Abstract
The main purpose of this paper is to investigate characterizations of composition operators on Bloch and Hardy type spaces. Initially, we use general doubling weights to study the composition operators from harmonic Bloch type spaces on the unit disc \({\mathbb {D}}\) to pluriharmonic Hardy spaces on the Euclidean unit ball \({\mathbb {B}}^n\). Furthermore, we develop some new methods to study the composition operators from harmonic Bloch type spaces on \({\mathbb {D}}\) to pluriharmonic Bloch type spaces on \({\mathbb {D}}\). Additionally, some application to new characterizations of the composition operators between pluriharmonic Lipschitz type spaces to be bounded or compact will be presented. The obtained results of this paper provide the improvements and extensions of the corresponding known results.
Similar content being viewed by others
Avoid common mistakes on your manuscript.
1 Introduction
The study of composition operators on various Banach spaces of holomorphic functions and planar harmonic functions is currently a very active field of complex and functional analysis (see [1, 6, 7, 13, 15, 19, 20, 28, 32, 37]). This paper continues the study of previous work of authors [6, 7] and is mainly motivated by the articles of Kwon [19], Pavlović [25], Wulan et al. [35] and Zhao [36]. First, we use general doubling weights to study the composition operators from harmonic Bloch type spaces on the unit disc \({\mathbb {D}}\) to pluriharmonic Hardy spaces on the Euclidean unit ball \({\mathbb {B}}^n\). In addition, we develop some new methods to study the composition operators from harmonic Bloch type spaces on \({\mathbb {D}}\) to pluriharmonic Bloch type spaces on \({\mathbb {D}}\). At last, some application to new characterizations of the composition operators between pluriharmonic Lipschitz type spaces to be bounded or compact will be given. The obtained results of this paper provide the improvements and extensions of the corresponding known results. In particular, we improve and extend the main results of Chen et al. [7] and Kwon [19], and we also establish a completely different characterization from Pavlović [25]. In order to state our main results, we need to recall some basic definitions and introduce some necessary terminologies.
Let \({\mathbb {C}}^{n}\) be the complex space of dimension n, and let \({\mathbb {C}}:={\mathbb {C}}^{1}\) be the complex plane, where n is a positive integer. For \(z=(z_{1},\ldots ,z_{n})\in {\mathbb {C}}^{n}\) and \(w=(w_{1},\ldots ,w_{n})\in {\mathbb {C}}^{n} \), we write \(\langle z,w\rangle := \sum _{k=1}^nz_k{\overline{w}}_k\) and \( |z|:={\langle z,z\rangle }^{1/2}.\) For \(a\in {\mathbb {C}}^n\), we set \({\mathbb {B}}^n(a, r)=\{z\in {\mathbb {C}}^{n}:\, |z-a|<r\}. \) In particular, let \({\mathbb {B}}^n:={\mathbb {B}}^n(0, 1)\) and \({\mathbb {D}}:={\mathbb {B}}^1\).
A twice continuously differentiable complex-valued function f defined on a domain \(\Omega \subset {\mathbb {C}}^{n}\) is called a pluriharmonic function if for each fixed \(z\in \Omega \) and \(\theta \in \partial {\mathbb {B}}^{n}\), the function \(f(z+\zeta \theta )\) is harmonic in \(\{\zeta : |\zeta |< d_{\Omega }(z)\}\), where \(d_{\Omega }(z)\) is the distance from z to the boundary \(\partial \Omega \) of \(\Omega \). Recently, pluriharmonic functions have been widely studied (cf. [4, 8, 10, 16, 29,30,31, 34]). If \(\Omega \subset {\mathbb {C}}^{n}\) is a simply connected domain, then a function \(f:\Omega \rightarrow {\mathbb {C}}\) is pluriharmonic if and only if f has a decomposition \(f=h+{\overline{g}},\) where h and g are holomorphic in \(\Omega \) (see [34]). This decomposition is unique up to an additive constant. From this decomposition, it is easy to know that the class of pluriharmonic functions is broader than that of holomorphic functions. Furthermore, a twice continuously differentiable real-valued function in a simply connected domain \(\Omega \) is pluriharmonic if and only if it is the real part of some holomorphic function on \(\Omega \). Obviously, all pluriharmonic functions are harmonic. In particular, if \(n=1\), then the converse holds (cf. [9]). Throughout this paper, we use \({\mathscr {H}}(\Omega )\) and \(\mathscr{P}\mathscr{H}(\Omega )\) to denote the set of all holomorphic functions of a domain \(\Omega \subset {\mathbb {C}}^{n}\) into \({\mathbb {C}}\) and that of all pluriharmonic functions of \(\Omega \) into \({\mathbb {C}}\), respectively.
1.1 Hardy Spaces
For \(p\in (0,\infty ]\), the pluriharmonic Hardy space \(\mathscr{P}\mathscr{H}^{p}({\mathbb {B}}^{n})\) consists of all those functions \(f\in \mathscr{P}\mathscr{H}({\mathbb {B}}^{n})\) such that, for \(p\in (0,\infty )\),
and, for \(p=\infty \),
where
and \(d\sigma \) denotes the normalized Lebesgue surface measure on \(\partial {\mathbb {B}}^{n}\). In particular, we use \({\mathscr {H}}^{p}({\mathbb {B}}^{n}):={\mathscr {H}}({\mathbb {B}}^{n})\cap \mathscr{P}\mathscr{H}^{p}({\mathbb {B}}^{n})\) to denote the holomorphic Hardy space. If \(f\in \mathscr{P}\mathscr{H}^{p}({\mathbb {B}}^{n})\) for some \(p\in (1,\infty )\), then the radial limits
exist for almost every \(\zeta \in \partial {\mathbb {B}}^{n}\) (see [3, Theorems 6.7, 6.13 and 6.39]). Moreover, if \(p\in [1,\infty )\), then \(\mathscr{P}\mathscr{H}^{p}({\mathbb {B}}^{n})\) is a normed space with respect to the norm \(\Vert \cdot \Vert _p\) and if \(p\in (0,1)\), then \(\mathscr{P}\mathscr{H}^{p}({\mathbb {B}}^{n})\) is a metric space with respect to the metric \(d(f,g)=\Vert f-g\Vert _p^p\).
1.2 Bloch Type Spaces
For a pluriharmonic function f of \({\mathbb {B}}^{n}\) into \({\mathbb {C}}\), let
and
for \(z=(z_{1},\ldots ,z_{n})\in {\mathbb {B}}^{n}\).
A continuous non-decreasing function \(\omega :[0,1)\rightarrow (0,\infty )\) is called a weight if \(\omega \) is unbounded (see [1]). Moreover, a weight \(\omega \) is called doubling if there is a constant \(C>1\) such that
for \(s\in (0,1]\).
For a weight \(\omega \), we use \({\mathscr {B}}_{\omega }({\mathbb {B}}^n)\) to denote the pluriharmonic Bloch type space consisting of all complex-valued pluriharmonic functions defined in \({\mathbb {B}}^n\) with the norm
where \({\mathscr {B}}^{f}_{\omega }(z)=\Lambda _{f}(z)/\omega (|z|).\) It is easy to know that \({\mathscr {B}}_{\omega }({\mathbb {B}}^n)\) is a complex Banach space. Furthermore, let
be the semi-norm. If \(f\in {\mathscr {B}}_{\omega }({\mathbb {B}}^n)\), then we call f a pluriharmonic Bloch function.
