Plasma-Enhanced Atomic Layer Deposition of Lithium Nickel Oxide Thin Film Model Systems

Abstract

The main bottleneck in the development of the next generation of Li-ion batteries with higher energy densities, as demanded for long-range electrical mobility, is the lack of understanding and control of the reactions at the electrode-electrolyte interfaces. Reactions at the cathode side cause rapid capacity loss during cycling and represent a major issue for batteries with high-voltage cathodes, which are crucial in reaching the required energy densities. Promising candidates for use in next generation of batteries are Ni-rich Li(Ni1-x-yMnxCoy)O2 (NMC) cathodes, which show high capacity, high potential, and contribute to address the geopolitical issue of decreasing the Co content in the cathode. Ni-rich NMCs exhibit similar electrochemical properties as pure LiNiO2 and therefore LiNiO2 thin films can be used as model systems to gain fundamental understanding of the interfacial processes that hinder the large-scale implementation of Ni-rich NMCs. Atomic layer deposition (ALD) is a suitable technique to fabricate these model systems, as it gives precise control over film thickness, uniformity, chemical composition, and crystal orientation.Previous work on LiNiO2 by ALD is not extensively reported in literature. Moreover, films grown by a supercycle process, in which two ALD subcycles are alternated, suffered from a low Ni at.% and a large C content because of precursor ligand residues, and films fabricated using a complex multilayered approach showed severe Si contamination from the Li precursor [1]. In this contribution we report on ALD lithium nickel oxide (LNO) thin films with stoichiometries close to LiNiO2 and low levels of contaminants.The ALD process is based on lithium hexamethyldisilazide (LiHMDS) and nickel bis(N,N’-di-tert-butylacetamidinate) (Ni(tBu-MeAMD)2) as precursors and O2 plasma as co-reactant. 1 ALD cycle of Li2O is combined with x cycles of NiO using a supercycle approach. In this contribution we explore the relation between the Ni/Li cycle ratio x and material properties of the LNO films.The growth of the LNO films is monitored by means of in-situ spectroscopic ellipsometry. The ellipsometric data of films with a high Ni/Li cycle ratio are fitted with a general oscillator model used for NiO films, while a Cauchy model was adopted in case of a low Ni/Li cycle ratio and Li2O films. The growth per supercycle (GPSC) of LNO shows an intermixing effect, as it does not follow the linear trend calculated from the growth per cycle (GPC) values of the individual Li2O and NiO ALD processes, which are 1.0 Å/cycle and 0.34 Å/cycle respectively.XPS surface measurements show that the Ni/Li atomic ratio is tuned from 0 to 10 as function of the Ni/Li cycle ratio x. Low levels of Si and N contamination, below 4% and 2% respectively, are observed. A shift of the main peak in the Ni 2p3/2 spectrum to higher binding energies indicates a shift in oxidation state from Ni2+ to Ni3+ as decreases [2]. According to surface measurements, the LNO film with x=5 shows a stoichiometry close to LiNiO2, but depth profile measurements indicate that the Ni at.% is significantly higher in the bulk than on the surface of the films, most certainly attributed to the formation of a Li2CO3 surface layer upon air exposure. Li2CO3 is also present in the bulk of the LNO films, but the bulk C at.% stays below 4%.These results show that we can use ALD to fabricate a high-voltage cathode model system, which we plan to characterize in-situ and post-mortem in combination with both liquid and solid electrolytes to study the cathode-electrolyte interface.[1] Maximov et al., Energies, 13 (2020) 2345[2] Fu et al., Applied Surface Science 441 (2018) 1048 This work is part of the ‘BatteryNL – Next Generation Batteries based on Understanding Materials Interfaces’ project (with project number NWA.1389.20.089) of the NWA research programme ‘Research on Routes by Consortia (ORC)’ funded by the Dutch Research Council (NWO). Figure:(a) The GPSC of LNO ALD process shows an intermixing effect when comparing the experimental results with the expected GPSC based on the two processes (b) Ni 2p3/2 XPS spectrum for different Ni/Li cycle ratios x. A shift to higher binding energies is observed with decreasing x, indicating a shift from Ni2+ to Ni3+. (c) XPS surface measurements of LNO films show low levels of contaminants and a Ni/Li atomic ratio varying with the Ni/Li cycle ratio. (d) Depth profile of LNO with x=5 shows an increased Ni at.% and a low C at.% in the bulk of the film. Figure 1