Human versus Robots in the Discovery and Crystallization of Gigantic Polyoxometalates

Abstract The discovery of new gigantic molecules formed by self‐assembly and crystal growth is challenging as it combines two contingent events; first is the formation of a new molecule, and second its crystallization. Herein, we construct a workflow that can be followed manually or by a robot to probe the envelope of both events and employ it for a new polyoxometalate cluster, Na6[Mo120Ce6O366H12(H2O)78]⋅200 H2O (1) which has a trigonal‐ring type architecture (yield 4.3 % based on Mo). Its synthesis and crystallization was probed using an active machine‐learning algorithm developed by us to explore the crystallization space, the algorithm results were compared with those obtained by human experimenters. The algorithm‐based search is able to cover ca. 9 times more crystallization space than a random search and ca. 6 times more than humans and increases the crystallization prediction accuracy to 82.4±0.7 % over 77.1±0.9 % from human experimenters.

[1]  Achim Müller,et al.  Inorganic chemistry goes protein size: a Mo368 nano-hedgehog initiating nanochemistry by symmetry breaking. , 2002, Angewandte Chemie.

[2]  Marwin H. S. Segler,et al.  Neural-Symbolic Machine Learning for Retrosynthesis and Reaction Prediction. , 2017, Chemistry.

[3]  Sarah L Price,et al.  Predicting crystal structures of organic compounds. , 2014, Chemical Society reviews.

[4]  Yutaka Yano,et al.  Photoinduced self-assembly to lanthanide-containing molybdenum-blue superclusters and molecular design , 2006 .

[5]  Taylor D. Sparks,et al.  High-Throughput Machine-Learning-Driven Synthesis of Full-Heusler Compounds , 2016 .

[6]  Nello Cristianini,et al.  An Introduction to Support Vector Machines and Other Kernel-based Learning Methods , 2000 .

[7]  K. Müller,et al.  Machine Learning Predictions of Molecular Properties: Accurate Many-Body Potentials and Nonlocality in Chemical Space , 2015, The journal of physical chemistry letters.

[8]  Jerome G. P. Wicker,et al.  Will it crystallise? Predicting crystallinity of molecular materials , 2015 .

[9]  Burr Settles,et al.  Active Learning Literature Survey , 2009 .

[10]  Leroy Cronin,et al.  "Molecular symmetry breakers" generating metal-oxide-based nanoobject fragments as synthons for complex structures: [[Mo(128)Eu(4)O(388)H(10)(H(2)O)(81)](2)](20-), a giant-cluster dimer. , 2002, Angewandte Chemie.

[11]  Marco Buongiorno Nardelli,et al.  The high-throughput highway to computational materials design. , 2013, Nature materials.

[12]  Yunshan Zhou,et al.  A Supramolecular Derivative of a Nanoporous Molybdenum Oxide Based Inorganic Keplerate with Self-Defocusing Nonlinear Optical Properties , 2010 .

[13]  Achim Müller,et al.  [Mo154(NO)14O420(OH)28(H2O)70](25 ± 5)−: ein wasserlösliches Riesenrad mit mehr als 700 Atomen und einer relativen Molekülmasse von ca. 24000 , 1995 .

[14]  Mohamed A. Korany,et al.  Experimental design and machine learning strategies for parameters screening and optimization of Hantzsch condensation reaction for the assay of sodium alendronate in oral solution , 2015 .

[15]  Leroy Cronin,et al.  Controlling the ring curvature, solution assembly, and reactivity of gigantic molybdenum blue wheels. , 2014, Journal of the American Chemical Society.

[16]  G. G. Stokes.  "J." , 1890, The New Yale Book of Quotations.

[17]  Alok Choudhary,et al.  Combinatorial screening for new materials in unconstrained composition space with machine learning , 2014 .

[18]  R. Davey,et al.  Nucleation of organic crystals--a molecular perspective. , 2012, Angewandte Chemie.

[19]  Paul Raccuglia,et al.  Machine-learning-assisted materials discovery using failed experiments , 2016, Nature.

[20]  Leroy Cronin,et al.  Discovery of gigantic molecular nanostructures using a flow reaction array as a search engine , 2014, Nature Communications.

[21]  Roger J. Davey,et al.  Keimbildung organischer Kristalle aus molekularer Sichtweise , 2013 .

[22]  Leroy Cronin,et al.  Solution-phase monitoring of the structural evolution of a Molybdenum Blue nanoring. , 2012, Journal of the American Chemical Society.

[24]  Achim Müller,et al.  Synthese und Struktur des ringförmigen, reduzierten „Metalloxids”︁ [(MoO3)176(H2O)80H32] , 1998 .

[25]  A. Müller,et al.  Influencing the size of giant rings by manipulating their curvatures: Na6[Mo120O366(H2O)48H12(Pr(H2O)5)6](approximately 200H2O) with open shell metal centers at the cluster surface. , 2000, Inorganic chemistry.

[26]  Taylor D. Sparks,et al.  Perspective: Web-based machine learning models for real-time screening of thermoelectric materials properties , 2016 .

[27]  Yutaka Yano,et al.  Coordination of {Mo142} Ring to La3+ Provides Elliptical {Mo134La10} Ring with a Variety of Coordination Modes , 2009, Materials.

[28]  Alexander J. Smola,et al.  Advances in Large Margin Classifiers , 2000 .

[29]  Achim Müller,et al.  [Mo154(NO)14O420(OH)28(H2O)70](25 ± 5)−: A Water‐Soluble Big Wheel with More than 700 Atoms and a Relative Molecular Mass of About 24000 , 1995 .

[30]  Marina Schmid,et al.  Structure And Bonding In Crystals , 2016 .

[31]  Bryce Meredig,et al.  A recommendation engine for suggesting unexpected thermoelectric chemistries , 2015, 1502.07635.

[32]  Martin Dressel,et al.  En route to coordination chemistry under confined conditions in a porous capsule: Pr3+ with different coordination shells. , 2004, Chemical communications.

[33]  Achim Müller,et al.  Formation of a Ring-Shaped Reduced "Metal Oxide" with the Simple Composition [(MoO3 )176 (H2 O)80 H32 ]. , 1998, Angewandte Chemie.