Open Access

New periodic solutions for a class of singular Hamiltonian systems

Boundary Value Problems20142014:42

https://doi.org/10.1186/1687-2770-2014-42

Received: 14 November 2013

Accepted: 10 February 2014

Published: 19 February 2014

Abstract

We use the variational minimizing method to study the existence of new nontrivial periodic solutions with a prescribed energy for second order Hamiltonian systems with singular potential V C 1 ( R n { 0 } , R ) , which may have an unbounded potential well.

MSC:34C15, 34C25, 58F.

Keywords

singular Hamiltonian systemsperiodic solutionsvariational methods

1 Introduction and main results

For singular Hamiltonian systems with a fixed energy h R ,
q ¨ + V ( q ) = 0 ,
(1.1)
1 2 | q ˙ | 2 + V ( q ) = h .
(1.2)

Ambrosetti-Coti Zelati [1, 2] used Ljusternik-Schnirelmann theory on an C 1 manifold to get the following theorem.

Theorem 1.1 (Ambrosetti-Coti Zelati [1])

Suppose V C 2 ( R n { 0 } , R ) satisfies

(A0)
V ( u ) , u 0 ,
(A1)
3 V ( u ) u + ( V ( u ) u , u ) 0 ,
(A2)
V ( u ) u > 0 , u 0 ,
(A3) α > 2 , s.t.
V ( u ) u α V ( u ) ,
(A4) β > 2 , r > 0 , s.t.
V ( u ) u β V ( u ) , 0 < | u | < r ,
(A5)
lim sup | u | + [ V ( u ) + 1 2 V ( u ) u ] 0 .

Then (1.1)-(1.2) have at least one non-constant periodic solution.

After Ambrosetti-Coti Zelati, a lot of mathematicians studied singular Hamiltonian systems. Here we only mention a related recent paper of Carminati-Sere-Tanaka [3], in which they used complex variational and geometrical and topological methods to generalize Pisani’s results [5]. They got the following theorems.

Theorem 1.2 Suppose h > 0 , L 0 > 0 and V C ( R n { 0 } , R ) satisfies (A0), (A4), and

(B1) V ( q ) 0 ;

(B2) V ( q ) + 1 2 V ( q ) q h , | q | e L 0 ;

(B3) V ( q ) + 1 2 V ( q ) q h , | q | e L 0 .

Then (1.1)-(1.2) have at least one periodic solution with the given energy h and whose action is at most 2 π r 0 with
r 0 = max { [ 2 ( h V ( q ) ) ] 1 2 ; | q | = 1 } .

Theorem 1.3 Suppose h > 0 , ρ 0 > 0 and V C ( R n { 0 } , R ) satisfies (B1), (A4), and

(B2′) lim | q | + V ( q ) = 0 ;

(B3′) V ( q ) + 1 2 V ( q ) q h , | q | ρ 0 .

Then (1.1)-(1.2) have at least one periodic solution with the given energy h and whose action is at most 2 π r 0 .

Using variational minimizing methods, we get the following theorem.

Theorem 1.4 Suppose V C 1 ( R n { 0 } , R ) satisfies

(V1) α > 0 , β > 2 , r > 0 , s.t.
V ( q ) α | q | β , 0 < | q | < r ;
(V2)
V ( q ) < 0 , q 0 ;
(V3)
V ( q ) = V ( q ) , q 0 .

Then for any h > 0 , (1.1)-(1.2) have at least one non-constant periodic solution with the given energy h.

2 A few lemmas

Let
H 1 = W 1 , 2 ( R / Z , R n ) = { u : R R n , u L 2 , u ˙ L 2 , u ( t + 1 ) = u ( t ) } .
Then the standard H 1 norm is equivalent to
u = u H 1 = ( 0 1 | u ˙ | 2 d t ) 1 / 2 + | 0 1 u ( t ) d t | .
Let
Λ = { u H 1 | u ( t ) 0 , t } .
By symmetry condition (V3), similar to Ambrosetti-Coti Zelati [1], let
Λ 0 = { u H 1 = W 1 , 2 ( R / Z , R n ) , u ( t + 1 / 2 ) = u ( t ) , u ( t ) 0 } .
We define the equivalent norm in E = { u H 1 = W 1 , 2 ( R / Z , R n ) , u ( t + 1 2 ) = u ( t ) } :
u = u E = ( 0 1 | u ˙ | 2 d t ) 1 / 2 .

Lemma 2.1 ([1, 4])

Let f ( u ) = 1 2 0 1 | u ˙ | 2 d t 0 1 ( h V ( u ) ) d t and u ˜ Λ be such that f ( u ˜ ) = 0 and f ( u ˜ ) > 0 . Set
1 T 2 = 0 1 ( h V ( u ˜ ) ) d t 1 2 0 1 | u ˜ ˙ | 2 d t .
(2.1)

Then q ˜ ( t ) = u ˜ ( t / T ) is a non-constant T-periodic solution for (1.1)-(1.2). Furthermore, if V ( x ) < h , x 0 , then f ( u ) 0 on Λ and f ( u ) = 0 , u Λ if and only if u is a nonzero constant.

If u ˜ Λ 0 such that f ( u ˜ ) = 0 and f ( u ˜ ) > 0 , then we find that q ˜ ( t ) = u ˜ ( t / T ) is a non-constant T-periodic solution for (1.1)-(1.2).

Lemma 2.2 (Gordon [6])

Let V satisfy the so-called Gordon Strong Force condition: There exist a neighborhood of 0 and a function U C 1 ( Ω , R ) such that:
  1. (i)

    lim s 0 U ( x ) = ;

     
  2. (ii)

    V ( x ) | U ( x ) | 2 for every x N { 0 } .

     
Let
Λ = { u H 1 = W 1 , 2 ( R / Z , R n ) , t 0 , u ( t 0 ) = 0 } .
Then we have
0 1 V ( u ) d t , u n u Λ .
Let
Λ 0 = { u H 1 = W 1 , 2 ( R / Z , R n ) , u ( t + 1 2 ) = u ( t ) , t 0 , u ( t 0 ) = 0 } .
Then we have
0 1 V ( u ) d t , u n u Λ 0 .
Lemma 2.3 Let X be a Banach space, and let E X be a weakly closed subset. Suppose that ϕ ( u ) is defined on an open subset Λ X and ϕ ( u ) for any u Λ . Let ϕ ( u ) = + for u Λ . Assume ϕ ( u ) + and is weakly lower semi-continuous on Λ ¯ E , and that it is coercive on Λ E :
ϕ ( u ) + , u +
and
ϕ ( u n ) + , u n u Λ .

Then ϕ attains its infimum in Λ E .

Proof We set
c = inf Λ E ϕ ( u ) .
Then
< c < + ,

in fact, by the assumptions, it is obvious that c < + . Now if c = , then there exists { u n } Λ E such that ϕ ( u n ) . Then we know that { u n } is bounded, since ϕ is coercive. By the Eberlein-Schmulyan theorem, { u n } has a weakly convergent subsequence. Finally, by the definition for c and the assumption for the weakly lower semi-continuity for ϕ ( u ) , we know ϕ ( u ) = . This is a contradiction.

Now we know that there exists minimizing sequence { u n } such that ϕ ( u n ) c . Furthermore by the coercivity of ϕ we know that { u n } is bounded; then { u n } has a weakly convergent subsequence. We claim the weak limit u Λ , since otherwise ϕ ( u ) = + by the assumption. On the other hand, by the definition of the infimum c and the assumption for the weak lower semi-continuity for ϕ ( u ) on Λ ¯ E , we know ϕ ( u ) = c < + . This is a contradiction. So the weak limit u Λ E and ϕ ( u ) = c . □

3 The proof of Theorem 1.4

Lemma 3.1 Assume (V1) hold, then for any weakly convergent sequence u n u Λ 0 , we have
f ( u n ) + .
Proof Notice that (V1) imply Gordon’s strong force condition. By the weak limit u Λ and V satisfying Gordon’s strong force condition, we have
0 1 V ( u n ) d t + , u n u Λ .
By u n u in the Hilbert space H 1 , we know that u n is bounded.
  1. (1)
    If u 0 , then by Sobolev’s embedding theorem, we have the uniform convergence property:
    | u n | 0 , n + .
     
By the symmetry of u ( t + 1 / 2 ) = u ( t ) , we have 0 1 u ( t ) d t = 0 , then we have Sobolev’s inequality:
0 1 | u ˙ ( t ) | 2 d t 12 | u ( t ) | 2 .
Then we have
f ( u n ) 6 | u n | 2 β + , n + .
So in this case we have
lim inf f ( u n ) = + f ( u ) .
  1. (2)
    If u 0 , then we have the following. By the weakly lower semi-continuity for the norm, we have
    lim inf u n u > 0 .
     
So, by Gordon’s lemma, we have
lim inf f ( u n ) = lim inf ( 1 2 0 1 | u ˙ n | 2 d t ) 0 1 ( h V ( u n ) ) d t = + 1 2 0 1 | u ˙ | 2 d t 0 1 ( h V ( u ) ) d t = f ( u ) .

 □

Lemma 3.2 f ( u ) is weakly lower semi-continuous on Λ ¯ 0 .

Proof For any { u n } Λ ¯ 0 : u n u , by Sobolev’s embedding theorem, we have uniform convergence:
| u n ( t ) u ( t ) | 0 .
  1. (i)
    If u Λ 0 , then by V C 1 ( R n { 0 } , R ) , we have
    | V ( u n ( t ) ) V ( u ( t ) ) | 0 .
     
By the weakly lower semi-continuity for norm, we have
lim inf u n u .
Hence
lim inf f ( u n ) = lim inf ( 1 2 0 1 | u ˙ n | 2 d t ) 0 1 ( h V ( u n ) ) d t 1 2 0 1 | u ˙ | 2 d t 0 1 ( h V ( u ) ) d t = f ( u ) .
  1. (ii)
    If u Λ 0 , then by Λ satisfying Gordon’s strong force condition, we have
    0 1 V ( u n ) d t + , u n u Λ 0 .
     
  2. (1)
    If u 0 , then
    | u n | 0 , n + .
     
Then we have
f ( u n ) 6 | u n | 2 β + , n + .
So in this case we have
lim inf f ( u n ) = + f ( u ) .
  1. (2)
    If u 0 . By the weakly lower semi-continuity for norm, we have
    lim inf u n u > 0 .
     
So by Gordon’s lemma, we have
lim inf f ( u n ) = lim inf ( 1 2 0 1 | u ˙ n | 2 d t ) 0 1 ( h V ( u n ) ) d t = + 1 2 0 1 | u ˙ | 2 d t 0 1 ( h V ( u ) ) d t = f ( u ) .

 □

Lemma 3.3 Λ ¯ 0 is a weakly closed subset of H 1 .

Proof By Sobolev’s embedding theorems, the proof is obvious. □

Lemma 3.4 The functional f ( u ) is coercive on Λ 0 .

Proof By the definition of f ( u ) and the assumption (V2), we have
f ( u ) = 1 2 0 1 | u ˙ | 2 d t 0 1 ( h V ( u ) ) d t h 2 0 1 | u ˙ | 2 d t , u Λ 0 .

 □

Lemma 3.5 The functional f ( u ) attains the infimum on Λ 0 ; furthermore, the minimizer is non-constant.

Proof By Lemma 2.2 and Lemmas 3.1-3.3, we know that the functional f ( u ) attains the infimum in Λ 0 ; furthermore, we claim that
inf Λ 0 f ( u ) > 0 ,

since otherwise, u 0 ( t ) = const attains the infimum 0, then by the symmetry of Λ 0 , we have u 0 ( t ) 0 , which contradicts the definition of Λ 0 . Now we know that the minimizer is non-constant. □

Declarations

Acknowledgements

This study was supported by the Scientific Research Fund of Sichuan Provincial Education Department (11ZA172) and the Scientific Research Fund of Science Technology Department of Sichuan Province (2011JYZ010).

Authors’ Affiliations

(1)
Department of Mathematics, Yibin University
(2)
Department of Mathematics, Sichuan University

References

  1. Ambrosetti A, Coti Zelati V: Closed orbits of fixed energy for singular Hamiltonian systems. Arch. Ration. Mech. Anal. 1990, 112: 339-362. 10.1007/BF02384078MathSciNetView ArticleGoogle Scholar
  2. Ambrosetti A, Coti Zelati V: Periodic Solutions for Singular Lagrangian Systems. Springer, Berlin; 1993.View ArticleGoogle Scholar
  3. Carminati C, Sere E, Tanaka K: The fixed energy problem for a class of nonconvex singular Hamiltonian systems. J. Differ. Equ. 2006, 230: 362-377. 10.1016/j.jde.2006.01.021MathSciNetView ArticleGoogle Scholar
  4. Tanaka K: Periodic solutions for singular Hamiltonian systems and closed geodesics on non-compact Riemannian manifolds. Ann. Inst. Henri Poincaré, Anal. Non Linéaire 2000, 17: 1-33. 10.1016/S0294-1449(99)00102-XView ArticleGoogle Scholar
  5. Pisani L: Periodic solutions with prescribed energy for singular conservative systems involving strong forces. Nonlinear Anal. TMA 1993, 21: 167-179. 10.1016/0362-546X(93)90107-4MathSciNetView ArticleGoogle Scholar
  6. Gordon WB: Conservative dynamical systems involving strong forces. Trans. Am. Math. Soc. 1975, 204: 113-135.View ArticleGoogle Scholar

Copyright

© Wang and Shen; licensee Springer. 2014

This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited.