1997

Simell, Pekka

The aim of this work was to study the catalytic cleaning of gasification gas from tars and ammonia. In addition, factors influencing catalytic activity in industrial applications were studied, as well as the effects of different operation conditions and limits. Also the catalytic reactions of tar and ammonia with gasification gas components were studied. The activities of different catalyst materials were measured with laboratory-scale reactors fed by slip streams taken from updraft and fluid bed gasifiers. Carbonate rocks and nickel catalysts proved to be active tar decomposing catalysts. Ammonia decomposition was in turn facilitated by nickel catalysts and iron materials like iron sinter and iron dolomite. Temperatures over 850 °C were required at 2 000 h-1 space velocity at ambient pressure to achieve almost complete conversions. During catalytic reactions H2 and CO were formed and H2O was consumed in addition to decomposing hydrocarbons and ammonia. Equilibrium gas composition was almost achieved with nickel catalysts at 900 °C. No deactivation by H2S or carbon took place in these conditions. Catalyst blocking by particulates was avoided by using a monolith type of catalyst. The apparent first order kinetic parameters were determined for the most active materials. The activities of dolomite, nickel catalyst and reference materials were measured in different gas atmospheres using laboratory apparatus. This consisted of nitrogen carrier, toluene as tar model compound, ammonia and one of the components H2, H2O, CO, CO2, CO2+H2O or CO+CO2. Also synthetic gasification gas was used. With the dolomite and nickel catalyst the highest toluene decomposition rates were measured with CO2 and H2O. In gasification gas, however, the rate was retarded due to inhibition by reaction products (CO, H2, CO2). Tar decomposition over dolomite was modelled by benzene reactions with CO2, H2O and gasification gas. Operation conditions free of external or internal mass transfer limitations were used. Apparent first order kinetic parameters were determined for all the studied gas mixtures. In addition, for the CO2 reaction, a mechanistic model of Langmuir-Hinshelwood type was derived and tested. The best model was based on benzene single site adsorption as the rate determining step and non-dissociable adsorption of CO2. To be active in gasification gas carbonate rocks have to be in a calcined state. When the catalyst was carbonated and water was present, the activity of the catalysts was lost almost completely. This decline of activity closely followed the equilibrium decomposition pressure and temperature of CaCO3.