consumer electronics batteries with graphite anodes.

The electrodes described herein can be incorporated into a variety of commercial battery/pack designs, such as prismatic cells, wound cylindrical cells, button cells, or other suitable battery/pack designs. A battery may comprise a single pair of electrodes or multiple pairs of electrodes assembled with one or more parallel and/or series electrical connections. For ease of placement in a container, the electrode stack may have additional electrodes so that the stack terminates with the same polarity as the other end of the stack. Although the electrode structures described herein can be used in batteries intended for single-charge or single-charge applications, the resulting batteries typically exhibit ideal cycle performance for rechargeable applications during multiple battery cycles.

In some implementations, the positive and negative electrodes can be stacked together with a separator between them, and the resulting stacked structure can be wound into a cylindrical or prismatic configuration to form a battery structure. Suitable conductive tabs can be soldered to or similarly connected to the current collector, and the resulting jellyroll structure can be placed in a metal can or polymer package, with the negative and positive tabs soldered to suitable external contacts. Electrolyte is added to the can, and the can is sealed to complete the battery. Some currently used commercially available rechargeable batteries include, for example, cylindrical 18650 batteries (18 mm in diameter and 65 mm in length) and 26700 batteries (26 mm in diameter and 70 mm in length), but other battery/pack sizes, as well as prismatic batteries and foil pouch batteries/packs of selected sizes, can be used.

Pouch cells can be particularly ideal for a variety of applications, including certain transportation applications, due to their ease of stacking and low container weight. Pouch cell designs for transportation batteries incorporating high-capacity cathode active materials are further described in U.S. Patent 8,187,752, entitled "High-Energy Lithium-Ion Secondary Battery," by Buckley et al., and in U.S. Patent 9,083,062B2, entitled "Battery Pack for Transportation and High-Capacity Pouch Cell for Incorporation into Compact Battery Packs," by Kumar et al., both of which are incorporated herein by reference. While pouch cell designs are particularly convenient for use in certain battery pack designs, they can also be used efficiently in other situations with high capacity. Ideal results are demonstrated in examples using prismatic pouch cells with electrode stacks.

Representative embodiments of the pouch battery are shown in Figures 1 through 4. In this embodiment, the pouch battery 100 includes a pouch housing 102, an electrode core 104, and a pouch cap 106. The electrode core is discussed further below. The pouch housing 102 includes a cavity 110 and an edge 112 surrounding the cavity. The cavity 110 has dimensions that allow the electrode core 104 to be mounted within the cavity 110. As shown in Figures 2 and 3, the pouch cap 106 can be sealed around the edge 112 to seal the electrode core 104 within the sealed battery. Terminal tabs 114, 116 extend outward from the sealed pouch to make electrical contact with the electrode core 104. Figure 3 is a schematic cross-section of the battery of Figure 2 taken along line 3-3. Several other embodiments of the pouch battery with different edge and seal configurations are possible.

Figure 4 illustrates one embodiment of an electrode core 104 that typically includes an electrode stack. In this embodiment, the electrode stack 130 includes negative electrode structures 132, 134, 136, positive electrode structures 138, 140, and separators 150, 152, 154, 156 disposed between adjacent positive and negative electrodes. The separators can be provided as single-folded sheets, with the electrode structures placed within the separator folds. The negative electrode structures 132, 134, 136 include negative electrodes 160, 162, 164, 166, and 168, 170, respectively disposed on both sides of current collectors 172, 174, 176. The positive electrode structures 138, 140 include positive electrodes 180, 182 and 184, 186, respectively disposed on opposite sides of current collectors 188, 190. Ear tabs 192, 194, 196, 198, and 200 are connected to current collectors 172, 188, 174, 190, and 176, respectively, to facilitate series or parallel connections of the electrodes. For vehicle applications, the ear tabs are typically connected in parallel so that ear tabs 192, 196, and 200 will be electrically connected to electrical contacts accessible outside the container, and ear tabs 194 and 198 will be electrically connected to electrical contacts outside the container that are of opposite polarity.

The electrode stack may have an additional negative electrode such that both external electrodes adjacent to the container are negative electrodes. Typically, a battery with stacked electrodes of the dimensions described herein has 5 to 40 negative electrode elements (with active material coated on both sides of the current collector) and, in further embodiments, 7 to 35 negative electrode elements, along with a corresponding number of positive electrode elements, typically one fewer than the number of negative electrode elements. Those skilled in the art will recognize that other ranges of electrode numbers within the aforementioned specific ranges are to be considered and are within the scope of this disclosure.

As described above, wound electrodes can be used accordingly for cylindrical or generally prismatic batteries. A wound battery for cylindrical lithium-ion batteries is further described in U.S. Patent 8,277,969, entitled "Lithium-ion Secondary Battery," by Kobayashi et al., which is incorporated herein by reference. A prismatic battery with wound electrodes is described in U.S. Patent 7,700,221 ('221 patent), entitled "Electrode Assembly and Lithium-ion Secondary Battery Using the Electrode Assembly," by Yeo, which is also incorporated herein by reference. Neither Kobayashi's '969 patent nor Yeo's '221 patent describes how to achieve reasonable cycling or high energy density using silicon-based active materials. For example, the '221 patent cited above further describes a design for a prismatic battery with wound electrodes. The specific design of the stacked electrode assembly or wound battery may be influenced by the target size and target capacity of the battery.

