Sandwich type results for meromorphic functions with respect to symmetrical points

Author(s): Kuldeep Kaur Shergill1, Sukhwinder Singh Billing1
1Department of Mathematics, Sri Guru Granth Sahib World University Fatehgarh Sahib-140407(Punjab), India.
Copyright © Kuldeep Kaur Shergill, Sukhwinder Singh Billing. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Abstract

In the present paper, we use the technique of differential subordination and superordination involving meromorphic functions with respect to symmetric points and also derive some sandwich results. As a consequence of main result, we obtain results for meromorphic starlike functions with respect to symmetrical points.

Keywords: Meromorphic function; Meromorphic starlike function; Differential subordination; Superordination.

1. Introduction

Let \(\Sigma\) denote the class of functions of the form \[f(z)=\frac{1}{z}+\sum_{k=1}^\infty a_kz^{k-1},\] which are analytic in the punctured unit disc \( \mathbb E_0=\mathbb E\setminus\{0\},\) where \(\mathbb E = \{z\in\mathbb C:|z|< 1\}. \) A function \(f \in \Sigma \) is said to be meromorphic starlike of order \(\alpha \) if \(f(z) \neq 0 \) for \(z \in \mathbb E_0 \) and

\[-\Re\left(\frac{zf'(z)}{f(z)}\right)>\alpha,\hspace{1.7cm}(\alpha< 1;z \in \mathbb E).\] The class of such functions is denoted by \(\mathcal {MS}^*(\alpha)\) and write \(\mathcal {MS}^*=\mathcal {MS}^*(0)\)-the class of meromorphic starlike functions.

In 1959, Sakaguchi [1] introduced and studied the class of starlike functions with respect to symmetric points in \(\mathbb E\). Further investigations into the class of starlike functions with respect to symmetric points can be found in [2,3].

Recently, Ghaffar et al., [4] introduced and investigated a class of meromorphic starlike functions with respect to symmetric points which satisfies the condition

\[-\frac{2zf'(z)}{f(z)-f(-z)}\prec \frac{1+Az}{1+Bz}, ~z\in\mathbb E_0\] and \(-1\leq B< A\leq1.\) We denote the above class by \(\mathcal {MS}^{s*}[A,B].\) If \(f\) is analytic and \(g\) is analytic univalent in open unit disk \(\mathbb E\), we say that \(f(z)\) is subordinate to \(g(z)\) in \(\mathbb E\) and written as \(f(z)\prec g(z)\) if \(f(0) = g(0)\) and \(f( \mathbb E) \subset g(\mathbb E)\). To derive certain sandwich-type results, we use the dual concept of differential subordination and superordination.

Let \(~\Phi:\mathbb C^2 \times \mathbb E\longrightarrow \mathbb C\) (\(\mathbb C\) is the complex plane) and \(h\) be univalent in \(\mathbb E\). If \(p\) is analytic in \(\mathbb E\) and satisfies the differential subordination

\begin{equation} \Phi(p(z), zp'(z); z) \prec h(z), \Phi(p(0), 0; 0) = h(0),\label{eq1} \end{equation}
(1)
then \(p\) is called a solution of the differential subordination (1). The univalent function \(q\) is called a dominant of differential subordination (1) if \(p\prec q\) for all \(p\) satisfying \(1.\) A dominant \(\tilde q \prec q \) for all dominants \(q\) of \(1,\) is said to be the best dominant of \(1.\)

Let \(\Psi:\mathbb C^2 \times \mathbb E\longrightarrow \mathbb C\) \((\mathbb C\) is the complex plane) be analytic and univalent in domain \(\mathbb C^2 \times \mathbb E, h\) be analytic in \(\mathbb E,\) \(p\) is analytic and univalent in \(\mathbb E,\) with \((p(z), zp'(z); z)\in \mathbb C^2 \times \mathbb E \) for all \(z\in \mathbb E.\) Then \(p\) is called a solution of first order differential superordination if it satisfies

\begin{equation} h(z) \prec \Psi(p(z), zp'(z); z) ,h(0)=\Psi(p(0), 0; 0).\label{eq2} \end{equation}
(2)
An analytic function \(q\) is called a subordinant of differential superordination (2) if \(q\prec p\) for all \(p\) satisfying \((2).\) A univalent subordinant \(\tilde q\) that satisfies \(q \prec \tilde q \) for all subordinants \(q\) for (2), is said to be the best subordinant of (2).

In this paper we study the concepts of subordination and superordination to obtain meromorphic starlikeness with respect to symmetric points.On the basis of the theory we also investigate some important sandwich results of symmtric meromorphic functions.

2. Preliminaries

We shall use the following lemmas to prove our result.

Lemma 1. [5] Let \(q\) be univalent in \(\mathbb E\) and let \(\theta\) and \(\phi\) be analytic in a domain \(\mathbb D\) containing \(q(\mathbb E)\), with \(\phi(w)\neq 0\), when \(w\in q (\mathbb E)\). Set \(Q_1(z)=zq'(z)\phi[q(z)],~h(z)=\theta[q(z)]+Q_1(z)\) and suppose that either

  • (i) \(h\) is convex, or
  • (ii) \(Q_1\) is starlike.
In addition, assume that
  • (iii) \(\displaystyle \Re\left(\frac{zh'(z)}{Q_1(z)}\right)>0\) for all \(z\) in \(\mathbb E\).
If \(p\) is analytic in \(\mathbb E\), with \(p(0)=q(0),~p(\mathbb E)\subset \mathbb D\) and \[\theta[p(z)]+zp'(z)\phi[p(z)]\prec\theta[q(z)]+zq'(z)\phi[q(z)],~z\in\mathbb E,\] then \(p(z)\prec q(z)\) and \(q\) is the best dominant.

Definition 1. We denote by \(Q\) the set of functions \(p\) that are analytic and injective on \(\overline{\mathbb E}\setminus\mathbb B(p),\) where \[\mathbb B(p)=\left\{\zeta\in \partial{\mathbb E}: \lim_{z\rightarrow\zeta} {p(z)=\infty} \right\},\] and are such that \(p'(\zeta)\neq 0\) for \(\zeta \in \partial \mathbb E\setminus\mathbb B(p).\)

Lemma 2. [6] Let \(q\) be the univalent in \(\mathbb E\) and let \(\theta\) and \(\phi\) be analytic in a domain \(\mathbb D\) containing \(q(\mathbb E).\) Set \(Q_1(z)=zq'(z)\phi[q(z)],~h(z)=\theta[q(z)]+Q_1(z)\) and suppose that (i) \( Q_1(z)\) is starlike in \(\mathbb E;\) and (ii) \( \Re\left(\frac{\theta'(q(z))}{\phi(q(z))}\right)> 0,\) for \(z \in \mathbb E.\) If \(p\in \mathcal H[q(0),1]\cap Q,\) with \(p(\mathbb E)\subset \mathbb D\) and \(\theta[p(z)]+zp'(z)\phi[p(z)]\) is univalent in \(\mathbb E\) and \[\theta[q(z)]+zq'(z)\phi[q(z)]\prec \theta[p(z)]+zp'(z)\phi[p(z)],z\in \mathbb E,\] then \(q(z)\prec p(z)\) and \(q\) is the best subordinant.

Lemma 3. [7] The function \(\displaystyle q(z)=\frac{1}{(1-z)^{2ab}}\) is univalent in \(E\) if and only if \(|2ab-1|\leq1\) or \(|2ab+1|\leq1.\)

3. Subordination results

Theorem 1. Let \(q\) be univalent in \(\mathbb E,\) with \(q(0)=1,\) and let

\begin{equation} \Re \left(1+\frac{zq”(z)}{q'(z)}\right)>max\left\{0;~-\Re \frac{1}{\gamma}\right\}, ~z\in\mathbb E,\label{e1}\end{equation}
(3)
where \(\gamma\in\mathbb C^*:=\mathbb C\setminus\{0\}.\) If \(f\in\Sigma\) satisfy the condition
\begin{equation} f(z)\neq f(-z),~z\in\mathbb E_0.\label{e2}\end{equation}
(4)
and
\begin{equation} \frac{-2(1+\lambda)zf'(z)}{f(z)-f(-z)}+\frac{2\lambda z^2 f'(z)(f'(z)+f'(-z))}{(f(z)-f(-z))^2}-\frac{2\lambda z^2 f”(z)}{f(z)-f(-z)}\prec q(z)+\lambda zq'(z), \label{e3}\end{equation}
(5)
then \[\frac{-2zf'(z)}{f(z)-f(-z)}\prec q(z),\] and \(q\) is the best dominant of (5).

Proof. Setting \[p(z)=\frac{-2zf'(z)}{f(z)-f(-z)},~z\in\mathbb E,\] from assumption (4) it follows that \(p\) is analytic in \(\mathbb E,\) with \(p(0)=1.\) A simple computation shows that \[\frac{-2(1+\lambda)zf'(z)}{f(z)-f(-z)}+\frac{2\lambda z^2 f'(z)(f'(z)+f'(-z))}{(f(z)-f(-z))^2}-\frac{2\lambda z^2 f”(z)}{f(z)-f(-z)}=p(z)+\lambda zp'(z),\] hence, the subordination (5) is equivalent to \(p(z)+\lambda zp'(z)\prec q(z)+\lambda zq'(z)\).

