Skip to main content

Infinitely many weak solutions for a fractional Schrödinger equation

Abstract

In this paper we are concerned with the fractional Schrödinger equation ( Δ ) α u+V(x)u=f(x,u), x R N , where 0<α<1, N>2α, ( Δ ) α stands for the fractional Laplacian of order α, V is a positive continuous potential, and f is a continuous subcritical nonlinearity. We obtain the existence of infinitely many weak solutions for the above problem by the fountain theorem in critical point theory.

Introduction

In this paper we consider the following fractional Schrödinger equation:

( Δ ) α u+V(x)u=f(x,u),x R N ,
(1.1)

where 0<α<1, N>2α, ( Δ ) α stands for the fractional Laplacian of order α, and the potential V: R N R is a continuous function satisfying

  1. (V)

    0< inf x R N V(x)= V 0 < lim inf | x | V(x)= V <.

The nonlinearity f: R N ×RR is a continuous function, satisfying the subcritical condition.

(H1) There exist d 1 >0, d 2 >0 and p(2, 2 α ) such that

| f ( x , s ) | d 1 |s|+ d 2 | s | p 1 ,(x,s) R N ×R,
(1.2)

where 2 α = 2 N N 2 α is the fractional critical exponent.

Recently, there have appeared plenty of works on the fractional Schrödinger equations; for example, see [[1]–[11]] and the references therein. In [[1]], Shang and Zhang considered the critical fractional Schrödinger equation

ε 2 α ( Δ ) α u+V(x)u= | u | 2 α 2 u+λf(u),x R N ,
(1.3)

where ε and λ are positive parameters, V and f satisfy (V) and (H1), respectively. They obtained the result that (1.2) has a nonnegative ground state solution and investigated the relation between the number of solutions and the topology of the set where V attains its minimum for all sufficiently large λ and small ε. In [[2]], Shang et al. considered the existence of nontrivial solutions for (1.2) with f(u)= | u | q 2 u, where 2<q< 2 α .

In [[8]], Hua and Yu studied the critical fractional Laplacian equation

{ ( Δ ) α 2 u = | u | 2 α 2 u + μ u in  Ω , u = 0 on  Ω ,
(1.3)

where 0<α<2, Ω R N , N>(1+ 2 )α is a bounded domain. They obtained the result that the problem (1.3) possesses a nontrivial ground state solution for any μ>0.

In [[7]], Secchi investigated the existence of radially symmetric solutions for (1.1) replacing f(t,u) by g(u), where g satisfies the following conditions.

(g1) g:RR is of class C 1 , γ for some γ>max{0,12α}, and odd,

(g2) < lim inf t 0 + g ( t ) t lim sup t 0 + g ( t ) t =m<0,

(g3) < lim sup t + g ( t ) t 2 α 1 0,

(g4) for some ξ>0 such that G(ξ)= 0 ξ g(t)dt>0.

Inspired by the mentioned papers, we first establish a compact embedding lemma via a fractional Gagliardo-Nirenberg inequality. Then by virtue of the fountain theorem in critical point theory, we get two existence results of infinitely many weak solutions for (1.1).

Preliminary results

In this section we offer some preliminary results which enable us to obtain the main existence theorems. First, we collect some useful facts of the fractional order Sobolev spaces.

For any 0<α<1, the fractional Sobolev space H α ( R N ) is defined by

H α ( R N ) = { u L 2 ( R N ) : | u ( x ) u ( y ) | | x y | N + 2 α 2 L 2 ( R N × R N ) } ,

endowed with the norm

u H α ( R N ) = ( R N | u | 2 d x + R 2 N | u ( x ) u ( y ) | 2 | x y | N + 2 α d x d y ) 1 2 ,

where [ u ] H α ( R N ) = ( R 2 N | u ( x ) u ( y ) | 2 | x y | N + 2 α d x d y ) 1 / 2 is the so-called Gagliardo semi-norm of u. Let S be the Schwartz space of rapidly decaying C functions in R N , for any uS and α(0,1), and let ( Δ ) α be defined as

( Δ ) α u(x)= k N , α P.V. R N u ( x ) u ( y ) | x y | N + 2 α dy= k N , α lim ε 0 C B ε ( x ) u ( x ) u ( y ) | x y | N + 2 α dy.
(2.1)

The symbol P.V. stands for the Cauchy principal value, and k N , α is a dimensional constant that depends on N and α, precisely given by k N , α = ( R N 1 cos ζ 1 | ζ | N + 2 α d ζ ) 1 .

Indeed, the fractional Laplacian ( Δ ) α can be viewed as a pseudo-differential operator of symbol | ξ | 2 α , as stated in the following.

