(560ia) Kinetic Modelling of the Ammonia Temperature Programmed Desorption of ZSM-5

Kinetic
Modelling of the Ammonia Temperature Programmed Desorption of ZSM-5

font-family:" times new roman>Rebecca L. Gibson^a,b, Prof Mark
Simmons^c, Prof Athanasios Tsolakis^d ,Prof
Hugh Stitt^a, Dr Stephen Schuyten^b, Robert Gallen^a

margin-bottom:.0001pt;text-align:center;text-indent:-18.0pt"> font-family:" times new roman>a.    
font-family:" times new roman>Johnson Matthey, Belasis Avenue,
Billingham, Stockton-on-Tees, TS23 1LB, UK.

margin-bottom:.0001pt;text-align:center;text-indent:-18.0pt">b.     Johnson
Matthey, Eli Whitney Blvd, Savannah, GA, 31408, USA.

text-align:center"> font-family:" times new roman>c. School of Chemical Engineering,
University of Birmingham, Edgbaston, B15 2TT, UK.

text-align:center"> font-family:" times new roman>d. School of Engineering, University of
Birmingham, Edgbaston, B15 2TT, UK.

Temperature
programmed desorption (TPD) is a commonly used thermal analysis technique for
investigating the strength of adsorption sites [1].  The
adsorption energies of these sites are however currently rarely quantified from
TPD data [2]. 
These adsorption sites are important features in the performance of a sorbent
or catalyst and can be influenced by formulation and chemical structure. 
Characterising these sites would offer improved understanding and enable predictive
modelling of performance.

" times new roman> 

It
is common for TPD results to feature more than one peak, which makes the
resolution of these peaks an important part of processing the results. 
Historically, the shapes of these peaks, as well as their overall contribution
to the extent of reaction would be estimated before mechanistic modelling was
carried out [3]. 
The inherent assumptions can be a source of error.  In the approach used in
this study, no assumptions are made about the peak shapes and contributions,
rather, the Sestak Berggren model (Eq. (1)) [4]
is used to fit the data directly,

where
 is
the extent of reaction, A is the pre-exponential factor, E_a
is the adsorption energy, R is the universal gas constant, T_b
is the base temperature of the peak, T is the temperature, n and m
are exponent parameters and E_v is a contribution term.

This
model is a numerical fit and the exponent parameters (n and m) indicate
which mechanisms could be plausible for the system under investigation [4]. 
These are analysed, and model discrimination carried out, allowing the statistically
adequate fitting of the kinetic parameters, such as adsorption energy. 
Additionally, this gives information about the rate limiting step occurring
during the process [2].

This
Sestak-Berggren equation, along with the mechanistic models it represents, captures
the behavior of kinetically-limited, irreversible chemical reactions. Transport
phenomena and reversible reactions are not represented by this modelling
scheme.

In
silico font-family:" times new roman> data has been previously used for model
validation, allowing rigorous methods to be developed [5].
This methodology has then been applied to a commercially available ZSM-5;
testing the method on experimental data. Although this is a widely studied class
of material, extracting kinetic information using ammonia TPD is still
uncommon.

The
results of the Sestak-Berggren modelling provide a good statistical fit for the
two overlapped thermal events present for these experiments. The statistical
metrics used to access the quality of fit for this parameter estimation study
are: R-squared, 95% confidence intervals and residual sum of squares. This
Sestak-Berggren fitting then allows for the prediction of a statistically relevant
kinetic triplet (pre-exponential factor, activation energy and kinetic
mechanism), as is the intention of the method.

In
order to discriminate between the many mechanistic models possible for this
desorption, multiple ramp rate experiments must be used [6].
In this case, five experiments were conducted at ramp rates of 2, 4, 6, 8 and
10 K min^-1. The parameter estimation was carried out using all
datasets.

Although
the Sestak-Berggren modelling provides a suitable statistical fit for these experiments,
a shift in the fitting has been observed which varies with ramp rate as shown
in Figure 1. This shift is not consistent with mass transport effects and has
been attributed to uncaptured reversibility in the system [7].

This
study concludes that the Sestak-Berggren model can successfully indicate
mechanisms for multiple overlapped thermal events, such as those observed in
the ammonia TPD of ZSM-5. However, the fit of each data set should be analysed individually
to check for trends. In this case it is believed that the drift in model fit
with ramp rate is a result of the adsorption process reversibility. Further
model development to capture reversibility is underway.  

6.0pt;margin-left:0cm;text-align:justify;line-height:12.0pt;page-break-after:
avoid">Acknowledgements

justify;line-height:12.0pt"> font-family:" times>Rebecca Gibson has been funded by the EPSRC Centre
for Doctoral Training in Formulation Engineering at the University of
Birmingham (EPSRC grant number EP/L015153/1) and Johnson Matthey.

References

[1]  P.
J. Barrie, ‘Analysis of temperature programmed desorption (TPD) data for the
characterisation of catalysts containing a distribution of adsorption sites’, Physical
Chemistry Chemical Physics, vol. 10, no. 12, p. 1688, 2008.

" times new roman>[2]  M. Niwa, N. Katada, M. Sawa, and Y. Murakami,
‘Temperature-Programmed Desorption of Ammonia with Readsorption Based on the
Derived Theoretical Equation’, The Journal of Physical Chemistry, vol.
99, no. 21, pp. 8812–8816, May 1995.

" times new roman>[3]  Q.-L. Yan et al., ‘Decomposition kinetics
and thermolysis products analyses of energetic diaminotriazole-substituted
tetrazine structures’, Thermochimica Acta, vol. 667, pp. 19–26, Sep.
2018.

" times new roman>[4]  J. Sestak and G. Berggren, ‘Study of the kinetics
of the mechanism of solid-state reactions at increasing temperatures’, Thermochimica
Acta, vol. 3, pp. 1–12, 1971.

" times new roman>[5]  R. Gibson, M. Simmons, A. Tsolakis, H. Stitt, J.
West, and R. Gallen, ‘Kinetic Modelling of Ammonia Temperature Programmed
Desorption Using the Sestak-Berggren Equation: An In Silico Study.’, Industrial
& Engineering Chemistry Research, vol. Under Review.

" times new roman>[6]  M. Maciejewski, ‘Computational aspects of kinetic
analysis. Part B: The ICTAC Kinetics Project the decomposition kinetics of
calcium carbonate revisited, or some tips on survival in the kinetic minefield’,
Thermochimica Acta, p. 10, 2000.

" times new roman>[7]  L. Lietti, I. Nova, S. Camurri, E. Tronconi, and
P. Forzatti, ‘Dynamics of the SCR-DeNOx reaction by the transient-response
method’, AIChE Journal, vol. 43, no. 10, pp. 2559–2570, Oct. 1997.