# A remark on the a-minimally thin sets associated with the Schrödinger operator

Gaixian Xue

Author Affiliations

School of Mathematics and Information Science, Henan University of Economics and Law, Zhengzhou, 450046, China

Boundary Value Problems 2014, 2014:133  doi:10.1186/1687-2770-2014-133

 Received: 23 February 2014 Accepted: 29 April 2014 Published: 23 May 2014

This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited.

### Abstract

The aim of this paper is to give a new criterion for a-minimally thin sets at infinity with respect to the Schrödinger operator in a cone, which supplement the results obtained by T. Zhao.

##### Keywords:
minimally thin set; Schrödinger operator; Green a-potential

### 1 Introduction and results

Let R and R + be the set of all real numbers and the set of all positive real numbers, respectively. We denote by R n ( n 2 ) the n-dimensional Euclidean space. A point in R n is denoted by P = ( X , x n ) , X = ( x 1 , x 2 , , x n 1 ) . The Euclidean distance between two points P and Q in R n is denoted by | P Q | . Also | P O | with the origin O of R n is simply denoted by | P | . The boundary and the closure of a set S in R n are denoted by ∂S and S ¯ , respectively. Further, intS, diamS, and dist ( S 1 , S 2 ) stand for the interior of S, the diameter of S, and the distance between S 1 and S 2 , respectively.

We introduce a system of spherical coordinates ( r , Θ ) , Θ = ( θ 1 , θ 2 , , θ n 1 ) , in R n which are related to cartesian coordinates ( x 1 , x 2 , , x n 1 , x n ) by x n = r cos θ 1 .

Let D be an arbitrary domain in R n and A a denote the class of non-negative radial potentials a ( P ) , i.e. 0 a ( P ) = a ( r ) , P = ( r , Θ ) D , such that a L loc b ( D ) with some b > n / 2 if n 4 and with b = 2 if n = 2 or n = 3 (see [[1], p.354] and [2]).

If a A a , then the stationary Schrödinger operator

S c h a = Δ + a ( P ) I = 0 ,

where Δ is the Laplace operator and I is the identical operator, can be extended in the usual way from the space C 0 ( D ) to an essentially self-adjoint operator on L 2 ( D ) (see [[1], Ch. 11]). We will denote it S c h a as well. This last one has a Green a-function G D a ( P , Q ) . Here G D a ( P , Q ) is positive on D and its inner normal derivative G D a ( P , Q ) / n Q 0 , where / n Q denotes the differentiation at Q along the inward normal into D.

We call a function u that is upper semi-continuous in D a subfunction with respect to the Schrödinger operator S c h a if its values belong to the interval [ , ) and at each point P D with 0 < r < r ( P ) we have the generalized mean-value inequality (see [[1], Ch. 11])

u ( P ) S ( P , r ) u ( Q ) G B ( P , r ) a ( P , Q ) n Q d σ ( Q )

satisfied, where G B ( P , r ) a ( P , Q ) is the Green a-function of S c h a in B ( P , r ) and d σ ( Q ) is a surface measure on the sphere S ( P , r ) = B ( P , r ) . If −u is a subfunction, then we call u a superfunction (with respect to the Schrödinger operator S c h a ).

The unit sphere and the upper half unit sphere in R n are denoted by S n 1 and S + n 1 , respectively. For simplicity, a point ( 1 , Θ ) on S n 1 and the set { Θ ; ( 1 , Θ ) Ω } for a set Ω, Ω S n 1 , are often identified with Θ and Ω, respectively. For two sets Ξ R + and Ω S n 1 , the set { ( r , Θ ) R n ; r Ξ , ( 1 , Θ ) Ω } in R n is simply denoted by Ξ × Ω . By C n ( Ω ) , we denote the set R + × Ω in R n with the domain Ω on S n 1 . We call it a cone. We denote the set I × Ω with an interval on R by C n ( Ω ; I ) .

From now on, we always assume D = C n ( Ω ) . For the sake of brevity, we shall write G Ω a ( P , Q ) instead of G C n ( Ω ) a ( P , Q ) . We shall also write g 1 g 2 for two positive functions g 1 and g 2 , if and only if there exists a positive constant c such that c 1 g 1 g 2 c g 1 .

Let Ω be a domain on S n 1 with smooth boundary. Consider the Dirichlet problem

( Λ n + λ ) φ = 0 on  Ω , φ = 0 on  Ω ,

where Λ n is the spherical part of the Laplace operata Δ n

Δ n = n 1 r r + 2 r 2 + Λ n r 2 .

