30.04.2013 Views

ENTANGLEMENT OF GAUSSIAN STATES Gerardo Adesso

ENTANGLEMENT OF GAUSSIAN STATES Gerardo Adesso

ENTANGLEMENT OF GAUSSIAN STATES Gerardo Adesso

SHOW MORE
SHOW LESS

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

248 14. Gaussian entanglement sharing in non-inertial frames<br />

with uniform accelerations aL and aN, respectively. They have single-mode detectors<br />

sensitive to modes λ and ν, respectively. We consider, that in the inertial<br />

frame all the field modes are in the vacuum except for modes λ and ν which are in<br />

the pure two-mode squeezed state σ p<br />

LN<br />

(s) of the form Eq. (2.22), with squeezing<br />

parameter s as before. Due to their acceleration, two horizons are created so the<br />

entanglement is redistributed among four parties: Leo, anti-Leo (living respectively<br />

in Rindler region I and II of Leo’s horizon), Nadia, anti-Nadia (living respectively<br />

in Rindler region I and II of Nadia’s horizon). These four (some real and some<br />

virtual) parties will detect modes λI, λII, νI, νII, respectively. By the same argument<br />

of Sec. 14.2, the four observers will share a pure, four-mode Gaussian state,<br />

with CM given by<br />

σLLN ¯ N ¯ (s, l, n) = SλI ,λII (l)SνI ,νII (n)SλI ,νI (s) · 8<br />

· S T λI ,νI (s)ST νI ,νII (n)ST λI ,λII (l) ,<br />

(14.16)<br />

where the symplectic transformations S are given by Eq. (2.24), 8 is the CM of<br />

the vacuum |0〉λII ⊗|0〉λI ⊗|0〉νI ⊗|0〉νII , while l and n are the squeezing parameters<br />

associated with the respective accelerations aL and aN of Leo and Nadia [see<br />

Eq. (14.2)]. Explicitly,<br />

⎛<br />

σL¯ εLL ¯ εLN ¯ εL¯ N¯<br />

⎜<br />

σLLN ¯ N ¯ = ⎜ ε<br />

⎝<br />

T LL ¯ σL εLN εLN¯ εT LN ¯ εT LN σN εNN¯ εT L¯ N¯ εT L¯n εT N ¯ ⎞<br />

⎟<br />

⎠ , (14.17)<br />

σ N N¯<br />

where:<br />

σ ¯ X = [cosh 2 (x) + cosh(2s) sinh 2 (x)]2 ,<br />

σX = [cosh 2 (x) cosh(2s) + sinh 2 (x)]2 ,<br />

ε ¯ XX = ε X ¯ X = [cosh 2 (s) sinh(2x)]Z2 ,<br />

ε ¯ XY = ε Y ¯ X = [cosh(y) sinh(2s) sinh(x)]2 ,<br />

ε ¯ X ¯ Y = [sinh(2s) sinh(x) sinh(y)]Z2 ,<br />

εXY = [cosh(x) cosh(y) sinh(2s)]Z2 ,<br />

with Z2 = 1 0<br />

0 −1 ; X, Y = {L, N} with X = Y , and accordingly for the lower-case<br />

symbols x, y = {l, n}.<br />

The infinite acceleration limit (l, n → ∞) can now be interpreted as Leo and<br />

Nadia both escaping the fall into a black hole by accelerating away from it with<br />

acceleration aL and aN, respectively. Their entanglement will be degraded since<br />

part of the information is lost through the horizon into the black hole [GA21]. Their<br />

acceleration makes part of the information unavailable to them. We will show that<br />

this loss involves both quantum and classical information.<br />

14.3.1. Bipartite entanglement<br />

We first recall that the original pure-state contangle Gτ (σ p<br />

L|N ) = 4s2 detected by<br />

two inertial observers is preserved under the form of bipartite four-mode entanglement<br />

Gτ (σ ( ¯ LL)|(N ¯ N)) between the two horizons, as the two Rindler change of<br />

coordinates amount to local unitary operations with respect to the ( ¯ LL)|(N ¯ N) bipartition.<br />

The computation of the bipartite Gaussian contangle in the various 1×1

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!