1: % ****** Start of file apssamp.tex ******
2: %
3: % This file is part of the APS files in the REVTeX 4 distribution.
4: % Version 4.0 of REVTeX, August 2001
5: %
6: % Copyright (c) 2001 The American Physical Society.
7: %
8: % See the REVTeX 4 README file for restrictions and more information.
9: %
10: % TeX'ing this file requires that you have AMS-LaTeX 2.0 installed
11: % as well as the rest of the prerequisites for REVTeX 4.0
12: %
13: % See the REVTeX 4 README file
14: % It also requires running BibTeX. The commands are as follows:
15: %
16: % 1) latex apssamp.tex
17: % 2) bibtex apssamp
18: % 3) latex apssamp.tex
19: % 4) latex apssamp.tex
20: %
21:
22: %\documentclass[osajnl,preprint,showpacs]{revtex4}
23: %\documentclass[onecolumn,showpacs,preprintnumbers,amsmath,amssymb]{revtex4}
24: \documentclass[twocolumn,showpacs,preprintnumbers,amsmath,amssymb]{revtex4}
25: %\documentclass[preprint,showpacs,preprintnumbers,amsmath,amssymb]{revtex4}
26:
27: % Some other (several out of many) possibilities
28: %\documentclass[preprint,aps]{revtex4}
29: %\documentclass[preprint,aps,draft]{revtex4}
30: %\documentclass[prb]{revtex4}% Physical Review B
31:
32: \usepackage{graphicx}% Include figure files
33: \usepackage{dcolumn}% Align table columns on decimal point
34: \usepackage{bm}% bold math
35:
36: %\nofiles
37:
38: \begin{document}
39:
40: %\preprint{APS/123-QED}
41:
42: \title{Stimulated emission of two photons in parametric amplification\\
43: and its interpretation as multi-photon interference}
44: % Force line breaks with \\
45:
46: \author{F. W. Sun$^{1}$}
47: \author{B. H. Liu$^{1}$}
48: \author{Y. X. Gong$^{1}$}
49: \author{Y. F. Huang$^{1}$}
50: \author{Z. Y. Ou$^{2,1}$}
51: \email{zou@iupui.edu}
52: \author{G. C. Guo$^{1}$}
53: \affiliation{$^1$Key Laboratory of Quantum Information,
54: University of Science and Technology of China,\\ CAS, Hefei, 230026, the People's Republic of China
55: \\$^2$Department of Physics, Indiana
56: University-Purdue University Indianapolis \\ 402 N. Blackford
57: Street, Indianapolis, IN 46202}%Lines break automatically or can be forced with \\
58:
59: \date{\today}% It is always \today, today,
60: % but any date may be explicitly specified
61:
62: \begin{abstract}
63: Stimulated emission of two photons is observed experimentally in
64: the parametric amplification process and is compared to a
65: three-photon interference scheme. We find that the underlying
66: physics of stimulated emission is simply the constructive
67: interference due to photon indistinguishability. So the observed
68: signal enhancement upon the input of photons is a result of
69: multi-photon interference of the input photons and the otherwise
70: spontaneously emitted photon from the amplifier.
71:
72: \end{abstract}
73:
74: \pacs{42.50.-p, 42.50.St, 42.25.Hz, 42.65.Yj}% PACS, the Physics and Astronomy
75: % Classification Scheme.
76: %\keywords{Suggested keywords}%Use showkeys class option if keyword
77: %display desired
78: \maketitle
79:
80:
81: Stimulated emission, first proposed by Einstein \cite{ein} to
82: explain the blackbody radiation spectrum, is the main process in
83: laser operation. It provides the optical gain of an active medium
84: and is responsible for the coherence of laser light \cite{eber}.
85: Although the process was studied extensively as an amplification
86: process of a classical wave field as early as in 1955 \cite{town},
87: its effect on the nonclassical state of light was only
88: investigated not long ago \cite{man}, especially in the contest of
89: quantum state cloning \cite{zei,bouw,how}.
90:
91: Fundamentally, stimulated emission occurs at single-photon level,
92: i.e., it is seen as the creation of an identical photon to an
93: incoming photon. However, the same photon can also be produced
94: even without the incoming photon, due to spontaneous emission.
