Ph.D Thesis

Ph.D StudentAbir Hadas
SubjectUsing DFT for Process Intensification by Simultaneous
Reactor and Catalyst Optimization: Scaled Down
Membrane Reformer
DepartmentDepartment of Chemical Engineering
Supervisor PROFESSOR EMERITUS Moshe Sheintuch
Full Thesis textFull thesis text - English Version


The recent interest in transportation fueled by hydrogen caused increased attention in the production of hydrogen from natural gas in small facilities. These reactions are traditionally performed in large facilities. It is of interest to develop a methodology for catalyst selection and kinetic modeling of hydrogen production reactions, in order to apply them under possible new situations. In this thesis we use first principle calculations, by applying DFT approaches with Transition State Theory, to describe the kinetics of three rates of significance in a Pd-membrane catalytic reactor for hydrogen production from methane. Two reactions are involved, Water Gas shift (WGS) and Methane Steam Reforming (SR), for which we study the kinetics of (111) faces of Cu, Ni, Pt, Pd, Ir and Ag. As well as hydrogen transfer rate in a Pd and Pd3Ag membranes. 

The WGS mechanism incorporates water dissociation, CO oxidation and carboxyl formation as its rate determining steps. Rates are limited by water dissociation and CO adsorption inhibition. Hydroxyl dissociation is found to be too small to account for experimental results; atomic oxygen is formed by hydroxyl disproportionation.

The SR mechanism employed incorporates water dissociation and CH4 dissociative adsorption as its rate determining steps. The mechanism includes CHO formation as an alternative path to the direct CH dissociation path which was found to have high activation energy.

Scaling and BEP relations were used for both reactions to predict reaction rate as a function of adsorption energies of two key components (e.g. C and O).

Hydrogen transfer rate through the selective membrane was evaluated using Fick 's law with the diffusivity constant calculated from first principles while accounting for adsorption and subsurface penetration. The inhibition due to surface adsorption of possible co-adsorbents was accounted for using DFT-calculated adsorption energies.

The main novelties are the use of an overall mechanism that accounts for the fact that the controlling mechanism may be different on different metals and incorporating the CO coverage effect. We suggest a fast and simple method to assess entropy change and transition state entropy.

For hydrogen transfer in the membrane we modeled the sub layer penetration up to 3 layers as well as the behavior deep within the bulk by using periodic conditions in all three directions. We have considered the option of different paths for hydrogen transfer in the Pd membrane.  It should be emphasize that we have not resorted to parameter fitting.