Spectral asymptotics for the semiclassical Bochner Laplacian of a line bundle with constant rank curvature
Abstract.
The goal of this paper is manyfold. Firstly, we want to give a short introduction to the Bochner Laplacian and explain why it acts locally as a magnetic Laplacian. Secondly, given a confining magnetic field, we use Agmon-like estimates to reduce its spectral study to magnetic Laplacians, in the semiclassical limit. Finally, we use this to translate already-known spectral asymptotics for the magnetic Laplacian to the Bochner Laplacian.
1 – Introduction
1.1 – Motivations and context
The spectral theory of the magnetic Laplacian, and the Bochner Laplacian, has given rise to many interesting questions. First motivated by the Ginzburg-Landau theory, bound states of the magnetic Laplacian on a Riemannian manifold in the semiclassical limit were studied in many works (see the books [2, 12]), and appears to have very various behaviours according to the variations of the magnetic field and the boundary conditions. The first main technique consisted in the construction of approximated eigenfunctions (see for instance the works of Helffer-Mohamed [6] and Helffer-Kordyukov [3, 4, 5]). More recently, an other approach was developped, which consists in an approximation of the operator itself, using semiclassical tools such as microlocalisation estimates and Birkhoff normal forms (As in Raymond-Vu Ngoc [13] and Helffer-Kordyukov-Raymond-Vu Ngoc [7]). In the semiclassical limit , we recover the classical behaviour of a particle exposed to the magnetic field , since the magnetic Laplacian is the quantification of the classical energy.
If we are given a magnetic field which is not exact, there is no potential and we cannot define the magnetic Laplacian. However, the Bochner Laplacian appears to be the suitable generalization in this case, since it acts locally as a magnetic Laplacian. In this context the semiclassical parameter is . Its spectral theory appears to be deeply related to holomorphic structures and to the Kodaira Laplacian (or the renormalized Bochner Laplacian more generaly). For instance, this is exploited in the works of Marinescu-Savale [9], and Kordyukov [8]. In this last paper, the case of non-degenerate magnetic wells with full-rank magnetic field is studied, and expansions of the ground states energies are given, using quasimodes. In [11], similar results were obtained in the special case of the magnetic Laplacian, using a Birkhoff normal form, but also giving a description of semi-excited states, and a Weyl law. In [10], these result are generalized to constant-rank magnetic fields.
The spectral theory of the Bochner Laplacian is also deeply related to the global geometry of complex manifolds. For instance, in [1], a Weyl law was proven for , and used to get Morse inequalities and Riemann-Roch formulas.
1.2 – The Bochner Laplacian on a line bundle
Let be a compact oriented manifold of dimension . We consider a complex line bundle over , endowed with a Hermitian metric . In other words, we associate to each a -dimensional complex vector space , and a Hermitian product on . is a -dimensional manifold such that . A smooth section of (or -valued function) is a smooth function such that . It is the generalisation of the notion of function , but here the target space can vary with . Similarily, -valued -forms are sections of . We denote by the set of smooth sections of , and the set of smooth -valued -forms.
We take a Hermitian connexion on . It is the generalisation of the exterior derivative . The underlying idea is that the "derivative" of a -valued function should be -valued too. satisfies:
(1.1) |
(1.2) |
One can prove that acts as a multiplication. There exists a real closed -form on such that:
(1.3) |
Example : The trivial line bundle. The line bundle , such that is called the trivial line bundle. We identify sections with functions by . Similarily, -valued -forms are identified with -valued -forms, and we recover the usual differential objects on . If is endowed with the Hermitian product , we call the trivial Hermitian line bundle. We write for short. Hermitian connexions on the trivial line bundle are given by where and is the exterior derivative. The curvature of is , as shown by the easy but enlightening calculation:
(1.4) |
Let us describe now the Bochner Laplacian associated to a Hermitian connexion on a Hermitian complex line bundle . First note that the spaces and are endowed with -norms. The norm of a section is:
(1.5) |
where denotes the volume form of the oriented Riemannian manifold . We denote by the completion of for this norm. The definition of the norm of a -valued -form is a little more involved. First, using a partition of unity, it is enough to define it for where is a small open subset of . If is small enough, there exists a section such that . Then for any , there exists a unique such that (we identify -forms with tangent vectors using the metric ). We define:
(1.6) |
The completion of for this norm is denoted by : it is the space of square-integrable -valued -forms. These norms are associated with scalar products, denoted by brackets .