The improved negative electrode can be used in a range of applications and battery/pack designs. For electrode stacks, the electrode area can be rationally selected based on the volume and design constraints of the specific application. The following discussion focuses on larger batteries typically designed for transportation applications such as drones, cars, trucks, or other vehicles. However, the improved negative electrode described herein can be effectively used in consumer electronics applications that can be based on smaller battery forms. Furthermore, it should be noted that vehicles can use smaller consumer electronics batteries, and Tesla vehicles are currently known for using thousands of small consumer electronics batteries in their battery packs. Generally, larger battery/pack forms can achieve higher energy densities within a specific range. Ideally, the positive electrode active material can be selected based on the specific application to balance various considerations, such as energy density.

For selecting electrode parameters, the design of a high-weight energy density battery can combine a balance of various factors, including electrode area, the number of electrode structures, and battery capacity. Electrode area refers to the spatial extent of one of the electrodes along one side of the current collector. Figure 1 depicts the length “L” of the electrode, and Figure 3 depicts the width “W” of the electrode. As shown in the figures, the area of the electrode can be defined as L x W. In some embodiments, the areas of the individual electrodes can be similar so that the dimensions of the battery comprising the electrode stack can have a length and width similar to the length and width of each electrode in the stack. In some embodiments, the separator can be sheet-like, with an area slightly larger than the electrode area, and in some embodiments, the separator can be folded, pleated, or formed into a pocket in which the electrode is placed.

For some consumer electronics devices, the length and width of a generally prismatic battery can independently range from about 15 mm to 500 mm, in a further embodiment from about 17.5 mm to about 400 mm, and in another embodiment from about 20 mm to about 350 mm. The thickness of the battery can range from about 1 mm to about 15 mm, in a further embodiment from 1.5 mm to about 13.5 mm, and in another embodiment from about 2 mm to about 12 mm. The volume of the battery can range from 500 ^mm³ to about 100,000 ^mm³ , in a further embodiment from about 750 ^mm³ to about 75,000 ^mm³ , and in other embodiments from about 1,000 ^mm³ to about 50,000 ^mm³ . For wound batteries, two electrodes and a separator are placed together and then typically wound along a mandrel or similar using suitable means. To obtain the corresponding volume, the length is typically substantially greater than the width. The width is typically from about 15 mm to about 150 mm, and in a further embodiment from about 20 mm to about 120 mm. The length-to-width ratio corresponding to the winding dimension can be from about 3 to about 25, and in a further embodiment from about 4 to about 20. After winding, the spiral-wound electrode can be prismatic, cylindrical, or other convenient shapes. The diameter of the cylindrical battery can be from about 5 mm to about 50 mm, in a further embodiment from about 7 mm to about 40 mm, and in another embodiment from about 8 mm to about 30 mm. The prismatic-wound electrode can have the same overall dimensions as the electrode stack described above. Those skilled in the art will recognize that other ranges of dimensional parameters within the above-defined ranges are to be considered and are within the scope of this disclosure.

Generally, larger battery/battery packs offer higher weight and volumetric energy density. For larger battery types typically ideal for transportation applications, the desired face area of a pouch battery can range from about 10,000 ^mm² to about 50,000 ^mm² , and in further embodiments from about 15,000 ^mm² to about 40,000 ^mm² . Furthermore, the volume of a pouch battery can range from about 30,000 ^mm³ to about 800,000 ^mm³ , in further embodiments from about 50,000 mm³ to about 750,000 ^mm³ , and in other embodiments from about 100,000 ^mm³ to about 600,000 ^mm³ . The width of a pouch battery can range from about 50 mm to about 500 mm, in further embodiments from about 65 mm to about 450 mm, and in other embodiments from about 75 mm to about 400 mm. Similarly, the height of the pouch battery can range from about 75 mm to about 750 mm, in further embodiments from about 85 mm to about 700 mm, and in other embodiments from about 100 mm to about 650 mm. The thickness of the pouch battery can range from about 3 mm to about 18 mm, in further embodiments from about 3.25 mm to about 16 mm, and in other embodiments from about 3.5 mm to about 15 mm. In some embodiments, the total capacity of the larger form battery can be from about 0.5 Ah to about 105 Ah, in other embodiments from about 2 Ah to about 80 Ah, and in other embodiments from about 5 Ah to about 65 Ah. Those skilled in the art will recognize that further ranges of battery size and capacity within the above-defined ranges are to be considered and are within the scope of this disclosure.

