1: \documentclass[prd,showpacs,preprintnumbers,floatfix,twocolumn]{revtex4}
2: \usepackage{amsmath}
3: \usepackage{graphicx}
4: \begin{document}
5: \title{Quantum Entanglement of Electromagnetic Field in Non-inertial Reference Frames}
6: \date{\today }
7: \author{Yi Ling}
8: \email{yling@ncu.edu.cn} \affiliation{Center for Gravity and
9: Relativistic Astrophysics, Department of Physics, Nanchang
10: University, Nanchang, 330047, China\\
11: CCAST (World Laboratory), P.O. Box 8730, Beijing,
12: 100080, China}
13: \author{Song He}
14: \affiliation{Institute of Theoretical Physics, School of Physics, Peking
15: University, Beijing, 100871, China}
16: \author{Weigang Qiu}
17: \affiliation{Department of Physics, Huzhou Teachers College, Huzhou,
18: 313000, China}
19: \author{Hongbao Zhang}
20: \affiliation{Department of Astronomy, Beijing Normal University, Beijing, 100875,
21: China\\
22: Department of Physics,
23: Beijing Normal University, Beijing, 100875, China\\
24: CCAST (World
25: Laboratory), P.O. Box 8730, Beijing,
26: 100080, China}
27: \begin{abstract}
28: Recently relativistic quantum information has received considerable
29: attention due to its theoretical importance and practical
30: application. Especially, quantum entanglement in non-inertial
31: reference frames has been studied for scalar and Dirac fields. As a
32: further step along this line, we here shall investigate quantum
33: entanglement of electromagnetic field in non-inertial reference
34: frames. In particular, the entanglement of photon helicity entangled
35: state is extensively analyzed. Interestingly, the resultant
36: logarithmic negativity and mutual information remain the same as
37: those for inertial reference frames, which is completely different
38: from that previously obtained for the particle number entangled
39: state.
40: \end{abstract}
41:
42: \pacs{03.67.Mn 03.65.Vf 03.65.Yz} \maketitle
43:
44: \section{Introduction}
45: Quantum entanglement is both the central concept and the major
46: resource in quantum information science such as quantum
47: teleportation and quantum computation\cite{BEZ}. In recent years,
48: tremendous progress has been made in the research on quantum
49: entanglement: not only have remarkable results been obtained in this
50: field, but also important techniques been applied to various
51: circumstances\cite{PV}.
52:
53: Especially, considerable effort has been expended on the
54: investigation of quantum entanglement in the relativistic framework
55: recently\cite{AET,PT,Shi}. A key issue in this intriguing and active
56: research direction is whether quantum entanglement is
57: observer-dependent. It has been shown that quantum entanglement
58: remains invariant between inertial observers with relative motion in
59: flat spacetime although the entanglement between some degrees of
60: freedom can be transferred to others\cite{Peres1,AM1,GA,He}.
61: However, for scalar and Dirac fields, the degradation of
62: entanglement will occur from the perspective of a uniformly
63: accelerated observer, which essentially originates from the fact
64: that the event horizon appears and Unruh effect results in a loss of
65: information for the non-inertial observer\cite{AM2,AM3,FM,AFMT}.
66:
67: As a further step along this line, this paper will provide an
68: analysis of quantum entanglement of electromagnetic field in
69: non-inertial reference frames. In particular, we here choose the
70: photon helicity entangled state
71: $\frac{1}{\sqrt{2}}(|\uparrow\rangle_A|\downarrow\rangle_B+|\downarrow\rangle_A|\uparrow\rangle_B)$
72: rather than the particle number entangled state
73: $\frac{1}{\sqrt{2}}(|0\rangle_A|0\rangle_B+|1\rangle_A|1\rangle_B)$
74: in an inertial reference frame as our main point for investigation
75: of quantum entanglement in non-inertial reference frames, where A
76: and B represent an inertial observer Alice, and a uniformly
77: accelerated observer Bob respectively, as is illustrated in
78: FIG.\ref{AliceBob}. It thus makes the present work acquire much
79: interest and significance: the former entangled state seems to be
80: more popular in quantum information science, but previous work only
81: restricts within the latter setting\cite{AM2,AM3,FM,AFMT}. In
82: addition, the result obtained here shows that although Bob is forced
83: to trace over a causally disconnected region of spacetime that he
84: can not access due to his acceleration, which also leads his
85: description of the helicity entangled state to take the form of a
86: mixed state; the corresponding logarithmic negativity and mutual
87: information both remain invariant against the acceleration of Bob.
88: Therefore our result is of remarkable novelty: it is completely
89: different from those obtained for the case of the particle number
90: entangled state, where the degradation of entanglement is dependent
91: on the acceleration of observer, namely, the larger the
92: acceleration, the larger the degradation\cite{AM2,AM3,FM,AFMT}.
