Visual smell fingerprinting with sensor arrays

In this communication, we propose a definition of a smell fingerprint as reliable and sufficient smell sensing data visualized for the easy and convenient distinguishing of the smell targets by a human. It would complement a strategy of collecting as much data as possible about particular smells (their volatile components) using exuberant analytical chemical and sensor techniques. The data for the smell fingerprinting were measured using a model MQ-sensor array detection system. The application of different non-specific sensors in the array helps to collect characteristic features of a particular smell, which includes the magnitude of responses, speed of the changes, and the relative ratios in the responses or response rates. The sensors can differentiate various odors of pure laboratory chemicals (ethanol and acetone) or complex mixtures, such as several kinds of alcohol-containing and alcohol-free beer. The smell fingerprints were visualized as semi-matrices containing colors presenting ratios of end-signals or sensor response rates. Both types of smell fingerprints allow visual distinguishing of beers studied.