arxiv: 2604.22248 · v1 · submitted 2026-04-24 · ✦ hep-th · hep-ph

Recognition: unknown

Vacuum structure of a scalar field on a torus with uniform magnetic flux

Mayumi Akamatsu , Hiroki Imai , Makoto Sakamoto , Maki Takeuchi

Authors on Pith no claims yet

Pith reviewed 2026-05-08 10:56 UTC · model grok-4.3

classification ✦ hep-th hep-ph

keywords complex scalar fieldtwo-dimensional torusquantized magnetic fluxvacuum expectation valuecritical arealowest-mode approximationsymmetry propertiesspontaneous symmetry breaking

0 comments

The pith

A complex scalar field on a torus with magnetic flux develops a nonzero, position-dependent vacuum expectation value only above a critical torus area.

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

The paper studies a complex scalar field on a two-dimensional torus threaded by quantized magnetic flux M. It establishes that a critical torus area exists: below this area the vacuum expectation value is zero, but above it the expectation value turns on and must vary with position on the torus. The authors employ the lowest-mode approximation to count the degenerate vacuum states, finding one configuration for M=1, two for M=2, and six for M=3. They further examine which symmetries of the torus and flux are preserved or spontaneously broken by these choices. The results clarify vacuum selection in a compact space with magnetic background.

Core claim

When the area of the torus exceeds a critical value, the vacuum expectation value of the complex scalar field becomes nonvanishing. Any nonzero vacuum expectation value necessarily exhibits nontrivial dependence on the coordinates of the torus. In the lowest-mode approximation a single vacuum configuration appears for M=1, while two and six degenerate vacuum configurations arise for M=2 and M=3, respectively, each with distinct symmetry properties under the underlying torus and flux symmetries.

What carries the argument

lowest-mode approximation for the scalar field modes in the uniform magnetic flux background on the torus, which reduces the vacuum search to minimizing an effective potential for the lowest modes.

If this is right

For M=1 the vacuum is unique.
For M=2 the vacuum has two-fold degeneracy.
For M=3 the vacuum has six-fold degeneracy.
Any nonzero vacuum expectation value breaks translation invariance on the torus.
The chosen vacua may preserve or spontaneously break the discrete symmetries of the magnetized torus.

Where Pith is reading between the lines

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

The position dependence of the vacuum expectation value could source additional effective potentials for moduli fields if the torus size is allowed to vary.
The pattern of degeneracy for successive M may generalize to higher flux quanta and suggest a counting rule tied to the flux.
Analogous critical-area phenomena might appear in lattice models of scalar fields coupled to U(1) gauge fields with background flux.

Load-bearing premise

The lowest-mode approximation is sufficient to determine the vacuum configurations, their degeneracy, and their symmetry properties for the values of M considered.

What would settle it

Evaluating the full effective potential including higher modes and checking whether a nonzero position-dependent vacuum expectation value appears exactly when the torus area surpasses the reported critical value would test the result.

Figures

Figures reproduced from arXiv: 2604.22248 by Hiroki Imai, Maki Takeuchi, Makoto Sakamoto, Mayumi Akamatsu.

**Figure 1.** Figure 1: A two-dimensional torus 2.1 Magnetized torus We define a two-dimensional torus T 2 as the quotient space R 2/Λ, where Λ ≡ {n1e1+n2e2|n1, n2 ∈ Z} is the lattice in R 2 generated by linearly independent basis vectors ei (i = 1, 2) given by e1 = (L, 0), e2 = (L Reτ, LImτ ), (2.1) where L denotes the overall length scale of T 2 (see view at source ↗

**Figure 2.** Figure 2: Critical area 3 Critical area In this section, we show the existence of a critical area above which the vacuum expectation value hϕ(z)i becomes nonvanishing. The vacuum expectation value hϕ(z)i is determined by minimizing the functional Ve[ϕ]. Substituting the mode expansion (2.18) into the potential (2.14), we obtain Ve[ϕ] = X∞ n=0 M X−1 j=0 4πM Imτ n + 1 2 − µ ′2 view at source ↗

**Figure 3.** Figure 3: The zero point of the vacuum solution hϕ (M=1)(z)i This configuration has a single zero determined by ϑ " 0 0 # (z, τ ) = 0, (5.28) which is located at z = 1 2 + 1 2 τ, (5.29) as depicted in view at source ↗

