Electronic Journal of Differential Equations, Vol. 2009(2009), No. 115, pp. 1–11. ISSN: 1072-6691. URL: http://ejde.math.txstate.edu or http://ejde.math.unt.edu ftp ejde.math.txstate.edu UPPER AND LOWER SOLUTIONS FOR A SECOND-ORDER THREE-POINT SINGULAR BOUNDARY-VALUE PROBLEM QIUMEI ZHANG, DAQING JIANG, SHIYOU WENG, HAIYIN GAO Abstract. We study the singular boundary-value problem u00 + q(t)g(t, u) = 0, u(0) = 0, t ∈ (0, 1), η ∈ (0, 1), γ > 0 u(1) = γu(η) . The singularity may appear at t = 0 and the function g may be superlinear at infinity and may change sign. The existence of solutions is obtained via an upper and lower solutions method. 1. Introduction Motivated by the study of multi-point boundary-value problems for linear second order ordinary differential equations, Gupta [7] studied certain three point boundary-value problems for nonlinear ordinary differential equations. Since then, more general nonlinear multi-point boundary-value problems have been studied by several authors using the Leray-Schauder theorem, nonlinear alternative of LeraySchauder or coincidence degree theory. We refer the reader to [3, 4, 5, 9, 12, 13, 14, 15] for some existence results of nonlinear multi-point boundary-value problems. Recently, Ma [14] proved the existence of positive solutions for the three point boundary-value problem u00 + b(t)g(u) = 0, u(0) = 0, t ∈ (0, 1) u(1) = αu(η), where η ∈ (0, 1), 0 < α < 1/η, b ≥ 0 and g ≥ 0 is either superlinear or sublinear. He applied a fixed point theorem in cones. In this paper, we study the singular three-point boundary-value problem u00 + q(t)g(t, u) = 0, u(0) = 0, t ∈ (0, 1), η ∈ (0, 1), γ > 0 u(1) = γu(η). (1.1) The singularity may appear at t = 0, and the function g may be superlinear at u = ∞ and may change sign. 2000 Mathematics Subject Classification. 34B15, 34B16. Key words and phrases. Singular boundary-value problem; upper and lower solutions; existence of solutions; superlinear. c 2009 Texas State University - San Marcos. Submitted January 27, 2009. Published September 12, 2009. Supported by grants 10571021 from NSFC of China, and and KLAS from Key Laboratory for Applied Statistics of MOE. 1 2 Q. ZHANG, D. JIANG, S. WENG, H. GAO EJDE-2009/115 Some basic results on the singular two point boundary-value problems were obtain in [1, 11, 17], in all these papers the arguments rely on the assumption that g(t, u) is positive. This implies that the solutions are concave. Recently, some authors have studied the case when g is allowed to change sign by applying the modified upper and lower solutions method; see for example [11]. The present work is a direct extension of some results on the singular two-point boundary-value problems. As in [11], our technique relies essentially on a modified method of upper and lower solutions method for singular three-point boundaryvalue problems which we believe is well adapted to this type of problems. 