AI-Driven Optimization of Pulsed DC Sputtering for Enhanced Indium Tin Oxide Films
Open Access
- 10 October 2025
- journal article
- Published by iLearning Journal Center in International Transactions on Artificial Intelligence (ITALIC)
- Vol. 4 (1) , 85-94
Abstract
The performance of heterojunction silicon solar cells strongly depends on the quality of the Transparent Conductive Oxide (TCO) layer, particularly Indium Tin Oxide (ITO), which governs both lateral charge transport and optical coupling. Despite its industrial relevance, achieving low-resistivity and high transmittance ITO at low thermal budgets remains a manufacturing challenge. This study aims to identify and optimize the key pulsed DC magnetron sputtering parameters that govern the structural, electrical, and optical properties of ultra-thin ITO films suitable for heterojunction applications. A systematic experimental approach was implemented by varying discharge power, oxygen partial pressure, deposition temperature, and post-deposition annealing conditions using an inline industrial sputtering system. The results show that moderate discharge power, controlled oxygen incorporation, and substrate temperatures around 180 ◦C significantly enhance film densification and electron transport. Under the optimized parameter set, the ITO films achieved low resistivity in the range of (4.6–5)×10−4 Ω·cm and visible-light transmittance above 90%, while annealing at 160 ◦C further reduced defect density and stabilized electrical per- formance. These findings demonstrate that high-quality ITO can be produced within the thermal constraints of heterojunction silicon technologies and provide a robust foundation for integration into automated or AI-assisted sputtering control frameworks. Overall, this work supports the development of scalable, energy-efficient, and industrially viable fabrication routes for next-generation photovoltaic devices.Keywords
This publication has 32 references indexed in Scilit:
- Universal 3D‐Printing of Suspended Metal Oxide Nanowire Arrays on MEMS for AI‐Optimized Combinatorial Gas FingerprintingAdvanced Science, 2025
- Investigation of the dependence of fill factor and efficiency on gallium-doped silicon wafer characteristics using unsupervised learningEngineering Applications of Artificial Intelligence, 2025
- Silicon‐Integrated Next‐Generation Plasmonic Devices for Energy‐Efficient Semiconductor ApplicationsAdvanced Materials Technologies, 2025
- Artificial Intelligence‐Driven Approaches in Semiconductor ResearchAdvanced Materials, 2025
- Design Strategies and Emerging Applications of Perovskite‐Based SensorsSmartMat, 2025
- Advancements in P‐type Wide and Ultra‐Wide Bandgap Oxide Semiconductors for Prospective ApplicationsPhysica Status Solidi (RRL) – Rapid Research Letters, 2025
- AI‐Generated Ammonium Ligands for High‐Efficiency and Stable 2D/3D Heterojunction Perovskite Solar CellsAdvanced Materials, 2025
- A Comparative Study of Silicon Heterojunction Solar Cells Utilizing a Double‐Layer Anti‐Reflection Coating on the Front SidePhysica Status Solidi (a), 2024
- Silicon heterojunction solar cells achieving 26.6% efficiency on commercial-size p-type silicon waferJoule, 2024
- Evaluating the Practical Efficiency Limit of Silicon Heterojunction–Interdigitated Back Contact Solar Cells by Creating Digital Twins of Silicon Heterojunction Solar Cells with Amorphous Silicon and Nanocrystalline Silicon Hole Contact LayersPhysica Status Solidi (a), 2024