552133 Packed foams: a novel reactor configuration for intensification of the methane steam reforming process

Thursday, June 6, 2019: 12:06 PM
Texas Ballroom D (Grand Hyatt San Antonio)
Riccardo Balzarotti1, Matteo Ambrosetti1, Tommaso Selleri1, Alessanda Beretta1, Gianpiero Groppi1 and Enrico Tronconi2, (1)Politecnico di Milano, Dipartimento di Energia, Via La Masa, Milan, Italy, (2)energy, politecnico di milano, milano, Italy

Packed foams: a novel reactor configuration for intensification of the methane steam reforming process


Riccardo Balzarotti1, Matteo Ambrosetti1, Tommaso Selleri1, Alessanda Beretta1, Gianpiero Groppi1 and Enrico Tronconi1,*

1 Politecnico di Milano, Dipartimento di Energia, Via La Masa, 34 - 20156 Milano (Italy)



Structured catalysts are widely considered a valuable solution for the intensification of non-adiabatic chemical processes, where reactor performance and process scalability are limited by radial heat transfer [1]. The adoption of highly conductive metallic internals has been proposed to enhance heat transfer in fixed bed reactors with respect to conventional packed beds [2]. These metallic structures are available in various shapes (i.e. honeycomb monoliths, open cell foams) and they are usually made catalytically active by washcoating a thin layer of catalytic material onto their surface [3]. Despite the high potential of coated structured supports, some drawbacks such as smaller catalyst inventory with respect to packed beds and issues associated with washcoat adhesion and catalyst loading/unloading have discouraged the application of this technology at the industrial scale [4].

In this work, a novel fixed bed reactor configuration is proposed and tested for the steam reforming of CH4. The concept consists in filling the void volume of conductive open-cell foams with small catalytic pellets. As demonstrated for exothermic processes [5], the aim of using this new reactor layout is to enhance the radial heat transfer, thanks to the high thermal conductivity of the solid interconnected matrix. In principle, this could be obtained with negligible losses in terms of catalyst inventory with respect to the traditional packed bed configuration.

Experimental tests were performed using a catalyst in the form of alumina egg-shell particles. The catalyst was prepared by incipient wetness impregnation using alumina particles with a diameter equal to 0.6 mm. Rhodium was used as active metal, aiming to a metal content of 0.3 % wt. with respect to the carrier. In addition, a dedicated 2D two-phase model of the packed foam reactor has been developed, assuming: i) a pseudo homogeneous phase comprising both solid catalytic pellets and flowing gas, as traditionally done for packed bed reactor models; ii) a solid phase to account for the additional contribution of the conductive open-cell foam to heat and mass transfer. For the gas-pellet phase we solve material and energy balances including the effects of axial and radial diffusion as well as a reaction term while, on the solid phase, the heat exchange between gas and solid and the conduction through the foam matrix are accounted for. The model incorporates an already validated kinetic scheme for SR and WGS reactions [6].

In the case of the packed foam layout, three different cellular matrices were tested, namely a 12 PPI FeCrAlY foam (Porvair) and two copper foams, with 10 and 40 PPI, respectively (ERG Aerospace Corporation). In the case of FeCrAlY material (labelled as FeCrAl10 in the following), cell size and void fraction were equal to 5.2 mm and 0.92, respectively. In the case of copper foams, the 10 PPI foam (labelled as Cu10 in the following) had similar properties to the FeCrAlY foam (i.e. porosity of 0.91 and cell size of 4.6), while the 40 PPI copper foam (Cu40 in the following) had values of 0.88 and 2 mm for void fraction and cell size, respectively. For all samples, diameter and height of the foam were set at 29 mm and 25 mm. In order to collect temperature profiles in the axial direction, holes were drilled for the thermocouple-wells. Temperature profiles were in fact recorded longitudinally across the catalytic bed in three different radial positions, namely at the centerline, at 8 mm from the center and at the external wall. In the packed foam configuration, the foam void volumes were filled with catalytic particles, as shown in Fig. 1.

Fig. 1. Close-up view of a packed copper foam into the reactor

For comparison, tests in a conventional packed bed system were also performed. In this case, the same mass of catalytic particles (5.75 g) was loaded in the reactor, mixed with SiC particles (SiC to catalyst weight ratio of 1.54) in order to have the same reactor volume as for the packed foam layout.

Catalytic tests were carried out in a tubular reactor (I.D. = 29.5 mm) externally heated by a furnace; tests were performed in the 600-800 °C range. The feed consisted of a steam/CH4 mixture with S/C ratio of 3.5. Water was condensed and separated downstream from the reactor. A small N2 flow was mixed to the dry product stream as an internal standard. The quantification of reaction products was performed using an on line micro-GC (model GCX by Pollution) equipped with MolSieve and Porapack columns connected to TCD detectors.


Fig. 2 - Maximum radial difference in temperature between wall temperature and central temperature in catalytic bed (a) and conversion as a function of furnace temperature (b) for all the tested configurations

At the GHSV of 10,000 h-1, the introduction of a conductive matrix resulted in a drop of the maximum temperature difference between the wall and the center of the reactor (Fig. 2-a). Such a decrease was more evident when copper foams were used, thanks to the higher thermal conductivity of copper.

Additionally, an improvement of CH4 conversion was observed in the packed foam configurations when compared with the packed bed at any furnace temperature (Fig. 2-b). Although small, this improvement is significant, given the irreducible role of thermodynamics at these high temperatures: it can be ascribed to the reduced heat transfer resistances from the reactor wall to the catalyst, which enabled smaller temperature gradients along the radial direction thanks to the conductive heat transfer mechanism favored by the presence of the metallic open cellular structure (Fig. 1).



[1] J. Gascon, J. R. van Ommen, J. A. Moulijn, F. Kapteijn, Catal. Sci. Technol. 5 (2015) 807-817 

[2] E. Tronconi, G. Groppi, C.G. Visconti, Curr. Opin. Chem. Eng. 5 (2014) 55-67

[3] A. Montebelli, C.G. Visconti, G. Groppi, E. Tronconi, C. Cristiani, C. Ferreira, S. Kohler, Catal. Sci. Technol., 4 (2014) 2846-2870 

[4] C.G. Visconti, G. Groppi, E. Tronconi, Cat. Today 273 (2016) 178-186

[5] L. Fratalocchi, C.G. Visconti, G. Groppi, L. Lietti, E. Tronconi, Chem. Eng. J. 349 (2018) 829-837

[6] A. Donazzi, A. Beretta, G. Groppi, P. Forzatti, J. Catal. 255 (2008) 241-258

Acknowledgment: This project has received funding from the European Research Council (694910/INTENT). Co-funding from Microgen30 project (Italian Ministry of Industrial Development) is also acknowledged.


Extended Abstract: File Not Uploaded
See more of this Session: OCM Reactors
See more of this Group/Topical: General Submissions