<![CDATA[The rapid expansion of silicon based Photovoltaic (PV) technologies continues to drive the global shift toward sustainable energy systems. However, the environmental implications across the full life cycle of PV modules particularly those associated with upstream silicon purification routes remain insufficiently examined. This study provides a comprehensive assessment of the environmental and process level impacts of Metallurgical Grade Silicon (MGS) and Upgraded Metallurgical Grade Silicon (UMGS), covering extraction, manufacturing, operation, and end of life stages. A process oriented Life Cycle Assessment (LCA) is conducted to analyze variations in carbon intensity, hazardous material use, and energy demand, complemented by comparative evaluations of monocrystalline and polycrystalline module production pathways. To enhance analytical precision, this study incorporates an AI-assisted predictive modeling framework using supervised machine learning to estimate Global Warming Potential (GWP) and identify key factors influencing emission variability. The AI-enhanced model reveals that electricity mix and purification route exert the strongest influence on GWP, and scenario simulations demonstrate that UMGS based processes can reduce upstream emissions by up to 89% under favorable energy conditions. Additionally, the study highlights future challenges related to increasing PV waste volumes between 2025 and 2030 and the need for improved recycling infrastructures. Overall, the integration of AI-based prediction with conventional LCA offers a more dynamic and adaptive evaluation of PV sustainability performance. The findings underscore the importance of renewable powered manufacturing, early adoption of low-energy purification technologies, and policy support to achieve long-term environmental and socio-economic benefits.]]>
Copyrights © 2025
