Skip to Content
You are currently on the new version of our website. Access the old version .
MathematicsMathematics
  • Article
  • Open Access

27 February 2026

15 Pages

Study of Second-Order Differential Operators on Graphs with Small Edges

and
1
Department of Mathematical and Computer Modeling, International Information Technologies University (IITU), Almaty 050000, Kazakhstan
2
Faculty of Mechanics and Mathematics, Al-Farabi Kazakh National University, Almaty 0500000, Kazakhstan
*
Author to whom correspondence should be addressed.

Abstract

In this work, we investigate a Schrödinger operator defined on a model graph containing small loops, under the assumption that the standard nonresonance condition—typically ensuring the holomorphic dependence of the resolvent for elliptic operators on graphs with short edges—is violated. Our analysis focuses on the behavior of those components of the resolvent that correspond to finite edges and small loops. It is shown that these components retain their holomorphic dependence on a small parameter characterizing the length of the loops. In contrast to the nonresonant case, however, the part of the resolvent associated with the small loops develops an additional contribution in the leading term of its Taylor expansion, which results in a certain localization of the resolvent on these loops.

1. Introduction

Over the past two decades, the theory of quantum graphs has undergone intensive development. One of its principal directions concerns the dependence of the resolvent and spectral properties of elliptic operators on the lengths of the graph edges. When the lengths of finite edges vary slightly, the corresponding perturbation is regular, and the resolvent remains holomorphic [1]. In contrast, when the edges themselves are short, the problem becomes singular, and the convergence of the resolvent in this singular regime has been investigated in a number of specific models (see, e.g., [2,3,4,5,6]). The most general results on such convergence were first obtained in [7] and later extended in [8,9,10,11,12,13], initially for simple model configurations and then for general elliptic operators on arbitrary graphs with small edges. It was established in these works that, under typical assumptions, the resolvents of elliptic operators on graphs with short edges are holomorphic with respect to a small parameter representing the edge lengths. The same property holds for isolated eigenvalues and the associated eigenfunctions. This typical situation is characterized by a certain nonresonance condition, which essentially corresponds to the generic case. When the nonresonance condition is violated, however, the analytical methods developed in [10,11,12,13] are no longer applicable, and the behavior of the resolvent and the spectrum remains an open problem. In the present paper, we address this issue by considering a model graph of relatively simple but sufficiently general structure that includes small loops attached to a common vertex (see Figure 1). On this graph, we study the Schrödinger operator. The simplicity of the underlying geometry makes it possible to carry out a detailed analysis of the resolvent. Our results show that the relevant parts of the resolvent retain their holomorphic dependence even when the nonresonance condition does not hold. At the same time, for the component of the resolvent corresponding to the small loops, an additional qualitatively new term appears in the leading part of its Taylor expansion. This new contribution causes a partial localization of the resolvent on the small loops, providing further insight into the spectral behavior of elliptic operators on such graph structures.

Statement of the Problem—The Main Results

Let Γ be a finite metric graph consisting of a set of vertices and edges. On each edge, an orientation and a local variable are introduced. Assume that the graph Γ has a distinguished vertex M, from which only one edge, denoted by e 0 , emanates. Let ε > 0 be a small parameter, and denote by Γ ε the graph obtained from Γ by attaching a finite number of loops of lengths a i ε , i = 1 , … , n , to the vertex M, where a i are fixed positive constants. Each loop is denoted by l i , ε , i = 1 , … , n . On every loop, an orientation and a local coordinate x i ∈ [ 0 , a i ε ] are introduced. An example of such a graph with six loops l i , ε is shown in Figure 1.
Figure 1. The perturbed graph Γ ε , containing loops of the length a i ε , i = 1 , … , n , attached to the vertex M, and converges to the limit graph Γ .
The graph Γ ϵ contains small loops attached to the vertex M, whose lengths are proportional to ε . These loops represent singular perturbations and disappear in the limit ε → 0 , leading to the limit graph Γ.
Let the graph Γ be equipped with a real potential V ( x ) that is bounded and continuous. Each small loop l i , ε is associated with a local variable x ε ( i ) . For these loops, we define scaled potentials of the form W i , ε ( x ε ) : = W i ( x i , ε ε ) , where the functions W i ( ξ ) are continuous and real-valued on the intervals l i : = ( 0 , a i ) , i = 1 , … , n . The main operator investigated in this paper is the Schrödinger operator H ε acting on the graph Γ ε . It is defined by the differential expression
H ε = − d 2 d x 2 + V ε ( x ) ,
where the potential V ε ( x ) is given by
V ε ( x ) = V ( x ) , o n Γ , ε − 2 W i ( x i , ε ε ) , o n l i , ε , i = 1 , … , n .
Here ε > 0 is a small parameter representing the characteristic length of the attached loops.
At the common vertex M , we impose the standard Kirchhoff matching conditions that ensure the continuity of the function and the conservation of the total flux. These conditions can be written as
u | e 0 ( M ) = u | l i , ε ( M − 0 ) = u | l i , ε ( M + 0 ) , i = 1 , … , n ,
and
u ′ | e 0 ( M ) + ∑ i = 1 n ( u ′ | l i , ε ( M − 0 ) + u ′ | l i , ε ( M + 0 ) ) = 0 .
The first relation expresses the continuity of the wave function at the vertex, while the second one represents the Kirchhoff flux conservation law.
Here, the functions u | e 0 and u | l i , ε represent the restrictions of the global function u to the corresponding edges of the graph. The symbols u | l i , ε ( M − 0 ) and u | l i , ε ( M + 0 ) denote, respectively, the values of these restrictions at the initial and terminal points of the loop l i , ε . The same interpretation applies to the terms involving the derivatives of u , where each derivative is taken in the direction outward from the vertex M. At the remaining vertices of the graph Γ, we impose the standard boundary conditions, ensuring that the operator H ε is self-adjoint in the Hilbert space L 2 ( Γ ε ) . Such types of vertex conditions guaranteeing self-adjointness are well described in the general theory of differential operators on graphs (see, for example, [14] (Chapter 1, Subsection 1.4.1, Theorem 1.4.4) and [11] (Lemma 4.1)). The domain D ( H ε ) of the operator H ε is defined as a subspace of W ˙ 2 2 ( Γ ε ) , consisting of all functions that satisfy the prescribed boundary conditions at the vertices of the graph Γ ε . For any metric graph, we denote by W ˙ 2 j ( · ) the direct sum of the Sobolev spaces W 2 j ( e ) taken over all edges e of the graph, that is,
W ˙ 2 j ( · ) : = ⨁ e ∈ · W 2 j ( e ) .
We impose the following key assumption on the potentials W i = W i ( ξ ) , i = 1 , … , n . For each i , the function W i is such that the boundary value problem
− d 2 Ψ i d ξ 2 + W i ( ξ ) Ψ i ( ξ ) = 0 , ξ ∈ ( 0 , a i ) , Ψ ( 0 ) = Ψ i ( a i ) = 0
admits a single nontrivial solution. Moreover, this solution satisfies the normalization condition
Ψ i ′ ( a i ) = Ψ i ′ ( 0 ) = 1
We denote by Φ i = Φ i ( ξ ) , i = 1 , … , n , another linearly independent solution of the differential equation given in (2). Using the normalization condition (3) and the fact that the Wronskian of two linearly independent solutions of Equation (2) remains constant along the interval, we can select the function Φ i in such a way that
Ψ i ( 0 ) = Ψ i ( a i ) = 1
The main objective of this work is to investigate the analytic behavior of the resolvent associated with the operator H ε in the limit as ε → + 0 . In order to present the principal results, we first introduce several auxiliary definitions and notations.
Let the collection of all small loops of the graph be denoted by l ε . The Hilbert space L 2 ( Γ ε ) can naturally be regarded as the direct sum
L 2 ( Γ ) ⊕ L 2 ( l ε ) = L 2 ( Γ ) ⊕ L 2 ( l 1 , ε ) … ⊕ L 2 ( l n , ε ) .
Let l = l 1 , … , l n be a graph consisting of the edges l i . In the space L 2 ( l i ) , we define the contraction operators that map functions from L 2 ( l i ) into L 2 ( l i , ε ) , by the rule
( S i ( ε ) u ) ( x ε ) : = u x i , ε ε .
It is straightforward to verify that each operator S i ( ε ) is bounded and invertible. Moreover, their operator norms satisfy the relations
‖ S i ( ε ) ‖ L 2 ( l i ) → L 2 ( l i ε ) = ε 1 2 ,   ‖ S i − 1 ( ε ) ‖ L 2 ( l ε ) → L 2 ( l ) = ε − 1 2 .
Throughout the paper, the notation ∥ · ∥ ( X → Y ) denotes the operator norm of a bounded linear mapping from a Banach space X into a Banach space Y. We introduce the operator
S ( ε ) = S 1 ( ε ) ⊕ · · · ⊕ S n ( ε ) ,
and denote by
S ε : l → l ε
the combined mapping acting between the spaces associated with the unscaled and scaled graphs, respectively.
We define two auxiliary operators, denoted by H D and H N , acting on the graph Γ and associated with the same differential expression
H ^ : = − d 2 d x 2 + V ( x ) ,
where V is a real, bounded, and continuous potential on the edges of Γ. At all vertices of the graph, except for the vertex M, we impose the same boundary conditions as those used for the operator H ε . At the distinguished vertex M, the operator H D is defined by the Dirichlet boundary condition
u ( M ) = 0 ,
whereas the operator H N corresponds to the Neumann condition
d u d x ( M ) = 0 .
The domains of these operators, denoted by D ( H D ) and D ( H N ) , are the subspaces of W ˙ 2 2 ( Γ ) consisting of functions that satisfy the respective boundary conditions at all vertices of Γ Both H D and H N are self-adjoint operators in L 2 ( Γ ) due to the imposed boundary conditions (see [11] (Lemma 4.1)).
Let us introduce the restriction operators P Γ and P l i , ε , which act from the space L 2 ( Γ ε ) onto L 2 ( Γ ) and L 2 ( l i , ε ) respectively. These operators are defined by the relations
P Γ u = u | Γ , P l i , ε u = u | l i , ε ,
where u | Γ and u | l ε denote the restrictions of a function u, defined on the entire graph Γ ε , to the subgraph Γ and to the loop l i , ε respectively. It is then convenient to define the combined restriction operator
P l ε : = P l 1 , ε ⊕ · · · ⊕ P l n , ε : L 2 ( Γ ε ) → L 2 ( l ε )
which maps functions from the space L 2 ( Γ ε ) to their restrictions on the subgraph Γ ε , consisting of all small loops attached to the vertex M.
Since the operators H ε and H D are self-adjoint, their resolvents ( H ε − λ ) − 1 , ( H D − λ ) − 1 , ( H N − λ ) − 1 are well defined for all complex values of λ such that Im λ ≠ 0 . For such values of λ , we introduce on the space L 2 ( Γ ) ⊕ L 2 ( l ) , a pair of linear operators defined by
R Γ ( ε , λ ) : = P Γ ( H ε − λ ) − 1 ( S Γ ⊕ S ( ε ) ) , R ( ε , λ ) : = S − 1 ( ε ) P ε ( H ε − λ ) − 1 ( S Γ ⊕ S ( ε ) ) ,
here, S Γ denotes the identity operator in the space L 2 ( Γ ) , and the direct sum is taken in accordance with the decomposition L 2 ( Γ ε ) = L 2 ( Γ ) ⊕ L 2 ( l ε ) . It follows directly from the definitions that the operator R Γ ( ε , λ ) acts as a bounded mapping from L 2 ( Γ ) ⊕ L 2 ( l ) into W ˙ 2 2 ( Γ ) , while the operator R ( ε , λ ) is bounded as an operator from L 2 ( Γ ) ⊕ L 2 ( l ) to W ˙ 2 2 ( l ) : = W 2 2 ( l 1 ) ⊕ · · · ⊕ W 2 2 ( l n ) .
For convenience, we introduce the notation q : = ∑ i = 1 n ( φ i ′ ( a i ) − φ ′ ( 0 ) ) , which will be used in the statement of the main theorem presented below.
Theorem 1.
Let Im λ ≠ 0 . Then the operators R Γ ( ε , λ ) and R l ( ε , λ ) are holomorphic with respect to the small parameter ε. If
q ≠ 0
then the leading terms of the Taylor expansions of R Γ ( ε , λ ) and R l ( ε , λ ) as ε → 0 are given by
R Γ ( ε , λ ) ( f Γ , f l ) = ( H D − λ ) − 1 f Γ + O ( ε ) , R l ( ε , λ ) ( f Γ , f l ) = − ⨁ i = 1 n ( f l , ψ i ) L 2 ( l ) λ ∥ ψ i ∥ L 2 ( l ) 2 ψ i + O ( ε ) ,
where the direct sum is taken in the sense of the decomposition
L 2 ( l ) = ⨁ i = 1 n L 2 ( l i )
If, on the other hand,
q = 0
then the first terms of the corresponding Taylor series have the following form:
R Γ ( ε , λ ) ( f Γ , f l ) = ( H N − λ ) − 1 f Γ + O ( ε ) , R l ( ε , λ ) ( f Γ , f l ) = − ⨁ i = 1 n ( f l , ψ i ) L 2 ( l ) λ ∥ ψ i ∥ L 2 ( l ) 2 ψ i − − ( ( H D − λ ) − 1 f Γ ) ′ ( M ) U D ′ ( M ) ⨁ i = 1 n φ i ⊥ + O ( ε ) ,
where
φ i ⊥ : = φ i − ( φ i , ψ i ) L 2 ( l i ) ∥ ψ i ∥ L 2 ( l i ) 2 ψ i .
Let us analyze the principal result in more detail, starting from the behavior of the operators R Γ ( ε , λ ) and R l ( ε , λ ) . The operator I Γ ⊕ S ( ε ) acts on a pair of functions f Γ ∈ L 2 ( Γ ) , f l ∈ L 2 ( l ) , where f Γ is defined on the main graph Γ and f on the loops i. The action is as follows: the component f Γ remains unchanged, while in the second component f, the variable is rescaled (contracted), so that the resulting function is defined on the small loops i ε . Hence, we obtain a function belonging to the space L 2 ( Γ ε ) . Applying the resolvent ( H ε − λ ) − 1 to this function and then restricting the result to the subgraph Γ and to the small loops E ε , we get the following. The restriction to Γ corresponds precisely to the action of the operator R Γ ( ε , λ ) . Similarly, applying the inverse contraction operator S − 1 ( ε ) to the restriction on the small loops yields the same result as the action of the operator R l ( ε , λ ) on the pair ( f Γ , f l ) . Therefore, it is natural to interpret the operators R Γ ( ε , λ ) and R l ( ε , λ ) as the components of the resolvent ( H ε − λ ) − 1 corresponding to the subgraph Γ and the system of small loops l ε , respectively. This interpretation is supported by the fact that the full resolvent can be reconstructed from its parts via the simple identity
( H ε − λ ) − 1 = ( R Γ ( ε , λ ) ⊕ S ( ε ) R l ( ε , λ ) ) ( I Γ ⊕ S − 1 ( ε ) ) .
In accordance with the general theory developed in [11], the components of the resolvent associated with operators on graphs containing small edges are analytic with respect to the parameter ε in the generic situation, that is, when the so-called nonresonance condition A from [11] is fulfilled.
The graph Γ ε considered in this paper, however, represents a simple instance where this condition is not satisfied. Following the analytical framework proposed in [11], we examine a model graph composed of several loops l i and a single semi-infinite edge, all joined at a common vertex M. The coordinate along the semi-infinite edge is denoted by ξ 0 , and the entire structure will be referred to as the graph l ∞ . On this graph, we define a self-adjoint operator H l acting according to the differential expressions
H ^ = − d 2 d ξ i 2 + W i ( ξ i ) on each loop l i H l ^ = − d 2 d ξ 0 2 on the semi-infinite edge ,
subject to the Kirchhoff matching condition at the vertex M. The essential spectrum of H l coincides with the semi-axis [ 0 , + ∞ ) . The nonresonance condition A from [11] asserts that there are no eigenvalues embedded at the lower bound of this essential spectrum. However, the framework allows for the existence of a virtual level, which corresponds to a nontrivial solution φ of the equation H ^ l φ = 0 on the graph l ∞ , such that φ takes constant values along the semi-infinite edge.
In the present situation, the nonresonance condition A is clearly violated, since the operator H l admits exactly n eigenfunctions associated with the lower edge of its essential spectrum. Each of these eigenfunctions coincides with the function ψ i on the corresponding loop i and vanishes identically on all other edges of the graph l ∞ . The boundary conditions for ψ i specified in (2)–(3) ensure that the Kirchhoff matching condition at the common vertex M is automatically satisfied. In addition, the operator H l may also exhibit the presence of a virtual level located at the bottom of the essential spectrum. The criterion for the existence of such a level is precisely the relation given by identity (8). In this resonant case, the associated generalized eigenfunction takes the form φ i ⊥ on each loop i and equals 1 identically on the semi-infinite edge.
According to Theorem 1, even though the nonresonance condition A is not fulfilled, the operators R Γ ( ε , λ ) and R l ( ε , λ ) nevertheless remain holomorphic with respect to ε . This constitutes the first key observation derived from our main theorem and demonstrates that the analyticity of these operators is, in fact, a universal feature. Furthermore, the leading term in the Taylor expansion of R Γ ( ε , λ ) coincides with that obtained in the standard situation where condition A is satisfied. Specifically, when inequality (6) holds, the operator H does not admit a virtual level, and the leading term corresponds to the resolvent with a Dirichlet condition at the vertex M. Conversely, if equality (8) is valid, a virtual level emerges, and the leading term involves the operator subject to a Neumann condition. These findings are in full agreement with the conclusions of [11] for the nonresonant case: as the small parameter tends to zero, one obtains the Dirichlet condition at M in the presence of a virtual level, and the Neumann condition otherwise.
The second key conclusion derived from Theorem 1 highlights the principal distinction between our results and those established in [11] for the nonresonant case. This difference concerns the behavior of the operator R l ( ε , λ ) , in particular, the structure of the leading term in its Taylor expansion. Specifically, an additional component of the form
⨁ i = 1 n ( f l , ψ i ) L 2 ( l ) λ ∥ ψ i ∥ L 2 ( l ) 2 ψ i
appears in the leading term, representing a qualitatively new contribution absent in the nonresonant setting. The additional term is characterized by the projection of f onto the eigenfunctions corresponding to the embedded eigenvalue at zero. It appears in both (7) and (9), indicating that its presence is independent of the existence of a virtual level. In the case where a virtual level is present, an additional term emerges
− ( ( H D − λ ) − 1 f Γ ) ′ ( M ) U D ′ ( M ) ⨁ i = 1 n φ i ⊥
This term appears in the leading order (see (9)) and is qualitatively distinct from the other term. Specifically, the projection onto the eigenfunctions corresponding to the embedded eigenvalue is determined solely by the function f ; that is, it depends on the graph Γ ε and the values of the right-hand side on the small loops. Moreover, if the right-hand side vanishes on the finite edges of the subgraph Γ ( f Γ = 0 ) , then the leading term in the Taylor series of the operator R Γ is identically zero, and the term associated with the virtual level in (9) also vanishes. A similar property was observed in [11], where the leading terms in the Taylor series for the operators R Γ and R l were entirely determined by the right-hand side on the finite edges. In the present case, however, due to the embedded eigenvalue, an additional term depending solely on f l arises in the leading term of the Taylor series of R l . When f Γ = 0 , only a first-order term remains in the restriction of the resolvent on small loops. This indicates that the resolvent is effectively localized on the small loops. A related phenomenon was previously noted in [2] (Theorem 2), although no statements regarding holomorphy were made there; furthermore, the graph Γ considered was of a very simple, star-shaped structure.
In summary, we conclude the following. For a graph Γ ε containing small loops, if an embedded eigenvalue exists at the bottom of the essential spectrum of the operator H l , the holomorphy of the operators R Γ and R l is preserved. The term representing the projection of f l onto the eigenfunctions corresponding to the embedded eigenvalue appears in the leading term of the Taylor expansion of R l , indicating an additional localization of the resolvent of H ε on the small loops. The remaining terms in the leading part of the Taylor series of R Γ and R l are analogous to those in the nonresonance case described in [11].

2. Analytic Behaviour of the Resolvent

In this section, we prove Theorem 1. Owing to the simple structure of Γ ε , specifically the small loops l i , s , the resolvent of H ε can be explicitly represented, allowing for a detailed analysis of the operators R ε and R l .
We select arbitrary functions f Γ ∈ L 2 ( Γ ) and f l i ∈ L 2 ( l i ) and define f ∈ L 2 ( Γ ε ) by and we consider the resolvent equation for the operator H ε :
( H ε − λ ) u ε = f , I m λ ≠ 0 .
Focusing on the loop i , ε , we introduce the scaled variable ξ i = x i , s / ε . Setting v i : = S i − 1 ( ε ) u ε , with v i = v i ( ξ i , ε ) we immediately obtain the ordinary differential equation
− d 2 v i d ξ i 2 + W i v i − ε 2 λ v i = ε 2 f i , l on l i , W i = W i ( ξ i ) .
Since λ is fixed and ε is small, the term ε 2 λ is correspondingly small.
We denote by L i the self-adjoint operator in L 2 ( l i ) generated by the differential expression
L ^ i = − d 2 d ξ i 2 + W i ( ξ i ) ,
subject to Dirichlet boundary conditions. The domain D ( L i ) consists of functions in W 2 2 ( l i ) that vanish at the end points of the interval l i . Let ψ i be the eigenfunction corresponding to the zero eigenvalue. Since this eigenvalue is simple, the standard resolvent expansion of a self-adjoint operator in a neighborhood of an isolated eigenvalue (see, e.g., [15]) applies. Consequently, for sufficiently small μ , the resolvent ( L i − μ ) − 1 admits the representation
( L i − μ ) − 1 = − 1 μ A i + B i ( μ ) ,
where A i denotes the orthogonal projection onto the eigenfunction ψ i in L 2 ( i ) ; its action is given by
A i h : = ( h , ψ i ) L 2 ( l i ) ∥ ψ i ∥ 2 L 2 ( l i ) ψ i .
The operator B i ( μ ) : L 2 ( i ) → W 2 2 ( i ) is bounded, and its dependence on μ is holomorphic in a neighborhood of μ = 0 . Moreover, the image of B i ( μ ) lies entirely in the domain D ( L i ) .
We further introduce the notation
z i : = ( L i − ε 2 λ ) − 1 f l i , z i = z i ( ξ i , ε ) .
We examine the behavior of the fundamental system of solutions to the differential equation
− d 2 v i d ξ i 2 + W v − μ v = 0 on l i
Assuming that the functions φ i and ψ i are known for all i = 1 , … , n , Equation (15) can be reformulated, in the usual manner, as a Volterra integral equation of the second kind:
v = μ M i v + c 1 ψ + c 2 φ , ( M i v ) ( x ) : = ψ ( x ) ∫ 0 x φ ( t ) h ( t ) d t − φ ( x ) ∫ 0 x ψ ( t ) h ( t ) d t ,
where c 1 and c 2 are arbitrary constants. The operator M i is clearly bounded when regarded as a mapping from C ( l i ¯ ) into C 2 ( l i ¯ ) . Consequently, the operator ( I − μ M i ) − 1 , with I denoting the identity operator, is well defined for sufficiently small μ and depends holomorphically on μ as an operator C ( l i ¯ ) → C 2 ( l i ¯ ) . This ensures that Equation (16) has a unique solution in C 2 ( l i ¯ ) whenever μ is sufficiently small. We denote by Ψ i = Ψ i ( ξ , μ ) the solution corresponding to c 1 = 1 and c 2 = 0 , and by Φ i = Φ i ( ξ , μ ) the one associated with c 1 = 0 and c 2 = 1 , that is Both functions Ψ i and Φ i depend holomorphically on μ for sufficiently small values of the parameter, when considered in the C 2 ( l i ¯ ) -norm. Moreover, they satisfy the identities
Ψ i ( ξ , 0 ) = ψ i ( ξ ) , Φ i ( ξ , 0 ) = ϕ i ( ξ ) .
Using these observations, the general solution of Equation (11) may be written in the form
v i ( ξ i , ε ) = ε 2 z i ( ξ i , ε ) + C Ψ i ( ε ) Ψ i ( ξ i , ε 2 λ ) + C Φ i ( ε ) Φ i ( ξ i , ε 2 λ ) .
We now examine Equation (10) restricted to the subgraph Γ of Γ ε . On Γ, this equation reduces to the differential problem
− d 2 u d x 2 + V u − λ u = f on Γ
supplemented with the prescribed boundary conditions at every vertex of Γ, except at the vertex M. At this vertex the function u is required to satisfy the Kirchhoff condition, modified to account for the presence of the small loop Γ. Because the resolvent ( H D − λ ) − 1 is well defined for the chosen value of λ , the function u D : = ( H D − λ ) − 1 f Γ solves Equation (19) and satisfies all the boundary conditions on Γ, except that it vanishes at the vertex M. Consequently, u D , ε belongs to the space W ˙ 2 2 ( Γ ) .
Let χ be a smooth cut-off function defined on the edge e 0 which equals 1 in a small neighborhood of the vertex M and vanishes outside a sufficiently large neighborhood of M. Extend χ by zero to all other edges of Γ, and introduce
U D : = χ − ( H D − λ ) − 1 − d 2 d x 2 + V − λ χ .
Using the standard mapping properties of the resolvent ( H D − λ ) − 1 , one verifies that U D ∈ W ˙ 2 2 ( Γ ) , satisfies all vertex boundary conditions on Γ except at M, and in addition solves the boundary value problem
