Multi-scale Modelling of Oxygen Carriers in Chemical Looping Combustion Process

Abstract

Climate change is one of the major concerns affecting society. Given the severe effects of global warming, efficient CO2 capture and storage (CCS) technologies have become an urgent necessity. One of the major challenges of CO2 disposal is the intensive energy consumption associated with current CO2 capture technologies. Chemical looping combustion (CLC) is an emerging technology that requires lower energy expenditure compared to other CO2 capture methods. The key is to use an oxygen carrier (OC), which avoids direct contact between air and a fossil fuel. OC development is thus key to improve CLC performance. Although numerous studies have been reported in the literature regarding OC development, the majority of these studies entail experimental investigations. Theoretical studies on this subject have been limited. As a result, the reaction mechanisms, the microscopic insights into the OC performance and the significant factors influencing OC performance are still not clear even for some of the most popular OCs such as NiO. In addition, multi-scale simulations combining density functional theory (DFT) analysis and microscopic modelling in this area are scarce. This study provides a comprehensive investigation of syngas adsorption and combustion, using NiO as the OC through developing multi-scale models those take into account the effects of oxygen vacancies and nearest neighbours. An analysis of the syngas adsorption principle on NiO while considering the neighbouring effects was considered first. In particular, this work described the adsorption principles of syngas (i.e. CO and H2) on a clean NiO (100) surface under single and multiple first nearest neighbouring effects using DFT analysis. The results showed that the adsorption stability of CO and H2 is mostly weakened by the first neighbour compared to the second, third and fourth neighbours. With the same species as nearest neighbours (i.e. uniform adsorption), syngas adsorption stability was reduced when the number of neighbours increased. However, when compared to uniform adsorption, the adsorption stability of CO and H2 was slightly stronger with neighbouring sites occupied with different species (hybrid adsorption). In addition, a lower degree of symmetry seemed to strengthen CO and H2 adsorption. Results from this DFT study showed that the adsorption stability of CO and H2 in the presence of neighbours is highly related to steric, hybrid and symmetry effects. This study is key for the development of a multi-scale model for this system. Next, DFT calculations of syngas combustion with NiO were conducted to reveal the elementary reaction mechanisms: CO oxidation proceeds through a one-step mechanism while H2 oxidation proceeds through a three-step mechanism. Among them, H2 decomposition was proven to be the controlling step that dominated the overall syngas combustion process. These results were used to build a DFT-based mean-field (MF) multi-scale model, which verified the accuracy of the proposed reaction kinetics. The results from this multi-scale model showed that high temperatures and low pressures will lead to high CO2/H2O product ratios. The reaction kinetics obtained from this study were used to further analyze other factors that affect OC performance. A theoretical analysis of the studied system was conducted to gain insights on the vacancy effects on syngas adsorption, syngas oxidation and oxygen migration. The adsorption analysis and the proposed reaction mechanisms showed that the presence of the defective sites benefit the syngas oxidation reactions. In particular, H2 oxidation changed from a 3-step process on a perfect surface (i.e. without vacancies) to a 2-step process on a defective surface. The CO oxidation reaction was shown to dominate the overall syngas oxidation process. In addition, the outward diffusion direction of oxygen migration was observed from the bulk side to the surface. The resulting reaction kinetics and the vacancy effects were validated against experimental data using the DFT-based MF multi-scale model. The neighbouring effects on syngas oxidation were studied next using DFT calculations. An analysis on the activation energy showed that CO oxidation is slightly weakened by the CO neighbours, but it is enhanced in the presence of H2 neighbours. Meanwhile, H2 decomposition, hydrogen migration and H2O formation were mostly enhanced by their neighbours, with the exception of the three H2 neighbour configuration. In addition, the CO neighbours resulted in more significant changes in the reaction equilibrium. The resulting neighbouring effects on syngas adsorption and combustion were used to establish a DFT-based kinetic Monte Carlo (kMC) multi-scale model. The results from this model indicated that CO adsorption is the most sensitive step to neighbouring effects while CO oxidation is the least sensitive. OC conversion is enhanced by the neighbouring effects on CO adsorption and H2O formation. Moreover, the neighbouring effects on the H2 chemisorption weakened the OC conversion. The significant changes observed in OC conversion with and without neighbouring effects implied that this studied phenomena are key to predict a realistic OC performance. Overall, the multi-scale models developed in this research revealed the adsorption principle and reaction mechanisms of the studied system, while also considering the critical influencing factors of vacancies and nearest neighbours. Electronic analyses were additionally conducted on each step in this investigation to further support the conclusions. The vacancies generally enhance the OC performance. The neighbouring effects, meanwhile, benefit the syngas oxidations but weaken the syngas adsorption process. The significant changes caused by the effects of vacancies and the nearest neighbours implied that it is critical to consider these effects to capture the OC performance. The developed multi-scale models related the electronic-distribution-based DFT results to the experimental observations at the macroscopic scale. Therefore, the challenge of validating the results from DFT analyses by the experimental observations were overcome. Additionally, the developed multi-scale models also estimated the reaction kinetics and their influencing factors under practical operating conditions. The proposed DFT-based kMC model provided a method to consider the dynamic effects caused by the changing surface environment. This multi-scale study served to fill some of the current gaps in the literature in this area.