Architector for high-throughput cross-periodic table 3D complex building

Area of Science:

• Computational chemistry
• Organometallic chemistry
• Materials science

Background:

• Rare-earth and actinide complexes are vital for clean-energy technologies.
• Predicting three-dimensional (3D) structures of these organometallic systems is computationally challenging.
• This limitation hinders computational discovery in the field.

Purpose of the Study:

• Introduce Architector, a high-throughput computational code for in-silico synthesis and structure prediction.
• Enable the design of novel mononuclear organometallic complexes across the periodic table.
• Expand the scope of computational chemical discovery for metal complexes.

Main Methods:

• Architector utilizes metal-center symmetry, interatomic force fields, and tight-binding methods.
• It generates diverse 3D conformers from 2D inputs, including metal oxidation and spin states.
• The code is designed for s-, p-, d-, and f-block elements.

Main Results:

• Architector accurately predicts structures for over 6,000 experimentally determined complexes (via x-ray diffraction).
• The code captures a wide range of known chemical space and can design novel complexes.
• It generates and ranks conformers, aiding in potential energy surface exploration and force field training.

Conclusions:

• Architector offers a transformative approach to computational design of metal complex chemistry.
• It significantly enhances the ability to discover and design organometallic compounds for various applications.
• The tool facilitates exploration beyond known chemical space for rare-earth and actinide complexes.