486a Toward the Reproducible Intrinsic Kinetic Measurement in Multi-Phase Microreactors

Pedro Castaņo1, Daniel Van Herk1, Michiel T. Kreutzer2, Michiel Makkee2, and Jacob A. Moulijn2. (1) Delft University of Technology, Julianalaan 136, Delft, 2628 BL, Netherlands, (2) Catalysis Engineering, Delft University of Technology, Julianalaan 136, Delft, 2628 BL, Netherlands

Reducing the scale of multi-phase reactors has a number of benefits in terms of safety, catalyst preparation, waste-production, chemicals consumption, and in the potential for drastic increase of throughput. It was found that scaling down such systems implies a major change in the hydrodynamic behavior. The most important reason is that the particle size has to be reduced to a low value and this has as a consequence that gravity forces are less important than capillary forces. Compared to beds packed with mm-range particles, hold up and hydrodynamic behavior are dramatically different. In catalyst testing, microreactors are very attractive and developing a good methodology (operating window, packing- and dilution-procedure) is crucial.

In this work we have delimited the conditions where intrinsic kinetic data can be obtained from multiphase micro-packed beds. Details of the parallel microreactor system are: six parallel reactors; continuous gas and liquid flow over a powder bed; 2 mm internal diameter; 20-50 mg of catalyst. The superficial liquid velocity is around 1 mm/min and the superficial gas velocity can be an order of magnitude higher. Our approach implies the comparison of the kinetic performance of a slurry-type reactor set of experiments (free of both internal and external mass-transfer limitations) with those obtained in the microreactor. The selected reaction was the liquid-phase hydrogenation of aromatics (biphenyl) over noble bimetallic catalyst under hydrotreating conditions; 120-160°C and 5 MPa.

The catalyst had been diluted with inert material (SiC). Our results pointed that the size (average and distribution), morphology and density of catalyst and diluent particles, as well as the packing procedure, in combination with both the superficial gas and liquid velocity determine whether reproducible and intrinsic kinetic data in micro packed-beds can be obtained or not. In fact it was found that non-optimal protocols lead to a persistent and frustrating lack of reproducibility.

We report in this contribution the key aspects to bear in mind when selecting particle size and loading reactors with powder catalyst in combination with the required (hydrodynamic) reaction conditions.