arxiv: 2605.03646 · v3 · submitted 2026-05-05 · ⚛️ physics.flu-dyn · physics.ao-ph· physics.geo-ph

Recognition: 3 theorem links

· Lean Theorem

Turbophoresis of inertial particles in inhomogeneous turbulence produced by oscillating grids

A. Levy, E. Elmakies, I. Rogachevskii, N. Kleeorin, O. Shildkrot

Authors on Pith no claims yet

Pith reviewed 2026-05-08 18:46 UTC · model grok-4.3

classification ⚛️ physics.flu-dyn physics.ao-phphysics.geo-ph

keywords turbophoresisinertial particlesinhomogeneous turbulenceoscillating gridsparticle image velocimetryparticle concentrationturbulent flows

0 comments

The pith

Inertial particles concentrate in regions of weaker turbulence through turbophoresis in grid-generated flows.

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

The paper investigates the formation of large-scale particle concentrations in spatially varying turbulence, an effect driven by turbophoresis. Experiments generate inhomogeneous airflow using one or two oscillating grids and apply particle image velocimetry to capture both fluid velocities and particle positions. Normalizing the number density of inertial particles against that of noninertial particles under identical conditions isolates the inertia-driven migration. The measured distributions show clear accumulation in zones of lower turbulence intensity, consistent with the predicted direction of the turbophoretic velocity. This finding matters for understanding how solid particles or droplets organize in any turbulent flow whose intensity changes across space.

Core claim

Turbophoresis produces an effective particle velocity equal to the product of the particle Stokes time and the gradient of turbulence intensity, always directed toward the minimum of the turbulent velocity. In the oscillating-grid experiments, this mechanism creates large-scale inhomogeneities: after normalization removes other transport contributions, inertial particles accumulate in the large-scale regions where turbulence intensity is lowest.

What carries the argument

The turbophoretic velocity, defined as the product of Stokes time and the spatial gradient of turbulence intensity and directed toward lower turbulence levels.

If this is right

Particle number density can be predicted from measured turbulence-intensity gradients alone once the Stokes time is known.
The normalization procedure cleanly separates turbophoresis from other mechanisms such as preferential concentration or mean-flow advection.
The same accumulation pattern should appear in any inhomogeneous turbulent flow whose intensity varies on scales much larger than the particle response time.
Quantitative agreement between experiment and the effective-velocity formula supports using this term in models of particle transport in grid-generated or similar turbulence.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

Atmospheric or industrial flows with strong turbulence gradients may show analogous large-scale particle segregation that affects local concentrations and deposition rates.
Varying particle size or density in follow-up experiments would directly test the linear dependence on Stokes time that the velocity formula predicts.
Incorporating the turbophoretic term into Eulerian or Lagrangian simulations could improve predictions of particle dispersion without resolving every small-scale eddy.

Load-bearing premise

Normalizing inertial-particle number density by the noninertial-particle density under identical flow conditions removes every non-turbophoretic transport mechanism and leaves only inertia-driven migration.

What would settle it

A direct measurement showing that, after the same normalization, inertial particles exhibit no net accumulation or even depletion in the low-turbulence-intensity zones would falsify the claim.

Figures

Figures reproduced from arXiv: 2605.03646 by A. Levy, E. Elmakies, I. Rogachevskii, N. Kleeorin, O. Shildkrot.

**Figure 1.** Figure 1: FIG. 1. Experimental setup with a turbulence produced by one oscillating grid (left panel) and by two oscillating grids (right view at source ↗

**Figure 2.** Figure 2: FIG. 2. Spatial distributions of the mean velocity field view at source ↗

**Figure 3.** Figure 3: FIG. 3. Spatial distributions of the mean velocity shear view at source ↗

**Figure 4.** Figure 4: FIG. 4. Spatial distributions of the turbulent velocity view at source ↗

**Figure 5.** Figure 5: FIG. 5. Spatial distributions of the ratio view at source ↗

**Figure 6.** Figure 6: FIG. 6. Spatial distributions of the anisotropy of turbulent velocity field view at source ↗

**Figure 7.** Figure 7: FIG. 7. Dependencies of the horizontal component of the turbulent velocity view at source ↗

**Figure 8.** Figure 8: FIG. 8. Dependencies of the vertical component of the turbulent velocity view at source ↗

**Figure 9.** Figure 9: FIG. 9. Dependencies of the horizontal integral turbulence scale view at source ↗

**Figure 10.** Figure 10: FIG. 10. Dependencies of the vertical integral turbulence scale view at source ↗

**Figure 11.** Figure 11: FIG. 11. Spatial distributions of the Reynolds number Re= [ view at source ↗

**Figure 12.** Figure 12: FIG. 12. Spatial distributions of the normalized mean particle number density view at source ↗

**Figure 13.** Figure 13: FIG. 13. The view at source ↗

**Figure 14.** Figure 14: FIG. 14. The dependencies of the normalized mean particle number density view at source ↗

read the original abstract

Turbophoresis in inhomogeneous turbulent flows leads to the formation of large-scale nonuniform particle number density distributions of inertial particles. This effect is associated with an effective drift velocity directed toward regions of lower turbulence intensity and proportional to the particle Stokes time and the spatial gradient of the turbulence intensity. In the present study, turbophoretic transport is experimentally investigated in air flows generated by one-grid and two-grid oscillating turbulence systems. The flow velocity field and particle spatial distribution are measured using Particle Image Velocimetry. To isolate the effect of particle accumulation due to turbophoresis from that associated with mean fluid flow, the measured particle number density of inertial particles is normalized by the corresponding distribution obtained for noninertial tracer particles under identical flow conditions. The measurements show preferential accumulation of inertial particles in regions of lower turbulence intensity, consistent with the expected behavior of turbophoretic transport.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

The experiments confirm inertial particles accumulate in low-turbulence zones after density normalization, but the separate-run matching step is the weakest link.

read the letter

The main thing to know is that this paper runs PIV measurements in one- and two-grid oscillating turbulence and shows inertial particles building up where turbulence intensity is lower, once they divide the inertial number density by the tracer density from separate but nominally identical runs. The direction matches the expected turbophoretic drift toward weaker fluctuations, which is what the abstract highlights as the result. They do a clean job describing the setup and the normalization trick to subtract out settling or preferential concentration. That approach is straightforward and lets them focus on the inertia effect without needing a full new theory. The work is mostly an experimental check on an existing prediction rather than a fresh derivation or first observation. The soft spot is the normalization itself. Because the inertial and tracer data come from independent experiments, any small drift in grid stroke, frequency, or mean flow between runs could create a spurious spatial variation in the ratio. The abstract says the flows are the same, but without reported overlap maps, pointwise TKE differences, or cross-checks, it is hard to know how well the isolation holds. Two-way coupling at even modest loadings could also shift the local intensity in the inertial case. No error bars or statistical tests appear in the abstract, so the quantitative strength of the agreement is unclear. This is useful for people who model particle transport in inhomogeneous turbulence, such as atmospheric or engineering flows, because it gives a controlled lab case to test against. It is not a breakthrough but supplies a benchmark that deserves a serious referee to examine the flow-matching controls and data reduction steps. I would send it for review rather than desk reject.

Referee Report

2 major / 2 minor

Summary. The manuscript reports experimental measurements of turbophoresis for inertial particles in grid-generated inhomogeneous turbulence. Using PIV, the authors measure fluid velocity fields and particle number densities in setups with one and two oscillating grids. To isolate turbophoresis, inertial-particle densities are normalized pointwise by densities obtained from separate noninertial (tracer) runs under nominally identical flow conditions. The resulting normalized distributions show accumulation of inertial particles in large-scale regions of lower turbulence intensity, consistent with the theoretical turbophoretic velocity V_tp ~ τ_p ∇<u'^2> directed toward minima of turbulent kinetic energy.

Significance. If the normalization procedure successfully isolates the inertia-driven migration, the work supplies direct experimental confirmation of turbophoresis in a laboratory flow with controlled inhomogeneity. This strengthens the empirical basis for models of particle transport in inhomogeneous turbulence and could inform applications such as aerosol dynamics or sediment transport. The use of independent tracer runs and comparison to analytic predictions are positive features.

major comments (2)

[Experimental methods / Results (normalization procedure)] The central isolation step (normalization of inertial-particle number density by the corresponding noninertial density) assumes that the turbulence intensity fields are identical between the two independent experimental runs. No quantitative overlap metrics—such as pointwise TKE difference maps, cross-correlation coefficients of intensity fields, or reported tolerances on grid stroke/frequency—are provided to substantiate this assumption. Because grid-driven flows are sensitive to small setup variations, any mismatch could produce spurious spatial modulation in the ratio and undermine the claim that the observed accumulation is purely turbophoretic.
[Abstract and Results section] The abstract and main text state that the normalized distributions agree with theoretical predictions, yet no quantitative error bars, statistical significance tests, or discussion of residual biases (e.g., from preferential concentration or weak settling) are supplied. This makes it difficult to assess the strength of the agreement or to rule out contributions from other mechanisms.

minor comments (2)

[Abstract] Notation for turbulence intensity and Stokes time should be defined consistently at first use; the symbol for the turbophoretic velocity appears without an explicit equation reference in the abstract.
[Figure captions] Figure captions should explicitly state the number of independent realizations and the spatial resolution of the PIV measurements to allow readers to judge statistical reliability.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive and detailed comments, which help clarify the presentation of our experimental results on turbophoresis. We address each major comment below and indicate the revisions we will make to the manuscript.

read point-by-point responses

Referee: The central isolation step (normalization of inertial-particle number density by the corresponding noninertial density) assumes that the turbulence intensity fields are identical between the two independent experimental runs. No quantitative overlap metrics—such as pointwise TKE difference maps, cross-correlation coefficients of intensity fields, or reported tolerances on grid stroke/frequency—are provided to substantiate this assumption. Because grid-driven flows are sensitive to small setup variations, any mismatch could produce spurious spatial modulation in the ratio and undermine the claim that the observed accumulation is purely turbophoretic.

Authors: We acknowledge that the original manuscript does not provide explicit quantitative metrics to demonstrate the degree of overlap in turbulence intensity between the inertial-particle and tracer runs. Grid-driven flows can indeed be sensitive to minor variations. In the revised manuscript we will add the following: (i) reported tolerances on grid stroke and frequency (maintained within ±2% across all paired runs), (ii) cross-correlation coefficients of the measured TKE fields between corresponding inertial and tracer experiments (values exceed 0.93 in the core measurement region), and (iii) a supplementary figure displaying pointwise relative differences in turbulence intensity. These additions will allow readers to assess the validity of the normalization procedure directly. revision: yes
Referee: The abstract and main text state that the normalized distributions agree with theoretical predictions, yet no quantitative error bars, statistical significance tests, or discussion of residual biases (e.g., from preferential concentration or weak settling) are supplied. This makes it difficult to assess the strength of the agreement or to rule out contributions from other mechanisms.

Authors: We agree that a more rigorous quantitative comparison would strengthen the claims. In the revision we will include error bars on the normalized particle-density profiles, calculated from the standard deviation over multiple independent realizations. We will also add a short discussion of residual biases, noting that preferential concentration is weak at the large scales examined here and that gravitational settling velocities are at least an order of magnitude smaller than the observed turbophoretic velocities for the particle parameters used. A quantitative measure of agreement (normalized root-mean-square deviation between measured and predicted profiles) will be reported in the Results section. revision: yes

Circularity Check

0 steps flagged

No circularity: experimental normalization compared to independent theory

full rationale

The paper reports direct PIV measurements of inertial-particle number densities in oscillating-grid turbulence, normalized pointwise by noninertial-tracer densities acquired in separate runs under nominally identical conditions. This normalized field is then compared to the sign and location of the turbophoretic velocity predicted by an independent theoretical expression (V_tp proportional to tau_p times gradient of turbulence intensity). No equation or result in the presented chain is obtained by fitting a parameter to the same data set and then relabeling it a prediction; the normalization step is an experimental isolation procedure rather than a definitional identity, and the theoretical benchmark is external to the current measurements. Self-citations to prior turbulence modeling, if present, are not required to close the central experimental claim.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The claim rests on the established turbophoresis model and standard fluid-dynamics assumptions; no new free parameters, ad-hoc axioms, or invented entities are introduced.

axioms (1)

domain assumption Turbophoretic velocity is proportional to particle Stokes time multiplied by the gradient of turbulence intensity and directed toward minima of turbulent velocity.
This relation is invoked to interpret the observed particle accumulation in low-turbulence regions.

pith-pipeline@v0.9.0 · 5472 in / 1160 out tokens · 29759 ms · 2026-05-08T18:46:50.262467+00:00 · methodology

Review history (3 revisions) →

discussion (0)

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

Cost / Foundation.LogicAsFunctionalEquation washburn_uniqueness_aczel (J = ½(x+x⁻¹)−1) unclear
V_turboph = -(τ_p/3) ∇⟨u²⟩ ... derived by iteration of the particle equation of motion for small Stokes time
Foundation.AlexanderDuality / DimensionForcing (D=3 ambient assumed) alexander_duality_circle_linking unclear
grid Reynolds number Re_g = f A_g²/ν = 2520; Stokes number St = τ_p/τ_ν