Lemma 2.1

(see [[12]])

Letα(0,1)and ( Δ ) α :S L 2 ( R N )be the fractional Laplacian operator defined by (2.1). Then for anyuS,

( Δ ) α u(x)= F 1 ( | ξ | 2 α ( F u ) ) (x),ξ R N ,

where is the Fourier transform, i.e.,

F(ϕ)(ξ)= 1 ( 2 π ) N / 2 R N exp{2πiξx}ϕ(x)dx.

Now we can see that an alternative definition of the fractional Sobolev space H α ( R N ) via the Fourier transform is as follows:

H α ( R N ) = { u L 2 ( R N ) : R N ( 1 + | ξ | 2 α ) | F u | 2 d ξ < + } .

It can be proved that

2 k N , α 1 R N | ξ | 2 α | F u | 2 dξ=2 k N , α 1 ( Δ ) α 2 u L 2 ( R N ) 2 = [ u ] H α ( R N ) 2 .
(2.2)

As a result, the norms on H α ( R N ),

u u H α ( R N ) , u ( u L 2 ( R N ) 2 + ( Δ ) α 2 u L 2 ( R N ) 2 ) 1 2 , u ( u L 2 ( R N ) 2 + R N | ξ | 2 α | F u | 2 d ξ ) 1 2 ,
(2.3)

are all equivalent.

In this paper, in view of the presence of potential V(x), we consider its subspace

E= { u H α ( R N ) : R N V ( x ) u 2 d x < } .

We define the norm in E by

u E = ( R N ( | ξ | 2 α u ˆ 2 + u ˆ 2 ) d ξ + R N V ( x ) u 2 d x ) 1 2 ,

where u ˆ =F(u). Moreover, by [[6]], E is a Hilbert space with the inner product

u , v E = R N ( | ξ | 2 α u ˆ ( ξ ) v ˆ ( ξ ) + u ˆ ( ξ ) v ˆ ( ξ ) ) dξ+ R N V(x)u(x)v(x)dx,u,vE.

Note that by (2.2) and (2.3), together with the condition (V), we know that E is equivalent to the norm

u= ( R N ( | ( Δ ) α 2 u | 2 + V ( x ) u 2 ) d x ) 1 2 .
(2.4)

The corresponding inner product is

u,v= R N ( ( Δ ) α 2 u ( x ) ( Δ ) α 2 v ( x ) + V ( x ) u ( x ) v ( x ) ) dx.

Throughout out this paper, we will use the norm in E.

Definition 2.2

We say that uE is a weak solution of (1.1), if

R N ( ( Δ ) α 2 u ( Δ ) α 2 ϕ + V ( x ) u ϕ ) dx= R N f(x,u)ϕdx,ϕE.

Lemma 2.3

(see [[7]] and [[12]])

E is continuously embedded into L p ( R N )forp[2, 2 α ]and compactly embedded into L loc p ( R N )forp[2, 2 α ).

Lemma 2.4

E is compactly embedded into L p ( R N )forp[2, 2 α )with 2 α = 2 N N 2 α .

Proof

By [[4]], we know E is compactly embedded into L 2 ( R N ), i.e., if there exists a sequence { u n }E and u 0 E such that u n u 0 weakly in E, passing to a subsequence if necessary, we have u n u 0 strongly in L 2 ( R N ). Therefore, we only consider p(2, 2 α ). In order to do this, we need the following fractional Gagliardo-Nirenberg inequality, see [[13], Corollary 2.3]. Let 1p, p 2 <, 0<s<p<, 0<α<N and 1< p 1 <N/α. Then

u ( x ) L p ( R N ) η s p ( Δ ) α / 2 u ( x ) L p 1 ( R N ) s p u ( x ) L p 2 ( R N ) 1 s p
(2.5)

with

s ( 1 p 1 α N ) + p s p 2 =1andη= 2 α π α / 2 Γ ( ( N α ) / 2 ) Γ ( ( N + α ) / 2 ) ( Γ ( N ) Γ ( N / 2 ) ) α / N .

Note that the dimension N>2α, we can take p 1 = p 2 =2, and then s( 1 2 α N )+ p s 2 =1 whence s= ( p 2 ) N 2 α (0,p) as p(2, 2 s ). Consequently, from (2.5) we have

u L p ( R N ) η s p ( Δ ) α / 2 u L 2 ( R N ) s p u L 2 ( R N ) 1 s p .

Furthermore, note that (2.4); we see that

u L p ( R N ) η s p u s p u L 2 ( R N ) 1 s p .
(2.6)

Then by (2.6) and E L 2 ( R N ), we find

u n u 0 L p ( R N ) η s p u n u 0 s p u n u 0 L 2 ( R N ) 1 s p η s p ( u n s p + u 0 s p ) u n u 0 L 2 ( R N ) 1 s p 0 .

Therefore, E is compactly embedded into L p ( R N ) for p[2, 2 α ) with 2 α = 2 N N 2 α , as required. This completes the proof. □

The functional associated with (1.1) is defined by

J(u)= 1 2 R N ( | ( Δ ) α 2 u | 2 + V ( x ) u 2 ) dx R N F(x,u)dx,uE,
(2.7)

where F(x,u)= 0 u f(x,s)ds.

