Our group research various functional materials that can be integrated into perovskite solar cells to further improve its performance in terms of power conversion efficiency and operational stability. One high potential material is rare-earth metal doped upconversion nanoparticles, which is known for its ability to convert near-infrared radiation photons which are non-useable for most perovskite photovoltaic devices into useable visible light photons, which improves power conversion efficiency of perovskite photovoltaic devices. These nanoparticles also can improve the performance of the photovoltaic devices by improving the crystallinity and morphology of the material films within the devices, making it a multi-functional additive. 

Various methods of enhancing the upconversion emission intensity will also be explored in hopes of further improving the performance gains for photovoltaic devices. Utilizing prior research work done as a foundation, the aim is to successfully integrate rare-earth metal doped upconversion nanoparticles directly into the perovskite layer or other functional layers and demonstrate the direct performance gain from the upconversion effect of these nanoparticles. 

Our research focuses on the fabrication and optimization of high-efficiency perovskite solar cells, emphasizing engineering of the charge-transport layer, interface modification, and integration of functional nanomaterials. Building on prior doctoral work, current efforts aim to advance device efficiency, scalable fabrication, and long-term stability for practical deployment. The goal is to develop scalable and robust perovskite photovoltaic devices suitable for practical indoor and outdoor operation applications.

In our research group, machine learning frameworks are proposed to optimize the performance of perovskite photovoltaic (POPV) devices. Different algorithm models and kernel functions will be developed and tailored for the perovskite solar cell domain to optimize various experimental variables, ranging from material composition to fabrication process parameters. The aim of incorporating machine learning frameworks is to find optimal solutions with minimal experimentation, thereby moving from trial-and-error to informed decision-making.

Our group employs Density Functional Theory (DFT) to investigate the properties of perovskite photovoltaic systems. To model the configurational complexity of doped and mixed-cation, we utilize the Alloy Theoretic Automated Toolkit (ATAT) alongside our first-principles calculations.

Since our approach is ab initio, we can simulate realistic material environments without empirical input, offering accurate predictions that directly guide the experimental design of the next generation of Perovskite Photovoltaic.