Batchelor's formula and infrared renormalization for sedimentation

Antoine Gloria; Mitia Duerinckx

Batchelor's formula and infrared renormalization for sedimentation

T0 review · 0 major / 4 minor · reviewed 2026-07-14 · grok-4.5

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

Pith's one-line read Batchelor’s dilute correction to the mean settling speed of a random suspension is rigorously justified after infrared renormalization of Stokes interactions.

desk verdict This is the first rigorous justification of Batchelor’s formula and of a shape-independent infinite-volume settling speed, via a new infrared renormalization that actually works. read the letter →

arxiv 2607.09995 v1 pith:CILSZU7R submitted 2026-07-10 physics.flu-dyn math-phmath.APmath.MP

Mitia Duerinckx , Antoine Gloria This is my paper · ORCID

classification physics.flu-dynmath-phmath.APmath.MP MSC 76T2035Q3060H1582D30 PACS 47.15.G47.57.E05.40.-a

keywords sedimentationStokesflowBatchelorformulainfraredrenormalizationclusterexpansionrandomsuspensionsmeansettlingspeedhydrodynamicinteractions

topics Navier-Stokes Regularity

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

When rigid particles settle under gravity in a viscous fluid, their hydrodynamic interactions are long-ranged. This creates infrared divergences that have blocked a rigorous definition of the infinite-volume mean settling speed and a proof of Batchelor’s famous first dilute correction. The paper shows that, in dimensions greater than two and for stationary suspensions with quantitative decorrelation, an infinite-volume mean settling speed exists, is independent of container shape, and equals the almost-sure limit of the relative local settling speed of particles inside large containers. The same quantity admits a renormalized cluster expansion whose two-particle term recovers Batchelor’s formula exactly, with a controlled remainder. The argument works by isolating the singular large-scale part of every hydrodynamic interaction, subtracting the counterterms that encode the mean backflow of the suspension, and controlling the regular remainder by elliptic estimates. For a sympathetic reader the result settles a classical foundational question: the mean settling speed is an intrinsic bulk quantity, and Batchelor’s correction is not merely a formal expression but a theorem.

What carries the argument

Infrared renormalization of hydrodynamic interactions: every infinite-volume observable is split into an explicit singular part carrying the non-integrable large-scale Stokes tail and a regular remainder controlled by elliptic estimates; the singular part is cancelled by counterterms that encode the mean backflow, implemented cluster-by-cluster via a finitary diagrammatic expansion into reflection blocks.

What would settle it

Compute the relative settling speed for a hardcore Poisson suspension of spheres at volume fractions φ≈0.01–0.05 inside large cylinders of different aspect ratios and check whether the measured dilute slope matches the absolutely convergent two-particle integral (1.18) and is independent of cylinder shape.

Watch

Extended reading notes

Core claim

In dimension d>2, for a stationary ergodic δ-hardcore point process whose two-point correlation decays slightly faster than |x|⁻², the infinite-volume mean settling speed exists, is independent of container geometry, and equals the Stokes velocity of a single particle plus Batchelor’s two-particle correction plus a remainder of higher order in the density.

Load-bearing premise

The particles must decorrelate fast enough that multi-point correlations decay slightly faster than inverse-square; without that decay the large-scale Green-function integrals do not converge and the counterterms fail.

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.

Desk Editor's Note

read the letter

This paper finally closes a classical gap: under d>2 and quantitative mixing (pair correlations slightly faster than |x|^{-2}, Ursell functions controlled up to order 8), the infinite-volume mean settling speed exists, equals the almost-sure relative local speed in large containers of any shape, and expands as Stokes velocity plus Batchelor’s two-particle correction plus a controlled remainder. That is the real news.

What is new is the infrared machinery. They isolate the non-integrable large-scale part of the Stokes interactions, cancel it with explicit counterterms that encode the mean backflow (the linear pressure gradient forced by the nonzero mean force density), and control the regular remainder by elliptic estimates. For the dilute expansion they invent a finitary “reflection-block” diagrammatic expansion that isolates the divergent substructures cluster by cluster and makes the cancellations of Lemma 4.7 visible. This is substantially harder than their earlier effective-viscosity work, because the Stokeslet is longer-range and the backflow must be renormalized at every order. The three main theorems are proved in full; the Green-function bounds, energy estimates, and diagrammatic bookkeeping are written out carefully.

The soft spots are minor and already flagged by the authors. The remainder estimates lose a square root from the nonlinear buckling argument (Remark 1.5); pushing the cluster expansion one order higher recovers the expected O(λ^{2}) for hardcore Poisson, but they stop at the order needed for Batchelor. The mixing assumption is load-bearing, yet it is essentially sharp: the Stokeslet decays as |x|^{2-d}, so you need slightly better than |x|^{-2} for the convolutions to converge. No free parameters, no circular fitting, and the self-citations supply technical lemmas that are used cleanly.

This is for people who work on stochastic homogenization of Stokes systems or on the mathematical theory of suspensions. Anyone who has ever wondered whether Batchelor’s renormalization can be made rigorous will get value from it. The math is solid, the citation pattern is appropriate, and the result is important enough that a serious editor should send it to referees without hesitation. I would engage with it and expect to cite the construction and the reflection-block idea.

Referee Report

0 major / 4 minor

Summary. The paper constructs an infinite-volume mean settling speed for stationary random hardcore suspensions of rigid particles in Stokes flow (d>2) under quantitative decorrelation of the two-point correlation, shows that this quantity is the almost-sure limit of the relative local mean settling speed in large containers independently of container shape (cylinder, snow-globe, finite cylinder), and derives a renormalized second-order cluster expansion that recovers Batchelor’s two-particle correction as an absolutely convergent integral involving the pair density and the one- and two-particle Stokes flows. The argument proceeds by infrared regularization of the Stokes system (massive term plus bulk-viscosity penalization), identification of the mean-backflow counterterm from the Fredholm alternative, a finitary reflection-block expansion of multi-particle flows that isolates divergent substructures, and diagrammatic cancellation identities that make the cluster integrals finite.

Significance. A rigorous justification of Batchelor’s formula has been open for decades because of the non-integrable Stokeslet. The paper supplies a complete, self-contained resolution under essentially sharp mixing assumptions: existence of the infinite-volume speed (Thm 1.1), shape-independent large-container limit for the relative velocity (Thm 1.2), and the renormalized expansion with explicit two-particle term and controlled remainder (Thm 1.4 / Prop 4.1). The reflection-block calculus and the integral cancellation identities (Lemmas 4.5–4.7) are reusable technical tools for other long-range hydrodynamic problems. The physical discussion in §5 correctly separates relative settling, intrinsic convection and fluctuations, clarifying what is and is not claimed.

minor comments (4)

Remark 1.5 notes that the remainder bound in Thm 1.4 is not optimal and can be improved by pushing the cluster expansion to fourth order. A short pointer in the statement of Thm 1.4 itself would help the reader who only consults the main theorems.
The diagrammatic notation of §4.6 is introduced carefully, but a one-line dictionary of the most frequent diagrams (two-legged J-edge, terminal G-edge, dashed correlation edge) placed near the first use of the rules would improve readability of the long expansions in §§4.7–4.9.
Appendix A.2 recovers Batchelor’s original spherical formula from (1.18). A brief numerical remark that the resulting coefficient α≈6.55 matches the classical value would make the equivalence more immediate for the fluid-mechanics audience.
A few minor typos appear (e.g., “detailss” near the end of Step 5 of Lemma 4.10; occasional missing spaces after punctuation). A light copy-edit pass would suffice.

Circularity Check

1 steps flagged · score 1.0 of 10

No significant circularity: Batchelor formula is derived from Stokes PDE + cluster expansion with explicit infrared counterterms; self-citations supply independent technical lemmas only.

self citation load bearing [§3 (proof of Thm 1.2), Step 4; citation of [13, Thm 3]]
"By the large-scale Lipschitz regularity property of [13, Theorem 3] (which holds under mere ergodicity of P), there exists an almost surely finite random variable r∗ such that for all r∗ ≤ r ≤ R, ∫_{Br} |∇(u_{R,i} − u_{∞,i})|^{2} ≲ ∫_{BR} |∇(u_{R,i} − u_{∞,i})|^{2}."

The large-container identification of relative settling speed with V̄_∞ uses the authors’ prior large-scale regularity result [13]. This is a genuine technical dependency, but [13] is an independent elliptic regularity theorem under ergodicity alone and does not assume or encode Batchelor’s formula; the central expansion (Thm 1.4) does not rely on it. Hence only minor, non-load-bearing self-citation.

full rationale

The derivation chain is self-contained. Theorem 1.1 constructs the infinite-volume settling speed from the Stokes system with a massive/divergence-penalized infrared cut-off; the counterterm α = λ|B|/(1−λ|B|) is identified from the mean force balance / Fredholm alternative (eqs. 2.2–2.5), not fitted. Theorem 1.2 compares finite-container solutions to this infinite-volume field via energy estimates and the authors’ prior large-scale regularity theory [13], which is an independent elliptic estimate under mere ergodicity and does not assume Batchelor’s formula. Theorem 1.4 / Proposition 4.1 then expands the centered field via a finitary reflection-block decomposition (Lemma 4.5, key dependence rule 4.43), applies the integral cancellation identities of Lemma 4.7, and controls remainders by diagrammatic convolution estimates under the stated Ursell decay; the two-particle term (1.18) is obtained by absolute convergence after subtracting the explicit far-field backflow, not by renaming an empirical fit. Self-citations ([13], [14], [16], [24]) provide technical tools (regularity, scalar analogues, viscosity-cluster methods) that do not include the target expansion as an assumption. No equation reduces a claimed prediction to a fitted input or to a self-definitional identity. Score 1 only for the presence of non-load-bearing self-citation of technical lemmas.

Assumptions & free parameters 0 free parameters · 4 assumptions · 2 invented entities

The central claims rest on the classical Stokes equations with rigid inclusions, stationarity/ergodicity of the point process, the δ-hardcore separation, quantitative decay of Ursell functions, and dimension d>2. No free parameters are fitted; the only 'invented' objects are technical devices (reflection blocks, infrared cut-offs) whose existence is proved, not postulated.

assumptions (4)

domain assumption The fluid velocity satisfies the steady Stokes system with no-slip on rigid particles, force/torque balance, and homogeneous boundary conditions on the container (or whole-space decay).
Standard continuum model for low-Reynolds sedimentation; invoked from the outset (eqs. 1.3, 1.12).
domain assumption The particle centers form a stationary ergodic δ-hardcore point process whose two-point (and higher) Ursell functions decay at least as |x|^{-2-eta}.
Minimal statistical assumption guaranteeing infrared cancellations; stated in Theorems 1.1 and 1.4 and used throughout §2–4.
domain assumption Space dimension d>2.
Ensures integrability of the Stokeslet after renormalization; critical dimension d=2 is excluded (explicitly noted).
standard math Existence and uniqueness of finite-energy solutions to the infrared-regularized Stokes problems with finitely many inclusions (Lemmas 4.2, B.1).
Classical elliptic theory for the Stokes operator with the chosen massive + bulk-viscosity regularization.

invented entities (2)

Reflection blocks (finitary diagrammatic expansion of multi-particle Stokes flows) independent evidence
purpose: Isolate leading infrared-divergent substructures so that cancellations of Lemma 4.7 can be applied term-by-term.
Technical device constructed by iterating Green representations; existence and decay are proved, not postulated as physical objects.
Infrared counterterms encoding mean backflow (α = λ|B|/(1-λ|B|)) independent evidence
purpose: Cancel the non-integrable zero mode of the Stokeslet so that infinite-volume observables remain finite.
Derived from the Fredholm alternative / mean force balance (eqs. 2.2–2.5); not an extra free parameter.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Batchelor's formula and infrared renormalization for sedimentation." pith.science (2026). https://pith.science/paper/CILSZU7R

@misc{pith2026260709995,
  author       = {Pith},
  title        = {Pith review of: Batchelor's formula and infrared renormalization for sedimentation},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/CILSZU7R}},
  note         = {Machine review of arXiv:2607.09995}
}

abstract

We study the sedimentation of stationary random suspensions of rigid particles in Stokes flow. Batchelor's formula predicts the first dilute correction to the infinite-volume mean settling speed due to hydrodynamic interactions between suspended particles. A rigorous derivation has long been obstructed by the long-range nature of the Stokes flow, which gives rise to infrared divergences in the large-volume limit. In dimension $d>2$, for stationary suspensions satisfying quantitative decorrelation assumptions, we construct the infinite-volume mean settling speed and show that it governs the relative settling speed of particles in large containers, independently of the container shape. We then establish a renormalized cluster expansion of this mean settling speed in the dilute regime and compute it up to the two-particle term, thereby justifying Batchelor's formula. The proof is based on the infrared renormalization of hydrodynamic interactions. Infinite-volume observables are decomposed into an explicit singular part, carrying the non-integrable large-scale contribution, and a regular remainder controlled by elliptic estimates. The singular part is renormalized through counterterms that encode the diverging mean backflow generated by the suspension. At the level of the dilute cluster expansion, the renormalization is implemented cluster by cluster and the singular-regular decomposition is achieved through a finitary diagrammatic expansion of hydrodynamic interactions, inspired by the method of reflections, which isolates the leading divergent substructures and exposes the key cancellations.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

42 extracted references

[1]

G. K. Batchelor. Sedimentation in a dilute dispersion of spheres.J. Fluid Mech., 52(2):245–268, 1972

1972
[2]

G. K. Batchelor. Brownian diffusion of particles with hydrodynamic interaction.Journal of Fluid Mechanics, 74(1):1–29, 1976

1976
[3]

C. W. J. Beenakker and P. Mazur. Is sedimentation container-shape dependent?The Physics of Fluids, 28(11):3203–3206, 11 1985

1985
[4]

J.Bourgain.Invariantmeasuresforthe2d-defocusingnonlinearschrödingerequation.Communications in Mathematical Physics, 176(2):421–445, 1996

1996
[5]

Bruneau, R

D. Bruneau, R. Anthore, F. Feuillebois, X. Auvray, and C. Petipas. Measurement of the average velocity of sedimentation in a dilute polydisperse suspension of spheres.J. Fluid Mech., 221:577–596, 1990

1990
[6]

Bruneau, F

D. Bruneau, F. Feuillebois, R. Anthore, and E. J. Hinch. Intrinsic convection in a settling suspension. Physics of Fluids, 8(8):2236–2238, 08 1996

1996
[7]

J. M. Burgers. On the influence of the concentration of a suspension upon the sedimentation velocity (in particular for a suspension of spherical particles).Proc. Kon. Nederl. Akad. Wet., 44:1045–1051 & 1177–1184, 1941

1941
[8]

J. M. Burgers. On the influence of the concentration of a suspension upon the sedimentation velocity (in particular for a suspension of spherical particles).Proc. Kon. Nederl. Akad. Wet., 45:9–16 & 126–128, 1942

1942

Show all 42 references

[9]

R. E. Caflisch and J. H. C. Luke. Variance in the sedimentation speed of a suspension.Phys. Fluids, 28(3):759–760, 1985

1985
[10]

Da Prato and A

G. Da Prato and A. Debussche. Strong solutions to the stochastic quantization equations.Ann. Probab., 31(4):1900–1916, 2003

1900
[11]

Duerinckx and A

M. Duerinckx and A. Gloria. Analyticity of homogenized coefficients under Bernoulli perturbations and the Clausius-Mossotti formulas.Arch. Ration. Mech. Anal., 220(1):297–361, 2016

2016
[12]

Duerinckx and A

M. Duerinckx and A. Gloria. Corrector equations in fluid mechanics: effective viscosity of colloidal suspensions.Arch. Ration. Mech. Anal., 239(2):1025–1060, 2021

2021
[13]

