Vasilios Duros+, Jonathan Grizou+, Weimin Xuan, Zied Hosni, De-Liang Long, Haralampos N. Miras and Leroy Cronin*
+ These authors contributed equally in this work
* Corresponding author
Abstract: The discovery of new gigantic molecules formed by self-assembly using single crystal X-ray crystallography is a challenging endeavor as it combines two contingent events; first is the formation of a new molecule, and second its crystallization. Herein, we constructed a workflow that can be followed manually or by a robot to probe the envelope of both events and employed it in the chemical space of a new polyoxometalate cluster, namely Na6[Mo120Ce6O366H12(H2O)78]·200H2O (1) which has a trigonal-ring type architecture (yield 4.3% based on Mo). This compound features a dodecameric ring with an inner diameter of 17 Å and an outer of 31 Å and the synthesis and crystallization was probed using an active machine-learning algorithm developed by us to explore the crystallization space and compare it to human experimenters. The algorithm covers ca. 9 times more space versus random and ca. 6 times more than humans and increases prediction accuracy to 82.4±0.7% over 77.1±0.9% from humans.
To be cited as:
- V. Duros, J. Grizou, W. Xuan, Z. Hosni, D.-L. Long, H. N. Miras, L. Cronin, Angew. Chem. Int. Ed. 2017, 56, 10815-10820
DOI: 10.1002/anie.201705721
&
- V. Duros, J. Grizou, W. Xuan, Z. Hosni, D.-L. Long, H. N. Miras, L. Cronin, Angew. Chem. 2017, 129, 10955-10960
DOI: 10.1002/ange.201705721
Respective links to VoR:
- analysis contains all the script for the analysis of both explored space and learning progress, it generates most plots used in the SI
- figures holds the script to generate the final figures in the paper
- real_experiments holds the data from the real experiments
- simulation folder contains the simulation script used to generate the results in the SI
- utils folder includes the various mathematical and procedural tools
The code is provided under GNU GPL v3 license. For more information read the LICENSE.md file.