Is the Cu/Zn disorder the main culprit for the voltage deficit in kesterite solar cells?

TLDR: In this article, the role of anionic (S/Se) distribution and cationic (Cu/Zn) disorder on the open-circuit voltage (Voc) and the ultimate photovoltaic performance of kesterite devices is clarified.

Abstract: Photovoltaic thin film solar cells based on kesterite Cu2ZnSn(Sx,Se1–x)4 compounds (CZTSSe) have reached >12% sunlight-to-electricity conversion efficiency. This is still far from the >20% record devices known in Cu(In1–y,Gay)Se2 and CdTe parent technologies. A selection of >9% CZTSSe devices reported in the literature is examined to review the progress achieved over the past few years. These devices suffer from a low open-circuit voltage (Voc) never better than 60% of the Voc max, which is expected from the Shockley-Queisser radiative limit (S-Q limit). The possible role of anionic (S/Se) distribution and of cationic (Cu/Zn) disorder on the Voc deficit and on the ultimate photovoltaic performance of kesterite devices, are clarified here. While the S/Se anionic distribution is expected to be homogeneous for any ratio x, some grain-to-grain and other non-uniformity over larger area can be found, as quantified on our CZTSSe films. Nevertheless, these anionic distributions can be considered to have a negligible impact on the Voc deficit. On the Cu/Zn order side, even though significant bandgap changes (>10%) can be observed, a similar conclusion is brought from experimental devices and from calculations, still within the radiative S-Q limit. The implications and future ways for improvement are discussed.