An Evolutionary Algorithm for <i>de Novo</i> Optimization of Functional Transition Metal Compounds

Chu, Yunhan; Heyndrickx, Wouter; Occhipinti, Giovanni; Jensen, Vidar R.; Alsberg, Bjørn K.

doi:10.1021/ja300865u

Cited by 98 publications

(153 citation statements)

References 85 publications

Supporting

Mentioning

144

Contrasting

Order By: Relevance

“…The systematic analysis of a large number of electronically and sterically modified precatalysts in various RCM reactions by using the methodology reported here has led to a more detailed understanding of the factors contributing to the success of ring‐closing metathesis reactions. This enhanced understanding should aid the quest for more powerful catalysts for such reactions 57…”

Section: Resultsmentioning

confidence: 99%

Ring‐Closing Metathesis Reactions: Interpretation of Conversion–Time Data

Thiel

Wannowius

Wolff

et al. 2013

Chemistry A European J

View full text Add to dashboard Cite

Conversion-time data were recorded for various ring-closing metathesis (RCM) reactions that lead to five- or six-membered cyclic olefins by using different precatalysts of the Hoveyda type. Slowly activated precatalysts were found to produce more RCM product than rapidly activated complexes, but this comes at the price of slower product formation. A kinetic model for the analysis of the conversion-time data was derived, which is based on the conversion of the precatalyst (Pcat) into the active species (Acat), with the rate constant k(act), followed by two parallel reactions: 1) the catalytic reaction, which utilizes Acat to convert reactants into products, with the rate k(cat), and 2) the conversion of Acat into the inactive species (Dcat), with the rate k(dec). The calculations employ two experimental parameters: the concentration of the substrate (c(S)) at a given time and the rate of substrate conversion (-dc(S)/dt). This provides a direct measure of the concentration of Acat and enables the calculation of the pseudo-first-order rate constants k(act), k(cat), and k(dec) and of k(S) (for the RCM conversion of the respective substrate by Acat). Most of the RCM reactions studied with different precatalysts are characterized by fast k(cat) rates and by the k(dec) value being greater than the k(act) value, which leads to quasistationarity for Acat. The active species formed during the activation step was shown to be the same, regardless of the nature of different Pcats. The decomposition of Acat occurs along two parallel pathways, a unimolecular (or pseudo-first-order) reaction and a bimolecular reaction involving two ruthenium complexes. Electron-deficient precatalysts display higher rates of catalyst deactivation than their electron-rich relatives. Slowly initiating Pcats act as a reservoir, by generating small stationary concentrations of Acat. Based on this, it can be understood why the use of different precatalysts results in different substrate conversions in olefin metathesis reactions.

show abstract

Section: Resultsmentioning

confidence: 99%

Ring‐Closing Metathesis Reactions: Interpretation of Conversion–Time Data

Thiel

Wannowius

Wolff

et al. 2013

Chemistry A European J

View full text Add to dashboard Cite

show abstract

“…Particularly the introduction of multidentate ligands leads to a stabilization of the scaffold around the transition metal center due to the chelate effect . In recent work, Jensen and coworkers emphasized the necessity of including chelate ligands in their molecular design approaches.…”

Section: Protocol For Shell‐wise Construction Of An Embedding Environmentioning

confidence: 99%

Stabilization of activated fragments by shell‐wise construction of an embedding environment

2017

View full text Add to dashboard Cite

An activated fragment which is structurally unstable when considered isolated can be stabilized through binding to a suitable molecular environment; for instance, to a transition-metal fragment. The metal fragment may be designed in a shell-wise build-up of a surrounding molecular environment. However, adding more and more atoms in a consecutive fashion soon leads to a combinatorial explosion of structures, which is unfeasible to handle without automation. Here, we present a fully automated and parallelized framework that constructs such an embedding environment atom-wise. Molecular realizations of such an environment are constructed based on simple heuristic rules intended to screen a sufficiently large portion of the possible compound space and are then subsequently optimized by electronic structure methods. (Constrained-optimized) structures are then evaluated with respect to a scoring function, for which we choose here the concept of gradient-driven molecule construction. This concept searches for structure modifications that reduce the forces on all atoms. We develop and analyze our approach at the example of CO activation by reproducing a known compound and mapping out possible alternative structures and their effect on the stabilization of a (bent) CO ligand. For all generated structures, the nuclear gradient on the activated fragment and its coordination energy are evaluated to steer the design process. © 2017 Wiley Periodicals, Inc.

show abstract

“…Molecular catalyst motifs remain an attractive target for computational screening efforts. Although some experimental and computational screening efforts have been carried out for inorganic complex discovery, robust, user‐friendly tools for the rapid generation and assessment of inorganic complexes are not widely available.…”

Section: Introductionmentioning

confidence: 99%

molSimplify: A toolkit for automating discovery in inorganic chemistry

Ioannidis¹,

Gani²,

Kulik³

2016

J Comput Chem

198

268

View full text Add to dashboard Cite

We present an automated, open source toolkit for the first-principles screening and discovery of new inorganic molecules and intermolecular complexes. Challenges remain in the automatic generation of candidate inorganic molecule structures due to the high variability in coordination and bonding, which we overcome through a divide-and-conquer tactic that flexibly combines force-field preoptimization of organic fragments with alignment to first-principles-trained metal-ligand distances. Exploration of chemical space is enabled through random generation of ligands and intermolecular complexes from large chemical databases. We validate the generated structures with the root mean squared (RMS) gradients evaluated from density functional theory (DFT), which are around 0.02 Ha/au across a large 150 molecule test set. Comparison of molSimplify results to full optimization with the universal force field reveals that RMS DFT gradients are improved by 40%. Seamless generation of input files, preparation and execution of electronic structure calculations, and post-processing for each generated structure aids interpretation of underlying chemical and energetic trends. © 2016 Wiley Periodicals, Inc.

show abstract

An Evolutionary Algorithm for de Novo Optimization of Functional Transition Metal Compounds

Cited by 98 publications

References 85 publications

Ring‐Closing Metathesis Reactions: Interpretation of Conversion–Time Data

Ring‐Closing Metathesis Reactions: Interpretation of Conversion–Time Data

Stabilization of activated fragments by shell‐wise construction of an embedding environment

molSimplify: A toolkit for automating discovery in inorganic chemistry

Contact Info

Product

Resources

About