The microstructure and properties of thin (~100 nm) SnO 2 films with noble metals Pt, Pd, Au additives, grown by dc magnetron deposition are studied. It is shown that the introduction of additives into the bulk and the deposition of dispersed catalysts on the semiconductor surface make it possible to control the sensor parameters in pure air and upon exposure to reduction (CO, H 2 , CH 4 ) and oxidation (NO 2 ) gases. Possible mechanisms for the effect of Pt, Pd, Au on the bulk and surface properties of tin dioxide are dis cussed. The technological conditions for film growth, which provide the selective detection of low concen trations (10-100 ppm) of CO and H 2 , below explosive concentrations (0.5-2.5 vol %) of methane, and trace concentrations (0.05-5 ppm) of NO 2 are determined.

No abstract

The results of theoretical and experimental studies into the effect of water vapor on the electrical conductance of a gas sensor and the sensor response to hydrogen action are discussed. A relation describing the dependence of electrical conductance G 0 on absolute humidity in the pure air is derived using a hypothesis of the presence of space-charge regions depleted of electrons between the SnO 2 grains in a polycrystalline tin dioxide film. Due to dissociative chemisorption of water molecules, the energy-band bending at the SnO 2 grain interfaces decreases and the oxygen-vacancy concentration in the grains increases, resuling in an increase in G 0 . An equation for the sensor response to hydrogen action is derived (the G 1 /G 0 , ratio, where G 1 is the sensor conductance in a gas mixture containing molecular hydrogen). The expression describes the dependence of G 1 /G 0 on the hydrogen concentration n H 2 in the interval 50-6·10 3 ppm, band bending at the SnO 2 grain interface, and sensor temperature. The dependences of the sensor conductance, highest possible conductance, and energy-band bending on temperature and absolute humidity resulting from processing of the experimental data are in good agreement with the theoretical predictions.

No abstract

No abstract