(600bj) Hydrogenation of α-Methylstyrene in a Mini Fixed-Bed Reactor Under External Mass Transfer Limitation

Introduction

For small and medium scale processes, the group of micro- and mini-reactors
has proven to be a viable solution for multiphase processes due to high mass
and heat transfer rates. For heterogeneous catalyzed gas-liquid reactions,
there are two options to fix the catalyst inside the reactor: a) coating of the
channel wall (e.g. monolith catalysts) and b) a catalyst packing (mini
fixed-bed). Thereby, a reproducible and long-term stable coating of long
channels is quite challenging. On the other side, the packing of micro- and
mini-reactors with commercial and industrial tested catalyst spheres and
extrudates is relatively easy [1].

The mini fixed-bed reactors can be found in particular in the field of
catalyst-screening of solid catalysts with a size of a few millimeters for
conventional large scale fixed-bed reactors. They show a similar hydrodynamic
behavior like the large scale unit and can therefore be used efficiently to
study the kinetic of catalyst particles [2]. By using this technology,
operation costs can be saved and the setup complexity as well as the
experimental time is reduced [3]. However, for catalyst screening it is
important to guarantee the absence of external mass transfer limitations, which
would affect the results drastically.

Numerous investigations concerning hydrodynamics in these reactors show,
that in particular the channel-to-particle-ratio (d_ch/d_p)
has a significant influence on flow pattern, residence time distribution and
pressure drop [2]. Until now, there is only less information about mass
transfer in these mini fixed-bed reactors.

Aim of this work is the investigation of mass transfer characteristic of
a mini fixed-bed reactor using the palladium catalyzed hydrogenation of
α-methlystyrene (AMS) to cumene. In particular the influence of flow
pattern, superficial gas and liquid velocities (0.002 ms^-1< u_G,S < 1,2 ms^-1, 0.005 ms^-1< u_L,S
< 0,2 ms^-1) and gas ratio at the reactor inlet (ε_G=
0,25; 0,50; 0,75) on overall mass transfer coefficient is examined.

Experimental work

The mini fixed-bed reactor consists of an inert,
pressure-resistant round capillary (d_ch=
2.0 mm) being filled with a bed of 400 spheres (5 wt.% Pd/Al₂O₃,
d_p= 0.8 mm). The
distribution of spheres is limited by two regular arrangements of spheres which
are shown below in figure 1 for d_ch/d_p= 2.5
under the consideration of a package of mono dispersed catalyst spheres.

Figure  SEQ Figure \*
ARABIC 1:              Limiting packing gaskets for a d_ch/d_p
ratio of 2.5. Left: Closest regular packing gasket with 16 spheres building
four layers of four spheres. Right: Least close regular gasket with 12 spheres
building six alternating layers of three and a single sphere.

The reactor is embedded in a pilot plant that is equipped with mass flow
controllers to provide the hydrogen, a pump for the reaction mixture (20 wt.%
AMS in cumene) and a temperature and pressure control. All experiments are
carried out in down-flow mode.

Mass transfer calculation

The mass transfer of hydrogen from the gas phase through the liquid to
the catalyst particles is calculated by measuring the consumption of AMS
between inlet and outlet of the reactor.

The surface concentration of hydrogen c_H2,SAT can be calculated by knowing the intrinsic
kinetic constant k_int of
the used catalyst and accordingly the following equation is obtained.

With this equation the overall mass transfer coefficient of hydrogen can
be calculated by the conversion of AMS, if it is ensured that the reaction is
external mass transfer limited.

Results

For identifying an external mass transfer limitation, the variation of
reaction temperature is an appropriate method. Figure 2 shows the conversion of
the liquid AMS (left side) and the Arrhenius plot (right side) in dependence on
reaction temperature.

Figure  SEQ Figure \* ARABIC 2                Left: Conversion
of AMS and right: apparent kinetic constants at different temperatures (p= 1.1
MPa).

It can be seen that the conversion of AMS rises linearly until a
temperature of T= 353 K due to the effect of pore diffusion in the porous
catalyst, which can be explained by the reduced apparent action energy of E_app2=
24 kJ mol^-1 instead of E_intr= 37 kJ mol^-1 for
a kinetic limited reaction. At higher temperature the
conversion of AMS almost stagnates indicating a further external limiting step.
According to this, the setup is applicable to measure mass transfer
coefficients.

After the temperature variation the effect of superficial gas and liquid
velocities on conversion of AMS and overall mass transfer coefficient of
hydrogen is examined. With raised superficial gas velocity the conversion of
AMS as well as the mass transfer coefficient increases. On the other side, an
elevated superficial liquid velocity has only a marginal influence on mass
transfer and leads to lower AMS conversion due to lower residence time of the
liquid within the reactor.

An investigation of the influence of the gas-to-liquid ratio at the
reactor entrance reveals that the conversion of AMS decreases with rising two-phase
velocity for all gas ratios due to the shorter residence time. The conversion
increases with elevated gas ratio in the whole range of two-phase velocities.
This effect is quite reasonable due to the smaller amount of liquid molecules
entering the reactor. The mass transfer coefficient increases with elevated
two-phase velocity showing that the effect of raised mass transfer is more
pronounced than the shortened residence time with increasing gas velocity. Interestingly,
the curve of a gas ratio of 0.75 crosses the two other curves in two points, as
illustrated in figure 3.

Figure  SEQ Figure \* ARABIC 3                Mass
transfer coefficients of hydrogen in dependence on superficial two-phase
velocity for different gas ratios (p=1.1 MPa, T= 353 K).

This effect must have its origin in hydrodynamics of the gas-liquid
two-phase flow. In figure 4 the flow map of the applied system is shown.

Figure  SEQ Figure \* ARABIC 4                Flow map with mass transfer
coefficients of hydrogen in dependence on superficial gas and liquid velocities
(p=1.1 MPa, T= 353 K).

It is obvious that the three experiments with a ε_G= 0.75 are possibly located
in different flow regimes. If additionally anticipating ka_GLS,OV values
from the investigation of the effect of superficial gas and liquid velocity,
the suspicion is confirmed easily. Only the mass transfer coefficients for u_TP,S=
0.04 ms^-1 can be compared with regard to the gas ratio, because
these three points are in the same flow regime. It is obvious that with
elevated gas ratio the mass transfer increases, which can be explained by a
higher gas-liquid surface area. Since the mass transfer of dispersed periodic
film is generally higher than in dispersed stagnant film, the values ε_G=
0.75 at u_TP,S=
0.02 ms^-1 and u_TP,S=
0.06 ms^-1 are lower than for the lower gas ratio.

Summary

In this work the influence of operations conditions on overall mass
transfer coefficient of hydrogen in a mini fixed-bed reactor has been
investigated. It is shown, that with higher superficial gas velocity the mass
transfer (ka_GLS,OV)
increases due to larger gas-liquid interfacial area and higher interaction
between the two phases. In contrast, a raised superficial liquid velocity would
increase k_GLS,OV
too, but would decrease the interfacial area a_GLS,OV.
Therefore, superficial liquid velocity has only a marginal effect on mass
transfer. Furthermore, it can be stated that mass transfer is dependent on flow
regime, which characterizes the interaction between the two phases.

References

[1]        Bauer,
T., Haase, S.: Comparison of structured trickle-bed and monolithic reactors
in Pd-catalyzed hydrogenation of alpha-methylstyrene. Chemical Engineering
Journal 169 (2011). pp. 263?269.

[2]        Hipolito
A., Rolland, M., Boyer, C., de Belefon, C.: Single Pellet String Reactor for
Intensification of Catalyst Testing in Gas/Liquid/Solid Configuration. Oil
& Gas Science and Technology 65 (2010). pp. 689-701.

[3]
       Kallinikos, L.E., Papayannakos,
N.G.: Operation of a Miniscale String Bed Reactor in Spiral Form at
Hydrotreatment Conditions. Industrial Engineering and Chemical Research 46
(2007). pp. 5531-5535.