93:
94: The paper is organized as follows. In the next section, we shall
95: briefly review the four disconnected sectors in Minkowski spacetime
96: and the accelerated observers in Rindler spacetime. In the
97: subsequent section, introducing the two sets of expansion bases for
98: quantizing the electromagnetic field in Minkowski spacetime, we have
99: developed the relationship between the corresponding annihilation
100: and creation operators in Minkowski spacetime. In Section \ref{QE},
101: we shall analyze quantum entanglement of electromagnetic field in
102: non-inertial reference frames, especially for the photon helicity
103: entangled state . Conclusions and discussions are presented in the
104: last section.
105:
106: System of natural units are adopted: $\hbar=c=1$. In addition, the
107: metric signature takes $(+,-,-,-)$, and the Lorentz gauge condition
108: $\nabla_aA^a=0$ is imposed onto the electromagnetic potential in
109: flat spacetime, where Maxwell equation reads
110: \begin{equation}
111: \nabla_a\nabla^aA_b=0.\label{Maxwell}
112: \end{equation}
113: Moreover, the well known inner product is reduced to
114: \begin{equation}
115: (A,A')=i\int_\Sigma[\nabla^a\bar{A}^b)A'_b-\bar{A}_b\nabla^aA'^b]\epsilon_{acde},\label{inner}
116: \end{equation}
117: which is gauge invariant and independent of the choice of Cauchy
118: surface $\Sigma$\cite{Moretti,HQZ}.
119: \section{Accelerated Observers in Minkowski Spacetime}
120: Start from Minkowski spacetime
121: \begin{equation}
122: ds^2=dt^2-dx^2-dy^2-dz^2.
123: \end{equation}
124: As is shown in FIG. \ref{AliceBob}, we perform the coordinate
125: transformations for the four disconnected sectors in Minkowski
126: spacetime, respectively, i.e.,
127: \begin{eqnarray}
128: &R&\nonumber\\
129: t=\rho\sinh\tau,&&x=\rho\cosh\tau,\nonumber\\
130: \rho=\sqrt{x^2-t^2},&&\tau=\tanh^{-1}(\frac{t}{x}),
131: \end{eqnarray}
132: \begin{eqnarray}
133: &L&\nonumber\\
134: t=\rho\sinh\tau,&&x=\rho\cosh\tau,\nonumber\\
135: \rho=-\sqrt{x^2-t^2},&&\tau=\tanh^{-1}(\frac{t}{x}),
136: \end{eqnarray}
137: \begin{eqnarray}
138: &F&\nonumber\\
139: t=\rho\cosh\tau,&&x=\rho\sinh\tau,\nonumber\\
140: \rho=\sqrt{t^2-x^2},&&\tau=\tanh^{-1}(\frac{x}{t}),
141: \end{eqnarray}
142: \begin{eqnarray}
143: &P&\nonumber\\
144: t=\rho\cosh\tau,&&x=\rho\sinh\tau\nonumber\\
145: \rho=-\sqrt{t^2-x^2},&&\tau=\tanh^{-1}(\frac{x}{t}).
146: \end{eqnarray}
147: In particular, the $R$($L$) sector, viewed as a spacetime in its own
148: right, is also called $R$($L$) Rindler spacetime, where the metric
149: reads
150: \begin{equation}
151: ds^2=\rho^2d\tau^2-d\rho^2-dy^2-dz^2,
152: \end{equation}
153: and the integral curves of boost Killing field
154: $(\frac{\partial}{\partial\tau})^a$ correspond to the worldlines of
155: accelerated observers with proper time $\rho\tau$ and acceleration
156: $\frac{1}{\rho}$.
157: \begin{figure}
158: \includegraphics[width=2.5inch]{AliceBob.eps}\\
159: \caption{The four disconnected patches in Minkowski spacetime with an inertial observer Alice and a uniformly accelerated observer Bob constrained in $R$ sector. }\label{AliceBob}
160: \end{figure}
161: \section{Quantum Electromagnetic Field in Minkowski Spacetime}
162: As is well known, the quantum fields can be expanded in terms of
163: various bases, but the corresponding vacua may be completely
164: different. For the quantum electromagnetic field in Minkowski
165: spacetime, we firstly choose the expansion basis as
166: \begin{eqnarray}
167: &&A_\mu(\omega\in R,p_y\in R,p_z\in R,
168: s=\pm 1)=\nonumber\\
169: &&\frac{1}{8\pi^2p_\bot}[(0,0,p_z\phi,-p_y\phi)+s(\partial_x\phi,\partial_t\phi,0,0)],
170: \end{eqnarray}
171: where $p_\bot=\sqrt{p_y^2+p_z^2}$, and
172: \begin{equation}
173: \phi=\int_{-\infty}^\infty d\lambda
174: e^{(-i\omega\lambda-ip_\bot\cosh\lambda t+ip_\bot\sinh\lambda
175: x+ip_yy+ip_zz)}
176: \end{equation}
177: satisfies Klein-Gordon equation in Minkowski spacetime, with
178: $\omega$ a dimensionless parameter\cite{Moretti,Colosi}.
