quant-ph0702264/PLA.tex
1: 
2: %\documentclass[twocolumn,aps,pra,floatfix,showpacs,noshowkeys,epsfig,graphics]{revtex4}%
3: \documentclass[elsart,floatfix,showpacs,noshowkeys,epsfig,graphics]{revtex4}%
4: \usepackage{graphicx}
5: \usepackage{amsmath}
6: \usepackage{amsfonts}
7: \usepackage{amssymb}
8: %
9: 
10: \newcommand{\newc}{\newcommand}
11: \newc{\beq}    {\begin{equation}}
12: \newc{\eeq}    {\end{equation}}
13: 
14: 
15: \newc{\beqa}    {\begin{eqnarray}}
16: \newc{\eeqa}    {\end{eqnarray}}
17: \newc{\bs}    {\section}
18: \newc{\no}    {\\ \nonumber}
19: \def\PL{{\em Phys. Lett.}}
20: \def\PLA{{ Phys. Lett.} {\bf A} }
21: \def\NPB{{\em Nucl. Phys.} {\bf B} }
22: %\def\PRL{{\em Phys. Rev. Lett. }}
23: %\def\PRB{{\em Phys. Rev.} {\bf B} }
24: %\def\PRA{{\em Phys. Rev.} {\bf A} }
25: \def\PRL{{ Phys. Rev. Lett. }}
26: \def\PRB{{ Phys. Rev.} {\bf B} }
27: \def\PRA{{ Phys. Rev.} {\bf A} }
28: 
29: \def\PR{{\em Phys. Rev.}  }
30: \def\PREP{{\em Phys. Rep.}  }
31: \def\RMP{{\em Rev. Mod. Phys. }  }
32: 
33: \topmargin -0mm \hoffset = -0.45cm \voffset = -1.5cm
34: %
35: %\hyphenation{KIAS-Manuscript}
36: %
37: \begin{document}
38: \title{Entanglement of  thermal scalar fields in a compact space}
39: \author{Jae-Weon Lee}
40: \email{scikid@kias.re.kr}
41: \author{Jaewan Kim}
42: \affiliation{School of Computational Sciences,
43:              Korea Institute for Advanced Study,
44:              207-43 Cheongnyangni 2-dong, Dongdaemun-gu, Seoul 130-012, Korea}
45: \author{Taeseung Choi}
46: %\email{tchoi@korea.ac.kr}
47: \affiliation{Department of Physics,
48:              Korea university,Anam-dong Seongbuk-Gu, Seoul, 136-701, Korea}
49: \date{\today}
50: \begin{abstract}
51: Using  the thermal  Green's function approach we propose a general method to investigate entanglement  of the vacuum state or
52: thermal ground states in an arbitrary dimensional space-time.
53: As an application we show quantum separability of
54: the massive thermal scalar field vacuum in the 1+1 dimensional cylindrical space-time. Separability is demonstrated
55: using the positive partial transpose criterion for effective
56: two-mode Gaussian states of collective operators. In this case, for all mass and temperature values
57: entanglement is absent between the collective operators.
58: \end{abstract}
59: 
60: \pacs{03.67.Mn, 03.67.-a, 03.65.Ud, 71.10.Ca}
61: %\keywords{entanglement; solid state qubits; quantum information}
62: \maketitle
63: 
64: Entanglement is now treated as a physical quantity,
65: as well as valuable resource allowing various quantum information processing.
66: This  brings a renewal of interest in studying the non-locality
67: and entanglement in many particle systems such as Bose Einstein condensations~\cite{bec},
68: fermion systems~\cite{Vedral03,oh}, and superconductors~\cite{ohsc}.
69:  Vedral~\cite{Vedral03}
70: studied the entanglement in many body systems at zero temperature
71: using the second quantization formalism. Following his works, Oh
72: and Kim~\cite{oh} studied the entanglement of  two electron spins
73: in a free electron gas, superconductivity~\cite{ohsc} and the Kondo model~\cite{kondo} at finite
74: temperature using thermal Green's function methods.
