Pith. sign in

REVIEW 4 major objections 5 minor 3 references

A finite viscoelastic constitutive model for low to high strain rate response of elastomers with application of strain rate-induced glass transition

T0 review · 4 major / 5 minor · reviewed 2026-08-03 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Two viscous mechanisms unify elastomer stress-strain response from slow creep to high-rate glassy yielding in one constitutive law.

desk verdict New two-mechanism viscoelastic model covers low-to-high rate response, but the glass-transition crossover and DMA predictions rest on a single poorly-fit parameter; worth review, not yet validated. read the letter →

arxiv 2601.05405 v4 pith:6GPMIN5J submitted 2026-01-08 cond-mat.soft cond-mat.mtrl-sci

Bibekananda Datta , Sushan Nakarmi , Nitin P. Daphalapurkar This is my paper · ORCID
classification cond-mat.softcond-mat.mtrl-sci MSC 74D1074A20
keywords finiteviscoelasticityglasstransitionstrainratesensitivityelastomerspolyborosiloxaneconstitutivemodelBergstrom-BoyceRee-Eyringflowrule
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper tries to show that a single thermodynamically consistent constitutive model can describe an amorphous elastomer from quasi-static rates (10^-3 s^-1) to high strain rates (5x10^3 s^-1), including the strain-rate-induced glass transition. The model combines two viscous mechanisms running in parallel: molecular relaxation of free chains at low rates, and intermolecular rearrangement plus alignment at high rates. Using data for polyborosiloxane, the authors fit the model and then show it reproduces rate-stiffening, yielding and hardening, cyclic hysteresis, stress relaxation, and the qualitative frequency dependence of storage and loss moduli. If right, this gives engineers a parameter-based bridge across six decades of loading rate for soft elastomers.

What carries the argument

The key machinery is a generalized Maxwell-type rheological setup built on the Reese-Govindjee multiplicative-decomposition framework, with the deformation gradient split into elastic and viscous parts for each branch. Mechanism I uses a Bergstrom-Boyce power-law flow rule driven by chain reptation and stretch; mechanism II uses a Ree-Eyring sinh flow rule with an evolving yield surface for frozen-chain rearrangement. The two branches run in parallel, so their non-equilibrium stresses add to the equilibrium Arruda-Boyce network stress, and the evolution equations are integrated with an exponential-map predictor-corrector scheme.

What would settle it

Measure the small-amplitude storage modulus and loss factor of polyborosiloxane at about 1 Hz: the model predicts a rubbery-state storage modulus near three times the fitted slow non-equilibrium shear modulus (roughly 750 kPa) with a large loss factor. If measured values are far lower, the large fitted modulus is a curve-fitting artifact rather than a physical resistance, and the predicted dissipation crossover near 10 s^-1 becomes suspect.

Watch

Extended reading notes

Core claim

The central claim is that elastomer response across the rubbery-to-glassy strain-rate transition can be captured by a finite viscoelastic theory with two simultaneously active viscous mechanisms: a low-rate Bergstrom-Boyce molecular-relaxation branch and a high-rate Ree-Eyring intermolecular-rearrangement branch with an evolving yield stress. Calibrated on polyborosiloxane, the model predicts that total dissipated energy rises with strain rate while the portion dissipated by molecular relaxation falls beyond a crossover rate near 10 s^-1, signaling the onset of glass transition. The model also reproduces the qualitative features of dynamic mechanical analysis: a smooth rise in storage modulu

Load-bearing premise

The model assumes that one Neo-Hookean non-equilibrium spring in the slow relaxation branch can, on its own, reproduce the very large residual strain of polyborosiloxane upon unloading, which forces that spring to be about 500 times stiffer than the equilibrium network and is fitted from a single unloading loop with poor agreement.