The formal adjoint of for these scalar products is denoted by . The Bochner Laplacian is the self-adjoint extension of . It is the operator associated with the quadratic form:
(1.7) |
We denote by its domain. is a dense subspace of and:
(1.8) |
Since is compact, one can prove that has compact resolvent, and we denote by
(1.9) |
the non-decreasing sequence of its eigenvalues. We will use the following notation for the Weil counting function:
In this paper, we are interested in the semiclassical limit, i.e. the high curvature limit "". We can increase the curvature using tensor products of . For any , we denote by the -th tensor power of . is still a complex line bundle of , with . It is endowed with the Hermitian product . The connexion induces a Hermitian connexion on by the formula:
The curvature of is
(1.10) |
Hence, the high curvature limit is . We want to invastigate the behaviour of and the corresponding eigensections in the limit .
1.3 – Main results
One can measure the "intensity" of the curvature (the magnetic field) in the following way. We denote by the linear operator defined by
(1.11) |
is real and skew-symmetric with respect to , and thus its eigenvalues lie on the imaginary axis and are symmetric with respect to the origin. We denote its non-zero eigenvalues by:
(1.12) |
with . Hence, the rank of is and might depend on , but we will soon assume that is constant, at least locally. The magnetic intensity is the function defined by:
(1.13) |
This function is continuous on , but not smooth in general. However, note that it is smooth on a neighborhood of any point where the are simple (if is locally constant near ).
One of the purposes of this article is to show that the eigensections of are localized near the minimum points of , and to deduce that the low-lying spectrum of is given by magnetic Laplacians on neighborhoods of the minimal points of .
We will do the following assumptions.
Assumptions.
-
(A1)
The minimal value of is only reached at non degenerate points . We denote by .
-
(A2)
The rank of is constant on small neighborhoods of . We denote by the rank of .
-
(A3)
We assume , which is equivalent to say that for any .
As noticed in several papers, one can prove using Agmon-like estimates that the eigensections of associated to low-lying eigenvalues are exponentially localized near in the limit . Now let us present the local model operators on .
Recall that the -form is closed: . Hence, if the open sets are small enough, is exact on : there exists such that on . We denote by the Dirichlet realization of on . It is the self-adjoint operator associated to the following sesquilinear form on :
(1.14) |
We prove the following Theorem.
Theorem 1.
Let . Under assumptions and , if are small enough, then:
(1.15) |
uniformly with respect to , where
and denotes the number of eigenvalues of an operator below , counted with multiplicities.
As a corollary, we can deduce spectral asymptotics for from already-known results for . Let us recall some of these results here.
1.3.1 – The full-rank case
Under the assumptions , we fix a , and we denote by . Hence, is just the restriction of to the small open set , where it admits a primitive . is the magnetic Laplacian with Dirichlet boundary conditions on , with magnetic field . We first focus on the full-rank case, when the rank of is maximal: . We define by the condition
(1.16) |
Note that, if the are pairwise distinct, we can choose . Moreover, if the open set is small enough we have, for all and ,
(1.17) |
The following Theorem was proved in [11].
Theorem 2.
We assume with and in (1.16). Let small enough. Then there exists a symplectomorphism such that:
(1.18) |
uniformly with respect to , where is a pseudo-differential operator with principal symbol:
and
Hence, we have a description of the semi-excited states of . In the same paper, Weyl estimates are proven for . We directly deduce the following Weyl estimates for . A Similar formula was proven in [1] using a local approximation of the magnetic field by a constant field.
Corollary 3.
Assume , and for any that , and that are pairwise distinct. Then, for small enough,
(1.19) |
in the limit , where .
Finally, we also deduce asymptotic expansions of the first eigenvalues.
Corollary 4.
Assume , and for any that and that are pairwise distinct, and . Then, for any and ,
for some coefficients .
Remark. We also have geometric interpretations of the coefficients. First, the full expansion comes from the effective operator , which is the reduction of to the lowest energy of the Harmonic oscillator describing the classical cyclotron motion. Moreover, is given by an eigenvalue of an other Harmonic oscillator whose symbol is the Hessian of at (for some ): it describes a slow drift of the classical particle arround . If the eigenvalues of this oscillator are simple, then a Birkhoff normal form can be used to show that if is odd.
1.3.2 – The constant-rank case
In the non-full-rank case, the kernel of (which corresponds to the directions of the field lines), has a great influence on the spectrum of . Fix . If the rank of is constant, equal to , then its kernel as dimension . The partial Hessian of at , in the directions of the Kernel of , is non-degenerate. we denote by
(1.20) |
its eigenvalues. For simplicity, we will make the following non-resonance assumptions (however, we can deal with resonances using a resonance order as in the full-rank case).
Assumptions.
(A4) For every , are non-resonnant in the following sense:
(A5) For every such that , are non resonnant in the following sense:
Applying the results of [10] to get spectral asymptotics for , we deduce from Theorem 1 the following corollary.
Corollary 5.
Assume , and let . Then admits a full asymptotic expansion in powers of :
Moreover:
-
•
If there is at least one such that , then .
-
•
If , then
2 – Some Remarks
2.1 – The Bochner Laplacian and the magnetic Laplacian are locally the same
If is any open subset of such that there exists a non-vanishing section , then any can be written for some . Hence,
with . Moreover,
and thus . Hence acts locally as , and as the magnetic Laplacian . This is the core of Theorem 1.
2.2 – On the quantization of a magnetic field
If we are given a closed -form (the magnetic field), the quantization question constist in finding a quantum operator associated to . If is exact, this question is answered by the semiclassical magnetic Laplacian , with . Here, is the semiclassical parameter (Planck’s constant) and the semiclassical limit is .
If is not exact, but if there exists an Hermitian line bundle with Hermitian connexion such that , then the Bochner Laplacian acts locally as the magnetic Laplacian and hence it is a good candidate. Moreover, we have locally
so that the semiclassical parameter is now (Also notice the factor which is important for the eigenvalue asymptotics). The limit is equivalent to exept that the semiclassical parameter becomes discrete ().
A new question arises : When does such an Hermitian line bundle exists ? Weil’s Theorem states that it exists if and only if satisfies the prequantization condition:
(2.1) |
where denotes the cohomology class of . This condition also enlightens the discreteness of the semiclassical parameter. Indeed, if one wants to quantize the magnetic field , then one must have , and thus , unless which means that is exact (and thus we can use the magnetic Laplacian !).
3 – Proof of Theorem 1
3.1 – Agmon-like estimates
In this section, we prove exponential decay on the eigensections of , away from the set . We need the following Lemma.
Lemma 6.
There exist and such that, for and ,
Proof.
Take a partition of unity on , such that with a small open subset of , and
(3.1) |
Then, for any we have
(3.2) |
and . If every are small enough, we can find a non-vanishing section on which we can use to trivialize the line bundle. Writting for some , we have
(3.3) |
where is such that . For the magnetic Laplacian , the desired inequality is well-known: There exist , such that, for every , and :
(3.4) |
Using (3.2), (3.3), and (3.4), we deduce that
(3.5) |
with , and for . Finally, (3.1) yields
(3.6) |
Hence, up to changing into , Lemma 6 is proved. ∎
Now we can use Lemma 6 to prove Agmon-like decay estimates.
Theorem 7.
Let , , and . There exist and such that, for all and all eigenpair of with ,
Proof.
Let be a Lipschitz function. The Agmon formula is:
(3.7) |
Using Lemma 6, we deduce that:
We split this integral into two parts.
We choose :
where for , for , and uniformly bounded with respect to . Since on and for large enough, we have:
Moreover, since and ,
for large enough. But on , so there is a and such that, for :
Since on , we get a new such that:
and we can use Fatou’s lemma in the limit to get the desired inequality. ∎
Corollary 8.
Let and be a smooth cutoff function, being 1 on a small neighborhood of
Then, for any eigenpair of , with we have:
and
uniformly with respect to .
Proof.
By Theorem 7, we have:
(3.8) |
which gives the first estimates. Moreover, we have with ,
and using Agmon’s formula 3.7 and Theorem 7:
Thus,
(3.9) |
We can use these Agmon estimates on to get our second result.
(3.10) |
The first term is dominated by
(3.11) |
where is a cutoff function such that on and on . We can apply (3.8) to to get:
(3.12) |
The second term of (3.10) is dominated as in (3.8), using (3.9):
(3.13) |
Finally, (3.10) with (3.12) and (3.13) yields
∎
3.2 – Comparison of the spectrum of and
We recall that the minimum of is reached at in a non-degenerate way. For small enough, the compact set has disjoint connected components such that . We fix the value of , and we take a neighborhood of . For sufficiently small, .
We denote by the restriction of to . is the Dirichlet realisation of , with such that . It is the self adjoint operator associated to the quadratic form:
(3.14) |
Let us denote by
(3.15) |
We split the proof of Theorem 1 into two Lemmas.
Lemma 9.
Let . We have:
uniformily with respect to .
Proof.
We prove this using the min-max principle. For , let be the normalized eigenfunction associated to . We will define the quasimode using a local trivialisation of on . Let be the non-vanishing local section of such that, for any ,
(3.16) |
Let be a smooth cutoff function, such that on . We define by , and
Then
Moreover, by (3.16),
Now, satisfies the assumptions of Corollary 8 (with instead of ). Thus,
uniformly with respect to . being the eigensection associated to , it remains:
This is true for every . Hence, for we have
and the Lemma follows from the min-max principle, because the vector space ranged by is -dimensional (and ). ∎
The reverse inequality is proven similarily.
Lemma 10.
Let . We have:
uniformily with respect to .
Proof.
The -th eigenvalue of is given by an eigenpair of for some . Let be a cutoff function equal to on . Then, Agmon estimates (Theorem 7) for imply that
uniformly with respect to . We define , where satisfies (3.16), and we extend by outside . Then,
Hence the min-max principle implies
which is the desired inequality. ∎
References
- [1] J.P. Demailly “Champs magnétiques et inégalités de Morse pour la d"-cohomologie” In Annales de l’Institut Fourier 35-4, 1985
- [2] S. Fournais and B. Helffer “Spectral methods in surface superconductivity” Progress in Nonlinear Differential Equationstheir Applications. Birkhäuser Boston Inc., Boston, MA, 2010
- [3] B. Helffer and Y. Kordyukov “Semiclassical spectral asymptotics for a magnetic Schrödinger operator with non-vanishing magnetic field” In Geometric Methods in Physics: XXXII Workshop, Bialowieza, Trends in Mathematics. Birkhäuser, Basel, 2014
- [4] B. Helffer and Y. Kordyukov “Semiclassical spectral asymptotics for a two-dimensional magnetic Schrödinger operator. II The case of degenerate wells” In Communications in Partial Differential Equations 37, 2012
- [5] B. Helffer and Y. Kordyukov “Semiclassical spectral asymptotics for a two-dimensional magnetic Schrödinger operator: The case of discrete wells” In Spectral Theory and Geometric Analysis 535, 2011
- [6] B. Helffer and A. Mohamed “Semiclassical Analysis for the Ground State Energy of a Schrödinger Operator with Magnetic Wells” In Journal of Functional Analysis 138, 1996
- [7] B. Helffer, Y. Kordyukov, N. Raymond and S. Vũ Ngoc “Magnetic Wells in Dimension Three” In Anal. and PDE 9, 2016
- [8] Y.A. Kordyukov “Semiclassical eigenvalue asymptotics for the Bochner Laplacian of a positive line bundle on a symplectic manifold” In arXiv: 1908.01756v1, 2019
- [9] G. Marinescu and N. Savale “Bochner Laplacian and Bergman kernel expansion of semi-positive line bundles on a Riemann surface” In arXiv: 1811.00992, 2018
- [10] L. Morin “A semiclassical Birkhoff normal form for constant-rank magnetic fields” In arXiv:2005.09386, 2020
- [11] L. Morin “A Semiclassical Birkhoff Normal Form for Symplectic Magnetic Wells” In arXiv:1907.03493, 2019
- [12] N. Raymond “Bound States of the Magnetic Schrödinger Operator” EMS Tracts in Mathematics, 2017
- [13] N. Raymond and S. Vũ Ngoc “Geometry and Spectrum in 2D Magnetic wells” In Annales de l’Institut Fourier 65 (1), 2015