75: Recently, in the context of AdS/CFT duality~\cite{Maldacena},
76: entanglement
77: entropy~~\cite{ryu:181602,emparan-2006-0606,PhysRevD.34.373} is used
78:  in studying the black hole entropy~\cite{PhysRevD.34.373,solodukhin:201601}.
79: It was also  suggested~\cite{entanglementDE,ourcoincidence}  that entanglement is an origin of
80: mysterious dark energy.
81: However, since entanglement entropy is a good measure of entanglement only for pure states,
82:  we need a reliable method to calculate entanglement for more general states.
83: Thus, studying entanglement of quantum fields vacuum or thermal ground states is of fundamental
84: importance~\cite{reznik}
85: in various physical fields.
86: In ~\cite{my}, the temperature Green's function approach
87: was used to
88: investigate the quantum entanglement of two non-interacting
89:  spin 1 boson particles in  thermal
90: equilibrium.
91: It is also well known that, in algebraic quantum field theory, due to the Reeh-Schlieder theorem~\cite{reeh}
92:  the vacuum
93: for  quantum fields violate Bell inequality and has quantum
94: nonlocality~\cite{werner,narnhofer1}.
95: In Ref. ~\cite{narnhofer1,narnhofer2} entanglements of  Bose and Fermi quasi-free (gaussian) states
96: are extensively studied, while spatial entanglement of
97: free thermal bosonic fields
98: was studied by  post-selecting certain momenta in Ref. ~\cite{heaney-2006}.
99:  On the other hand,  for decades the nature of the  vacuum, concerning the
100: Casimir energy~\cite{casimir,casimir2}, in a bounded space has been
101: extensively studied using the Hadamard Green's function
102: method~\cite{dewitt}.
103: In Ref. ~\cite{our} we showed that this Hadamard Green's function is also useful
104: to study entanglement of vacuum and used the method to investigate entanglement of massless scalar field
105: vacuum with
106: a Dirichlet boundary.
107: Entanglement of the scalar field could be experimentally tested by the scheme with trapped ions~\cite{reznik,reznik:042104}
108:   or Bose-Einstein condensates~\cite{Kaszlikowski}.
109: The entanglement measures such as the purity~\cite{purity} or the
110: negativity has been also suggested~\cite{negativity} for two-mode
111: Gaussian states.
112: Since
113: the scalar fields are continuous variables we need a separability
114: criterion for continuous variables such as Gaussian states.
115:  To utilize positive partial transpose (PPT) criterion for Gaussian mode, we  averaged the fields over two
116: tiny boxes to make the vacuum state an effective two-mode Gaussian
117: state.
118: 
119: In this letter, we first generalize the methods proposed in Ref. ~\cite{our}
120: to arbitrary space-time dimension and use this   Green's function
121: approach  to investigate the  quantum entanglement of
122: the vacuum
123: for
124: a massive thermal scalar field in the 1+1 dimensional cylindrical space-time.
125: Studying the nature  of quantum fields in a compact space is especially important in the string physics.
126: Our methods could be used to study entanglement of continuum limit of harmonic chains
127: in various situations.
128: 
129:  We start by introducing  massive real Klein-Gordon  fields, which are described by a Lagrangian
130:  density,
131:  \beq
132:  \label{L}
133:  \mathcal{L} (\vec{x},t)=\frac{1}{2}( \partial_\alpha\phi
134: \partial^\alpha \phi +m^2 \phi^2),
135: \eeq
136:  in  $(D+1)$ dimensional space-time.
