INFINITELY MANY SOLUTIONS FOR A CLASS OF SUBLINEAR SCHRÖDINGER EQUATIONS. Jing Chen and X. H. Tang 1. INTRODUCTION

Transcription

1 TAIWANESE JOURNAL OF MATHEMATICS Vol. 19, No. 2, pp , April 2015 DOI: /tjm This paper is available online at INFINITELY MANY SOLUTIONS FOR A CLASS OF SUBLINEAR SCHRÖDINGER EQUATIONS Jing Chen and X. H. Tang Abstract. In this paper, we deal with the existence of infinitely many solutions for a class of sublinear Schrödinger equation { u + V x)u = fx, u), x, u H 1 ). Under the assumptions that inf V x) > 0 and fx, t) is indefinite sign and sublinear as t +, we establish some existence criteria to guarantee that the above problem has at least one or infinitely many nontrival solutions by using the genus properties in critical point theory. 1. INTRODUCTION Consider the following semilinear Schrödinger equation 1.1) { u + V x)u = fx, u), x, u H 1 ), where V : R and f : R R. In the past several decades, the existence and multiplicity of nontrivial solutions for problem 1.1) have been extensively investigated in the literature with the aid of critical point theory and variational methods. Many papers deal with the autonomous case where the potential V and the nonlinearity f are independent of x, or with the radially symmetric case where V and f depend on x, see for instance [1, 2, 5, 12, 18, 19] and the references therein. If the radial symmetry is lost, the problem becomes very different because of the lack of compactness. Ever since the work of Ding and Ni [7], Received November 17, 2013, accepted June 20, Communicated by Eiji Yanagida Mathematics Subject Classification: 35J10, 35J20. Key words and phrases: Schrödinger equation, Sublinear, Genus, Indefinite sign. This work is partially supported by the NNSF No and No ) of China and supported by Scientific Research Fund of Hunan Provincial Education Department 14C0465). 381

2 382 Jing Chen and X. H. Tang Li [11] and Rabinowitz [16], this situation has been treated in a great number of papers under various growth conditions on V and f. When the nonlinearity f is superlinear as t, the so-called Ambrosetti-Rabinowitz superquadratic condition is usually assumed, see [1, 2, 5, 7, 11, 12, 16, 18, 19]. Weaker superlinear conditions on f were obtained in [3, 21, 22, 23, 24, 27] and the references therein. Compared to the superlinear case, as far as the authors are aware, there are few papers [6, 8-10, 14, 26] concerning the case where f is sublinear as t. In 1992, Brezis and Kamin [6] gave a sufficient and necessary condition for the existence of bounded positive solutions of problem 1.1) with V x) =0. With a strong coercive condition on the potential V x), Ding and Li [8] 1994) and Ding [9] 1997) proved the existence and multiplicity of nontrivial solutions for a class of sublinear elliptic systems corresponding to 1.1). In 2007, Kristály [10] considered 1.1) with parameter λ in fx, u) and proved the finite multiplicity of the solutions for some uncertain values of λ. This result was improved in the recent paper [14]. When fx, t) =μax) t μ 2 t,whereμ 1, 2) is a constant and a : R is a positive continuous function such that a L 2/2 μ), [0, + )), byusingvariant fountain theorem [27], Zhang and Wang [26] established the following theorem on the existence of infinitely many nontrivial solutions of problem 1.1) under the assumptions that V satisfies some weaker conditions than those in [3], which have been given in [4]. Theorem 1.1. [26]). Assume that V and f satisfy the following conditions: S1) V C, R) and inf V x) =β>0; S2) There exists a constant l 0 > 0 such that lim meas { x : x y l 0,Vx) M } =0, y + M>0, where meas ) denotes the Lebesgue measure in ; S3) fx, t) =μax) t μ 2 t,whereμ 1, 2) is a constant and a : R is a positive continuous function such that a L 2/2 μ), [0, + )). Then problem 1.1) possesses infinitely many nontrivial solutions. In the above theorem, assumption S2) is a coercive condition on L, and S3) is a strict restriction on f. There are much sublinear functions in mathematical physics in problem like 1.1) except for fx, t) =μax) t μ 2 t in S3). In the present paper, motivated by paper [20, 26], we will use the genus properties in critical point theory to generalize Theorem 1.1 by removing assumption S2) and relaxing assumption S3). Theorem 1.2. Assume that V and f satisfy S1) and the following conditions:

3 Infinitely Many Solutions for a Class of Sublinear Schrödinger Equations 383 F1) f C R, R) and there exist two constants 1 <γ 1 <γ 2 < 2 and two functions a 1 L 2/2 γ1), [0, + )) and a 2 L 2/2 γ2), [0, + )) such that fx, t) γ 1 a 1 x) t γ γ 2 a 2 x) t γ 2 1, x, t) R; F2) There exist an open set J and three constants δ, η > 0 and γ 3 1, 2) such that F x, t) η t γ 3, x, t) J [ δ, δ], where F x, t) := t 0 fx, s)ds, x RN,t R. Then problem 1.1) possesses at least one nontrivial solution. Theorem 1.3. Assume that V and f satisfy S1), F1), F2) and the following condition: F3) fx, t) = fx, t), x, t) R. Then problem 1.1) possesses infinitely many nontrivial solutions. It is easy to see that assumption F2) is satisfied if the following condition holds: F2 ) There exist an open set J and three constants δ, η > 0 and γ 3 1, 2) such that tfx, t) γ 3 η t γ 3, x, t) J [ δ, δ]. Hence, by Theorems 1.1 and 1.2, we have the following corollary. Corollary 1.4. In Theorems 1.2 and 1.3, if assumption F2) is replaced by F2 ), then the conclusions still hold. Remark 1.5. If fx, t) =μax) t μ 2 t,thenf x, t) =ax) t μ. Hence, assumption S3) implies that F1), F2) and F3) with γ 1 = γ 3 = μ<γ 2 < 2, a 1 x) =ax) and a 2 x) =0. Remark 1.6. Our results can be applied to the following indefinite sign sublinear functions: 1.2) fx, t) = 1+sin2 x ) 1 1+ x N/2 t 3/4 t 3 t 1/2 t and 1.3) fx, t) = 4cosx 1 3 3sinx 2 t + 1+ x N/2) t 2/3 2 t, 1+ x N/3) t 1/2

4 384 Jing Chen and X. H. Tang where x =x 1,x 2,,x N ). See Examples 4.1 and 4.2 in Section 4. The remainder of this paper is organized as follows. In Section 2, some preliminary results are presented. The proofs of our main results are given in Section 3. Some examples to illustrate our results are given in Section PRELIMINARIES Let { E = u H 1 [ ) : u 2 + V x)u 2] } dx<+ equipped with the norm { [ u = u 2 + V x)u 2] } 1/2 dx, u E, and the inner product u, v)= [ u, v)+v x)uv]dx, u, v E. Then E is a Hilbert space with this inner product. As usual, for 1 p<+, welet ) 1/p u p = u p dx, u L p ), and u = ess sup u, u L ). x Lemma 2.1. Assume that S1) and F1) hold. Then the functional ϕ : E R defined by 2.1) ϕu) = 1 2 u 2 F x, u)dx is well defined and of class C 1 E,R) and 2.2) ϕ u),v =u, v) fx, u) vx)dx. Furthermore, the critical points of ϕ in E are solutions of problem 1.1). Proof. By virtue of F1), one has 2.3) F x, t) a 1 x) t γ 1 + a 2 x) t γ 2, x, t) R.

