Vapor-Phase Incorporation of Ge in CZTSe Absorbers for Improved Stability of High-Efficiency Kesterite Solar Cells

Abstract

We report an approach to incorporate Ge into Cu2ZnSnSe4 using GeSe vapor during the selenization step of alloyed metallic precursors. The vapor incorporation slowly begins at T ≈ 480 °C and peaks at 530 °C, resulting in a Ge-based composition shift inside the previously formed kesterite layer. We initially observe the formation of a Ge-rich surface layer that merges into a homogeneous distribution of the incorporated element during the further dwelling stage of the annealing. This approach is very versatile and could be used in many similar fabrication processes for incorporating Ge into CZTSe-absorber layers. Because the vapor-based composition shift in the layer happens after the formation of the absorber film towards the end of the fabrication process, most process parameters and the precursor structure may not need any significant re-optimization. The careful integration of this step could help to reduce Sn-related deep defects and accompanying VOC losses. The best CZTGSe-power-conversion efficiency obtained in this series is 10.4 % (with EG = 1.22 eV, FF = 54%, JSC = 36 mA/cm2, VOC = 540 mV, VOCdef,SQ = 417 mV). These results demonstrate the potential of this approach for Ge incorporation into kesterite absorbers.

Ge-alloyed kesterite thin-film solar cells: previous investigations and current status – a comprehensive review

“…71 Besides that, one study reproduced similar improvements related to Ge doping as those brought by the IREC Ge nanolayer approach, 74 namely an increased doping concentration and enhanced morphology. This improved structural quality was jointly observed by two different teams investigating an alternative strategy that consists in integrating Ge only during the selenization step of a sputtered stack, either in a GeSe 2 -Se, 72 or a pure Se-Sn-Ge environment 73 as shown in Fig. 5(a).…”

“…The former focused on Ge doping ( x < 0.02) and highlighted the widely seen bi-layered absorber with a Ge gradient towards the back promoting alkali diffusion, no secondary phases and reduced band tails. 72 The latter discovered two dependencies on the value of x ranging from 0 to 1: 73 first, an enlarging spread of the Sn–Se binary phase in both the bulk and at the surface, the evaporation of which may likely explain the voids observed in between the Ge-enlarged grains (Fig. 5(b)).…”

“…Many references report investigations about CZTGSe absorbers deposited by two-step processes under vacuum, most of them attributed to one research group from IREC and close collaborations, 60–70 and a few other investigations carried out by other teams. 71–74…”

“…Many references report investigations about CZTGSe absorbers deposited by two-step processes under vacuum, most of them attributed to one research group from IREC and close collaborations, [60][61][62][63][64][65][66][67][68][69][70] and a few other investigations carried out by other teams. [71][72][73][74] The IREC group implemented and extensively used one approach to process Ge-doped CZTSe, or CZTSe:Ge, thin lms on glass/Mo, i.e. sputtering/evaporation of a Cu/Sn/Cu/Zn precursor stack with the addition of a thin thermally evaporated Ge nanolayer at a certain position within the stack, followed by reactive annealing in a Se (and Sn) atmosphere.…”

“…Still, from the few reports available, 36,37,41 one can notice that one-step co-evaporation processes do not seem to suffer from the same secondary phase issues widely observed for two-step deposition, leading for instance to poor adhesion and strong delamination related to IV-(S,Se) 2 evaporation. 47,48,57,59,62,63,68,70,73,75,76 This may likely be explained by the fact that evaporating the necessary elements near the phase transition temperature without leaving a vacuum restricts the formation of potentially unstable binary phases. In contrast, sequential processes involve complex intraprecursor diffusion mechanisms inuenced by the stack design, physical state of elements, annealing conditions, and air exposure or time delays before selenization and sulfurization.…”