ABSTRACT
This paper uses deep learning to present a proof-of-concept for data-driven chemistry in single-molecule magnets (SMMs). Previous discussions within SMM research have proposed links between molecular structures (crystal structures) and single-molecule magnetic properties; however, these have only interpreted the results. Therefore, this study introduces a data-driven approach to predict the properties of SMM structures using deep learning. The deep-learning model learns the structural features of the SMM molecules by extracting the single-molecule magnetic properties from the 3D coordinates presented in this paper. The model accurately determined whether a molecule was a single-molecule magnet, with an accuracy rate of approximately 70% in predicting the SMM properties. The deep-learning model found SMMs from 20 000 metal complexes extracted from the Cambridge Structural Database. Using deep-learning models for predicting SMM properties and guiding the design of novel molecules is promising.
ABSTRACT
The mol-ecular structure of the title compound, [Cu(C12H13N2O3)(H2O)2]·[Cu(C12H13N2O3)(H2O)], consists of two different mol-ecules in the asymmetric unit. Both of the structures consist of a tridentate ligand synthesized from l-valine and salicyl-aldehyde, and one water mol-ecule or two water mol-ecules coordinating to CuII. They have a square-planar (mol-ecule 1) or a square-pyramidal (mol-ecule 2) coordination geometry. In the crystal, the mol-ecules form intra- and inter-molecular O-Hâ¯O hydrogen bonds involving the coordinated water mol-ecules and other sites. A Hirshfeld surface analysis indicated that the most important contributions to the packing are from Hâ¯H [52.9% (mol-ecule 1) and 51.1% (mol-ecule 2)] and Hâ¯O/ Oâ¯H [21.2% (mol-ecule 1) and 25.8% (mol-ecule 2)] contacts. In addition, an electrostatic potential map was also obtained from DFT calculations to support the discussion of the inter-molecular inter-actions.
ABSTRACT
In the title complex, [Sm(NO3)3(C30H28N2O4)], the Sm atom is surrounded by ten O atoms. The (S,S)-2,2'-{[(1,2-di-phenyl-ethane-1,2-di-yl)bis-[(aza-niumylyl-idene)methanylyl-idene]}bis-(6-meth-oxy-phenolate) ligand, obtained from o-vanillin and (1S,2S)-(-)-1,2-di-phenyl-ethyl-enedi-amine, exhibits a slightly distorted planar arrangement of the four coordinated O atoms. In the crystal, the complex shows intra-molecular N-Hâ¯O hydrogen bonds and weak inter-molecular C-Hâ¯O hydrogen bonds. The Hirshfeld surface analysis indicates that the most important contributions to the packing are from Hâ¯H (33.5%), Oâ¯H (34.1%) and Câ¯H (21.7%) contacts.
ABSTRACT
The crystal structures of two azobenzene derivative Schiff base metal complexes (new C44H40CuN6O2 of P-1 and known C44H38MnN6O7 of P21/c abbreviated as Cu and Mn, respectively) were (re-)determined experimentally using conventional X-ray analysis to obtain electron density using a PLATON program. Cu affords a four-coordinated square planar geometry, while Mn affords a hexa-coordinated distorted octahedral geometry whose apical sites are occupied by an acetate ion and water ligands, which are associated with hydrogen bonds. The π-π or CH-π and hydrogen bonding intermolecular interactions were found in both crystals, which were also analyzed using a Hirshfeld surface analysis program. To compare these results with experimental results, a density functional theory (DFT) calculation was also carried out based on the crystal structures to obtain calculated electron density using a conventional Gaussian program. These results revealed that the axial Mn-O coordination bonds of Mn were relatively weaker than the in-plane M-N or M-O coordination bonds.