Date of Award

Spring 5-22-2021

Degree Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Biomedical and Chemical Engineering

Advisor(s)

Jesse Q. Bond

Keywords

biphasic system, Brønsted acid site, Lewis acid site, Solid acids, γ-valerolactone

Subject Categories

Chemical Engineering | Engineering

Abstract

Lignocellulosic biomass is an alternative carbon source for industrial applications. γ-valerolactone (GVL) is an example of biomass-derived platform chemicals that provides multiple commercial potentials. Of interest here is the acid-catalyzed ring opening of GVL to form pentenoic acid (PEA) isomers and following decarboxylation to produce butenes. In this study, GVL ring opening, and decarboxylation are examined in both liquid and gas phases over various catalytic systems aiming at improving the efficiency in selective preparation of target product. An initial attempt was made in a biphasic reactor to improve the yield to PEA catalyzed by mineral acids (H2SO4). A wide range of solvents was tested for their affinity towards GVL and PEA. And alkanes appeared to have the best resolution to separate PEA from GVL. A two-parameter Margules model was further applied to regress the activity coefficient, and it successfully described the phase distribution at any given compositions in the biphasic system. Furthermore, an activity-based model was proposed to simulate the reaction in the biphasic reactor. Finally, some experiments were performed to verify the simulation results. The results turned out that theocratically, the biphasic system can significantly improve the PEA yield. However, due to practical limitations and uncertainty in simulation, the PEA yield can be enhanced by introducing a second phase but not as much as predicted in the simulation. Furthermore, four solid acids abundant in Lewis acid sites were considered for GVL/PEA interconversion in the presence of water. SiO2/Al2O3, γ-Al2O3, TiO2, and ZrO2 are characterized and tested in a pack bed reactor. Catalytic performances were examined over various contact times and temperatures. γ-Al2O3 proved to be an optimistic catalyst that can realize a high yield to PEA with high selectivity at a wide operation window. SiO2/Al2O3 can also achieve a high yield to PEA at certain conditions. But the selectivity to PEA was relatively low. ZrO2 exhibited a high selectivity to PEA, but its overall activity was limited due to a lack of active sites on the surface. TiO2 was not practical for this application. Proximity to equilibrium was found to be the critical element influencing selectivity to PEA. Meanwhile, the GVL ring opening activity was more controlled by Lewis acidity instead of Brønsted acidity in this system. Deactivation was observed on these materials, and calcination can only partially regenerate the catalysts. Finally, different framework zeolites that had a high density of Brønsted acid sites were examined. GVL decarboxylation was performed over MFI, FAU, BEA, FER, and MOR zeolites in the gas phase. These zeolites were well characterized and were found that Brønsted acid site density, Lewis acid site density as well as BAS: LAS ratio increased as the aluminum content decreased in zeolites. Brønsted acid site was the primary active site for GVL decarboxylation instead of the Lewis acid site. Brønsted acidity in different framework zeolites seemed to be identical. However, the intrinsic activity of materials was not only dependent on the strength of Brønsted acid sites. Significantly high activity was observed in MFI framework zeolites. The confinement effect and the local environment of acid sites may contribute to the increased activity in MFI zeolites. Severe deactivation was found in zeolites due to coking formation. But the catalysts were fully regenerable after calcination. Pore diameter and micropore area were the most critical elements for designing an efficient and stable zeolite catalyst.

Access

Open Access

Download

Share

COinS