Now, in order to prove our result we will use Lemma 1. Consider the functions \(\theta(w)=w\) and \(\phi(w)=\gamma\) analytic in \(\mathbb C,\) and set

\begin{equation} Q(z)=zq'(z) \phi(q(z))=\gamma zq'(z) \&nbsp \&nbsp and \&nbsp \&nbsp h(z)=\theta(q(z))+Q(z)=q(z)+\gamma z q'(z). \end{equation} Since \(Q(0)=0\) and \(Q'(0)=\gamma q'(0)\neq0\), the assumption (3) implies that \(Q\) is starlike in \(\mathbb E\) and \[\Re\left(\frac{zh'(z)}{Q(z)}\right)=\Re \left(1+\frac{zq”(z)}{q'(z)}+\frac{1}{\gamma}\right)>0, ~z \in \mathbb E.\] Therefore, Lemma 1 and assumption (5) imply \(p(z)\prec q(z)\) and the function \(q\) is the best dominant of (5).

If we take \(\displaystyle q(z)=\frac{1+Az}{1+Bz},-1\leq B< A\leq 1\) in Theorem 1 then the condition (3) reduces to

\begin{equation} \Re\frac{1-Bz}{1+Bz}>max\left\{0;~-\Re \frac{1}{\lambda}\right\}, ~z\in\mathbb E.\label{e4} \end{equation}
(6)
It is easy to see that the function \(\displaystyle \chi(\zeta)=\frac{1-\zeta}{1+\zeta}\) with \(|\zeta|< |B|\) is convex in \(E\) and since \(\chi(\overline{\zeta}) =\overline{\chi(\zeta) }\) for all \(|\zeta|< |B|\), it follows that \(\chi(\mathbb E)\) is a convex domain symmetric with respect to the real axis. Hence
\begin{equation} \inf\left(\Re\frac{1-Bz}{1+Bz}:~z\in\mathbb E\right)>\frac{1-|B|}{1+|B|}>0\,.\label{e5} \end{equation}
(7)
Thus, the inequality (6) is equivalent to \(\displaystyle\Re\frac {1}{\lambda}\geq \frac{|B|-1}{|B|+1}\), hence we deduce the following corollary:

Corollary 1. Let \(\lambda\in\mathbb C^*\) and \(-1\leq B< A\leq 1\) with \(\displaystyle\frac{1-|B|}{1+|B|}\geq max\left\{0;~-\Re \frac{1}{\lambda}\right\}\). If \(f\in\Sigma\) satisfy the condition (4) and

\begin{equation} \frac{-2(1+\lambda)zf'(z)}{f(z)-f(-z)}+\frac{2\lambda z^2 f'(z)(f'(z)+f'(-z))}{(f(z)-f(-z))^2}-\frac{2\lambda z^2 f”(z)}{f(z)-f(-z)}\prec \frac{1+Az}{1+Bz}+\lambda\frac{(A-B)z}{(1+Bz)^2}, \label{e6} \end{equation}
(8)
then \[\frac{-2zf'(z)}{f(z)-f(-z)}\prec \frac{1+Az}{1+Bz},\] i.e., \(f\in\mathcal{MS}^{s*}[A,B].\) Moreover, the function \(\displaystyle\frac{1+Az}{1+Bz}\) is the best dominant of (8).

For \(A=1\) and \(B=-1\), the above corollary reduces to the next special case:

Remark 1. Let \(\lambda\in\mathbb C^*\) with \(\Re \frac{1}{\lambda}\geq0\). If \(f\in\Sigma\) satisfy the condition (4) and

\begin{equation} \frac{-2(1+\lambda)zf'(z)}{f(z)-f(-z)}+\frac{2\lambda z^2 f'(z)(f'(z)+f'(-z))}{(f(z)-f(-z))^2}-\frac{2\lambda z^2 f”(z)}{f(z)-f(-z)}\prec \frac{1+z}{1-z}+\lambda\frac{2z}{(1-z)^2}, \label{e7}\end{equation}
(9)
then \[\frac{-2zf'(z)}{f(z)-f(-z)}\prec \frac{1+z}{1-z},\] i.e., \(f\in\mathcal{MS}^{s*},\) or \(f\) is meromorphic starlike with respect to symmetrical points in \(\mathbb E\). Moreover, the function \(\displaystyle\frac{1+z}{1-z}\) is the best dominant of (9).