137: The equal time commutation relations of the scalar field are
138: \beqa
139: \label{commutation}
140:  [ \phi(\vec{x},t),\phi(\vec{x}',t)]&=&0 , [ \pi(\vec{x},t),\pi(\vec{x}',t)]=0, \no
141:  [ \phi(\vec{x},t), \pi(\vec{x}',t) ] &=& i \delta(\vec{x}-\vec{x}'),
142:  \eeqa
143: where $\pi(\vec{x},t)\equiv \partial_t \phi(\vec{x},t)$ is a
144: momentum operator for $\phi(\vec{x},t)$. Consider a vector
145: of the field and its momentum operator at two points $\vec{x}$ and
146: $\vec{x}'$
147:  at a given time $t$, i. e., $\xi\equiv (\phi(\vec{x},t),\pi(\vec{x},t),\phi(\vec{x}',t),\pi(\vec{x}',t))$.
148: The variance matrix for the vacuum or the thermal ground state of the scalar field  ($|0\rangle$)
149:  is defined as
150:  \beq
151:  V_{\alpha \beta}\equiv \frac{1}{2}\langle 0|\{\triangle \xi_\alpha,  \triangle \xi_\beta  \}|0\rangle,
152:  \eeq
153:  where $\{A,B\}=AB+BA$  and $\triangle \xi_\alpha\equiv \xi_\alpha-\langle 0| \xi_\alpha|0\rangle$
154:  with $\langle 0| \xi_\alpha|0\rangle=0$ in this paper.
155:  Following Ref. \cite{our}, one can obtain  the variance matrix
156: \begin{eqnarray}
157: V =            \begin{bmatrix}
158:             a & 0 & c  &0 \\
159:             0 & b & 0 &d \\
160:             c & 0 & a'  &0 \\
161:             0 & d & 0 & b'
162:            \end{bmatrix},
163: \end{eqnarray}
164: where $a=\langle0|\{\phi(\vec{x},t),\phi(\vec{x},t)\}|0\rangle/2$,
165: $b=\langle
166: 0|\{\pi(\vec{x},t),\pi(\vec{x},t)\}|0\rangle/2, ~a'=\langle0|\{\phi(\vec{x'},t),\phi(\vec{x'},t)\}|0\rangle/2,$
167: and so on.
168:  Usually  $a$ and $a'$ terms have  a divergence which can be removed
169:  by  subtracting a
170: free-space Green's function $G_0$ from a Green's function $G$ called the Hadamard's
171: elementary function ~\cite{birrell}.  Thus we can define a regularized  Green's
172: function; $ G_R\equiv G-G_0$ which is not divergent.
173: 
174: In Ref. \cite{our} we showed that,
175: once we know the Green's function we can calculate the components of the variance matrix $V$ from it, i.e.,
176:  \beqa
177:  \label{cd}
178: c(\vec{x},\vec{x}')&=&\frac{1}{2} \langle
179: 0|\{\phi(\vec{x},t),\phi(\vec{x}',t)\}|0\rangle =\lim_{t\rightarrow
180: t'}\frac{1}{2} G_R(\vec{x},t;\vec{x}',t'),\no
181: d(\vec{x},\vec{x}')&=&\frac{1}{2}
182: \langle 0|\{\partial_t\phi(\vec{x},t),\partial_t\phi(\vec{x}',t)\}|0\rangle =
183: \lim_{t\rightarrow t'} \partial_t \partial_{t'}
184: \frac{1}{2}G_R(\vec{x},t;\vec{x}',t').
185:  \eeqa
186: Then, $ a= \lim_{\vec{x}' \rightarrow \vec{x}}
187: c(\vec{x},\vec{x}')$, $ b= \lim_{\vec{x}'\rightarrow \vec{x}}$
188: $ d(\vec{x},\vec{x}')$ and so on.
189: 
190: 
191: \begin{figure}[htbp]
192: \includegraphics[width=0.4\textwidth]{fig1.eps}
193: \caption{Cylindrical 1+1 dimensional space time with spatial circumstance $R$.
194:  we average  fields over two tiny boxes $B$ and $B'$  centered at $x$ and $x'$, respectively.
