Pith. sign in

REVIEW 3 major objections 5 minor 54 references

The four-color theorem can govern how many-component fluids phase-separate in two dimensions, suppressing coalescence when four or more phases coexist.

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

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

In 2D, phase separation of four or more fluid components coarsens by diffusion rather than coalescence, a change the authors link to the four-color theorem; 3D systems only approach this behavior for many components.

T0 review reviewed 2026-08-03 challenge →

load-bearing objection The N=3 vs N≥4 coarsening dichotomy and the master-curve collapse are real and worth engaging; the four-color-theorem story is a plausible heuristic but the stated graph-theory premise is imprecise. the 3 major comments →

arxiv 2511.20215 v4 pith:WNSAKPJB submitted 2025-11-25 physics.flu-dyn

Topology Controls the Phase Separation Dynamics of Many Component Fluid Mixtures

Michael Rennick , Xitong Zhang , Halim Kusumaatmaja This is my paper
classification physics.flu-dyn
keywords four-color theoremphase separationcoarsening dynamicsmulti-component fluidsOstwald ripeningtopological constraintscoalescence suppressionplanar maps
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

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 argues that the coarsening of many-component fluid mixtures is shaped by mathematical coloring constraints. In two dimensions, when four or more immiscible fluids separate, the four-color theorem guarantees that no two domains of the same fluid need to touch, so coalescence is suppressed and growth proceeds by diffusion along a single master curve. With only three fluids, odd-neighbor configurations force same-fluid contacts, so coalescence cannot be avoided and the dynamics remain hydrodynamic. The paper further shows that tuning interfacial tensions can break this topological protection, and that thin films recover the two-dimensional rule once domains grow taller than the film thickness. If correct, this gives a predictive, design-oriented framework for multi-phase soft materials.

Core claim

The central claim is that in two dimensions, the number of coexisting fluid phases N splits the coarsening behavior into distinct classes: N=2 shows familiar hydrodynamic scaling, N=3 is intermediate with coalescence-driven growth, and N≥4 exhibits diffusion-limited coarsening L* ∝ (t*)^{1/3} with a universal prefactor a_N. The explanation is topological: the four-color theorem ensures any planar arrangement of domains can be colored with four colors so that adjacent domains have different colors; for N≥4 the system can always rearrange after a domain shrinks to avoid same-color contact, while for N=3 the requirement that every face have even degree forces coalescence. The authors derive a_N

What carries the argument

The key object is the four-color theorem: any planar map can be colored with four colors so that no adjacent regions share a color. Applied to the adjacency graph of fluid domains, it means that with at least four phases, a stable coloring exists regardless of local neighbor counts, so same-phase contact can be avoided after each domain-shrink event. For three phases, the paper invokes the criterion that only maps in which every face has an even number of neighbors are three-colorable, making coalescence unavoidable. A second element is a standard steady-state droplet-size distribution for diffusion-limited coarsening, used to compute the prefactors that collapse the data.

Load-bearing premise

The paper assumes that the adjacency graphs of coarsening fluid domains in 2D always satisfy the condition that three-colorability is equivalent to every domain having an even number of neighbors; if that equivalence fails for the actual fluid networks, the claimed topological necessity of the N=3 versus N≥4 distinction collapses.

What would settle it

Find or simulate a 2D arrangement of three immiscible fluids where a domain has an odd number of neighbors yet is still three-colorable without same-color contact; if that configuration remains stable and coarsens without coalescence, the claimed topological suppression for N=3 is undercut.

Watch this falsifier. Get emailed when new claim-graph text bears on it.

If this is right

  • In flat geometries, a mixture of four or more phases will coarsen by diffusion at a predictable rate, not by the faster hydrodynamic coalescence seen in binary mixtures.
  • The universal master curve gives a direct quantitative test: measure domain growth in a 2D multi-phase system and check whether L* follows a_N times t^{1/3}.
  • Thin-film confinement can impose the two-dimensional topological rule on an otherwise three-dimensional system, as long as domains are thicker than the film.
  • Cloaking provides a way to tune individual component growth laws, potentially allowing designed heterogeneous microstructures.
  • In bulk 3D, the benefit of adding more components is gradual rather than a sharp threshold; at least about seven phases are needed for noticeable coalescence suppression.