2. Upper and lower solutions Consider the three-point boundary-value problem u00 + f (t, u) = 0, t ∈ (0, 1), η ∈ (0, 1), γ ∈ (0, 1/η) u(0) = A, u(1) − γu(η) = B. (2.1) We use the following assumption: (A1) f : (0, 1] × R → R is a continuous function, there exist two functions α, β ∈ C([0, 1], R) and α(t) ≤ β(t), for all t ∈ [0, 1], if there exist a function h ∈ C( (0, 1], (0, ∞)), such that |f (t, u)| ≤ h(t) for α(t) ≤ u ≤ β(t), Z 1 lim+ t2 h(t) = 0, th(t)dt < ∞. t→0 (2.2) (2.3) 0 We call a function α(t) a lower solution for (2.1), if α ∈ C([0, 1], R)∩C 2 ((0, 1), R), and α00 + f (t, α) ≥ 0, α(0) ≤ A, for t ∈ (0, 1), α(1) − γα(η) ≤ B. Similarly, we call a function β(t) an upper solution for (2.1), if β ∈ C([0, 1], R) ∩ C 2 ((0, 1), R), and β 00 + f (t, β) ≤ 0, β(0) ≥ A, for t ∈ (0, 1), β(1) − γβ(η) ≥ B. A function u(t) is said to be a solution to (2.1), if it is both a lower and an upper solution to (2.1). Our first result reads as follows. Theorem 2.1. Assume (A1) and let α, β be, respectively, a lower solution and an upper solution for (2.1) such that α(t) ≤ β(t) on [0, 1]. Then (2.1) has at least one solution u(t) such that α(t) ≤ u(t) ≤ β(t), for t ∈ [0, 1]. Consider now the modified boundary-value problem u00 + f1 (t, u) = 0, u(0) = A, for t ∈ (0, 1), u(1) − γu(η) = B, (2.4) EJDE-2009/115 UPPER AND LOWER SOLUTIONS 3 where f (t, α(t)), if u < α(t), f1 (t, u) = f (t, u), if α(t) ≤ u ≤ β(t), f (t, β(t)), if u > β(t). Lemma 2.2. Assume that (2.3) holds. Then the boundary-value problem y 00 = −h(t), y(0) = A, 0 < t < 1, y(1) − γy(η) = B (2.5) has a unique solution y(t) in C([0, 1], [0, ∞)) ∩ C 2 ((0, 1), R), which can be written as Z 1 B − A(1 − γ) t+ G(t, s)h(s)ds, 0 ≤ t ≤ 1, y(t) = A + 1 − γη 0 where G(t, s) is Green’s function of the boundary-value problem −y 00 = 0, y(0) = 0, y(1) = γy(η). The function G is explicitly given by: when 0 ≤ s ≤ η, ( s[1−t−γ(η−t)] , s ≤ t, 1−γη G(t, s) = t[1−s−γ(η−s)] , s > t; 1−γη when η < s ≤ 1, ( s(1−t)+γη(t−s) G(t, s) = 1−γη t(1−s) 1−γη , , s ≤ t, s > t. Proof. Uniqueness. The proof of the uniqueness of a solution is standard and hence omitted. Existence. Let Z 1 B − A(1 − γ) y(t) := A + t+ G(t, s)h(s)ds, 0 ≤ t ≤ 1; 1 − γη 0 i.e., Rt A + B−A(1−γ) t + 0 s[1−t−γ(η−t)] h(s)ds 1−γη 1−γη R R + η t[1−s−γ(η−s)] h(s)ds + 1 t(1−s) h(s)ds, 0 ≤ t ≤ η, 1−γη t η 1−γη y(t) = R η s[1−t−γ(η−t)] B−A(1−γ) A + 1−γη t + 0 h(s)ds 1−γη R 1 t(1−s) R t s(1−t)+γη(t−s) + η h(s)ds + t 1−γη h(s)ds, η < t ≤ 1. 1−γη Then we have Rt B−A(1−γ) + 0 s(γ−1) 1−γη 1−γη h(s)ds R η 1−s−γ(η−s) + h(s)ds 1−γη Rt1 1−s 0 0 < t ≤ η, y (t) = + η 1−γη h(s)ds, R η B−A(1−γ) s(γ−1) + 0 1−γη h(s)ds 1−γη + R t γη−s h(s)ds + R 1 1−s h(s)ds, η < t ≤ 1. η 1−γη t 1−γη R1 Rt and y 00 (t) = −h(t) for all t ∈ (0, 1). Since 0 th(t)dt < ∞, limt→0+ 0 sh(s)ds = 0; so we have Z η 1 − s − γ(η − s) y(0) = A + lim+ t h(s)ds. 1 − γη t→0 t 4 Q. ZHANG, D. JIANG, S. WENG, H. GAO EJDE-2009/115 R1 R1 If 0 1−s−γ(η−s) h(s)ds < ∞, then y(0) = A. If 0 1−s−γ(η−s) h(s)ds = ∞, then by 1−γη 1−γη (2.3) we obtain R η 1−s−γ(η−s) h(s)ds 1 − γη + t(γ − 1) 1−γη t y(0) = A + lim+ = A + lim+ t2 h(t) = A. 1/t 1 − γη t→0 t↓0 We have also y(1) − γy(η) Z 1 sγ(1 − η) γη(1 − s) h(s)ds + h(s)ds 1 − γη 1 − γη 0 η Z η Z 1 s(1 − η) η(1 − s) B − A(1 − γ) − γ( η+ h(s)ds + h(s)ds) = B. 1 − γη 1 − γη 1 − γη 0 η B − A(1 − γ) = + 1 − γη Z η This shows that y(t) is a positive solution of (2.5), and y ∈ C([0, 1], [0, ∞)) ∩ C 2 ((0, 1), R). Let us define an operator Φ : X → X by (Φu)(t) = A + B − A(1 − γ) t+ 1 − γη Z 1 G(t, s)f1 (s, u(s))ds, (2.6) 0 where X = {u ∈ C([0, 1], R) with the norm kuk} is a Banach space, with kuk := sup{|u(t)| : 0 ≤ t ≤ 1}. Without loss of generality, we assume that A = B = 0. To prove the existence of a solution to (2.4), we need the following Lemma. Lemma 2.3. The function Φ is continuous from X to X and Φ(X) is a compact subset of X. Proof. As in the proof of Lemma 2.2, from the definition of f1 and from (2.6), we have Z 1 Z 1 |(Φu)(t)| ≤ G(t, s)|f1 (s, u(s))|ds ≤ G(t, s)h(s)ds = y(t), t ∈ [0, 1]. (2.7) 0 0 So we have Φu ∈ C([0, 1], R) ∩ C 2 ((0, 1), R), and kΦuk ≤ kyk. (2.8) This shows that Φ(X) is a bounded subset of X. Noting the facts that y(0) = 0 and the continuity of y(t) on [0, 1], we have from (2.7) that for any > 0, one can find a δ1 > 0 (independent with u) such that 0 < δ1 < 1/8 and (2.9) (Φu)(t) < , t ∈ [0, 2δ1 ]. 2 On the other hand, from (2.6), since |f1 (s, u(s))| ≤ h(s), s ∈ (0, 1), we can obtain |(Φu)0 (t)| ≤ L, t ∈ [δ1 , 1]. 2L , then for t1 , t2 ∈ [δ1 , 1], |t2 − t1 | < δ2 , we have |(Φu)(t1 ) − (Φu)(t2 )| ≤ L|t1 − t2 | < . 2 Define δ = min{δ1 , δ2 }, then using (2.9), (2.10), we obtain Let δ2 = |(Φu)(t1 ) − (Φu)(t2 )| < , (2.10) (2.11) EJDE-2009/115 UPPER AND LOWER SOLUTIONS 5 for t1 , t2 ∈ [0, 1], |t1 − t2 | < δ. This shows that {(Φu)(t) : u ∈ X} is equicontinuous on [0, 1]. We can obtain the continuity of Φ in a similar way as above. In fact, if un , u ∈ X and kun − uk → 0 as n → ∞, then we have Z 1 G(t, s)h(s)ds = 2y(t), t ∈ [0, 1], (2.12) |(Φun )(t) − (Φu)(t)| ≤ 2 0 Noting the facts that y(0) = 0 and the continuity of y(t) on [0, 1], then for any > 0, one can find a δ1 > 0 (independent of un ) such that 0 < δ1 < 1/8 and |(Φun )(t) − (Φu)(t)| < , t ∈ [0, δ1 ]. (2.13) On the other hand, from the