195: \label{ccorrelation} }
196: \end{figure}
197:  Since `if and only
198: if ' separability test for infinite-mode states are unknown, we
199: need to
200:  reduce the infinite-mode states to effective two-mode Gaussian states by defining collective
201: operators. To define collective operators, we follow
202: the approach in Ref. ~\cite{kofler:052107}, i.e., at a given time $t$
203: we spatially average the field operator over  two tiny spatial boxes $B$ and $B'$  centered at $\vec{x}_0$ and $\vec{x'}_0$,
204: respectively (See Fig. 1).
205: Defining collective operators in this manner is physically reasonable, because
206: in a real situation probes always have  finite spatial resolution.
207: We do not average but integrate the momentum operators within the box, since momentum is additive.
208: ( However, averaging the momentum gives the same final results in this work.)
209:   Thus the collective operators are
210:   \beqa
211: \Xi&\equiv& \left(\Phi(\vec{x},t),\Pi(\vec{x},t),\Phi(\vec{x'},t),\Pi(\vec{x'},t)\right)\no
212: &=&\left(\frac{1}{L^D}\int_Bd^D \vec{y} \phi(\vec{x}+\vec{y},t),
213:   \int_Bd^D \vec{y} \pi(\vec{x}+\vec{y},t),
214:   \frac{1}{L^D}\int_{B'}d^D\vec{y'} \phi(\vec{x'}+\vec{y'},t),
215:   \int_{B'}d^D\vec{y'} \pi(\vec{x'}+\vec{y'},t)\right).
216:   \eeqa
217: 
218: Here $\int_B d^D\vec{y} f(\vec{x}+\vec{y})$ denotes an integration of $f(\vec{x})$ over a box $B$
219:  centered at $\vec{x}$ with volume $L^D$.
220:   Then, the commutation relations in Eq. (\ref{commutation})
221:   reduce to the  canonical commutation relations for the collective operators  ~\cite{kofler:052107}
222: \beq
223: [{\Xi}_\alpha,{\Xi}_\beta]=i\Omega_{\alpha\beta},
224: \eeq
225: where
226: \begin{eqnarray}
227: \Omega =            \begin{bmatrix}
228:             J & 0  \\
229:             0 & J \\
230:            \end{bmatrix}\,,
231:            J =            \begin{bmatrix}
232:             0 & 1  \\
233:             -1 & 0 \\
234:            \end{bmatrix}\,.
235: \end{eqnarray}
236: Similarly we define a variance matrix for the collective operators
237: $\tilde{V}_{\alpha \beta}\equiv \frac{1}{2}\langle 0|\{ {\Xi}_\alpha,  {\Xi}_\beta  \}|0\rangle.$
238: 
239: 
240: Now we discuss separability of the vacuum.
241: Using the fundamental theorem of calculus~\cite{calculus},
242: one can calculate the  variance matrix $\tilde{V}$ for the two-mode Gaussian states
243: with those of $V$ in the limit $L\rightarrow 0$ (but still non-zero).
244: For example,
245: \beqa
246: \tilde{V}_{13}&=&\lim_{L\rightarrow 0}\frac{1}{2}\langle 0|\{ \Phi(\vec{x},t),  \Phi(\vec{x'},t)\}|0\rangle \no
247: &=&\lim_{L\rightarrow 0} \frac{1}{2L^{2D}}  \int_Bd^D\vec{y}\int_{B'}d^D\vec{y}'\langle 0|\{ \phi(\vec{x}+\vec{y},t),
248:  \phi(\vec{x'}+\vec{y}',t)\}|0\rangle \no
249: &=&\frac{1}{2}\langle 0|\{ \phi(\vec{x},t), \phi(\vec{x'},t)\}|0\rangle
250: =V_{13}=c.
251: \eeqa
252: This is always possible when  the integrand of the second line
253:  is continuous.
254:  Similarly,
255: \beqa
256: \tilde{V}_{24}&=&
257: \lim_{L\rightarrow 0}  L^{2D}\frac{1}{2L^{2D}}  \int_Bd^D\vec{y}\int_{B'}d^D\vec{y}'\langle 0|\{ \pi(\vec{x}+\vec{y},t),
258:  \pi(\vec{x'}+\vec{y}',t)\}|0\rangle =L^{2D} V_{24}=L^{2D} d.
259: \eeqa
260: Hence,
261: \begin{eqnarray}
262: \label{Vnew}
263: \tilde{V} =
264:              \begin{bmatrix}
265:             a & 0 & c  &0 \\
266:             0 &L^{2D} b & 0 & L^{2D} d \\
267:             c & 0 & a'  &0 \\
268:             0 & L^{2D} d & 0 & L^{2D} b'
269:            \end{bmatrix}\,
270: \equiv
271:               \begin{bmatrix}
272:             A & G  \\
273:             G^T & B  \\
274:            \end{bmatrix}.
275: \end{eqnarray}
276: For $L$ infinitesimally small the above equation
277: gives an exact  result, while for $L \ll 1$ it provides us a good approximation.
278: The separability  criterion we use in this paper is  PPT criterion~\cite{peres,ppt}
279: for two-mode Gaussian states~\cite{simon}
280: which is equivalent to
281: \beq
282: F\equiv \tilde{\Sigma}-(\frac{1}{4}+4 det \tilde{V})\leq 0,
283: \label{F}
284: \eeq
285: where $\tilde{\Sigma}=det A +det B-2det G$.
286: By inserting the components of $\tilde{V}$ into Eq. (\ref{F}) we obtain
287: \beq
288: F=-\frac{1}{4}+L^{2D}[ab+a'b'-2cd]-4L^{4D}[aa'bb'-aa'd^2-c^2bb'+c^2d^2].
289: \eeq
290: Now the quantum separability can be tested for an arbitrary space-time
291: using this quantity once the Hadamard's elementary function is known.
292: Furthermore, if $a=a'$ and $b=b'$ one can immediately obtain an entanglement measure called the negativity
293: using Eq. (9) in Ref. ~\cite{PhysRevA.66.030301};
294: \beq
295: E(\rho)=max\left\{0,\frac{1}{L^{2D}(a-|c|)(b-|d|)}-1\right\}.
296: \eeq
297: Thus, we can easily determine how much a given state has entanglement, once we know the Green's function
298: at least for the collective operators. This is very useful results since the Hadamard's function is well known
299: for many situations.
300: 
301: As an application, we test the quantum separability of the massive thermal scalar fields vacuum
302: in 1+1 dimensional cylindrical space-time using the methods described above.
303: This model is related to one-dimensional harmonic chain systems with periodic boundary condition
304: \cite{Cramer:2005mx,Plenio:2004he,kofler:052107}.
305: First let us choose two
306: points $x$ and $x'$ (see Fig. 1). Using the
307: method of images~\cite{image} one can obtain the thermal Hadamard Green's function for this field
308: ~\cite{birrell}
309: \beq
310: G = \frac{-1}{4 \pi} ln\left(16 \left[ sin(\pi (\Delta u+i \beta m)/R)sin(\pi (\Delta v+i \beta m)/R)\right]^2\right),
311:  \eeq
312: where $u \equiv -x + t $,
313: $ v \equiv x  + t$
314: are the light-cone coordinates, and $\Delta u=-(x'-x)+(t'-t)\equiv -\Delta x +\Delta t$.
315:  This Green's function diverges as $R \rightarrow \infty$;
316: \beq
317: \lim_{R\rightarrow \infty} G\sim -\left( \frac{\ln (\frac{1}{R})}{\pi } \right)  - \frac{\ln (16\,
318:       {\left( \pi{\Delta t}  - \pi{\Delta x}  + i \pi\,m\, \,\beta  \right) }^2\,
319:       {\left( \pi{\Delta t}  + \pi {\Delta x} + i \pi\,m\, \,\beta  \right) }^2)}{4\,\pi }\equiv G_0.
