arxiv: 2604.04945 · v1 · submitted 2026-03-28 · ⚛️ physics.class-ph · cond-mat.mes-hall· physics.flu-dyn

Recognition: 2 theorem links

· Lean Theorem

Induced-current magnetophoresis

V. Kumaran

Authors on Pith no claims yet

Pith reviewed 2026-05-14 22:24 UTC · model grok-4.3

classification ⚛️ physics.class-ph cond-mat.mes-hallphysics.flu-dyn

keywords magnetophoresiseddy currentsLorentz forceconducting particlesoscillating magnetic fieldanisotropic diffusionparticle alignment

0 comments

The pith

Oscillating inhomogeneous magnetic fields exert steady forces on electrically conducting non-magnetic particles through induced eddy currents.

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

The paper establishes that a spatially varying oscillating magnetic field induces eddy currents inside conducting particles. The Lorentz force from these currents has a steady component that produces net force and, for rods, torque even though the particles carry no permanent magnetism. For a sphere the force is proportional to minus mu zero times radius cubed times the dot product of the field gradient and the field amplitude. For a thin rod the force and aligning torque are proportional to radius squared times length with coefficients that depend on the ratio of particle size to magnetic penetration depth. Particle interactions enter the number-density equation as an anisotropic diffusion term whose coefficient is negative perpendicular to the field, amplifying concentration fluctuations.

Core claim

When an electrically conducting non-magnetic particle is subjected to a spatially varying and oscillating applied magnetic field of amplitude H + G · x and frequency omega, an oscillating eddy current is induced. The Lorentz force density has a steady component that produces a net force on a sphere of radius R proportional to -mu0 R^3 G · H and on a thin rod of radius R and length L proportional to -mu0 R^2 L (G · H - 1/2 (G · o-hat)(H · o-hat)). A torque mu0 R^2 L (o-hat × H)(o-hat · H) acts on the rod and tends to align it with the field. The coefficients depend on the dimensionless ratio beta R = sqrt(mu0 omega kappa R^2). The effect of particle interactions appears as an anisotropic term

What carries the argument

The time-averaged Lorentz force density formed by the cross product of the induced current density and the applied magnetic field, which yields a nonzero steady component only when the applied field is spatially inhomogeneous.

Load-bearing premise

The applied magnetic field is taken to vary linearly in space while the particle is treated as non-magnetic and perfectly conducting at the driving frequency, with no back-reaction on the external field or fluid flow.

What would settle it

Place a conducting sphere or rod of known radius in a controlled oscillating magnetic field with measured gradient G and field H, record its steady velocity or torque, and check whether the observed force matches the predicted scaling with R^3 G · H or the rod expressions.

Figures

Figures reproduced from arXiv: 2604.04945 by V. Kumaran.

**Figure 2.** Figure 2: FIG. 2. The coefficients [PITH_FULL_IMAGE:figures/full_fig_p017_2.png] view at source ↗

**Figure 3.** Figure 3: FIG. 3. Schematic of the oscillating magnetic field [PITH_FULL_IMAGE:figures/full_fig_p019_3.png] view at source ↗

**Figure 4.** Figure 4: FIG. 4. The coefficients [PITH_FULL_IMAGE:figures/full_fig_p022_4.png] view at source ↗

**Figure 5.** Figure 5: FIG. 5. Schematic for calculating the torque on a conducting [PITH_FULL_IMAGE:figures/full_fig_p023_5.png] view at source ↗

**Figure 6.** Figure 6: FIG. 6. The scaled magnitude of the diffusion coefficient for (a) [PITH_FULL_IMAGE:figures/full_fig_p032_6.png] view at source ↗

read the original abstract

When an electrically conducting non-magnetic particle is subjected to a spatially varying and oscillating applied magnetic field of amplitude $\mathcal{H} + \mathcal{G} \cdot x$ and frequency $\omega$, an oscillating eddy current is induced. The Lorentz force density, the cross product of the current density and the magnetic field, consists of a steady component and a component with frequency $2 \omega$. If there is a spatial variation in the applied field, there is a steady force on a sphere of radius $R$ proportional to $- \mu_0 R^3 \mathcal{G} \cdot \mathcal{H} $, and a steady force on a thin rod of radius $R$ and length $L$ proportional to $- \mu_0 R^2 L (\mathcal{G} \cdot \mathcal{H} - \tfrac{1}{2} (\mathcal{G} \cdot \hat o)(\mathcal{H} \cdot \hat o))$, where $\mu_0$ is the magnetic permeability. There is torque proportional to $\mu_0 R^2 L (\hat o \times \mathcal{H} ) (\hat o \cdot \mathcal{H} )$ on a thin rod which tends to align the rod direction of the magnetic field. The coefficients in the force and torque expressions are functions of the dimensionless ratio of the radius and the penetration depth of the magnetic field, $\beta R = \sqrt{\mu_0 \omega \kappa R^2}$, where $\kappa$ is the electrical conductivity. It is shown that the effect of particle interactions can be expressed as an anisotropic diffusion term in the equation for the particle number density. The diffusion coefficient is negative, and concentration fluctuations are amplified, in the plane perpendicular to the magnetic field.

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 paper derives steady forces and torques on conducting particles from induced eddy currents in oscillating gradient fields, plus negative anisotropic diffusion from interactions, but the back-reaction omission needs checking.

read the letter

The main takeaway is that this paper derives steady forces and torques on non-magnetic conducting particles in an oscillating magnetic field with a linear gradient, arising from the Lorentz force on induced eddy currents. For a sphere the force scales as -μ0 R^3 G · H, while a thin rod gets both a modified force and a torque that aligns it with the field. Interactions between particles are cast as an anisotropic diffusion term whose coefficient is negative perpendicular to the field, which would amplify concentration fluctuations rather than damp them. The coefficients depend on the dimensionless parameter βR that measures particle size against the magnetic penetration depth. This is new relative to standard magnetophoresis, which usually starts from magnetic susceptibility instead of conductivity-driven currents. The derivation begins from Maxwell equations and the Lorentz force density without fitted parameters or circular definitions, and the expressions are dimensionally consistent. The paper does a clean job of integrating the force density over the particle volume and then showing how the resulting single-particle forces feed into a mean-field interaction model for the number density. A reader gets explicit, usable formulas for both isolated particles and their collective effect. The central soft spot is the assumption that the magnetic field inside each particle is exactly the imposed external form H + G · x. This leaves out the first-order field correction produced by the eddy currents themselves. Because the interaction term already relies on the perturbation one particle creates for its neighbors, the same correction enters at leading order and could shift the force coefficients or flip the sign of the perpendicular diffusion term. The abstract gives no indication that this correction was estimated or shown to be parametrically small. The work is aimed at researchers in electromagnetism and soft matter who study particle manipulation in fluids. Someone looking for contactless control methods based on conductivity rather than magnetism would find the explicit expressions and the negative-diffusion prediction useful. It deserves a serious referee because the derivations are grounded in standard equations, the results are specific enough to test, and the potential application is clear even if the back-reaction issue requires attention in revision. I would recommend sending it to peer review.

Referee Report

1 major / 2 minor

Summary. The manuscript derives the steady component of the Lorentz force arising from induced eddy currents on electrically conducting non-magnetic particles in a linearly varying oscillating magnetic field H + G · x. For a sphere the steady force is proportional to −μ₀ R³ G · H; for a thin rod of radius R and length L the force is proportional to −μ₀ R² L (G · H − ½ (G · ô)(H · ô)) with an aligning torque μ₀ R² L (ô × H)(ô · H). Both sets of coefficients are functions of the dimensionless parameter βR = √(μ₀ ω κ R²). Particle interactions are shown to produce an anisotropic diffusion term in the number-density equation whose coefficient is negative in the plane perpendicular to the applied field, implying amplification of concentration fluctuations.