continuity of f1 , one has |(Φun )(t) − (Φu)(t)| → 0, t ∈ [δ1 , 1], (2.14) as n → ∞. This together with (2.13) implies that kΦun − Φuk → 0 as n → ∞. Therefore, Φ : X → X is completely continuous. The proof is complete. Lemma 2.4. Let u(t) be a solution to (2.4). Then α(t) ≤ u(t) ≤ β(t) for all t ∈ [0, 1]; i.e., u(t) is a solution to (2.1). Proof. We first prove that u(t) ≤ β(t) on [0, 1]. Let x(t) := u(t) − β(t). Assume that u(t) > β(t) for some t ∈ [0, 1]. Since u(0) = 0 ≤ β(0), it follows that x(0) ≤ 0, x(1) = u(1) − β(1) ≤ γu(η) − γβ(η) = γx(η). Let σ ∈ (0, 1] be such that x(σ) = maxt∈[0,1] x(t). Then x(σ) > 0. Case(i): σ ∈ (0, 1). So there exists an interval (a, σ] ⊂ (0, 1) such that x(t) > 0 in (a, σ], and x(a) = 0, x(σ) = max x(t) > 0, x0 (σ) = 0. t∈[0,1] For t ∈ (a, σ] we have that f1 (t, u(t)) = f (t, β(t)) and therefore u00 (t) + f1 (t, u(t)) = u00 (t) + f (t, β(t)) = 0 for all t ∈ (a, σ]. On the other hand, as β is an upper solution for (2.1), we have β 00 (t) + f (t, β(t)) ≤ 0 00 for all t ∈ (a, σ]. 00 Thus, we obtain u (t) ≥ β (t) for all t ∈ (a, σ], and hence, x00 (t) ≥ 0. Then x0 (t) ≤ 0 on (a, 1] which is a contradiction. Case(ii): σ = 1. So there exists (a, 1] ⊂ (0, 1] such that x(a) = 0, x(1) = max x(t), x(1) − γx(η) ≤ 0. t∈[0,1] In the same way as in Case(i), we can obtain that x(t) > 0, x00 (t) ≥ 0, t ∈ (a, 1]. Since x(η) ≥ γ1 x(1) > 0, then η > a. Consider the three-point boundary-value problem x00 = h(t) > 0, x(a) = 0, a < t < 1, x(1) − γx(η) = b1 ≤ 0. (2.15) Then this equation has a unique solution x(t) ∈ C([a, σ], [0, ∞)) ∩ C 2 ((a, 1), R), which can be represented as Z 1 b1 (t − a) x(t) = − G[a,1] (t, s)h(s)ds, a ≤ t ≤ 1, 1 − a − γ(η − a) a 6 Q. ZHANG, D. JIANG, S. WENG, H. GAO EJDE-2009/115 where G[a,1] (t, s) is the Green’s function of the boundary-value problem −y 00 = 0, y(a) = 0, y(1) = γy(η), which is explicitly given by: when a ≤ s ≤ η, (s−a)[1−t−γ(η−t)] , s ≤ t, 1−a−γ(η−a) G[a,1] (t, s) = (t−a)[1−s−γ(η−s)] , s > t; 1−a−γ(η−a) when η < s ≤ 1, G[a,1] (t, s) = (s−a)(1−t)+γ(t−s)(η−a) , 1−a−γ(η−a) (t−a)(1−s) ; s > t. 1−a−γ(η−a) s ≤ t, 1−a Since 0 < γ < η1 < η−a , then G[a,1] (t, s) ≥ 0, and hence x(t) ≤ 0 on [a, 1], which is a contradiction. In very much the same way, we can prove that u(t) ≥ α(t) on [0, 1]. 3. Main results Let g : [0, 1] × (0, ∞) → R be a continuous function and q ∈ C((0, 1], R+ 0 ). Consider the three-point boundary-value problem u00 + q(t)g(t, u) = 0, t ∈ (0, 1), η ∈ (0, 1), γ ∈ (0, 1] u(0) = 0, u(1) = γu(η). (3.1) Theorem 3.1. Assume that (H1) |g(t, x)| ≤ F (x) + Q(x) on [0, 1] × (0, ∞) with F > 0 continuous and nonincreasing on (0, ∞), Q ≥ 0 continuous on [0, ∞), and Q F nondecreasing on (0, ∞); (H2) there exist constants L > 0 and ε > 0 such that g(t, x) > L for all (t, x) ∈ [0, 1] × (0, ε], and F (x) > L, x ∈ (0, ε]; (H3) Z 1 tq(t)dt < ∞, (3.2) lim t2 q(t) = 0, t→0+ 0 Z c 1 du sup > b0 , (3.3) Q(c) c∈(0,∞) 1 + 0 F (u) F (c) where b0 = R1 0 rq(r)dr. Then (3.1) has at least one solution u ∈ C([0, 1], [0, ∞))∩C 2 ((0, 1), R) with u(t) > 0 on (0, 1]. From Lemma 2.2, we obtain the following result. Lemma 3.2. There exists an unique solution W ∈ C([0, 1], [0, ∞)) ∩ C 2 ((0, 1), R), with W (t) > 0 on (0, 1] to the problem W 00 + q(t) = 0, W (0) = 0, 0 < t < 1, W (1) = γW (η). Choose M > 0, δ > 0 (δ < M ) such that Z M du 1 > b0 . Q(M ) F (u) 1+ δ F (M ) (3.4) (3.5) EJDE-2009/115 UPPER AND LOWER SOLUTIONS 7 Let n0 ∈ {1, 2, . . . } be chosen so that 1/n0 < min{ε − mkW k, δ}, where W is the solution of (3.4), and 0 < m < min{L, ε/kW k, 1} is chosen and fixed. Let N + = {n0 , n0 + 1, . . . }. We first show that the boundary-value problem u00 + q(t)g(t, u) = 0, 0 < t < 1, (3.6) 1 1−γ u(0) = , u(1) − γu(η) = , n ∈ N+ n n has a solution un for each n ∈ N + with un (t) ≥ n1 for t ∈ [0, 1] and kun k < M . We have the following Claim Claim: Let αn (t) = mW (t) + n1 , t ∈ [0, 1], then αn (t) is a (strict) lower solution for problem (3.6). Proof. For the choice of m and n, we have mW (t) + n1 ≤ mkW k + n10 < ε, then from (H2), 1 g(t, mW (t) + ) > L > m for all t ∈ [0, 1]. n Then we obtain 1 1 αn00 (t) + q(t)g(t, αn (t)) = (mW (t) + )00 + q(t)g(t, mW (t) + ) n n 1 = mW 00 (t) + q(t)g(t, mW (t) + ) n 1 = q(t)(g(t, mW (t) + ) − m) > 0, 0 < t < 1. n We obtain αn (0) = mW (0) + n1 = n1 , and 1 1 − γ(mW (η) + ) n n 1−γ 1−γ = m(W (1) − γW (η)) + = . n n Thus the proof of Claim is complete. αn (1) − γαn (η) = mW (1) + To find the upper solution of (3.6), we consider the problem u00 + q(t)F (u)(1 + Q(M ) ) = 0, F (M ) 0 < t < 1, (3.7) 1 1−γ , u(1) − γu(η) = . n n To show that this problem has a solution we study u(0) = u00 + q(t)F ∗ (u)(1 + Q(M ) ) = 0, F (M ) 0 < t < 1, (3.8) 1 1−γ , u(1) − γu(η) = , n n ( F (u), u ≥ 1/n, ∗ F (u) = F ( n1 ), u < 1/n. u(0) = where Then F ∗ (u) ≤ F (u) for u > 0. In the same way as in the Claim, we can easily prove αn (t) = a (strict) lower solution of (3.8). 1 n + mW (t) is also 8 Q. ZHANG, D. JIANG, S. WENG, H. GAO EJDE-2009/115 By Lemma 2.2, let βn0 ∈ C([0, 1], R) ∩ C 2 ((0, 1), R) be the unique solution of the boundary-value problem Q(M ) ) = 0, 0 < t < 1, F (M ) 1−γ u(1) − γu(η) = . n u00 + q(t)F (αn (t))(1 + 1 u(0) = , n (3.9) Since βn0 is a solution of this equation, 00 βn0 + q(t)F (αn (t))(1 + βn0 (0) = 1 , n Q(M ) ) = 0, F (M ) βn0 (1) − γβn0 (η) = 0 < t < 1, 1−γ . n On the other hand, as αn is a lower solution of (3.8), and αn ≥ 1/n, we have αn 00 + q(t)F (αn (t))(1 + αn (0) = 1 , n Q(M ) ) ≥ 0, F (M ) αn (1) − γαn (η) = 0 < t < 1, 1−γ . n So we obtain αn (t) ≤ βn0 (t) for t ∈ [0, 1]. Thus Q(M ) ) F (M ) Q(M ) Q(M ) ) + q(t)F (βn0 )(1 + ) = −q(t)F (αn )(1 + F (M ) F (M ) Q(M ) = q(t)(1 + )(F (βn0 ) − F (αn )) ≤ 0, F (M ) 00 βn0 + q(t)F ∗ (βn0 )(1 + so that βn0 is an upper solution for problem (3.8). If we now take αn0 ≡ αn , we have that αn0 and βn0 are, respectively, a lower and an upper solution of (3.8) with αn0 (t) ≤ βn0 (t), for all t ∈ [0, 1]. So by the Lemma 2.4, we know that there exists a solution βn ∈ C([0, 1], R) ∩ C 2 ((0, 1), R) of (3.8) such that αn (t) = αn0 (t) ≤ βn (t) ≤ βn0 (t), ∀t ∈ [0, 1]. Now we claim that kβn k < M . Suppose this is false; i.e., suppose kβn k ≥ M . Since βn (1)− n1 = γ(βn (η)− n1 ) ≤ βn (η)− n1 , βn00 (t) ≤ 0 on (0,1) and βn ≥ n1 on [0, 1], there exists σ ∈ (0, 1) with βn0 (t) ≥ 0 on (0, σ), βn0 (t) ≤ 0 on (σ, 1) and βn (σ) = kβn k. Then for z ∈ (0, 1), we have −βn00 (z) ≤ F (βn (z))(1 + Q(M ) )q(z). F (M ) Integrate from t(0 < t ≤ σ) to σ to obtain Z σ Q(M ) ) F (βn (z))q(z)dz; βn0 (t) ≤ (1 + F (M ) t so we have βn0 (t) Q(M ) ≤ (1 + ) F (βn (t)) F (M ) Z σ q(z)dz . t (3.10) EJDE-2009/115 UPPER AND LOWER SOLUTIONS 9 Then integrate from 0 to σ to obtain Z βn (σ) Z Z Z Q(M ) σ dy Q(M ) σ σ q(z)dz dt = 1 + tq(t)dt. ≤ 1+ 1 F (y) F (M ) 0 F (M ) 0 t n Consequently M Z dy Q(M ) 1 tq(t)dt. (3.11) ≤ 1+ F (y) F (M ) 0 δ This contradicts (3.5) and consequently kβn k < M . It follows from the fact βn ≥ 1/n, we can obtain βn is a solution of (3.7) also. Since F is nonincreasing on (0, ∞), similar to the proof of Lemma 2.4, we can obtain the uniqueness of solutions to (3.7). Next we show that βn is an upper solution of (3.6). Observe that Z |g(t, x)| ≤ F (x) + Q(x) on [0, 1] × (0, ∞). We have Q(M ) + q(t)|g(t, βn (t))| βn00 (t) + q(t)g(t, βn (t)) ≤ −q(t)F (βn (t)) 1 + F (M ) Q(β (t)) Q(M ) n ≤ q(t)F (βn (t)) − ≤ 0, t ∈ (0, 1). F (βn (t)) F (M ) Thus βn is an upper solution for problem (3.6). This together with the Claim yields that αn and βn are, respectively, a lower and an upper solution for (3.6) with αn ≤ βn for all t ∈ [0, 1]. So we conclude (3.6) has a solution un ∈ C([0, 1], R) ∩ C 2 ((0, 1), R) such that 1 mW (t) + = αn (t) ≤ un (t) ≤ βn (t) ≤ M, ∀t ∈ [0, 1]. n Consider now the pointwise limit z(t) := lim un (t), ∀t ∈ [0, 1]. (3.12) Let e = [a, 1] ⊂ (0, 1], Let tn ∈ (a, 1) such that u0n (tn ) = un (1) − un (a) /(1 − a). We obtain Z tn un (1) − un (a) 0 un (t) = + q(s)g(s, un (s)ds, t ∈ e. 1−a t n→+∞ Since mW (t) ≤ un (t) ≤ M , then we have Z 2M Q(M ) 1 + 1+ q(t)F (mW (t))dt := C(a, 1), |u0n (t)| ≤ 1−a F (M ) a t ∈ e. (3.13) Set vn = maxt∈e |u0n (t)|, which implies vn is bounded. That means u0n (t) is bounded on e. Then, by the Ascoli-Arzela theorem, it