320: \eeq
321: Interestingly, the Green's function is a function of $m\beta\equiv M$.
322: As mentioned above by subtracting  $G_0$
323: from $G$ one can define a regularized  Green's
324: function $ G_R\equiv G-G_0$, from which one can obtain
325: \beqa
326: & &(a,b,c,d)= \no
327: & &\left( \frac{1}{2\,\pi }{\ln \left(\frac{\pi m\, \,\beta \,csch(\frac{\pi m\, \,\beta }{R})}{R}\right)},\frac{-\pi }{6\,R^2},
328:  \frac{1}{8\,\pi }{\ln \left(\frac{16\,{\pi }^4\,{\left( {{\Delta x}}^2 + m^2\,{\beta }^2 \right) }^2}{4R^4\,
329: {\left( \cos (\frac{2\,{\Delta x}\,\pi }{R}) -
330:   \cosh (\frac{2\,m\,\pi \,\beta }{R}) \right) }^2}\right)},
331:   \frac{1}{2\,\pi }\left({{\Delta x}}^{-2} - \frac{{\pi }^2\,{\csc^2 (\frac{{\Delta x}\,\pi }{R})}}{R^2}\right)\right)
332: \eeqa
333: using Eq. (\ref{cd}) after some tedious calculation.
334: By inserting these components into Eq. (\ref{F}) we obtain
335: \beq
336:  F=-\frac{1}{4} +L^2 f_2(\Delta x, M) +L^4 f_4(\Delta x, M) ,
337: \eeq
338: where
339: \beq
340: f_2(\Delta x, M) \equiv -\frac{ \left( -1 + 2\,{\pi }^2\,{{\Delta
341: x}}^2 + \cos (2\,\pi \,{\Delta x}) \right) \,
342:        {\csc^2 (\pi \,{\Delta x})}\,\ln \left(\frac{{\left( \cos (2\,\pi \,{\Delta x}) -
343:               \cosh (2\,M\,\pi ) \right) }^2}{4\,{\pi }^4\,{\left( M^2 + {{\Delta x}}^2 \right) }^2}\right)
344:         }{16\,{\pi }^2\,{{\Delta x}}^2} - \frac{\ln (M\,\pi \,{csch}(M\,\pi ))}{6}
345: \eeq
346: and
347: \beq
348: f_4(\Delta x, M) \equiv
349: \frac{-1}{576\,{\pi }^4}{\left(  9{\left( {{\Delta x}}^{-2} - {\pi }^2\,{\csc (\pi \,{\Delta x})}^2
350:            \right) }^2 -{\pi }^4\right) \,
351:        \left( \left[\ln \left(\frac{{\left( \cos (2\,\pi \,{\Delta x}) -
352:                \cosh (2\,M\,\pi ) \right) }^2}{4\,{\pi }^4\,{\left( M^2 + {{\Delta x}}^2 \right) }^2}\right)\right]^2 -
353:       16\,[\ln (M\,\pi \,{csch}(M\,\pi ))]^2 \right) }.
354: \eeq
355: Here we have set $R=1$, hence from now on all length scales are in units of $R$.
356:  \begin{figure}[htbp]
357: \includegraphics[width=0.4\textwidth]{fig2.eps}
358: \caption{$F$ as a function of $\Delta x$ and $M$ for L=0.01.
359:  \label{Fig2} }
360: \end{figure}
361: 
362: Our numerical study shows that $f_1$ has a
363: maximum $f_1(1/2,0)=0.134$ and  $f_2$ has a
364: maximum $f_2(1/2,1.107)=0.0164$.
365:  Hence, one can see that
366: \beq
367: F<-\frac{1}{4} +0.134 ~ L^2 +0.0164 L^4 <0
368: \eeq
369: for $0<L<1/2$.
