Formulation Design of Frontal Polymerization Systems for Epoxy Resins Based on Machine Learning

Journal of Physics Conference Series, Journal Year: 2025, Volume and Issue: 2956(1), P. 012015 - 012015

Published: March 1, 2025

Abstract Frontal polymerization (FP) is a self-propagating reaction, which propagates through material in self-sustaining manner, resembling wave front. It considered promising alternative curing strategy for epoxy resins due to its low energy requirements only the initiation process. However, achieving stable and controlling properties remain challenging. We utilize combination of machine learning experimental validation analyze frontal system predict their properties. Our findings indicate that incorporating comonomers with higher radical reactivity or adding appropriate diluents can significantly enhance reduce viscosity. When combined carbon fiber, Bisphenol A diglycidyl ether demonstrates remarkably high elastic modulus, underscoring suitability as an optimal matrix filler choice. The developed model achieves coefficient determination (R 2 ) 0.90, affirming applicability effectiveness algorithms FP resins. This approach offers robust enhancing performance

Language: Английский

---

Measurement and QSPR modeling of Flory–Huggins parameter for solvent-swollen gels, and gel catalyst informatics

Journal of the Taiwan Institute of Chemical Engineers, Journal Year: 2025, Volume and Issue: 173, P. 106159 - 106159

Published: April 27, 2025

Language: Английский

---

Machine learning in constructing structure–property relationships of polymers

Chemical Physics Reviews, Journal Year: 2025, Volume and Issue: 6(2)

Published: May 1, 2025

The properties of polymer materials are closely related to their structures. A deep understanding quantitative relationships between the structures and polymers is crucial for design preparation high-performance materials. However, these inherently complex difficult model with limited trial error experimental data. In recent years, machine learning (ML) has become an effective multidimensional relationship modeling method, playing important role in construction This review first provides overview ML workflow, a focus on feature engineering commonly used algorithms application processes. Afterward, progress was summarized evaluated from aspects mechanical properties, thermal conductivity, glass transition temperature (Tg), compatibility, dielectric refractive index polymers. Finally, prospects material research were proposed.

Language: Английский

---

Machine-Learning-Assisted Design of Mechanically Robust Room-Temperature Self-Healing Epoxy Resins

Macromolecules, Journal Year: 2025, Volume and Issue: unknown

Published: May 1, 2025

Language: Английский

---

Interpretable Machine Learning Combined TD-DFT Calculations for the Study of Colorless Transparency Polyimides

Macromolecules, Journal Year: 2025, Volume and Issue: unknown

Published: Feb. 25, 2025

Language: Английский

---

Design of Co‐Cured Multi‐Component Thermosets with Enhanced Heat Resistance, Toughness, and Processability via a Machine Learning Approach

Macromolecular Rapid Communications, Journal Year: 2024, Volume and Issue: 45(19)

Published: July 17, 2024

Abstract Designing heat‐resistant thermosets with excellent comprehensive performance has been a long‐standing challenge. Co‐curing of various high‐performance is an effective strategy, however, the traditional trial‐and‐error experiments have long research cycles for discovering new materials. Herein, two‐step machine learning (ML) assisted approach proposed to design co‐cured resins composed polyimide (PI) and silicon‐containing arylacetylene (PSA), that is, poly(silicon‐alkyne imide) (PSI). First, two ML prediction models are established evaluate processability PIs their compatibility PSA. Then, another developed predict thermal decomposition temperature flexural strength PSI resins. The optimal molecular structures compositions high‐throughput screened. screened experimentally verified exhibit enhanced heat resistance, toughness, processability. framework in this work can be generalized rational other advanced multi‐component polymeric

Language: Английский

---

Deep Learning‐Assisted Design of Novel Donor–Acceptor Combinations for Organic Photovoltaic Materials with Enhanced Efficiency

Advanced Materials, Journal Year: 2024, Volume and Issue: unknown

Published: Dec. 8, 2024

Abstract Designing donor (D) and acceptor (A) structures discovering promising D‐A combinations can effectively improve organic photovoltaic (OPV) device performance. However, to obtain excellent power conversion efficiency (PCE), the trial‐and‐error structural design in infinite chemical space is time‐consuming costly. Herein, a deep learning (DL)‐assisted framework for OPV materials proposed. To digitally represent D A structures, structure representation method, polymer fingerprints, developed, database of constructed. By applying an end‐to‐end graph neural network modeling high‐precision DL models predicting performance are established. After combining existing ≈0.6 million virtual generated. Then, these candidate predicted by well‐trained models, numbers novel with high identified. Experimental validations confirm that prediction accuracy greater than 93% one screened (i.e., D18:BTP‐S11) exhibits above 19.3% single‐junction solar cells. Finally, based on gene analysis, rules guide experimental explorations suggested. The developed DL‐assisted approach accelerate ultrahigh bring property breakthroughs devices.

Language: Английский

---

Computer-aided design of thermosetting benzoxazoles containing bis-endoalkynyl groups: Low melting points and high thermal stability

European Polymer Journal, Journal Year: 2024, Volume and Issue: 220, P. 113503 - 113503

Published: Oct. 12, 2024

Language: Английский

---