Bottom-up designing nanostructured oxide libraries under a lab-on-chip paradigm towards a low-cost highly-selective E-nose

Background: The multisensor concept has been developed as a powerful alternative to well-known gas-analytical instrumentation for applications where a fast but accurate and reliable assessment of the environment is required. The concept follows a biology-inspired approach where the selectivity towards various gases/odors is attained via pattern recognition of multisensory signal vectors. Herein, we discuss how to design a selective multisensor library based on various metal oxide nanostructures like a lab-on-chip using a simple but efficient bottom-up growth of materials over the multi-electrode chip under robust dc electrochemical protocols. Results: In addition to a conventional growth of oxide layers over the metal electrodes, we show that the fine nanowall-like oxide structures appear as a quasi-matrixed percolation film over the SiO2 substrate surface in the inter-electrode gaps to constitute a chemiresistive film. We have tested two directions while applying the technique to grow Co, Ni, Mn, and Zn oxides to develop on-chip sensor arrays of, (i) monoxide type employing the oxide films with gradual change of growth time, and (ii) multi-oxide type based on the four oxides. The materials were thoroughly characterized by electron microscopy, X-ray diffraction, thermogravimetric analysis, and X-ray photoelectron spectroscopy/mapping to prove the composition and structure. Among tested oxides, ZnO readily appears not only at the electric potential-targeted chip zone but also in other areas to dope the films for yielding heterojunctions with other oxides that enhances a variability of functional properties in the on-chip sensor array. The gas-sensing performance of the chips has been tested versus various chemically akin alcohol vapors at the sub- and low ppm range of concentrations in a mixture with air. Significance: We show that the grown oxide nanostructures exhibit a high-sensitive chemiresistive signal which allows one to build a multisensor vector signal, selective to the kind of alcohols, even at sub-ppm concentrations. Moreover, the multi-oxide library yields options for a superior selectivity under LDA metrics than the gradient-grown mono-oxide one due to the versatility of materials while the low-cost growth protocols remain to be the same in both cases. The delivered method to produce multisensor arrays allows one producing low-cost but efficient electronic nose units for numerous applications.