370: Therefore, the effective two-mode scalar field vacuum with the periodic boundary is PPT and, hence, separable,
371: for all $M$ and $\Delta x$.
372: Fig. 2 shows $F$ as a function of $\Delta x$ and $M$ for $L=0.01$.
373: 
374: 
375: Our results  implies that changing mass, temperature, and periodic boundary condition
376: does not induce additional entanglement to the possible entanglement of the
377:  zero temperature free space scalar fields vacuum.
378: Our results also indicates that the entanglement of generic vacuum resides in the
379: free space Green's function $G_0$ which usually gives rise to infinite quantities.
380: Our approach
381: provides a new and generic method  to
382: investigate entanglement of the quantum fields vacuum or thermal ground states in a various situation
383:  using the Green's function method. Since the Hadamard's functions are known for many cases
384: our approach can be useful.
385: 
386: 
387: \vskip 15pt
388: \paragraph*{Acknowledgments.--}
389: \indent
390: Authors are thankful to Jinhyoung Lee for helpful discussions.
391: J. Lee and
392: J. Kim
393: was supported by the Korea Ministry of Information and Communication
394: with the``Next Generation Security Project".
395:  T. Choi was supported by the SRC/ERC program of MOST/KOSEF
396: (R11-2000-071), the Korea Research Foundation Grant
397: (KRF-2005-070-C00055), the SK Fund, and the KIAS.
398: 
399: 
400: 
401: %\bibliographystyle{elsart-num}
402: %\bibliography{entanglement}% Produces the bibliography via BibTeX.
403: \begin{thebibliography}{10}
404: \expandafter\ifx\csname url\endcsname\relax
405:   \def\url#1{\texttt{#1}}\fi
406: \expandafter\ifx\csname urlprefix\endcsname\relax\def\urlprefix{URL }\fi
407: 
408: \bibitem{bec}
409: A.~Sorensen, L.-M. Duan, J.~I. Cirac, P.~Zoller, Nature 409 (2001) 63.
410: 
411: \bibitem{oh}
412: S.~Oh, J.~Kim, Phys. Rev. A 69 (2004) 054305.
413: 
414: \bibitem{Vedral03}
415: V.~Vedral, Central Eur. J. Phys. 1 (2003) 289.
416: 
417: \bibitem{ohsc}
418: S.~Oh, J.~Kim, Phys. Rev. B 71 (2005) 144523.
419: 
420: \bibitem{kondo}
421: S.~Oh, J.~Kim, Phys. Rev. B 73 (2006) 052407.
422: 
423: \bibitem{Maldacena}
424: O.~Aharony, S.~S. Gubser, J.~Maldacena, H.~Ooguri, Y.~Oz, Phys. Rep. 323~(3-4) (2000) 183--386.
425: 
426: \bibitem{PhysRevD.34.373}
427: L.~Bombelli, R.~K. Koul, J.~Lee, R.~D. Sorkin, Phys. Rev. D 34~(2) (1986) 373--383.
428: 
429: \bibitem{emparan-2006-0606}
430: R.~Emparan,  JHEP 0606 (2006) 012.
431: 
432: \bibitem{ryu:181602}
433: S.~Ryu, T.~Takayanagi, Phys. Rev. Lett.
434:   96~(18) (2006) 181602.
435: 
436: \bibitem{solodukhin:201601}
437: S.~N. Solodukhin, Phys. Rev. Lett. 97~(20) (2006)
438:   201601.
439: 
440: \bibitem{ourcoincidence}
441: H.-C. Kim, J.-W. Lee, J.~Lee, eprint:hep-th/0702121.
442: 
443: \bibitem{entanglementDE}
444: J.-W. Lee, J.~Lee, , H.-C. Kim,  eprint:hep-th/0701199.
445: 
446: \bibitem{kofler:052107}
447: J.~Kofler, V.~Vedral, M.~S. Kim, C.~Brukner, Phys. Rev. A 73~(5) (2006) 052107.
448: 
449: \bibitem{reznik}
450: A.~Retzker, J.~I. Cirac, B.~Reznik, Phys. Rev. Lett. 94 (2005) 050504.
451: 
452: \bibitem{my}
453: J.-W. Lee, S.~Oh, J.~Kim, Phy. Lett. A 363,374.
454: 
455: \bibitem{reeh}
456: H.~Reeh, S.~Schlieder, Nuovo Cimento 22 (1961) 1051.
457: 
458: \bibitem{narnhofer1}
459: H.~Narnhofer, Phys. Lett. A 310 (2003) 423.
460: 
461: \bibitem{werner}
462: S.~Summers, R.~Werner, Phys. Lett. A 110 (1985) 257.
463: 
464: \bibitem{narnhofer2}
465: H.~Narnhofer, Phys. Rev. A 71 (2005) 052326.
466: 
467: \bibitem{heaney-2006}
468: L.~Heaney, J.~Anders, V.~Vedral,
469:   eprint:quant-ph/0607069.
470: 
471: \bibitem{casimir2}
472: M.~Bordag, U.~Mohideen, V.~Mostepanenko, Phys. Rept. 353 (2001) 1.
473: 
474: \bibitem{casimir}
475: H.~Casimir, Proc. K. Ned. Akad. Wet. 51 (1948) 793.
476: 
477: \bibitem{dewitt}
478: B.~S. DeWitt, Phys. Rept. 19 (1975)
479:   295.
480: 
481: \bibitem{our}
482: J.-W. Lee, J.~Kim, T.~Choi,  Phys. Rev. A 74~(4) (2006) 044304.
483: 
484: \bibitem{purity}
485: G.~Adesso, A.~Serafini, F.~Illuminati, Phys. Rev. Lett. 92 (2004) 087901.
486: 
487: \bibitem{negativity}
488: M.~S. Kim, J.~Lee, W.~J. Munro,  Phys. Rev. A 66~(3) (2002) 030301.
489: 
490: \bibitem{birrell}
491: N.~Birrell, P.~Davies, Quantum fields in curved space, Cambridge University
492:   Press, 1982.
493: 
494: \bibitem{calculus}
495: W.~Rudin, Principles of Mathematical Analysis, McGraw-Hill, 1976.
496: 
497: \bibitem{ppt}
498: M.~Horodecki, P.~Horodecki, R.~Horodecki, Phys. Lett. A 223 (1996) 1.
499: 
500: \bibitem{peres}
501: A.~Peres, Phys. Rev. Lett. 77 (1996) 1413.
502: 
503: \bibitem{simon}
504: R.~Simon,  Phys. Rev. Lett. 84~(12) (2000) 2726--2729.
505: 
506: \bibitem{PhysRevA.66.030301}
507: M.~S. Kim, J.~Lee, W.~J. Munro,  Phys. Rev. A 66~(3) (2002) 030301.
508: 
509: \bibitem{Cramer:2005mx}
510: M.~Cramer, J.~Eisert, M.~B. Plenio, J.~Dreissig,  Phys. Rev. A73 (2006) 012309.
511: 
512: \bibitem{Plenio:2004he}
513: M.~B. Plenio, J.~Eisert, J.~Dreissig, M.~Cramer,  Phys. Rev. Lett. 94
514:   (2005) 060503.
515: 
516: \bibitem{image}
517: L.~S. Brown, G.~J. Maclay, Phys. Rev. 184~(5) (1969) 1272--1279.
518: 
519: \bibitem{reznik:042104}
520: B.~Reznik, A.~Retzker, J.~Silman,
521:   Phys. Rev. A 71~(4) (2005) 042104.
522: 
523: \bibitem{Kaszlikowski}
524: D.~Kaszlikowski, V.~Vedral, eprint:quant-ph/0606238.
525: 
526: \end{thebibliography}
527: 
528: 
529: \end{document}
530: