Towards quantification of electric fields at interfaces in battery materials

Abstract text (incl. figure legends and references)

Introduction:

The characterization of long-range electric fields and space-charge layer effects at the few to tens of nanometer scale is of supreme importance at various interfaces in Lithium-ion battery materials. Relevant interfaces include grain boundaries in cathode materials as well as the interface between cathode materials and solid electrolytes,[1,2] etc. The momentum transfer of the electric field to a scanning electron probe can be used to derive the electric field as it can be quantitatively measured using fast electron detectors. This method has been shown to yield quantitative data for GaAs p-n-homojunctions.[3] Characterizing long-range electric fields across junctions of different materials or materials in different orientations, however, requires employing a low semi-convergence angle (<2 mrad) in scanning beam mode. Furthermore, applying precession on top of this helps to reduce dynamical effects.[4] Typically, the scanning transmission electron microscope (STEM) modes of most microscopes with aberration correctors are not optimized to form a low semi-convergence angle (<2 mrad). One possibility is to use the TEM mode of the microscope with the smallest available condenser lens aperture. However, two major difficulties are usually associated with the TEM mode. First, the TEM modes of most microscopes do not allow the scanning of the electron beam. In such cases, the electron beam can be scanned by external scanners. Secondly, the TEM modes of most microscopes do not allow the insertion of annular dark field and bright field detectors, which makes the simple tasks of searching the region quite tedious.

Objectives

Being able to form a low semi-convergence angle with precession in the STEM mode allows us to use the microscope's internal scanner with annular detectors.

Methods and Materials:

Here we discuss methods to achieve a low semi-convergence angle in STEM mode and compare its pros and cons with the TEM mode with the external scanner. We also apply precession using the NanoMegas P2000 unit. Since aligning precession requires switching between diffraction and imaging modes, the precession alignment procedures in STEM mode are also outlined. We also comment on the highest attainable precession frequency, which is of supreme importance when taking datasets with a fast pixelated direct electron detector.

Results:

Finally, we acquire 4D STEM datasets using a pnCCD by employing a low semi-convergence angle (1.73 mrad) and precession (0.1⁰) at the grain boundary (GB) region in a Ni-rich lithium-ion battery layered cathode material (NCM622). The initial results are shown in Figure 1.

Fig. 1: Scanning-precession images at a GB-region in NCM622, (a) High-angle annular dark field (HAADF), (b) Virtual Bright Field (VBF), (c) Center-of-Mass (COM) x-component & (d) COM y-component.

References:

[1] L. Wang, R. Xie, B. Chen, X. Yu, J. Ma, C. Li, Z. Hu, X. Sun, C. Xu, S. Dong, T.-S. Chan, J. Luo, G. Cui, L. Chen, Nat. Commun. 2020, 11, 5889.

[2] J. Haruyama, K. Sodeyama, L. Han, K. Takada, Y. Tateyama, Chem. Mater. 2014, 26, 4248.

[3] A. Beyer, M. S. Munde, S. Firoozabadi, D. Heimes, T. Grieb, A. Rosenauer, K. Müller-Caspary, K. Volz, Nano Lett. 2021, 21, 2018.

[4] T. Mawson, A. Nakamura, T. C. Petersen, N. Shibata, H. Sasaki, D. M. Paganin, M. J. Morgan, S. D. Findlay, Ultramicroscopy 2020, 219, 113097.