Snaith group

Abstract:

Multi-band gap heterojunction solar cells fabricated from a blend of 1.2 eV and 1.4 eV PbS colloidal quantum dots (CQDs) show poor device performance due to non-radiative recombination. To overcome this, a CdS shell is epitaxially formed around the PbS core using cation exchange. From steady state and transient photoluminescence measurements, we understand the nature of charge transfer between these quantum dots. Photoluminescence decay lifetimes are much longer in the PbS/CdS core/shell blend compared to PbS only, explained by a reduction in non-radiative recombination resulting from CdS surface passivation. PbS/CdS heterojunction devices sustain a higher open-circuit voltage and lower reverse saturation current as compared to PbS-only devices, implying lower recombination rates. Further device performance enhancement is attained by modifying the composition profile of the CQD species in the absorbing layer resulting in a three dimensional quantum cascade structure.

Abstract:

The vast majority of perovskite solar cell research has focused on organic-inorganic lead trihalide perovskites. Herein, we present working inorganic CsPbI3 perovskite solar cells for the first time. CsPbI3 normally resides in a yellow non-perovskite phase at room temperature, but by careful processing control and development of a low-temperature phase transition route we have stabilised the material in the black perovskite phase at room temperature. As such, we have fabricated solar cell devices in a variety of architectures, with current-voltage curve measured efficiency up to 2.9% for a planar heterojunction architecture, and stabilised power conversion efficiency of 1.7%. The well-functioning planar junction devices demonstrate long-range electron and hole transport in this material. Importantly, this work identifies that the organic cation is not essential, but simply a convenience for forming lead triiodide perovskites with good photovoltaic properties. We additionally observe significant rate-dependent current-voltage hysteresis in CsPbI3 devices, despite the absence of the organic polar molecule previously thought to be a candidate for inducing hysteresis via ferroelectric polarisation. Due to its space group, CsPbI3 cannot be a ferroelectric material, and thus we can conclude that ferroelectricity is not required to explain current-voltage hysteresis in perovskite solar cells. Our report of working inorganic perovskite solar cells paves the way for further developments likely to lead to much more thermally stable perovskite solar cells and other optoelectronic devices.