179:
180: It is easy to check that $A_\mu(\omega,p_y,p_z,s)$ is the
181: simultaneous eigensolution of boost, transverse momentum, and
182: helicity operators with the corresponding eigenvalues $\{\omega,p_y,
183: p_z, s\}$ in Minkowski spacetime\cite{HQZ,Ashtekar}. Furthermore, it
184: is orthonormal with respect to the inner product (\ref{inner}),
185: i.e.,
186: \begin{eqnarray}
187: &&(A(\omega,p_y,p_z,s),A(\omega',p'_y,p'_z,s'))=\nonumber\\
188: &&\delta(\omega-\omega')\delta(p_y-p'_y)\delta(p_z-p'_z)\delta_{ss'}.
189: \end{eqnarray}
190: Thus in terms of this basis, the quantum electromagnetic field can
191: be expanded as
192: \begin{eqnarray}
193: \hat{A}_\mu=&&\int_{-\infty}^\infty d\omega\int_{-\infty}^\infty
194: dp_y\int_{-\infty}^\infty dp_z\sum_{s=\pm1}\nonumber\\
195: &&[c(\omega,p_y,p_z,s)A_\mu(\omega,p_y,p_z,s)\nonumber\\
196: &&+c^\dagger(\omega,p_y,p_z,s)\bar{A}_\mu(\omega,p_y,p_z,s)],
197: \end{eqnarray}
198: where $c$ and $c^\dagger$ are the corresponding annihilation and
199: creation operators, respectively, adjoint to each other, and
200: satisfying the following commutation relations
201: \begin{equation}
202: [c(\omega,p_y,p_z,s),c(\omega',p'_y,p'_z,s')]=0,
203: \end{equation}
204: \begin{equation}
205: [c^\dagger(\omega,p_y,p_z,s),c^\dagger(\omega',p'_y,p'_z,s')]=0,
206: \end{equation}
207: \begin{eqnarray}
208: &&[c(\omega,p_y,p_z,s),c^\dagger(\omega',p'_y,p'_z,s')]=\nonumber\\
209: &&\delta(\omega-\omega')\delta(p_y-p'_y)\delta(p_z-p'_z)\delta_{ss'}.
210: \end{eqnarray}
211:
212: Next we can also employ Unruh expansion basis for the quantum
213: electromagnetic field, i.e.,
214: \begin{eqnarray}
215: &&R_\mu(\omega\in R^+,p_y,p_z,s)=
216: \frac{1}{\sqrt{2\sinh(\pi\omega})}\nonumber\\
217: &&[e^{(\frac{\pi\omega}{2})}A_\mu(\omega,p_y,p_z,s)-e^{(-\frac{\pi\omega}{2})}\bar{A}_\mu(-\omega,-p_y,-p_z,s)],\nonumber\\
218: \\
219: &&L_\mu(\omega\in R^+,p_y,p_z,s)=\frac{1}{\sqrt{2\sinh(\pi\omega})}\nonumber\\
220: &&[e^{(\frac{\pi\omega}{2})}A_\mu(-\omega,p_y,p_z,s)-e^{(-\frac{\pi\omega}{2})}\bar{A}_\mu(\omega,-p_y,-p_z,s)],\nonumber\\
221: \end{eqnarray}
222: where $R_\mu$ vanishes in the $L$ sector, and $L_\mu$ vanishes in
223: the $R$ sector. It is noteworthy that $R_\mu(\omega\in
224: R^+,p_y,p_z,s)$($L_\mu(\omega\in R^+,p_y,p_z,s)$) is the
225: simultaneous eigenstate of energy, transverse momentum, and helicity
226: operators with eigenvalues of $\{a\omega,p_y,p_z,s\}$ detected by an
227: observer with uniform acceleration $a$ in the $R$($L$) Rindler
228: spacetime\cite{HQZ,Ashtekar}. Moreover, with respect to the inner
229: product (\ref{inner}), Unruh basis is orthonormal, i.e.,
230: \begin{eqnarray}
231: &&(R(\omega,p_y,p_z,s),R(\omega',p'_y,p'_z,s'))=\nonumber\\
232: &&\delta(\omega-\omega')\delta(p_y-p'_y)\delta(p_z-p'_z)\delta_{ss'},\\
233: &&(L(\omega,p_y,p_z,s),L(\omega',p'_y,p'_z,s'))=\nonumber\\
234: &&\delta(\omega-\omega')\delta(p_y-p'_y)\delta(p_z-p'_z)\delta_{ss'},\\
235: &&(R(\omega,p_y,p_z,s),L(\omega',p'_y,p'_z,s'))=0.