Significance. If the central expressions hold, the work supplies a parameter-free derivation, starting from Maxwell’s equations and the Lorentz force, of a magnetophoretic mechanism that does not require intrinsic particle magnetism. The explicit force, torque, and interaction-induced diffusion expressions constitute falsifiable predictions that could be tested in microfluidic or low-Reynolds-number experiments. The predicted perpendicular instability is a distinctive feature with potential implications for controlled assembly or separation of conducting particles.

major comments (1)

[Derivation of steady force and interaction model] The force and torque expressions (abstract and the derivations leading to the sphere and rod results) are obtained by integrating the Lorentz force density J × B while taking the internal magnetic field to be exactly the imposed linear form H + G · x. The first-order correction to B produced by the eddy currents themselves is omitted. Because the interaction model constructs the effective field on one particle from the perturbation generated by its neighbors, the same omitted correction appears at leading order in the mean-field interaction; if it is not parametrically smaller than the retained G · H term, both the force coefficients and the sign of the perpendicular diffusion coefficient can be altered.

minor comments (2)

[Notation and parameter definitions] The definition βR = √(μ₀ ω κ R²) is introduced in the abstract but should be restated explicitly when the coefficients are first plotted or tabulated in the main text.
[Abstract] The abstract states that particle interactions 'can be expressed as an anisotropic diffusion term' but does not give the explicit form of the diffusion tensor or the section in which it is derived.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for their careful reading and insightful comments. We address the concern regarding the approximation used in deriving the force, torque, and interaction terms below.

read point-by-point responses

Referee: The force and torque expressions (abstract and the derivations leading to the sphere and rod results) are obtained by integrating the Lorentz force density J × B while taking the internal magnetic field to be exactly the imposed linear form H + G · x. The first-order correction to B produced by the eddy currents themselves is omitted. Because the interaction model constructs the effective field on one particle from the perturbation generated by its neighbors, the same omitted correction appears at leading order in the mean-field interaction; if it is not parametrically smaller than the retained G · H term, both the force coefficients and the sign of the perpendicular diffusion coefficient can be altered.

Authors: We acknowledge that our derivation approximates the internal magnetic field as the externally imposed linear field H + G · x when computing the induced current density J and the subsequent Lorentz force. This neglects the first-order correction δB generated by the eddy currents themselves. As noted, this correction is of relative order (βR)^2, where βR is the dimensionless parameter already appearing in our coefficients. For moderate values of βR (βR ≲ 1), the approximation remains accurate to leading order. In the interaction model, the effective field from neighboring particles is computed using the same consistent approximation for the induced perturbations, ensuring that the mean-field interaction is treated at the same level of accuracy. The anisotropic nature of the diffusion, leading to negative diffusion perpendicular to the field, arises from the directional dependence of the induced currents and torques, which we expect to persist even with higher-order corrections. Nevertheless, to address the referee's concern, we will add a dedicated paragraph in the discussion section clarifying the approximation, its validity range, and the potential impact of O((βR)^2) corrections on the force coefficients and diffusion sign for large βR. We do not anticipate changes to the central expressions but will include this caveat. revision: partial

Circularity Check

0 steps flagged

No significant circularity in derivation chain

full rationale

The paper derives the steady forces and torques by direct integration of the Lorentz force density J × B after solving the eddy-current problem for the imposed linear field H + G · x inside non-magnetic conducting particles. The interaction contribution is obtained by superposing the perturbation fields from neighboring particles to produce an anisotropic diffusion term in the number-density equation. No step reduces by construction to a fitted parameter renamed as a prediction, a self-definitional relation, or a load-bearing self-citation; the coefficients are explicit functions of the dimensionless ratio βR obtained from the Maxwell solution. The derivation is therefore self-contained against the stated assumptions.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The central claim rests on standard electromagnetic theory with no free parameters fitted to data; the only dimensionless group βR emerges directly from the governing equations. No new entities are postulated.

axioms (2)

standard math Maxwell's equations in the quasi-static limit for conducting media
Used to obtain the induced current density from the oscillating applied field.
domain assumption Linear spatial variation of the applied magnetic field amplitude
Stated explicitly as H + G · x; required for the force expressions to be proportional to G · H.

pith-pipeline@v0.9.0 · 5622 in / 1354 out tokens · 70692 ms · 2026-05-14T22:24:18.586936+00:00 · methodology

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.

IndisputableMonolith/Cost/FunctionalEquation.lean washburn_uniqueness_aczel unclear

?

unclear
Relation between the paper passage and the cited Recognition theorem.

steady force on a sphere … proportional to −μ₀ R³ G·H … obtained by integrating the Maxwell stress … over SI
IndisputableMonolith/Foundation/AbsoluteFloorClosure.lean absolute_floor_iff_bare_distinguishability unclear

?

unclear
Relation between the paper passage and the cited Recognition theorem.

effect of particle interactions … anisotropic diffusion term … negative … perpendicular to the magnetic field

What do these tags mean?

matches: The paper's claim is directly supported by a theorem in the formal canon.
supports: The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends: The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses: The paper appears to rely on the theorem as machinery.
contradicts: The paper's claim conflicts with a theorem or certificate in the canon.
unclear: Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

39 extracted references · 39 canonical work pages

[1]

In the following analysis, the accent ˜ is used to denote complex varia bles, while real variables are written without the accent

Therefore, the amplitude of the oscillatory force and torque a re also obtained by the substitution ˜H ∗ → ˜H in the resulting expression. In the following analysis, the accent ˜ is used to denote complex varia bles, while real variables are written without the accent. The calligraphic font is use d for the applied magnetic ﬁeld ( H) and the magnetic ﬁeld...

work page
[2]

There is a steady force on an electrically conducting spherical pa rticle of radius R in a spatially varying and oscillating applied magnetic ﬁeld of amplitude H + G · x and frequency ω , where x is the position vector from the center of the particle. The steady magnetophoretic force on the particle is of the form ¯F = − ¯Γ sµ 0R3 ˜H · G, where the positi...

work page
[3]

The force is in the direction of decreasing magnetic ﬁeld components parallel an d perpendicular to the orientation vector

For a thin rod with radius R length L, high aspect ratio L ≫ R and orientation vector ˆo, the magnetophoretic force is ¯F = − ¯Γ rµ 0R2L(G · H − 1 2(G · ˆo)(H · ˆo)). The force is in the direction of decreasing magnetic ﬁeld components parallel an d perpendicular to the orientation vector. In the viscous limit, the mobility of a particle (ratio of velocit...

work page
[4]

In a viscous ﬂow , the induced angular velocity is proportional to µ 0|H|2R2 ¯Γ r log (L/R )/ηL 2

There is a torque on a thin rod T = 1 2 µ 0R2L¯Γ r( ˆo × H)( ˆo ·H), which tends to align the rod in the direction parallel to the magnetic ﬁeld. In a viscous ﬂow , the induced angular velocity is proportional to µ 0|H|2R2 ¯Γ r log (L/R )/ηL 2. 36 The time scale for the alignment of the thin rod is compared to the tra nslation time over a distance compara...

work page
[5]

The eﬀect of far-ﬁeld magnetic particle interactions and the mod iﬁcation of the applied magnetic ﬁeld due to particle magnetisation is considered. For spher ical particles, when there is a small spatial variation in the particle number density, the eﬀect of interactions reduces to an anisotropic diﬀusion term in the conserv ation equation 117 for the num...

work page
[6]

However, in this case, it is shown that number density variations are damped along the magnetic ﬁeld and ampliﬁed perpendicular to the magnetic ﬁeld

For a suspension of thin rods, the eﬀect of interactions can not be reduced to an anisotropic diﬀusion term in the conservation equation 132 for the n umber density. However, in this case, it is shown that number density variations are damped along the magnetic ﬁeld and ampliﬁed perpendicular to the magnetic ﬁeld. The diﬀusion coeﬃcients are proportional ...