Theorem 2. Suppose that \(q\) be univalent in \(\mathbb E\) with \(q(0)=1\) and \(q(z)\neq0\) for all \(z\in\mathbb E\) such that

\begin{equation} \Re \left(1+\frac{zq”(z)}{q'(z)}-\frac{zq'(z)}{q(z)}\right)>0, ~z\in\mathbb E,\label{e8} \end{equation}
(10)
where \(\gamma,\mu\in\mathbb C^*\) and \(\nu,\eta\in\mathbb C\) with \(\nu+\eta\neq 0\) and let \(f\in\Sigma\) satisfy the conditions
\begin{equation} \nu z(f'(z)+ f'(-z)) +\eta (f(z)- f(-z))\neq0,~z\in\mathbb E_0\label{e9} \end{equation}
(11)
and
\begin{equation} \frac{-2(\nu+\eta) zf'(z)}{\nu z(f'(z)+ f'(-z)) +\eta (f(z)- f(-z))},~z\in\mathbb E_0\,.\label{e10} \end{equation}
(12)
If
\begin{equation}1+\gamma\mu\left[1-\frac{\nu z[z(f'(z)+ f'(-z))]’ +\eta z(f'(z)+ f'(-z))}{\nu z(f'(z)+ f'(-z)) +\eta (f(z)- f(-z))}+\frac{zf”(z)}{f'(z)}\right]\prec 1+\gamma\frac{zq'(z)}{q(z)},\label{e11} \end{equation}
(13)
then \[-\left[\frac{2(\nu+\eta) zf'(z)}{\nu z(f'(z)+ f'(-z)) +\eta (f(z)- f(-z))}\right]^\mu\prec q(z)\] and \(q\) is the best dominant of (13) (the power is the principal one).

Proof. Define a function

\begin{equation} p(z)= -\left[\frac{2(\nu+\eta) zf'(z)}{\nu z(f'(z)+ f'(-z)) +\eta (f(z)- f(-z))}\right]^\mu, ~z\in\mathbb E\,.\label{e12}\end{equation}
(14)
According to the assumptions (11) and (12), the multivalued power function \(p\) has an analytic branch in \(\mathbb E\) with \(p(0)=1\) and from the (14) it follows that \[\mu\left[1-\frac{\nu z[z(f'(z)+ f'(-z))]’ +\eta z(f'(z)+ f'(-z))}{\nu z(f'(z)+ f'(-z)) +\eta (f(z)- f(-z))}+\frac{zf”(z)}{f'(z)}\right]=\frac{zp'(z)}{p(z)}.\] To prove our desired result we will use Lemma 1. Thus, let the functions \(\theta(w)=1\) and \(\phi(w)=\displaystyle\frac{\gamma}{w}\). Also, if we let \( Q(z)=zq'(z) \phi(q(z))=\gamma zq'(z)\) and \(h(z)=\theta(q(z))+Q(z)=1+\gamma\displaystyle\frac{z q'(z)}{q(z)}.\)

Since \(Q(0)=0\) and \(Q'(0)=\gamma q'(0)\neq0\), the assumption (10) implies that \(Q\) is starlike in \(\mathbb E\) and

\[\Re\left(\frac{zh'(z)}{Q(z)}\right)=\Re \left(1+\frac{zq”(z)}{q'(z)}+\frac{z q'(z)}{q(z)}\right)>0, ~z \in \mathbb E.\] Therefore, using Lemma 1 and assumption (13) implies \(p(z)\prec q(z)\) and the function \(q\) is the best dominant of (13).

In particular, taking \(\nu=0, \eta=\gamma=1\) and \(\displaystyle q(z)=\frac{1+Az}{1+Bz}\) in the above theorem, it is easy to check that the inequality (10) holds whenever \(-1\leq B< A\leq 1\). Hence, we deduce the following corollary:

Corollary 2. Let \(-1\leq B< A\leq 1\) and \(\mu\in\mathbb C^*\). Let \(f\in\Sigma\) satisfy the conditions (4) and

\begin{equation}\frac{-2zf'(z)}{f(z)-f(-z)}\neq0,~z\in\mathbb E.\label{e13}\end{equation}
(15)
If
\begin{equation}1+\mu\left[1-\frac{ z(f'(z)+ f'(-z))} {f(z)-f(-z)}+\frac{zf”(z)}{f'(z)}\right]\prec 1+\frac{(A-B)z}{(1+Az)(1+Bz)}\,,\label{e14}\end{equation}
(16)
then \[-\left[\frac{2zf'(z)}{f(z)-f(-z)}\right]^\mu\prec \frac{(1+Az)}{(1+Bz)}, \] and the function \( \displaystyle\frac{(1+Az)}{(1+Bz)}\) is the best dominant of (16) (the power is principal one).

Using \(\nu=0, \eta=1, \gamma=\displaystyle \frac{1}{ab}\) with \(a,b\in\mathbb C^*, \mu=a\) and \( q(z)=\displaystyle \frac{1}{(1-z)^{2ab}}\) in Theorem 2, then using this result together with Lemma 3, we deduce the following result:

Corollary 3. Let \(a,b\in\mathbb C^*\) such that \[|2ab-1|\leq1 \hspace{1cm}or \hspace{1cm}|2ab+1|\leq1\] and suppose that \(f\in\Sigma\) satisfy the conditions (4) and (15). If \[1+\frac{1}{b}\left[1-\frac{z(f'(z)+f'(-z))}{f(z)-f(-z)}+\frac{zf”(z)}{f'(z)}\right]\prec\frac{1+z}{1-z},\] then

\begin{equation}-\left[\frac{2zf'(z)}{f(z)-f(-z)}\right]^a \prec\displaystyle \frac{1}{(1-z)^{2ab}}\label{e15}\end{equation}
(17)
and the function \(\displaystyle \frac{1}{(1-z)^{2ab}}\) is the best dominant of (17) (the power is the principal one).

By using \(\nu=0, \eta=\gamma=1\) and \( q(z)= \displaystyle(1+Bz)^{\frac{\mu(A-B)}{B}}\), \(-1\leq B< A\leq 1\), \(B\neq0\) in Theorem 2, then using Lemma 3 we have the next result:

Corollary 4. Let \(-1\leq B< A\leq 1\) with \(B\neq0\), and suppose that \[\left|\frac{\mu(A-B)}{B-1}\right|\leq1 \hspace{1cm}or \hspace{1cm}\left|\frac{\mu(A-B)}{B+1}\right|\leq1,\] where \(\mu\in\mathbb C^*\). If \(f\in\Sigma\) satisfy the conditions (4), (15) and

\begin{equation}1+\mu\left[1-\frac{ z(f'(z)+ f'(-z))} {f(z)-f(-z)}+\frac{zf”(z)}{f'(z)}\right]\prec \frac{1+[B+\mu(A-B)]z}{(1+Bz)},\label{e16}\end{equation}
(18)
then \[-\left[\frac{2zf'(z)}{f(z)-f(-z)}\right]^\mu\prec (1+Bz)^{\frac{\mu(A-B)}{B}} \] and the function \((1+Bz)^{\frac{\mu(A-B)}{B}}\) is the best dominant of (18) (the power is the principal one).

By taking \(\nu=0, \eta=1\), \(\gamma=\frac{e^{\iota\lambda}}{ab \cos \lambda}, a,b\in\mathbb C^*, |\lambda|< \displaystyle\frac{\pi}{2}, \mu=a\) and \(q(z)=\displaystyle\frac{1}{(1-z)^{2abe^{-\iota\lambda} \cos \lambda}}\) in Theorem 2, we obtain the following:

Corollary 5. Let \(a,b\in\mathbb C^*\) and \(|\lambda|< \frac{\pi}{2}\) and suppose that \[ |2abe^{-\iota\lambda} \cos \lambda-1|\leq1 \hspace{1cm} \text{or} \hspace{1cm} |2abe^{-\iota\lambda} \cos \lambda+1|\leq1 .\] If \(f\in\Sigma\) satisfy the conditions (4), (15) and \[1+\frac{e^{\iota\lambda}}{b \cos \lambda}\left[1-\frac{z(f'(z)+f'(-z))}{f(z)-f(-z)}+\frac{zf''(z)}{f'(z)}\right]\prec\frac{1+z}{1-z},\] then

\begin{equation}-\left[\frac{2zf'(z)}{f(z)-f(-z)}\right]^a \prec\displaystyle \frac{1}{(1-z)^{2abe^{-\iota\lambda} \cos \lambda}}\label{e17}\end{equation}
(19)
and the function \(\displaystyle \frac{1}{(1-z)^{2abe^{-\iota\lambda} \cos \lambda}}\) is the best dominant of (19) (the power is the principal one).

Theorem 3. Suppose that \(q\) be univalent in \(\mathbb E\) with \(q(0)=1\) and \(q(z)\neq0\) for all \(z\in\mathbb E\), such that

\begin{equation} \Re \left(1+\frac{zq”(z)}{q'(z)}\right)>max\left\{0;~-\Re \frac{\delta}{\gamma}\right\}, ~z\in\mathbb E,\label{e18}\end{equation}
(20)
where \(\gamma,\mu\in\mathbb C^*\) and \(\delta,\nu,\eta\in\mathbb C\) with \(\nu+\eta\neq 0\) and let \(f\in\Sigma\) satisfy the conditions (11), (12) and
\begin{equation} \phi(z)=-\left[\frac{2(\nu+\eta) zf'(z)}{\nu z(f'(z)+ f'(-z)) +\eta (f(z)- f(-z))}\right]^\mu.\label{e19}\end{equation}
(21)
\[\left[\delta+\gamma\mu\left(1-\frac{\nu z[z(f'(z)+ f'(-z))]’ +\eta z(f'(z)+ f'(-z))}{\nu z(f'(z)+ f'(-z)) +\eta (f(z)-f(-z))}+\frac{zf”(z)}{f'(z)}\right)\right]\,.\] If
\begin{equation} \phi(z)\prec \delta q(z)+\gamma z q'(z),\label{e20}\end{equation}
(22)
then \[-\left[\frac{2(\nu+\eta) zf'(z)}{\nu z(f'(z)+ f'(-z)) +\eta (f(z)- f(-z))}\right]^\mu\prec q(z)\] and \(q\) is the best dominant of (13) (the power is the principal one).

