Skip to main content

The estimation and profile of the critical value for a Schrödinger equation

Abstract

In this paper, we are concerned with the following Schrödinger problem: Δu+V(x)u=f(u), u>0, in R N , where f:RR is of class C 1 . The estimation and profile of the critical value of the corresponding functional is proved, which entails the relationship between the critical value on the balls and the least-energy value on the whole space. Our results are also true for three cases of the potential function V(x).

1 Introduction

The main subject of this paper is the following problem:

Δu+V(x)u=f(u),u>0, in  R N ,
(1.1)

where Δ= i = 1 N 2 / x i 2 is the Laplace operator. Compactness and noncompactness assumptions posed on the potential function V(x) are also discussed.

The nonlinear Schrödinger equation (1.1) serves as a model for various problems in physics. For the last 20 years, (1.1) has received considerable attention as its solutions seem both mathematically intriguing and scientifically useful. We would like to mention earlier results on the existence of entire solutions of Schrödinger type equations with or without potentials, which was studied in [1]–[3] (see references therein).

A more general form of nonlinearity, i.e.

Δu+V(x)u=f(x,u),u H 1 ( R N ) ,
(1.2)

is also studied by many authors. Kryszewski-Szulkin [4] considered the existence of a nontrivial solution of (1.2) in a situation when f(x,u), V(x) are periodic in the x-variables, f(x,u) is superlinear at u=0 and |u|=, and 0 lies in a spectral gap of Δ+V. In addition, if f(x,u) is odd in u, they proved that (1.2) has infinitely many solutions. The result from Bartsch-Wang [5] suggested that (1.2) should have one sign changing solution. Bartsch-Liu-Weth [6] further proved the existence of sign changing solutions of (1.2) in H 1 (R) with superlinear and subcritical nonlinearity f(x,u), and the number of nodal domains can be controlled. If f(x,u) is odd, they obtained an unbounded sequence of sign changing solutions u k (k1), and they have at most k+1 nodal domains.

For the power nonlinearity f(t)= t p (1<p< N + 2 N 2 ), since V(x)>0, it is well known that this problem has a positive solution which goes to zero at infinity. This solution is, besides, radially symmetric around some point and unique up to translations; see [7] and [8]. Moreover, the linearized equation around w is nondegenerate in the sense that the equation

ΔuV(x)u+p w p 1 u=0,in  R N ,

has linear combinations of the functions w/ x i as its only solutions which go to zero at infinity [8], [9]. These facts are crucial in the formulation of a Lyapunov-Schmidt type procedure, which was first introduced by Floer-Weinstein [10] for the one-dimensional case, and then was extended by Oh [11], [12] to higher dimensions. Ni-Wei [13] studied the critical value of the energy functional

E(u)= 1 2 Ω ( ε 2 | u | 2 + u 2 ) 1 p + 1 Ω u p + 1 ,u H 1 (Ω),

of the classical singular perturbation equation ε 2 Δu+u= u p (u>0) on Ω, where Ω is a bounded smooth domain in R N .

Throughout this paper the following hypotheses on f C 1 and V(x) will be assumed.

( f 1 ): f(0)=0, f(t)=o(|t|), as |t|0, uniformly in x.

( f 2 ): There exists a number p>1, with p< N + 2 N 2 if N3, such that lim | t | F ( t ) | t | p <+, where F(t)= 0 t f(s)ds.

( f 3 ): There exists a number μ>2 such that 0<μF(t)tf(t), for all |t|>0.

( f 4 ): For any x, f ( t ) t t f ( t ) t is nonincreasing in t.

( V 1 ): VC( R N ,R) and inf R N V(x)>0.

( V 2 ): V(x)=V(|x|).

Theorem 1.1

Under assumptions ( f 1 ) ( f 3 ) and ( V 1 ), there exists a least-energy ground state solution of (1.1).