Performance properties

The combination of design features described herein provides longer cycle stability while maintaining desirable battery performance. Achieving long-term cycling involves balancing the improved electrode design described above with battery design parameters. With some trade-offs in energy density, the battery design of this invention significantly improves cycle life. A gravimetric energy density, as illustrated herein, comparable to that described in '535 application, is achieved in a smaller form factor battery, exhibiting modest cycle efficiency at C/10 rate and a voltage window of 4.35V to 3V. In embodiments of particular interest, the cathode herein typically comprises a nickel-rich lithium nickel manganese cobalt oxide, which may be blended with a lithium-rich + manganese-rich lithium nickel manganese cobalt oxide. The use of N-NMC active materials is consistent with good cycling and high energy density at the applied voltage window.

The negative electrode can be tested with a lithium foil electrode in a half-cell configuration to evaluate its performance independently of the positive electrode characteristics. Specifically, the negative electrode can be cycled at a selected rate relative to lithium metal over a voltage range of 0.005V to 1.5V. The specific capacity of the negative electrode is as described above. The improved negative electrode described in this paper cycles better relative to a reference electrode in a half-cell configuration, but exhibits improved cycle stability if more explicitly quantified in the following full-cell configuration. Positive electrode performance also affects battery performance, but positive electrode performance is evaluated in the case of a full cell with silicon-based electrodes.

Batteries and battery packs in high-energy forms suitable for commercial applications, such as electric vehicle applications based on high-capacity positive and high-capacity negative electrodes, have been developed. As mentioned above, electrode designs have been developed to utilize high-capacity materials. The lithium-ion secondary batteries disclosed herein can have a discharge energy density of at least about 235 Wh/kg at C/3 when discharged from a selected charging voltage to 2.5V at 30°C, at least about 240 Wh/kg in further embodiments, at least about 245 Wh/kg in other embodiments, and at least about 250 Wh/kg at C/3 when discharged from a selected charging voltage to 2.5V at 30°C. Some batteries can be discharged to slightly different discharge voltages, such as 2.4V or 2.3V, with little difference in the resulting performance because there is almost no remaining capacity at these voltages, and the actual discharge voltage can be selected for the specific application. The selected charging voltage may be affected by the positive electrode active material. Typically, these batteries have a selected charging voltage of about 4.05V to 4.4V, with exemplified values including 4.15V (NMC811), 4.2V, 4.3V, and 4.35V. The batteries can exhibit very good cycle performance. In some embodiments, when discharged from the selected charging voltage to 2.5V at 30°C with C/3, the battery can exhibit a discharge capacity of at least about 75% of the capacity of the 6th cycle after 500 cycles, at least about 80% in other embodiments, and in yet another embodiment, when cycled from the selected charging voltage to 2.5V at 30°C with C/3, the discharge capacity of the 500th cycle is at least about 82% of the discharge capacity of the 6th cycle. Similarly, when discharged from a selected charging voltage to 2.5V at 30°C with a C/3 ratio, the battery can exhibit a discharge capacity of at least about 70% of the capacity of the 6th cycle at the 600th cycle, at least about 73% in other embodiments, and in yet another embodiment, when cycled from a selected charging voltage to 2.5V at 30°C with a C/3 ratio, the discharge capacity at the 600th cycle is at least about 75% of the discharge capacity of the 6th cycle. Those skilled in the art will recognize that additional scopes beyond the explicit foregoing are to be considered and are within the scope of this disclosure.

Example

Example 1 - Active Material Composition

This embodiment illustrates how silicon oxide composite active materials can be cyclically stabilized by incorporating a significant amount of electrochemically active graphite along with conductive nanoscale carbon and a suitable polymer binder.

To evaluate the negative electrode formulation, coin cells were formed using lithium foil counter electrodes or positive electrodes containing lithium metal oxide blends as active compositions. To form a negative electrode with silicon-based active materials, powdered commercial silicon oxide/silicon/carbon composite material (hereinafter referred to as SiO _{_x} /Si/C) and a selected amount of graphite were thoroughly mixed with 1% to 7% by weight of nanoscale carbon conductive additives to form a homogeneous powder mixture. Four sample silicon-based electrodes were formed using different amounts of graphite (KS 6 synthetic graphite, Timcal): Sample 1 – 70% by weight SiO _{_x} /Si/C + 30% by weight graphite; Sample 2 – 78% by weight SiO _{_x} /Si/C + 22% by weight graphite; Sample 3 – 85% by weight graphite + 15% by weight graphite; and Sample 4 – 100% by weight SiO _{_x} /Si/C. Separately, a blend of a polymer binder, namely a polyimide binder and a lower elastic modulus binder, is mixed with N-methylpyrrolidone (“NMP”) (Sigma-Aldrich) and stirred overnight to form a polymer binder-NMP solution. The homogeneous powder mixture is then added to the polymer binder-NMP solution and mixed for approximately 2 hours to form a homogeneous slurry. The slurry is coated onto a copper foil current collector to form a thin wet film, and the laminated current collector is dried in a vacuum oven to remove NMP and cure the polymer. The laminated current collector is then pressed between the rolls of a sheet metal mill to obtain the desired lamination thickness. The dried laminate contains 2% to 20% by weight of binder, with the remainder of the electrode contributed by the powder. The negative electrode is electrochemically pre-lithiated with sufficient lithium to compensate for 100% to 160% of the lithium loss due to irreversible capacity loss at the anode.