Where Pith is reading between the lines

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

  • The dichotomous N=3 versus N≥4 behavior rests on a restricted characterization of three-colorable planar maps; if coarsening adjacency graphs violate the even-degree condition, some N=3 configurations might avoid coalescence, softening the claimed cutoff.
  • The master-curve collapse suggests a possible universal scaling function for all planar systems with at least four phases, testable experimentally in droplet monolayers of synthetic DNA nanostars or colloids.
  • The coloring/coarsening connection may extend to other interface-topology-driven coarsening problems, such as grain growth in polycrystalline films, where the number of 'colors' is not fixed a priori.
  • Cloaking could be exploited as a design tool for hierarchical emulsions, where one phase encapsulates others, producing intentionally heterogeneous coarsening dynamics useful for microencapsulation.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit.

Referee Report

3 major / 5 minor

Summary. The manuscript reports lattice-Boltzmann simulations of N-component immiscible-fluid phase separation (N=2–8) in two-dimensional, thin-film, and three-dimensional geometries, with both equal and tuned interfacial tensions. The central claim is that in two dimensions the four-color theorem imposes a topological constraint that, for N≥4 in the absence of cloaking, suppresses coalescence hydrodynamics, so that coarsening follows a universal diffusion-limited master curve L* = a_N (t*)^{1/3}, with prefactors a_N computed from Marqusee's 2D Ostwald-ripening theory using the volume fraction V_N = 1/N. For N=3, coalescence is claimed to be topologically necessary, and the dynamics show a crossover to diffusive scaling only at late times. In full three dimensions there is no finite-N cutoff, and diffusive scaling is approached only asymptotically; thin films reproduce the 2D master curve once the domain size exceeds the film thickness. Tuning spreading parameters can induce cloaking and heterogeneous component-wise coarsening.

Significance. The empirical master-curve collapse in Fig. 1(b) is a genuine test rather than a fit: the prefactors are obtained from Marqusee's theory with V_N=1/N, and the same master curve is recovered for thin films. The direct coalescence-probability measurements (Fig. 1(c)) and the cloaking examples (Fig. 3) are valuable additions. If the topological mechanism were rigorously established, the paper would open a novel connection between chromatic graph theory and multiphase coarsening, with sharp falsifiable predictions for N=3 versus N≥4. However, the derivation as written rests on a graph-theoretic statement that is false in the form stated, and the four-color theorem alone does not imply that the actual component labeling avoids same-color adjacency. The central causal claim therefore needs substantial revision or reframing before publication.

major comments (3)
  1. [After Fig. 1(d)] The statement 'Only planar maps in which every domain has an even number of neighbours are three-colorable' (ref [47]) is false for general planar maps. For example, a star graph K_{1,3} is planar and 3-colorable with a degree-3 vertex. The even-degree necessary condition holds only for restricted families, such as plane triangulations in which exactly three regions meet at each vertex. The paper neither states this restriction nor shows that the adjacency graph of coarsening fluid domains is such a map. Since this premise is the basis for the claimed topological necessity of coalescence for N=3, the central mechanism is not established. The authors should verify the required condition from their simulation data (e.g., by measuring adjacency graphs and degree parity) or soften the claim to a phenomenological observation.
  2. [Four-color theorem argument, §1 and Fig. 1(d)] The four-color theorem guarantees that every planar map admits some 4-coloring, but it does not imply that the particular assignment of N≥4 component labels to a coarsening configuration avoids adjacent same-color domains. The paper's statement that N≥4 'allows' arrangements that avoid coalescence is an existence statement, not a dynamical derivation. The step from 'a conflict-free coloring exists' to 'the system will realize it after Ostwald ripening' is not argued. This logical gap is independent of the parity issue and affects the sufficiency of the topological explanation.
  3. [Eqs. (9)–(11)] The 'theoretical model for N≥3' is Marqusee's classical 2D Ostwald-ripening theory with the volume fraction set to V_N=1/N; there is no chromatic or topological input in the derivation of a_N. Thus the master-curve collapse demonstrates diffusion-limited coarsening, but it does not by itself explain why hydrodynamics is suppressed. The abstract's phrase 'using chromatic graph theory, we derive a theoretical model' overstates the role of graph theory. The paper should distinguish the empirical collapse from the proposed topological mechanism.
minor comments (5)
  1. [Fig. 1(b) caption] The caption says L*/a_N is plotted for varying numbers of components, but a_N values are only given for N=3–6. Please clarify the normalization used for N=2.
  2. [Reference [47]] The classical theorem that a plane triangulation is 3-colorable if and only if all vertices have even degree is due to Heawood; Steinberg's survey is not the standard source for this result. Please cite the theorem precisely and state the hypotheses.
  3. [Fig. 2(b) methodology] The thin-film length scale is computed from a single slice through the domain. This may bias the structure factor. Please justify the choice or show sensitivity to the slice position.
  4. [Abstract and introduction] The phrase 'N≥4 (absent cloaking)' is used before cloaking is defined. Consider defining the cloaking condition (positive spreading parameter) earlier in the text.
  5. [Author affiliation and text formatting] Minor typographical issues: 'U niversity' in the affiliation line and missing spaces in expressions such as 'Fluid structure forN= 2'.

