Proceedings of nanoGe International Conference on Perovskite Solar Cells, Photonics and Optoelectronics (NIPHO19)
DOI: https://doi.org/10.29363/nanoge.nipho.2019.009
Publication date: 21st November 2018
A combination of high-resolution mapping techniques was used to probe defects and homogeneity in perovskite solar cells with different architectures. First, the localized effect of excess PbI2 on the photophysical and photoelectrical properties of perovskite solar cells with inverted structure was investigated by photoluminescence (PL), photocurrent and Raman mapping with micrometre resolution. On the contrary to other works in which excess PbI2 is obtained by varying the composition of perovskite precursors solution or by annealing the already formed perovskite film, we use laser irradiation to generate localized PbI2 in a full device, which amount can be controlled by tuning the laser intensity or irradiation time. We show that whereas a thick PbI2 film at the perovskite/hole transport layer interface has a detrimental effect on the local photocurrent, a thin PbI2 film (<20 nm) leads to a significant photocurrent increase, which is ascribed to the passivation of non-radiative defects and reduced charge recombination at the interface1.
Then, the infiltration mechanisms and homogeneity of mesoscopic perovskite solar cells with structure mesoporous TiO2/mesoporous ZrO2/mesoporous carbon were investigated by a combination of photoluminescence, photocurrent, Raman and electroluminescence mapping performed on large and small sample areas. Three different types of cells prepared using a one-step infiltration process with MAPbI3 or AVAI-MAPbI3 solution or two-step process with MAPbI3 were investigated. It is found that the one-step MAPbI3 cell has very limited infiltration which results in poor device performance. On the contrary, high loading of the mesopores of the TiO2 and ZrO2 scaffold is observed when using AVAI-MAPbI3 solution, but some micrometre-sized areas are not efficiently infiltrated due to the presence of dense carbon flakes hindering perovskite infiltration. Quite differently, the two-step cell has a complex morphology with several types of defects having beneficial or detrimental effects on the local photocurrent.
This work shows how complementary mapping techniques can be used to correlate materials and devices properties on micrometre length scales. PL intensity analysis can be difficult to interpret when measuring a full solar device because of competing photoelectronic effects (quenching, recombination…). Hence, the correlation with photocurrent, electroluminescence and Raman mapping provides a better understanding of the interplay between perovskite infiltration/crystallisation, defects, local PbI2 excess and charges extraction for perovskite solar cells with various architectures.
The authors acknowledge funding from the EPSRC (grant no EP/M025020/1), Welsh Assembly Government funded Ser Cymru Solar Project. V. K and C. R. would like to thank EPSRC for financial support under projects EP/N008065/1, EP/ M025020/1 and EP/L017563/1 and for the use of the University of Sheffield electron microscopy facilities.