314 29 13MB
English Pages 218 [231] Year 1988
P H O T O N S A N D Q U A N T U M F L U C T U A T IO N S
Other books in series Chaos, Noise and Fractals
edited by E R Pike and L A Lugiato Frontiers in Quantum Optics
edited by E R Pike and S Sarkar Basic Methods of Tomography and Inverse Problems
edited by P C Sabatier Nonlinear Phenomena and Chaos
edited by S Sarkar
MALVERN PHYSICS SERIES General Series Editor: Professor E R Pike FRS
PHOTONS AND QUANTUM FLUCTUATIONS Edited by E R Pike Royal Signals and Radar Establishment, Malvern and Department of Physics, King’s College, London and H Walther Sektion Physik, Universitat Miinchen and Max-Planck-Institut fur Quantenoptik, Garching
CRC Press Taylor & Francis Group Boca Raton London New York CRC Press is an imprint of the Taylor & Francis Group, an i nf or ma business
CRC Press Taylor & Francis Group 6000 Broken Sound Parkway NW, Suite 300 Boca Raton, FL 33487-2742 First issued in paperback 2019 © 1988 by Taylor & Francis Group, LLC CRC Press is an imprint o f Taylor & Francis Group, an Informa business No claim to original U.S. Government works ISB N -13: 978-0-85274-240-2 (hbk) ISBN-13: 978-0-367-40338-6 (pbk) This book contains inform ation obtained from authentic and highly regarded sources. Reasonable efforts have been made to publish reliable data and information, but the author and publisher cannot assume responsibility for the validity o f all materials or the consequences o f their use. The authors and publishers have attempted to trace the copyright holders o f all m aterial reproduced in this publication and apologize to copyright holders if perm ission to publish in this form has not been obtained. If any copyright m aterial has not been acknowledged please write and let us know so we may rectify in any future reprint. Except as perm itted under U.S. Copyright Law, no part o f this book may be reprinted, reproduced, transmitted, or utilized in any form by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying, m icrofilming, and recording, or in any inform ation storage or retrieval system, without written permission from the publishers. For perm ission to photocopy or use material electronically from this work, please access w w w .copyright.com (http://www.copyright.com/) or contact the Copyright Clearance Center, Inc. (CCC), 222 Rosewood Drive, Danvers, MA 01923, 978-7508400. CCC is a not-for-profit organization that provides licenses and registration for a variety o f users. For organizations that have been granted a photocopy license by the CCC, a separate system o f paym ent has been arranged.
Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe.
British Library Cataloguing in Publication Data Photons and quantum fluctuations. 1. Quantum optics I. Pike, E. R. (Edward Roy). II. Walther, H. (Herbert), 1935- III. Series 535'. 15 ISBN 0-85274-240-1 US Library o f Congress Cataloging-in-Publication Data are available Visit the Taylor & Francis Web site at http://www.taylorandfrancis.com and the CRC Press Web site at http://www.crcpress.com
CONTENTS Preface
vii
List of Contributors
ix
Nonclassical Light H P Yuen
1
Single-atom Oscillators H Walther
10
Photodetection and Photostatistics M Le Berre
31
Nonclassical Photon Interference Effects C K Hong, Z Y Ou and L Mandel
51
Photons and Approximate Localisability E R Pike and S Sarkar
66
Quantum Noise Reduction on Twin Laser Beams E Giacobino, C Fabre, S Reynaud, A Heidmann and R Horowicz
81
Nonclassical Effects in Parametric Downconversion J G Rarity and P R Tapster
122
Moving Mirrors and Nonclassical Light S Sarkar
151
