Chemical Sensing

Artificial olfaction has the potential to perform as well as or even outperform canine olfaction because it can be manufactured inexpensively using microfabrication techniques with very high repeatability. Adding to their advantages over canine olfaction, these devices can be distributed onto networks with identical "training" being imprinted upon manufacturing, creating arrays that can facilitate plume modeling and chemical imaging. It has also been demonstrated that these devices are capable of sensing and identifying mixtures of odorants in complex backgrounds (Benkstein et al. 2014; Rogers, Benkstein, and Semancik 2012). It is important to state that the power of a dog's nose is not only in its sensitivity, but also in the fact that it is directly linked to the animal's brain, which processes, identifies, and contains guidance rules for following a dispersed scent. Cortana has experience developing an algorithm that can process electronic nose (e-nose) data. Further research would allow us to incorporate transport and dispersion models in a similar way as the dog brain-nose system for use in various applications such as identification and tracking of individuals.

Any material of interest, which has a vapor pressure at ambient conditions and can interact with other materials at an interface, has the potential to be sensed and identified by e-nose technology utilizing algorithms. It is also possible that these sensors can be designed to sensitively detect chemicals with lower vapor pressures than typical odorants, such as those found in modern and homemade explosives.


Benkstein, K. D., Rogers, P. H., & Semancik, S. Analytical Capabilities of Chemiresistive Microsensor Arrays in a Simulated Martian Atmosphere. Sensors Actuators B Chem.197, 280 – 291 (2014). doi:10.1016/j.snb.2014.02.088

Rogers, P. H., Benkstein, K. D. & Semancik, S. Machine Learning Applied to Chemical Analysis: Sensing Multiple Biomarkers in Simulated Breath Using a Temperature-pulsed Electronic-nose. Anal. Chem. 84, 9774 – 81 (2012). doi:10.1021/ac301687j