Circularity Check

0 steps flagged

No significant circularity: master-curve prefactors are computed from Marqusee theory with V_N=1/N and independently compared; topology claim is post-hoc and explicitly qualified.

full rationale

The central quantitative prediction is not circular. The prefactor a_N in the claimed master law L* = a_N (t*)^(1/3) is computed from Marqusee's steady-state Ostwald theory (Eqs. 9-11), with the only input V_N=1/N; the values a_3=0.975, a_4=0.923, a_5=0.887, a_6=0.859 are theory outputs, not fits to the simulated L* data. The subsequent comparison rescales simulation data by L*/a_N without adjusting any parameter, so the collapse of N>=4 data is an independent test. The four-color/even-neighbor argument is presented as a post-hoc explanation of measured coalescence probabilities, and the paper explicitly disclaims sufficiency ('the four-color theorem only provides a necessary but not completely sufficient condition for total hydrodynamic arrest'). The possible mathematical misstatement of the 3-colorability condition (the even-neighbor criterion is correct for plane triangulations, not general planar maps) is a correctness risk about the applicability of an external theorem (ref [47]), not a self-referential derivation. The only author self-citation (LBM textbook [36]) supports numerical implementation, not the scaling law, and is not load-bearing. No equation-level reduction or fitted-parameter-renamed-prediction is present.

Axiom & Free-Parameter Ledger

0 free parameters · 4 axioms · 0 invented entities

Central claim uses no fitted constants; prefactors a_N are computed from Marqusee theory. The main assumptions are mathematical/graph-theoretic (validity of the 3-colorability condition as applied to fluid maps) and physical (triple-junction topology, Marqusee theory in 2D). No new entities are invented.

axioms (4)
  • standard math Four-color theorem: every planar map can be colored with four colors so adjacent regions differ.
    Invoked to assert N≥4 removes topological constraints on planar domain arrangements; cited as [34,35].
  • domain assumption A planar map is three-colorable only if every domain has an even number of neighbours (even-neighbour condition).
    Used to argue N=3 with odd-neighbour domains forces same-component contact and coalescence (paragraph after Fig. 1(d)). As stated this is not true for general maps; it holds for restricted classes such as plane triangulations if the fluid adjacency graph is one.
  • domain assumption Marqusee's steady-state droplet size distribution and screening-length equation describe 2D diffusion-limited Ostwald ripening for N components with V_N=1/N.
    The theoretical prefactors a_N are obtained by solving Eqs. (9)-(11) from ref [46]; the model's applicability to multicomponent systems with equal volume fractions is assumed.
  • domain assumption The LBM scheme with free energy (4) and reduction-consistency terms (6)-(7) reproduces the continuum Navier-Stokes/Cahn-Hilliard equations (1)-(3).
    All numerical results rest on this; implementation details are deferred to Supplemental Section I, unavailable in the reviewed text.

reviewed 2026-08-03 · how reviews work

0 comments
Cite this review

Pith. "Pith review of Topology Controls the Phase Separation Dynamics of Many Component Fluid Mixtures." pith.science (2026). https://pith.science/paper/WNSAKPJB

