We demonstrate the application of deep neural networks as a machine-learning tool for the analysis of a large collection of crystallographic data contained in the crystal structure repositories. Using input data in the form of multiperspective atomic fingerprints, which describe coordination topology around unique crystallographic sites, we show that the neural-network model can be trained to effectively distinguish chemical elements based on the topology of their crystallographic environment. The model also identifies structurally similar atomic sites in the entire data set of ∼50000 crystal structures, essentially uncovering trends that reflect the periodic table of elements. The trained model was used to analyze templates derived from the known crystal structures in order to predict the likelihood of forming new compounds that could be generated by placing elements into these structural templates in a combinatorial fashion. Statistical analysis of predictive performance of the neural-network model, which was applied to a test set of structures never seen by the model during training, indicates its ability to predict known elemental compositions with a high likelihood of success. In ∼30% of cases, the known compositions were found among the top 10 most likely candidates proposed by the model. These results suggest that the approach developed in this work can be used to effectively guide the synthetic efforts in the discovery of new materials, especially in the case of systems composed of three or more chemical elements.
The nitroxyl radical 1-methyl-2-azaadamantane N-oxyl (Me-AZADO) exhibits magnetic bistability arising
from a radical/dimer
interconversion. The transition from the rotationally disordered paramagnetic
plastic crystal, Me-AZADO, to the ordered diamagnetic crystalline
phase, (Me-AZADO)2, has been conclusively demonstrated
by crystal structure determination from high-resolution powder diffraction
data and by solid-state NMR spectroscopy. The phase change is characterized
by a wide thermal hysteresis with high sensitivity to even small applied
pressures. The molecular dynamics of the phase transition from the
plastic crystal to the conventional crystalline phase has been tracked
by solid-state (1H and 13C) NMR and EPR spectroscopies.
Heteroleptic complexes [Fe(bpte)(bim)]X and [Fe(bpte)(xbim)]X (bpte = S,S'-bis(2-pyridylmethyl)-1,2-thioethane, bim = 2,2'-biimidazole, xbim = 1,1'-(α,α'-o-xylyl)-2,2'-biimidazole, X = ClO, BF, OTf) were prepared by reacting the corresponding Fe(II) salts with a 1:1 mixture of the ligands. All mononuclear complexes exhibit temperature-induced spin crossover (SCO) with the onset above room temperature. The SCO is rather gradual, due to low cooperativity of interactions between the cationic complexes, as revealed by crystal structure analyses. These complexes expand the range of the recently discovered Fe(II) SCO materials with {NS} coordination environment.
Unique dependence on the nature of metal salt and reaction conditions for coordination assembly reactions of varying architecture and nuclearity have been identified in V-shaped [Co3L4] and planar disc-like [Co7L6] compounds: [CoL2(μ-L)2(μ-OH2)2(CF3CO2)2] (1) and [Co(μ-L)6(μ-OMe)6]Cl2 (2) (HL = 2-{(3-ethoxypropylimino)methyl}-6-methoxyphenol). At room temperature varying reaction conditions, cobalt-ligand ratios and use of different bases allowed unique types of coordination self-assembly. The synthetic marvel lies in the nature of aggregation with respect to the two unrelated cores in 1 and 2. Complex 1 assumes a V-shaped arrangement bound to L(-), water and a trifluoroacetate anion, while 2 grows around a central Co(II) ion surrounded by a {Co} hexagon bound to methoxide and L(-). Magnetic measurements revealed that the intermetallic interactions are antiferromagnetic in nature in the case of complex 1 and ferromagnetic in the case of 2 involving high spin cobalt(ii) ions with stabilization of the high-spin ground state in the latter case. In MeCN solutions complexes 1 and 2 showed catalytic oxidation of 3,5-di-tert-butylcatechol (3,5-DTBCH2) to 3,5-di-tert-butylbenzoquinone (3,5-DTBQ) in air. The kinetic study in MeCN revealed that with respect to the catalytic turnover number (kcat) 2 is more effective than 1 for oxidation of 3,5-DTBCH2.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.