Can Raman Spectra of NMC Material Family be Detangled?

Abstract

Raman spectroscopy is an advanced tool for investigating the structural and chemical properties of Li-ion battery materials. With the ability to probe the vibrational modes of crystals, it can provide valuable insights into the structural evolution and chemical reactions that occur during battery cycling. This technique has been used extensively to study electrode materials, solid electrolytes, and interfaces in Li-ion batteries, providing a deeper understanding of the mechanisms governing their performance and degradation. [1]Raman spectroscopy plays a crucial role in the characterization of LiNi1-x-yMnxCoyO2 (NMC) materials, which are widely used as positive electrodes. It can provide information on the phase transitions, crystal symmetry, and degradation that occur during battery cycling. [2] The Raman spectrum of LiMO2 is composed of peaks that correspond to the vibrational modes of the crystal lattice. The most prominent line in Raman spectrum of the R-3m space group symmetry should correspond to the A1g mode, which in case of LiMO2 crystal arises from the symmetric stretching vibration of the metal-oxygen (M-O) bonds in the octahedral coordination. The second predicted peak in the spectrum is Eg which involves the movement of oxygens in opposite directions across adjacent O-layers. [2] The interpretation of the NMC Raman spectra can be challenging due to several factors. One of the main difficulties is the complexity of its crystal structure, giving rise to a large number of vibrational modes in the spectrum. The presence of multiple metal-oxygen (M-O) bond types in the octahedral coordination can lead to overlapping Raman peaks, making it difficult to distinguish between the different modes. Another challenge is the presence of defects and disorder, which can lead to changes in the crystal symmetry and vibrational modes of the material, resulting in additional peaks or shifts in the Raman spectrum. Factors such as particle size, morphology, and orientation might also affect the Raman spectrum, as might the excitation wavelength and polarization of the incident light. To overcome these challenges, a combination of experimental and theoretical techniques, such as X-ray diffraction and density functional theory calculations, is typically used to assist in the interpretation of the Raman spectra. [3] Despite the efforts, the full interpretation of NMC Raman spectra, particularly with respect to its charged states, remains undone.In our investigation, we found noticeable changes in the Raman spectra across samples with similar, but non-identical, compositions. We also see that partially and fully charged NMC samples have significantly more than two or three Raman modes varied or shifted, likewise suggesting more complex spectrum. The systematic comparison of a larger number of spectra allowed us to identify these variations and correlate them with specific material compositions. We developed a novel approach for interpreting Raman spectra of NMC materials that involves an extensive collection of spectral data. This method requires us to gather Raman spectra from diverse NMC samples with varying compositions, particle sizes, and morphologies. We use advanced analysis to identify correlations and trends between the spectral features and the material properties. This demanding approach gives a more comprehensive understanding of the relationship between the Raman spectra and the composition of NMC samples, which can facilitate better interpretation and design of these materials for use in Li-ion batteries. In this work, we will highlight the potential of this method to enable us to more accurately interpret Raman spectra of charge and discharge states that might further help in developing easy monitoring of NMC electrodes during their operation. A detailed data analysis will be presented at the conference. Funding for this work was provided by the National Science Center in Poland under the Sonata BIS 11 program (No. UMO-2021/42/E/ST5/00390).[1] P.P.R.M.L. Harks et al. J. Power Sources 288 (2015) 92-105[2] E. Flores et al. Front. Energy Res., 6 (2018) 82[3] E. Flores et al. Chem. Mater. 2020, 32, 1, 186–194