Pith. sign in

REVIEW 3 major objections 6 minor 2 references

Deformation of ellipsoidal droplets in homogeneous and isotropic turbulence

T0 review · 3 major / 6 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 establishes that reduced-order ellipsoidal droplet models, specifically the modified and extended Maffettone-Minale closures, reproduce the deformation statistics of sub-Kolmogorov droplets in homogeneous isotropic turbulence…

desk verdict First FRS-vs-ellipsoidal-model benchmark in HIT with a credible qualitative story, but the Ca=0.4 censoring rule needs a common survivor set before the headline claim is watertight. read the letter →

arxiv 2506.09692 v1 pith:RQKLNW3M submitted 2025-06-11 physics.flu-dyn

Fabio Guglietta , Diego Taglienti , Mauro Sbragaglia This is my paper · ORCID
classification physics.flu-dyn
keywords dropletdeformationhomogeneousisotropicturbulenceellipsoidalmodelsMaffettone-Minalemodelcapillarynumbersub-Kolmogorovskewnessfullyresolvedsimulation
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 asks whether cheap reduced-order ellipsoidal models can reproduce the statistics of sub-Kolmogorov droplet deformation in homogeneous isotropic turbulence. It tests four models against fully resolved simulations driven by the same turbulent velocity-gradient trajectories, over capillary numbers 0.025, 0.15, and 0.4. It finds that the linear theory works only at low capillary number, the standard Maffettone-Minale model underestimates extreme elongation, and the modified and extended MM models reproduce the skewed, heavy-tailed distributions of all three semiaxes, including the sign change in width skewness near breakup. If true, these models become reliable substitutes for expensive simulations in parameter-scanning and statistical studies.

What carries the argument

The central object is the morphological tensor S, a second-order tensor whose eigenvalues are the squared semiaxes of the droplet ellipsoid. Its evolution is governed by an ordinary differential equation (the Maffettone-Minale model) that couples S to the turbulent velocity-gradient tensors through two coefficients f1 and f2. The extended model (EMM) upgrades f1 and f2 to polynomials in Ca fitted to fully resolved shear-flow simulations, and this Ca-dependent closure is what lets the model capture the heavy tails and negative width skewness observed in fully resolved turbulence simulations.

What would settle it

Simulate the same droplets at Ca=0.4 in a larger numerical domain (or with a higher L/R cutoff) and recompute the skewness of W/R in both the fully resolved and EMM data; if the negative skewness disappears or the model and simulation diverge once the cutoff is changed, the reported skewness transition and model agreement would be artifacts of the exclusion threshold.

Watch

Extended reading notes

Core claim

In homogeneous isotropic turbulence with sub-Kolmogorov droplets, the linear theory derived in the Ca→0 limit fails at Ca≥0.15, and the original Maffettone-Minale model systematically underestimates large deformations and overestimates the critical capillary number. The modified MM model and, especially, the extended MM model reproduce the full probability distributions of the three normalized semiaxes L/R, B/R, and W/R across the entire range of capillary numbers studied, including the transition from positive to negative skewness in W/R that signals cigar-like shapes approaching breakup. This establishes that the MM and EMM closures are quantitatively useful reduced-order tools for droplet deformation statistics in HIT up to Ca=0.4.

Load-bearing premise

The central assumption is that the EMM coefficients, fitted to steady shear-flow simulations in earlier work, transfer unchanged to the unsteady and intermittent turbulent flow studied here; if that transfer fails, the claimed agreement with fully resolved simulations is not meaningful.

Editorial extensions

If this is right

  • The linear theory and quasi-static approximation are only valid for Ca→0; using them at Ca=0.15 or higher underestimates the probability of large droplet elongation.
  • The original MM model is artificially stable, overestimating the critical capillary number and missing breakup precursors; it should not be used for rare-event statistics at high Ca.
  • Modified MM and EMM reproduce the full distribution of all three semiaxes across Ca=0.025 to 0.4, so they can generate large statistical ensembles in a fraction of the computational cost of fully resolved simulations.
  • The appearance of negative skewness in W/R is captured by MM and EMM, tying a statistical signature of near-breakup cigar-like shapes to a closed-form model.
  • EMM exhibits threshold-crossing events at a frequency comparable to fully resolved simulations, making it suitable for studying extreme deformations up to the imposed L/R=5 cutoff.

Reading between the lines

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

  • If the EMM coefficients transfer to other viscosity ratios or Reynolds numbers, the model could replace fully resolved simulations for broad breakup-rate surveys; that portability has not been demonstrated in this paper.
  • The L/R>5 cutoff applied to both FRS and EM data may alter the tail and skewness comparisons at Ca=0.4; a domain-size study could test whether the reported negative width skewness survives with a higher cutoff.
  • The low-Ca quasi-static limit enslaves droplet shape to the turbulent strain eigenframe, so the positive skewness of the intermediate strain eigenvalue directly imprints on deformation statistics; at high Ca, nonlinearity and volume conservation become the controlling mechanisms.
  • The same model hierarchy could be extended to droplets with surface viscosity or elasticity by modifying the relaxation coefficients, but those cases are not addressed here.
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

3 major / 6 minor

Summary. The manuscript compares statistics of deformation of sub-Kolmogorov droplets in homogeneous isotropic turbulence (HIT) obtained from fully resolved simulations (FRS) with predictions from four ellipsoidal models: linear theory (LT), the Maffettone–Minale (MM) model, a modified MM model with a Ca-dependent f2 (denoted MM), and an extended MM model (EMM) whose coefficients were fitted to shear-flow FRS data in a prior work. Using velocity-gradient trajectories from the TURB-Lagr database, the authors evaluate the probability distributions of the semiaxes L, B, W for Ca=0.025, 0.15, and 0.4. They report that at low Ca all models agree with FRS and the quasi-static approximation holds; at higher Ca, LT and MM miss the heavy tails of L/R, while MM and EMM reproduce the tails and the transition to negative skewness in W/R, supporting the use of these reduced-order models up to Ca=0.4.

Significance. The paper addresses a practical question—whether cheap ellipsoidal models can replace expensive fully resolved simulations for sub-Kolmogorov droplet deformation in turbulence—and provides a systematic comparison on a shared set of Lagrangian trajectories. The use of an open-source trajectory database, the side-by-side test of several closure strategies, and the explicit focus on rare-event statistics are strengths. If the main conclusions hold, the EMM and modified-MM models would be validated as efficient tools for generating droplet-statistics ensembles, which is of interest for multiphase flow modeling and data-driven studies. The main weakness is the absence of statistical uncertainty quantification and the potentially biased censoring at Ca=0.4, which need to be addressed before the quantitative claims can be fully accepted.

major comments (3)
  1. [Sec. 4 (threshold L/R>5, Fig. 7)] The comparison at Ca=0.4 is compromised by the differential application of the L/R>5 exclusion threshold: the manuscript states that FRS trajectories exceed this threshold more frequently than MM predictions and that EMM crosses at a comparable rate, but it does not report survival probabilities. Because the PDFs are conditioned on survival and the censoring event is model-dependent, the FRS and EM distributions in Fig. 7 are not directly comparable; in particular, the reported negative skewness of W/R could be biased by the removal of high-L/R events that are correlated with small W/R. The authors should report the fraction of excluded trajectories for each model, show the EM PDFs without the threshold, and provide a robustness check using a common conditioning event (e.g., trajectories that survive in all models).
  2. [Sec. 4 and Sec. 5] The central claims that MM and EMM 'reproduce the skewed and heavy-tailed distributions' and the 'transition from positive to negative skewness in W/R' are not supported by quantitative measures. The PDFs in Figs. 4–7 are presented without error bars or a goodness-of-fit statistic, and the skewness values are never computed. Given that the differences between MM and EMM at Ca=0.4 are subtle, a quantitative comparison (e.g., KL divergence, skewness coefficient, bootstrap confidence intervals) is needed to substantiate the ranking of models and the skewness transition.
  3. [Sec. 2.3 and Sec. 4] The EMM coefficients were fitted to fully resolved shear-flow simulations with the same in-house lattice Boltzmann–immersed boundary code (Ref. [41]), and the validation here uses FRS data from the same code family (Ref. [32]). Although the test is not circular because the flows differ, the agreement between EMM and FRS may partly reflect systematic errors common to the numerical method. The authors should discuss this limitation explicitly and, where possible, validate against independent simulation data or analytical small-Ca asymptotics to demonstrate that the transferability of the coefficients is not an artifact of the shared numerical platform.
minor comments (6)
  1. [Throughout] The notation for the modified MM model is visually identical to the original MM in the extracted text; the manuscript should ensure a clear typographical distinction (e.g., an overline) and define it once at first use.
  2. [Sec. 2.3] The footnote in Sec. 2.3 ('reported here 1') is awkward; the coefficients should be presented in the main text or a properly formatted table, and the fact that f2^EMM=1 for all Ca (i.e., the entire Ca-dependence is in f1) should be stated explicitly.
  3. [Fig. 7] The PDFs in Fig. 7 are truncated at L/R=5 by construction; the authors should annotate the threshold in the figure and comment on the resulting shape at the cutoff.
  4. [Sec. 4] The statistical convergence of the tails is not discussed; with 1000 trajectories of about one eddy turnover time each, the authors should at least estimate the number of effectively independent samples in the tail and mention it when interpreting the PDFs.
  5. [Sec. 2.4] The reference to 'Fig. 2' in Sec. 2.4 is a forward reference to a figure in Sec. 3; please adjust the citation or reorder the figure placement.
  6. [Sec. 2.1] Typo: 'descibed' should be 'described'.

Circularity Check

0 steps flagged · score 2.0 of 10

The paper tests previously fitted EMM coefficients against new, out-of-sample HIT fully resolved simulation data; no equation reduces to its own output, so circularity is at most a minor self-citation concern.

full rationale

The central claim is that EMM (and MM) reproduce FRS statistics across Ca values in HIT. This is not circular: the EMM coefficients were fitted to stationary shear-flow FRS data in Taglienti et al. (2023, Ref. [41]), while the present comparisons use independent, newly generated HIT FRS data driven by TURB-Lagr velocity-gradient trajectories. The target quantities (PDFs of L/R, B/R, W/R and their skewness) are not used to define any model parameter in this paper. The quasi-static relation S_QS = I R^2 + (2 f2 tau_sigma / f1) R^2 E is derived from the MM equation in the Ca->0 limit rather than fitted to the reported statistics, and the reappearance of strain-eigenvalue skewness in droplet semiaxis PDFs is a consequence of that derivation, not a renaming. The paper's self-citations are to the in-house immersed boundary-lattice Boltzmann method (Ref. [32]) and to the prior EMM calibration (Ref. [41]); these are fixed inputs validated against external shear benchmarks and do not make the HIT comparison statistically forced. The L/R>5 exclusion threshold is applied to both FRS and EM data, and the paper notes that FRS trajectories cross it more often than MM but comparably to EMM; this is a legitimate sampling-condition comparability concern, not a definitional circularity. There is therefore no exhibited reduction of a prediction to its own input, and the appropriate circularity score is low, reflecting only the self-reliance on the authors' prior codes and fits.

Assumptions & free parameters 2 free parameters · 6 assumptions · 0 invented entities

The central comparison rests on the standard MM model family, the sub-Kolmogorov linear-flow assumption, the passive-tracer approximation, and the transferability of EMM coefficients from shear to HIT. The only fitted quantities entering the analysis are the EMM coefficients from prior work and the deformation threshold.

free parameters (2)
  • EMM coefficients a_i^(1) and a_i^(2) (i=0..5) at lambda=1 = a0^1=0.457, a1^1=0.235, a2^1=-4.546, a3^1=20.536, a4^1=-34.798; a0^2=1, a_i^2=0 for i=1..5
    Fitted to fully resolved shear-flow simulations in Taglienti et al. (2023) (ref [41]); used as fixed inputs here, but they are model parameters fitted to prior data.
  • Deformation threshold L/R = 5 = 5
    Chosen by hand to mimic breakup and keep FRS droplets inside the finite domain; applied to both FRS and EM statistics in Sec. 3 and Sec. 4. Affects tail statistics, especially at Ca=0.4.
assumptions (6)
  • domain assumption MM family of ellipsoidal models (Eq. 1) with f1, f2 from linear perturbation theory
    Standard reduced model from Maffettone and Minale [34], assumed to track droplet shape via tensor S.
  • domain assumption Sub-Kolmogorov droplet experiences locally linear flow
    Sec. 3: outer flow is a linear flow constructed from turbulent gradients, valid for R/eta = 0.472 < 1.
  • domain assumption Droplet is neutrally buoyant, passive tracer, viscosity ratio lambda=1
    Sec. 3: droplets follow Lagrangian trajectories from TURB-Lagr without feedback to the flow.
  • domain assumption TURB-Lagr velocity-gradient database represents HIT at Re_lambda ~ 130
    Sec. 3: trajectories are drawn from the public database [38]; claim depends on the database's fidelity.
  • domain assumption EMM coefficients fitted in shear flow transfer to HIT
    The EMM model's polynomial coefficients were fitted to steady shear-flow FRS data in ref [41]; the paper assumes they remain valid in unsteady HIT.
  • domain assumption Immersed boundary-lattice Boltzmann FRS method accurately captures droplet deformation
    The ground-truth FRS methodology is that of ref [32] by the same group; its accuracy is taken as given here.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Deformation of ellipsoidal droplets in homogeneous and isotropic turbulence." pith.science (2026). https://pith.science/paper/RQKLNW3M

@misc{pith2026250609692,
  author       = {Pith},
  title        = {Pith review of: Deformation of ellipsoidal droplets in homogeneous and isotropic turbulence},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/RQKLNW3M}},
  note         = {Machine review of arXiv:2506.09692}
}
abstract

We study the statistics of deformation of neutrally buoyant droplets in homogeneous isotropic turbulence (HIT), wherein the characteristic droplet size $R$ is smaller than the characteristic Kolmogorov scale $\eta$ of the turbulent flow. We systematically focus on the characterization of droplet statistics obtained with various phenomenological ellipsoidal models (EMs) -- assuming that the droplet preserves the ellipsoidal shape at all times - with the droplet moving as a passive tracer in the turbulent flow. The predictions of the EMs are compared with ground-truth data obtained with three-dimensional fully resolved simulations (FRSs) without any ad-hoc assumption on the droplet shape. Our work helps in elucidating the applicability of the EMs in describing droplet deformation in HIT at changing the capillary number $\text{Ca}=\tau_{\sigma}/\tau_{\eta}$, weighting the relative importance of the droplet characteristic time $\tau_{\sigma}$ with respect to the turbulent flow characteristic time $\tau_{\eta}$.

Figures

Figures reproduced from arXiv: 2506.09692 by the authors.

Figure 1
Figure 1. Panel (a): fully resolved simulations (FRSs) of sub-Kolmogorov droplet dynamics along a homogeneous [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. PDFs of the eigenvalues Λ1,2,3 of the rate of strain tensor E of the HIT turbulent trajectories used for the analysis of droplet deformation. Eigenvalues are normalized with the norm ||E|| of the rate-of-strain tensor E. matches a linear flow constructed from the turbulent gradients [17, 29]. The analysis for droplet deformation is conducted for three different capillary numbers: Ca = 0.025, Ca = 0.15, and Ca = 0.4.… view at source ↗
Figure 3
Figure 3. Comparison between fully resolved simulations (FRSs) and different ellipsoidal models (EMs) for two values [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figures from the paper (4 more)
Figure 4
Figure 4. Figure 4: Probability distribution functions (PDFs) of normalized droplet semiaxes [PITH_FULL_IMAGE:figures/full_fig_p007_4.png]
Figure 5
Figure 5. Figure 5: Probability distribution functions (PDFs) of normalized droplet semiaxes [PITH_FULL_IMAGE:figures/full_fig_p007_5.png]
Figure 6
Figure 6. Figure 6: Probability distribution functions (PDFs) of normalized droplet semiaxes [PITH_FULL_IMAGE:figures/full_fig_p008_6.png]
Figure 7
Figure 7. Figure 7: Same as Fig. 6, for Ca [PITH_FULL_IMAGE:figures/full_fig_p008_7.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

2 extracted references · 1 canonical work pages

  1. [1]

    Particle-turbulence interactions in atmospheric clouds

    1R. A. Shaw, “Particle-turbulence interactions in atmospheric clouds”, Annual Review of Fluid Mechanics35, 183–227 (2003). 2M Carl, T. Behrendt, C Fleing, M Frodermann, J Heinze, C Hassa, U Meier, D Wolff-Gaßmann, S Hohmann, and N Zarzalis, “Experimental and numerical investigation of a planar combustor sector at realistic operating conditions”, J. Eng. G...

    arXiv 2003

  2. [2]

    Suspensions of viscoelastic capsules: effect of membrane viscosity on transient dynamics

    Mathematical, Physical and Engineering Sciences379, 20200395 (2021). 48F. Guglietta, F. Pelusi, M. Sega, O. Aouane, and J. Harting, “Suspensions of viscoelastic capsules: effect of membrane viscosity on transient dynamics”, Journal of Fluid Mechanics971, A13 (2023). 49J. Rallison, “Note on the time-dependent deformation of a viscous drop which is almost s...

    work page 2021

Pith tools

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