An initial set of coin cells, called a half-cell, is formed using lithium foil as the electrode. A portion of the negative electrode, along with the separator, a portion of the lithium foil, and a corresponding current collector for the lithium foil are cut to the desired size. The separator for the coin cells described herein comprises a commercially available three-layer polyolefin separator. An electrolyte comprising dimethyl carbonate and fluoroethylene carbonate is placed in the cell and the cell is sealed. The coin cells are then cycled in the cell from 0.005V to 1.5V for the first charge-discharge cycle at C/10, the second charge-discharge cycle at C/5, and the remaining cycles at C/3. The initial specific capacity is given in Table 1. In half-cell form, the lithium foil is initially discharged to load (intercalate or alloy) lithium into the silicon oxide electrode, and the subsequent charging step reverses this reaction to remove lithium from the silicon oxide electrode (deintercalate or dealloy).

Table 1

Figure 5 plots the specific capacity for both charging and discharging (as a function of cycle number). The charging and discharging capacities essentially converge after approximately 10 cycles. The specific capacity decreases with increasing graphite content, as expected. Batteries containing graphite exhibit significantly improved cycle stability compared to batteries with only silicon-based low-valence oxide active materials. In half-cell form, batteries with different amounts of graphite exhibit similar cycle stability.

Another series of batteries was formed using four silicon oxide composite electrode samples and a positive electrode made from a blend of nickel-rich lithium nickel manganese cobalt oxide (N-NMC) and lithium + manganese-rich NMC (see above '901 patent). These full cells are referred to according to their negative electrode samples. The assembled batteries were cycled between 4.35V and 2.3V. The batteries were charged and discharged at a rate of C/10 in the first cycle, at a rate of C/5 in the second cycle, and then cycled at a rate of C/3. Cycling results regarding specific capacity are plotted in Figure 6, and normalized capacity results are given in Figure 7. Batteries with graphite in the negative electrode exhibited improved cycling performance after more than 600 cycles and a capacity retention of more than 60%. Table 2 summarizes the cycling performance. For this battery form, battery performance becomes similar with or without graphite active material during longer cycling periods exceeding 600 cycles.

Table 2

Another series of coin cells were formed using three different nanoscale carbon conductive additives: carbon black, carbon nanofibers, or carbon nanotubes. The silicon-based electrodes were the same as described above. A first set of coin cells was formed using lithium metal foil as the electrode. The coin cells were then formed as described above. The half-cell performance is summarized in Table 3 for the indicated rate performance, namely the first discharge/charge at C/10, the second discharge/charge at C/5, and the subsequent discharge/charge at C/3.

Figure 8 shows the charge and discharge specific capacities. For low cycle counts, carbon black exhibits a higher specific capacity, but batteries with carbon nanofibers demonstrate a higher discharge specific capacity at higher cycle counts. The specific capacities for rate performance of these batteries are shown in Table 3.

Table 3

A set of full cells was also formed using negative electrodes with different conductive additives and the same positive electrode used to form sample full cells 1-4. Similarly, the full cells were named according to the negative electrode sample designation (5 = carbon nanotubes, 6 = carbon black, 7 = carbon nanofibers). The initial performance of these cells is summarized in Table 4. Specific capacity results are plotted as a function of cycling in Figures 9 (specific capacity) and 10 (normalized capacity). For these cells, those with carbon black and carbon nanotubes performed similarly, while those with carbon nanofibers exhibited poorer performance at higher cycle counts.

Table 4

Example 2 - Binder Composition for Negative Electrode

This embodiment illustrates the improved cycle performance of silicon-based electrodes using a hybrid binder formulation.

Four sets of batteries were formed to test the binder compositions. Two sets of batteries were formed using a first active composition ratio, one set being a half-cell and the other a full-cell, and two sets of batteries were formed using a second active composition ratio, one set being a half-cell and the other a full-cell. The first active composition ratio consisted of 70 wt% SiO composite material and 30 wt% graphite, and included 2 wt% to 6 wt% carbon nanotubes as a conductive additive. Five half-cells were assembled with lithium foil electrodes within a coin cell as described in Example 1 above. Samples 8, 9, and 12 had 1 wt% to 7 wt% binder with a lower elastic modulus and 7 wt% to 15 wt% polyimide, wherein samples 9 and 12 had the same amount of binder with a lower elastic modulus, and sample 8 had a larger amount of binder with a lower elastic modulus (binder 2) relative to samples 9 and 12. Sample 12 had a larger amount of polyimide and a correspondingly smaller amount of active material relative to sample 9. The weight ratios of binder 2 to polyimide in samples 8, 9, and 12 were 0.714, 0.333, and 0.250, respectively. Samples 10 and 11 contained no binder 2, and sample 10 had a lower amount of polyimide and a correspondingly higher amount of conductive carbon nanotubes.

The assembled half-cells were cycled between 0.005V and 1.5V, with the first cycle at C/10, the second at C/5, and subsequent cycles at C/3. Initial cycle performance is shown in Table 5. Charge and discharge specific capacities (as a function of cycle number) are plotted in Figure 11. For longer cycles, the sample with the polymer blend exhibited improved capacity.

Table 5

A set of full cells was also formed using a positive electrode and a coin cell structure with a lithium metal oxide blend as described in Example 1. These cells were designated according to their negative electrode samples. The cells were cycled, with the first cycle at C/10, the second cycle at C/5, and from the third cycle onwards at C/3. The charge and discharge capacities are plotted in Figures 12 (specific capacity) and 13 (normalized capacity). The performance is also summarized in Table 6. Similarly, the cells with polymer binder blends exhibited improved cycle performance.

Table 6

A second set of two batteries, one half-cell and one full-cell, were formed using a second active composition ratio. The second active composition ratio consisted of 85 wt% SiO _{_x} /Si/C composite material and 15 wt% graphite, and the batteries also contained 2 wt% to 6 wt% carbon nanotubes as conductive additives. Five half-cells were assembled with lithium foil electrodes within a coin cell as described above. Samples 13, 14, and 17 contained 1 wt% to 7 wt% of a binder (binder 2) with a lower elastic modulus and 7 wt% to 15% polyimide, wherein samples 14 and 17 contained the same amount of binder 2, and sample 13 contained a larger amount of binder 2 compared to samples 14 and 17. Sample 17 contained a larger amount of polyimide and a correspondingly lower amount of active material compared to sample 14. The weight ratios of binder 2 to polyimide in samples 13, 14, and 17 were 0.714, 0.333, and 0.250, respectively. Samples 15 and 16 do not contain any binder 2, and sample 15 has a lower amount of polyimide and a correspondingly higher amount of conductive carbon nanotubes.

The assembled half-cells were cycled between 0.005V and 1.5V, with the first cycle at C/20, the second at C/5, and subsequent cycles at C/3. Initial cycle performance is given in Table 7. Charge and discharge specific capacities (as a function of cycle number) are depicted in Figure 14. For longer cycles, the sample with the polymer blend exhibited the best cycling performance with decreasing active material loading, while the battery containing only polyimide and without any added conductive additives exhibited the worst cycle capacity.

Table 7

For the second set of negative electrodes, a set of full cells was formed using the positive electrode and coin cell structure as described in Example 1. The cells were designated according to the negative electrode sample number. The cells were cycled, with the first cycle at C/10, the second at C/5, and from the third cycle onwards at C/3. Charge and discharge capacities are plotted in Figures 15 (specific capacity) and 16 (normalized capacity). Performance is also summarized in Table 8. Similarly, cells with polymer blends exhibited improved cycle performance.

Table 8

Example 3 - Battery with NMC Positive Electrode

This example illustrates good cycle performance in the case of non-blended positive electrode active materials.

In the first group of batteries, the negative electrode used in these batteries was substantially the same as that used in sample 8 above. The positive electrode was essentially equivalent to the electrodes in Examples 1 and 2, except that LiNi _{0.8Mn 0.1Co 0.1O 2} (NMC811) or LiNi _{0.6Mn 0.2Co 0.2O 2} (NMC622) was used as the only active material used in the battery. However, batteries were formed using different loadings of positive electrode active material. The NMC 622 battery was cycled within a voltage range of 4.30V to 2.5V or 4.35V to 2.5V, and the NMC 811 battery was cycled within a voltage range of 4.20V to 2.5V.

Initial charge-discharge cycles were performed at a rate of C/10. Figure 17 plots the voltage versus specific capacity of the first charge/discharge cycle for a representative cell with each cathode material. Results for all samples are also summarized in Table 9.

Table 9

In addition, the batteries corresponding to samples 18-25 were cycled. Specifically, the batteries were charged and discharged at a rate of C/10 in the first cycle, at a rate of C/5 in the second cycle, and then cycled at a rate of C/3. The rate performance of the battery samples based on the discharge capacity at the specified discharge rates is summarized in Table 10.

Table 10

Figure 18 plots the discharge capacity (as a function of cycles) of samples 19, 21, and 24, and Figure 19 plots the normalized capacity (as a function of cycles). The best cycle performance for these coin cells is the NMC622-based sample cycling from 4.30V to 2.5V, while the worst cycle performance is the NMC622-based sample cycling from 4.35V to 2.5V throughout the cycle range.

Another battery pack was formed using NMC811 as the positive electrode active material. For this battery pack, the negative electrode comprised a _SiO₂x /Si/C composite material without any graphite, and included a polyimide binder and a mixture of carbon nanotubes and acetylene black as a conductive additive. The electrodes were designed to achieve high energy density.

Figure 20 plots the battery voltage (as a function of specific capacity) from 4.3V to 2.5V at a charge/discharge rate of C/20. The battery exhibits a discharge capacity of 207.6 mAh/g at C/20. Referring to Figure 21, a graph of the battery voltage (as a function of specific capacity) is shown for the second charge/discharge cycle, which is charged to 4.3V at a rate of C/6 and discharged at a rate of C/10, resulting in a slightly lower specific capacity. For the initial charge, the battery is charged at a constant current until the charging voltage is reached, and then charged at a constant voltage until the current drops to a low value.

