Stability of electrodynamically levitated one or many charged droplets in the presence of noise

Mohit Singh; Rochish Thaokar; Y.S. Mayya

REVIEW 3 major objections 4 minor 16 references

Stability of electrodynamically levitated one or many charged droplets in the presence of noise

T0 review · 3 major / 4 minor · reviewed 2026-08-14 · deepseek-v4-flash

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

Pith's one-line read Gaussian white noise in an electrodynamic balance makes the inter-drop separation of two charged droplets fluctuate with exactly one third the variance of a single droplet, and drives an order-disorder transition near L0 ≈ 0.1.

desk verdict Useful two-droplet variance formula and a plausible disordering threshold, but the noise projection onto the relative coordinate is underived and the L0 conversion has an algebraic slip. read the letter →

arxiv 1908.06762 v1 pith:GDHQAHVH submitted 2019-08-19 physics.flu-dyn physics.app-phphysics.data-an

Mohit Singh , Y.S. Mayya , Rochish Thaokar This is my paper · ORCID

classification physics.flu-dynphysics.app-phphysics.data-an PACS 05.40.-a47.65.-d36.40.Ei

keywords chargeddropletselectrodynamicbalancethermalnoisepseudo-potentialapproximationmodifiedMathieuequationCoulombiccrystalorderparameterphasetransition

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 claims that when charged droplets levitated in an electrodynamic balance are driven by Gaussian white noise, the positional variance of a single droplet is $\sigma_r^2 = 2 k_B T (1+c^2)/(a^2 m \omega^2)$, while for two mutually repelling droplets the variance of the inter-drop separation is exactly one third of that: $\sigma_r^2 = 2 k_B T (1+c^2)/(3 a^2 m \omega^2)$. It further claims that above a noise strength $L_0 \approx 0.1$ the ordered Coulombic crystal of roughly a hundred droplets transforms into a disordered, liquid-like structure. These formulas matter because they turn the worry that noise destabilizes levitated droplet arrays into a quantitative, testable prediction. The paper's approach is to apply the classical pseudo-potential method to the noisy many-droplet problem and to compare the resulting variances with numerical solutions of the non-homogeneous modified Mathieu equation.

What carries the argument

The central machinery is the Dehmelt (adiabatic) pseudo-potential approximation, which splits the droplet motion into a fast micro-oscillation at the drive frequency and a slow secular drift, replacing the oscillating quadrupole force by a time-averaged conservative potential. The paper couples this deterministic potential to a Boltzmann weight with an effective temperature, so that the probability distribution of the slow coordinate is Gaussian and its variance is read off by comparison with the standard Gaussian form. For the two-drop system the Coulomb repulsion term $b^2/r^2$ adds to the pseudo-potential, and the curvature of the combined well at the equilibrium separation $\bar{r}^3 = 2 b^2 (1+c^2)/a^2$ is $E''(r_0) = \frac{3}{2} a^2/(1+c^2)$, which is the factor of three that produces the $1/3$ variance ratio. The structural transition is diagnosed by an order parameter, either $\langle \sigma \rangle / \langle r \rangle$ for small drop numbers or the radial distribution function $g(r)$ for large ones.

What would settle it

A direct numerical solution of the full stochastic equations of motion without the adiabatic decomposition, run over many trajectories with controlled Gaussian noise strength $L_0$, would settle the claim: if the ratio of two-drop inter-drop variance to single-drop variance is not close to $1/3$ over the claimed range of $a$, or if the disordering threshold in the hundred-drop radial distribution function appears at a noise strength far from $L_0 \approx 0.1$, the central claim fails.

Watch

Extended reading notes

Core claim

The central discovery is that mutual Coulomb repulsion between two levitated droplets deepens the effective trapping well at their equilibrium separation by a factor of three, so that noise-induced fluctuations of the inter-drop distance have one third the variance of a single droplet's position at the same noise level. In dimensionless terms the paper obtains $\sigma_r^2 = \frac{1}{3} (L_0/a)^2 h (1+c^2)/a$, and this prediction agrees with simulations of the stochastic equation of motion for stability parameters $a \lesssim 0.5$. The paper also reports that for a hundred-droplet system the radial distribution function loses its sharp peaks when $L_0$ exceeds about 0.1, and that the same threshold appears as a change in the slope of the order parameter $\langle \sigma \rangle / \langle r \rangle$ in the two-drop case. This is presented as a noise-driven transition from a well-ordered Coulombic cluster to an amorphous or fluid-like arrangement.