1.3 Composition Operators
Given a holomorphic function \(\phi \) of \({\mathbb {B}}^{n}\) into \({\mathbb {D}}\), the composition operator \(C_{\phi }:\mathscr{P}\mathscr{H}({\mathbb {D}})\rightarrow \mathscr{P}\mathscr{H}({\mathbb {B}}^{n})\) is defined by
where \(f\in \mathscr{P}\mathscr{H}({\mathbb {D}})\).
Shapiro [32] gave a complete characterization of compact composition operators on \({\mathscr {H}}^{2}({\mathbb {D}})\), with a number of interesting consequences for peak sets, essential norm of composition operators, and so on. Recently, the studies of composition operators on holomorphic function spaces have been attracted much attention of many mathematicians (see [1, 6, 7, 13, 15, 19, 20, 28, 37]). In particular, Kwon [19] investigated some characterizations of composition operators from the holomorphic Bloch spaces to the holomorphic Hardy spaces to be bounded or compact. Let us recall the main result in [19] as follows.
For \(\zeta \in \partial {\mathbb {B}}^{n}\) and \(\alpha \in (1,\infty )\), we use \(D_{\alpha }^{n}(\zeta )\) to denote the Koranyi approach domain defined by
For \(\phi :{\mathbb {B}}^{n}\rightarrow {\mathbb {D}}\) holomorphic, let
be the maximal function (see [17]). Let \(\text{ Aut }({\mathbb {B}}^n)\) denote the group of holomorphic automorphisms of \({\mathbb {B}}^n\).
Theorem A
[19, Theorems 5.1 and 5.10]. Let \(p\in (0,\infty )\), \(1<\alpha ,\beta <\infty \) and \(\omega (t)=1/(1-t^{2})\) for \(t\in [0,1)\). If \(\phi :{\mathbb {B}}^{n}\rightarrow {\mathbb {D}}\) is holomorphic, then the followings are equivalent:
-
(1)
\(\sup _{r\in (0,1)}\int _{\partial {\mathbb {B}}^{n}}\left( \frac{1}{2}\log \frac{1+|\phi (r\zeta )|}{1-|\phi (r\zeta )|}\right) ^{\frac{p}{2}}d\sigma (\zeta )<\infty \);
-
(2)
\(\int _{\partial {\mathbb {B}}^{n}}\left( M_{\beta }\phi (\zeta )\right) ^{\frac{p}{2}}d\sigma (\zeta )<\infty \);
-
(3)
\(\int _{\partial {\mathbb {B}}^{n}}\left( \int _{0}^{1}\frac{|\nabla (\phi \circ \varphi _{r\zeta })(0)|^{2}}{(1-|\phi (r\zeta )|^{2})^{2}} \frac{dr}{1-r}\right) ^{\frac{p}{2}}d\sigma (\zeta )<\infty ,\) where \(\varphi _{r\zeta }\in \text{ Aut }({\mathbb {B}}^{n})\) with \(\varphi _{r\zeta }(0)=r\zeta \);
-
(4)
\(\int _{\partial {\mathbb {B}}^{n}}\left( \int _{D_{\alpha }^{n}(\zeta )} \frac{|\nabla (\phi \circ \varphi _{z})(0)|^{2}}{(1-|\phi (z)|^{2})^{2}} \frac{dV(z)}{(1-|z|^{2})^{n+1}}\right) ^{\frac{p}{2}}d\sigma (\zeta )<\infty ,\) where dV denotes the Lebesgue volume measure of \({\mathbb {C}}^{n}\), and \(\varphi _{z}\in \text{ Aut }({\mathbb {B}}^{n})\) with \(\varphi _{z}(0)=z\);
-
(5)
\(C_{\phi }:{\mathscr {B}}_{\omega }({\mathbb {D}})\cap {\mathscr {H}}({\mathbb {D}})\rightarrow {\mathscr {H}}^{p}({\mathbb {B}}^{n})\) is a bounded operator;
-
(6)
\(C_{\phi }:{\mathscr {B}}_{\omega }({\mathbb {D}})\cap {\mathscr {H}}({\mathbb {D}})\rightarrow {\mathscr {H}}^{p}({\mathbb {B}}^{n})\) is compact.
Also, the study of composition operators and linear operators on holomorphic function spaces by using weight has aroused great interest of many mathematicians (cf. [7, 13, 22, 27]). However, there are few literatures on the theory of composition operators of harmonic functions. In the following, by using general doubling weights, we will establish the characterizations of composition operators from harmonic Bloch type spaces on \({\mathbb {D}}\) to pluriharmonic Hardy spaces on \({\mathbb {B}}^n\) to be bounded or compact.
Theorem 1.1
Let \(p\in (0,\infty )\), \(\alpha \in (1,\infty )\) and \(\omega \) be a doubling function. If \(\phi :{\mathbb {B}}^{n}\rightarrow {\mathbb {D}}\) is holomorphic, then the followings are equivalent:
-
(1)
\(C_{\phi }:{\mathscr {B}}_{\omega }({\mathbb {D}})\rightarrow \mathscr{P}\mathscr{H}^{p}({\mathbb {B}}^{n})\) is a bounded operator;
-
(2)
\( \int _{\partial {\mathbb {B}}^{n}}\left( \int _{0}^{1}|\nabla \phi (r\zeta )|^{2}\omega ^{2}(|\phi (r\zeta )|)(1-r) dr\right) ^{\frac{p}{2}}\,d\sigma (\zeta )<\infty ; \)
-
(3)
\( \int _{\partial {\mathbb {B}}^{n}}\left( \int _{D_{\alpha }^{n}(\zeta )}|\nabla \phi (z)|^{2}\omega ^{2}(|\phi (z)|)(1-|z|)^{1-n} dV(z)\right) ^{\frac{p}{2}}\,d\sigma (\zeta )<\infty ;\)
-
(4)
\(C_{\phi }:{\mathscr {B}}_{\omega }({\mathbb {D}})\rightarrow \mathscr{P}\mathscr{H}^{p}({\mathbb {B}}^{n})\) is compact.
If we take \(\omega (t)=1/(1-t^{2})\) for \(t\in [0,1)\) in Theorem 1.1, then we extend Theorem A into the following form.
