Predicting glass transition of amorphous polymers by application of cheminformatics and molecular dynamics simulations

Published on Mar 18, 2021in Polymer4.231
· DOI :10.1016/J.POLYMER.2021.123495
Anas karuth (NDSU: North Dakota State University), Amirhadi Alesadi5
Estimated H-index: 5
(NDSU: North Dakota State University)
+ 1 AuthorsBakhtiyor Rasulev23
Estimated H-index: 23
(NDSU: North Dakota State University)
Sources
Abstract
abstract Predicting glass-transition temperature ( T g ) of glass-forming polymers is of critical importance as it governs the thermophysical properties of polymeric materials, such as relaxation dynamics, modulus, specific heat, dielectric properties. The cheminformatics approaches based on machine learning algorithms are becoming very useful in predicting the quantitative relationships between key molecular descriptors and various physical properties of materials. In this work, we developed a modeling framework by integrating cheminformatics methods and coarse-grained molecular dynamics (CG-MD) simulations to predict T g of diverse sets of polymers. The best predictive machine learning-based QSPR model identified the most prominent molecular descriptors influencing the T g of a hundred of polymers. Informed by the model, CG-MD simulations are performed to further delineate mechanistic interpretation and systematic dependence of these influential molecular features on T g by investigating three major CG model parameters, namely the cohesive interaction, chain stiffness, and grafting density. The CG-MD simulations reveal that the higher intermolecular interaction and chain stiffness elevates the T g of CG polymers, where their relative influences are coupled with the existence of side chains grafted on the backbone. This synergistic modeling approach provides valuable insights into the roles of key molecular features influencing the T g of polymers, paving the way to establish a materials-by-design framework for polymeric materials via molecular engineering.
📖 Papers frequently viewed together
12 Citations
1 Author (Sylvie Neyertz)
37 Citations
References67
Newest
#1Yangchao Liao (NDSU: North Dakota State University)H-Index: 3
#2Zhaofan Li (NDSU: North Dakota State University)H-Index: 3
Last. Wenjie Xia (NDSU: North Dakota State University)H-Index: 20
view all 4 authors...
Abstract Understanding the complex structural behaviors of crumpled graphene at a fundamental level is of critical importance in various engineering and technological applications. Here, we present a coarse-grained molecular dynamics (CG-MD) study for investigating structural behaviors of graphene sheets having varying sizes (or masses) during the crumpling process. The simulation results reveal that larger size graphene sheets at the initial two-dimension (2D) state tends to become more self-ad...
7 CitationsSource
#1Zhaofan Li (NDSU: North Dakota State University)H-Index: 3
#2Wenjie Xia (NDSU: North Dakota State University)H-Index: 20
Abstract Owing to the impressive mechanical properties and renewability, nanocellulose has been increasingly used for the development of lightweight materials in various applications. Here, we present a coarse-grained (CG) modeling study for investigations of mechanical performance of cellulose-based bulk material consisting of disordered cellulose nanocrystals (CNCs), forming a porous network microstructure. The simulation results reveal that increasing density and cohesive interaction between ...
7 CitationsSource
#1Luis A. Miccio (Donostia International Physics Center)H-Index: 11
#2Gustavo Ariel Schwartz (Donostia International Physics Center)H-Index: 20
Abstract We used fully connected artificial neural networks (ANN) to localize and quantify, based on the monomer structure of several polymers, the specific features responsible for their observed glass transition temperatures (Tg). The use of ANNs allows us not only to successfully predict the Tg of the polymers but, even more important, to understand what parts of the monomer are mainly contributing to it. For this task, we used the weights of a trained ANN as obtained after fitting the input ...
6 CitationsSource
#1Wen-Sheng Xu (CAS: Chinese Academy of Sciences)H-Index: 13
#2Jack F. Douglas (NIST: National Institute of Standards and Technology)H-Index: 91
Last. Xiaolei Xu (CAS: Chinese Academy of Sciences)H-Index: 7
view all 4 authors...
The generalized entropy theory (GET) of glass formation provides an analytic framework for the structural relaxation time and thermodynamic properties of glass-forming polymers in terms of molecula...
11 CitationsSource
#1Song Zhang (USM: University of Southern Mississippi)H-Index: 13
#2Amirhadi Alesadi (NDSU: North Dakota State University)H-Index: 5
Last. Xiaodan Gu (USM: University of Southern Mississippi)H-Index: 29
view all 15 authors...
14 CitationsSource
#1Wen-Sheng Xu (CAS: Chinese Academy of Sciences)H-Index: 13
#2Jack F. Douglas (NIST: National Institute of Standards and Technology)H-Index: 91
Last. Xiaolei Xu (CAS: Chinese Academy of Sciences)H-Index: 7
view all 3 authors...
The key role of chain flexibility in polymer glass formation has long been appreciated, but a fundamental understanding of how chain stiffness influences polymer glass formation remains elusive. In...
10 CitationsSource
#1Luis A. Miccio (Donostia International Physics Center)H-Index: 11
#2Gustavo Ariel Schwartz (Donostia International Physics Center)H-Index: 20
Abstract In this work convolutional-fully connected neural networks were designed and trained to predict the glass transition temperature of polymers based only on their chemical structure. This approach has shown to successfully predict the Tg of unknown polymers with average relative errors as low as 6%. Several networks with different architecture or hiperparameters were successfully trained using a previously studied glass transition temperatures dataset for validation, and then the same met...
12 CitationsSource
#1Amirhadi Alesadi (NDSU: North Dakota State University)H-Index: 5
#2Wenjie Xia (NDSU: North Dakota State University)H-Index: 20
Understanding the mechanical behavior of glassy polymers at a fundamental molecular level is of critical importance in engineering and technological applications. Among various molecular parameters...
9 CitationsSource
#1Nazila Golmohammadi (Persian Gulf University)H-Index: 1
#2Mohadeseh Boland-Hemmat (Persian Gulf University)H-Index: 1
Last. Hossein Eslami (Persian Gulf University)H-Index: 19
view all 4 authors...
We have constructed efficient coarse-grained (CG) models of poly(ethylene terephthalate) (PET), using three mapping schemes, in which a repeat unit is lumped into either three or four beads. The CG potentials are parameterized to reproduce target distributions of an underlying accurate atomistic model [H. Eslami and F. Muller-Plathe, Macromolecules 42, 8241-8250 (2009)]. The CG simulations allow equilibration of long PET chains at all length scales. The CG results on the density of PET in melt a...
5 CitationsSource
#1Wen-Sheng Xu (CAS: Chinese Academy of Sciences)H-Index: 13
#2Wenjie Xia (NDSU: North Dakota State University)H-Index: 20
6 CitationsSource
Cited By2
Newest
#1Chaofu Wu (Hunan University of Humanities, Science and Technology)H-Index: 11
#2Li Kewen (Hunan University of Humanities, Science and Technology)
Last. Zhang Lei (Hunan University of Humanities, Science and Technology)
view all 4 authors...
While multiscale modeling significantly enhances the capability of molecular simulations of polymer systems, it is well realized that the systematically derived coarse-grained (CG) models generally underestimate the thermomechanical properties. In this work, a charge-based mapping scheme has been adopted to include explicit electrostatic interactions and benchmarked against two typical polymers, atactic poly(methyl methacrylate) (PMMA) and polystyrene (PS). The CG potentials are parameterized ag...
Source
#1Xiaolei Xu (CAS: Chinese Academy of Sciences)H-Index: 7
#2Jack F. Douglas (NIST: National Institute of Standards and Technology)H-Index: 91
Last. Wen-Sheng Xu (CAS: Chinese Academy of Sciences)H-Index: 13
view all 3 authors...
1 CitationsSource