95: Thus, the existence of the input photon will enhance the
96: production rate, as compared to the case without the input photon.
97: Indeed, in recent study of stimulated emission by single photons,
98: a doubled rate is observed in photon production that is correlated
99: to the input photon \cite{ou,bouw,how}. But the above picture is
100: only phenomenological and does not tell the underlying physics. So
101: what fundamental physical principle governs this phenomenon?
102:
103: If we make a detailed analysis of the enhancement due to
104: stimulated emission, we find that it bears some resemblance to the
105: famous Hanbury Brown-Twiss effect \cite{hbt}, i.e., the photon
106: bunching effect of thermal light \cite{ou,orw2}: the enhancement
107: factor in both cases is one-fold and the temporal profile is the
108: same. The photon bunching effect, as pointed out by Glauber
109: \cite{glau2} in 1965, is in essence a two-photon interference
110: effect. This suggests that the underlying physics in stimulated
111: emission is simply multi-photon quantum interference. Recently,
112: Khan and Howell \cite{how2} and Irvine {\it et al.} \cite{irv}
113: utilized a beam splitter and the two-photon Hong-Ou-Mandel
114: interference effect \cite{hom,rar} to emulate the photon cloning
115: process observed in stimulated emission \cite{bouw}. This further
116: demonstrates the connection between stimulated emission and the
117: multi-photon interference.
118:
119: \begin{figure}
120: \includegraphics[width = 3in]{Fig1.eps}
121: \caption{Comparison between (a) the stimulated emission and (b)
122: the multi-photon interference. }
123: \end{figure}
124:
125:
126: In this letter, we wish to report on an experiment in support of
127: the above claim. In the experiment, we inject a two-photon state
128: into a parametric amplifier and observe an enhancement by a factor
129: of nearly two in the photon production that is correlated to the
130: input photons. Furthermore, we mimic the same phenomenon with a
131: beam splitter, in a similar manor to Ref.\cite{rar} for the
132: single-photon input case. These two phenomena can be viewed as a
133: generalized three-photon bunching effect and are a result of
134: three-photon constructive interference due to photon
135: indistinguishability.
136:
137: To understand the connection between the stimulated emission and
138: multi-photon interference, we consider the two situations in
139: Fig.1. The process of stimulated emission of an $N$-photon state
140: is shown in Fig.1(a), where $N$ photons interact with an atom in
141: an excited state. The atom will emit one photon regardless of the
142: input. Total photon number is $N+1$. Assume that the spontaneous
143: emission rate is $R$ into the same mode of the input photons. It
144: is known that the emission rate stimulated by one photon is the
145: same. Then since each input photon may stimulate the excited atom,
146: the total rate is then $(N+1)R$. The extra $R$ is from spontaneous
147: emission. There is an enhancement factor of $N$ in the photon
148: production rate. The case of $N=1$ was observed in
149: Ref.\cite{bouw}.
150:
151: In multi-photon interference in Fig.1(b), on the other hand, a
152: single photon and $N$ photons are combined by a 50:50 beam
153: splitter. The probability of detecting all $N+1$ photons in one
154: side is easily calculated to be $(N+1)/2^{N+1}$ (see below).
155: However, when the single photon is distinguishable from the $N$
156: photons and no multi-photon interference occurs, we find the
157: detection probability is simply $1/2^{N+1}$. The $N+1$ factor is a
158: result of constructive interference of $N+1$ possibilities in
159: detecting the $N+1$ photons. Each possibility corresponds to the
160: situation when the single input photon is detected by a specific
161: detector. (Fig.1 shows two such possibilities in the $N+1$-photon
162: detection.) We add the amplitudes of the $N+1$ possibilities
163: before taking the absolute value for the indistinguishable case
164: but we add the absolute values of the amplitude of each
165: possibility for the distinguishable case. The ratio between the
166: two cases is then $N+1$. The case of $N=1$ is the Hong-Ou-Mandel
167: photon bunching effect \cite{rar}. A similar interference effect
168: was observed by Ou {\it et al.} \cite{orw} with a $|2,2\rangle$
169: input state. Notice that the enhancement factor here is the same
170: as the stimulated emission in Fig.1(a). Therefore, the spontaneous
171: emission rate $R$ corresponds to the situation when the input $N$
172: photons to the atom are distinguishable from the emitted photon by
173: the atom, so that the atom is not influenced by the input photons
174: and only emits spontaneously. This case is exactly the same as the
175: case in Fig.1(b) but when the single photon is distinguishable
176: from the $N$ photons. When the input $N$ photons are
177: indistinguishable from the emitted photon by the atom,
178: constructive multi-photon interference leads to a factor of $N+1$
179: enhancement in photon detection rate. Thus, the underlying physics
180: in the stimulated emission is the photon indistinguishability that
181: results in multi-photon interference.
182:
183: In the following, we will use parametric amplifier to study the
184: stimulated emission by multiple photons. Mathematically, any phase
185: insensitive linear amplifier can be modelled as a single mode
186: parametric amplifier, which is described quantum mechanically by
187: \cite{cav}
188: \begin{eqnarray}
189: \hat a_s^{(out)} = G\hat a_s +g \hat a_i^{\dag}, \label{1}
190: \end{eqnarray}
191: where $\hat a_i$ corresponds to the internal modes of the
192: amplifier and is the idler mode for the parametric amplifier. It
193: is usually independent of $\hat a_s$ and is in vacuum. To preserve
194: the commutation relation, we need $|G|^2 - |g|^2 =1$.
195:
196: At microscopic level of atoms, we have a small value of $|g|<<1$
197: or $|G|\sim 1$. The unitary evolution operator for Eq.(\ref{1})
198: then has the form of
199: \begin{eqnarray}
200: \hat U \approx 1 + (g \hat a_s^{\dag}\hat a_i^{\dag} +
201: h.c.)\label{2}
202: \end{eqnarray}
203: With a vacuum input of $|0\rangle$, we have the output state
204: \begin{eqnarray}
205: |\Phi\rangle_{out}^{(0)} =\hat U |0\rangle \approx |0\rangle + g
206: |1_s\rangle\otimes|1_i\rangle.\label{3}
207: \end{eqnarray}
208: The last term gives the spontaneous emission with a probability
209: of $|g|^2$. When the input is a single-photon state
210: $|1_s\rangle\otimes|0_i\rangle$, we have
211: \begin{eqnarray}
212: |\Phi\rangle_{out}^{(1)} &\approx & |1_s\rangle|0_i\rangle + g
213: (\hat a_s^{\dag}|1_s\rangle)\otimes(\hat a_i^{\dag}|0_i\rangle
214: )\cr &= &|1_s\rangle|0_i\rangle + \sqrt{2}g
215: |2_s\rangle\otimes|1_i\rangle.\label{4}
216: \end{eqnarray}
217: The probability for the emission from the amplifier is then
218: $2|g|^2$. The extra emission probability of $|g|^2$ is usually
219: attributed to the stimulated emission, which is similar to the
220: photon bunching effect \cite{hbt,ou,orw2,rar}.
221:
222: When the input state is a two-photon state of
223: $|2_s\rangle|0_i\rangle$, we have the output state as
224: \begin{eqnarray}
225: |\Phi\rangle_{out}^{(2)} &\approx & |2_s\rangle|0_i\rangle + g
226: (\hat a_s^{\dag}|2_s\rangle)\otimes(\hat a_i^{\dag}|0_i\rangle )
227: \cr &= &|2_s\rangle|0_i\rangle + \sqrt{3}g
228: |3_s\rangle\otimes|1_i\rangle.\label{5}
229: \end{eqnarray}
230: The photon emission rate from the amplifier is now three times the
231: spontaneous rate. In fact, it is straightforward to find that,
232: with an $N$-photon state as the input, the rate of photon emission
233: from the amplifier is $N+1$ times the spontaneous emission rate.
234: As the single-photon input case, each fold of increase in the rate
235: can be attributed to the stimulated emission from one individual
236: photon in the input $N$-photon state.
237:
238: Notice that when the input photons are not in the same mode as the
239: amplifier and thus are distinguishable from the photon emitted by
240: the amplifier, the output state becomes
241: \begin{eqnarray}
242: |\Phi\rangle_{out}^{(N)'} &\approx &
243: |0_s\rangle|N_{s'}\rangle|0_i\rangle + g (\hat
244: a_s^{\dag}|0_s\rangle) \otimes|N_{s'}\rangle\otimes(\hat
245: a_i^{\dag}|0_i\rangle ) \cr &=
246: &|0_s\rangle|N_{s'}\rangle|0_i\rangle + g
247: |1_s\rangle|N_{s'}\rangle |1_i\rangle, \label{6}
248: \end{eqnarray}
249: where $N \ge 1$. So the photon production rate is exactly the same
250: as the spontaneous emission.
251:
252: The above analysis with the parametric amplifier confirms the
253: previous results obtained from the pictorial argument based on
254: Fig.1. Next, we will demonstrate experimentally the enhancement
255: effect for a two-photon input and compare it with a three-photon
256: interference scheme with a beam splitter.
257:
258: \begin{figure}
259: \includegraphics[width = 2.9in]{Fig2.eps}
260: \caption{Schematics for studying the stimulated emission of an
261: input of $N$-photon state with parametric amplification. SMF:
262: single-mode fiber; IF: interference filter; H,V: horizontal and
263: vertical polarizations; $T_H$: adjustable delay.}
264: \end{figure}
265:
266: The experimental arrangement for studying the stimulated emission
267: is sketched in Fig.2. A mode-locked Ti:sapphire laser operating at
268: 780 nm is frequency doubled and the harmonic field serves as the
269: pump field for a parametric amplifier made of a 1-mm long
270: $\beta-$Barium Borate (BBO) crystal. The crystal is so oriented
271: that it satisfies the type-II phase matching condition and
272: beam-like fields are generated \cite{tsu}. A small portion is
273: split from the Ti:sapphire laser and serves as the input field to
274: the signal port of the parametric amplifier. The injected coherent
275: field is heavily attenuated down to a rate much less than one
276: photon per pulse. But even so, the coherent state consists of
277: vacuum, one-photon state, two-photon state, and more. So the
278: output state is a superposition of the states in
279: Eqs.(\ref{3}-\ref{5}). Therefore, in order to observe the
280: enhancement effect in stimulated emission by a specific number of
281: photons, we need to make a projection measurement to the
282: corresponding states in Eqs.(\ref{4}, \ref{5}). For example, for a
283: two-photon state input, the projection is to the second term in
284: Eq.(\ref{5}). This is achieved by a four-photon coincidence
285: measurement, as depicted in Fig.2. Joint measurement with the
286: idler photon is necessary to discriminate against the three-photon
287: contribution directly from the injected coherent field. Photon
288: (in)distinguishability between the input photons and the photon
289: emitted from parametric down-conversion is realized by an
290: adjustable delay $T_H$ on the coherent injection field. A
291: single-mode fiber (SMF) is used to collect the signal field from
292: the amplifier, in order to ensure a good spatial mode match.
293: Interference filters of bandpass of 3 $nm$ are used for temporal
294: mode cleaning.
295:
296:
297: \begin{figure}
298: \includegraphics[width = 2.8in]{Fig3.eps}
299: \caption{(a) Four-photon coincidence of ABCD detectors in 100s and
300: (b) three-photon coincidence of ABD detectors in 10s as a function
301: of the delay $cT_H$.}
302: \end{figure}
303:
304: The conditions for the situations in Eqs.(\ref{5}, \ref{6}) are
305: satisfied by adjusting the delay $T_H$. When the delay is right,
306: the injected coherent field pulse arrives in time with the pump
307: pulse to the amplifier and the photon emitted by the amplifier is
308: indistinguishable from the incoming photons in the coherent state.
309: But when the delay is either too large or too small, there is no
310: overlap between the coherent pulse and the pump pulse. This is the
311: situation described in Eq.(\ref{6}). Therefore, as we scan the
312: delay $T_H$, the four-photon coincidence of A, B, C, D detectors
313: should exhibit a bunching effect with a peak-to-wing ratio close
314: to three. Fig.3(a) shows the result of the measurement. The error
315: bars are the statistical errors of one standard deviation. The
316: solid curve is a least square fit of the data to a Gaussian of the
317: form
318: \begin{eqnarray}
319: F(T_H) = A\Big[1+v e^{-(T_H-T_0)^2/T_c^2}\Big],
320: \end{eqnarray}
321: where $T_0$ is the center position of the peak and $T_c$ is
322: related to the width of the peak. We obtain $v_2=1.81 \pm 0.15$ as
323: the enhancement factor for the data in Fig.3(a), which gives 2.81
324: as the ratio between the peak and the wing. This value is close to
325: the ideal value of three in Eq.(\ref{5}) for the stimulated
326: emission by two photons.
327:
328: In the meantime, three-photon coincidences of ABD detectors are
329: also registered and are shown in Fig.3(b). This measurement
330: corresponds to the second term in Eq.(\ref{4}) and gives the
331: stimulated emission by one input photon. The Gaussian fit gives an
332: enhancement factor of $v_1=0.88\pm0.14$. The peak-to-wing ratio of
333: 1.88 is close to the ideal value of two in Eq.(\ref{4}).
334:
335: The reason for the imperfection in the experiment is due to mode
336: mismatch between the input field and the amplifier mode, {\it
337: i.e.}, mismatch between $|N\rangle$ and the mode for which the
338: operator $\hat a_s$ represents. Although the spatial mode is
339: matched by the single-mode fiber (SMF in Fig.2), the temporal mode
340: is hard to match because the temporal coherence of the parametric
341: down-conversion process is very complicated and the fields are not
342: transform-limited even if the pump field is. Nevertheless, we use
343: interference filters to clean up the temporal profile. The full
344: widths of the peaks in Fig.3 is approximately $2T_c= 660ps$, close
345: to the coherence time of the interference filters (IF in Fig.2) of
346: width 3 $nm$.
347:
348: Next we consider the situation depicted in Fig.1(b) where an
349: $N$-photon state is superposed with a single-photon state by a
350: 50:50 beam splitter. The output state for the beam splitter is
351: given by \cite{cam}
352: \begin{eqnarray}
353: |\Phi\rangle_{out}^{(BS)} = \sqrt{N+1\over
354: 2^{N+1}}|N+1\rangle_1|0\rangle_2 + ...~,\label{8}
355: \end{eqnarray}
356: where we only write down the state for which all the $N+1$ photons
357: exit at one port (port 1) of the beam splitter. On the other hand,
358: if the input $N$ photons are distinguishable from the single
359: photon from the other input port, they behave like classical
360: particles and follow the Bernoulli distribution. The output state
361: becomes
362: \begin{eqnarray}
363: |\Phi\rangle_{out}^{(BS)'} = \sqrt{1\over
364: 2^{N+1}}|N\rangle_{1'}|1\rangle_1|0\rangle_2 + ...~.\label{9}
365: \end{eqnarray}
366: Therefore, the rate of detecting $N+1$ photons in port 1 is $N$
367: times bigger when the photons are all indistinguishable than when
368: the $N$ photons are distinguishable from the one photon. As
369: discussed before, this increase stems from a constructive
370: $N+1$-photon interference.
371:
372: \begin{figure}
373: \includegraphics[width = 2.9in]{Fig4.eps}
374: \caption{A simple scheme for interference of $N$ photons and one
375: photon with a beam splitter. Same notations as Fig.2.}
376: \end{figure}
377:
378: \begin{figure}
379: \includegraphics[width = 2.9in]{Fig5.eps}
380: \caption{Same as Fig.2 but with data obtained from Fig.4.}
381: \end{figure}
382:
383:
384: The experimental arrangement with a beam splitter is similar to
385: Fig.2 and is shown in Fig.4, where the split weak coherent field
386: is directed to a beam splitter to combine with the signal field
387: (H) from the parametric down-conversion. In order to mimic the
388: stimulated emission process shown in Fig.2, all the experimental
389: parameters such as pump power, the strength of coherent field,
390: etc. are the same as those in Fig.2. We adjust the delay $T_H$ on
391: the coherent field to ensure the temporal overlap between the
392: coherent field and the down-converted photon. When gated on the
393: detection of the V-photon by detector D, the H-field of the
394: down-conversion is in a single-photon state. But because of the
395: complicated dispersion in the parametric down-conversion process,
396: the single-photon state is not transform-limited. Again,
397: interference filters are used to clean up some of the temporal
398: modes but impossible to achieve the perfect match.
399:
400:
401: We record both the four-photon coincidence of ABCD detectors and
402: the three-photon coincidence of ABD detectors. The former
403: corresponds to the $N=2$ case in Eq.(\ref{9}), whereas the latter
404: to the $N=1$ case. The results are shown in Fig.5. The fitted
405: curves are very similar to Fig.3 but with $v_2= 1.78\pm 0.14$ and
406: $v_1 =0.86\pm0.09$.
407:
408: As can be seen, Figs.3 and 5 show the same result within the
409: statistical errors. This confirms our claim that the underlying
410: physics in stimulated emission is nothing but multi-photon
411: interference. The interference effect is a result of
412: indistinguishability between the input photons and the photon
413: emitted by the amplifier.
414:
415: \begin{acknowledgments}
416: This work was funded by National Fundamental Research Program of
417: China, the Innovation Funds of Chinese Academy of Sciences. ZYO is
418: supported by the US National Science Foundation under Grant No.
419: 0245421 and No. 0427647.
420: \end{acknowledgments}
421:
422: \begin{thebibliography}{99}
423: \bibitem{ein} A. Einstein, Phys. Z. {\bf 18}, 121 (1917).
424:
425: \bibitem{eber} P. W. Milonni and J. H. Eberly, {\it Lasers} (Wiley, New York, N.Y., 1988).
426:
427: \bibitem{town} J. P. Gordon, H. J. Zeiger, and C. H. Townes, Phys. Rev. {\bf 99},
428: 1264 (1955).
429:
430: \bibitem {man} S. Friberg and L. Mandel, Opt. Commun. {\bf 46} 141
431: (1983).
432:
433: \bibitem {zei} C. Simon, G. Weihs, and A. Zeilinger,
434: Phys. Rev. Lett. {\bf 84}, 2993 (2000).
435:
436: \bibitem {bouw} A. Lamas-Linares, J. C. Howell, and D. Bouwmeester, Nature, {\bf
437: 412} 6850 (2001).
438:
439: \bibitem {how} A. Lamas-Linares, C. Simon, J. C. Howell, and D. Bouwmeester, Science {\bf 296}, 712
440: (2002).
441:
442: \bibitem {ou} Z. Y. Ou, L. J. Wang, and L. Mandel, J. Opt. Soc. Am. B {\bf 7}, 211 (1990).
443:
444: \bibitem {hbt} R. Hanbury-Brown and R. W. Twiss, Nature {\bf 177}, 27
445: (1956).
446:
447: \bibitem {orw2} Z. Y. Ou, J.-K. Rhee, and L. J. Wang, Phys. Rev. A {\bf 60},
448: 593 (1999).
449:
450: \bibitem {glau2} R. J. Glauber, in {\it Quantum Optics and Electronics
451: (Les Houches Lectures)}, p.63, edited by C. deWitt, A. Blandin,
452: and C. Cohen-Tannoudji (Gordon and Breach, New York, 1965).
453:
454: \bibitem {how2} I. A. Khan and J. C. Howell, \pra {\bf 70}, 010303(R)
455: (2004).
456:
457: \bibitem {irv} W. T. M. Irvine, A. Lamas-Linares, M. J. A. de Dood, and D. Bouwmeester, \prl {\bf 92},
458: 047902 (2004).
459:
460: \bibitem{hom} C. K. Hong, Z. Y. Ou, and L. Mandel, \prl {\bf 59}, 2044 (1987).
461:
462: \bibitem {rar} J. G. Rarity and P. R. Tapster, J. Opt. Soc. Am. B {\bf 6},
463: 1221 (1989).
464:
465: \bibitem{orw} Z. Y. Ou, J.-K. Rhee, and L. J. Wang, Phys. Rev. Lett. {\bf 83}, 959
466: (1999).
467:
468:
469: \bibitem{cav} C. M. Caves, Phys. Rev. D {\bf 26}, 1817
470: (1982).
471:
472: \bibitem{tsu} S. Takeuchi, Opt. Lett. {\bf 26}, 843 (2001).
473:
474: \bibitem{cam} R. A. Campos, B. E. A. Saleh, and M. C. Teich, Phys. Rev. A{\bf 40}, 1371 (1989).
475:
476:
477: \end{thebibliography}
478:
479:
480: \end{document}
481: %
482: % ****** End of file apssamp.tex ******
483: