International Journal of Mathematics and Mathematical Sciences Volume 2011, Article ID 801313,5pages
doi:10.1155/2011/801313
Research Article
Almost
α
-Hyponormal Operators with
Weyl Spectrum of Area Zero
Vasile Lauric
Department of Mathematics, Florida A&M University, Tallahassee, FL 32307, USA
Correspondence should be addressed to Vasile Lauric,[email protected]
Received 27 December 2010; Accepted 20 March 2011
Academic Editor: Ra ¨ul Curto
Copyrightq2011 Vasile Lauric. 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.
We define the class of almostα-hyponormal operators and prove that for an operatorT in this class,T∗Tα−TT∗αis trace-class and its trace is zero whenα∈0,1and the area of the Weyl spectrum is zero.
This note is dedicated to Professor Carl M. Pearcy with the occasion of his 75th birthday.
LetHbe a complex, separable, infinite-dimensional Hilbert space, and letLHdenote the
algebra of all linear bounded operators onH, and for 1 ≤ p < ∞, let CpH denote the
p-Schatten class on H. For K ∈ CpH, the expression ||K||p : ∞n1μnKp1/p, where
μ1K≥μ2K≥ · · · are the singular values ofK, is a norm forp≥1, and is only a quasinorm
for 0< p <1it does not satisfy the triangle inequality. Nevertheless, the latter case will be
used in what follows.
For T ∈ LH,σT and σwT will denote the spectrum and the Weyl spectrum,
respectively. Recall that Weyl spectrum is the union of the essential spectrum, σeT, and
all bounded components of \σeTassociated with nonzero Fredholm index. An operator
T ∈ LHis calledCp, α-normalnotation:T ∈ Nα
pHifCαT : T∗Tα−TT∗αbelongs
to CpH, andT is calledCp, α-hyponormalnotation: T ∈ Hα
pHif CαT is the sum of a
positive definite operator and an operator inCpH, or equivalently,CαT−the negative part
of Cα
Tbelongs to CpH, whereαis a positive number. This note will be concerned with
the particular classHα
1H, which by some parallelism with some terminology used in1,
would be appropriate to be referred as almostα-hyponormal operators.
Voiculescu’s 1generalization of Berger-Shaw inequality gives an estimate for the
trace ofC1
T. The result was extended in2. The combination of these results will be stated
T inLH, denoted bymT, is the smallest cardinal numbermwith the property that there
aremvectorsx1, . . . , xminHsuch that
∨fTxj|1≤j≤m, f ∈RatσTH, 1
where RatσTis the algebra of complex-valued rational functions with poles offσT.
For a Borel subsetE⊆ andα >0, denoteμαE α/2Eρα−1 dρdθ. In particular,
μ2is the planar Lebesgue measure.
Theorem Asee1,2. SupposeT ∈H11H. If there existsK ∈ C2Hsuch that eithermT
K<∞orμ2σTK 0, thenT ∈N1
1H. Moreover, whenmTK<∞,
trC1T
≤ mTK
π · μ2σTK, 2
and whenμ2σTK 0, trC1
T≤0, and consequently, trC1T 0.
In fact, it was observed in2that the inequality can be improved by replacingmTK
withτTK, where
τS:lim infrankI−PSP, 3
and the lim inf is taken over all sequences of finite-rank orthogonal projections such that
P → Iin the strong operator topology.
Corollary Bsee2. LetT ∈H1
1Hsuch thatμ2σwT 0. ThenT ∈N11Hand trCT1 0.
On the other hand, Berger-Shaw inequality was extended to operators inH1αHusing
similar circle of ideas used in1. This was done in3for the caseα∈1/2,1and later on
in4for the caseα∈0,1/2.
Theorem Csee3,4. Let 0< α≤1, and letT ∈H1αHandK∈ C2αHwithmTK<∞.
ThenT ∈Nα
1Hand
trCα
T ≤
mTK
π ·μ2ασTK. 4
The case in whichmTK ∞andμ2ασTK 0 was not discussed in4or3.
It is the goal of this note to make some progress towards this case. We have the following.
Theorem 1. Letα ∈ 0,1and letT ∈ Hα
1HandK ∈ CαHwithμ2ασT K 0. Then
T ∈Nα1Hand trCTα 0.
Remark. It would have been desirable that Theorem1 be proved with the hypothesis that
K∈ C2αH.
Corollary 2. Letα∈0,1and letT ∈H1αHsuch thatμ2σwT 0. ThenT ∈N1αHand
trCαT 0.
Proof. Ifα1, then conclusion holds according to Corollary B. Letα∈0,1. First, a careful
inspection of the proof of a result of Stampfli5leads to the following. ForT ∈ LHand
α >0, there existsKα∈ CαHsuch thatσTKα\σwTconsists of a countable set which
clusters only onσwT. Thereforeμ2σTKα 0 and thus Theorem1applies.
The proof of Theorem1makes use of the following three inequalities.
Proposition DHansen’s inequality6. IfA, B∈LH,A≥0,||B|| ≤1, andα∈0,1, then
B∗AαB≤B∗ABα.
Proposition E Lowner’s inequality 7. IfA, B ∈ LH,A ≥ B ≥ 0, and α ∈ 0,1, then
Aα≥Bα.
The following is a consequence of Theorem 3.4 of8.
Proposition FJocic’s inequality8. LetA, B∈LH,A, B≥0,α∈0,1, and 1≤p <∞. If
A−B∈ CαpH, thenAα−Bα∈ CpHand||Bα−Aα||p≤ || |B−A|α ||p.
Proof of Theorem1. Letα ∈0,1,T ∈Hα
1H, andK ∈ CαHwithμ2ασT K 0, and
assumemTK ∞, otherwise Theorem C impliesT∈Nα
1H.
Let{en}n∈be an orthonormal basis ofHand let
Hn∨rTKej| j1, . . . , n, r∈RatσTK. 5
Assume that with respect to the decompositionHHn⊕ H⊥n, operatorsTandKare written
as
T
T1n T2n
T3n T4n
, K
K1n K2n
K3n K4n
. 6
SinceHnis a rationally invariant subspace forTK, we haveT3nK3n 0, and thusT3n
−K3n∈ CαHn⊆ C2αHn, andσT1nK1n⊆σTK, which impliesμ2ασT1nK1n 0.
LetPnbe the orthogonal projection ontoHn, and thusPn ↑ Istrongly. We will prove
next thatT1n∈Hα
1Hnby first establishing that
PnCTαPn−CαT1n −QnKn, 7a
whereQn∈LHnis positive semidefinite andKn ∈ C1Hn.
Assuming that equality7awas already proved and writingCαT QKwithQ≥0
andK∈ C1H, then we have
CαT1n PnQPnPnKPnQn −Kn, 7b
that is,Cα
T1nis the sum ofPnQPnQ
n, which is a positive semidefinite operator, and ofPnKPn−
Indeed, the expressionPnCαTPn−CTα1ncan be written asD1−D2, where
D1PnT∗TαPn−T1n∗ T1n
α ,
D2PnTT∗αPn−T1nT1n∗ α.
8
We can writeD1−QnKn, where
Qn
PnT∗TPnα−PnT∗Tα
Pn, 9
which according to Hansen’s inequality is a positive semidefinite operator, and
Kn
PnT∗TPnα−PnT∗PnTPnα
, 10
which according to Jocic’s inequality is a trace-class operator that satisfies
K
n1≤|PnT∗TPn−PnT∗PnTPn|
α
1T3n∗ T3nα1
T3n∗ T3n
α
α≤T3n∗
α· T3nα
α≤ Tα· T3nαα.
11
Concerning operatorD2, we can writeD2Qn Kn, where
Qn PnTT∗αPn−PnTPnT∗αPn, 12
which according to Lowner’s inequality is a positive semidefinite operator, and
KnPnTPnT∗αPn−PnTPnT∗PnαPn
TPnT∗α−PnTPnT∗Pnα
Pn, 13
which is also a trace-class operator since
TPnT∗−PnTPnT∗Pn TPnT∗−TPnT∗Pn TPnT∗Pn−PnTPnT∗Pn
TPnT∗I−Pn I−PnTPnT∗Pn
TT3n∗ T3nT∗Pn∈ CαH,
14
and according to Jocic’s inequality
K
n1≤TPnT∗
α−PnTPn
T∗Pnα1 ≤TT3n∗ T3nT∗Pn
α
1
TT3n∗ T3nT∗Pn
α
α≤CTT3n∗
α
αT3nT∗Pn
α α
≤CTαT3n∗ααT3nαα 2CTαT3nαα.
Therefore,
D2Qn Kn, withQn ≥0, Kn∈C1H, 16
and consequently, D1−D2 −QnKn−Qn Kn −Qn Qn Kn−Kn, where
QnQn : Qnis positive semidefinite andKn−Kn : Kn is trace-class, which establishes
equality7a.
According to7b,T1n ∈ H1αHn, and sincemT1n K1n ≤ nand σT1nK1n ⊆
σT K, Theorem C implies that trCα
T1n≤0, and furthermore, by replacingT1nwithT
∗
1n,
trCα
T1n 0. Furthermore, equality7aimplies
PnCαTPn≤CαT1nKn, 17
which further implies
trPnCαTPn ≤tr
Kn . 18
Similar utilization of Lowner’s and Hansen’s inequalities implies that Kn and −Kn are
positive semidefinite, and thus so isKn Kn−Kn. Therefore
trKn ≤Kn 1Kn 1≤12CT
αT3nα
α. 19
SinceT3n−K3n∈ CpHnandK3n → 0 weakly and both|T3n|and |T3n∗ | ≤ ||T||I, we have
||T3n||α → 0, and thus trCα
T≤0. ReplacingTwithT∗we conclude that trCαT 0.
References
1 D. Voiculescu, “A note on quasitriangularity and trace-class self-commutators,” Acta Scientiarum
Mathematicarum, vol. 42, no. 1-2, pp. 1303–1320, 1980.
2 D. Hadwin and E. Nordgren, “Extensions of the Berger-Shaw theorem,” Proceedings of the American
Mathematical Society, vol. 102, no. 3, pp. 517–525, 1988.
3 R. E. Curto, P. S. Muhly, and D. Xia, “A trace estimate forp-hyponormal operators,” Integral Equations
and Operator Theory, vol. 6, no. 4, pp. 507–514, 1983.
4 A. Aluthge and D. Xia, “A trace estimate ofT∗Tp−TT∗p,” Integral Equations and Operator Theory,
vol. 12, no. 2, pp. 300–303, 1989.
5 J. G. Stampfli, “Compact perturbations, normal eigenvalues and a problem of Salinas,” Journal of the
London Mathematical Society, vol. 9, no. 2, pp. 165–175, 1974-1975.
6 F. Hansen, “An operator inequality,” Mathematische Annalen, vol. 246, no. 3, pp. 249–250, 1980. 7 E. Heinz, “Beitr¨age zur st ¨orungstheorie der spektralzerlegung,” Mathematische Annalen, vol. 123, pp.
415–438, 1951.
8 D. R. Joci´c, “Integral representation formula for generalized normal derivations,” Proceedings of the
Submit your manuscripts at
http://www.hindawi.com
Hindawi Publishing Corporation
http://www.hindawi.com Volume 2014
Mathematics
Journal ofHindawi Publishing Corporation
http://www.hindawi.com Volume 2014
Hindawi Publishing Corporation http://www.hindawi.com
Differential Equations International Journal of
Volume 2014
Applied MathematicsJournal of
Hindawi Publishing Corporation
http://www.hindawi.com Volume 2014
Hindawi Publishing Corporation
http://www.hindawi.com Volume 2014
Hindawi Publishing Corporation
http://www.hindawi.com Volume 2014
Mathematical PhysicsAdvances in
Complex Analysis
Journal of Hindawi Publishing Corporationhttp://www.hindawi.com Volume 2014
Optimization
Journal of Hindawi Publishing Corporationhttp://www.hindawi.com Volume 2014
Combinatorics
Hindawi Publishing Corporation
http://www.hindawi.com Volume 2014
International Journal of
Hindawi Publishing Corporation
http://www.hindawi.com Volume 2014
Journal of
Hindawi Publishing Corporation
http://www.hindawi.com Volume 2014
Function Spaces
Abstract and Applied Analysis Hindawi Publishing Corporation
http://www.hindawi.com Volume 2014
International Journal of Mathematics and Mathematical Sciences
Hindawi Publishing Corporation http://www.hindawi.com Volume 2014
The Scientific
World Journal
Hindawi Publishing Corporation
http://www.hindawi.com Volume 2014
Hindawi Publishing Corporation
http://www.hindawi.com Volume 2014
Discrete Dynamics in Nature and Society
Hindawi Publishing Corporation
http://www.hindawi.com Volume 2014
Hindawi Publishing Corporation
http://www.hindawi.com Volume 2014
Discrete Mathematics
Journal ofHindawi Publishing Corporation
http://www.hindawi.com Volume 2014
Hindawi Publishing Corporation
http://www.hindawi.com Volume 2014