Cited by 2,961Quantum Information and CryptographyQuantum Mechanics and ApplicationsQuantum Computing Algorithms and Architecture

Purification of Noisy Entanglement and Faithful Teleportation via Noisy Channels

Open access
Bennett, Charles H., Brassard, Gilles, Popescu, Sandu, Schumacher, Benjamin, Smolin, John A., Wootters, William K.·1996·Physical Review Letters

Abstract

Two separated observers, by applying local operations to a supply of not-too-impure entangled states (e.g., singlets shared through a noisy channel), can prepare a smaller number of entangled pairs of arbitrarily high purity (e.g., near-perfect singlets). These can then be used to faithfully teleport unknown quantum states from one observer to the other, thereby achieving faithful transmission of quantum information through a noisy channel. We give upper and lower bounds on the yield $D\left(M\right)$ of pure singlets $(|{\ensuremath{\Psi}}^{\ensuremath{-}}〉)$ distillable from mixed states $M$, showing $D\left(M\right)>0$ if $〈{\ensuremath{\Psi}}^{\ensuremath{-}}|M|{\ensuremath{\Psi}}^{\ensuremath{-}}〉>\frac{1}{2}$.

AI snapshot

Written from the paper’s full text.

TL;DR

Two separated observers can purify noisy shared entanglement by local operations and classical communication, then use the purified pairs for faithful teleportation through noisy channels. The paper gives protocols showing distillable entanglement is nonzero for mixed states with singlet fidelity greater than \(1/2\), with upper and lower bounds on yield.

Key findings

  • For a general mixed two-spin state \(M\), the distillable singlet yield satisfies \(D(M)>0\) if \(\langle \Psi^-|M|\Psi^-\rangle > 1/2\).
  • Given two Werner pairs of fidelity \(F>1/2\), the recurrence protocol produces one Werner pair of fidelity \(F'>F\) with probability greater than \(1/4\).
  • The breeding method has yield \(1-S(W)\), producing more pure pairs than consumed if the mixed state’s von Neumann entropy \(S(W)<1\); for Werner states this yield is positive for \(F>0.8107\).
  • Werner states with \(F>1/2\) are regarded as nonlocal, even though only those with \(F>(2+3\sqrt{2})/8\approx0.78\) violate the CHSH inequality.

Methods

  • Theoretical quantum-information protocol using local unitary operations, local measurements, and classical communication between Alice and Bob; no experimental data are reported.
  • Random bilateral rotations symmetrize an arbitrary two-spin mixed state \(M\) into a Werner state \(W_F\) with the same fidelity \(F\), where \(F=1-3P_\parallel/2\).
  • Purification uses bilateral XOR (BXOR) tests, unilateral Pauli rotations, bilateral \(\pi/2\) rotations, and \(z\)-axis measurements; analysis includes recurrence relations, entropy-counting arguments, and an upper bound from entanglement of formation.

Limitations

  • The paper states that the optimal asymptotic yield \(D(M)\) is not known for general mixed states \(M\), nor even for Werner states, and that no \(M\) is known for which distillable entanglement and entanglement of formation provably differ.
  • The recurrence method’s yield is “rather poor” and tends to zero as \(F_{\text{out}}\to1\); the breeding method requires prepurified \(\Phi^+\) pairs as targets and \(S(W)<1\).
  • Evident from the text: the results are asymptotic and probabilistic, relying on large numbers of pairs and high-probability error location; no experimental implementation is presented.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.1103/physrevlett.76.722

Cites

11 of 11 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Claims the paper makes

0 of 14 cited sources checked · 0 supported · 0 flagged

  • Findingreasoning
    § Purification of Noisy Entanglement and Faithful Teleportation via Noisy Channels · p. 3

    Asymptotically, S(W) BXOR tests per impure pair suffice to find all the errors.

    “Combining these results, and recalling that log 2 ( N 1 N 2 ) = nS ( W ) + O ( √ n ), we obtain the desired result, viz. that asymptotically S ( W ) BXOR tests per impure pair suffice to find all the errors.”
    Quote in the paperSubstantial claim
  • Findingreasoning
    § Purification of Noisy Entanglement and Faithful Teleportation via Noisy Channels · p. 3

    By performing a number of BXOR tests on different subsets of the original impure pairs, all the Ψ states can be found and corrected to Φ states.

