MPI für Dynamik komplexer technischer Systeme / Physical and Chemical Foundations of Process Engineering |
|Pilot-plant study on membrane reactors for catalytic oxidation of hydrocarbons|
|Authors:||Klose, F.; Wolff, T.; Hamel, C.; Tota, A.; Ahchieva, D.; Heinrich, S.; Seidel-Morgenstern, A.|
|Name of Conference/Meeting:||XXXX. Jahrestreffen Deutscher Katalytiker|
|Place of Conference/Meeting:||Weimar, Germany|
|(Start) Date of Event |
|End Date of Conference/Meeting |
|Abstract / Description:||Membrane reactors are often reported to be promising for enhancement of productivity in selective oxidation of hydrocarbons. Herein, the membrane is used as oxidant distributor (e.g. [1-3]). This is meaningful because the order of deep oxidation with respect to oxygen is often higher than that of the desired formation of olefins/oxygenates [4,5]. This means that oxygen availability influences both, alkane conversion and selectivity to olefins/oxygenates and this in opposite manner. Beside changes in axial oxygen concentration profile, membrane assisted oxidant dosing causes also changes in residence time of the reactants . The possible variation of the feed dosed via the membrane is an additional degree of freedom.
The focus of this study is to evaluate the membrane reactor concept in pilot-plant scale in combination with packed and fluidized catalyst beds (Fig. 1). As model reaction the oxidation of ethane to ethylene (1.4 % V/γ-Al2O3 catalyst) was selected. The obtained results reveal that both membrane reactors outperform the reference reactors with co-feed reactant supply (fixed-bed reactor - FBR and fluidized bed reactor - FLBR) by improved ethylene selectivity or ethane conversion; although at higher conversion this advantage is only within a few percent. For compensation of the effect of oxygen on alkane conversion it was found to be beneficial to operate the membrane reactors at overall O2/C2H6 ratios near 2, i.e. at higher levels than conventional FBRs.
The packed-bed membrane reactor (PBMR, with microporous ceramic or macroporous metal membranes) shows its best performance at moderate contact times (W/F =
200 gs/l) and high fractions of membrane dosed feed. Beside better performance it is characterized by smoother axial temperature profiles ("Hot spots" e.g. 51 K (PBMR) vs. 121 K (FBR) at 550 °C, 200 gs/l, O2/C2H6 = 1). Radial temperature gradients were typically below 20 K.
A crucial point for PBMR operation is diffusion of reactants from the catalyst bed onto the side of oxidant insertion resulting in a loss of reactor performance . For its suppression a minimum transmembrane pressure drop of 0.5 bar was found to be mandatory. Another drawback of the PBMR is an enhanced probability of soot formation, especially at the reactor inlet, where local oxygen concentration and gas velocity are low.
For smoothing temperature gradients especially fluidization of the catalyst bed is very effectively. However, then the catalyst has not only to show a good performance; it has additionally to meet the demands due to fluidization regarding particle size and mechanical stability. Also the bubble size in fluidization has to be kept low to avoid losses of conversion. That is why for the studies on the fluidized-bed membrane reactor (FLBMR with macroporous metal membrane, [8,9]) two batches of catalyst particles were used with diameters of 0.4 and 1.8 mm.
The FLBMR was found to reach yields exceeding 30 %. In principal, the FLBMR shows the same tendencies regarding the impact of process parameters. However, due to the thermal homogenization the FLBMR demonstrates a lower sensitivity against parameter variation than FBR, FLBR and PBMR. Catalyst abrasion was < 0.2 % for operation over several weeks.
Comparing the membrane reactors, the FLBMR seems to be the more promising concept to perform exothermic oxidation reactions (see Fig. 2). However, this has to be balanced with lower space time yields and higher efforts on the investment site. Further, the difference to the PBMR is not really high, so that also the PBMR concept can be considered as an attractive alternative for process intensification.
Financial support by the Deutsche Forschungsgemeinschaft is gratefully acknowledged (DFG research unit 447, "Membrane supported reaction engineering").
 J. G. S. Macano, T. T. Tsotsis, Catalytic Membranes and Membrane Reactors, Wiley-VCH, Weinheim, 2002
 A. Julbe, D. Farruseng, C. Guizard, J. Membr. Sci. 181 (2001), 3-20
 A. G. Dixon, Int. J. Chem. Reactor Eng. 1 (2003), 1-35 (Review R6)
 O. Levenspiel , Chemical reaction engineering, Wiley, New York, 1972
 Y. Lu, A. G. Dixon, W. R. Moser, Y. H. Ma, Catal. Today, 35 (1997), 443-450
 F. Klose, T. Wolff, S. Thomas, A. Seidel-Morgenstern, Catal. Today., 82 (2003), 25-40
 R. Ramos, M. Menéndez, J. Santamaría, Catal. Today, 56 (2000), 239-245
 D. Ahchieva, M. Peglow, S. Heinrich, L. Mörl, T. Wolff, F. Klose, Appl. Catal. A – Gen., 296 (2005), 176-185
 D. Ahchieva, Ph.D thesis, Otto von Guericke University Magdeburg, Germany, 2007
|Communicated by:||Andreas Seidel-Morgenstern|
|Affiliations:||MPI für Dynamik komplexer technischer Systeme/Physical and Chemical Foundations of Process Engineering|
|External Affiliations:||Institut für Verfahrenstechnik, Otto-von-Guericke-Universität Magdeburg, Universitätsplatz 2, 39106 Magdeburg, Germany
Institut für Apparate- und Umwelttechnik, Otto-von-Guericke-Universität Magdeburg, Universitätsplatz 2, 39106 Magdeburg, Germany
The scope and number of records on eDoc is subject
to the collection policies defined by each institute
- see "info" button in the collection browse view.