Boundedness and Oscillation of Third Order Neutral Differential Equations with Deviating Arguments ()
1. Introduction
In this paper we consider third order neutral differential equations of the form
(1)
where
and the following conditions are satisfied
(A1)
and
,
(A2)
,
is strictly increasing,
and we define 
(A3) 
(A4)
, f is non-decreasing and
for
,
(A5)
and
is not zero on any half line 
(A6)
,
for
and
,
is continuous, has positive partial derivative on
with respect to t, nondecreasing with respect to
and ![]()
(A7)
,
is nondecreasing and the integral of Equation (1) is in the sense Riemann-stieltijes.
We mean by a solution of Equation (1) a function
,
such that
,
,
and
exist and are continuous on
. A nontrivial solution of (1) is called oscillatory if it has arbitrarily large zeros, otherwise it is called non-oscillatory.
Asymptotic properties of solutions of differential equations of the second and third order have been subject of intensive studying in the literature. This problem for neutral differential equations has received considerable attention in recent years (see [1] - [11] ).
Recently, in [12] by using Riccati technique, have established some general oscillation criteria for third-order neutral differential equation
![]()
In [3] , Candan presented several oscillation criteria for third order neutral delay differential equation
![]()
[9] and [13] obtained some oscillation criteria for study third order nonlinear neutral differential equations
![]()
and
![]()
In this paper, we establish some oscillation criteria for Equation (1), which complement and extend the results in [3] [13] .
We begin with analyzing of the asymptotic behavior of possible non-oscillatory solutions of the Equation (1) in the case when
. Let
be a non-oscillatory solution of (1) on
. From (1) it follows that the function
has to be eventually of constant sign, so either
(a) ![]()
or
(b) ![]()
for all sufficiently large t. Denote by
[or
] the set of all non-oscillatory solutions
of the Equation (1) such that (a) [or (b)] is satisfied. We begin with some useful lemmas.
Lemma 1.1 Let
. Assume that (A1) and (A2) hold and x be continuous non-oscillatory solution of the functional inequality (a). Then
![]()
Lemma 1.2 Let
. Assume that (A1) and (A2) hold and x be continuous non-oscillatory solution of the functional inequality (b). If
then
![]()
These lemmas are modifications of the Lemma 1 in the paper [14] and the Lemma 2 in the paper [13] .
2. Main Results
In this part, for the sake of convenience, we introduce the following notation:
![]()
2.1. Oscillation Criteria If ![]()
In this section, we will establish some oscillation criteria for Equation (1) in the case when
and
.
Lemma 2.1 Let x be a bounded positive solution of Equation (1) on the interval I. Then there exists a
such that
has the following properties:
(2)
Proof. Let x be a bounded positive solution of Equation (1) on the interval I. From (A1), (A2) and (A6), there exists a
such that
,
and
for
. Then
is bounded and non-oscillatory. Thus, Equation (1) implies that
![]()
Hence, the function
is a non-increasing and of one sign. We claim that
for
. Suppose that
for
. Then there exists a
and constant
such that
![]()
By integrating the last inequality from
to t, we get
![]()
Letting
, from (A3), we have
. Then there exists a
and constant
such that
![]()
By integrating this inequality from
to t and using (A3), we get
. This yields that
and this contradicts the Lemma 1.1. Now we have
for
. Hence
is increasing function and we have two possible cases for
either
eventually or
eventually for
. If
for
, then there exist a
and a constant
such that
![]()
By integrating this inequality from
to t and using (A3), we get
. This means that
and we get
for all sufficiently large t. Then
, which contradicts the boundedness of
. Hence,
for
. ![]()
Theorem 2.1 if
(3)
Then every bounded solution
of Equation (1) is either oscillatory or tends to zero.
Proof. Let x be a bounded non-oscillatory solution of Equation (1) on the interval I. Without loss of generality we may assume that
. From Lemma 2.1, we get that (2) holds. New, we have
![]()
for all sufficiently large t. Repeating this procedure and the monotonicity of
, we obtain that there exists an
integer
such that
and
![]()
where
. Hence, we get
(4)
Thus, from Equation (1), we obtain
(5)
Now, since
is bounded decreasing function, then there exist
such that
![]()
If
for
, then
and which contradicts the Lemma 1.1. Therefore
for
and
. We shall prove that
. Let
. For
, we obtain
![]()
Thus, form Lemma 2.1, we get
(6)
So, for
, we have
![]()
Hence, from (6), we get
(7)
where
. Let us define function
![]()
We note that
. Deriving
partially with respect to s and using Lemma 2.1, (A4) and (A6), we get
![]()
From (5), we have
. Hence, we obtain
(8)
By (A4) and (A6), we get
![]()
Thus, from (7), we have
(9)
Then, substituting (8) in (9), it follows that
![]()
By integrating this inequality from
to t with respect to s, we obtain
(10)
where
. Since
, we get
![]()
Hence, from (10), we have
![]()
which contradicts (3). Therefore,
and according to the Lemma 1.2 we have that
. ![]()
In the following Theorem, we establish some sufficient conditions for boundedness and oscillation of Equation (1) under the condition
(11)
Theorem 2.2 Let (11) holds. If there exist an integer
such that
(12)
then every bounded solution
of Equation (1) is oscillatory.
Proof. Let x be a bounded non-oscillatory solution of Equation (1) on the interval I. Without loss of generality we may assume that
. We can proceed exactly as in the proof of Theorem 2.1 and we use the fact that (12) implies (3). Hence, we get a non-oscillatory solution with the properties
,
, ![]()
and
for
,
and
. New, from (4), there exists
such that
![]()
Thus, Equation (1) implies that
![]()
By integrating this inequality from
to t, we get
![]()
where
. Thus, we obtain
(13)
where
. Since
, from the Inequality (7), we get
(14)
Combining (13) and (14), we have
![]()
Hence, we get
![]()
for
and this contradicts the condition (12).
Corollary 2.1 Let (11) holds. If
(15)
then every bounded solution
of Equation (1) is oscillatory.
Example 2.1 Consider the differential equation
![]()
where
. We have
![]()
and
. Thus, all conditions of Corollary 2.1 are satisfied then all bounded solutions
of the above equation are oscillatory.
Remark 2.1 If
and
then, our results extend the results in [13] .
2.2. Oscillation Criteria If ![]()
In this section, we will present some oscillation criteria for Equation (1) under the case
and the condition
(16)
Lemma 2.2 If
is an eventually positive solution of (1), then for sufficiently large t, there are only two possible cases:
(i) ![]()
(ii) ![]()
Proof. The proof of this lemma is similar to the proof Lemma 1 in [9] and we omit the details. ![]()
Theorem 2.3 Let (16) holds. If
(17)
and there exist a positive real function
such that
(18)
Then every solution of Equation (1) is either oscillatory or tends to zero.
Proof. Let x be a non-oscillatory solution of Equation (1) on the interval I. Without loss of generality we may assume that
. Then there exists a
such that
,
and
for
. By Lemma 2.2, we have two cases for
. In the Case (i), since
and
, we get
. Let
, then we have
for all
and t enough large. Choosing
, we obtain
![]()
where
. Hence, from (1), (A6) and (16), we have
![]()
By integrating two times from t to
, we get
![]()
Integrating the last inequality from
to
, we obtain
![]()
This contradicts to the condition (17), then
, which implies that
. In the Case (ii),
since
and
. Then there exist a
such that
![]()
for
. Thus, from (1), (A4) and (A6), we get
(19)
Also, we have
![]()
Since
, we obtain
(20)
Now, we define
![]()
By differentiating and using (19) and (20), we get
![]()
Hence, we obtain
![]()
By integrating the above inequality from
to t we have
![]()
Taking the superior limit as
and using (18), we get
which contradicts that
. This completes the proof of Theorem 2.3. ![]()
Remark 2.2 We can rewrite the condition (17) in the Theorem 2.3 as following
![]()
Remark 2.3 If
and
, then our results extend the results in [3] .
Example 2.2 Consider the differential equation
![]()
where
and
. Choosing
and
. Thus, all conditions of Theorem 2.3
are satisfied then every solutions of this equation is either oscillatory or tends to zero.