5 Infinitely Many Solutions for a Class of Sublinear Schrödinger Equations 385 For any u E, it follows from S1), 2.3) and the Hölder inequality that F x, u) dx [a 1 x) u γ 1 + a 2 x) u γ 2 ]dx β γ i/2 a i x) 2/2 γ i) dx ) 2 γi )/2 V x)u 2 dx ) γi /2 β γi/2 a i 2/2 γi ) u γ i, and so ϕ defined by 2.1) is well defined on E. Next, we prove that 2.2) holds. For any function θ : 0, 1), byf1)and the Hölder inequality, we have max fx, u + θx)hv)v dx h [0,1] max fx, u + θx)hv) v dx h [0,1] 2.4) γ i a i x) u + v ) R γi 1 v dx N γ i a i x) R u γi 1 + v γ i 1 ) v dx N + γ i β γ i/2 V x)u 2 dx γ i β γ i/2 a i x) 2/2 γ i) dx ) γ i 1 2 ) 2 γ i 2 V x)v 2 dx a i x) 2/2 γ i) dx ) 1 2 ) 2 γ i 2 V x)v 2 dx γ i β γi/2 a i 2/2 γi ) u γ i 1 + v γ i 1 ) v < +. Then by 2.1), 2.4) and Lebesgue s Dominated Convergence Theorem, we have ) γ i 2

6 386 Jing Chen and X. H. Tang ϕ ϕu + hv) ϕu) u),v = lim h 0 + h { 1 u + hv 2 u 2 } = lim [F x, u + hv) F x, u)]dx h 0 + h 2 ] = lim [u, v)+ h v 2 fx, u + θx)hv)vdx h R N =u, v) fx, u)vdx. This shows that 2.2) holds. Furthermore, by a standard argument, it is easy to show that the critical points of ϕ in E are solutions of problem 1.1). Let s prove now that ϕ is continuous. Let u k u in E. Thenu k u in L 2 ), and so 2.5) lim kx) =ux), k a.e. x. We claim that 2.6) lim k fx, u k ) fx, u) 2 dx =0. If 2.6) is not true, then there exist a constant ε 0 > 0 and a subsequence {u ki } such that 2.7) fx, u ki ) fx, u) 2 dx ε 0, i N. Since u k u in L 2 ), passing to a subsequence if necessary, it can be assumed that u k i u 2 2 < +. Set wx) =[ u k i x) ux) 2] 1/2, x. Then w L 2 ).Notethat fx, u ki ) fx, u) 2 2.8) 2 fx, u ki ) 2 +2 fx, u) 2 [ ] 4γ1 a 2 1 x) 2 u ki 2γ1 1) + u 2γ 1 1) [ ] +4γ2 2 a 2x) 2 u ki 2γ2 1) + u 2γ 2 1) [ ] 4 γ j +4)γj 2 a j x) 2 u ki u 2γj 1) + u 2γ j 1) j=1 [ ] 4 γ j +4)γj 2 a j x) 2 w 2γj 1) + u 2γ j 1) j=1 := gx), i N, x

7 Infinitely Many Solutions for a Class of Sublinear Schrödinger Equations 387 and 2.9) = gx)dx 4 γ j +4)γj 2 a j x) 2 4 γ j +4)γj 2 a j 2 2/2 γ j ) j=1 j=1 [ w 2γ j 1) + u 2γ j 1) ] dx ) w 2γ j 1) 2 + u 2γ j 1) 2 < +. Then by 2.5), 2.8), 2.9) and Lebesgue s Dominated Convergence Theorem, we have lim fx, u ki ) fx, u) 2 dx =0, i which contradicts 2.7). Hence 2.6) holds. From 2.2), 2.6) and the Hölder inequality, we have ϕ u k ) ϕ u),v = u k u, v) [fx, u k ) fx, u)]vdx R N u k u v + fx, u k ) fx, u) v dx 1/2 u k u v + β R 1/2 fx, u k ) fx, u) dx) 2 v N o1), k +, which implies the continuity of ϕ. The proof is complete. Lemma 2.2. [13]). Let X be a real Banach space and ψ C 1 X, R) satisfy the PS)-condition. If ψ is bounded from below, then c =inf X ψ is a critical value of ψ. To order to find nontrivial critical points of ψ, we will use the genus properties, so we recall the following definitions and results see [15] and [17]). Let X be a Banach space, ψ C 1 X, R) and c R. Weset Σ={A X {0} : A is closed in X and symmertric with respect to 0}, K c = {u X : ψu)=c, ψ u) =0}, ψ c = {u X : ψu) c}. Definition 2.3. [15]). For A Σ, we say genus of A is n denoted by γa)=n) if there is an odd map φ CA, R n \{0}) and n is the smallest integer with this property.

