Electronic Journal of Differential Equations, Vol. 2001(2001), No. 61, pp. 15. ISSN: 1072-6691. URL: http://ejde.math.swt.edu or http://ejde.math.unt.

edu ftp ejde.math.swt.edu (login: ftp)

Orders of solutions of an n-th order linear dierential equation with entire coecients
Benharrat Bela di & Saada Hamouda

Abstract We study the solutions of the dierential equation f (n) + An1 (z )f (n1) + + A1 (z )f + A0 (z )f = 0, where the coecients are entire functions. We nd conditions on the coecients so that every solution that is not identically zero has innite order.

Introduction
f (n) + An1 (z )f (n1) + + A1 (z )f + A0 (z )f = 0 , (1.1)

For n 2, we consider the linear dierential equation

where A0 (z ), . . . , An1 (z ) are entire functions with A0 (z ) 0. Let (f ) denote the order of the growth of an entire function f as dened in [4]: (f ) = lim sup
r

log log max|z|=r |f (z )| log r

The value T (r, f ) = log(max|z|=r |f (z )|) is known as the Nevanlinna characteristic of f [4]. It is well known that all solutions of (1.1) are entire functions and when some of the coecients of (1.1) are transcendental, (1.1) has at least one solution with order (f ) = . The question which arises is: What conditions on A0 (z ), . . . , An1 (z ) will guarantee that every solution f 0 has innite order? In this paper we prove two results concerning this question. When A0 (z ), . . . , An1 (z ) are polynomials with A0 (z ) 0, every solution of (1.1) is an entire function with nite rational order; see for example [3], [5, pp. 199-209], [6, pp. 106-108], and [7, pp. 65-67].
Mathematics Subject Classications: 30D35, 34M10, 34C10, 34C11. Key words: Linear dierential equations, entire functions, order of growth. c 2001 Southwest Texas State University. Submitted July 23, 2001. Published September 17, 2001.

Orders of solutions

EJDE2001/61

In the study of the dierential equation f + A(z )f + B (z )f = 0 (1.2)

where A(z ) and B (z ) 0 are entire functions, Gundersen proved the following results. Theorem 1.1 ([1, p. 418]) Let A(z ) and B (z ) 0 be entire functions such that for real constants , , 1 , 2 with > 0, > 0, and 1 < 2 , we have |B (z )| exp{(1 + o(1))|z | } and |A(z )| exp{o(1)|z | } (1.4) as z with 1 arg z 2 . Then every solution f 0 of (1.2) has innite order. Theorem 1.2 ([1, p. 419]) Let {k } and {k } be two nite collections of real numbers satisfying 1 < 1 < 2 < 2 < < n < n < n+1 , where n+1 = 1 + 2 , and set = max (k+1 k ).
1kn

(1.3)

(1.5)

Suppose that A(z ) and B (z ) are entire functions such that for some constant 0, |A(z )| = O(|z | ) (1.6) as z with k arg z k for k = 1, . . . , n and where B (z ) is transcendental with (B ) < . Then every solution f 0 of (1.2) has innite order.

Statement and proof of results

In this paper we prove the following two theorems: Theorem 2.1 Let A0 (z ), . . . , An1 (z ), A0 (z ) 0 be entire functions such that for real constants , , , 1 , 2 , where 0 < , > 0 and 1 < 2 we have |A0 (z )| e|z| and |Ak (z )| e |z| ,

(2.1)

k = 1, . . . , n 1

(2.2)

as z with 1 arg z 2 . Then every solution f 0 of (1.1) has innite order.

EJDE2001/61

Benharrat Bela di & Saada Hamouda

Theorem 2.2 Let {k } and {k } be two nite collections of real numbers satisfying 1 < 1 < 2 < 2 < < m < m < m+1 where m+1 = 1 + 2 , and set = max (k+1 k ). (2.3)
1km

Suppose that A0 (z ), . . . , An1 (z ) are entire functions such that for some constant 0, |Aj (z )| = O(|z | ), j = 1, . . . , n 1 (2.4) as z with k arg z k for k = 1, . . . , m and where A0 (z ) is transcendental with (A0 ) < /. Then every solution f 0 of (1.1) has innite order. Next, we provide a lemma that is used in the proofs of our theorems. Lemma 2.3 ([2, p. 89]) Let w be a transcendental entire function of nite order . Let = {(k1 , j1 ), (k2 , j2 ), . . . , (km , jm )} denote a nite set of distinct pairs of integers satisfying ki > ji 0 for i = 1, . . . , m, and let > 0 be a given constant. Then there exists a set E [0, 2 ) that has linear measure zero, such that if 0 [0, 2 ) E , then there is a constant R0 = R0 (0 ) > 0 such that for all z satisfying arg z = 0 and |z | R0 and for all (k, j ) , we have w(k) (z ) |z |(kj )(1+) . w(j ) (z )

