OPTIMAL DESIGN OF CH4 CATALYTIC PARTIAL OXIDATION REFORMERS ON Rh BASED CATALYSTS FOR SHORT-CONTACT-TIME SYNGAS PRODUCTION Matteo Maestri, Alessandra Beretta*, Pio Forzatti, Gianpiero Groppi, Enrico Tronconi, Dion Vlachos§ Laboratory of Catalysis and Catalytic Processes – Dipartimento di Energia Politecnico di Milano – 20133 Milano, Italy § Center for Catalytic Science and Technology – Department of Chemical Engineering University of Delaware, 19716, Newark, DE (USA) *corresponding author. Email: alessandra.beretta@polimi.it TOPIC 4: Modeling and scale-up of structured reactors and microreactors Temperature [°C] mmol/s The possibility to produce H2/CO mixtures without external energy input in small reactors with reduced heat capacity makes the catalytic partial oxidation (CPO) of methane and higher hydrocarbons a promising solution for on-board and in situ generation of syngas for energy applications. Though very flexible, the process is very complex: CH4 CPO short-contact-time reactors work under severe conditions, with very high gas hourly space velocity and temperatures, complex fluid patterns and a strong coupling of heat and mass transfer with surface kinetics. These severe operating conditions pose a challenge [1] for process stability requiring a suitable trade-off between low hot spot temperature and high syngas productivity. In this work, we have coupled the previously derived 1D dynamic model of the adiabatic CH4 CPO reformer [2] with our recently developed microkinetic model for CH4 activation on Rh based catalysts [3] for the model-based analysis of spatially resolved CH4 CPO experimental data on foams in adiabatic conditions [4]. This improved model allows us to analyse in detail the behaviour of the reactor (including transport phenomena at the macroscale and surface chemistry at the microscale) and propose strategies for process optimization. Figure 1 compares spatially 1100 2 100 ppi resolved experimental data [4] and 1050 H 2 model predictions for a typical 1.5 CH4 70 ppi 1000 CPO experiment on foam. The 950 model reveals that along the entire 1 O 900 40 ppi reactor the most abundant reactive CO 2 50 ppi HO intermediates (MARI) are CO* 850 2 0.5 and H* and the surface turns out to 800 CO 2 be reasonably clean. Under those 750 0 conditions of surface coverages, 0 2 4 6 8 10 -2 0 2 4 6 8 10 Axial length [mm] Axial length [mm] methane consumption occurs via Figure 1. Species flows along the axial Figure 2. Effect on catalyst temperature pyrolysis followed by oxidation of length. Solid lines: model predictions – of decreasing the transport properties of C* species by OH* and O2 Square symbols: experimental data [4] the bed, by changing the cubic cell consumption is always mass length of the foam. transfer controlled. In line with previous results [2], this is confirmed by the analysis of the species concentration at the catalyst interface, that points out that O2 concentration at the surface interface is nearly zero. Consequently, the heat release on the catalyst surface is strictly governed by mass transfer. On the other hand, methane consumption has a mixed chemical-mass transfer character and consequently the catalyst loading and distribution along the reactor affect the rates of the endothermic reactions. Overall, the final temperature profile is determined by the combination of several physical phenomena in addition to the chemical reactions (including quenching effect of the feed cold gas, the rate of the O2 mass transfer, and the axial thermal dispersion). As an example, we investigated (Figure 2) the effect of varying the pore foam cell density, while keeping constant the Rh load: the increase in interphase mass transfer resistances results in a more gradual dosing of O 2 along the axial coordinate and consequently in a reduction of the hot-spot temperature (from 1080°C for a 100 ppi foam to 950°C for a 40 ppi foam). The great sensitivity of the reactor behaviour upon those characteristics offers a large number of degrees of freedom for reactor design and optimization, by proper tailoring of the type, the geometry and the material of the catalyst support with a twofold purpose: reduction in hot spot temperature and enhanced production of syngas. Variation of other parameters leading to reactor optimization will also be discussed. Experimental validation is currently on-going. References [1] I. Tavazzi, A. Beretta, G. Groppi, M. Maestri, E. Tronconi and P. Forzatti, Catalysis Today, 129 (2007) 372-379 [2] M. Maestri, A. Beretta, G. Groppi, E. Tronconi and P. Forzatti, Catalysis Today, 105 (2005) 709-717 [3] M. Maestri, D.G. Vlachos, A. Beretta, G. Groppi and E. Tronconi, AIChE J., doi:AIC11767 (2009) [4] A. Bitsch-Larsen, R. Horn and L.D. Schmidt, Applied Catalysis A: General, 348 (2008) 165-172