Focused Ion Beams for Lithiation of High Capacity Host Materials for Negative Electrodes

Abstract

High capacity negative electrode Li ion battery materials such as Sn and Si are attracting great attention as higher capacity replacements for conventional graphite-based electrodes. Although they are very attractive materials in terms of their theoretical capacity (993 mAh/g for Sn and 4200 mAh/g for Si), and are nontoxic and naturally abundant, the challenges lie in their huge volume expansion and subsequent cracking, pulverization and electrochemical degradation [1]. Atomic level structural studies of lithiated nanostructured Sn- and Si-based electrodes using transmission electron microscopy (TEM), synchrotron radiation, X-ray diffraction (XRD), neutron diffraction and other instrumental methods under static conditions and in siturepresent an important approach for understanding the mechanisms of ion transport and associated volume expansion and cracking of the prospective host materials [2,3]. However, in these methods, multi-step phase transformations during lithiation involving cracking are only observed together with the transformations and movements of interfaces and boundaries of various expanding lithiated phases. To discern the various processes, it is necessary to have efficient control of the lithiation kinetics and the quantities of the transformed material. Techniques that enable the probing of lithiation at the nanoscale are critical for deeper insights into the complex mechanisms of lithiation processes and phase transformations of the host electrode materials. We introduce focused low-energy Li-ion beam (Li-FIB) as a new tool for direct lithiation of Sn and Si-based host materials with precise control of the lithiation speed as well as location and size of the chosen area [4]. In the Li-FIB, a focused beam with spot size of order 50 nm is scanned in a pattern across the sample. The rate of Li-ion injection (in the range of 5×106 ion·μm-2 s-1 for this measurement) can be controlled by adjusting the beam current (I ≈ 1 pA), and the area of lithiation can be precisely defined over a wide range, up to several micrometers, by controlling the scan pattern. Unlike with electrochemical lithiation, a solid electrolyte interface (SEI) is not formed with Li-FIB lithiation since the process takes place in a vacuum of ~2×10-4Pa. This enables us to study solely the structural evolution and associated stress accumulation in the host materials without any influence from the formation of the SEI. As an initial study, we compared Li-FIB lithiation with conventional electrochemical methods. The electrochemical lithiation was carried out in a three electrode cell with Li metal foil used for counter and reference electrodes. The electrolyte was composed of 1 mol dm-3 LiClO4in a mixture of ethylene carbonate (EC): diethyl carbonate (DEC) (1:1 by volume). The processed electrodes were washed with dimethyl carbonate (DMC) after lithiation and could be transferred for further TEM examination in a glove bag purged with Ar to prevent air exposure. To analyze the nanoscale morphological, crystallographic and compositional differences in the host materials lithiated by the two different methods, we employed Ga+ ion beam cross sectioning and field-emission scanning electron microscopy (FESEM). In this talk, we will compare a few models of Sn and Si electrodes lithiated with Li-FIB and electrochemically. For single crystalline β-Sn microspheres deposited onto a hard carbon substrate, the results are shown in Fig. 1. Figure 1 shows cross-sectioned Sn samples of (a) lithiated Sn with Li-FIB and (b) electrochemically lithiated Sn. The surface roughness in Fig. 1(a) indicates the volume expansion of Sn due to lithiation. In Fig. 1(b), non-uniform corrugations were observed on the surfaces of electrochemically lithiated electrodes, which may be caused by volume expansion and SEI formation during processing in the organic electrolyte. We think the band contrasts in Fig. 1 are associated with lithiated volume, since no contrast was observed for unlithiated spheres. It is interesting to note that the distribution of the band of contrast is different in Fig. 1 (a) and (b). The obtained results indicate that nanoscale processes that occur during Li+implantation involve a series of overlapping physico-chemical transformations, which will be discussed in more details in the talk. [1] D. Ma et al, Nano-Micro Lett.(2014) 347. [2] M. T. Janish et al, J. Mater. Sci., 51 (2016) 589 [3] X. H. Liu et al, Nat. Nanotechnol. 7 (2012) 479. [4] S. Takeuchi et al, J. Electrochem. Soc., 163 (2016) A1010 Figure 1