(694d) Flux Response Technology (FRT) Applied in Zero Length Column Diffusivity Measurements

Intracrystalline (or
intraparticle) diffusion within porous materials is widely studied for its
useful and important applications in molecular separation, heterogeneous
catalysis, membrane technology, fuel cells, soil mechanics and petroleum
engineering to name but a few. The well known automotive catalytic converter
makes use of a catalytic monolith reactor in a ?honeycomb' configuration to
oxidise carbon monoxide and unburned hydrocarbons while reducing oxides of
nitrogen. Understanding the intracrystalline diffusion that occurs within the
washcoat and substrate material of monolith supported catalysts is vital to the
modeling and scale-up of reactors and consequently affects both performance and
cost effectiveness of such systems.

The purpose of this paper
is to report on the measurement of the intracrystalline diffusivity in the
washcoat of monolith samples using Flux Response Technology (FRT). In previous
studies, we have shown the versatility of the FRT method in measuring in
situ adsorption, reaction and desorption processes in flow reactors. The
FRT apparatus is an extremely sensitive tool which allows the
measurement of flowrate changes of the order of 10^-2
μl/min. By applying the Zero Length Column (ZLC) method to the analysis of
concentration perturbation induced ad- and desorption transients recorded by
the FRT technique, accurate rates of diffusion within washcoats can be arrived
at.

Shown in Figure 1a
is an example of a flux response profile obtained for a cordierite sample
coated with a washcoat of alumina/CeZrOx when subjected to repeated
concentration perturbations of propane in helium. For the investigation of
diffusivities within the washcoat of the cordierite samples, it is necessary to
monitor the release of the adsorbed species (C₃H₈ in this
instance) into the effluent stream in order to apply the ZLC analysis. The signal
of the DPT pressure responses in the flux response profile have been shown to
be directly proportional to the amount of probe molecules ad/desorbed onto or
from the surface of the porous adsorbent.

ZLC
response curves (nonlinear
regression) were fitted to experimental, dispersion corrected FRT data (shown in Figure
1b) and allow the
extraction of effective diffusivities of propane within the washcoat. At 25^oC an effective
diffusivity of 7.04 × 10^-10 m²/swas
determined for the above sample which
agrees well with published data measured by other macroscopic
techniques. In
particular, it is reassuring to observe that the calculated activation energy for
propane diffusion in alumina/CeZrOx compares well to that predicted by Granato et
al. based on theoretical molecular dynamics simulations of propane
diffusion in 13X, and pulse field gradient nuclear magnetic resonance (PFG-NMR)
data by Schwartz et al. The results also compare
well to the values of Sun et al. who
employed a Wicke-Kallenbach type method using a zeolite membrane and report diffusivities of propane
in silicalite in
the temperature
range
between 30 to 70^oC
as 4-6 10^-10m²/s (Table 1).

It
has been observed that propane diffusivity in 13X is much smaller when measured
with the ZLC approach as compared to PFG-NMR and it has been suggested that
this may be due to the PFG-NMR technique returning intra-crystalline
diffusivities whereas ZLC experiments result in the observation of combined
diffusion and desorption processes.

We
will show that with FRT these two processes can be compared and contrasted in
one experiment by analysing the dispersion corrected adsorption and desorption
transients in the FRT profile.