CAREER: Machine Learned Coarse-grained Modeling for Mechanics of Thermoplastic Elastomers

职业：热塑性弹性体力学的机器学习粗粒度建模

• 批准号: 2323108
• 负责人: Ying Li
• 金额: $ 59.29万
• 依托单位: University of Wisconsin-Madison
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2022
• 资助国家: 美国
• 起止时间: 2022-10-01 至 2026-10-31
• 项目状态: 未结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2323108&HistoricalAwards=false
• 关键词: CAREER; Machine; Learned; Coarse; grained

This Faculty Early Career Development (CAREER) grant will support fundamental research to understand complex mechanical behaviors of thermoplastic elastomers (TPEs). Biodegradable TPEs are promising recyclable and sustainable polymers with minimal environmental impact. They have the potential to be used as protective coatings for cell phones, artificial muscles for soft robotics, and polymer electrolytes for batteries. However, few of these applications have been effectively realized due to the limited understanding of TPEs' synthesis-structure-property relation. This research project aims to understand and quantify the link between synthesis, microstructure, and mechanical property of TPEs, with the help of multi-scale computational modeling, data science (machine learning), and experimental validation. With tailored mechanical properties, these biodegradable and environmentally friendly polymers can be widely used to enable an array of novel structural and device applications, alleviating the plastic pollution crisis. The project includes an education and outreach plan to train diverse groups of next-generation engineers through a variety of avenues: production of educational movies for the general public and K-12 students, engineering education for K-12 students through Pre-Engineering and Explore Engineering Programs, and providing research experience for undergraduate and graduate students, especially the underrepresented groups, through internships at industries and national labs. The research objective of this project is to formulate a machine-learned coarse-grained model for TPEs with thermodynamic consistency, temperature transferability, and representability. TPEs are segmented copolymers composed of hard segments and soft segments, forming a two-phase microstructure. Thus, the machine-learned coarse-grained model can be used to understand microphase separation and its contribution to mechanical behaviors of TPEs, leading to the well-defined synthesis-structure-property relation. Specifically, this project aims to: 1) establish the machine-learned coarse-grained model for TPEs through deep neural networks and an active learning scheme; 2) integrate coarse-grained molecular simulations and constitutive modeling to achieve a meaningful structure-property relation of TPEs; 3) explore a novel class of sequence-defined TPEs with tailored microstructures and mechanical properties. The fundamental understanding of the synthesis-structure-property relationship will highlight a clear design path for experimentalists to utilize biodegradable TPEs for a broad range of applications, e.g., sound and vibration damping materials, shape-memory materials, and adaptive solar control materials. This computational framework can be readily adapted and generalized to many other polymeric materials for understanding their structure-property relations, such as fatigue-resistant hydrogels and protein-mimetic polymers, with phase-separated microstructures.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

该学院早期职业发展 (CAREER) 资助将支持基础研究，以了解热塑性弹性体 (TPE) 的复杂机械行为。可生物降解的 TPE 是一种有前途的可回收和可持续聚合物，对环境影响最小。它们有潜力用作手机的保护涂层、软机器人的人造肌肉以及电池的聚合物电解质。然而，由于对 TPE 的合成-结构-性能关系的了解有限，这些应用很少得到有效实现。该研究项目旨在借助多尺度计算建模、数据科学（机器学习）和实验验证，了解和量化 TPE 的合成、微观结构和机械性能之间的联系。凭借定制的机械性能，这些可生物降解且环保的聚合物可广泛用于实现一系列新颖的结构和设备应用，从而缓解塑料污染危机。该项目包括一项教育和推广计划，通过多种途径培训不同群体的下一代工程师：为公众和 K-12 学生制作教育电影，通过预工程和探索为 K-12 学生提供工程教育工程项目，并通过在行业和国家实验室实习为本科生和研究生，特别是代表性不足的群体提供研究经验。该项目的研究目标是建立一个具有热力学一致性、温度可传递性和可表征性的机器学习粗粒度 TPE 模型。 TPE是由硬链段和软链段组成的嵌段共聚物，形成两相微观结构。因此，机器学习的粗粒度模型可用于理解微相分离及其对 TPE 机械行为的贡献，从而产生明确的合成-结构-性能关系。具体来说，该项目的目标是：1）通过深度神经网络和主动学习方案建立TPE的机器学习粗粒度模型； 2）整合粗粒度分子模拟和本构建模，以获得有意义的TPE结构-性能关系； 3) 探索一类新型序列定义的 TPE，具有定制的微观结构和机械性能。对合成-结构-性能关系的基本理解将为实验者提供清晰的设计路径，以将可生物降解的 TPE 应用于广泛的应用，例如声音和振动阻尼材料、形状记忆材料和自适应阳光控制材料。该计算框架可以很容易地适应和推广到许多其他聚合物材料，以了解它们的结构-性能关系，例如具有相分离微结构的抗疲劳水凝胶和蛋白质模拟聚合物。该奖项反映了 NSF 的法定使命，并被视为值得通过使用基金会的智力优点和更广泛的影响审查标准进行评估来支持。

项目成果

Investigating structure and dynamics of unentangled poly(dimethyl- co -diphenyl)siloxane via molecular dynamics simulation

通过分子动力学模拟研究未缠结的聚（二甲基-共-二苯基）硅氧烷的结构和动力学

• DOI: 10.1039/d3sm00509g
• 发表时间: 2023-06
• 期刊: Soft Matter
• 影响因子: 3.4
• 作者: Xian, Weikang;He, Jinlong;Maiti, Amitesh;Saab, Andrew P.;Li, Ying
• 通讯作者: Li, Ying

Unified machine learning protocol for copolymer structure-property predictions

用于共聚物结构-性能预测的统一机器学习协议

• DOI: 10.1016/j.xpro.2022.101875
• 发表时间: 2022-12-16
• 期刊: STAR Protocols
• 作者: Tao, Lei;Arbaugh, Tom;Byrnes, John;Varshney, Vikas;Li, Ying
• 通讯作者: Li, Ying