The ODIN CIGS project was carried out over a multi‑year period under the auspices of the German Ministry of Education and Research. The consortium comprised the Zentrum für Sonnenenergie‑ und Wasserstoff‑Forschung (ZSW), several industrial partners specialising in thin‑film deposition, and academic collaborators from universities with expertise in materials science and device physics. The project’s primary aim was to deepen the physical understanding of Cu(In,Ga)Se₂ (CIGS) solar cells and to translate this knowledge into tangible efficiency gains through process optimisation, interface engineering and module integration.

A central technical outcome was the systematic optimisation of the post‑deposition treatment (PDT) of the absorber layer. The introduction of rubidium fluoride (RbF) as a standard PDT step on the inline CIS4 deposition line yielded a measurable improvement for conventional CIGS cells, whereas the addition of sodium fluoride (NaF) as a seed or PDT layer did not produce a significant benefit and, in some cases, degraded the gallium‑gradient profile. For the Ag‑doped variant (ACIGS), the optimal silver content was found to be between 5 % and 10 at % (AAC 5–10 %). Under these conditions, ACIGS cells achieved efficiencies of 18.9 % without an antireflection coating and 20.4 % with a MgF₂ ARC, accompanied by fill factors above 80 %. Bulk recombination increased at higher silver loadings, as revealed by recombination loss analyses.

The project also explored alternative n‑type buffer layers to replace the conventional CdS. Zinc‑based sulfide buffers, Zn(O,S), were deposited by chemical bath deposition (CBD), sputtering and atomic‑layer deposition (ALD). The best CBD‑CdS reference cell delivered an efficiency of 19.6 % (Voc = 715 mV, FF = 77 %, Jsc = 35.6 mA cm⁻²). The best Zn(O,S) cells, fabricated by sputtering (dc‑Zn(O,S) = 43 nm) and by radio‑frequency deposition (rf‑Zn(O,S) = 85–95 nm), reached efficiencies of 18.4 % and 18.0 % respectively, with Voc values 31 mV lower and fill factors 3.5 % lower than the CdS reference. The loss in efficiency was mainly due to reduced open‑circuit voltage and fill factor, while the short‑circuit current remained comparable. For ACIGS cells, the Zn(O,S) buffers produced a 1–4 % absolute efficiency loss relative to CdS, depending on the absorber type and Ag content.

Oxide‑based ternary buffers of the form (Zn,M)O (M = Mg, Sn, Ti) were also investigated. ZnSnO and ZnTiO, previously reported for high‑bandgap CIGS, were applied to standard absorbers and yielded open‑circuit voltages above 1 V, confirming their potential for high‑efficiency devices. The ZnMgO buffer, deposited by CBD, achieved a maximum efficiency of 16 % in a 22 nm thick layer, with a corresponding Voc of 620 mV and a fill factor of 60 %.

Device integration studies demonstrated that the best Zn(O,S) cells could be combined with perovskite top cells to form four‑terminal tandem devices. A 0.5 cm² tandem cell reached 27.0 % efficiency, setting a world record for perovskite/CIGS tandems. Scaling to a 10.4 cm² module produced a 22.0 % efficiency, again a world‑leading value for this module size. Long‑term stability tests on small modules revealed a degradation rate of 0.5 % per year under accelerated ageing conditions, and a degradation model was developed to predict module lifetime.

Throughout the project, the consortium maintained close collaboration between deposition engineers, materials scientists and device physicists. The ZSW provided the deposition cluster and analytical facilities, while industrial partners supplied large‑area substrates and contributed to module fabrication. The project’s outcomes were disseminated through peer‑reviewed publications and conference presentations, ensuring that the advances in CIGS technology are accessible to the broader photovoltaic community.