− d 2 U D d x 2 + V U D − λ U D = 0 on Γ , U D ( M ) = 1
We observe that
U D ′ ( M ) ≠ 0 ,
since, otherwise, λ (having a nonzero imaginary part) would have to be an eigenvalue of the self-adjoint operator H N , which is impossible.
Using the known properties of u D and U D , consider the function w ( x , ε ) = u D ( x ) + K ( ε ) U D ( x ) , where K ( ε ) is an arbitrary constant. This function lies in the space W ˙ 2 2 ( Γ ) , solves Equation (19), and satisfies all vertex boundary conditions on the graph Γ except at the vertex M. At M the function w obeys
w ( M , ε ) = K ( ε ) , w ′ ( x , ε ) = u D ′ ( x ) + K ( ε ) U D ′ ( M )
We seek a solution of Equation (10) in the form
u ε = w , on Γ S i ( ε ) v i , on l i , ε
where the functions w and v i are given by relations (18) and (21).
Due to the previously established properties of these functions, the constructed u ε belongs to the space W ˚ 2 2 ( Γ ε ) and satisfies the differential equation
− d 2 u ε d x 2 + V u ε − λ u ε = f on Γ ε
together with all vertex boundary conditions on Γ ε except at the vertex M. In what follows, we demonstrate that an appropriate choice of the unknowns C Ψ i and C Φ i ensures that the Kirchhoff condition (1) is fulfilled at M.
Using the first set of identities in (1) together with (21), while also accounting for the boundary conditions satisfied by z i , we obtain
K ( ε ) = C Ψ i ( ε 2 λ ) Ψ i ( 0 , ε 2 λ ) + C Φ i ( ε 2 λ ) Φ i ( 0 , ε 2 λ ) = = C Ψ i ( ε 2 λ ) Ψ i ( a i , ε 2 λ ) + C Φ i ( ε 2 λ ) Φ i ( a i , ε 2 λ )
Combining the first identity in (23) with relations (23) and (25), we immediately conclude that
C Φ 1 ( ε ) = … = C Φ n ( ε ) = K ( ε ) .
The second equality in (23) enables us to express each of the quantities C Ψ i and C Φ i in terms of the other. To proceed, we analyze the asymptotic behavior of the differences Φ i ( a i , μ ) − Φ i ( 0 , μ ) , Ψ i ( a i , μ ) − Ψ i ( 0 , μ ) , as μ → 0 . It is convenient to present the required properties in the following form.
Lemma 1.
Each of the functions Ψ i ( 0 , μ ) , Ψ i ′ ( 0 , μ ) , Ψ i ( a i , μ ) , Ψ i ′ ( a i , μ ) , Φ i ( 0 , μ ) , Φ i ′ ( 0 , μ ) , Φ i ( a i , μ ) , and Φ i ′ ( a i , μ ) is holomorphic for sufficiently small values of μ. Moreover, the following relations are valid:
Ψ i ( 0 , μ ) = 0 , Ψ i ′ ( 0 , μ ) = 1 , Φ i ( 0 , μ ) = 1 , Φ i ′ ( 0 , μ ) = ϕ i ′ ( 0 ) ,
Ψ i ( a i , μ ) = μ ∥ Ψ ∥ L 2 ( l i ) 2 + O ( μ 2 ) ,
Φ i ( a i , μ ) = 1 + μ ( ψ i , ϕ i ) L 2 ( l i ) + O ( μ 2 ) ,
Ψ i ′ ( a i , μ ) = 1 + O ( μ 2 ) , Φ i ′ ( 0 , μ ) = ϕ i ′ ( a i ) + O ( μ 2 ) .
Proof. 
The analyticity of all listed functions follows directly from the fact that Ψ i and Φ i depend holomorphically on μ in the C 2 ( l i ¯ ) -norm. If we expand ( I − μ M i ) − 1 in (17) into its Neumann series, we obtain
Ψ i = ψ i − μ M i ψ i + μ 2 M i ( I − μ M i ) − 1 ψ i , Φ i = ϕ i − μ M i ϕ i + μ 2 M i ( I − μ M i ) − 1 ϕ i
Using these representations, together with the explicit form of the operator M i in (16) and the boundary conditions stated in (2)–(4), one directly derives all expansions and identities collected in (25).
From the second relation in (23), we can rewrite the coefficient C Ψ i in terms of C Φ i as
C Ψ i ( ε ) = C Φ i ( ε ) Φ i ( a i , ε 2 λ ) − Φ i ( 0 , ε 2 λ ) Ψ i ( 0 , ε 2 λ ) − Ψ i ( a i , ε 2 λ ) , i = 1 , … , n .
Using the expansions in (25), one sees that the map
μ ↦ Φ i ( a i , μ ) − Φ i ( 0 , μ ) Ψ i ( 0 , μ ) − Ψ i ( a i , μ )
is well-defined for small μ and is holomorphic in this parameter.
Furthermore, the leading terms of its Taylor expansion are given by
Φ i ( a i , μ ) − Φ i ( 0 , μ ) Ψ i ( 0 , μ ) − Ψ i ( a i , μ ) = − ( ϕ i , ψ i ) L 2 ( l i ) ∥ ψ i ∥ L 2 ( l i ) + O ( μ ) .
To obtain the second identity, we make use of the Kirchhoff condition (1). Substituting the representation of u ε from (22) and using (21), the definition of the operators S i ( ε ) , as well as Formula (18), we arrive at
ε u D ′ ( M , ε ) + ε K ( ε ) U D ′ ( ε ) = = ∑ i = 1 n ( z i ′ ( a i ) − z i ′ ( 0 ) ) + ∑ i = 1 n C Ψ i ( ε ) ( Ψ i ′ ( a i , ε 2 λ ) − Ψ i ′ ( 0 , ε 2 λ ) ) + + ∑ i = 1 n C Φ i ( ε ) ( Φ i ′ ( a i , ε 2 λ ) − Φ i ′ ( 0 , ε 2 λ ) )
By substituting the obtained formulas for C Ψ i and C Φ i into the above identity, we arrive at an equation that determines K. Solving this equation yields
K ( ε ) = ε u ′ D ( M , ε ) − ∑ i = 1 n ( z i ′ ( a i ) − z i ′ ( 0 ) ) Q ( ε 2 λ ) − ε U D ′ ( M ) ,
where
