Existence of periodic solutions

for some second order quasilinear

Hamiltonian systems
by
Mario Girardi
Dipartimento di Matematica
Universit`
a degli Studi di Roma Tre
Largo S. Leonardo Murialdo 00146 Roma, Italy
Michele Matzeu
Dipartimento di Matematica
Universit`
a Roma Tor Vergata
Viale della Ricerca Scientifica 00133 Roma, Italy
Supported by MURST, Project Variational Methods and Nonlinear Differential
Equations

2005, December

Abstract
A class of second order nonautonomous quasilinear Hamiltonian systems (S) is
considered. One states that, for any T < T0 , depending on the growth coefficients
of the Hamiltonian function H, there exists a T periodic and T /2antiperiodic
solution of (S), in case that two symmetry conditions hold for H.
Keywords. Quasilinear Hamiltonian systems, periodic solutions, Mountain Pass
techniques, iteration methods.
Subject classification 34C25, 49R99, 58E99, 47H15

Introduction
In the present paper we are interested in the existence of periodic solutions to the
following second order quasilinear Hamiltonian system

u(t) = b(t)H(u(t), u(t))

(S)

tR

where H denotes the gradient of H arth respect to its first variable in RN , and b
is a periodic function.
In [2] the authors, in collaboration with D. De Figueredo, introduced a method
in order to solve a quasilinear elliptic equaiton by a variational approach using
Mountain Pass techniques and some estimates for Mountain Pass solutions of some
semilinear problems suitably approximating the quasilinear problem.
A similar kind of approach allows to solve another quasilinear problem wiht
more general assumptions (see [4]), a semilinear integrodifferential equation with
nonsymmetric kernel (see [5]), finally a fully nonlinear elliptic equation (see [3]).
The aim of the present paper is to use some basic ideas of [2], [3], [4], [5], in
order to find periodic solutions of (S).
As a matter of fact, in [2], [3], [4], [5] it is very important tha the Poincare
inequality holds in the variational approach in order to state that the approximating
solutions actually converge to a solution of the initial poblem. In the present
case, we consider a variational approach in which, due to some suitable symmetry
assumption on H, one case use the Wirtinger inequality, which still allows to prove
the convergence of the approximating solutions.
The T periodicity of the solution founded is assured for any T < T0 , where T0
depends on the growth coefficients of H with respect to its two variables and on
the maximum of b(), which is further supposed to be T /2periodic. Indeed one
finds that this solution is even T /2antiperiodic.
Obviously, as a particular case, one can consider also autonomous Hamiltonian
systems, if one chooses b(t) 1.

1. The result
Let H : RN RN be a continuously differentiable function on the first variable
(let H denote the corresponding gradient) and continuous on the second variable.
Let b a continuous periodic function on R.
The problem is to find a nonzero periodic solution to the following second order
quasilinear Hamiltonian system

u(t) = b(t)H(u(t), u(t))

(S)

tR

One can state the following

Theorem 1. Let H be as above and let the further assumptions be satisfied:
(H0) H is Lipschitz continuous w.r. to its two variables
(H1) H(x, y) = H(x, y) x, y RN
(H2) H(x, y) = H(x, y) x, y RN
= 0 uniformly w.r. to y RN
(H3) lim H(x,y)
|x|2
x0

(H4) a1 > 0, p > 1, r (0, 1) such that

|H(x, y)| a1 (1 + |x|p )(1 + |y|r ) x, y RN
(H5) > 2 such that
0 < H(x, y) xH(x, y) x RN \{0} ,

y RN

(H6) a2 , a3 > 0 : H(x, y) a2 |y| a3 x, y RN

Then there exists a positive number T0 , explicitely depending on p, r, , a1 , a2 ,
a3 , N , B 1 , such that, for all T (0, T0 ) and for all T /2periodic functions b(),
there exists a nonzero classical T periodic and T /2antiperiodic solution of (S).
Remark 1. From (H4) and (H5) it follows p + 1.
Remark 2. Let us point out that one can always choose, for any T0 > 0, b(t) 1,
then Theorem 1 assures the existence a T periodic and T /2antiperiodic nonzero
solution, for any T < T0 , of the autonomous Hamiltonian system

u(t) = H(u(t), u(t))

tR

if (H0), . . . , (H6) hold.

Remark 3. Let us note that, as a consequence of the T periodicity and T /2
antiperiodicity of the solution u(), one gets
Z

u(t)dt = 0
0

As a standard example of a function H satisfying (H0), . . . , (H6), one can consider

H(x, y) = h(x, y) with

h(x, y) =

a
1 x|x|p1 (1 + |y| ) ,
x|x|p1 |y|r ,

|x| 1 , |y| 2
|x| 1 , |y| 2

where p, > 1, r (0, 1), moreover a

i j j (j = 1, 2) are fixed positive constants
with j > j .

1B

= max{|b(t)| : t [0, T ]}.

2. A variational approach to the problem

Let us consider the space
V = {v H 1 ([0, T ]; RN ) : v(0) = v(T )}
and its L2 orthogonal decomposition
V = V 0 V1 V 2
where

(
V 0 = RN =

1
T

v(t)dt : v V
0

X
i2k
t
V1 = v V : v(t) =
ak e T , ak = ak

k=2h+1
hZ

V2 = v V : v(t) =

ak e

i2k
T

, ak = ak

k=2h
hZ\{0}

Let us fix w in V1 and consider the following functional

1
Iw (v) =
2

|v(t)| dt
0

b(t)H(v(t), w(t))dt

v V

Proposition 1. Let b() be T /2periodic, then any critical point uw of Iw on V1 is

a weak solution of the following second order Hamiltonian system with T periodic
boundary conditions

(Sw )

uw (t) = b(t)H(uw (t), u w (t))

uw (0) = uw (T ) ,

u w (0) = u w (T )

Moreover uw is T /2antiperiodic.
Proof . Let uw be critical on V1 , that is
hIw0 (uw ), v1 i = 0 v1 V1
We have to prove that
hIw0 (uw ) , v0 + v1 + v2 i = 0 v = v0 + v1 + v2 V = V0 V1 V2
(as the critical points of Iw on the whole space V are the weak solutions of (Sw ),
and, if uw V1 , then uw is T /2antiperiodic).
It is enough to prove that
hIw0 (uw ), v0 i = hIw0 (uw ), v2 i = 0 v0 V0 ,

v1 V2

On the other side,

Iw0 (uw (t)) = u00w (t) b(t)H(uw (t), w(t))

and one has, by (H1) (H2) and the T /2periodicity of b(),

b(t + T /2)H(uw (t + T /2), w(t
+ T /2)) = b(t + T /2)H(uw (t), w(t))

=
= b(t)H(uw (t), w(t))

= b(t)H(uw (t)), w(t))

Therefore the function

b(t)H(uw (t), w(t))

is T /2antiperiodic, so it is orthogonal to V2 as well as to V0 , as it has zero mean.

Since u00w obviously belongs to V2 , the proof is complete. 
In the following we will put
!1/2

2
|v(t)|

dt

kvk =

v V1

which is an equivalent norm to the H 1 norm of V , as the Wirtinger inequality holds

in the space V1 .
Therefore the functional Iw , for any w V1 , has the form
1
Iw (v) = kvk2
2

b(t)H(v(t), w(t))dt

v V1

3. Proof of Theorem 1
First of all, let us fix R > 0 and put
CR = {v V1 C 2 ([0, T ]) : kvkC 2 (0,T ) R}
Proposition 2. For any w CR , there exists a Mountain Pass critical point of
uw CR for Iw on V1 (as defined in Step 3).
We prove Proposition 2 by steps.
Step 1. Let w CR . Then there exists R , R > 0 depending on R, but not on
w such that
Iw (v) R v V1 : kvk = R
Proof . From (H1) it follows that, for any > 0, there exists > 0 such that
H(v(t), w(t))

<

1
|v(t)|2 t [0, T ] ,
2

t [0, T ] ,

|v(t)|

hence
Z

T
2

|v(t)|

dt + K(1 + R)

H(v(t), w(t))dt

< /2

|v(t)|p+1 dt

K 0 (/2 + K(1 + R)r kvkp1 )kvk2

with a constant K 0 depending on the Wirtinger and Sobolev inequalities. Choosing

kv| =

2K(1 + R)r

1/p1
= R

one gets
Z

H(v, w)
K 0 kvk2

Recalling that B = max |b(t)|, if one chooses <

t[0,T ]

1
2BK 0

and R =

1
2

K 0 2R ,

the thesis follows. 

Step 2. Let w CR and let us fix v in V1 with k
v k = 1. Then there exists s > 0,
independent of w and R, such that
(1)

Iw (s
v) 0

s s ,

then v = sv satisfies the relations

kvk > R

Iw (v) 0

Proof . It follows from (H6) that

1
Iw (s
v ) s2 a2 |s|
2

|v| dt + a3 T

By the Sobolev embedding theorem as p + 1 (see Remark 1), one gets

Iw (s
v)

1 2
s a2 |s| (S ) + a3 T
2

where S is the embedding constant of V1 in L ([0, T ]). Since > 2, one gets some
s such that (1) holds.
Step 3. Let w CR . Then there exists a Mountain Pass critical pont uw for Iw on
V1 , that is
(2)

Iw (uw ) = inf max I((t))

T t[0,1]

where
= { C 0 ([0, 1]; RN ) : (0) = 0 ,

(1) = v}

and
Iw (uw ) R > 0

( uw 6 0)

Proof . It is a consequence of the theorem by Ambrosetti and Robinowitz (see [1]),

as Iw (0) = 0, Step 1 and Step 2 hold, and the PalaisSmale condition is trivially
satisfied, due to the continuous embedding of V1 into L (0, T ; RN ) and the fact
that p > 2. 
Proposition 3. Let w CR and let uw be a Mountain Pass solution given by
Proposition 2. Then there exists a positive constant c1 (R) depending on R, but
not on w, such that
kuw k c1 (R)
Proof . Actually the estimate holds for any critical point uw of Iw on V1 with uw 6 0
and one does not use the Mountain Pass nature of uw . Indeed, if one puts in the
relation
Z
Z
T

u w v =

v V1

b(t)H(uw , w)v

v = uw , one gets
T

kuw k =

(3)

b(t)H(uw , w)u
w dt
0

From (H3), (H4), (H5), it follows that, for any > 0, there exists a positive
constant c , R, depending on and R, but not on w, such that
|H(uw , w)|
|uw | + c,R |uw |p
Using this inequality, (3) yields
kuw k2 B

|uw |2 + c,R

|uw |p+1

hence, by the Wirtinger inequality and the continuous Sobolev embedding,

(1 (T /2)2 )kuw k2 c,R kuw kp+1


2 2
,
T

which implies the thesis choosing <

as p + 1 > 2. 

Proposition 4. Let w CR and let uw be a Mountain Pass solution given by

Proposition 2. Then there exists a constant c2 > 0, independent of w and R, such
that
kuw k c2
Proof . From the infmaximum characterization of uw , by choosing in as the
segment line joining 0 and v, one gets
Iw (uw ) sup Iw (sv) ,
s0

hence, by (H6),
(
Iw (uw ) B sup
s0

s2
2

Z
0

T
2

a2 |s|
|v|

|v| + a3 T
0

Since > 2, such an upper bound is a maximum and it does not depend on R and
w, hence
Iw (uw ) const R > 0 , w CR
At this point, using the criticality of uw for Iw , (H5) and (3), one gets
1
1
kuw k2 const +
2

b(t)H(uw , w)u
w = const +
0

1
kuw k2

by which the estimate follows, as > 2. 

Actually, for any w CR , any Mountain Pass solution of (Sw ) is not only weak,
but just of class C 2 , so a classical solution, since it solves a problem of the type

(4)

uw (t) = (t)(= b(t)H(uw (t), w(t)))

t [0, T ]
uw (0) = uw (T ) ,

u w (0) = u w (T )

where belongs to C 0 ([0, T ]).

Proposition 5. There exists a constant > 0 such that the Mountain Pass
solutions uw of (4) verify, for any R > 0, and any w CR ,
kuw kC 2 ([0,T ]) (1 + Rr )
Proof . It is a consequence of (H4), of Proposition 4 and the fact that uw solves
(4) in the classical sense. 
Proposition 6. There exists a constant R > 0 such that
w CR uw CR
for any Mountain Pass solution uw of (4).
Proof . It easily follows from Proposition 5 and the fat that r < 1. 
At this point it is very natural to introduce an iterative scheme in the following
way. Let R be given by Proposition 6 and let u0 be arbitrarily fixed in CR . Let us
define un as a Mountain Pass solution the following problem, for any n N,


un (t) = b(t)H(un (t), u n1 (t)) t [0, T ]

(Sn )
un (0) = un (T ) , u n (0) = u n (T )
Obviously, by Proposition 6, one has, for any n N,
un CR
Now we are in a position to give the
Proof of Theorem 1 . Let us consider un given as a Mountain Pass solution of (Sn )
for any n N. First of all we prove that there exists some positive number T0 , such
that, if T < T0 , then the whole sequence {un } strongly converges in V1 .

Indeed, using (sn ) and (sn+1 ), one gets

Z T
T
b(t)H(un+1 (t), u n (t))(un+1 (t) un (t))dt
u n+1 (t)(u n+1 (t) u n (t))dt =
0
T

Z
u n (t)(u n+1 (t) u n (t))dt =

b(t)H(un (t), u n1 (t))(un+1 (t) un (t))dt

0

which yields
Z

|H(un+1 (t), u n (t)) H(un (t), u n1 (t))kun+1 (t) un (t)|dt

kun+1 un k B
0

(5)

|H(un+1 (t), u n (t)) H(un (t), u n (t))kun+1 (t) un (t)|dt+

0
Z T

|H(un (t), u n (t)) H(un (t), u n1 (t))| |un+1 (t) un (t)|dt

+B
0
0
00
cR , cR

Denoting by
the best Lipschitz constants of H w.r. to its two variables in
the set BR BR where
BR = {x RN : |x| R} ,
one gets from (5) the relation

Z T
Z
2
00
2
0

|un+1 un | dt + cR kun un1 k

kun+1 un k B cR
0

!1/2

|un+1 un |2 dt

Using the Wirtinger inequality (as un and un+1 are T periodic with zero mean),
one obtains
!
 2
 
T
T
kun+1 un k2 B c0R
kun+1 un k2 + c00R
kun un1 k kun+1 un k
2
2
from which
(6)

kun+1 un k

00
BcR

T
2

T 2
2

1 Bc0R

if

kun un1 k = kun un1 k

2
T <p 0
BcR

Actually, by putting

(7)

2
T0 = min p 0 ,
BcR

B 2 (c00R )2 4Bc0R Bc00R

2Bc0R

and if T < T0 , the constant in (6) is less than 1. Therefore if T < T0 , given by
(7), then (6) implies that {un } is a Cauchy sequence in V1 , so it strongly converges
to some u in V1 .
At this point, from the AscoliArzelas thereom and the fact that {un } is contained in CR , it follows that the whole sequence {un } converges in C 2 ([0, T ]), then,
as it easily verified, to a classical solution u of (S). The fact that u is not identically
zero is an immediate consequence of Proposition 3, putting R = R. 

10

References
[1] A. Ambrosetti, P.H. Rabinowitz, Dual variational methods in critical point
theory and applications, J. Funct. Analysis 14, (1973), 349381.
[2] D. De Figuereido, M. Girardi, M. Matzeu, Semilinear elliptic equations
with dependence on the gradient via mountainpass tecniques, Differential and
Integral Equations 17, 12 (2004), 119126.
[3] M. Girardi, M. Matzeu, Positive and negative solutions of a quasilinear
equation by a MountainPass method and truncature techniques, Nonlinear
Analysis 59 (2004), 199210.
[4] M. Girardi, S. Mataloni, M. Matzeu, Mountain Pass techniques for some
classes of nonvariational problems, to appear on Proceedings ISAAC 2005.
[5] S. Mataloni, M. Matzeu, Semilinear integrodifferential problems with non
symmetric kernels via mountainpass techniques, Advanced Nonlinear Sutides
5 (2005), 2331.