Development of Silicon-Based Materials as Anode for Li-Ion Batteries


Student thesis: Doctoral Thesis

View graph of relations


Related Research Unit(s)


Awarding Institution
Award date1 Feb 2019


Silicon has been widely studied as a next-generation anode material for Li-ion batteries, as it has a high gravimetric capacity of up to 3570 mAh g-1 and a low average discharge voltage of about 0.4V versus Li/Li+. The biggest challenge is to maintain reversibility of the electrode despite the large volume expansion of more than 280% from the alloy reaction. In the presented thesis, the failure mechanism of silicon(Si)-based electrode was investigated as an anode in lithium ion battery, with the purpose of understanding the origin of the irreversible lithium storage, regarding to the irreversible capacity from solid electrolyte interphase (SEI) formation, Li-trapping inside the conductive carbon and binder, and also the mechanical failure of the electrode. This aims to provide further information on the method to improve the electrochemical stability of the Si-based electrode. Based on this mechanism, several strategies on improving the cyclability and reversibility of Si-materials are also presented. The research started with using micron-sized Si particle fabricating electrodes with different amount of conductive carbon and binder to study the different contribution of irreversible capacity. A systematic approach is established to separate the different contributions, in particularly SEI formation, lithium accommodation in carbon and binder, and lithiation and de-lithiation of the active material from the charge-discharge capacities of silicon electrodes. This is possible because the three different contributions have different characteristics. Firstly, the SEI formation is a well-known contributor on irreversible capacity, since it involves the oxidation of the salt and electrolyte solvents on the surface on the Si particle. This process occurs prior to the lithiation of Si particle. In principle, the irreversible capacity from the SEI formation is the same regardless of the lithiation degree of Si particle. Therefore, by comparing the Coulombic efficiency of electrodes discharged to different capacities, SEI contribution can be identified. Secondly, the lithium trapping in carbon and binder depends on their amount inside the electrode. Therefore, by comparing electrodes with different compositions, this contribution can also be distinguished. Third, the mechanical failure of the Si electrode is highly related to the Si expansion level, which also can be controlled by the lithiation degree. Thus, comparing electrode with different lithiated capacity can identify the mechanical critical point of Si electrode.

With the information provided by this model, it is discovered that the Si particle size and the mechanical property of binder are very crucial factors affecting the cycle stability of Si electrode. Throughout literature, nanostructured Si has been proven to be beneficial on cycle performance, therefore, silicon anode has improved significantly in the past couple of years. However, three major shortcomings associated with nanostructures still need to be addressed, namely their high surface area, low tap density, and poor scalability. In the second part, a facile and practical method to produce micron-sized Si secondary particle cluster (SiSPC) with high tap density and low surface area from bulk Si by high energy ball-milling is present. By coupling SiSPC with a mechanically robust polyimide binder, more than 95% of the initial capacity is retained after 500 cycles at 3500 mA g-1 (1 C rate). Reversibility of electrode thickness change is confirmed by in situ dilatometry. In addition, the polyimide binder suppresses surface reaction of the particles with electrolyte, resulting in a high coulombic efficiency of 99.7%. Excellent cycle performance is obtained even for thick electrodes with an areal capacity of 3.57 mAh cm-2, similar to those in commercial lithium-ion batteries. The presented Si electrode system has a high volumetric capacity of 598 mAh cm-3, which is higher than that of commercial graphite anode materials (330-430 mAh cm-3).

Even though the electrochemical stability is significantly improved with the combination of SiSPC and robust PI binder. The improvement in terms of the volumetric capacity is still not impressive enough. The main reason for that is the intrinsic volume expansion of Si, which results in drastic electrode expansion. In order to further increase the volumetric capacity, the intrinsic volume expansion of Si has to be suppressed. In the third part, we demonstrate that titanium atoms inside the silicon matrix can act as an atomic binding agent to hold the silicon atoms together during lithiation and mend the structure after de-lithiation. Direct evidence from in situ dilatometry of co-sputtered silicon-titanium thin films reveals significantly smaller electrode thickness change during lithiation, compared to a pure silicon thin film. In addition, the thickness change is fully reversible with lithium extraction and ex situ post-mortem microscopy shows that film cracking is suppressed. Furthermore, Raman spectroscopy measurements indicate that the Si-Ti interaction remains intact after cycling. Optimized Si-Ti thin films can deliver a stable capacity of 1000 mAh g-1 at a current of 2000 mA g-1 for more than 300 cycles, demonstrating the effectiveness of titanium in stabilizing the material structure. A full cell with a Si-Ti anode and LiFePO4 cathode is demonstrated, which further validates the readiness of the technology.

The role of titanium inside silicon is clearly illustrated, other alternatives such as silicon monoxide (SiO) is very interesting to investigate. SiO is a potential high-capacity anode material (theoretical capacity=2615 mAh g-1) for lithium-ion batteries. Comparing with elemental Si, SiO has a lower theoretical expansion value of 160%. However, its low initial Coulombic efficiency hinders its adoption in commercial batteries. Here, we use an electrochemical approach, depth of discharge test, to study the origin of the irreversibility in SiO electrodes. We find two contributions to the irreversible capacity of SiO, depending on the reaction products during lithiation. Carbon-coating the material improves the reversible capacity as it increases its electrical conductivity. Disproportionation is an effective way to decrease irreversibility in the initial cycle with the formation of large cluster of SiO2 which is electrochemically inactive. Though, the reversible capacity is reduced. Even if the theoretical expansion is less for SiO, it still undergoes large volume changes during charge and discharge, leading to fast capacity fading. So far there has not been a good method to use micron-sized powders in battery electrode while maintaining its stability. In the last section, a multilayer coating system is designed to achieve the goal. It starts with introducing a self-assembled monolayer as an intermediate layer to cross-link the active material with the polyimide coating. Then, a high-modulus polyimide outer coating is also introduced which can accommodate the volume change and also exerts a compressive force when SiO particle contracts upon de-lithiation. It is demonstrated the multilayer system on SiO materials, with a stable capacity of 1310.7 mAh g-1 after 100 cycles at 150 mA g-1. Even at a rate of 1 A g-1, the SiO composite material is stable for 300 cycles.