Editorial extensions

If this is right

  • If the model is correct, a single constitutive law with two flow rules can replace separate rubbery and glassy sub-models in finite-element simulations of impact, damping, and soft armor.
  • The predicted dissipation crossover strain rate (about 10 s^-1 for polyborosiloxane) provides a quantitative, mechanism-based index for the strain-rate-induced glass transition.
  • The model qualitatively reproduces DMA master-curve features over a wide frequency sweep, so it can be used to infer storage and loss moduli from transient loading data.
  • Cyclic-loading predictions imply that high-rate hysteresis loops shrink and peak stress rises over the first few cycles as chains align, which matters for fatigue and energy-absorption predictions.
  • The model's stress-relaxation predictions show that relaxation time depends strongly on the prior loading rate, not just on holding strain, which is testable experimentally.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The same two-mechanism architecture could plausibly be transferred to other rate-stiffening elastomers, such as polyurea or thermoplastic polyurethane, with the crossover strain rate serving as a material-specific glass-transition descriptor.
  • Because the model omits a separate residual-strain internal variable, its calibration relies on an unusually large slow-branch modulus; adding a plastic-like back-stress variable might bring the low-rate fit quality up and simultaneously fix the overestimated rubbery-state modulus.
  • The paper's temperature-field assumptions (isothermal, fixed thermal activation energy) suggest a natural extension: coupling the Ree-Eyring branch with temperature would let the model predict frequency-temperature superposition and master curves without changing its core evolution laws.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 5 minor

Summary. The manuscript proposes a finite viscoelastic constitutive model for elastomers spanning roughly 10^-3 to 10^3 s^-1 strain rates. It uses the Reese-Govindjee multiplicative-split framework with two parallel non-equilibrium mechanisms: a Bergstrom-Boyce type flow rule for molecular relaxation and a Ree-Eyring flow rule with an evolving yield stress for intermolecular rearrangement/alignment. The model is calibrated to polyborosiloxane (PBS) data from Konale et al. (2023): equilibrium parameters from 0.001/s compression, molecular-relaxation parameters from a 0.01/s loading-unloading loop, and intermolecular parameters from 4500/s SHPB data. Section 4 presents simulations of single and cyclic loading-unloading, stress relaxation, and small-amplitude oscillation, from which the paper claims rate stiffening, a dissipation crossover near 10/s, and qualitative prediction of storage and loss moduli. The paper explicitly acknowledges that the large fitted G_neq(1) is tied to residual strain and that no residual-strain internal variable is included.

Significance. If the central claims are borne out, the model would be a useful single thermodynamically consistent framework for elastomer response from quasi-static to high strain rates, including strain-rate-induced glass transition, with a modest parameter set and a clear algorithmic implementation (Appendix A). The high-rate SHPB fit is strong (R^2=0.993), the thermodynamic derivation is standard, and the use of mechanism-specific flow rules is physically motivated. However, the current evidence is not sufficient to establish the paper's strongest assertions. The low-rate calibration is weak (R^2=0.7093), the DMA 'prediction' is partly anchored to its own input (nu_0 is computed from G' and G'' at 100 rad/s), the dissipation crossover is an untested model extrapolation, and no held-out validation is given (notably, the 8500/s SHPB curve is unused). The paper's honest admission of the G_neq(1) shortcoming in Section 4.4 underscores, rather than resolves, the identifiability problem.

major comments (4)
  1. [Section 3.5 Step 2 and Section 4.4] The low-rate mechanism is calibrated against a single loading-unloading loop at 0.01/s with R^2=0.7093. The fitted G_neq(1)=248.6 kPa is about 500 times G_eq and, as the paper admits in Section 4.4, it controls the predicted rubbery-state storage modulus and the large tan-delta. Because the model has no separate residual-strain internal variable (acknowledged in Section 5), G_neq(1) can absorb the permanent set rather than representing a physical relaxation modulus. Consequently, the DMA curves in Figure 7 and the dissipation crossover in Figure 4(b) are not independent tests of the two-mechanism hypothesis. The authors should demonstrate identifiability by fitting multiple low-rate loops/rates or by introducing a residual-strain variable, and compare predicted DMA against experimental DMA data.
  2. [Section 3.5 Step 3 and Section 4.4] The reference viscosity nu_0 is computed from measured G' and G'' at omega_0=100 rad/s via nu_0=(G'^2+G''^2)/(omega_0 G''), and Section 4.4 then presents the frequency sweep of E', E'', and tan-delta as a model prediction. This is circular: the linear-viscoelastic point at 100 rad/s is an input, not an independent validation. Please reframe Figure 7 as a consistency check or, preferably, validate against DMA data at frequencies not used in the calibration, or against the unused 8500/s SHPB curve.
  3. [Section 4.1, Figure 4(b)] The crossover strain rate near 10/s is the paper's central new claim, but it is obtained by simulating cycles with parameters calibrated only at 0.01/s and 4500/s. No experimental dissipation data in the intermediate regime (roughly 0.1-1000/s) are used. The cusp in mechanism-I dissipation is therefore a model extrapolation, not a measured phenomenon. It should be explicitly labeled as a falsifiable prediction and, if possible, tested against intermediate-rate experiments or DMA data before being presented as evidence of the glass transition.
  4. [General validation] The source data set (Konale et al., 2023) includes an 8500/s SHPB curve, but only the 4500/s curve is used for calibration. For a model claiming predictive capability up to about 5x10^3 s^-1, a held-out prediction at 8500/s is a natural and necessary validation. Please add this comparison, or another unseen strain rate, without refitting the model parameters.
minor comments (5)
  1. [Introduction] Typographical and grammatical errors: 'resolved some of the issues issues' and 'was the developed to describe' should be corrected.
  2. [Equations (23)-(24)] The parameter c in Eq. (23) is relabeled c2 in Eq. (24) without explicit comment. Please define the correspondence clearly to avoid confusion.
  3. [Table 1 and Section 3.5] Several parameters are set by hand (kappa_eq/G_eq ratio, kappa_neq ratios, Delta G, delta, volumetric viscosities). A sensitivity analysis showing how the predicted dissipation crossover and DMA response depend on these choices would strengthen the paper.
  4. [Figure 3(b)] The inset for the 0.01/s loading-unloading fit is too small to judge. A separate panel with the experimental data and model curve, plus a residual or error plot, would better convey the quality (or lack thereof) of the low-rate fit.
  5. [Data statement] The statement that code/data are available 'upon request' is not reproducible. Depositing the implementation and calibration data in a public repository would be more appropriate for a computational constitutive paper.

Circularity Check

2 steps flagged · score 5.0 of 10

DMA and dissipation 'predictions' are partly anchored to parameters fit to the same hysteresis/DMA inputs; stress-strain rate-stiffening and high-rate yield remain independent.

  1. fitted input called prediction [Section 3.5 Step 2 (Eqs. 24-25) and Section 4.4 (Fig. 7)]
    "A shortcoming of our proposed model is that it relies on the resistance to molecular relaxation, G neq(1), to capture the residual strain upon unloading. Since polyborosiloxane (PBS) showed large residual strain in the experiments performed by Konale et al. (2023), our calibrated non-equilibrium modulus for molecular relaxation, Gneq(1), is a few orders of magnitude larger than the equilibrium shear modulus, Geq. This large value of Gneq(1) is responsible for the larger rubbery-state modulus at low oscillation frequency we see in Figure 7 as well as the large loss factor, tanδ, we obtained in"

    G_neq(1) was fit in Step 2 to the 0.01 s^-1 loading-unloading loop (R^2=0.7093), explicitly to reproduce PBS's large residual strain because the model has no separate residual-strain variable (Section 5 admits this). Section 4.4 then presents the low-frequency rubbery modulus and large tanδ as a model 'prediction', but the quoted paragraph states these quantities are controlled by the same G_neq(1). Thus the low-frequency DMA response is not an independent output of the two mechanisms; it is the fitted hysteresis/residual-strain parameter re-expressed in the frequency domain. The high-frequency glassy plateau and the transition between plateaus still have independent content.

  2. fitted input called prediction [Section 3.5 Step 3 and Section 4.4 (Fig. 7)]
    "We then estimated the reference viscosity, ν0 = G′2+G′′2 / ω0G′′ = 3.53×10^3 Pa·s, where G′ = 33270.46 and G′′ = 3162.28 at ω0 = 100 rad/s from oscillatory shear data reported by the authors."

    ν0 is computed from measured G' and G'' at 100 rad/s, so this measured DMA point is an input to the model. Section 4.4 then presents storage and loss moduli over 1-10^5 Hz as a prediction ('we numerically studied the linear viscoelastic behaviors... The computed storage modulus, E′, increases...'). At the calibration frequency (100 rad/s ≈ 16 Hz) the predicted small-strain moduli are forced by the chosen ν0; the DMA curve is anchored to its own input there. The curve at other frequencies and the rubbery-to-glassy transition are still emergent, so the circularity is partial.

full rationale

The derivation is mostly self-contained: the stress-strain calibration uses three external experimental data sets from Konale et al. (2023) (0.001/s, 0.01/s, 4500/s), and the intermediate-rate stress-strain curves, cyclic stabilization, and stress relaxation are genuine extrapolations rather than fits to their target outputs. There is no load-bearing self-citation chain: the Reese-Govindjee framework, Bergstrom-Boyce flow rule, and Ree-Eyring flow rule are all external prior work, and no author-overlapping uniqueness theorem is invoked. However, two 'predictions' are partially anchored to their own inputs. Section 3.5 Step 2 fits G_neq(1)=248.6 kPa to the 0.01/s loading-unloading loop (R^2=0.7093), explicitly to absorb PBS's large residual strain; Section 4.4 then presents the low-frequency rubbery modulus and large tanδ as a prediction while admitting that these are controlled by that same fitted G_neq(1). The low-frequency DMA quantities are therefore a restatement of the fitted hysteresis parameter in the frequency domain. Independently, the reference viscosity used in the DMA simulation is computed from measured G',G'' at 100 rad/s, so the predicted DMA curve is anchored at that frequency by construction. The dissipation-crossover rate and the high-rate yield/hardening retain independent content, so the circularity is partial rather than total.

Assumptions & free parameters 16 free parameters · 9 assumptions · 0 invented entities

The model introduces no new physical entities, but it depends on 17 fitted or hand-set parameters and several domain assumptions about which flow rule applies in each regime. The most fragile inputs are nu_0 and Delta G for the high-rate mechanism, and the large G_neq(1) that absorbs residual strain. Several ratios and volumetric viscosities are chosen without direct experimental support.

free parameters (16)
  • G_eq = 497.2 Pa
    Fit to lowest-rate (0.001/s) compressive stress-strain curve; R^2=0.9832.
  • lambda_L = 2.22
    Fit together with G_eq to the same 0.001/s curve.
  • kappa_eq ratio = 50*G_eq
    Chosen by hand to enforce near-incompressibility; no bulk data used.
  • G_neq(1) = 248.6 kPa
    Fit to 0.01/s loading-unloading; R^2=0.7093. Strongly influenced by residual strain.
  • c1 = 3.39 MPa^-1.315/s
    Bergstrom-Boyce stretch pre-factor fit in Step 2.
  • c2 = -0.246
    Stretch exponent in molecular-relaxation flow rule, fit in Step 2.
  • m = 1.315
    Stress exponent in molecular-relaxation flow rule, fit in Step 2.
  • kappa_neq(1) ratio = 50*G_neq(1)
    Chosen by hand; no bulk relaxation data.
  • G_neq(2) = 9.72 MPa
    Fit to initial stress-strain response at 4500/s in Step 3.
  • tau_0^y = 11.83 MPa
    Initial yield strength fit to 4500/s SHPB data; table lists 11.183 MPa, text says 11.83 MPa.
  • h = 21.12 MPa
    Hardening modulus fit to 4500/s data.
  • Q_s = 5e4 K^-1
    Stress activation energy fit to 4500/s data.
  • nu_0 = 3.53e3 Pa*s
    Computed from G' and G'' at 100 rad/s from oscillatory data, two orders of magnitude below the high-rate regime of interest.
  • Delta G = 1000 J/mol
    Set from literature range (10^2-10^5 J/mol) because no temperature-dependent data; not measured for PBS.
  • delta = 1e-3
    Stretch regularization constant set to avoid numerical singularity, not calibrated.
  • nu_vol(1), nu_vol(2) = 1e15 Pa*s each
    Volumetric viscosities set to 'reasonable values' due to lack of bulk relaxation experiments.
assumptions (9)
  • standard math The deformation gradient admits a multiplicative split F = F^e_k F^v_k for each Maxwell branch.
    Standard finite viscoelasticity/plasticity kinematics from Lee (1969) and Reese-Govindjee (1998), used throughout Section 2.
  • standard math Helmholtz free energy is additively decomposed into one equilibrium and multiple non-equilibrium parts.
    Standard thermodynamic internal-variable ansatz, Eq. (7).
  • standard math Coleman-Noll and Coleman-Gurtin thermodynamic restrictions yield the stress and dissipation relations used.
    Section 2.2; standard continuum thermodynamics.
  • domain assumption Rubbery flow follows the Bergstrom-Boyce flow rule with stretch pre-factor, stretch exponent, and stress exponent.
    Section 3.2; adopted from prior reptation-based polymer dynamics literature, not re-derived.
  • domain assumption Glassy flow follows the Ree-Eyring flow rule with an evolving yield stress.
    Section 3.3; adopted from glassy-polymer literature (Mulliken-Boyce, Nguyen et al.), assumed valid for elastomers at high strain rate.
  • domain assumption The two viscous mechanisms are simultaneously active and their stresses add.
    Rheological representation in Figure 2 and Eq. (13); no direct microstructural evidence is provided for PBS.
  • domain assumption Deformation is isothermal, with temperature rise estimated at only about 5 degrees C.
    Section 4.1 footnote; thermal softening is neglected, which may fail for other loading paths or thicker specimens.
  • domain assumption Uniaxial stress boundary condition plus homogeneous material-point simulation is sufficient for parameter identification.
    Section 3.5 and Appendix A; no finite-element validation of inhomogeneous boundary conditions.
  • domain assumption The experimental data of Konale et al. (2023) for PBS are accurate and representative.
    All calibrations depend on this dataset; no independent measurements are used.

how reviews work

0 comments
Cite this review

Pith. "Pith review of A finite viscoelastic constitutive model for low to high strain rate response of elastomers with application of strain rate-induced glass transition." pith.science (2026). https://pith.science/paper/6GPMIN5J

@misc{pith2026260105405,
  author       = {Pith},
  title        = {Pith review of: A finite viscoelastic constitutive model for low to high strain rate response of elastomers with application of strain rate-induced glass transition},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/6GPMIN5J}},
  note         = {Machine review of arXiv:2601.05405}
}
read the original abstract

Amorphous elastomers exhibit significant rate-stiffening and unique viscous flow characteristics across a wide range of strain rates, often undergoing glass transition above a strain rate threshold. We have developed a thermodynamically-consistent and micromechanically-inspired constitutive model for soft elastomers to capture the rate-dependent stress-strain behavior and hysteresis when subjected to low to high strain rates. Our proposed constitutive model encapsulates the viscous flow of materials through molecular motion at low strain rates and intermolecular rearrangement and alignment of the molecules at high strain rates, essentially covering the glass transition. We applied our constitutive model to uniaxial compression experiments performed at low and high strain rates for polyborosiloxane (PBS) to identify the material parameters, and subsequently, performed numerical simulations of single and multi-cycle compression, stress relaxation, and small amplitude oscillatory tension-compression. Our analyses indicate that the model predicts higher total energy dissipation with increasing strain rate; however, dissipation associated with molecular relaxation decreases because, beyond a crossover strain rate, intermolecular rearrangement and alignment become dominant, which is consistent with the onset of the glass transition. For cyclic loading-unloading, we observed that dissipation over a cycle remains constant at low strain rates but decreases non-monotonically at high strain rates before becoming constant, with the peak stress over the cycle becoming higher, which can be interpreted as more loading being carried elastically by the polymer network as the intermolecular rearrangement process occurs. Additionally, our model was able to qualitatively predict the storage modulus and loss modulus in the limit of small strain over a wide range of frequency sweeps.

Figures

Figures reproduced from arXiv: 2601.05405 by the authors.

Figure 1
Figure 1. (a) Multiplicative decomposition of the deformation gradient into elastic and viscous [PITH_FULL_IMAGE:figures/full_fig_p005_1.png] view at source ↗
Figure 2
Figure 2. Rheological representation of dual-mechanism finite viscoelasticity model for quasi-static [PITH_FULL_IMAGE:figures/full_fig_p010_2.png] view at source ↗
Figure 3
Figure 3. Identification of material parameters: (a) calibration of the equilibrium shear modulus, [PITH_FULL_IMAGE:figures/full_fig_p015_3.png] view at source ↗
Figures from the paper (4 more)
Figure 4
Figure 4. Figure 4: Rate sensitivity study of the material model in uniaxial compressive loading-unloading: [PITH_FULL_IMAGE:figures/full_fig_p018_4.png]
Figure 5
Figure 5. Figure 5: Stress-strain response under cyclic compressive loading and unloading performed at (a) [PITH_FULL_IMAGE:figures/full_fig_p020_5.png]
Figure 6
Figure 6. Figure 6: Stress relaxation characteristic of the material for the applied true strain of (a) 25%, (b) [PITH_FULL_IMAGE:figures/full_fig_p022_6.png]
Figure 7
Figure 7. Figure 7: Variation of the storage modulus, E′ , loss modulus, E′′, and loss factor, tan δ, over 1 Hz to 105 Hz for small amplitude (ε0 = 0.01) oscillatory tensile-compressive loading. The computed storage modulus, E ′ , increases from one equilibrium to another equi￾librium sta…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

3 extracted references · 1 canonical work pages

  1. [1]

    −∆t ˙γv,dev k τdev neq(k) ||τ dev neq(k) || + ˙ϵv,vol k 3 1 ! trial # (be k)trial , ⇒ Ve k 2 t+∆t = exp

    At the current time,t+ ∆t, compute the kinematic quantities: Jt+∆t = det(Ft+∆t), Ct+∆t = (F⊤)t+∆t(F)t+∆t, bt+∆t = (F)t+∆t(F⊤)t+∆t, Ft+∆t =J −1/3 t+∆t Ft+∆t, Ct+∆t =J −2/3 t+∆t Ct+∆t, bt+∆t =J −2/3 t+∆t bt+∆t. (A.1) 2.Equilibrium stress:Use the total deformation tensors from Step 1 to compute the equilibrium stress,σeq = ˆσeq(bt+∆t, Jt+∆t)using Eq. (20). 3...

  2. [178]

    Dal, H., Kaliske, M., 2009

    doi:10.1007/BF01262690. Dal, H., Kaliske, M., 2009. Bergström–Boyce model for nonlinear finite rubber viscoelasticity: Theoretical aspects and algorithmic treatment for the FE method. Computational Mechanics 44, 809–823. doi:10.1007/s00466-009-0407-2. de Gennes, P.G., 1971. Reptation of a Polymer Chain in the Presence of Fixed Obstacles. The Journal of Ch...

    doi:10.1007/bf01262690 2009

  3. [1751]

    Green, M.S., Tobolsky, A.V., 1946

    doi:10.1016/0020-7683(92)90167-R. Green, M.S., Tobolsky, A.V., 1946. A new approach to the theory of relaxing poly- meric media. The Journal of Chemical Physics 14, 80–92. doi:10.1063/1.1724109. Holzapfel, G.A., Simo, J.C., 1996. A new viscoelastic constitutive model for contin- uous media at finite thermomechanical changes. International Journal of Solid...

    work page doi:10.1016/0020-7683(92)90167-r 1946

Pith tools

Reviewed August 3, 2026 · model on record in the stance chip above.