Machine learning plays an increasingly important role in many areas of chemistry and materials science, being used to predict materials properties, accelerate simulations, design new structures, and predict synthesis routes of new materials. Graph neural networks (GNNs) are one of the fastest growing classes of machine learning models. They are of particular relevance for chemistry and materials science, as they directly work on a graph or structural representation of molecules and materials and therefore have full access to all relevant information required to characterize materials. In this Review, we provide an overview of the basic principles of GNNs, widely used datasets, and state-of-the-art architectures, followed by a discussion of a wide range of recent applications of GNNs in chemistry and materials science, and concluding with a road-map for the further development and application of GNNs.

While materials based on organic molecules usually have either superior optoelectronic or superior chiral properties, the combination of both is scarce. Here, a crystalline chiroptical film based on porphyrin with homochiral side groups is presented. While the dissolved molecule has a planar, thus, achiral porphyrin core, upon assembly in a metal-organic framework (MOF) film, the porphyrin core is twisted and chiral. The close packing and the crystalline order of the porphyrin cores in the MOF film also results in excellent optoelectronic properties. By exciting the Soret band of porphyrin, efficient photoconduction with a high On-Off-ratio is realized. More important, handednessdependent circularly-polarized-light photoconduction with a dissymmetry factor g of 4.3 × 10 À 4 is obtained. We foresee the combination of such assembly-induced chirality with the rich porphyrin chemistry will enable a plethora of organic materials with exceptional chiral and optoelectronic properties.

High‐entropy alloys (HEAs) or complex concentrated alloys (CCAs) offer a huge research area for new material compositions and potential applications. Since the combination of several elements sometimes leads to unexpected and unpredictable material properties. In addition to the element combinations, the optimization of the element proportions in CCAs and HEAs is also a decisive factor in tailoring desired material properties. However, it is almost impossible to achieve the composition and characterization of CCAs and HEAs with a sufficient number of compositions by conventional experiments. Therefore, an optimized high‐throughput magnetron sputtering synthesis to fabricate a new HEA gradient layer of six elements is presented. With this approach, the compositional space of the HEA system CrMoNbTaVW can be studied in different subsections to determine the influence of the individual elements and their combinations on the structure, morphology, and physical properties (hardness and resistivity). It is found that the Cr‐, Ta‐, and W‐rich phases, which have a grain size of 10–11 nm, exhibit the hardest mechanical properties, whereas V‐, Ta‐, and Cr‐rich compounds exhibit the highest electrical resistivity. The combination of high‐throughput synthesis, automated analysis tools, and automated data interpretation enables rapid and time‐efficient characterization of the novel CrMoNbTaVW gradient film.

Während Materialien auf der Basis organischer Moleküle in der Regel entweder exzellente optoelektronische oder exzellente chirale Eigenschaften aufweisen, ist die Kombination aus beiden selten. Hier wird ein kristalliner chiroptischer Film auf Basis von Porphyrin mit homochiralen Seitengruppen vorgestellt. Während das gelöste Molekül einen planaren, also achiralen Porphyrin-Kern hat, ist der Porphyrin-Kern beim Einbau in einen metallorganischen Gerüstfilm (MOF) verdreht und chiral. Die dichte Packung und die kristalline Ordnung der Porphyrinmoleküle im MOF-Film führen auch zu hervorragenden optoelektronischen Eigenschaften. Durch Anregung des Soret-Bandes von Porphyrin wird eine effiziente Fotoleitung mit einem hohen An-Aus-Verhältnis erreicht. Noch wichtiger ist, dass eine von der Händigkeit abhängige Fotoleitfähigkeit von zirkular polarisiertem Licht mit einem Dissymmetriefaktor g von 4.3 × 10 À 4 erreicht wird. Wir gehen davon aus, dass die Kombination solcher chiralen Filme mit der reichhaltigen Porphyrinchemie eine Fülle von organischen Materialien mit außergewöhnlichen chiralen und optoelektronischen Eigenschaften ermöglichen wird.

Machine learning techniques have successfully been used to extract structural information such as the crystal space group from powder X-ray diffractograms. However, training directly on simulated diffractograms from databases such...