236: \end{eqnarray}
237: Whence the quantum electromagnetic field can be reformulated as
238: \begin{eqnarray}
239: \hat{A}_\mu=&&\int_0^\infty d\omega\int_{-\infty}^\infty
240: dp_y\int_{-\infty}^\infty dp_z\sum_{s=\pm1}\nonumber\\
241: &&[r(\omega,p_y,p_z,s)R_\mu(\omega,p_y,p_z,s)\nonumber\\
242: &&+r^\dagger(\omega,p_y,p_z,s)\bar{R}_\mu(\omega,p_y,p_z,s)\nonumber\\
243: &&+l(\omega,p_y,p_z,s)L_\mu(\omega,p_y,p_z,s)\nonumber\\
244: &&+l^\dagger(\omega,p_y,p_z,s)\bar{L}_\mu(\omega,p_y,p_z,s)].
245: \end{eqnarray}
246: Here $r$ and $r^\dagger$ are the corresponding annihilation and
247: creation operators for the $R$ Rindler spacetime; similarly, $l$ and
248: $l^\dagger$ are the corresponding annihilation and creation
249: operators for the $L$ Rindler spacetime. They satisfy the ordinary
250: commutation relations as $c$ and $c^\dagger$ do. Furthermore, they
251: can be related to $c$ and $c^\dagger$ by Bogoliubov transformation,
252: i.e.,
253: \begin{eqnarray}
254: &&r(\omega,p_y,p_z,s)=
255: \frac{1}{\sqrt{2\sinh(\pi\omega})}\nonumber\\
256: &&[e^{(\frac{\pi\omega}{2})}c(\omega,p_y,p_z,s)+e^{(-\frac{\pi\omega}{2})}c^\dagger(-\omega,-p_y,-p_z,s)],\nonumber\\
257: \\
258: &&l(\omega,p_y,p_z,s)=\frac{1}{\sqrt{2\sinh(\pi\omega})}\nonumber\\
259: &&[e^{(\frac{\pi\omega}{2})}c(-\omega,p_y,p_z,s)+e^{(-\frac{\pi\omega}{2})}c^\dagger(\omega,-p_y,-p_z,s)];\nonumber\\
260: \end{eqnarray}
261: or vice versa
262: \begin{eqnarray}
263: &&c(\omega,p_y,p_z,s)=
264: \frac{1}{\sqrt{2\sinh(\pi\omega})}\nonumber\\
265: &&[e^{(\frac{\pi\omega}{2})}r(\omega,p_y,p_z,s)-e^{(-\frac{\pi\omega}{2})}l^\dagger(\omega,-p_y,-p_z,s)],\nonumber\\
266: \\
267: &&c(-\omega,p_y,p_z,s)=\frac{1}{\sqrt{2\sinh(\pi\omega})}\nonumber\\
268: &&[e^{(\frac{\pi\omega}{2})}l(\omega,p_y,p_z,s)-e^{(-\frac{\pi\omega}{2})}r^\dagger(\omega,-p_y,-p_z,s)].\nonumber\\
269: \end{eqnarray}
270: Note that the vacuum state killed by the annihilation operator $c$
271: is equivalent to the ordinary Minkowski one\cite{Colosi}. Hence one
272: obtains the expression for the ordinary Minkowski vacuum in the mode
273: $A_\mu(\omega,p_y,p_z,s)$ as a Rindler state, i.e.,
274: \begin{eqnarray}
275: |0\rangle^M_{\omega,p_y,p_z,s}=&&\sqrt{\frac{2\sinh(\pi\omega)}{e^{(\pi\omega)}}}\sum_{n=0}^\infty
276: e^{(-n\pi \omega)}
277: \nonumber\\
278: &&|n(\omega,p_y,p_z,s)\rangle^R\otimes|n(\omega,-p_y,-p_z,s)\rangle^L,\nonumber\\
279: \end{eqnarray}
280: where
281: $|n(\omega,p_y,p_z,s)\rangle^R$($|n(\omega,p_y,p_z,s)\rangle^L$)
282: denotes the state with $n$ particles in Unruh mode
283: $R_\mu(\omega,p_y,p_z,s)$($L_\mu(\omega,p_y,p_z,s)$). Furthermore,
284: we have
285: \begin{eqnarray}
286: &&|1\rangle^M_{\omega,p_y,p_z,s}=c^\dagger(\omega,p_y,p_z,s)|0\rangle^M=\nonumber\\
