Wet and dry desulphurization technologies: modelling and simulation.

Abstract

Energy demand continues to increase with population growth and economic development therefore numerous technologies have been proposed over the years to convert coal to electricity. Turbine cycles have received a great deal of interest in the last decade, especially the integration of the gasification with the gas turbine/steam turbine combined cycle (lGCC). The major components of a coal-based IGCC cycle are coal handling and gasification, sulphur removal, gas turbine cycle, steam turbine cycle. Sulphur removal processes are generally classified into amine absorption processes (wet-based systems), and hot desulphurization processes using metal oxide sorbents (dry-based systems). Amine absorption processes are proven commercially available technologies but are energy intensive. Hot desulphurization processes have a cost advantage over amine absorption processes, but the degree of long-term reliability is still to be proven. The main aim of this study was to compare the commercial operability and economic feasibility between an amine absorption process (using an MDEA-based amine solution) and two hot desulphurization processes (the first using a free zinc oxide sorbent and the second using a silica supported zinc oxide sorbent). Aspen Plus simulation package was used to develop the simulation models. A coupled gasification and sulphur removal system was designed for each process. The gasifier model takes the processes of gasification into account i.e. coal drying, coal pyrolysis, char gasification and char combustion. A fluidized bed reactor was chosen to carry out the gasification due to its good temperature control. Fluidization behaviour and reaction kinetics for char gasification and combustion were included in the model. A coal flow rate of 84 tons/day was selected (a typical flowrate of coal processed in an industrial gasifier) with an operating pressure of 30 atm and 900°C. A built-in amine data package was used to calculate electrolyte capabilities and kinetic reactions in the liquid phase, of the amine absorption process. MDEA/H2O was the amine solution used, due to its high affinity to sulphur compounds. This process required a pre-cooling and tar removal step prior to desulphurization and regeneration, and was achieved by a heat exchanger and a water spray tower respectively, at 30 atm. An absorber tower was used for desulphurization with operating temperatures between 40° and 60°C, and pressures around 55 atm. A distillation column was used for the regeneration process with operating temperatures between 110° and 120°C, and pressures around 1 atm. Multiple auxiliary units were set up along the process in order for process streams to meet process conditions. Pumps, compressors and turbines were added to cater for any pressure changes and heat exchangers along with its utility for any temperature changes. 10% of the regenerated amine solution was purged and remainder was recycled. For the hot desulphurization processes free zinc oxide sorbent and zinc oxide sorbent supported on silica. Fluidized bed reactor was chosen to carry out the desulphurization and regeneration, due to its good temperature control. Fluidization behaviour, elutriation characteristics and reaction kinetics from literature were incorporated in these models. Both the desulphurization and regeneration had an operating temperature of 550°C and 30 atm. Cyclone separators were used to separate solids carried over by the fluidized

beds. 90% of the regenerated free sorbent was purged and remainder was recycled, while 50% of the regenerated supported sorbent was purged with the remainder being recycled. A total annual cost (TAC) analysis was developed to compare the economic feasibility of each process, consisting of the capital and operating costs. The intrinsic rate constant was regressed from available experimental data for the supported sorbent. The intrinsic kinetics obtained from the regression was found to be six orders of magnitude lower than that reported in the literature due to the low intrinsic rate constant extracted from the regression may not be a true reflection of the reaction behaviour, if mass transfer limitations through the product layer diffusion were significant. The intrinsic rate constant and product layer diffusion rate constant from literature were used to represented the kinetics for the supported sorbent in the simulation. The hot desulphurization processes did not require pre-cooling or a tar removal step, which made these processes more thermally efficient than the amine absorption process. The amine absorption unit had a lower purge rate as compared to the hot desulphurization processes, resulting in better utilization of the Sulphur removing agent, therefore making the amine absorption process more effective. The capital cost for the amine absorption process was 105.657 mil.R, which was much higher than capital costs for the hot desulphurization processes (32.633 mil.R and 22.403 mil.R for the free sorbent and supported sorbent processes respectively), due to the larger amounts of process equipment required for amine absorption. The operating cost for the amine absorption process was R63.943 mil.R/yr, which was much higher than TAC for the hot desulphurization processes (R38.038 mil.R/yr and 24.991 mil.R/yr for the free sorbent and supported sorbent processes respectively). This was due to the utility requirements present in the amine absorption process. The TAC for the amine absorption process was R99.162 mil.R/yr, which was much higher than TAC for the hot desulphurization processes (R48.916 mil.R/yr and 32.459 mil.R/yr for the free sorbent and supported sorbent processes respectively). It was concluded that hot desulphurization processes are economically less intensive than amine absorption processes with lower waste production, and supported sorbents provide cheaper and more effective operation than free metal oxide sorbents.