From Abstract to Concrete : Advancing the Globally Optimal Catalyst Method

This Thesis explores avenues of extension and improvement to the Globally Optimal Catalyst (GOCAT) approach. In this inverse design method, the goal is to find in silico an optimal catalytic embedding for arbitrary chemical reactions, in the shape of abstract interaction entities. These have, in the past, mostly taken the form of point charges with fixed spatial positioning. This general idea is extended here to include van-der-Waals interactions and even molecular fragments in a QM/MM setting. Additionally, new implementational work is done in the framwork of OGOLEM, in which the GOCAT method is implemented. It allows a new way of optimizing a frame-wise embedding for sterically involved reactions, and the inclusion of interaction entities close to reaction centers in adaptive re-optimization of the reaction path, which was not possible before.

Furthermore, the complexity required in the interaction entities was previously unknown, i.e. how many point charges are needed for optimal catalysis. By employing a fitting scheme in OGOLEM using its evolutionary algorithm (EA), it could be shown that very complex GOCAT structures can be reduced to a minimum number of required interactions.

One of the most pressing issues of inverse design methods is,  however, their transferability to actual chemical structures. With the new additions to the method, especially the ability to include molecular fragments and to reduce the complexity of GOCATs to their bare minimum, the approach is inching closer to a wider applicability, e.g. in biocatalyst design. Early results also show that method can be used successfully as a guidance in the design of catalytic surfaces and organic frameworks.

Vorschau

Rechte

Nutzung und Vervielfältigung:


CC BY 4.0

Bitte beachten Sie, dass einzelne Bestandteile der Publikation anderweitigen Lizenz- bzw. urheberrechtlichen Bedingungen unterliegen können.

Zitieren

Zitierform:
Zitierform konnte nicht geladen werden.