Patent Application
20250125362
The electrode matrix is usually comprised of carbonaceous conductive additives and non-conducting polymeric binders. Over the last two decades, the most commonly used electrode matrix is polyvinylidene fluoride/carbon black (PVDF/C) due to its excellent electrochemical/chemical stability and ease of processing[1]. In the development of new high-energy-density battery materials, some intrinsic drawbacks of PVDF/C are becoming more prominent[2-4]. Firstly, processing of battery electrodes with PVDF/C requires toxic N-Methyl-2-pyrrolidone (NMP) and electrode drying and solvent recovery are energy intensive[5]. Secondly, fluoropolymers, such as PVDF, exhibit famously weak intermolecular interactions with other substances, which makes them well suited as lubricants[6]. Simultaneously, they are very weak adhesives in electrodes, which significantly contributes to battery performance degradation due to electrode disintegration upon cycling[2,4,7]. Strong interaction between the electrode matrix and active materials is vital to suppress electrode cracking, maintain electrode architecture as well as enhance electrode stability[8]. Thirdly, carbonaceous additives exhibit little polarity at their surface. Given highly polar oxide intercalation materials and electrolytes, this lack of polar interactions within the matrix increases the occurrence of contact loss and carbon agglomeration[4,9]. The development of conductive electrode matrices with polar surfaces improves ease, cost, and environmental impact of electrode processing, and addresses longevity of electrodes with large volume changes by increasing adhesion of the conductor to the active materials[2,4,10].
Several strategies have been reported to prepare alternative electrode matrices to enhance battery performance. Most studies have focused on using aqueous battery binders and other carbonaceous additives[10-12]. Moving to aqueous binders, such as carboxymethyl cellulose (CMC)[12,13], alginate[8,14], and polyacrylic acid (PAA)[15-17], electrode processing becomes cheaper and more environmentally friendly[5,10,13]. At the same time, it exploits intermolecular interactions and chemical bonding with active materials[8,18,19], resulting in stronger adhesion. As a result, the electrode integrity and good performance in high-energy-density electrodes, such as silicon, can be maintained over more cycles. An alternative approach exploits the combination of electrical conductivity, mechanical flexibility, and extended microstructure of conducting polymers to boost the electrical conductivity of battery electrodes with promising results[20-25]. By adding a small amount of conducting polymer into a mixture of active materials, conventional binders, and carbon additives, the conducting polymers can bridge connections between conductive particles that would otherwise be lost, leading to the maintenance of a continuous conductive network. Other electrode matrix designs are based on functional group-modified conducting polymers[26-28] and three-dimensional conducting polymer gels[29-33], in which carbon additives and/or additional binders (PVDF, CMC) are involved during electrode fabrication.
Some studies have reported the fabrication of carbon-additive-free electrodes with only active materials and poly(3,4-ethylenedioxythiophene):polystyrene sulfonate (PEDOT:PSS)[9.34]. No additional conductive additives and binders are required to fabricate electrodes, which suggest the possibility of using conducting polymers as single-component multifunctional electrode matrices. Despite their promising results, however, the high cost of PEDOT:PSS hinders the wide-scale application as a battery electrode matrix. The success of PEDOT:PSS is based on the combination of the conducting polymer PEDOT with a water-dispersible polymer PSS. As PEDOT carries positive charges (electronic holes) along its backbone in its conductive state and PSS contains negatively charged sulfonate groups, both polymers are permanently intertwined, forming a molecular composite[35]. In this composite, PEDOT provides electronic conductivity, whereas PSS increases adhesion, gives the composite film-forming properties and makes it dispersible in water. Inspired by the PEDOT:PSS structure, other combinations of conducting polymers and polyelectrolytes can be developed, as demonstrated herein.
This work introduces a design concept for a new class of electrode matrices for Li-ion batteries. By polymerizing conducting polymer (CP) monomers in the aqueous solutions of polyanionic binders, molecular composites are formed and could be used as multifunctional electrode matrices for Li-ion battery electrodes. Conducting polymer monomers are the smallest constructive units of conducting polymers, which contain conjugated backbone. Common conducting polymer monomers are aniline, pyrrole, 3,4-ethylenedioxythiophene, and the like. Polyanionic binders are polymeric compounds containing negatively charged groups such as carboxylate and sulfonate. Examples for polyanionic binders are carboxymethyl cellulose (CMC), polyacrylic acid (PAA), poly styrene sulfonate, and the like. One of the examples for multifunctional CP-based battery electrode matrices is PPy:CMC composite. As a conductor, PPy:CMC composite provides electrical conduction pathways between electrode active materials, allowing batteries to function at high C-rates without carbon additives added. As an adhesive binder, PPy:CMC composite was found to have strong interactions with LiCoO2 (LCO) and LiNi1/3Mn1/3Co1/3O2 (NMC111) cathode active materials.
Strong interactions are also expected between PPy:CMC composites and other electrode materials such as silicon/tin-based materials (Li-ion battery anodes), and sulfur (Li—S battery cathodes), to name a few. Such unique features could be beneficial for the development of other rechargeable batteries including Li-ion batteries, Li—S batteries and multivalent metal-ions batteries. Aqueous electrode processing was also achieved by using this class of electrode matrices, thus contributing to the development of a greener battery fabrication process.
According to an aspect of the invention, there is provided an electrode matrix comprising: an electrically conductive polymer; and a polyanionic binder.
According to another aspect of the invention, there is provided a method of activating an electrode matrix comprising: mixing an electrically conductive polymer, a polyanionic binder and an oxidant; fabricating an electrode matrix from the mixture of the electrically conductive polymer, the polyanionic binder and the oxidant; and subjecting the electrode matrix to a charging voltage at or above a typical upper cut off voltage for the electrode matrix until at least an expected electrode capacity is reached.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the invention belongs. Although any methods and materials similar or equivalent to those described herein can be used in the practice or testing of the present invention, the preferred methods and materials are now described. All publications mentioned hereunder are incorporated herein by reference.
