Search for: All records

This benchmark study evaluates deep learning-based molecular generative models on various polymer datasets. Selected models were further refined with reinforcement learning to generate hypothetical heat-resistant polymers. 

Graphene-based nanostructures hold immense potential as strong and lightweight materials, however, their mechanical properties such as modulus and strength are difficult to fully exploit due to challenges in atomic-scale engineering. This study presents a database of over 2,000 pristine and defective nanoscale CNT bundles and other graphitic assemblies, inspired by microscopy, with associated stress–strain curves from reactive molecular dynamics (MD) simulations using the reactive INTERFACE force field (IFF-R). These 3D structures, containing up to 80,000 atoms, enable detailed analyses of structure-stiffness-failure relationships. By leveraging the database and physics- and chemistry-informed machine learning (ML), accurate predictions of elastic moduli and tensile strength are demonstrated at speeds 1,000 to 10,000 times faster than efficient MD simulations. Hierarchical Graph Neural Networks with Spatial Information (HS-GNNs) are introduced, which integrate chemistry knowledge. HS-GNNs as well as extreme gradient boosted trees (XGBoost) achieve forecasts of mechanical properties of arbitrary carbon nanostructures with only 3 to 6% mean relative error. The reliability equals experimental accuracy and is up to 20 times higher than other ML methods. Predictions maintain 8 to 18% accuracy for large CNT bundles, CNT junctions, and carbon fiber cross-sections outside the training distribution. The physics- and chemistry-informed HS-GNN works remarkably well for data outside the training range while XGBoost works well with limited training data inside the training range. The carbon nanostructure database is designed for integration with multimodal experimental and simulation data, scalable beyond 100 nm size, and extendable to chemically similar compounds and broader property ranges. The ML approaches have potential for applications in structural materials, nanoelectronics, and carbon-based catalysts. 

The structural integrity of MXene and MXene-based materials is important across applications from sensors to energy storage. While MXene processing has received significant attention, its structural integrity for real-world applications remains challenging due to its flake-like structure. Here the mechanical response of layered MXene-polymer nanocomposites (MPC) with high MXene concentration (>70 %) and bioinspired nacre-like brick-and-mortar architecture is investigated to offer insights for MPC design and processing. An automated finite element analysis (FEA) framework is developed to analyze MPC models with randomized geometries and multiple combinations of the parameter space. Specifically, the influence of concentration, aspect ratio (AR), flake thickness, flake distribution, and interfacial strength is investigated. The results reveal property trends such as increasing elastic modulus, strength, and toughness with increasing cohesive strength and concentration for lower AR (=40, 60) but a decreasing trend at higher AR of 75. Local structural features like flake distribution, overlapping MXene lengths, and interconnected polymers in adjacent layers was found a critical determinant of performance. For example, stronger cohesive interaction showed 6X high toughness (291 226 ) compared to weaker case (50 24 ), but the large scatter highlighted the impact of microstructural features. The results are compared and validated with theoretical, computational, and experimental work. The findings provide valuable guidance for optimizing MPC design and their processing. Finally, the automation of the framework allows the design to be extended beyond the current system and chosen material combinations.