We denote the least positive eigenvalue of this boundary value problem by λ and the normalized positive eigenfunction corresponding to λ by φ ( Θ ) . In order to ensure the existence of λ and a smooth φ ( Θ ) , we put a rather strong assumption on Ω: if n 3 , then Ω is a C 2 , α -domain ( 0 < α < 1 ) on S n 1 surrounded by a finite number of mutually disjoint closed hypersurfaces (e.g. see [[3], pp.88-89] for the definition of C 2 , α -domain).

For any ( 1 , Θ ) Ω , we have (see [[4], pp.7-8])

δ ( P ) r φ ( Θ ) , (1)

where P = ( r , Θ ) C n ( Ω ) and δ ( P ) = dist ( P , C n ( Ω ) ) .

Solutions of an ordinary differential equation (see [[5], p.217])

Q ( r ) n 1 r Q ( r ) + ( λ r 2 + a ( r ) ) Q ( r ) = 0 , 0 < r < . (2)

It is well known (see, for example, [6]) that if the potential a A a , then equation (2) has a fundamental system of positive solutions { V , W } such that V and W are increasing and decreasing, respectively.

We will also consider the class B a , consisting of the potentials a A a such that there exists the finite limit lim r r 2 a ( r ) = k [ 0 , ) , and, moreover, r 1 | r 2 a ( r ) k | L ( 1 , ) . If a B a , then the (sub)superfunctions are continuous (see [7]). In the rest of paper, we assume that a B a and we shall suppress this assumption for simplicity.

Denote

ι k ± = 2 n ± ( n 2 ) 2 + 4 ( k + λ ) 2 ,

then the solutions to equation (2) have the asymptotic (see [3])

V ( r ) r ι k + , W ( r ) r ι k , as  r . (3)

It is well known that the Martin boundary of C n ( Ω ) is the set C n ( Ω ) { } , each of which is a minimal Martin boundary point. For P C n ( Ω ) and Q C n ( Ω ) { } , the Martin kernel can be defined by M Ω a ( P , Q ) . If the reference point P is chosen suitably, then we have

M Ω a ( P , ) = V ( r ) φ ( Θ ) and M Ω a ( P , O ) = c W ( r ) φ ( Θ ) , (4)

for any P = ( r , Θ ) C n ( Ω ) .

In [[8], p.67], Zhao introduce the notations of a-thin (with respect to the Schrödinger operator S c h a ) at a point, a-polar set (with respect to the Schrödinger operator S c h a ) and a-minimal thin sets at infinity (with respect to the Schrödinger operator S c h a ). A set H in R n is said to be a-thin at a point Q if there is a fine neighborhood E of Q which does not intersect H { Q } . Otherwise H is said to be not a-thin at Q on C n ( Ω ) . A set H in R n is called a polar set if there is a superfunction u on some open set E such that H { P E ; u ( P ) = } . A subset H of C n ( Ω ) is said to be a-minimal thin at Q C n ( Ω ) { } on C n ( Ω ) , if there exists a point P C n ( Ω ) such that

R ˆ M Ω a ( , Q ) H ( P ) M Ω a ( P , Q ) ,

where R ˆ M Ω a ( , Q ) H is the regularized reduced function of M Ω a ( , Q ) relative to H (with respect to the Schrödinger operator S c h a ).

Let H be a bounded subset of C n ( Ω ) . Then R ˆ M Ω a ( , ) H ( P ) is bounded on C n ( Ω ) and hence the greatest a-harmonic minorant of R ˆ M Ω a ( , ) H is zero. When by G Ω a μ ( P ) we denote the Green a-potential with a positive measure μ on C n ( Ω ) , we see from the Riesz decomposition theorem that there exists a unique positive measure λ H a on C n ( Ω ) such that

R ˆ M Ω a ( , ) H ( P ) = G Ω a λ H a ( P )

for any P C n ( Ω ) and λ H a is concentrated on I H , where

I H = { P C n ( Ω ) ; H  is not a-thin at  P } .

The Green a-energy γ Ω a ( H ) (with respect to the Schrödinger operator S c h a ) of λ H a is defined by

γ Ω a ( H ) = C n ( Ω ) G Ω a λ H a d λ H a .

Also, we can define a measure σ Ω a on C n ( Ω )

σ Ω a ( H ) = H ( M Ω a ( P , ) δ ( P ) ) 2 d P .

In [[8], Theorem 5.4.3], Long gave a criterion that characterizes a-minimally thin sets at infinity in a cone.

Theorem AA subsetHof C n ( Ω ) is a-minimally thin at infinity on C n ( Ω ) if and only if

j = 0 γ Ω a ( H j ) W ( 2 j ) V 1 ( 2 j ) < ,

where H j = H C n ( Ω ; [ 2 j , 2 j + 1 ) ) and j = 0 , 1 , 2 ,  .

In recent work, Zhao (see [[2], Theorems 1 and 2]) proved the following results. For similar results in the half space with respect to the Schrödinger operator, we refer the reader to the papers by Ren and Su (see [9,10]).

Theorem BThe following statements are equivalent.

(I) A subsetHof C n ( Ω ) is a-minimally thin at infinity on C n ( Ω ) .

(II) There exists a positive superfunction v ( P ) on C n ( Ω ) such that

inf P C n ( Ω ) v ( P ) M Ω a ( P , ) = 0 (5)

and

H { P C n ( Ω ) ; v ( P ) M Ω a ( P , ) } .

(III) There exists a positive superfunction v ( P ) on C n ( Ω ) such that even if v ( P ) c M Ω a ( P , ) for any P H , there exists P 0 C n ( Ω ) satisfying v ( P 0 ) < c M Ω a ( P 0 , ) .

Theorem CIf a subsetHof C n ( Ω ) is a-minimally thin at infinity on C n ( Ω ) , then we have

H d P ( 1 + | P | ) n < . (6)

Remark From equation (3), we immediately know that equation (6) is equivalent to

H V ( 1 + | P | ) W ( 1 + | P | ) ( 1 + | P | ) 2 d P < . (7)

This paper aims to show that the sharpness of the characterization of an a-minimally thin set in Theorem C. In order to do this, we introduce the Whitney cubes in a cone.

A cube is the form

[ l 1 2 j , ( l 1 + 1 ) 2 j ] × × [ l n 2 j , ( l n + 1 ) 2 j ] ,

where j, l 1 , , l n are integers. The Whitney cubes of C n ( Ω ) are a family of cubes having the following properties:

(I) k W k = C n ( Ω ) .

(II) int W j int W k = ( j k ).

(III) diam W k dist ( W k , R n C n ( Ω ) ) 4 diam W k .

Theorem 1IfHis a union of cubes from the Whitney cubes of C n ( Ω ) , then equation (7) is also sufficient forHto be a-minimally thin at infinity with respect to C n ( Ω ) .

From the Remark and Theorem 1, we have the following.

Corollary 1Let v ( P ) be a positive superfunction on C n ( Ω ) such that equation (5) holds. Then we have

{ P C n ( Ω ) ; v ( P ) M Ω a ( P , ) } V ( 1 + | P | ) W ( 1 + | P | ) ( 1 + | P | ) 2 d P < .

Corollary 2LetHbe a Borel measurable subset of C n ( Ω ) satisfying

H V ( 1 + | P | ) W ( 1 + | P | ) ( 1 + | P | ) 2 d P = + .

If v ( P ) is a non-negative superfunction on C n ( Ω ) andcis a positive number such that v ( P ) c M Ω a ( P , ) for all P H , then v ( P ) c M Ω a ( P , ) for all P C n ( Ω ) .

### 2 Lemmas

To prove our results, we need some lemmas.

Lemma 1Let W k be a cube from the Whitney cubes of C n ( Ω ) . Then there exists a constant cindependent ofksuch that

γ Ω a ( W k ) c σ Ω a ( W k ) .

Proof If we apply a result of Long (see [[8], Theorem 6.1.3]) for compact set W ¯ k , we obtain a measure μ on C n ( Ω ) , supp μ W ¯ k , μ ( W ¯ k ) = 1 such that

{ C n ( Ω ) | P Q | 2 n d μ ( Q ) = { Cap ( W ¯ k ) } 1 if  n 3 , C 2 ( Ω ) log | P Q | d μ ( Q ) = log Cap ( W ¯ k ) if  n = 2 (8)

for any P W ¯ k . Also there exists a positive measure λ W ¯ k a on C n ( Ω ) such that

R ˆ M Ω a ( , ) W ¯ k ( P ) = G Ω a λ W ¯ k a ( P ) (9)

for any P C n ( Ω ) .

Let P k = ( r k , Θ k ) , ρ k , t k be the center of W k , the diameter of W j , the distance between W k and C n ( Ω ) , respectively. Then we have ρ k t k 4 ρ k and ρ k r k . Then from equation (1) we have

r k M Ω a ( P , ) V ( r k ) ρ k (10)

for any P W ¯ k . We can also prove that

G Ω a ( P , Q ) { | P Q | 2 n if  n 3 , log ρ k | P Q | if  n = 2 (11)

for any P W ¯ k and any Q W ¯ k . Hence we obtain

r k λ W ¯ k a ( C n ( Ω ) ) { V ( r k ) ρ k Cap ( W ¯ k ) if  n 3 , V ( r k ) ρ k { log ρ k Cap ( W ¯ k ) } 1 if  n = 2 (12)

from equations (8), (9), (10), and (11). Since

γ Ω a ( W k ) = G Ω a λ W ¯ k a d λ W ¯ k a W ¯ k M Ω a ( P , ) d λ W ¯ k a ( P ) r k ι k + 1 ρ k λ W ¯ k a ( C n ( Ω ) )

from equations (3), (9), and (10), we have from (12)