@misc{pith2026251120215,
  author       = {Pith},
  title        = {Pith review of: Topology Controls the Phase Separation Dynamics of Many Component Fluid Mixtures},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/WNSAKPJB}},
  note         = {Machine review of arXiv:2511.20215}
}
Share X Bluesky LinkedIn Reddit HN
abstract

Fluid mixtures, ranging from the cellular cytoplasm to synthetic DNA nanostar systems, can spontaneously compartmentalize into many ($N$) coexisting liquid phases through liquid-liquid phase separation. While such systems exhibit a remarkable diversity of spatial organizations, the physical principles governing their non-equilibrium dynamics remain poorly understood. Here, combining simulations and analytical theory, we show that the coarsening dynamics of many component phase separation are fundamentally linked to mathematical coloring problems. For planar phase organization, relevant to synthetic droplet monolayers and simple biological structures, we identify distinct topological constraints for $N=2$, $N=3$, and $N=4$, with no further change for $N>4$, consistent with the four-color theorem. These constraints govern the coarsening dynamics, and, using chromatic graph theory, we derive a theoretical model for $N\geq 3$ that quantitatively captures the diffusive-like coarsening. By contrast, classical theories based solely on Ostwald ripening underestimate the observed dynamics. We further show that tuning interfacial tensions modifies the set of admissible phase arrangements, enabling highly heterogeneous coarsening dynamics across different phases. For unconfined systems with nonplanar phase organization, different coloring constraints apply, with no analogue of the four-color theorem, and coalescence suppression emerges only when the number of phases exceeds $N\gtrsim 7$. More broadly, our work establishes coloring theory as a topological framework for understanding and predicting the dynamics of many component phase-separating fluids.

Figures

Figures reproduced from arXiv: 2511.20215 by Halim Kusumaatmaja, Michael Rennick, Xitong Zhang.

Figure 1
Figure 1. Figure 1: FIG. 1. (a) Fluid structure for [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: FIG. 2. (a) [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: FIG. 3. (a) Fluid structure during [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

54 extracted references · 2 linked inside Pith

  1. [1]

    Tadros, P

    T. Tadros, P. Izquierdo, J. Esquena, and C. Solans, Advances in Colloid and Interface Science108-109, 303 (2004)

    2004

  2. [2]

    D. J. McClements, Critical Reviews in Food Science and Nutrition47, 611–649 (2007)

    2007

  3. [3]

    Huang, E

    Y. Huang, E. J. Kramer, A. J. Heeger, and G. C. Bazan, Chemical Reviews114, 7006–7043 (2014)

    2014

  4. [4]

    Nelson, Materials Today14, 462–470 (2011)

    J. Nelson, Materials Today14, 462–470 (2011)

    2011

  5. [5]

    Y. Cai, Q. Li, G. Lu, H. S. Ryu, Y. Li, H. Jin, Z. Chen, Z. Tang, G. Lu, X. Hao, H. Y. Woo, C. Zhang, and Y. Sun, Nature Communications13, 2369 (2022)

    2022

  6. [6]

    Allabar and M

    A. Allabar and M. Nowak, Earth and Planetary Sci- ence Letters501, 192 (2018)

    2018

  7. [7]

    Sahagian and T

    D. Sahagian and T. L. Carley, Geochemistry, Geo- physics, Geosystems21, e2019GC008898 (2020)

    2020

  8. [8]

    A. J. Parnell, A. L. Washington, O. O. Mykhaylyk, C. J. Hill, A. Bianco, S. L. Burg, A. J. C. Dennison, M. Snape, A. J. Cadby, A. Smith, S. Prevost, D. M. Whittaker, R. A. L. Jones, J. P. A. Fairclough, and A. R. Parker, Scientific Reports5, 18317 (2015)

    2015

  9. [9]

    Saranathan, S

    V. Saranathan, S. Narayanan, A. Sandy, E. R. Dufresne, and R. O. Prum, Proceedings of the Na- tional Academy of Sciences118, e2101357118 (2021)

    2021

  10. [10]

    Narasimhan, R

    V. Narasimhan, R. H. Siddique, J. O. Lee, S. Ku- mar, B. Ndjamen, J. Du, N. Hong, D. Sretavan, and H. Choo, Nature Nanotechnology13, 512 (2018)

    2018

  11. [11]