287: &&[1-e^{(-2\pi\omega)}]\sum_{n=0}^\infty e^{-n\pi\omega}\sqrt{n+1}\nonumber\\
288: &&|(n+1)(\omega,p_y,p_z,s)\rangle^R\otimes|n(\omega,-p_y,-p_z,s)\rangle^L\nonumber\\
289: &&\prod_{\{\omega',p'_y,p'_z,s'\}\neq\{\omega,p_y,p_z,s\}}|0\rangle^M_{\omega',p'_y,p'_z,s'}.\nonumber\\
290: \label{expansion}
291: \end{eqnarray}
292:
293:
294:
295:
296: %\section{Quantum Electromagnetic Field in Rindler Spacetime}
297: %In the Rindler spacetime, the basis for quantizing electromagnetic
298: %field can be gauge chosen as
299: %\begin{eqnarray}
300: %&&|\omega,p_y, p_z, s=\pm1\rangle^R=A_\mu^R(\omega,p_y,p_z,
301: %s)=\nonumber\\
302: %&&\frac{\sqrt{\sinh(\pi\omega)}}{2\sqrt{2}\pi^2p_\bot}[(0,0,p_z\phi,-p_y\phi)+s(\rho\partial_\rho\phi,-i\frac{\omega}{\rho}\phi,0,0)].\nonumber\\
303: %\end{eqnarray}
304: %Here $p_\bot=\sqrt{p_y^2+p_z^2}$, and
305: %\begin{equation}
306: %\phi=K_{i\omega}(p_\bot\rho)e^{(-i\omega\tau+ip_yy+ip_zz)}
307: %\end{equation}
308: %satisfies the Klein-Gorden equation
309: %\begin{eqnarray}
310: %&&\nabla_a\nabla^a\phi=\frac{1}{\sqrt{-g}}\partial_\mu(\sqrt{-g}g^{\mu\nu}\partial_\nu\phi)=\nonumber\\
311: %&&(-\frac{\omega^2}{\rho}-\partial_\rho\rho\partial_\rho+\rho
312: %p_\bot^2)\phi=0,
313: %\end{eqnarray}
314: %where $K$ is the MacDonald function of imaginary index.
315:
316: %It is easy to check that the basis $|\omega,p_y, p_z,
317: %s=\pm1\rangle^R$ is the simultaneous eigensolution of energy,
318: %transverse momentum, and helicity operator with the corresponding
319: %eigenvalue $\{\omega,p_y, p_z, s\}$ in the Rindler spacetime.
320: %Furthermore, with respect to the well defined inner product
321:
322: %the basis is orthonormal, i.e.,
323: %\begin{eqnarray}
324: %&&^R\langle\omega,p_y, p_z,
325: %s|\omega',p'_y,p'_z,s'\rangle^R=\nonumber\\
326: %&&\delta(\omega-\omega')\delta(p_y-p'_y)\delta(p_z-p'_z)\delta_{ss'}.
327: %\end{eqnarray}
328: %Thus the quantum electromagnetic field could be expanded as
329: %\begin{eqnarray}
330: %\hat{A}_\mu^R=&&\int_0^\infty d\omega\int_{-\infty}^\infty
331: %dp_y\int_{-\infty}^\infty dp_z\sum_{s=\pm1}\nonumber\\
332: %&&[a(\omega,p_y,p_z,s)A_\mu^R(\omega,p_y,p_z,s)\nonumber\\
333: %&&+a^\dagger(\omega,p_y,p_z,s)\bar{A}_\mu^R(\omega,p_y,p_z,s)],
334: %\end{eqnarray}
335: %where $a$ and $a^\dagger$ are the familiar annihilation and creation
336: %operators, respectively, satisfying the following commutation
337: %relations
338: %\begin{equation}
339: %[a(\omega,p_y,p_z,s),a(\omega',p'_y,p'_z,s')]=0,
340: %\end{equation}
341: %\begin{equation}
342: %[a^\dagger(\omega,p_y,p_z,s),a^\dagger(\omega',p'_y,p'_z,s')]=0,
343: %\end{equation}
344: %\begin{eqnarray}
345: %&&[a(\omega,p_y,p_z,s),a^\dagger(\omega',p'_y,p'_z,s')]=\nonumber\\
346: %&&\delta(\omega-\omega')\delta(p_y-p'_y)\delta(p_z-p'_z)\delta_{ss'}.