Proof. Define a function

\begin{equation} p(z)= -\left[\frac{2(\nu+\eta) zf'(z)}{\nu z(f'(z)+ f'(-z)) +\eta (f(z)- f(-z))}\right]^\mu, ~z\in\mathbb E\,.\label{e21}\end{equation}
(23)
From the assumptions (11) and (12) it follows that the multivalued power function \(p\) has an analytic branch in \(\mathbb E\) with \(p(0)=1\) and from the (23) it follows that \[\mu p(z) \left[1-\frac{\nu z[z(f'(z)+ f'(-z))]’ +\eta z(f'(z)+ f'(-z))}{\nu z(f'(z)+ f'(-z)) +\eta (f(z)- f(-z))}+\frac{zf”(z)}{f'(z)}\right]=zp'(z).\] Consider the functions \(\theta(w)=\delta w\) and \(\phi(w)=\gamma\) that are analytic in \(\mathbb C^*\). Also, if we let \( Q(z)=zq'(z) \phi(q(z))=\gamma zq'(z)\) and \(h(z)=\theta(q(z))+Q(z)=\delta q(z)+\gamma\frac{z q'(z)}{q(z)}.\)

Since \(Q(0)=0\) and \(Q'(0)=\gamma q'(0)\neq0\), the assumption (20) implies that \(Q\) is starlike in \(\mathbb E\) and

\[\Re\left(\frac{zh'(z)}{Q(z)}\right)=\Re \left(\frac{\delta}{\gamma}+1+\frac{zq”(z)}{q'(z)}+\frac{z q'(z)}{q(z)}\right)>0, ~z \in \mathbb E.\] Therefore, using Lemma 1 and assumption (22) implies \(p(z)\prec q(z)\) and the function \(q\) is the best dominant of (22).

Taking \(\displaystyle q(z)=\frac{1+Az}{1+Bz}\) in Theorem 3, where \(-1\leq B< A\leq1\), according to (7) the condition (20) becomes \[max\left\{0;~-\Re \frac{\delta}{\gamma}\right\}\leq\frac{1-|B|}{1+|B|}.\] Hence, for the particular cases \(\eta=0\) and \(\mu=\gamma=1\), we have the following result:

Corollary 6. Let \(-1\leq B< A\leq 1, \mu\in\mathbb C^*\) and \(\delta\in\mathbb C\) with \[max\left\{0;~-\Re \frac{\delta}{\gamma}\right\}\leq\frac{1-|B|}{1+|B|}.\] If \(f\in\Sigma\) satisfy the conditions (4), (15) and

\begin{equation}-\left[\frac{2f'(z)}{z(f'(z)+ f'(-z))}\right]^\mu \left[\delta+\mu\left(1-\frac{(z(f'(z)+ f'(-z)))’}{(f'(z)+f'(-z))}\right)\right] \prec \delta\frac{1+Az}{1+Bz}+\frac{(A-B)z}{(1+Bz)^2},\label{e22}\end{equation}
(24)
then \[ -\left[\frac{2f'(z)}{f'(z)+ f'(-z)}\right]^\mu \prec \frac{1+Az}{1+Bz}\] and the function \(\displaystyle \frac{1+Az}{1+Bz}\) is the best dominant of (24) (all the powers are principal ones).

Taking \(\eta=\gamma=1,\mu=0\) and \(q(z)=\displaystyle\frac{1+z}{1-z}\) in Theorem 3, we get the following corollary:

Corollary 7. Let \(\mu\in\mathbb C^*\) and \(\delta\in\mathbb C\) with \(\Re \delta \geq 0.\) If \(f\in\Sigma\) satisfy the conditions (4), (15) and

\begin{equation} -\left[\frac{2zf'(z)}{f(z)-f(-z))}\right]^\mu \left[\delta+\mu\left(1-\frac{z(f'(z)+ f'(-z))}{(f(z)-f(-z))}+\frac{zf”(z)}{f'(z)}\right)\right] \prec \delta\frac{1+z}{1-z}+\frac{2z}{(1-z)^2},\label{e23}\end{equation}
(25)
then \[ -\left[\frac{2f'(z)}{f'(z)+ f'(-z)}\right]^\mu \prec \frac{1+z}{1-z},\] and the function \(\displaystyle \frac{1+z}{1-z}\) is the best dominant of (25) (all the powers are principal ones).

4. Superordination and sandwich theorems

Theorem 4. Let \(q\) be convex in \(\mathbb E\) with \(q(0)=1\) and \(\lambda\in\mathbb C\) with \(\Re~ \lambda>0\). Let \(f\in\Sigma\) satisfy the condition (4) such that \(\displaystyle-\frac{2zf'(z)}{f(z)-f(-z)}\in\mathbb Q\) and suppose that the function \[ \frac{-2(1+\lambda)zf'(z)}{f(z)-f(-z)}+\frac{2\lambda z^2 f'(z)(f'(z)+f'(-z))}{(f(z)-f(-z))^2}-\frac{2\lambda z^2 f”(z)}{f(z)-f(-z)}\] is univalent in \(\mathbb E\). If

\begin{equation} q(z)+\lambda z q'(z) \prec \frac{-2(1+\lambda)zf'(z)}{f(z)-f(-z)}+\frac{2\lambda z^2 f'(z)(f'(z)+f'(-z))}{(f(z)-f(-z))^2}-\frac{2\lambda z^2 f”(z)}{f(z)-f(-z)}\,,\label{e41}\end{equation}
(26)
then \[q(z)\prec-\frac{2zf'(z)}{f(z)-f(-z)}\] and \(q\) is the best subordinant of (26).

Proof. Setting \[p(z)=\frac{-2zf'(z)}{f(z)-f(-z)},~z\in\mathbb E,\] then \(p\) is analytic in \(\mathbb E\) with \(p(0) = 1.\) Taking logarithmic differentiation of the above relation with respect to \(z,\) we have \[\frac{zp'(z)}{p(z)}=1-z\left(\frac{f'(z)+f'(-z)}{f(z)-f(-z)}-\frac{f”(z)}{f'(z)}\right)\] and a simple calculation yields that the assumption (26) is equivalent to \[ q(z)+\lambda zq'(z)\prec p(z)+\lambda zp'(z).\] Now, in order to prove our result we will use Lemma 2. Consider the functions \(\theta(w)=w\) and \(\phi(w)=\lambda\) analytic in \(\mathbb C\) and set \[h(z)=zq'(z)\phi(q(z))=\lambda z q'(z).\] Since \(h(0)=0\), \(h'(0)=\lambda q'(0)\neq0\) and \(q\) is convex in \(\mathbb E\), it follows that \(h\) is starlike in \(\mathbb E\) and \[\Re \frac{\theta’q(z)}{\phi(q(z))}=\Re \frac{1}{\gamma}>0, ~z \in \mathbb E.\] Therefore, Lemma 2 and assumption (26) imply \(q(z)\prec p(z)\) and the function \(q\) is the best subordinant of (26).

Taking \(\displaystyle q(z)=\frac{1+Az}{1+Bz}\) in Theorem 4, where \(-1\leq B< A\leq1\), we get the following corollary:

Corollary 8. Let \(q\) be convex in \(\mathbb E\) with \(q(0)=1\) and \(\lambda\in\mathbb C\) with \(\Re~ \lambda>0\). Let \(f\in\Sigma\) satisfy the condition (4) such that \(\displaystyle-\frac{2zf'(z)}{f(z)-f(-z)}\in\mathbb Q\) and suppose that the function \[ \frac{-2(1+\lambda)zf'(z)}{f(z)-f(-z)}+\frac{2\lambda z^2 f'(z)(f'(z)+f'(-z))}{(f(z)-f(-z))^2}-\frac{2\lambda z^2 f”(z)}{f(z)-f(-z)}\] is univalent in \(\mathbb E\). If

\begin{equation} \frac{1+Az}{1+Bz}+\lambda\frac{(A-B)z}{(1+Bz)^2}\prec \frac{-2(1+\lambda)zf'(z)}{f(z)-f(-z)}+\frac{2\lambda z^2 f'(z)(f'(z)+f'(-z))}{(f(z)-f(-z))^2}-\frac{2\lambda z^2 f”(z)}{f(z)-f(-z)}\,,\label{e42}\end{equation}
(27)
then \[\displaystyle\frac{1+Az}{1+Bz}\prec-\frac{2zf'(z)}{f(z)-f(-z)}\] and \(q\) is the best subordinant of (27).

Using the same techniques as in proof of Theorem 3 and then applying Lemma 2, we could prove the next theorem:

Theorem 5. Let \(\gamma,\mu\in\mathbb C^*\) and \(\delta,\nu,\eta\in\mathbb C\) with \(\nu+\eta\neq 0\) and \(\Re\left (\displaystyle\frac{\delta}{\gamma}\right)>0\). Suppose that \(q\) is convex in \(\mathbb E\), with \(q(0)=1\) and let \(f\in\Sigma\) satisfy the conditions (11), (12) and \[-\left[\frac{2(\nu+\eta) zf'(z)}{\nu z(f'(z)+ f'(-z)) +\eta (f(z)- f(-z))}\right]^\mu\in\mathbb Q. \] If the function \(\phi \) given by (21) is univalent in \(\mathbb E\) and

\begin{equation}\delta q(z)+\gamma z q'(z)+\sigma\prec \phi(z),\label{e43}\end{equation}
(28)
then \[q(z)\prec -\left[\frac{2(\nu+\eta) zf'(z)}{\nu z(f'(z)+ f'(-z)) +\eta (f(z)- f(-z))}\right]^\mu\] and \(q\) is the best subordinant of (28) (the power is the principal one).

