Patent Application
20250210760
An electrode is a conductor that is used to establish electrical contact with a non-metallic part of a circuit. Electrodes are used in a wide range of applications, including in batteries, electrochemical cells, and various types of sensors. In medical applications, electrodes are used to measure electrical signals in the body, such as in electrocardiograms (ECGs) and electroencephalograms (EEGs).
Electrodes can be of various shapes and sizes, depending on their intended use. Some common shapes include rods, plates, wires, and discs.
According to various aspects, an electrochemical storage device includes an electrode. The electrode includes TiNb2O7, LiCoO2, LiNi0.5Mn0.5O2, lithium metal, LiMn2O4, Li4Ti5O12, a mixture of LiMn2O4 and LiCoO2, or mixture thereof.
The drawings illustrate generally, by way of example, but not by way of limitation, various aspects of the present invention.
Reference will now be made in detail to certain aspects of the disclosed subject matter, examples of which are illustrated in part in the accompanying drawings. While the disclosed subject matter will be described in conjunction with the enumerated claims, it will be understood that the exemplified subject matter is not intended to limit the claims to the disclosed subject matter.
Throughout this document, values expressed in a range format should be interpreted in a flexible manner to include not only the numerical values explicitly recited as the limits of the range, but also to include all the individual numerical values or sub-ranges encompassed within that range as if each numerical value and sub-range is explicitly recited. For example, a range of “about 0.1% to about 5%” or “about 0.1% to 5%” should be interpreted to include not just about 0.1% to about 5%, but also the individual values (e.g., 1%, 2%, 3%, and 4%) and the sub-ranges (e.g., 0.1% to 0.5%, 1.1% to 2.2%, 3.3% to 4.4%) within the indicated range. The statement “about X to Y” has the same meaning as “about X to about Y,” unless indicated otherwise. Likewise, the statement “about X, Y, or about Z” has the same meaning as “about X, about Y, or about Z,” unless indicated otherwise.
In this document, the terms “a,” “an,” or “the” are used to include one or more than one unless the context clearly dictates otherwise. The term “or” is used to refer to a nonexclusive “or” unless otherwise indicated. The statement “at least one of A and B” has the same meaning as “A, B, or A and B.” In addition, it is to be understood that the phraseology or terminology employed herein, and not otherwise defined, is for the purpose of description only and not of limitation. Any use of section headings is intended to aid reading of the document and is not to be interpreted as limiting; information that is relevant to a section heading may occur within or outside of that particular section.
All publications, patents, and patent documents referred to in this document are incorporated by reference herein in their entirety, as though individually incorporated by reference. In the event of inconsistent usages between this document and those documents so incorporated by reference, the usage in the incorporated reference should be considered supplementary to that of this document; for irreconcilable inconsistencies, the usage in this document controls.
In the methods described herein, the acts can be carried out in any order without departing from the principles of the invention, except when a temporal or operational sequence is explicitly recited. Furthermore, specified acts can be carried out concurrently unless explicit claim language recites that they be carried out separately. For example, a claimed act of doing X and a claimed act of doing Y can be conducted simultaneously within a single operation, and the resulting process will fall within the literal scope of the claimed process.
The term “about” as used herein can allow for a degree of variability in a value or range, for example, within 10%, within 5%, or within 1% of a stated value or of a stated limit of a range, and includes the exact stated value or range. The term “substantially” as used herein refers to a majority of, or mostly, as in at least about 50%, 60%, 70%, 80%, 90%, 95%, 96%, 97%, 98%, 99%, 99.5%, 99.9%, 99.99%, or at least about 99.999% or more, or 100%.
The term “substantially free of” as used herein can mean having none or having a trivial amount of, such that the amount of material present does not affect the material properties of the composition including the material, such that about 0 wt % to about 5 wt % of the composition is the material, or about 0 wt % to about 1 wt %, or about 5 wt % or less, or less than or equal to about 4.5 wt %, 4, 3.5, 3, 2.5, 2, 1.5, 1, 0.9, 0.8, 0.7, 0.6, 0.5, 0.4, 0.3, 0.2, 0.1, 0.01, or about 0.001 wt % or less, or about 0 wt %.
Lithium-ion (Li-ion) batteries are an indispensable technology which have been applied in many applications including automobiles, portable electronic devices, and the like. Compared to other rechargeable batteries, Li-ion cells have high energy and power density. However, the performance demands for energy storage are still increasing and necessitate further development. Conventional Li-ion battery electrodes are composites consisting of electroactive material, conductive additive, and polymer binders. Recently, electrodes composed of porous thin films containing only electroactive material that has undergone a thermal treatment (“sintered electrodes”) have been studied. Sintered electrodes can be made much thicker than composite electrodes, and thus the energy density at the cell level can be increased. In some examples a “sintered electrode” can alternatively be refereed to as a “all active material electrode”. This can mean that there is no electrochemically inactive material added to the electrode.
Reversible electrochemical cycling of sintered electrodes has been accomplished with LiCoO2 (LCO) cathodes and Li4Ti5O12 (LTO) anodes. However, these very thick electrodes and are found to result in ion transport resistances, which limits the ability to charge/discharge at high rates. Routes to mitigate the transport restrictions have included processing electrodes with reduced tortuosity and incorporating electrolytes with increased ion concentration/conductivity. Cell improvements (especially energy density) can be achieved by changing the active material components. Alternative materials evaluated for sintered cathodes include LiMn2O4 (LMO) and LiFePO4 (LFP). LMO and LFP have environmental and cost benefits, though not necessarily higher energy density, than LCO but have electronic conductivity limitations. LTO is a popular anode option due to minimal strain during Li+ insertion/extraction and an operating voltage within the stability window of carbonate electrolytes. However, the gravimetric capacity of LTO (175 mAh g−1) is relatively low compared to other Li-ion anode materials.
As disclosed herein, an alternative anode with electrochemical capacity within the stability window of carbonate electrolytes is TiNb2O7 (TNO). TNO has higher theoretical gravimetric capacity (388 mAh g−1) and density (4.3 g cm−3 TNO vs. 3.5 g cm−3 LTO) than LTO, thus providing a significant opportunity to increase volumetric capacity as a sintered electrode anode.
According to an aspect of this disclosure, TNO was synthesized via sol-gel method. As further disclosed, TNO can be processed into both composite and sintered electrodes and evaluated electrochemically when paired in half and full cell configurations. The outcomes disclosed herein show that TNO is a suitable material as a sintered anode, providing stable and reversible electrochemical cycling.
More specifically, the instant disclosure shows that the energy density of lithium-ion batteries at a cell level can be improved via increasing thickness and reducing inactive material content of electrodes. One system which achieves both attributes are sintered electrodes comprised of porous thin films of only electroactive material. Li4Ti5O12 has often been the used as a sintered anode, however, higher energy density anodes could significantly improve cell energy density. As disclosed herein, TiNb2O7 (TNO) was synthesized and evaluated as a sintered anode material. Sintered TNO had stable cycling and relatively high volumetric energy density, suggesting TNO has promise as a sintered anode.
As mentioned previously, examples of two routes to achieve high energy densities at the cell level are to reduce the mass fraction of inactive materials such as conductive additives and binders, and to increase electrode thickness. For both routes there are tradeoffs: electronic conductivity and mechanical flexibility/strength limitations for reducing/eliminating inactive materials and increased mass transfer resistances from using thick electrodes. However, one way to reduce inactive materials in the electrode and to increase thickness is to use electrodes comprised solely of electroactive materials. Such electrodes in some cases undergo a mild thermal treatment to improve the mechanical properties and will be referred to as “sintered electrodes” herein.
Sintered electrodes have been reported which are free of inactive conductive additives and binders, with high loadings exceeding 150 mg cm−2 and thicknesses over 500 μm. The electrode microstructure for sintered electrodes does not contain inactive additives in the interstitial regions between particles, and thus sintered electrodes have lower tortuosity than conventional composite electrodes. However, the electrodes are still very thick, and thus ion transport limitations through the microstructure and electron transport through the electrode matrix can result in high polarization and rate capability limitations. Because electron conduction through the electrode matrix must proceed through the electroactive material itself in sintered electrodes, materials with relatively high electronic conductivity across the range of extents of lithiation experienced during charge/discharge of the cell are desirable. Thus, cathode materials such as LiCoO2 (LCO) have been used in multiple sintered electrode studies. LCO is well suited to such electrodes due to its relatively high electronic conductivity: reported to range from 10−2 to 102 S cm−1 from Li1CoO2 to Li0.55CoO2.
Previous studies with composite cathodes have investigated using multiple cathode materials within the same electrode to improve electrode capabilities or properties. For example, a composite electrode blend using LiMn2O4 (LMO) and LiNixCo1-x-yAlyO2 (NCA) demonstrated combined benefits of lower cost, higher operating voltage, and better rate capability from the LMO component and higher capacity, longer storage life, and greater stability from the NCA component.
As disclosed herein, the concept of combining multiple electrode materials is explored in a sintered electrode system. Composite electrodes experience relatively low temperatures during processing, and the individual active material particles (at least before calendaring) are generally separated from one another. In contrast, sintered electrode electroactive material particles have many contact points due to compression during processing, and then are subjected to temperatures which are mild for sintering but much higher than composite electrode solvent removal temperatures. Ideally, the benefits previously observed for multicomponent composite electrodes would translate directly to sintered electrodes; however, the dissimilar materials processed in direct contact with one another may result in unique considerations as will be discussed herein.
In addition, as described above for thick sintered electrodes, the electronic conductivity through the electrode matrix is dependent upon the electrode active material itself. Thus, LCO was chosen as one of the constituents due to its relatively high electronic conductivity as discussed earlier. For the second material in this disclosure, the cathode material LMO was chosen. LMO has been used commercially in composite electrodes, and has benefits relative to LCO with regards to cost due to the higher relative abundance of Mn compared to Co, and may also have environmental benefits. However, the low electronic conductivity of LMO results in high polarization and limited rate capability in a thick sintered electrode system. Thus, combinations of LMO and LCO will be explored as a multicomponent thick sintered electrode in this disclosure.
According to various examples of the instant disclosure, sintered cathodes containing both LMO and LCO, three different situations will be described. The first is a homogeneous blend, where powders of the two materials are blended together and then processed into a sintered electrode, which will be referred to as “Blend” (
In view of the foregoing, in accordance with various aspects, an electrochemical storage device can include an electrode. The electrode can include many suitable materials. Examples of suitable materials can include TiNb2O7, LiCoO2, LiNi0.5Mn0.5O2, lithium metal, LiMn2O4, Li4Ti5O12, a mixture of LiMn2O4 and LiCoO2, or mixture thereof. Depending on various aspects, the electrode can be an anode or a cathode. In specific examples, a cathode can include LiCoO2. In some examples, the LiCoO2 includes at least one dopant, at least two dopants, or any suitable number. The at least one dopant has a 1+, 2+, or 3+ oxidation state. As non-limiting examples, the at least one dopant comprises copper, aluminum, sulfur, potassium, or a mixture thereof.
The benefits of the electrodes described herein can be achieved when the electrode is a sintered electrode or a composite electrode. In examples where the electrode is a composite, the electrode can include a metallic substrate to which the TiNb2O7, LiCoO2, LiNi0.5Mn0.5O2, lithium metal, LiMn2O4, Li4Ti5O12, a mixture of LiMn2O4 and LiCoO2, or mixture thereof is applied. While not so limited, the metallic substrate can include a stainless steel, aluminum, alloys thereof, or mixtures thereof.
As explained herein, electrode thickness is an important parameter of the electrode. As an example, a thickness of the electrode can be in a range of from about 50 μm to about 2000 μm, about 70 μm to about 400 μm, less than, equal to, or greater than about 50 μm, 60, 70, 80, 90, 100, 110, 120, 130, 140, 150, 160, 170, 180, 190, 200, 210, 220, 230, 240, 250, 260, 270, 280, 290, 300, 310, 320, 330, 340, 350, 360, 370, 380, 390, 400, 410, 420, 430, 440, 450, 460, 470, 480, 490, 500, 510, 520, 530, 540, 550, 560, 570, 580, 590, 600, 610, 620, 630, 640, 650, 660, 670, 680, 690, 700, 710, 720, 730, 740, 750, 760, 770, 780, 790, 800, 810, 820, 830, 840, 850, 860, 870, 880, 890, 900, 910, 920, 930, 940, 950, 960, 970, 980, 990, 1000, 1110, 1120, 1130, 1140, 1150, 1160, 1170, 1180, 1190, 1200, 1210, 1220, 1230, 1240, 1250, 1260, 1270, 1280, 1290, 1300, 1310, 1320, 1330, 1340, 1350, 1360, 1370, 1380, 1390, 1400, 1410, 1420, 1430, 1440, 1450, 1460, 1470, 1480, 1490, 1500, 1510, 1520, 1530, 1540, 1550, 1560, 1570, 1580, 1590, 1600, 1610, 1620, 1630, 1640, 1650, 1660, 1670, 1680, 1690, 1700, 1710, 1720, 1730, 1740, 1750, 1760, 1770, 1780, 1790, 1800, 1810, 1820, 1830, 1840, 1850, 1860, 1870, 1880, 1890, 1900, 1910, 1920, 1930, 1940, 1950, 1960, 1970, 1980, 1990, or about 2000 μm.
A porosity of the electrode can be in a range of from about 30% to about 50%, about 35% to about 40%, less than, equal to, or greater than about 30%, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, or about 50%. The porosity can alternatively be expressed as a void space of the portion of the electrode that is empty. The pores can be through pores in some examples.
According to various examples, the electrode is substantially free of any conductive additives. For example, the electrode can include about 0 wt % to about 5 wt % conductive additive, about 0 wt % to about 0.50 wt %, less than, equal to, or greater than about 0 wt %, 0.1, 0.2, 0.3. 0.4,.0.5, 0.6, 0.7, 0.8, 0.9, 1, 1.1, 1.2, 1.3, 1.4, 1.5, 1.6, 1.7, 1.8, 1.9, 2, 2.1, 2.2, 2.3, 2.4, 2.5, 2.6, 2.7, 2.8, 2.9, 3, 3.1, 3.2, 3.3, 3.4, 3.5, 3.6, 3.7, 3.8, 3.9, 4, 4.1, 4.2, 4.3, 4.4, 4.5, 4.6, 4.7, 4.8, 4.9, or about 5 wt %. Non limiting examples of conductive additives that the electrode is free of include a polymer binder, a conductive carbon, or a mixture thereof.
The material(s) and construction of the electrode can control the capacity of the electrode. As an example, the capacity of the electrode can be in a range of from about 7 mAh/cm2 to about 100 mAh/cm2, about 10 mAh/cm2 to about 50 mAh/cm2, less than, equal to, or greater than about 7 mAh/cm2, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, or about 100 mAh/cm2.