8 388 Jing Chen and X. H. Tang Lemma 2.4. [15]). Let ψ be an even C 1 functional on X and satisfy the PS)- condition. For any n N, set Σ n = {A Σ:γA) n}, c n = inf sup A Σ n u A ψu). i) If Σ n and c n R, thenc n is a critical value of ψ; ii) If there exists r N such that c n = c n+1 = = c n+r = c R, and c ψ0), thenγk c ) r PROOFS OF THEOREMS Proof of Theorem 1.2. In view of Lemma 2.1, ϕ C 1 E,R). In what follows, we first show that ϕ is bounded from below. By 2.1), 2.3) and the Hölder inequality, we have ϕu) = 1 2 u 2 F x, u)dx 1 2 u 2 a 1 x) u γ 1 dx a 2 x) u γ 2 dx 3.1) 1 ) 2 γi )/2 2 u 2 β γ i/2 a i x) 2/2 γi) dx R ) N γi /2 V x)u 2 dx 1 2 u 2 β γi/2 a i 2/2 γi ) u γ i. Since 1 <γ 1 <γ 2 < 2, 3.1) implies that ϕu) + as u +. Consequently, ϕ is bounded from below. Next, we prove that ϕ satisfies the PS)-condition. Assume that {u k } k N E is a sequence such that {ϕu k )} k N is bounded and ϕ u k ) 0 as k +. Thenby 3.1), there exists a constant A>0 such that 3.2) u k 2 β 1/2 u k A, k N. So passing to a subsequence if necessary, it can be assumed that u k u 0 in E. For any given number ε>0, by F1), we can choose R ε > 0 such that ) 2 γi )/2 3.3) a i x) 2/2 γi) dx <ε,, 2. x >R ε

9 Infinitely Many Solutions for a Class of Sublinear Schrödinger Equations 389 We claim that 3.4) lim k u k u 0 2 dx =0. x R ε In fact, if 3.4) is not true, then there exist a constant ε 0 > 0 and a subsequence {u ki } such that 3.5) u ki u 0 2 dx ε 0, x R ε i N. {u ki } has a convergent subsequence in L 2 B Rε ), which is still denoted by {u ki } for the sake of simplicity. Let {u ki } converge to u in L 2 B Rε ),i.e. 3.6) lim i u ki u 2 dx =0. x R ε Since u ki u 0 in E L 2 B Rε ), it follows that u ki u 0 in L 2 B Rε ),which, together with 3.6), implies that u x) =u 0 x), a.e.x B Rε. Hence, lim i u ki u 0 2 dx =0, x R ε which contradicts 3.5), and so 3.4) holds. By 3.4), there exists k 0 N such that 3.7) u k u 0 2 dx<ε 2 x R ε for k k 0. Hence, by F1), 3.2), 3.7) and the Hölder inequality, we have fx, u k ) fx, u 0 ) u k u 0 dx x R ε ) 1/2 ) 1/2 fx, u k ) fx, u 0 ) 2 dx u k u 0 2 dx x R ε x R ε [ 2 ] 1/2 fx, u k ) 2 + fx, u 0 ) 2) dx ε x R ε 3.8) [ 1/2 2 γi 2 a i 2 u k 2γi 1) + u 0 2γ i 1) dx] ε x R ε [ ) ] 1/2 2 γi 2 a i 2 2/2 γ i ) u k 2γ i 1) 2 + u 0 2γ i 1) 2 ε [ ) ] 1/2 2 γi 2 a i 2 2/2 γ i ) A 2γi 1) + u 0 2γ i 1) 2 ε, k k 0.

