A Statistical Approach to Materials Evaluation and Selection for Chemical Sensor Arrays
We present a generic approach for designing sensor arrays for a given chemical sensing task. First, we present a correlation-based metric to systematically assess the analytical information obtained from the conductometric responses of chemiresistive films as a function of their operating temperatures and material composition. We illustrate how this measure can also be used to test the reproducibility of signals obtained from sensors of equal manufacture. Next, complementing the correlation-based analysis, we employ a statistical dimensionality-reduction algorithm to visualize the multivariate sensor response obtained from sensor arrays. We adapt this method to quantify the discriminability of chemical fingerprints. Finally, we show how to determine an optimal set of material compositions to be incorporated within an array for individual species' recognition when practical constraints/tradeoffs on fabrication are also considered. We validate our approach by designing a microsensor array for the task of recognizing a chemical hazard at sub-lethal concentrations in complex environments.
KeywordsLinear Discriminant Analysis TiO2 Film Sensor Array Array Size Chemical Vapor Deposition Process
We acknowledge partial financial support of this project by the U.S. Department of Homeland Security, Science and Technology Directorate. BR was supported by a NIH(NIBIB)-NIST Joint Postdoctoral Associateship Award administered through the National Research Council. We thank Kurt Benkstein, Mike Carrier, Steve Fick, Jim Melvin, Wyatt Miller, Chip Montgomery, Casey Mungle, Jim Yost, Blaine Young, and Li Zhang for their valuable contributions to this project. We are grateful to Mark Stopfer for his helpful comments on an earlier version of this manuscript.
- 1.Mention of these and any other commercial products is strictly for provision of proper experimental definition, and does not constitute an endorsement by the National Institute of Standards and TechnologyGoogle Scholar
- 3.Pearce, T. C.; Sanchez-Montanes, M., Chemical sensor array optimization: geometric and information theoretic approaches, In Handbook of Machine Olfaction: Electronic Nose Technology; Pearce, T. C.; Schiffman, S. S.; Nagle, H. T.; Gardner, J. W., Eds. Wiley-VCH, Weinheim, 2002, 347–376.CrossRefGoogle Scholar
- 20.Meier, D. C.; Semancik, S., Effects of Materials Chemistry on Conductometric Sensor Signals. In 2005 Materials Research Society Meeting Boston, 2005 Google Scholar
- 22.Duda, R. O.; Hart, P. E.; Stork, D. G., Pattern Classification, 2nd edn.; Wiley-Interscience, New York, 2000, 115–121Google Scholar