Exploring the role of surface and porosity in CO2 capture by CaO-based adsorbents through response surface methodology (RSM) and artificial neural networks (ANN)

Abstract

In this study, synthetic CaCO3 materials were utilized as precursors for CaO-based CO2 sorbents. The investigation examined how various operating parameters—such as synthesis temperature (ST), stirring rate (SR), and surfactant percentage (SP)—impact the properties of the adsorbents. Samples were firstly characterized by X-ray diffraction and Scanning Electron Microscopy (SEM), which revealed that the prevalence of calcite or aragonite crystal phases in the synthetic CaCO3 precursors can be tuned by adequately choosing the dose of surfactant (Triton-X100®), so that it can be used as a crystal habit and growth modifier. The calcination process applied to the CaCO3 precursors leads to the formation of partially sinterized cubic crystals of CaO, accompanied by minor quantities (< 5 %) of additional compounds like Ca(OH)2 or CaSO4. Specific surface area (SBET) and porosity were determined by measuring the N2 adsorption isotherms. A CaCO3 sample with an unprecedented value of SBET as large as 116 m2/g was prepared operating under optimal conditions. SBET and pore volumes were successfully correlated with the CO2 uptake capacity of the samples. SBET is more influential for experiments carried out under diluted CO2 atmosphere. When pure CO2 is used, the influence of meso- and micropore volumes (Vme and Vmi) is clearly predominant, which suggests that in this latter case diffusion through the porous texture of the samples plays a more remarkable role. A double-way approach through Response Surface Methodology (RSM) and the use of Artificial Neural Networks (ANNs) has been used to analyze the CO2 uptake capacity of the samples. Within the operational interval, excellent results were obtained for pure and diluted CO2 flow, and RSM and ANNs have demonstrated to be a very efficient tool to correlate the behavior of the CaO-based materials as CO2 sorbents with the surface area and pore volumes of the samples. Valuable information on (i) the importance of the different factors under study; (ii) their influence on the surface and porosity of the CaO-derived sorbents; and (iii) the subsequent CO2 capture performance of the sorbents has been obtained. The results suggest that four parameters have a statistically significant influence on CO2 uptake. These parameters are SR, the square of SR, its interaction with SP, and the square of SP. Additionally, the study assessed the stability of the CaO-based sorbents over 11 consecutive calcination-carbonation cycles. By adequately choosing the synthesis strategy and conditions, an almost negligible shrinkage effect can be achieved, resulting in a more sustained uptake capacity throughout the cycles.

Keywords

• Calcium looping
• CO2 capture
• Surface and porosity
• Response surface methodology
• Artificial neural networks