Abstract
Abstract
Two-dimensional materials exhibit extraordinary properties for next-generation electronics, optoelectronics, and biosensors, yet industrial adoption remains limited by challenges in scalable chemical vapor deposition (CVD) synthesis and consistent quality control. This review examines how artificial intelligence (AI) and machine learning (ML) are shifting CVD synthesis from empirical trial-and-error toward data-driven manufacturing. Key engineering bottlenecks address large-area uniformity, nucleation control, batch-to-batch reproducibility, and precursor delivery. ML workflows for feature engineering, predictive modeling, and real-time in situ monitoring were also reviewed, while AI-assisted quality control via Raman/photoluminescence classification (random forest, XGBoost) and CNN-based image analysis is discussed alongside emerging autonomous laboratories, closed-loop feedback, and FAIR data ecosystems, offering a practical roadmap for reliable, reproducible 2D material manufacturing.