Machine learning‐enabled biomimetic electronic olfaction using graphene single-channel sensors

Thu-P1-005

Presented by: Shirong Huang

Shirong Huang ¹, Alexander Croy ⁴, Bergoi Ibarlucea ^{1, 2}, Gianaurelio Cuniberti ^{1, 2, 3}

¹ Institute for Materials Science and Max Bergmann Center for Biomaterials, TU Dresden, 01062 Dresden, Germany, ² Center for Advancing Electronics Dresden (cfAED), TU Dresden, 01062 Dresden, Germany, ³ Dresden Center for Computational Materials Science (DCMS), TU Dresden, 01062 Dresden, Germany, ⁴ Institute for Physical Chemistry, Friedrich Schiller University Jena, 07743 Jena, Germany

Background: Olfaction is an evolutionary old sensory system, which provides sophisticated access to information about our surroundings. ¹Inspired by the biological example, electronic noses (e-noses) in combination with efficient machine learning techniques aim to achieve similar performance and thus to digitize the sense of smell. ^{2, 3} Objectives: In this work, the discriminative recognition of odors using graphene single-channel nanosensor-based electronic olfaction in conjunction with machine learning were investigated at room temperature. Experimental methods: Functionalized graphene-based single-channel nanosensor were prepared and sensing signal was acquired towards exposure to various odors. The fingerprint information of odors was then represented by feature vector extracted from sensing signal and applied to discriminate as well as identify odors by machine learning. Results: The developed prototype exhibits excellent odor discrimination (83.3%) and identification performance (97.5%) at room temperature, maximizing the obtained results from a single nanosensor. Upon exposure to binary odor mixture, the response features behave similarly to one of the existing individual odor component, mimicking the overshadowing effect in human olfactory perception. Conclusions: We present the excellent performance of graphene single-channel nanosensor based electronic olfaction in conjunction with machine learning. The developed platform may facilitate miniaturization of e-noses, digitization of odors, and distinction of volatile organic compounds (VOCs) in various emerging applications.

Sources of funding: We appreciate the funding support from VolkswagenStiftung (grant no. 9B396).

References:
1. Sarafoleanu, C.et.al. J Med Life 2009, 2 (2), 196-8.
2. Gardner, J. W.et.al. Sensors and Actuators B: Chemical 1994, 18 (1-3), 210-211.
3. Covington, J. A.et.al. IEEE Sensors Journal 2021, 1-1.