Continue from this paper
Indexing is not verification. Extracted statements should be checked against the source.
Abstract
In the present article, we determine some subordination
and superordination results involving Wanas operator for certain nor-
malized analytic functions defined in the unit disk U. These results are
applied to establish sandwich results. Our results extend corresponding
previously known results.
Results & Lemmas (12)
Each result is auto-extracted from the paper via OCR - the math is shown verbatim and may carry occasional transcription artifacts. Click to expand the full statement.
Lemma 1
Lemma 1 ([8]). Let q be univalent in the unit disk U and let θ and φ be analytic in a domain D containing q (U) with φ (w) ̸= 0 when w ∈q…
Lemma 1 ([8]). Let q be univalent in the unit disk U and let θ and φ be analytic in a domain D containing q (U) with φ (w) ̸= 0 when w ∈q (U). Set Q (z) = zq′ (z) φ (q (z)) and h (z) = θ (q (z)) + Q (z). Suppose that (1) Q (z) is starlike univalent in U, (2) ℜ zh′(z) Q(z) > 0 for z ∈U. If ξ is analytic in U, with ξ (0) = q (0), ξ (U) ⊂D and (5) θ (ξ (z)) + zξ′ (z) φ (ξ (z)) ≺θ (q (z)) + zq′ (z) φ (q (z)) , then ξ ≺q and q is the best dominant of (5).
Lemma 2
Lemma 2 ([9]). Let q be a convex univalent function in Uand let µ ∈C, ν ∈C 0 with ℜ 1 + zq′′ (z) q′ (z) > max n 0, −Re µ ν o.
Lemma 2 ([9]). Let q be a convex univalent function in Uand let µ ∈C, ν ∈C\{0} with ℜ 1 + zq′′ (z) q′ (z) > max n 0, −Re µ ν o .
Lemma 3
Lemma 3 ([9]). Let q be convex univalent in U and let ν ∈C. Further assume that ℜ(ν) > 0. If ξ ∈H [q (0), 1]∩Q and ξ (z)+νzξ′ (z) is…
Lemma 3 ([9]). Let q be convex univalent in U and let ν ∈C. Further assume that ℜ(ν) > 0. If ξ ∈H [q (0) , 1]∩Q and ξ (z)+νzξ′ (z) is univalent in U, then (7) q (z) + νzq′ (z) ≺ξ (z) + νzξ′ (z) , which implies that q ≺ξ and q is the best subordinant of (7).
Lemma 4
Lemma 4 ([5]). Let q be convex univalent in the unit disk U and let θ and φ be analytic in a domain D containing q (U). Suppose that (1) ℜ…
Lemma 4 ([5]). Let q be convex univalent in the unit disk U and let θ and φ be analytic in a domain D containing q (U). Suppose that (1) ℜ θ′(q(z)) φ(q(z)) > 0 for z ∈U, (2) Q (z) = zq′ (zφ (q (z))) is starlike univalent in U. If ξ ∈H [q (0) , 1]∩Q, with ξ (U) ⊂D, φ (ξ (z))+zξ′ (z) φ (ξ (z)) is univalent in U and (8) θ (q (z)) + zq′ (z) φ (q (z)) ≺θ (ξ (z)) + zξ′ (z) φ (ξ (z)) , then q ≺ξ and q is the best subordinant of (8). 2. Main Results
Theorem 1.
Theorem 1. Let q be convex univalent in U with q (0) = 1, σ ∈C 0, γ > 0 and suppose that q satisfies (9) ℜ 1 + zq′′ (z) q′ (z) > max n…
Theorem 1. Let q be convex univalent in U with q (0) = 1, σ ∈C\{0}, γ > 0 and suppose that q satisfies (9) ℜ 1 + zq′′ (z) q′ (z) > max n 0, −ℜ γ σ o .
Theorem 2.
Theorem 2. Let η, τ ∈C, γ > 0, λ ∈C 0 and q be convex univalent in U with q (0) = 1, q (z) ̸= 0 (z ∈U) and assume that q satisfies (13) ℜ …
Theorem 2. Let η, τ ∈C, γ > 0, λ ∈C\{0} and q be convex univalent in U with q (0) = 1, q (z) ̸= 0 (z ∈U) and assume that q satisfies (13) ℜ 1 + τ λq (z) + zq′′ (z) q′ (z) −zq′ (z) q (z) > 0. Suppose that zq′(z) q(z) is starlike univalent in U. If f ∈A satisfies (14) Ω(η, τ, γ, λ, k, δ, α, β; z) ≺η + τq (z) + λzq′ (z)
Theorem 3.
Theorem 3. Let q be convex univalent in U with q (0) = 1, γ > 0 and ℜ(σ) > 0. Let f ∈A satisfies
Theorem 3. Let q be convex univalent in U with q (0) = 1, γ > 0 and ℜ(σ) > 0. Let f ∈A satisfies
Theorem 4.
Theorem 4. Let η ∈C, γ > 0, λ ∈C 0 and q be convex univalent in U with q (0) = 1, q (z) ̸= 0 (z ∈U) and assume that q satisfies (22) ℜ τ λq…
Theorem 4. Let η ∈C, γ > 0, λ ∈C\{0} and q be convex univalent in U with q (0) = 1, q (z) ̸= 0 (z ∈U) and assume that q satisfies (22) ℜ τ λq (z) > 0. Suppose that zq′(z) q(z) is starlike univalent in U. If f ∈A satisfies
Theorem 5.
Theorem 5. Let q1 and q2 be convex univalent in U with q1 (0) = q2 (0) = 1. Suppose q2 satisfies (9), γ > 0 and ℜ(σ) > 0. Let f ∈A satisfies
Theorem 5. Let q1 and q2 be convex univalent in U with q1 (0) = q2 (0) = 1. Suppose q2 satisfies (9), γ > 0 and ℜ(σ) > 0. Let f ∈A satisfies
Theorem 6.
Theorem 6. Let q1 and q2 be convex univalent in U with q1 (0) = q2 (0) = 1. Suppose q1 satisfies (22) and q2 satisfies (13). Let f ∈A satisfies
Theorem 6. Let q1 and q2 be convex univalent in U with q1 (0) = q2 (0) = 1. Suppose q1 satisfies (22) and q2 satisfies (13). Let f ∈A satisfies
Theorem 3.6
Theorem 3.6, Theorem 3.9], (3) Setting α = 0 and k = β = 1 in Theorems 1, 3 and 5, we get the results obtained by Răducanu and Nechita [11,…
Theorem 3.6, Theorem 3.9], (3) Setting α = 0 and k = β = 1 in Theorems 1, 3 and 5, we get the results obtained by Răducanu and Nechita [11, Corollary 3.3,
Corollary 3.8
Corollary 3.8, Corollary 3.11]. References [1] J. W. Alexander, Functions which map the interior of the unit circle upon simple region,…
Corollary 3.8, Corollary 3.11]. References [1] J. W. Alexander, Functions which map the interior of the unit circle upon simple region, Annals of Mathematics, 17 (1) (1915), 12–22. [2] F. M. Al-Oboudi, On univalent functions defined by a generalized Sălăgean opera- tor, International Journal of Mathematics and Mathematical Sciences, 27 (2004), 1429–1436. [3] A. A. Attiya, M. F. Yassen, Some subordination and superordination results associated with generalized Srivastava-Attiya operator, Filomat,
Related Papers