Q ( μ ) : = ∑ i = 1 n 1 Ψ i ( a i , μ ) − Ψ i ( 0 , μ ) det Ψ i ( a i , μ ) − Ψ i ( 0 , μ ) Φ i ( a i , μ ) − Φ i ( 0 , μ ) Ψ i ′ ( a i , μ ) − Ψ i ′ ( 0 , μ ) Φ i ′ ( a i , μ ) − Φ i ′ ( 0 , μ ) .
Using the expansions in (25), we observe that the function Q is holomorphic in μ , and its Taylor expansion at μ = 0 begins with
Q ( μ ) = q + O ( μ ) .
To analyze the behavior of the differences z i ′ ( a i , ε ) − z i ′ ( 0 , ε ) for i = 1 , … , n , we multiply the differential equation satisfied by z i by ϕ i and integrate by parts twice, making use of the boundary conditions for both z i and ϕ i . This yields
z i ′ ( a i , ε ) − z i ′ ( 0 , ε ) = − ε 2 ( f i + ε 2 λ z i , ϕ i ) L 2 ( i ) , i = 1 , … , n
Invoking representations (12) and (13), the above identity can be rewritten as
z i ′ ( a i , ε ) − z i ′ ( 0 , ε ) = − ε 2 ( B i ( ε 2 λ ) f l i + A i ⊥ f i , ϕ i ) L 2 ( l i )
where A i ⊥ = I − A i
Using the last identity together with (28), (24), (26), and (14), we can finally obtain the expressions for C Φ i , C Ψ i , and K. Specifically, we have
K ( ε ) = ε u D ′ ( M , ε ) − ε 2 ∑ i = 1 n ( B i ( ε 2 λ ) f l i + A i ⊥ f i , ϕ i ) L 2 ( l i ) Q ( ε 2 λ ) − ε U D ′ ( M ) ,
and therefore
C Φ i ( ε ) = K ( ε ) , C Ψ i ( ε ) = Φ i ( a i , ε 2 λ ) − Φ i ( 0 , ε 2 λ ) Ψ i ( 0 , ε 2 λ ) − Ψ i ( a i , ε 2 λ ) · ε u D ′ ( M ) + ε 2 ∑ i = 1 n ( B i ( ε 2 λ ) f l i + A i ⊥ f i , φ i ) L 2 ( l i ) Q ( ε 2 λ ) − ε U D ′ ( M ) .
Since the operator B i ( μ ) and the functions Q , Φ i , Ψ i , Φ i ′ , Ψ i ′ are holomorphic in μ , the functions K , C Ψ i , C Φ i are meromorphic in ε . They can therefore be interpreted as the values of meromorphic bounded linear functionals in ε , acting on the space L 2 ( Γ ) ⊕ L 2 ( l ) . We denote these functionals by the same symbols C Ψ i and K. Namely,
K ( ε ) ( f Γ , f l ) : = K ( ε ) = C Φ i ( ε ) , C Ψ i ( ε ) ( f Γ , f l ) : = C Ψ i ( ε ) , f l : = ( f 1 , … , f l n ) .
Using Formulas (27) and (29) and inequality (20), we obtain that the functionals K and C Ψ i are holomorphic in ε , and the first terms of their Taylor expansions satisfy
K ( ε ) ( f Γ , f l ) = ( ( H D − λ ) − 1 f Γ ) ′ ( M ) q + O ( ε 2 ) ,
C Ψ i ( ε ) ( f Γ , f l ) = − ε ( ϕ i , ψ i ) L 2 ( M ) q ∥ ϕ i ∥ L 2 ( M i ) 2 ( ( H D − λ ) − 1 f Γ ) ′ ( M ) + O ( ε 2 )
if condition (6) holds, whereas
K ( ε ) ( f Γ , f l ) = ( ( H D − λ ) − 1 f Γ ) ′ ( M ) U D ′ ( M ) + O ( ε ) ,
C Ψ i ( ε ) ( f Γ , f l ) = ( ϕ i , ψ i ) L 2 ( M ) ∥ ϕ i ∥ L 2 ( M i ) 2 ( ( H D − λ ) − 1 f Γ ) ′ ( M ) U D ′ ( M ) + O ( ε )
if condition (8) holds. Here the remainder terms are understood in the norms of bounded linear functionals on L 2 ( Γ ) ⊕ L 2 ( l ) .
Thus, the functions K ( ε ) , C Ψ i ( ε ) and C Φ i ( ε ) have been determined, which confirms the validity of the Kirchhoff condition (1). As a result, the function u ε introduced in (22) indeed represents the action of the resolvent ( H ε − λ ) − 1 on the function f. Using the definitions of the operators R Γ ( ε , λ ) and R l ( ε , λ ) given in (5), we directly obtain w ε = R Γ ( ε , λ ) ( f Γ , f l ) , v ε = R l ( ε , λ ) ( f Γ , f l ) . Substituting (12) into (14), then inserting the resulting identity into (18), and taking into account the known properties of the functions Φ and Ψ as well as those of the functionals K and C Ψ i ( ε ) , we derive explicit formulas for these operators. Consequently,
R Γ ( ε , λ ) ( f Γ , f l ) = ( H D − λ ) − 1 f Γ + + ε ( ( H D − λ ) − 1 f Γ ) ′ ( M ) + ε 2 ∑ i = 1 n ( B i ( ε 2 λ ) f l i + A i ⊥ f i , φ i ) L 2 ( l i ) Q ( ε 2 λ ) − ε U D ′ ( M ) U D ′ R Γ ( ε , λ ) ( f Γ , f l ) = − − ⨁ i = 1 n ( f l i , ψ i ) L 2 ( l i ) λ ∥ ψ i ∥ L 2 ( l i ) 2 ψ i + ε 2 ⨁ i = 1 n B i ( ε 2 λ ) f l i + + ε ( ( H D − λ ) − 1 f Γ ) ′ ( M ) + ε 2 ∑ i = 1 n ( B i ( ε 2 λ ) f l i + A i ⊥ f i , φ i ) L 2 ( l i ) Q ( ε 2 λ ) − ε U D ′ ( M ) ⨁ i = 1 n Υ i ( · , ε 2 λ ) ,
where L 2 ( l ) = ⨁ i = 1 n L 2 ( l i ) , and
Υ i ( ξ i , μ ) : = Φ i ( a i , μ ) − Φ i ( 0 , μ ) Ψ i ( 0 , μ ) − Ψ i ( a i , μ ) Ψ i ( ξ , ε 2 λ ) + Φ i ( ξ i , ε 2 λ ) .
From the explicit representations obtained above, it follows that the operators R Γ ( ε , λ ) and R l ( ε , λ ) depend holomorphically on the parameter ε . Making use of identities (29) and (25), together with the elementary equalities
( H D − λ ) − 1 f Γ − ( ( H D − λ ) − 1 f Γ ) ′ ( M ) U D ′ ( M ) U D = ( H N − λ ) − 1 f Γ , φ i ⊥ ( ξ i ) = Υ i ( ξ , 0 ) ,
one can directly check that the leading terms of the Taylor expansions of these operators agree with Formula (7) when condition (6) is satisfied, and with Formula (9) when condition (8) is fulfilled. This completes the proof of the theorem. □

