Pith. sign in

REVIEW 4 major objections 5 minor 11 references

The Stress-Force-Fabric relation across shear bands

T0 review · 4 major / 5 minor · reviewed 2026-08-07 · deepseek-v4-flash

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

Pith's one-line read This paper shows that the stress-force-fabric sum rule predicts bulk friction for an entire sheared granular assembly, while individual shear-band regions systematically overestimate it.

desk verdict New experimental observation of fabric/force orientation decoupling in shear bands, but the paper's central claim rests on comparing band-level SFF sums to a global friction coefficient without computing local stresses. read the letter →

arxiv 2506.09741 v1 pith:JGHLPPZZ submitted 2025-06-11 cond-mat.soft

Carmen Lee , Émilien Azéma , Karen Daniels This is my paper
classification cond-mat.soft
keywords granularmaterialsstress-force-fabricrelationshearbandfabricanisotropybulkfrictioncoefficientphotoelasticdisksannularcellforcetransmission
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

The paper aims to establish that the stress-force-fabric (SFF) relation, $\mu = (a+a_n+a_t)/2$, remains a valid predictor of bulk friction in a granular material with strong spatial gradients, provided it is applied over the whole assembly. Using a photoelastic annular shear cell, it measures contact orientations and forces in three radial bands and compares the SFF prediction with the friction coefficient from the stress tensor. The sum rule matches the measured bulk friction for the entire annulus, but each individual band — inside the shear band, the middle region, and outside it — overestimates friction when evaluated locally. The paper concludes that fabric connectivity varies sharply across a shear band without contributing to the direct loading of the material, so global anisotropy, not local band structure, controls bulk strength.

What carries the argument

The central machinery is the second-order Fourier description of angular distributions — contact density $E^c(\theta) = \frac{1}{2\pi}[1 + a\cos 2(\theta-\theta_a)]$, average normal force $\langle f_n\rangle(\theta) = f_0[1 + a_n\cos 2(\theta-\theta_a)]$, and average tangential force $\langle f_t\rangle(\theta) = f_0 a_t\sin 2(\theta-\theta_a)$ — together with the Stress-Force-Fabric sum rule $\mu = (a+a_n+a_t)/2$ that these amplitudes feed into. The anisotropy amplitudes convert orientation statistics into a prediction for the bulk friction coefficient, and comparing the sum rule computed for the whole annulus with its value inside each radial band is the operation that exposes the decoupling of fabric orientation from load bearing.

What would settle it

Compute the fourth and higher Fourier coefficients of the contact, normal-force, and tangential-force angular distributions within the inner and outer bands; if those coefficients are comparable to the second-harmonic amplitudes, the local overestimation of the sum rule is a truncation artifact, and the claim that shear-band fabric does not contribute to loading would need to be re-evaluated.

Watch

Extended reading notes

Core claim

Within a slowly sheared annulus of photoelastic disks, the contact fabric is strongly band-dependent: the inner band is dominated by azimuthal contacts, the middle band is nearly isotropic, and the outer band retains a radial fabric that appears frozen after early shear-induced particle migration. The average force distributions, by contrast, remain aligned near the shear direction in all bands, showing that the most numerous contacts are not the load-bearing ones. Computing the SFF sum rule from the anisotropy amplitudes $a$, $a_n$, and $a_t$ reproduces the measured bulk friction coefficient over the full strain window when the entire annulus is used, while every single radial band overestimates it because strong local fabric anisotropy inflates $(a+a_n+a_t)/2$. The paper argues that fabric connectivity can change sharply across a shear band without changing the direct loading, and that bulk properties like friction should be predicted from whole-assembly anisotropy.

Load-bearing premise

The load-bearing assumption is that each angular distribution of contacts, normal forces, and tangential forces is fully captured by one cosine or sine of twice the angle, with no significant finer angular structure; if sharper orientational structure exists inside the shear band, the reported local overestimates could be fitting artifacts rather than evidence of decoupling.

Editorial extensions

If this is right

  • Local contact-fabric anisotropy inside a shear band should not be used as a direct measure of local strength, because the SFF sum rule overestimates friction there.
  • Bulk friction in an annular shear geometry can be predicted from global Fourier anisotropies even when strong spatial gradients are present.
  • The nearly uniform force direction across all bands indicates that the stress-bearing structure is more homogeneous than the contact network, so connectivity statistics alone can mislead.
  • The frozen radial fabric in the outer band acts as a structural record of the early transient migration, while the current load is carried by a different set of contacts.
  • The SFF framework, previously verified across loading histories and geometries, extends to systems with localization as long as the averaging is over the full assembly.

Reading between the lines

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

  • If the single-harmonic truncation is checked and found insufficient inside the strongly anisotropic bands, a natural extension is to add fourth-order Fourier modes to the local sum rule; this would test whether the band-level overestimation is a fitting artifact rather than a physical decoupling.
  • The frozen outer-band fabric suggests fabric orientation could act as a strain-history memory even while carrying little current load; reloading the packing along a different direction would reveal whether this passive fabric becomes load bearing.
  • A mesoscale averaging window smaller than the band width may be why local predictions fail; varying the averaging radius across the shear band could identify the minimum scale at which the sum rule matches the measured friction.
  • A three-dimensional shear-band analogue using contact-force anisotropy data would show whether the decoupling between fabric connectivity and loading is special to two-dimensional photoelastic disks or generic to granular localization.
Share X Bluesky LinkedIn Reddit HN

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 paper studies the Stress-Force-Fabric (SFF) relation in a sheared annular cell of photoelastic disks, splitting the assembly into three radial bands (inner, middle, outer) to examine the effect of shear-band localization. The authors measure angular distributions of contact density, average normal force, and average tangential force, fit them to second-order Fourier harmonics (Eq. 3), and compare the SFF sum rule (a+a_n+a_t)/2 (Eq. 4) with the bulk friction coefficient μ computed from the microstructural stress tensor (Eqs. 1–2). They find excellent agreement for the entire annulus, while each radial band's SFF prediction overestimates the bulk μ. They interpret this as evidence that fabric connectivity varies strongly across the shear band but does not contribute to the direct loading of the material, and they claim to disentangle packing-fraction gradients from boundary constraints.

Significance. If substantiated, the result would provide valuable experimental evidence on the limits of the SFF framework in inhomogeneous flows and on the role of fabric versus force anisotropy in shear-band localization. The paper benefits from high-quality photoelastic force measurements, a direct experimental check of the SFF sum rule for the global assembly, and a clear visualization of strong band-to-band fabric variations. However, the central claim about fabric not contributing to local loading is not yet supported by the analysis as presented, because the band-level predictions are never compared with band-level stresses, and the underlying assumptions of the Fourier representation are not validated in the strongly anisotropic bands.

major comments (4)
  1. [Sec. 5, Fig. 3] The band-level SFF predictions (a+a_n+a_t)/2 are compared with a single bulk μ computed from the stress tensor of the entire annulus (Eqs. 1–2). No stress tensor is computed for the inner, middle, or outer band individually. To support the claim that fabric connectivity does not contribute to direct loading within a band, the authors must compute the local stress tensor (summing Eq. 1 over contacts within each band and dividing by that band's area) and the corresponding local μ. Without this, the observed 'overestimation' may simply reflect that the local μ in a band is higher than the global μ, a point the authors themselves acknowledge by citing radial variations of μ ([11]). The central conclusion therefore rests on an invalid global-vs-local comparison.
  2. [Sec. 3, Eq. 3 and Sec. 4, Fig. 2] The Fourier representation in Eq. 3 assumes a single principal angle θ_a common to E_c(θ), ⟨f_n⟩(θ), and ⟨f_t⟩(θ). The data in Fig. 2 show that the contact orientation varies strongly with band (inner contacts primarily azimuthal, outer contacts primarily radial), while the force orientation remains near π/4. For the inner and outer bands, the principal axes of the contact and force distributions are therefore approximately orthogonal, so the common-axis assumption of Eq. 3 is violated. In this regime the scalar sum rule Eq. 4, derived under this assumption, is not the appropriate predictor; the stress tensor involves the orientation-averaged product of fabric and force anisotropies, and misaligned or anti-correlated distributions cannot be captured by a single scalar. The reported band-level overestimates are thus ambiguous: they could reflect a real physical decoupling or an artifact of applying a small-anisotropy, single-axis sum rule to strongly anisotropic, out-of-phase subsystems.
  3. [Sec. 4 and Sec. 5] No uncertainties are reported for the fitted Fourier amplitudes a, a_n, and a_t, and the validity of the second-order truncation in Eq. 3 is not checked in the strongly anisotropic inner and outer bands. If higher harmonics are significant in these bands, the amplitudes extracted from the fits would be unreliable, and the overestimation reported in Fig. 3 could be a truncation artifact rather than a physical result. The authors should report fit residuals, confidence intervals on the amplitudes, and a test of whether including the fourth harmonic changes the sum-rule prediction.
  4. [Abstract and Sec. 4] The abstract and conclusion claim that the paper disentangles the effects of packing-fraction gradients and boundary constraints on fabric orientation. However, the manuscript contains no quantitative analysis separating these influences: the discussion in Sec. 4 (migration to the outer boundary, freezing of radial fabric) is speculative and not supported by controlled variations of packing fraction or boundary conditions. This claim should be either removed or substantiated with additional measurements or simulations.
minor comments (5)
  1. [Sec. 5, first sentence] The sentence is broken: 'Finally, to compare difference between the radial positions and the bulk properties of the annular We begin cell, we plot...' should be rewritten, e.g., 'Finally, to compare differences between the radial positions and the bulk properties of the annular cell, we plot...'.
  2. [Eq. 3 and notation] The notation in Eq. 3(c) has a missing space ('f 0at sin 2(θ−θ a)') and the sign convention for a_t is not defined; please clarify.
  3. [References] The name 'Rothenbug' in the reference to Rothenburg and Bathurst is a typo; also, references [5] and [6] are arXiv preprints and should be updated to their published versions if available.
  4. [Fig. 2 caption] The caption says 'The angular distribution correspond to the angle from the radial coordinate'; grammar should be corrected and the definition of θ (radial vs azimuthal) made more explicit.
  5. [General] The manuscript would benefit from a brief statement of the assumptions underlying the SFF sum rule (small anisotropy, common principal axes, contact-force correlations) so that the reader can evaluate the applicability to the band-resolved data.

Circularity Check

0 steps flagged · score 2.0 of 10

No significant circularity: the SFF sum rule is an external theoretical benchmark and the bulk agreement is a consistency check, not a fitted prediction; only minor self-citations are present.

full rationale

The paper's central test compares the microstructural stress tensor (Eqs. 1-2) with the SFF sum rule (Eq. 4), which is imported from the independent prior derivations of Rothenburg and Bathurst and Li and Yu. The anisotropy amplitudes a, a_n, and a_t are fitted to the measured contact and force angular distributions, but the sum rule itself is not fitted to the stress; therefore the bulk agreement is a genuine consistency check of the SFF framework rather than a tautology. The paper does not claim to predict stress from variables that are themselves defined by the stress. Refs. [5,6] are self-citations used to assert prior verification and to justify not plotting tangential forces separately, but the current paper's bulk comparison is self-contained against its own measured stress, so these self-citations are not load-bearing. The band-level overestimation is a comparison between per-band SFF sums and the global friction coefficient, not a per-band stress; this is a control-volume limitation and a possible source of the overestimation, but it is not a circular reduction because the SFF sum is not constructed from the global mu. No equation in the paper reduces to its own input by definition. The mild self-citation burden is the only reason the score is not zero.

Assumptions & free parameters 3 free parameters · 3 assumptions · 0 invented entities

The central claim rests on the SFF sum rule from prior literature and on the adequacy of the second-order Fourier expansion of the angular distributions. No new free parameters beyond the measured anisotropy amplitudes, and no invented entities, are introduced.

free parameters (3)
  • a (contact anisotropy amplitude) = not tabulated
    Fit to E_c(theta) in Eq. 3a for each radial band and the whole annulus.
  • a_n (normal force anisotropy amplitude) = not tabulated
    Fit to <f_n>(theta) in Eq. 3b.
  • a_t (tangential force anisotropy amplitude) = not tabulated
    Fit to <f_t>(theta) in Eq. 3c.
assumptions (3)
  • domain assumption The SFF sum rule mu = (a+a_n+a_t)/2 is valid for the granular system
    Taken from Rothenburg and Bathurst [2] and Li and Yu [3]; assumes the angular distributions are well described by the second-order Fourier expansion and that the anisotropy is small.
  • ad hoc to paper Second-order Fourier truncation adequately represents the angular distributions
    Eq. 3 truncates after the leading harmonic; no validation of higher harmonics is provided in the paper.
  • standard math The microstructural stress tensor (Eq. 1) correctly measures the stress in each radial bin
    Standard result from Bagi [10]; assumes the contact force description and area S are appropriate for the annular geometry.

how reviews work

0 comments
Cite this review

Pith. "Pith review of The Stress-Force-Fabric relation across shear bands." pith.science (2026). https://pith.science/paper/JGHLPPZZ

@misc{pith2026250609741,
  author       = {Pith},
  title        = {Pith review of: The Stress-Force-Fabric relation across shear bands},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/JGHLPPZZ}},
  note         = {Machine review of arXiv:2506.09741}
}
read the original abstract

The strength of granular materials is highly dependent on grain connectivity (fabric), force transmission, and frictional mobilization at the particle scale. Furthermore, these bulk properties are strongly dependent on the geometry and history of loading. It is well established that anisotropy in fabric and force transmission through a granular packing directly relates to the bulk scale strength of the packing via the Stress-Force-Fabric (SFF) relation. We have recently verified the validity of this framework for a broad variety of loading histories and geometries in experimental granular packings, using photoelastic disks to measure individual interparticle contact forces. By tracking both particle positions and interparticle contact force vectors, we mapped the anisotropy of the fabric and forces to the macroscale stress and strain and found excellent agreement between the anisotropic particle-scale measures and the macroscale responses in experiments. Here, we present an analysis of the effect of strong spatial gradients (shear bands) using the SFF framework in a sheared annular geometry, finding that there are strong variations in contact orientation depending on the location within or outside the shear band, even though the principal loading direction is uniform. This highlights that the fabric connectivity significantly changes across the shear band but does not contribute to the direct loading of the material. We disentangle the effects of packing fraction gradients and boundary constraints on the differences in fabric orientation.

Figures

Figures reproduced from arXiv: 2506.09741 by the authors.

Figure 1
Figure 1. (a) Top view of photoelastic particles viewed through a darkfield polariscope (white = higher stress) inside of an annular shear cell with inner boundary rotating at a constant strain rate ˙γ and fixed outer boundary. Each particle provides a measurement of its own contact forces, which we measure relative to the radial polar coordinate R and angular coordinate θ. Overlaid on this image are coloured bands that corre… view at source ↗
Figure 2
Figure 2. Angular distributions of the con￾tact density E c (θ) (top row) and normalized force magnitude ⟨| f |⟩/ f0 for the three radial bands and the entire annulus. The normal￾izing force scale f0 is the average normal force in the system. The angular distribu￾tion is calculated after the system comes to steady state γ > 1. The angular distribution correspond to the angle from the radial co￾ordinate, with θ = 0 and π corre… view at source ↗
Figure 3
Figure 3. Comparison of the bulk friction coefficient µ calculated from the stress tensor to the pre￾diction from the SFF relation, as a function of strain γ. Each colour represents the calculated values for the inner, middle, and outer ra￾dial bands, with the bars indicat￾ing 1 standard deviation in the fits. tributions vary radially, each radial band presents a higher anisotropy than in the bulk and leads to an elevated a a… view at source ↗

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

11 extracted references · 7 canonical work pages

  1. [11]

    Fazelpour, K.E

    F. Fazelpour, K.E. Daniels, Controlling rheology via boundary conditions in dense granular flows, Soft Matter19, 2168 (2023). 10.1039/D2SM00683A

    work page doi:10.1039/d2sm00683a 2023

  2. [1]

    Jaeger, S.R

    H.M. Jaeger, S.R. Nagel, R.P. Behringer, The physics of granular materials, Physics Today49, 32–38 (1996). 10.1063/1.881494

    work page doi:10.1063/1.881494 1996

  3. [2]

    Rothenburg, R.J

    L. Rothenburg, R.J. Bathurst, Analytical study of induced anisotropy in idealized granu- lar materials, Géotechnique39, 601 (1989). 10.1680/geot.1989.39.4.601

    doi:10.1680/geot.1989.39.4.601 1989

  4. [3]

    X. Li, H.S. Yu, On the stress–force–fabric rela- tionship for granular materials, International Jour- nal of Solids and Structures50, 1285 (2013). 10.1016/j.ijsolstr.2012.12.023

    work page doi:10.1016/j.ijsolstr.2012.12.023 2013

  5. [4]

    Schall, M

    P. Schall, M. van Hecke, Shear bands in matter with granularity, Annual Review of Fluid Mechanics 42, 67–88 (2010). 10.1146/annurev-fluid-121108- 145544

    work page doi:10.1146/annurev-fluid-121108- 2010

  6. [5]

    C.L. Lee, E. Bililign, E. Azéma, K.E. Daniels, Particle scale anisotropy controls bulk properties in sheared granular materials (2024),https:// arxiv.org/abs/2405.00653

    work page Pith review arXiv 2024

  7. [6]

    C.L. Lee, E. Bililign, E. Azéma, K.E. Daniels, Loading-dependent microscale measures control bulk properties in granular material: an experimental test of the stress-force-fabric relation (2024),https: //arxiv.org/abs/2409.08140

    arXiv 2024

  8. [7]

    Abed Zadeh, J

    A. Abed Zadeh, J. Bares, T.A. Brzinski, K.E. Daniels, J. Dijksman, N. Docquier, H.O. Everitt, J.E. Kollmer, O. Lantsoght, D. Wang et al., Enlight- ening force chains: a review of photoelasticimetry in granular matter, Granular Matter21, 83 (2019). 10.1007/s10035-019-0942-2

    doi:10.1007/s10035-019-0942-2 2019

Show all 11 references
  1. [8]

    Daniels, J.E

    K.E. Daniels, J.E. Kollmer, J.G. Puckett, Photoelas- tic force measurements in granular materials, Re- view of Scientific Instruments88, 051808 (2017). 10.1063/1.4983049

    2017 doi

  2. [9]

    C.L. Lee, L. McCabe, B. McMillan, A. Naseer, D. Xie, T. Brzinski, K.E. Daniels, T. Murthy, K. Nordstrom, Photo-Elastic Granular Solver2 (PEGS2) https://github.com/photoelasticity/pegs2 (2024)

    2024

  3. [10]

    Bagi, Microstructural Stress Tensor of Granular Assemblies With V olume Forces, Journal of Applied Mechanics66, 934 (1999)

    K. Bagi, Microstructural Stress Tensor of Granular Assemblies With V olume Forces, Journal of Applied Mechanics66, 934 (1999). 10.1115/1.2791800

    1999 doi

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.