bibkey: kopel2026sharpnorm authors: E. Kopel year: 2026 title: “A sharp norm inequality for entanglement-breaking channels” doi: 10.48550/arXiv.2609.27906 url: https://arxiv.org/abs/2609.27906v1 claim: “Theorem 1: the Bloch data (A, c) of an entanglement-breaking channel on M_d satisfy ‖A‖_*² + (d(d−1)/2)|c|² ≤ (d−1)²; open question 3 asks whether the coefficient d(d−1)/2 is optimal.” strata_touched:
- D5/S3/Quantum/QuantumChannels/EntanglementBreakingNormCoefficient license: citation-only triage: anchor
Kopel, A sharp norm inequality for entanglement-breaking channels
E. Kopel, A sharp norm inequality for entanglement-breaking channels, arXiv:2609.27906v1 (submitted 21 August 2026; quant-ph, cross-listed math-ph and math.OA).
Verified locator
DOI: 10.48550/arXiv.2609.27906.
Primary version: https://arxiv.org/abs/2609.27906v1 (the only arXiv version).
The TeX source of v1 supplies the Holevo form (Eq. eq:holevo, Section 1), the Bloch
parameterisation (Section 2), Theorem 1 (thm:main, Section 3), the equality discussion
(Section 4) and the list of open questions (Section 8); the numbering is the printed one.
Source statements
The source macros \Tr, \I and \nuc{·} are written out as , and .
Section 1: “A completely positive trace-preserving map is \emph{entanglement breaking} (EB) if is separable for every input, equivalently if admits a Holevo form with a POVM and states”.
Section 2: “Fix a Hermitian traceless basis of normalised by . […] Every state is , , with equality precisely for pure states. A channel acts affinely on the Bloch vector, .”
Theorem 1: “Let be an entanglement-breaking channel on with Bloch data . Then .”
Section 4: “Whether equality with is achievable for is not settled here; condition (ii) becomes restrictive as the number of rank-one terms grows.”
Section 8, item 3: “Is the coefficient optimal? It combines two separately sharp bounds which may not be simultaneously saturable with in higher dimension.”
Scope
In the basis the Bloch data of a channel are and , and is the nuclear (trace) norm. The paper proves Theorem 1 from the Holevo form, shows equality for the completely dephasing channel in every dimension and for a family with at , and relates the inequality to the covariance-matrix separability criterion.