Phase-Dependent NO₂ gas-sensing performance of WO₃ nanostructures:An evidence-based quantitative re-analysis of published experimentaldata (2020–2025)

Metal-oxide chemiresistors often exhibit trade-offs among sensitivity, operating temperature, humidity tolerance, and recovery speed. This study provides a focused secondary quantitative re-analysis of published WO₃ NO₂-sensing data, centered on the open dataset. A common WO₃·H₂O precursor was converted at 200°C to monoclinic m-WO₃ (TF200) in a tubular furnace and cubic c-WO₃ (MF200) in a muffle furnace. Reported XRD, Raman, SEM, BET, optical, and gas-sensing results were re-examined without generating or imputing laboratory observations. m-WO₃ had a larger BET surface area (40.6 vs. 25.0 m² g⁻¹) and higher responses at 0.25, 0.50, 1.0, and 2.5 ppm NO₂ (116, 366, 1061, and 1322 vs. 83, 183, 511, and 780). Exploratory regressions based on only four published concentration-level means per phase (n = 4) gave linear R² values of 0.797 and 0.912 and power-law R² values of 0.895 and 0.953 for m-WO₃ and c-WO₃, respectively. Because raw replicate-level measurements were unavailable, these fits and their p-values are descriptive and exploratory rather than confirmatory statistical inference. At 2.5 ppm, response times were 239 s for m-WO₃ and 204 s for c-WO₃. Overall, the published dataset shows phase-dependent trade-offs among surface area, response magnitude, detection limit, and response/recovery kinetics.