    Chung and R

    H.-J. Chung and R. J. Composto, Physical Review Letters92, 185704 (2004)

    2004

  12. [12]

    Furukawa, Advances in Physics34, 703 (1985)

    H. Furukawa, Advances in Physics34, 703 (1985)

    1985

  13. [13]

    A. J. Wagner and M. E. Cates, Europhysics Letters 56, 556 (2001)

    2001

  14. [14]

    A. J. Wagner and J. M. Yeomans, Physical Review Letters80, 1429 (1998)

    1998

  15. [15]

    R. A. L. Jones, L. J. Norton, E. J. Kramer, F. S. Bates, and P. Wiltzius, Physical Review Letters66, 1326–1329 (1991)

    1991

  16. [16]

    Tanaka, Physical Review Letters70, 53 (1993)

    H. Tanaka, Physical Review Letters70, 53 (1993)

    1993

  17. [17]

    Shin and C

    Y. Shin and C. P. Brangwynne, Science357, eaaf4382 (2017)

    2017

  18. [18]

    J.-M. Choi, A. S. Holehouse, and R. V. Pappu, Annual Review of Biophysics49, 107 (2020)

    2020

  19. [19]

    Berry, C

    J. Berry, C. P. Brangwynne, and M. Haataja, Reports on Progress in Physics81, 046601 (2018)

    2018

  20. [20]

    N. A. Erkamp, T. Sneideris, H. Ausserw¨ oger, D. Qian, S. Qamar, J. Nixon-Abell, P. St George-Hyslop, J. D. Schmit, D. A. Weitz, and T. P. J. Knowles, Nature Communications14, 684 (2023)

    2023

  21. [21]

    Mehta and J

    S. Mehta and J. Zhang, Nature Reviews Cancer22, 239 (2022)

    2022

  22. [22]

    Zwicker and L

    D. Zwicker and L. Laan, Proceedings of the National Academy of Sciences119, e2201250119 (2022)

    2022

  23. [23]

    W. M. Jacobs and D. Frenkel, The Journal of Chem- ical Physics139, 024108 (2013)

    2013

  24. [24]

    W. M. Jacobs and D. Frenkel, Biophysical Journal 112, 683 (2017)

    2017

  25. [25]

    D. W. Sanders, N. Kedersha, D. S. Lee, A. R. Strom, V. Drake, J. A. Riback, D. Bracha, J. M. Eeftens, A. Iwanicki, A. Wang, M.-T. Wei, G. Whitney, S. M. Lyons, P. Anderson, W. M. Jacobs, P. Ivanov, and C. P. Brangwynne, Cell181, 306 (2020)

    2020

  26. [26]

    A. S. Chaderjian, S. Wilken, and O. A. Saleh, arXiv:2508.18574 (2025)

    Pith/arXiv arXiv 2025

  27. [27]

    G. R. Abraham, A. S. Chaderjian, A. B. N Nguyen, S. Wilken, and O. A. Saleh, Reports on Progress in Physics87, 066601 (2024)

    2024

  28. [28]

    W. M. Jacobs, Physical Review Letters126, 258101 (2021)

    2021

  29. [29]

    R. B. Teixeira, G. Carugno, I. Neri, and P. Sartori, Proceedings of the National Academy of Sciences121, e2320504121 (2024)

    2024

  30. [30]

    R. B. Teixeira, I. Neri, and P. Sartori, arXiv:2509.10705 (2025)

    Pith/arXiv arXiv 2025

  31. [31]

    Shrinivas and M

    K. Shrinivas and M. P. Brenner, Proceedings of the National Academy of Sciences118, e2108551118 (2021)

    2021

  32. [32]

    S. Mao, M. S. Chakraverti-Wuerthwein, H. Gaudio, and A. Koˇ smrlj, Physical Review Letters125, 218003 (2020)

    2020

  33. [33]

    S. Mao, D. Kuldinow, M. P. Haataja, and A. Koˇ smrlj, Soft Matter15, 1297 (2019)

    2019

  34. [34]

    The four-color theorem states that any planar map can be colored with four colors so that no adjacent regions share a color

  35. [35]

    Robertson, D

    N. Robertson, D. Sanders, P. Seymour, and R. Thomas, Journal of Combinatorial Theory, Series B70, 2 (1997)

    1997

  36. [36]

    Krueger, H

    T. Krueger, H. Kusumaatmaja, A. Kuzmin, O. Shardt, G. Silva, and E. Viggen,The Lattice Boltz- mann Method: Principles and Practice, Graduate Texts in Physics (Springer, 2016)

    2016

  37. [37]

    Zheng, S

    L. Zheng, S. Zheng, and Q. Zhai, Physica A: Sta- tistical Mechanics and its Applications574, 126015 (2021)

    2021

  38. [38]

    X. Yuan, B. Shi, C. Zhan, and Z. Chai, Physics of Fluids34, 023311 (2022)

    2022

  39. [39]

    L. Ju, Z. Guo, B. Yan, and S. Sun, Physical Review E109, 045307 (2024)

    2024

  40. [40]

    Z. Guo, C. Zheng, and B. Shi, Physical Review E65, 046308 (2002)

    2002

  41. [41]

    Boyer and S

    F. Boyer and S. Minjeaud, Mathematical Models and Methods in Applied Sciences24, 2885 (2014). 6

    2014

  42. [42]

    Dong, Journal of Computational Physics361, 1 (2018)

    S. Dong, Journal of Computational Physics361, 1 (2018)

    2018

  43. [43]

    See Supplemental Material at URL-will-be-inserted- by-publisher for (i) specific details of lattice Boltz- mann algorithm, (ii) discussion of the approach to ensure reduction consistency, (iii) description of the method to measure coalescence events, (iv) unnor- malised data from Fig. 1(b), (v) thin film results with varying thicknesses, (vi) study of arr...

  44. [44]

    V. M. Kendon, M. E. Cates, I. Pagonabarraga, J.-C. Desplat, and P. Bladon, Journal of Fluid Mechanics 440, 147–203 (2001)

    2001

  45. [45]

    Lifshitz and V

    I. Lifshitz and V. Slyozov, Journal of Physics and Chemistry of Solids19, 35 (1961)

    1961

  46. [46]

    J. A. Marqusee, The Journal of Chemical Physics81, 976 (1984)

    1984

  47. [47]

    Steinberg, inQuo Vadis, Graph Theory?, Annals of Discrete Mathematics, Vol

    R. Steinberg, inQuo Vadis, Graph Theory?, Annals of Discrete Mathematics, Vol. 55, edited by J. Gimbel, J. W. Kennedy, and L. V. Quintas (Elsevier, 1993) pp. 211–248

    1993

  48. [48]

    Mai and A

    Y. Mai and A. Eisenberg, Chemical Society Reviews 41, 5969 (2012)

    2012

  49. [49]

    C. Lang, E. C. Lloyd, K. E. Matuszewski, Y. Xu, V. Ganesan, R. Huang, M. Kumar, and R. J. Hickey, Nature Nanotechnology17, 752 (2022)

    2022

  50. [50]

    Peinemann, V

    K.-V. Peinemann, V. Abetz, and P. F. W. Simon, Nature Materials6, 992 (2007)

    2007

  51. [51]

    Agudo-Canalejo, S

    J. Agudo-Canalejo, S. W. Schultz, H. Chino, S. M. Migliano, C. Saito, I. Koyama-Honda, H. Stenmark, A. Brech, A. I. May, N. Mizushima, and R. L. Knorr, Nature591, 142 (2021)

    2021

  52. [52]

    Mannattil, H

    M. Mannattil, H. Diamant, and D. Andelman, Phys- ical Review Letters135, 108101 (2025)

    2025

  53. [53]

    Tanaka, Communications Physics5, 167 (2022)

    H. Tanaka, Communications Physics5, 167 (2022)

    2022

  54. [140]

    This work used the Cirrus UK National Tier-2 HPC Service at EPCC

    and UKRI Engineering and Physical Sciences Re- search Council (EP/V034154/2). This work used the Cirrus UK National Tier-2 HPC Service at EPCC. We also acknowledge compute resources on ARCHER2 via the UK Consortium on Mesoscale Engineering Sci- ences (EP/L00030X/1). ∗ halim.kusumaatmaja@ed.ac.uk

This paper was first reviewed by deepseek-v4-flash on August 3, 2026.