Carbon dioxide capture from flue gases using dry sorbents

Abstract

The project entails capturing CO2 produced from industrial flue gases with dry adsorbents through gas adsorption to mitigate the CO2 emissions. Four commercially available activated carbon samples (CQ006, CQ30P, CQ650 and PCX1) (available in South Africa, from ChemQuest), derived from coal, coconut fibre and wood, were investigated in this study. The samples were comprehensive characterized using methods such as proximate analysis, ultimate analysis, surface area analysis, pore size distribution and volume analysis, as well as scanning electron microscopy analysis. The CO2 adsorption isotherms for each activated carbon sample was evaluated at low pressures, ranging from 0 to 114 kPa, at 0, 10, 20, 30, 40 and 55 °C. In efforts to determine the optimal adsorption isotherm model suitable for CO2 adsorption isotherm modelling on the activated carbon samples, the individual adsorption isotherms were modelled with eight adsorption isotherm models: Langmuir, BET, Dubinin–Radushkevich (D-R), Dubinin–Astakhov (D-A), Toth, Freundlich, Temkin and SIPS. The goodness of fit for each adsorption isotherm model was evaluated with quality of fit and average relative error. D-R presented as the best fitting adsorption isotherm model to describe the experimental adsorption isotherm data of the activated carbon samples. Thermodynamic analysis was conducted on the activated carbon samples to determine the change in enthalpy, entropy, Gibbs free energy and isosteric heat of adsorption. The adsorption rates of the activated carbon samples were evaluated at 40, 55, 70 and 85 °C with inlet CO2 concentrations of 5, 15 and 25 vol% at a pressure of 1 bar. A fixed bed reactor was designed and built to measure the adsorption rates of the activated carbon adsorbent at each selected temperature and pressure. The individual adsorption rates were modelled with five adsorption rate models: Pseudo first order (P1O), pseudo second order (P2O), Elovich, Avrami and the fractional order adsorption rate models, to find the optimal adsorption rate model suitable for CO2 adsorption rate modelling on the activated carbon samples. The Avrami adsorption rate model presented as the best fitting adsorption rate model on the experimental adsorption rate data. CQ650 was found to be the most suitable adsorbent for CO2 adsorption in terms of adsorption capacity and rate. The CQ650 sample is derived from coconut fibres, making it robust and resistant to attrition, which is ideal for the use in a dry carbon capture process.