(38a) Low and High Molecular Weight Hydrogenations in Structured, Pulsed Flow Reactors

We have studied two model hydrogenations
- alpha-methyl styrene (AMS) and polystyrene (PS) ? in novel reactors and under
conditions where pulsed flows might be expected to greatly enhance observed
rates of reaction. The PS was hydrogenated in a twin screw extruder modified
with a custom-made die that can hold either a catalyst monoliths (100 cpsi) or
a conventional packed bed (20-35 mesh). The polymer solution (2 or 10 wt% PS
dissolved in 10 vol% THF/cyclohexane) is pre-mixed with hydrogen in an
autoclave and pumped to the extruder, where it can be mixed with a pulsing gas
flow (pulsed solenoid valves) prior to the die entrance. A photocell records
the exit flow behavior. Because extruders also exhibit inherently unsteady-state
operation in liquid-starved operation,¹ natural flow oscillations (similar
to slug flow) are also observed, albeit at different frequency. The effects of
these natural ?unforced? pulses were compared to the performance of the superimposed
?forced? pulses.

The catalyst for all these studies
was a typical hydrogenation catalyst composed of 0.5 wt% Pd/γ-Al₂O₃
prepared by an ion exchange technique from Pd(NH₃)₄(NO₃)₂.
The dispersion was ~73% measured by H₂ chemisorption with a BET
surface area of 290 m²/g and an average pore size of 10 nm. The
observed reaction rate constants were determined by modeling the system in plug
flow with first order dependences in aromatic group and hydrogen
concentrations.²

As has been previously reported³, for
2 wt% PS solutions at low to average liquid space velocities (0.14-0.48 mL/s/g
Pd), the observed pseudo first-order rate constants (k_obs) (1-3 x 10^-5
L/s/g Pd) are below those obtained from batch autoclave studies where
intraparticle and liquid film diffusional resistances were minimized. However,
increasing liquid and gas flow rates simultaneously increases k_obs
in qualitative agreement with correlations for gas-limited mass transfer in
slug flow monolith systems.^4,5 At high flow rates (0.8-1.8 mL/s/g
Pd), the rate constants are within 30% of those observed for an agitated vessel
at comparable conditions (~1 x 10^-4 L/s/g Pd), suggesting significant
increases in the mass transfer rates for H₂, both gas to liquid and
liquid film to catalyst surface. Forced pulsing at 0.1 Hz and higher had no effect
on the k_obs, indicating the unforced oscillations are already
optimal at this low PS concentration.

Increasing the PS concentration to 10 wt% results
in much greater external mass transfer resistance - the viscosity is ~15 times
that of the 2 wt% solution.  The k_obs values for the pulsed extruder
are approximately an order of magnitude less than for the 2 wt% PS solution.
Similar to the 2 wt% PS solution, k_obs increases as the flow rates
increase. However, forced pulsing at 0.1 Hz did increase k_obs by
~35% compared to the unforced oscillations, indicating a greater effect of
forced pulsing with increasing resistances to mass transfer. Pulsing at 0.5 Hz
decreased the observed rates, as the catalyst then operated under liquid-starved
conditions.   There is a clearly observable optimal frequency for this process.

The AMS hydrogenation reaction
was carried out in vertically stacked monoliths mounted above a piston/cam
arrangement, thereby allowing oscillatory frequencies much higher (0-20 Hz)
than the solenoid configuration on the reactive extruder. It has already been
shown that this configuration can significantly enhance rates of mass transfer
? by as much as 700% - for the air-water system. The behavior with respect to
frequency can be attributed to resonance effects that can be predicted
theoretically.⁶ Reactor studies on AMS hydrogenation in this
oscillating monolith reactor are currently underway to determine optimum
pulsing frequency and to compare results to similar work on alternating gas-
and liquid-rich conditions in packed  and trickle beds.⁷

References

1. Mudalamane, R.; Bigio, D.I., ?Process variations and the transient
behavior of extruders? AIChE J, 2003, 49, 3150-3160.

2. Xu, D.; Carbonell, R.G.; Kiserow D.J.; Roberts, G.W.
?Kinetic and Transport Processes in the Heterogeneous Catalystic Hydrogenation
of Polystyrene.? Ind. Eng. Chem. Res., 2003, 42, 3509-3515.

3. Bussard, A.; Dooley, K. ?Polymer Hydrogenation by
Reactive Extrusion ? Pulsed and Continuous Flow Systems? AIChE Annual
Meeting, San Francisco, 2006, 508d.

4. Bercic, G.; Pintar, A. ?The role
of gas bubbles and liquid slug lengths on mass transport in the Taylor flow through capillaries? Chem. Eng. Sci., 1997, 52, 3709-3719.

5. Kreutzer, M.T.; Du, P.;
Heiszwolf, J.J.; Kapteijn, F.; Moulijn, J.A. ?Mass transfer characteristics of
three phase monolith reactors? Chem. Eng. Sci., 2001, 56,  6015-6023.

6. Knopf, F.C., Waghmare, Y.G.,
Ma, J., Rice, R.G. ?Pulsing to improve bubble column performance: II. Jetting
gas rates?  AIChE Journal. 52, 1116 (2006).

7. Urseanu, M.I., Boelhouwer,
J.G., Bosman, H.J.M., Schroijen, J.C. ?Induced pulse operation of high-pressure
trickle bed reactors with organic liquids: hydrodynamics and reaction study? Chem.
Eng. and Proc. 43, 1411 (2004).