Thin film solid ionic conductors have been identified as a key component for miniaturised all-solid-state sensors and batteries since the 1980’s. Among them, amorphous Li-based materials prepared by vacuum deposition techniques are particularly attractive as electrolytes for making lithium microbatteries and integrating the latter into smart autonomous microsystems. In that respect, lithium phosphorus oxynitride, prepared by reactive sputtering of a Li₃PO₄ target under a pure nitrogen atmosphere, exhibits a combination of interesting properties such as an ionic conductivity of ~ 2.10^-6 S.cm^-1 at room temperature, a very low electronics conductivity < 10^-14 S.cm^-1, and passivating behaviour on lithium metal, which led to the market launch of first microbatteries. Since then, energy density specifications for IoT and medical applications have led to changes in conventional cell design with more stringent requirements for ionic conductivity. In recent years, there has also been renewed interest in these thin-film amorphous electrolytes as part of the development of next-generation batteries for automotive applications. Used in the form of coatings, they can solve reactivity problems at the electrode/electrolyte interface and improve the cycling behaviour of metallic lithium [1].

Amorphisation and then partial nitridation of lithium orthophosphate, leading to the formation of a disordered material containing condensed units with bridging nitrogen [2, 3], was found to be an effective means of boosting ionic conductivity. The introduction of a second glass former, such as in Li-Si-P-O-N, can also provide an additional tenfold increase through a mixed former effect [4], while maintaining exceptional mechanical properties [5]. This evolution illustrates the need for research into materials that are more and more complex in their composition, and not confined to ‘simple’ systems, to identify the best ionic conductors. In this respect, high-throughput experimentation is an attractive approach for screening multi-element compositions [6]. It usually involves a combinatorial synthesis method capable of rapidly producing a large number of samples with different compositions, i.e. material libraries, combined with automated characterisation methods that allow various specific properties to be rapidly measured on all the samples produced.

The development of a specific experimental approach for the high-throughput screening of thin film solid electrolytes based on the synthesis of material libraries by magnetron co-sputtering on 4’’ wafers and the rapid characterisation of their thickness (stylus profilometry), composition (LIBS, ICP-OES), structure (Raman spectroscopy) and conduction properties (Impedance spectroscopy) will be presented [7]. Mapping the composition of these thin films containing low-Z elements, particularly Li, on large substrates is undoubtedly the most difficult stage in this process. Laser-induced breakdown spectroscopy (LIBS) appears to be the most appropriate technique for achieving this objective, as it meets the key requirements of speed, spatial resolution and sensitivity [8]. However, further developments are required to achieve elemental quantification. To this end, combined ICP-OES, LIBS and ion beam (RBS, NRA) analyses were carried out.

Ultimately, screening large parts of the compositional space of Li(Si,P)(O,N) reveals general relationships between composition, local structure and ion transport properties, but also raises questions about the influence of synthesis conditions on the degree of disorder in these amorphous materials