Structure−Mechanics Relation of Natural Rubber: Insights from Molecular Dynamics Simulations

  • Qionghai Chen
  • , Zhiyu Zhang
  • , Yongdi Huang
  • , Hengheng Zhao
  • , Zhudan Chen
  • , Ke Gao
  • , Tongkui Yue
  • , Liqun Zhang
  • , Jun Liu

Research output: Contribution to journalArticlepeer-review

48 Scopus citations

Abstract

Attributed to its strain-induced crystallization (SIC), natural rubber (NR) exhibits more excellent mechanical properties compared to other elastomeric materials and has been attracting numerous scientific and technological attention. However, a systematical understanding of the structure−mechanics relation of NR is still lacking. Herein, for the first time, we employ molecular dynamics simulation to examine the effects of the key structural factors on the SIC and mechanical properties at the molecular level. We examine the effects of phospholipid and protein mass fraction (ω), the strength of hydrogen-bond interaction (εH), and the strength of non-hydrogen-bond interaction (εNH) on structural morphology, dynamic behavior, and mechanical properties. NR tends to form local clusters due to the hydrogen-bond interaction formed between phospholipids or proteins and chain ends, which is absent in the case of cis-1,4-polyisoprene (PIP). The polymer chain mobility of NR is retarded due to the formed clusters or even physical network at great εH and high ω. Interestingly, we find that the stress−strain behavior of NR is greatly manipulated by εH and ω, as evidenced by the increase of the chain orientation and the SIC, compared with the cases of PIP. This underlying mechanism results from the alignment of the molecular chains induced by the formed clusters along the deformed direction, and the clusters during the deformation become more stable, particularly at great εH. Lastly, we adopt a machine learning algorithm named extreme gradient boosting via data augmentation, finding that εH has the most significant influencing weight factor on the stress−strain behavior of NR. In general, this work demonstrates a detailed molecular-level structure−mechanics relation of NR and provides some rational guidelines for experimentally designing and synthesizing biomimetic NR.

Original languageEnglish
Pages (from-to)3575-3586
Number of pages12
JournalACS Applied Polymer Materials
Volume4
Issue number5
DOIs
StatePublished - 13 May 2022
Externally publishedYes

Keywords

  • extreme gradient boosting
  • machine learning
  • molecular dynamics simulation
  • natural rubber
  • strain-induced crystallization
  • stress−strain behavior

Fingerprint

Dive into the research topics of 'Structure−Mechanics Relation of Natural Rubber: Insights from Molecular Dynamics Simulations'. Together they form a unique fingerprint.

Cite this