Now, we list our assumptions on f and F.

(H2) lim | s | F ( x , s ) | s | 2 =+ uniformly for x R N .

(H3) There exist d 3 >0 and φ 1 >0 such that R N φ 1 (x)dx<+ and

tf(x,t)2F(x,t)sf(x,s)2F(x,s)+ d 3 φ 1 ,0<t<s or s<t<0,x R N .

(H4) F(x,s)0 for (x,s) R N ×R.

(H5) There exist β>2, r 0 >0 such that

βF(x,s)+sf(x,s)0,|s| r 0 ,uniformly for x R N .

(H6) F(x,s)=F(x,s) for all (x,s) R N ×R.

Remark 2.5

  1. (1)

    Let F(x,s)= s 2 ln(|s|+1), for all x R N and sR. Then (H1), (H2), (H4), and (H6) hold. Moreover, we easily have f(x,s)=2sln(|s|+1)+ s 3 [ | s | ( | s | + 1 ) ] 1 and sf(x,s)2F(x,s)= | s | 3 ( | s | + 1 ) 1 , so (H3) is satisfied.

However, we can see that F(x,s) does not satisfy the Ambrosetti-Rabinowitz condition (see [[6], ( f 4 )]):

(AR) there is a constant μ>2 such that

0<μF(x,s)sf(x,s)for all x R N  and sR{0}.

Indeed, sf(x,s)μF(x,s)= s 2 [(2μ)ln(|s|+1)+|s| ( | s | + 1 ) 1 ]0 is impossible for all x R N and sR{0}.

  1. (2)

    Let β>2 and F(x,s)= | s | β ln(|s|+1), for all x R N and sR. Then (H1), (H2), (H4), and (H6) hold. Moreover, from βF(x,s)+sf(x,s)= | s | β + 1 ( | s | + 1 ) 1 , and (H5) holds.

Note that from Theorem 4 in [[14]] we have (H4) and (H5) imply (H2).

Lemma 2.6

(see [[15], Lemma 1])

Let (V) and (H1) hold. ThenJ C 1 (E,R)and its derivative

( J ( u ) , ϕ ) = R N ( ( Δ ) α 2 u ( Δ ) α 2 ϕ + V ( x ) u ϕ ) dx R N f(x,u)ϕdx,u,ϕE.

Moreover, the critical points of J are weak solutions of (1.1).

To complete the proofs of our theorems, we need the following critical point theorems in [[16]–[19]].

Definition 2.7

Let (X,) be a real Banach space, J C 1 (X,R). We say that J satisfies the ( C c ) condition if any sequence { u n }X such that J( u n )c and J ( u n )(1+ u n )0 as n has a convergent subsequence.

Let X be a Banach space equipped with the norm and X= j N X j ¯ , where dim X j < for any jN. Set Y k = j = 1 k X j and Z k = j = k X j ¯ .

Lemma 2.8

Let(X,)be a real reflexive Banach space, J C 1 (X,R)satisfies the ( C c ) condition for anyc>0and J is even. If for each sufficiently largekN, there exist ρ k > r k >0such that the following conditions hold:

  1. (i)

    a k := inf { u Z k , u = r k } J(u)+ as k,

  2. (ii)

    b k := max { u Y k , u = ρ k } J(u)0,

then the functional J has an unbounded sequence of critical values, i.e., there exists a sequence{ u k }Xsuch that J ( u k )=0andJ( u k )+ask.

In the following, we will introduce a variant fountain theorem by Zou [[16]]. Let X and the subspace Y k and Z k be defined above. Consider the following C 1 -functional J λ :XR defined by

J λ (u):=A(u)λB(u),λ[1,2].
(2.8)

Lemma 2.9

If the functional J λ satisfies

(T1) J λ maps bounded sets to bounded sets uniformly forλ[1,2], and, moreover, J λ (u)= J λ (u)for all(λ,u)[1,2]×X,

(T2) B(u)0for alluX; moreover, A(u)orB(u)asu,

(T3) there exist r k > ρ k >0such that

a k (λ):= inf u Z k , u = ρ k J λ (u)> b k (λ):= max u Y k , u = r k J λ (u),λ[1,2],

then

a k (λ) ζ k (λ)= inf γ Γ k max u B k J λ ( γ ( u ) ) ,λ[1,2],

where B k ={u Y k :u r k }and Γ k ={γC( B k ,X):γ is odd,γ | B k =id}. Moreover, for a.e. λ[1,2], there exists a sequence { u n k ( λ ) } n = 1 such that

sup n u n k ( λ ) <, J λ ( u n k ( λ ) ) 0and J λ ( u n k ( λ ) ) ζ k (λ)as n.

Remark 2.10

As mentioned in [[6]], E is a Hilbert space. Let { e j } be an orthonormal basis of E and define X j :=span{ e j }, Y k := j = 1 k X j , and Z k := j = k + 1 X j ¯ , kN. Clearly, E= j N X j ¯ with dim X j < for all jN.

Existence of weak solutions for (1.1)

Theorem 3.1

Assume that (V), (H1)-(H4), and (H6) hold. Then (1.1) has infinitely many weak solutions{ u k }satisfying

1 2 R N ( | ( Δ ) α 2 u k | 2 + V ( x ) u k 2 ) dx R N F(x, u k )dx+as k.

Proof

We first prove that J satisfies the ( C c ) condition for any c>0. Let { u n }E be a ( C c ) sequence, i.e.,

J( u n )c>0, J ( u n ) ( 1 + u n ) 0when n,
(3.1)

which implies that

c=J( u n )+o(1), J ( u n ) u n =o(1)as n.
(3.2)

In what follows, we shall show that { u n } is bounded. Otherwise, up to a subsequence, { u n } is unbounded in E, and we may assume that u n as n. We define the sequence { w n } by w n = u n u n , n=1,2, . Clearly, { w n }E and w n =1 for any n. Going over, if necessary, to a subsequence, we may assume that

w n w  weakly in  E , w n w  strongly in  L p ( R N )  for  p [ 2 , 2 s ) , w n ( x ) w ( x )  a.e.  x R N .
(3.3)

Suppose that w0 in E. Dividing by u n 2 in both sides of (2.7), noting that J( u n )c, we obtain

R N F ( x , u n ) u n 2 dx= 1 2 +o ( u n 2 ) <+.
(3.4)

On the other hand, denote Ω :={x R N :w(x)0}, by (H2), for all x Ω , and we find

F ( x , u n ) u n 2 = F ( x , u n ) | u n | 2 | u n | 2 u n 2 = F ( x , u n ) | u n | 2 | w n | 2 +when n.

If | Ω |>0, using Fatou’s lemma, we obtain

R N F ( x , u n ) u n 2 dx+as n.

This contradicts (3.4). Hence, Ω has zero measure, i.e., w=0 a.e. in R N . Let t n [0,1] such that

J( t n u n )= max t [ 0 , 1 ] J(t u n ).

Then we claim J( t n u n ) is bounded. If t n =0, J(0)=0; if t n =1, J( t n u n )=J( u n )c. Therefore, J( t n u n ) is bounded when t n =0,1. If 0< t n <1 for n large enough

R N ( ( Δ ) α 2 t n u n ( Δ ) α 2 t n u n + V ( x ) t n u n t n u n ) d x R N f ( x , t n u n ) t n u n d x = ( J ( t n u n ) , t n u n ) = t n d d t | t = t n J ( t u n ) = 0 .

Consequently, by (H3), noting that (3.1) and (3.2) hold, we have

J ( t u n ) J ( t n u n ) 1 2 ( J ( t n u n ) , t n u n ) = 1 2 R N f ( x , t n u n ) t n u n d x R N F ( x , t n u n ) d x R N [ ( 1 2 u n f ( x , u n ) F ( x , u n ) ) + d 3 2 φ 1 ( x ) ] d x = J ( u n ) 1 2 ( J ( u n ) , u n ) + R N d 3 2 φ 1 ( x ) d x d 4 , t [ 0 , 1 ] ,
(3.5)

where d 4 is a positive constant. But fixing any m> d 4 , we let w ¯ n = 2 m u n u n = 2 m w n . Note that from (H1) we see that there exist d 5 >0, d 6 >0 such that

F(x,u) d 5 | u | 2 + d 6 | u | p ,(x,u) R N ×R.
(3.6)

Then by (3.3) we have

lim n R N F(x, w ¯ n )dx lim n R N ( d 5 | w ¯ n | 2 + d 6 | w ¯ n | p ) dx=0.

Then for n large enough,

J( t n u n )J ( 2 m u n u n ) =J( w ¯ n )=m R N F(x, w ¯ n )dxm.

This also contradicts (3.5).

Now the sequence { u n } is bounded, as required. Next, we verify that { u n } has a convergent subsequence. Without loss of generality, we assume that

u n u  weakly in  E , u n u  strongly in  L p ( R N )  for  p [ 2 , 2 s ) .
(3.7)

Combining this with (H1) and the Hölder inequality, we see

| R N [ f ( x , u n ) f ( x , u ) ] ( u n u ) d x | R N [ d 1 ( | u n | + | u | ) + d 2 ( | u n | p 1 + | u | p 1 ) ] | u n u | d x d 1 ( u n 2 + u 2 ) u n u 2 + d 2 ( u n p p 1 + u p p 1 ) u n u p 0 as  n .

Consequently,

u n u 2 = ( J ( u n ) J ( u ) , u n u ) + R N [ f ( x , u n ) f ( x , u ) ] ( u n u)dx0

with the fact that ( J ( u n ) J (u), u n u)0 when n. Therefore, we prove that J satisfies the ( C c ) condition for any c>0.

Clearly, J(u)=J(u) by (H6). It remains to prove that the conditions (i) and (ii) of Lemma 2.8 hold. Let β r (k):= sup { u Z k , u = 1 } u r with r[2, 2 α ), where Z k is defined in Remark 2.10. Then by Lemma 3.8 of [[19]], β r (k)0 as k for the fact that E L r ( R N ).

Now for u Z k with u= r k = ( β 2 ( k ) + β p ( k ) ) 1 , we obtain

J ( u ) 1 2 u 2 R N ( d 5 | u | 2 + d 6 | u | p ) d x = 1 2 u 2 d 5 u 2 2 d 6 u p p 1 2 u 2 d 5 β 2 2 ( k ) u 2 d 6 β p p ( k ) u p 1 2 r k 2 d 5 d 6 as  k .

Hence,

a k := inf u Z k , u = r k J(u)+as k.
(3.8)

Next we shall prove that, for any finite dimensional subspace XE, we have

J(u)as u,uX.
(3.9)

Suppose the contrary. For some sequence { u n }X with u n , there is a M>0 such that J( u n )M for all nN. Put v n = u n u n and then v n =1. Up to a subsequence, assume that v n v weakly in E. Since dimX<, v n vX in E, v n v a.e. on R N , and v=1. Denote Ω:={x R N :v(x)0}, then meas(Ω)>0 and for a.e. xΩ, lim n | u n (x)|. It follows from (2.7) that

lim n R N F ( x , u n ) d x u n 2 = lim n 1 2 u n 2 J ( u n ) u n 2 d 7 with a constant  d 7 >0.
(3.10)

But, for large n, on account of F being nonnegative, (H2) and Fatou’s Lemma enable us to obtain

lim n R N F ( x , u n ) d x u n 2 lim n Ω F ( x , u n ) v n 2 u n 2 d x lim inf n Ω F ( x , u n ) v n 2 u n 2 d x Ω lim inf n F ( x , u n ) v n 2 u n 2 d x = Ω lim inf n F ( x , u n ) u n 2 [ χ Ω ( x ) ] v n 2 d x

as n. This contradicts (3.10). Consequently, (3.9) holds, as required. Note that dim Y k < in Remark 2.10, and there exist positive constants d k such that

J(u)0,for each u Y k  and u d k .
(3.11)

Combining this and (3.8), we can take ρ k :=max{ d k , r k +1}, and thus b k := max { u Y k , u = ρ k } J(u)0. Until now, we have proved the functional J satisfies all the conditions of Lemma 2.8. Hence, J has an unbounded sequence of critical values, i.e., there exists a sequence { u k }E such that J ( u k )=0 and J( u k )+ as k. This completes the proof. □

We prove that there exists a 0 = a 0 ( r 0 )>0 ( r 0 is determined in (H5)) such that

| β F ( x , s ) + s f ( x , s ) | a 0 | s | 2 ,|s| r 0 , for all x R N .
(3.12)

Indeed, by (3.6) we see

| F ( x , s ) | d 5 | s | 2 + d 6 | s | p ( d 5 + d 6 r 0 p 2 ) | s | 2 ,|s| r 0 ,x R N .

This, together with (H1), implies that

| β F ( x , s ) + s f ( x , s ) | β | F ( x , s ) | + | s f ( x , s ) | β ( d 5 + d 6 r 0 p 2 ) | s | 2 + | s ( d 1 | s | + d 2 | s | p 1 ) | ( d 1 + β d 5 + ( d 2 + β d 6 ) r 0 p 2 ) | s | 2 , if  | s | r 0 .

Clearly, (3.12) holds true with a 0 = d 1 +β d 5 +( d 2 +β d 6 ) r 0 p 2 . In the following theorem, we make the following assumption instead of (V):

(V′): VC( R N ,R), inf x R N V(x)= V 0 [ 1 2 a 0 + 1 ( β 2 1 ) ] 1 >0, where a 0 in (3.12), β in (H5).Especially, by (V′), we obtain

u 2 2 1 2 a 0 + 1 ( β 2 1 ) u 2 ,uE.
(3.13)

Now, we define a class of functionals on E by

J λ (u)= 1 2 u 2 λ R N F(x,u)dx=A(u)λB(u),λ[1,2].

It is easy to know that J λ C 1 (E,R) for all λ[1,2] and the critical points of J 1 correspond to the weak solutions of problem (1.1). Note that J 1 =J, where J is the functional defined in (2.7).

Theorem 3.2

Assume that (V′), (H1), and (H4)-(H6) hold. Then (1.1) possesses infinitely many weak solutions.

Proof

We first prove that there exist a positive integer k 1 and two sequences r k > ρ k as k such that

a k (λ)= inf u Z k , u = ρ k J λ (u)>0,k k 1 ,
(3.14)
b k (λ)= max u Y k , u = r k J λ (u)<0,kN,
(3.15)

where Y k and Z k are defined in Remark 2.10.

Step 1. We claim that (3.14) is true.

By (3.6) and (H4) we have

J λ ( u ) = 1 2 u 2 λ R N F ( x , u ) d x 1 2 u 2 2 R N F ( x , u ) d x 1 2 u 2 2 R N ( d 5 | u | 2 + d 6 | u | p ) d x 1 2 u 2 2 d 5 u 2 2 2 d 6 u p p .
(3.16)

Since E L r ( R N ) with r[2, 2 α ), and from Theorem 3.1 we have

J λ (u) 1 2 u 2 2 d 5 β 2 2 (k) u 2 2 d 6 β p p (k) u p .

Let ρ k = 1 β 2 ( k ) + β p ( k ) as k. Then there exists k 1 such that 1 2 ρ k 2 2 d 5 2 d 6 >0, k k 1 . Therefore,

a k (λ)= inf u Z k , u = ρ k J λ (u) 1 2 ρ k 2 2 d 5 2 d 6 >0,k k 1 .

Step 2. We show that (3.15) is true.

We apply the method in Lemma 2.6 of [[20]] to verify the claim. First, we prove that there exists ε>0 such that

meas ( x R N : | u ( x ) | ε u ) ε,uX{0},XE and dimX<.
(3.17)

There would otherwise exist a sequence { u n } n N X{0} such that

meas ( x R N : | u n ( x ) | u n n ) < 1 n ,nN.
(3.18)

For each nN, let v n := u n u n X. Then v n =1, nN and

meas ( x R N : | v n ( x ) | 1 n ) < 1 n ,nN.
(3.19)

Passing to a subsequence if necessary, we may assume v n v 0 in E for some v 0 X since X is of finite dimension. We easily find v 0 =1. Consequently, there exists a constant σ 0 >0 such that

meas ( x R N : | v 0 ( x ) | σ 0 ) σ 0 .
(3.20)

Indeed, if not, then we have

meas ( x R N : | v 0 ( x ) | 1 n ) =0,nN,
(3.21)

which implies

0 R N | v 0 ( x ) | 3 dx v 0 2 2 n 0as n.

This leads to v 0 =0, contradicting v 0 =1. In view of E L p ( R N ) and the equivalence of any two norms on X, we have

R N | v n v 0 | 2 dx0as n.
(3.22)

For every nN, denote

N:= { x R N : | v n ( x ) | < 1 n } and N c := { x R N : | v n ( x ) | 1 n } ,

and N 0 :={x R N :| v 0 (x)| σ 0 }, where σ 0 is defined by (3.20). Then for n large enough, by (3.20), we see

meas(N N 0 )meas( N 0 )meas ( N c ) σ 0 1 n 3 σ 0 4 .

Consequently, for n large enough, we find

R N | v n v 0 | 2 d x N N 0 | v n v 0 | 2 d x 1 4 N N 0 | v 0 | 2 d x N N 0 | v n | 2 d x ( σ 0 2 4 1 n 2 ) meas ( N N 0 ) 9 σ 0 3 64 > 0 .

This contradicts (3.22). Therefore, (3.17) holds. For the ε given in (3.17), we let

N u := { x R N : | u ( x ) | ε u } ,uX{0}.

Then by (3.17), we find

meas( N u )ε,uX{0}.
(3.23)

As is well known, (H5) implies (H2), and hence for any kN, there is a constant S k >0 such that

F(x,u) | u | 2 ε 3 ,|u| S k ,

where ε is determined in (3.17). Therefore,

J λ ( u ) = 1 2 u 2 λ R N F ( x , u ) d x 1 2 u 2 N u | u | 2 ε 3 d x ( 1 2 1 ) u 2 .

Now for any kN, if we take r k >max{ ρ k , S k ε }, so u= r k is large enough, we have

b k (λ)= max u Y k , u = r k J λ (u)<0,kN.

Step 3. Clearly, J λ C 1 (E,R) implies that J λ maps bounded sets to bounded sets uniformly for λ[1,2]. In view of (H6), J λ (u)= J λ (u) for all (λ,u)[1,2]×E. Thus the condition (T1) of Lemma 2.9 holds. Besides, the condition (T2) of Lemma 2.9 holds for the fact that A(u)= 1 2 u 2 as u and B(u)0 since F(x,u)0. Evidently, Step 1 and Step 2 imply that the condition (T3) of Lemma 2.9 also holds for all k k 1 . Consequently, Lemma 2.9 implies that for any k k 1 and a.e. λ[1,2], there exists a sequence { u n k ( λ ) } n = 1 such that

sup n u n k ( λ ) <, J λ ( u n k ( λ ) ) 0and J λ ( u n k ( λ ) ) ζ k (λ)as n,

where

B k = { u Y k : u r k } , Γ k = { γ C ( B k , W ) : γ  is odd , γ | B k = id } , ζ k ( λ ) = inf γ Γ k max u B k J λ ( γ ( u ) ) , λ [ 1 , 2 ] .

Furthermore, we easily have ζ k (λ)[ a ¯ k , ζ ¯ k ], k k 1 , where ζ ¯ k := max u B k J λ (γ(u)) and a ¯ k := 1 2 ρ k 2 2 d 5 2 d 6 as k.

Claim 1. { u n k ( λ ) } n = 1 E possesses a strong convergent subsequence in E, a.e. λ[1,2] and k k 1 . In fact, by the boundedness of { u n k ( λ ) } n = 1 , passing to a subsequence, as n, we may assume u n k (λ) u k (λ) in E. By the method of Theorem 3.1, we easily prove that u n k (λ) u k (λ) strongly in E.

Thus, for each k k 1 , we can choose λ l 1 such that for the sequence { u n k ( λ l ) } n = 1 we have obtained a convergent subsequence, and passing again to a subsequence, we may assume

lim n u n k ( λ l )= u l k  in E,lN and k k 1 .

Thus we obtain

J λ l ( u l k ) =0and J λ l ( u l k ) [ a ¯ k , ζ ¯ k ],lN and k k 1 .
(3.24)

Claim 2. { u l k } is bounded in E and has a convergent subsequence with the limit u k E for all k k 1 . For convenience, we set u l k = u l for all lN. Consequently, (3.12) and (H5) imply that

β J λ l ( u l ) ( J λ l ( u l ) , u l ) = ( β 2 1 ) u l 2 + λ l R N [ β F ( x , u l ( x ) ) + f ( x , u l ( x ) ) u l ( x ) ] d x = ( β 2 1 ) u l 2 + λ l | u l | r 0 [ β F ( x , u l ( x ) ) + f ( x , u l ( x ) ) u l ( x ) ] d x + λ l | u l | > r 0 [ β F ( x , u l ( x ) ) + f ( x , u l ( x ) ) u l ( x ) ] d x ( β 2 1 ) u l 2 λ l | u l | r 0 | β F ( x , u l ( x ) ) + f ( x , u l ( x ) ) u l ( x ) | d x ( β 2 1 ) u l 2 λ l a 0 u l 2 2 ( β 2 1 ) u l 2 2 a 0 u l 2 2 ( β 2 1 ) u l 2 2 a 0 2 a 0 + 1 ( β 2 1 ) u l 2 = 1 2 a 0 + 1 ( β 2 1 ) u l 2 .

Therefore, { u l } l = 1 is bounded in E. By Claim 1, we see that { u l } l = 1 has a convergent subsequence, which converges to an element u k W for all k k 1 .

Hence, passing to the limit in (3.24), we see J 1 ( u k )=0 and J 1 ( u k )[ a ¯ k , ζ ¯ k ], lN and k k 1 . Since a ¯ k as k, we get infinitely many nontrivial critical points of J 1 =J. Therefore (1.1) possesses infinitely many nontrivial solutions by Lemma 2.9. This completes the proof. □

References

  1. Shang X, Zhang J: Ground states for fractional Schrödinger equations with critical growth. Nonlinearity 2014, 27: 187-207. 10.1088/0951-7715/27/2/187

    Article  MathSciNet  Google Scholar 

  2. Shang X, Zhang J, Yang Y:On fractional Schrödinger equation in R N with critical growth. J. Math. Phys. 2013., 54: 10.1063/1.4835355

    Google Scholar 

  3. Cabré X, Sire Y: Nonlinear equations for fractional Laplacians I: regularity, maximum principles, and Hamiltonian estimates. Ann. Inst. Henri Poincaré, Anal. Non Linéaire 2014, 31: 23-53. 10.1016/j.anihpc.2013.02.001

    Article  Google Scholar 

  4. Chang X: Ground state solutions of asymptotically linear fractional Schrödinger equations. J. Math. Phys. 2013., 54: 10.1063/1.4809933

    Google Scholar 

  5. Chang X, Wang Z: Ground state of scalar field equations involving a fractional Laplacian with general nonlinearity. Nonlinearity 2013, 26: 479-494. 10.1088/0951-7715/26/2/479

    Article  MathSciNet  Google Scholar 

  6. Secchi S:Ground state solutions for nonlinear fractional Schrödinger equations in R N . J. Math. Phys. 2013., 54: 10.1063/1.4793990

    Google Scholar 

  7. Secchi, S: On fractional Schrödinger equations in N without the Ambrosetti-Rabinowitz condition (2012) , [http://arxiv.org/abs/arXiv:1210.0755]

    Google Scholar 

  8. Hua Y, Yu X: On the ground state solution for a critical fractional Laplacian equation. Nonlinear Anal. 2013, 87: 116-125. 10.1016/j.na.2013.04.005

    Article  MathSciNet  Google Scholar 

  9. Barrios B, Colorado E, de Pablo A, Sánchez U: On some critical problems for the fractional Laplacian operator. J. Differ. Equ. 2012, 252: 6133-6162. 10.1016/j.jde.2012.02.023

    Article  Google Scholar 

  10. Autuori G, Pucci P:Elliptic problems involving the fractional Laplacian in R N . J. Differ. Equ. 2013, 255: 2340-2362. 10.1016/j.jde.2013.06.016

    Article  MathSciNet  Google Scholar 

  11. Dipierro S, Pinamonti A: A geometric inequality and a symmetry result for elliptic systems involving the fractional Laplacian. J. Differ. Equ. 2013, 255: 85-119. 10.1016/j.jde.2013.04.001

    Article  MathSciNet  Google Scholar 

  12. Di Nezza E, Palatucci G, Valdinoci E: Hitchhiker’s guide to the fractional Sobolev spaces. Bull. Sci. Math. 2012, 136: 521-573. 10.1016/j.bulsci.2011.12.004

    Article  MathSciNet  Google Scholar 

  13. Hajaiej H, Yu X, Zhai Z: Fractional Gagliardo-Nirenberg and Hardy inequalities under Lorentz norms. J. Math. Anal. Appl. 2012, 396: 569-577. 10.1016/j.jmaa.2012.06.054

    Article  MathSciNet  Google Scholar 

  14. Wu X:Existence of nontrivial solutions and high energy solutions for Schrödinger-Kirchhoff-type equations in R N . Nonlinear Anal., Real World Appl. 2011, 12: 1278-1287. 10.1016/j.nonrwa.2010.09.023

    Article  MathSciNet  Google Scholar 

  15. Ye Y, Tang C:Multiple solutions for Kirchhoff-type equations in R N . J. Math. Phys. 2013., 54: 10.1063/1.4819249

    Google Scholar 

  16. Zou W: Variant fountain theorems and their applications. Manuscr. Math. 2001, 104: 343-358. 10.1007/s002290170032

    Article  Google Scholar 

  17. Struwe M: Variational Methods, Applications to Nonlinear Partial Differential Equations and Hamiltonian Systems. Springer, New York; 2000.

    Google Scholar 

  18. Ambrosetti A, Rabinowitz P: Dual variational methods in critical point theory and applications. J. Funct. Anal. 1973, 14: 349-381. 10.1016/0022-1236(73)90051-7

    Article  MathSciNet  Google Scholar 

  19. Willem M: Minimax Theorems. Birkhäuser, Boston; 1996.

    Book  Google Scholar 

  20. Zhang Q, Xu B: Multiplicity of solutions for a class of semilinear Schrödinger equations with sign-changing potential. J. Math. Anal. Appl. 2011, 377: 834-840. 10.1016/j.jmaa.2010.11.059

    Article  MathSciNet  Google Scholar 

Download references

Acknowledgements

The authors are highly grateful for the referees’ careful reading of this paper and their comments. Supported by the NNSF-China (11371117), Shandong Provincial Natural Science Foundation (ZR2013AM009) and Hebei Provincial Natural Science Foundation (A2012402036).

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Wei Dong.

Additional information

Competing interests

The authors declare that they have no competing interests.

Authors’ contributions

WD and JX obtained the results in a joint research. All the authors read and approved the final manuscript.

Authors’ original submitted files for images

Below are the links to the authors’ original submitted files for images.

Authors’ original file for figure 1

Authors’ original file for figure 2

Authors’ original file for figure 3

Authors’ original file for figure 4

Authors’ original file for figure 5

Authors’ original file for figure 6

Authors’ original file for figure 7

Authors’ original file for figure 8

Authors’ original file for figure 9

Authors’ original file for figure 10

Authors’ original file for figure 11

Authors’ original file for figure 12

Authors’ original file for figure 13

Authors’ original file for figure 14

Authors’ original file for figure 15

Authors’ original file for figure 16

Authors’ original file for figure 17

Authors’ original file for figure 18

Authors’ original file for figure 19

Authors’ original file for figure 20

Authors’ original file for figure 21

Authors’ original file for figure 22

Authors’ original file for figure 23

Authors’ original file for figure 24

Authors’ original file for figure 25

Authors’ original file for figure 26

Authors’ original file for figure 27

Authors’ original file for figure 28

Authors’ original file for figure 29

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

Dong, W., Xu, J. & Wei, Z. Infinitely many weak solutions for a fractional Schrödinger equation. Bound Value Probl 2014, 159 (2014). https://doi.org/10.1186/s13661-014-0159-6

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/s13661-014-0159-6

Keywords