The interaction between heterogeneous and homogeneous chemistries is a crucial issue for high-temperature catalytic processes. In particular, the assessment of the main routes that control the selectivity to the desired products is essential for the design and safe operation of reaction units. This assessment is particularly true for the ultrafast conversion of hydrocarbons in short-contact-time (SCT) reactors that play a pivotal role in the effort to cope with the worldwide growing demand for more efficient exploitation of energy andmaterial resources. Examples are the catalytically assisted combustion for gas turbines with ultralow emissions, the catalytic partial oxidation (CPO) of hydrocarbons to H2 or CO/H2 mixtures (i.e., syngas), and the oxidative dehydrogenation (ODH) of light alkanes to olefins. As a common feature, these processes operate in autothermal and compact reactors, with noble-metal catalysts (palladium, rhodium, and platinum). An enormous energy intensity is peculiar to the SCT autothermal conversion of hydrocarbons over noble metals. Strongly exothermic and endothermic reactions proceed on the catalyst surface at extremely high rates. As a consequence, sharp gradients of temperature (up to 200 8Cmm ) and concentration are established within the small reactor volumes. Temperatures ranging from 250 to 1100 8C are generally experienced within a few millimeters. Such a level of severity—in terms of power density and extent of temperature and concentration gradients—is typical of flames and gas-phase oxidation processes in general. For a catalytic process, however, these conditions represent a thoroughly unconventional kinetic regime. To best grasp the intensity and the speed of the involved phenomena, we need to think of SCT conversion of light alkanes as the catalytic equivalent of a flame. This analogy can clearly depict the complexity of the process and emphasize the related scientific issues: To what extent does a catalytic process “stick” to the catalyst surface at these very high temperatures, where the adsorption of species is thermodynamically unfavored? Can the gas-phase activation of C H bonds (e.g., the formation and propagation of radicals) cooperate or compete with the catalytic process? In this respect, it is largely accepted that in the case of CH4 CPO over rhodium the gas-phase paths are negligible at atmospheric pressure and the catalytic route dominates. Conversely, the SCT-ODH of short alkanes over platinum proceeds mainly in the gas phase, thus giving rise to the production of olefins and other hydrocarbon species. On the basis of these examples, we could conclude that either catalytic or gas-phase chemistry governs the SCT conversion of hydrocarbons, depending only on the stability of the C H bond in the gas phase. Herein, by using novel techniques for collecting spatially resolved temperature and concentration profiles, we show that the partial oxidation of short-chain alkanes over rhodium breaks the paradigmatic compartments of heterogeneous processes and gas-phase processes, revealing the real complexity of these “flame-like” processes. Specifically, we examine the reaction of C3H8 CPO (C3H8+ 3/2O2!3CO+ 4H2) as a case study, and we apply novel techniques for collecting spatially resolved gas-phase and solid temperature and concentration profiles within a rhodium-coated honeycomb monolith to monitor the evolution of a propane/air mixture fed at high flow rate. The temperature and the composition of the reacting system were monitored from the inlet reactor section where the mixture was fed to the outlet section of the catalytic unit where the syngas stream was delivered. The results are reported in Figure 1 and Figure 2 as spatially resolved profiles of temperature and molar fraction of reactants and products. In the first 5 mm of the honeycomb, a sharp drop of O2 and C3H8 concentration was observed, accompanied by the formation of total oxidation products (CO2 and H2O) and partial oxidation products (H2 and CO). Correspondingly, a hot spot formed on the catalyst surface (980 8C, measured by the pyrometer) and a steep rise was observed in the gas temperature (up to 945 8C, measured by the thermocouple). In line with the occurrence of the endothermic steam reforming reaction, the evolution of H2O showed a maximum. Moreover, the temperature of the solid surface and in the gas phase decreased toward the exit of the honeycomb. Qualitatively, this behavior is what our and other research groups have observed also in the case of a CH4CPO experiment on rhodium, and can be fully explained by the catalytic production of syngas on rhodium. Thus, integral measurements (i.e., measurements of temperature and composition collected exclusively at the reactor outlet) would have suggested the unique existence of a heterogeneous process. A purely catalytic process was, for instance, [*] Dr. A. Donazzi, D. Livio, Dr. M. Maestri, Prof. A. Beretta, Prof. G. Groppi, Prof. E. Tronconi, Prof. P. Forzatti Laboratory of Catalysis and Catalytic Processes Dipartimento di Energia, Politecnico di Milano Piazza Leonardo da Vinci 32, 20133 Milano (Italy) Fax: (+39)0223993284 E-mail: alessandra.beretta@polimi.it Homepage: http://www.lccp.polimi.it
Read full abstract