Benchmarking Strategies of Sustainable Process Chemistry Development: Human-Based, Machine Learning, and Quantum Mechanics

• Published in: Organic process research & development Vol. 28; no. 7; pp. 2885 - 2895
• Main Authors: Almeida, A. Filipa, Branco, Sofia, Carvalho, Luisa C. R., Dias, Andre Raposo Moreira, Leitão, Emília P. T., Loureiro, Rui M. S., Lucas, Susana D., Mendonça, Ricardo F., Oliveira, Rudi, Rocha, Inês L. D., Sardinha, Joao, Silva, Saúl, Sobral, Luís M. S., Lourenço, Nuno M. T., Valente, Pedro C.
• Format: Journal Article
• Language: English
• Published: American Chemical Society 19.07.2024
• Subjects: quantum mechanics; reaction optimization; algorithmic process optimization; organic chemistry; random forest; Bayesian optimization; machine learning; artificial intelligence
• ISSN: 1083-6160; 1520-586X
• DOI: 10.1021/acs.oprd.4c00164

This study benchmarks diverse strategies in sustainable process chemistry development, ranging from human subject matter expertise to advanced computational models, including machine learning, Bayesian optimization, and quantum mechanics simulations. Through a “virtual laboratory” case study simulating a Pd-catalyzed C–H arylation reaction, the efficiency, sustainability, and practical application of these methodologies were compared. The study highlights the nuanced interplay between traditional expertise and computational tools, offering insights into their complementary roles in accelerating development and achieving green-by-design principles in pharmaceutical synthesis. Our findings suggest that no single approach universally outperforms others; instead, a hybrid strategy leveraging both human intuition and computational power appears to be the most promising approach when combining powerful tools in the complex field of modern organic synthesis.