**Figure 4.** Figure 4: The zero points of the vacuum solution ϕ (M=2) ± (z) Each vacuum configuration has two zeros, as shown in view at source ↗

**Figure 5.** Figure 5: The zero points of the vacuum solution hϕ (M=3)(z)i Thus, the two vacuum configurations ϕ (I) vac(z) and ϕ (II) vac (z) are connected each other by the broken symmetry transformations belonging to the coset G/H, and hence they are physically equivalent. 5.2.3 M = 3 We finally analyze the symmetry properties of the vacuum configurations for M = 3 and τ = i. From Eq.(5.25), the discrete symmetry group G of t… view at source ↗

read the original abstract

We investigate the vacuum expectation value of a complex scalar field on a two-dimensional torus with quantized magnetic flux $M$. A characteristic feature of this system is the emergence of a critical area: when the area of the torus exceeds this critical value, the vacuum expectation value becomes nonvanishing. Furthermore, any nonzero vacuum expectation value necessarily exhibits nontrivial dependence on the coordinates of the torus. Employing the lowest-mode approximation, we find a single vacuum configuration for $M=1$, whereas two and six degenerate vacuum configurations arise for $M=2$ and $M=3$, respectively. We then analyze the symmetry properties of these vacuum configurations and determine whether they preserve or spontaneously break the symmetry of the underlying system.

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

This paper works out explicit vacuum degeneracies and coordinate-dependent VEVs for a scalar on a fluxed torus using the lowest-mode approximation, but leaves the truncation error unquantified.

read the letter

The main thing to know is that the authors compute a critical torus area above which a complex scalar develops a nonzero VEV that must vary with position, and they count the resulting degenerate vacua as 1 for M=1, 2 for M=2, and 6 for M=3, then classify which symmetries those vacua preserve or break. That counting and the symmetry analysis are the concrete new pieces; they are not just restating prior torus Landau-level results but applying them to the vacuum problem with explicit small-M cases. The calculation itself is a standard effective-potential minimization inside the lowest Landau level, done cleanly enough that the degeneracy pattern follows directly from the symmetry of the flux and the torus lattice. Credit is due for keeping the presentation straightforward and for stating the approximation up front rather than burying it. The limitation is the one the stress-test note flags. All the reported features—onset of the VEV, its coordinate dependence, and the exact degeneracy numbers—are read off from the truncated energy functional. No bound on the contribution of higher modes appears, nor is there a comparison of the effective potential with the next level kept or a check of stability against mode mixing right at the critical area where the VEV is parametrically small. If higher modes mix appreciably they could lift some of the reported degeneracies or smooth out the coordinate dependence, so the strongest claims rest on an assumption whose error is not estimated. This is a niche but honest calculation in the subfield of fields on compact manifolds with background flux. A reader who needs a worked example with small integer M and explicit symmetry breaking patterns will find it useful; someone looking for a general reorganization of the subject will not. The work is coherent on its own terms and the math is reproducible in principle, so it is worth a serious referee even though the approximation needs tightening. I would send it to peer review rather than desk-reject.

Referee Report

1 major / 0 minor

Summary. The paper investigates the vacuum expectation value of a complex scalar field on a two-dimensional torus with quantized magnetic flux M. It reports a critical area above which the VEV becomes nonvanishing and necessarily exhibits nontrivial coordinate dependence on the torus. Employing the lowest-mode approximation, the work finds a single vacuum for M=1, two degenerate vacua for M=2, and six degenerate vacua for M=3, then analyzes whether these configurations preserve or spontaneously break the underlying symmetries.

Significance. If the results are robust beyond the truncation, the identification of a critical area and the mandatory coordinate dependence of any nonzero VEV would provide a concrete example of how magnetic flux on a compact space enforces nontrivial vacuum structure and controls degeneracy patterns. This could be relevant to effective descriptions in QFT on tori or Landau-level physics.

major comments (1)

[lowest-mode approximation (as stated in the abstract and employed for all quantitative results)] The central claims—the existence and value of the critical area, the exact degeneracy counts (1/2/6 vacua for M=1/2/3), the mandatory nontrivial coordinate dependence, and the symmetry-breaking patterns—are obtained exclusively inside the lowest-mode truncation. No error estimate, bound on truncation error, comparison of the effective potential with the next Landau level, or stability analysis against mode mixing near the critical area (where the VEV is parametrically small) is supplied. Because the reported degeneracy and spontaneous symmetry breaking are read off from the truncated energy functional, any significant admixture of higher modes could lift degeneracies or alter the coordinate dependence, directly affecting the strongest claims.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for the careful reading of our manuscript and for highlighting the importance of assessing the robustness of the lowest-mode approximation. We address this point in detail below and will revise the manuscript accordingly.

read point-by-point responses

Referee: [lowest-mode approximation (as stated in the abstract and employed for all quantitative results)] The central claims—the existence and value of the critical area, the exact degeneracy counts (1/2/6 vacua for M=1/2/3), the mandatory nontrivial coordinate dependence, and the symmetry-breaking patterns—are obtained exclusively inside the lowest-mode truncation. No error estimate, bound on truncation error, comparison of the effective potential with the next Landau level, or stability analysis against mode mixing near the critical area (where the VEV is parametrically small) is supplied. Because the reported degeneracy and spontaneous symmetry breaking are read off from the truncated energy functional, any significant admixture of higher modes could lift degeneracies or alter the coordinate dependence, directly affecting the strongest claims.

Authors: We agree that all quantitative results, including the reported critical area and degeneracy counts, are obtained within the lowest-mode truncation, as stated in the abstract and throughout the text. This truncation is motivated by the Landau-level structure of the problem, where the energy gap to the next level grows with the magnetic flux and the inverse area; however, we did not provide explicit error estimates or a stability analysis against mode mixing. In the revised manuscript we will add a dedicated subsection discussing the regime of validity of the approximation. This will include (i) a perturbative estimate of the admixture of the first excited Landau level near the critical area, (ii) a qualitative argument that the degeneracy pattern remains stable when the VEV is small because the higher-mode corrections are parametrically suppressed by the Landau-level gap, and (iii) a brief comparison of the effective potential evaluated with and without the next level for a representative set of parameters. We will also explicitly state that the mandatory coordinate dependence of any nonzero VEV follows from the topology of the lowest Landau level and is therefore robust beyond the truncation, while the precise degeneracy counts are approximation-dependent. revision: yes

Circularity Check

0 steps flagged

No significant circularity; derivation self-contained in lowest-mode truncation

full rationale

The paper derives the critical area, nonvanishing VEV, coordinate dependence, and vacuum degeneracies (1/2/6 for M=1/2/3) by minimizing the effective potential under the lowest-mode approximation applied to the scalar field on the torus with flux M. This follows directly from the model's Lagrangian and the truncation choice without any self-definitional loops, fitted inputs renamed as predictions, or load-bearing self-citations that reduce the central claims to unverified priors. The steps are independent applications of standard methods on compact manifolds, with the approximation stated explicitly rather than smuggled in or justified circularly.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The central claim rests on the assumption that the lowest-mode truncation captures the true vacuum structure and on the standard quantization of magnetic flux on the torus; no additional free parameters or invented entities are introduced in the abstract.

axioms (2)

domain assumption Magnetic flux through the torus is quantized to an integer M
Standard topological quantization for U(1) bundle on torus; invoked to label the sectors studied.
ad hoc to paper The lowest-mode approximation suffices to determine the vacuum expectation value and its degeneracy
Explicitly employed to obtain the reported single, double, and six-fold degenerate vacua.

pith-pipeline@v0.9.0 · 5421 in / 1402 out tokens · 45310 ms · 2026-05-08T10:56:52.068684+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

40 extracted references · 34 canonical work pages

[1]

Broken Symmetries and the Masses of Gauge Boso ns,

P. W. Higgs, “Broken Symmetries and the Masses of Gauge Boso ns,” Phys. Rev. Lett. 13 (1964) 508–509

1964
[2]

Spontaneous Symmetry Breakdown without Mas sless Bosons,

P. W. Higgs, “Spontaneous Symmetry Breakdown without Mas sless Bosons,” Phys. Rev. 145 (1966) 1156–1163

1966
[3]

