Skip to main content

On some geometric properties of multivalent functions

Abstract

We prove some new sufficient conditions for a function to be p-valent or p-valently starlike in the unit disc.

1 Introduction

Let \(\mathcal{A}_{p}\) be the class of analytic functions of the form

$$ f(z)=z^{p}+\sum_{n=1}^{\infty}a_{p+n}z^{p+n}\quad (z\in \mathbb{D}). $$
(1.1)

A function \(f(z)\) which is analytic in a domain \(D \subset C\) is called p-valent in D if for every complex number w, the equation \(f(z)= w\) has at most p roots in D, and there will be a complex number \(w_{0}\) such that the equation \(f(z) = w_{0}\) has exactly p roots in D. Further, a function \(f\in\mathcal{A}_{p}\), \(p\in\mathbb{N}\setminus\{1\}\), is said to be p-valently starlike of order α, \(0\le\alpha< p\), if

$$ \mathfrak{Re} \biggl\{ \frac{zf'(z)}{f(z)} \biggr\} >\alpha \quad (z\in \mathbb{D}). $$

The class of all such functions is usually denoted by \(\mathcal{S}^{*}_{p}(\alpha)\). For \(p=1\), we receive the well-known class of normalized starlike univalent functions \(\mathcal{S}^{*}(\alpha)\) of order α, \(\mathcal{S}_{p}^{*}(0)=\mathcal{S}_{p}^{*}\). If \(zf'(z)\in\mathcal{S}^{*}(\alpha)\), then \(f(z)\) is said to be p-valently convex of order α, \(0\le\alpha< p\). The class of all such functions is usually denoted by \(\mathcal{C}_{p}(\alpha)\). For \(p=1\), we receive the well-known class of normalized convex univalent functions \(\mathcal{C}(\alpha)\) of order α, \(\mathcal{C}_{p}(0)=\mathcal{C}_{p}\).

The well-known Noshiro-Warschawski univalence condition (see [1] and [2]) indicates that if \(f(z)\) is analytic in a convex domain \(D\subset\mathbb{C}\) and

$$ \mathfrak{Re}\bigl\{ e^{i\theta}f(z)\bigr\} >0\quad (z\in D), $$

where θ is a real number, then \(f(z)\) is univalent in D. In [3] Ozaki extended the above result by showing that if \(f(z)\) of the form (1.1) is analytic in a convex domain D and for some real θ we have

$$ \mathfrak{Re} \bigl\{ e^{i\theta}f^{(p)}(z)\bigr\} > 0,\quad |z|< 1, $$

then \(f(z)\) is at most p-valent in D. Applying Ozaki’s theorem, we find that if \(f(z)\in\mathcal{A}_{p}\) and

$$ \mathfrak{Re}\bigl\{ f^{(p)}(z)\bigr\} >0 \quad(z\in \mathbb{D}), $$

then \(f(z)\) is at most p-valent in \(|z|<1\). In [4] it was proved that if \(f(z)\in\mathcal{A}_{p}\), \(p\geq2\), and

$$ \arg\bigl|f^{(p)}(z)\bigr|< \frac{3\pi}{4},\quad |z|< 1, $$

then \(f(z)\) is at most p-valent in \(|z|<1\).

2 Preliminary lemmata

Lemma 2.1

[5]

Let \(f(z) = z + a_{2} z^{2} + \cdots\) be analytic in the unit disc and suppose that

$$ \bigl|f''(z)\bigr|< 1,\quad |z|< 1, $$
(2.1)

then \(f(z)\) is univalent in \(|z| < 1\).

