%L discovery1310248
%T Modelling of gas transport in porous zeolite-modified discriminating gas sensors
%D 2011
%X The ability to distinguish effectively between a range of gases in a reliable, repeatable
manner is of major interest with both scientific and commercial relevance.
Semiconducting metal oxide gas sensors have a long life-span, are inexpensive and are
highly sensitive; however, they are generally found to lack a desired level of selectivity.
One highly viable approach for enhancing the selective power of such devices is the
addition of a transformation layer. This will typically be a micro- or meso-porous, solid
which will act to transform the analyte gas stream by some means. Here the use of
zeolite compounds for this purpose is investigated.
Different theoretical models are used to probe the dependency of the response of a
porous metal oxide sensor on the transport properties of gas through the device,
including through an additional zeolite layer. Through the use of a force-field based
method, shape and size selective adsorption is predicted and used to justify
experimental results of zeolite modified sensors, for example, the reduction of response
to linear hydrocarbons as the chain-length is increased. However, the limit of such
calculations is also realised such that this approach is unlikely to provide an adequate
predictive tool for selecting a suitable zeolite for a particular gas sensing task.
Following this, a model based on the method of diffusion eigenstates has been
developed to calculate bulk effective diffusivities and rate constants for porous systems
representing both the sensor and zeolite porous layers. The effective properties are
found to depend strongly on the microstructure, the partitioning between phases and
diffusion coefficients of the different phases. The effective parameters are then
interpreted in terms of sensor response by solving the one-dimensional diffusionreaction
equation for a simple two-layered macroscopic geometry. The method of finite
differences is used to find the concentration profile which generates a response on
interaction with an electric field established between two electrodes. The concentration
profile and hence the response depends on the balance of diffusion and reaction of the
analyte gas within both the sensor and zeolite layers. It is shown how the response can
be explored to expose such differences by firstly looking at both the steady state
response and response time and also by varying the positioning of the electrodes used
to measure the response. Good correlation with experimental response data is
demonstrated, supporting the importance of the diffusion-reaction properties modelled
to the sensing mechanism, and the potential of developing a predictive tool based on the
models presented is discussed.
%I UCL (University College London)
%A S.J. Dungey