In recent years, metal halide perovskites have emerged as a game changer in solar cell technology, achieving a remarkable power conversion efficiency of over 26.7%, comparable to the commercial silicon solar cell. Beyond their impressive performance in photoelectric conversion, metal halide perovskite materials have also been employed as semiconductors and ionic conductors, making them suitable for use in artificial synapses for information processing. Benefiting from low cost, simple fabrication process, and excellent optoelectronic properties, metal halide perovskite is a promising solution to the current energy crisis and the rapidly growing information technology. However, the significant stability challenge of perovskites should be addressed before they can be widely employed.

Originating from the inherently defective nature and soft ionic nature of polycrystalline perovskites, instability results from chemical degradation reactions, ion migration, and lattice strain under the attack of extrinsic environment factors such as moisture, oxygen, UV light, and temperature change. In addition to improving the encapsulation techniques to shield perovskite from extrinsic factors, intensive efforts have been devoted to improving the intrinsic stability of perovskites, using techniques including reduction of defect density by improving film quality and passivation, inhibition of chemical reaction by intrinsic encapsulation and elimination of reactive species, suppression of ion migration by physical barrier and relaxation of strain by stress absorption. This thesis aims to deepen our understanding of perovskite degradation pathways and explore strategies to improve perovskite stability, and ultimately applying these perovskites in solar cells and artificial synapses.

Among various strategies, defect passivation through interaction with additives is convenient and effective. However, the selection of additives is limited to high purity chemicals, resulting in a complex synthesis process and restricting material availability. In this thesis, based on the defect passivation approach, a natural material, soy protein with abundant functional groups of -C=O and -NH, was employed as a prototype for investigating its effect on PSC. The interaction between soy protein and perovskite was manifested in the coordination with Pb²⁺ and hydrogen bonding with I^-, which facilitate the high-quality film formation and suppress ion migration. Consequently, the phase and structure of perovskite remained stable under external stress of high humidity and temperature change. Furthermore, solar cells fabricated with these perovskites exhibited enhanced photovoltaic performance and improved stability under different aging conditions.

In a complex environment of UV light, moisture, and oxygen, severe chemical reactions between active species generated from external factors and perovskite cause significant damage to perovskite. Considering the pathways of degradation reaction, a radical scavenging strategy learned from nature was proposed, wherein active radicals are scavenged by melatonin, preventing the decomposition reaction of perovskite. Combining the effects of UV light absorption to mitigate the attack of detrimental UV light, defect passivation for reducing the sites for reaction, and energy level modulating for suppressing charge accumulation, melatonin enables perovskite to resist UV aging in air. Further, the application of anti-aging perovskite in solar cells promotes the device’s resistance to both UVA and UVB in air with a relative humidity of 30%.

Significant efforts have been dedicated to enhancing the stability of perovskite materials over the years. However, mainstream research is focused on optoelectronic devices, while the development of neuromorphic devices has lagged considerably. To address this gap, we constructed a novel two-terminal lateral synapse with stable triple cation-based perovskite. Due to excellent photoelectric properties, simple fabrication process, and intrinsic stability of the triple cation perovskite, the synapse exhibits typical synaptic characteristics with high light sensitivity, low energy consumption, wider substrate selection, and high stability. Moreover, the potential application of the synapse was demonstrated by emulating the classical conditioning and learning experience behavior and simulation of neural networks.

In summary, this thesis offers insights into the proposed approaches to improve the stability of perovskite through multifunctional passivating agents. Besides, this work opens up new perspectives on the development of stable perovskite-based artificial synapses. These findings provide guidance on the design and fabrication of robust perovskite-based optoelectronic and neuromorphic devices for practical future applications.