Lemma 2.2

[6]

Let \(p(z)\) be an analytic function in \(|z|<1\) with \(p(0)=1\), \(p(z)\neq0\). If there exists a point \(z_{0}\), \(|z_{0}|<1\), such that

$$ \bigl|{\arg} \bigl\{ p(z) \bigr\} \bigr|< \frac{\pi\alpha}{2} \quad\textit{for } |z|< |z_{0}| $$

and

$$ \bigl|{\arg} \bigl\{ p(z_{0}) \bigr\} \bigr|=\frac{\pi\alpha}{2} $$

for some \(0<\alpha<2\), then we have

$$ \frac{z_{0}p'(z_{0})}{p(z_{0})}=is\alpha, $$

where

$$ s\geq\frac{1}{2} \biggl(a+\frac{1}{a} \biggr)\quad \textit{when }{\arg} \bigl\{ p(z_{0}) \bigr\} =\frac{\pi\alpha}{2} $$

and

$$ s\leq-\frac{1}{2} \biggl(a+\frac{1}{a} \biggr)\quad \textit{when }{\arg} \bigl\{ p(z_{0}) \bigr\} =-\frac{\pi\alpha}{2}, $$

where

$$ \bigl\{ p(z_{0}) \bigr\} ^{1/\alpha}=\pm ia, \textit{ and }a>0. $$

Lemma 2.3

[7]

Let p be a positive integer. If \(f(z)=z^{p}+\sum_{n=p+1}^{\infty}a_{n}z^{n}\) is analytic in \(\mathbb{D}\) and if it satisfies

$$ \mathfrak{Re} \biggl\{ \frac{zf^{(p)}(z)}{f^{(p-1)}(z)} \biggr\} >0,\quad z\in\mathbb{D}, $$
(2.2)

then \(f(z)\) is p-valently starlike in \(\mathbb{D}\) and

$$ \mathfrak{Re} \biggl\{ \frac{zf^{(k)}(z)}{f^{(k-1)}(z)} \biggr\} >0,\quad z\in\mathbb{D} $$
(2.3)

for \(k=1,2,\ldots,(p-1)\).

Lemma 2.4

([7], p.282)

Let \(f\in\mathcal{A}_{p}\). If there exists a \((p-k+1)\)-valent starlike function \(g(z)=\sum_{n=p-k+1}^{\infty}b_{n}z^{n}\) (\(b_{p-k+1}\neq0\)) that satisfies

$$ \mathfrak{Re} \biggl\{ \frac{zf^{(k)}(z)}{g(z)} \biggr\} > 0,\quad |z| < 1, $$
(2.4)

then \(f(z)\) is p-valent in \(|z|<1\).

3 Main results

Now we state and prove the main results. The first theorem poses a growth condition to a higher derivative of an analytic function. Hence, its proof, among others, uses the method of integrating the derivatives, which is frequently used in complex analysis, especially in estimating point evaluation operators (see, e.g., Lemma 7 in [8] and Lemma 4 in [9]).

Theorem 3.1

Let \(f\in\mathcal{A}_{p}\) and suppose that

$$ \bigl| f^{(p+1)}(z) \bigl| < \alpha(\beta_{0}) (p!),\quad |z| < 1, $$
(3.1)

where \(\beta_{0}=0.38\cdots\) is the positive root of the equation

$$ 2\beta+ \frac{2}{\pi}\tan^{-1}\beta= 1,\quad 0 < \beta< 1 $$
(3.2)

and

$$\begin{aligned} \alpha(\beta_{0}) =& \sin \biggl\{ \frac{\pi}{2} \bigl( \beta_{0} + (2/\pi )\tan^{-1}\beta_{0} \bigr) \biggr\} \\ =&\sin \biggl\{ \frac{(1-\beta_{0})\pi}{2} \biggr\} \\ =&0.82\cdots. \end{aligned}$$

Then we have

$$ \mathfrak{Re} \biggl\{ \frac{zf^{(p)}(z)}{f^{(p-1)}(z)} \biggr\} > 0,\quad |z| < 1, $$
(3.3)

and, therefore, we have

$$ \mathfrak{Re} \biggl\{ \frac{zf'(z)}{f(z)} \biggr\} > 0,\quad |z| < 1, $$
(3.4)

or \(f(z)\) is p-valently starlike in \(|z| < 1\).

