화학공학소재연구정보센터
Industrial & Engineering Chemistry Research, Vol.36, No.3, 592-604, 1997
Hydrogenation of 2,4-Dinitrotoluene Using a Pd/Al2O3 Catalyst in a Slurry Reactor - A Molecular-Level Approach to Kinetic Modeling and Nonisothermal Effects
The kinetics of hydrogenation of 2,4-dinitrotoluene (2,4-DNT) using a 5% Pd/Al2O3 catalyst was studied in a semibatch slurry reactor in a temperature range of 323-363 K. Experimental data on the concentration-time and H-2 consumption-time profiles were obtained, and the effects of 2,4-DNT concentration, H-2 pressure, and catalyst loading were studied under both isothermal and nonisothermal conditions. A fundamental approach based on a molecular level description of the catalytic cycle has been used to derive the rate models. Several rate forms were derived considering different types of interactions, but the rate equations derived assuming that the reaction between the transient molecular species formed due to the interactions of H-2 and liquid phase components on different sites of Pd catalyst were found to best represent experimental data. The overall hydrogenation rate was found to vary by approximately second order with respect to catalyst loading, and this trend is adequately explained by the kinetic model proposed. It was found that the intraparticle diffusional effects were important for particle sizes (d(p)) > 3 x 10(-4) m, but the external mass-transfer (gas-liquid and liquid-solid) effects were unimportant. For a complex rate equation observed, an approximate expression for the overall effectiveness factor was derived and the experimental data for different particle sizes were found to agree with the predictions of the model incorporating intraparticle diffusion effects. Under certain conditions, a significant temperature rise was observed and the increase in temperature was found to vary with time and the initial set of conditions. A mathematical model to predict the temperature and concentration profiles in a semibatch reactor under nonisothermal conditions has been proposed. A comparison of the experimental data with model predictions showed an excellent agreement.