1:
2: \documentstyle[amssymb,pra,aps,tighten]{revtex}
3: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
4: %TCIDATA{TCIstyle=article/art2.lat,aps,revtex}
5:
6: %TCIDATA{Created=Wed Aug 16 18:36:20 2000}
7: %TCIDATA{LastRevised=Thu Oct 10 14:50:59 2002}
8: %TCIDATA{Language=American English}
9:
10: \newcommand{\twoket}[2]{\ket{{#1}}\!\!\ket{{#2}}}
11: \newcommand{\twobra}[2]{\bra{{#1}}\!\!\bra{{#2}}}
12: \newcommand{\ket}[1]{\left| {#1} \right\rangle}
13: \newcommand{\bra}[1]{\left\langle {#1} \right|}
14: \newcommand{\op}[1]{\hat{#1}^{}{}}
15:
16: \begin{document}
17: \title{Experimental preparation of Werner state via spontaneous parametric
18: down-conversion}
19: \author{Yong-Sheng Zhang\thanks{%
20: Electronic address: yshzhang@ustc.edu.cn}, Yun-Feng Huang, Chuan-Feng Li,
21: and Guang-Can Guo\thanks{%
22: Electronic address: gcguo@ustc.edu.cn}}
23: \address{Laboratory of Quantum Information, University of Science and Technology of\\
24: China, CAS, Hefei 230026, People's Republic of China\vspace{0.5in}}
25: \maketitle
26:
27: \begin{abstract}
28: We present an experiment of preparing Werner state via spontaneous
29: parametric down-conversion and controlled decoherence of photons in this
30: paper. In this experiment two independent BBO (beta-barium borate) crystals
31: are used to produce down-conversion light beams, which are mixed to prepare
32: Werner state.
33:
34: PACS number(s): 03.65.Ud, 42.50.-p, 03.67.-a
35: \end{abstract}
36:
37: \section{Introduction}
38:
39: Entanglement is not only one of the most striking features of quantum
40: mechanics, it also plays a crucial role in the field of quantum information
41: \cite{Shor94,Ben84,Ekt91,Ben93}. Entangled states are important resources
42: for most applications of quantum information such as quantum key
43: distribution \cite{Ekt91}, superdense coding \cite{Ben922}, quantum
44: teleportation \cite{Ben93} and quantum error correction \cite{QECC} etc.
45: Although the best performance of such tasks requires maximally entangled
46: states (Bell states), the decoherence effects due to the environment make
47: the pure entangled state into a statistical mixture and degrade quantum
48: entanglement in the real word. For a practical purpose, a purification
49: scheme may be applied to the degraded entanglement \cite{Ben96,Deu96}.
50:
51: One of the most important degraded Bell states is Werner state \cite
52: {Werner89}, A Werner state in $2\times 2$ system takes the following form
53: \cite{Ben96}:
54: \begin{equation}
55: \rho _W=\frac{1-F}3I_4+\frac{4F-1}3\left| \Psi ^{-}\right\rangle
56: \left\langle \Psi ^{-}\right| . \eqnum{1}
57: \end{equation}
58: where $I_n$ denotes the $n\times n$ identity matrix and $\left| \Psi
59: ^{-}\right\rangle $ is the singlet state of the four Bell states
60: \begin{eqnarray*}
61: \left| \Phi ^{\pm }\right\rangle &=&\frac 1{\sqrt{2}}\left( \left|
62: 00\right\rangle \pm \left| 11\right\rangle \right) , \\
63: \left| \Psi ^{\pm }\right\rangle &=&\frac 1{\sqrt{2}}\left( \left|
64: 01\right\rangle \pm \left| 10\right\rangle \right) .
65: \end{eqnarray*}
66: The Werner state $\rho _W$ is characterized by a single real parameter $F$
67: called fidelity. This quantity measures the overlap of Werner state with a
68: Bell state. In the case where $F\leqslant 1/2$, the state is separable, and
69: thus has no entanglement to recover or maintain. And a Werner state with $%
70: F>\left( 2+3\sqrt{2}\right) /8\approx 0.78$ violates the
71: Clauser-Horne-Shimony-Holt (CHSH) inequality \cite{CHSH69,Pop94,Aravind95}.