347: %\end{eqnarray}
348:
349: \section{Entanglement for Electromagnetic Fields in Non-inertial Reference Frames\label{QE}}
350: In order to analyze quantum entanglement for electromagnetic field
351: in non-inertial reference frames, firstly following previous work
352: \cite{AM1,AM2,FM,AFMT}, we can also take into account the particle
353: number entangled state in the inertial reference frame associated
354: with Alice, i.e.,
355: \begin{equation}
356: |\varphi\rangle=\frac{1}{\sqrt{2}}(|0\rangle^M_A|0\rangle^M_B+|1\rangle^M_A|1\rangle^M_B).
357: \end{equation}
358: It is easy to show that the helicity structure of photon has no
359: influence in this case, and the corresponding calculation goes
360: straightforward, exactly the same as that for scalar particle, which
361: thus justifies modeling photon with scalar particle in investigation
362: of quantum entanglement in non-inertial reference frames for the
363: particle number entangled state\cite{AM1,AM2,FM}.
364:
365: We would next like to concentrate onto two photons' maximally
366: helicity entangled state in the inertial reference frame, i.e.,
367: \begin{eqnarray}
368: |\psi\rangle&=&\frac{1}{\sqrt{2}}(|1\rangle^M_{\omega,p_y,p_z,1A}|1\rangle^M_{\omega,-p_y,-p_z,-1B}\nonumber\\
369: &&+|1\rangle^M_{\omega,p_y,p_z,-1A}|1\rangle^M_{\omega,-p_y,-p_z,1B}),\label{helicity}
370: \end{eqnarray}
371: which also seems to be more popular than the particle number
372: entangled state in quantum information science. For later
373: convenience, we shall rewrite (\ref{helicity}) as
374: \begin{equation}
375: |\psi\rangle=\frac{1}{\sqrt{2}}(|1\rangle^M_{+\uparrow
376: A}|1\rangle^M_{-\downarrow B} +|1\rangle^M_{+\downarrow
377: A}|1\rangle^M_{-\uparrow B}).
378: \end{equation}
379: To describe this state from the viewpoint of the non-inertial
380: observer Bob, firstly we shall employ (\ref{expansion}) to expand
381: this state. Since Bob is causally disconnected from the $L$ sector,
382: we must take trace over all of the $L$ sector modes, which results
383: in a mixed density matrix between Alice and Bob, i.e.,
384: \begin{eqnarray}
385: \rho_{AB}=&&\frac{[1-e^{(-\frac{2\pi E}{a})}]^2}{2}\sum_{n=0}^\infty e^{(-\frac{2n\pi E}{a})}(n+1)\nonumber\\
386: &&(\ \ |1\rangle^M_{+\uparrow A}|n+1\rangle^R_{-\downarrow B}\langle1|^M_{+\uparrow A}\langle n+1|^R_{-\downarrow B}\nonumber\\
387: &&+|1\rangle^M_{+\uparrow A}|n+1\rangle^R_{-\downarrow B}\langle1|^M_{+\downarrow A}\langle n+1|^R_{-\uparrow B}\nonumber\\
388: &&+|1\rangle^M_{+\downarrow A}|n+1\rangle^R_{-\uparrow B}\langle1|^M_{+\uparrow A}\langle n+1|^R_{-\downarrow B}\nonumber\\
389: &&+|1\rangle^M_{+\downarrow A}|n+1\rangle^R_{-\uparrow
390: B}\langle1|^M_{+\downarrow A}\langle n+1|^R_{-\uparrow B}\
391: ),\nonumber\\\label{joint}
392: \end{eqnarray}
393: where $a$ denotes Bob's acceleration, and $E=a\omega$ is the energy
394: sensitive to Bob's detector.
395:
396: To determine whether this mixed state is entangled or not, we here
397: use the partial transpose criterion\cite{Peres2}. It states that if
398: the partial transposed density matrix of a system has at least one
399: negative eigenvalue, it must be entangled, otherwise it has no
400: distillable entanglement, but may have other types of entanglement.
