Oleksandr Selyshchev, Yevhenii Havryliuk, Mykhailo Ya. Valakh, Volodymyr O. Yukhymchuk, Oleksandra Raievska, Oleksandr L. Stroyuk, Volodymyr Dzhagan*, Dietrich R.T. Zahn
Copyright © 2020 American Chemical Society
In this work, the authors outline the use of XPS and RAMAN as a combined approach to deducing chemical structure in complex quaternary solids, using binary and ternary materials as a reference.
Numerous materials and concepts have been developed as alternative to silicon solar cells currently dominating the global market, however many suffer from high costs or contain toxic elements.
Quaternary Cu2ZnSnS4 (CZTS) and mixed Cu2ZnSn(S,Se)4 (CZTSSe) compounds are among the best candidates which do not have the above deficiencies. CZTS consists of non-toxic and cheap elements, has a direct bandgap at 1.4-1.5 eV with an absorption coefficient above 104 cm-1, as well relatively good intrinsic (p-type) conductivity. Despite the great potential and the intense technological and physical investigations in the last decade, the photovoltaic efficiencies achieved up to now, at most 12-13 %, are not
sufficient for CZTS and CZTSSe to be employed commercially.
The main problem of CZTS-based solar cells, the low open-circuit voltage, is supposed to have two physical reasons. One of which is a very small stability region of the CZTS compound in the rather complex phase diagram of Cu-Zn-Sn-S. This results in a high probability of formation of secondary
phases during growth of CZTS. Due to this, it is vital to improve experimental capabilities of identification of secondary phases, in order to optimize the growth conditions and quality of CZTS nanocrystals and thin films.
The Role of XPS
XPS was used to determine the elemental composition and verify the valence states of the elements, however due to the nature of the material many aspects of the XPS spectra are not straightforward.
While XPS was able to identify many species formed, such as Sn(II) and Sn(IV), it is not always able to identify the precise species. For example, SnS and SnO may appear indistinguishable even when considering the auger parameter. It was therefore important to account for the possibility of the species being present as an oxidized surface.
Raman spectroscopy however revealed no significant SnO population and the species was determined to be SnS with SnS2 present as a surface species.
By using the Raman spectra in combination with the XPS, the authors were able to easily deduce the formation of relevant binary and tertiary compounds within both the tertiary references and within the eventual quaternary CZTS material and therefore identify a synthetic route to maximising the quantity of the desired kesterite CZTS.