Graduation Year


Document Type




Degree Name

MS in Chemical Engineering (M.S.C.H.)

Degree Granting Department

Chemical Engineering

Major Professor

Venkat R. Bhethanabotla, Ph.D.

Co-Major Professor

John N. Kuhn, Ph.D.

Committee Member

Scott W. Campbell, Ph.D.


carbon dioxide conversion, DFT, oxygen vacancy, reverse water gas shift-chemical looping


There is a pressing need for carbon dioxide (CO2) mitigation technologies as increasing emissions continue to threaten our environment. Carbon capture and storage (CCS) alone would not be able to achieve temperature rise below 2 °C. In addition, CCS cost increases considerably for power generation and other industries where CO2 separation is necessary. Conversion of CO2 to usable products is the most economical and feasible solution as it could offset CO2 separation cost. Liquid fuels are an attractive route since it is a commodity used globally and demand increases with population and economic growth. Conversion of CO 2 to liquid fuels would also reduce dependence on fossil fuels, increasing fuel economy security. This way CO2 would be recycled from emissions and transformed into fuel and then recycled again, potentially generating zero CO2 emissions, creating a closed carbon cycle.

CO2 can be converted to liquid fuels using various technologies. For instance, biomass, algae, and photochemical technologies can be used to directly convert CO2 to fuels, while, photochemical and thermochemical can be used to convert CO2 to CO, which can subsequently be converted to liquid fuels. The latter has been regarded as the most efficient route currently and potentially in the future. Currently, thermochemical routes outperform photochemical in conversion rates and material recyclability over several cycles.

Thermochemical CO2 conversion technologies include reverse water gas shift (RWGS), thermochemical cycles, and reverse water gas shift-chemical looping (RWGSCL). The differences between these technologies will be discussed in Section 1.2. RWGS-CL is a promising technology since it has achieved higher conversion rates than thermochemical cycles at lower operating temperatures. This is a two-step cyclic process, that first reduces a perovskite oxide (ABO 3) with hydrogen (H2) and subsequently oxidizes the oxygen vacant perovskite (ABO3-δ), thereby converting CO2 to CO and regenerating its structure. CO can subsequently be converted to hydrocarbons via Fischer Tropsch Synthesis. Performance of different perovskites compared to the state-of-the art material ceria (CeO2) are discussed in Section 1.3.

Energy requirements are reduced by doping the A-site and B-site with alkaline earth and transition metals. Perovskites containing cobalt (Co), iron (Fe), and manganese (Mn) have exhibited either enhanced CO2 conversion, lower operating temperatures or higher stability. The specific studies showcasing these properties are discussed in Section 1.4. We aim to introduce these notable properties to achieve a high performing material by simultaneous doping of Co, Fe, and Mn into the B-site of a lanthanum perovskite.

Three compositions, LaCo0.50Fe0.25Mn0.25O 3 (Co50), LaCo0.25Fe0.50Mn0.25O 3 (Fe50), and LaCo0.25Fe0.25Mn0.50O 3 (Mn50), were synthesized experimentally (Section 2.1) and modeled computationally (Section 2.5). Characterization techniques such as Xray diffraction (XRD), scanning electron microscopy (SEM), and X-ray photo-electron spectroscopy (XPS) were used to understand the effects of the three metals doped in the B-site. Temperature programmed reduction in hydrogen (TPR-H2) and oxidation with carbon dioxide (TPO-CO2) experiments were used to experimentally quantify the amount of oxygen vacancies and carbon monoxide generated (Section 2.4). Oxygen vacancy formation and CO2 conversion capability were studied using bulk oxygen vacancy formation energy (E vac) as descriptor.

The three compositions achieved an orthorhombic perovskite phase with well dispersed metals on the bulk structure. Evac decreased with transition metal, Co50 having the lowest and Mn50 the highest. Experimental oxygen vacancy formation and CO yield were achieved as expected from E vac results. Co50 had the highest CO production (1780 μmol/g-perovskite) but resulted in an unfeasible isothermal process. Fe50 on the other hand had similar CO production (1420 μmol/g-perovskite) and ideal reduction and oxidation temperatures for isothermal RWGS-CL. A RWGS-CL experiment showed Fe50 produced ∼900 μmol/g-perovskite stably over five cycles. The high performance of Fe50 is attributed to the promotion of multiple valence states on the surface caused by the triple transition metal doping on the B-site.