If we combine Theorem 1 with Theorem 4 and Theorem 3 with Theorem 5, we deduce the following sandwich results, respectively:

Theorem 6. Let \(q_1\) and \(q_2\) be two convex functions in \(\mathbb E\) with \(q_1(0)=q_2(0)=1\) and \(\lambda\in\mathbb C\) with \(\Re~ \lambda>0\). Let \(f\in\Sigma\) satisfy the condition (4), such that \(\displaystyle-\frac{2zf'(z)}{f(z)-f(-z)}\in\mathbb Q\) and suppose that the function \[ \frac{-2(1+\lambda)zf'(z)}{f(z)-f(-z)}+\frac{2\lambda z^2 f'(z)(f'(z)+f'(-z))}{(f(z)-f(-z))^2}-\frac{2\lambda z^2 f”(z)}{f(z)-f(-z)}\] is univalent in \(\mathbb E\). If \[ q_1(z)+\lambda zq_1′(z)\prec \frac{-2(1+\lambda)zf'(z)}{f(z)-f(-z)}+\frac{2\lambda z^2 f'(z)(f'(z)+f'(-z))}{(f(z)-f(-z))^2}-\frac{2\lambda z^2 f”(z)}{f(z)-f(-z)}\,.\]

\begin{equation}\prec q_2(z)+\lambda zq_2′(z)\,,\label{e44}\end{equation}
(29)
then \[q_1(z)\prec\frac{-2zf'(z)}{f(z)-f(-z)}\prec q_2(z),\] where \(q_1\) and \(q_2\) are respectively the best subordinant and the best dominant of (29).

Theorem 7. Let \(q_1\) and \(q_2\) be two convex functions in \(\mathbb E\) with \(q_1(0)=q_2(0)=1\) and let \(\gamma,\mu\in\mathbb C^*\) and \(\nu,\eta\in\mathbb C\) with \(\nu+\eta\neq 0\) and \(\Re\displaystyle \frac{\delta}{\gamma}>0\). Let \(f\in\Sigma\) satisfy the condition (11), (12) and such that \(\displaystyle-\frac{2zf'(z)}{f(z)-f(-z)}\in\mathbb Q\) and \[-\left[\frac{2(\nu+\eta) zf'(z)}{\nu z(f'(z)+ f'(-z)) +\eta (f(z)-f(-z))}\right]^\mu\in\mathbb Q. \] If the function \(\phi \) given by (21) is univalent in \(\mathbb E\) and

\begin{equation}\delta q_1(z)+\gamma z q_1′(z)\prec \phi(z)\prec\delta q_2(z)+\gamma z q_2′(z)\,,\label{e45}\end{equation}
(30)
then \[q_1(z)\prec -\left[\frac{2(\nu+\eta) zf'(z)}{\nu z(f'(z)+ f'(-z)) +\eta (f(z)- f(-z))}\right]^\mu\prec q_2(z)\] and \(q_1\) and \(q_2\) are respectively the best subordinant and the best dominant of (29) (the power is the principal one).

Author Contributions

All authors contributed equally to the writing of this paper. All authors read and approved the final manuscript.

Conflicts of Interest

”The authors declare no conflict of interest.”

References:

  1. Sakaguchi, K. (1959). On a certain univalent mapping. Journal of the Mathematical Society of Japan, 11(1), 72-75. [Google Scholor]
  2. Chand, R., & Singh, P. (1979). On certain schlicht mappings. Indian Journal of Pure and Applied Mathematics, 10,(9), 1167-1174. [Google Scholor]
  3. Thangamani, J. (1980). On starlike functions with respect to symmetric points. Indian Journal of Pure and Applied Mathematics, 11, 392-405. [Google Scholor]
  4. Khan, M. G., Darus, M., Ahmad, B., Murugusundaramoorthy, G., Khan, R., & Khan, N. (2020). Meromorphic starlike functions with respect to symmetric points. International Journal of Analysis and Applications, 18(6), 1037-1047. [Google Scholor]
  5. Miller, S. S., & Mocanu, P. T. (2000). Differential Subordinations: Theory and Applications. CRC Press. [Google Scholor]
  6. Bulboaca, T. (2002). Classes of first-order differential superordinations. Demonstratio Mathematica, 35(2), 287-292. [Google Scholor]
  7. Royster, W. C. (1965). On the univalence of a certain integral. Michigan Mathematical Journal, 12(4), 385-387.[Google Scholor]