Correlating the influence of porosity, tortuosity, and mass loading on the energy density of LiNi_0.6Mn_0.2Co_0.2O₂ cathodes under extreme fast charging (XFC) conditions

Highlights

• Correlating mass loading, porosity and charging protocols to energy density.

• Optimal case— 11.5 mg cm⁻² with 35% porosity and 5C (CCCV, 10 min).

• Lower energy density under XFC with increasing mass loading.

• Li-ion diffusion in the electrolyte is the limiting factor at higher loadings.

• Li metal is the rate limiting step in half cells under high current density.

Abstract

Extreme fast charging capabilities along with high energy density of Li-ion batteries are the key factors to increase the adoption of electric vehicles while eliminating the problem of range anxiety. The U.S Department of Energy has a goal of <12 min charging time with energy density of >200 Wh kg⁻¹. A combined improvement in the electrode architecture, electrolyte properties, and separator membrane is necessary to achieve this goal. Cells with thin electrodes are capable of extreme fast charging at the expense of low energy density and high cost. Electrode engineering can maximize energy density. Here, the influence of porosity, mass loading and charging protocols on capacity and energy density and electrode kinetics are investigated under extreme fast charging conditions. Increasing the mass loading from 11.5 mg cm⁻² to 25 mg cm⁻² compromises the rate performance due to the mass transport limitation and underutilization of thick electrodes. Reducing the electrode porosity from 50% to 35% improves the rate performance ascribed to shorter Li ion diffusion length. Symmetric cells are cycled to verify the performance of the half cells, suggesting that Li metal plating is the rate limiting step under high current density.

Introduction

The rapid development of Li-ion batteries (LIBs) and the significant drop in its cost by 80% has led to the emerging rise in the market of electric vehicles (EVs) [[1], [2], [3]] although the EVs still only account for ~1.7% of annual vehicle sales [4,5]. Range anxiety and longer charging time have been reported as the key factors for the wide scale adoption of EVs [6]. Presently, the charging time for battery electric vehicles is significantly longer than the refueling of the conventional internal combustion engine vehicles [3]. For the wide scale adoption of EVs to increase, it is necessary to reduce the charging time. The United States Department of Energy has defined a specific goal for extreme fast charging (XFC) of electric vehicles where the charging time should be ≤ 10 min with a gravimetric energy density of >200 Wh kg⁻¹ and volumetric energy density of >550 Wh L⁻¹ at cell level [[7], [8], [9]]. The gravimetric and volumetric energy density metric are critical based on different applications that require high/low power density [10]. For EV application, both energy densities are important due to the constraint in the weight and volume of a battery pack [11].

The main components of a LIB are anode, cathode, liquid electrolyte, and a porous separator membrane. During the charging process, lithium ions travel from the cathode to the anode via liquid electrolyte, which includes several diffusion steps 1) solid diffusion through active material particles, 2) diffusion across the electrode/electrolyte interface, and 3) through porous electrodes via electrolyte. Each step contributes to the barrier of enabling extreme fast charging. Cells with LiNi_xMn_yCo_zO₂ (NMC) and graphite chemistry are capable for fast charging but the electrodes need to be very thin [[12], [13], [14]]. This suggests that the rate limiting step in extreme fast charging of high-energy batteries with thick electrodes is the lithium ion diffusion in electrolyte which was due to electrolyte depletion and long diffusion length induced from the porous electrodes and separator [15]. Enhancement of electrolyte properties such as transference number and ionic conductivity could improve the fast charging capabilities [12,16,17].

Graphite is the state-of-the-art anode material in LIBs. However, it is susceptible to Li plating due to the proximity of the LiC_x potential to that of Li⁺/Li, which limits the charging current density and also results in capacity fade [18,19]. The plated metallic lithium can cause electrolyte decomposition and Li inventory loss from cathodes [9,20]. The plated lithium can further form dendrites and cause internal short circuits as well as form dead lithium on the anode, that could pose safety concerns for a LIB. Thus, extensive efforts have been made to improve the fast charging capability of graphite anode and alleviate Li plating via surface coating [21], utilizing graphene-like-graphite [22], and minimizing electrode tortuosity [23].

Common electrode materials, such as graphite and LiNi_xMn_yCo_1-x-yO₂, are capable of fast charging when the electrode thickness is very thin, which results in low energy density and high cost [24]. To increase electrode thickness and cell energy density, other manufacturing processes such as freeze casting [25,26] and magnetic templating [27] have been adapted to tailor the electrode architecture and enable fast charging, but they are not readily scaled up and yet economically viable. These efforts to achieve fast charging with high energy density have shown slightly detrimental influence on the cell cycle life [19] due to Li plating and increased temperature leading to accelerated solid electrolyte interface growth [[28], [29], [30]]. However, some studies also showed that high temperatures (45 °C) were useful in improving the cycle life for thick anodes [31,32] or while charging the cell at 2C rate [33].

While graphite anode is problematic in XFC, the cathode also deserves some attention especially if other anode materials with fast charging capability, such as lithium titanium oxide [34] and niobium titanium oxide [35], are utilized. Previous studies on cathodes were mainly focused on optimizing electrode formulation including conductive additive [11], surface modification of the active material particles [36], and optimizing the stoichiometry and crystallite size of cathode materials [37]. A few studies also evaluated the correlation of electrode design (areal loading and porosity), and energy and power density. Gallagher et al. and Appiah et al. have systematically investigated on optimizing the areal capacities, electrode thickness and porosity of LiNi_0.6Co_0.2Mn_0.2O₂ (NMC622) cathodes with help of modeling tools [18,38]. However, fewer insights were provided on the experimental aspect of varying the electrode architecture, and its influence on high rate capability [38,39]. A recent study by Huebner et al. focused on understanding the influence of electrode design (porosity and thickness) on electrochemical performance and lithium insertion kinetics of NMC622-based cathodes for high energy density cells at high rates. However, no work has considered the XFC [[40], [41], [42]].

This study reports a systematic investigation on optimizing NMC622 cathodes for XFC application. A matrix of 4 mass loadings and porosities were investigated. All electrodes were calendered to different electrode thicknesses to obtain the target porosity and mass loading. In addition, the influence of charging protocol on the energy density under XFC was also evaluated. An optimal NMC622 cathode was identified as 11.5 mg cm⁻² in mass loading, 35% in porosity and under 5C charging protocol.