Corollary 1.2
Let \(p\in (0,\infty )\) and \(1<\alpha ,\beta <\infty \). If \(\phi :{\mathbb {B}}^{n}\rightarrow {\mathbb {D}}\) is holomorphic, then the followings are equivalent:
-
(1)
\(\sup _{r\in (0,1)}\int _{\partial {\mathbb {B}}^{n}}\left( \frac{1}{2}\log \frac{1+|\phi (r\zeta )|}{1-|\phi (r\zeta )|}\right) ^{\frac{p}{2}}d\sigma (\zeta )<\infty \);
-
(2)
\(\int _{\partial {\mathbb {B}}^{n}}\left( M_{\beta }\phi (\zeta )\right) ^{\frac{p}{2}}d\sigma (\zeta )<\infty \);
-
(3)
\(\int _{\partial {\mathbb {B}}^{n}}\left( \int _{0}^{1}\frac{|\nabla (\phi \circ \varphi _{r\zeta })(0)|^{2}}{(1-|\phi (r\zeta )|^{2})^{2}} \frac{dr}{1-r}\right) ^{\frac{p}{2}}d\sigma (\zeta )<\infty ,\) where \(\varphi _{r\zeta }\in \text{ Aut }({\mathbb {B}}^{n})\) with \(\varphi _{r\zeta }(0)=r\zeta \);
-
(4)
\(\int _{\partial {\mathbb {B}}^{n}}\left( \int _{D_{\alpha }^{n}(\zeta )} \frac{|\nabla (\phi \circ \varphi _{z})(0)|^{2}}{(1-|\phi (z)|^{2})^{2}} \frac{dV(z)}{(1-|z|^{2})^{n+1}}\right) ^{\frac{p}{2}}d\sigma (\zeta )<\infty ,\) where \(\varphi _{z}\in \text{ Aut }({\mathbb {B}}^{n})\) with \(\varphi _{z}(0)=z\);
-
(5)
\( \int _{\partial {\mathbb {B}}^{n}}\left( \int _{0}^{1}|\nabla \phi (r\zeta )|^{2}\frac{(1-r)}{(1-|\phi (r\zeta )|^{2})^{2}} dr\right) ^{\frac{p}{2}}\,d\sigma (\zeta )<\infty ; \)
-
(6)
\( \int _{\partial {\mathbb {B}}^{n}}\left( \int _{D_{\alpha }^{n}(\zeta )}\frac{|\nabla \phi (z)|^{2}}{(1-|\phi (z)|^{2})^{2}}(1-|z|)^{1-n} dV(z)\right) ^{\frac{p}{2}}\,d\sigma (\zeta )<\infty ; \)
-
(7)
\(C_{\phi }:{\mathscr {B}}_{\omega }({\mathbb {D}})\rightarrow \mathscr{P}\mathscr{H}^{p}({\mathbb {B}}^{n})\) is a bounded operator;
-
(8)
\(C_{\phi }:{\mathscr {B}}_{\omega }({\mathbb {D}})\rightarrow \mathscr{P}\mathscr{H}^{p}({\mathbb {B}}^{n})\) is compact.
In the case \(\phi \) maps \({\mathbb {D}}\) into \({\mathbb {D}}\), Chen et al. [6] proved the following result.
Theorem B
[6, Theorem 6]. Let \(\omega (t)=1/\left( (1-t^{2})^{\alpha }\left( \log \frac{e}{1-t^{2}}\right) ^{\beta }\right) \) and \(\phi :{\mathbb {D}}\rightarrow {\mathbb {D}}\) be an analytic function, where \(\alpha \in (0,\infty )\) and \(\beta \le \alpha \). Then the followings are equivalent:
-
(1)
\(C_{\phi }:{\mathscr {B}}_{\omega }({\mathbb {D}})\cap {\mathscr {H}}({\mathbb {D}})\rightarrow \mathscr{P}\mathscr{H}^{2}({\mathbb {D}})\) is a bounded operator;
-
(2)
\(\frac{1}{2\pi }\int _{0}^{2\pi }\int _{0}^{1} \frac{|\phi '(re^{i\theta })|^{2}}{(1-|\phi (re^{i\theta })|)^{2\alpha }\left( \log \frac{e}{1-|\phi (re^{i\theta })|}\right) ^{2\beta }} (1-r)\,dr\,d\theta <\infty . \)
A continuous increasing function \(\psi :[0,\infty )\rightarrow [0,\infty )\) with \(\psi (0)=0\) is called a majorant if \(\psi (t)/t\) is non-increasing for \(t>0\) (see [11, 24]). By using a special doubling function \(\omega (t)=1/\psi \left( (1-t^{2})^{\alpha }\left( \log \frac{e}{1-t^{2}}\right) ^{\beta }\right) \) for \(t\in [0,1)\), the characterization of composition operators from \({\mathscr {B}}_{\omega }({\mathbb {D}})\) to \(\mathscr{P}\mathscr{H}^{p}({\mathbb {D}})\) to be bounded or compact was established in [7] as follows, which is the improvement of Theorem B, where \(p\in (0,\infty )\), \(\alpha \in (0,\infty )\) and \(\beta \in (-\infty ,\alpha ]\) are constants.
Theorem C
[7, Theorem 2.4]. Let \(p\in (0,\infty )\), \(\alpha \in (0,\infty )\), \(\beta \in (-\infty ,\alpha ]\) and \(\omega (t)=1/\psi \left( (1-t^{2})^{\alpha }\left( \log \frac{e}{1-t^{2}}\right) ^{\beta }\right) \) for \(t\in [0,1)\), where \(\psi \) is a majorant. If \(\phi :{\mathbb {D}}\rightarrow {\mathbb {D}}\) is a holomorphic function, then the followings are equivalent:
-
(1)
\(C_{\phi }:{\mathscr {B}}_{\omega }({\mathbb {D}})\rightarrow \mathscr{P}\mathscr{H}^{p}({\mathbb {D}})\) is a bounded operator;
-
(2)
\(\int _{0}^{2\pi }\left( \int _{0}^{1}|\phi '(re^{i\theta })|^{2}\omega ^{2}(|\phi (re^{i\theta })|)(1-r) dr\right) ^{\frac{p}{2}}\frac{d\theta }{2\pi }<\infty ;\)
-
(3)
\(C_{\phi }:{\mathscr {B}}_{\omega }({\mathbb {D}})\rightarrow \mathscr{P}\mathscr{H}^{p}({\mathbb {D}})\) is compact.
In the following, by using a general doubling weight, we give the characterizations of composition operators from harmonic Bloch type spaces on \({\mathbb {D}}\) to harmonic Hardy spaces on \({\mathbb {D}}\) to be bounded or compact, which is an improvement of Theorem C.
Also, the characterizations (3), (4) and (5) are new.
Theorem 1.3
Let \(p\in (0,\infty )\), \(\omega \) be a doubling function and \(\phi : {\mathbb {D}}\rightarrow {\mathbb {D}}\) be a holomorphic function. Then the followings are equivalent:
-
(1)
\(C_{\phi }:{\mathscr {B}}_{\omega }({\mathbb {D}})\rightarrow \mathscr{P}\mathscr{H}^{p}({\mathbb {D}})\) is a bounded operator;
-
(2)
\(\int _{0}^{2\pi }\left( \int _{0}^{1}|\phi '(re^{i\theta })|^{2}\omega ^{2}(|\phi (re^{i\theta })|)(1-r) dr\right) ^{\frac{p}{2}}\frac{d\theta }{2\pi }<\infty ;\)
-
(3)
\(\int _{0}^{2\pi }\left( \sum _{k=0}^{\infty }2^{-2k}|\phi '(r_{k}e^{i\theta })|^{2} \omega ^{2}(|\phi (r_{k}e^{i\theta })|)\right) ^{\frac{p}{2}}\frac{d\theta }{2\pi }<\infty ,\) where \(r_{k}=1-2^{-k}\);
-
(4)
\(\int _{0}^{2\pi }\left( \int _{0}^{1}(1-r)\sup _{0<\rho<r} \left( |\phi '(\rho \,e^{i\theta })|^{2}\omega ^{2}(|\phi (\rho \,e^{i\theta })|)\right) dr\right) ^{\frac{p}{2}}\frac{d\theta }{2\pi }<\infty ;\)
-
(5)
\( \int _{0}^{2\pi }\left( \int _{D_{\alpha }^{1}(\zeta )}|\nabla \phi (z)|^{2}\omega ^{2}(|\phi (z)|) dA(z)\right) ^{\frac{p}{2}}\,\frac{d\theta }{2\pi }<\infty ,\) where dA denotes the Lebesgue area measure of \({\mathbb {C}}\);
-
(6)
\(C_{\phi }:{\mathscr {B}}_{\omega }({\mathbb {D}})\rightarrow \mathscr{P}\mathscr{H}^{p}({\mathbb {D}})\) is compact.
Recently, the studies of composition operators between the classical analytic Bloch spaces have attracted much attention of many mathematicians (cf. [20,21,22, 35, 36]). In particular, Zhao [36] gave characterizations of composition operators from the analytic \(\alpha \)-Bloch space \({\mathscr {B}}_{\omega _{1}}({\mathbb {D}})\cap {\mathscr {H}}({\mathbb {D}})\) to the analytic \(\beta \)-Bloch space \({\mathscr {B}}_{\omega _{2}}({\mathbb {D}})\cap {\mathscr {H}}({\mathbb {D}})\) to be bounded or compact, where \(\alpha ,\beta \in (0,\infty )\), \(\omega _{1}(t)=1/(1-t^2)^{\alpha }\) and \(\omega _{2}(t)=1/(1-t^2)^{\beta }\) for \(t\in [0,1)\). Pavlović [25] gave some derivative-free characterizations of bounded composition operators between analytic Lipschitz spaces. In the following, we will develop some new methods to give new characterizations of the composition operators between the Bloch type spaces with weights to be bounded or compact, and give some application to new characterizations of the composition operators between the Lipschitz spaces to be bounded or compact.
For \(k\in \{1,2,\ldots \}\) and a weight \(\omega \), let
where \(\mu _{\omega ,k}(x)=x^{k-1}/\omega (x), x\in [0,1)\).
Proposition 1.4
Let \(r_1=0\in {\mathscr {E}}_{\omega }(1)\) and let \(r_k \in {\mathscr {E}}_{\omega }(k)\), \(k\in \{2,3,\ldots \}\), be arbitrarily chosen. Then \(\{ r_{k}\}\) is a non-decreasing sequence.
Theorem 1.5
For \(k\in \{1,2,\ldots \}\), suppose that \(\omega _{1}\) is a weight so that a point \(r_{k}\) can be selected from each set \({\mathscr {E}}_{\omega _{1}}(k)\) with \(r_1=0\) and
Let \(\omega _{2}\) be a weight, and let \(\phi \) be a holomorphic function of \({\mathbb {B}}^n\) into \({\mathbb {D}}\). Then
-
(1)
\(C_{\phi }: {\mathscr {B}}_{\omega _{1}}({\mathbb {D}})\rightarrow {\mathscr {B}}_{\omega _{2}}({\mathbb {B}}^n)\) is bounded if and only if
$$\begin{aligned} \sup _{k\ge 1}\frac{\omega _{1}(r_{k})}{k}\Vert \phi ^{k}\Vert _{{\mathscr {B}}_{\omega _{2}}({\mathbb {B}}^n)}<\infty . \end{aligned}$$(1.1) -
(2)
\(C_{\phi }: {\mathscr {B}}_{\omega _{1}}({\mathbb {D}})\rightarrow {\mathscr {B}}_{\omega _{2}}({\mathbb {B}}^n)\) is compact if and only if
$$\begin{aligned} \lim _{k\rightarrow \infty }\frac{\omega _{1}(r_{k})}{k}\Vert \phi ^{k}\Vert _{{\mathscr {B}}_{\omega _{2}}({\mathbb {B}}^n)}=0.\end{aligned}$$(1.2)
For \(\alpha \in (0,1]\), the Lipschitz space \({\mathscr {L}}_{\alpha }({\mathbb {B}}^n)\) consists of all those functions \(f\in {\mathcal {C}}({\mathbb {B}}^n)\) satisfying
where \({\mathcal {C}}({\mathbb {B}}^n)\) is a set of all continuous functions defined in \({\mathbb {B}}^n\).
By using Theorem 1.5, we give a characterization of the composition operators between pluriharmonic Lipschitz type spaces to be bounded or compact which is completely different from Pavlović [25] as follows.
Theorem 1.6
Suppose that \(k\in \{1,2,\ldots \}\), \(\alpha ,\beta \in (0,1)\) and \(\omega (t)=1/(1-t)^{1-\beta }\) for \(t\in [0,1)\). Let \(\phi \) be a holomorphic function of \({\mathbb {B}}^n\) into \({\mathbb {D}}\). Then,
-
(1)
\(C_{\phi }: {\mathscr {L}}_{\alpha }({\mathbb {D}})\cap \mathscr{P}\mathscr{H}({\mathbb {D}})\rightarrow {\mathscr {L}}_{\beta }({\mathbb {B}}^n)\cap \mathscr{P}\mathscr{H}({\mathbb {B}}^n)\) is bounded if and only if
$$\begin{aligned} \sup _{k\ge 1}\left\{ k^{-\alpha }\Vert \phi ^{k}\Vert _{{\mathscr {B}}_{\omega }({\mathbb {B}}^n)}\right\} <\infty ; \end{aligned}$$ -
(2)
\(C_{\phi }:{\mathscr {L}}_{\alpha }({\mathbb {D}})\cap \mathscr{P}\mathscr{H}({\mathbb {D}})\rightarrow {\mathscr {L}}_{\beta }({\mathbb {B}}^n)\cap \mathscr{P}\mathscr{H}({\mathbb {B}}^n)\) is compact if and only if
$$\begin{aligned} \lim _{k\rightarrow \infty }\left\{ k^{-\alpha }\Vert \phi ^{k}\Vert _{{\mathscr {B}}_{\omega }({\mathbb {B}}^n)}\right\} =0. \end{aligned}$$
The proofs of Theorems 1.1–1.6 and Proposition 1.4 will be presented in Sect. 2.
2 The Proofs of the Main Results
Denote by \(L^{p}(\partial {\mathbb {B}}^{n})\) \((p\in (0,\infty ))\) the set of all measurable functions F of \(\partial {\mathbb {B}}^{n}\) into \({\mathbb {C}}\) with
Given \(f\in {\mathscr {H}}^{p}({\mathbb {B}}^{n})\), the Littlewood–Paley type \({\mathscr {G}}\)-function is defined as follows
Then
for \(p\in (0,\infty )\) (see [2, 18, 33]). The conclusion of (2.1) also can be rewritten in the following form. There exists a positive constant C, depending only on p, such that
for \(p\in (0,\infty )\) (see [2, 18, 33]). For \(\alpha \in (1,\infty )\) and \(p\in (0,\infty )\), it follows from [2, Theorem 3.1] (or [18, Theorem 1.1]) that there is a positive constant C such that
where \(f\in {\mathscr {H}}^{p}({\mathbb {B}}^{n})\) and
It follows from (2.2) and (2.3) that there is a positive constant C such that
The following result easily follows from [1, Lemma 1] and [1, Theorem 2].
Lemma 2.1
Let \(\omega \) be a doubling function. Then there exist functions \(f_{j}\in {\mathscr {B}}_{\omega }({\mathbb {D}})\cap {\mathscr {H}}({\mathbb {D}})\) \((j\in \{1,2\})\) such that, for \(z\in {\mathbb {D}}\),
The following result is well-known.
Lemma D
cf. [6, Lemma 5]. Suppose that \(a,b\in [0,\infty )\) and \(q\in (0,\infty )\). Then
2.1 The Proof of Theorem 1.1
We first prove \((1)\Rightarrow (2)\). By Lemma 2.1, there exist functions \(f_{j}\in {\mathscr {B}}_{\omega }({\mathbb {D}})\cap {\mathscr {H}}({\mathbb {D}})\) \((j\in \{1,2\})\) such that, for \(z\in {\mathbb {D}}\),
Since \(C_{\phi }:{\mathscr {B}}_{\omega }({\mathbb {D}})\rightarrow \mathscr{P}\mathscr{H}^{p}({\mathbb {B}}^{n})\) is a bounded operator, by (2.1), we see that
for \(j\in \{1,2\}\). It follows from (2.5), (2.6) and Lemma D that
where \({\mathscr {M}}_{p}=2^{-p/2-\max \{p/2-1,0\}}\).
Next, we prove \((2)\Rightarrow (1)\). We split the proof of this case into two steps.
Step 1. We first prove \(C_{\phi }(f)\in \mathscr{P}\mathscr{H}^{p}({\mathbb {B}}^{n})\). Since \({\mathbb {D}}\) is a simply connected domain, we see that f admits the canonical decomposition \(f=f_{1}+\overline{f_{2}}\), where \(f_{1}\) and \(f_{2}\) are analytic in \({\mathbb {D}}\) with \(f_{2}(0)=0\). For \(j\in \{1,2\}\), elementary calculations lead to
For \(j\in \{1,2\}\), let
Consequently, there is a constant \(C=\Vert f\Vert _{{\mathscr {B}}_{\omega }({\mathbb {D}})}^p\) such that
which, together with (2.1), implies that \(C_{\phi }(f_{j})\in {\mathscr {H}}^{p}({\mathbb {B}}^{n})\) for \(j\in \{1,2\}\). Hence \(C_{\phi }(f)\in \mathscr{P}\mathscr{H}^{p}({\mathbb {B}}^{n})\).
Step 2. In this step, we will show that \(C_{\phi }\) is a bounded operator. Without loss of generality, we assume that \(\Vert f\Vert _{{\mathscr {B}}_{\omega }({\mathbb {D}})}\ne 0\), and \(f_{j}\) are not constant functions for \(j\in \{1,2\}\). Since, for \(j\in \{1,2\}\),
by (2.2) and (2.8), we see that there is a positive constant C such that
where
Combining (2.10) and Lemma D gives
Therefore, \(C_{\phi }:{\mathscr {B}}_{\omega }({\mathbb {D}})\rightarrow \mathscr{P}\mathscr{H}^{p}({\mathbb {B}}^{n})\) is a bounded operator.
Now, we prove \((1)\Rightarrow (3)\). By Lemma 2.1, there exist functions \(f_{j}\in {\mathscr {B}}_{\omega }({\mathbb {D}})\cap {\mathscr {H}}({\mathbb {D}})\) \((j\in \{1,2\})\) such that, for \(z\in {\mathbb {D}}\),
Since \(C_{\phi }:{\mathscr {B}}_{\omega }({\mathbb {D}})\rightarrow \mathscr{P}\mathscr{H}^{p}({\mathbb {B}}^{n})\) is a bounded operator, by (2.3), we see that
for \(j\in \{1,2\}\). It follows from (2.11), (2.12) and Lemma D that
where \({\mathscr {V}}(z)=|\nabla \,\phi (z)|^{2}(1-|z|)^{1-n}\).
The proof of \((3)\Rightarrow (1)\) is similar to the proof of \((2)\Rightarrow (1)\). We only need to replace “(2.2)” and “\(\int _{\partial {\mathbb {B}}^{n}}\big ({\mathscr {G}}(C_{\phi }(f_{j}))(\zeta )\big )^{p}d\sigma (\zeta )\)” by “(2.4)” and
respectively, in the proof of \((2)\Rightarrow (1)\), where \(j\in \{1,2\}\).
At last, we prove \((1)\Leftrightarrow (4)\). Since \((4)\Rightarrow (1)\) is obvious, we only need to prove \((1)\Rightarrow (4)\). Let \(C_{\phi }:{\mathscr {B}}_{\omega }({\mathbb {D}})\rightarrow \mathscr{P}\mathscr{H}^{p}({\mathbb {B}}^{n})\) be a bounded operator. Then \(f\circ \phi \in \mathscr{P}\mathscr{H}^{p}({\mathbb {B}}^{n})\) for all \(f\in {\mathscr {B}}_{\omega }({\mathbb {D}})\), and
exists for almost every \(\zeta \in \partial {\mathbb {B}}^{n}\). Suppose that \(\{f_{k}=h_{k}+\overline{g_{k}}\}\) is a sequence of \({\mathscr {B}}_{\omega }({\mathbb {D}})\) such that \(\Vert f_{k}\Vert _{{\mathscr {B}}_{\omega }({\mathbb {D}})}\le 1\), where \(h_{k}\) and \(g_{k}\) are analytic in \({\mathbb {D}}\) with \(g_{k}(0)=0\). Next, we will show that \(\{C_{\phi }(f_{k})\}\) has a convergent subsequence in \(\mathscr{P}\mathscr{H}^{p}({\mathbb {B}}^{n})\). Since
\(\{ h_{k}\}\) and \(\{ g_k\}\) are normal families. Then there are subsequences of \(\{ h_{k}\}\) and \(\{ g_k\}\) that converge uniformly on compact subsets of \({\mathbb {D}}\) to holomorphic functions h ang g, respectively. Without loss of generality, we assume that the sequences \(\{ h_{k}\}\) and \(\{ g_{k}\}\) themselves converge to h and g, respectively. Then the sequence \(f_{k}=h_{k}+\overline{g_{k}}\) itself converges uniformly on compact subsets of \({\mathbb {D}}\) to the harmonic function \(f=h+{\overline{g}}\) with \(g(0)=0\). Consequently,
which gives that \(f\in {\mathscr {B}}_{\omega }({\mathbb {D}})\) with \(\Vert f\Vert _{{\mathscr {B}}_{\omega }({\mathbb {D}})}\le 1\). It follows from \((1)\Rightarrow (2)\) that
and
Also, we have
for \(\zeta \in \partial {\mathbb {B}}^n\).
Applying (2.13), (2.14), (2.15) and the control convergence theorem twice, we have
which, together with (2.2), implies that
Consequently,
as \(k\rightarrow \infty \). By using similar reasoning as in the proof of (2.16), we have
as \(k\rightarrow \infty \). Therefore, by (2.16) and (2.17), we conclude that \(C_{\phi }(f)=C_{\phi }(h)+\overline{C_{\phi }(g)}\in \mathscr{P}\mathscr{H}^{p}({\mathbb {B}}^{n})\), and \(C_{\phi }(f_{k})\rightarrow \,C_{\phi }(f)\) in \(\mathscr{P}\mathscr{H}^{p}({\mathbb {B}}^{n})\) as \(k\rightarrow \infty \). The proof of this theorem is finished. \(\square \)
Theorem E
[26, Theorem 1.1]. Let \(p\in (0,\infty )\) and f be a holomorphic function in \({\mathbb {D}}\). Then the followings are equivalent:
-
(1)
\(f\in {\mathscr {H}}^{p}({\mathbb {D}})\);
-
(2)
\({\mathcal {G}}[f]\in L^p({\mathbb {T}}),\)where
$$\begin{aligned} {\mathcal {G}}[f](\zeta )=\left( \int _{0}^{1}(1-r)|f'(r\zeta )|^{2}dr\right) ^{\frac{1}{2}}, \quad \zeta \in {\mathbb {T}}; \end{aligned}$$ -
(3)
\({\mathcal {G}}_*[f]\in L^p({\mathbb {T}}),\) where
$$\begin{aligned} {\mathcal {G}}_*[f](\zeta )=\left( \int _{0}^{1}(1-r)\sup _{\rho \in (0,r)}|f'(\rho \,\zeta )|^{2}dr\right) ^{\frac{1}{2}}, \quad \zeta \in {\mathbb {T}}; \end{aligned}$$ -
(4)
\({\mathcal {G}}_d[f]\in L^p({\mathbb {T}}),\) where
$$\begin{aligned} {\mathcal {G}}_d[f](\zeta )=\left( \sum _{k=0}^{\infty }2^{-2k}|f'(r_{k}\zeta )|^{2}\right) ^{\frac{1}{2}}, \quad \zeta \in {\mathbb {T}} \end{aligned}$$and \(r_{k}=1-2^{-k}\).
Furthermore, there are constants \(C_1\), \(C_2\), \(C_3\) and \(C_4\) independent of f such that
2.2 The Proof of Theorem 1.3
Since \((1)\Leftrightarrow (2)\Leftrightarrow (5)\Leftrightarrow (6)\) easily follows from Theorem 1.1, we only need to prove \((1)\Leftrightarrow (3)\Leftrightarrow (4)\). We first prove \((1)\Rightarrow (3)\). By Lemma 2.1, there are two analytic functions \(f_{j}\in {\mathscr {B}}_{\omega }({\mathbb {D}})\cap {\mathscr {H}}({\mathbb {D}})\) \((j\in \{1,2\})\) such that, for \(z\in {\mathbb {D}}\),
Since \(C_{\phi }:{\mathscr {B}}_{\omega }({\mathbb {D}})\rightarrow \mathscr{P}\mathscr{H}^{p}({\mathbb {D}})\) is a bounded operator, by Theorem E, we see that
for \(j\in \{1,2\}\). It follows from (2.18), (2.19) and Lemma D that
where \({\mathscr {M}}_{p}\) is defined in the proof of Theorem 1.1.
Next, we prove \((3)\Rightarrow (1)\). We first show \(C_{\phi }(f)\in \mathscr{P}\mathscr{H}^{p}({\mathbb {D}})\). Let \(f\in {\mathscr {B}}_{\omega }({\mathbb {D}})\). Since \({\mathbb {D}}\) is a simply connected domain, we see that f admits the canonical decomposition \(f=f_{1}+\overline{f_{2}}\), where \(f_{1}\) and \(f_{2}\) are analytic in \({\mathbb {D}}\) with \(f_{2}(0)=0\). For \(j\in \{1,2\}\), let
Then, by (2.7), we see that there is a constant \(C=\Vert f\Vert _{{\mathscr {B}}_{\omega }({\mathbb {D}})}^p\) such that
which, together with Theorem E, implies that \(C_{\phi }(f_{j})\in \mathscr{P}\mathscr{H}^{p}({\mathbb {D}})\). Hence
Now we come to prove \(C_{\phi }\) is a bounded operator. By (2.9) and Theorem E, we see that there is a positive constant C, depending only on p, such that, for \(j\in \{1,2\}\),
where
It follows from (2.20) and Lemma D that
which implies that \(C_{\phi }\) is a bounded operator.
Now we prove \((1)\Rightarrow (4)\). By Lemma 2.1, there are two analytic functions \(f_{j}\in {\mathscr {B}}_{\omega }({\mathbb {D}})\cap {\mathscr {H}}({\mathbb {D}})\) \((j\in \{1,2\})\) such that, for \(z\in {\mathbb {D}}\),
Since \(C_{\phi }\) is a bounded operator, by (2.21), Theorem E and Lemma D, we see that
where \(\delta (r)=1-r\) and \({\mathscr {M}}_{p}\) is defined in the proof of Theorem 1.1.
At last, we prove \((4)\Rightarrow (1)\). Let \(f\in {\mathscr {B}}_{\omega }({\mathbb {D}})\). It is not difficult to know that, for \(z\in {\mathbb {D}}\), f has the canonical decomposition \(f(z)=f_{1}(z)+\overline{f_{2}(z)}\), where \(f_{1}\) and \(f_{2}\) are analytic in \({\mathbb {D}}\) with \(f_{2}(0)=0\). Then, by (2.7), we see that
where \(j\in \{1,2\}\) and \(M_{j}=\Vert f_{j}\Vert _{{\mathscr {B}}_{\omega }({\mathbb {D}})}^p\). It follows from (2.22) and Theorem E that \(C_{\phi }(f_{j})\in \mathscr{P}\mathscr{H}^{p}({\mathbb {D}})~(j=1,2)\), which implies that
Next, we prove \(C_{\phi }\) is also a bounded operator. By (2.9) and Theorem E, we see that there is a positive constant C, depending only on p, such that, for \(j\in \{1,2\}\),
where
It follows from (2.23) and Lemma D that
which implies that \(C_{\phi }:{\mathscr {B}}_{\omega }({\mathbb {D}})\rightarrow \mathscr{P}\mathscr{H}^{p}({\mathbb {D}})\) is a bounded operator. The proof of this theorem is finished. \(\square \)
2.3 The Proof of Proposition 1.4
If \(k=1\), then we can take \(r_1=0\). Since, for \(k\in \{2,3,\ldots \}\),
we see that \({\mathscr {E}}_{\omega }(k)\) is not an empty set. Consequently, for \(k\in \{2,3,\ldots \}\), we can choose points \(r_{k}>0\) and \(r_{k+1}>0\) from the sets \({\mathscr {E}}_{\omega }(k)\) and \({\mathscr {E}}_{\omega }(k+1)\), respectively, such that \(\mu _{\omega ,k}(r_{k})\ge \mu _{\omega ,k}(r_{k+1})\) which is equivalent to
On the other hand, for \(k\in \{2,3,\ldots \}\), we have \(\mu _{\omega ,k+1}(r_{k+1})\ge \mu _{\omega ,k+1}(r_{k})\) which implies that
Combining (2.24) and (2.25) gives \(r_{k}\le \,r_{k+1}\). \(\square \)
2.4 The Proof of Theorem 1.5
(1) We first prove the sufficiency. By (1.1), we have
For \(f\in {\mathscr {B}}_{\omega _{1}}({\mathbb {D}})\), we have
So, it suffices to show that there exists a constant independent of f such that
We split the remaining proof into two cases.
Case 1. If
then there is a constant \(\rho _{0}\in (0,1)\) such that
For \(f\in {\mathscr {B}}_{\omega _{1}}({\mathbb {D}})\), it follows from (2.26) and (2.27) that
which implies that \(C_{\phi }:{\mathscr {B}}_{\omega _{1}}({\mathbb {D}})\rightarrow {\mathscr {B}}_{\omega _{2}}({\mathbb {B}}^n)\) is bounded.
Case 2. If
then, for \(k\in \{1,2,\ldots \}\), let
where \(r_{1}=0\). Since \(\{r_k\}\) is a non-decreasing sequence satisfying
we see that
Let m be the smallest positive integer such that \(\Omega _{m}\) is not an empty set. By (2.28) and (2.30), we see that \(\Omega _{k}\) is not empty for every integer \(k\in \{m,m+1,\ldots \}\), and \({\mathbb {B}}^n=\cup _{k=m}^{\infty }\Omega _{k}.\) Then, for \(k\in \{m,m+1,\ldots \}\), we have
It follows from \(\mu _{\omega _{1},k}(r_{k})\ge \mu _{\omega _{1},k}(r_{k+1})\) and \(\mu _{\omega _{1},k+1}(r_{k+1})\ge \mu _{\omega _{},k+1}(r_{k})\) that
which, together with (2.29) and (2.30), yields that
Combining (2.31) and (2.33) gives
Hence there is a positive integer \(m_{0}\ge \,m\) such that, for all \(k\in \{m_{0},m_{0}+1,\ldots \}\),
which implies that, for \(f\in {\mathscr {B}}_{\omega _{1}}({\mathbb {D}})\),
where
and
By (2.34), we see that
and
which imply that \(C_{\phi }:{\mathscr {B}}_{\omega _{1}}({\mathbb {D}})\rightarrow {\mathscr {B}}_{\omega _{2}}({\mathbb {B}}^n)\) is bounded.
Next, we begin to prove the necessity. For \(k\in \{2,3,\ldots \}\), let \(f(w)=w^{k},w\in {\mathbb {D}}\). Since
we see that
From (2.36), we see that there is a positive constant, independent of k, such that
which is equivalent to
For \(w\in {\mathbb {D}}\), let \(F(w)=w^{k}/\Vert w^{k}\Vert _{{\mathscr {B}}_{\omega _{1}}({\mathbb {D}})}\). Then \(\Vert F\Vert _{{\mathscr {B}}_{\omega _{1}}({\mathbb {D}})}=1\). Therefore,
which implies (1.1) is true.
(2) Assume that \(C_{\phi }\) is compact. For \(k\in \{1,2,\ldots \}\) and \(w\in {\mathbb {D}}\), let \(F_k(w)=w^{k}/\Vert w^{k}\Vert _{{\mathscr {B}}_{\omega _{1}}({\mathbb {D}})}\). Then \(\Vert F_k\Vert _{{\mathscr {B}}_{\omega _{1}}({\mathbb {D}})}=1\). For \(k\in \{2,3,\ldots \}\) and \(m\in \{1,2,\ldots \}\), it follows from (2.32) that
which implies that
For \(k\in \{2,3,\ldots \}\) and \(w\in {\mathbb {D}}\), elementary calculations lead to
which, together with (2.29), (2.30) and (2.37), yields that \(F_k\rightarrow 0\) locally uniformly in \({\mathbb {D}}\) as \(k\rightarrow \infty \). Since \(C_{\phi }\) is compact, we deduce that
Consequently, (1.2) follows from
Conversely, assume that (1.2) holds. Suppose that \(\{f_{j}=h_{j}+\overline{g_{j}}\}\) is a sequence of \({\mathscr {B}}_{\omega _1}({\mathbb {D}})\) such that \(\Vert f_{j}\Vert _{{\mathscr {B}}_{\omega _1}({\mathbb {D}})}\le 1\), where \(h_{j}\) and \(g_{j}\) are analytic in \({\mathbb {D}}\) with \(g_{j}(0)=0\). Then there are subsequences of \(\{ h_{j}\}\) and \(\{ g_{j}\}\) that converge uniformly on compact subsets of \({\mathbb {D}}\) to holomorphic functions h ang g, respectively. Without loss of generality, we assume that the sequences \(\{ h_{j}\}\) and \(\{ g_{j}\}\) themselves converge to h and g, respectively. Then the sequence \(f_{j}=h_{j}+\overline{g_{j}}\) itself converges uniformly on compact subsets of \({\mathbb {D}}\) to the harmonic function \(f=h+{\overline{g}}\) with \(g(0)=0\). Consequently,
which gives that \(f\in {\mathscr {B}}_{\omega _1}({\mathbb {D}})\) with \(\Vert f\Vert _{{\mathscr {B}}_{\omega _1}({\mathbb {D}})}\le 1\). Let \(\varepsilon >0\) be arbitrarily fixed. Since (1.2) holds, there exists a positive integer \(N>m_0\) such that, for \(k\in \{N, N+1,\ldots \}\),
which implies that, for \(j\ge 1\),
Since the sequence \(\{ \Lambda _{f_{j}-f}\}\) converges to 0 locally uniformly in \({\mathbb {D}}\), there exists an integer \(j_0\) such that, for \(j\ge j_0\),
Combining (2.35), (2.38) and (2.39) gives that, for \(j\ge j_0\),
which implies that \(C_{\phi }\) is compact. The proof of this theorem is finished. \(\square \)
2.5 The Proof of Theorem 1.6
For \(\alpha ,\beta \in (0,1)\), let \(f\in \mathscr{P}\mathscr{H}({\mathbb {B}}^n)\cap {\mathscr {L}}_{\varpi }({\mathbb {B}}^n)\), where \(\varpi =\alpha ~\text{ or }~\beta \). Since \({\mathbb {B}}^n\) is a simply connected domain, we see that f admits the canonical decomposition \(f=f_{1}+\overline{f_{2}}\), where \(f_{1}\) and \(f_{2}\) are analytic in \({\mathbb {B}}^n\) with \(f_{2}(0)=0\). Next, we prove \(f_{1},f_{2}\in {\mathscr {L}}_{\varpi }({\mathbb {B}}^n)\). Let \(f=f_{1}+\overline{f_{2}}=u+iv\), where \(f_{1}=u_{1}+iv_{1}\) and \(f_{2}=u_{2}+iv_{2}\). Then \(f\in {\mathscr {L}}_{\varpi }({\mathbb {B}}^n)\) if and only if \(u,~v\in {\mathscr {L}}_{\varpi }({\mathbb {B}}^n)\). Let \(F=f_{1}+f_{2}\) and \({\widetilde{v}}=\text{ Im }(F)\), where “\(\text{ Im }\)” is the imaginary part of a complex number. Then iF is holomorphic in \({\mathbb {B}}^n\). Since \(\text{ Re }(iF)=-{\widetilde{v}}\), where “\(\text{ Re }\)” is the real part of a complex number, and \(\text{ Im }(iF)=\text{ Im }(if)=u\), u and \({\widetilde{v}}\) satisfy \(\text{ div }(\nabla u)=\text{ div }(\nabla {\widetilde{v}})=0\) and \(|\nabla u|=|\nabla {\widetilde{v}}|\). Then by [23, Corollary 3.11], we see that
By (2.40), we have
and
Applying [23, Corollary 3.11] to \(f_{1}\) and \(f_{2}\) again, we see that
and
Consequently, \(f_{1}\in {\mathscr {L}}_{\varpi }({\mathbb {B}}^n)\) follows from (2.41) and (2.43), and \(f_{2}\in {\mathscr {L}}_{\varpi }({\mathbb {B}}^n)\) follows from (2.42) and (2.44). It follows from [31, Section 6.4] that \(f_{1},~f_{2}\in {\mathscr {B}}_{\omega }({\mathbb {B}}^n)\), where \(\omega (t)=1/(1-t)^{1-\varpi }\) for \(t\in [0,1)\). Hence \(f\in \mathscr{P}\mathscr{H}({\mathbb {B}}^n)\cap {\mathscr {L}}_{\varpi }({\mathbb {B}}^n)\) is equivalent to \(f\in {\mathscr {B}}_{\omega }({\mathbb {B}}^n)\). Hence we only need to show that \(C_{\phi }:{\mathscr {B}}_{\omega _{1}}({\mathbb {D}})\rightarrow {\mathscr {B}}_{\omega _{2}}({\mathbb {B}}^n)\) is bounded if and only if
and \(C_{\phi }\) is compact if and only if
where \(\omega _{1}(t)=1/(1-t)^{1-\alpha }\) and \(\omega _{2}(t)=1/(1-t)^{1-\beta }\) for \(t\in [0,1)\). By taking \(r_{k}=(k-1)/(k-\alpha )\) in Theorem 1.5, we can obtain the desired result. The proof of this theorem is finished. \(\square \)
Data Availability
Our manuscript has no associated data.
References
Abakumov, E., Doubtsov, E.: Reverse estimates in growth spaces. Math. Z. 271, 399–413 (2012)
Ahern, P., Bruna, J.: Maximal and area integral characterization of Hardy–Sobolev spaces in the unit ball of \({\mathbb{C} }^{n}\). Rev. Mat. Iberoam. 4, 123–153 (1988)
Axler, S., Bourdon, P., Ramey, W.: Harmonic Function Theorem. Springer, New York (2001)
Chen, S.L., Hamada, H.: Some sharp Schwarz-Pick type estimates and their applications of harmonic and pluriharmonic functions. J. Funct. Anal. 282, 109254 (2022)
Chen, S.L., Hamada, H.: On (Fejér-)Riesz type inequalities, Hardy-Littlewood type theorems and smooth moduli. Math. Z. 305, 64 (2023)
Chen, S.L., Ponnusamy, S., Rasila, A.: On characterizations of Bloch-type, Hardy-type, and Lipschitz-type spaces. Math. Z. 279, 163–183 (2015)
Chen, S.L., Hamada, H., Zhu, J.-F.: Composition operators on Bloch and Hardy type spaces. Math. Z. 301, 3939–3957 (2022)
Chen, S.L., Hamada, H., Ponnusamy, S., Vijayakumar, R.: Schwarz type lemmas and their applications in Banach spaces. J. Anal. Math. (2023). https://doi.org/10.1007/s11854-023-0293-0
Duren, P.: Harmonic Mappings in the Plane. Cambridge Univ Press, Cambridge (2004)
Duren, P., Hamada, H., Kohr, G.: Two-point distortion theorems for harmonic and pluriharmonic mappings. Trans. Am. Math. Soc. 363, 6197–6218 (2011)
Dyakonov, K.M.: Equivalent norms on Lipschitz-type spaces of holomorphic functions. Acta Math. 178, 143–167 (1997)
Frazer, H.: On the moduli of regular functions. Proc. Lond. Math. Soc. 36, 532–546 (1934)
Guo, Y.T., Zhang, X.J.: Composition operators from normal weight general function spaces to Bloch type spaces, submitted
Hardy, G.H., Littlewood, J.E.: Some properties of fractional integrals II. Math. Z. 34, 403–439 (1932)
Hosokawa, T., Ohno, S.: Differences of weighted composition operators acting from Bloch space to \(H^{\infty }\). Trans. Am. Math. Soc. 363, 5321–5340 (2011)
Izzo, A.J.: Uniform algebras generated by holomorphic and pluriharmonic functions. Trans. Am. Math. Soc. 339, 835–847 (1993)
Korányi, A., Vagi, S.: Singular integrals in homogeneous spaces and some problems of classical analysis. Ann. Scuola Normale Superiore Pisa 25, 575–648 (1971)
Krantz, S.G., Li, S.Y.: Area integral characterizations of functions in Hardy spaces on domains in \({\mathbb{C} }^{n}\). Complex Var. 32, 373–399 (1997)
Kwon, E.G.: Hyperbolic mean growth of bounded holomorphic functions in the ball. Trans. Am. Math. Soc. 355, 1269–1294 (2003)
Madigan, K., Matheson, A.: Compact composition operators on the Bloch space. Trans. Am. Math. Soc. 347, 2679–2687 (1995)
Montes-Rodríguez, A.: The essential norm of a composition operator on Bloch spaces. Pac. J. Math. 188, 339–351 (1999)
Montes-Rodríguez, A.: Weighted composition operators on weighted Banach spaces of analytic functions. J. Lond. Math. Soc. 61, 872–884 (2000)
Nolder, C.A.: Hardy-Littlewood theorems for solutions of elliptic equations in divergence form. Indiana Univ. Math. J. 40, 149–160 (1991)
Pavlović, M.: On Dyakonov’s paper Equivalent norms on Lipschitz-type spaces of holomorphic functions. Acta Math. 183, 141–143 (1999)
Pavlović, M.: Derivative-free characterizations of bounded composition operators between Lipschitz spaces. Math. Z. 258, 81–86 (2008)
Pavlović, M.: On the Littlewood–Paley \(g\)-function and Calderón’s area theorem. Expo. Math. 31, 169–195 (2013)
Peláez, J.A., Rättyä, J.: Generalized Hilbert operators on weighted Bergman spaces. Adv. Math. 240, 227–267 (2013)
Pérez-González, F., Xiao, J.: Bloch–Hardy pullbacks. Acta Sci. Math. (Szeged) 67, 709–718 (2001)
Ramey, W.: Local boundary behavior of pluriharmonic functions along curves. Am. J. Math. 108, 175–191 (1986)
Ramey, W., Ullrich, D.: The pointwise Fatou theorem and its converse for positive pluriharmonic functions. Duke Math. J. 49, 655–675 (1982)
Rudin, W.: Function Theory in \({\mathbb{C} }^{n}\). Springer, New York (1980)
Shapiro, J.H.: The essential norm of a composition operator. Ann. Math. 125, 375–404 (1987)
Stein, E.: Some problems in harmonic analysis. Proc. Symp. Pure Math. 35, 3–19 (1979)
Vladimirov, V.S.: Methods of the Theory of Functions of Several Complex Variables. M.I.T. Press, Cambridge (1966). (in Russian)
Wulan, H., Zheng, D., Zhu, K.: Compact composition operators on BMOA and the Bloch space. Proc. Am. Math. Soc. 137, 3861–3868 (2009)
Zhao, R.H.: Essential norms of composition operators between Bloch type spaces. Proc. Am. Math. Soc. 138, 2537–2546 (2010)
Zhu, K.: Operator Theory in Function Spaces. Monographs and Textbooks in Pure and Applied Mathematics, vol. 139. Marcel Dekker Inc, New York (1990)
Acknowledgements
The authors would like to thank the referee for many valuable comments.
Funding
The research of the first author was partly supported by the National Science Foundation of China (grant no. 12071116), the Hunan Provincial Natural Science Foundation of China (grant no. 2022JJ10001), the Key Projects of Hunan Provincial Department of Education (grant no. 21A0429); the Double First-Class University Project of Hunan Province (Xiangjiaotong [2018]469), the Science and Technology Plan Project of Hunan Province (2016TP1020), and the Discipline Special Research Projects of Hengyang Normal University (XKZX21002); The research of the second author was partly supported by JSPS KAKENHI Grant Number JP22K03363.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
The authors declare that they have no conflict of interest.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Chen, S., Hamada, H. Characterizations of Composition Operators on Bloch and Hardy Type Spaces. Results Math 79, 95 (2024). https://doi.org/10.1007/s00025-024-02125-3
Received:
Accepted:
Published:
DOI: https://doi.org/10.1007/s00025-024-02125-3
Keywords
- Bloch type space
- complex-valued harmonic function
- composition operator
- hardy space
- pluriharmonic functions