    “By performing a number of BXOR tests, on different subsets of the original impure pairs, all the Ψ states can be found and corrected to Φ states.”
    Quote in the paperSubstantial claim
  • Findingreasoning
    § Purification of Noisy Entanglement and Faithful Teleportation via Noisy Channels · p. 3

    For the choice k(F) ≈ 1/√(1-F), the expected fraction of pairs discarded at each step, to lowest order, is 2/3(1-F)^{1/2}.

    “The expected fraction of the pairs discarded at each step, also to lowest order, is 2 3 (1 -F ) 1 2 .”
    Quote in the paperSubstantial claim
  • Findingreasoning
    § Purification of Noisy Entanglement and Faithful Teleportation via Noisy Channels · p. 2

    A random bilateral rotation on each shared pair transforms any initial general two-spin mixed state M into a rotationally symmetric Werner state W_F of the same purity F.

    “Because of the singlet's invariance under bilateral rotations, the symmetrized state W F , which we shall call a Werner state [7] of purity F , has the same F as the initial mixed state M from which it was derived.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [7] R.F. Werner, Phys. Rev. A 40 , 4277 (1989). source

      Not checked in the free run.

  • Findingnone given
    § Purification of Noisy Entanglement and Faithful Teleportation via Noisy Channels · p. 2

    The purity F of a two-spin mixed state relative to a perfect singlet can be computed from the probability P|| of obtaining parallel outcomes if the two spins are measured locally along the same random axis as F = 1 - 3P||/2.

    “one finds that F = 1 -3 P ‖ / 2.”
    Quote in the paperSubstantial claim
  • May combine two claims. 6
  • The quote only partly states this claim. 1
  • Classified as probably not a substantial claim. 3
  • May combine two claims.

    Findingreasoning
    § Purification of Noisy Entanglement and Faithful Teleportation via Noisy Channels · p. 3

    The breeding method has a yield 1 - S(W), producing more pure pairs than consumed if the mixed state's von Neumann entropy S(W) is less than 1.

    “The breeding method has a yield 1 -S ( W ), producing more pure pairs than consumed if the mixed state's von Neumann entropy, S ( W ), is less than 1.”
    Quote in the paperSubstantial claim
  • May combine two claims.

    Findingreasoning
    § Purification of Noisy Entanglement and Faithful Teleportation via Noisy Channels · p. 3

    By BXORing a variable number k(F) ≈ 1/√(1-F) of source pairs into each target pair before measuring it, the yield of the recurrence method can be increased and made to approach a positive limit as F_out → 1.

    “but by BXORing a variable number k ( F ) ≈ 1 / √ 1 -F of source pairs, rather than 1, into each target pair before measuring it, the yield can be increased and made to approach a positive limit as F out → 1.”
    Quote in the paperSubstantial claim
  • May combine two claims.

    Findingreasoning
    § Purification of Noisy Entanglement and Faithful Teleportation via Noisy Channels · p. 4

    F = 1/2 is a threshold below which Werner states can be made from unentangled ingredients, and above which they can be used as a starting material to make pure singlets.

    “We have seen that F = 1 2 is a threshold below which Werner states can be made from unentangled ingredients, and above which they can be used as a starting material to make pure singlets.”
    Quote in the paperSubstantial claim
  • The quote only partly states this claim.

    Findingnone given
    § Purification of Noisy Entanglement and Faithful Teleportation via Noisy Channels · p. 4

    For F > 1/2, W_F can be expressed as an equal mixture of eight pure states each having entropy of entanglement equal to the right side of Eq. (9a).

    “each having entropy of entanglement equal to the right side of eq. 9a”
    Quote in the paperSubstantial claim
  • Classified as probably not a substantial claim.

    Premisecitation
    § Purification of Noisy Entanglement and Faithful Teleportation via Noisy Channels · p. 1

    Quantum information cannot be cloned.

    “Since quantum information cannot be cloned [4]”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [4] W.K. Wootters and W.H.Zurek, Nature 299 , 802 (1982). source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation
    § Purification of Noisy Entanglement and Faithful Teleportation via Noisy Channels · p. 1

    Quantum data compression allows quantum data to be transmitted with asymptotically perfect fidelity using a number of qubits asymptotically approaching the source's von Neumann entropy.