72:
73: There are many theoretical investigations on Werner state \cite{Wer}, and
74: the Werner state plays an important role in entanglement purification \cite
75: {Ben96}, nonlocality \cite{Werner89}, entanglement measure \cite{Shor01} and
76: etc. However, there is no report of experimental realization of Werner state
77: to date. At present, the most accessible and controllable source of
78: entanglement is obtained from the process of spontaneous parametric
79: down-conversion in a nonlinear crystal. In this paper, we present an
80: experimental preparation of bi-photon Werner state via spontaneous
81: parametric down-conversion and the prepared state is in the form
82: \begin{equation}
83: \rho _W^{\prime }=x\left| \Phi ^{-}\right\rangle \left\langle \Phi
84: ^{-}\right| +\left( 1-x\right) \frac{I_4}4, \eqnum{2}
85: \end{equation}
86: which can be transformed to $\rho _W$ by a local unitary transformation of $%
87: \sigma _x\otimes I$, where $\sigma _x$ is one of Pauli operators and $I$ is
88: the identity operator.
89:
90: There are previous works on preparing mixed single photon or bi-photon state
91: like in Ref. \cite{Kwiat00,White02,Thew01,ZhangC}. A. G. White {\it et al.}
92: \cite{White02} have reported an optical two-qubit source that can produce a
93: wide range of states in the form
94: \[
95: \rho =x\left| \Phi ^{+}\right\rangle \left\langle \Phi ^{+}\right| +\left(
96: 1-x\right) \left| 01\right\rangle \left\langle 01\right| .
97: \]
98: C. Zhang \cite{ZhangC} has proposed a theoretical protocol to produce an
99: arbitrary two-bit mixed state by using beam splitters with variable
100: polarization transmission coefficients and single-mode optical fibers.
101: However, in this paper, we propose a different way to prepare Werner states
102: conveniently in experiment.
103:
104: This paper is organized as follows. In Sec. II, the experimental set-up is
105: described. The experimental results is given in Sec. III and the discussion
106: and summary is given in Sec. IV.
107:
108: \section{Experimental Set-up}
109:
110: In our experiment, the Werner state is bi-photon's polarization state, which
111: is obtained by mixing a complete mixed state and an entangled state.
112:
113: The experimental set-up is depicted in Fig. 1. The 1.7-mm-diam pump beam at
114: 351.1 nm (single line, 100 mW) is produced by an argon ion laser (Coherent,
115: Sabre, model DBW25/7), and directed to the first BBO (beta-barium borate)
116: crystal (cut for type-I phase matching, optic axis cut at $\theta
117: =35.0^{\circ }$, 1 mm thick) after passing through a polarizing beam
118: splitter (PBS) to give a pure horizontal polarization state. Two entangled
119: photon beams (polarized in vertical direction) produced by the BBO crystal
120: passing through a sequence of quartz plates, with an optic axis set in the
121: diagonal direction. It is pointed out that the photon polarization state
122: will decohere in such an environment \cite{Kwiat00}. The thickness of quartz
123: plates is set in such a way that the photon's polarization decoheres
124: completely (See Ref. \cite{Zhang}, the difference between optical path of
125: horizontal and vertical optical polarizations in the quartz plates is $%
126: 153\lambda _0$, and $\lambda _0$ is the central wave length of the
127: down-converted light.). The remained pump beam passes through the BBO
128: crystal and is rotated to $\frac 1{\sqrt{2}}\left( \left| H\right\rangle
129: +\left| V\right\rangle \right) $ by a half-wave plate (HWP) and is
130: transmitted through the second BBO crystal which is the same as that
131: proposed by Kwiat in Ref. \cite{Kwiat99,White99} to produce
132: polarization-entangled photons in such a state. The down-converted light
133: beams produced by the first BBO were reflected by four reflectors and passed
134: through the second BBO crystal to be mixed with the down-converted beams
135: produced in the second BBO. Note the reason that the pump was set in single
136: line instead of single frequency is to decrease the coherent length (about 4
137: cm in our experiment) of the pump light and avoid the interference between
138: the down-converted light beams from the two BBO crystals.
139:
140: \begin{center}
141: {\bf Figure 1.}
142: \end{center}
143:
144: At the end, the bi-photon polarization state is measured tomgraphically \cite
145: {James01,White99} by quarter-wave plates (QWP), HWP and PBS. We use 16
146: analyzer settings as listed in Table I. The photons are detected by using
147: silicon avalanche photodiodes (EG\&G, SPCM-AQR) operated in the geiger mode.
148: Each detector is preceded by a small iris (the diameter is 1.5 mm) to define
149: the spatial mode, a narrowband interference filter (IF) centered at 702 nm
150: (Andover, 050FC46-25/7022 \cite{Andover}, full width at half is equal to
151: 4.62 nm) to reduce background and define the bandwidth of the photons, and a
152: collection lens. The detector outputs are recorded in coincidence with a
153: time-to-amplitude converter and a single-channel analyzer, leading to an
154: effective coincidence window of 5 ns. The resulting rate of accidental
155: coincidences is less than $1$s$^{-1}$, which can be neglected compared with
156: the typical rate of the true coincidences, which is about $300$s$^{-1}$.
157:
158: We also test whether the states produced by this set-up violate the CHSH
159: inequality. This inequality shows that $\left| S\right| \leq 2$ for any
160: local realistic theory, where
161: \begin{equation}
162: S=E\left( \theta _1,\theta _2\right) +E\left( \theta _1^{\prime },\theta
163: _2\right) +E\left( \theta _1,\theta _2^{\prime }\right) -E\left( \theta
164: _1^{\prime },\theta _2^{\prime }\right) \eqnum{3}
165: \end{equation}
166: and $E\left( \theta _1,\theta _2\right) $ is given by
167: \[
168: \frac{C\left( \theta _1,\theta _2\right) +C\left( \theta _1^{\bot },\theta
169: _2^{\bot }\right) -C\left( \theta _1^{\bot },\theta _2\right) -C\left(
170: \theta _1,\theta _2^{\bot }\right) }{C\left( \theta _1,\theta _2\right)
171: +C\left( \theta _1^{\bot },\theta _2^{\bot }\right) +C\left( \theta _1^{\bot
172: },\theta _2\right) +C\left( \theta _1,\theta _2^{\bot }\right) }
173: \]
174: and $C\left( \theta _1,\theta _2\right) $ is the coincidence rate of two
175: detectors when the polarization analyzer angels are set in $\theta _1$ and $%
176: \theta _2$. In this experiment, we selected the settings: $\theta
177: _1=-22.5^{\circ }$, $\theta _1^{\bot }=67.5^{\circ }$, $\theta _1^{\prime
178: }=22.5^{\circ }$, $\theta _1^{\prime \bot }=112.5^{\circ }$, $\theta
179: _2=0^{\circ }$, $\theta _2^{\bot }=90^{\circ }$, $\theta _2^{\prime
180: }=45^{\circ }$, $\theta _2^{\prime \bot }=135^{\circ }$.
181:
182: \section{Experimental Results}
183:
184: We can adjust the intensity proportion (it can be tuned by adjusting the
185: position of the reflectors) between the down-converted light from two BBO
186: crystals to obtain Werner states with different coefficients $x$. Two output
187: states have been produced in this experiment, for one of them the CHSH
188: inequality is violated while the other is not.
189:
190: The tomographic results are shown in Table I.
191:
192: \begin{center}
193: {\bf Table I.}
194: \end{center}
195:
196: From the data in the third column of Table I we can obtain the density
197: matrix of the first output state directly, however, it is not non-negative
198: definite \cite{James01}. We have used the maximum likelihood estimation \cite
199: {James01} to construct a non-negative definite density matrix
200: \begin{equation}
201: \rho _1=\left(
202: \begin{array}{cccc}
203: 0.4169 & 0.0203+0.0022i & 0.0094-0.0237i & -0.3476+0.0296i \\
204: 0.0203-0.0022i & 0.0531 & -0.0122-0.0527i & -0.0163+0.0005i \\
205: 0.0094+0.0237i & -0.0122+0.0527i & 0.0559 & -0.0134-0.0191i \\
206: -0.3476-0.0296i & -0.0163-0.0005i & -0.0134+0.0191i & 0.4741
207: \end{array}
208: \right) . \eqnum{4}
209: \end{equation}
210: The fit Werner state
211: \begin{equation}
212: \rho _1^{\prime }=x_1\left| \Phi ^{-}\right\rangle \left\langle \Phi
213: ^{-}\right| +\left( 1-x_1\right) \frac{I_4}4, \eqnum{5}
214: \end{equation}
215: (where $x_1=0.801\pm 0.005$) satisfies
216: \begin{eqnarray}
217: F\left( \rho _1,\rho _1^{\prime }\right) &=&Tr^2\left( \sqrt{\rho _1^{\prime
218: 1/2}\rho _1\rho _1^{\prime 1/2}}\right) \eqnum{6} \\
219: &=&\max F\left( \rho _1,\rho _W\right) , \nonumber
220: \end{eqnarray}
221: where $\rho _W$ is an arbitrary two-bit Werner state and the fidelity \cite
222: {Joz94} $F\left( \rho _1,\rho _1^{\prime }\right) $ is equal to $0.932$ in
223: this experiment. The CHSH correlation value [Eq. (3)] $\left| S\right|
224: =2.198\pm 0.004>2$, (the uncertainty of the HWP is $\Delta \theta \simeq
225: 0.2^{\circ }$) which violates the CHSH inequality. The theoretical value of $%
226: \left| S\right| $ of state $\rho _1^{\prime }$ is equal to $2.266$.
227:
228: The tomographic results of the second output are shown in the sixth column
229: of Table I. From these data we can obtain the density matrix of the second
230: output state
231: \begin{equation}
232: \rho _2=\left(
233: \begin{array}{cccc}
234: 0.3949 & -0.0217+0.0398i & -0.0080+0.0024i & -0.1657+0.0150i \\
235: -0.0217-0.0398i & 0.1285 & 0.0165-0.0011i & -0.0075+0.0504i \\
236: -0.0080-0.0024i & 0.0165+0.0011i & 0.1311 & -0.0152+0.0228i \\
237: -0.1657-0.0150i & -0.0075-0.0504i & -0.0152-0.0228i & 0.3455
238: \end{array}
239: \right) \eqnum{7}
240: \end{equation}
241: The fit Werner state is
242: \begin{equation}
243: \rho _2^{\prime }=x_2\left| \Phi ^{-}\right\rangle \left\langle \Phi
244: ^{-}\right| +\left( 1-x_2\right) \frac{I_4}4, \eqnum{8}
245: \end{equation}
246: where $x_2=0.405\pm 0.005$ and the fidelity $F\left( \rho _2,\rho _2^{\prime
247: }\right) $ is equal to $0.982$. The CHSH correlation value $\left| S\right|
248: =1.380\pm 0.008<2$, which satisfies the CHSH inequality. The theoretical
249: value of $\left| S\right| $ of state $\rho _2^{\prime }$ is equal to $1.146$%
250: . We should note that there is significant difference in count rates between
251: the two sets of results. In our experiment, the intensity proportion of
252: down-converted light beams between two BBO crystals is tuned by adjusting
253: the position of the reflectors. So the length of down-converted photon's
254: optical path between the source and the detector has been changed for
255: preparing the second state. Therefore, the count rates have been decreased
256: significantly.
