Development of a PointNet for Detecting Morphologies of Self-Assembled Block Oligomers in Atomistic Simulations

Abstract

Molecular simulations with atomistic or coarse-grained force fields are a powerful approach for understanding and predicting the self-assembly phase behavior of complex molecules. Amphiphiles, block oligomers, and block polymers can form mesophases with different ordered morphologies describing the spatial distribution of the blocks, but entirely amorphous nature for local packing and chain conformation. Screening block oligomer chemistry and architecture through molecular simulations to find promising candidates for functional materials is aided by effective and straightforward morphology identification techniques. Capturing 3-dimensional periodic structures, such as ordered network morphologies, is hampered by the requirement that the number of molecules in the simulated system and the shape of the periodic simulation box need to be commensurate with those of the resulting network phase. Common strategies for structure identification include structure factors and order parameters, but these fail to identify imperfect structures in simulations with incorrect system sizes. Building upon pioneering work by DeFever et al. [Chem. Sci. 2019, 10, 7503–7515] who implemented a PointNet (i.e., a neural network designed for computer vision applications using point clouds) to detect local structure in simulations of single-bead particles and water molecules, we present a PointNet for detection of nonlocal ordered morphologies of complex block oligomers. Our PointNet was trained using atomic coordinates from molecular dynamics simulation trajectories and synthetic point clouds for ordered network morphologies that were absent from previous simulations. In contrast to prior work on simple molecules, we observe that large point clouds with 1000 or more points are needed for the more complex block oligomers. The trained PointNet model achieves an accuracy as high as 0.99 for globally ordered morphologies formed by linear diblock, linear triblock, and 3-arm and 4-arm star-block oligomers, and it also allows for the discovery of emerging ordered patterns from nonequilibrium systems.

Stabilizing a Double Gyroid Network Phase with 2 nm Feature Size by Blending of Lamellar and Cylindrical Forming Block Oligomers

“…A convolutional neural network for image recognition in Fourier space (FTCNN, see Methods ) designed to distinguish various ordered morphologies and trained on ordered and disordered structures of diblock oligomers and of star triblock oligomers (but not including any blends and only synthetic data for network morphologies) 54 also distinguishes among the PL structures (see Figure 4 ). It should be noted that the PL morphology with disordered perforations was not included as a class during the FTCNN training.…”

“…The FTCNN is a three-dimensional convolutional neural network (CNN) that encodes representations of the discrete volumes to extract more discriminative features of each sample. For the training, we use the same point cloud data as in our previous work of a PointNet, 54 a deep neural network for simulation morphology detection using point clouds. These point clouds contain the coordinates representing repeat units of the minority component for nine distinct morphologies: body-centered cubic micelles, double diamond, double gyroid, disordered, hexagonally packed cylinders, hexagonally perforated lamellar, lamellar, plumber’s nightmare, and single gyroid.…”

“…Batch normalization and dropout with a 0.85 keep ratio are applied before the last fully connected layer. For each morphology, 3000 voxels are obtained by preprocessing the point clouds from the MD simulation trajectories and the synthetic network structures 54 and are split into training and test data in a 4:1 ratio. The model is trained over 1000 epochs with a batch size of 128 voxels.…”

“…These two composite layers contain 16 and 64 filters, respectively. In contrast to the PointNet layers, ,, which make the prediction results invariant to the order of the input points, the convolution operation used in a CNN is index-ordered. This is because, for the Fourier transformed occupancy grids, the relative positions of the high-intensity grids are crucial to distinguish among different original structures.…”

Inverse Design of Complex Block Copolymers for Exotic Self-Assembled Structures Based on Bayesian Optimization

“…How to develop an efficient approach to realize rapid search for the target structure and to roughly determine its stable region is crucial for the inverse design of BCPs. Recently, there have been some pioneering reports about the development of approaches for “inverse design” of BCPs. − Fredrickson and co-workers have coupled bioinspired optimization algorithms (e.g., particle swarm optimization) with SCFT to explore some known structures efficiently in two-dimensional (2D) and three-dimensional (3D) spaces. , Nevertheless, the inverse design of BCPs is still in its infancy, and it needs urgently to develop more efficient methods. , …”