(16e) Computational Investigation of Solvent Effects On the Hydrolysis of Ether Linkages | AIChE

(16e) Computational Investigation of Solvent Effects On the Hydrolysis of Ether Linkages

Authors 

Fleming, K. - Presenter, University of Washington
Pfaendtner, J., University of Washington


Computational Investigation of Solvent Effects on the Hydrolysis of Ether Linkages

 

Kelly Fleming, Jim Pfaendtner

Department of Chemical Engineering, University of Washington

Seattle, Washington, 98105, USA

ABSTRACT

            Cellulosic biomass components are bonded by ether linkages that must be broken in the first steps of preparing them for fermentation or catalytic upgrading. Environmentally benign routes for biomass processing usually involves enzymatic hydrolysis of the ether bonds. The solvent, reaction conditions, substrate form, and the catalyst or enzyme used influence the efficiency that the ether linkage is hydrolyzed.1,2 We seek to gain molecular level insights into the specific role of the solvent on biomass degradation rates. Gaining such understanding will enable new rational design strategies as we continue to seek to make biofuels competitive with conventional energy sources.

            Dimethyl ether (DME) is the simplest form of an ether linkage between two alkyl groups. To model the reaction mechanism of ether hydrolysis, density functional theory (DFT) calculations were done on DME with and without an acid catalyst in water using Gaussian 09. The mechanisms between the catalyzed and uncatalyzed hydrolysis reactions were qualitatively and quantitatively quite different. The uncatalyzed reaction is a single step, where the catalyzed one is a double displacement reaction. As expected, the energy barrier for the uncatalyzed reaction was larger by a factor of about three. The systems were modeled with a range of implicit solvent models to relate the energy barriers to the solvent properties. We observed that using implicit solvents with the features of organic acids lowers the activation energy significantly. The system was then modeled with ab initio molecular dynamics (QM/MM) using CPMD and metadynamics. The quantum region in the CPMD calculation was the same geometry as the DME reaction modeled in Gaussian, and the molecular mechanics region being the ionic liquid solvents with water and ionic liquid systems.

            The system was expanded and modeled with a pyranose dimer, to include the presence of ring strain and more closely model structural features found in cellulose and hemicellulose. The pyranose dimer was modeled with an acid catalyst in water and in dilute acid solvents using DFT. The acid solvents showed a decrease in the energy barrier and a direct correlation between bond distances in the transition states and the dielectric constant of the solvent. The pyranose dimer was then modeled in ionic liquid solvents using CPMD and the geometric features were compared to those in water.

            (1)      Liang, X.; Montoya, A.; Haynes, B. S. J. Phys. Chem. B 2011, 115, 8199.

            (2)      Huber, G. W.; Iborra, S.; Corma, A. Chem. Rev. 2006, 106, 4044.

Topics