Tin dioxide gas sensors. Part 2.—The role of surface additives
It has been shown earlier (J. F. McAleer, P. T. Moseley, J. O. W. Norris and D. E. Williams, J. Chem. Soc., Faraday Trans. 1, 1987, 83, 1323) that the gas response of porous pellets of tin dioxide depends crucially on electronic surface states involving adsorbed oxygen, and on the rates of combustion reactions involving the gases to be detected. It is expected, therefore, that the use of surface additives capable of either pinning the Fermi level of the tin dioxide or altering the rate of combustion (depending on choice of material and temperature) would profoundly affect the gas sensor response of the material. This is found to be the case. Studies of the influence of surface additions of precious metals and of metal oxide particles on the gas response behaviour of tin dioxide are described. The use of precious metals at temperatures near to ambient imposes an oxygen-independent Schottky barrier on the tin dioxide surface and results in a distinct low-temperature response to carbon monoxide, probably by adsorption on the precious metal modifying the surface potential (and hence the Schottky barrier). This particular effect is expected only when the precious metal particles distributed over the oxide surface are extremely small (1–10 nm): the effect disappears if the catalyst particles are aggregated by heating. The distribution of particles of a foreign oxide (here Ag2O or ZnO) on the surface of the tin dioxide also appears to pin the surface-energy levels to those of the additives, and tin dioxide treated in this way exhibits gas-sensing properties that are modified to resemble those of the additive oxide in bulk. At higher temperatures, especially in the presence of a precious-metal catalyst, the gas is completely oxidised within a thin outer shell of the specimen and consequently the response to the gas of the measured resistance of the pellet disappears. Indeed, a response to the reaction products (H2O and CO2) is obtained.