257:
258: We can use the ``linear entropy'' to quantify the degree of mixture of a
259: quantum state and use the tangle to measure the degree of entanglement of a
260: two-bit quantum state \cite{James01,Wootters}. The linear entropy for a
261: two-bit system is defined by
262: \begin{equation}
263: P=\frac 43\left( 1-Tr\left\{ \rho ^2\right\} \right) . \eqnum{9}
264: \end{equation}
265: The tangle is defined as $T=C^2$, where $C$ is the concurrence in Ref. \cite
266: {Wootters}. In this experiment, for the first density matrix in Eq. (4), $%
267: P=0.46\pm 0.03$ and $T=0.35\pm 0.01$, and for the second density matrix in
268: Eq. (7), $P=0.83\pm 0.03$ and $T=0.01\pm 0.00$.
269:
270: \section{Discussion and Conclusion}
271:
272: We have prepared two typical Werner states in this experiment and measured
273: the density matrix tomographically. Two independent BBO crystals are used to
274: produce parametric down-conversion light beams which can be mixed to prepare
275: a wide range of two-bit quantum states. In our experiment, the intensity
276: proportion of down-converted light beams between two BBO crystals are tuned
277: by adjusting the position of the reflectors. In fact, the proportion can be
278: tuned with linear optical elements, which is shown in Fig. 2.
279:
280: \begin{center}
281: {\bf Figure 2.}
282: \end{center}
283:
284: Although the output light beams were produced by classically mixing (not
285: quantum superposing) the down-converted light beams, it is hard (but not in
286: principle) to distinguish from which crystals the output photon is produced
287: since the pump light is continuous. However, if we use the pulse pump we can
288: distinguish which crystal the output photon from by measuring the photon's
289: arrival time, that is, we can distill the classical information of the mixed
290: state and it can be regarded as a pure state.
291:
292: We expect that this method can be modified to prepare arbitrary two-bit
293: states conveniently and this source can be used in experimental
294: investigation of entanglement purification or other quantum information
295: tasks.
296:
297: \begin{center}
298: {\bf ACKNOWLEDGMENTS}
299: \end{center}
300:
301: This work was supported by the National Fundamental Research Program
302: (2001CB309300), the National Natural Science Foundation of China, the
303: Innovation Funds from Chinese Academy of Sciences (CAS), and was also funded
304: by the outstanding Ph. D. thesis award and the CAS's talented scientist
305: award rewarded to Lu-Ming Duan.
306:
307: \begin{center}
308: {\bf Appendix}
309: \end{center}
310:
311: For a single photon, the non-dissipative coupling between photon frequency
312: and polarization in a birefringent media leads to decoherence of
313: polarization, provided we regard the freedom of frequency as ``environment''%
314: \cite{Kwiat00,James01,White02}.
315:
316: The single-photon's initial state is described by a pure product state of
317: polarization and frequency,
318: \begin{equation}
319: \left| \Psi \right\rangle =\left( a_1\left| H\right\rangle +a_2\left|
320: V\right\rangle \right) \otimes \int d\omega A\left( \omega \right) \left|
321: \omega \right\rangle . \eqnum{A-1}
322: \end{equation}
323: with basis $\left| H\right\rangle $ (horizontal polarization) and $\left|
324: V\right\rangle $ (vertical polarization) denoted by
325: \[
326: \left| H\right\rangle =\left(
327: \begin{array}{l}
328: 1 \\
329: 0
330: \end{array}
331: \right) ,\left| V\right\rangle =\left(
332: \begin{array}{l}
333: 0 \\
334: 1
335: \end{array}
336: \right)
337: \]
338: respectively, and $A\left( \omega \right) $ is the complex amplitude
339: corresponding to $\omega $, normalized so that
340: \[
341: \int d\omega \left| A\left( \omega \right) \right| ^2=1.
342: \]
343: After being transmitted by the birefringent medium whose length is $x$, the
344: final state of the photon is
345: \begin{equation}
346: \left| \Psi ^{\prime }\left( x\right) \right\rangle =a_1\left|
347: H\right\rangle \otimes \int d\omega A\left( \omega \right) e^{-\frac 12%
348: i\omega \left( n_H-n_V\right) x/c}\left| \omega \right\rangle +a_2\left|
349: V\right\rangle \otimes \int d\omega A\left( \omega \right) e^{\frac 12%
350: i\omega \left( n_H-n_V\right) x/c}\left| \omega \right\rangle . \eqnum{A-2}
351: \end{equation}
352: Where $n_H$ ($n_V$) is the refractive index of horizontal (vertical)
353: polarization. It is necessary to note that although $n_H$ and $n_V$ are
354: dependent on frequency $\omega $, the value of $n_H-n_V$ does not vary
355: obviously with frequency varying in a small scale and we regard it as
356: constant. The density operator of the polarization state can be obtained by
357: tracing over frequency degrees of freedom from the complete density operator
358: \begin{equation}
359: \rho \left( x\right) =\left(
360: \begin{array}{ll}
361: a_1a_1^{*} & a_1a_2^{*}\Gamma \left( x\right) \\
362: a_1^{*}a_2\Gamma ^{*}\left( x\right) & a_2a_2^{*}
363: \end{array}
364: \right) , \eqnum{A-3}
365: \end{equation}
366: where
367: \begin{equation}
368: \Gamma \left( x\right) =\int d\omega A\left( \omega \right) A^{*}\left(
369: \omega \right) e^{-i\omega \left( n_H-n_V\right) x/c}. \eqnum{A-4}
370: \end{equation}
371:
372: From Eq. (A-3) it can be seen that the evolution of the polarization's
373: states depend on the character of the ``environment'' which is described by
374: the function $A\left( \omega \right) $.
375:
376: The spectrum of the photon's frequency is defined by the narrowband
377: interference filter (IF). In this experiment, the spectrum of the photon's
378: frequency is rectangular function. To obtain this environment, we use an
379: interference filter whose spectrum's shape is approximate to rectangle and
380: the full width at half maximum (FWHM) is $\delta =4.62$ nm. (Andover,
381: 050FC46-25/7022 \cite{Andover}). The theoretical value of $\Gamma \left(
382: x\right) $ is
383: \[
384: \Gamma \left( x\right) =\frac{2ic}{x\left( n_H-n_V\right) \Delta \omega }%
385: e^{i\omega _0x\left( n_H-n_V\right) /c}\sin \frac{x\left( n_H-n_V\right)
386: \Delta \omega }{2c}
387: \]
388: and
389: \begin{equation}
390: \left| \Gamma \left( x\right) \right| =\frac{2c}{x\left( n_H-n_V\right)
391: \Delta \omega }\left| \sin \frac{x\left( n_H-n_V\right) \Delta \omega }{2c}%
392: \right| , \eqnum{A-5}
393: \end{equation}
394: where we assume that $n_H>n_V.$
395:
396: \begin{center}
397: {\bf Figure 3}.
398: \end{center}
399:
400: In the experiment \cite{Zhang} (See Fig. 3, the BBO crystal in the figure is
401: cut at a degenerate type-I phase matching angle), we first measure the
402: density matrix of the decohered photon's polarization tomgraphically \cite
403: {James01}:
404: \[
405: \rho ^{\prime }=\left(
406: \begin{array}{ll}
407: a_1 & a_2 \\
408: a_2^{*} & a_3
409: \end{array}
410: \right) .
411: \]
412: The experimental value of $\left| \Gamma \left( x\right) \right| $ can be
413: calculated as
414: \begin{equation}
415: \left| \Gamma \left( x\right) \right| =\left| a_2/\sqrt{a_1a_3}\right| .
416: \eqnum{A-6}
417: \end{equation}
418: The results are shown in Fig. 4. The coherence of the photon's polarization
419: degenerates as theoretical expected.. It is interesting that the coherence
420: is recovered to some extent after it degenerated completely, which is
421: similar to the Fraunhofer diffraction at a rectangular aperture (or slit)
422: \cite{Born}. However, since the spectrum of the filter is not strictly
423: rectangular, there are some deviations at the tail of curve.