Load-bearing premise

The load-bearing assumption is that the motion can still be cleanly split into fast and slow parts when random noise is present, so the noise only enters through a Boltzmann factor of the deterministic pseudo-potential; if the noise couples the fast and slow scales strongly, the variance formulas and the $L_0 \approx 0.1$ threshold do not follow.

Editorial extensions

If this is right

The $1/3$ variance ratio means that relative droplet spacing is more robust to noise than absolute position, within the validity of the adiabatic approximation.
Inter-drop Coulomb repulsion does not change the stability boundary, so two droplets remain trapped up to the same stability parameter as a single droplet when noise is present.
Above $L_0 \approx 0.1$, the mean inter-drop separation grows roughly as $L_0^{0.6}$, giving a concrete scaling law for the swelling of the cluster under noise.
A hundred-droplet levitated cluster loses its sharp radial correlation peaks at the same threshold, predicting a measurable solid-to-liquid-like crossover in experiments.
Because only Gaussian white noise statistics enter the theory, athermal mechanical or electrical fluctuations with the same statistics can be treated with the same variance formulas.

Reading between the lines

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

If the adiabatic split degrades at high noise, the exact threshold may be non-universal; a useful extension would be to solve the full stochastic Mathieu system without the Dehmelt approximation and compare thresholds.
The $1/3$ factor likely follows from the curvature of the effective two-body well and may generalize to other harmonic-plus-Coulomb trap configurations, giving a family of variance ratios for different cluster modes.
The transition at $L_0 \approx 0.1$ is a crossover rather than a true thermodynamic phase transition; finite-size scaling with $N$ could show whether the threshold sharpens as the number of droplets grows.
For the contactless-membrane application, the paper's threshold supplies a design criterion: keep dimensional noise levels below the equivalent $L_0 \approx 0.1$ to preserve ordered arrays for particle capture.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

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

The useful bit here is the two-droplet result: eq. 22/23, the variance of the separation being one third of the single-drop variance, plus the simulation evidence that a levitated Coulomb cluster disorders above L0 ~ 0.1. If those survive scrutiny they give a simple design rule for contactless membranes and aerosol traps. The single-droplet part is mostly a re-derivation of Arnold and others, so the novelty is the interacting case.

What the paper does well: the Dehmelt/pseudo-potential treatment is standard and applied carefully; the simulations are independent Langevin runs, not fits to the theory; the radial distribution function comparison for 100 drops is a nice way to visualize the transition. The authors also correctly flag the difficulty of separating slow and fast motion under noise.

Now the soft spots. First, the most important: eqs. (4)-(6) use the same symbol f(τ) for the noise on every droplet coordinate. The paper never says whether the droplets see the same realization or independent ones. When you reduce to the relative coordinate, subtracting the two equations gives f1 - f2 (independent noise) or zero (common noise), not the single f(τ) that appears in eq. (14). The 1/3 variance is the equipartition result for a relative mode in a thermal bath, so it implicitly assumes independent thermal noise on each droplet. But the equations as written don't say that, and the derivation of eq. (14) is not shown. This needs to be stated and fixed; as written, it is an equation-level gap.

Second, the noise-strength conversion is off. Plugging L0^2 = 2K_BT/(h m ω^2) into eq. (9) gives σ^2 = (L0/a)^2 h (1+c^2), but eq. (10) writes (L0/a)^2 h/a (1+c^2). The extra factor of a is never explained. Eq. (23) inherits it. That's a typo or a definitional mismatch, but it affects any design use of the formula.

Third, the L0 ~ 0.1 threshold and the L0^0.6 scaling are empirical read-offs from finite-size simulations, with no error bars, no system-size dependence, and no derived condition. The word "phase transition" is loose. The authors do call it loosely termed, so that is partially mitigated. No code or data are released, which makes this hard to reproduce.

The central idea — that interaction modifies the variance by a factor, and that there is a noise threshold for structural order — holds up in broad strokes. The specific claims need correction and more careful support before they become reliable numbers. This is a legitimate niche paper for levitated droplet and aerosol physics. It deserves a serious referee, but the referee should send it back for major revision.

