Artwork by Sampson Wilcox
Jaekang Song, Dong-Ha Kim, Jan Tiepelt, Young-Moo Jo, Graham McGrath, Michael Song, Tianyang Chen, Jiande Wang, Alicia Coto, Sundar Palani, Dooyong Koh, Samuel Fuller, Max Shulaker, Mircea Dincă & Marc Baldo
DOI: 10.1038/s44460-026–00037‑z
Abstract:
Chemical gas sensing is essential for healthcare, environmental monitoring and industrial safety, yet current sensors lack sensitivity, selectivity and scalability. Carbon nanotube field-effect transistors (CNFETs) offer high surface area and low-power operation, but they traditionally provide limited chemical discrimination. Here we report an integrated sensing platform that combines CNFETs with conductive metal–organic frameworks and catalytic metal nanoparticles to achieve tunable selectivity and enhanced sensitivity. The hybrid architecture boosts response by up to two orders of magnitude and enables on-chip pattern generation for robust gas classification. As proof of concept, we apply the platform to the classification of clinically relevant bacteria and yeast species by analysing the volatile organic compounds emitted from cultures grown on agar plates with 95% accuracy, using a portable measurement set-up. By integrating functionalized CNFETs into a commercial foundry-derived system, this work introduces a rapid, cost-effective and scalable gas sensing approach for real-world biomedical and industrial sensing applications.