Flux-explicit Cheeger bounds for magnetic Laplacians on compact metric graphs

Bj\"orn Dahlke

arxiv: 2606.12453 · v1 · pith:QL2WIGY3new · submitted 2026-06-05 · 🧮 math-ph · math.MP· math.SP

Flux-explicit Cheeger bounds for magnetic Laplacians on compact metric graphs

Bj\"orn Dahlke This is my paper

Pith reviewed 2026-06-27 20:30 UTC · model grok-4.3

classification 🧮 math-ph math.MPmath.SP

keywords magnetic LaplacianCheeger constantmetric graphsfrustration indexAharonov-Bohm fluxspectral lower boundheat semigroup decay

0 comments

The pith

The magnetic Cheeger constant equals the ℓ¹ distance of the global cycle flux vector from the integral flux lattice.

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

The paper shows that for a magnetic potential on a compact metric graph, the frustration index measuring how much the potential twists phases on subgraphs reduces exactly to an ℓ¹ distance in the space of cycle fluxes. Adding this to the standard Cheeger definition produces an explicit lower bound on the lowest eigenvalue of the magnetic Laplacian that depends only on how far the total fluxes are from integer multiples. This bound also controls the decay of the magnetic heat semigroup in L². A reader cares because it turns an abstract optimization over subgraphs into a direct calculation from the periods of the potential around cycles, making the effect of the magnetic field transparent and computable.

Core claim

Defining a magnetic Cheeger constant by augmenting the usual boundary term with the frustration index of the potential on subgraphs, the paper proves that this index is precisely the ℓ¹ distance determined by the periods of A on cycles. Consequently the Cheeger constant equals the distance of the global cycle flux vector from the integral flux lattice, yielding an explicit lower bound for the bottom of the spectrum of H_A together with L² decay bounds for the magnetic heat semigroup.

What carries the argument

The magnetic Cheeger constant, which augments the standard Cheeger expression with the frustration index of the magnetic potential on subgraphs and equals the ℓ¹ distance of the flux vector to the integer lattice.

If this is right

The bottom of the spectrum of H_A is bounded below by a positive quantity determined by the flux distance.
The magnetic heat semigroup satisfies an L² decay bound with rate given by that quantity.
The magnetic energy functional decays at a corresponding rate.
The bound depends explicitly on the Aharonov-Bohm fluxes and requires no further optimization over subgraphs.

Where Pith is reading between the lines

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

If the equality holds, numerical evaluation of spectral bounds reduces to computing cycle integrals of A rather than searching over all possible subgraphs.
The approach may generalize to other differential operators on graphs where frustration can be localized to topological cycles.
Similar flux-distance reductions could appear in discrete magnetic Laplacians or in quantum graphs with different boundary conditions.

Load-bearing premise

The frustration index of the magnetic potential on any subgraph equals exactly the ℓ¹ distance in flux space determined by the periods of A on the cycles of that subgraph.

What would settle it

Finding a metric graph and magnetic potential where the minimal frustration index over subgraphs differs numerically from the ℓ¹ distance of the cycle flux vector to the nearest integer vector would disprove the claimed equality.

read the original abstract

Let (\Gamma) be a finite compact connected metric graph and let (A\in L^\infty(\Gamma)) be a real magnetic potential. The magnetic Laplacian (H_A) with standard vertex conditions is defined by the closed quadratic form [ q_A[u]=\sum_e\int_e |(-i\partial_x-A_e)u_e|^2,dx. ] A magnetic Cheeger constant is introduced by adding to the usual boundary term the frustration index of the potential on subgraphs. The first point of the paper is that, on a metric graph, this frustration index is exactly a finite dimensional (\ell^1) flux distance determined by the periods of (A) on cycles. Consequently the Cheeger constant can be written directly in terms of Aharonov Bohm fluxes. We prove a Cheeger type lower bound for the bottom of the spectrum and derive the corresponding explicit lower estimate in terms of the distance of the global cycle flux vector from the integral flux lattice. The estimate also gives (L^2) decay bounds for the magnetic heat semigroup and for the magnetic energy. The constants are not asserted to be sharp; the emphasis is on the flux dependence and on the self contained metric graph formulation.

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 reduces the magnetic Cheeger constant to an explicit l1 distance of cycle fluxes to the integer lattice and derives a lower bound from it.

read the letter

The core move here is turning the frustration index of the magnetic potential into an l1 distance in flux space on the cycle basis, so the Cheeger constant becomes the distance of the global flux vector to the integer lattice. That gives a direct lower bound on the ground state energy of H_A and some L2 decay for the semigroup. The abstract presents this as a clean identification from the quadratic form and the periods of A.

What is actually new is the explicit flux-organized form of the bound. Standard Cheeger results on graphs do not have this magnetic version written out in terms of Aharonov-Bohm fluxes, so the reduction itself is the contribution. The paper stays inside the metric-graph setting and keeps the constants explicit rather than asymptotic, which is useful for anyone who needs to track the dependence on A.

The load-bearing step is the claim that the infimum of the frustration index over subgraphs equals exactly the global l1 distance. The stress-test note flags the possibility that dependent cycles could let a proper subgraph achieve a strictly smaller value by partial cancellation. The abstract treats the equality as immediate after the definition, so the proof must show why the infimum cannot drop below the global quantity. If that argument is complete and handles the linear dependence correctly, the rest follows without further issues. No other gaps are visible from the given material.

This is for people already working on spectral problems for magnetic Laplacians on graphs. A reader in that area gets a usable explicit estimate and the semigroup bounds. It is narrow enough that it will not interest a broad audience, but the result is concrete enough to warrant referee time. I would send it out for review and ask specifically for a clear write-up of the subgraph-infimum step.

Referee Report

2 major / 2 minor

Summary. The paper defines a magnetic Cheeger constant on a finite compact connected metric graph by augmenting the standard Cheeger ratio with the frustration index of a real magnetic potential A. It asserts that this frustration index on any subgraph equals the ℓ¹ distance of the vector of periods of A around a cycle basis to the integer lattice. Consequently the magnetic Cheeger constant is expressed directly in terms of Aharonov-Bohm fluxes, yielding an explicit lower bound on the bottom of the spectrum of the magnetic Laplacian H_A together with L² decay estimates for the associated heat semigroup.

Significance. If the central identification holds, the result supplies the first flux-explicit, parameter-free Cheeger bound for magnetic Laplacians on metric graphs. The self-contained formulation, the reduction to a finite-dimensional ℓ¹ distance, and the explicit semigroup decay estimates are genuine strengths that would be of interest to spectral geometers working on graphs.

major comments (2)

[Section 3 (or the section containing the identification of frustration index with ℓ¹ flux distance)] The proof that the infimum of the frustration index over all subgraphs equals the global ℓ¹ distance of the cycle-flux vector to ℤ^k (the step labeled 'consequently' in the abstract) is load-bearing for every subsequent claim. The argument must explicitly rule out the possibility that, on graphs with linearly dependent cycles, a proper subgraph could achieve a strictly smaller frustration index by partial cancellation of periods. Please supply the precise statement and proof of this equality, including the choice of cycle basis and the handling of dependent cycles.
[Theorem 4.1 (or the theorem stating the spectral lower bound)] The Cheeger-type lower bound for λ₁(H_A) is stated in terms of the magnetic Cheeger constant; the constant in front of this bound must be verified to be independent of the magnetic potential A. If the constant depends on A through the choice of test functions or through the metric, the claimed flux-explicit character is weakened.

minor comments (2)

[Abstract and §2] Notation for the cycle basis and the period map should be introduced once and used consistently; the current abstract uses both 'periods of A on cycles' and 'global cycle flux vector' without cross-reference.
[Introduction] The statement that the constants are 'not asserted to be sharp' is appropriate, but a brief remark on whether equality can be attained on a cycle graph would help the reader.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the careful reading and the constructive major comments. We respond point-by-point below. Both concerns can be addressed by expanding the relevant proofs and adding explicit remarks; we will revise the manuscript accordingly.

read point-by-point responses

Referee: [Section 3 (or the section containing the identification of frustration index with ℓ¹ flux distance)] The proof that the infimum of the frustration index over all subgraphs equals the global ℓ¹ distance of the cycle-flux vector to ℤ^k (the step labeled 'consequently' in the abstract) is load-bearing for every subsequent claim. The argument must explicitly rule out the possibility that, on graphs with linearly dependent cycles, a proper subgraph could achieve a strictly smaller frustration index by partial cancellation of periods. Please supply the precise statement and proof of this equality, including the choice of cycle basis and the handling of dependent cycles.