Recommendation: send to peer review, conditional on the noise-projection and L0 conversion issues being addressed.

Referee Report

3 major / 4 minor

Summary. The paper develops an analytical pseudo-potential (Dehmelt) theory for the position variance of a single charged droplet and for the inter-droplet separation variance of two charged droplets in an electrodynamic balance subject to Gaussian white noise. The theoretical formulas are compared with Langevin simulations of the modified Mathieu equations, and the simulations are used to identify a noise-strength threshold L0 approximately 0.1 above which an ordered Coulomb cluster transforms into an amorphous, liquid-like structure. The paper also reports radial distribution functions for 100 droplets and discusses implications for contactless membrane applications.

Significance. If the central formulas hold, the paper provides compact, parameter-free predictions for thermal and athermal fluctuations in levitated charged-droplet structures, including the notable two-drop result that the relative-separation variance is one third of the single-drop variance (Eqs. 22-23). The authors are transparent about the adiabatic approximation and about the difficulty of separating slow and fast motions in the presence of noise. However, the significance is offset by an equation-level gap in the reduction to relative coordinates and by an unclear conversion from thermal energy to the noise parameter; these issues affect the central two-drop formula and the inferred transition threshold.

major comments (3)

[Two drop system, Eqs. (4)-(6) and (14)] The stochastic differential equations (4)-(6) contain a single scalar noise f(τ) added to every coordinate equation. When the two single-drop equations are subtracted to obtain the relative-coordinate equation, the noise term becomes either f(τ)-f(τ)=0 if the droplets share the same noise realization, or f1(τ)-f2(τ) if the realizations are independent. Neither case yields the f(τ) appearing in Eq. (14). The factor 1/3 in Eqs. (22)-(23) is the equipartition variance of the relative mode and is correct only if that mode is driven by white noise of the same amplitude L0 as a single drop. The manuscript never specifies whether droplets experience common or independent noise, and this is load-bearing: common noise would leave the relative mode undriven, while independent noise would change the amplitude and hence the variance formula and the threshold. Please state the noise statistics explicitly and re-derive Eq. (14) from Eqs. (4)-(6) accordingly.
[Single drop system, definition of L0 and Eq. (10)] The conversion from thermal energy to the noise parameter L0 is not internally consistent. The text defines L0 = sqrt(2 K_B T / (h m ω^2)) with h the time step, which introduces a numerical discretization parameter into a supposedly physical noise strength. Equation (10) implies σ_r^2 is proportional to L0^2 and inversely proportional to a^3, but the text states that σ_r^2 is proportional to L0. This discrepancy affects the single-drop formula and is carried over to the two-drop formula (23). Please provide a dimensionally consistent, time-step-free definition of L0 and reconcile the stated functional dependence with the derived expression.
[Two drop system, Eqs. (13)-(14) and (16)-(17)] The deterministic slow equation for the relative coordinate is written with the force term (1/2) a r/(1+c^2), but the potential (16) and the equilibrium condition (17) imply that this force should be (1/2) a^2 r/(1+c^2). The missing factor a makes Eq. (13) dimensionally inconsistent with the rest of the derivation. This appears to be a typographical error, but it should be corrected because Eq. (14) is the starting point for the noise analysis and the factor enters the variance formula.

minor comments (4)

[Equations (4)-(6)] In Eq. (5), the inter-particle coupling term is written with (x_i - x_j) in the numerator, but the equation is for the y-coordinate; this should presumably be (y_i - y_j).
[Title and text] The title and several places in the text use 'droplet' where 'droplets' is intended (e.g., 'one or many charged droplet'). Please correct the grammar.
[Abstract and conclusions] The abstract says theory and simulations are in 'fair agreement', while the conclusions state 'close agreement' with a 10% deviation at a < 0.5. Please make the characterization consistent.
[Figure 2(c)] The caption of Fig. 2(c) mentions a comparison with theory, but the main text does not explain how the theoretical order-parameter curve is computed. Please describe the theoretical curve and the manner of comparison.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the variance formulas are parameter-free consequences of the stated pseudo-potential and Boltzmann model, and the L0≈0.1 transition is an empirical simulation observation, not a fitted input recycled as prediction.

full rationale

The central variance results are derived, not fitted. For the single drop, Eq. (8) gives the slow effective force, the potential is E(r) = (1/2) a^2 \bar r^2 m \omega^2/(1+c^2), and equating the Boltzmann weight to a Gaussian gives \sigma_r^2 = 2 K_B T (1+c^2)/(a^2 m \omega^2), Eq. (9). For two drops, the effective potential from Eq. (13) has curvature E''(r0) = (3/2) a^2/(1+c^2), leading to Eq. (22), \sigma_r^2 = 2 K_B T (1+c^2)/(3 a^2 m \omega^2). No parameter is fitted to the simulation data and then renamed as a prediction; the factor 1/3 is the analytic curvature ratio, not an input. The simulations are independent Langevin calculations used to test the theory, and the threshold L0≈0.1 and the L0^0.6 scaling are read off the simulations rather than inserted into the variance derivation. The self-citations, e.g. refs. [7] and [15] for the numerical scheme and the two-drop stationary-state geometry, provide context and methods from prior published work; they are not the load-bearing source of the variance formulas. The paper itself concedes that separating slow and fast motion is difficult in the presence of noise, and it defers the detailed derivation of Eq. (13) to a supplementary file; those are evidentiary or correctness limitations, not circular substitutions. A separate equation-level concern is that Eqs. (4)–(6) place the same f(\tau) on every coordinate while the relative-coordinate Eq. (14) keeps f(\tau) after subtracting equations; this is an inconsistency in the derivation, but it is not circular because the formula does not reduce by definition to its input. Accordingly, no circular step meeting the quoted-reduction standard is present, and the circularity score is 0.

Assumptions & free parameters 2 free parameters · 5 assumptions · 0 invented entities

No new particles, forces, dimensions, or conserved quantities are introduced. The 'effective temperature' is a standard mapping from fluctuation strength to thermal energy, and the order parameter sigma/<r> is a diagnostic, not an invented entity. The main inputs are the physical parameters a, c, b, and the empirical threshold and scaling exponent.

free parameters (2)

threshold noise strength L0_c = ~0.1
The transition from ordered to disordered structure is read off from the slope change in fig. 2c and from the disappearance of g(r) peaks in fig. 3. It is an empirical observation, not a derived quantity.
scaling exponent for mean separation vs noise = ~0.6
For L0>0.1, the mean inter-drop separation is observed to scale as <r>~L0^0.6 (fig. 2b). The exponent is fit to simulation data and used to predict the order-parameter slope of 0.4 in fig. 2c.

assumptions (5)

domain assumption Dehmelt pseudo-potential approximation: fast and slow motion separate and slow motion follows the time-averaged effective potential.
Used to derive the slow equations (8) and (13); the paper notes this separation is difficult to justify in the presence of noise.
domain assumption Noise is Gaussian, zero-mean, isotropic white noise with <f(t)f(t')> = L0' delta(t-t').
Stated in the governing equations section; this is the noise model for all simulations and theory.
standard math Stationary distribution of the slow coordinate follows the Boltzmann factor exp(-E(r)/kBT) with the effective potential E(r).
Standard statistical mechanics; applied in eqs. (9) and (21).
domain assumption Two drops settle into symmetric coplanar oscillations with equilibrium separation r0^3 = 2b^2(1+c^2)/a^2.
Taken from prior work [7,15]; used to expand the potential near equilibrium in eqs. (17)-(20).
domain assumption Gravitational and dielectrophoretic forces are negligible for small drops.
Stated before eq. (7); the dielectric force is approximated as O(a^3).

how reviews work

0 comments

Cite this review

Pith. "Pith review of Stability of electrodynamically levitated one or many charged droplets in the presence of noise." pith.science (2026). https://pith.science/paper/GDHQAHVH

@misc{pith2026190806762,
  author       = {Pith},
  title        = {Pith review of: Stability of electrodynamically levitated one or many charged droplets in the presence of noise},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/GDHQAHVH}},
  note         = {Machine review of arXiv:1908.06762}
}

read the original abstract

The theory of the effect of external fluctuation force on the stability and spatial distribution of mutually interacting and slowly evaporating charged drops, levitated in an electrodynamic balance, is presented using classical pseudo-potential approach. The theory is supplemented with numerical simulations where the non-homogeneous modified Mathieu equation is solved for single droplet as well as many droplets. The transition from the well ordered Coulombic crystal to randomly distributed liquid like structure is observed above a threshold value of the order parameter. The theory and simulations are found to be in fair agreement with each other. The simulation is aimed at studying the stability of structures for capturing the pollutant particle form the air streams using contactless membrane.

