Optimization of Printed Polyaniline Composites for Gas Sensing Applications

Polyaniline (PANI) films are promising candidates for electronic nose-based IoT applications, but device performances are influenced by fabrication parameters and ambient conditions. Affinities of different PANI composites to analytes for gas sensing applications remain elusive. In this study, we investigate the material properties in detail for two different dopant systems: F4TCNQ and carbon black. Using a reproducibility-driven approach, we investigate different dopant concentrations in regard to their sensitivity and specificity towards five relevant markers for breath cancer diagnosis. We benchmark the system using ammonia measurements and evaluate limits of detection. Furthermore, we provide statistical analysis on reproducibility and pave the way towards machine learning discrimination via principal component analysis. The influence of relative humidity on sensor hysteresis is also investigated. We find that F4TCNQ-doped PANI films show improved reproducibility compared to carbon black-doped films. We establish and quantify a tradeoff between sensitivity, reproducibility, and environmental stability by the choice of dopant and concentrations ratios.