    “quantum data compression [2,3] allows quantum data to be transmitted, with asymptotically perfect fidelity, using a number of 2-state quantum systems or qubits ( e.g. spin1 2 particles) asymptotically approaching the source's von Neumann entropy”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites [2] B. Schumacher, Phys. Rev. A 51 , 2738 (1995). source

      Not checked in the free run.

    • cites [3] R. Jozsa and B. Schumacher, J. Modern Optics 41 , 2343 (1994). source

      Not checked in the free run.

  • May combine two claims.

    Interpretationcitation
    § Purification of Noisy Entanglement and Faithful Teleportation via Noisy Channels · p. 3

    The non-breeding protocol requires only one-way classical communication, allowing it to be used to protect quantum information from errors during storage as well as during transmission.

    “This non-breeding protocol requires only oneway classical communication, allowing it to be used to protect quantum information from errors during storage (cf [5,11]) as well as during transmission.”
    Quote in the paperSubstantial claim0 of 2 sources checked
    • cites [5] P. Shor Phys. Rev. A 52, R2493 (1995). source

      Not checked in the free run.

    • cites [11] C.H. Bennett, D. DiVincenzo, J.A. Smolin, and W.K. Wootters 'Mixed State Entanglement and Quantum Error Correcting Codes', preprint (1995).

      Not checked in the free run.

  • May combine two claims.

    Interpretationreasoning
    § Purification of Noisy Entanglement and Faithful Teleportation via Noisy Channels · p. 1

    The purified entangled pairs can be used to faithfully teleport unknown quantum states from one observer to the other, achieving faithful transmission of quantum information through a noisy channel.

    “These can then be used to faithfully teleport unknown quantum states from one observer to the other, thereby achieving faithful transmission of quantum information through a noisy channel.”
    Quote in the paperSubstantial claim
  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation
    § Purification of Noisy Entanglement and Faithful Teleportation via Noisy Channels · p. 1

    Quantum error-correcting codes have been discovered that operate by embedding the quantum information to be protected in a subspace so oriented in a larger Hilbert space as to leak little or no information to the environment, within a given noise model.

    “Nevertheless, quantum error-correcting codes have recently been discovered [5] which operate in a subtler way, essentially by embedding the quantum information to be protected in a subspace so oriented in a larger Hilbert space as to leak little or no information to the environment, within a given noise model.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [5] P. Shor Phys. Rev. A 52, R2493 (1995). source

      Not checked in the free run.

  • May combine two claims.

    Premisecitation
    § Purification of Noisy Entanglement and Faithful Teleportation via Noisy Channels · p. 4

    These are the least entangled ensembles realizing W_F; therefore E(W_F) may be viewed as the Werner state's entanglement of formation.

    “In fact [11] these are the least entangled ensembles realizing W F ; therefore, E ( W F ) may be viewed as the Werner state's 'entanglement of formation'-the asymptotic number of singlets required to prepare one W F by local actions.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [11] C.H. Bennett, D. DiVincenzo, J.A. Smolin, and W.K. Wootters 'Mixed State Entanglement and Quantum Error Correcting Codes', preprint (1995).

      Not checked in the free run.

10 of 10
  • The paper relies on this without support.

    Two mixed states having the same density matrix are physically indistinguishable.

    “At this point it should be recalled that two mixed states having the same density matrix are physically indistinguishable, even though they may have had different preparations.”
  • The paper relies on this without support.

    The probability distribution P_X over n-bit strings x has a typical set containing N1 = 2^{H(X)+O(√n)} members, where H(X) is the Shannon entropy of P_X.

    “The probability distribution P X over n -bit strings x , where x represents the original sequence of Φ / Ψ values of the impure pairs, receives almost all its weight from a set of 'typical' strings containing N 1 = 2 H ( X )+ O ( √ n ) members, where H ( X ) is the Shannon entropy of P X .”
  • The paper relies on this without support.

    The conditional distribution P_{Y|X=x} of n-bit strings y has a typical set containing N2 = 2^{H(Y|X=x)+O(√n)} members.