The batteries (two identical batteries in duplicate) were initially charged/discharged at a rate of C/20, followed by charging at a rate of C/6 and discharging at a rate of C/10. Specific capacity (as a function of cycles) is plotted in Figure 22, and normalized specific capacity is plotted in Figure 23. The batteries retained at least 80% of their capacity during 130 to 140 cycles.

Example 4 - High-capacity, long-cycle battery

This embodiment illustrates the long-term cycle stability of an improved negative electrode based on its integration into a large-form battery.

Two pairs of pouch cells are formed using identical electrodes of two different cell sizes. The negative electrode comprises a _SiO₂x /Si/C composite material with 30% by weight of graphite active material. The negative electrode also contains a blend of polyimide and a binder with a low elastic modulus, as well as a carbon nanotube conductive material. The positive electrode is a blend of nickel-rich NMC and HCMR. Based on the low rate (C/20) capacity, the initial capacity of the negative electrode is approximately 150% of the capacity of the positive electrode. A fluorinated electrolyte as described above is also used.

For the two battery forms, a first form is shown in Figures 24 and 25, in which the prismatic pouch battery has approximate dimensions of 225 mm x 165 mm x 4.5 mm (thickness) (ignoring the tabs), as indicated in the figures (the indicated figures are approximate dimensions in mm). A second form is shown in Figures 26 and 27, in which the prismatic pouch battery has approximate dimensions of 145 mm x 64 mm x 7.7 mm (thickness) (ignoring the tabs), as indicated in the figures (the indicated figures are approximate dimensions in mm). Electrodes (10 to 25 cathode layers and 11 to 26 anode layers) are formed as described in Example 1, and the separator sheet is pleated, with the plated electrodes located within the separator folds. The separator for the pouch battery is a porous polymer sheet with a gel-forming polymer coating. Lithium supplementation is provided by coating the negative electrode surface with lithium powder (FMC Corp.) prior to assembly to substantially compensate for 100% to 160% of the IRCL of the silicon-based negative electrode active material. The battery is designed to have a total capacity of approximately 21 Ah (first form) or 11 Ah (second form) at a discharge rate of C/3 at 30 degrees Celsius. The battery is cycled in one formation cycle at a charge and discharge rate of C/20. The battery is then cycled at a charge and discharge rate of C/3 at 30 degrees Celsius.

The 11Ah battery design achieves a higher energy density of 280Wh/kg, while the 21Ah battery achieves 245Wh/kg. Cycling performance for both battery sizes is plotted in Figure 28. The higher energy density battery (11Ah design) exhibits approximately 550 cycles before reaching 80% capacity, while the 21Ah battery design exhibits approximately 750 cycles with 80% capacity retention. However, both batteries outperform the known performance of any battery in this energy density range in terms of observed cycling performance.

Furthermore, the target 11Ah battery is formed with the same anode as the above 11Ah battery and an NMC811-based cathode, and with an NMC622-based cathode and an anode containing 85 wt% _SiO₂x /Si/C composite material and 15 wt% graphite. Except for the indicated active material substitutions, this battery is comparable to the above battery with a cathode blend active material. For the battery with the NMC622 cathode, two comparable representative batteries were cycled from 4.3V to 2.5V. The cycling results are plotted in Figures 29-32. In Figures 29 and 30, the specific energy and normalized specific energy are plotted as a function of cycling. For more than 300 cycles, the specific energy still exceeds 250Wh/kg. Figures 31 and 32 are graphs of the corresponding specific capacity and normalized specific capacity within the same cycling range.

For batteries with NMC811 cathodes, equivalent batteries were cycled from 4.15V to 2.5V and from 4.20V to 2.5V. The cycling results are plotted in Figures 33-36. In Figures 33 and 34, the specific energy and normalized specific energy are plotted as a function of cycling. For batteries cycled from 4.20V to 2.5V, the specific energy exceeds 300Wh/kg in almost the first 100 cycles at a C/3 discharge rate. Even with lower charging voltages for these materials to achieve more cycles, the average voltage and capacity remain high, resulting in a high specific energy value. Figures 35 and 36 are graphs of the corresponding specific capacity and normalized specific capacity within the same cycling range.

In summary, large-form batteries can achieve initial energy densities as high as and exceeding 300 Wh/kg. Batteries with N-NMC active materials without blending achieved high initial energy density values. Batteries with blends of positive electrode active materials achieved ideal cycle stability, but when extrapolated to larger cycle numbers, NMC811 batteries cycled at a charging voltage of 4.15 V yielded very promising results.

The above embodiments are intended to be illustrative rather than restrictive. Other embodiments are also included within the scope of the claims. Furthermore, although the invention has been described with reference to specific embodiments, those skilled in the art will recognize that changes in form and detail may be made without departing from the spirit and scope of the invention. Any combination of references cited above is limited so that the subject matter incorporated herein does not contradict the explicit disclosure herein. The extent to which specific structures, compositions, and/or processes are described herein using terms such as components, elements, ingredients, or other divisions should be understood that, unless otherwise expressly stated, the disclosure herein covers specific embodiments, embodiments comprising specific components, elements, ingredients, other divisions, or combinations thereof, and embodiments primarily composed of such specific components, ingredients, or other divisions, or combinations thereof, which may include additional features that do not alter the essential nature of the subject matter, as indicated in the discussion.

Claims (29)

1. A lithium-ion battery, the lithium-ion battery comprising: The negative electrode comprises 75% to 92% by weight of an active material, 1% to 7% by weight of nanoscale conductive carbon, and 6% to 20% by weight of a polymer binder, wherein the active material comprises 40% to 95% by weight of a silicon oxide-based material and 5% to 60% by weight of graphite, and wherein the polymer binder comprises a blend of 60% to 95% by weight of a polyimide with an elongation of at least 50% and at least 5% by weight of a second binder polymer with an elastic modulus not greater than 2.4 GPa. The positive electrode comprises a nickel-rich lithium nickel cobalt manganese oxide, conductive carbon, and a polymer binder. The nickel-rich lithium nickel cobalt manganese oxide is represented by the formula LiNi _x Mn _y Co _z O ₂ , where x + y + z = 1, 0.45 ≤ x, 0.025 ≤ y ≤ 0.35, and 0.025 ≤ z ≤ 0.35. A diaphragm is provided between the negative electrode and the positive electrode. Electrolytes, the electrolyte comprising lithium salts and non-aqueous solvents; and A container that encapsulates other battery components. 2. The lithium-ion battery of claim 1, wherein the silicon oxide material comprises a silicon oxide-silicon-carbon composite material. 3. The lithium-ion battery of claim 1 or claim 2, wherein the negative electrode active material comprises 50% to 90% by weight of a silicon oxide-based material and 10% to 50% by weight of graphite, wherein the graphite has a BET surface area of 2 ^m² /g to 100 ^m² /g. 4. The lithium-ion battery of claim 1 or claim 2, wherein the second binder polymer is selected from the group consisting of: polyvinylidene fluoride, carboxymethyl cellulose, styrene-butadiene rubber, lithium-ionized polyacrylic acid, copolymers thereof, and mixtures thereof. 5. The lithium-ion battery of claim 1 or claim 2, wherein the nickel-rich lithium nickel manganese cobalt oxide is represented by the formula LiNi _x Mn _y Co _z O ₂ , where x + y + z = 1, 0.50 ≤ x, 0.03 ≤ y ≤ 0.325, and 0.03 ≤ z ≤ 0.325. 6. The lithium-ion battery of claim 1 or claim 2, wherein the positive electrode active material further comprises 20% to 80% by weight of a lithium metal oxide rich in (lithium + manganese), said lithium metal oxide rich in (lithium + manganese) is represented by the formula Li _1+b Ni _α Mn _β Co _γ A _δ O _2-z F _z , wherein b+α+β+γ+δ＝1, b is in the range of 0.04 to 0.3, α is in the range of 0 to 0.4, β is in the range of 0.2 to 0.65, γ is in the range of 0 to 0.46, δ is in the range of 0 to 0.15, and z is in the range of 0 to 0.2, provided that α and γ are not both 0, and wherein A is a metal different from lithium, manganese, nickel, and cobalt. 7. The lithium-ion battery of claim 1 or claim 2, wherein the lithium-ion battery further comprises supplemental lithium in an amount of 80% to 180% of the irreversible capacity loss of the negative electrode during the first cycle, wherein the ratio of the negative electrode capacity to the positive electrode capacity is 1.10 to 1.95 during the fourth cycle at a discharge rate of C/3, and wherein the lithium-ion battery has an energy density of at least 235 Wh/kg when discharged at 30°C at a C/3 rate from a selected charging voltage to 2.5V, wherein the selected charging voltage is 4.15V to 4.35V. 8. The lithium-ion battery according to claim 1 or claim 2, wherein the lithium-ion battery further comprises: Multiple negative electrodes; Multiple positive electrodes The container is prism-shaped, and the assembled battery has a capacity of at least 2Ah when cycled from 4.35V to 2V at a rate of C/3. 9. The lithium-ion battery of claim 1 or claim 2, wherein when cycled at a rate of C/3 between 2.5V and a selected charging voltage from the 10th cycle to the 450th cycle, the capacity of the lithium-ion battery at the 450th cycle is at least 80% of the capacity at the 5th cycle, wherein the selected charging voltage is 4.15V to 4.35V. 10. A negative electrode for a lithium-ion battery, the negative electrode comprising 78% to 92% by weight of an active material, 1% to 7% by weight of nanoscale conductive carbon, and 6% to 20% by weight of a polymer binder, wherein the polymer binder comprises a blend of 60% to 95% by weight of a polyimide with an elongation of at least 50% and at least 5% by weight of a second polymer binder with an elastic modulus not greater than 2.4 GPa. 11. The negative electrode of claim 10, wherein the active material comprises a silicon-based composition. 12. The negative electrode of claim 11, wherein the silicon-based composition comprises a silicon oxide-silicon-carbon composite composition. 13. The negative electrode of claim 11 or claim 12, wherein the active material comprises 5% to 60% by weight of graphite blended with the silicon-based composition. 