Unraveling the morphological complexity of two-dimensional macromolecules

Summary: 2D macromolecules, such as graphene and graphene oxide, possess a rich spectrum of conformational phases. However, their morphological classification has only been discussed by visual inspection, where the physics of deformation and surface contact cannot be resolved. We employ machine learning methods to address this problem by exploring samples generated by molecular simulations. Features such as metric changes, curvature, conformational anisotropy and surface contact are extracted. Unsupervised learning classifies the morphologies into the quasi-flat, folded, crumpled phases and interphases using geometrical and topological labels or the principal features of the 2D energy map. The results are fed into subsequent supervised learning for phase characterization. The performance of data-driven models is improved notably by integrating the physics of geometrical deformation and topological contact. The classification and feature extraction characterize the microstructures of their condensed phases and the molecular processes of adsorption and transport, comprehending the processing-microstructures-performance relation in applications. The bigger picture: Resolving morphological complexity of macromolecules is the stepping stone to the design and fabrication of high-performance, multi-functional materials and to understanding the soft matter behaviors in biology and engineering. To extract the physics of lattice distortion and surface contact beyond the conformation is critical, yet challenging. Here, we show that, by labeling the simulation data using the 2D map of potential energies, the 3D geometry, and the topology of contact, morphological classification can be achieved with high accuracy. The well-trained model can be used to decipher the microstructural complexity using simulation or experimental data, which may include the geometrical representation only. This data-driven approach extracts the key geometrical and topological features of 2D macromolecules that are directly responsible for the material performance in relevant applications and can be extended to study other complex surfaces such as red blood cells and the brain.