A Model of Leptons,

S. Weinberg, “A Model of Leptons,” Phys. Rev. Lett. 19 (1967) 1264–1266

1967
[4]

Weak and Electromagnetic Interactions,

A. Salam, “Weak and Electromagnetic Interactions,” Conf. Proc. C 680519 (1968) 367–377

1968
[5]

Plasmons, Gauge Invariance, and Mass,

P. W. Anderson, “Plasmons, Gauge Invariance, and Mass,” Phys. Rev. 130 (1963) 439–442

1963
[6]

Spontaneou sly broken translational invariance of compactiﬁed space,

M. Sakamoto, M. Tachibana, and K. Takenaga, “Spontaneou sly broken translational invariance of compactiﬁed space,” Phys. Lett. B 457 (1999) 33–38 , arXiv:hep-th/9902069

work page arXiv 1999
[7]

Novel phase structure of twi sted O(N) phi**4 model on M**(D-1) x S**1,

K. Ohnishi and M. Sakamoto, “Novel phase structure of twi sted O(N) phi**4 model on M**(D-1) x S**1,” Phys. Lett. B 486 (2000) 179–185 , arXiv:hep-th/0005017

work page arXiv 2000
[8]

Vacuum structure of twisted scalar ﬁeld theories on M**(D-1) x S**1,

H. Hatanaka, S. Matsumoto, K. Ohnishi, and M. Sakamoto, “ Vacuum structure of twisted scalar ﬁeld theories on M**(D-1) x S**1,” Phys. Rev. D 63 (2001) 105003 , arXiv:hep-th/0010283

work page arXiv 2001
[9]

Spontaneou s breaking of the rotational symmetry induced by monopoles in extra dimensions,

S. Matsumoto, M. Sakamoto, and S. Tanimura, “Spontaneou s breaking of the rotational symmetry induced by monopoles in extra dimensions,” Phys. Lett. B 518 (2001) 163–170 , arXiv:hep-th/0105196

work page arXiv 2001
[10]

Spontaneous breaking of t he C, P, and rotational symmetries by topological defects in extra two dimensions,

M. Sakamoto and S. Tanimura, “Spontaneous breaking of t he C, P, and rotational symmetries by topological defects in extra two dimensions, ” Phys. Rev. D 65 (2002) 065004 , arXiv:hep-th/0108208

work page arXiv 2002
[11]

Pha se Structure of Gauge Theories on an Interval,

Y. Fujimoto, T. Nagasawa, S. Ohya, and M. Sakamoto, “Pha se Structure of Gauge Theories on an Interval,” Prog. Theor. Phys. 126 (2011) 841–854 , arXiv:1108.1976 [hep-th] . 23

work page arXiv 2011
[12]

Cremades, L

D. Cremades, A. Ibáñez, and F. Marchesano, “Computing y ukawa couplings from magnetized extra dimensions,” JHEP 05 (2004) 079 , arXiv:hep-th/0404229

work page arXiv 2004
[13]

Non-Abelian discrete gauge symmetries in 4d string models ,

M. Berasaluce-Gonzalez, P. G. Camara, F. Marchesano, D. Regalado, and A. M. Uranga, “Non-Abelian discrete gauge symmetries in 4d string models ,” JHEP 09 (2012) 059 , arXiv:1206.2383 [hep-th]

work page arXiv 2012
[14]

Noncom mutative compactiﬁcations of type I strings on tori with magnetic background ﬂux,

R. Blumenhagen, L. Goerlich, B. Kors, and D. Lust, “Noncom mutative compactiﬁcations of type I strings on tori with magnetic background ﬂux,” JHEP 10 (2000) 006 , arXiv:hep-th/0007024

work page arXiv 2000
[15]

Abe, K.-S

H. Abe, K.-S. Choi, T. Kobayashi, and H. Ohki, “Three gen eration magnetized orbifold models,” Nucl. Phys. B 814 (2009) 265–292 , arXiv:0812.3534 [hep-th]

work page arXiv 2009
[16]

Classiﬁcation of three-generation models on m agnetized orbifolds,

T.-h. Abe, Y. Fujimoto, T. Kobayashi, T. Miura, K. Nishi waki, M. Sakamoto, and Y. Tatsuta, “Classiﬁcation of three-generation models on m agnetized orbifolds,” Nucl. Phys. B 894 (2015) 374–406 , arXiv:1501.02787 [hep-ph]

work page arXiv 2015
[17]

