We show that for adjointable morphisms on Hilbert C*-module over a commutative unital C*-algebra, a version of breakthrough numerical radius inequalities obtained by Kittaneh [\textit{{Stud. Math.}, 2005}] hold.
Introduction
The numerical range or the field of values[11, 12, 13, 14, 15, 16, 18, 19, 36] of a continuous linear operator T acting on a complex Hilbert space H (with the inner product linear in the first variable and conjugate linear in the second variable) is defined as
WC(T):={⟨Th,h⟩:h∈H,∥h∥=1}⊆C
and the numerical radius of T is defined as
wC(T):=h∈H,∥h∥=1sup∣⟨Th,h⟩∣∈[0,∞).
Using the Cauchy-Schwarz inequality and the generalized polarization identity [13], we get
2∥T∥≤wC(T)≤∥T∥(1)
Let B(H) be the C*-algebra of all continuous linear operators acting on H. Inequalities (1) say that the map wC(⋅):B(H)→R is a norm which is equivalent to the operator norm. In 2005, Kittaneh made a far reaching generalization of Inequalities (1) [22].
Theorem 1.1.[22] (Kittaneh Numerical Radius Inequalities) Let T∈B(H). Then
21∥T∗T+TT∗∥≤wC(T)≤21∥T∗T+TT∗∥(2)
Since
2∥T∥≤21∥T∗T+TT∗∥≤21∥T∗T+TT∗∥≤∥T∥,∀T∈B(H),
Inequalities (2) improve Inequalities (1). Our fundamental motivation comes from the following question: What is the noncommutative analogue of Theorem 1.1? This is then naturally connected with the notion of Hilbert C*-modules which are first introduced by Kaplansky [21] for modules over commutative C*-algebras and later developed for modules over arbitrary C*-algebras by Paschke [27] and Rieffel [32].
Definition 1.2.[21, 27, 32] Let A be a unital C*-algebra. A right module E over A is said to be a semi-inner product C*-module if there exists a map ⟨⋅,⋅⟩:E×E→A such that the following hold.
Theorem 1.3.[23] (Cauchy-Schwarz-Kaplansky-Paschke-Rieffel Inequality) Let E be a semi-inner product C*-module. Then
⟨y,x⟩⟨x,y⟩≤∥⟨x,x⟩∥⟨y,y⟩,∀x,y∈E.
In particular,
∥⟨y,x⟩∥2≤∥⟨x,x⟩∥∥⟨y,y⟩∥,∀x,y∈E.
Definition 1.4.[23] A semi-inner product C*-module E is said to be inner product C*-module if x∈E satisfies ⟨x,x⟩=0, then x=0.
Definition 1.5.[23] An inner product C*-module E is said to be Hilbert C*-module if E is complete w.r.t. the norm ∥x∥:=∥⟨x,x⟩∥, ∀x∈E.
In this article, we derive noncommutative version of classical numerical radius inequalities and modular version of Kittaneh numerical radius inequalities.
Modular Kittaneh Numerical Radius Inequalities
Let E be a Hilbert C*-module over a unital C*-algebra A. Similar to Hilbert space case, we define the noncommutative numerical range of an adjointable morphism T:E→E as
WA(T):={⟨x,Tx⟩:x∈E,∥x∥=1}⊆A
and the noncommutative numerical radius of T is defined as
wA(T):=x∈E,∥x∥=1sup∥⟨x,Tx⟩∥∈[0,∞).
Note that our definition of numerical range and radius on Hilbert C∗-modules differ from the existing same notion on Hilbert C∗-modules [26, 30, 39]. We give various examples. In all following examples, A is a unital C∗-algebra.
Example 2.1.
WA(1)={a∗a:a∈A,∥a∥=1}
and
wA(1)=1.
Example 2.2. Let d∈N and Id be the identity matrix of size d by d. Then we have
Given r>0 and v∈A, the closed disc centered at v∈A of radius r is defined as
DA[v,r]:={z∈A:∥z−v∥≤r}.
Following are basic properties of noncommutative numerical range and radius.
Proposition 2.8. Let E be a Hilbert C∗-module over a unital C∗-algebra A. Let T,S:E→E be adjointable morphisms and a,b∈A. Let I be the identity morphism on E.
(i) WA(T)⊆DA[0,∥T∥].