The electrode, in some aspects, can be associated with an electrically conductive current collector to facilitate the flow of electrons between the electrode and an exterior circuit. The current collector can include metal, such as a metal foil or a metal grid. In some aspects, the current collector can be formed from nickel, aluminum, stainless steel, copper or the like. The electrode material can be cast as a thin film onto the current collector. The electrode material with the current collector can then be dried, for example in an oven, to remove solvent from the sintered electrode. In some aspects, the dried sintered electrode material in contact with the current collector foil or other structure can be subjected to a pressure, such as, from about 2 to about 10 kg/cm2 (kilograms per square centimeter).
The separator is located between the positive electrode and the negative electrode. The separator is electrically insulating while providing for at least selected ion conduction between the two electrodes. A variety of materials can be used as separators. Commercial separator materials are generally formed from polymers, such as polyethylene and/or polypropylene that are porous sheets that provide for ionic conduction. Commercial polymer separators include, for example, the Celgard® line of separator material from Hoechst Celanese, Charlotte, N.C. Also, ceramic-polymer composite materials have been developed for separator applications. These composite separators can be stable at higher temperatures, and the composite materials can significantly reduce the fire risk. In some further examples the separator can include glass fibers. Glass fiber separators can be especially useful when a sintered electrode is used.
The electrolyte includes solvated ions as electrolytes, and ionic compositions that dissolve to form solvated ions in appropriate liquids are referred to as electrolyte salts. Electrolytes for lithium ion batteries can comprise one or more selected lithium salts. Appropriate lithium salts generally have inert anions. Suitable lithium salts include, for example, lithium hexafluorophosphate, lithium hexafluoroarsenate, lithium bis(trifluorometh yl sulfonyl imide), lithium trifluoromethane sulfonate, lithium tris(trifluoromethyl sulfonyl) methide, lithium tetrafluoroborate, lithium perchlorate, lithium tetrachloroaluminate, lithium chloride, lithium difluoro oxalato borate, and combinations thereof. Traditionally, the electrolyte comprises a 1 M concentration of the lithium salts, although greater or lesser concentrations can be used.
A non-aqueous liquid can be used to dissolve the lithium salt(s). The solvent generally does not dissolve the electroactive materials. Appropriate solvents include, for example, propylene carbonate, dimethyl carbonate, diethyl carbonate, 2-methyl tetrahydrofuran, dioxolane, tetrahydrofuran, methyl ethyl carbonate, γ-butyrolactone, dimethyl sulfoxide, acetonitrile, formamide, dimethyl formamide, triglyme (tri(ethylene glycol) dimethyl ether), diglyme (diethylene glycol dimethyl ether), DME (glyme or 1,2-dimethyloxyethane or ethylene glycol dimethyl ether), nitromethane and mixtures thereof.
The electrodes described herein can be incorporated into various commercial battery designs, such as prismatic shaped batteries, wound cylindrical batteries, coin batteries or other reasonable battery shapes. The batteries can comprise a single sintered electrode stack or a plurality of sintered electrodes of each charge assembled in parallel and/or series electrical connection(s). Appropriate electrically conductive tabs can be welded or the like to the current collectors, and the resulting jellyroll or stack structure can be placed into a metal canister or polymer package, with the negative tab and positive tab welded to appropriate external contacts. Electrolyte is added to the canister, and the canister is sealed to complete the battery. Some presently used rechargeable commercial batteries include, for example, the cylindrical 18650 batteries (18 mm in diameter and 65 mm long) and 26700 batteries (26 mm in diameter and 70 mm long), although other battery sizes can be used.
The sintered electrode can be formed according to many suitable methods. For example, the electroactive material of the electrode is sintered. Sintering can be performed at a temperature in range of from about 500° C. to about 1100° C., about 700° C. to about 900° C., less than, equal to, or greater than about 500° C., 550, 600, 650, 700, 750, 800, 850, 900, 950, 1000, 1050, or about 1100° C. Sintering can be conducted at a constant temperature or at a variable temperature during the sintering process. Sintering can last for any amount of time in a range of from about 0 hours to about 20 hours, about 12 hours to about 18 hours, less than, equal to, or greater than about 0 hours, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, or about 20 hours. Sintering can be conducted in one step or across multiple sintering steps.
Alternatively, a composite electrode can be formed by disposing the electroactive material on a substrate. The electroactive material can be in the form of a slurry mixed with a solvent, carbon source, and any other suitable component. The substrate can be an aluminum foil. After the slurry is disposed on the substrate, the slurry is allowed to dry in an oven to form the electrode.
As mentioned previously, in some aspects the electrode can include a degree of porosity. The structure and/or arrangement of pores in the sintered electrode microstructure can lead to additional improvements in the energy and power density of the energy storage device to which the electrode is incorporated. As an example, aligning the pores can improve these properties. One method that can be used to accomplish this is to fabricate an electrode using reactants that are capable of forming an anisotropic structure. This can be accomplished by producing an anisotropic precipitate particle that includes a single transition metal or multiple transition metals. The precipitate can be used as a precursor template in formulating battery active material having an anisotropic morphology. Any suitable transition metal can be used to form the precursor template.
In producing the sintered electrode, there are different strategies to improve or overcome inherent electronic conductivity limitations of electroactive materials. One route is to dope the electroactive material with elements that result in increased electronic conductivity, and this method has even been applied to sintered electrodes with an example being the material LiMn2O4 (LMO) or TNO. There are limits to the electronic conductivity improvements via doping methods and at too high of a doping level secondary phases can result. Another option is to carbon coat the electroactive material, and this strategy has been attributed to improving rate capability of other low electronic conductivity electroactive materials such as LiFePO4. Carbon coating of electroactive materials has been broadly reported for improved interfacial stability and electronic conductivity. Sucrose is a common carbon source that also can serve as a particulate binder material. As described herein a carbon coated sintered electrode can be produced using sucrose both as a sacrificial binder for sintered electrode processing and as a carbon source, where the carbon coating process does not require an extra step for treating the electroactive material powder or the sintered porous pellet relative to processing the electrodes without the coating.
The carbon coating processing is schematically illustrated as an example in
The electrochemical storage device described herein can be disposed in many suitable articles. For example, the electrochemical storage device can be disposed in vehicle (e.g., an electric vehicle) or an electronic device.
Examples
Various aspects of the present inventive subject matter can be better understood by reference to the following Examples which are offered by way of illustration. The present inventive subject matter is not limited to the Examples given herein.
TiNb2O7 (TNO) was synthesized via sol-gel method adapted from literature. First, 0.01 mol Ti(OC3H7)4 (Sigma-Aldrich) and 0.02 mol NbCl5 (Sigma-Aldrich) was added into 40 mL ethanol with 300 rpm stirring. The solution was heated to 60° C. for 2 hours. Then, the solution was air dried in a fume hood overnight at room temperature to evaporate ethanol. The resulting gel was transferred to an 80° C. oven and dried for 24 hours in air. The resulting powder was washed with deionized (DI) water. Next, the material was fired at 700° C. for 2 h in a Carbolite CWF 1300 box furnace with the heating and cooling rate set to be 1° C. min−1. Powder X-ray diffraction (Empyrean) of the product confirmed the crystal structure of the TNO material was consistent with previous reports.
LiCoO2 (LCO) was synthesized via coprecipitation of CoC2O4·2H2O followed by lithiation of the precursor. First, 1800 mL of 62.8 mmol·L−1 Co(NO3)2·6H2O (Fisher Reagent Grade) solution and 1800 mL of 87.9 mmol L−1 (NH4)2C2O4·H2O (Fisher Certified ACS) solution was prepared and heated to 50° C. Then the Co(NO3)2 was poured into (NH4)2C2O4 solution all at once with 800 rpm stirring to facilitate the mixing. After maintaining at the same temperature for 30 minutes, the solid precipitate was collected using vacuum filtration and rinsed with 4 L DI water. The powder was dried in an 80° C. oven for 24 h in air. The oxalate precursor was mixed with Li2CO3 (Fisher Chemical) with a molar ratio of 1.02:1 for Li:Co using mortar and pestle. Then, the powder mixture was fired at 800° C. in air in a Carbolite CWF 1300 box furnace with a ramp rate of 1° C. min−1. The sample was cooled to room temperature without control of cooling rate after reaching 800° C. The LCO product was then milled in a Fritsch Pulverisette 7 planetary ball mill using 57 zirconia beads (5 mm diameter) at 300 rpm for 5 hours.
LiFePO4 (LFP) cathode powder was purchased from a commercial supplier (Xiamen TOB New Energy Technology) and used as received. Characterization and electrochemical properties of LCO and LFP material used in this study can be found in previous publications.
The composite electrode was made via slurry casting. The active material was mixed with Super P carbon (Alfa Aesar) and polyvinylidene difluoride (PVDF, Alfa Aesar) dissolved in n-methyl-2-pyrrolidone (NMP) solvent. The weight ratio of active material:carbon:PVDF was 8:1:1. The slurry was cast onto an aluminum foil and dried in an 80° C. oven in air atmosphere for 12 hours. The electrode sheet was transferred to a vacuum oven and further dried in 80° C. in vacuum for 3 hours. The resulting electrode was stored in a dry box.
The composite LiNi0.8Co0.15Al0.05O2 (NCA) electrode was provided by Argonne CAMP facility with a loading of 9.9 mg cm−2. For cell fabrication, a 9/16″ discs were punched from electrodes.
TNO or LCO powder was mixed with 1 wt. % polyvinyl butyral (Pfaltz & Bauer) solution dissolved in ethanol (Acros). 1 g powder was blended with 2 mL polymer binder solution with a mortar and pestle. After the solvent evaporated, 0.2 g of the mixture was loaded into a 13 mm diameter Carver pellet die and pressed with 12,000 lbf for 2 minutes in a Carver hydraulic press. The pellets were then heated in a Carbolite CWF 1300 box furnace in an air atmosphere at a ramp rate of 1° C. min−1 from 600° C., held at 600° C. for 1 hour, then cooled at 1° C. min−1 to 25° C.
All electrochemical characterization was conducted using CR2032 coin cells. For TNO composite half cell, the TNO electrode was paired with a 100 μm thick, 9/16″ diameter Li foil disc. For composite full cells, the TNO anode was paired with a composite cathode. 8 drops of electrolyte (1.2 mol L−1 LiPF6 in 3:7 EC:EMC, Gotion) were added for each composite cell. For sintered full cell, the LCO pellets were pasted onto the bottom plate of the coin cell and TNO pellets were pasted on a stainless steel spacer (15.5 mm in diameter and 0.5 mm in thickness). The paste was composed of 1:1 weight ratio Super P carbon to PVDF dissolved in NMP. After drying in an 80° C. oven for 12 hours in air, the pellets were fabricated in to full cells with 16 drops of electrolyte added. Cell fabrication was conducted in an Ar atmosphere glove box with both O2 and H2O content <1 ppm and Celgard 2325 separators were used for all cells.
Galvanostatic charge/discharge cycling was conducted at different C rates using a MACCOR battery cycler. The C rate was based on the mass loading of the cathode material with an assumed capacity (150 mAh g−1 for LCO, 165 mAh g−1 for LFP and 180 mAh g−1 for NCA).
The charge and discharge capacity of a Li/TNO cell with a composite TNO cathode at different rates of charge/discharge can be found
Composite full cell electrochemical capacity (on a LCO cathode basis) for full cells of TNO composite anodes paired with LCO composite cathodes can be found in
The irreversible first cycle capacity of 9% for the TNO/LCO cell likely had significant contribution from the presence of the TNO (with 6% first cycle irreversible capacity,
Stability of TNO in composite electrodes led to further assessment of its suitability as a sintered electrode active material. Sintered electrodes do not have conductive carbon and polymer binder as additives. The electrodes consist of only active materials and were much thicker than the composite electrodes (for TNO ˜70 μm for composite and ˜400 μm for sintered electrodes). TNO sintered anodes were paired with LCO sintered cathodes. Two cell variations will be described: one was a cathode limited condition, where the LCO was 191 mg and TNO was 177 mg (˜38.9 mAh anode and ˜28.6 mAh cathode based on composite electrode capacities). The second condition was anode limited where 147 mg TNO was paired with 250 mg LCO (˜32.3 mAh anode and ˜37.5 mAh cathode based on composite electrode capacities). The gravimetric capacities on both TNO and LCO bases at increasing rates can be found in
The capacity was relatively stable for each cell at each rate (
The rate capability results in
Another difference between the cells was capacity fade with cycling. Although for the cycles within each rate the capacity was stable, comparison of the series of cycles at C/20 (e.g., cycles 6-10 compared with cycles 31-35) revealed different capacity fade. The underlying cause of capacity fade will be the subject of future reports; however, several factors may have contributed to the fade. One factor was that the volume change of TNO material during lithiation/delithiation process has been reported as 7.22%, which was much larger than the volume change of LTO (0.2%) used in previous studies, and also higher than the LCO material (1.9%) used as cathode in this work. This volume change could result in internal stress and impact particle contacts necessary for electronic conductivity through the electrode matrix and access to electrode capacity. Another contributor may have been the different extents and progression of lithiation/delithiation during the cycling. Although the overall cell voltage window was carefully controlled, TNO does not have a flat voltage plateau and thus it is challenging to be confident the overpotential experienced by the active material particles was the same both conditions. There may have been regions within the sintered electrodes where the local potential could have resulted in extra irreversible capacity loss due to exceeding electrolyte stability limits or reversible lithiation extents of the TNO. Overall, sintered TNO/LCO cells showed promising cycling performance.
The use of TNO can improve the energy density at a cell level and ion transport polarization relative to using LTO. For example, a 0.5 mm thick, 35% porosity LCO pellet has a capacity of 32.2 mAh. And thus, a balanced LTO (40% porosity) anode requires a thickness of 0.66 mm. However, a balanced TNO anode needs a thickness of only 0.43 mm, which is a 35% reduction in anode thickness and a 20% reduction in total cell electrode thickness. Assuming a constant cathode, a thinner sintered anode will reduce ion transport resistance and increase volumetric energy density-increasing both the power and energy of the cell.
As demonstrated herein, TNO was synthesized via sol-gel method and processed into composite and sintered electrodes. Composite TNO electrodes showed good cycling stability both in half cells paired with Li metal anodes and full cells paired with composite LCO cathodes. TNO/LCO sintered electrode full cells also had promising reversible electrochemical capacity, although some fade was observed which warrants further investigation. TNO has a relatively high gravimetric and volumetric capacity, and this work demonstrated the cycling of TNO anodes in sintered full cells, which shows promise for providing high cell level energy density when coupling the electrochemical properties of TNO materials with the large thicknesses achievable with sintered electrode processing.
