Tailoring Molecular Space to Navigate Phase Complexity in 2D and Quasi-2D Halide Perovskites

2D and quasi-2D halide perovskites (HPs) exhibit promising optoelectronic functionalities and stability as well as thermodynamic feasibility in low-temperature synthesis suited to photovoltaics (PVs) applications – particularly as a front sub-cell in tandem devices. However, the successful synthesis of quasi-2D HPs is exceptionally challenging, due to difficulty in controlling the multiple phase emergence and their spatial distributions across the film matrices. Resolving this multi-faceted problem – vital for utilization of quasi-2D HPs – necessitates a novel approach for understanding and regulating the chemistry between the molecular spacers and inorganic building blocks during the crystallization process. However, the chemical variability of the spacer cations offers an infinite molecular space, making the single-loop materials design and optimization difficult. Additionally, the chemistry in the vast molecular space has an inherently discrete landscape rather than continuous, as each spacer has a distinctive molecular structure that is distinguishable from others at an atomic level. This indicates that optimizing the optoelectronic functionalities of the quasi-2D HPs requires seamless navigation of such complex chemical space. From the machine learning (ML) perspective, manifesting the desired chemistry by navigating such discrete molecular spaces with the classical, single-directional optimization strategies remains challenging. In this talk, I will present a novel approach to navigate spacer chemistry from the vast molecular space to design high-performance Cs based quasi-2D HPs PVs by leveraging high throughput exploration and advanced ML techniques. This allows us to overcome the inherent challenges in designing complex materials, thereby realizing accelerated materials discovery. Additionally, I show that with judicious selection of spacer cations, 2D HP can manifest self-assembly of twisted Moire structure, which has not been observed from conventional 2D HP systems with linear spacers. These studies exemplify how a high-throughput autonomous experimental workflow effectively expedites discoveries and processing optimizations in complex materials systems with multiple functionalities, facilitating their realization in scalable optoelectronic manufacturing processes.