Propagation of Nonclassical Light I Abram
173
Models for Phase-insensitive Quantum Amplifiers G L Mander, R Loudon and T J Shepherd
190
Quasiprobabilities based on Squeezed States F Haake and M Wilkens
205
Index
217
PREFACE
There has been considerable progress in the field of quantum optics in recent years. New experimental techniques have enlarged our basic understanding of radiation-matter coupling. Single particle events can now be studied in detail and it has become possible to explore non-classical properties of radiation not only in photon correlation measurements of antibunching (silent light!) but also via the dynamics of the radiation-atom coupling. In cavities electromagnetic fields characterized by fixed photon numbers (Fock states) can be generated and it is possible to study the interaction of atoms in such fields. Furthermore, methods of nonlinear optics allow, via parametric downconversion, two photons to be created simultaneously; by use of the photoelec tric detection of one of them as a trigger, a good approximation to the ideal localized one-photon state can be achieved in the other. By means of nonlinear optical processes squeezed states can also be generated which allow measure ments with precision beyond the limit set by the zero-point or vacuum fluctuations of the optical field. Among the potential applications of such fields is their use in a large laser interferometer for gravitational wave detection. These exciting and recent developments together with other fundamental theoretical contributions in quantum optics are discussed in this book in detail. The different contributions review talks which were given at a special ONR Seminar held on January 21st and 22nd, 1988 at the Istituto d’Arte in Cortina d’Ampezzo, Italy, and at which we were honoured to join their 100th anniversary celebrations. The spectacular natural environment of Cortina was as inspiring and exciting as the topics of the seminar. This Special Seminar was followed by a NATO ARW on ‘Squeezed and Non-Classical Light’, the proceedings of which will be published by Plenum Press. The two volumes may be regarded as complementary. We would like to acknowledge, on behalf of all the participants, the generous support and encouragement of the US Office of Naval Research in London. We would also like to acknowledge additional financial assistance provided by Ministero Pubblica Istruzione, Universita di Roma ‘La Sapienza’, Consiglio Nazionale delle Ricerche (CNR), Gruppo Nazionale di Struttura della Materia (CNR) and local sponsorship from Olivetti Spa, Consorzio per lo Sviluppo e Turismo di Cortina d’Ampezzo, Municipio di Cortina d’Ampezzo, Istituto Statale d’Arte and Hotel Europa. We express our gratitude for the invaluable assistance of Professor Paolo Tombesi of the University of Rome and the inspiration of Professor Danny Walls to suggest such a timely meeting. Special thanks are due to Professor M Spamponi, Mr G Milani, Ing E Cardazzi, Professor G Demenego, Mr E Demenego and, in particular, Professor G Olivieri. Jim Revill of Adam Hilger provided expert assistance with these
Viii
PREFACE
proceedings and, last but by no means least, we thank our willing hard-worked secretaries Beverley James, Angela Di Silvestro and Marcella Mastrofini. Roy Pike Herbert Waither
May 1988
LIST OF CONTRIBUTORS
I ABRAM
CNET 196 Avenue Henri Ravera 92200 Bagneux Cedex France
M LE BERRE
Laboratoire de PPM Universite de Paris Sud Batiment 213 91405 Orsay Cedex France
C FABRE
Laboratoire de Spectroscopie Hertzienne de TENS Universite Pierre et Marie Curie Tour 12, Premier Etage 4 Place de Jussieu 75252 Paris Cedex 05 France
