Molecular modeling has become a necessary tool in state-of-the-art catalysis research, and it has been prominent in the CATMAT collaboration. Reaction and activation energies may be computed, allowing for predictions of thermodynamics, kinetics, and reaction mechanisms. Modelling is most useful when it is combined with experimental methods in a close interdisciplinary approach, and this is how we apply modelling in CATMAT.

Methods:

• Methods based on density functional theory (DFT). These can be complemented with ab initio calculations (HF, MP2, CCSD), which are useful for quality assurance or when high accuracy is required.
• Band structure calculations for crystalline solids. These may be based on DFT or ab initio approaches.
• Methods based on molecular mechanics have proven useful in our work (diffusion, sorption, structures of certain classes of crystalline solids)

Platinum diimine complex with oxidatively added methane

Computed activation energies for different reaction paths. Oxidative addition is clearly more feasible

Some cases:

Heterolytic activation of methane by a homogeneous platinum catalyst
Diimine complexes of platinum may activate methane at very mild conditions. Through a combination of experimental studies and quantum chemical calculations, we elucidated the reaction mechanism for this reaction. (see publication list)

Methanol to hydrocarbons
The collaboration leading to the CATMAT centre goes back to the 1980’s, when the (still ongoing) studies of the methanol-to-hydrocarbons (MTH) reaction commenced. In a series of papers, we have described the detailed mechanism for activation of methanol and the subsequent formation of alkenes, which proceeds through a “carbon pool” consisting of alkylated cationic arenes. In all, the comprehensive description afforded by molecular modelling has meant new possibilities for improving catalysts for the MTH process. (see publication list)

The Figure shows the transition state structure for the ring contraction of the heptamethylbenzenium ion

Some aspects of heterogeneous catalysis may be modelled by gas-phase reactions. An example is the reconstruction of protonated alkylbenzenes. This is an important part of the carbon pool mechanism. The Figure shows the transition state structure for the ring contraction of the heptamethylbenzenium ion.

The figure illustrates the transition state for direct dimerization of ethene

Sometimes, a surface may be fruitfully modelled by a small cluster of atoms. This has been used in a study of acid-catalyzed dimerization of alkenes.  Here, we were able to discriminate between different reaction mechanisms. The figure illustrates the transition state for direct dimerization of ethene.

Other aspects of the mechanisms require explicit modelling of the solid.  In these cases, we resort to so-called band structure models. We are in the process of investigating acid catalysis with such extended models. The picture illustrates 9 unit cells of the catalytic material SAPO-34. (See publication list)

Complex hydrides as hydrogen storage materials
An important global challenge is the safe and efficient storage of hydrogen as an energy carrier for automotive applications. We have studied the thermodynamics of different complex hydrides in an attempt to find new candidates for hydrogen storage materials.  (see publication list) The Figure shows the crystal structure of NaMgH3 as calculated using DFT-based band structure methods. Note the (MgH3)n  chains and the solitary sodium ions.

The computational screening of different materials markedly improves the efficiency of the search for the optimal hydrogen storage medium, as experimental work may be focussed on the candidates that are predicted to have the most desirable properties.

Metal Organic framework as selective sorption of CO₂
CO₂ capture is another challenge. Working both experimentally and computationally, we investigate Metal Organic Framework (MOF) materials as, among other things, solid sorbents for the selective sorption of CO₂. (see publication list) The Figure shows a carbonic acid molecule adsorbed on a model for a MOF material that has been synthesized in our laboratories. Again, computational work elucidates reaction mechanisms and helps to docus experimental investigations.

See publications list for Theoretical tools