Theoretical Investigation of Lattice-Matched III-N-V/Si Double-Junction Solar Cells

Xiaobin Zhang,Liming Liu,Feng Chi,Ke Wang,Kaiwen Lin,Yuehui Wang
DOI: https://doi.org/10.1088/1361-6463/ac9537
2022-01-01
Abstract:The lattice-matched III-N-V/Si double-junction (DJ) solar cells are designed with GaNAsP and GaInNP top cells, respectively. Under AM1.5G condition, the efficiencies of III-N-V/Si DJ cells are calculated with variable electron lifetime (tau(e) ) and electron surface recombination velocity (S-e ) in top cell. When S-e is 100 cm s(-1) and tau(e) rises from 1 to 1000 ns, the optimal efficiency of GaNAsP/Si cell increases from 31.12% to 36.13% due to the increasing short-circuit current and open-circuit voltage. With tau(e) of 100 ns, the optimal efficiency keeps at a high value of similar to 35% when S-e changes from 10 to 1000 cm s(-1), but drops obviously with S-e of 10 000 cm s(-1). In comparison, the optimal efficiency of GaNAsP/Si cell is less sensitive to S-e than to tau(e) . With fixed S-e of 100 cm s(-1), GaNAsP/Si cell shifts the optimal top-cell bandgap from 1.716 to 1.787 eV when raising tau(e) from 1 to 1000 ns. However, the effect of S-e on optimal top-cell bandgap is negligible. For III-N-V/Si cell with 100 ns tau(e) and 100 cm s(-1) S-e , an optimal efficiency is obtained as similar to 35.1%, which would be closer to the experimental limit owing to the expectable values of tau(e) and S-e . Furthermore, the optimal efficiency of GaNAsP/Si cell drops slightly when thinning Si substrate from 300 to 150 mu m, but has a maximum of 35.95% with substrate doping of 1 x 10(16) cm(-3) when the doping concentration varies from 1 x 10(15) to 1 x 10(18) cm(-3). The results and discussion in this work may act as a guidance for studying III-N-V/Si DJ cell.
What problem does this paper attempt to address?