Oxygen vacancies (VO) in metal oxide semiconductors
play an important role in improving gas-sensing performance of chemiresistive
gas sensors. Nonetheless, there is still a lack of clear understanding
of the inherent mechanism of the influence of oxygen vacancies on
gas sensing due to generally focusing on the concentration of VO. Herein, oxygen vacancies were rationally modulated in WO3 nanoflower structures via an annealing process, resulting
in a transformation of VO from neutral (VO
0) to a doubly ionized (VO
2+) state. Density
functional theory (DFT) calculations indicate that VO
2+ is significantly more efficient
than VO
0 for
NO2 detection in competition with atmospheric O2. Benefiting from a high concentration of VO
2+, the WO3-450 (WO3 annealed at 450 °C) sensor exhibits excellent sensing performance
with an ultrahigh sensitivity (3674.1 to 5 ppm NO2), superior
selectivity, and long-term stability (one month). Furthermore, the
sensor with the wide range of concentration detection not only can
detect NO2 gas with parts per million (ppm) but also can
detect NO2 with parts per billion (ppb) level concentration,
with a high sensibility reaching 2.8 to 25 ppb NO2 and
over 100 to 100 ppb NO2. This study elucidates the oxygen
vacancy mediated sensing mechanism toward NO2 and provides
an effective strategy for the rational design of gas sensors with
high sensing performance.