Duerinckx and A

M. Duerinckx and A. Gloria. Quantitative homogenization theory for random suspensions in steady Stokes flow.J. Éc. polytech. Math., 9:1183–1244, 2022

2022
[14]

Duerinckx and A

M. Duerinckx and A. Gloria. Sedimentation of random suspensions and the effect of hyperuniformity. Ann. PDE, 8(1):Paper No. 2, 66, 2022

2022
[15]

Anal., 134(3-4):437–453, 2023

M.DuerinckxandA.Gloria.TheClausius-Mossottiformula.Asymptot. Anal., 134(3-4):437–453, 2023

2023
[16]

Duerinckx and A

M. Duerinckx and A. Gloria.On Einstein’s effective viscosity formula, volume 7 ofMemoirs of the European Mathematical Society. EMS Press, Berlin, 2023

2023
[17]

Duerinckx, A

M. Duerinckx, A. Gloria, and F. Otto. The structure of fluctuations in stochastic homogenization. Comm. Math. Phys., 377:259–306, 2020

2020
[18]

Geigenmüller and P

U. Geigenmüller and P. Mazur. Sedimentation of homogeneous suspensions in finite vessels.J. Statist. Phys., 53(1-2):137–173, 1988

1988
[19]

Gérard-Varet

D. Gérard-Varet. Derivation of the Batchelor-Green formula for random suspensions.J. Math. Pures Appl. (9), 152:211–250, 2021

2021
[20]

Gérard-Varet

D. Gérard-Varet. A simple justification of effective models for conducting or fluid media with dilute spherical inclusions.Asymptot. Anal., 128(1):31–53, 2022

2022
[21]

Gérard-Varet and M

D. Gérard-Varet and M. Hillairet. Analysis of the viscosity of dilute suspensions beyond Einstein’s formula.Arch. Ration. Mech. Anal., 238(3):1349–1411, 2020. 4Views and opinions expressed are however those of the authors only and do not necessarily reflect those of the Europe...

2020
[22]

Gérard-Varet and R

D. Gérard-Varet and R. M. Höfer. Mild assumptions for the derivation of Einstein’s effective viscosity formula.Comm. Partial Differential Equations, 46(4):611–629, 2021

2021
[23]

Gérard-Varet and A

D. Gérard-Varet and A. Mecherbet. Analysis of a sedimenting suspension near a vertical wall.J. Éc. polytech. Math., 12:523–546, 2025

2025
[24]

A. Gloria. A scalar version of the Caflisch-Luke paradox.Comm. Pure Appl. Math., 74(7):1403–1452, 2021

2021
[25]

Gloria and F

A. Gloria and F. Otto. An optimal variance estimate in stochastic homogenization of discrete elliptic equations.Ann. Probab., 39(3):779–856, 2011

2011
[26]

Goldman, M

M. Goldman, M. Huesmann, and F. Otto. Quantitative linearization results for the Monge-Ampère equation.Comm. Pure Appl. Math., 74(12):2483–2560, 2021

2021
[27]

Guazzelli and J

E. Guazzelli and J. Hinch. Fluctuations and Instability in Sedimentation.Annu. Rev. Fluid Mech., 43:97–116, 2011

2011
[28]

Guazzelli and J

E. Guazzelli and J. Morris.A Physical Introduction to Suspension Dynamics. Cambridge University Press, 2011

2011
[29]

Gubinelli, P

M. Gubinelli, P. Imkeller, and N. Perkowski. Paracontrolled distributions and singular PDEs.Forum Math. Pi, 3:e6, 75, 2015

2015
[30]

B. M. Haines and A. L. Mazzucato. A proof of Einstein’s effective viscosity for a dilute suspension of spheres.SIAM J. Math. Anal., 44(3):2120–2145, 2012

2012
[31]

M. Hairer. A theory of regularity structures.Invent. Math., 198(2):269–504, 2014

2014
[32]

Hasimoto

H. Hasimoto. On the periodic fundamental solutions of the Stokes equations and their application to viscous flow past a cubic array of spheres.J. Fluid Mech., 5(2):317–328, 1959

1959
[33]

