SiO is regarded to be one of the promising anode materials to replace graphite in lithium-ion batteries (LIB) because of its higher specific capacity than graphite and better cycle stability than Si [1]. However, the large irreversible capacity is a serious problem which consumes charged capacity in vain during initial cycles. Therefore, Li pre-dope to dose Li to SiO beforehand has been proposed. In this study, our previous achievement on Li pre-doping to Si electrodes was much helpful. The highlights of the achievement are as follows [2, 3].

1. Li-naphthalenide (Li-NTL) was selected as a Li pre-doping agent.
2. Li pre-doped capacity was much dependent on the solvent of Li-NTL solution.
3. 2-methyltetrahydrofuran (MeTHF) delivered the largest pre-doped capacity.

Hence, we conducted this study on Li pre-doping to SiO electrodes by referring to the achievement on Si electrodes.

SiO electrodes coated on Cu foil consisted of SiO, Ketjen Black, and polyimide binder in the ratio of 80 : 5 : 15 (m/m). Two types of SiO powder with an average particle size of 5 μmφwere used: carbon-coated SiO denoted as C-SiO and bare SiO which were purchased from OSAKA Titanium technologies Co., Ltd. We adopted 0.5 M NTL/MeTHF as a standard pre-doping solution based on our result of Li pre-doping to Si. SiO electrodes were confined in an air-tight cells with the pre-doping solution for 24 h to be pre-doped. The electrodes were taken out of the cells, rinsed in DMC, and used to assemble coin cells with Li counter electrodes, separators and electrolyte solution of 1.0 M LiPF₆ / EC + DMC (1:1, v/v) + 10 wt% FEC. All of these processes were conducted in an Ar-filled glove box. Discharge/charge cycle test mode was constant-curent at 0.1 mA cm⁻² between 0.02 and 1.5 V at 30 °C.

Figure 1 shows charge/discharge performance of C-SiO electrodes with and without Li pre-doping. In Figure 1a, the OCV of the Li pre-doped electrode was as low as 0.2 V, which was much lower than that of the electrode without pre-doping, exceeding 2.0 V, and the charge capacity was nearly missing. However, the discharge capacity was not less than that of the electrode without pre-doping. This result reveals that SiO can be pre-doped effectively by Li-NTL/MeTHF, which is the same as Si. Figure 1b shows dQ/dV analysis results, in which the anodic curves for the with and without pre-doping are nearly identical. However, almost no cathodic curve is observed for the with pre-doping as shown in Figure 1a, while a sharp peak at 0.46 V assigned to formation of L_xSiO_y is observed for the without pre-doping [4]. A dashed curve inserted in the Figure 1b is the cathodic curve for the second cycle of the with pre-doping to complement the missing first cycle one. This curve shows no peak at 0.46 V and suggests formation of L_xSiO_y during pre-doping. We further analyzed the pre-doping reaction to elucidate the mechanism based on the result shown above. The detail of the results will be shown and discussed in our presentation.

[1] M. N. Obrovac, V. L. Chevrier, Chem. Rev., 114, 11444 (2014).

[2] M. Saito et al., Abstracts of PRiME 2020, A02-0410 (2020).

[3] M. Saito et al., J. Electrochem. Soc., 166, A5174 (2019).

[4] C. P. Grey et al., J. Am. Chem. Soc., 141, 7014 (2019).