A detrimental reaction between a Mo back contact and CZTS

CZTS(e) Compiled description of CZTS-solar cells with either sulfur (S) or selenium (Se). Likewise for other compounds with S(e).
Annealing Heat treatment of films to get a certain crystal structure.
Phases Here: the different structures that can be created from the elements in CZTS.
Passivation When surfaces are covered with a protective layer to avoid reaction and improve optical and/or electrical properties.

For CZTS-solar cells to be improved and contend with the similar CIGS-solar cells, more knowledge for the chemical processes in the material is needed – especially when this differs from the more well-researched CIGS. For example, it is known that the surface of the CZTS layer is heat sensitive and decomposes at high temperatures, especially at low pressures, where Sn(IV) in CZTS is reduced to Sn(II). In this study, another consequence of this instability was investigated, which provided proof for a decomposition of CZTS(e) in the interface to the back contact of Mo where MoS(e)2 is formed.

It has already been shown that a layer of MoS(e)2 is normally formed after annealing of CZTS(e)-samples, although it has not earlier been proven that the formation is due to a reaction between Mo and CZTS(e). The reaction was here suggested as a reaction between CZTS(e) and Mo which results in Cu2S(e), ZnS(e), SnS(e) and MoS(e)2. Thermodynamic calculations on the reaction showed that it was favorable in energy, but the study also investigated the existence of the reaction experimentally. For this, samples with CZTS (with sulfur) were sputtered on a glass covered with Mo as back contact.

To indicate the existence of the reaction the presence of MoS2 at the interface between CZTS and Mo as well as a separation of CZTS into Cu2S, ZnS and SnS close to the back contact was needed. To exclude other reactions that could form MoS2, all other sources of S were eliminated – both during sputtering and annealing. Presence of the other phases simultaneously at the back contact would be difficult to refer to any other process, although all processes that generally lead to formation of other phases were minimized. After fabrication and annealing, the phases in the top surface of the CZTS layer, the back surface of the CZTS layer and in the Mo layer were investigated. At the top surface of the CZTS layer no other phases than the CZTS were found, while Cu2S, ZnS and SnS were indicated at the back surface and MoS2 was also found at the back contact. All these were seen to increase in thickness for samples that had been annealed for a longer time. Other processes that could form these cannot explain the results combined, therefore it can be concluded that the suggested reaction has taken place. Due to the similarities with CZTSe-solar cells (with selenium), there is probably a similar reaction for these as well.

This process has a negative impact on the solar cell’s efficiency since the other phases increase the recombination and decrease the material quality. To produce solar cells with a high performance, processes that decrease the quality of the CZTS layer must be eliminated. In this case, this could be done by replacing Mo with another material or using a barrier layer to passivate the interface.

Read the original article here