424:
425: \begin{center}
426: {\bf Figure 4.}\bigskip
427: \end{center}
428:
429: \begin{references}
430: \bibitem{Shor94} P. W. Shor, in {\it Proceedings of the 35th Annual
431: Symposium on the Foundations of Computer Science}, p. 124-133 (IEEE Computer
432: Society Press, Los Alamitos, California, 1984).
433:
434: \bibitem{Ben84} C. H. Bennett and G. Brassard, in{\it \ Proceedings of the
435: IEEE International Conference on Computers, Systems, and Signal Processing,
436: Bangalore, India, }1984 (IEEE, New York, 1984), p. 175.
437:
438: \bibitem{Ekt91} A. K. Ekert, Phys. Rev. Lett. {\bf 67}, 661 (1991).
439:
440: \bibitem{EPR35} A. Einstein, B. Podolsky and N. Rosen, Phys. Rev. {\bf 47},
441: 777 (1935).
442:
443: \bibitem{Ben93} C. H. Bennett, G. Brassard, C. Cr\'{e}peau, R. Jozsa, A.
444: Peres and W. K. Wootters, Phys. Rev. Lett. {\bf 70}, 1895 (1993).
445:
446: \bibitem{Ben922} C. H. Bennett and S. J. Wiesner, Phys. Rev. Lett. {\bf 69}%
447: , 2881 (1992).
448:
449: \bibitem{QECC} D. Gottesman, e-print quant-ph/0004072.
450:
451: \bibitem{Ben96} C. H. Bennett, G. Brassard, S. Popescu, B. Schumacher, J.
452: A. Smolin and W. K. Wootters, Phys. Rev. Lett. {\bf 76}, 722 (1996).
453:
454: \bibitem{Deu96} D. Deutsch, A. Ekert, R. Jozsa, C. Macchiavello, S. Popescu
455: and A. Sanpera, Phys. Rev. Lett. {\bf 77}, 2818 (1996).
456:
457: \bibitem{Werner89} R. F. Werner, Phys. Rev. A {\bf 40}, 4277 (1989).
458:
459: \bibitem{CHSH69} J. F. Clauser, M. A. Horne, A. Shimony and R. A. Holt,
460: Phys. Rev. Lett. {\bf 23}, 880 (1969).
461:
462: \bibitem{Pop94} S. Popescu, Phys. Rev. Lett. {\bf 72}, 797 (1994).
463:
464: \bibitem{Aravind95} P. K. Aravind, Phys. Lett. A {\bf 200}, 345 (1995); R.
465: Horodecki, P. Horodecki, and M. Horodecki, {\it ibid.} {\bf 200}, 340 (1995).
466:
467: \bibitem{Wer} J. Lee and M. S. Kim, Phys. Rev. Lett. {\bf 84}, 4236 (2000);
468: T. Hiroshima and S. Ishizaka, Phys. Rev A {\bf 62}, 044302 (2000); A. O.
469: Pittenger and M. H. Rubin, e-print quant-ph/0001110 (2000); S. Bose and V.
470: Vedral, Phys. Rev. A {\bf 61}, 040101(R) (2000); S. Ishizaka and T.
471: Hiroshima, Phys. Rev. A {\bf 62}, 022310 (2000); A. Felicetti, S. Mancini,
472: and P. Tombesi, Phys. Rev. A {\bf 65}, 062107 (2002).
473:
474: \bibitem{Shor01} P. W. Shor, J. A. Smolin, and B. M. Terhal, Phys. Rev.
475: Lett. {\bf 86}, 2681 (2001).
476:
477: \bibitem{Kwiat00} P. G. Kwiat, A. J. Berglund, J. Altepeter, and A. G.
478: White, Science {\bf 290}, 498 (2000); A. J. Berglumd, e-print
479: quant-ph/0010001.
480:
481: \bibitem{Kwiat99} P. G. Kwiat, E. Waks, A. G. White, I. Appelbaum, and P.
482: H. Eberhard, Phys. Rev A {\bf 60}, R773 (1999).
483:
484: \bibitem{White99} A. G. White, D. F. V. James, P. H. Eberhard, and P. G.
485: Kwiat, Phys. Rev. Lett. {\bf 83}, 3103 (1999).
486:
487: \bibitem{James01} D. F. V. James, P. G. Kwiat, W. J. Munro, and A. G.
488: White, Phys. Rev. A {\bf 64}, 052312 (2001).
489:
490: \bibitem{White02} A. G. White, D. F. V. James, W. J. Munro, and P. G.
491: Kwiat, Phys. Rev. A {\bf 65}, 012301 (2002).
492:
493: \bibitem{Thew01} R. T. Thew and W. J. Munro, Phys. Rev. A {\bf 64}, 022320
494: (2001).
495:
496: \bibitem{ZhangC} C. Zhang, e-print quant-ph/0107145.
497:
498: \bibitem{Andover} Part number: 050FC46-25/7022, http://www.andcorp.com.
499:
500: \bibitem{Joz94} R. Jozsa, J. Mod. Opt. {\bf 41}, 2315 (1994).
501:
502: \bibitem{Wootters} W. K. Wootters, Phys. Rev. Lett. {\bf 80}, 2245 (1998).
503:
504: \bibitem{Born} M. Born and E. Wolf, {\it Principles of Optics} (7th
505: edition), (Press of University of Cambridge, Cambridge, England, 1999).
506:
507: \bibitem{Zhang} Y.-S. Zhang, Y.-F. Huang, C.-F. Li, and G.-C. Guo,
508: unpublished.\bigskip
509: \end{references}
510:
511: Figure caption
512:
513: {\bf Figure 1}. Experimental set-up to produce bi-photon Werner state.
514:
515: {\bf Figure 2}. Experimental set-up to tune the intensity proportion of
516: down-conversion light beams from two BBO crystals.
517:
518: {\bf Figure 3}. Experimental setup of controlled decoherence of single
519: photon's polarization. The 351.1 nm line of an argon ion laser (Coherent,
520: Sabre) is used to pump a BBO crystal (3 mm thick), which is cut at a
521: degenerate type-I phase matching angle to produce a pair of entangled
522: photons. Each photon of the entangled pair is polarized in state $\left|
523: H\right\rangle $. Then the polarization of one photon of the entangled pair
524: is rotated to $\frac 1{\sqrt{2}}\left( \left| H\right\rangle +\left|
525: V\right\rangle \right) $ by the half-wave plate (HWP). After decohering in
526: the quartz plates, this photon's polarization state is measured
527: tomographically by HWP, quarter-wave plate (QWP) and polarization beam
528: splitter (PBS). The other photon of the pair is used to do coincidence
529: detection. We measure the decohered photon's polarization with 4 analysis
530: settings ($\left| H\right\rangle ,\left| V\right\rangle ,\left|
531: H\right\rangle +\left| V\right\rangle ,\left| H\right\rangle +i\left|
532: V\right\rangle $), allowing reconstruction of the density matrix.
533:
534: {\bf Figure 4}. Experimental results vs theoretical curve. The FWHM of the
535: IF is $4.62$nm and the spectrum of the frequency is assumed to be
536: rectangular. $x$ is the optical path difference between horizontal and
537: vertical polarizations and $\lambda _0$ is the central wavelength --- $702.2$%
538: nm --- of photons produced by SPDC.
539:
540: {\bf Table I}. Settings for measuring bi-photon Stokes parameters. $H$, $V$,
541: and $D$ are horizontal, vertical, and diagonal ($45^{\circ }$) linear
542: polarization, respectively. $R$ is right circular polarization. The data are
543: for state $\rho _1$ (counted over 100 s).
544:
545: {\bf Table II}. The data are for state $\rho _2$ (counted over 100 s).
546:
547: \end{document}
548: