Hydrogen production by steam reforming of hydrocarbons from biomass gasification modeling and experimental study

Abstract

Hydrogen and synthesis gas are the most important energy carriers in the sustainable energy supply system of the future and play a key role in production of many chemical and petrochemicals in oil and energy industries. Steam reforming of light hydrocarbons, like methane, ethylene and propylene, produced from biomass gasification, is proven to have many advantages and environmental benefits. It is an efficient and economical option in industry and the most common process for hydrogen production. Steam reforming process converts light hydrocarbons from biomass into hydrogen and carbon monoxide at high temperatures in the presence of a nickel or noble metal based catalysts. The process uses fixed bed reactors, filled with catalyst particles in order to increase the reactive surface area between the flow phases and improve the efficiency of the reactions. Due to the strongly endothermic nature of the process, a large amount of heat is supplied to the reactor and therefore the tube wall and the catalyst particles are exposed to significant axial and radial temperature gradients. In developing such reactors, the knowledge of the temperature profile within the reactor is important for designing and optimizing the catalysts structure and the reactor geometry to achieve the best performance. However, due to insufficient room for a thermocouple or to potential interference with local fluid dynamics, temperature measurement within the reactor becomes difficult. Therefore, to evaluate not only the local temperature profiles but the whole reactor performance in terms of conversion and selectivity, accurate descriptive and predictive models are necessary. A steady state experimental and theoretical model is developed to investigate the performance of the catalytic steam reforming tubular reactor. A packed bed reactor consisting of cylindrical catalyst particles is simulated numerically to investigate the reforming chemistry and evaluate the diffusional resistances. The mathematical modeling and solutions consist of both pseudo-heterogeneous model and a pseudo-homogeneous model. Pseudo-homogeneous reactor models do not distinguish between the gas and solid phases in the fixed bed reactor. Averaged properties are used and the temperature, pressure and concentrations are equal inside the catalyst and in the gas phase. Heterogeneous models distinguish between the fluid phase and the solid phase to correctly account for the heat transfer in the reactor. Reaction kinetics, reaction rates, effectiveness factors and partial differential equations including mass transfer and energy balance equations are studied in detail. This thesis includes the extensive review of the already published research papers in this field to find empirically correlations for packed bed reactors, suitable for different cases. Wide varieties of correlations are used to calculate effectiveness factors, dispersion and diffusion coefficients according to the type and shape of the catalysts. Hydrogen sulfide is known to deactivate nickel based steam reforming catalysts by chemisorptions on the metal surface. The metal sulfur bond is so strong and even at extremely low gas phase concentrations of hydrogen sulfide the catalytic activity is substantially reduced. Using conventional sulfur cleaning has a negative effect on process efficiency and steam reforming has to be run without cleaning the gas prior to the reactor. This thesis aim is to describe and model sulfur deactivation effects and steam reforming process in a combined kinetic equation. All chemical reactions are mathematically solved and modeled by the state of the art, COMSOL Multiphysics 4.3 software, which represents a powerful and general class of finite element methods and techniques for approximate solution of partial differential equations. Choosing similar operating conditions in terms of inlet composition and temperature as in experiments allows a direct comparison between experimental and modeling results to validate and see how well the simulation predicts the experimental results. The result of the modeling shows temperature and pressure distribution, and gas composition along the reactor and predicts accurate results when the inlet and process conditions are changed. They are also able to find the optimum characteristics and geometry dimensions for a steam reformer used for a specific process. The effect of varying operating parameters like steam to carbon molar ratio, temperature, gas composition, space velocity, different amount of sulfur on the hydrocarbon conversion, hydrogen yield and reforming efficiency are investigated carefully.