401: After a straightforward calculation, the partial transposed density
402: matrix can be obtained as
403: \begin{eqnarray}
404: \rho_{AB}^T=&&\frac{[1-e^{(-\frac{2\pi E}{a})}]^2}{2}\sum_{n=0}^\infty e^{(-\frac{2n\pi E}{a})}(n+1)\nonumber\\
405: &&(\ \ |1\rangle^M_{+\uparrow A}|n+1\rangle^R_{-\downarrow B}\langle1|^M_{+\uparrow A}\langle n+1|^R_{-\downarrow B}\nonumber\\
406: &&+|1\rangle^M_{+\downarrow A}|n+1\rangle^R_{-\downarrow B}\langle1|^M_{+\uparrow A}\langle n+1|^R_{-\uparrow B}\nonumber\\
407: &&+|1\rangle^M_{+\uparrow A}|n+1\rangle^R_{-\uparrow B}\langle1|^M_{+\downarrow A}\langle n+1|^R_{-\downarrow B}\nonumber\\
408: &&+|1\rangle^M_{+\downarrow A}|n+1\rangle^R_{-\uparrow B}\langle1|^M_{+\downarrow A}\langle n+1|^R_{-\uparrow B}\ ),\nonumber\\
409: \end{eqnarray}
410: whose eigenvalues are easy to be computed, specifically those
411: belonging to the \textit{n}th diagonal block are
412: $\frac{[1-e^{[-\frac{2\pi E}{a})}]^2}{2}e^{(-\frac{2n\pi
413: E}{a})}(n+1)(1,1,1,-1)$. Thus the state as seen by Bob will be
414: always entangled if only the acceleration is finite. However,
415: quantification of the distillable entanglement can not be carried
416: out in this case. Therefore we only provide an upper bound of the
417: distillable entanglement by the logarithmic negativity\cite{VW}. It
418: is defined as $N(\rho)=\log_2||\rho^T||_1$, where$||\ ||_1 $is the
419: trace norm of a matrix. Whence the logarithmic negativity is given
420: by
421: \begin{equation}
422: N(\rho_{AB})=\log_2\{2[1-e^{(-\frac{2\pi E}{a})}]^2\sum_{n=0}^\infty
423: e^{(-\frac{2n\pi E}{a})}(n+1)\}=1,
424: \end{equation}
425: which is independent of the acceleration of Bob.
426:
427: Further, we can also make an estimation of the total correlation in
428: the state by employing the mutual information, i.e.,
429: $I(\rho_{AB})=S(\rho_A)+S(\rho_B)-S(\rho_{AB})$ where
430: $S(\rho)=-Tr(\rho\log_2\rho)$ is the entropy of the matrix $\rho$.
431: According to (\ref{joint}), the entropy of the joint state reads
432: \begin{eqnarray}
433: S(\rho_{AB})&&=-[1-e^{(-\frac{2\pi E}{a})}]^2\sum_{n=0}^\infty e^{(-\frac{2n\pi E}{a})}(n+1)\nonumber\\
434: &&\log_2\{[1-e^{(-\frac{2\pi E}{a})}]^2e^{(-\frac{2n\pi
435: E}{a})}(n+1)\}.
436: \end{eqnarray}
437: Tracing over Alice's states yields Bob's density matrix as
438: \begin{eqnarray}
439: \rho_B=&&\frac{[1-e^{(-\frac{2\pi E}{a})}]^2}{2}\sum_{n=0}^\infty
440: e^{(-\frac{2n\pi E}{a})}(n+1)\nonumber\\
441: &&(\ |n+1\rangle^R_{-\downarrow B}\langle n+1|^R_{-\downarrow
442: B}+|n+1\rangle^R_{-\uparrow B}\langle
443: n+1|^R_{-\uparrow B}),\nonumber\\
444: \end{eqnarray}
445: whose entropy is
446: \begin{eqnarray}
447: S(\rho_B)=&&1-[1-e^{(-\frac{2\pi E}{a})}]^2\sum_{n=0}^\infty e^{(-\frac{2n\pi E}{a})}(n+1)\nonumber\\
448: &&\log_2\{[1-e^{(-\frac{2\pi E}{a})}]^2e^{(-\frac{2n\pi
449: E}{a})}(n+1)\}.
450: \end{eqnarray}
451: Similarly, tracing over Bob's states, we obtain Alice's density
452: matrix as
453: \begin{equation}
454: \rho_A=\frac{1}{2}(|1\rangle^M_{+\uparrow A}\langle1|^M_{+\uparrow
455: A}+|1\rangle^M_{+\downarrow A}\langle1|^M_{+\downarrow A}),
456: \end{equation}
457: which has an entropy $S(\rho_A)=1$. As a result, the mutual
458: information is $I(\rho_{AB})=2$, which is the same for any uniformly
459: accelerated observer, no matter how much the magnitude of
460: acceleration is.
461:
462: Therefore, as seen by Bob, the helicity entanglement in non-inertial
463: reference frames shows a remarkably interesting behavior, which is
464: obviously different from the case for the particle number
465: entanglement. In particular, the calculable logarithmic negativity
466: and mutual information both remain constant for the photon helicity
467: entangled state, which is in strong contrast to the particle number
468: entangled state, where they both degrade with the increase of
469: acceleration. All of this seems to imply that the photon helicity
470: entangled state is more robust against the perturbation of
471: acceleration or gravitation than the particle number entangled
472: state, thus can be used as a more effective resource for performing
473: some quantum information processing technology.
474: \section{Conclusions and Discussions}
475: In this paper we have attempted to provide an analysis of quantum
476: entanglement of electromagnetic field in non-inertial reference
477: frames. In particular, we find that the maximally helicity entangled
478: state is a stable state under acceleration in the sense of its
479: logarithmic negativity and mutual information, which is obviously a
480: novel result, completely different from the case for the particle
481: number entangled state.
482:
483: As is mentioned in the beginning, the major difference between our
484: work and previous ones concerning quantum entanglement in
485: non-inertial frames is that we have considered the helicity
486: entanglement while previous ones only focus on the entanglement in
487: particle number. The helicity structure is special to photons, which
488: is a completely new trait that can not be presented in the case of
489: scalar particles. It is tempting to say that the entanglement of the
490: discrete degrees of freedom is generally different from the particle
491: number entanglement. Especially, the entangled state seems more
492: immune to the destruction of the acceleration or gravitation in
493: discrete degrees of freedom than particle number. To confirm this
494: conjecture, the spin entanglement of Dirac field in non-inertial
495: reference frames is a necessary and important task worthy of further
496: investigation. Since Dirac particle is constrained by Pauli
497: exclusion principle, it is a qubit-qubit system and the evaluation
498: of the corresponding entanglement is much easier, especially the
499: entanglement of formation can be explicitly
500: calculated\cite{Wootters}. Such a detailed analysis of the spin
501: entanglement in non-inertial reference frames and related problems
502: is expected to be reported elsewhere.
503: \section*{Acknowledgements}
504: We would like to give much gratitude to Chopin Soo for his
505: stimulating suggestion on this work and Bo Hu for his figure plotted
506: here. In addition, we gratefully acknowledge Steven J. van Enk for
507: his insightful and helpful comments. Valuable discussions from Paul
508: Alsing, Robert Mann, and Tracy Tessier are also much appreciated. Y.
509: Ling's work is partly supported by NSFC(Nos.10205002 and 10405027)
510: and SRF for ROCS. S. He's work is supported by NSFC(Nos.10235040 and
511: 10421003). W. Qiu's work is supported by NSFC(No.10547116), the
512: Science Research Fund of Huzhou Teachers College(No.KX21001) and the
513: Science Research Fund of Huzhou City(No.KY21022). H. Zhang's work is
514: supported in part by NSFC(Nos.10373003 and 10533010).
515: \begin{thebibliography}{99}
516: \bibitem{BEZ}The Physics of Quantum Information,
517: Springer-Verlag, 2000, D. Bouwmeester \emph{et al.}(Eds.).
518: \bibitem{PV}M. B. Plenio and S. Virmani, quant-ph/0504163.
519: \bibitem{AET}S. J. van Enk and T. Rudolph, Quant. Inf. Comput. 3: 423(2003).
520: \bibitem{PT}A. Peres and D. R. Terno, Rev. Mod. Phys. 76: 93(2004).
521: \bibitem{Shi}Y. Shi, Phys. Rev. D70: 105001(2004).
522: \bibitem{Peres1}A. Peres \emph{et al.}, Phys. Rev. Lett. 88:
523: 230402(2002).
524: \bibitem{AM1}P. M. Alsing and G. J. Milburn, Quant. Inf. Comput. 2: 487(2002).
525: \bibitem{GA}R. M. Gingrich and C. Adami, Phys. Rev. Lett. 89:
526: 270402(2002).
527: \bibitem{He}S. He \emph{et al.}, quant-ph/0701233.
528: \bibitem{AM2}P. M. Alsing and G. J. Milburn, Phys. Rev. Lett. 91:
529: 180404(2003).
530: \bibitem{AM3}P. M. Alsing \emph{et al.}, J. Optics. B6: S834
531: (2004).
532: \bibitem{FM}I. Fuentes-Schuller and R. B. Mann, Phys. Rev. Lett. 95:
533: 120404(2005).
534: \bibitem{AFMT}P. M. Alsing \emph{et al.}, Phys. Rev. A74: 032326(2006).
535: \bibitem{Moretti}V. Moretti, J. Math. Phys. 38: 2922(1997).
536: \bibitem{HQZ}Y. Hu \emph{et al.}, J. Math. Phys. 47: 052304(2006).
537: \bibitem{Colosi}D. Colosi, Nuovo Cim. B115: 1101(2000).
538: \bibitem{Ashtekar}A. Ashtekar, J. Math. Phys. 27: 824(1986).
539: \bibitem{Peres2}A. Peres, Phys. Rev. Lett. 77: 1413(1996).
540: \bibitem{VW}G. Vidal and R. F. Werner, Phys. Rev. A65: 032314(2002).
541: \bibitem{Wootters}W. K. Wootters, Phys. Rev. Lett. 80: 2245(1998).
542: \end{thebibliography}
543: \end{document}
544: