(255c) Micro-Structured Catalysts for Process Intensification

Abstract

The first part of this talk deals with the investigation of open-celled foams (or sponges) as potentially enhanced catalyst supports. Ceramic and metallic open-celled foams are in fact micro-structured materials with interesting structural properties, which make them attractive as low pressure drop catalyst carriers: they come with high porosities (from 70 to 95%), and may have very high surface areas per unit volume. The adoption of foams as catalyst supports has been suggested for chemical processes (e.g. steam reforming of methane), as well as for the fast diffusion-limited processes of environmental catalysis (like e.g. in catalytic combustion, in DeNOx processes for aftertreatment of vehicle exhausts, or in catalytic partial oxidation of CH₄ for H₂ production). However, methods for making the foams catalytically active, and most of all engineering correlations of their transport properties, have been scarcely reported in the scientific literature so far.

After briefly touching the development of a procedure for coating open celled foams with thin layers of Pd-Al₂O₃, this talk will present measurements of gas-solid mass and heat transfer rates in metallic and ceramic foams, using CO oxidation and transient cooling experiments, respectively, and leading to generalized engineering correlations. In this respect, three types of catalyst support (foams, honeycomb monoliths and spherical pellets) will be compared, first on a generalized basis, then by simulating their application to the autothermal partial oxidation of methane to syngas: both cold start-up and steady-state behaviours are analyzed in the latter case. Finally, very recent experimental results concerning the overall (non-adiabatic) heat transfer performances of foams with different geometrical (porosity, ppi) and structural (FeCrAlloy, aluminium) properties will be illustrated.

The second part of the presentation will focus on a parallel investigation of conductive honeycomb monolith catalysts. We have shown in the past that there is potential for significant enhancement of radial heat transfer rates in industrial multitubular packed-bed reactors with external cooling if the random packings of catalyst pellets are replaced by structured catalysts with highly conductive honeycomb supports. This concept has been proposed on the basis of simulation studies, confirmed by heat transfer and reactive experiments at the lab scale, and claimed in patents. We present herein a proof-of-concept at an industrial scale based on a campaign (1600 hours) of o-xylene oxidation runs (c/o Polynt, Italy) in a single-tube pilot reactor (i.d. = 24.6 mm) loaded with washcoated (V₂O₅/TiO₂) monolithic catalysts with Al honeycomb supports (Corning Inc., USA) and operated at typical industrial conditions for PA (phthalic anhydride) production. The conductive monolithic supports afforded substantially reduced T-gradients in comparison with reference runs in the same pilot reactor loaded with conventional ring catalysts, the maximum T-difference with the salt bath being halved at the same hot spot temperature (440 °C) and the mean bed temperature being about 20 °C higher. Temperature gradients were still moderate at the highest investigated o-xylene feed load (400 g/h at Q_air = 4 Nm³/h), which on the contrary represents an upper limit for current industrial PA packed-bed reactors. The twofold enhancement of radial heat transfer rates can be exploited e.g. either to increase the o-xylene feed load above 100 g/Nm³ (and the PA productivity accordingly) within a retrofitting strategy, or to design new reactors with larger tube diameters. Ongoing related activities addressing the strongly exothermic Fischer-Tropsch Synthesis in fixed-bed reactors will be also reported.