Modeling and Experimental Verification of Physical and Chemical Processes during Pyrolysis of a Refuse-Derived Fuel Valerio Cozzani, † Cristiano Nicolella, ‡ Mauro Rovatti, ‡ and Leonardo Tognotti* ,† Dipartimento di Ingegneria Chimica, Chimica Industriale e Scienza dei Materiali, Universita ´ degli Studi di Pisa, via Diotisalvi n. 2, 56126 Pisa, Italy, and Istituto di Scienze e Tecnologie dell’Ingegneria Chimica, Universita ´ degli Studi di Genova, via Opera Pia n. 15, 16145 Genova, Italy A model for refuse-derived fuel (RDF) conventional pyrolysis in a fixed-bed reactor is presented. The model investigates the influence of the heat- and mass-transfer processes on the pyrolysis product yields. Solid degradation reactions have been modeled by assuming that the interactions between the main RDF components during pyrolysis are negligible and that the RDF pyrolysis behavior may be considered as the sum of the separate behaviors of “primary reacting species”. The model accounts for conductive and convective heat transfer within the solid matrix and secondary tar-cracking reactions, as well as for variability in physical properties and in the void fraction of the pyrolyzing material. Quite good agreement was found between model results and experimental data obtained for conventional pyrolysis of a RDF in a laboratory-scale fixed- bed reactor. The model is able to predict the temperature transients, the rate of gas generation, and the product final yields during conventional pyrolysis of RDF. Introduction Disposal and energy recovery from municipal solid wastes (MSW) through economically viable technologies is of worldwide importance. Among the conventional means of disposal, thermal treatments of solid wastes may offer several benefits, providing a captive energy source, reducing the quantity of waste material to be deposited in landfills, and reducing pollutant-generating problems (Buekens and Shoeters, 1986). Furthermore, the fundamental knowledge of the phenomena that occur during the thermal decomposition of waste- derived materials is also important since pyrolysis is an important stage for other thermochemical processes, such as gasification and incineration. The use of refuse-derived fuels (RDF) as starting materials for pyrolysis and gasification processes pre- sents several advantages over the use of municipal or other solid wastes. RDF are produced by selecting the combustible fraction of MSW by mechanical sorting and processing. This results in a relatively constant com- position and good transportation and storage possibili- ties, since putrescible components are eliminated (Muh- len et al., 1989). The implementation of pyrolysis and gasification processes depends on the reliable design of large-scale units, in which the pyrolysis reactor plays an important role. For this reason, the understanding of pyrolysis chemistry and physics is of great importance. No specific fundamental model for RDF pyrolysis is present in the literature. Nevertheless, the physical and chemical problems are similar to those found in modeling biomass thermal degradation, since cellulose and lignin are the main components of both biomass and RDF (Rampling and Hickey, 1988). Starting from the study of Bamford et al. (1946), considerable work has been done on the development of reliable models of biomass pyrolysis. Most of the models presented con- sider the solid as a single homogeneous species, include lumping of the reaction products into three main categories (char, tar, and gas), and consider single-step reaction kinetics. Obtaining predictions of product yields on a wider range of operating conditions required the introduction of more complex kinetic models, based on single-component multireaction schemes (Panton and Rittman, 1972; Chan et al., 1985). A further improve- ment has been the inclusion of secondary tar-cracking reactions in the kinetic scheme (Curtis and Miller, 1988; Hastouglu and Berruti, 1989; Di Blasi, 1993a). Different assumptions have been used in the descrip- tion of the heat- and mass-transfer phenomena coupled to the thermal degradation of the solid matrix. Several models considered conductive heat transfer, a pseudo- steady-state gas phase, and no pressure gradients within the solid (Kung, 1972; Panton and Rittman, 1972; Pyle and Zaror, 1984; Capart et al., 1985; Viller- maux et al., 1986; Wichman and Atreya, 1987; Curtis and Miller, 1988; Koufopanos et al., 1991). A few models attempted to include convective heat transfer within the solid due to the gas-phase flow of the volatiles generated by the pyrolysis process (Chan et al., 1985; Alves and Figueiredo, 1989; Bilbao et al., 1993), while the radiative contribution to heat transfer was found to be of limited importance in the range of temperatures of interest for the pyrolysis process (Curtis and Miller, 1988). In some models the pseudo-steady-state as- sumption has been removed and the non-steady-state convective flow of volatiles toward a solid surface has been accounted for by the Darcy law (Kansa et al., 1977; Di Blasi, 1993a) or the Dusty gas equation (Hastaouglu and Berruti, 1989). A survey of the main results of these works is not within the scope of this paper. Details may be found in an exaustive review (Di Blasi, 1993b). The experimental work on RDF pyrolysis has pointed out that the product yields and temperature range of pyrolysis reactions are qualitatively similar for RDF and biomass pyrolysis processes (Evans and Milne, 1988; Mallya and Helt, 1988; Lai and Krieger-Brockett, 1992). However, RDF is a more complex substrate, yielding products that are not present in biomass pyrolysis, mainly due to the presence of plastic components. * Author to whom correspondence should be sent: tele- phone, (39)-50-511111; fax, (39)-50-511266; e-mail, tognotti@ ccii.unipi.it. † Universita ´ degli Studi di Pisa. ‡ Universita ´ degli Studi di Genova. 90 Ind. Eng. Chem. Res. 1996, 35, 90-98 0888-5885/96/2635-0090$12.00/0 © 1996 American Chemical Society