Simulation of The Industrial Fixed Bed Catalytic R
Simulation of The Industrial Fixed Bed Catalytic R
Simulation of The Industrial Fixed Bed Catalytic R
net/publication/231366287
CITATIONS READS
42 1,875
3 authors, including:
SEE PROFILE
Some of the authors of this publication are also working on these related projects:
Kinetics and Chemical Reactors Modeling, Simulation and Optimization View project
All content following this page was uploaded by Professor Babiker Karama Abdalla on 19 August 2018.
Ronald Hughes
Chemical and Gas Engineering Department, University of Salford, Salford, England
C6H5CH,CH3 - C6H5CHCH, + H,
the fluxes by solving the material balance equations
simultaneously with the dusty gas model equations.
While problems of multicomponent diffusion and re-
action in porous catalyst generally give rise to pressure
- PsTPH,/KEB)
ri = ~ ( P E B gradients within the catalyst pellet, it is possible to neglect
the resulting viscous flow contributions to the component
KEB = exp(-AFo/R T ) flux equations. It was also concluded by many investi-
gators that for realistic parameter values viscous flow does
AFo=a+bT+cF not contribute appreciably to pressure variation and
effectiveness factor calculations in porous catalyst pellets
(Evans et al., 1961,1962; Mason et al., 1967; Kehoe and
side reactions Aris, 1973; Haynes, 1978; Burghardt and Aerts, 1988).
Solution of the reactor model differential equations (24- factors are used to evaluate the actual rates of the reactions
27) simulates the molar flow rates, pressure drop, and in the overall reactor model (eqs 24-27).
energy balance of the reactor, while the solution of the The reactor model differential equations (24-27) were
catalyst pellet equations (15, 17, and 23) provides the integrated. Subroutine DGEAR (ISML Math/PC-Li-
effectiveness factors vi’s for eqs 24, 25, and 27. brary) (Hindmarsh, 1974) was used for solving these
differential equations. DGEAR uses a Runge-Kutta-
Solution of the Model Equations Verner fifth- and sixth-order method with automatic step
size to ensure accuracy.
The main objective of the catalyst pellet model is to
calculate the effectiveness factors for the six reactions Extracting Intrinsic Rate Constants
taking place inside the pellet. The effectiveness factors
are defined as the ratios of the actual rates taking place Pseudohomogeneous models do not account explicitly
in the catalyst pellet to the rates taking place under bulk for the catalyst packing the reactor; however, heteroge-
conditions. neous models account for the catalyst by considering the
The rate equations obtained from the literature for the diffusion of reactants and products through the pores of
calculation of the rates in the pellet and in the bulk are the catalyst pellet.
functions of the partial pressures of the components. The In this investigation industrial reactor data from the
partial pressures can be calculated from the mole fractions Polymer Corporation, Sarnia, Ontario, Canada were used
and the total pressure. The mole fractions of the com- (Sheel and Crowe, 1969). The specification of the reactor
ponents in the bulk and at the collocation points in the and the feed conditions to the reactor are given in Table
pellet are thus required, in order to compute the effec- I. These data were obtained from Sheel and Crowe (1969),
tiveness factors. as were the kinetic data in columns 2 and 4 of Table 11.
This requires the simultaneous solution of the set of Two corrections in the published values, noted on these
differential equations (15 and 17) together with the flux tables, were provided by Crowe (1989,1992). These kinetic
relations (19-22). The subroutine ZSPOW (IMSL Math/ constants were effective, not intrinsic, rate constants.
PC-Library) (More et al., 1980) for solving nonlinear These data were used as starting values in an iteration
algebraic equations was used to solve the algebraic scheme to obtain intrinsic kinetics suitable for heteroge-
equations resulting from applying the global orthogonal neous models.
collocation technique to the differential equations of the The heterogeneous model is used to extract the intrinsic
pellet. The mole fractions obtained were then used to kinetic constants from the industrial data. The rate
calculate the rates of the reactions. Rates inside the pellet constants of the pseudohomogeneous model (nonintrinsic
were then divided by rates at the surface to give the rates), given in Table 11, were first used as starting values
effectiveness factors of the reactions. The effectiveness in the heterogeneous model. The results obtained are
2540 Ind. Eng. Chem. Res., Vol. 32, No. 11, 1993
Table 111. Results of the Heterogeneous Model with 1 .o
Intrinsic Rate Constants (Last Iteration) As Compared
with Those of the Industrial Reactor (Sheel and Crowe,
e
_I
1969)
4 0.8
industrial reactor heterogeneous model N
m
item MFRa conversion, % MFR conversion, %
I--
EB 19.45 47.25 19.29 47.45 cn
H2O b b 447.32 1.26 $0.6
L
2! 0.2
0
a MFR, molar flow rate (kmouh); p E , outlet pressure (bar); TE,
6
*
outlet temperature (K).Not measured in plant operation. 0.0
J
j , , , , , , , , , ,, , , , , , , , , , , , , , , I , , I I , , , , , (
apparent intrinsic
~