Theorem 1.2

Under assumptions ( f 1 ) ( f 4 ) and ( V 2 ), there exists a least-energy ground state of (1.1), which is radially uniqueness solution, with the corresponding least-energy value c .

Remark 1.3

The assumptions ( V 1 ) and ( V 2 ) are adopted in [14], which means that the potential functions possess certain compactness conditions. ( V 2 ) indicates that the x dependence is radially symmetric. For this case, {u H 1 ( R N )| R N V(x) u 2 <,u is radial} is compactly embedded in L p ( R N ) (2p< 2 =2N/(N2)).

Remark 1.4

By standard variational arguments, the assumptions ( f 1 ) ( f 3 ) and ( V 1 ) guarantee the results of Theorem 1.1, which can be proved by the traditional Minmax Theory.

In fact, the critical value of the energy functional

I(u)= 1 2 R N ( | u | 2 + V ( x ) u 2 ) R N F(u),u H 1 ( R N )
(1.3)

can be characterized as

c = inf u 0 sup t > 0 I(tu).

The associated critical point actually solves (1.1) and is called a least-energy solution. It decays exponentially at infinity.

Remark 1.5

As far as we know, the most general result of uniqueness of (1.1) type is obtained by Serrin-Tang [15], which would guarantee radial uniqueness in (1.1) if additionally one assumes ( f 4 ). The proof of Theorem 1.2 is similar to the steps in [15]; we omit the details.

Now we define the energy functional of (1.1) on B ρ ={x R N ||x|ρ}:

I ρ (u)= 1 2 B ρ ( | u | 2 + V ( x ) u 2 ) B ρ F(u).
(1.4)

The main result of this paper is the following.

Theorem 1.6

Under assumptions ( f 1 ) ( f 4 ) and ( V 2 ), the critical value c ρ of the functional I ρ satisfies

c ρ = c + γ 2 exp ( 2 ρ ( 1 + o ( 1 ) ) ) ,
(1.5)

where c is the least-energy value in Theorem  1.2or Remark  1.4, γ is defined as

γ= 1 2 R N f(w)V>0,
(1.6)

where w is the unique solution in Theorem  1.2.

Remark 1.7

The assumptions ( f 1 ) ( f 4 ) and ( V 2 ) can guarantee the existence of c ρ on the bounded domain B ρ . In fact, it can be proved by the Minmax Theory as in Remark 1.4.

Remark 1.8

Assuming the conditions ( V 3 ) or ( V 4 ) on V(x), we can get a similar equality as (1.5).

( V 3 ): There exists r 0 >0 such that, for any M>0,

lim | y | m ( { x R N : | x y | r 0 } { x R N : V ( x ) M } ) =0,

where m denotes the Lebesgue measure on R N .

( V 4 ): VC( R N ,R), inf R N V(x)>0, and V(x) as |x|.

The assumptions ( V 3 ) and ( V 4 ) are certain compactness conditions, listed in [14]. ( V 3 ) is a more general condition, which gives a compact embedding. For ( V 4 ), we have a compact embedding from {u H 1 ( R N )| R N V(x) u 2 <} in L p ( R N ) for 2p< 2 .

Remark 1.9

There are two cases of noncompactness conditions that are posed on the potential functions V(x)[16], and the assumption ( V 6 ) is also adopted in [17]:

( V 5 ): VC( R N ,R), inf R N V(x) V 0 >0. V(x) is 1-periodic in each of x 1 , x 2 ,, x N .

( V 6 ): 0< inf R N V(x) lim x V(x)= sup R N V(x)<.

Assumption ( V 5 ) is periodic, i.e., the x-dependence is periodic. ( V 6 ) means that V(x) has a bounded potential well in the sense that lim | x | V(x) exists and is equal to sup R N V(x).

Remark 1.10

A particularly interesting case is whether one can come to the same conclusion as Theorem 1.6 under the noncompact assumptions ( V 5 ) and ( V 6 ).

