Performance Enhancements for Chemiresistive Electronic Noses Based upon Materials, Temperature Modulation and Signal Processing

Chemical sensors capable of detecting, identifying and quantifying trace chemical species in the air would be beneficial for a broad range of application areas from the laboratory, to workplace safety, to medical diagnostics, to homeland security and environmental monitoring, and even extending to planetary exploration. Our recent research has focused on studying the analytical capabilities of electronic nose-type sensors that utilize arrays of chemiresistive microsensor elements. In particular, we have explored the utility of microsensor arrays populated with varied sensing materials and employing temperature modulation to further enhance the analytical content of the measured signal streams. It is important that these aspects work together to improve the performance of the electronic nose, and we will illustrate that with several examples.

One scenario demonstrates the discrimination of several volatile organic compounds at varied concentrations using a temperature-modulated electronic nose. We have studied both inorganic and polymer-based sensing materials in the past, which were integrated with the microsensor array using a variety of approaches [1]. Here, the materials that we employed in the arrays are based upon metal oxide semiconductors with varied morphologies (including: tin oxide nanowire clusters, copper oxide films, and porous indium oxide films) deposited using semi-automated microcapillary pipetting. While each material is cross-selective to the targeted analytes, their ensemble performance with temperature modulation shows enhancements for analyte and concentration discrimination when evaluated using Linear Discriminant Analysis. Concentrations for the analytes (acetone, toluene and chlorobenzene) range from 1 µmol/mol to 80 µmol/mol, and these chemicals are presented to the microsensor array in air backgrounds either alone or in mixtures with varied levels of humidity. 

A second example demonstrates different temperature-modulation approaches to improve the performance characteristics of the microsensors. We have expanded our use of a relatively simple bi-level program to enhance the speed of the sensor responses [2], testing the approach for use with different materials and different types of analyte molecules. The utility of the approach for attaining faster responses is shown in Figure 1. Here the raw data collected from a microsensor element based upon a porous indium oxide film are shown for two operating conditions: fast temperature modulation and isothermal. As seen, the response time for the modulated operation is ≈ 2× faster (t₉₀ = (2.0 ± 0.2) s vs. (4.1 ± 0.2) s) than for the isothermal operation when responding to 40 µmol/mol of chlorobenzene in dry air.

Finally, we conclude by touching on several examples of using more complex ramp/base temperature programs that cover a broader range of temperature-induced sensor-analyte interactions. Here, we examine the potential of the different temperatures in the modulation program for yielding sensor responses that show resilience versus confounding chemicals (e.g., humidity, other analytes) or aging and drift, or those temperatures that maximize the overall response magnitude for particular analytes. This work emphasizes our collaborative research efforts, which have examined a variety of approaches, including those inspired by biological olfaction, to enhance the analytical capabilities of the chemiresistive electronic noses. Ultimately, for the flexibility needed to impact different applications, as well as for obtaining optimized performance, it is important the materials selections, operational mode and data analysis work together synergistically.

References
[1] K.D. Benkstein, C.J. Martinez, G.F. Li, D.C. Meier, C.B. Montgomery, S. Semancik, "Integration of nanostructured materials with MEMS microhotplate platforms to enhance chemical sensor performance," J. Nanopart. Res., Vol. 8, 2006, pp. 809-822.
[2] A. Vergara, K.D. Benkstein, C.B. Montgomery, S. Semancik, "Demonstration of Fast and Accurate Discrimination and Quantification of Chemically Similar Species Utilizing a Single Cross-Selective Chemiresistor," Anal. Chem., Vol. 86, 2014, pp. 6753-6757.