Polypyrrole:carboxymethyl cellulose (PPy:CMC) composites were synthesized by in situ chemical oxidative polymerization. Several characterization techniques were used to understand the morphology, structure, and physical/chemical properties of PPy:CMC composites. Following that, carbon-additive-free LiCoO2/PPy:CMC cathodes were fabricated by using water as a processing solvent. Carbon-additive-free cathodes were then cycled to study the performance of PPy:CMC electrode matrices.
Furthermore, to facilitate the adoption of CP-based electrode matrices in Li-ion batteries, their compatibilities with commercially relevant LiNixMnyCo1−x−yO2 (NMC) cathode materials were investigated. Herein, LiNi0.33Mn0.33Co0.33O2 (NMC111) was synthesized by the sol-gel method. Following that, for the first time, carbon-additive-free NMC cathodes were made available by using PPy:CMC electrode matrices. Regardless of eliminating carbon conductive additives from electrode composition, the NMC111/PPy:CMC cathode is able to operate at high C-rate, confirming the capability of PPy:CMC composites to provide enough electrical conductivity for battery electrodes.
As will be known by those of skill in the art, prior art LiCoO2/PPy:CMC cathodes suffer from capacity fading. It is important to investigate the degradation mechanisms to understand the compatibility of PPy:CMC composites in Li-ion batteries, which could also proliferate the use of other CP-based battery electrode matrices. The causes of capacity fading were investigated by means of electrochemical and post-mortem chemical analyses. The result indicate that PPy:CMC composites were electrochemically stable within the cathode operating voltage window. As the cycle number increased, electrolyte anions became dopants for PPy units in PPy:CMC composites. The sharp spike in cell voltage of LiCoO2/PPy:CMC cathodes in the first charging cycle indicated that undoped/neutral PPy units in PPy:CMC composite were oxidized and doped to become fully conductive. This unique phenomena teaches an activation procedure for using other CP-based electrode matrices in Li-ion batteries such as polyaniline:carboxymethyl cellulose (PANI:CMC) composites, as discussed herein.
According to an aspect of the invention, there is provided an electrode matrix comprising: an electrically conductive polymer; and a polyanionic binder.
The designed electrode matrices can be used for all types of cathode active materials such as for example but by no means limited to lithium transition metal oxides LiMO2 (M are transition metals such as Co, Ni, Mn:LiMnO2, LiNiO2, LiNixMn1−xO2, LiNiO2, LiNixMnyCo1−x−yO2, LiNixCoyAl1−x−yO2, . . . ), Li-rich xLi2MnO3·(1−x)LiMO2 (M are transition metals). Based on our stability measurements, the electrode matrix will also work at the anode with graphite. We believe that the electrode matrix can be used beyond intercalation materials as well, for example with alloying materials (tin, silicon, aluminum) and conversion materials (sulfur, oxygen).
In some embodiments, the electronically conductive polymer is selected from the group consisting of: polyacetylene, polyphenylene sulphide, polyphenylene vinylene, polyisothianaphthene, polythiophene, poly(3-alkylthiophene), poly(3,4-ethylenedioxythiophene), polyaniline and polypyrrole.
In some embodiments, the electronically conductive polymer is selected from the group consisting of: polythiophene, poly(3-alkylthiophene), poly(3,4-ethylenedioxythiophene), polyaniline and polypyrrole.
In some embodiments, the electronically conductive polymer is polyaniline or polypyrrole.
In some embodiments, the polyanionic binder is selected from the group consisting of: polystyrene sulfonate, sodium carboxymethyl cellulose, sodium polyacrylate, and sodium alginate.
In some embodiments, the polyanionic binder is selected from the group consisting of: sodium carboxymethyl cellulose, sodium polyacrylate, and sodium alginate.
In some embodiments, the electrically conductive polymer and the polyanionic binder are present at 30-70% electrically conductive polymer and 30-70% polyanionic binder.
In some embodiments, the electrically conductive polymer and the polyanionic binder are present at 5-95% electrically conductive polymer and 5-95% polyanionic binder.
In some embodiments, the electrically conductive polymer and the polyanionic binder are present at 40-60% electrically conductive polymer and 40-60% polyanionic binder.
As will be known by those of skill in the art, the design concept of electrode matrices from an in situ combination of electrically conductive polymer and polyanionic binder is new. No other studies claimed to make electrode matrices from the above-mentioned components. Instead, they add the two components solely as electrode additives in their composite forms. This work introduces a way to form a molecular composite, where the electrically conductive polymer is synthesized in the presence (=in situ) of the polyanion, as discussed herein. PEDOT:PSS composite is the closest relative the PPy:CMC composite, which is the representative of the new electrode matrices introduced. While PEDOT is well-known conductive polymer, PSS has not been used as polyanionic binder for battery electrodes.
This work is inspired by PEDOT:PSS structure, but the design concept is driven by the use of cheap conductive polymer monomers and relevant carboxylate-containing polyanionic binders such as carboxymethyl cellulose.
Specifically, previous studies did not synthesize molecular composites by polymerizing conducting polymer monomers in the solution of a polyanionic binder. Instead, those studies mixed pre-formed conducting polymers and polyanionic binders. As a result, no doping of conducting polymer by polyanionic binders form at a molecular level.
According to another aspect of the invention, there is provided a method of activating an electrode matrix comprising:
Specifically, in the example provided herein, the battery is prepared as usual. During the first charging cycle, the charging is cut off at the typical potential of around 4.2V only after a minimum amount of charged has passed. The battery voltage will initially rise and pass an activation voltage that is typically around 4.5V. This high-voltage step will normally last from a few seconds to 10 minutes. This activation could also be achieved by holding the charging voltage at approximately 4.5V for about 10 minutes or any other suitable method for applying a high potential to battery cells for short periods of time.
As will be appreciated by one of skill in the art and as discussed herein, surprisingly, by forming molecular composites as described herein and then following with the appropriate activation step(s) as discussed herein, the composites become sufficiently conductive that electrode formulations without carbon additives can be used.
As discussed herein, some conducting polymer units are not sufficiently oxidized and doped during synthesis as the oxidation is limited by the rate of oxidation and the oxidizing potential of the oxidant used. However, as discussed herein, the activation increases the oxidation potential beyond that of oxidant.
While not wishing to be bound to a particular theory or hypothesis, adding pre-formed conducting polymers to the mix might not work because of the low dispersibility of most conducting polymers. Specifically, increasing the content of conducting polymers in the conducting polymer composite composition would improve the electrical conductivity of conducting polymer composite; however, this will also reduce the water dispersibility and adhesion of the electrode matrix. Increasing the amount of polyanionic binders, on the other hand, improves adhesion and water-dispersibility, but decreases the amount of charge-storing materials, which in turn lowers energy per volume and mass of the electrode.
For example, as discussed herein, in one embodiment of the invention, the in-situ chemically oxidative polymerization is performed in an ice-bath to slow down the growth of the polymer chain, allowing polyanionic binder to dope along the backbone of conducting polymers and increasing PPy chain length/conductivity.
As known by those of skill in the art, carbon additives are electrically conductive. Without the presence of carbon additives in battery electrodes, the electrical connection within battery electrodes relies on the conductivity of the conducting polymer composites. Conducting polymers need to be activated to provide sufficient electrical conductivity for the battery to operate. With carbon additives present, the activation would work differently—for example, the potential may not rise at the beginning of the charging process to the high potential, but rather may rise at the very end. However, with carbon additives present, it is not clear whether an activation would be necessary at all, since carbon would likely provide sufficient conductivity.
Furthermore, adding carbon additives and conventional binders to the carbon-additive-free electrodes composed of conducting polymer composites and active materials would not make any significant difference. Specifically, while it was believed that carbon additives must be used to make operational battery electrodes, this study confirms that conducting polymer composites could provide sufficient electrical conductivity for a battery electrode.
Accordingly, in some embodiments of the invention, there is the proviso that the mixture consists essentially of an electrically conductive polymer, a polyanionic binder and an oxidant.
In some embodiments of the invention, there is provided the proviso that the mixture and the electrode matrix are substantially free of carbon additives or are free of carbon additives.
In some embodiments of the invention, the charging voltage is above the typical cut off voltage for at least a first 10% of charging.
In some embodiments of the invention, the electrode matrix is subjected to the charging voltage above the typical cut off voltage and then subjected to a standard first charge cycle.
In some embodiments of the invention, the charging voltage is held at the upper cut off voltage at the end of a first charge until theoretical electrode capacity is reached.
In some embodiments of the invention, the electrode matrix is subjected first to a minimum amount of charge at the typical upper cut off voltage and then subjected to a charging voltage above the typical cut off voltage for the electrode matrix until theoretical electrode capacity is reached.
As will be appreciated by one of skill in the art, versions and/or variations of the above-described activation protocol can be applied to any conducting polymer composites that are used in battery electrodes. By using the activation protocol, conducting polymers in conducting polymer composites could be oxidized and then doped by available anions in battery electrolyte. However, in electrode matrixes without a conducting polymer in the electrode, this activation protocol would not have the beneficial effect and may in fact cause some degradation of the electrode or electrolyte.
In some embodiments of the invention, the electronically conductive polymer is selected from the group consisting of: polyacetylene, polyphenylene sulphide, polyphenylene vinylene, polyisothianaphthene, polythiophene, poly(3-alkylthiophene), poly(3,4-ethylenedioxythiophene), polyaniline and polypyrrole.
In some embodiments of the invention, the electronically conductive polymer is selected from the group consisting of: polythiophene, poly(3-alkylthiophene), poly(3,4-ethylenedioxythiophene), polyaniline and polypyrrole.
In some embodiments of the invention, the electronically conductive polymer is polyaniline or polypyrrole.
In some embodiments of the invention, the polyanionic binder is selected from the group consisting of: polystyrene sulfonate, sodium carboxymethyl cellulose, sodium polyacrylate, and sodium alginate.
In some embodiments of the invention, the polyanionic binder is selected from the group consisting of: sodium carboxymethyl cellulose, sodium polyacrylate, and sodium alginate.
In some embodiments of the invention, the electrically conductive polymer and the polyanionic binder are mixed at 5-95% electrically conductive polymer and 5-95% polyanionic binder.
In some embodiments of the invention, the electrically conductive polymer and the polyanionic binder are mixed at 30-70% electrically conductive polymer and 30-70% polyanionic binder.
In some embodiments of the invention, the electrically conductive polymer and the polyanionic binder are mixed at 40-60% electrically conductive polymer and 40-60% polyanionic binder.
In some embodiments of the invention, the oxidant is selected from the group consisting of: chromic acid, perchloride acid, hydrogen peroxide, dibenzoyl peroxide, ammonium perchlorate, ferric chloride and ammonium persulfate.
In some embodiments of the invention, the oxidant is selected from the group consisting of: ammonium perchlorate, ferric chloride and ammonium persulfate.
In some embodiments of the invention, the oxidant is ferric chloride or ammonium persulfate.
The invention will now be further described and/or elucidated by way of examples; however, the invention is not necessarily limited to or by the examples.
For the synthesis of polypyrrole:carboxymethyl cellulose (PPy:CMC) composites, pyrrole (Sigma-Aldrich, 98%), FeCl3 (Fisher, 98%), sodium carboxymethyl cellulose (Na-CMC) (Sigma-Aldrich, MW=250000 g/mol, degree of substitution 0.9), and ethanol (Fisher, 98%) were used as purchased without further purification. For electrode fabrication, LiCoO2 (Sigma-Aldrich, 99.8%), PVDF (Sigma-Aldrich, 99%, MW=534 000 g/mol), C-black (Cabot, black pearls 2000), and N-Methyl-2-pyrrolidone solvent (NMP) (Sigma-Aldrich, 99%) were used. Reverse osmosis (RO) water was used throughout the experiment.
PPy:CMC composites were synthesized via in situ polymerization. Firstly, Na-CMC was completely dissolved in water. After that, 400 μl (˜5.8 mmol) pyrrole was added to the viscous Na-CMC solution. The mixed precursor solution was then placed in an ice bath. The mass ratio between pyrrole and CMC was varied as follow: 1:0 (0 wt % CMC), 1:0.25 (˜25 wt % CMC), 1:0.5 (˜33.33 wt % CMC), 1:0.75 (˜42.85 wt % CMC), 1:1 (˜50 wt % CMC), and 1:1.25 (˜55.5 wt % CMC), which were denoted as PPy, PPy:CMC 1:0.25, PPy:CMC 1:0.5, PPy:CMC 1:0.75, PPy:CMC 1:1 and PPy:CMC 1:1.25, respectively. FeCl3 was dissolved in water and added dropwise into the above precursor solution. The molar ratio between pyrrole and FeCl3 was initially fixed at 1:2.5 (denoted as R2.5) and then increased to 1:2.75 and 1:3.0, which were denoted as R2.75 and R3.0, respectively. The concentration of pyrrole was 0.6 mol L−1. The polymerization reaction was carried out for 4 hours in an ice bath. The product suspensions were immersed in ethanol overnight with the suspension/ethanol volume ratio of 1:4 to induce the precipitation of PPy:CMC composites. The precipitates were filtered by vacuum filtration and washed with ethanol until a colorless filtrate was observed. The products were then dried at 80° C. under vacuum for two days.
To characterize the structure of conducting polymer composites, transmission electron microscopy coupled with energy-dispersive X-ray spectroscopy (TEM/EDX), scanning electron microscopy (SEM), X-ray photoelectron spectroscopy (XPS) and scanning transmission X-ray microscopy (STXM) were used. PPy:CMC samples were dispersed in isopropanol and drop coated on carbon-film-coated copper grids for TEM and EDX measurements on the FEI Talos F200X microscope at an accelerating voltage of 80 keV. SEM imaging was performed on an FEI Nova NanoSEM 450 microscope. SEM imaging for electrodes was performed by imaging 2×2 mm electrode pieces coated on Al substrate. XPS measurements were performed on a Kratos Axis Ultra spectrometer at a pass energy of 160 eV for survey scans and 20 eV for N 1s, C 1s, O 1s narrow scans. Charge neutralization of 2.5 eV was applied for each measurement. STXM imaging on PPy:CMC 1:1 composite was performed at the 10ID-1 SM beamline of the Canadian Light Source (CLS).
To measure the electrical conductivity of PPy:CMC composites, samples were compressed into PPy:CMC pellets and measured employing the four-point probe method on a Miller Design FPP-5000 instrument. The 0.6 mm-thick pellets were prepared by grinding 100 mg of PPy:CMC sample and compressing it at 200 MPa in a hydraulic press. In order to evaluate electrode cohesion and adhesion, scratch testing was performed on 15 mm diameter electrode pieces, which contained 20-25 μm-thick electrode coatings on a 25 μm-thick Al substrate. Scratch tests were performed with assistance from Anton-Paar on an Anton Paar MST3 Micro Scratch Tester with feed-back loop. A Rockwell diamond tip with a tip radius of 20 μm was scanned across an immobilized electrode piece at a constant speed of 3 mm min−1 and a loading rate of about 45 N min−1. Two failure events were recorded. An initial detachment was marked at the smallest force at which a first penetration to the Al sublayer is observed, and full delamination was marked by the continuous delamination of the electrode film from the Al substrate.
LiCoO2/PPy:CMC electrode slurries were prepared by ball-milling LiCoO2 and PPy:CMC composites in water with a solid content of ˜30 wt %. The mass ratio between LiCoO2 and PPy:CMC composites was fixed at 90:10 (wt %:wt %). The electrode slurries were cast on 25 μm thick aluminum (Al) foil (Goodfellow, USA) using a doctor blade. The electrode thickness and mass loading were approximately 25 μm and 3 mg cm-2 respectively. After drying in a vacuum oven for two days at 80° C., electrode sheets were cut into 15 mm disks. Standard LiCoO2/PVDF/C electrodes were prepared by a similar procedure in NMP with 5 wt % PVDF and 5 wt % carbon black unless otherwise specified. In order to evaluate electrode performance, half cells of R2032-format (MTI, USA) were assembled in an argon-filled glove box with oxygen and moisture level below 0.1 ppm. LiCoO2/PPy:CMC or LiCoO2/PVDF/C electrodes were used as cathodes. LiPF6 in DMC/EC (50:50 v/v. Sigma-Aldrich) was used as the electrolyte. 15 mm lithium anode disks were cut from a lithium ribbon (Alfar Aesar, 99.9%, 0.75 mm thick). Whatman™ glass fiber (Fisher) was used as the separator. Coin cells were galvanostatically cycled with cut-off voltages of 2.8 V and 4.2 V vs. Li/Li+ on Neware Battery Cyclers. A current density of 274 mA g−1 and 27.4 mA g−1 was applied to cycle coin cells as denoted as 1 C and 0.1 C-rate, respectively.
To confirm the successful synthesis of PPy:CMC composites, the chemical structure and nanoscopic morphology of the prepared composites were investigated. PPy:CMC composites were synthesized in their oxidized state by chemical oxidative polymerization of pyrrole in aqueous CMC solution. In the oxidized state, an anionic dopant is necessary to charge-balance positive charges on PPy. In the synthesis solution, those negative charges are mainly carried by carboxylate groups on CMC, which encourages the formation of a molecular composite between PPy and CMC, as presented in
A comparison between in situ polymerized and mechanically mixed PPy:CMC 1:1 R.2.5 composites morphology was carried out. In contrast to the in situ polymerized sample, PPy in the mechanically mixed composite is already oxidized and doped when it is mixed with CMC. Consequently, such a composite would exhibit a different microstructure and potentially separate phases.
TEM reveals a nano-morphology of 50.3±2.1 nm spheres in the PPy:CMC 1:1 R.2.5.
To verify the chemical structure of PPy:CMC 1:1 composites, X-ray absorption and photoelectron measurements were taken. The spatially-resolved STXM image reveals a complex structural arrangement of components in PPy:CMC 1:1 R.2.5 composite (
Similar structural complexity is observed in XPS measurements (
Together, chemical and microstructural analysis of the composites demonstrate that PPy and CMC are co-located and well distributed at the nanoscale. The obtained degree of oxidation of PPy is sufficient for significant electron conduction[44]. While these methods are not able to determine whether CMC acts as a dopant in PPy, the observations confirm the successful synthesis of a composite of PPy and CMC with homogenous distribution at the nanoscale.
To act as the sole electrode matrix, PPy:CMC composites must have sufficient electrical conductivity to support electron conduction during charging and discharging. In order to achieve a balanced electrode charging throughout the thickness of the electrode, ionic and electronic conductivity should be similar. Assuming a typical electrode with approximately 50% porosity, for standard carbonate electrolytes, at an electrode matrix weight fraction of 10% and ideal application of the Bruggeman relation, a minimum composite matrix conductivity of approximately 20 mS cm−1 is required. A significantly larger conductivity than this benchmark can be achieved by PPy alone (
According to the effective medium approximation (EMA) theory, the electrical conductivity of composites, composed of spherical insulating particles embedded in a conductive matrix, would follow the Bruggeman relationship. Similar behaviors have been found for randomly distributed spherical conductive and non-conductive particles[45-47]. A simplified Bruggeman equation is described as:
In which, σeff, σo, ε, and α are effective conductivity, intrinsic conductivity, conductive volume fraction, and Bruggeman exponent, respectively. The expected Bruggeman exponent for a random mixture of near-spherical PPy and CMC particles is close to 1.5. Yet, a fit of the data reveals an exponent of 13.08 (
The matrix should not only be conductive but be easily processed, ideally through tape casting. Aqueous processing requires a stable dispersion of the matrix in water. While a PPy suspension settles quickly, PPy:CMC dispersions are stable for an hour (
To evaluate the performance of PPy:CMC composites as electrode matrices, electrodes with different PPy:CMC or PVDF/C matrices were subjected to galvanostatic cycling. LiCoO2/(PPy:CMC 1:1 R2.75) electrodes can be cycled at 27.4 mA g−1 (0.1 C) and 274 mA g−1 (1 C) with similar initial capacity as the benchmark LiCoO2/PVDF/C electrodes (
It should be noted that lower electrode performance is observed with the typically suggested Py:Fe ratio of 2.5 (denoted as R2.5) during the polymerization[36,51,52], but an increased equivalent of FeCl3 was required of 2.75 (denoted as R2.75). This is likely due to a small amount of ferric chloride being captured by CMC[53], as observed in the XPS results. As a result, PPy:CMC 1:1 R2.75 composite exhibited higher electrical conductivity than PPy:CMC 1:1 R2.5 composite (
While the initial cycling performance of PPy:CMC electrodes is similar to standard PVDF/C electrodes, capacity fading is observed almost immediately with the PPy:CMC matrices. This fade is correlated with a swift increase in internal resistance. There are several possible reasons for this performance decay. While plenty of publications show long performance stability of conducting polymer-containing electrodes[54], there is little targeted research on the compatibility and stability of conducting polymers in Li-ion batteries[55]. It is possible that conducting polymer performance decay is underestimated in many studies, where carbon additives and conducting polymers are used together, due to the reduced impact that conducting polymer conductivity has on electrode performance. In targeting the full replacement of carbon additives to avoid agglomeration, long-term conducting polymer performance becomes critical and is now the subject of ongoing work in our group. LiCoO2/(PPy:CMC) cells also exhibited an initial spike in cell voltage followed by a descent in the first charging step. A number of processes could lead to this behavior including slow electrolyte infiltration or doping of PPy by the electrolyte.
This work demonstrates the design concept of water-dispersible, self-conductive electrode matrices from conducting polymer composites synthesized via a simple and scalable in situ polymerization. Multifunctional PPy:CMC composites, that exhibit adhesion, electronic conductivity, and contribute to charge storage are a promising replacement for PVDF and carbon additives in electrodes to reduce carbon agglomeration and achieve a conductive electrode network that actively adheres to intercalation materials during cycling. We demonstrated that a LiCoO2/PPy:CMC electrode can be cycled at a high current density of 274 mA·g−1 without carbon additives, which is a unique achievement within carbon-free Lithium intercalation cathodes. At the same time, the conductive network within the electrode appears to deteriorate during cycling, which is the subject of ongoing work.
Notwithstanding the stability limitations, this study highlights the performance potential of battery electrodes containing conducting polymer composite matrices. Those matrices are produced from low-cost, high-volume raw materials, are easily synthesized and processed in low-cost, low-impact aqueous solutions. As new electrode materials are developed with ever-increasing energy densities, the potential for larger volume change increases as well. This will require a solution to the intrinsic lack of adhesion between carbonaceous conductors and active materials, of which conducting polymer composites are shown here to be a promising candidate.
In brief, pyrrole (Sigma-Aldrich, 98%), was chemically polymerized by FeCl3 oxidant (Fisher, 98%) in the aqueous solution of Na-CMC (Sigma-Aldrich, MW=˜250.000 g/mol, degree of substitution 0.9) as described in the EXAMPLE 1. In EXAMPLE 2, pyrrole:Na-CMC mass ratio and pyrrole:FeCl3 molar ratio were fixed at 1:1 and 1:2.75, respectively. To simplify the notation, the PPy:CMC11-R275 composite as denoted in the EXAMPLE 1 will be denoted simply as PPy:CMC composite. After synthesis, PPy:CMC composite was purified and dried as described detail in EXAMPLE 1.
Synthesis of LiNi0.33Mn0.33Co0.33O2 (NMC111)
NMC111 was synthesized by a sol-gel method that followed the synthesis procedure reported by previous work[56]. CH3COOLi·2H2O (Sigma-Alrich, 98%), (CH3COO)2Ni·4H2O (Sigma-Alrich, 98%), (CH3COO)2Mn·4H2O (Sigma-Alrich, 98%), (CH3COO)2Co·4H2O (Sigma-Aldrich, 98%) and citric acid (Sigma-Aldrich, 98%) were used as received without any purification. Firstly, acetate salts were dissolved together in DI water with the Li:Ni:Mn:Co ratio of 1.1:0.33:0.33:0.33. Citric acid (Sigma-Aldrich, 98%) was dissolved completely in DI water, mixed with ethylene glycol (Sigma-Aldrich, 99.8%), and then slowly added to the NMC precursor solution. The molar ratio between citric acid and total metal ions was 1:1. The solution was heated to 70° C. overnight to yield a pink gel that was then kept at 70° C. for 2 days. The dried gel was ground and transferred to a ceramic crucible before putting in a furnace. The sample was pre-calcinated at 450° C. for 2 hours to burn out organic components. The powder was ground and compressed into pellets at 25 MPa. The calcination was performed at 900° C. for 12 hours in air. A heating rate of 2° C.·min−1 was used throughout the experiment. The final product was ground by mortar and pestle into fine powders and dried in a vacuum oven before use.
NMC111/PPy:CMC cathode was prepared by mixing NMC111 powder with PPy:CMC composite (PPy:CMC 1:1 R2.75) at a mass ratio of 90:10 wt %. The solid mixture was dispersed in water with a solid content of 30 wt % and then stirred vigorously on a magnetic stirrer for 6 hours. A homogenous electrode slurry was cast on aluminum to yield ˜25 μm-thick electrode sheets on 25 μm-thick Al foil (Goodfellow, USA) by using a doctor blade. For comparison, NMC111/PVDF/C (90:5:5 wt %) reference cathode was prepared in N-Methyl-2-Pyrrolidone (NMP) solvent by a similar procedure. Electrode mass loading was controlled at approximately 3 mg·cm−2. Electrode sheets were carefully dried in a vacuum oven at 80° C. for two days and cut into 15-mm in diameter electrode disks before use.
CR2032-type coin cells (MTI, USA) were used to fabricate half-cell Li-ion batteries for testing cathode performance. Coin cell fabrication was performed in an argon-filled glove box (Unilab Mbraun) with oxygen and moisture level below 0.1 ppm. Each coin cell contains 15 mm disc cathode, 20 mm disc glass fiber separator (Whatman™), 15 mm lithium disc (Alfar Aesar, 99.9%, 0.75 mm thick, LiPF6 in DMC/EC (50:50 v/v) electrolyte (Sigma-Aldrich).
The crystal structure of NMC111 cathode materials was confirmed by powder X-ray diffraction (XRD) measurement on the D4 Endeavor instrument with Cu Kα source at a working voltage of 40 kV. Data treatment was performed on QuaIX software with Crystallography Open Database[57]. The morphologies of NMC111 were studied by SEM (FEI Nova NanoSEM 450) and TEM/EDX (FEI Talos F200X S/TEM).
Coin cells were galvanostatically charged and discharged within the fixed potential window of 3.0 V-4.3 V Li/Li+ on Neware battery testers. The applied current density of 275 mA·g−1, 27.5 mA·g−1, and 13.75 mA·g−1 were denoted as 1 C, 0.1 C, and 0.05 C-rate, respectively.
After cycling for 100 cycles at 1 C, a coin cell was opened in an argon-filled glovebox. Cathode material was disassembled and then washed in propylene carbonate (PC) solvent. The washed cathode was dried in a vacuum oven at 85° C. for 2 days. The sample was then stored in the argon-filled glovebox prior to mount on the sample holder for SEM measurement.
The structure of lab-synthesized LiNi0.33Mn0.33Co0.33O2 (NMC111) cathode materials was confirmed by powder X-ray diffraction.
A comparison between the morphologies of NMC111-based electrodes with PVDF/C and PPy:CMC electrode matrices was demonstrated in
The NMC111/PVDF/C reference cathode showed great capacity retention, maintaining a specific discharge capacity of approximately 135 mAh·g−1 after 100 cycles at 0.1 C (27.5 mA·g−1). In comparison to the NMC111/PVDF/C reference cathode, NMC111/PPy:CMC cathode demonstrated a higher initial discharge capacity of 150 mAh·g−1 (
Nonetheless, carbon-additive-free NMC111/PPy:CMC cathode was able to operate at 1 C with great capacity retention. After cycling for 100 cycles at 1 C, they still delivered ˜90 mAh·g−1 (
To investigate changes in the morphology of NMC/PPy:CMC cathode upon repeated charge/discharge cycling at 1 C for 100 cycles, the NMC/PPy:CMC coin cell was disassembled. The cycled NMC/PPy:CMC cathode was characterized by SEM.
The sudden capacity fading of NMC111/PPy:CMC cathode after operating for more than 20 cycles at 0.1 C could have resulted from unknown side reactions. However, the cell that cycled at 1 C demonstrated relatively stable performance. It is worth noting that cycling at 1 C is theoretically 10 times faster than cycling at 0.1 C. For that reason, the sudden cell failure after cycling for 20 cycles at 0.1 C might not be observed after cycling for 100 cycles at 1 C. Unlike the cycle life, which is closely related to the performance degradation due to electrochemical reactions during the repeated charge/discharge cycles, the calendar life depends on the operating time of battery cells rather than the cycle number. The discrepancy in the performance of NMC111/PPy:CMC cathode cycled at 1 C and 0.1 C rate could be attributed to their limited calendar life, indicating some unwanted chemical reactions occurred between battery components.
The study has shown that PPy:CMC composites are versatile for usage as electrode matrices for different types of intercalation materials ranging from traditional LiCoO2 to NMC111. PPy:CMC composite plays dual roles as electrode binder and conductor. Despite having very low intrinsic electrical conductivity as mentioned in EXAMPLE 1, PPy:CMC composite was capable of providing a sufficient conductive matrix for NMC111 particles thanks to their strong adhesion with NMC111 particles. As a result, carbon-additive-free NMC111/PPy:CMC cathode could cycle at a high C-rate. The result suggests that the electrical conductivity of battery electrodes depends not only on the intrinsic electrical conductivity of conductive agents but also on how conductive agents interact with active materials.
PPy:CMC composites were synthesized by chemically in-situ polymerizing pyrrole in aqueous sodium carboxymethyl cellulose solution with FeCl3 as oxidant as reported in the EXAMPLE 1. The PPy:CMC composite that was synthesized with Py:Na-CMC 1:1 mass ratio and Py:FeCl3 1:2.75 molar ratio yielded the best electrode performance. Therefore, PPy:CMC11 R2.75 composite was chosen to study in this example. The PPy:CMC 1:1 R2.75 composite reported in the EXAMPLE 1 was now denoted as PPy:CMC composite for simplicity.
Carbon-additive-free LiCoO2/PPy:CMC cathodes (90:10 wt %) were prepared by ball-milling a mixture of LiCoO2 and PPy:CMC composite in water. The electrode slurry was then cast on a hard temper Al current collector[59]. Reference LiCoO2/PVDF/C (90:5:5 wt %) cathodes were prepared by a similar procedure in NMP solvent. PPy:CMC-only cathodes were made by compressing 25 mg of PPy:CMC composite into ˜0.1 mm-thick, 13 mm diameter pellets. R2032 coin cell fabrication was carried out in an argon-filled glovebox. LiCoO2/PPy:CMC, LiCoO2/PVDF/C, or PPy:CMC-pellet were used as cathodes. Lithium metal (Alfar Aesar, 99.9%, 0.75 mm thick) and glass-fiber (Whatman™) were used as anode, and separator, respectively. All coin cells used LiPF6 in DMC/EC (50:50 v/v) (Sigma-Aldrich) as liquid electrolyte unless otherwise specified.
LiCoO2/PPy:CMC coin cells underwent galvanostatic charge/discharge at 0.1 C (27.4 mA·g−1) for 1 cycle, 10 cycles, and 100 cycles within a voltage range of 2.8 V-4.2 V vs Li/Li+ on Neware battery testers. After undergoing certain cycling numbers, coin cells were disassembled in an argon-filled glovebox. Cathode materials, that were attached to Al foil, were immersed in propylene carbonate 3 times (5 mins/each) and then dried in a vacuum oven at 85° C. for 2 days. Samples were stored in the argon-filled glovebox prior to mount on the sample holder for SEM (FEI Nova NanoSEM 450) and XPS (Kratos Axis Nova spectrometer, Al X-ray source) measurements.
Galvanostatic electrochemical impedance spectroscopy (EIS) measurement was carried out on coin cells assembled with LiCoO2/PVDF/C reference cathode or LiCoO2/PPy:CMC cathode. Frequencies were scanned from 100 kHz to 10 mHz by Interface 1010E (Gamry Instrument). EIS equivalent circuit fitting was performed on Gamry Echem Analyst by the simplex method. Cyclic voltammetry (CV) was carried out within two potential ranges of 2.8-4.2 V vs Li/Li+ and 0-5 V vs Li/Li+ on CR2032 coin cells assembled with ˜0.1 mm thick PPy:CMC pellets as cathode materials.
According to the voltage profiles of LiCoO2/PVDF/C reference cathode in
The electrochemical behavior of PPy:CMC composite at the cathode working potentials was tested by performing cyclic voltammetry measurement on PPy:CMC-pellet coin cells. Within the cathode operating potential window, there was no redox reaction observed as shown in
The evolution of electrode impedance after each charge/discharge process would provide information about electrochemical processes that occurred in battery cells. In contrast to the coin cell assembled with LiCoO2/PVDF/C cathode, interestingly, the as-prepared coin cell assembled with LiCoO2/PPy:CMC cathode did not yield any charge and discharge capacities during the first charge/discharge coupled with EIS measurement. As shown in
At the fully charged state, there were two clear depressed semi-circles at high and medium frequencies corresponded to two Randles circuits at anode and cathode, respectively. The impedance evolution routes for the two coin cells were different. As the cycle number increased, the anode charge transfer resistance (Rct) of the reference LiCoO2/PVDF/C coin cell increased relatively fast (
In the fully discharged state, It was clear that LiCoO2/PVDF/C reference cathode showed higher cathode impedance than LiCoO2/PPy:CMC cathode. As discussed in the EXAMPLE 1, the intrinsic electrical conductivity of as-prepared PPy:CMC composite was 10 000 times lower than that of carbon black additives. However, PPy:CMC composite seemed to offer better electrical conductivity for LiCoO2 electrode than PVDF/C electrode matrix. There are several possible explanations. Firstly, undoped PPy molecules in PPy:CMC composite were likely to be oxidized and doped by polyanions and anions from battery electrolyte during the first charging process of battery cells. As a result, the actual electrical conductivity of PPy:CMC composite in Li-ion battery cells was higher than that of their pristine state. Secondly, PPy:CMC composite adhered strongly on the surface of LiCoO2 particles, thus providing good electrical conduction from particle to particle. In contrast, most carbon additives do not adhere to LiCoO2 particles and are trapped inside the non-conductive PVDF binder, which reduces the effective electrical conductivity of the PVDF/C electrode matrix.
After cycling for several charge/discharge cycles, coin cells were disassembled to characterize morphological and structural changes of LiCoO2/PPy:CMC cathode upon cycling. XPS survey spectra (
As the cycle number increased, the peak intensity of Cl 2p decreased while the peak intensity of F 1s increased simultaneously. According to the previous study on PPy:CMC composite, positively charged PPy was doped by carboxyl groups (from CMC structure) and residual chloride ions (from residual FeCl3 oxidant)[59]. Initially, chloride ions were key dopants for PPy, which explained the high peak intensity of Cl 2p in the XPS spectrum of the electrode that cycled for 1 cycle. During the continuous charge/discharge process, PF6− ions were likely to substitute Cl ions as anionic dopants for PPy due to the excess use of LiPF6 electrolyte. Free chloride ions were easily washed out during the electrode washing prior to XPS measurement. In contrast, doping Cl− and PF6− ions were trapped within PPy:CMC electrode matrix. As a result, the XPS spectra for the cathode that cycled 100 times exhibited strong peak intensity for F 1s (
After cycling for 1 cycle, the Co 2p XPS spectrum of LiCoO2/PPy:CMC cathode was identical to pristine LiCoO2 as reported in the previous studies[66]. However, there was a significant change in the Co 2p XPS spectrum of LiCoO2/PPy:CMC cathode following the 100th cycle. Such changes in the specification of Co 2p peaks indicate changes in the surface structure of LiCoO2. The question remained whether PPy:CMC composite contribute to the degradation of LiCoO2 because the magnitude of capacity fading of LiCoO2/PPy:CMC cathode and LiCoO2/PVDF/C reference cathode were similar.
At the 100th cycle, two C 1s peaks appeared at approximately 287 eV and 289 eV that could be attributed to ester and ether groups of CMC (
The morphologies of LiCoO2/PPy:CMC cathode became rougher as the cycle number raised (
The performance degradation of Li-ion battery cells could be attributed to many factors ranging from the intrinsic properties of electrode materials, to impurities in the electrolyte and other electrode components. It is worth noting that the magnitude of capacity fading was the same for both LiCoO2/PVDF/C reference cathode and LiCoO2/PPy:CMC cathode. The capacity fading mechanism might originate from the intrinsic degradation of LiCoO2 in the investigated system. For example, some electrolyte additives were reported to suppress the degradation of LiCoO2 in carbonate electrolytes. This study, however, used commercial 1 M LiPF6 DMC:EC (50:50 v:v) electrolyte without electrolyte additives, which might contribute to the degradation observed. One possible cause for the degradation of LiCoO2/PPy:CMC cathode could be impurities in liquid electrolytes used. However, as shown in EXAMPLE 2, the NMC111/PVDF/C cathode exhibited excellent capacity retention regardless of using the same bottle of LiFP6-based liquid electrolyte. Therefore, electrolyte impurities would not be the reason for the capacity fading of these LiCoO2 cathodes.
Residual chloride ions were reported in the composition of PPy:CMC composites that were previously purified by vacuum filtration. However, the chloride contamination would not affect electrochemical stability of PPy:CMC composites as explained in the cyclic voltammograms. The impact of residual chloride ions on the performance of LiCoO2/PPy:CMC cathodes was not well-understood. In order to get rid of residual chloride ions in PPy:CMC composites, centrifugation was used to purify PPy:CMC composites.
As for LiPF6 electrolyte, it is well-known that traces of moisture in electrode could react with LiPF6, forming corrosive HF gas that attacks LiCoO2 and degrades overall cell performance. Therefore, one could argue that the hygroscopic nature of the CMC component in PPy:CMC composite might lead to moisture absorption during the handling of LiCoO2/PPy:CMC electrodes. To reject this hypothesis, several coin cells with LiCoO2/PPy:CMC cathodes were made with 1M LiClO4 in polypropylene carbonate electrolyte, which is insusceptible to moisture contamination. However, the electrode still suffered from capacity fading as shown in
To sum up, the degradation of the LiCoO2/PPy:CMC cathode was likely due to the intrinsic problem of LiCoO2 active materials. Nevertheless, more studies should have been done to fully understand the properties of CP-based electrode matrices such as PPy:CMC composites in the working environment of Li-ion batteries.
It is imperative to emphasize the importance of activating CPs in Li-ion batteries by setting a high charging potential limit in the first charging stage. This unique protocol of activating CPs in Li-ion batteries has not been reported elsewhere. Without activation, the electrical conductivity of CPs would be too low to conduct electrons in carbon-additive-free electrodes. As a result, most of the studies still added carbon additives to CP-containing electrodes. This intriguing activation phenomenon suggests a good strategy to activate CPs in Li-ion batteries. For example, PANI:CMC composite was able to function carbon-additive-free LiCoO2/PANI:CMC cathode as depicted in
The study proved that PPy:CMC composite demonstrated good electrochemical stability within the potential range of cathode. Having a high number of carboxyl and hydroxyl groups, CMC offered a great affinity towards the surface of LiCoO2, forming a good coverage on LiCoO2 particles. During the first charging step, undoped PPy in PPy:CMC composite was oxidized to become fully-charged, which explained the abnormally sharp increase in the voltage profile within few seconds of the charging process. When the amount of residual chloride ions in PPy:CMC composites decreases by more careful purification, the activation potential was observed to increase accordingly. Once the activation potential goes beyond the upper working potential range of cathode (4.2 V vs Li/Li+ for LiCoO2), setting a high potential limit (4.5 V vs Li/Li+ for LiCoO2) in the first charging step is necessary to allow neutral PPy unit to be oxidized and doped, thus activating PPy:CMC composites. The same activation protocol has been applied successfully to activate PANI:CMC composites as electrode matrices. Based on the XPS measurement, as the cycle number increased, PF6− ions continued to substitute carboxyl groups and residual chloride ions to become one of the main dopants for positively charged PPy molecules. The degradation of LiCoO2/PPy:CMC cathodes was likely to originate from the degradation of LiCoO2.
While the preferred embodiments of the invention have been described above, it will be recognized and understood that various modifications may be made therein, and the appended claims are intended to cover all such modifications which may fall within the spirit and scope of the invention.
The instant application claims the benefit of U.S. Provisional Patent Application Ser. No. 63/238,917, filed Aug. 31, 2021 and entitled “CONDUCTING POLYMER-BASED ELECTRODE MATRICES FOR LITHIUM-ION BATTERIES”, the entire contents of which are incorporated herein by reference for all purposes.
| Filing Document | Filing Date | Country | Kind |
|---|---|---|---|
| PCT/CA2022/051216 | 8/9/2022 | WO |
| Number | Date | Country | |
|---|---|---|---|
| 63238917 | Aug 2021 | US |