3. Discussion: Comparison with the Nonresonant Case and an Illustrative Example

3.1. Comparison with the Nonresonant Case

We briefly compare the results obtained in the present (resonant) setting with those established previously in the nonresonant case.
In the nonresonant situation, the eigenvalues involved in the asymptotic construction do not satisfy any resonance relations. Consequently, the perturbation procedure proceeds in a regular manner: at each step of the asymptotic expansion, the correction terms are uniquely determined by solving linear problems with nonvanishing spectral gaps. No additional compatibility conditions arise, and the amplitudes corresponding to different modes remain decoupled. In particular, the leading-order term preserves the qualitative structure of the associated linear problem, and the reduced model retains a scalar (or diagonally decoupled) character.
In contrast, in the resonant configuration considered in this paper, certain eigenvalues satisfy resonance relations. As a result, the higher-order equations generate additional solvability (orthogonality) conditions. These conditions lead to a coupled amplitude system governing several interacting modes simultaneously. The asymptotic construction can no longer be performed mode by mode independently, and the structure of the leading-order term is modified by nonlinear interaction effects.
Therefore, unlike the nonresonant case, the resonant regime produces a qualitatively different limiting behavior and alters the structure of the reduced problem. This distinction underlines the mathematical novelty of the present analysis.

3.2. A Simple Illustrative Example

To clarify the abstract construction underlying the main results, we present a simplified model example illustrating the difference between the nonresonant and resonant mechanisms.
Consider the perturbed problem
L u = ε N ( u ) ,
where L is a linear self-adjoint operator with eigenvalues λ 1 and λ 2 and corresponding eigenfunctions ϕ 1 and ϕ 2 , and ε > 0 is a small parameter. We seek a solution in the form
u = a 1 ϕ 1 + a 2 ϕ 2 + ε u 1 + O ( ε 2 ) .

3.2.1. Nonresonant Case

If λ 1 ≠ λ 2 , substitution of the expansion and projection onto the eigenspaces generated by ϕ 1 and ϕ 2 yield independent equations for the amplitudes a 1 and a 2 . Since the eigenvalues are separated, the associated linear operators are invertible on the orthogonal complements, and no additional compatibility conditions arise. The correction term u 1 is uniquely determined, and the construction proceeds in a decoupled manner.

3.2.2. Resonant Case

If λ 1 = λ 2 , the projection of the higher-order terms onto the resonant eigenspace leads to compatibility conditions of the form
⟨ N ( a 1 ϕ 1 + a 2 ϕ 2 ) , ϕ i ⟩ = 0 , i = 1 , 2 .
These conditions produce a coupled system for the amplitudes a 1 and a 2 . The interacting modes must therefore be determined simultaneously, and the reduced problem acquires a genuinely coupled structure. This mechanism reflects, in a simplified form, the phenomenon responsible for the qualitative differences established in the main results of this paper.

4. Conclusions

We have analyzed the resolvent of a Schrödinger operator on a graph with small loops in the resonant regime, where the standard nonresonance condition is violated. The results show that the relevant components of the resolvent admit a holomorphic expansion in the small parameter ϵ, with an additional leading-order term arising from eigenfunctions localized on the shrinking loops. This term reveals a threshold localization effect associated with embedded eigenvalues or virtual levels and highlights the influence of small-scale geometry on the operator.
Although the present work focuses on the resolvent level, the obtained results provide a solid foundation for further investigations of spectral and dynamical properties of quantum graphs with small edges. In particular, they suggest possible extensions to the study of threshold resonances, scattering behavior, and related time-dependent problems. We expect that the approach developed here may be useful in addressing these questions in more general graph settings.

Author Contributions

Software, G.A.; Validation, M.K.; Formal analysis, G.A.; Investigation, M.K.; Data curation, M.K.; Writing—original draft, M.K. and G.A.; Writing—review & editing, M.K. and G.A.; Project administration, M.K. and G.A.; Funding acquisition, M.K. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by the Science Committee of the Ministry of Education and Science of the Republic of Kazakhstan (Grant No. AP22685191).

Data Availability Statement

The original contributions presented in this study are included in the article. Further inquiries can be directed to the corresponding author.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Kurasov, P.; Nowaczyk, M. Inverse spectral problem for quantum graphs. J. Phys. A Math. Gen. 2005, 38, 4901–4915. [Google Scholar] [CrossRef] [Scilit]
  2. Kuchment, P. Quantum graphs: An introduction and a brief survey. Proc. Symp. Pure Math. 2008, 77, 291–314. [Google Scholar]
  3. Kostrykin, V.; Schrader, R. Kirchhoff’s rule for quantum wires. J. Phys. A Math. Gen. 1999, 32, 595–630. [Google Scholar] [CrossRef] [Scilit]
  4. Bolte, J.; Endres, S. Trace formulae for quantum graphs. Ann. Henri Poincaré 2009, 10, 189–223. [Google Scholar] [CrossRef] [Scilit]
  5. Gnutzmann, S.; Smilansky, U. Quantum graphs: Applications to quantum chaos and universal spectral statistics. Adv. Phys. 2006, 55, 527–625. [Google Scholar] [CrossRef] [Scilit]
  6. Post, O. Spectral Analysis on Graph-like Spaces; Lecture Notes in Mathematics; Springer: Berlin/Heidelberg, Germany, 2012. [Google Scholar]
  7. Kuchment, P.; Zeng, H. Asymptotics of spectra of Neumann Laplacians on thin domains. Contemp. Math. 2003, 327, 199–213. [Google Scholar]
  8. Berkolaiko, G.; Keating, J. Two-point spectral correlations for star graphs. J. Phys. A Math. Gen. 1999, 32, 7827–7841. [Google Scholar] [CrossRef] [Scilit]
  9. Friedlander, L. Genericity of simple eigenvalues for a metric graph. Israel J. Math. 2005, 146, 149–156. [Google Scholar] [CrossRef] [Scilit]
  10. Molchanov, S.; Vainberg, B. Scattering solutions in networks of thin fibers: Small diameter asymptotics. Commun. Math. Phys. 2007, 273, 533–559. [Google Scholar] [CrossRef] [Scilit]
  11. Exner, T.; Post, O. Convergence of spectra of graph-like thin manifolds. J. Geom. Phys. 2005, 54, 77–115. [Google Scholar] [CrossRef] [Scilit]
  12. Kuchment, P.; Post, O. On the spectra of carbon nano-structures. Commun. Math. Phys. 2007, 275, 805–826. [Google Scholar] [CrossRef] [Scilit]
  13. Alon, L.; Band, R.; Berkolaiko, G. Universality of nodal count distribution in large metric graphs. Exp. Math. 2022, 33, 301–335. [Google Scholar] [CrossRef] [Scilit]
  14. Barra, F.; Gaspard, P. Transport and dynamics on open quantum graphs. Phys. Rev. E 2002, 65, 016205. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  15. Post, O. Spectral convergence of quasi-one-dimensional spaces. Ann. Henri Poincaré 2006, 7, 933–973. [Google Scholar] [CrossRef] [Scilit]
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Article Metrics

Citations

Article Access Statistics

Multiple requests from the same IP address are counted as one view.