University of Cambridge
Developing Ternary Bismuth Chalcogenide Nanocrystals for Photovoltaics and Photodetectors
Abstract
dc:description.abstractOver the past decade, lead-halide perovskites (LHPs) have demonstrated significant potential in terms of their performance across a wide range of optoelectronic devices, including solar cells, photodetectors and light-emitting diodes. However, the toxicity of lead and instability issue of LHPs are still concerns for their widespread implementation. These successes, but also the challenges of LHPs have motivated great efforts across multiple disciplines to search for lead-free and stable alternatives that can have similar optoelectronic properties to LHPs, namely ‘perovskite-inspired materials (PIMs)’. With the deeper understanding of defect tolerance displayed in LHPs, a large number of PIMs have been identified until now. Among all the identified PIMs, ternary chalcogenides or ABZ<sub>2</sub> materials, are believed to be one of the most promising alternatives so far, owing to their simple fabrication protocols, strong absorption and high stability in air. Particularly, AgBiS<sub>2</sub> solar cells have demonstrated the highest efficiency (9.17%) among all bismuth-based solar cells. Nevertheless, studies into ternary chalcogenides are mostly limited to AgBiS<sub>2</sub> photovoltaics, and the investigations into other potential ABZ<sub>2</sub> materials or broader applications are rare so far. Therefore, this thesis will aim to investigate the optoelectronic properties of another promising while rarely investigated ABZ<sub>2</sub> material – NaBiS<sub>2</sub>, and also the potential of AgBiS<sub>2</sub> as near-infrared (NIR) photodetectors. In the first project of this thesis, NaBiS<sub>2</sub> nanocrystals (NCs) have been shown to exhibit extremely strong absorption, along with a comparatively sharp absorption onset. However, optical-pump-terahertz-probe (OPTP) measurements indicated that most free charge-carriers in NaBiS<sub>2</sub> NCs will be localised within a few picoseconds. These localised charge-carriers only exhibited low mobility of around 0.03 cm<sup>2</sup> V<sup>-1</sup> s<sup>-1</sup> and could not transport effectively even though they might be rather long-lived in NaBiS<sub>2</sub> and unaffected by intentionally-introduced defects. With help from density functional theory (DFT) calculations, all of these unusual characteristics in NaBiS<sub>2</sub> have been shown to closely associate with intrinsic cation disorder, which was also observed in AgBiS<sub>2</sub>. Although post-annealing is effective for improving cation inhomogeneity and enhancing absorption in AgBiS<sub>2</sub>, its effect on NaBiS<sub>2</sub> was found to be rather minor, which also indicated that the charge-carrier localisation process in NaBiS<sub>2</sub> could not be significantly mitigated after annealing. Based on the fundamental insights acquired in the first project, the possibility of further improving charge-carrier transport in NaBiS<sub>2</sub> NCs through ligand exchange treatment was investigated in my second project. Using a variety of correlated spectroscopic characterisation techniques, I found that NaBiS<sub>2</sub> NCs treated by inorganic iodide ligands had enhanced sum mobility and surface photovoltage (SPV) signals, which implies an improvement in the macroscopic charge-carrier transport. However, the ultrafast localisation process was still observed in these iodide-treated NaBiS<sub>2</sub> NCs, suggesting that their cation disorder was not greatly changed. At the same time, the defect capture rates were also found to be lower in the iodide-treated NaBiS<sub>2</sub> NCs based on my two proposed models for describing charge-carrier dynamics. As a result, solar cells based on these iodide-treated NaBiS<sub>2</sub> NCs could exhibit a peak external quantum efficiency (EQE) value over 50%, along with a power conversion efficiency exceeding 0.7%. Although this is an order of magnitude larger than previous reports, I found ion migration to be a limiting factor for NaBiS<sub>2</sub> devices from temperature-dependent transient current measurements, where a low activation energy of only 88 meV was extracted. In my third project, AgBiS<sub>2</sub> photodetectors were fabricated and characterised in depth. Aside from the broadband photo-response across from ultra-violet (UV) to near-infrared (NIR) region, AgBiS<sub>2</sub> photodetectors have demonstrated an extremely high cut-off frequency (f<sub>-3dB</sub>) on MHz order, indicating their great potential in applications requiring fast device response such as optical communications. The mechanism behind this fast response was studied, and a relatively long drift length compared to the AgBiS<sub>2</sub> film thickness is believed to be the key reason. Similar to NaBiS<sub>2</sub> devices, ion migration was also found easy in AgBiS<sub>2</sub> devices with an activation energy of 124 meV, which could lead to their increasing noise currents with time. Importantly, these noise currents could be also effectively suppressed when optimising the AgBiS<sub>2</sub> film thickness, in which a balance between large shunt resistant and cumulative quantity of defects should be reached. Finally, owing to the small bandgap of AgBiS<sub>2</sub> NCs (~1.2 eV), AgBiS<sub>2</sub> photodetectors could effectively monitor the heartbeat rates by probing the transmission change of blood vessels illuminated by NIR light, which has been widely used in the medical field owing to its deeper penetration in tissues. These three projects not only uncovered several remarkable optoelectronic characteristics of ABZ<sub>2</sub> materials, but also investigated possible methods to further alter these characteristics. Although ABZ<sub>2</sub> materials have shown great potential as light harvesters, it can be seen that both cation disorder (or charge-carrier localisation) and ion migration are still limiting the performance. More studies on the root causes of both phenomena, and how to effectively suppress their effects on the materials, would be hence crucial in the future work. With more understandings on this material class, we could expect more efficient, stable, and cleaner optoelectronic devices to be realised in the future.
Degree
thesis:*- Name dc:type.qualificationname
- Doctor of Philosophy (PhD)
- Level dc:type.qualificationlevel
- Doctoral
- Grantor dc:publisher.institution
- University of Cambridge
- Year dc:date.issued
- 2023
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Huang, Yi-Teng
- Advisor dc:contributor.advisor
-
- Rao, Akshay
Subjects
dc:subject × 5Rights
dc:rightsIdentifiers
dc:identifier.*- DOI dc:identifier.doi
- https://doi.org/10.17863/CAM.96957
- OAI identifier oai:identifier
- oai:www.repository.cam.ac.uk:1810/350430