Selective detection of naphthalene with nanostructured WO 3 gas sensors prepared by pulsed laser deposition

2016 
Abstract. Pulsed laser deposition (PLD) at room temperature with a nanosecond laser was used to prepare WO 3 layers on both MEMS microheater platforms and Si/SiO 2 substrates. Structural characterization showed that the layers are formed of nanoparticles and nanoparticle agglomerates. Two types of layers were prepared, one at an oxygen partial pressure of 0.08 mbar and one at 0.2 mbar. The layer structure and the related gas sensing properties were shown to be highly dependent on this deposition parameter. At an oxygen pressure of 0.2 mbar, formation of e -phase WO 3 was found, which is possibly contributing to the observed increase in sensitivity of the sensor material. The gas sensing performance of the two sensor layers prepared via PLD was tested for detection of volatile organic compounds (benzene, formaldehyde and naphthalene) at ppb level concentrations, with various ethanol backgrounds (0.5 and 2 ppm) and gas humidities (30, 50 and 70 % RH). The gas sensors were operated in temperature cycled operation. For signal processing, linear discriminant analysis was performed using features extracted from the conductance signals during temperature variations as input data. Both WO 3 sensor layers showed high sensitivity and selectivity to naphthalene compared to the other target gases. Of the two layers, the one prepared at higher oxygen partial pressure showed higher sensitivity and stability resulting in better discrimination of the gases and of different naphthalene concentrations. Naphthalene at concentrations down to 1 ppb could be detected with high reliability, even in an ethanol background of up to 2 ppm. The sensors show only low response to ethanol, which can be compensated reliably during the signal processing. Quantification of ppb level naphthalene concentrations was also possible with a high success rate of more than 99 % as shown by leave-one-out cross validation.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    35
    References
    15
    Citations
    NaN
    KQI
    []