Figures

Figures reproduced from arXiv: 1908.06762 by the authors.

**Figure 2.** (a) Numerical simulation of change in the mean [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗

**Figure 3.** The radial density distribution of 100 number of [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

16 extracted references · 16 canonical work pages

[1]

and Dehmelt H

Wineland D., Ekstrom P. and Dehmelt H. , Physical review letters, 31 (1279) 1973

work page 1973
[2]

Brown L. S. and Gabrielse G. ,, Reviews of Modern Physics, 58 (233) 1986

work page 1986
[3]

Dehmelt, H. G. ,, Advances in Atomic and Molecular Physics, 3 (72) 1968

work page 1968
[4]

Paul W.,, Reviews of modern physics , 62 (531) 1990

work page 1990
[5]

and Ortjohann H

Winter H. and Ortjohann H. W. , American Journal of Physics, 59 (813) 1991

work page 1991
[6]

F., Shelton H

Wuerker R. F., Shelton H. and Langmuir R. V. , Jour- nal of Applied Physics , 30 (349) 1959

work page 1959
[7]

S., Gaware J

Singh M., Mayya Y. S., Gaware J. and Thaokar R. M., Journal of Applied Physics , 121 (054503) 2017

work page 2017
[8]

Arnold, Stephen and Hessel, N ,, Review of scientiﬁc instruments, 56 (2069) 1985

work page 1985

Show all 16 references

[9]

Arnold S., Folan L. M. and Korn A. ,, Journal of ap- plied physics, 74 (4297) 1993

1993
[10]

H., Holler S., Korn A

Arnold S., Li J. H., Holler S., Korn A. and Izmailov A. F. , Journal of applied physics , 78 (3571) 1995

1995
[11]

and Siemers I

Blatt R., Zoller P., Holzmuller G. and Siemers I. , Zeitschrift f¨ ur Physik D Atoms, Molecules and Clusters, 4 (126) 1986

1986
[12]

and H¨anggi P

Zerbe C., Jung P. and H¨anggi P. , Physical Review E , 49 (3626) 1994

1994
[13]

and Lindner A., Zeitschrift fur Physik D Atoms, Molecules and Clusters , 11 (300) 1989

Joos E. and Lindner A., Zeitschrift fur Physik D Atoms, Molecules and Clusters , 11 (300) 1989

1989
[14]

L., Vehring R., Davis E

Aardahl C. L., Vehring R., Davis E. J., Schweiger G. and Swanson B. D. , Journal of aerosol science , 28 (1505) 1997

1997
[15]

and Mayya, Y

Singh M., Thaokar R., Khan A. and Mayya, Y. S. , Physical Review E, 98 (032202) 2018

2018
[16]

and An- derson J

Tepper G., Fenn J., Kessick R., Pestov D. and An- derson J. , In 2006 Sixth IEEE Conference on Nanotech- nology, 2 (782) 2006 p-7

2006

Pith tools

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

[1] [1]

and Dehmelt H

Wineland D., Ekstrom P. and Dehmelt H. , Physical review letters, 31 (1279) 1973

work page 1973

[2] [2]

Brown L. S. and Gabrielse G. ,, Reviews of Modern Physics, 58 (233) 1986

work page 1986

[3] [3]

Dehmelt, H. G. ,, Advances in Atomic and Molecular Physics, 3 (72) 1968

work page 1968

[4] [4]

Paul W.,, Reviews of modern physics , 62 (531) 1990

work page 1990

[5] [5]

and Ortjohann H

Winter H. and Ortjohann H. W. , American Journal of Physics, 59 (813) 1991

work page 1991

[6] [6]

F., Shelton H

Wuerker R. F., Shelton H. and Langmuir R. V. , Jour- nal of Applied Physics , 30 (349) 1959

work page 1959

[7] [7]

S., Gaware J

Singh M., Mayya Y. S., Gaware J. and Thaokar R. M., Journal of Applied Physics , 121 (054503) 2017

work page 2017

[8] [8]

Arnold, Stephen and Hessel, N ,, Review of scientiﬁc instruments, 56 (2069) 1985

work page 1985