E G IA CO B IN O
L a b o ra to ire de S pectroscopie H ertzien n e de FEN S
Universite Pierre et Marie Curie Tour 12, Premier Etage 4 Place de Jussieu 75252 Paris Cedex 05 France F HAAKE
Fachbereich Physik Universitat Essen - Gesamthochschule Postfach 103764 4300 Essen 1 FRG
A HEIDMANN
Laboratoire de Spectroscopie Hertzienne de FENS Universite Pierre et Marie Curie Tour 12, Premier Etage 4 Place de Jussieu 75252 Paris Cedex 05 France
X
LIST OF CONTRIBUTORS
C K HONG
Department of Physics and Astronomy University of Rochester Rochester New York NY 14627 USA
R HOROWICZ
Laboratoire de Spectroscopie Hertzienne de PENS Universite Pierre et Marie Curie Tour 12, Premier Etage 4 Place de Jussieu 75252 Paris Cedex 05 France
R LOUDON
Department of Physics University of Essex Wivenhoe Park Colchester C 0 4 3SQ UK
L MANDEL
Department of Physics and Astronomy University of Rochester Rochester New York NY 14627 USA
G L MANDER
Department of Physics University of Essex Wivenhoe Park Colchester C 04 3SQ UK
Z YOU
Department of Physics and Astronomy University of Rochester Rochester New York NY 14627 USA
LIST OF CONTRIBUTORS
E R PIKE
Centre for Theoretical Studies Royal Signals and Radar Establishment St Andrews Road Malvern Worcestershire WR14 3PS UK
and
Department of Physics King’s College Strand London WC2R 2LS UK
J G RARITY
Royal Signals and Radar Establishment St Andrews Road Malvern Worcestershire WR14 3PS UK
S REYNAUD
Laboratoire de Spectroscopie Hertzienne de TENS Universite Pierre et Marie Curie Tour 12, Premier Etage 4 Place de Jussieu 75252 Paris Cedex 05 France
SARBEN SARKAR
Centre for Theoretical Studies Royal Signals and Radar Establishment St Andrews Road Malvern Worcestershire WR14 3PS UK
T J SHEPHERD
Royal Signals and Radar Establishment St Andrews Road Malvern Worcestershire WR14 3PS UK
x ii
LIST OF CONTRIBUTORS
P R TAPSTER
Royal Signals and Radar Establishment St Andrews Road Malvern Worcestershire WR14 3PS UK
H WALTHER
Sektion Physik Universitat Miinchen Am Coulombwall 1 8046 Garching FRG
and
Max-Planck-Institut fur Quantenoptik 8046 Garching FRG
M WILKENS
Fachbereich Physik Universitat Essen - Gesamthochschule Postfach 103764 4300 Essen 1 FRG
H P YUEN
Department of Electrical Engineering and Computer Science Northwestern University Evanston Illinois 60620 USA
NONCLASSICAL LIGHT
H P Y U EN
1. IN T R O D U C T IO N
A synopsis on nonclassical light will be presented, em phasizing the roles of dif ferent quan tu m amplifiers, th e problem of overcoming the detrim ental effect of loss, and highlighting certain points not explicitly brought out before. It is very far from a com prehensive review. In particu lar, no experim ent will be described. A useful recent review on squeezed light w ith considerable scope has been given by Loudon and K night (1987). No corresponding review on sub-Poissonian light is available, p artly because it is as yet m uch less developed com pared to squeezed light. 2 . NONCLASSICAL LIG H T — SQUEEZING AND ANTIBUNCHING
T he “vacuum ” is filled w ith free electrom agnetic field in its ground state. For a single m ode w ith annihilation operator a and num ber operator N = a^a, the ground state |0 > is an eigenstate of N as well as a. W hen a m ean am plitude a = oq + i a 2 for the field quad ratu res cq, a 2, a = cq -f ia2, is added to the vacuum , we have a “classically” excited m ode in a coherent sta te \a > , a\a > = a \a > (G lauber 1963). A classical state is, by definition, a coherent state or a random superposition of coherent states, i.e., a sta te w ith a tru e probability density P -representation.
Such states
are obtained in all th e conventional light sources. T here are su b stan tial quantum fluctuations even in a pure coherent state. T he q u ad ratu re fluctuation in \a > is ( A a l) = j ,
= (iia + va*)\fiva > , |/i |2 - \v\2 = 1
(5)
T he photocount statistics of |pit/a > can be sub-Poissonian depending on the p aram eters. It is generally spiked at two photons apart; in p articular, (fivo\n) ^ 0 only for even n. T he m inim al q u ad ratu re fluctuation in \pLva > is (A a l ) = 7 (W - M )2
(6)
w ith th e corresponding m axim al (Aa^o+i ) = \(\p\ + \v\)2 so th a t the uncertainty p ro d u ct ( A a l ) ( A a l +«_) > ^
(7)
achieves its m inim um value a t o, which is a function of pi and v (Yuen 1976b). T he im portance of TCS com pared to other squeezed states in low noise application derives from th e fact th a t th e q u ad ratu re signal-to-noise ratio (—\ U / a , “ (A al)
(8)
is m axim ized, und er th e fixed power constraint t r p N < S', by an ap p ro p riate TCS am ong all states (Yuen 1976a). In particu lar, the m axim um value ( § ) a ^ S = 4 5 ( 5 + 1 ) greatly exceeds th e coherent sta te level (jy )J^ = 4 5 for large 5 . Since ( A N 2) = 0 in a num ber state \n > , it m ay appear th a t as
S N R » s (££>
(9)
3
H P YUEN
approaches infinity, near-num ber states are vastly superior to TCS. However, the discrete n a tu re of \n > greatly reduces this apparent advantage in m ost applications. In m any ways, TCS and num ber states are com parable and com plem entary. 4. C O H ER EN C E P R O P E R T IE S T he characterization of the coherence properties of a field depends on the (quan tu m ) m easurem ent we are interested in perform ing on the field. In m any applications, w hether a field is coherent or not is irrelevant, particularly in regard to the so-called higher order coherence. (We do generally need a field traveling in a relatively welldefined direction.) Here, we lim it ourselves to the usual first-order coherence de scribing interference, which is m athem atically characterized by the factorization of the correlation function G {riti) r2t 2) = t r p E ^ i r i t ^ E i r f a )
(10)
into the p roduct of two c-num ber fields. Physically, this coherence condition expresses the sp atial-tem poral coherence of the field. It is equivalent (T itu laer and G lauber 1966) to having only one (general) space-tim e m ode excited to an arb itra ry state, while th e other modes rem ain in vacuum. Thus, a num ber state field can be perfectly space-tim e coherent, as can a TCS field. It is also well known th a t a m ultim ode excited coherent sta te field is coherent to all orders (G lauber 1963). How does this reconcile w ith the above single m ode condition? It tu rn s out th a t a m ultim ode coherent sta te field can be described, via a m odal tranform ation, as a single m ode coherent state field. This possibility can be traced to th e fact th a t th e vacuum sta te is a coherent state. In a squeezed vacuum where all the modes are in \fxu0 > w ith the sam e //, v , a m ultim ode excitation |fivcti > would be equivalent to a single m ode excitation in a sim ilar fashion, while an ordinary m ultim ode coherent state field would no longer be space-tim e coherent. This is ju st one illustration of the im p o rtan t fact th a t the vacuum sta te is a coherent state. On the other hand, although the vacuum sta te is also a num ber state, a m ultim ode num ber sta te field is not equivalent to a single-mode one, i.e., not space-tim e coherent. This fact can be understood in term s of the unique linear transform ation properties of G aussian states, i.e., states w ith G aussian characteristic functions.
A detailed
developm ent will be given elsewhere. 5. G EN ERA TIO N , PRO PA G A TIO N , AND D E T E C T IO N T he generation of TCS and NNS via nonlinear optical processes are, in some sense, surprisingly similar. This sim ilarity is in addition to th a t between TCS and
4
N O N C L A S S IC A L L I G H T
num ber-phase m inim um uncertain ty states when the nonclassical effects are sm all (Y. Y am am oto, N. Im oto, and S. M achida 1986). It seems to arise from an underlying intrinsic q u an tu m correlation th a t is w orth fu rth er exploring to great depth. Nonclassical light do not p ropagate well. This is because significant loss, be it radiative, absorption, or otherw ise, is often incurred during propagation and as we will see in th e following section, ordinary linear loss re-introduces coherent state fluctuation into th e field. Each ordinary optical detection scheme corresponds to the qu an tu m m easurem ent of a certain observable — in direct detection one m easures iV, in hom odyne detection cty, and in heterodyne detection a (Yuen and Shapiro 1980).
T hus to utilize the
advantage of a large ( 8 ) one hom odynes, or balance-hom odynes (Yuen and C han 1983) for com pletely elim inating th e local oscillator noise. To get a large (9) one counts photons. However, th e effect of a nonunity quantum efficiency is equivalent to linear loss (Yuen and Shapiro 1978a). It seriously degrades the nonclassical effect as follows. 6 . LOSS AND A M PLIFIC A TIO N
T he effect of linear loss on a m ode a can be represented by the transform ation to an o th er m ode b (Yuen 1975, Yuen and Shapiro 1978b, Yuen 1983), b = rjza + (1 — rj)*d
( 11)
w here th e d-m ode would be in vacuum except when a special m edium is involved (Yuen 1975). It follows im m ediately from ( 11) th a t the resulting q u ad ra tu re fluctu ation is, for a rb itra ry a-m ode state, (A b l )
=
tj {
A al) +
(1 - »?)/4 > (1 - »?)/4
(12)
T hus, a noise floor is introduced in addition to the usual signal atten u atio n , thereby greatly reducing th e SNR advantage of a TCS a-m ode. Similarly, the d-m ode would in troduce p a rtitio n noise into th e b-m ode when the a-m ode is in a num ber state. From (11) we have, for any a-m ode state, (A7V62) = 772 2 . A straightforw ard com putation from (14) leads to (A N 2) - {N b) = G 2{ A N 2) + G (G - 2){Na) + (G - l )2
(16)
which shows th a t any sub-Poissonian statistics in a would disappear in b for G > 2 also. These features are, of course, confirmed in detailed calculations involving more explicit linear am plifier models (Hong, Friberg, and M andel 1985; M ander, Loudon, and Shepherd, this volume). It is clear th a t PIA would seriously degrade the SNR for TCS or NNS. From (14)-(15), S N R b. = S N R J
1+
G- 1 1 G 4 1/G. Sim ilar to the PSA case, let the PNA o u tp u t b pass through a linear atten u ato r of final o u tp u t c, w ith the resulting (Yuen 1986b) S N R Nc = S N R n J
(26)
From (26), S N R Nc ~ S N R Na for a sufficiently large G.
(27)
If a PIA is used in place of the PNA, we would have
S N R Nc < S N R jv a for NNS inp u t and S N R jy c ~ ^ S N R mq for coherent state input. Again, detection device noise can be suppressed in a sim ilar m anner. E quations (19), (21) and (24) show th a t different amplifiers are suited fo r different detection schemes, regardless of the nature of the source. Of course, different detec tion schemes are n atu rally suited for different nonclassical sources. We sum m arize this situ atio n in the following table: SOURCE Coherent state TCS NNS
D E T E C T IO N heterodyne hom odyne direct
A M PL IFIE R PIA PSA PNA
It should be clear th a t PNA is indeed a n a tu ra l amplifier, com pleting the triad for m atching the three sta n d a rd detection schemes.
We need P SA to preserve the
advantage of squeezing, and P N A that of antibunching. In addition, PSA and PNA are useful even for coherent state sources, as they are still 3db superior to PIA in hom odyne or direct detection th a t are som etim es usefully employed for detecting
8
N O N C L A S S IC A L L I G H T
coherent sta te light. It should also be em phasized th a t they are necessary if one w ants to overcome loss or detection noise by pream plification. 10.
A PPLIC A TIO N S Nonclassical light w ith m atching detection leads to significant im provem ent in
S N R , as we have seen. Even in situations where S N R is not the m ost ap propriate perform ance m easure, it is clear th a t the low noise characteristics of TCS and NNS, or sim ply th eir difference from classical states, could be useful. T he m any applications th a t have been suggested for them , to my knowledge, are listed in the following: • C om m unications (Yuen 1975; Yuen and Shapiro 1978b, 1980; Shapiro, Yuen and M achado-M ata 1979; Yuen 1987). • F ib er T apping (Shapiro 1980; Yuen 1987). • Gyros (D orschner, Haus, et. al. 1980). • Interferom etry (Caves 1981; B ondurant and Shapiro 1985; Yuen 1986b; Yurke, M cCall and K lauder 1986). • O ptical C om puting (Y am am oto). • O ptical M em ory (Levenson).