γ Ω a ( W k ) { r k 2 ι k + 2 ρ k 2 Cap ( W ¯ k ) if  n 3 , r k 2 ι k + 2 ρ k 2 { log ρ k Cap ( W ¯ k ) } 1 if  n = 2 . (13)

Since

{ Cap ( W ¯ k ) ρ k n 2 if  n 3 , Cap ( W ¯ k ) ρ k if  n = 2 ,

we obtain from equation (13)

γ Ω a ( W k ) r k 2 ι k + 2 ρ k n . (14)

On the other hand, we have from equation (1)

σ Ω a ( W k ) r k 2 ι k + 2 ρ k n ,

which, together with equation (14), gives the conclusion of Lemma 1. □

### 3 Proof of Theorem 1

Let { W k } be a family of cubes from the Whitney cubes of C n ( Ω ) such that H = k W k . Let { W k , j } be a subfamily of { W k } such that W k , j ( H j 1 H j H j + 1 ) , where j = 1 , 2 , 3 ,  .

Since γ Ω a is a countably subadditive set function (see [[8], p.49]), we have

γ Ω a ( H j ) k γ Ω a ( W k , j ) (15)

for j = 1 , 2 ,  . Hence for j = 1 , 2 , we see from Lemma 1

k γ Ω a ( W k , j ) k σ Ω a ( W k , j ) , (16)

which, together with equation (1), gives

k σ Ω a ( W k , j ) ( H j 1 + H j + H j + 1 ) V 2 ( r ) r 2 d P ( H j 1 + H j + H j + 1 ) r 2 ( ι k + 1 ) d P r 2 ( j 1 ) ( ι k + 1 ) | H j 1 | + r 2 j ( ι k + 1 ) | H j | + r 2 ( j + 1 ) ( ι k + 1 ) | H j + 1 | (17)

for j = 1 , 2 ,  . Thus equations (15), (16), and (17) give

γ Ω a ( H j ) r 2 ( j 1 ) ( ι k + 1 ) | H j 1 | + r 2 j ( ι k + 1 ) | H j | + r 2 ( j + 1 ) ( ι k + 1 ) | H j + 1 |

for j = 1 , 2 ,  . Finally we obtain from equation (1)

j = 0 γ Ω a ( H j ) W ( 2 j ) V 1 ( 2 j ) γ Ω a ( H 0 ) + j = 0 2 j ( 2 ι k + 2 ) 2 j ( ι k + + ι k ) | H j | γ Ω a ( H 0 ) + j = 0 2 2 j W ( 2 j ) V 1 ( 2 j ) | H j | γ Ω a ( H 0 ) + H V ( 1 + | P | ) W ( 1 + | P | ) ( 1 + | P | ) 2 d P < ,

which shows with Theorem A that H is a-minimally thin at infinity with respect to C n ( Ω ) .

### Competing interests

The author declares that they have no competing interests.

### Acknowledgements

This work was supported by the National Natural Science Foundation of China under Grants Nos. 11301140 and U1304102.

### References

1. Levin, B, Kheyfits, A: Asymptotic behavior of subfunctions of time-independent Schrödinger operator. Some Topics on Value Distribution and Differentiability in Complex and P-Adic Analysis, pp. 323–397. Science Press, Beijing (2008)

2. Zhao, T: Minimally thin sets at infinity with respect to the Schrödinger operator. J. Inequal. Appl.. 2014, (2014) Article ID 67

3. Gilbarg, D, Trudinger, NS: Elliptic Partial Differential Equations of Second Order, Springer, Berlin (1977)

4. Courant, R, Hilbert, D: Methods of Mathematical Physics, Interscience, New York (2008)

5. Miranda, C: Partial Differential Equations of Elliptic Type, Springer, London (1970)

6. Verzhbinskii, GM, Maz’ya, VG: Asymptotic behavior of solutions of elliptic equations of the second order close to a boundary. I. Sib. Mat. Zh.. 12, 874–899 (1971)

7. Simon, B: Schrödinger semigroups. Bull. Am. Math. Soc.. 7, 447–526 (1982). Publisher Full Text

8. Long, PH: The Characterizations of Exceptional Sets and Growth Properties in Classical or Nonlinear Potential Theory. Dissertation of Beijing Normal University, Beijing (2012)

9. Ren, YD: Solving integral representations problems for the stationary Schrödinger equation. Abstr. Appl. Anal.. 2013, (2013) Article ID 715252

10. Su, BY: Dirichlet problem for the Schrödinger operator in a half space. Abstr. Appl. Anal.. 2012, (2012) Article ID 578197