(705e) An Industrial Reactor Design for Heterogeneously Catalyzed Ethene Oligomerization

An Industrial Reactor
Design for Heterogeneously Catalyzed Ethene Oligomerization

K. Toch, J.W.
Thybaut, G.B. Marin

Laboratory for Chemical
Technology, Ghent University, Krijgslaan 281 ? S5, 9000 Ghent, Belgium

Heterogeneously
catalyzed ethene oligomerization has, compared to the current homogeneous based
technologies, the advantages of being a solvent free process, a decreased
energy requirement and an increase in product flexibility. By tailoring the
reaction conditions simultaneously with the catalyst properties, the product
distribution can be altered from chemicals, e.g., α-olefins, to
liquid fuels, e.g., diesel and gasoline. This research fits within OCMOL,
a large-scale European FP7 project funded by the European Commission [1]. OCMOL
is the acronym for Oxidative Coupling of Methane followed by the
Oligomerization to Liquids. The main advantages of this integrated process are
an economic operation at relative small to medium scale, i.e., 100kTon/y, the
low to zero CO₂ emission, the use of either methane or biogas as a
process feedstock and the flexibility of the product stream containing either
chemicals or fuels.

An
industrial reactor is designed for ethene oligomerization employing bifunctional,
heterogeneous catalysts that comprise nickel as active metal on an acid support
such as amorphous silica-alumina or zeolites, i.e., MCM-41 and Beta. The
bifunctional nature of the catalyst, i.e., having both metal-ion as acid
functions, has as advantage that the reaction proceeds at much milder
conditions then when only an acid functionality would be present on the
catalyst. The Ni metal ion dimerizes ethene to butene, which is more easily
protonated and subsequently alkylated and/or isomerized on the acid sites, see Figure
1.

Figure 1: Reaction scheme for ethene oligomerization
on a bifunctional heterogeneous catalyst

For
industrial applications, three reactor types are possible: fixed bed, slurry or
monolith reactors. The industrial reactor modeled in this work is of the fixed
bed type. The model is capable to describe an isothermal, adiabatic or even
cooled/heated multi catalytic bed with/without cooling/heating between the
catalytic beds. Additionally, throughout the reactor more heavier
oligomerization products are formed which are in liquid phase. Therefore, a
gradual transition from gas-solid to gas-liquid-solid operation is included. It
is also possible to simulate the use of a solvent in the feed in which the
reactants are dissolved. At industrial scale, transport limitations at the
pellet scale are more common compared to laboratory scale where it is commonly
aimed at to obtain intrinsic kinetics. To account for these transport
limitations, intraparticle mass and heat transport phenomena are included in
the industrial reactor model. Axial and radial dispersion is not accounted for.

The
individual components' mass balances over the reactor is given by the plug flow
reactor equation:

in
which F_i is the molar flow rate of component i, W
the catalyst mass and R_i the net rate of formation of
component i. The net rate of formation corresponds to the intrinsic, chemical
kinetics as calculated by the Single-Event MicroKinetics (SEMK) [2]. Kinetic
and catalyst descriptors have been determined by regression to experimental
data acquired on amorphous silica-alumina and zeolites MCM-41 and Beta.

To
be able to describe the adiabatic behavior of the reactor, the following energy
balance over the reactor is implemented:

T is the temperature of the catalyst bed, ΔH_f,i
is the enthalpy of formation of every component, u_s is the
linear velocity through the reactor and ρ_g and c_p
are resp. the molar volume and the thermal capacity of the gas, all at the
reactor conditions. u_s is determined via the volumetric gas
flow rate assuming it is an ideal gas and the cross sectional area of the
reactor. c_p and its temperature dependency is determined
using thermodynamic data available from literature [3].

To
incorporate intraparticle diffusion, for every component i, a
one-dimensional unsteady-state mass balance over an infinitesimal volume of the
crystallite is considered. Even it is expected not to have a significant
influence, intraparticle conductivity has also been included in the model using
an analogous balance.

C_sat is the saturation concentration, L is the
crystallite dimension in the direction of the diffusion path, ξ is
a dimensionless diffusion length, i.e., , s
is the particle shape factor, i.e., 0, 1 or 2 for resp. a slab, cylinder or
sphere, D is the intracrystalline diffusion coefficient and θ
is the fractional occupancy of the catalyst surface by the component
considered. R, the net rate of formation, is affected by the shape of
the crystallite assumed and is determined as follows:

For
this set of partial differential equations, the following boundary and initial
conditions were considered:

-for all t,
except t=0

-for t=0

θ^s is the fractional occupancy of the catalyst
surface at the outer surface of the crystallite.

This
set of partial differential equations is solved using finite differences by
dividing the crystallite in a number of meshes (n_mesh). From
the initial conditions, this set of equations is solved until convergence,
i.e., steady state.

Figure
2 shows the comparison between an adiabatic multi bed reactor to an isothermal
and adiabatic single fixed bed reactor. All three reactor types have an equal
total amount of catalyst mass loaded. The criterion for determining the
catalyst bed size in the multi bed reactor was a maximum temperature increase by
30 K. They were not dimensioned based on the attainment of thermodynamic
equilibrium since this is not reached. As can be observed in Figure 2, a total
of 4 catalytic beds is required to obtain a conversion of 96%. The conversion
and, hence, temperature, increases gradually less steep with increasing bed
number, e.g., the last bed used for the final 30% conversion is 20 times larger
than the first one. The isothermal single fixed bed reactor has a slightly
lower conversion at the reactor exit, i.e., 94%. A conversion amounting to 96%
is already obtained halfway the catalyst bed in the single bed adiabatic
reactor, but is accompanied with a temperature increase exceeding 100 K, which may
be detrimental for the product selectivity, catalyst stability or even process equipment.

Figure
2:
Comparison of an adiabatic multi fixed bed reactor with an isothermal and
adiabatic single fixed bed reactor for ethene oligomerization. Temperature
(left) and ethene conversion (right) as function of the axial reactor
coordinate given by the catalyst mass. The inlet temperature at each catalytic
bed is 473 K. The inlet partial pressure and molar flow rate of ethene is equal
to resp. 0.35 MPa and 100 mol s^-1.

Figure
3 shows the effect of the pellet geometry and efficacy on the fractional
coverage of ethene on the catalyst surface as a function of the dimensional
length of the pellet. It is clear that a spherical geometry has a negative
effect on the catalyst efficacy (45.2%) compared to the slab (91.2%) and
cylindrical geometry (79.8%).

Figure 3: Influence of the pellet geometry (left) and
efficacy (right) on the fractional coverage of ethene on the catalyst surface
as function of the dimensional length of the pellet, i.e., from the outside to
the center of the pellet at 473 K, 0.35 MPa ethene

The
transition of fixed bed to trickle bed behavior has been implemented by
accounting for the vapor pressure of every component throughout the axial
coordinate of the reactor. If only gasses are present, the components
concentration at the catalyst outer surface are directly given by a Langmuir
isotherm including competitive adsorption. If liquids are formed, they will (partially)
wet the catalyst surface, and hence, the concentration on the catalyst outer
surface will be determined by the composition of the liquid phase. The
solubility of the gasses, mainly ethene, in the liquid phase is accounted for.

?This abstract reports work undertaken in the context
of the project ?OCMOL, Oxidative Coupling of Methane followed by
Oligomerization to Liquids?. OCMOL is a Large Scale Collaborative Project
supported by the European Commission in the 7^th Framework Programme
(GA n°228953). For further information about OCMOL see: http://www.ocmol.eu ?

[1] http://www.ocmol.eu

[2] G.F. Froment, Catalysis
Today. 52 (1999) 153-163

[3]
Reid, Prausnitz and Poling, The Properties of Gases and Liquids