BENDABLE, CREASABLE, AND PRINTABLE BATTERIES WITH ENHANCED SAFETY AND HIGH TEMPERTURE STABILITY - METHODS OF FABRICATION, AND METHODS OF USING THE SAME
A method of applying a separator ink onto a dried electrode of a lithium ion battery. The method includes preparing a separator ink suspension, applying the separator ink suspension onto the dried electrode, and drying the applied separator ink suspension. The separator ink suspension includes a binder comprising 20 wt % to 50 wt % of a total weight of the separator ink suspension, the binder being selected from the group consisting of PVDF, PVDF-HFP, PTFE, PEO, PMMA, PAN, CNC, SBR, and combinations thereof; a solvent selected from the group consisting of NMP, DMF, acetone, DMAc, DMSO, trimethyl urea, triethyl phosphate, and combinations thereof; a non-solvent selected from the group consisting of glycerol, water, ethanol, methanol, ethylene glycol, diethylene glycol, triethylene glycol, hexane, heptane, and combinations thereof; and a ceramic filler comprising 50 wt % to 80 wt % of the total weight of the composite electrolyte the ceramic filler being selected from the group consisting of Al2O3, SiO2, TiO2, MgO, Li2O, LiAlO2, BaTiO3, LAGP, LATP, LLTO, and combinations thereof.
Latest Government of the United States as represented by the Secretary of the Air Force Patents:
- ENDOVASCULAR VARIABLE AORTIC CONTROL CATHETER
- ENDOVASCULAR VARIABLE AORTIC CONTROL CATHETER
- Systems and methods for determining feature resolution of image data
- METHODS FOR TOPOLOGICAL DATA ANALYSIS AI/ML PIPELINE (TDAML) WITH ALGORITHM FOR MULTIMODAL SENSOR DATA FUSION IN AUTONOMY APPLICATIONS
- Urinary Metabolites as Predictors of Acute Mountain Sickness Severity
This application is a continuation of U.S. application Ser. No. 15/623,044, filed Jun. 14, 2017, which claimed the benefit of and priority to prior filed Provisional Application Ser. No. 62/353,918, filed Jun. 23, 2016. The disclosure of each of these applications is expressly incorporated herein by reference, each in its entirety.
RIGHTS OF THE GOVERNMENTThe invention described herein may be manufactured and used by or for the Government of the United States for all governmental purposes without the payment of any royalty.
FIELD OF THE INVENTIONThe present invention relates generally to batteries and, more particularly, to flexible, creasable, printable, high temperature batteries, components of the same, and methods of preparing components of the same.
BACKGROUND OF THE INVENTIONFlexible energy storage is a widely recognized necessity to powering the next generation of portable and flexible electronic devices, such as sensors, smart skins for human performance monitoring, radio frequency ID tags, and wearable electronics. Conventional batteries based on metal foil current collectors (“CCs”) are too rigid and incompatible to sustain operation with repeated flexing and folding. The metal foil CC contributes to a significant proportion of a conventional battery's total weight (anywhere from 15% to 80%), which reduces overall energy density and performance, and is prone to corrosion.
To overcome these difficulties, alternative designs and substrate materials have been proposed, including carbon nanotubes (“CNTs”), graphene, textiles, and paper. These materials are flexible, light weight, have high mechanical strength, are generally chemically-stable, and some have good electrical conductivity. While promising progress is being made, the development of a flexible, bendable, and creasable device that maintains high performance, even when exposed to harsh environmental and mechanical conditions, is still a significant challenge.
There have been two conventional approaches to creating flexible batteries: (1) fabricating the battery components to be thin or (2) building free-standing, composite electrodes by growing, functionalizing, or embedding active material into a flexible, conductive scaffold. The first approach is limited because merely reducing the thickness of conventional battery components also reduces power capacity and capability. The second approach is advantageous in that the rigid, metal foil substrate may, in some circumstances, be altogether removed. For example,
The electrode material 56, shown in greater detail in
There has been some steady progress in the fabrication of composite electrodes using a variety of methods, such as vacuum filtration, hot pressing, drop casting, doctor blading, electrospinning, or freeze-drying; however, most are manufactured by batch processes that often limit deposition to two-dimensional substrates. Additionally, materials fabricated by such methods are often inherently brittle, presumptively due to a lack of binder and higher active material loading (often greater than 90 wt %), which leads to limited utility in flexible or stretchable applications.
Another difficulty associated with conventional batteries is directed to form: specifically, cylindrical or prismatic. One proposed solution has been to adopt unique spatial arrangements of rigid battery packs to enable some deformation without performance loss. However, the range of motion of these devices is restricted, and the conventional metal foil CCs are susceptible to crack formation and damage.
Further, conventional polyolefin separators require multiple processing steps (i.e., melt extrusion and stretching) for proper functionality. Such processing renders these conventional polyolefin separators unsuitable for batteries prepared exclusively with additive manufacturing.
An alternative approach is to build printable energy storage solutions. Printing capabilities could greatly simplify the manufacturing process for conventional batteries (for example, lithium ion batteries) and may enable direct integration of a power source into its corresponding device during the fabrication process. Direct write manufacturing techniques are an effective method to creating complex, multifunctional structures. And, unlike most printing techniques commonly adopted for batteries (such as stencil printing, screen printing, and spray printing), solution cast-direct write printing does not require masking, material removal, or processes to aerosolize or otherwise reduce particle size of solids in the ink. If composite electrode materials could retain suitable rheological properties, then these printing techniques (filamentary, ink jet, or aerosol jet printing) could be used. Such template-free printing techniques offer an economical, scalable approach to rapid prototyping of battery electrodes and architectures that can be patterned to fit a specific application or even directly printed on a device. Printing offers the utilization of confined or nonplanar substrates as power sources.
Beyond ease of fabrication, there is an increasing demand for secondary energy storage devices operable in high temperature environments (ranging from 60° C. to 300° C.) for applications in grid storage, automotive, aviation, medical and oil industries. For example, power sources are needed for high temperature electronics operated near engines (where temperatures can reach 150° C.,) for medical devices that can withstand a 120° C. autoclave, and for energy storage for desert photovoltaic devices (where battery temperatures can reach temperatures in excess of 60° C.). Lithium-thionyl chloride batteries are typically used for high temperature application today (operational temperature range of −60° C. to 150° C.); however, lithium-thionyl chloride batters are not rechargeable and thionyl chloride is both toxic and reactive with water. Conventional, rechargeable Li-ion batteries have the potential to meet the needs of these applications due to their high energy density, high operating voltage, and long lifetime, but several of the components of these conventional Li-ion batteries are not suited for the aforementioned, thermally-demanding conditions that often lead to accelerated cell failure. Conventional Li-ion batteries have been used in some harsh, thermal environments applications, such as solar grid storage or electric vehicles, but such applications require thermal management systems to prevent the battery from overheating. The thermal management systems not only take up unnecessary space, but are expensive and account for 8% to 12% of the total battery cost.
Despite such efforts to compensate for overheating, components of conventional Li-ion batteries are generally unable to achieve stable, long-term, high temperature (greater than 50° C.) operation. Organic electrolytes are limited to temperatures not exceeding 50° C. due to decomposition of the lithium hexafluoro-phosphate (“LiPF6”) salt. Many of the carbonates used in common liquid electrolyte blends, such as dimethyl carbonate (“DMC”), diethyl carbonate (“DEC”), and ethylmethyl carbonate (“EMC”), have low boiling points (less than 130° C.) and flash points (about room temperature). Therefore flammability and high internal cell pressure, especially at elevated temperature, are a concern with these carbonates. Conventional polyolefin separators typically possess a thermal shutdown temperature (less than 120° C.) due to thermally induced shrinkage or melting to close pores and the cell will not function above this temperature. If a temperature of the polyolefin separator reaches higher temperatures (e.g., greater than 200° C.), or the flammable polyolefin structure catches fire, catastrophic shrinkage ensues, which brings the electrodes into contact and initiates thermal runaway. Large shrinkage of commercial polyolefin separators can be explained by shape recovery behavior resulting from the stretching during manufacturing to generate desired porosity for ion transport.
Therefore, there remains a need for improved methods and designs of flexible, creasable, printable battery components and composite devices, or components or devices having combinations of such improved features. Moreover, there is a further need for such components and composite devices to be operational, safely, at high temperatures.
SUMMARY OF THE INVENTIONThe present invention overcomes the foregoing problems and other shortcomings, drawbacks, and challenges of conventional batteries by offering flexible, creasable, printable, batteries and components batteries, or combinations thereof. Additionally, the batteries and components of batteries as described herein are operable at a wide range of temperatures with enhanced safety properties. While the invention will be described in connection with certain embodiments, it will be understood that the invention is not limited to these embodiments. To the contrary, this invention includes all alternatives, modifications, and equivalents as may be included within the spirit and scope of the present invention.
According to an embodiment of the present invention, a composite electrode includes an active material, a conductive additive, a binder, and a solvent. The composite electrode may be cast or printed.
Other embodiments of the present invention include a battery having a cathode, an anode, a separator, and an electrolyte. The cathode is a composite electrode having a LiFePO4 active material, a conductive additive, a binder, and a solvent. The separator is positioned between the cathode and the anode.
For still other embodiments of the present invention, a method of fabricating a composite electrode includes preparing a suspension having an active material, a conductive additive, a binder, and a solvent. The suspension is cast or printed and dried.
Additional objects, advantages, and novel features of the invention will be set forth in part in the description which follows, and in part will become apparent to those skilled in the art upon examination of the following or may be learned by practice of the invention. The objects and advantages of the invention may be realized and attained by means of the instrumentalities and combinations particularly pointed out in the appended claims.
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments of the present invention and, together with a general description of the invention given above, and the detailed description of the embodiments given below, serve to explain the principles of the present invention.
It should be understood that the appended drawings are not necessarily to scale, presenting a somewhat simplified representation of various features illustrative of the basic principles of the invention. The specific design features of the sequence of operations as disclosed herein, including, for example, specific dimensions, orientations, locations, and shapes of various illustrated components, will be determined in part by the particular intended application and use environment. Certain features of the illustrated embodiments have been enlarged or distorted relative to others to facilitate visualization and clear understanding In particular, thin features may be thickened, for example, for clarity or illustration.
DETAILED DESCRIPTION OF THE INVENTIONWith reference now to the figures, materials, composites, methods for processing, and methods for fabricating flexible battery components and flexible batteries according to the various embodiments of the present invention are described. With particular reference to
A suitable electrode material slurry 78 may then be prepared for application to the porous substrate 74 (Block 80). According to the illustrated embodiment of the present invention, the slurry 78 comprises an active material 81 (here, the active material 81 is a lithium-based nanoparticle; however, other active materials 81 for other applications would be understood by those of ordinary skill in the art having the benefit of the disclosure herein), a conductive additive 83, a binder 85 (for example, a polymer), and an electrolyte (not shown). For purposes of illustration herein, lithium ion cathodes may be prepared from a slurry 78 that includes lithium iron phosphate (LiFePO4) active material with a conductive additive and a binder in an organic solvent; anodes may be prepared via a slurry 78 that includes lithium titanate (Li4Ti5O12) active material with a conductive additive and a binder in an organic solvent. Other appropriate active materials may include, for example, lithium cobalt oxide (LiCoO2), lithium manganese oxide (LiMn2O4), (LiFePO4), lithium nickel manganese oxide (LiNiMnCoO2), graphite, silicon, and so forth.
Conductive additives 83 of the slurry 78 facilitate electronic transport and should, therefore, be conductive. Exemplary conductive additives 83 may include, for example, any carbon-based materials (for example, carbon black or graphite) or metallic fillers (for example, Al, Cu, Ag, or Ni) having high conductivity and configured to provide a percolative network. For the particular embodiments of the present invention here described, suitable conductive additives may include graphite, carbon black, CNTs, CNFs, and so forth
The binder 83 of the slurry 78 may be any chemical system configured to provide adhesion, mechanical support, and flexibility. Appropriate examples of the binder 83 may include PVDF, polyvinylidene fluoride-co-hexafluoropropylene (“PVDF-HFP”), polytetrafluoroethylene (“PTFE”), styrene-butadiene rubber (“SBR”), carboxymethyl cellulose (“CMC”), and so forth.
Suitable organic solvents of the slurry 78 may include N-methyl-2-pyrrolidone (“NMP”), dimethylformamide (“DMF”), acetone, dimethylacetamide (DMAc), tetramethyl urea, dimethyl sulfoxide (“DMSO”), triethyl phosphate, and mixtures thereof.
Generally, the slurry 78 is prepared with a ratio of active material/conductive additive/binder, suitable ratios may be, for example, 93/4/3 or 70/20/10. However, the amounts of active material 81 may range from 70 wt % to 99 wt %, the amounts of conductive additive 83 may range from 0.5 wt % to 20 wt %, and the amounts of binder 85 may range from 0.5 wt % to 10 wt %. According to yet other embodiments, the amounts of active material 81 may range from 70 wt % to 99 wt %, the amounts of conductive additive 83 may range from 1 wt % to 20 wt %, and the amounts of binder 85 may range from 0 wt % to 10 wt %.
The slurry 78 may then be applied to the porous substrate 74 (Block 82). As illustrated in
The electrolyte is not specifically shown in
Referring now to
Turning now to
With desired functionality determined, a composite suspension 114 may be prepared (Block 116). While specific compositions and ratios of the elements comprising the composite suspension 114 may vary, the suspension 114 may generally include an active material 118, a conductive additive 120, a binder 122, and a solvent (not shown).
Composition of the active material 118 depends, at least in part, on a functionality of the electrode 110 being prepared and whether an anode or a cathode is desired. For purposes of illustration herein, the active material 118 may include Li4Ti5O12, LiFePO4, LiCoO2, and so forth. Amounts of the active material may range from 25 wt % to 80 wt %.
The conductive additive 120 facilitates electronic transport and should, therefore, be conductive. Exemplary conductive additives may include, for example, any carbon-based material or metallic filler (for example, Al, Cu, Ag, and Ni) having high conductivity and configured to provide a percolative network. For purposes of illustration herein, the conductive additive 120 includes CNFs, providing high conductivity (greater than 1000 S/cm), a large aspect ratio (50 nm to 150 nm diameter with lengths of up to 100 μm), and large tensile strength (greater than 2 GPa). The conductive additive 120, particularly CNFs, may further aid in establishing a porous scaffolding that promotes ion diffusion. An amount of the conductive additive may range from 10 wt % to 40 wt %.
While a composition of the binder 122 may vary, as would be known by those of ordinary skill in the art, the binder 122 may generally include any chemical system configurable to provide adhesion, mechanical support, and flexibility. According to particular exemplary embodiments of the present invention, as described herein, the binder 122 may include those described above with reference to the electrode 72 (
Suitable solvents include those in which the binder 122 may be dissolved and stable over a period to time (such as for use as an ink, if desired). Specifically, as described herein, NMP, may be preferred. While non-solvent elections are more applicable to electrolyte fabrication, described in greater detail below, if a non-solvent is selected for preparing the composite electrode 110, then the weight fraction of non-solvent to solvent may range from 0.0 wt % to 30 wt %.
In preparing the suspension 114, the binder 122 may be dissolved in the solvent. The conductive additive 120, may be added to the solution. The manner by which the suspension is mixed may include shaking, sonication, and so forth. Selection of the active material 118, as described in this specific embodiment, is dependent on whether the suspension 114 is to produce a cathode or an anode, may then be added to the well-dispersed suspension. Again, the suspension 114 may be mixed.
Referring still to
If printing is desired (“Print” branch of Decision Block 124), and as shown in
Regardless of the final shape, whether the cast disc 138 (
The composite electrode 110 may then be used in preparing of a battery cell 145 (Block 146), which is shown in
Turning now to
Good solvents are those solvents that dissolve the binder well and by energetically favorable interactions between the binder and the solvent molecules; non-solvents are those solvents that do not dissolve the binder as well as the good solvents because of a preference of the binder to self-interact in the presence of the non-solvent. According to some embodiments of the present invention, non-solvents have a boiling point that is higher than a boiling point of the good solvent. In such instances, the non-solvent remains during a drying process, leading to pore formation via a phase inversion (“PI”) process. Generally, an amount of the good solvent may range from 90 wt % to 95 wt % while an amount of the non-solvent may range from 0.1 wt % to 30 wt % or, more specifically for some embodiments, from 5 wt % to 10 wt %.
Once the solution is mixed, a ceramic filler may, optionally, be introduced (Block 152) and the suspension mixed (Block 154). The ceramic filler may be any suitable inert particulate suitable for disrupting dendrite formation. Suitable ceramic fillers may include, for example, alumina (Al2O3), silica (SiO2), titania (TiO2), magnesia (MgO), lithia (Li2O), lithium aluminate (LiAlO2), barium titanate (BaTiO3), lithium aluminum germanium phosphate (“LAGP”), lithium aluminum titanium phosphate (“LATP”), lithium lanthanum titanate (“LLTO”). An amount of the ceramic filler may range from 50 wt % to 80 wt %.
In Block 156, the suspension may be printed, such as in the manner described above with reference to
According to still other embodiments of the present invention, materials for the method of preparing an electrolyte according to
For high temperature applications, selection of materials comprising the electrolyte, electrode, composite electrode, and so forth may be limited as compared to the embodiments described above, including those at room temperature. For instance, a LiFePO4//Li4Ti5O12 full-cell prepared using a CPE-PI electrolyte prepared according to embodiments described above does not cycle at high temperature when lithium bis(trifluoromethylsulfonyl)imide (“LiTFSI”) is used as the salt. While not wishing to be bound by theory, it is believed that failure of these full-cells may be due to liquid electrolyte (trifluoromethylsulfonyl)imide, “TFSI,” intercalation into CNFs comprising the composite electrodes. Such intercalation may cause electrolyte and Li+ consumption, which may render the full-cells unable to achieve a single full charge.
For instance, cells suitable for operating at temperatures ranging from room temperature up to about 120° C. may comprise electrolyte comprising a thermally stable salt dissolved in a high boiling point liquid (i.e., a liquid having a boiling point greater than 200° C.). Suitable salts may include LiTFSI or lithium bix(oxoalato)borate (“LiBOB”). The electrolyte liquid may include, but is not limited to, high boiling point carbonates (such as EC, PC, or dibuty carbonate (“DBC”)), room temperature ionic liquids (“RTILs”), other high boiling point liquids (such as γ-butyrolactone (“GBL”) and tetraethylene glycol dimethyl ether (“TEGDME”)), or combinations thereof. Such electrolyte may facilitate the formation of a stable solid electrolyte interphase (“SEI”) layer. The SEI is a passivating, solid layer typically formed on the surface of the anode (e.g., graphite or lithium metal) that is electrically insulating but facilitates ionic conduction. The SEI layer is necessary to minimize irreversible liquid electrolyte degradation and capacity fade. Electrolyte additives, such as vinylene carbonate (“VC”) and ethylene sulfite (“ES”) may aid in the formation of the SEI layer. An amount of the electrolyte additives may range from 0.0 wt % to 10 wt %.
LiBOB in PC or EC provides thermal stability suitable for certain embodiments of the present invention. However, use of LiBOB and SEI forming materials may limit the selection of conductive additives as such liquids may react with additives having high surface area carbon (such as carbon black or CNFs), a phenomenon that may be accelerated at elevated temperatures. Therefore, lower surface area conductive additives, such as graphite, may be preferred.
Flexible or creasable batteries suitable for operating at temperatures ranging from room temperature up to about 120° C., according to embodiments of the present invention may be prepared by from a slurry comprising a suitable conductive additives including low surface area carbons (such as graphite) or metallic fillers (for example, Al, Cu, Ag, or Ni) with a suitable binder (PVDF, PVDF-HFP, PTFE, CMC, or SBR), and an active material (graphite, LiFePO4, LiMn2O4, LiNiMnCoO2, Li4Ti5O12, silicon, or LiCoO2) applied to a flexible substrate.
The flexible or creasable and high-temperature battery may further include a CPE-PI separator and a liquid electrolyte having 1 M LiBOB in 1/1 EC/PC and 5% VC.
The flexible or creasable and high-temperature battery may further include an anode having an anode active material on a flexible substrate or a cathode having a cathode active material on a flexible substrate.
In some instances, formulations for components according to one or more embodiments described herein may require alterations to render the formulations suitable for certain print techniques. For instance, slot die coating techniques require fast dry times. As such, and by way of example, a fast drying electrolyte according to an embodiment of the present invention may include a combination of acetone and water as the solvent and non-solvent, respectively. More particularly, PVDF-HFP could be dissolved in a 95/5 w/w solution of acetone and water. Once dissolved, Al2O3 may be introduced and the solution sonicated. The low boiling points of acetone and water facilitate faster drying times by allowing removal of solvent within minutes at room temperature.
Referring now to
In use, and with reference now to
In Equation 1 (schematically illustrated in
The following examples illustrate particular properties and advantages of some of the embodiments of the present invention. Furthermore, these are examples of reduction to practice of the present invention and confirmation that the principles described in the present invention are therefore valid but should not be construed as in any way limiting the scope of the invention.
EXAMPLE 1MWNT CCs according to embodiments of the present invention were prepared from slurries comprising LiFePO4 (cathode) or Li4Ti5O12 (anode) as an active material, graphite powder as a conductive additive, and PVDF as a binder in NMP solvent. Both LiFePO4 and Li4Ti5O12 slurries were applied, separately, to MWNT mats and conventional metal foils (Al and Cu) by a doctor blade method (Gardco Inc., Pompano Beach, Fla.) using a 6 mil path depth, and subsequently dried at 120° C. under vacuum for at least 12 hr to yield an active coating. Samples (0.375 in diameter discs) were punched from each of the MWNT CCs and the foil-based CCs. On average, the weights of Li4Ti5O12 and LiFePO4 in each disc ranged from 2.0 mg to 2.5 mg for the foil-based and MWNT CCs, respectively, corresponding to coating thicknesses ranging between 50 μm and 60 μm.
Wetting properties of these MWNT CCs and Li4Ti5O12/MWNT electrodes were compared to commercially-available and conventional copper foil CCs and Li4Ti5O12/Cu electrodes. Commercially-available CC and electrodes included copper (EQ-bccf-25u) or aluminum (EQ-bcaf-15u-280) foil CC s (MTI Corp., Richmond, Calif.). Spinel Li4Ti5O12 nanopowder, PVDF, and NMP were purchased from Sigma Aldrich Co. (St. Louis, Mo.). LiFePO4 was purchased through BASF SE (Ludwigshafen, Germany) (HED LFP-400), and graphite was purchased from Asbury Graphite Mills (Ashbury, N.J.) (HPM 850 powder). 1 M LiPF6 in 1/1 EC/DEC by wt. was used as received from Novolyte Technologies Inc. To evaluate electrolyte wetting, samples were taped to a glass slide and 2.0 μL of liquid electrolyte (1 M LiPF6 in 1/1 EC/DEC by wt.) was applied to the center of each sample with a micropipetter.
Liquid electrolyte wetting was evaluated using electrodes prepared in the manner described in Example 1 by depositing a 2 μL, electrolyte (1 M LiPF6 in 1/1 EC/DEC by wt.) drop on each cell. For CCs having the active coating and for neat CCs (that is, the substrates), the drop quickly wicks through the MWNT mat and is visible on the bottom side of the substrate. On the electrodes having foil-based CCs, the electrolyte drop did not wick through the substrate thickness.
The electrolyte droplet on the neat MWNT mat demonstrated a significantly lower contact angle as compared to the contact angle of the electrolyte drop on the neat Cu foil. Such result indicates better wettability in the MWNT mat.
Mechanical testing was performed using an H10K-S UTM benchtop tester (Tinius Olsen, Horsham, Pa.) with a HTE-100 N load cell at an extension rate of 0.5 mm/min. Surface area and porosity data of the MWNT CC were calculated by the Brunauer-Emmett-Teller (“BET”) method based on N2 adsorption-desorption isotherms at 77 K obtained using an ASAP 2020 surface area and porosimetry analyzer (Micromeritics Instrument Corp., Norcross, Ga.). Each sample was degassed at 200° C. under vacuum for 12 hr before acquiring measurements. Electrical conductivity was performed with an SP4, four point probe head (Lucas Labs, Gilroy, Calif.) controlled with a 2410 SourceMeter (Keithley Instruments, Solon, Ohio) and a computer-operated LabVIEW program (National Instruments, Austin, Tex.).
Samples from MWNT CCs (
Samples of the MWNT CCs were sufficiently porous such that electrolyte (1 M LiPF6 in 1/1 EC/DEC by wt.) could access the active material from both sides the MWNT mat, which promotes homogenous wetting. Similar wetting is not possible with the foil-based CCs. Proper wetting is a critical factor for Li-ion batteries as insufficient or inhomogeneous electrolyte wetting has been shown to accelerate cell degradation and shorten cell life.
The high porosity of the MWNT CCs yielded a much lower density as compared to the foil-based CCs (0.52 g/cm3, 2.56 g/cm3, and 8.87 g/cm3 for MWNT CCs, Al-based CCs, and Cu-based CCs, respectively). This, as illustrated in
Electrochemical performance of half-cells composed of either LiFePO4 or Li4Ti5O12 slurries on both MWNT mats and conventional metal foils (Al and Cu) were fabricated versus a lithium counter electrode. Electrode samples were assembled into a 2325 coin cell configuration under argon environment (less than 1 ppm of each of H2O and O2). As illustrated in
For half-cell experiments, each electrode sample (having either a foil-based CC or a MWNT CC) was assembled with lithium foil as counter and reference electrodes and a porous polymer membrane (CELGARD 2325, Celgard, LLC, Charlotte, N.C.) as the separator. The half-cells were soaked in a liquid electrolyte solution of 1 M LiPF6 in 1/1 w/w EC/DEC (Novolyte Technologies Inc., Independence, Ohio).
In
Half- and full-cells were assembled using both MWNT CCs and foil-based CCs. Performance and energy density at a C/5 current rate were compared. All half- and full-cells comprising MWNT CCs exhibited extremely stable charge/discharge profiles (
Electrochemical cycling was performed using a series 4000 battery test system (Maccor, Inc., Tulsa, Okla.). LiFePO4 and Li4Ti5O12 half-cells were investigated for cycling stability through 100 cycles at a constant C/5 current rate between 2.0 V and 4.3 V versus Li/Li+ and 1.0 V and 2.1 V versus Li/Li+, respectively. The rate studies were carried out in a similar manner, with each half-cell tested for five cycles at the following C-rates: C/5, C/2, 1 C, 2 C, 5 C, and repeat of C/5. All C-rates were calculated with 1 C being defined as 170 mA/g and 175 mA/g for LiFePO4 and Li4Ti5O12, respectively.
Full-cell electrochemical cycling was carried out in a similar fashion, with LiFePO4 and Li4Ti5O12 used as cathode and anode, respectively. These cells were cycled at a constant C/5 current rate (calculated using the active mass of the limiting electrode, Li4Ti5O12) between 1.2 V and 2.4 V, with 1 C being defined as 175 mA/g.
EXAMPLE 3In situ durability analysis of full-cells (see Example 2) began with cutting battery components and encapsulating the components between two layers of 75 μm thick Surlyn (GLTE/M, Europack, Inc., Wilmington, Del.). The electrode samples were cut to a size of 3×3 cm2 with electrical lead dimensions of approximately 1×2 cm2. The actual anode capacity/cathode capacity ratio was adjusted to between 0.80 and 0.90 for the Li4Ti5O12/LiFePO4 full-cells. A CELGARD 2325 separator was cut to a size of 4×5 cm2. Copper wires were placed in electrical contact with the leads of each electrode sample while a remainder of the copper wires extended externally from the encapsulation layers. Optionally, some copper wires were laminated between two sheets of 75 μm Surlyn using a GBC 9″ Personal Desktop Laminator (General Binding Corp., Lake Zurich, Ill.) to prevent electrolyte leakage. A perimeter of each cell was sealed using a ZIPLOCK V151 vacuum sealer system (S.C. Johnson & Son, Inc., Racine, Wis.). Sealing was performed between two non-porous TEFLON sheets (E. I. du Pont de Nemours and Co.) to prevent sticking to the heating element. Using a needle, 0.8 mL of liquid electrolyte (1 M LiPF6 in 1/1 EC/DEC by wt.) was added to each cell while in an argon glovebox. The excess region of the cell was sealed off using the method described above.
Contact between all layers was improved by carefully kneading the fully constructed cell to ensure electrolyte uptake and to remove any bubbles.
Testing was accomplished using a flex tester according to embodiments of the present invention, as exemplified in
Bending conditions were chosen such that ϵ ranged from 2.1% (r=7.88 mm) to 2.6% (r=6.35 mm) or to 4.2% (r=4.00 mm). Given that the total thickness of each cell was about 335 μm, shifting of the neutral axis could be neglected, and its location was taken to be the middle layer of the cell (i.e., the separator). A pure-bending set-up and linear strain field were assumed for calculations. Cells with dimensions of approximately 12 cm×8 cm were rolled a linear distance of 3 cm at a linear velocity of 10 mm/sec using Motion Planner v. 4.3.2 software (Parker Hannifin, Corp., Rohnert Park, Calif.). Each complete cycle (roll/unroll) took 6 sec for completion.
The electrochemical properties were galvanostatically tested in each cell with potential ranging from 1.2 V to 2.4 V. Each cell was charged in an argon environment to 2.4 V at a C/3 current rate. The bend test was then performed under ambient conditions while each cell was discharged at a C/5 current rate (calculated using the active mass of Li4Ti5O12). Every 20 min, the mandrel was replaced with one having a smaller radius.
For better comparison, representative 1.5 min increments from each complete cycle were selected to highlight the amplitude of voltage noise on cells having MWNT CCs or foil-based CCs (illustrated inserts provided in
An ideal flexible power source should experience minimal voltage noise during continuous deformation in order to maintain a steady supply voltage, a point which is especially significant for reliable microsystem performance. To evaluate such voltage noise, strain graphically illustrated in
Stability of cells prepared in the manner described in Example 3 was evaluated by repeatedly folding and creasing each cell along a selected fold line. The fold line was selected to correspond to an extreme amount of mechanical stress and strain at the crease, which is illustrated in
By contrast,
Failure of the cells having foil-based CCs resulted from either severe buckling or fracturing along the fold line due to permanent plastic deformation, fatigue, and catastrophic failure of the metal foil during testing. Although failure from tearing is easy to conceptualize, buckling is also highly detrimental to performance as it decreases compliancy, exacerbates interlayer separation, and leads to severe voltage spikes and capacity degradation. On the other hand, the cells having MWNT CCs (
The stability of cells prepared in the manner described in Example 3 was evaluated by repeatedly folding each cell along two, perpendicular axes. As shown in
The discharge voltage of the cells having MWNT CCs (graphically illustrated in
Resultantly, cells having MWNT CCs provide particular advantage in origami-based applications.
EXAMPLE 6A MWNT CC having pristine LiFePO4 active coatings (hereafter, LiFePO4/MWNT) was subjected to 50 cycles of the ±180° folding, which is illustrated in
To complement the macroscopic observations, both the MWNT and Al-based electrodes were cryo-fractured in a direction perpendicular to the folding line for cross-sectional SEM examination. Resulting images are shown in
The mechanical behavior of CCs was assessed by uni-axial tensile testing. Results of the testing are graphically illustrated in
The MWNT mat, therefore, should have sufficient tensile strength to act as a mechanical support for battery electrodes. The increase in maximum strain should yield improved performance under flex, bend, or crease testing.
Composite electrodes according to various embodiments of the present invention were prepared with Li4Ti5O12, LiFePO4, and LiCoO2 active material powders purchased from the MTI Corp. CNFs (PYROGRAF-III PR-19-HT) were supplied by Applied Sciences Inc. (Cedarville, Ohio). Graphite was purchased through Asbury Graphite Mills (HPM 850 powder) and Super P-Li carbon black was from TIMCAL Graphite & Carbon. KYNAR HSV 900 PVDF was provided from Arkema Inc. (King of Prussia, Pa.). NMP was purchased through Sigma Aldrich Co. Ratios of these materials for various samples are provided in Table 4, below.
For composite electrodes having an active material, a conductive additive, and a binder, each batch included 100 mg of the selected active material and 5 mL solvent (conductive additive and binder were scaled according to the ratios in Table 4). The procedure for each composite electrode included first dissolving the binder in NMP. Next, CNFs were added and probe sonicated for 1 min at 17% amplitude using a Sonics Vibra-Cell VCX 750 Probe Sonicator (Sonics and Materials, Inc., Newton, Conn.) with a 0.25 in microtip. The active material was then added, and the mixture was bath sonicated for 10 min using a 2510 bath sonicator (Branson Ultrasonics Inc., Danbury, Conn.). The suspensions were shaken by hand several times during the sonication period for more thorough mixing. Finally, the suspension was transferred to a 6 cm diameter PTFE dish and dried at 90° C. Once the solvent had evaporated, the suspensions were dried under vacuum at 120° C. overnight to yield composite electrodes.
Samples (9.5 mm diameter discs) were punched from the composite electrodes with typical weights of Li4Ti5O12, LiFePO4, and LiCoO2 comprising each sample ranging from 1.5 mg to 3.0 mg. These weights corresponded to sample thicknesses ranging between 100 μm and 200 μm, as measured by a digital micrometer (Mitutoyo America Corp.). Half-cells were assembled versus lithium foil under argon environment (less than 1 ppm H2O and O2) as a full-cell having the 2325 coin cell configuration illustrated in
The electrochemical performances of composite electrode compositions selected from Table 4 were studied and evaluated. For example, electrochemical performance of four electrodes comprising Li4Ti5O12/CNF/PVDF (Sample Nos. A, B, D, and F) are graphically illustrated in
Where a) sample ratios were by wt %, b) four-point probe electrical conductivity, c) tensile strength at failure, d) carbon black, and e) the sample was too brittle to measure electrical conductivity or mechanical properties.
All electrochemical cycling was performed using a Maccor series 4000 battery test system (Maccor, Inc.). LiFePO4, LiCoO2, Li4Ti5O12, and CNFs were investigated for cycling stability at a constant C/5 current rate between 2.0 V and 4.3 V, between 3.0 V and 4.2 V, between 1.0 V and 2.1 V, and between 0.01 V vs. 3.0 V Li/Li+, respectively. Rate studies were carried out in a similar manner, with each half-cell tested for five cycles at C/5, C/2, 1 C, 2 C, 5 C, and C/5 for all active materials. All C-rates were calculated with 1 C being defined as 170 mA/g, 155 mA/g, 175 mA/g, and 372 mA/g for LiFePO4, LiCoO2, Li4Ti5O12, and CNFs, respectively.
Electrochemical Impedance Spectroscopy (“EIS”) was used to investigate losses affecting electrode performance and were conducted in a three-electrode split cell (MTI Corp.) with Li foil as the counter and reference electrodes. An area of the working electrodes (i.e., Li4Ti5O12 composites) was 1.6 cm2. The liquid electrolyte was the same as the liquid electrolyte used in the cycling experiments.
Measurements were performed with a Gamry Reference 600 Potentiosat/Galvanostat/ZRA (Gamry Instruments, Philadelphia, Pa.) at open circuit potential by applying an AC amplitude of 10 mV over a frequency ranging from 50 mHz to 300 kHz at room temperature. Prior to performing the EIS measurements, the electrodes were subjected to one discharge/charge cycle at C/5 between 1.0 V and 2.1 V to activate the electrode/electrolyte interface, followed by a discharge cycle to 50% depth-of-discharge. The samples were then allowed to equilibrate for 2 hr before the EIS measurements were acquired to minimize the drift in open circuit voltage. A stable open circuit voltage was typically attained at 1.58 V.
Curve fitting of Nyquist plots was used for quantitative analysis of the variation in impedance between the composites. An equivalent circuit diagram illustrating the investigative process is shown in
Impedance spectra were fitted to the circuit illustrated in
The increased impedance values in composite electrodes comprising Sample No. A vs. Sample No. B were consistent with the electrochemical data in
Referring still to
The combination of a semicircle and straight line in the lowest frequency region represents the phase transformation from Li4Ti5O12 (spinel) into Li4Ti5O12 (rock-salt) and solid-phase diffusion of Li+ within a bulk of the composite electrode material, respectively. Overall, the impedance curves of the coin cell having a composite electrode comprising Sample No. B suggest comparable or improved electronic properties relative to similar works performed with Li4Ti5O12 coatings on Cu foil, which would suggest that embedded CNFs are sufficient to replace conventional metal foil CCs.
One of the biggest factors dictating electrochemical performance of the composite electrodes prepared in accordance with embodiments of the present invention is binder loading, which does not have a significant impact on the electrical conductivity of the composites. All of the Li4Ti5O12/CNF/PVDF composite electrodes of Table 4 possessed acceptable electrical conductivities (ranging between 3 S/cm and 10 S/cm), since the CNF loading is well above the percolation threshold (which is known to be less than about 1 wt % when using CNFs embedded in a polymer matrix).
The range of conductivity is sufficient for lithium ion batteries without a dedicated metal foil CC. Therefore, the binder may be affecting an intrinsic composite property, such as lithium diffusivity. Lithium ion diffusivity (DLi
where R is the gas constant, T is the absolute temperature, A is the area of the composite electrode, n is the number of electrons transferred, F is the Faraday constant, C is the molar concentration of lithium, and σ is the Warburg coefficient.
Based on the evaluation of these parameters, illustrated in
Because polymer loading correlated with degraded performance, capacity as a function of polymer density throughout the composite for both low and high C-rates was evaluated.
Mechanical properties of the composite electrodes of Example 8 were also investigated. SEM imaging was performed using an FEI Sirion XL-30 FEG-SEM (FEI, Co.). Samples for cross-sectional SEM were prepared through cryo-fracturing in liquid N2. All tensile tests were conducted on a DMA Q800 Dynamic Mechanical Analyzer (TA Instruments, New Castle, Del.) in displacement control mode with a constant strain rate of 1%/min at room temperature. The composite electrodes were cut into rectangular tensile specimens of 20 mm length and 3 mm width using a razor. The finished tensile specimens were examined using a Zeiss Stemi DV4 optical microscope (Carl Zeiss, AG) to ensure that there were no visible flaws from cutting. The thickness of each tensile specimen was determined from the average of three measurements taken along the gauge length with a Mitutoyo micrometer (Mitutoyo America Corp.). The surface area and porosity data of the MWNT mat were calculated by BET method based on N2 adsorption-desorption isotherms at 77 K obtained using an ASAP 2020 surface area and porosimetry analyzer (Micromeritics Instrument Corp.). The sample was degassed at 100° C. under vacuum for 16 hr before running the measurement. Electrical conductivity was performed with a Lucas Labs SP4 four-point probe head controlled with a Keithley 2410 SourceMeter (Keithley Instruments) and a computer operated LabVIEW program (National Instruments).
To understand the observed improved mechanical strength and electrochemical performance of electrodes prepared in Example 8, SEM images were acquired to characterize composite electrodes comprising Sample Nos. A and B. Resultant images are shown in
The composition of Sample No. A (
The 40% loading of active material composite electrodes comprising either Sample No. A or Sample No. B is comparable to commercially-available Li-ion battery electrodes, which typically consist of about 30% to 35% active material when considering conductive additive, binder, and metal foil CC.
Electrodes having other, more common conductive additives for Li-ion batteries, such as graphite or carbon black, were also fabricated at the optimum loading found for the Li4Ti5O12/CNF/PVDF (Samples Nos. G and H). However, both electrodes comprising Sample Nos. G and H presented undesirable issues during casting, such as inhomogeneous settling in the case of graphite or cracking and curling in the case of carbon black.
EXAMPLE 12To examine versatility of methods of preparation, composite electrodes were prepared in accordance with the method described in Example 8, using LiFePO4 (Sample No. J) and LiCoO2 (Sample No. I), with optimized ratios determined using Li4Ti5O12. Electrical and mechanical characterizations were performed in a manner similar to those described in Examples 8-10.
The composite electrodes were tested for 100 cycles at a C/5 current rate, as shown in
Composites were prepared using only CNF and PVDF, according to the method described in Example 8 using a 50/50 ratio of CNF to PVDF (Sample No. K) and a 67/33 of CNF to PVDF (Sample No. L).
While the composite electrodes comprising Sample Nos. K and L showed excellent C/5 cyclability over 50 cycles, the composite electrode with Sample No. K (with 50% CNFs) achieved a reversible capacity of only 115 mAh/g, whereas the composite electrode with Sample No. L (with 67% CNFs) showed a 250 mAh/g CNF-specific capacity (
Using the testing methods described above, composite electrodes comprising Sample Nos. K and L were formed as free-standing, flexible films; however, the electrochemical performance and mechanical strength both suffered. At higher electrochemical windows, the CNFs only act as a conductive additive and do not participate electrochemically within the electrodes. Only minimal CNF contributions (0.6 mAh/g to 1.7 mAh/g) were observed (
In reviewing the composite electrodes from Example 8, those comprising Sample No. B were found to retain a casting shape without shrinkage, which is a desirable trait for a printable ink. By comparison, those comprising Sample No. A led to significant shrinkage and out-of-plane buckling around the edges during the casting process. Any shrinkage or out-of-plane deformation would severely complicate printing, particularly for building up sequential layers in the z-direction. The 40/40/20 ratio of active/CNF/PVDF provides a good balance between electrochemical performance, mechanical strength, and shape retention while drying.
In that regard, composite inks configured for printing according to embodiments of the present invention were prepared using the procedure described above in Example 8, except solid loading of each composition was doubled to increase viscosity. For anodes consisting of only CNFs, 150 mg of CNFs were added to a 6 mL polymer/NMP solution (the amount of binder was scaled according to the ratios in Table 4). The anode mixture was probe sonicated for 2 min at 22% amplitude, transferred to a 6 cm PTFE dish, dried at 90° C. to remove solvent, then vacuumed at 120° C. overnight.
Printing was performed by extrusion from a syringe mounted on a custom Aerotech 3D-motion gantry (Pittsburgh, Pa.). Typical motion speeds ranged from 5 mm/s to 20 mm/s. Extrusion was controlled using an Ultimus V pressure pump system from Nordson EFD (East Providence, R.I.) at pressures ranging from 0.5 psi to 5 psi. The syringe tips, also from Nordson EFD, ranged from 18 gauge to 25 gauge, of 0.25 in stainless steel. Needle geometry, pressure, and print speed were tuned to minimize feature size and maximize print reliability (reliable extrusion and wetting of ink). Tool paths were developed in AeroBasic scripting language.
A printed Li4Ti5O12 disc (not shown) of the prepared anode composite ink was electrochemically cycled in the half-cell configuration. The printed anode's rate performance was on par with the performance of the casted composite electrodes comprising Sample No. B (graphically illustrated in
To emphasize the versatility of the composite electrodes, a composite cathode ink comprising LiFePO4 ink and having ratios similar to the composite electrode was prepared and used to print several complex patterns onto a substrate (here, transparency paper). Two of such patterns are shown in
A dry phase inversion ink for printing porous electrolytes for use in Li ion batteries according to embodiments of the present invention was prepared with KYNARHSV 900 PVDF from Arkema Inc.; 40-50 nm APS powder Al2O3 from NanoDur purchased through Alfa Aesar (Ward Hill, Mass.); and NMP and glycerol purchased from Sigma Aldrich Co.
Nanocomposite electrolytes were prepared by dissolving 0.1726 g PVDF in 3 mL NMP (or NMP/glycerol 95/5 w/w for electrolytes made with the phase inversion process). For electrolytes containing ceramic filler, 0.4030 g Al2O3 (70/30 Al2O3/PVDF w/w) was added and bath sonicated for 3 hr using a 2510 bath sonicator (Branson Ultrasonics Inc.).
For electrodes, the CNFs were probe sonicated in solvent for 1 min at 17% amplitude using a Vibra-Cell VCX 750 Probe Sonicator (Sonics and Materials, Inc.) with a 0.25 in microtip. Active material was then added, and the mixture was bath sonicated for 10 min. The suspensions were shaken by hand several times. The suspension was then either cast or printed. Generally, each suspension included 100 mg of active material and 5 mL solvent (conductive additive and binder were scaled accordingly).
For casting, the suspensions were transferred to a 6 cm diameter PTFE dish and dried at 90° C. Once NMP evaporated, the suspensions were subjected to vacuum at 120° C. overnight for better drying and to remove glycerol. Electrolyte samples (19 mm discs) were punched from cast electrolyte membranes; composite electrode samples (prepared in a manner similar to those of Example 8) (9.5 mm diameter discs) were punched from composite electrodes. Typical weights of Li4Ti5O12 and LiFePO4 in each composite electrode sample ranged from 1.5 mg to 3.0 mg, with corresponding thicknesses ranging from 100 μm to 200 μm (measured by a Mitutoyo digital micrometer). Composite electrode samples were assembled versus lithium foil under argon environment (less than 1 ppm H2O and O2) in a 2325 coin cell configuration (see
For printing, direct ink write was performed by extrusion from a syringe mounted on a custom Aerotech 3D-motion gantry. Typical motion speeds ranged from 5 mm/sec to 20 mm/sec. Extrusion was controlled using an Ultimus V pressure pump system from Nordson EFD at pressures ranging from 0.5 psi to 5 psi. The syringe tips, also from Nordson EFD, ranged from 18 gauge to 25 gauge of 0.25 in stainless steel. Needle geometry, pressure, and print speed were tuned to minimize feature size and maximize print reliability (reliable extrusion and wetting of ink). Tool paths were developed in AeroBasic scripting language.
EXAMPLE 15SEM imaging was performed using a FEI Quanta 600F (FEI, Co.) using the method described in Example 17 (below). All tensile tests were conducted on a TA Instruments DMA Q800 Dynamic Mechanical Analyzer in displacement control mode with a constant strain rate of 1%/min at room temperature (ASTM D882). Rectangular tensile specimen were cut from composite electrodes with dimensions of 20 mm by 3 mm (length by width) using a razor. The finished tensile specimen were examined using a Zeiss Stemi DV4 optical microscope to ensure that there were no visible flaws due to cutting. The thickness of each tensile specimen was determined from the average of three measurements taken along the gauge length with a Mitutoyo micrometer. The surface area and porosity data of the MWNT mat were calculated by a BET method based on N2 adsorption-desorption isotherms at 77 K obtained using an ASAP 2020 surface area and porosimetry analyzer (Micromeritics Instrument Corp.).
The samples were degassed at 100° C. under vacuum for 16 hr before running the measurement. Electrical conductivity was performed with a Lucas Labs SP4 four-point probe head controlled with a Keithley 2410 SourceMeter and a computer operated Lab VIEW program. For wettability experiments, a 10 μL drop of liquid electrolyte was placed onto the samples and spreading of the electrolyte drop was observed.
The coin cells were cycled using a Maccor series 4000 battery test system (Maccor, Inc.). LiFePO4, and Li4Ti5O12 were investigated for cycling stability at a constant C/5 current rates ranging between 2.0 V and 4.3 V and between 1.0 V and 2.1 V vs. Li/Li+, respectively. The rate studies were carried out in a similar manner, with half-cells tested for 5 cycles at C/5, C/2, 1 C, 2 C, 5 C, and repeated C/5 for all active materials. All C-rates were calculated with 1 C being defined as 170 mA/g and 175 mA/g for LiFePO4 and Li4Ti5O12, respectively.
EXAMPLE 16A composite electrolyte according to an embodiment of the present invention was prepared by sonicating Al2O3 nanoparticles (diameter ranging from 40 nm to 50 nm) and PVDF in a mixture of NMP (good solvent) and glycerol (non-solvent). More specifically, KYNARHSV 900 PVDF was provided from Arkema Inc. Al2O3 powder (40 nm to 50 nm) from NanoDur was purchased through Alfa Aesar. NMP and glycerol were purchased from Sigma Aldrich Co. Glycerol was added as a non-solvent to increase porosity through phase inversion to aid in the dispersion of Al2O3 nanoparticles, and to protect PVDF from dehydrofluorination (glycerol has been shown to interact strongly with Al2O3 by forming a bridging alkoxy bond with one of the primary alcohols and a hydrogen bond with the secondary alcohol group). The solvation layer minimized nanoparticle aggregation and aided in more uniform dispersion throughout the electrolyte membrane.
The composite electrolytes were prepared by dissolving 0.1726 g PVDF in 3 mL NMP (or NMP/glycerol 95/5 w/w for composite electrolytes made with the phase inversion process). For composite electrolytes comprising ceramic filler, 0.4030 g Al2O3 (70/30 Al2O3/PVDF w/w) was added and bath sonicated for 3 hr using a Branson 2510 bath sonicator. The suspension was then suitable for casting or printing.
The suspension had a high boiling point (greater than about 200° C.) and was suitable for casting or printing.
For casting, the suspensions were transferred to a 6 cm diameter PTFE dish and dried at 90° C. under nitrogen. Once the NMP had evaporated, the samples were placed in vacuumed at 120° C. overnight to complete drying and to remove glycerol.
Casted electrolyte membranes were punched to 19.0 mm discs and composite electrode samples were punched to 9.5 mm diameter discs. Typical thicknesses of electrolytes used in this study were about 100 μm, although similar performance was observed in thicknesses as low as 30 μm. Electrode samples were assembled versus Li foil under argon environment (less than 1 ppm H2O and O2) in 2325 coin cell configuration (see
For composite electrodes prepared using phase inversion, PVDF was dissolved in 5 mL NMP (or NMP/glycerol 95/5 w/w). Next, the CNFs were added and probe sonicated for 1 min at 17% amplitude using a Sonics Vibra-Cell VCX 750 Probe Sonicator with a 0.25 in microtip. 100 mg active material was then added (CNF and PVDF loading was determined using the composite wt. ratios listed in the text and figure captions), and the mixture was bath sonicated for 10 min. The samples were shaken by hand several times during the sonication period for more thorough mixing. Finally, the mixtures were transferred to a 6 cm diameter PTFE dish and dried at 90° C. Once NMP was evaporated, the samples were dried under vacuum at 120° C. overnight. Composite inks for printing were prepared using the same procedure as above except the solids loading was doubled to increase the viscosity of the mixture and sonication intensity was increased to 21% for 2 min.
Direct ink write printing was performed by extrusion from a syringe mounted on a custom Aerotech 3D-motion gantry. Typical motion speeds ranged from 5 mm/s to 20 mm/s. Extrusion was controlled using an Ultimus V pressure pump system from Nordson EFD at pressures ranging from 0.5 psi to 5 psi and 5 psi to 25 psi for the composite electrode and composite electrolyte inks, respectively. The stainless steel syringe tips, also from Nordson EFD, were 18 gauge to 25 gauge and 0.25 in to 0.50 in in length. Needle geometry, pressure, and print speed were tuned to maximize print reliability and fidelity (reliable extrusion and wetting of the composite electrolyte ink). Composite electrolyte inks were deposited onto glass substrates that were previously scrubbed with acetone and dried with compressed air. Tool paths were developed in the AeroBasic scripting language.
Multilayer electrolyte prints required specific drying conditions to prevent a first printed layer from swelling subsequent printed layers were dispensed thereon. Each printed layer was partially dried by covering with a glass dish and heating to 90° C. for 10 min to 20 min, depending on the amount of ink suspension dispensed. Subsequently printed layers were then dispensed atop the previously printed, partially dried ink.
EXAMPLE 17When NMP alone is used to prepare a PVDF/Al2O3 composite electrolyte, a color of the composite turns from white to orange/brown with heated drying. Such color change may be explained by an Al2O3-catalyzed dehydrofluorination reaction in PVDF that forms conjugated double bonds along the main chain and crosslinking between polymer chains. Dehydrofluorination of PVDF has been associated with undesirable shrinkage and embrittlement, which was observed in composite electrolytes comprising PVDF/Al2O3 without glycerol.
SEM imaging was used to characterize the cross-sections of three different PVDF-based, electrolytes (PE-PI, CPE, and CPE-PI) prepared in accordance with Example 16. Exemplary images are provided in
Thermal stability was studied by placing the composite electrolyte and CELGARD 2325 on a hotplate for 5 min while increasing the temperature. Exact temperatures were determined using a non-contact IR thermometer. Shrinkage was monitored by taking photos after each temperature increase and measuring dimensions of each sample.
Based on the SEM images, the phase inversion process and the inclusion of Al2O3 has a significant impact on the resulting morphology. The pure polymer electrolyte (PE-PI;
In contrast, composite electrolyte morphology became less porous for the CPE having a ceramic and polymer using only good solvent, NMP (
When phase inversion was used with Al2O3 in preparing the composite electrolyte membrane (CPE-PI), a balance between the two aforementioned microstructures was achieved, wherein the resulting composite electrolyte membrane included many small, submicron pores (
The neat polymer membrane prepared with good solvent alone (for example, PE) demonstrated limited porosity (21%), and the addition of ceramic in CPE had an insignificant effect on the porosity value (25%). Both electrolyte films prepared by phase inversion displayed excellent porosities (greater than 50%), exceeding the porosity of commercially-available CELGARD 2325 (38%).
The ionic conductivity of each electrolyte membrane was observed to correlate with both the materials used and the solvents present. The use of phase inversion resulted in a two order of magnitude increase in conductivity when applied to similar material compositions. For example, electrolyte membranes comprising CPE exhibited a low conductivity (0.0080 mS/cm) compared to 0.82 mS/cm for electrolyte membranes comprising CPE-PI. Interestingly, while the porosity was the same for electrolyte membranes comprising CPE-PI and PE-PI, the ionic conductivity was 4-times less in the latter. Even though the majority of its morphology consisted of large, open pores, the smaller, cellular pores in the electrolyte membrane comprising PE-PI likely reduced the ionic conductance. In contrast, electrolyte membranes comprising CPE-PI had uniform porosity throughout its entire thickness, resulting in the highest observed conductivities.
where a) sample ratios by wt %, b) room temperature ionic conductivity, c) activation energy, and d) tensile strength, and e) strain-to-failure.
Ductility, which is closely related to strain tolerance, is of particular concern for applications requiring battery flexibility. Mechanical properties of electrolyte membranes were characterized and displayed in
The thermal properties of the composite electrolyte membranes were also evaluated. The composite electrolyte membranes comprising CPE-PI and commercially-available CELGARD 2325 were subjected to identical thermal treatments. As shown in
Half-cells having composite electrode or electrolyte membranes comprising LiFePO4 or Li4Ti5O12 were cycled using a Maccor series 4000 battery test system for investigating cycling stability at a constant C/5 current rate between 2.0 V to 4.3 V and 1.0 V to 2.1 V vs. Li/Li+, respectively. The rate studies were carried out in a similar manner, with each half-cell tested for five cycles at C/5, C/2, 1 C, 2 C, 5 C, and repeat of C/5. All C-rates were calculated with 1 C being defined as 170 mA/g and 175 mA/g for LiFePO4 and Li4Ti5O12, respectively.
The conductivity cells consisted of two parallel, stainless steel blocking electrodes and a membrane (A=2.85 cm2) soaked with liquid electrolyte. Cells were allowed to age for at least 8 hr in order to activate the membrane before measurements were recorded. Temperature was controlled by a Tenney environmental chamber and recorded with a k-type thermocouple. The ionic conductivity was measured every 10° C. from 20° C. to 60° C. The conductivity cells were allowed to equilibrate for 1 hr at each temperature before the next measurement was recorded. Measurements were performed with a Gamry Reference 3000 at open circuit potential by applying an AC-amplitude of 5 mV over a frequency range from 1 Hz to 1 MHz. The ionic conductivity was calculated according to Equation 3:
where t, Rb, and A are thickness (cm), bulk resistance (Ω), and area (cm2) of the sample, respectively. Symmetric Li/electrolyte/Li cells for dendritic failure testing were prepared in 2325 coin cells. Electrolyte membranes (12.7 mm diameter) were saturated with liquid electrolyte (1 M LiPF6 in 1/1 EC/DEC by wt.) and placed between 10.0 mm diameter Li foil discs. After aging for 8 hr, lithium metal plating/stripping was achieved by alternating a +0.15 mA/cm2 current density (about C/5 rate) for 4.5 hr with a 30 min rest between.
The electrochemical performances of the electrolyte membranes prepared in Example 16 in half-cell configurations using a CC-embedded 40/40/20 LiFePO4/CNF/PVDF electrode (Example 8) were evaluated. The rate performances and corresponding voltage profiles (
The electrolyte membranes comprising CPE-PI were tested for 100 cycles at a C/5 current rate, which is graphically illustrated in
Wettability of electrolyte membranes comprising CPE-PI and a commercially-available separator were compared by applying the same amount of liquid electrolyte thereon. The electrolyte membrane showed near complete wet-out within 30 min; the commercially-available separator did not wet beyond the initial application site. From a commercial point-of-view, good wettability may shorten the electrolyte filling step during assembly and extend the battery's lifetime, whereas poor wettability may do the exact opposite by increasing the internal ionic resistance.
EXAMPLE 19Printed electrode membrane assembly (“PEMAs”) were prepared by printing a composite electrolyte (Example 8), surrounding the printed composite electrode with a collar of the composite electrolyte (Example 14), and covering the collared electrode-electrolyte with at least one layer of the composite electrolyte (Example 14). Challenges in PEMA fabrication arose from the inherent roughness of the dried electrode ink, which had a surface roughness of over 10 μm. This required the composite electrode ink to be dispensed above the highest feature, causing an uneven surface-to-nozzle distance. However, the higher boiling point solvent and the moderate viscosity of the composite electrolyte ink permitted leveling prior to drying. Printed electrolytes and PEMAs were fully dried under vacuum at 120° C., overnight, before use.
In route to batteries fabricated exclusively by an additive manufacturing process, a PEMA was prepared by depositing a composite electrolyte membrane comprising CPE-PI directly over a composite cathode comprising LiFePO4. From SEM images (exemplary image provided in
The dry phase inversion approach used to prepare high performance electrolytes in Example 14 was also directly amenable to printable electrode inks. When only NMP was used, a composite electrode comprising 40/20/40 Li4Ti5O12/CNF/PVDF displayed a sharp drop off in both 2 C and 5 C performances.
High-temperature Li-ion batteries, according to embodiments of the present invention, were prepared and investigated. LiFePO4 was purchased from either MTI Corp. or BASF. Li4Ti5O12 and commercially-available graphite electrodes on copper foil were purchased from MTI Corp. CNFs, commercially-available as PYROGRAF-III (PR-19-HT), were supplied by Applied Sciences, Inc., and graphite was purchased from Asbury Graphite Mills (Ashbury, N.J.) (HPM 850 powder). KYNAR HSV 900 PVDF was supplied by Arkema, Inc. Al2O3 powder (diameters ranging from 40 nm to 50 nm) were purchased from NanoDur through Alfa Aesar. LiTFSI, NMP, glycerol, PC, and EC were purchased from Sigma Aldrich Co. 1-methyl-1-propylpyrrolidinium TFSI (“[MPPyr][TFSI]”) was purchased from IoLiTec, Inc. Conventional liquid electrolyte (LP40, 1 M LiPF6 in 1/1 EC/DEC by wt.), LiBOB, and VC were purchased through BASF.
CPE-PI ink was prepared by dissolving 0.863 g PVDF in 15 mL NMP and 0.64 mL glycerol (95/5 w/w). 2.015 g Al2O3 (70/30 Al2O3/PVDF w/w) was added and bath sonicated for 3 hr using a Branson 22510 bath sonicator. The ink was then suitable for casting or printing. CPE-PI membranes were prepared by doctor blading 3 mL of ink on glass using an adjustable film applicator (MTI Corp.) with an 800 μm height and dried, overnight, at 90° C. under nitrogen. The CPE-PI membranes were then dried, under vacuum, at 120° C. overnight.
Composite electrodes constructed from 40/40/20 active material/CNF/PVDF by wt % were prepared by dissolving 50 mg PVDF in 5 mL NMP. 100 mg CNFs were added and probe sonicated for 1 min at 17% amplitude using a Sonics Vibra-Cell VCX 750 Probe Sonicator with a 0.25 in microtip. 100 mg active material was added and the mixture sonicated for 10 min. The mixtures were shaken by hand throughout sonication. Finally, the mixtures were transferred to a 6 cm diameter PTFE dish and dried at 90° C. Once NMP was evaporated, the samples were dried, overnight, under vacuum at 120° C.
LiFePO4 cathodes with active loading (ranging from 13 mg/cm2 to 15 mg/cm2) were prepared on aluminum foil using a ratio of 75/20/5 LiFePO4/graphite/PVDF by wt %. 56.7 mg PVDF (dissolved in NMP) was combined with 226.7 mg graphite and ground using a mortar and pestle. 850 mg LiFePO4 was added and the mixture was again ground using a mortar and pestle for 10 min. The slurry was transferred to an 8×8 cm2 area on aluminum foil. The coating was dried under vacuum at 90° C. until NMP evaporated, and then under vacuum at 120° C. overnight.
Electrolyte membrane samples (19 mm diameter discs) were punched from the doctor bladed CPE-PI membranes; thicknesses of the electrolyte samples ranged from 50 μm to 90 μm. Composite electrode samples (9.5 mm diameter discs) were punched from the cast mixtures; weights of Li4Ti5O12 and LiFePO4 of samples ranged from 1.5 mg to 3.0 mg, corresponding to sample thicknesses ranging from 100 μm to 200 μm as measured by a Mitutoyo digital micrometer. Foil-based graphite/Cu or LiFePO4/Al electrodes were punched to 12.7 mm diameter discs with active loading ranging from 5.5 mg/cm2 to 5.8 mg/cm2 for graphite and from 13 mg/cm2 to 15 mg/cm2 for LiFePO4.
Liquid electrolyte solutions comprising 1 M LiTFSI solutions were prepared by adding 0.4 g LiTFSI to 1.09 mL PC or RTIL and stirring overnight at 50° C. under argon. 1 M LiBOB in 1/1 EC/PC (v/v) was prepared by adding 0.252 g LiBOB to 0.545 mL of EC and 0.545 mL of PC under argon. The mixture was stirred at 120° C. until dissolved. Optionally, 5 wt % VC relative to EC and PC (0.053 mL) was added, which may aid in SEI formation and cell reliability.
Cells for electrochemical cycling were prepared under argon environment (less than 1 ppm H2O) and O2) in 2325 coin cell configurations. CPE-PI or CELGARD 2325 films were infiltrated with either 1 M LiTFSI in PC, 1 M LiTFSI in [MPPyr][TFSI], or 1 M LiBOB in 1/1 EC/PC and used as the separator. After fabrication, the coin cells were allowed to age for at least 8 hr before electrochemical cycling, so as to allow for homogeneous electrolyte wetting.
Coin cells prepared from component described in Example 20 were cycled using a Maccor series 4000 battery test system. Half-cells based on composite electrodes were assembled vs. lithium and cycled for 5 cycles at 0.17 C, 0.33 C, 1 C, and 3 C, and then for 30 cycles at 0.33 C. All C-rates were calculated with 1 C being defined as 170 mA/g and 175 mA/g for LiFePO4 and Li4Ti5O12, respectively. LiFePO4 was cycled between 2.5 V and 4.0 V versus Li/Li+ at all temperatures except 120° C., where a 1.35 V to 1.8 V window was used. Foil-based electrode half-cells were assembled vs. lithium and cycled at 0.17 C and 0.33 C using a voltage window ranging from 2.5 V to 4.0 V for LiFePO4 and from 0.01 V to 1.0 V for graphite. Foil-based graphite//LiFePO4 full-cells were cycled between 2.5 V and 4.0 V at rates of 0.17 C, 0.33 C, and 1 C. All C-rates for foil-based cells were calculated using observed capacities with 1 C being defined as 330 mA/g for graphite and 150 mA/g for LiFePO4. An anode/cathode ratio of 1.10 was desired to avoid lithium plating on a surface of the graphite. Elevated temperature cycling was done using the Maccor battery test system with battery holders extended into an oven (Memmert Beschickung loading model 100-800) using PTFE-coated wires. For tests using the same cells at different temperatures, cycling was paused to change the temperature and the cells were permitted to equilibrate for 1 hr at the new temperature before cycling was resumed.
EXAMPLE 21Undesirable reactions between LiTFSI and aluminum were avoided by using a CC-embedded composite electrode, such as those having 40/40/20 ratio of active material/CNF/PVDF. While cycling at high temperatures, SEI on lithium decomposes and results in constant SEI reformation, consumption of liquid electrolyte, and increased likelihood of dendritic lithium deposition.
LiFePO4 (
Referring still to
Improved functionality of the composite electrolyte membrane for high temperature cycling is easily appreciated when comparing the results of
The CPE-PI electrolyte membrane of Example 19 matrix was infiltrated with 1 M LiTFSI in [MPPyr][TFSI]. This room temperature ionic liquid (“RTIL”) possesses low viscosity, high ionic conductivity, and has been shown to have potential as a high temperature Li+ transport medium. LiFePO4 composite electrode half-cells were then prepared with this RTIL electrolyte and electrochemically cycled from room temperature to 120° C.
Because full-cells prepared using LiTFSI in PC were unable to achieve one full charge (such as the LiFePO4//Li4Ti5O12 cells), alternative electrode and electrolyte formulations were evaluated. As to electrolyte, a similar cell (using a commercial graphite anode on copper) was prepared with 1 M LiBOB in PC/EC 1/1 v/v and 5% VC as the electrolyte. Both EC and PC have boiling points over 240° C., are thermally stable at the elevated temperatures tested. LiBOB, EC, and VC help to form a stable SEI on graphite anodes and the low melting point of PC (about −50° C.) keeps the electrolyte solution liquid at room temperature.
As shown in
When compared previously discussed electrolyte formulations according to other embodiments of the present invention (such as 1 M LiTFSI in PC), the LiBOB electrolyte gives rise to exceptional coulombic efficiencies (greater than 99%) for both graphite and LiFePO4 half-cells, which may indicate the formation of a more stable SEI on both graphite and lithium.
Flammability of conventional separators (such as CELGARD 2325) and separators prepared in accordance with embodiments of the present invention was evaluated. In
In
While the present invention has been illustrated by a description of one or more embodiments thereof and while these embodiments have been described in considerable detail, they are not intended to restrict or in any way limit the scope of the appended claims to such detail. Additional advantages and modifications will readily appear to those skilled in the art. The invention in its broader aspects is therefore not limited to the specific details, representative apparatus and method, and illustrative examples shown and described. Accordingly, departures may be made from such details without departing from the scope of the general inventive concept.
Claims
1. A method of applying a separator ink onto a dried electrode of a lithium ion battery, the method comprising:
- preparing a separator ink suspension comprising: a binder comprising 20 wt % to 50 wt % of a total weight of the separator ink suspension, the binder being selected from the group consisting of PVDF, PVDF-HFP, PTFE, PEO, PMMA, PAN, CNC, SBR, and combinations thereof;
- a solvent selected from the group consisting of NMP, DMF, acetone, DMAc, DMSO, trimethyl urea, triethyl phosphate, and combinations thereof;
- a non-solvent selected from the group consisting of glycerol, water, ethanol, methanol, ethylene glycol, diethylene glycol, triethylene glycol, hexane, heptane, and combinations thereof; and
- a ceramic filler comprising 50 wt % to 80 wt % of the total weight of the composite electrolyte the ceramic filler being selected from the group consisting of Al2O3, SiO2, TiO2, MgO, Li2O, LiAlO2, BaTiO3, LAGP, LATP, LLTO, and combinations thereof;
- applying the separator ink suspension onto the dried electrode; and
- drying the applied separator ink suspension.
2. The method of claim 1, wherein the dried electrode is uncalendared.
3. The method of claim 2, wherein the separator ink coated dried electrode is calendared.
4. The method of claim 1, wherein the dried electrode is calendared.
5. The method of claim 4, wherein the separator ink coated dried electrode is calendared.
6. The method of claim 1, wherein the dried electrode includes a metal foil current collector.
7. The method of claim 1, wherein the dried electrode is a composite electrode fabricated by a method comprising:
- preparing the composition as a suspension, the suspension comprising:
- an active material selected from the group consisting of Li4Ti5O12, LiCoO2, LiMn2O4, LiFePO4, LiNiMnCoO2, carbon nanofibers, silicon, graphite, and combinations thereof; a conductive additive comprising a carbon based material or a metallic material; and a binder;
- a solvent; and
- optionally, a non-solvent having a boiling point that is greater than a boiling point of the solvent, is configured to cause binder self-interaction inducing porosity, wherein an amount of the active material of the solids ranges from 25 wt % to 80 wt % of the suspension, an amount of the conductive additive of the solids ranges from 10 wt % to 40 wt % of the suspension, and an amount of the binder of the solids ranges from 10 wt % to 50 wt % of the suspension,
- wherein a weight fraction of non-solvent to solvent ranges from 0 wt % to 30 wt %, and
- wherein a viscosity of the composition is selected for printing or casting,
- casting, printing, slot die, or doctor blading the suspension; and
- drying the suspension
8. The method of claim 7, wherein applying the separator ink includes ink jet printing.
Type: Application
Filed: Jun 11, 2020
Publication Date: Oct 1, 2020
Applicant: Government of the United States as represented by the Secretary of the Air Force (Wright Patterson AFB, OH)
Inventors: Michael F. Durstock (West Chester, OH), Ryan R. Kohlmeyer (Dayton, OH), Aaron J. Blake (Huber Heights, OH)
Application Number: 16/899,257
