Sign in Subscribe
Sensors

Breakthrough in Toxic Gas Detection Using Machine Learning

Strategic Insights

— 1 min read

Researchers at Dartmouth College have made a groundbreaking discovery in the development of low-power, sensitive, and selective gas sensors. A chemiresistive sensor array, comprising of three conductive hexahydroxytriphenylene-based metal-organic frameworks (MOFs), has been designed to detect and differentiate parts-per-million (ppm) levels of toxic gases such as carbon monoxide, ammonia, sulfur dioxide, hydrogen sulfide, and nitric oxide. The sensor array has been paired with machine learning algorithms, enabling it to accurately predict gas compositions and identify individual materials responsible for analyte discrimination.

Key Takeaways:

  • The chemiresistive sensor array consists of three conductive MOFs (Ni, Cu, Zn) capable of detecting ppm levels of CO, NH, SO, HS, and NO.
  • Machine learning techniques such as principal component analysis and random forest classification confirm the ability to discriminate toxic gases.
  • Feature importance method applied to the classifier assigns importance scores to each sensor in the array, quantifying the impact of individual materials on analyte discrimination.
  • Spectroscopic investigations reveal how structural features of MOFs influence sensing performance and ascertain material-analyte interactions.
  • The research has been peer-reviewed and published in the journal ACS Sensors.

Statistics:

  • The sensor array can detect ppm levels of toxic gases such as CO, NH, SO, HS, and NO.
  • Machine learning algorithms enable accurate prediction of gas compositions and individual material identification.
  • 3 conductive MOFs (Ni, Cu, Zn) comprise the sensor array.
  • 90% accuracy achieved in discriminating toxic gases using machine learning techniques.
  • 10-fold cross-validation used to evaluate the performance of machine learning models.

Sources:

  • Metal-Organic Framework-Based Chemiresistive Array Paired with Machine Learning Algorithms for the Detection and Differentiation of Toxic Gases by Patrick Damacet et al., ACS Sensors, 2025.
  • Journal of Engineering, October 20, 2025, p 2381.
  • Amer Chemical Soc, 1155 16TH St, NW, Washington, DC 20036, USA (publisher contact information).
  • Dartmouth College, Dept. of Chemistry, Hanover, New Hampshire 03755, USA (contact for additional information).

Read more