Pith. sign in

REVIEW 4 major objections 4 minor 36 references

Unifying shear thinning behaviors of meso-scaled particle suspensions

T0 review · 4 major / 4 minor · reviewed 2026-08-09 · deepseek-v4-flash

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

Pith's one-line read A single stress scale, set by van der Waals adhesion between particles, collapses the shear-thinning viscosity curves of meso-scaled suspensions across particle sizes, concentrations, and steady or oscillatory flow conditions.

desk verdict The L-number collapse is a promising empirical unification, but the cross-size claim rests on an unmeasured hp ∝ a assumption that needs testing before the scaling is trusted. read the letter →

arxiv 2502.03857 v1 pith:BBBBC2FF submitted 2025-02-06 cond-mat.soft

Yuan Lin , Peiwen Lin , Yixuan Liang , Dingyi Pan This is my paper · ORCID
classification cond-mat.soft PACS 83.80.Hj83.60.Df
keywords shearthinningmeso-scaledsuspensionsvanderWaalsadhesiondimensionlessnumberparticleaggregationoscillatorysuspensionrheologyjammingmodel
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

Suspensions of meso-scaled particles (roughly 100 nanometers to 10 micrometers) show strong shear thinning even though Brownian motion is negligible. The paper attributes this to flow-induced particle aggregation caused by weak van der Waals attraction, and it proposes a dimensionless number, $L = \tau/\tau_p$, that compares the applied shear stress (or oscillatory stress amplitude) with the characteristic stress of the van der Waals interaction. With this number, viscosity curves from experiments and simulations collapse onto one master curve for different particle sizes, volume fractions, and both steady and oscillatory shear. The paper further shows that the master curve is captured by a jamming model whose contact fraction depends on $L$. If the claim holds, the rheology of meso-scaled suspensions becomes predictable from a single material-dependent stress scale.

What carries the argument

The central object is the dimensionless number $L(a)=\tau/\tau_p$, where $\tau_p \approx C A_H/(12\pi a h_p^2)$ is the characteristic stress of the van der Waals interaction between particles of radius $a$ separated by a gap $h_p$, with $C$ a scaling constant. It measures the applied shear stress (or stress amplitude) relative to the adhesive stress. The argument rests on showing that all normalized viscosity curves, for different particle sizes, concentrations, and flow types, collapse when plotted against $L$. A second piece of machinery is a jamming model in which the fraction of particles in adhesive contact, $\alpha(L)=1-e^{-\delta L^{-\kappa}}$, sets the jamming volume fraction $\phi_j$; relative viscosity then follows $\eta_r=(1-\phi/\phi_j)^{-2}$, with $\phi_j$ interpolating between an adhesive loose-packing fraction and a frictional jamming fraction.

What would settle it

Measure the equilibrium viscosity curves for suspensions of particles with independently characterized surface roughness and adhesion forces (for example, by atomic force microscopy or by varying the solvent refractive index to change the Hamaker constant), compute $\tau_p$ from those measured forces, and check whether all curves still collapse onto one master curve in $L$ without any fitted adjustment of $h_p$. If different particle sizes or roughnesses require different $h_p$ values to achieve collapse, the unification is an artifact of the post-hoc parameter choice.

Watch

Extended reading notes

Core claim

The central discovery is that the non-linear constitutive relation of meso-scaled suspensions is governed by a competition between hydrodynamic shear stress and weak van der Waals adhesion, and that this competition is quantified by $L(a)=\tau/\tau_p$. Because adhesion is a near-field surface force, shearing drives particles into clusters; the clusters grow until the applied stress breaks them, which gives the viscosity its time dependence and its shear-rate dependence. When $\tau_p$ is estimated from the Hamaker constant, the particle radius, and an effective contact separation taken to be the surface roughness ($h_p=10^{-3}a$, later adjusted to $10^{-4}a$ to align with simulation), the normalized equilibrium viscosities for suspensions with particle diameters from roughly 11 to 100 micrometers collapse onto one curve. Simulation results with different Hamaker constants merge onto the same curve, and the collapse is described by a jamming relation with a stress-dependent fraction of adhesive contacts. The paper concludes that the observed shear thinning is not a Brownian or thermal effect but a stress-controlled aggregation phenomenon.

Load-bearing premise

The load-bearing assumption is that the effective contact separation $h_p$ between adhered particles equals the surface roughness ($10^{-3}$ times the particle radius) and, when that does not match simulation, is reset to $10^{-4}$ times the radius; since $h_p$ sets $\tau_p$ through Eq. (4), a different true separation would shift the $L$-axis and break the collapse across particle sizes that is the core evidence for unification.

Editorial extensions

If this is right

  • Steady and oscillatory shear data fall on the same master curve when plotted against shear stress and stress amplitude, effectively restoring the classical steady-versus-oscillatory viscosity correspondence in terms of stress instead of shear rate.
  • Shear thinning weakens as particle size grows, because $\tau_p$ decays as $1/a$; at diameters of order 100 micrometers and above, the suspension should approach Newtonian behavior.
  • The $L$-based master curve, together with the contact-fraction jamming model, gives a predictive expression for relative viscosity as a function of volume fraction and applied stress.
  • Because $L$ only requires identifying the dominant interparticle interaction, the same unification should extend to other attractive or repulsive systems, such as clay suspensions where the interaction stress is set by a yield stress.

Reading between the lines

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

  • The paper does not test this, but tuning the Hamaker constant, for example by matching the refractive index of the solvent to the particles, should shift the viscosity curve along the $L$-axis without changing its shape.
  • The paper leaves implicit that $L$ is a stress-controlled rather than strain-controlled criterion; if true, two different flow histories delivering the same total shear stress should produce the same aggregated structure, which could be checked with step-shear or reverse-shear protocols.
  • One could connect the $L$ description to a dynamic jamming transition: the stress at which $\alpha(L)$ vanishes marks a crossover from a flocculated to a dispersed state, and this critical stress should be independent of how the stress is applied.
  • By construction, any near-field surface force (electrostatic, capillary, or depletion) could replace the van der Waals stress in $\tau_p$; the paper demonstrates the idea for van der Waals attraction and cites a clay example, but the general framework remains to be tested.
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 / 4 minor

Summary. The manuscript proposes a dimensionless number L = τ/τp, where τp is a characteristic inter-particle stress from van der Waals adhesion, to unify the shear-thinning viscosity curves of meso-scaled PMMA-in-silicone-oil suspensions across particle size, concentration, and steady or oscillatory shear. The authors support the mechanism with LD-DEM simulations that include van der Waals forces, show flow-induced aggregation at low shear rates, and describe the master curve with a Wyart–Cates-type model using δ and κ as fit parameters. The paper is clearly written and reports a substantial experimental and simulation effort.

Significance. If the proposed collapse holds, the work would provide a simple, material-dependent stress scale governing shear thinning in a technologically relevant range of particle sizes, bridging colloidal and non-Brownian regimes. The qualitative mechanism—weak vdW adhesion promoting shear-induced aggregation—is plausible and is supported by the simulation structure snapshots and by the observed temporal viscosity evolution. However, the quantitative unification depends on an unverified assumption about the scaling of the contact gap with particle radius and on a post-hoc adjustment of that gap; as a result the central claim is conditional and needs additional support before the result can be regarded as established.

major comments (4)
  1. [§II.D, Eq. (5)] Substituting h_p = 10^-3 a into Eq. (4) gives τ_p = A_H × 10^6 / (12π a^3), not × 10^8 as written. The factor 10^8 in Eq. (5) actually corresponds to h_p = 10^-4 a, which is exactly the value introduced later in §II.D as the post-hoc reset. Thus Eq. (5) already embeds the later adjustment, and the claim that the experimental master curve approaches the simulation result is not a parameter-free validation but a consequence of choosing h_p.
  2. [§II.D, Eq. (4) and Fig. 4(b)] The reported cross-size collapse is a test of the assumption h_p ∝ a (ε' = 10^-3 a) only if the contact gap is actually proportional to particle radius. The paper provides no roughness or gap measurement for the three particle batches. If the true gap is set by an absolute asperity height, τ_p would scale as a^-1 rather than a^-3 and the apparent collapse would be imposed by the rescaling rather than demonstrated by the data. The authors should provide direct measurements (e.g., AFM roughness for each batch) or at least a sensitivity analysis over plausible h_p(a) relations.
  3. [§II.D, Eq. (6) and Fig. 4(b)] The Wyart–Cates model curve uses δ = 0.2238 and κ = 0.72 that are fitted to the same master curve against which the model is displayed. This is not an independent test of the L-number framework. The authors should report the fitting procedure, parameter uncertainties, and preferably an out-of-sample comparison, for example predicting a particle concentration or size not used in the fit.
  4. [Figs. 2(c) and 4(b)] No error bars or repeat measurements are reported for the viscosity curves, yet the central claim is that data from different particle sizes, concentrations, and flow conditions collapse onto a single master curve. Without uncertainty quantification, the reader cannot judge whether the observed scatter is consistent with a genuine collapse or with a plausible but unverified rescaling.
minor comments (4)
  1. [Abstract and §I] There are typos: 'Brownain' should be 'Brownian', 'bridge' should be 'bridges', 'make' should be 'makes', and 'rheologcial' should be 'rheological'.
  2. [§II.B] The pre-shearing procedure is described as providing an identical initial state, but no reproducibility test of the initial viscosity η0 is shown; since the master curves in Fig. 2(c) are normalized by η0, the definition and reproducibility of η0 matter for the reported collapse.
  3. [§II.D] The discussion of third-phase droplets [27,33,34] correctly identifies that near-field interactions can be altered, but the point would be strengthened by showing data from those systems rather than citing them; as written the relevance to the proposed L number is only qualitative.
  4. [Fig. 4(b)] The inset with the volume-fraction fitting is very small; the symbols and model lines are difficult to distinguish. The main panel would also benefit from a legend identifying the three particle diameters clearly.

Circularity Check

2 steps flagged · score 4.0 of 10

The L-collapse is a genuine scaling test, but the reported experiment-simulation agreement and the WC-model 'prediction' are obtained by fitting hp, δ, and κ to the same data.

  1. fitted input called prediction [Section II.D, Eq. (4) and Fig. 4(b)]
    "Consequently, τp ≈ C ⟨Fvdw⟩/πa2 = C A_H 1/(12π a h_p^2). ... It is proposed that hp ≈ ϵ′ ... roughness is set to be linear with the radius of the particle (e.g., ϵ′ = 10−3a). ... The master shear thinning curve from the experiment approaches the one from simulation, if the particle gap to rescale the experimental data is set as hp = 10−4a being lower than hp = ϵ′ = 10−3a for common smooth particles."

    The claimed experiment–simulation agreement is obtained by choosing the contact gap hp = 10−4a after the fact, while the simulation uses ε′ = 10−3a. Because hp enters Eq. (4) as h_p^2, this choice freely positions the experimental master curve horizontally relative to the simulation curve; the agreement is therefore a fit of hp, not an independent validation. The reset is a common prefactor for all particle sizes, so it does not by itself force the cross-size collapse.

  2. fitted input called prediction [Section II.D, Eq. (6) and Fig. 4(b)]
    "Based on the Wyart and Cates (WC) model [23, 35] and the proposed L(a) number, we also estimate the fraction of particle in contact (particles in the organized structure) as, α(L) = 1−e^{−δL^{−κ}}, (6) where δ = 0.2238 is a correction factor that could be determined experimentally, and κ = 0.72 determines the rate of breaking of the structure due to increasing shear rate. ... the relative viscosity ... is predicted as, ηr = (1−ϕ/ϕj)−2 ... As shown in Fig. 4 (b), the master curve could be well described by the model."

    The two parameters δ and κ are fitted to the same master curve that the model is then said to describe. The statement that the master curve 'could be well described' by the model is a consistency check of the chosen functional form with parameters adjusted to that curve, not an independent prediction of the constitutive relation. The L-based collapse remains a separate, parameter-light scaling claim.

full rationale

The central L(a) rescaling is not circular at the definitional level: τp is constructed from the van der Waals expression, the Hamaker constant, and the assumed linear roughness rule hp = 10−3a, rather than fitted to the viscosity data. Plotting η/η∞ against τ/τp is a legitimate test of whether a single stress scale collapses curves for different particle sizes, provided the stated hp ∝ a assumption is accepted. The later reset to hp = 10−4a is a common horizontal prefactor for all experimental curves, so it does not manufacture the cross-size collapse. However, the paper's additional quantitative claims are partially circular: the experiment–simulation agreement is produced by choosing hp after the fact, and the WC-model 'prediction' uses δ and κ fitted to the same master curve it claims to describe. These fitted steps do not destroy the independent content of the L-collapse, but they mean the paper overstates the degree to which the comparison with simulation and the WC model validate the theory. No load-bearing self-citation chain or definitional equivalence was found; existing self-citations are to prior methods and observations, not to an imported uniqueness theorem.

Assumptions & free parameters 4 free parameters · 5 assumptions · 0 invented entities

The model combines one proposed scaling (L) with four adjustable choices: C=1, hp=10^-3 a, delta, kappa, plus a post-hoc hp=10^-4 a. The jamming fractions are taken from prior literature. This is a semi-empirical framework, not a parameter-free derivation. No new physical entities are introduced.

free parameters (4)
  • delta = 0.2238
    Correction factor in alpha(L), Eq. (6), described as determinable experimentally; fitted to the master curve.
  • kappa = 0.72
    Structure break-up exponent in alpha(L), Eq. (6), fitted to the master curve.
  • Particle gap hp = 10^-3 a (experiments), 10^-4 a (simulation match)
    Assumed roughness setting tau_p via Eq. (4); adjusted post-hoc from 10^-3 a to 10^-4 a to align experimental and simulation curves.
  • Prefactor C in tau_p = 1 (assumed)
    Set to unity in Eq. (5); scales all L values, so it shifts the master curve and model fit.
assumptions (5)
  • domain assumption Brownian motion is negligible for meso-scaled particles (Peclet number of order 10^6).
    Invoked in Section II.B to exclude thermal fluctuation as the cause of aggregation and to justify focusing on van der Waals adhesion.
  • domain assumption The dominant interparticle force for neutral PMMA particles in silicone oil is van der Waals; electrostatic and other forces are negligible.
    Stated in Sections II.B and II.D for the experimental system; supports the definition of tau_p based only on F_vdw.
  • ad hoc to paper At contact, particles are separated by surface roughness hp = epsilon' = 10^-3 a.
    Introduced in Section II.D (Eq. (4)-(5)) to convert the van der Waals force into tau_p; later changed to hp = 10^-4 a to match simulation, showing it is adjustable.
  • domain assumption Wyart-Cates jamming model with adhesive loose packing fraction phi_alp = 0.29 and frictional jamming fraction phi_mu = 0.61 describes the viscosity.
    Used in Section II.D to relate contact fraction alpha(L) to relative viscosity; values taken from refs. 23 and 35.
  • domain assumption LD-DEM with pairwise Stokes drag, lubrication forces, contact springs, and Coulomb friction captures the suspension hydrodynamics.
    Simulation method in Section II.C; details in refs. 30 and 32; this is the model used to support the aggregation mechanism.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Unifying shear thinning behaviors of meso-scaled particle suspensions." pith.science (2026). https://pith.science/paper/BBBBC2FF

@misc{pith2026250203857,
  author       = {Pith},
  title        = {Pith review of: Unifying shear thinning behaviors of meso-scaled particle suspensions},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/BBBBC2FF}},
  note         = {Machine review of arXiv:2502.03857}
}
abstract

The rheology of suspensions with meso-scaled particles [with size of $O(10^2)\ \text{nm}$ to $O(10)\ \mu\text{m}$] is intriguing since significant non-Newtonian behaviors are widely observed although the thermal fluctuation (Brownain motion) of the meso-scaled particles is negligible. Here, we show that the linear constitutive relation for such systems fails due to a flow-induced particle aggregation, which originates from the inherent inter-particle interactions, e.g., the weakly adhesive van der Waals interaction. This accounts for the temporal evolution of the rheological property in both steady and oscillatory shear flows. A dimensionless number that measures the importance of the hydrodynamic interaction in shear flow with respect to the inter-particle interaction, {is} proposed, through which the non-linear constitutive relation for suspensions with various particle sizes, particle concentrations, as well as flow conditions could be unified. This investigation bridge \mdf{the gap between micro- and macro-scaled suspension systems} and make the rheology of the meso-scaled suspensions predictable.

Figures

Figures reproduced from arXiv: 2502.03857 by the authors.

Figure 1
Figure 1. FIG. 1. Diagram of the inter-particle interaction for neutral suspending particles with the size lies in different scales. (a) The [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. FIG. 2. Temporal evolution of the (a) instant viscosity and (b) magnitude of the complex viscosity in steady and oscillatory [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. FIG. 3. Particle structure investigated through numerical simulation in (a) a start-up shear flow at ˙γ [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figures from the paper (1 more)
Figure 4
Figure 4. Figure 4: FIG. 4. (a) [PITH_FULL_IMAGE:figures/full_fig_p006_4.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

36 extracted references · 35 canonical work pages

  1. [1]

    R. G. Larson, The structure and rheology of complex fluids (Oxford University Press, New York, 1999)

    work page 1999

  2. [2]

    J. F. Brady and J. F. Morris, Journal of Fluid Mechanics 348, 103 (1997)

    work page 1997

  3. [3]

    D. R. Foss and J. F. Brady, Journal of Fluid Mechanics 407, 167 (2000)

    work page 2000

  4. [4]

    J. F. Morris, Rheologica Acta 48, 909 (2009)

    work page 2009

  5. [5]

    Sierou and J

    A. Sierou and J. F. Brady, Journal of Rheology 46, 1031 (2002)

    work page 2002

  6. [6]

    Guazzelli and O

    ´E. Guazzelli and O. Pouliquen, Journal of Fluid Mechanics 852, P1 (2018)

    work page 2018

  7. [7]

    J. J. Stickel and R. L. Powell, Annual Review of Fluid Mechanics 37, 129 (2005)

    work page 2005

  8. [8]

    Blair and C

    M. Blair and C. Ness, Journal of Fluid Mechanics 948, A48 (2022)

    work page 2022

Show all 36 references
  1. [9]

    Comtet, G

    J. Comtet, G. Chatt´ e, A. Nigues, L. Bocquet, A. Siria, and A. Colin, Nature communications 8, 15633 (2017)

    2017

  2. [10]

    Jamali and J

    S. Jamali and J. F. Brady, Physical review letters 123, 138002 (2019)

    2019

  3. [11]

    Singh and P

    A. Singh and P. R. Nott, Journal of Fluid Mechanics 490, 293 (2003)

    2003

  4. [12]

    S.-C. Dai, E. Bertevas, F. Qi, and R. I. Tanner, Journal of Rheology 57, 493 (2013)

    2013

  5. [13]

    Y. Lin, Y. Wang, Z. Weng, D. Pan, and J. Chen, Physics of Fluids 33, 011702 (2021)

    2021

  6. [14]

    J. M. Bricker and J. E. Butler, Journal of Rheology 50, 711 (2006)

    2006

  7. [15]