is standard to conclude that z(t) is a solution of (3.1) on the interval e = [a, 1]. Since e is arbitrary, we find that z ∈ C((0, 1], [0, ∞)) ∩ C 2 ((0, 1), R), and z 00 (t) + q(t)g(t, z(t)) = 0, t ∈ (0, 1). Also, we have 1 1−γ = 0, z(1) − γz(η) = lim = 0. n→+∞ n n The same argument as in [11] works, we can prove the continuity of z(t) at t = 0 and t = 1. The proof is complete. z(0) = lim n→+∞ 10 Q. ZHANG, D. JIANG, S. WENG, H. GAO EJDE-2009/115 By essentially the same argument as in Theorem 3.1 and [2, Theorem 4.2], we have the following result. Theorem 3.3. Assume that (H1*) for any r > 0 there is hr ∈ C((0, 1], (0, ∞)): |q(t)g(t, x)| ≤ hr (t) for all (t, x) ∈ (0, 1] × [r, ∞), such that Z 1 thr (t)dt < +∞; lim+ t2 hr (t) = 0, t→0 0 (H2*) there exist constants L > 0 and ε > 0 such that g(t, x) > L for all (t, x) ∈ [0, 1] × (0, ε]. Then (3.1) has at least one solution u ∈ C([0, 1], [0, ∞) ∩ C 2 ((0, 1), R). Moreover, if g(t, x) is non-increasing in x > 0, then the solution is unique. 4. An example Consider the singular boundary-value problem u00 + σt−m (u−α + uβ − T sin(8πt)) = 0, u(0) = 0, zu(1) = γu(η), t ∈ (0, 1) η ∈ (0, 1), γ ∈ (0, 1] (4.1) with 0 ≤ m < 2, σ > 0, α > 0, β ≥ 0. Set F (u) = u−α , Q(u) = uβ + 1, q(t) = σt−m , Z 1 σ b0 = rq(r)dr = . 2−m 0 Applying Theorem 3.1, we find that (4.1) has a positive solutions if xα+1 . α α+β ) x∈(0,∞) (α + 1)(1 + x + x σ < (2 − m) sup (4.2) Obviously, (H1)-(H3) in Theorem 3.1 are satisfied. Thus, (4.1) has a solution u ∈ C([0, 1], [0, ∞) ∩ C 2 ((0, 1), R) with u > 0 on (0, 1]. We remark that if 0 ≤ β < 1, then (4.1) has at least one positive solution for all σ > 0, since the right-hand side of (4.2) is infinity. References [1] R. P. Agarwal and D. O’Regan: Nonlinear superlinear singular and nonsingular second order boundary-value problems. J. Differential Equations, 143 (1998), 60-95. [2] R. P. Agarwal, H. Gao, D. Q. Jiang, D. O’Regan, and X. Zhang: Existence principles and a upper and lower solution theory for the one-dimension p-Laplacian singular boundary-value problem with sign changing nonlinearities. Dynamic Systems and Applications, 15 (2006), 3-20. [3] W. Feng and J. R. L. Webb: Solvability of a three-point Nonlinear boundary-value problems at resonance. Nonlinear Analysis TMA, 30 (1997) No. 6, 3227-3238. [4] W. Feng and J. R. L. Webb: Solvability of a m-point boundary-value problems with nonlinear growth. J. Math. Anal. Appl., 212 (1997), 467-480. [5] W. Feng: On a m-point Nonlinear boundary-value problem. Nonlinear Analysis TMA, 30 (1997), No. 6, 5369-5374. [6] Y. P. Guo, W. R. Shan, W. G. Ge: Positive solutions for a second order m-point boundaryvalue problems. J. comput. Appl. Math. , 151 (2003), 415-424. [7] C. P. Gupta: Solvability of a three-point nonlinear boundary-value problem for a second order ordinary differential equations. J. Math. Anal. Appl., 168 (1992), 540-551. EJDE-2009/115 UPPER AND LOWER SOLUTIONS 11 [8] C. P. Gupta, S. K. Ntouyas, P. C. Tsamatos: Solvability of a m-point boundary-value problem for second order ordinary differential equations. J. Math. Anal. Appl., 189 (1995), 575-584. [9] C. P. Gupta: A sharper condition for the solvability of a three-point second-order boundaryvalue problem. J. Math. Anal. Appl., 205 (1997), 586-597. [10] X. M. He, W. G. Ge: Triple solutions for second order three-point boundary-value problems. J. Math. Anal. Appl., 268 (2002), 256-265. [11] D. Q. Jiang: Upper and lower solutions for a superlinear singular boundary-value problem. Computers and Mathematics with Applications, 41 (2001), 563-569. [12] R. Ma: Existence theorems for a second order three-point boundary-value problem. J. Math. Anal. Appl., 212 (1997), 430-442. [13] R. Ma: Existence theorems for a second order m-point boundary-value problem. J. Math. Anal. Appl., 211 (1997), 545-555. [14] R. Ma: Positive solutions of a nonlinear three-point boundary-value problem. Electronic Journal of Differential Equations, 1999 (1999), No. 34, 1-8 [15] R. Ma and N. Cataneda: Existence of solution for nonlinear m-point boundary-value problem. J. Math. Anal. Appl., 256 (2001), 556-567. [16] R. Ma and Wang Haiyan: Positive solutions of nonlinear three-point boundary-value problems. J. Math. Anal. Appl. , 279 (2003), 1216-1227. [17] D. O’Regan: Positive solutions to singular and nonsingular second-order boundary-value problems. J. Math. Anal. Appl., 142 (1989), 40-52. [18] Xu Xian: Positive solutions of Singular m-point Boundary Value Problems with Positive Parameter. J. Math. Anal. Appl. , 291 (2004), 352-367. [19] Z. X. Zhang and J. Y. Wang: The upper and lower solution method for a class of singular nonlinear second-order three-point boundary-value problems. J. Comput. Appl., 147 (2002), 41-52. [20] Z. L. Wei, Z. T. Zhang: A necessary and sufficient condition for the existence of positive solutions of singular superlinear boundary value problems. Acta Mathematica Sinica Chinese Series, 48 (2005), No. 1, 25-34. Qiumei Zhang School of Science, Changchun University, Changchun 130022, China. Department of Mathematics, Northeast Normal University, Changchun 130024, China E-mail address: zhangqm1110@yahoo.com.cn Daqing Jiang Department of Mathematics, Northeast Normal University, Changchun 130024, China E-mail address: jiangdq067@nenu.edu.cn School of Science, Changchun University, Changchun 130022, China E-mail address, Haiyin Gao: gaohaiyinhealthy@yahoo.com.cn E-mail address, Shiyou Weng: wengshiyou2001@yahoo.com.cn