10 390 Jing Chen and X. H. Tang On the other hand, it follows from F1), 3.2), 3.3) and the Hölder inequality that fx, u k ) fx, u 0 ) u k u 0 dx x >R ε γ i a i x) u k x >R γi 1 + u 0 γi 1 ) u k + u 0 )dx ε 3.9) γ i a i x) u k x >R γ i + u 0 γ i )dx ε γ i a i x) 2/2 γi) dx x >R ε γ i a i x) 2/2 γi) dx x >R ε γ i A γ i + u 0 γ i 2 ) ε, k N. ) 2 γi )/2 ) 2 γi )/2 u k γ i 2 + u 0 γ i 2 ) A γ i + u 0 γ i 2 ) Since ε is arbitrary, combining 3.8) with 3.9), one gets 3.10) [fx, u k ) fx, u 0 ][u k u 0 ]dx 0 as k. It follows from 2.2) that 3.11) ϕ u k ) ϕ u 0 ),u k u 0 = u k u 0 2 [fx, u k ) fx, u 0 )][u k u 0 ]dx. Since ϕ u k ) ϕ u 0 ),u k u 0 0, it follows from 3.10) and 3.11) that u k u 0 in E. Hence, ϕ satisfies the PS)-condition. By Lemma 2.2, c =inf E ϕu) is a critical value of ϕ, that is there exists a critical point u E such that ϕu )=c. Finally, we show that u 0. Let u 0 W 1,2 0 J) ) E \{0} and u 0 1, then by 2.1) and F2), we have 3.12) ϕsu 0 )= s2 2 u 0 2 F x, su 0 )dx R N = s2 2 u 0 2 F x, su 0 )dx J s2 2 u 0 2 ηs γ 3 J u 0 γ 3 dx, 0 <s<δ.

11 Infinitely Many Solutions for a Class of Sublinear Schrödinger Equations 391 Since 1 <γ 3 < 2, it follows from 3.12) that ϕsu 0 ) < 0 for s>0 small enough. Hence ϕu )=c<0, therefore u is a nontrivial critical point of ϕ, andsou is a nontrivial solution of problem 1.1). The proof is complete. Proof of Theorem 1.3. In view of Lemma 2.1 and the proof of Theorem 1.2, ϕ C 1 E,R) is bounded from below and satisfies the PS)-condition. By F3), it is obvious that ϕ is even and ϕ0) = 0. In order to apply Lemma 2.4, we prove now that 3.13) for any n N there exists ε>0 such that γϕ ε ) n. For any n N, wetaken disjoint open sets J i such that n J i J. For i =1, 2,...,n,letu i W 1,2 0 J i ) ) E \{0}, u i < + and u i =1,and E n = span{u 1,u 2,,u n }, S n = {u E n : u =1}. For any u E n,thereexistλ i R,, 2,...,nsuch that 3.14) u = n λ i u i for x. Then ) 1/γ3 n ) 1/γ3 3.15) u γ3 = u γ 3 dx = λ i γ 3 u i γ 3 dx, J i and 3.16) u 2 = = [ u 2 + V x)u 2] n dx = λ 2 [ i ui 2 + V x)u 2 ] i dx J i n λ 2 [ i ui 2 + V x)u 2 ] n n i dx = λ 2 i u i 2 = λ 2 i. Since all norms of a finite dimensional normed space are equivalent, so there is a constant c > 0 such that 3.17) c u u γ3 for u E n. By F3), 2.1), 3.14), 3.15), 3.16) and 3.17), we have

12 392 Jing Chen and X. H. Tang 3.18) ϕsu) = s2 RN 2 u 2 F x, su)dx = s2 2 u 2 s2 2 u 2 ηs γ 3 n λ i γ 3 s2 2 u 2 ηc s) γ 3 u γ 3 n J i F x, sλ i u i )dx u i Ji γ 3 dx = s2 2 u 2 ηs γ 3 u γ 3 γ 3 ) 1 = s2 2 ηc s) γ 3, u S n, 0 <sδ max u i. 1in 3.18) implies that there exist ε>0 and σ>0 such that 3.19) ϕσu) < ε for u S n. Let { Sn σ = {σu : u S n }, Ω= λ 1,λ 2,,λ n ) R n : } n λ 2 i <σ 2. Then it follows from 3.19) that ϕu) < ε for u S σ n, which, together with the fact that ϕ C 1 E,R) and is even, implies that 3.20) S σ n ϕ ε Σ. On the other hand, it follows from 3.14) and 3.16) that there exists an odd homeomorphism mapping φ CSn σ, Ω). By some properties of the genus see 3 of Proposition 7.5 and Proposition 7.7 in [15]), we have 3.21) γϕ ε ) γs σ n )=n, so the proof of 3.13) follows. Set c n = inf sup A Σ n u A ϕu). It follows from 3.21) and the fact that ϕ is bounded from below on E that < c n ε<0, thatisforanyn N, c n is a real negative number. By Lemma 2.4, ϕ has infinitely many nontrivial critical points, and so problem 1.1) possesses infinitely many nontrivial solutions. The proof is complete.

13 Infinitely Many Solutions for a Class of Sublinear Schrödinger Equations 393 Example ). Then fx, t) 4. EXAMPLES In problem 1.1), let V x) =1+sin 2 x 1,andletfx, t) be as in 2 1+ x N/2 t 1/4 +3 t 1/2), x, t) R, ) F x, t) = 1+sin2 x x N/2 5 t 5/4 2 t 3/2 3 N+2)/ N/2 + π N/2) t 5/4, x, t) B π/3 [ 4 4, 4 4]. Thus all conditions of Theorem 1.3 are satisfied with 5 4 = γ 1 = γ 3 <γ 2 = 3 2, a 1x) = 2 1+ x N/2, a 2x) = δ = 1 4 4, η = 3 N+2)/ N/2 + π N/2), J = B π/3. By Theorem 1.3, problem 1.1) has infinitely many nontrivial solutions. Example ). Then fx, t) 6 1+ x N/2, In problem 1.1), let V x) =1+cos 2 x 1,andletfx, t) be as in x N/2) t 1/ x N/3) t 1/2, x, t) R, F x, t) = cos x 1 1+ x N/2 t 4/3 + sin x 2 1+ x N/3 t 3/ N N/4) t 4/3, x, t) 1, 1) N [ 1, 1]. Thus all conditions of Theorem 1.3 are satisfied with 4 3 = γ 1 = γ 3 <γ 2 = 3 2, a 1x) = x N/2), a 2x) = 1 δ =1, η = 2 1+N N/4), J = 1, 1)N. By Theorem 1.3, problem 1.1) has infinitely many nontrivial solutions x N/3),

14 394 Jing Chen and X. H. Tang Example 4.3. In problem 1.1), let V x) =ln3+ x 1 ), andlet Then fx, t) = fx, t) 3sinx x N/3) 1+ t 3/2) t 1/2 z x N/3) t 1/2, x, t) R, F x, t) = sin x 2 1+ x N/3 ln 1+ t 3/2) 1 4 [ 1+4N) N/6] t 3/2, x, t) 1, 2) N [ 1, 1]. These show that all conditions of Theorem 1.2 are satisfied, where γ 1 = γ 3 = 3 2 <γ 2 < 2, a 1 x) = x N/3), a 2x) =0, 1 δ =1, η = 4 [ 1+4N) N/6], J =1, 2)N. By Theorem 1.3, problem 1.1) has infinitely many nontrivial solutions. REFERENCES 1. T. Bartsch and M. Willem, Infinitely many nonradial solutions of a Euclidean scalar field equation, J. Funct. Anal., ), T. Bartsch and M. Willem, Infinitely many radial solutions of a semilinear elliptic problem on, Arch. Ration. Mech. Anal., ), T. Bartsch and Z.-Q. Wang, Existence and multiplicityresults for some superlinear elliptic problems on, Comm. Partial Differential Equations, ), T. Bartsch, Z.-Q. Wang and M. Willem, The Dirichlet problem for superlinear elliptic equations, in: M. Chipot, P. Quittner Eds.), Handbook of Differential Equations- Stationary Partial Differential Equations, vol. 2, Elsevier, 2005, pp. 1-5 Chapter 1). 5. H. Berestycki and P. L. Lions, Nonlinear scalar field equations, Arch. Ration. Mech. Anal., ), H. Brezis and S. Kamin, Sublinear elliptic equations in, Manuscripta Math., ), W.-Y. Ding and W.-M. Ni, On the existence of positive entire solutions of a semilinear elliptic equation, Arch. Ration. Mech. Anal., ), Y. H. Ding and S. J. Li, Existence of entire solutions for some elliptic systems, Bull. Aust. Math. Soc., ),

15 Infinitely Many Solutions for a Class of Sublinear Schrödinger Equations Y. H. Ding, Infinitely many entire solutions of an elliptic system with symmetry, Topol. Methods Nonlinear Anal., ), A. Kristály, Multiple solutions of a sublinear Schröinger equation, NoDEA Nonlinear Differential Equations Appl., ), Y. Li, Remarks on a semilinear elliptic equation on, J. Differential Equations, ), Y. Li, Nonautonomous scalar field equations, Indiana Univ. Math. J., ), J. Mawhin and M. Willem, Critical point theory and Hamiltonian systems, in: Applied Mathematical Sciences, Vol. 74, Springer-Verlag, New York, D. Moschetto, Existence and multiplicity results for a nonlinear stationary Schröinger equation, Ann. Polon. Math., ), P. H. Rabinowitz, Minimax methods in critical point theory with applications to differential equations, in: CBMS Reg. Conf. Ser. in Math., Vol. 65, Amer. Math. Soc., Providence, RI, P. H. Rabinowitz, On a class of nonlinear Schröinger equations, Z. Angew. Math. Phys., ), A. Salvatore, Homoclinic orbits for a special class of nonautonomous Hamiltonian systems, in: Proceedings of the Second World Congress of Nonlinear Analysis, Part 8 Athens, 1996), Nonlinear Anal., 308) 1997), W. A. Strauss, Existence of solitary waves in higher dimensions, Comm. Math. Phys., ), M. Struwe, Multiple solutions of differential equations without the Palais-Smale condition, Math. Ann., ), X. H. Tang and X. Y. Lin, Infinitely many homoclinic orbits for Hamiltonian systems with indefinite sign subquadratic potentials, Nonlinear Analysis, ), X. H. Tang, Infinitely many solutions for semilinear Schrödinger equations with signchanging potential and nonlinearity, J. Math. Anal. Appl., ), X. H. Tang, New conditions on nonlinearity for a periodic Schrödinger equation having zero as spectrum, J. Math. Anal. Appl., ), X. H. Tang, Ground state solutions for superlinear Schrödinger equation, Advance Nonlinear Studies, ), X. H. Tang, Non-Nehari manifold method for superlinear Schrödinger equation, Taiwan J. Math., 2014, DOI: /tjm M. Willem, Minimax Theorems, Birkhäuser, Boston, Q. Zhang and Q. Wang, Multiple solutions for a class of sublinear Schrödinger equations, J. Math. Anal. Appl., ),

16 396 Jing Chen and X. H. Tang 27. W. Zou, Variant fountain theorems and their applications, Manuscripta Math., ), Jing Chen Department of Mathematics Hunan University of Science and Technology Xiangtan, Hunan P. R. China X. H. Tang School of Mathematics and Statistics Central South University Changsha, Hunan P. R. China