Predictive modelling and experimental analysis of radiation-resistant MXene-silicon heterojunction solar cells

MXene-silicon heterojunction solar cells (Ti₃C₂T_x-Si) have become a potential candidate for radiation-tolerant photovoltaic (PV) applications, specifically in harsh space conditions. In this study, a combination of predictive modeling and experimental verification is used to evaluate the degradation characteristics of Ti₃C₂T_x - Si solar cells under proton and electron irradiation. The study indicates that Ti₃C₂T_x and Si heterojunction devices can be used to achieve a superior yield, maintaining over 84% of the original power conversion efficiency (PCE), as compared to conventional Si solar cells that can only maintain 55% efficiency under the same exposure. X-ray diffraction (XRD), Raman spectroscopy, and Atom Probe Tomography (APT) reveal the interfacial stability of the MXene layers that alleviates the radiation induced structural damage. A physics-based digital twin (DT) framework is applied to the synthetic degradation data, which achieved a high prediction accuracy of 90% (R² > 0.96) using a training model based on Random Forest Regression (RFR), which in turn can aid to forecast the device performance with better accuracy. The PV performance parameters like the open circuit voltage (V_OC), short circuit current density (J_SC), fill factor (FF) and PCE are studied as functions of mission-relevant exposure durations. This hybrid approach of machine learning (ML)-aided modeling, combined with in situ experimental verification develops a solid framework for real-time health monitoring and lifetime prediction of PV devices, making it highly suitable for deployment in radiation-intensive orbital environments.Clinical Trial Registration: Not applicable. This study is not a clinical trial and does not involve any medical intervention or human subjects.