Estimating yield-strain via deformation-recovery simulations.

Area of Science:

• Computational materials science
• Polymer physics
• Materials characterization

Background:

• Predicting polymer yield strain is crucial but challenging.
• Molecular dynamics (MD) simulations often yield noisy stress-strain data, complicating yield point identification.

Purpose of the Study:

• To develop an alternative method for identifying polymer yield strain.
• To improve the accuracy and reliability of yield strain prediction in crosslinked polymers.

Main Methods:

• Utilizing deformation-recovery simulations to generate residual strain data.
• Analyzing the transition in residual strain behavior using non-linear regression to a hyperbolic model.
• Implementing uncertainty quantification techniques to assess data informativeness.

Main Results:

• Residual strain data offers a sharper, more reliable signal for yield strain compared to traditional methods.
• The proposed hyperbolic model effectively captures the yield transition.
• Uncertainty quantification provides insights into the reliability of simulated yield predictions.

Conclusions:

• The deformation-recovery simulation method offers a robust alternative for predicting polymer yield strain.
• This approach directly identifies permanent deformation, a key indicator of yield.
• The method shows promise for accurate predictions, supported by favorable comparisons with experimental data.