Stable Charge/Discharge Cycle Performance of LiFePO4 Cathode Prepared with Carboxymethly Cellulose Binder

    Carboxymethyl cellulose (CMC) has recently attracted attention as a binder in lithium ion batteries (LIBs) due to its hydrophilic nature, which has led to its promising electrochemical performance in Si anodes. A water-soluble CMC-based binder has attracted significant interest as it is expected to reduce the cost of LIBs by replacing the conventional organic solvent-based binders. In recent years, the application of CMC has also been demonstrated in a cathode binder. Several groups have attempted to extend the practical applications of the CMC cellulose derivative to the cathode binder. However, the surfaces of an Al foil cathode current collector easily repel the water slurries containing the active cathode materials that include a conductive additive and the CMC binder. Thus, homogeneous, adhesive cathode films cannot form on Al current collector surfaces. Therefore, unlike conventional cathode binders, no stable cell performance could be observed. We altered the Al foil surfaces via chemical treatment, turning a water-slurry-passive surface into an adhesive surface, and used the chemically treated surfaces as the cathode current collector to investigate the effects of the treatment on the cell performance of CMC-based cathodes. The charge/discharge cycle performance of the cathode was compared with that of chemically and mechanically treated Al foils.

　The Al foils used as current collectors are 99.99% pure with a 100-μm thickness. The chemical treatment of Al foils was referred to in Japanese patents acquired by the Japan Surface Treatment Institute Co., Ltd. ^1,2 All slurries were prepared using Milli-Pore water (> 18 MΩ). The CMC1 (degree of etherification = 0.7, viscosity = 1000–2000 mPa·s at 1% AQ, 25 °C) binder purchased from Polyscience Inc., the CMC2 (CMC-2200, degree of etherification = 0.8–1.0, viscosity = 1500–2500 mPa·s at 1% AQ, 25 °C) binder received from Daicel, Japan and the polyvinylidene difluoride (PVdF, KF9130, Kureha, Japan) were used as received without any further treatment to prepare the LiFePO₄ cathode. To avoid agglomeration of the cathode material and binder in the cathode film, the mixture containing the LiFePO₄ active material and acetylene black was first dispersed in water; then, a CMC aqueous solution was added to this suspension, and the mixture was subsequently magnetically agitated for at least 4 h prior to producing the electrode film. The cathodes were prepared via doctor-blade (100-μm gap) coating. Electrochemical tests were performed using a CR2032 coin-type cell. The test cell was comprised of a cathode and a lithium metal anode separated by a porous polypropylene film (Celgard 3401). The electrolyte used in the tests contained a mixture of 1 M LiPF₆-ethylene carbonate (EC)/dimethylcarbonate (DMC) (1:2 v/v, Ube Chemicals, Japan).  All tests were performed at room temperature. A constant-current (CC) mode was applied to the pre-treatment process and cycle tests. 

    The cathodes prepared with the chemically treated Al foils exhibited a stable charge/discharge capacity for 100 cycles compared with the mechanically treated Al foils that exhibited poor stability in their charge/discharge capacity (Fig. １). Two of the CMC samples ((a) and (b)) did not exhibit a large difference in their cycle performance, indicating that the degree of etherification and viscosity of the CMC affects the cycle performance. The charge capacities observed during first several cycles were below the stable charge/discharge capacities after several cycles. The decrease in the initial charge capacity can be explained by the deintercalation of the Li⁺ ions from the LiFePO₄particles during their exposure to the aqueous solutions used to prepare the slurry containing the CMC binder.　The rate performance is important in the application of LiFePO₄ cathodes. All cells were charged at a current of 0.1 C to insure identical initial conditions for each discharge. The discharge currents were 0.1, 0.2, 0.4, 0.8, 1.6 and 3.2 C, indicating that the LiFePO₄cathodes prepared with CMC1 and CMC2 exhibit discharge capacities comparable to those obtained using a conventional PVdF binder even at high rate current. Therefore, the CMC binders do not exhibit and negative influence on the rate performance.