Three fermionic gener ations on a topological defect in extra dimensions,

M. V. Libanov and S. V. Troitsky, “Three fermionic gener ations on a topological defect in extra dimensions,” Nucl. Phys. B 599 (2001) 319–333 , arXiv:hep-ph/0011095

work page arXiv 2001
[18]

Three gen erations on a local vortex in extra dimensions,

J. M. Frere, M. V. Libanov, and S. V. Troitsky, “Three gen erations on a local vortex in extra dimensions,” Phys. Lett. B 512 (2001) 169–173 , arXiv:hep-ph/0012306

work page arXiv 2001
[19]

Fermion masses and quantum numbers from ex tra dimensions,

A. Neronov, “Fermion masses and quantum numbers from ex tra dimensions,” Phys. Rev. D 65 (Jan, 2002) 044004 . https://link.aps.org/doi/10.1103/PhysRevD.65.044004

work page doi:10.1103/physrevd.65.044004 2002
[20]

Fermion generations, mas ses, and mixings in a 6d brane model,

S. Aguilar and D. Singleton, “Fermion generations, mas ses, and mixings in a 6d brane model,” Phys. Rev. D 73 (Apr, 2006) 085007 . https://link.aps.org/doi/10.1103/PhysRevD.73.085007

work page doi:10.1103/physrevd.73.085007 2006
[21]

Fermion Generations from ’Apple-Shaped’ Extra Dimensions,

M. Gogberashvili, P. Midodashvili, and D. Singleton, “ Fermion Generations from ’Apple-Shaped’ Extra Dimensions,” JHEP 08 (2007) 033 , arXiv:0706.0676 [hep-th]

work page arXiv 2007
[22]

Fermion Families from Two Layer W arped Extra Dimensions,

Z.-q. Guo and B.-Q. Ma, “Fermion Families from Two Layer W arped Extra Dimensions,” JHEP 08 (2008) 065 , arXiv:0808.2136 [hep-ph]

work page arXiv 2008
[23]

Spacetime as a topological insul ator: Mechanism for the origin of the fermion generations,

D. B. Kaplan and S. Sun, “Spacetime as a topological insul ator: Mechanism for the origin of the fermion generations,” Phys. Rev. Lett. 108 (May, 2012) 181807 . https://link.aps.org/doi/10.1103/PhysRevLett.108.181807

work page doi:10.1103/physrevlett.108.181807 2012
[24]

Arkani-Hamed and M

N. Arkani-Hamed and M. Schmaltz, “Hierarchies without symmetries from extra dimensions,” Phys. Rev. D 61 (2000) 033005 , arXiv:hep-ph/9903417

work page arXiv 2000
[25]

Families as neighbors in e xtra dimension,

G. R. Dvali and M. A. Shifman, “Families as neighbors in e xtra dimension,” Phys. Lett. B 475 (2000) 295–302 , arXiv:hep-ph/0001072

work page arXiv 2000
[26]

Bulk Fields and Supersymmetry in a Slice of AdS

T. Gherghetta and A. Pomarol, “Bulk ﬁelds and supersymme try in a slice of AdS,” Nucl. Phys. B 586 (2000) 141–162 , arXiv:hep-ph/0003129. 24

work page Pith review arXiv 2000
[27]

Supersymmetry breaking, fermion masses and a small extra dimension,

D. E. Kaplan and T. M. P. Tait, “Supersymmetry breaking, fermion masses and a small extra dimension,” JHEP 06 (2000) 020 , arXiv:hep-ph/0004200

work page arXiv 2000
[28]

Fermion masses, mixings and pro ton decay in a Randall-Sundrum model,

S. J. Huber and Q. Shaﬁ, “Fermion masses, mixings and pro ton decay in a Randall-Sundrum model,” Phys. Lett. B 498 (2001) 256–262 , arXiv:hep-ph/0010195

work page arXiv 2001
[29]

Quark mass hierarchy and mixing via geometry of extra dimension with point interacti ons,

Y. Fujimoto, T. Nagasawa, K. Nishiwaki, and M. Sakamoto , “Quark mass hierarchy and mixing via geometry of extra dimension with point interacti ons,” PTEP 2013 (2013) 023B07 , arXiv:1209.5150 [hep-ph]

work page arXiv 2013
[30]

D ynamical generation of fermion mass hierarchy in an extra dimension,

Y. Fujimoto, T. Miura, K. Nishiwaki, and M. Sakamoto, “D ynamical generation of fermion mass hierarchy in an extra dimension,” Phys. Rev. D 97 (Jun, 2018) 115039 . https://link.aps.org/doi/10.1103/PhysRevD.97.115039

work page doi:10.1103/physrevd.97.115039 2018
[31]

Gaussi an froggatt-nielsen mechanism on magnetized orbifolds,

H. Abe, T. Kobayashi, K. Sumita, and Y. Tatsuta, “Gaussi an froggatt-nielsen mechanism on magnetized orbifolds,” Phys. Rev. D 90 (Nov, 2014) 105006 . https://link.aps.org/doi/10.1103/PhysRevD.90.105006

work page doi:10.1103/physrevd.90.105006 2014
[32]

cp phase from twisted higgs vacuum expectation value in extra dimension,

Y. Fujimoto, K. Nishiwaki, and M. Sakamoto, “ cp phase from twisted higgs vacuum expectation value in extra dimension,” Phys. Rev. D 88 (Dec, 2013) 115007 . https://link.aps.org/doi/10.1103/PhysRevD.88.115007

work page doi:10.1103/physrevd.88.115007 2013
[33]

Flavor mixings in ﬂux co mpactiﬁcations,

W. Buchmuller and J. Schweizer, “Flavor mixings in ﬂux co mpactiﬁcations,” Phys. Rev. D 95 no. 7, (2017) 075024 , arXiv:1701.06935 [hep-ph]

work page arXiv 2017
[34]

Flavor physics without ﬂa vor symmetries,

W. Buchmuller and K. M. Patel, “Flavor physics without ﬂa vor symmetries,” Phys. Rev. D 97 (Apr, 2018) 075019 . https://link.aps.org/doi/10.1103/PhysRevD.97.075019

work page doi:10.1103/physrevd.97.075019 2018
[35]

Kobayashi, Y

T. Kobayashi, Y. Nishioka, H. Otsuka, and M. Tanimoto, “ More about quark Yukawa textures from selection rules without group actions,” JHEP 05 (2025) 177 , arXiv:2503.09966 [hep-ph]

work page arXiv 2025
[36]

Operator analysis of physical states on magnetized T 2/Z N orbifolds,

T.-h. Abe, Y. Fujimoto, T. Kobayashi, T. Miura, K. Nishi waki, and M. Sakamoto, “Operator analysis of physical states on magnetized T 2/Z N orbifolds,” Nucl. Phys. B 890 (2014) 442–480 , arXiv:1409.5421 [hep-th]

work page arXiv 2014
[37]

Bachas,A Way to break supersymmetry,arXiv:hep-th/9503030

C. Bachas, “A Way to break supersymmetry,” arXiv:hep-th/9503030

work page arXiv
[38]

Open strings in background gauge ﬁelds,

A. Abouelsaood, C. G. Callan, Jr., C. R. Nappi, and S. A. Y ost, “Open strings in background gauge ﬁelds,” Nucl. Phys. B 280 (1987) 599–624

1987
[39]

Fermion Wavefunc tions in Magnetized branes: Theta identities and Yukawa couplings,

I. Antoniadis, A. Kumar, and B. Panda, “Fermion Wavefunc tions in Magnetized branes: Theta identities and Yukawa couplings,” Nucl. Phys. B 823 (2009) 116–173 , arXiv:0904.0910 [hep-th]

work page arXiv 2009
[40]

An extension of Fourier an alysis for the n-torus in the magnetic ﬁeld and its application to spectral analysis of th e magnetic Laplacian,

M. Sakamoto and S. Tanimura, “An extension of Fourier an alysis for the n-torus in the magnetic ﬁeld and its application to spectral analysis of th e magnetic Laplacian,” Journal of Mathematical Physics 44 no. 11, (11, 2003) 5042–5069 . https://doi.org/10.1063/1.1616203. 25

work page doi:10.1063/1.1616203 2003