Factors affecting low temperature performance of zirconia gas sensors.

A reduction in the operation temperature of zirconia ceramic gas sensors is highly desirable for a number of practical reasons. This work seeks to investigate the factors that prevent a reduction in operation temperature and propose methods by which these may be resolved.
A novel approach to sensor fabrication has been developed and employed with the advantage of reduced device complexity that should lead to subsequent cost and reliability benefits. Leakage rates in these devices have been shown to be small and electrochemical in origin. Leakage was greater than reported for gold seal devices, partly due to increased electrode activity. The flexibility of device configuration allows a variety of sensor
geometries and functions to be realised. This flexibility led to the characterisation of sensors at the upper and lower ends of measurement range and the identification of
deviations from theoretical performance. These deviations have been reconciled with theory extended to cover these limits.
Such sensors are known to be sensitive to reducible gas species such as CO2 and H2O with a second limiting plateau allowing quantification of these gases. Such analysis capabilities have been found to be extended by incorporating a second pair of electrodes. These effects
have not previously been reported. Sensors have been shown to be more sensitive to H2O than to CO2.
To investigate the low temperature response of sensors, a variety of techniques and analyses have been developed and are employed with varying success. Impedance spectroscopy was by far the most useful and revealing tool but this is a function of the highly developed hardware and sophisticated control and analysis software bought as a complete system. Gas step changes and current / voltage sweeps were also useful as comparative techniques but could not separate out component mechanisms. Scanning electron microscopy proved to be a vital tool as it allowed vital information to be obtained concerning electrode and electrolyte microstructure. Again this is a function of a highly
developed and sophisticated instrument. The techniques of pressure and concentration modulation were limited in terms of ease of use, measurement range and results interpretation. The main drawbacks were limited frequency ranges and laborious data collection and analysis. They do both however show large potential for improvement. Both
amperometric and potentiometric sensors response rates were analysed with a variety of noble metal electrodes using each technique.
Electrode material proved to have a marked effect on sensor performance with the best results obtained with silver and electro-deposited platinum. Scanning electron microscopy
of silver showed that a finely divided and openly porous electrode was not required for high performance contrary to expectations. This is thought to be due to the solubility of
oxygen in this metal. With platinum however, the improved microstructure is thought to be a signifîcant factor in electro-deposited and cermet electrode performance.
Response rates in amperometric sensors did not show any significant temperature dependence although a restriction in measurement range was observed. Response rates were suspected to be mainly influenced by sensor geometry whilst measurement range was a function of sensor geometry, electrolyte conductivity and electrode activity. Improved
electrolytes will provide improvements and may come in the form of attention to the YSZ system or by employing an alternative ion conductor such as ceria. Close attention to
sensor dimensions provides possibility for enhancements. In amperometric devices for instance a long, thin diffusion barrier is required leading to a small internal cavity with a large electrode surface area and a thin electrolyte membrane.

Thesis submitted in partial fulfilment of the requirements for the degree of Doctor of Philosophy.