Paper Number
SM22                         My Program 

Session
Polymer Solutions, Melts and Blends

Title
Quantitative metrics to assess evidence of time-temperature superposition (tTS)

Presentation Date and Time
October 21, 2025 (Tuesday) 11:30

Track / Room
Track 3 / Coronado + DeVargas

Authors (Click on name to view author profile)

  1. Modi, Atharva S. (University of Illinois Urbana-Champaign, Mechanical Science and Engineering)
  2. Ramlawi, Nabil (University of Illinois Urbana-Champaign, Mechanical Science and Engineering)
  3. Hedegaard, Aaron T. (3M Company)
  4. Breedlove, Evan L. (3M Company)
  5. McAllister, John W. (3M Company)
  6. Lee, Hansol (3M Company)
  7. Rajabifar, Bahram (3M Company)
  8. Ewoldt, Randy H. (University Of Illinois Urbana-champaign, Mechanical Science and Engineering)

Author and Affiliation Lines (in printed abstract book)
Atharva S. Modi1, Nabil Ramlawi1, Aaron T. Hedegaard2, Evan L. Breedlove2, John W. McAllister2, Hansol Lee2, Bahram Rajabifar2 and Randy H. Ewoldt1
1Mechanical Science and Engineering, University Of Illinois Urbana-champaign, Urbana, IL 61801; 23M Company, Saint Paul, MN 55144

Speaker / Presenter 
Modi, Atharva S.

Keywords
theoretical methods; computational methods; methods; polymer melts; techniques

Text of Abstract
Time-temperature superposition (tTS) has been one of the most ubiquitous techniques in rheology since its introduction 70 years ago, with applications ranging from broadening the time range of experimental data to understanding the interplay of complex relaxation mechanisms in linear viscoelasticity. However, validation of superposition has mostly relied on visual inspection, leading to subjective, binary labels of thermorheologically “simple” or “complex”- a distinction especially contentious for pseudo-tTS. We reconsider the problem with a Bayesian perspective and propose a continuous credibility measure of the tTS model. Unlike conventional model fitting, where there are natural metrics for credibility, the tTS model assumption does not involve an analytical function or residual form. Hence, we introduce quantitative criteria that compute the credibility based on the experimental scope, shift factor variability across material functions, and curve-similarity measures, e.g. between frequency sweeps. We validate the framework on synthetic datasets spanning canonical models of thermorheological behavior and identify key metrics for each model. The quantitative nature of the criteria also enables the propagation of experimental uncertainties through the criteria values. These criteria can be calculated on any material functions of linear viscoelasticity, reflecting the function's own sensitivity to thermorheological complexity. Additionally, they can be applied to superposed curves to identify localized regions of reliable superposition. By translating subjective judgments into a reproducible numerical gauge, the proposed method enables more rigorous reporting of thermorheological behavior through a low-dimensional assessment of the superposition.