Machine Learning Framework for Characterizing Processing–Structure Relationship in Block Copolymer Thin Films
Open Access
- 11 January 2026
- journal article
- research article
- Published by American Chemical Society (ACS) in Macromolecules
- Vol. 59 (2) , 714-727
Abstract
The morphology of block copolymers (BCPs) critically influences material properties and applications. This work introduces a machine learning (ML)-enabled, high-throughput framework for analyzing grazing incidence small-angle X-ray scattering (GISAXS) data and atomic force microscopy (AFM) images to characterize BCP thin film morphology. A convolutional neural network was trained to classify AFM images by surface features, achieving 97% testing accuracy. Classified images were then analyzed to extract 2D grain size measurements from the samples in a high-throughput manner. ML models were trained to predict domain orientation based on processing parameters such as solvent ratio, additive type, and additive ratio. GISAXS-based properties were predicted with strong performances (R2 > 0.75), while AFM-based property predictions were less accurate (R2 < 0.60), likely due to the localized nature of AFM measurements compared to the bulk information captured by GISAXS. Beyond model performance, interpretability was addressed using SHapley Additive exPlanations (SHAP). SHAP analysis revealed that the additive ratio had the largest impact on morphological predictions, where additive provides the BCP chains with increased volume to rearrange into thermodynamically favorable morphologies. This interpretability helps validate model predictions and offers insight into parameter importance. Altogether, the presented framework combining high-throughput characterization and interpretable ML offers an approach to exploring and optimizing BCP thin film morphology across a broad processing landscape.Keywords
All Related Versions
Funding Information
- University of Southern Mississippi
- Basic Energy Sciences (DE-SC0024432)
- Office of Integrative Activities (2229686)
- Lawrence Berkeley National Laboratory
This publication has 81 references indexed in Scilit:
- Application of Digital Methods in Polymer Science and EngineeringAdvanced Functional Materials, 2023
- Machine learning in polymer informaticsInfoMat, 2021
- Nanoscale Block Copolymer Self‐Assembly and Microscale Polymer Film Dewetting: Progress in Understanding the Role of Interfacial Energies in the Formation of Hierarchical NanostructuresAdvanced Materials Interfaces, 2019
- Machine learning in materials scienceInfoMat, 2019
- Unraveling the Main Chain and Side Chain Effects on Thin Film Morphology and Charge Transport in Quinoidal Conjugated PolymersAdvanced Functional Materials, 2018
- Light‐Modulation of the Charge Injection in a Polymer Thin‐Film Transistor by Functionalizing the Electrodes with Bistable Photochromic Self‐Assembled MonolayersAdvanced Materials, 2016
- 25th Anniversary Article: Bulk Heterojunction Solar Cells: Understanding the Mechanism of OperationAdvanced Materials, 2013
- An In Situ Grazing Incidence X‐Ray Scattering Study of Block Copolymer Thin Films During Solvent Vapor AnnealingAdvanced Materials, 2013
- Accurate Prediction of θ (Lower Critical Solution Temperature) in Polymer Solutions Based on 3D Descriptors and Artificial Neural NetworksMacromolecular Theory and Simulations, 2008
- Gas separation using polymer membranes: an overviewPolymers for Advanced Technologies, 1994