Hillairet and R

M. Hillairet and R. M. Höfer. Hindered settling of well-separated particle suspensions.Pure Appl. Anal., 6(2):581–609, 2024

2024
[34]

R. M. Höfer. Convergence of the method of reflections for particle suspensions in Stokes flows.J. Differ. Equ., 297:81–109, 2021

2021
[35]

Huesmann and K.-T

M. Huesmann and K.-T. Sturm. Optimal transport from Lebesgue to Poisson.Ann. Probab., 41(4):2426–2478, 2013

2013
[36]

D. L. Koch and E. S. G. Shaqfeh. Screening in sedimenting suspensions.J. Fluid Mech., 224:275–303, 1991

1991
[37]

Nozières

P. Nozières. A local coupling between sedimentation and convection: Application to the Beenakker- Mazur effect.Physica A: Statistical Mechanics and its Applications, 147(1):219–237, 1987

1987
[38]

G. C. Papanicolaou and S. R. S. Varadhan. Boundary value problems with rapidly oscillating random coefficients. InRandom fields, Vol. I, II (Esztergom, 1979), volume 27 ofColloq. Math. Soc. János Bolyai, pages 835–873. North-Holland, Amsterdam, 1981

1979
[39]

Pertinand.Effective properties of random media : expansions in dilute regimes

J. Pertinand.Effective properties of random media : expansions in dilute regimes. PhD thesis, Sor- bonne université, 2022

2022
[40]

Peysson and E

Y. Peysson and E. Guazzelli. An experimental investigation of the intrinsic convection in a sedimenting suspension.Physics of Fluids, 10(1):44–54, 01 1998

1998
[41]

von Smoluchowski

M. von Smoluchowski. Zur kinetischen Theorie der Brownschen Molekularbewegung und der Suspen- sionen.Ann. Phys., 21:756–780, 1906

1906
[42]

von Smoluchowski

M. von Smoluchowski. Über die Wechselwirkung von Kugeln, die sich in einer zähen Flüssigkeit bewegen.Bull. Acad. Sci. Cracovie A, 1:28–39, 1911. (MitiaDuerinckx)Université Libre de Bruxelles, Département de Mathématique, 1050 Brus- sels, Belgium Email address:mitia.duerinckx@u...

1911

Pith tools

Reviewed July 14, 2026 · model on record in the stance chip above.

[1] [1]

G. K. Batchelor. Sedimentation in a dilute dispersion of spheres.J. Fluid Mech., 52(2):245–268, 1972

1972

[2] [2]

G. K. Batchelor. Brownian diffusion of particles with hydrodynamic interaction.Journal of Fluid Mechanics, 74(1):1–29, 1976

1976

[3] [3]

C. W. J. Beenakker and P. Mazur. Is sedimentation container-shape dependent?The Physics of Fluids, 28(11):3203–3206, 11 1985

1985

[4] [4]

J.Bourgain.Invariantmeasuresforthe2d-defocusingnonlinearschrödingerequation.Communications in Mathematical Physics, 176(2):421–445, 1996

1996

[5] [5]

Bruneau, R

D. Bruneau, R. Anthore, F. Feuillebois, X. Auvray, and C. Petipas. Measurement of the average velocity of sedimentation in a dilute polydisperse suspension of spheres.J. Fluid Mech., 221:577–596, 1990

1990

[6] [6]

Bruneau, F

D. Bruneau, F. Feuillebois, R. Anthore, and E. J. Hinch. Intrinsic convection in a settling suspension. Physics of Fluids, 8(8):2236–2238, 08 1996

1996

[7] [7]

J. M. Burgers. On the influence of the concentration of a suspension upon the sedimentation velocity (in particular for a suspension of spherical particles).Proc. Kon. Nederl. Akad. Wet., 44:1045–1051 & 1177–1184, 1941

1941

[8] [8]

J. M. Burgers. On the influence of the concentration of a suspension upon the sedimentation velocity (in particular for a suspension of spherical particles).Proc. Kon. Nederl. Akad. Wet., 45:9–16 & 126–128, 1942

1942