(ii) 0≤wA(T)≤∥T∥.
(iii) WA(T+S)⊆WA(T)+WA(S).
(iv) wA(S+T)≤wA(S)+wA(T).
(v) WA(Ta)=WA(T)a.
(vi) wA(Ta)≤∥a∥wA(T).
(vii) WA(T∗)=(WA(T))∗.
(viii) wA(T∗)=wA(T).
(ix) WA(Ta+b)=WA(T)a+b.
(x) WA(Re(T))=Re(WA(T)) and WA(Im(T))=Im(WA(T)).
(xi) If T≥0, then WA(T)⊆A+.
(xii) Let E0 be a Hilbert C∗-module over a unital C∗-algebra A such that E is orthogonally complementable in E0. If an adjointable morphism V:E0→E0 is a dilation of T, then WA(T)⊆WA(V).
(xiii) Let E1 be an orthogonally complementable closed submodule in E. Let P:E→E1 be an onto orthogonal projection. Then WA(PT∣E1)⊆WA(T).
(xiv) WA(U∗TU)=WA(T) for every unitary morphism U:E→E.
(xv) Let a∈A be such that there exists a x∈E with Tx=xa and ⟨x,x⟩=1. Then a∈WA(T). In other words,
σ1(T):={a∈A:∃x∈E,Tx=xa,⟨x,x⟩=1}⊆WA(T)
and
∥a∥≤wA(T),∀a∈σ1(T).
(xvi)
wA(ST+TS)≤2wA((S+T)2)+wA((S−T)2).
In particular, if ST=TS, then
wA(ST)≤4wA((S+T)2)+wA((S−T)2).
(xvii) If a sequence {Tn}n=1∞ of adjointable morphisms on E converges to an adjointable morphism T on E in the morphism norm, then the sequence {wA(Tn)}n=1∞ converges to wA(T) in the norm.
We next derive a noncommutative analogue of Inequalities (1).
Theorem 2.9. Let E be a Hilbert C∗-module over a unital C∗-algebra A. Let T:E→E be an adjointable morphism. Then
Therefore Inequalities (5) improve Inequalities (3) over commutative C∗-algebras. We wish to note that Abu-Omar and Kittaneh [1] and Bhunia, Bag and Paul [3] improved Theorem 1.1 but the techniques used by them do not seem to carry over to Hilbert C∗-modules.
Noncommutative Toeplitz-Hausdorff Problem
The celebrated Toeplitz-Hausdorff theorem says that the numerical range is always a convex subset of C[11, 13, 15, 16, 18, 25, 36]. We formulate the following problem.
Problem 3.1. What is noncommutative version of Toeplitz-Hausdorff Theorem (fundamental theorem of the numerical range)?
Noncommutative Crouzeix and von Neumann Inequality Problems
Breakthrough Crouzeix theorem says following.
Theorem 4.1.[4, 6, 7, 8, 31](Crouzeix Theorem) For every d∈N and for every matrix M∈Md(C), we have
Problem 4.2.(Noncommutative Crouzeix Problem) Let A be a unital C∗-algebra. Whether there is a universal constant RA (which may depend upon A) satisfying following: For every d∈N and for every matrix M∈Md(A), we have
Based on Theorem 4.3 we formulate following problem.
Problem 4.4. (Noncommutative von Neumann Inequality Problem) Let A be a unital C∗-algebra. Whether there is a universal constant RA (which may depend upon A) satisfying the following: For every d∈N and for every A∈Md(A) with ∥A∥≤1, we have
Theorem 4.3 has been extended by Ando for two variables [2, 35]. Note that Ando theorem cannot be extended to more than two commuting matrices [5, 37]. We observe that the Halmos dilation [14], Egervary dilation [10, 24] and Sz.-Nagy dilation (Schaffer construction) [33] carry over to adjointable morphisms on Hilbert C∗-modules [29] but this result will not give noncommutative von Neumann inequality. We also observe that the proof of Ando dilation [2] for commuting operators on Hilbert space will not carry over to commuting adjointable morphisms on Hilbert C∗-modules (mainly because closed submodules need not be orthogonally complementable and C∗-algebras need not have the invariant basis number property).
Noncommutative Power Inequality Problem
Surprising power inequality for numerical radius says the following.
Theorem 5.1.[28] For every bounded linear operator T on a Hilbert space,
(Power Inequality)wC(Tn)≤(wC(T))n,∀n∈N.
We formulate the following problem.
Problem 5.2. (Noncommutative Power Inequality Problem) Let T:E→E be an adjointable morphism. Whether
wA(Tn)≤(wA(T))n,∀n∈N?
Noncommutative Berger Power Dilation Problem
Berger obtained the following result connecting numerical radius and dilation theory.
Theorem 6.1.[13, 35] For a continuous linear operator T:H→H, the following are equivalent.
wC(T)≤1.
There exist a Hilbert space H0 which contains H isometrically, PH:H0→H an onto orthogonal projection and unitary operator U:H0→H0 such that
Tnh=2PHUnh,∀n∈N,∀h∈H.
We ask for the following.
Problem 6.2. What is noncommutative version of Berger power dilation?
Noncommutative Johnson Numerical Range Problem
In 1974, Johnson derived following important result.
Theorem 7.1.[20] (Johnson Numerical Range Theorem) For every A=[aj,k]1≤j≤d,1≤k≤d∈Md(C),
Problem 7.2. (Noncommutative Johnson Numerical Range Problem) What is noncommutative version of Johnson numerical range theorem?
Noncommutative Holbrook Numerical Radius Inequality Problem
In 1969, Holbrook, derived following result for commuting product.
Theorem 8.1.[17] Let A, B be continuous linear operators on the same Hilbert space such that AB=BA. Then
(Holbrook Numerical Radius Inequality for Commuting Product) wC(AB)≤2wC(A)wC(B).
For arbitrary operators, Holbrook derived the following result.
Theorem 8.2.[17] Let A, B be continuous linear operators on the same Hilbert space. Then
(Holbrook Numerical Radius Inequality for Product) wC(AB+BA)≤4wC(A)wC(B).
We formulate the following problem.
Problem 8.3. What is noncommutative version of Holbrook numerical radius inequality for product?
Acknowledgments
This paper was partly developed at the Lodha Mathematical Sciences Institute (LMSI), Mumbai, India, while the author was attending the thematic program “Surface Group Representation and Analytic Group Theory” from 9 to 27 November 2026. The author would like to thank the LMSI and its founders for this opportunity.
References
[1]Amer Abu-Omar and Fuad Kittaneh. Upper and lower bounds for the numerical radius with an application to involution operators. Rocky Mt. J. Math., 45(4):1055–1064, 2015.
[2]T. Andō. On a pair of commutative contractions. Acta Sci. Math., 24:88–90, 1963.
[3]Pintu Bhunia, Santanu Bag, and Kallol Paul. Numerical radius inequalities and its applications in estimation of zeros of polynomials. Linear Algebra Appl., 573:166–177, 2019.DOI
[4]Kelly Bickel, Pamela Gorkin, Anne Greenbaum, Thomas Ransford, Felix L. Schwenninger, and Elias Wegert. Crouzeix’s conjecture and related problems. Comput. Methods Funct. Theory, 20(3-4):701–728, 2020.
[5]M. J. Crabb and A. M. Davie. Von Neumann’s inequality for Hilbert space operators. Bull. Lond. Math. Soc., 7:49–50, 1975.DOI
[6]M. Crouzeix and C. Palencia. The numerical range is a (1 + √2)-spectral set. SIAM J. Matrix Anal. Appl., 38(2):649–655, 2017.DOI
[7]Michel Crouzeix. Bounds for analytical functions of matrices. Integral Equations Oper. Theory, 48(4):461–477, 2004.DOI
[8]Michel Crouzeix. Numerical range and functional calculus in Hilbert space. J. Funct. Anal., 244(2):668–690, 2007.DOI
[9]Bernard Delyon and François Delyon. Generalization of von Neumann’s spectral sets and integral representation of operators. Bull. Soc. Math. Fr., 127(1):25–41, 1999.DOI
[10]E. Egerváry. On the contractive linear transformations of n-dimensional vector space. Acta Sci. Math., 15:178–182, 1954.
[11]Hwa-Long Gau and Pei Yuan Wu. Numerical ranges of Hilbert space operators, volume 179 of Encycl. Math. Appl. Cambridge: Cambridge University Press, 2021.DOI
[12]Moshe Goldberg and Eitan Tadmor. On the numerical radius and its applications. Linear Algebra Appl., 42:263–284, 1982.DOI
[13]Karl E. Gustafson and Duggalira K. M. Rao. Numerical range. The field of values of linear operators and matrices. Universitext. New York, NY: Springer, 1996.
[14]P. R. Halmos. Numerical ranges and normal dilations. Acta Sci. Math., 25:1–5, 1964.
[15]Paul R. Halmos. A Hilbert space problem book, volume 19 of Grad. Texts Math. Springer, Cham, 1982.
[16]F. Hausdorff. Der Wertvorrat einer Bilinearform. Math. Z., 3:314–316, 1919.
[17]J. A. R. Holbrook. Multiplicative properties of the numerical radius in operator theory. J. Reine Angew. Math., 237:166–174, 1969.DOI
[18]Roger A. Horn and Charles R. Johnson. Topics in matrix analysis. Cambridge University Press, 1991.DOI
[19]Vasile I. Istratescu. Introduction to linear operator theory, volume 65 of Pure Appl. Math., Marcel Dekker. Marcel Dekker, Inc., New York, NY, 1981.
[20]Charles R. Johnson. Gersgorin sets and the field of values. J. Math. Anal. Appl., 45:416–419, 1974.DOI
[21]Irving Kaplansky. Modules over operator algebras. Amer. J. Math., 75:839–858, 1953.DOI
[22]Fuad Kittaneh. Numerical radius inequalities for Hilbert space operators. Stud. Math., 168(1):73–80, 2005.DOI
[23]E. Christopher Lance. Hilbert C*-modules. A toolkit for operator algebraists, volume 210 of Lond. Math. Soc. Lect. Note Ser. Cambridge University Press, 1995.
[24]Eliahu Levy and Orr Moshe Shalit. Dilation theory in finite dimensions: the possible, the impossible and the unknown. Rocky Mt. J. Math., 44(1):203–221, 2014.arxiv.org/abs/1012.4514
[25]Chi-Kwong Li. A simple proof of the elliptical range theorem. Proc. Am. Math. Soc., 124(7):1985–1986, 1996.DOI
[26]Marzieh Mehrazin, Maryam Amyari, and Mohsen Erfannian Omidvar. A new type of numerical radius of operators on Hilbert C*-module. Rend. Circ. Mat. Palermo (2), 69(1):29–37, 2020.
[27]William L. Paschke. Inner product modules over B*-algebras. Trans. Amer. Math. Soc., 182:443–468, 1973.
[28]C. Pearcy. An elementary proof of the power inequality for the numerical radius. Mich. Math. J., 13:289–291, 1966.DOI
[29]D. Popovici. On the geometric structure of minimal dilations on Hilbert C*-modules. Z. Anal. Anwend., 17(2):379–392, 1998.
[30]Rajna Rajic. On the algebra range of an operator on a Hilbert C*-module over compact operators. Proc. Am. Math. Soc., 131(10):3043–3051, 2003.DOI
[31]Thomas Ransford and Felix L. Schwenninger. Remarks on the Crouzeix-Palencia proof that the numerical range is a (1 + √2)-spectral set. SIAM J. Matrix Anal. Appl., 39(1):342–345, 2018.
[32]Marc A. Rieffel. Induced representations of C*-algebras. Advances in Math., 13:176–257, 1974.DOI
[33]J. J. Schäffer. On unitary dilations of contractions. Proc. Am. Math. Soc., 6:322, 1955.
[34]Béla Sz.-Nagy. Sur les contractions de l’espace de Hilbert. Acta Sci. Math., 15:87–92, 1953.
[35]Béla Sz.-Nagy, Ciprian Foias, Hari Bercovici, and László Kérchy. Harmonic analysis of operators on Hilbert space. Universitext. New York, NY: Springer, 2 edition, 2010.
[36]O. Toeplitz. Das algebraische Analogon zu einem Satze von Fejér. Math. Z., 2:187–197, 1918.DOI
[37]N. Th. Varopoulos. On an inequality of von Neumann and an application of the metric theory of tensor products to operators theory. J. Funct. Anal., 16:83–100, 1974.DOI
[38]Johann von Neumann. Eine Spektraltheorie für allgemeine Operatoren eines unitären Raumes. Math. Nachr., 4:258–281, 1951.