    “Similarly, the conditional distribution P Y | X = x of n -bit strings y , representing the + / -values of a sequence of impure pairs whose Φ / Ψsequence is x , receives almost all its weight from a set of typical strings containing N 2 = 2 H ( Y | X = x )+ O ( √ n ) members.”
  • The paper relies on this without support.

    Measuring both spins in a given pair along the z spin axis reliably distinguishes Ψ states from Φ states, but cannot distinguish + from - states.

    “measuring both spins in a given pair along the z spin axis. This reliably distinguishes Ψ states from Φ states, but cannot distinguish + from -states.”
4 of 4

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

  • By using entanglement purification, it becomes possible to teleport the quantum state of a particle even if the available quantum channels are of very poor quality.

    “Furthermore, by using entanglement purification [22] - a scheme of improving the quality of entanglement if it was degraded by decoherence during storage or transmission of the particles over noisy channels - it becomes possible to teleport the quantum state of a particle to some place, even if the available quantum channels are of very poor quality and thus sending the particle itself would very”

    From Experimental quantum teleportation (1997) · cites this paper as [22]

  • Logical purification can be used to remove errors from injected states and is effective if the error of the injected state is less than 0.141.

    “To remove errors from the injected states, logical purification can be used 27, 28 and is effective if the error of the injected state is less than 0 . 141 . 28”

    From Quantum computing with realistically noisy devices (2005) · cites this paper as [27]

  • To complete the Gmin gate set so that arbitrary quantum computations can be implemented, it suffices to add HAD and preparation of the state |π/8〉.

    “To complete the G min gate set so that we can implement arbitrary quantum computations, it suffices to add HAD and preparation of the state | π/ 8 〉 = cos( π/ 8) | 0 〉 + sin( π/ 8) | 1 〉 . 26, 27”

    From Quantum computing with realistically noisy devices (2005) · cites this paper as [27]

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

Each contradiction between two passages listed here was confirmed by an independent check.

  • The stated σy correction reverses the Bell-state signs

    Minor

    The paper first states that σy maps Ψ± to Φ∓. Step B2 instead says it converts Ψ± respectively to Φ±.

1 of 1

Cites

Flagged first, then by citations

Cited by

Sorted by citations

Authors
Bennett, C. H., Brassard, G., Popescu, S., Schumacher, B., Smolin, J. A., & Wootters, W. K.
Affiliations
  • IBM (United States) (US)
  • Tel Aviv University (IL)
  • Kenyon College (US)
  • University of California, Los Angeles (US)
  • Williams College (US)
Venue
Physical Review Letters
Published
1996-01-29
Volume
76
Issue
5
Pages
722-725
ISSN
  • 0031-9007 (ISSN-L)
  • 1079-7114
OpenAlex
W2062258079
Cited by
2,961
References
11
Citations by year
2026: 119 · 2025: 166 · 2024: 150 · 2023: 167 · 2022: 111 · 2021: 107 · 2020: 110 · 2019: 150 · 2018: 105 · 2017: 103 · 2016: 108 · 2015: 105 · 2014: 108 · 2013: 90 · 2012: 110
FWCI
66.84
Type
journal article
Language
en
Open access
Locations
  • Physical Review Letters (publishedVersion)
  • PubMed (publishedVersion)
  • Explore Bristol Research (publishedVersion)
  • arXiv (Cornell University) (submittedVersion)
  • Digital Kenyon (Kenyon College) (submittedVersion)
  • CERN Document Server (European Organization for Nuclear Research) (submittedVersion)
  • arXiv (Cornell University)
Topics
  • Physical Sciences › Computer Science › Artificial Intelligence › Quantum Information and Cryptography
  • Physical Sciences › Physics and Astronomy › Atomic and Molecular Physics, and Optics › Quantum Mechanics and Applications
  • Physical Sciences › Computer Science › Artificial Intelligence › Quantum Computing Algorithms and Architecture
Keywords
entanglement distillation, quantum teleportation, noisy channels, singlet state, mixed states, quantum information transfer, local operations

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Claims read by deepseek-v4.1-flash under definition version 3 (extract-v3) on 2026-10-07.

Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-10-07.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review