Silicon-dominant (Si) anodes with microscale silicon particles meet the requirements of high energy density and low costs due to its ten times higher theoretical electrochemical capacity of 3579 mAh/gSi (Li15Si4) compared to graphite,1 lifetime extension due to partial lithiation,1-2 and high abundance and economic availability due to existing industrial infrastructure.1-2 The major drawback of silicon is the large volume expansion of nearly 300% upon full lithiation. This hinders the broad application of silicon as an active anode material due to continuous SEI (re-) formation and electrochemical milling of particles. Additionally, the volume expansion impacts the overall electrode stability, leading to thickness changes, the disruption of the electronic pathways, and the decoupling of particles. Although the approach of partial lithiation of microscale1 Si eases these effects by lowering the overall volume expansion, these problems still exist.2,3 In this work, the effect of different mechanical cell pressures on lithium-ion batteries with partially lithiated silicon-dominant anodes is investigated with 5.4 Ah multilayer pouch cells as a follow-up study of our previous work with laboratory T-Cells.3 Methodologically, a novel cell holder and pressure device were developed and validated to apply different mechanical cell pressures. Our investigation covers a pressure range from uncompressed as zero (ZP, 0.00 MPa) to high (HP, 0.50 MPa) external cell pressure to determine the optimal pressure for high rate capacity, cyclic lifetime, energy density, and low thickness gain. The cells were tested by checkup cycles at C/10, rate tests up to 3C, and cycled at C/2 until 70% state of health (SoH). Electrochemical impedance spectroscopy was conducted at different SoHs to quantify the effect of mechanical pressures on the impedance response.4 The post-mortem analysis after reaching 70% SoH and operando thickness measurements in a compression test5 bench give insights into the electrode and cell thickness gain.In the electrochemical results, the discharge capacity Q DCH was immediately reduced by decreasing the mechanical pressure from 0.20 MPa to 0.00 MPa or 0.05 MPa, both for the cells in the cell holder (see inset in Figure 1a) and the compression test bench (not shown). This pressure change was also quantified by an increased charge-averaged full cell potential during charging and a decreased charge-averaged full cell potential during discharging. The impedance was reduced for higher mechanical pressure, especially the cathode contact resistance initially and over aging, resulting in lower temperature increases during the rate tests. The capacity retention Q DCH and the total charge throughput were increased to 360 cycles until 70% SoH for higher mechanical pressures at NP (normal pressure, 0.20 MPa) and HP (see Figure 1) compared to 181 cycles for ZP. Overall, applying higher mechanical pressures reduced the thickness increase, as shown by operando thickness measurements and the post-mortem analysis. A dependency of the reversible thickness on the charged capacity was found. Moreover, the porosity decreased with increasing pressure, which was measured with mercury intrusion porosimetry (MIP) and pycnometry. The anode mass increase correlates to the total charge throughput, which is pressure-dependent and the highest for NP.Our findings demonstrate that applying the optimal mechanical cell pressure is significant when using microscale silicon-dominant anodes. We recommend a mechanical pressure of 0.50 MPa for increased rate capacity, minimizing impedance, heat evolution, and thickness gain, and 0.20 MPa for increased energy density. Acknowledgments: S.F. gratefully acknowledges the financial support from the BMBF (Federal Ministry of Education and Research, Germany) under the auspices of the ExZellTUM III project (grant number 03XP0255). The authors want to thank the research battery production team of iwb at TUM for the multi-layer pouch cell production. We also thank Wacker Chemie for kindly providing the microscale silicon material.
Read full abstract