Authors: We agree that a fully explicit argument is needed. The frustration index on any subgraph equals the ℓ¹ distance of its restricted period vector to ℤ^{k'}, where the periods are obtained by integrating A along a cycle basis of the subgraph. Because A is a globally defined 1-form, the periods on any collection of cycles (including those of a proper subgraph) are integer-linear combinations of the periods on a fixed fundamental cycle basis of the whole graph. Consequently the minimal ℓ¹ distance over all possible subgraphs cannot be strictly smaller than the global distance: any apparent partial cancellation on a subgraph corresponds to choosing a different integer vector in the ambient lattice, whose distance is already bounded below by the global minimum. We will insert a new lemma in Section 3 that states the equality precisely, fixes a fundamental cycle basis, and proves the claim by exhibiting the linear dependence relations explicitly. This revision will also contain a short paragraph ruling out the cancellation scenario the referee describes. revision: yes
Referee: [Theorem 4.1 (or the theorem stating the spectral lower bound)] The Cheeger-type lower bound for λ₁(H_A) is stated in terms of the magnetic Cheeger constant; the constant in front of this bound must be verified to be independent of the magnetic potential A. If the constant depends on A through the choice of test functions or through the metric, the claimed flux-explicit character is weakened.

Authors: The prefactor in the Cheeger inequality of Theorem 4.1 is the standard combinatorial constant arising from the co-area formula and the isoperimetric inequality on the underlying metric graph; it depends only on the total length and the vertex degrees and is therefore independent of A. The test functions used in the proof are the characteristic functions of the Cheeger sets (or their smoothed versions), whose construction does not involve A. The magnetic contribution appears solely inside the definition of the magnetic Cheeger constant itself. We will add a short paragraph immediately after the statement of Theorem 4.1 that records this independence and recalls the origin of the numerical prefactor. revision: yes

Circularity Check

0 steps flagged

No circularity; derivation from quadratic form to flux-explicit Cheeger bound is self-contained

full rationale

The paper defines the magnetic Laplacian via the quadratic form q_A and introduces a magnetic Cheeger constant that augments the standard boundary term with a frustration index on subgraphs. It then states that this index equals the ℓ¹ distance of the global cycle flux vector to the integer lattice, allowing the constant and the resulting spectral lower bound to be written explicitly in terms of Aharonov-Bohm periods. This identification is presented as a proved property of metric graphs rather than an input assumption, a fitted parameter, or a self-citation; no equation reduces the claimed equality to a prior definition or to the target spectral quantity by construction. The subsequent Cheeger-type estimate and L² decay bounds therefore rest on an independent derivation step rather than on renaming or circular reuse of the input data.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 2 invented entities

The paper rests on standard domain assumptions for metric graphs and quadratic forms; it introduces the magnetic Cheeger constant and the identification of its frustration index with flux distance as new constructs without external independent evidence.

axioms (2)

domain assumption Γ is a finite compact connected metric graph equipped with standard vertex conditions.
Stated at the opening of the abstract as the setting for the magnetic Laplacian H_A.
domain assumption A belongs to L^∞(Γ) and is real-valued.
Given in the definition of the magnetic potential.

invented entities (2)

magnetic Cheeger constant no independent evidence
purpose: To extend the classical Cheeger constant by incorporating the frustration index of the magnetic potential.
Introduced in the abstract to obtain the flux-explicit bound.
frustration index no independent evidence
purpose: To quantify the additional cost imposed by the magnetic potential on subgraphs.
Defined by adding the index to the usual boundary term; identified with ℓ¹ flux distance.

pith-pipeline@v0.9.1-grok · 5741 in / 1459 out tokens · 25032 ms · 2026-06-27T20:30:51.275925+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

14 extracted references · 9 canonical work pages

[1]

Berkolaiko, J

G. Berkolaiko, J. B. Kennedy, P. Kurasov and D. Mugnolo, Edge connectivity and the spectral gap of combinatorial and quantum graphs,J. Phys. A: Math. Theor.50 (2017), 365201, 29 pp. DOI: 10.1088/1751-8121/aa8125

work page doi:10.1088/1751-8121/aa8125 2017
[2]

Berkolaiko and P

G. Berkolaiko and P. Kuchment,Introduction to Quantum Graphs, Mathematical Surveys and Monographs, vol. 186, American Mathematical Society, Providence, RI,
[3]

DOI: 10.1090/surv/186

work page doi:10.1090/surv/186
[4]

Berkolaiko and T

G. Berkolaiko and T. Weyand, Stability of eigenvalues of quantum graphs with respect to magnetic perturbation and the nodal count of the eigenfunctions,Philos. Trans. R. Soc. A372(2014), 20120522, 17 pp. DOI: 10.1098/rsta.2012.0522

work page doi:10.1098/rsta.2012.0522 2014
[5]

Colin de Verdiere, N

Y. Colin de Verdiere, N. Torki-Hamza and F. Truc, Essential self-adjointness for combinatorial Schrodinger operators III: magnetic fields,Ann. Fac. Sci. Toulouse Math.(6)20(2011), 599–611

2011
[6]

J. B. Kennedy and D. Mugnolo, The Cheeger constant of a quantum graph,PAMM Proc. Appl. Math. Mech.16(2016), 875–876. DOI: 10.1002/pamm.201610426

work page doi:10.1002/pamm.201610426 2016
[7]

Kuchment, Quantum graphs: an introduction and a brief survey, inAnalysis on Graphs and its Applications, Proc

P. Kuchment, Quantum graphs: an introduction and a brief survey, inAnalysis on Graphs and its Applications, Proc. Sympos. Pure Math., vol. 77, American Mathemat- ical Society, Providence, RI, 2008, pp. 291–312. DOI: 10.1090/pspum/077/2459870

work page doi:10.1090/pspum/077/2459870 2008
[8]

Kurasov, On the spectral gap for Laplacians on metric graphs,Acta Phys

P. Kurasov, On the spectral gap for Laplacians on metric graphs,Acta Phys. Polon. A124(2013), 1060–1062. DOI: 10.12693/APhysPolA.124.1060

work page doi:10.12693/aphyspola.124.1060 2013
[9]

Frustration index and Cheeger inequalities for discrete and continuous magnetic Laplacians

C. Lange, S. Liu, N. Peyerimhoff and O. Post, Frustration index and Cheeger inequal- ities for discrete and continuous magnetic Laplacians,Calc. Var. Partial Differential Equations54(2015), 4165–4196. DOI: 10.1007/s00526-015-0935-x

work page doi:10.1007/s00526-015-0935-x 2015
[10]

Nicaise, Spectre des reseaux topologiques finis,Bull

S. Nicaise, Spectre des reseaux topologiques finis,Bull. Sci. Math.(2)111(1987), 401–413

1987
[11]

Post, Spectral analysis of metric graphs and related spaces, inLimits of Graphs in Group Theory and Computer Science, G

O. Post, Spectral analysis of metric graphs and related spaces, inLimits of Graphs in Group Theory and Computer Science, G. Arzhantseva and A. Valette (eds.), EPFL Press, Lausanne, 2009, pp. 109–140

2009
[12]

Shigekawa, Eigenvalue problems for the Schrodinger operator with the magnetic field on a compact Riemannian manifold,J

I. Shigekawa, Eigenvalue problems for the Schrodinger operator with the magnetic field on a compact Riemannian manifold,J. Funct. Anal.75(1987), 92–127. DOI: 10.1016/0022-1236(87)90108-X

work page doi:10.1016/0022-1236(87)90108-x 1987
[13]

M. A. Shubin, Discrete magnetic Laplacian,Commun. Math. Phys.164(1994), 259–275. DOI: 10.1007/BF02101702

work page doi:10.1007/bf02101702 1994
[14]

Sunada, A discrete analogue of periodic magnetic Schrodinger operators, inGe- ometry of the Spectrum, Contemp

T. Sunada, A discrete analogue of periodic magnetic Schrodinger operators, inGe- ometry of the Spectrum, Contemp. Math., vol. 173, American Mathematical Society, Providence, RI, 1994, pp. 283–299. 10

1994

[1] [1]

Berkolaiko, J

G. Berkolaiko, J. B. Kennedy, P. Kurasov and D. Mugnolo, Edge connectivity and the spectral gap of combinatorial and quantum graphs,J. Phys. A: Math. Theor.50 (2017), 365201, 29 pp. DOI: 10.1088/1751-8121/aa8125

work page doi:10.1088/1751-8121/aa8125 2017

[2] [2]

Berkolaiko and P

G. Berkolaiko and P. Kuchment,Introduction to Quantum Graphs, Mathematical Surveys and Monographs, vol. 186, American Mathematical Society, Providence, RI,

[3] [3]

DOI: 10.1090/surv/186

work page doi:10.1090/surv/186

[4] [4]

Berkolaiko and T

G. Berkolaiko and T. Weyand, Stability of eigenvalues of quantum graphs with respect to magnetic perturbation and the nodal count of the eigenfunctions,Philos. Trans. R. Soc. A372(2014), 20120522, 17 pp. DOI: 10.1098/rsta.2012.0522

work page doi:10.1098/rsta.2012.0522 2014

[5] [5]

Colin de Verdiere, N

Y. Colin de Verdiere, N. Torki-Hamza and F. Truc, Essential self-adjointness for combinatorial Schrodinger operators III: magnetic fields,Ann. Fac. Sci. Toulouse Math.(6)20(2011), 599–611

2011

[6] [6]

J. B. Kennedy and D. Mugnolo, The Cheeger constant of a quantum graph,PAMM Proc. Appl. Math. Mech.16(2016), 875–876. DOI: 10.1002/pamm.201610426

work page doi:10.1002/pamm.201610426 2016

[7] [7]

Kuchment, Quantum graphs: an introduction and a brief survey, inAnalysis on Graphs and its Applications, Proc

P. Kuchment, Quantum graphs: an introduction and a brief survey, inAnalysis on Graphs and its Applications, Proc. Sympos. Pure Math., vol. 77, American Mathemat- ical Society, Providence, RI, 2008, pp. 291–312. DOI: 10.1090/pspum/077/2459870

work page doi:10.1090/pspum/077/2459870 2008

[8] [8]

Kurasov, On the spectral gap for Laplacians on metric graphs,Acta Phys

P. Kurasov, On the spectral gap for Laplacians on metric graphs,Acta Phys. Polon. A124(2013), 1060–1062. DOI: 10.12693/APhysPolA.124.1060

work page doi:10.12693/aphyspola.124.1060 2013

[9] [9]

Frustration index and Cheeger inequalities for discrete and continuous magnetic Laplacians

C. Lange, S. Liu, N. Peyerimhoff and O. Post, Frustration index and Cheeger inequal- ities for discrete and continuous magnetic Laplacians,Calc. Var. Partial Differential Equations54(2015), 4165–4196. DOI: 10.1007/s00526-015-0935-x

work page doi:10.1007/s00526-015-0935-x 2015

[10] [10]

Nicaise, Spectre des reseaux topologiques finis,Bull

S. Nicaise, Spectre des reseaux topologiques finis,Bull. Sci. Math.(2)111(1987), 401–413

1987

[11] [11]

Post, Spectral analysis of metric graphs and related spaces, inLimits of Graphs in Group Theory and Computer Science, G

O. Post, Spectral analysis of metric graphs and related spaces, inLimits of Graphs in Group Theory and Computer Science, G. Arzhantseva and A. Valette (eds.), EPFL Press, Lausanne, 2009, pp. 109–140

2009

[12] [12]

Shigekawa, Eigenvalue problems for the Schrodinger operator with the magnetic field on a compact Riemannian manifold,J

I. Shigekawa, Eigenvalue problems for the Schrodinger operator with the magnetic field on a compact Riemannian manifold,J. Funct. Anal.75(1987), 92–127. DOI: 10.1016/0022-1236(87)90108-X

work page doi:10.1016/0022-1236(87)90108-x 1987

[13] [13]

M. A. Shubin, Discrete magnetic Laplacian,Commun. Math. Phys.164(1994), 259–275. DOI: 10.1007/BF02101702

work page doi:10.1007/bf02101702 1994

[14] [14]

Sunada, A discrete analogue of periodic magnetic Schrodinger operators, inGe- ometry of the Spectrum, Contemp

T. Sunada, A discrete analogue of periodic magnetic Schrodinger operators, inGe- ometry of the Spectrum, Contemp. Math., vol. 173, American Mathematical Society, Providence, RI, 1994, pp. 283–299. 10

1994