LNMO was synthesized via high temperature lithiation and calcination of a transition metal oxalate precursor. The precursor was synthesized using precipitation methods. 200 mM of sodium oxalate (Na2C2O4) was dissolved into 400 mL of deionized (DI) water within a 1000 ml beaker. 100 mM of manganese sulfate monohydrate (MnC2O4·2H2O, Fisher Chemical) and 100 mM of nickel sulfate hexahydrate (NiSO4·6H2O, Fisher Chemical) were dissolved into 400 mL DI water using a separate 1000 ml beaker. Both solutions were heated to 60° C. before pouring the Mn/Ni sulfate solution into the oxalate solution all at once. The precipitation reaction proceeded for 30 minutes at 60° C. at 300 RPM stirring. Then, the precipitate was collected using vacuum filtration and rinsed with 1.6 L DI water before drying overnight at 80° C. in air. The resulting Ni0.5Mn0.5C2O4·2H2O was mixed with LiOH with a target molar ratio of 1:1.1 transition metal:Li using mortar and pestle by hand for 10 minutes. The powder mixture was then transferred to a furnace (Carbolite CWF 1300) for calcination. The temperature profile for the thermal process included a hold at 480° C. for 3 hours and then 950° C. for 10 hours. Temperature increases occurred at a rate of 1° C. min−1, but the cooling rate back to room temperature was not controlled.
For LCO, 200 mM each of Na2C2O4 and cobalt sulfate heptahydrate (CoSO4·7H2O, Acros Organics) were separately dissolved into 400 mL of DI water into two 1000 ml beakers. The reaction conditions were otherwise the same as the LNMO precursor synthesis. The CoC2O4·2H2O was mixed with Li2CO3 (Fisher Chemical) with a target molar ratio of 2:1.05 oxalate precursor:lithium carbonate. The mixture was heated to 800° C. without a hold at the top temperature using a heating rate of 1° C. min−1. Upon reaching the set temperature, cooling proceeded to room temperature without temperature control. The synthesized powder was then ball-milled (Fritsch Pulverisette 7 planetary ball miller) with zirconia beads of 4.8 mm diameter for 5 hours at 300 RPM using a powder to beads mass ratio of 1:5.
To fabricate sintered electrode cathodes, the synthesized active material powders of LNMO and LCO were used in isolation or in blends. For sintered electrode anodes, the material was purchased from a commercial vendor and was Li4Ti5O12 (LTO, NEI corporation). Material and electrochemical properties for the LCO and LTO materials, both in composite and sintered electrodes, have been reported previously. To prepare a sintered electrode, the active material powder of 1 g was blended with 2 mL of 1 wt % polyvinyl butyral (PVB, Pfaltz & Bauer) in ethanol. The suspension was mixed using mortar and pestle by hand until the ethanol evaporated and the powder appeared dry. PVB-coated powder was used for fabricating all the sintered electrodes.
For sintered electrode cathodes, PVB-coated LNMO and LCO powder was used and blended at the desired mass ratio by hand using mortar and pestle. The compositions used on a LCO mass percentage basis were 0%, 15%, 30%, 45%, 60%, and 100% (denoted as 0% LCO, 15% LCO, 30% LCO, 45% LCO, 60% LCO, and 100% LCO). 0.2 g of resulting cathode powder was loaded into a pellet die (Carver) with diameter of 13 mm before being hydraulically pressed at 430 MPa for 2 minutes (Carver). The pressed pellet was thermally treated by heating to 600° C. at a rate of 1° C. min−1 without a hold and then allowed to cool without control over the cooling rate to room temperature. The sintered pellets had thicknesses ranging from 450 to 550 μm as measured using calipers and loadings ranging from 145 to 150 mg cm−2. Due to the morphology and particle size differences, the pellet geometric void/pore volume fraction for pure LNMO was 0.41 and linearly decreased to 0.33 for pure LCO (see
For sintered LTO anode fabrication, 0.2 g of PVB-coated LTO was pressed following the same procedures as the cathode powders. For thermal treatment of the electrode, the temperature was increased at a rate of 1° C. min−1 to 600° C., held for 1 h, and the temperature was ramped down at a rate of 1° C. min−1 back to room temperature. The sintered LTO electrodes had thickness ranging from 710-720 μm and loadings ranging from 148-152 mg cm−2, with geometric porosity/void fractions of 0.40.
Scanning electron micrographs (SEM) were collected using a FEI Quantum 650. Primary particle sizes were measured for 20 particles in SEM images to calculate mean primary particle size. SEMs were taken for both LNMO and LCO as both loose active material powder and after processing into sintered pellets. X-ray diffraction (XRD) patterns were collected on both powders and pellets using a PANalytical X′pert ProMPD.
For composite electrode cells, the synthesized cathode active material powder (LNMO or LCO) was blended with acetylene carbon black (CB, Alfa Aesar), and polyvinyl pyrrolidone (PVP, Sigma Aldrich, 360 kDa molecular weight) using ethanol (Fisher) solvent with a mass ratio of 8:1:1 electroactive material:CB:PVP. The blend was then processed in a slurry mixer (Thinky AR-100) at 2000 RPM for 4 minutes, sonicated for 5 minutes, and blended in the slurry mixer for an additional 4 minutes at 2000 RPM. The resulting slurry was coated onto an aluminum foil with a doctor blade. The coated composite electrode was vacuum dried at 80° C. before punching into circular discs with an area of 1.33 cm2 and transferring the electrodes into a glove box. The LNMO and LCO composite cathode loadings were approximately 5.1 mg cm−2 and 3.6 mg cm−2. Celgard 2325 punched into circular discs with an area of 1.98 cm2 was used as separator and Li foil discs with an area of 1.60 cm2 were used as anodes. 1.2 M LiPF6 in 3:7 ethylene carbonate:ethyl methyl carbonate (Gotion) was used as electrolyte. Coin cells (2032-type) were assembled and cycled using a multichannel battery cycler (MACCOR) with a voltage range of 2.5 V to 4.4 V (vs. Li/Li+).
For sintered electrode cells, sintered pellet cathode and anode were adhered to a stainless steel bottom plate and a spacer, respectively, using a custom carbon paste. The carbon paste was made by blending a slurry consisting of by weight 4.76%:4.76%:90.48% CB:PVP:ethanol. The attached pellet electrodes were vacuum dried at 80° C. for 1 h before transferring to the glove box. Glass fiber (Fisher, type G6 circles) was used as a separator and the same electrolyte (Gotion) as for the composite electrode cells was used. Cells were cycled between 1.0 V and 2.85 V, where 1.13 mA cm−2 was used as C/20 rate, assuming the gravimetric capacity of cathode material to be 150 mAh g−1 based on a loading of ˜150 mg cm−2. Each individual charge and discharge cycle used the same current density, although the current density/rate applied was systematically varied for rate capability testing. Rate capability tests were conducted using C rates varying from C/20 to C/2.5.
The XRD patterns for both LNMO and LCO exhibited no impurity peaks (for the patterns see
XRD patterns for all sintered electrodes can be found in,
SEM images of the surfaces for each sintered electrode composition can be found in
Disordered jammed sphere packing, which may be a reasonable representation of the sintered electrodes, has been reported to have a critical volume fraction to achieve percolation for a given sphere material in the system of 0.199 for a three-dimensional system. Based on the reported crystal density for LNMO and LCO, the LCO volume fraction reached 0.17 for 30% LCO (slightly below the threshold) and 0.27 for 45% LCO (well above the threshold). The calculated volume fraction of LCO for each sample can be found in
To obtain data most representative of the intrinsic electroactive particle charge/discharge profiles without the electronic conductivity and large thickness complications of the sintered electrodes, LCO was first evaluated in conventional composite electrodes paired with Li metal anodes. With a voltage window of 2.5 V to 4.4 V, LCO reached 140 mAh g−1 cathode discharge for the first cycle at C/10 (
For the sintered LCO electrode (100% LCO) paired with sintered LTO using a voltage window from 1.0 V to 2.85 V (corresponding to 2.56 V to 4.41 V vs. Li/Li+, assuming the OCV of LTO was 1.56 V vs. Li/Li+), the first cycle at C/80 attained nearly the same discharge capacity (139 mAh g−1 cathode) as the composite cathode cells. The sintered electrode cell dQ/dV had a single peak at 2.3 V, which was consistent with previous reports. At higher rates of C/20, C/10, C/5, and C/2.5, the sintered electrode cell discharge capacity was 112 mAh g−1, 62 mAh g−1, 29 mAh g−1, and 12 mAh g−1, respectively, on an LCO mass basis. Such decrease in capacity was expected to have originated from the ionic transport limitations though the thick electrode microstructure for this material pairing, as has been described thoroughly in previous reports. The voltage profiles and rate capability results can be found in
For LNMO material evaluated in a composite cell paired with Li metal anode, the same voltage window as with LCO of 2.5 V to 4.4 V was used. The first discharge cycle at C/10 reached 163 mAh g−1 cathode. The dQ/dV of LNMO for that cycle compared to that of LCO had a broader peak located at slightly lower voltage of 3.75 V, consistent with previous literature.
For the sintered LNMO cathode cell, the first discharge at C/80 (0.282 mA cm−2) only reached 97 mAh g−1 cathode. Based on the composite electrode dQ/dV result and the voltage offset from using LTO in the sintered cell, a peak was expected at 2.16 V; however, a clear dQ/dV peak was not observed for this discharge. The discharge capacity retention of the sintered cell relative to the composite cathode cell for LNMO was 60%, while for LCO it was 99% at C/80. Such differences were hypothesized to result from the much lower electronic conductivity of the LNMO, which has been reported with a range from 10−5 to 10−6 S cm−1 for the pristine material before cycling. These electronic conductivity values are more than 4 orders of magnitude lower than LCO, where for over 95% of the lithiation range typically accessed for LCO the electronic conductivity is over 10−1 S cm−1. Cells cycled at C/200 and C/500 delivered 117 mAh g−1 LNMO and 138 mAh g−1 LNMO on the first discharge cycle, which was 72% and 85% retention relative to composite electrode initial discharge gravimetric capacity.
Interestingly, the first charge capacity at C/200 and C/500 were quite similar (169 mAh g−1 and 171 mAh g−1), where the differences (2 mAh g-1 cathode) were much smaller than that of the discharge (21 mAh g−1 cathode). This outcome would be consistent with the LNMO material having a variable electronic conductivity as a function of lithiation, where the electronic conductivity increases upon delithiation similar to other layered materials such as LCO. Thus, the charging process may have extracted all of the available capacity from the LNMO active material, facilitated by the improved electronic conductivity as the material was delithiated. However, as more Li intercalated into the LNMO crystal during discharge, the electronic conductivity would have decreased. This severe drop would have resulted in electronic overpotential which likely limited the electrochemical capacity that was able to be discharged. The later P2D simulations section contains a more detailed discussion of the impacts of LNMO electronic conductivity.
Although LNMO has lower electronic conductivity than LCO, at C/10 the sintered LNMO discharged 63 mAh g−1, which was greater than the pure sintered LCO. This was likely due to a few of other contributing factors. First, as mentioned earlier it was suspected that LNMO has an increasing electronic conductivity with delithiation, and thus was able to extract that level of capacity at C/10 before reaching extents of lithiation during discharge where the electronic conductivity dropped and the resulting overpotential limited the ability to deliver additional capacity. Second, another factor beyond electronic conductivity was the potential of the electrochemical reactions for the different cathode materials (e.g., thermodynamic or OCV factors). LNMO had an overall lower OCV for electrochemical charge/oxidation compared to LCO. As an example, a large charging voltage plateau region for LCO was ˜0.14 V higher than that of LNMO. The lower charging potential for extracting lithium and capacity means that after reaching an equivalent charging capacity that LNMO has greater potential range remaining before hitting the cutoff voltage, where the cutoff was the same for all sintered electrode full cells. Thus, the relative thermodynamic driving force for extracting additional capacity on charge at greater charging extents was expected to be greater for LNMO relative to LCO. In addition, porosity of sintered LCO electrodes (˜0.33) was lower than that of LNMO electrode (˜0.41), likely due to morphology differences for the powders. Thus, the overpotential from ion transport was greater for the LCO electrodes relative to LNMO electrodes. Ion transport through the electrode microstructure has previously been demonstrated to be a major limiting factor for thick sintered electrodes, and the higher porosity for the LNMO would improve the relative ionic conductivity of these electrodes. At higher rates such as C/5 the sintered LMNO delivered very low capacity (6 mAh g−1), suggesting the low electronic conductivity of LNMO (0% LCO) was too resistive at that rate to facilitate much of the electrode achieving high extents of lithiation. Voltage profiles and rate capability results starting from C/20 is shown in
As the relative LCO content in the sintered electrodes increased, at the slow rate of C/80, the discharge capacity became closer to the composite electrode gravimetric limit/line. This outcome was attributed to the increasing electronic matrix conductivity expected as the LCO relative content was increased. As the relative LCO content was increased from 0% LCO to 15% LCO at the slow rate of C/80, the discharge capacity dramatically increased from ˜97 to ˜141 mAh g−1 cathode, although the increase in capacity with further increases in LCO was minimal-the discharge capacity more or less was the same for the different blend compositions. At the higher rates, C/10 and C/5 in particular, the greatest discharge capacity was observed for the 45% LCO. This result was speculated to result from a combination of (i) reaching above a threshold LCO fraction for a well-connected LCO electronically conductive percolated network; (ii) still having significant LNMO content with its OCV function that had greater lower voltage capacity and resulted in greater driving force when approaching the higher voltage charging cutoff; and (iii) an intermediate porosity to the pure LCO and LNMO sintered electrodes to at least provide lower ionic overpotential relative to the pure LCO sintered electrodes. In summary, the combination of material (LCO) with relative higher electronic conductivity and lower gravimetric capacity with material (LNMO) with relative higher gravimetric capacity and porosity but lower electronic conductivity resulted in increased gravimetric discharge capacity than either of these materials used alone in sintered electrodes.
Closer inspection of the first discharge voltage profile at C/80 revealed that as the LCO content in the sintered electrodes increased, the average discharge voltage also increased. This was a consequence not only of the higher electronic conductivity of LCO compared to LNMO, but the intrinsic properties of the material with regards to the potentials at which the intercalation reactions occur. The relative impact of the voltage of the electrochemical reactions of the different materials can be further informed by examining the dQ/dV of the first discharge voltage profile at C/80. As mentioned previously, pure LNMO (0% LCO) had a very broad dQ/dV across the voltage range without a clear peak. However, even the 15% LCO sintered cathode had a clear dQ/dV peak assigned as arising from the LNMO component at 2.16 V (vs. LTO/anode). As the LCO content in the electrodes was increased from 15% LCO towards 100% LCO, the dQ/dV peak that was due to the LCO component (at 2.3 V) increased while the LNMO dQ/dV peak decreased, as would be expected for their changing relative compositions in the electrodes. However, the LNMO peak intensity change was not proportional to the LNMO content, in particular when comparing the dQ/dV peaks for 15% LCO and 30% LCO. This observation may have been due to the more limited improvement in matrix electronic conductivity at these extents of LCO loading, which then did not enable the LNMO to be fully accessible in the electrode for delivering capacity at these discharge rates.