14. The negative electrode of claim 10 or claim 11, wherein the second polymer binder comprises polyvinylidene fluoride, carboxymethyl cellulose, styrene-butadiene rubber, lithium-ionized polyacrylic acid, copolymers thereof, and mixtures thereof, and has an elongation of at least 35%. 15. The negative electrode of claim 10 or claim 11, wherein the polyimide has a tensile strength of at least 60 MPa. 16. The negative electrode of claim 10 or claim 11, wherein the nanoscale conductive carbon comprises carbon black, carbon nanofibers, carbon nanotubes, or combinations thereof. 17. The negative electrode of claim 10 or claim 11, wherein the negative electrode comprises 78% to 88% by weight of active material, 2% to 6% by weight of nanoscale conductive carbon and 8% to 18% by weight of polymer binder. 18. The negative electrode of claim 10 or claim 11, wherein the polymer binder comprises at least 10% by weight of a second polymer binder with an elastic modulus not greater than 2 GPa. 19. The negative electrode of claim 10 or claim 11, wherein the active material comprises 50% to 90% by weight of a silicon oxide-based material and 10% to 50% by weight of graphite, wherein the graphite has a BET surface area of 2 ^m² /g to 100 ^m² /g. 20. The negative electrode of claim 10 or claim 11, wherein the negative electrode has a specific capacity of at least 1000 mAh/g when cycled relative to lithium metal at a rate of C/3 from 5 mV to 1.5 V. 21. A lithium-ion battery, the lithium-ion battery comprising: The negative electrode comprises a silicon oxide-based material, graphite, nanoscale conductive carbon, and a polymer binder, wherein the polymer binder comprises a blend of 60% to 95% by weight of a polyimide with an elongation of at least 50% and at least 5% by weight of a second binder polymer with an elastic modulus of not more than 2.4 GPa. The positive electrode comprises lithium nickel cobalt manganese oxide, conductive carbon, and a polymer binder; A diaphragm is provided between the negative electrode and the positive electrode. Electrolytes, the electrolyte comprising lithium salts and non-aqueous solvents; and A container that encapsulates other battery components; The lithium-ion battery has an energy density of at least 235 Wh/kg when discharged from a selected charging voltage to 2.5V at a rate of C/3 at 30°C, and when cycled from the 5th cycle to the 450th cycle at a rate of C/3 between 2.3V and the selected charging voltage, the capacity of the lithium-ion battery at 450 cycles is at least 80% of the capacity at the 5th cycle, wherein the selected charging voltage is from 4.05V to 4.375V. 22. The lithium-ion battery of claim 21, wherein the negative electrode comprises 75% to 92% by weight of active material, 1% to 7% by weight of nanoscale conductive carbon and 7% to 20% by weight of polymer binder, wherein the active material comprises 40% to 95% by weight of silicon oxide-based material and 5% to 60% by weight of graphite. 23. The lithium-ion battery of claim 21, wherein the negative electrode active material comprises 50% to 90% by weight of a silicon oxide-based material and 10% to 50% by weight of graphite, and wherein the second binder polymer is selected from the group consisting of: polyvinylidene fluoride, carboxymethyl cellulose, styrene-butadiene rubber, lithium-ionized polyacrylic acid, copolymers thereof, and mixtures thereof. 24. The lithium-ion battery of claim 21, wherein the positive electrode comprises a nickel-rich lithium nickel cobalt manganese oxide represented by the formula LiNi _x Mn _y Co _z O ₂ , wherein x + y + z = 1, 0.45 ≤ x, 0.025 ≤ y ≤ 0.35, and 0.025 ≤ z ≤ 0.35. 25. The lithium-ion battery of claim 24, wherein 0.525 ≤ x ≤ 0.7, and wherein the selected charging voltage is from 4.25V to 4.375V. 26. The lithium-ion battery of claim 24, wherein 0.7 ≤ x ≤ 0.9, and wherein the selected charging voltage is from 4.05V to 4.285V. 27. The lithium-ion battery of claim 21 or claim 22, the lithium-ion battery having a total capacity of at least 2 Ah, and wherein the container has a prism shape with an area of 10,000 ^mm² to 50,000 ^mm² and a volume of 30,000 ^mm³ to 800,000 ^mm³ . 28. The lithium-ion battery of claim 21 or claim 22, further comprising a plurality of negative electrodes and a plurality of positive electrodes, and having an energy density of at least 245 Wh/kg when discharged from a selected voltage to 2.5V at a rate of C/3 at 30°C. 29. The lithium-ion battery of claim 21 or claim 22, wherein the lithium-ion battery further comprises supplemental lithium in an amount of 80% to 180% of the irreversible capacity loss of the negative electrode during the first cycle, and the ratio of the negative electrode capacity to the positive electrode capacity at a discharge rate of C/3 during the fourth cycle is 1.10 to 1.95, and when cycled at a discharge rate of C/3 between 2.5V and a selected voltage from the 10th cycle to the 450th cycle, the capacity of the lithium-ion battery at the 450th cycle is at least 80% of the capacity at the 5th cycle.