Proof of Theorem 2.1

Suppose that f 0 is a solution of (1.1) with (f ) < . Set = (f ). Then from Lemma 1, there exists a real constant 0 where 1 0 2 , such that f (k) (z ) = o(1)|z |k , f (z ) k = 1, . . . , n (2.5)

as z with arg z = 0 . Then from (2.5) and (1.1), we obtain that |A0 (z )| o(1)|z | |A1 (z )| + + o(1)|z |(n1) |An1 (z )| + o(1)|z |n (2.6)

as z with arg z = 0 . However this contradicts (2.1) and (2.2). Therefore, every solution f 0 of (1.1) has innite order. Next we give an example that illustrates Theorem 2.1. Example 1. Consider the dierential equation f (3 + 6ez )f + (2 + 6ez + 11e2z )f 6e3z f = 0 In this equation, for z = rei , r +,
6

(2.7)

we have
2 2 r

|A0 (z )| = | 6 e3z | = 6e3r cos > e3

, < e2r

|A1 (z )| = 2 + 6ez + 11e2z | 19e2r cos 19e |A2 (z )| = | (3 + 6ez )| 9er cos 9e

3r

3 2 r

< e2r .

Orders of solutions

EJDE2001/61

As we see, conditions (2.1) and (2.2) of Theorem 2.1 are veried. The three z z z linearly independent functions f1 (z ) = ee , f2 (z ) = e2e , f3 (z ) = e3e are solutions of (2.7) with (f1 ) = (f2 ) = (f3 ) = . Next we give a generalization of Example 1. Example 2. f Consider the dierential equation
(n)

+ Pn1 (ez )f

(n1)

+ + P1 (ez )f + ez f = 0 ,

(2.8)

where R, > 0, C, | | 1, and P1 , . . . , Pn1 are polynomials. If we take the sector 1 arg z 2 , 1 , 2 ]0, 2 [ with 1 near enough to 2 such cos 2 that max1kn1 deg(Pk ) < cos 1 , then conditions (2.1) and (2.2) of Theorem 2.1 are satised as z with 1 arg z 2 . From Theorem 2.1, it follows that every solution f 0 of (2.8) has innite order.

Proof of Theorem 2.2

Suppose that f 0 is a solution of (1.1) where (f ) < and we set = (f ). From Lemma 1, there exists a set E [0, 2 ) that has linear measure zero, such that if 0 [k , k ) E for some k , then | f (l) (z ) | = O(|z |l ), f (z ) l = 1, . . . , n (2.9)

as z with arg z = 0 . From (2.9) ,(2.4) and (1.1), we obtain that |A0 (z )| | f (n) f (n1) f | + |An1 (z )|| | + + |A1 (z )|| | = O(|z | ) f f f
/

(2.10)

as z with arg z = 0 , where = + n . Let > 0 be a small constant that satises (A0 ) < +2 (this is possible since (A0 ) < ). By using the Phragm en-Lindel of theorem on (2.10), it can be deduced that for some integer s>0 |A0 (z )| = O(|z |s ) (2.11) as z with k + arg z k for k = 1, . . . , m. Now for each k , we have from (2.3) that k+1 + (k ) + 2, and so (A0 ) < k+1 en-Lindel of theorem on (2.11) k +2 . Hence using the Phragm s we can deduce that |A0 (z )| = O(|z | ) as z in the whole complex plane. This means that A0 (z ) is a polynomial which contradicts our hypothesis and completes the proof of Theorem 2.2. Next we give an example that illustrates Theorem 2.2. Example 3. If A0 (z ) is transcendental with (A0 ) < 2, then from Theorem 2.2, every solution f 0 of the equation f (n) + Pn1 (z )f (n1) + + P2 (z )f + (ez + ei z )f + A0 (z )f = 0 , where Pn1 , . . . , P2 are polynomials, is of innite order.
3 3

EJDE2001/61

Benharrat Bela di & Saada Hamouda

Acknowledgement. The authors would like to thank the referee for his/her helpful remarks and suggestions.

References
[1] G. Gundersen, Finite order solutions of second order linear dierential equations, Trans. Amer. Math. Soc. 305 (1988), pp. 415-429. [2] G. Gundersen, Estimates for the logarithmic derivative of a meromorphic functions, plus similar estimates, J. London Math. Soc. (2) 37 (1988), pp. 88-104. [3] G. Gundersen, M. Steinbart M., and S. Wang, The possible orders of solutions of linear dierential equations with polynomial coecients, Trans. Amer. Math. Soc. 350 (1998), pp. 1225-1247. [4] W. K. Hayman W. K, Meromorphic functions, Clarendon Press, Oxford, 1964. [5] G. Jank and L. Volkmann, Einf uhrung in die Theorie der ganzen und meromorphen Funktionen mit Anwendungen auf Dierentialgleichungen, Birkh auser, Basel-Boston-Stuttgart, 1985. [6] G. Valiron, Lectures on the general theory of integral functions, translated by E. F. Collingwood, Chelsea, New York, 1949. [7] H. Wittich, Neuere Untersuchungen u ber eindeutige analytishe Funktionen, 2nd Edition, Springer-Verlag, Berlin-Heidelberg-New York, 1968. Benharrat Bela di (e-mail:belaidi.benharrat@caramail.com) Saada Hamouda (e-mail: hamouda.saada@caramail.com) Department of Mathematics, University of Mostaganem B. P. 227 Mostaganem, Algeria