A major challenge in materials design is how to efficiently search the vast chemical design space to find the materials with desired properties. One effective strategy is to develop sampling algorithms that can exploit both explicit chemical knowledge and implicit composition rules embodied in the large materials database. Here, we propose a generative machine learning model (MatGAN) based on a generative adversarial network (GAN) for efficient generation of new hypothetical inorganic materials. Trained with materials from the ICSD database, our GAN model can generate hypothetical materials not existing in the training dataset, reaching a novelty of 92.53% when generating 2 million samples. The percentage of chemically valid (charge-neutral and electronegativity-balanced) samples out of all generated ones reaches 84.5% when generated by our GAN trained with such samples screened from ICSD, even though no such chemical rules are explicitly enforced in our GAN model, indicating its capability to learn implicit chemical composition rules to form compounds. Our algorithm is expected to be used to greatly expand the range of the design space for inverse design and large-scale computational screening of inorganic materials.

The fluorescence of luminescent emitters is often quenched in the solid state, because of the typical aggregation-caused quenching (ACQ) effect, which is a thorny obstacle to high-performance organic optoelectronic materials. The exploration of solid-state enhanced long wavelength, red-emitting chromophores, especially possessing one-dimensional (1D) assembly features, is of great importance. Interestingly, an excellent solid-state enhanced red emission system (denoted as ED) based on quinolinemalononitrile has been developed via the delicate modification of the conventional ACQ dicyanomethylene-4H-pyran (DCM) derivative (denoted as BD) through crystal engineering. ED exhibits extraordinary self-assembly property in a variety of solvents, even realizing the "waving ribbons" with a length of 6 mm and a diameter of 10 μm. Crystal analysis shows that the CH···π and CH···N supramolecular interactions of ED contribute to the twisted self-assembly solid-state enhanced emission phenomenon. However, for BD, strong face-to-face stacking leads to fluorescence quenching in the solid state. Because of such easy assembly and strong solid-state emission properties, application for optical waveguides of ED is realized with a low optical loss. Stimuli-responsive behavior is also elaborated with color change between orange and red by grinding/fuming or pressing/heating.

One of the major design issues in machine learning (ML) models for materials property prediction(MPP) is how to enable the models to learn property related physicochemical features. While many composition...

crown-6)] + (KC) cations are used for cocrystallization with manganese halides, producing isostructural single crystals of organic− inorganic hybrid complexes, [K(dibenzo-18-crown-6)] 2 MnX 4 (abbreviated (KC) 2 MnX 4 ) (X = Cl, Br), which feature one-dimensional morphology and green phosphorescence with considerable photoluminescence quantum yields accompanied by excellent optical waveguide behavior with a low loss coefficient. More interestingly, (KC) 2 MnX 4 crystallizes in the monoclinic space group Cc belonging to the achiral point group m (C s ), where the non-centrosymmetric arrangement of racemic units, with right-and left-handed rotating optical axes, endows these achiral single crystals with circularly polarized luminescence, observed for the first time.