The purpose of this paper is to introduce and study the structure of p-tuple of \((n,m)\)-\(\mathcal{D}\)-normal operators. This is a generalization of the class of p-tuple of n-normal operators. We consider a generalization of these single variable n-\(\mathcal{D}\)-normal and \((n,m)\)-\(\mathcal{D}\)-normal operators and explore some of their basic properties.
Hinweise
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
1 Introduction
Let K be a complex Hilbert space, \({\mathbf{B}}[\mathbf{K} ]\) be the algebra of all bounded linear operators defined in K. For every N in \(\mathbf{B}[\mathbf{K}]\), denote \(\ker (N)\) as the null space and \(N^{*}\) as the adjoint of R, respectively.
The Drazin inverse of bounded linear operators on complex Banach spaces was introduced by Caradus [14] and King [26]. For more detailed study and applications of the concepts of Drazin invertibility, we invite the interested readers to refer to ([11, 12, 32]). It is well known that the Drazin inverse of the operator \(N\in{ \mathbf{B}}[\mathbf{K}]\) is the unique operator \(N^{\mathcal{D}}\in {\mathbf{B}}[\mathbf{K}]\) if it exists and satisfies the following conditions
Moreover, it was observed that if \(N \in {\mathbf{B}}_{d}[\mathbf{K}]\) and \(T \in {\mathbf{B}}[\mathbf{K}]\) is an invertible operator, then \(T^{-1}NT\in {\mathbf{B}}_{d}[\mathbf{K}]\) and \((T^{-1}NT )^{\mathcal{\mathcal{D }}}= T^{-1}N^{\mathcal{D}}T\).
Lemma 1.1
([14, 34]) Let\(N, T\in { \mathbf{B}}_{d}[\mathbf{K}]\). Then the following properties hold.
(1) NTis Drazin invertible if and only ifTNis Drazin invertible. Moreover
The success of the theory of normal operators on Hilbert spaces has led to several attempts to generalize it to large classes of operators, including normal operators.
For \(N,T \in {\mathbf{B}}[\mathbf{K}]\), we set \([N, T ]=NT-TN\). An operator \(N \in {\mathbf{B}}[\mathbf{K}]\) is called
(ii) n-normal if \([N^{n},N^{*} ]=0\) ([2, 24, 25]),
(iii) \((n,m)\)-normal if \([N^{n}, (N^{m} )^{*} ]=0\), where n, m are two nonnegative integers ([1, 3, 4]).
These concepts of normality, studied for \(N \in \mathbf{B}[\mathbf{K}]\), have been extended to the class of Drazin inverse of bounded linear operators on K as follows: For \(R\in{ \mathbf{B}}_{d}[{ \mathbf{K}}]\), R is said to be
(i) \(\mathcal{D}\)-normal if \([N^{\mathcal{D}},N^{*} ]=0 \) ([19]),
(ii) n-power \(\mathcal{D}\)-normal if \([ (N^{\mathcal{D}} )^{n},N^{*} ]=0\) ([19]),
(iii) \((n,m)\)-power \(\mathcal{D}\)-normal if \([ (N^{\mathcal{D}} )^{n}, (N^{*} )^{m} ]=0\) for some positive integers n and m ([28]),
The study of p-tuples of operators has received great interest from many authors in recent years. Some developments in this field have been presented in [7, 9, 10, 13, 15, 16, 23, 27, 29], and further references can be found therein.
Given a p-tuple \({\mathbf{\large N}} := (N_{1}, \ldots , N_{p} )\in {\mathbf{B}}[{ \mathbf{K}}]^{p}\), we define \([{\mathbf{\large N}}^{*},{\mathbf{\large N}}]\in {\mathbf{B}}[{\mathbf{K}} \oplus \cdots \oplus{\mathbf{K}}]\) as the self-commutator of N, which is given by
Recently, in [5], the author has introduced the concept of jointly n-normal tuple as follows: \({\mathbf{\large N}}=(N_{1},\ldots ,N_{p})\in { \mathbf{B}}[{ \mathbf{K}}]^{p}\) is said to be joint n-normal operators if R satisfying
Let \({\mathbf{\large N}}=(N_{1},\ldots ,N_{p})\in {\mathbf{B}}_{d}[\mathbf{K}]^{p}\). We set \({\mathbf{\large N}}^{\mathcal{D}}:= (N_{1}^{\mathcal{D}},\ldots ,N_{p}^{ \mathcal{D}} )\).
The present paper proposes and studies the concept of p-tuples of \((n, m)\)-\(\mathcal{D}\)-normal operators. These are natural generalizations of \(\mathcal{D}\)-normal, n-power \(\mathcal{D}\)-normal, and \((n, m)\)-power \(\mathcal{D}\)-normal single operators as done in [19, 28]. For more details on some classes of Drazin inverse operators, the reader is invited to consult [20, 21, 33].
This paper has been organized into two sections. In section two, we introduce the class of p-tuples of \((n, m)\)-\(\mathcal{D}\)-normal operators associated with Drazin invertible operators using their Drazin inverses. Some properties of this class are studied along with examples. In the third section, the tensor product of some members of this class is discussed.
2 p-tuple of \((n,m)\)-Drazin normal oreators
In this section, we introduce and study the class of jointly \((n,m)\)-power D-normal multioperators.
Definition 2.1
Let \({\mathbf{\large N}}:=(N_{1},\ldots ,N_{p}) \in {\mathbf{B}}_{d}[ \mathbf{K}]^{p}\). We said that N is p-tuple of \((n, m)\)-Drazin normal operators for some positive integers n and m if N satisfies the following conditions
When \(n=m=1\), we said that N is p-tuple of \(\mathcal{D}\)-normal operators and if \(m=1\), N is p-tuple of n-\(\mathcal{D}\)-normal operators.
Example 2.1
Let \(N\in { \mathbf{B}}_{d}[\mathbf{K}]\) be an \((n,m)\)-\(\mathcal{D}\)-normal operator, then \({\mathbf{\large N}}=(N,\ldots ,N)\in { \mathbf{B}}_{d}[\mathbf{K}]^{p}\) is p-tuple of \((n,m)\)-\(\mathcal{D}\)-normal operators.
Example 2.2
Let \({\mathbf{\large N}}:= (N_{1},\ldots ,N_{p} ) \in {\mathbf{B}}_{d}[ \mathbf{K}]^{p}\) be commuting operators. If each \(N_{k}\) is \((n,m)\)-\(\mathcal{D}\)-normal single operator, then N is p-tuple of \((n,m)\)-\(\mathcal{D}\)-normal operators.
The following example shows that there exists a p-tuple of operators \({\mathbf{\large N}}=(N_{1},\ldots ,N_{p})\in { \mathbf{B}}_{d}({ \mathbf{K}})^{p}\) such that each \(N_{k}\) is \((n,m)\)-\(\mathcal{D}\)-normal for \(k=1,\ldots ,p\), however N is not p-tuple of \((n,m)\)-\(\mathcal{D}\)-normal operators. This means that the study of these concepts is not trivial.
Example 2.3
Let \({\mathbf{\large N}}=(N_{1}, N_{2})\in { {\mathbf{B}}}[{\mathbb{C}}^{4}]\) where
It is easy to check that \([N_{1}, N_{2} ]\neq0\) and \([ (N_{j}^{\mathcal{D}} )^{n},N_{j}^{*m} ]=0\) for \(j=1,2\). This means that, each \(N_{j}\) is \((n,m)\)-power \(\mathcal{D}\)-normal, while that N is not p-tuple of \((n,m)\)-\(\mathcal{D}\)-normal operators.
In the the following theorem we collect some properties of p-tuple of \((n,m)\)-\(\mathcal{D}\)-normal operators.
Theorem 2.1
Let\({{\mathbf{\large N}}}=(N_{1},\ldots ,N_{p})\in {\mathbf{B}}_{d}[ \mathbf{K}]^{p}\)bep-tuple of\((n,m)\)-\(\mathcal{D}\)-normal operators, then the following properties hold.
(1) Nisp-tuple of\((rn,sm)\)-\(\mathcal{D}\)-normal operators for some positive integersrands.
By looking that \(N_{k}\) is an \((n,m)\)-\(\mathcal{D}\)-normal, then from [28, Proposition 2.10], we obtain that \(N_{k}^{q_{k}}\) is an \((n,m)\)-\(\mathcal{D}\)-normal for all \(k \in \{1,\ldots ,q\}\). This means that \((N_{1}^{q_{1}},\ldots ,N_{p}^{q_{p}} )\) is a p-tuple of \((n,m)\)-\(\mathcal{D}\)-normal operators.
(3) From Definition 2.1, we have under the condition that N is a p-tuple of \((n,m)\)-power \(\mathcal{D}\)-normal operators that is
By taking into account the Fugled–Putnam theorem ([31]), it follows that \([ (N_{k}^{\mathcal{D}} )^{n},N_{k}^{*m} ]=0\) for each \(k=1,\ldots ,p\). Therefore N is a p-tuple of \((n,m)\)-\(\mathcal{D}\)-normal operators. □
Proposition 2.2
Let\({{\mathbf{\large N}}}=(N_{1},\ldots ,N_{p})\in {\mathbf{B}}_{d}[ \mathbf{K}]^{p}\). The following assertions hold.
(3) Nis ap-tuple of\((n,m)\)-\(\mathcal{D}\)-normal operators, if and only if\(({\mathbf{\large N}}^{\mathcal{D}} )^{n}\)commutes with\({\mathbf{\large N}}^{\prime}\).
(4) Nis ap-tuple of\((n,m)\)-\(\mathcal{D}\)-normal operators if and only if\(({\mathbf{\large N}}^{\mathcal{D}} )^{n}\)commutes with\({\mathbf{\large N}}^{\prime \prime}\).
Proof
Obviously, \([N_{k},N_{l} ]=0\forall (k,l)\in \{1,\ldots ,p\}^{2}\). On the other hand,
Suppose N is p-tuple of \((n,m)\)-\(\mathcal{D}\)-normal operators for \(n\geq m\). It is easy to see that each \(N_{k}\) is \((n,m)\)-D-normal for \(1\leq k \leq d\). Under the hypothesis that \(N_{k}^{m}\) is a partial isometry, it follows from [28, Theorem 2.4] that \(N_{k}\) is \((n+m,m)\)-\(\mathcal{D}\)-normal operator for \(k=1,\ldots ,p\). Consequently, N is a p-tuple of \((n+m,m)\)-\(\mathcal{D}\)-normal operators. □
The following proposition shows that the class of p-tuple of \((n,m)\)-\(\mathcal{D}\)-normal operators is closed subset of \({ \mathbf{B}}_{d}[{\mathbf{K}}]^{p}\).
Proposition 2.5
The class ofp-tuple of\((n,m)\)-\(\mathcal{D}\)-normal operators is a closed subset of\({\mathbf{B}}_{d}[\mathbf{K}]^{p}\).
Proof
Suppose that \(( {\mathbf{\large N}}_{k}= (N_{1}(k),\ldots ,N_{p}(k) ) )_{k} \in {\mathbf{B}}_{d}[\mathbf{K}]^{p}\) is a sequence of p-tuple of \((n,m)\)-power \(\mathcal{D}\)-normal operators for which
Since \((N_{j}(k)^{\mathcal{D}} )^{n}N_{j}(k)^{*m}=N_{j}(k)^{*m} (N_{j}(k)^{ \mathcal{D}} )^{n}\) for each \(j=1,\ldots ,p\), it follows from [28, Theorem 2.4] that
On the other hand, since \(\{{\mathbf{\large N}}_{k}\}_{k}=\{ (N_{1}(k),\ldots ,N_{p}(k))\}_{k}\) is a sequence of p-tuple of \((n,m)\)-\(\mathcal{D}\)-normal operators, then
This implies that NS is a p-tuple of \((n,m)\)-\(\mathcal{D}\)-normal operators. In same way, we show that SN is a p-tuple of \((n.m)\)-\(\mathcal{D}\)-normal operators.
(2) For all \((i,j)\in \{1,\ldots ,p\}^{2}\), we have
So, \({\mathbf{\large N}+\mathbf{S}}\) is a p-tuple of \((n,m)\)-\(\mathcal{D}\)-normal operators. □
3 Tensor product
Let \({\mathbf{\large N}}=(N_{1},\ldots ,N_{p})\in { \mathbf{B}}[\mathbf{K}]^{p}\) and \({\mathbf{\large S}}=(S_{1},\ldots ,S_{p})\in { \mathbf{B}}[\mathbf{K}]^{p}\). We denote by
If \(N,S \in { \mathbf{B}}[\mathbf{K}]\), then \(N\otimes S\) is n-normal if and only if N and S are n-normal (see [6]) However, If \(N ,S\in { \mathbf{B}}_{d}[\mathbf{K}]\) such that N and S are \((n,m)\)-\(\mathcal{D}\)-normal operators, then \(N\otimes S\) is \((n,m)\)-\(\mathcal{D}\)-normal (see in [28]). The following theorem studied the tensor product of two p-tuples of \((n,m)\)-\(\mathcal{D}\)-normal operators.
Since \({\mathbf{\large N}}= (N_{1},\ldots ,N_{p} )\) and \({\mathbf{\large S}} = (S_{1},\ldots ,S_{p} )\) are p-tuples of \((n,m)\)-\(\mathcal{D}\)-normal operators, we have all \((k,l) \in \{ 1,\ldots ,p\}^{2}\)
Let \({\mathbf{\large N}}= (N_{1},N_{2} )\) and \({\mathbf{\large N}}\otimes {\mathbf{\large N}}= (N_{1}\otimes N_{1},N_{2} \otimes N_{2}) ) \). We observe that N is not 2-tuple of \((2,3)\)-\(\mathcal{D}\)-normal operators since \(N_{1}N_{2}\neq N_{2}N_{1}\). However
Hence, \({\mathbf{\large N}}\otimes {\mathbf{\large N}}\) is 2-tuple of \((2,3)\)-\(\mathcal{D}\)-normal pairs.
In the following theorem we give the conditions under which the converse of Theorem 3.1 is true.
Theorem 3.2
Let\({\mathbf{\large N}}=(N_{1},\ldots ,N_{p})\in { \mathbf{B}}_{d}[\mathbf{K}]^{p}\)and\({\mathbf{\large S}}=(S_{1},\ldots ,S_{p})\in { \mathbf{B}}_{d}[\mathbf{K}]^{p}\)be a commutingp-tuple of operators. Then, if\({\mathbf{\large N}}\otimes {\mathbf{\large S}}\)is ap-tuple of\((n,n)\)-\(\mathcal{D}\)-normal operators, then and only thenNandSarep-tuples of\((n,n)\)-\(\mathcal{D}\)-normal operators.
Proof
Assume that \({\mathbf{\large N}}\otimes {\mathbf{\large S}}\) is a p-tuple of \((n,n)\)-\(\mathcal{D}\)-normal operators. By taking into account the statement (1) of Proposition 2.1 it follows that
is normal for each \(k=1,\ldots ,p\). By [19, Propositon 3.2] it is well known that
$$ \bigl(N_{k}^{\mathcal{D}} \bigr)^{n}\otimes \bigl(S_{k}^{D} \bigr)^{n} \text{ is normal if and only if } \bigl(N_{k}^{\mathcal{D}} \bigr)^{n} \text{ and } \bigl(S_{k}^{\mathcal{D}} \bigr)^{n} \text{ are normal operators}. $$
However, According to [19, Propositon 3.2] it is well known that \((N_{k}^{\mathcal{D}} )^{n}\) is normal if and only if that \(N_{k}\) is n-power \(\mathcal{D}\)-normal and similarly, \((S_{k}^{ \mathcal{D}} )^{n}\) is normal if and only if that \(S_{k}\) is n-\(\mathcal{D}\)-normal. Therefore, N and S are p-tuple of \((n,n)\)-\(\mathbb{D}\)-normal operators.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.