Our theorems generalize the results in [18] to three cases of compactness potential function entailing a type of nonlinear Schrödinger equation. The existence of the least-energy ground state solution of (1.1) is essential. Our results show the relationship between the critical value on the balls and the least-energy value on the whole space. The estimation of the critical value can be used to locate the geometrical shape of the solution.

2 Preliminaries

In this section, we give some preliminary lemmas, which will be adopted in the proof of the theorems.

Lemma 2.1

Assume w is a solution of (1.1), and w ρ is a solution of

{ Δ u + V ( x ) u = f ( u ) , u > 0 , in B ρ , u = 0 , on B ρ .
(2.1)

Then

w ρ ( ρ 1 ) = γ exp ( ρ ( 1 + o ( 1 ) ) ) , w ( ρ 1 ) = γ exp ( ρ ( 1 + o ( 1 ) ) ) ,

where γ is defined in (1.6).

Proof

The proof of the two equalities is similar, we only prove the latter. Let w be the unique positive solution of (1.1), then the function

w ¯ =w ( x V ( x ) ) V ( x ) ln u ln ( u / f ( u ) )

satisfies the equation

Δu+u=f(u),u>0, in  R N .
(2.2)

Next we consider the solution of the equation

{ u ( 1 τ 2 ) u = 0 , in  ( R , ρ ) , u ( R ) = 1 , u ( ρ ) = 0 ,
(2.3)

where ρ>R.

Note s= 1 τ 2 , h= e s R e 2 s ρ s R ; then the solution of (2.3) is

u ¯ (x)= e 2 s ρ h e s x + 1 h e s x .
(2.4)

We have s1, as τ0. For ρ big enough,

1 h ( e s ρ s e s ρ + s ) e ( 1 τ ) ρ ,

so

e 2 s ρ h e s ( ρ 1 ) + 1 h e s ( ρ 1 ) e ( 1 τ ) ρ ,

that is,

u ¯ (ρ1) e ( 1 τ ) ρ .

The u ¯ in (2.4) is the supersolution of (2.2) on [R,ρ). So on [R,ρ), w ¯ u ¯ , we deduce that

w ( x V ( x ) ) V ( x ) ln u ln ( u / f ( u ) ) e ( 1 τ ) ρ ,

so

lim r w(r) e ( 1 τ ) r =γ>0,

where γ is defined in (1.6); then w(ρ1)γ e ( 1 τ ) ρ .

For the lower boundary estimation, given R>0, consider the equation

{ u + N 1 R u u = 0 , in  ( R , ρ ) , u ( R ) = w ρ ( R ) , u ( ρ ) = 0 .

Similarly to the computation of (2.3), we get, for ρ big enough, u ̲ (ρ1) e ( 1 τ ) ρ , and u ̲ is a subsolution of the above equation. So w ¯ u ̲ . Therefore, for ρ big enough, w( x V ( x ) )V ( x ) ln u ln ( u / f ( u ) ) e ( 1 τ ) ρ , so w(ρ1)γ e ρ ( 1 + o ( 1 ) ) . We conclude that w(ρ1)=γ e ρ ( 1 + o ( 1 ) ) . □

Lemma 2.2

Let u be a solution of

{ u u = 0 , in ( ρ 1 , ) , u ( ρ 1 ) = 1 , u ( + ) = 0 .
(2.5)

Let v be a solution of

{ v h ( ρ ) v = 0 , in ( ρ 1 , ρ ) , v ( ρ 1 ) = 1 , v ( ρ ) = 0 ,
(2.6)

whereh(ρ)is a function of ρ. Then v (ρ1) u (ρ1)1.

Proof

By computation u= e ρ 1 e x is a solution of (2.5), and u (ρ1)=1. Similarly, it can be checked that

v= e h ( ρ ) ρ e h ( ρ ) e h ( ρ ) e h ( ρ ) x + 1 e h ( ρ ) ρ ( e h ( ρ ) e h ( ρ ) ) e h ( ρ ) x

is a solution of (2.6), and

v = e h ( ρ ) ρ h ( ρ ) e h ( ρ ) e h ( ρ ) e h ( ρ ) x + h ( ρ ) e h ( ρ ) ρ ( e h ( ρ ) e h ( ρ ) ) e h ( ρ ) x .

So v (ρ1)0= u (ρ1)+1, i.e. v (ρ1) u (ρ1)1. □

3 The estimation of the critical value

This section is devoted to the proof of Theorem 1.6.

Proof of Theorem 1.6

c ρ is the critical value of the functional I ρ , c is the least-energy value of I(u).

First we find the upper bound of c ρ . Let v ρ be the solution of the equation

{ Δ u + V ( x ) u = 0 , in  B ρ B ρ 1 , u ( ρ 1 ) = w ( ρ 1 ) , u ( ρ ) = 0 ,

where w is the solution of (1.1) in Theorem 1.2, then

B ρ B ρ 1 V(x) v ρ 2 = B ρ B ρ 1 Δ v ρ v ρ = v ρ v ρ | B ρ 1 B ρ B ρ 1 | v ρ | 2 ,

so

B ρ B ρ 1 | v ρ | 2 + B ρ B ρ 1 V(x) v ρ 2 = v ρ v ρ | B ρ 1 .
(3.1)

Define

w ¯ ρ (r)={ w ( r ) , 0 r ρ 1 , v ρ ( r ) , ρ 1 r ρ .

We have

c ρ = I ρ ( w ρ )= max t 0 I ρ (t w ρ ) max t 0 I ρ (t w ¯ ρ )= I ρ ( t ρ w ¯ ρ ),
(3.2)

and

t ρ 1,as ρ.
(3.3)

Moreover, by the definition of w ¯ ρ ,

I ρ ( t ρ w ¯ ρ ) = 1 2 B ρ t ρ 2 ( | w ¯ ρ | 2 + V ( x ) w ¯ ρ 2 ) B ρ F ( t ρ w ¯ ρ ) = 1 2 B ρ 1 t ρ 2 ( | w ¯ ρ | 2 + V ( x ) w ¯ ρ 2 ) B ρ 1 F ( t ρ w ¯ ρ ) + ( 1 2 B ρ B ρ 1 t ρ 2 ( | w ¯ ρ | 2 + V ( x ) w ¯ ρ 2 ) B ρ B ρ 1 F ( t ρ w ¯ ρ ) ) 1 2 B ρ 1 t ρ 2 ( | w | 2 + V ( x ) w 2 ) B ρ 1 F ( t ρ w ) + 1 2 B ρ B ρ 1 t ρ 2 ( | v ρ | 2 + V ( x ) v ρ 2 ) c + t ρ 2 v ρ v ρ | B ρ 1 c + t ρ 2 v ρ ( r ) v ρ ( r ) | r = ρ 1 ,
(3.4)

where (3.1) is used in the second inequality.

Similar to the computation in Lemma 2.1, we have

v ρ (ρ1)γexp ( ρ ( 1 + o ( 1 ) ) ) ,
(3.5)
v ρ (ρ1)γexp ( ρ ( 1 + o ( 1 ) ) ) .
(3.6)

Combine (3.2), (3.3), (3.4), (3.5), and (3.6), then

c ρ c + γ 2 exp ( 2 ρ ( 1 + o ( 1 ) ) ) .

Next we find the lower bound of c ρ . Let v ¯ ρ be the solution of

{ Δ u + V ( r ) u = 0 , in  R N B ρ 1 , u ( ρ 1 ) = w ρ ( ρ 1 ) , u ( + ) = 0 ,

then

R N B ρ 1 V v ¯ ρ 2 = R N B ρ 1 Δ v ¯ ρ v ¯ ρ = v ¯ ρ v ¯ ρ | B ρ 1 R N B ρ 1 | v ¯ ρ | 2 ,
(3.7)

so

R N B ρ 1 | v ¯ ρ | 2 +V v ¯ ρ 2 = v ¯ ρ v ¯ ρ | B ρ 1 .

Define

w ˆ ρ (r)={ w ρ ( r ) , 0 r ρ 1 , v ¯ ρ ( r ) , ρ 1 r .

For t0,

I ρ ( w ρ ) = 1 2 B ρ ( | w ρ | 2 + V w ρ 2 ) B ρ F ( w ρ ) = ( 1 2 B ρ 1 ( | w ρ | 2 + V w ρ 2 ) B ρ 1 F ( w ρ ) ) + ( 1 2 B ρ B ρ 1 ( | w ρ | 2 + V w ρ 2 ) B ρ B ρ 1 F ( w ρ ) ) .
(3.8)

For the second part in (3.8), by I ρ ( w ρ )= max t > 0 I ρ (t w ρ ), for t>0,

1 2 B ρ B ρ 1 ( | w ρ | 2 + V w ρ 2 ) B ρ B ρ 1 F ( w ρ ) t 2 2 B ρ B ρ 1 ( | w ρ | 2 + V w ρ 2 ) B ρ B ρ 1 F ( t w ρ ) 1 2 B ρ B ρ 1 ( t 2 | w ρ | 2 + max ρ 1 r ρ { V F ( t w ρ ) ( t w ρ ) 2 } t 2 w ρ 2 ) .

So

I ρ ( t w ρ ) = t 2 2 B ρ ( | w ρ | 2 + V w ρ 2 ) B ρ F ( t w ρ ) t 2 2 B ρ 1 ( | w ρ | 2 + V w ρ 2 ) B ρ 1 F ( t w ρ ) + 1 2 B ρ B ρ 1 ( t 2 | w ρ | 2 + max ρ 1 r ρ { V F ( t w ρ ) ( t w ρ ) 2 } t 2 w ρ 2 ) = t 2 2 B ρ 1 ( | w ρ | 2 + V w ρ 2 ) B ρ 1 F ( t w ρ ) + t 2 2 R N B ρ 1 ( | v ¯ ρ | 2 + V v ¯ ρ 2 ) t 2 2 R N B ρ 1 ( | v ¯ ρ | 2 + V v ¯ ρ 2 ) + 1 2 B ρ B ρ 1 t 2 | w ρ | 2 + 1 2 B ρ B ρ 1 max ρ 1 r ρ { V F ( t w ρ ) ( t w ρ ) 2 } t 2 w ρ 2 t 2 2 B ρ 1 ( | w ρ | 2 + V w ρ 2 ) B ρ 1 F ( t w ρ ) + t 2 2 R N B ρ 1 ( | v ¯ ρ | 2 + V v ¯ ρ 2 ) R N B ρ 1 F ( t v ¯ ρ ) t 2 2 R N B ρ 1 ( | v ¯ ρ | 2 + V v ¯ ρ 2 ) + 1 2 B ρ B ρ 1 t 2 | w ρ | 2 + 1 2 B ρ B ρ 1 max ρ 1 r ρ { V F ( t w ρ ) ( t w ρ ) 2 } t 2 w ρ 2 = I ( t w ˆ ρ ) t 2 2 R N B ρ 1 ( | v ¯ ρ | 2 + V v ¯ ρ 2 ) + 1 2 B ρ B ρ 1 t 2 | w ρ | 2 + 1 2 B ρ B ρ 1 max ρ 1 r ρ { V F ( t w ρ ) ( t w ρ ) 2 } t 2 w ρ 2 ,

where I is defined in (1.3). Take (3.7) in the last equality of the above, then

I ρ ( t w ρ ) I ( t w ˆ ρ ) t 2 2 v ¯ ρ v ¯ ρ | B ρ 1 + 1 2 B ρ B ρ 1 t 2 | w ρ | 2 + 1 2 B ρ B ρ 1 max ρ 1 r ρ { V F ( t w ρ ) ( t w ρ ) 2 } t 2 w ρ 2 .
(3.9)

Choose t ρ such that I( t ρ w ˆ ρ ) c , and let w ρ w in H 1 , then as ρ, t ρ 1. Moreover, max ρ 1 r ρ {V F ( t ρ w ρ ) ( t ρ w ρ ) 2 } max ρ 1 r ρ V, as ρ.

Next we consider v ˆ ρ , which gives the solution of

{ Δ u + max { V F ( t ρ w ρ ) ( t ρ w ρ ) 2 } u = 0 , in  B ρ B ρ 1 , u ( ρ 1 ) = w ρ ( ρ 1 ) , u ( ρ ) = 0 .

Here v ˆ ρ works as the comparison function. Similarly to the computation of (3.1) and (3.7),

B ρ B ρ 1 max ρ 1 r ρ { V F ( t ρ w ρ ) ( t ρ w ρ ) 2 } v ˆ ρ 2 = B ρ B ρ 1 Δ v ˆ ρ v ˆ ρ = v ˆ ρ v ˆ ρ | B ρ 1 B ρ B ρ 1 | v ˆ ρ | 2 ,

that is,

B ρ B ρ 1 ( | v ˆ ρ | 2 + max ρ 1 r ρ { V F ( t ρ w ρ ) ( t ρ w ρ ) 2 } v ˆ ρ 2 ) = v ˆ ρ v ˆ ρ | B ρ 1 .
(3.10)

Take (3.10) in (3.9); by the definition of v ¯ ρ and v ˆ ρ , then

I ρ ( t w ρ ) I ( t ρ w ˆ ρ ) t ρ 2 2 v ¯ ρ v ¯ ρ | B ρ 1 + t ρ 2 2 B ρ B ρ 1 | v ˆ ρ | 2 + t ρ 2 2 B ρ B ρ 1 max ρ 1 r ρ { V F ( t ρ w ρ ) ( t ρ w ρ ) 2 } v ˆ ρ 2 = I ( t ρ w ˆ ρ ) t ρ 2 2 v ¯ ρ v ¯ ρ | B ρ 1 + t ρ 2 2 v ˆ ρ v ˆ ρ | B ρ 1 c + t ρ 2 2 w ρ ( ρ 1 ) ( v ˆ ρ ( ρ 1 ) v ¯ ρ ( ρ 1 ) ) .
(3.11)

By Lemma 2.2, v ˆ ρ (ρ1) v ¯ ρ (ρ1) w ρ (ρ1). And by Lemma 2.1 and (3.11),

c ρ = I ρ ( w ρ )= max t > 0 I ρ (t w ρ ) c + γ 2 exp ( 2 ρ ( 1 + o ( 1 ) ) ) .

So we conclude

c ρ = c + γ 2 exp ( 2 ρ ( 1 + o ( 1 ) ) ) .

 □

Author’s contributions

The work was carried out by the author.

References

  1. Berestycki H, Lions PL: Nonlinear scalar field equations. I. Existence of a ground state. Arch. Ration. Mech. Anal. 1983, 82: 313-345.

    MathSciNet  Google Scholar 

  2. Ding WY, Ni WM: On the existence of positive entire solutions of a semilinear elliptic equation. Arch. Ration. Mech. Anal. 1986, 91: 283-308. 10.1007/BF00282336

    Article  MathSciNet  Google Scholar 

  3. Rabinowitz PH: On a class of nonlinear Schrödinger equations. Z. Angew. Math. Phys. 1992, 43: 270-291. 10.1007/BF00946631

    Article  MathSciNet  Google Scholar 

  4. Kryszewski W, Szulkin A: Generalized linking theorem with an application to semilinear Schrödinger equation. Adv. Differ. Equ. 1998, 3: 441-472.

    MathSciNet  Google Scholar 

  5. Bartsch T, Wang ZQ: On the existence of sign changing solutions for semilinear Dirichlet problems. Topol. Methods Nonlinear Anal. 1999, 7: 115-131.

    MathSciNet  Google Scholar 

  6. Bartsch T, Liu ZL, Weth T: Sign changing solutions of superlinear Schrödinger equations. Commun. Partial Differ. Equ. 2004, 29: 25-42. 10.1081/PDE-120028842

    Article  MathSciNet  Google Scholar 

  7. Gidas B, Ni WM, Nirenberg L:Symmetry of positive solutions of nonlinear elliptic equations in R N . Adv. Math. Stud. A 1979, 7: 209-243.

    MathSciNet  Google Scholar 

  8. Kwong MK:Uniqueness of positive solutions of Δuu+ u p =0 in R n . Arch. Ration. Mech. Anal. 1989, 105: 243-266.

    Article  MathSciNet  Google Scholar 

  9. Ni WM, Takagi I: On the shape on least-energy solutions to a semilinear Neumann problem. Commun. Pure Appl. Math. 1991, XLIV: 819-851. 10.1002/cpa.3160440705

    Article  MathSciNet  Google Scholar 

  10. Floer A, Weinstein A: Nonspreading wave packets for the cubic Schrödinger equation with a bounded potential. J. Funct. Anal. 1986, 69: 397-408. 10.1016/0022-1236(86)90096-0

    Article  MathSciNet  Google Scholar 

  11. Oh YJ:Existence of semi-classical bound states of nonlinear Schrödinger equations with potential on the class ( V ) a . Commun. Partial Differ. Equ. 1988, 13: 1499-1519. 10.1080/03605308808820585

    Article  Google Scholar 

  12. Oh YJ: On positive multi-lump bound states nonlinear Schrödinger equations under multiple well potential. Commun. Math. Phys. 1990, 131: 223-253. 10.1007/BF02161413

    Article  Google Scholar 

  13. Ni WM, Wei J: On the location and profile of spike-layer solutions to singularly perturbed semilinear Dirichlet problems. Commun. Pure Appl. Math. 1995, 48: 731-768. 10.1002/cpa.3160480704

    Article  MathSciNet  Google Scholar 

  14. Liu Z, Wang ZQ: On the Ambrosetti-Rabinowitz superlinear condition. Adv. Nonlinear Stud. 2004, 4: 561-572.

    Google Scholar 

  15. Serrin J, Tang M: Uniqueness of ground states for quasilinear elliptic equations. Indiana Univ. Math. J. 2000, 49: 897-923. 10.1512/iumj.2000.49.1893

    Article  MathSciNet  Google Scholar 

  16. Li YQ, Wang ZQ, Zeng J: Ground states of nonlinear Schrödinger equations with potentials. Ann. Inst. Henri Poincaré, Anal. Non Linéaire 2006, 23: 829-837. 10.1016/j.anihpc.2006.01.003

    Article  MathSciNet  Google Scholar 

  17. Jeanjean L, Tanaka K:A positive solution for a nonlinear Schrödinger equation on R N . Indiana Univ. Math. J. 2005, 54: 443-464. 10.1512/iumj.2005.54.2502

    Article  MathSciNet  Google Scholar 

  18. Del Pino M, Felmer PL: Spike-layered solutions of singularly perturbed elliptic problems in a degenerate setting. Indiana Univ. Math. J. 1999, 48: 883-898.

    Article  MathSciNet  Google Scholar 

Download references

Acknowledgements

The author is supported by China Postdoctoral Science Foundation (2014M551830) and the Nonlinear Analysis Innovation Team (IRTL1206) funded by Fujian Normal University.

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Jing Zeng.

Additional information

Competing interests

The author declares to have no competing interests.

Rights and permissions

Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (https://creativecommons.org/licenses/by/4.0), which permits use, duplication, adaptation, distribution, and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Zeng, J. The estimation and profile of the critical value for a Schrödinger equation. Bound Value Probl 2014, 202 (2014). https://doi.org/10.1186/s13661-014-0202-7

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/s13661-014-0202-7

Keywords