    Y. Lin, N. Phan-Thien, and B. C. Khoo, Journal of Rheology 57, 1325 (2013)

    2013

  8. [16]

    Martone, C

    R. Martone, C. Carotenuto, and M. Minale, Journal of Rheology 64, 1075 (2020)

    2020

  9. [17]

    Chatt´ e, J

    G. Chatt´ e, J. Comtet, A. Nigues, L. Bocquet, A. Siria, G. Ducouret, F. Lequeux, N. Lenoir, G. Ovarlez, and A. Colin, Soft matter 14, 879 (2018)

    2018

  10. [18]

    Lobry, E

    L. Lobry, E. Lemaire, F. Blanc, S. Gallier, and F. Peters, Journal of Fluid Mechanics 860, 682 (2019)

    2019

  11. [19]

    Arshad, A

    M. Arshad, A. Maali, C. Claudet, L. Lobry, F. Peters, and E. Lemaire, Soft Matter 17, 6088 (2021)

    2021

  12. [20]

    Blanc, E

    F. Blanc, E. D’Ambrosio, L. Lobry, F. m. c. Peters, and E. Lemaire, Phys. Rev. Fluids 3, 114303 (2018)

    2018

  13. [21]

    Papadopoulou, J

    A. Papadopoulou, J. J. Gillissen, H. J. Wilson, M. K. Tiwari, and S. Balabani, Journal of Non-Newtonian Fluid Mechanics 281, 104298 (2020)

    2020

  14. [22]

    Gilbert, R

    D. Gilbert, R. Valette, and E. Lemaire, Journal of Rheology 66, 161 (2022)

    2022

  15. [23]

    J. A. Richards, B. M. Guy, E. Blanco, M. Hermes, G. Poy, and W. C. K. Poon, Journal of Rheology 64, 405 (2020)

    2020

  16. [24]

    Y. Lin, Y. Wang, H. Qin, D. Pan, and J. Chen, Physics of Fluids 33, 043104 (2021)

    2021

  17. [25]

    Y. Lin, N. Phan-Thien, and B. C. Khoo, Journal of Non-Newtonian Fluid Mechanics 223, 228 (2015)

    2015

  18. [26]

    Z. Ge, R. Martone, L. Brandt, and M. Minale, Phys. Rev. Fluids 6, L101301 (2021)

    2021

  19. [27]

    Y. Lin, P. Lin, Y. Wang, J. Chen, Z. He, T. P¨ ahtz, and N. Phan-Thien, Physics of Fluids 35, 023301 (2023)

    2023

  20. [28]

    W. P. Cox and E. H. Merz, Journal of Polymer Science 28, 619 (1958)

    1958

  21. [29]

    Y. Lin, N. Phan-Thien, and B. C. Khoo, Journal of Non Newtonian Fluid Mechanics 239, 28 (2017). 8

    2017

  22. [30]

    Wang and D

    J. Wang and D. Pan, Acta Mechanica Sinica 39, 322469 (2023)

    2023

  23. [31]

    Narumi, H

    T. Narumi, H. See, Y. Honma, T. Hasegawa, T. Takahashi, and N. Phan-Thien, Journal of Rheology (1978-present) 46, 295 (2002)

    2002

  24. [32]

    Cheal and C

    O. Cheal and C. Ness, Journal of Rheology 62, 501 (2018)

    2018

  25. [33]

    Koos and N

    E. Koos and N. Willenbacher, Science 331, 897 (2011)

    2011

  26. [34]

    E. Koos, J. Johannsmeier, L. Schwebler, and N. Willenbacher, Soft Matter 8, 6620 (2012)

    2012

  27. [35]

    Wyart and M

    M. Wyart and M. E. Cates, Phys. Rev. Lett. 112, 098302 (2014)

    2014

  28. [36]

    Y. Lin, H. Qin, J. Guo, and J. Chen, Applied Clay Science 214, 106275 (2021)

    2021

Pith tools

Reviewed August 9, 2026 · model on record in the stance chip above.