Mr Vincent J.-Y. Lim

Abstract:

Mixed-halide perovskites offer ideal bandgaps for tandem solar cells, but photoinduced halide segregation compromises photovoltaic device performance. This study explores the influence of a hole-transport layer, necessary for a full device, by monitoring halide segregation through in situ, concurrent X-ray diffraction and photoluminescence measurements to disentangle compositional and optoelectronic changes. This work demonstrates that top coating FA0.83Cs0.17Pb(Br0.4I0.6)3 perovskite films with a poly(triaryl)amine (PTAA) hole-extraction layer surprisingly leads to suppression of halide segregation because photogenerated charge carriers are rapidly trapped at interfacial defects that do not drive halide segregation. However, the generation of iodide-enriched regions near the perovskite/PTAA interface enhances hole back-transfer from the PTAA layer through improved energy level offsets, increasing radiative recombination losses. It is further found that while passivation with a piperidinium salt slows halide segregation in perovskite films, the addition of a PTAA top-coating accelerates such effects, elucidating the specific nature of trap types that are able to drive the halide segregation process. This work highlights the importance of selective passivation techniques for achieving efficient and stable wide-bandgap perovskite photovoltaic devices.

Abstract:

Metal-halide perovskites have been under intense research over the past decade due to their potential to compete with and complement silicon in solar photovoltaic devices, owing to their affordability and better device performances. However, their commercialisation has been hindered mainly due to their problems with stability, which requires more thorough fundamental investigation of these materials. This thesis focuses on understanding their photophysics and structural dynamics through various spectroscopic techniques. More specifically, metal-halide perovskites have issues regarding halide segregation and tin oxidation, as well as intriguing low-frequency Raman response, which are addressed in this thesis.

The effects of a hole-transport layer and defect passivation on halide segregation under illumination in mixed-halide perovskites have been investigated using in-situ photoluminescence and X-ray diffraction. It is demonstrated that using photoluminescence on its own may be misleading when studying halide segregation, especially when a charge-transport layer is present. The introduction of a hole-transport layer slows down halide segregation due to hole depletion, but increases the photoluminescence intensity through hole back-transfer into iodide-rich regions near the interface between the perovskite and the hole-transport layer. It is shown that defect passivation has a profound and complex effect on halide segregation, and the nature of the trap states determines their role in halide segregation.

The degradation mechanisms in tin-only and mixed lead-tin iodide perovskites in ambient air have been examined. Although these two different types of perovskites both undergo severe optoelectronic degradation in ambient air over a few hours, their degradation mechanisms have been shown to differ; tin-only perovskites undergo heavy doping of the valence band when exposed to ambient air, but mixed lead-tin perovskites do not undergo any significant doping during exposure. It was concluded that tin-only perovskites degrade via introduction of shallow defect states which contribute to hole-doping of the valence band, while mixed lead-tin perovskites degrade via formation of deep trap states which do not significantly dope either of the electronic bands. Possible degradation products during ambient air exposure of these perovskites have also been identified.

Finally, low-frequency vibrational properties of metal-halide perovskites were investigated. Metal-halide perovskites have been reported to have intriguing responses in low-frequency Raman spectra, where the response is extremely broad. However, this is not the case for IR spectra, where only well-defined contribution from phonon modes are visible. The origin of the broad Raman response, as well as the discrepancy between Raman and IR spectra, has been an actively debated topic in the literature, with various hypotheses being presented. Through Raman and IR spectra of various metal-halide semiconductors and literature data, a number of hypotheses have been ruled out as the main cause of such Raman response. Those that are ruled out include: extrinsic defects, octahedral tilting, cation lone pairs, and ‘liquid-like’ Boson peaks. Instead, an alternative explanation for such broad Raman response and the discrepancy with IR spectra was presented: the broad Raman response is a result of an interplay of significant broadening of Raman-active vibrational modes and a Bose-Einstein population factor, while IR-active vibrational modes are subject to different and slower decay channels, and such Bose-Einstein population factor is not applicable to IR spectra.