Representative simulation of the discharge of sintered cathodes relies on an accurate electronic conductivity as a function of lithiation for the active materials. The lack of LNMO electronic conductivity as a function of lithiation resulted in a poor match between the simulated and experimental voltage profiles. Using the measured value of 9.5×10−4 S m−1 for as synthesized/fabricated LMNO pellets as a constant electronic conductivity in simulation, the predicted voltage curve at C/20 was not representative of the experimental curve, where the predicted capacity and average discharge voltage were only 18 mAh g−1 cathode and 1.70 V, compared to the experimental measurements of 91 mAh g−1 cathode and 2.15 V. Both LCO and LNMO have a similar layered oxide crystal structure. For LCO, upon delithiation the production of Co4+ and the shortened distance between Co—Co has been reported to result in high electronic conductivity. Although for LNMO the transition metals were Mn and Ni instead of Co, it was speculated that decreasing transition metal-transition metal distances would similarly increase the electronic conductivity of LNMO upon delithiation. The LNMO electronic conductivity at high levels of delithiation was assumed to have a plateau similar to LCO, and then taper down to the experimental value measured at full lithiation. The LNMO value for the plateau region (the “high” conductivity) was adjusted until the error relative to the discharge profile at C/20 was minimized. The corresponding fitted electronic conductivity was plotted and is shown in
An overpotential distribution analysis throughout the discharge simulation was performed and a detailed introduction of the process associated with such analysis can be found in a previous report. In brief, the overpotentials associated with ionic resistance, electronic resistance, interfacial resistance, and OCV differences are calculated at different depths within the cell during the discharge simulation. These overpotentials are then weighted by the electrochemical current being produced at each location, and then they are summed up. Thus, at any given point in the discharge the relative overpotential for each of the resistances mentioned above for the cathode, anode, and separator were calculated. Such analysis provides insights into the factor(s) responsible for deviations between the discharge potential and the thermodynamic maximum potential for a given extent of discharge capacity delivered. In addition, the electrolyte and electrode concentration at selected discharged capacities is shown in
Due to the limited electronic conductivity of the solid phase (smaller than ionic conductivity of electrolyte phase), the initial condition from discharge simulation included a more delithiated region near the current collector for the LNMO electrode (
For the 45% LCO sintered electrode, it was assumed that 1) the LCO was at sufficient volume fraction to form a percolated network, and 2) the LCO electronic conductivity as a function of lithiation was the same as for pure LCO. Note that the matrix electronic conductivity effectively followed the LCO material conductivity because the electronic resistance was so much lower relative to the LNMO material. It is noted that the second assumption would require minimal incorporation of the Ni or Mn into the LCO material in the electrode, which was consistent with the XRD results. The simulated discharge voltage profile at C/20 matched with the experiment well, except that at later points in the discharge (over 80 mAh g−1 cathode) the simulated voltage was slightly higher. Such deviation was possibly due to (i) inhomogeneous distribution of LCO in the electrode; (ii) the electronic conductivity enhancement from LCO was not effective because further in the discharge the LCO was at near full lithiation state with relatively low electronic conductivity (will be elaborated later); and (iii) the co-diffusion of Mn, Ni, and Co from mild sintering was not detected in the bulk but still sufficiently reduced the electronic conductivity of the percolated LCO network relative to pure LCO.
The electrochemical potential of the electrode materials provides the driving force for the electrochemical reactions to occur. This can be analyzed in the context of the OCV of the materials in the electrode, which varies as a function of charge/discharge time and location as the reactions proceed. LCO had ˜90% of its capacity above ˜3.8 V (versus Li/Li+), where LNMO had only ˜56% of its capacity above this voltage. Before moving to the same analysis as the one done to the pure LNMO cell, a capacity contribution separately from each material (LNMO and LCO) was extracted from the simulation. For the first ˜30 mAh g−1 cathode capacity delivered, both LCO and LNMO contributed roughly similar capacity, which could be attributed to the initially sloped OCV functions for both materials. From ˜30 until ˜60 mAh g−1 cathode capacity delivered, the LCO started to contribute more capacity because it had much more capacity at ˜2.3 V (cell voltage, corresponding to the LCO voltage plateau at ˜3.9 V vs. Li/Li+). From ˜60 until ˜115 mAh g−1 cathode capacity delivered, the LNMO replaced LCO as the major capacity contributor due to its voltage plateau at ˜2.1 V (cell voltage, corresponding to the LNMO voltage plateau at ˜3.7 V vs. Li/Li+). For the final portion of delivered capacity until the end of discharge, both LCO and LNMO approximately equally contributed capacity, due to the sloped OCV function and similar to the early portion of the discharge capacity.
The same overpotential distribution analysis as was conducted for the LNMO sintered electrode cell was performed for the cell with the 45% LCO cathode. The Li concentration in the electrode for the solid (averaged based on the wt % of LNMO and LCO) and electrolyte phase can be found in
From 110 mAh g−1 cathode capacity delivered until the end of discharge, a reaction front in the LNMO similar to that observed for LCO earlier was observed. There was also a similar peak that occurred with regards to the ionic overpotential, and more Li was intercalated into the LNMO at the current collector side. Note that the electronic overpotential throughout the discharge was dramatically suppressed from the electronic conductivity enhancement of the blended LCO. The electronic overpotential started to increase at ˜120 mAh g−1 cathode capacity due to the decreasing electronic conductivity of LCO phase. Such decrease in electronic conductivity of LCO led to a decreased voltage, and the resulting voltage was lower than observed experimentally suggesting that the electronic conductivity in the simulation may have been underestimated at high extents of lithiation for the LCO.
The model used to generate data in Example 2 is 1-D implicit numerical, The following are the governing equations used this work:
For pure LNMO cell:
Lithium Flux across Electrode & Electrolyte Interface:
Lithium Flux across Electrode & Electrolyte Interface:
LiMn2O4 (LMO) active material powder was synthesized. 100 mM of sodium oxalate (Na2C2O4, Fisher Chemical) and 10 mM of sodium citrate dihydrate
(Na3C6O7H5·2H2O, Sigma-Aldrich), were dissolved into 400 mL of deionized (DI) water at the same time using a 1000 ml beaker. Within a separate 1000 ml beaker 100 mM of manganese sulfate monohydrate (MnC2O4·2H2O, Fisher Chemical) was dissolved into 400 mL DI water. The two solutions were heated to 60° C. followed by pouring the manganese sulfate solution into the oxalate solution all at once. The reaction was allowed to proceed for 30 minutes at 60° C. and 300 RPM. The precipitate was then collected via vacuum filtration before rinsing with 1.6 L of DI water. The powder cake was dried at 80° C. overnight in air. The resulting precipitate MnC2O4·2H2O was then mixed with Li2CO3 (Fisher Chemical) with a targeted molar ratio of Li:Mn of 1.05:2. The calcination temperature profile used was to ramp the temperature up to 900° C. at a rate of 1° C. min−1, hold at 900° C. for 6 h, ramp the temperature down to 700° C. at 1° C. min−1, hold at 700° C. for 10 h, and then ramp the temperature down to room temperature at a cooling rate of 1° C. min−1.
For LiCoO2 (LCO), the electroactive material was also synthesized from an oxalate precursor precipitate. 200 mM of each sodium oxalate and cobalt sulfate heptahydrate (CoSO4·7H2O, Acros Organics) were dissolved separately into 400 mL of deionized (DI) water in 2 beakers with volumes of 1000 mL. Both solutions were preheated to 60° C., followed by adding the cobalt sulfate solution into the oxalate solution all at once. The reaction solution was maintained at 60° C. and stirred at 300 RPM for the entire 30 minutes reaction duration. The solution containing the precipitates was then processed using vacuum filtration to collect the solid particles. After rinsing with 1.6 L of DI water, the powder cake was dried at 80° C. overnight in an air atmosphere. The resulting CoC2O4·2H2O was blended with Li2CO3 (Fisher Chemical) at a targeted molar ratio of 1.05:1 Li:Co using a mortar and pestle by hand for 10 minutes. After the blended powder was transferred into the furnace (Carbolite CWF 1300), it was heated with a ramp rate of 1° C. min−1 to 800° C. without a hold at the high temperature and allowed to cool down to room temperature without control over the cooling rate. The resulting LCO was then ball-milled (Fritsch Pulverisette 7 planetary ball miller) with 4.8 mm zirconia beads for 5 h and 300 RPM with a powder to bead mass ratio ˜1:5. Typical quantities loaded into the jar were ˜4 g of LCO and ˜20 g of beads (corresponding to 57 beads).
Scanning electron micrographs (SEM) and energy dispersive X-ray spectroscopy (EDS) were conducted using a FEI Quantum 650. Primary particle sizes were measured for 20 particles in SEM images for both active material powder and sintered pellets to calculate the mean primary particles size. Pellets comprised of two layers (LCO and LMO) were mounted such that EDS analysis could be performed through the thickness dimension on the edge of the entire pellet. Powder x-ray diffraction (XRD) patterns were collected using a PANalytical X′pert ProMPD. Crystalline size and strain were calculated from Debye-Scherrer equation using the largest XRD peaks, and Williamson-Hall relations for strain were analyzed using the three largest XRD peaks.
1 g of each electrochemically active material powder (LMO or LCO) was mixed with 2 mL of 1 wt % polyvinyl butyral (PVB, Pfaltz & Bauer) in ethanol, separately, and the suspension was blended by hand with mortar and pestle until it appeared dry. All sintered electrodes were prepared with the PVB-coated active material powders.
For the Blend sintered cathode case, 0.116 g of PVB-coated LCO and 0.100 g of PVB-coated LMO were mixed with mortar and pestle by hand for 5 minutes. The masses were chosen to result in an approximately 50% by volume blend of the two materials (LCO crystal density is 5.0 g cm−3, while LMO crystal density is 4.3 g cm−3). The resulting blended powder was loaded into a circular pellet die (Carver) with an area of 1.33 cm2 and pressed at 430 MPa for 2 mins with a hydraulic press (Carver). For both the CC:LMO:LCO and CC:LCO:LMO cases, 0.116 g of LCO was first loaded in to the pellet die with mild pressure by hand to flatten the powder surface, and then 0.1 g of LMO was loaded above the LCO powder before using the same pressure treatment with the hydraulic press as for all other samples. The pressed pellet was then transferred into the furnace with air atmosphere and heated at a ramp rate of 1° C. min−1 to 600° C., held at 600° C. for 1 h, then cooled back to room temperature at a rate of 1° C. min−1. After heating, pellet thicknesses were measured with calipers and varied between of 510-530 m, and the mass for all samples ranged between 0.204-0.209 g. The approximate geometric pore/void fraction was 34% for the total pellet thickness (it is noted that pellets pressed with only LMO or LCO powders and heated with the same furnace profile also had an approximate geometric pore/void volume of 34%). The 1.16 mass ratio resulted in the same total thickness for all pellets and the same thickness for the LCO and LMO individual layers for the two-layer cathodes. For pure LCO and LMO pellets, 0.2 g of coated powder was loaded into the pellet die and then followed the same procedure with regards to hydraulic compression and heat treatment. The final pure material pellets had porosity of ˜34%, mass of 0.197 g, and thicknesses of 470 μm and 540 μm, respectively. It is noted here that 600° C. was chosen as the processing temperature for the sintered cathodes in an effort to keep the temperature as low as possible to avoid any possible new material/composition interface formation. Temperatures of 400 and 500° C. were also attempted but resulted in fragile pellets which collapsed during handling.
For the anode, Li4Ti5O12 (LTO, NEI corporation) was used as the electroactive material for all sintered electrode cells. 0.2 g of PVB-coated LTO was processed following the same procedure described for the cathode materials above, and the final pellets had thicknesses ranging from 700-720 μm and masses ranging from 0.194-0.199 g, with a geometric pore/void fraction of 40%.
Cathode material powders (100% LCO, 100% LMO, or a mix of 54% LCO 46% LMO by mass) were mixed with polyvinyl pyrrolidone (PVP, Sigma Aldrich, 360 kDa molecular weight) dissolved in ethanol (Fisher) and acetylene carbon black (Alfa Aesar) with a mass ratio of 8:1:1 active material:PVP:carbon black. While PVP is not as commonly used as a Li-ion electrode binder, in this work PVP was used as the binder for a conductive layer adhered between the sintered electrodes and current collectors (described in the next paragraph). PVP was chosen as the binder because it enabled the use of ethanol as a solvent, which resulted in much faster drying and processing of the sintered electrode cells. PVP was then used for processing of composite electrodes to ensure consistency of the binder used across different cell systems. The components were combined via a slurry mixer (Thinky AR-100) at 2000 RPM for 4 minutes, followed by 5 minutes of sonication, and then 4 additional minutes in the slurry mixer at 2000 RPM. The slurry was coated onto an aluminum foil using a 400 μm gap doctor blade. The electrodes were dried in air for 0.5 h followed by vacuum drying at 50° C. for 1 h. Circular shaped 1.33 cm2 cathode discs were punched out, with resulting loadings across all electrodes ranging between 1.4-4.5 mg cm−2.
For sintered electrode cells, to reduce interfacial resistance between the sintered pellet electrode and current collector, conductive carbon paste comprised of acetylene carbon black (Alfa Aesar) and PVP with a mass ratio of 1:1 in ethanol was used between the electrode and current collector (bottom plate for the cathode and metal disc spacer for the anode) of the 2032-type cell. The attached pellet sintered electrodes were dried at 50° C. for 1 hour under vacuum to drive off the ethanol. To assemble the coin cell, 1.2 M LiPF6 in 3:7 ethylene carbonate:ethyl methyl carbonate (Gotion) was used as the electrolyte and glass fiber (Fisher) was used as separator. The cell was cycled between 1.0 V to 2.8 V (cell voltage) with a multichannel battery cycler (MACCOR), where 142.5 mA g−1 cathode electroactive material was assumed to be 1 C, and C rates were adjusted based on the actual mass of cathode material in the coin cell. For the sintered electrode full cells, 1 C would correspond to ˜21 mA cm−2.
For composite electrode cells, Li foil and Celgard 2325 were used as anode and separator, and the same electrolyte was used as that in the sintered electrode cells. All composite cathode cells were cycled between 2.5 V to 4.3 V (versus Li/Li+) where 137, 148,and 142.5 mA g−1 cathode material were assumed to be 1 C for 100% LCO, 100% LMO, or 54% LCO 46% LMO by mass, respectively.
It is noted here that for rate capability evaluation, sintered electrodes were in general cycled at lower C rates compared to composite electrodes. This was due to the major differences between the two types of electrodes with regards to electrode thickness and electroactive material loading. Even with the removal of inactive components from the electrode microstructure the sintered electrodes are much thicker, which limits higher C-rate cycling. Sintered electrodes also have approximately an order of magnitude greater areal loading of active material, which means the areal current densities are approximately and order of magnitude higher for the same C-rate.
For the two-layered sintered cathode simulations, the cathode was divided into 2 regions of LCO and LMO with equal thickness of 255 μm. The initial composition of the assembled cell electroactive materials was assumed to correspond to Li4Ti5O12, LiMn2O4, and LiCoO2 for the respective materials. The simulation then proceeded with charging the cell to 2.8 V at a current density that would correspond to C/50. This was the initial condition before simulating discharge at C/50. The end of discharge simulation at C/50 was then used as the initial condition for simulating the charge at C/20. Then the end of charge at C/20 was used as the initial condition for simulating the discharge at C/20. All physical properties of LCO, LMO, LTO, separator and electrolyte can be found in Supporting Information, Tables 5 and 6. As will be discussed in the following sections, the Blend electrode was not simulated. The voltage plateau below 3 V (versus Li/Li+) of LMO was also not considered in the simulation even though the voltage window used in simulation was 1.0 V to 2.8 V (cell, equivalent to 2.86 V-4.36 V versus Li/Li+).
For sintered electrodes, generally small (<1,000 nm) primary particles have been used in efforts to increase electrochemically active surface area and decrease solid-state transport resistances. Thus, for LCO the active material powder was ball-milled before pressing into a pellet and sintering. The ball-milled LCO powder had primary particle lengths of 220±70 nm, and after sintering the size was determined to be 240±90 nm (averages and standard deviations from measurements of 20 independent particles in the SEM images), suggesting the primary particle size was maintained after the thermal treatment. For LMO active material powder before sintering, the primary particle length was 780±170 nm, and after sintering the size was determined to be 760±190 nm, again suggesting the primary particle size was not impacted by the mild sintering process. Note that the after sintering primary particle measurements reported were for pure LCO or LMO particle contacts, and not for the Blend system.
When looking at the pellet before sintering, the XRD pattern reflected a blend of both LMO and LCO materials, with distinct peaks present for each material and consistent with previous reports. LMO had an Fd-3m spinel structure with a crystalline size of ˜50 nm and strain of 0.0027 determined from XRD analysis; LCO had a R-3m layered structure with a crystalline size of ˜72 nm and strain of 0.0016 determined from XRD analysis, and the peaks were consistent with contributions from each of these phases. Although the heat treatment was relatively mild (heated to 600° C. and held for 1 h), after heat treatment the LCO peaks shifted towards lower values, suggesting an increased lattice size for layered LCO, which could possibly be attributed to the substitution of Co (ionic radius of 56 pm) by Mn (ionic radius of 63 pm).
Similarly, all the LMO features shifted to increasing values, suggesting a decreased lattice size for spinel LMO, where Co may have been incorporated into the LMO lattice as well. In addition, an impurity peak was observed at 31.2° C., which was attributed to the formation of Co3O4 phase. One possible cause of the impurity in the Blend pellet could be diffusion of Li+ from the LCO to the LMO during the heat treatment process, resulting in loss of Li from the LCO and formation of the Co3O4 impurity. The results of the XRD analysis suggested that a large fraction of the electroactive material had undergone significant modifications of its material properties for the Blend pellet, and that new phases and interfaces likely resulted.
Such interface has not been observed after the heat treatment of LCO without LMO directly in contact with the LCO and was not observed for the LCO layer for the two-layer pellet. Heat treated pure LMO had a crystalline size of ˜57 nm and strain of 0.0013 determined from XRD analysis. These results suggest that there was not an obvious change in crystalline size, but there was a reduction in strain. The heat treated pure LCO had a crystalline size of ˜49 nm and strain of 0.0013 determined from XRD analysis. Both the crystalline size and strain were decreased for the heat treatment of the LCO powder.
To further investigate LCO-LMO contact regions after the thermal treatment, SEM and EDS analysis was conducted at the interface region of the two-layered LCO-LMO pellet (
Before examining the electrochemical properties of multi-material cathodes, single material cathodes of only LMO and LCO in both conventional composite half cells and sintered full cells were investigated. For LCO in a composite half cell paired with Li metal and cycled between 2.5 and 4.3 V (vs. Li/Li+), the material achieved 131 mAh g−1 LCO for the first C/20 discharge cycle after a C/20 charge cycle. It is noted that 86% of the discharge capacity was above 3.8 V. After the same heat treatment was applied to LCO powder as was used for sintering of LCO sintered electrode pellets, the LCO capacity was observed to increase to ˜140 mAh g−1 LCO for the first C/20 discharge cycle and had a slight increase in initial cycle coulombic efficiency. The rate capability was also improved after heat treatment. The origin of this moderate improvement in electrochemical properties with heat treatment was not investigated further, but may have been due to slight improvements in cation ordering or defect density with the mild heat treatment.
For the LCO processed into a sintered electrode and paired with a sintered LTO anode, the cell achieved 144 mAh g−1 LCO for the first C/50 discharge cycle after a C/50 charge cycle. A lower rate was used for the sintered electrode cell to minimize impacts of ion transport resistance during initial cell cycling, and the voltage window was 1.0 to 2.8 V (cell). The voltage plateau was distinct and its voltage (2.31 V versus LTO, assuming 1.56 V versus Li/Li+, roughly 3.87 V versus Li/Li+) was close to that in a composite half cell (˜3.9 V versus Li/Li+), suggesting that the single LCO sintered cathode did not have high electronic polarization even in the absence of conductive additives, consistent with previous results. The extra 7 mAh g-1 LCO in the sintered electrode compared to the composite electrode could possibly originate from a higher relative charge voltage window for the sintered electrode, and/or possibly more irreversible capacity loss in the composite electrode due to slightly higher electroactive area per mass particles. At C/20, C/10, and C/5, the sintered LCO cell reached 124 mAh g−1 LCO, 83 mAh g−1 LCO, and 39 mAh g−1 LCO, respectively.
LMO in a composite half cell was paired with Li metal and cycled between 2.5 and 4.3 V (vs. Li/Li+) at a charge and discharge rate of C/20. The electroactive material achieved a discharge capacity of 120 mAh g−1 LMO at 3.5 V (before overlithiating beyond lithium extracted during charge into Li1+xMn2O4 phase), and 184 mAh g−1 LMO at the end of first C/20 discharge. Three distinct voltage plateaus were observed at ˜4.1 V (transition between λ-MnO2 and Li0.5Mn2O4), ˜4.0 V (transition between Li0.5Mn2O4 and LiMn2O4), and ˜2.8 V (transition between LiMn2O4 and Li2Mn2O4), consistent with previous reports for a LMO spinel electrode material paired with Li metal. After the same heat treatment was applied to LMO powder as was used for sintering of LMO sintered electrode pellets, the LMO capacity was not noticeably impacted across all cycles and rates relative to the material that did not undergo the additional heat treatment. All LMO composite electrodes also had severe capacity fade especially at the first three slow charge/discharge cycles due to the Jahn-Teller distortion upon overlithiation when the lower voltage cutoff of 2.5 V was used. When the voltage window was restricted to 3.0 V and 4.3 V, initial discharge capacity was reduced, but the capacity fade was mitigated by avoiding LMO overlithiation.
However, when LMO was used as a sintered electrode paired with sintered LTO and charged and discharged at C/50, the initial C/50 discharge capacity only achieved 106 mAh g−1 LMO. There was a flat voltage plateau at ˜2.45 V (which would correspond to ˜4.01 V in the Li half cell with the composite electrode assuming a flat LTO potential of 1.56 V), however, sintered LMO cathode cell had a significant slope to the polarization curve between ˜2.4 and ˜2.0 V, compared to a much flatter plateau at the corresponding ˜4.0 V region in the Li half cell with the composite electrode. This behavior has been previously reported for LMO sintered electrodes, and has been attributed to the limited electronic conductivity especially near full lithiation of the LMO material—note that LMO has been reported to be ˜4 orders of magnitude lower than LCO in electronic conductivity. The capacity from below 3 V for the composite LMO cell was not present at all in the corresponding regions for the sintered electrode cell, although for the sintered full cell system there was only as much Li+ available from the anode as was intercalated during charge, whereas with the Li metal anode there was an excess Li+ source available to provide the lower potential redox reaction. The sintered LMO electrode also had discharge capacity fade from 106 to 87 mAh g−1 LMO even during the first three cycles at C/50 and down to ˜71 mAh g−1 LMO in the final (15th) cycle which also was at C/50. The capacity fade and limited rate capability of LMO in sintered electrode without conductive carbon was likely due to some combination of the limited electronic conductivity, Jahn-teller distortion, and Mn dissolution. The heat treatment during the sintering process not expected to negatively impact the electrochemical properties of the cathode based on the outcomes for similar processed material in composite electrodes. At C/20, the capacity was only 22 mAh g−1 LMO, and there was almost no capacity at higher rates, which was attributed to the limited electronic conductivity of the material.
For the blended cathode in a composite electrode, the discharge voltage profile had features from both LCO and LMO. The three plateaus from LMO (˜2.9 V, ˜4.0 V and ˜4.1V versus Li/Li+) and one plateau region from LCO (˜3.9 V versus Li/Li+) were distinct, suggesting that physical blending without sintering retained the electrochemical properties of both materials proportional to their loading. The rate capabilities also matched an average between the individual pure cathode material cells. Heat treatment of the individual LMO and LCO powders using the same heat treatment which was applied to during sintered electrode pellet processing and combined in a multicomponent electrode resulted in a combination of what was observed for the pure material electrodes that underwent identical processing. The blend of LCO and LMO powders which had undergone heat treatment had slightly higher discharge capacity at all rates, consistent with the higher capacity for the relative fraction of the LCO component and its electrochemical properties observed for the material in isolation.
The hydraulically pressed and thermally treated Blend sintered electrode had very different electrochemical properties relative to the composite electrode. As shown in
One additional item of note was that although the Blend sintered electrode had very low gravimetric capacity at slow rate, there was relatively high retention of that capacity at higher rates for a sintered electrode. One possible contributor to this outcome may have been the Co diffusion into the LMO lattice from the LCO, which could increase the electronic conductivity of the LMO. The higher conductivity LCO would also be expected to increase the electronic conductivity of the sintered electrode more generally, at least relative to LMO sintered electrodes.
For the sintered electrode cell with a two-layer cathode of CC:LMO:LCO (LMO in contact with the current collector side), for the first cycle at charge and discharge of C/50 the initial discharge capacity was 124 mAh g−1 cathode (
For the CC:LCO:LMO (LCO in contact with the current collector side), for the first cycle at charge and discharge of C/50 the initial discharge capacity was 127 mAh g−1 cathode (
For sintered electrodes, the matrix electronic conductivity was provided by the electroactive material only, and the electronic conductivity of intercalation materials has been reported to be dependent upon the degree of lithiation. Accurate simulations using the P2D model thus required information on the electronic conductivity as a function of lithiation. For LCO sintered cathodes, the electronic conductivity as a function of degree of lithiation has been reported with relatively consistent values. However, for LMO, the electronic conductivity as a function of lithiation has been reported with conflicting trends. A trend of decreasing electronic conductivity with increasing extent of lithiation was found to be consistent with experimental observations for LMO sintered electrode materials. The general trend was that fully delithiated LMO has the highest electronic conductivity, with both voltage plateaus having a corresponding flat electronic conductivity, and LMO lithiated to Li1Mn2O4 had the lowest electronic conductivity. Li1+xMn2O4 region was not considered due to lack of existing literature data and because the relevant capacity region was not significantly observed for sintered full cells. The electronic conductivity was thus aligned with the OCV of the LMO, or the different phases involved during LMO lithiation/delithiation: the two-phase reaction between spinel λ-MnO2 (˜0.64 mS cm−1) and spinel Li0.5Mn2O4 (˜0.14 mS cm−1) followed by a single-phase reaction proceeding to the LiMn2O4 (˜0.07 mS cm−1). Another assumption for the P2D simulation of two-layer sintered electrodes was the treatment of the interface layer between the LMO and LCO. This was treated as a single discretized point between the LCO and LMO layers and was assumed to be electrochemically inactive and having the same electronic conductivity as the adjacent discretized point towards the current collector direction.
For the CC:LMO:LCO two-layer electrode, at a discharge rate of C/50 the simulated and experimental profiles matched well for the first 90 mAh g−1 cathode (
To better assess the overpotentials during discharge for the two-layered CC:LMO:LCO sintered electrodes, a dQ/dV analysis was performed and can be found in
Above 2.6 V (4.16 V vs Li/Li+) LMO would be expected to hardly provide any capacity based on low rate composite electrode dQ/dV profiles. In this voltage region, there was a small amount of capacity (˜10 mAh g−1) that mostly should have been contributed by the LCO. The experimental dQ/dV plot had peaks in differential capacity that were shifted to a slightly lower voltage relative to the OCV dQ/dV, but the dQ/dV capacity peaks for the simulation were shifted to much lower voltages. This outcome suggested an underestimated electronic conductivity of LMO for the λ-MnO2 phase, where the matrix electronic resistance of the LMO was expected to account for the lower peak positions in the dQ/dV experiments.
Even though the onset of the first experimental dQ/dV peak was at ˜2.6 V (vs. anode/LTO), similar to the OCV calculation, the magnitude (˜270 mAh g−1 V−1) and position (2.50 V) of this peak were much lower than the OCV calculation (˜1290 mAh g−1 V−1 and 2.57 V). The OCV dQ/dV LCO contribution above 2.45 V had differential capacity less than ˜100 mAh g−1 V−1. LMO provided almost all of its discharge capacity associated with its higher voltage plateau between ˜2.6 V and ˜2.5 V (from λ-MnO2 to Li0.5Mn2O4 reaction). This result suggested that the smaller observed experimental peak in this voltage region was indeed from the first higher voltage LMO reaction, but that it was delivered at a lower and more sloped voltage profile due to overpotential in the cell. A probable source of that overpotential would be the low electronic conductivity of the Li0.5Mn2O4 phase. For the dQ/dV curve from the P2D simulation, the LMO started delivering capacity at 2.56 V, a lower voltage than observed experimentally. This result supported an underestimated electronic conductivity of the λ-MnO2 phase because that capacity was delivered experimentally at a higher voltage than simulation.
The second lower voltage discharge dQ/dV peak associated with the LMO material can be observed at 2.44 V (vs. anode/LTO) from the OCV calculation. The corresponding experimental differential capacity was a much broader peak of a smaller magnitude at 2.36 V, which initiated capacity at ˜2.40 V. The offset relative to the OCV dQ/dV peak location for the experimental lower voltage LMO peak compared to the higher voltage one was consistent with the Li1Mn2O4 phase being even more resistive than Li0.5Mn2O4 phase and causing increasing polarization as the discharge proceeded and more LiMn2O4 phase was formed. This outcome was also consistent with previous observations for sintered LMO electrodes. From the P2D simulation, the second lower voltage dQ/dV peak attributed to the LMO reaction occurred at 2.40 V and capacity initiated at ˜2.42 V (
In the OCV calculation (
The P2D simulation used an electronic conductivity function that had a decrease near the state of full lithiation of LMO. This decrease in LMO electronic conductivity shifted the LCO capacity down and resulted in a dQ/dV peak at 2.14 V; however, a second lower dQ/dV peak such as that observed experimentally at 2.00 V was not observed in the simulation. To further explore the possibility of the LCO lithiation resulting in two dQ/dV peaks due to changes in LMO electronic conductivity, the discharge simulations were conducted again with an alternative function for LMO electronic conductivity. A change was made where the conductivity at near full lithiation of LMO was reduced to a much lower value of 0.012 mS cm−1 (Table 5). With the modified electronic conductivity function for LMO, the simulated voltage had a plateau at 2.0 V, consistent with the experimental outcome. Although the simulation with the modified LMO electronic conductivity did not perfectly match the experimental discharge profile and dQ/dV, qualitatively the P2D simulation result supported a more dramatic and sudden drop in LMO electronic conductivity at nearly full lithiation during discharge.
The experimental lowest voltage dQ/dV peak provided further insights by monitoring its progression with continued cycling. With each cycle the capacity above ˜2.2 V decreased, and the capacity associated with the lowest voltage dQ/dV peak region increased and shifted to lower voltages with each cycle. Each successive discharge proceeded down to 1 V (˜2.65 V vs Li/Li+), which could have provided sufficient potential to convert some small amount of LMO to the tetragonal Li2Mn2O4 phase (which would be expected at ˜1.2 V; e.g. ˜2.8 V vs Li/Li+). Formation of this phase would result in Mn valence closer to 3+, where the Jahn-Teller effect becomes more pronounced. Thus, the degradation of LMO would be consistent with even lower electronic conductivity from the LMO layer in the electrode, which would result in more and more capacity in the lowest dQ/dV peak region.
Although obtaining the most accurate electronic conductivity as a function of lithiation for LMO was a challenge, the lithium intercalation position during simulations was expected to be semi-quantitatively or at least qualitatively reflective of the experimental system, because a major determining factor of the electrochemical reaction was the OCV difference of the electroactive materials.
During the process with LMO discharge (
For the CC:LCO:LMO two-layer electrode, at C/50 discharge rate, the simulated and experimental profiles matched well for the first 115 mAh g−1 cathode (
dQ/dV analysis was also performed for the CC:LMO:LCO two-layer electrode and can be found in
Above 2.6 V, the experimental dQ/dV curve resembled the OCV dQ/dV curve. The experimental would be expected to have slightly lower voltages because even if the LCO had minimal electronic resistance, the cell voltage should be smaller due to the overpotentials from the ionic resistance of the cathode and overpotentials contributed from anode. It was speculated that this relatively high voltage discharge capacity resulted as a consequence of the LCO in the sintered electrode being charged to 2.8 V (˜4.36 V vs Li/Li+), because the OCV function did not access above 4.3 V vs Li/Li+ and thus did not have capacity at as high of a potential as the sintered electrode cell likely experienced experimentally. This speculation was also consistent with the observation of a small amount of capacity below 1.2 V, where LCO may have underwent mild irreversible structural collapse at the higher potential the sintered cathode was subjected to relative to the composite cathodes.
The first experimental dQ/dV peak was at 2.52 V, which was lower than the OCV calculation (2.57 V) and the discrepancy was attributed to the electronically resistive Li0.5Mn2O4 phase. The first experimental dQ/dV peak position was slightly higher than that of experimental dQ/dV of the CC:LMO:LCO case (2.50 V), which possibly originated from the reduced ionic overpotential from a reduced ionic path with LMO being next to separator. The corresponding P2D simulated dQ/dV peak had a much greater peak magnitude but initiated later than the first experimental dQ/dV peak, suggesting an underestimated electronic conductivity of λ-MnO2 phase and overestimated electronic conductivity from Li0.5Mn2O4 phase, which was consistent with the trend observed from the CC:LMO:LCO simulation.
The second experimental dQ/dV peak had an onset at ˜2.45 V, but a peak position was difficult to define and was possibly overlapped with capacity contributed from LCO at lower voltage, and the capacity was delivered at a much lower voltage than the OCV calculation (2.44 V). This observation was consistent with the CC:LMO:LCO case and a previous sintered LMO report where the Li1Mn2O4 phase was suggested to be even more electronically resistive than the values in the electronic conductivity function. The P2D simulated dQ/dV had much greater capacity and more pronounced peak in this voltage region compared to experiment, suggesting an overestimated Li1Mn2O4 phase electronic conductivity in simulations consistent with the CC:LMO:LCO observations.
The experimental LCO peak was assigned at 2.29 V, which had a much greater magnitude, sharper profile, and closer agreement with the corresponding voltage from the OCV calculation (2.34 V) than for the CC:LMO:LCO case. This result suggested the electronic conductivity overpotential contributing to the location of this peak was much smaller, as expected due to the absence of an LMO layer in between the LCO and current collector. The simulated dQ/dV peak position matched the experimental well, suggesting that the LCO electronic conductivity used in the simulation was a more accurate reflection of the LCO material within the cell.
Examining the progression of the first five discharge experimental dQ/dV curves, the peak at below 2.2 V was not observed for the case where LCO was the layer in contact with the current collector. The LMO did still undergo some mild capacity fade, which may have been due to factors such as Mn dissolution or Jahn-Teller distortion. For LCO, from the first cycle until the fifth one, the peak shifted from 2.30 V to 2.29 V, the intensity decreased from ˜480 mAh g−1 V−1 to ˜380 mAh g−1 V−1, and the peak became broader. Such capacity fading was consistent with previous reports for structural collapse of the LCO due to overcharge, which could have resulted due to the relatively high potential experienced by the cathodes in the sintered electrode cell.
Analysis of lithium intercalation as a function of cell depth was also extracted from simulations of the CC:LCO:LMO case and can be found in
While homogeneously blended composite electrodes with multiple different electroactive material compositions/phases have been reported with beneficial properties, sintered electrodes can be more challenging to implement with multiple compositions and phases. As demonstrated in this study, differences in phase and/or concentration can lead to formation of new interfacial regions. In some cases, such a region could be beneficial if, for example, a third phase with higher electronic conductivity was formed and the electrochemical capacity and OCV of the constituent desired phases was not dramatically impacted. However, in this study and in other cases interfacial regions tend to result in increased resistances and/or loss of electrochemical capacity.
Composite electrodes with layers of different materials/particle sizes/phases have been reported to result in improved rate capability. For sintered electrodes with multiple active materials in a multiple layer architecture, there are potential benefits originating from the spatial arrangement of materials with different potentials associated with their electrochemical reactions and electronic conductivities. From an electronic conductivity perspective, a layer with relatively high electronic conductivity next to the current collector does not limit the rate capability (e.g., CC:LCO:LMO). However, for the scenario where a material layer with low electronic conductivity is next to the current collector, the conductivity needs to approach the same order of magnitude as the ionic conductivity of the electrolyte used or else the electronic conductivity of this material will limit the rate capability of the cell. With this higher electronic conductivity, the rate capability at rates of C/20 and C/50 was no longer limited by the electronic conductivity of the LMO material layer and nearly all the electrode capacity was accessed, consistent with previous analysis. This suggests that there may be processing methods such as the addition of electronically conductive coatings to materials such as LMO to aid in achieving higher utilization of the electrode materials.
When using single materials with relatively high electronic conductivity in sintered electrodes such as LCO or LTO, lithiation/delithiation tended to initiate from the separator and propagated towards the current collector side, in some cases via a front with a pronounced gradient. Such electrode depth/position dependence of the electrochemical reactions may not be desirable. For example, the reaction selectively occurring near the separator might accelerate additional cathode-electrolyte interphase (CEI) formation in this region. The CEI could then potentially impede the pores and restrict the transport of lithium further towards the current collector where there would be additional capacity remaining in the electrode. Also, regardless of CEI complications at high discharge rates the electrolyte concentration will become depleted near the current collector side of the cathode due to ion transport limitations. As a result, electrode regions with remaining capacity nearer the current collector can have situations where the electrolyte concentration becomes too low for Li to continue to intercalate, limiting the voltage and capacity of the cell. Thus, electrode designs with multiple layers may enable strategies to have more capacity from the current collector regions earlier in the discharge process to avoid the later ion transport restrictions that can more severely limit capacity.
This study investigated multicomponent sintered electrodes, where the two materials combined in the cathode architecture were LCO and LMO. Homogeneous blending of the two material powders resulted in new interfacial compositions forming during the thermal processing of the electrode and severe reductions in electrochemical capacity. This highlights the importance of considering the formation of new phases during multicomponent sintered electrode processing. When the LCO and LMO were processed as two separate layers, the interfacial component was a relatively small contributor to overall electrode electrochemical properties. Instead, the electrochemical properties resulting from the location of the electrodes on either the current collector or separator side were investigated.
Depending on which layer (LCO or LMO) was near the current collector dramatically changed electrochemical properties of the cells, which was attributed to the very different electronic conductivities of the two materials. Simulations provided further insights into the progression of the electrochemical reactions in the cell. For thick sintered electrodes, careful consideration must be given to the thermodynamics (e.g., OCV at different locations at any given point in the discharge progression), and electronic and ionic transport pathways which dramatically influence the spatial location of the lithiation and delithiation reactions. These results provide insights into the design considerations for multicomponent sintered electrode batteries.
P2D simulation was used to generate some of the data herein, with the systems of equations used below.
Lithium Flux across Electrode & Electrolyte Interface:
List of symbols are shown below.
TNO electroactive material powder was synthesized using a sol-gel method. Briefly, 0.5 M of niobium chloride (NbCl5, Sigma-Aldrich) and 0.25 M of titanium isopropoxide (Ti(OC3H7)4, Sigma-Aldrich) were dissolved in 40 mL ethanol at 40° C. and stirred at 300 RPM to facilitate dissolution and mixing. The solution was dried at room temperature overnight in air, then redissolved in 40 mL deionized (DI) water and dried at 80° C. overnight in air within a fume hood. The resulting precursor was then heated to 700° C. and held at that temperature for 2 hours, where the ramp rates for heating and cooling were both 1° C. min−1.
LCO electroactive material powder was synthesized via precipitation of a precursor followed by solid state reaction.23 0.2 M of cobalt sulfate heptahydrate (CoSO4·7H2O, Acros Organics) and 0.2 M of sodium oxalate (Na2C2O4, Fisher) were separately dissolved into 400 mL of DI water each and heated to 60° C. The solutions were mixed together via pouring all at once, and the mixture was stirred at 300 RPM. The reaction was allowed to proceed for 30 min before collection of the precipitate via vacuum filtration. The filter cake was rinsed with 1.6 L of DI water. The resulting powder cake was then dried at 80° C. overnight in air, and blended with lithium carbonate (Li2CO3, Fisher Chemical) with a molar ratio of 1.05:1 Li:Co using a mortar and a pestle by hand for 10 min. The blended powder was then heated at a ramp rate of 1° C. min−1 to 800° C. without a hold at the top temperature, and then allowed to cool to room temperature without control over the cooling rate.
Thermogravimetric analysis (TGA) was conducted using a TA Instruments Q50. Heating initiated from room temperature to 800° C. at a ramp rate of 10° C. min−1 in air atmosphere. Scanning electron micrographs (SEMs) were collected using a FEI Quantum 650. Powder X-ray diffraction (XRD) was conducted using a PANalytical X′pert ProMPD.
Electronic conductivity of materials was measured using a direct current (DC) technique. The as-prepared sintered pellet electrodes were coated with silver paste (Sigma-Aldrich) by hand on each flat side of the porous disc. The coated pellets were sandwiched between stainless steel and compressed by a clamp. A constant voltage at 10 mV was applied and current was recorded using a Gamry Reference 600 to measure the DC resistance necessary to calculate the pellet material electronic conductivity.
During fabrication of sintered TNO electrodes, 4 concentrations of sucrose binder were prepared in this work. The sucrose contents were 0.0%, 2.5%, 5.0%, and 7.5% by mass (denoted by 0.0%-sucrose, 2.5%-sucrose, 5.0%-sucrose, and 7.5%-sucrose). The percentage indicates the mass percent of sucrose relative to the total combined mass of sucrose and TNO. For the 0.0%-sucrose sample, 1 g of TNO powder was blended with 2 mL of 1 wt % polyvinyl butyral (PVB) solution (Pfaltz & Bauer) in ethanol using a mortar and pestle by hand until dried. For the other three samples which contained sucrose, a total mass of 1 g consisting of TNO and sucrose powder was blended with 0.5 mL of DI water and 1.5 mL of ethanol using a mortar and a pestle by hand until dried. For all cases, the coated powder was next loaded into a circular pellet die (Carver) with a diameter of 1.3 cm, followed by pressing at 420 MPa for 2 min using a hydraulic press (Carver). The pressed pellet was sintered in argon atmosphere at 800° C. for 1 hour at both ramp up and down rates of 2° C. min−1. The as-prepared pellets had thicknesses ranging between 580-610 μm (measured using digital calipers), areal loadings of 142-148 mg cm−2, and geometric pore/void fraction of 0.43-0.45 (calculated assuming the crystal density to be 4.3 g cm−3). Note that carbon coated (and carbon free) TNO powders were also used in composite electrodes, and for TNO powders used for composite electrodes all processing steps were the same as for sintered electrodes except that there was not a hydraulic compression of the powder.
For sintered LCO electrodes, 1 g of LCO powder was blended with 2 mL of 1 wt % PVB solution in ethanol using a mortar and a pestle by hand until dried. The pressing procedure was the same as above for TNO materials. The pressed pellet was sintered in air at 600° C. for 1 hour at both ramp up and down rates of 1° C. min−1. The as-prepared pellets had measured thicknesses ranging between 810-850 μm, areal loadings of 268-273 mg cm−2, and geometric pore/void fractions of 0.32-0.34 (calculated assuming the crystal density to be 5.0 g cm−3).
For composite electrode cells, the TNO powder was mixed with acetylene carbon black (CB, Alfa Aesar) and 3.33 wt % polyvinyl pyrrolidone (PVP, Sigma Aldrich, 360 kDa molecular weight) in ethanol solution with a mass ratio TNO:PVP:CB of 8:1:1. The resulting slurry was coated onto an aluminum foil current collector using a doctor blade, which resulted in final areal active material loadings measured for punched electrodes to range between 2.0-3.5 mg cm−2 after drying. Circular electrodes with an area of 1.33 cm2 were punched by hand using a die and transferred into a glove box filled with argon. Punched lithium metal foil was used as anodes and Celgard 2325 was the separator, and the electrodes were assembled into electrochemical cells using 2032-type coin cell parts, with assembly all in the glove box. The electrolyte was 1.2 M LiPF6 in 3:7 ethylene carbonate:ethyl methyl carbonate (Gotion). The as-prepared cells were evaluated using a multichannel battery cycler (MACCOR) between 1.0-2.5 V at C/20, where 387.6 mAh g−1 was assumed to be the TNO theoretical capacity used for adjusting the currents for the C rate determination. It is noted that for Li/TNO cells, the cycling starts on a discharge.
For sintered electrode cells, the as-prepared sintered cathode (LCO pellet) and sintered anode (TNO pellet) were attached to the bottom plate and the spacer of the 2032-type coin cell, respectively, with a custom carbon paste applied between the sintered electrode and stainless steel to reduce contact resistance. The custom carbon paste was applied as a homogenous slurry consisting of 4.76 wt % of CB, 4.76 wt % of PVP, and 90.48 wt % of ethanol. Pellets adhered to the cell components with the paste were then dried in vacuum at 80° C. for 20 min before transferring into the glove box. 2032-type coin cells were assembled with a single spacer which was attached to the anode, as the cathode was directly attached to the bottom piece. The same electrolyte was used for the composite electrodes described above. Glass fiber (Fisher, type G6 circles) was used as separator. The sintered cells were cycled between 1.0-3.2 V, and as a reference point for C rates 1.69 mA cm−2 corresponded to a rate of C/20.
The initial sol-gel synthesized powder with the targeted TNO composition had broad peaks in collected XRD patterns, suggesting relatively low crystallinity. The broad peaks may have in part been due to the relatively low calcination temperature, and were consistent with previous reports. Explicit assignment of each individual peak was challenging to affirm unambiguously; however, there were TixNbyOz phases that could be associated with all peaks. These micrographs suggested the morphology of the material was secondary aggregates that had length dimensions of approximately 20 μm, while the primary particles that formed the larger aggregates were less than 1 μm in length (with many primary particles in the size range of ˜100 nm).
The DC electronic conductivity measured using TNO porous pellets can be seen in
XRD patterns for the TNO pellets after hydraulic compression and heating in argon atmosphere can be found in
All samples had a similar appearance. Many of the primary particle sizes were still on the order of ˜100 nm as was the case for the powder before hydraulic compression and thermal treatment. The retention of the primary particle size was also consistent with a mild sintering process. These primary particles were part of larger aggregates that were 10-20 μm in size. Many of the aggregates had an appearance of being flattened, which likely resulted from the hydraulic compression processing step.
Initial electrochemical analysis on the TNO materials was conducted using composite electrodes paired with Li metal anodes. As the sucrose content that was added to the TNO before heat treatment was increased, the gravimetric capacity of the resulting TNO electroactive material decreased. The decrease in gravimetric capacity was likely due to increasing inactive carbon content in the powder and also due to increasing relative amounts of TiNbO4 phase. The voltage where TiNbO4 has electrochemical capacity is between 1.0-2.0V and thus overlaps with TNO; however, TiNbO4 has a reported capacity of 200 mAh g−1 and thus increase in its content would lower the overall material gravimetric capacity relative to TNO.
Next, electrochemical characterization was done of cells containing the TNO sintered electrodes directly. The TNO materials were used as sintered anodes paired with LCO sintered cathodes in a full cell configuration. Sintered electrodes were not evaluated in half cells due to the limitations of using lithium metal in such high areal capacity systems, where each cycle would have significant amount of plating and stripping that make the Li electrode the limiting factor. As for sintered cells paired with LCO, when looking at the discharge voltage profiles (
While the rate capability analysis above reflected the retention of electrochemical capacity, another outcome that changed for the different TNO electrode cells was the average discharge voltage. The average discharge voltage as a function of the discharge C rate for the TNO sintered anode cells can be found in
According to the Butler-Volmer equation, for most Li-ion battery electroactive materials the current has an exponential dependence on the overpotential if only considering the interfacial kinetics. Thus, in the ideal case galvanic charge and discharge of thin composite electrode batteries with low loadings would often primarily reflect the interfacial kinetics at different C rates. dQ/dV plots show the voltage regions corresponding to the delivery of electrochemical capacity. An additional use of dQ/dV is the ΔdQ/dV calculated for shifts in the peak positions between the charge and discharge cycle at the same current density (shift in voltage location of peaks in differential capacity), where the shift is the ΔdQ/dV values as a function of current densities. For batteries with very thick electrodes, such as the sintered electrode cells evaluated herein, the overpotential contribution from interfacial compared to ionic and/or electronic overpotential is much smaller according to previous analysis. For such thick electrodes, the electronic and ionic conductivity of the solid electrode and liquid electrolyte phase can be reflected in the ΔdQ/dV peak position-where the baseline is the dQ/dV peak at low cycling rates and the ΔdQ/dV is the shift in the peak locations at increasing C rates. For increasing C rates in a cell where the overpotential is primarily due to electronic and/or ionic conductivity through the matrix, it would be expected that the ΔdQ/dV peak positions would shift linearly as a function of increasing current density (or C-rate for identically processed electrodes/cells). Similar analysis was previously reported for indirect evaluation of the impact of the electronic conductivities of sintered LiMn2O4-type (LMO) materials with different dopants and lithiation states. The prior analysis of LMO materials demonstrated that dQ/dV not only serves well to analyze the redox potentials in thin composite electrode with negligible electronic and ionic overpotentials, but also to analyze the electronic and ionic conductivities in thick sintered electrodes with relatively low contributions from interfacial overpotentials to the overall cell overpotential. When comparing electroactive materials with different compositions/dopants/treatments, the use of ΔdQ/dV peak positions can exclude the impact from difference in open circuit voltage (OCV). In addition, with thicker electrodes where lithium depletion in general will occur, the peak dQ/dV position well represents the voltage where the electrochemical reactions occur most favorably well before lithium depletion in the cathode region. The Lit depletion likely resulted in the more dramatic decrease in voltage that resulted in lower capacity for the TNO sintered electrodes with the highest sucrose content during processing.
As expected for cells with ionic or electronic electrode conductivity being the primary contributors to overpotential, all cells had a linear dependence between the ΔdQ/dV peak positions and current density. As the sucrose content increased, the slope decreased, with these results summarized in
where σ is the overall conductivity, A is the electrode area, L is the total battery thickness including anode, cathode, and separator, and R is the resistance. Furthermore, the approximate slope of the ΔdQ/dV peak position with a unit of V cm2 mA−1 is approximated as
where k is the slope. Combining both equations, the approximate overall conductivity can be calculated by
The overall conductivities of all TNO sintered anode cells are shown in
The electronic conductivity improvement from carbon coating was beneficial for overall desirable electrochemical energy storage outcomes for TNO materials, but only up until a certain threshold of carbon added. More generally, many coating materials including carbon do not participate in the electrochemical battery reaction and sometimes exacerbate the interfacial kinetics if the coating layer is too thick. In a thick electrode architecture, another disbenefit of excess carbon material is that the excess material can potentially take up volume within the electroactive particle interstitial space of the electrode. Additional material within the pore/void volume which becomes laden with electrolyte will decrease effective ionic transport properties through the electrode. As mentioned earlier, so long as the effective electrode electronic conductivity exceeds approximately 1 S m−1, or roughly the ionic conductivity of the electrolyte (LiPF6 in carbonates), the matrix electronic conductivity of the sintered electrode no longer limits the capacity of the cell. In the case of higher conductivity electrolytes such as those containing lithium bis(fluorosulfonyl)imide (LIFSI), the threshold for electronic conductivity would be increased to match a value similar in conductivity of the relevant electrolyte.
TiNb2O7 was successfully synthesized using a sol-gel method. Sucrose was incorporated into the processing of the TNO material into porous sintered electrodes, where the sucrose played a dual role of binder during powder processing and carbon source for generating carbon during thermal treatment. As the relative weight fraction of sucrose was increased, the relative amount of carbon in the electrode also increased and the resulting initial measured electrode matrix electronic conductivity also increased. The carbon addition resulted in improvements in rate capability and average discharge voltage for cells which used sintered anodes containing the TNO materials. This report provides both a robust route for improving electrochemical outcomes when incorporating relatively low electronic conductivity electroactive material in sintered electrodes and supports general guidelines for targeting a carbon amount that achieves comparable conductivity to the electrolyte to avoid unnecessary ion transport restrictions from excessive carbon.
The synthesis of LiCoO2 (LCO) electroactive material powder was based on a previously reported method of oxalate precursor precipitation followed by solid state reaction. 200 mM of cobalt sulfate heptahydrate (CoSO4·7H2O, Acros Organics) and 200 mM of sodium oxalate (Na2C2O4, Fisher) were separately dissolved into 2 beakers of 400 mL of deionized (DI) water each and preheated to 60° C. The sulfate solution was poured all at once into the oxalate solution, and the solution was kept at 60° C. with stirring at 300 RPM for half an hour. The resulting precipitate was collected via vacuum filtration, followed by 1.6 L of DI water rinsing, before being transferred to an oven and dried at 80° C. overnight in air. The dried powder was then blended with lithium carbonate (Li2CO3, Fisher Chemical) by hand with target Li:Co stoichiometry of 1.05:1 using a mortar and a pestle for 10 min. The mixed powder was placed in a crucible and heated up to 800° C. at a ramp up rate of 1° C. min−1 in air without a hold. The furnace was then allowed to cool to room temperature without control over cooling rate.
TNO electroactive material precursor was synthesized using a sol-gel method. Briefly, 0.5 M of niobium chloride (NbCl5, Fisher) was dissolved in 40 mL of ethanol at 40° C. and stirred at 300 RPM, followed by addition of 0.25 M of titanium isopropoxide (Ti(OC3H7)4, Sigma-Aldrich) to the same solution. When the solution became fully transparent (typically <30 seconds), it was poured into a drying dish and dried at room temperature overnight in air. The resulting gel was redissolved in 40 mL deionized (DI) water using a spatula by hand until fully dissolved, and then dried at 80° C. overnight in air within a fume hood. The dried precursor was then ground using mortar and pestle by hand for 5 mins and heated in a furnace in air with ramp rates for heating and cooling of 1° C. min−1. The highest target temperature and hold time at that target was 1000° C. for 24 h, 800° C. for 2 h, 700° C. for 2 h, and 600° C. for 2 h (referred as 1000 C, 800 C, 700 C, and 600 C, respectively, when referenced herein).
All cathodes in composite cells were composite TNO electrodes. The 1000° C., 800° C., 700° C., and 600° C. powder was blended with acetylene carbon black (CB, Alfa Aesar) as conductive additive and 3.33 wt % polyvinyl pyrrolidone (PVP, Sigma Aldrich, 360 kDa molecular weight) as polymer binder with a TNO:PVP:CB weight ratio of 8:1:1. The slurry mixture was then casted onto an aluminum foil current collector using a doctor blade with a gap of 200 μm, and areal electroactive material loadings were 0.9-2.0 mg cm−2 after drying. Circular electrodes with an area of 1.33 cm2 were punched by hand using a die and transferred into a glove box filled with argon.
All anodes in composite cells were Li foils with thicknesses of 100 μm and punched into circular shape with an area of 1.6 cm2.
All cathodes in sintered cells were sintered LCO electrodes. 1 g of LCO powder was blended with 2 mL of 1 wt % PVB solution in ethanol using a mortar and a pestle by hand until dried. The coated LCO powder was transferred into a circular pellet die (Carver) with an area of 1.33 cm2, followed by pressing at 420 MPa for 2 min using a hydraulic press (Carver). The mass of PVB-coated LCO was targeted to be 350 mg for each electrode pellet. The pressed LCO pellet was then sintered in air at 600° C. for 1 hour with ramp rates of 1° C. min−1 for both heating and cooling. The as-prepared pellets had measured thicknesses ranging between 770-800 μm, areal loadings of 253-258 mg cm−2, and geometric pore/void fractions of 0.32-0.34 (calculated assuming the crystal density to be 5.0 g cm−3). Detailed characterization of sintered LCO electrodes can be found elsewhere for using the same procedures.
All anodes in sintered cells were sintered TNO electrodes. The coating and pressing were the same procedure as used for LCO. During the thermal treatment of the pressed TNO pellets, the top target temperature was 1000° C., 800° C., 700° C., and 600° C. for the 1000° C., 800° C., 700° C., and 600° C. source powders, respectively. In all cases the hold at the target temperature was 2 h and the heating and cooling ramp rates were 2° C. min−1. Resulting TNO pellets had measured thicknesses of 440-450 μm and areal loadings of 109-110 mg cm−2.
Powder X-ray diffraction (XRD) was performed using a PANalytical X′pert ProMPD. Scanning electron micrographs (SEM) were conducted using a FEI quantum 650.
All electrochemical cell assembly was performed in a glove box with water and oxygen levels <1 ppm. All cells used 2032-type coin cell parts. For composite cells, Celgard 2325 with an area of 1.98 cm2 was used as separator, and electrolyte was 1.2 M LiPF6 in 3:7 ethylene carbonate:ethyl methyl carbonate (Gotion) was used as electrolyte. The assembled cells were evaluated using a multichannel battery cycler (MACCOR) between 1.0-2.5 V (vs. Li/Li+) at C/10, where 200 mAh g−1 TNO was assumed for calculating C rates and the current was adjusted based on measured TNO mass loading in the electrodes.
For sintered cells, the as prepared LCO and TNO pellet electrodes were adhered to the bottom plate and the spacer of the 2032-type coin cell, respectively, using a custom carbon paste to reduce contact resistance. The carbon paste consisted of 4.76 wt % of CB, 4.76 wt % of PVP, and 90.48 wt % of ethanol. After attaching the pellets, they were dried in vacuum at 80° C. for 20 min to remove ethanol before transferring into the glove box. Electrolyte was the same used in composite cells, but glass fiber (Fisher, type G6 circles) with an area of 1.98 cm2 was used as separator for sintered electrode cells. The sintered cells were cycled between 1.0-3.1 V (cell voltage), and 0.85 mA cm−2 was assumed to be C/50 for all sintered cells. The mass ration of LCO:TNO was ˜2.5:1, resulting in what was expected to be anode limited cells for all materials.
The XRD patterns for all four synthesized TNO materials are shown in
The 1000° C. sample had large internal porosity, and the primary particles were distinct and exhibited particle size of ˜700 nm. The 800 C sample had slightly less distinguishable internal voids in the secondary aggregates, and the primary particles were much smaller, in the ˜100 nm range. The 700° C. and 600° C. samples appeared similar, and the surface of the secondary particle aggregates was smoother with less defined pore regions compared to the 800° C. and 1000° C. samples. The primary particles did not have pronounced morphology, which possibly originated from the low crystallinity in these materials and would be consistent with the XRD results. The pellet surfaces for all samples were flat due to the hydraulic compression step, and the primary particle morphologies were not changed noticeably relative to the source powders.