Proof

From the hypothesis (3.1), we have

$$\begin{aligned} \bigl|f^{(p)}(z)-p!\bigr| =& \biggl\vert \int_{0}^{z}f^{(p+1)}(t)\, \mathrm{d}t\biggr\vert \\ \leq& \int_{0}^{|z|}\bigl\vert f^{(p+1)}(t)\bigr\vert |\mathrm{d}t|< p!\int_{0}^{|z|} \alpha(\beta_{0})|\mathrm{d}t| \\ =& p! \alpha(\beta_{0})|z|< p! \alpha(\beta_{0}). \end{aligned}$$

This shows that

$$ \bigl\vert \arg\bigl\{ f^{(p)}(z)\bigr\} \bigr\vert < \sin^{-1}\alpha(\beta_{0}),\quad |z| < 1. $$
(3.5)

Let us put

$$ q(z)=\frac{f^{(p-1)}(z)}{p!z},\qquad q(0)=1. $$
(3.6)

Then it follows that

$$ q(z)+zq'(z)=\frac{f^{(p)}(z)}{p!}. $$

If there exists a point \(z_{0}\), \(|z_{0}|<1\), such that

$$ \bigl|\arg\bigl\{ q(z)\bigr\} \bigr|< \frac{\pi\beta_{0}}{2} \quad\mbox{for } |z|< |z_{0}| $$

and

$$ \bigl|\arg\bigl\{ q(z_{0})\bigr\} \bigr|=\frac{\pi\beta_{0}}{2}, $$

then by Lemma 2.2 we have

$$ \frac{z_{0}q'(z_{0})}{q(z_{0})}=i\beta_{0}k, $$

where

$$ k\geq1 \quad\mbox{when } \arg\bigl\{ q(z_{0})\bigr\} = \frac{\pi\beta_{0}}{2} $$
(3.7)

and

$$ k\leq-1 \quad\mbox{when } \arg\bigl\{ q(z_{0})\bigr\} =- \frac{\pi\beta_{0}}{2}. $$
(3.8)

For the case (3.7), we have

$$\begin{aligned} \arg\bigl\{ f^{(p)}(z)\bigr\} =& \arg \biggl\{ \frac{f^{(p)}(z)}{p!} \biggr\} \\ =& \arg\bigl\{ q(z_{0})+z_{0}q'(z_{0}) \bigr\} \\ =& \arg \biggl\{ q(z_{0}) \biggl(1+\frac{z_{0}q'(z_{0})}{q(z_{0})} \biggr) \biggr\} \\ =& \frac{\pi\beta_{0}}{2}+\arg\{1+i\beta_{0}k\} \\ \geq& \frac{\pi\beta_{0}}{2}+\tan^{-1}\beta_{0} \\ =&\frac{\pi}{2} \bigl\{ \beta_{0} + (2/\pi)\tan^{-1}( \beta_{0}) \bigr\} \\ =&\sin^{-1}\alpha(\beta_{0}). \end{aligned}$$

This contradicts (3.5), and for the case (3.8), applying the same method as above, we have

$$\begin{aligned} \arg\bigl\{ f^{(p)}(z)\bigr\} \leq-\sin^{-1}\alpha( \beta_{0}). \end{aligned}$$

This also contradicts (3.5) and, therefore, it shows that

$$ \biggl\vert \arg \biggl\{ \frac{f^{(p-1)}(z)}{p!z} \biggr\} \biggr\vert < \frac{\pi\beta _{0}}{2},\quad |z| < 1. $$
(3.9)

Applying (3.5) and (3.9), we have

$$\begin{aligned} \biggl\vert \arg \biggl\{ \frac{zf^{(p)}(z)}{f^{(p-1)}(z)} \biggr\} \biggr\vert =& \biggl\vert \arg \bigl\{ f^{(p)}(z) \bigr\} +\arg \biggl\{ \frac {z}{f^{(p-1)}(z)} \biggr\} \biggr\vert \\ \leq& \bigl\vert \arg \bigl\{ f^{(p)}(z) \bigr\} \bigr\vert +\biggl\vert \arg \biggl\{ \frac{z}{f^{(p-1)}(z)} \biggr\} \biggr\vert \\ < &\sin^{-1}\alpha(\beta_{0})+\frac{\pi\beta_{0}}{2} \\ =&\frac{\pi}{2} \biggl(2\beta_{0}+\frac{2}{\pi} \tan^{-1}\beta_{0} \biggr) \\ =&\frac{\pi}{2}. \end{aligned}$$

This shows that

$$ \mathfrak{Re} \biggl\{ \frac{zf^{(p)}(z)}{f^{(p-1)}(z)} \biggr\} > 0,\quad |z| < 1, $$

and by Lemma 2.3 we have

$$ \mathfrak{Re} \biggl\{ \frac{zf'(z)}{f(z)} \biggr\} > 0,\quad |z| < 1, $$

or \(f(z)\) is p-valently starlike in \(|z| < 1\). □

Remark

Note that if \(m(x)=2x+\frac{2}{\pi}\tan^{-1}x\), then

$$ m(0.383)=0.9988537761\cdots,\qquad m(0.384)=1.001408771\cdots. $$

Hence, if

$$ 2\beta_{0} + \frac{2}{\pi}\tan^{-1} \beta_{0} = 1, $$

then \(\beta_{0}=0.38\cdots\). Moreover,

$$ 0.9673808495\cdots< \sin^{-1}\alpha(\beta_{0})= \frac{\pi}{2} \biggl(\beta_{0} + \frac{2}{\pi} \tan^{-1}\beta_{0} \biggr)=\frac{(1-\beta_{0})\pi }{2}< 0.96982343 \cdots $$

and

$$ \alpha(\beta_{0})= \sin \biggl\{ \frac{\pi}{2} \biggl( \beta_{0} + \frac{2}{\pi}\tan^{-1}\beta_{0} \biggr) \biggr\} =0.824669\cdots. $$

For \(p=1\), Theorem 3.1 becomes the following corollary which extends the result contained in Lemma 2.1.

Corollary 3.2

Let \(f\in\mathcal{A}(1)\) and suppose that

$$ \bigl| f''(z) \bigr| < \alpha( \beta_{0}),\quad |z| < 1, $$
(3.10)

where \(\beta_{0}=0.38\cdots\) is the positive root of the equation

$$ 2\beta+ \frac{2}{\pi}\tan^{-1}\beta= 1, \quad 0 < \beta< 1, $$
(3.11)

and

$$\begin{aligned} \alpha(\beta_{0}) =&\sin \biggl\{ \frac{(1-\beta_{0})\pi}{2} \biggr\} \\ =&0.82\cdots. \end{aligned}$$

Then we have

$$ \mathfrak{Re} \biggl\{ \frac{zf'(z)}{f(z)} \biggr\} > 0,\quad |z| < 1, $$
(3.12)

or \(f(z)\) is starlike univalent in \(|z| < 1\).

An analytic function \(f(z)\) is said to be typically real if the inequality \(\mathfrak{Im}z\mathfrak{Im} f(z)\geq0\) holds for all \(z\in\mathbb{D}\). From the definition of a typically real function it follows that \(z\in\mathbb{D}^{+}\Leftrightarrow f(z)\in\mathbb{C}^{+}\) and \(z\in\mathbb{D}^{-}\Leftrightarrow f(z)\in\mathbb{C}^{-}\). The symbols \(\mathbb{D}^{+}\), \(\mathbb{D}^{-}\), \(\mathbb{C}^{+}\), \(\mathbb{C}^{-}\) stand for the following sets: the upper and the lower halves of the disk \(\mathbb{D}\), the upper half-plane and the lower half-plane, respectively.

Theorem 3.3

Let \(f(z)\in\mathcal{A}_{p}\) and suppose that

$$ \biggl\vert \arg \biggl\{ \frac{ zf^{(p)}(z)}{g(z)} \biggr\} \biggr\vert < \frac{\pi}{2}+ \tan^{-1}\frac{1-|z|}{1+|z|},\quad z\in\mathbb{D}, $$
(3.13)

where \(g(z)\) is univalent starlike in \(\mathbb{D}\) and the functions

$$ \frac{zg'(z)}{g(z)} \quad \textit{and} \quad \frac{z (\frac{ zf^{(p-1)}(z)}{g(z)} )'}{\frac{ zf^{(p-1)}(z)}{g(z)}} $$
(3.14)

are typically real in \(\mathbb{D}\). Then we have

$$ \biggl\vert \arg\frac{ zf^{(p-1)}(z)}{g(z)} \biggr\vert < \frac{\pi}{2},\quad z\in\mathbb{D}. $$
(3.15)

Proof

Let us put

$$ q(z) = \frac{zf^{(p-1)}(z)}{p! g(z)}, \qquad q(0) = 1. $$

Then it follows that

$$ \frac{zf^{(p)}(z)}{ g(z)}=p!q(z) \biggl(\frac{zg'(z)}{g(z)}+ \frac{zq'(z)}{q(z)} \biggr). $$
(3.16)

From the hypothesis

$$ \frac{zg'(z)}{g(z)}+\frac{zq'(z)}{q(z)} $$

is typically real in \(\mathbb{D}\). If there exists a point \(z_{0}\), \(|z_{0}|<1\), such that

$$ \bigl|{\arg} \bigl\{ q(z) \bigr\} \bigr|< \frac{\pi}{2} \quad\mbox{for } |z|< |z_{0}| $$

and

$$ \bigl|{\arg} \bigl\{ q(z_{0}) \bigr\} \bigr|=\frac{\pi}{2},\quad q(z_{0})=\pm ia, \mbox{ and } a>0, $$

then by Lemma 2.2 we have

$$ \frac{z_{0}q'(z_{0})}{q(z_{0})}=is, $$
(3.17)

where

$$ s\geq\frac{1}{2} \biggl(a+\frac{1}{a} \biggr) \quad\mbox{when } {\arg} \bigl\{ q(z_{0}) \bigr\} =\frac{\pi}{2} $$

and

$$ s\leq-\frac{1}{2} \biggl(a+\frac{1}{a} \biggr) \quad\mbox{when } {\arg} \bigl\{ q(z_{0}) \bigr\} =-\frac{\pi}{2}. $$

For the case

$$ {\arg} \bigl\{ q(z_{0}) \bigr\} =\frac{\pi}{2},\quad q(z_{0})= ia, a>0, $$

we have

$$ \frac{z_{0}q'(z_{0})}{q(z_{0})}=is, \quad s\geq\frac{1}{2} \biggl(a+ \frac{1}{a} \biggr)>1, $$
(3.18)

hence

$$ \mathfrak{Im} \{z_{0} \}>0 $$
(3.19)

because \(zq'(z)/q(z)\) is typically real. Therefore, (3.19) yields that

$$ \mathfrak{Im} \biggl\{ \frac{z_{0}g'(z_{0})}{g(z_{0})} \biggr\} >0, $$
(3.20)

because \(zg'(z)/g(z)\) is typically real. Moreover,

$$ \frac{1-|z_{0}|}{1+|z_{0}|}\leq\mathfrak{Re} \biggl\{ \frac {z_{0}g'(z_{0})}{g(z_{0})} \biggr\} \leq\frac{1+|z_{0}|}{1-|z_{0}|} $$
(3.21)

because \(g(z)\) is a univalent starlike function, see [10]. Applying (3.18), (3.20) and (3.21) in (3.16), we have

$$\begin{aligned} \arg \biggl\{ \frac{ z_{0}f^{(p)}(z_{0})}{g(z_{0})} \biggr\} =& \arg \bigl\{ q(z_{0}) \bigr\} +\arg \biggl\{ \frac {z_{0}g'(z_{0})}{g(z_{0})}+\frac{z_{0}q'(z_{0})}{q(z_{0})} \biggr\} \\ =& \arg \bigl\{ q(z_{0}) \bigr\} +\arg \biggl\{ \frac {z_{0}g'(z_{0})}{g(z_{0})}+is \biggr\} \\ \geq& \frac{\pi}{2}+\tan^{-1}\frac{1-|z_{0}|}{1+|z_{0}|}. \end{aligned}$$

This contradicts (3.13), and for the case

$$ {\arg} \bigl\{ q(z_{0}) \bigr\} =-\frac{\pi}{2}, $$

applying the same method as above, we have

$$\begin{aligned} \arg \biggl\{ \frac{ z_{0}f^{(p)}(z_{0})}{g(z_{0})} \biggr\} =& \arg \bigl\{ q(z_{0}) \bigr\} +\arg \biggl\{ \frac {z_{0}g'(z_{0})}{g(z_{0})}+\frac{z_{0}q'(z_{0})}{q(z_{0})} \biggr\} \\ \leq& - \biggl\{ \frac{\pi}{2}+\tan^{-1}\frac{1-|z_{0}|}{1+|z_{0}|} \biggr\} . \end{aligned}$$

This also contradicts (3.13) and, therefore, it shows that (3.15) holds. □

Corollary 3.4

Let \(f(z)\in\mathcal{A}_{p}\) and all the coefficients of \(f(z)\) are real and suppose that

$$ \biggl\vert \arg \biggl\{ \frac{ zf^{(p)}(z)}{g(z)} \biggr\} \biggr\vert < \frac{\pi}{2}+ \tan^{-1}\frac{1-|z|}{1+|z|},\quad z\in\mathbb{D}, $$

where \(g(z)\) is univalent starlike and typically real in \(\mathbb{D}\). Then we have

$$ \biggl\vert \arg\frac{ zf^{(p-1)}(z)}{g(z)} \biggr\vert < \frac{\pi}{2},\quad z\in \mathbb{D}. $$

References

  1. Noshiro, K: On the theory of Schlicht functions. J. Fac. Sci., Hokkaido Univ., Ser. 1 2(3), 129-135 (1934)

    Google Scholar 

  2. Warschawski, S: On the higher derivatives at the boundary in conformal mapping. Trans. Am. Math. Soc. 38, 310-340 (1935)

    Article  MathSciNet  Google Scholar 

  3. Ozaki, S: On the theory of multivalent functions. Sci. Rep. Tokyo Bunrika Daigaku, Sect. A 2, 167-188 (1935)

    Google Scholar 

  4. Nunokawa, M: A note on multivalent functions. Tsukuba J. Math. 13(2), 453-455 (1989)

    MATH  MathSciNet  Google Scholar 

  5. Ozaki, S, Ono, I, Umezawa, T: On a general second order derivative. Sci. Rep. Tokyo Bunrika Daigaku, Sect. A 124(5), 111-114 (1956)

    MathSciNet  Google Scholar 

  6. Nunokawa, M: On the order of strongly starlikeness of strongly convex functions. Proc. Jpn. Acad., Ser. A, Math. Sci. 69(7), 234-237 (1993)

    Article  MATH  MathSciNet  Google Scholar 

  7. Nunokawa, M: On the theory of multivalent functions. Tsukuba J. Math. 11(2), 273-286 (1987)

    MATH  MathSciNet  Google Scholar 

  8. Krantz, S, Stević, S: On the iterated logarithmic Bloch space on the unit ball. Nonlinear Anal. TMA 71, 1772-1795 (2009)

    Article  MATH  Google Scholar 

  9. Stević, S: On an integral-type operator from logarithmic Bloch-type and mixed-norm spaces to Bloch-type spaces. Nonlinear Anal. TMA 71, 6323-6342 (2009)

    Article  MATH  Google Scholar 

  10. Goodman, AW: Univalent Functions, Vols. I and II. Mariner Publishing Co., Tampa (1983)

    Google Scholar 

Download references

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Janusz Sokół.

Additional information

Competing interests

The authors declare that they have no competing interests.

Authors’ contributions

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

Rights and permissions

Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Nunokawa, M., Sokół, J. On some geometric properties of multivalent functions. J Inequal Appl 2015, 300 (2015). https://doi.org/10.1186/s13660-015-0818-x

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/s13660-015-0818-x

MSC

Keywords