We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:


Current browse context:


Change to browse by:

References & Citations


(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Physics > Computational Physics

Title: Determination of Glass Transition Temperature of Polyimides from Atomistic Molecular Dynamics Simulations and Machine-Learning Algorithms

Abstract: Glass transition temperature ($T_{\text{g}}$) plays an important role in controlling the mechanical and thermal properties of a polymer. Polyimides are an important category of polymers with wide applications because of their superior heat resistance and mechanical strength. The capability of predicting $T_{\text{g}}$ for a polyimide $a~priori$ is therefore highly desirable in order to expedite the design and discovery of new polyimide polymers with targeted properties and applications. Here we explore three different approaches to either compute $T_{\text{g}}$ for a polyimide via all-atom molecular dynamics (MD) simulations or predict $T_{\text{g}}$ via a mathematical model generated by using machine-learning algorithms to analyze existing data collected from literature. Our simulations reveal that $T_{\text{g}}$ can be determined from examining the diffusion coefficient of simple gas molecules in a polyimide as a function of temperature and the results are comparable to those derived from data on polymer density versus temperature and actually closer to the available experimental data. Furthermore, the predictive model of $T_{\text{g}}$ derived with machine-learning algorithms can be used to estimate $T_{\text{g}}$ successfully within an uncertainty of about 20 degrees, even for polyimides yet to be synthesized experimentally.
Comments: 12 pages, 10 figures, 25-page Supporting Information
Subjects: Computational Physics (physics.comp-ph); Materials Science (cond-mat.mtrl-sci); Soft Condensed Matter (cond-mat.soft)
Journal reference: Journal of Polymer Science 58 (11), 1521-1534 (2020)
DOI: 10.1002/pol.20200050
Cite as: arXiv:2001.08889 [physics.comp-ph]
  (or arXiv:2001.08889v1 [physics.comp-ph] for this version)

Submission history

From: Shengfeng Cheng [view email]
[v1] Fri, 24 Jan 2020 05:46:31 GMT (855kb,D)

Link back to: arXiv, form interface, contact.