First cycle of discharge and charge voltage profiles for TNO composite electrodes paired with Li foil anodes at rates of C/10, and the corresponding dQ/dV profiles from those cycles, can be found in
For the 800° C. TNO material, the first charge achieved 259 mAh−1 TNO with an initial CE of 111%. For the first discharge there were two small voltage plateaus at ˜1.7 V, and correspondingly two dQ/dV peaks were observed. The smaller dQ/dV peak appeared at ˜1.75 V and was not present in the 1000° C. sample. The following charge cycle had a relatively broad and shallow peak at ˜1.9 V. Coupling the dQ/dV peak position and the low reversibility, it is speculated that such capacity originated from the Ti4+/Ti3+ redox in the TiO2 phase, which would have been consistent with previous reports. The last charge cycle capacity retention (83.6%) was much improved compared to the 1000° C. sample.
For the 700° C. sample, the first charge achieved 233 mAh−1 TNO with an initial CE of 114%. The capacity was more evenly distributed throughout the voltage window. In the dQ/dV plot, as was the case for the 800° C. material, there were two peaks (at ˜1.75 V and ˜1.6 V). Compared to those of the 800° C. sample, the first discharge peak at ˜1.75 V initiated at a higher voltage and the peak was broader. Also in comparison to the 800 C material the second peak from discharge at ˜1.6 V had smaller intensity but was much wider, and the dQ/dV plot had greater absolute values in the regions outside the peaks as well. These observations were consistent with previous electrochemical reports for T-Nb2O5. The last charge cycle capacity retention (84.4%) was improved compared to the 800° C. sample.
For the 600° C. sample, the first charge achieved 190 mAh−1 TNO with an initial CE of 122%. This outcome may have been due to the low crystallinity of the material and/or the inherent smaller capacity of the contributing phase of TT-Nb2O5. In addition, the discharge profile had a significant slope with only a very shallow plateau observed. The last charge cycle capacity retention (94.8%) was the best.
In brief, as the synthesis temperature decreased, the last charge cycle capacity retention increased. However, the average charge voltage increased, which would limit the voltage if used as anode in a full cell.
For the first charge and discharge voltage profiles and dQ/dV at C/50 (
For the 800° C. TNO sintered anode, the first discharge delivered 231 mAh g−1 TNO capacity with an average discharge voltage of 2.25 V, cell. The first cycle voltage efficiency and CE were 90.8%, and 92.1%, both slightly lower than those of the 1000° C. sample. In the dQ/dV plot, the first charge had two peaks, but the first discharge had only one peak, consistent with the composite dQ/dV outcomes. The stability of this sample was much improved with a capacity retention of 91% and energy retention of 85% at 25th cycle, relative to the 1000° C. sample that faded quickly.
For the 700° C. sample, the first discharge delivered 233 mAh g−1 TNO capacity and an average voltage of 2.23 V. The first cycle voltage efficiency and CE were 90.2%, and 88.6%, both slightly lower than those of the 800 C sample. The number and location of peaks in the dQ/dV profile were also similar to those observed in the 800 C composite cell (when accounting for the offsets with regards to the different electrodes involved). The charge/discharge cycling capacity retention/stability was the best among the sintered TNO anodes evaluated, with a capacity retention of 95.3% and energy retention of 91% at 25th cycle. As mentioned earlier, the volume change during cycling was speculated to be a critical factor to sintered electrode cycling stability. The primary phase assigned to the 700 C TNO material was T-Nb2O5, with volume change reported after lithiation to be around 2-3%, much smaller than the volume change reported for the higher temperature TNO phase with high crystallinity phases. The reduced volume change for the T-Nb2O5 phase in the 700° C. TNO electrode was suspected as the cause of its increased cycling stability.
For the 600° C. sample, at the end of first charge another voltage plateau appeared (also shown in the dQ/dV plot). This outcome was attributed to the inherently lower capacity of this material compared to the other TNO materials, and may have resulted from overlithiating the anatase or TT-Nb2O5. Such overlithiation has been reported to destabilize the electrode architecture and create unfavorable solid electrolyte interphase (SEI). The initial intent was not to utilize the capacity at voltages where SEI formation would be significant, but in order to characterize and compare more thoroughly with the other three materials, the same relative loadings and voltage windows were used to maintain comparisons. The 600° C. TNO first discharge capacity reached 234 mAh g−1 TNO and average voltage was 2.12 V, the lowest among all sintered TNO anode materials evaluated. It also had the lowest energy efficiency and CE of 85.4% and 88.2%, respectively. The cell with this material anode had the lowest capacity retention of 78.8% and energy retention of 77% at the 25th cycle relative to the other TNO materials.
With respect to the rate capabilities, discharge voltage profiles for all samples except for the 1000° C. material can be found in
The terms and expressions that have been employed are used as terms of description and not of limitation, and there is no intention in the use of such terms and expressions of excluding any equivalents of the features shown and described or portions thereof, but it is recognized that various modifications are possible within the scope of the aspects of the present invention. Thus, it should be understood that although the present invention has been specifically disclosed by specific aspects and optional features, modification and variation of the concepts herein disclosed may be resorted to by those of ordinary skill in the art, and that such modifications and variations are considered to be within the scope of aspects of the present invention.
The following exemplary aspects are provided, the numbering of which is not
to be construed as designating levels of importance:
Aspect 1 provides an electrochemical storage device comprising:
Aspect 2 provides the electrochemical storage device of Aspect 1, wherein the electrode is a sintered electrode.
Aspect 3 provides the electrochemical storage device of any one of Aspects 1 or 2, wherein the electrode is a composite electrode.
Aspect 4 provides the electrochemical storage device of Aspect 3, wherein the composite electrode comprises a metallic substrate to which the TiNb2O7, LiCoO2, LiNi0.5Mn0.5O2, lithium metal, LiMn2O4, Li4Ti5O12, a mixture of LiMn2O4 and LiCoO2, multiple phase thereof, or mixture thereof is applied.
Aspect 5 provides the electrochemical storage device of Aspect 4, wherein the metallic substrate comprises a stainless steel, aluminum, alloys thereof, or mixtures thereof.
Aspect 6 provides the electrochemical storage device of any one of Aspects 1-5, wherein a thickness of the electrode is in a range of from about 50 μm to about 2000 μm.
Aspect 7 provides the electrochemical storage device of any one of Aspects 1-6, wherein a thickness of the electrode is in a range of from about 70 μm to about 400 μm.
Aspect 8 provides the electrochemical storage device of any one of Aspects 1-7, wherein the electrode is substantially free of a conductive additive.
Aspect 9 provides the electrochemical storage device of Aspect 8, wherein the conductive additive comprises a polymer binder, a conductive carbon, or a mixture thereof.
Aspect 10 provides the electrochemical storage device of any one of Aspects 1-9, wherein a porosity of the electrode is in a range of from about 30% to about 50%.
Aspect 11 provides the electrochemical storage device of any one of Aspects 1-10, wherein a porosity of the electrode is in a range of from about 35% to about 40%.
Aspect 12 provides the electrochemical storage device of any one of Aspects 1-11, wherein a capacity of the electrode is in a range of from about 7 mAh/cm2 to about 100 mAh/cm2.
Aspect 13 provides the electrochemical storage device of any one of Aspects 1-12, wherein a capacity of the electrode is in a range of from about 10 mAh/cm2 to about 50 mAh/cm2.
Aspect 14 provides the electrochemical storage device of any one of Aspects 1-13, wherein the electrode is an anode.
Aspect 15 provides the electrochemical storage device of any one of Aspects 1-13, wherein the electrode is a cathode.
Aspect 16 provides the electrochemical storage device of Aspect 15, wherein the cathode comprises the LiCoO2.
Aspect 17 provides the electrochemical storage device of any one of Aspects 1-16, wherein the electrode comprises one dopant.
Aspect 18 provides the electrochemical storage device of Aspect 17, wherein the doped electrode comprises at least two dopants.
Aspect 19 provides the electrochemical storage device of any one of Aspects 17 or 18, wherein the at least one dopant has a 1+, 2+, or 3+ oxidation state.
Aspect 20 provides the electrochemical storage device of any one of Aspects 1-19, wherein the at least one dopant comprises copper, aluminum, sulfur, potassium, or a mixture thereof.
Aspect 21 provides the electrochemical storage device of any one of Aspects 1-20, wherein the electrode is a battery anode.
Aspect 22 provides the electrochemical storage device of any one of Aspects 1-21, wherein the electrode is a battery cathode.
Aspect 23 provides the electrochemical storage device of any one of Aspects 1-22, wherein the electrode is free of a binder, an inactive solid additive, or a mixture thereof.
Aspect 24 provides the electrochemical storage device of any one of Aspects 1-23, further comprising a carbon coating at least partially disposed over the electrode.
Aspect 25 provides the electrochemical storage device of Aspect 24, wherein the carbon coating is disposed over about 30% to about 100% of a total surface area of the electrode.
Aspect 26 provides the electrochemical storage device of Aspect 24, wherein the carbon coating is disposed over about 60% to about 95% of a total surface area of the electrode.
Aspect 27 provides a lithium-ion battery comprising:
Aspect 28 provides the lithium-ion battery of Aspect 27, wherein the electrolyte comprises a salt.
Aspect 29 provides the lithium-ion battery of any one of Aspects 27 or 28, wherein the separator comprises a porous polymer or glass fiber impregnated with electrolyte.
Aspect 30 provides the lithium-ion battery of Aspect 29, wherein the polymer comprises a polyurethane, a polypropylene, a polyethylene, a copolymer thereof, or a mixture thereof.
Aspect 31 provides the electrochemical storage device of any one of Aspects 1-30, wherein the electrochemical storage device comprises a lithium-ion battery.
Aspect 32 provides an article comprising the electrochemical storage device of any one of Aspects 1-31.
Aspect 33 provides the article of Aspect 32, wherein the article comprises a vehicle or an electronic device.
Aspect 34 provides a method of making the electrode of any one of Aspects 1-33, the method comprising:
Aspect 35 provides the method of Aspect 34, wherein the sintering is performed at a temperature in range of from about 500° C. to about 1100° C.
Aspect 36 provides the method of any one of Aspects 34 or 35, wherein the sintering is performed at a temperature in range of from about 700° C. to about 900° C.
Aspect 37 provides the method of any one of Aspects 34-36, wherein sintering is performed at a variable temperature.
Aspect 38 provides the method of any one of Aspects 34-37, wherein sintering is performed over a period of time in a range of from about 0.5 hours to about 20 hours.
Aspect 39 provides the method of any one of Aspects 34-38, wherein sintering is performed over a period of time in a range of from about 16 hours to about 18 hours.
Aspect 40 provides the method of any one of Aspects 34-39, wherein at least one component of the electrode precursor is sucrose.
Aspect 41 provides a method of making the composite electrode of any one of Aspects 3-40, the method comprising:
Aspect 42 provides a method of using a battery including the electrode of any one of Aspects 1-41, to generate electricity.
Aspect 43 provides a method of using the electrode of any one of Aspects 1-42, electrode to perform electrochemical reactions.
Aspect 44 provides a method of providing or performing one or more of the following: a) communication-sintered electrode full cells incorporating TiNb2O7 anode materials, b) communication-TiNb2O7 anode for sintered electrode full cells, c) enhancing low electronic conductivity materials in sintered electrodes through multicomponent architecture, d) percolated LiCoO2 for improved electronic conductivity in multicomponent thick sintered lithium-ion battery electrodes, or e) multicomponent two-layered cathode for thick sintered lithium-ion batteries, as described herein.
Aspect 45 provides The method according to Aspect 44, including each and every novel feature or combination of features disclosed herein.
Aspect 46 provides a system configured to provide one or more of the following: a) communication-sintered electrode full cells incorporating TiNb2O7 anode materials, b) communication-TiNb2O7 anode for sintered electrode full cells, c) enhancing low electronic conductivity materials in sintered electrodes through multicomponent architecture, d) percolated LiCoO2 for improved electronic conductivity in multicomponent thick sintered lithium-ion battery electrodes, or e) multicomponent two-layered cathode for thick sintered lithium-ion batteries, as described herein.
Aspect 47 provides The system according to 46, including each and every novel feature or combination of features disclosed herein.
Aspect 48 provides a composition comprising (or aspects therefrom or thereof), one or more of the following: a) communication-sintered electrode full cells incorporating TiNb2O7 anode materials, b) communication-TiNb2O7 anode for sintered electrode full cells, c) enhancing low electronic conductivity materials in sintered electrodes through multicomponent architecture, d) percolated LiCoO2 for improved electronic conductivity in multicomponent thick sintered lithium-ion battery electrodes, or e) multicomponent two-layered cathode for thick sintered lithium-ion batteries, as described herein.
Aspect 49 provides The composition according to Aspect 48, including each and every novel feature or combination of features disclosed herein.
Aspect 50 provides an article of manufacture comprising (or aspect thereof) one or more of the following: a) communication-sintered electrode full cells incorporating TiNb2O7 anode materials, b) communication-TiNb2O7 anode for sintered electrode full cells, c) enhancing low electronic conductivity materials in sintered electrodes through multicomponent architecture, d) percolated LiCoO2 for improved electronic conductivity in multicomponent thick sintered lithium-ion battery electrodes, or e) multicomponent two-layered cathode for thick sintered lithium-ion batteries, as described herein.
Aspect 51 provides The article of manufacture of Aspect 50, including each and every novel feature or combination of features disclosed herein.
Aspect 52 provides the electrochemical storage device of any one of Aspects 1-31, wherein the electrode further comprises lithium metal.
This application claims the benefit of priority to U.S. Provisional Patent Application Ser. No. 63/320,657 entitled “Electrode Cells and Related Systems and Methods Thereof,” filed Mar. 16, 2022; U.S. Provisional Patent Application Ser. No. 63/423,956 entitled “Coating Processing Using Sucrose for Thick Sintered Electrode Titanium Niobium Oxide Lithium-Ion Battery Anodes and Related Methods and Compositions Thereof,” Nov. 9, 2022; U.S. Provisional Patent Application Ser. No. 63/423,995 entitled “Processing Temperature Impact on TiNb2O7 Thick Sintered Lithium-Ion Battery Electrodes and Related Methods and Compositions Thereof,” Nov. 9, 2022, the disclosures of which are incorporated herein in their entirety by reference.
This invention was made with government support under Grant No. 1825216 awarded by the National Science Foundation. The government has certain rights in the invention.
| Filing Document | Filing Date | Country | Kind |
|---|---|---|---|
| PCT/US2023/064509 | 3/16/2023 | WO |
| Number | Date | Country | |
|---|---|---|---|
| 63320657 | Mar 2022 | US | |
| 63423956 | Nov 2022 | US | |
| 63423995 | Nov 2022 | US |