work page
[7]

These 37 produce velocity ﬁelds that inﬂuence the dynamics of neighbouring p articles

There are also hydrodynamic interactions between the particle, because the Maxwell stress generates a force moment for each particle (equations 45 , 70 and 71). These 37 produce velocity ﬁelds that inﬂuence the dynamics of neighbouring p articles. However, the convective term in the number density conservation equation, 90, is zero because the velocity ﬁe...

work page 2006
[8]

The cross section of the rod is a disk of unit radius in scaled co-ordinates centered at the origin

Dipole moment perpendicular to the axis In the plane perpendicular to the axis, two-dimensional polar harmo nics are used to determine the induced dipole moment. The cross section of the rod is a disk of unit radius in scaled co-ordinates centered at the origin. The x− y co-ordinate system is used, where r = √ x2 + y2 is the distance from the origin. The ...

work page
[9]

Within the cylinder, there is a variation in the magnet ic ﬁeld with radial position

Dipole moment parallel to the axis The component of the magnetic ﬁeld H∥ parallel to the axis is uniform outside the con- ducting cylinder. Within the cylinder, there is a variation in the magnet ic ﬁeld with radial position. The magnetic ﬁeld has the same form as B8, with H ⊥ replaced by H ∥ . Since H ∥ is perpendicular to ˜ζ (2), the magnetic ﬁeld is ex...

work page
[10]

H. K. Moﬀatt. On the behaviour of a suspension of conductin g particles subjected to a time-periodic magnetic ﬁeld. J. Fluid Mech. , 218:509–529, 1990. 45

work page 1990
[11]

H. K. Moﬀatt. Electromagnetic stirring. Phys. Fluids , 3:1336–1343, 1991

work page 1991
[12]

V. Kumaran. Rheology of a suspension of conducting parti cles in a magnetic ﬁeld. J. Fluid Mech., 871:139–185, 2019

work page 2019
[13]

V. Kumaran. A suspension of conducting particles in a mag netic ﬁeld - the maxwell stress. J. Fluid Mech. , 901:A36, 2020

work page 2020
[14]

V. Kumaran. Eddies driven by eddy currents: Magnetokine tic ﬂow in a conducting drop due to an oscillating magnetic ﬁeld. EPL, 148:63001, 2024

work page 2024
[15]

V. Kumaran. Flow in an electrically conducting drop due t o an oscillating magnetic ﬁeld. J. Fluid Mech. , 1014:A36, 2025

work page 2025
[16]

M. Bu, T. B. Christensen, K. Smistrup, A. Wolﬀ, and M. F. Mik kel F. Hansen. Characteriza- tion of a microﬂuidic magnetic bead separator for high-thro ughput applications. Sensors and Actuators A: Physical , 145-146:430–436, 2008

work page 2008
[17]

Nameni, M

A. Nameni, M. H. Kayhani, V. V. Mashayekhi, and M. M. Nazar i. Magnetophoresis-based particle separation in microﬂuidic channels using grooved current-carrying conductor: Design, simulation, and optimization. Journal of Magnetism and Magnetic Materials , 624:173026, 2025

work page 2025
[18]

Bruno Frazier, and A

Ozgun Civelekoglu, A. Bruno Frazier, and A. Fatih Sariog lu. The origins and the current applications of microﬂuidics-based magnetic cell separat ion technologies. Magnetochemistry, 8:10, 2022

work page 2022
[19]

Reiter, J

N. Reiter, J. Auchter, M. Weber, S. Berensmeier, and S. P . Schwaminger. Magnetophoretic cell sorting: Comparison of diﬀerent 3d-printed milliﬂuidi c devices. Magnetochemistry, 8:113, 2022

work page 2022
[20]

S. N. Murthy, S. M. Sammeta, and C. Bowers. Magnetophore sis for enhancing transdermal drug delivery: Mechanistic studies and patch design. Journal of Controlled Release , 148:197– 203, 2010

work page 2010
[21]

L. Shao, Z. Feng, Q. Du, L. Guangzheng Zou, and Z. Tan. Res earch advances in magne- tophoretic separation: Fundamentals and methodologies. Separation & Puriﬁcation Reviews , 55:56–71, 2026

work page 2026
[22]

S. S. Leong, Z. Ahmad, S. C. Low, J. Camacho, J. Faraudo, a nd J.-K. Lim. Uniﬁed view of magnetic nanoparticle separation under magnetophoresis. Langmuir, 36:8033–8055, 2020

work page 2020
[23]

Munaz, Shiddiky M

A. Munaz, Shiddiky M. J. A., and N. T. Nguyen. Recent adva nces and current challenges in magnetophoresis based micro magnetoﬂuidics. Biomicroﬂuidics, 12:031501, 2018

work page 2018
[24]

Alnaimat, S

F. Alnaimat, S. Dagher, B. Mathew, A. Hilal-Alnqbi, and S. Khashan. Microﬂuidics based magnetophoresis: A review. The Chemical Record, 18:1596–1612, 2018

work page 2018
[25]

V. Kumaran. The eﬀect of inter-particle hydrodynamic an d magnetic interactions in a mag- netorheological ﬂuid. J. Fluid Mech. , 944:A49, 2022

work page 2022
[26]

Toner and Y

J. Toner and Y. Tu. Long-range order in a two-dimensiona l dynamical XY model: How birds ﬂy together. Phys. Rev. Lett. , 75:4326–4329, 1995

work page 1995
[27]

R. A. Simha and S. Ramaswamy. Hydrodynamic ﬂuctuations and instabilities in ordered suspensions of self-propelled particles. Phys. Rev. Lett. , 89:058101, 2002

work page 2002
[28]

Ramaswamy

S. Ramaswamy. The mechanics and statistics of active ma tter. Annual Review of Condensed Matter Physics , 1:323–345, 2010. 46

work page 2010
[29]

L. P. Dadhichi, S. K. Sahoo, K. V. Kumar, and S. Ramaswamy . Superdiﬀusion and antidiﬀu- sion in an aligned active suspension. arXiv preprint, arXiv:2519.00740v2, 2025

work page arXiv 2025
[30]

A. Kolin. An electromagnetokinetic phenomenon involv ing migration of neutral particles. Science, 117:134–137, 1953

work page 1953
[31]

H. K. Moﬀatt and A. Sellier. Migration of an insulating pa rticle under the action of uniform ambient electric and magnetic ﬁelds. part 1. general theory . J. Fluid Mech. , 464:279–286, 2002

work page 2002
[32]

Yariv and T

E. Yariv and T. Miloh. Electro-magneto-phoresis of sle nder bodies. J. Fluid Mech. , 577:331–340, 2007

work page 2007
[33]

V. G. Levich. Physicochemical Hydrodynamics. Prentice-Hall, Englewood Cliﬀs, New Jersey, USA., 1962

work page 1962
[34]

T. M. Squires and M. Z. Bazant. Induced-charge electro- osmosis. J. of Fluid Mech. , 509:217– 252, 2004

work page 2004
[35]

D. J. Griﬃths. Introduction to electrodynamics; 4th ed. Pearson, Boston, MA, 2013

work page 2013
[36]

L. D. Landau, E. M. Lifshitz, and L. P. Pitaevskii. Electrodynamics of Continuous Media . Butterworth-Heinemann, Oxford, UK, 2014

work page 2014
[37]

H. Brenner. Rheology of a dilute suspension of axisymme tric brownian particles. Int. J. Multiphase Flow , 1:195–341, 1974

work page 1974
[38]

The direction reverses upon inversion between right- a nd left-handed co-ordinates

work page
[39]

The direction does not change upon inversion between ri ght- and left-handed co-ordinates. 47

work page