• Precision M easurem ents (Yuen 1976b; Kimble; Slusher). • Spectroscopy (G ardiner 1986; M ilburn 1986; Yurke and W h ittak er 1987).
REFERENCES B o n d u ran t, R. S. and Shaprio, J. H. 1984, Phys. Rev. D, 30, 2548. Caves, C. M., 1981, Phys. Rev. D, 23, 1693. D orschner, T. A. H aus, H. A., et al, 1980, IE E E J. Q uant. E lectron., 16. 1376. G ardiner, C.W . 1986, Phys. Rev. L ett., 56, 1917. G lauber, R. J., 1963, Phys. Rev., 131. 2766. Haus, H.A., and M ullen, J. A., 1963, Phys. Rev., 128, 2407. Hong, C. K., Friberg, S., and M andel, L., 1985, JO SA -B, 2, 494. Loudon, R., and K night, P. L., 1987, J. Mod. O pt., 34, 709.
H P YUEN
M ander, G., Loudon, R., and Shepherd, T ., this volume. M ilburn G. J., 1986, Phys. Rev. A, 24, 4882. Shapiro, J. H., 1980, O ptics L ett., 5, 351. Shapiro, J. H., Yuen, H. P., and M achado-M ata, J. A., 1979, IE E E Trans. Inform. Theory, 25, 179. T itu laer, U. M., and G lauber, R. J., 1966, Phys. Rev., 145. 1041. Yam am oto, Y., Im oto, N., and M achida, S., 1986, Phys. Rev. A, 33, 3243. Yuen, H. P., 1975, Proceedings of the 1975 Conference on Inform ation Sciences and Systems, John Hopkins U niversity Press, B altim ore, pp. 171-177. Yuen, H. P., 1976a, Phys. L ett. A, 56, 101. Yuen, H. P., 1976b, Phys. Rev. A, 13, 2226. Yuen, H. P., 1983, Q uantum O ptics, E xperim ental G ravitation, and M ea surem ent Theory, P. M eystre and M. 0 . Scully, Eds., Plenum , New York, pp.
249-268.
Yuen, H. P.,
1986a, Phys.
L ett. A, 113. 405.
Yuen, H. P.,
1986b, Phys.
Rev. L ett., 56,2176.
Yuen, H. P.,
1987, O ptics
L ett., 12, 789.
Yuen, H. P.,
and C han, V. W . S., 1983, O ptics L ett., 8 , 177.
Yuen, H. P.,
and Shapiro, J. H., 1978a, Proc. of the F ourth R ochester
Conf. on Coherence and Q uantum O ptics, L. M andel and E. Wolf, Eds., Plenum , New York, pp. 719-727. Yuen, H. P., and Shapiro, J. H., 1978b, IE E E Trans. Inform. Theory, 24. 657. Yuen, H. P., and Shapiro, J. H., 1980, IE E E Trans. Inform. Theory, 26, 78. Yurke, B., McCall, S. L., and K lauder, J. R., 1986, Phys. Rev. A, 23, 4033. Yurke, B., and W h ittaker, E. A., 1987, O ptics L ett., 12 , 236.
SINGLE-ATOM OSCILLATORS
H W ALTHER
M o d e rn m e th o d s o f l a s e r s p e c t r o s c o p y a l l o w t h e s t u d y o f s i n g l e a to m s o r
io n s in an u n p e r tu r b e d e n v iro n m e n t.
up i n t e r e s t i n g
new e x p e r i m e n t s ,
o f ra d ia tio n -a to m of th is
c o u p lin g .
In th e
ty p e a r e re v ie w e d : t h e s in g le - a to m
s im p le s t
and
m o st
fu n d a m e n ta l
r a d i a t i o n - m a t t e r c o u p lin g i s
a
a c tin g
o f an
w ith a s i n g l e
c a v ity . s h o rtly th e
T h is
m ode
p ro b le m w as o f
p u re ly
so
s u f f ic ie n t to
th a t
th e
le a d to
g re a t deal
th a n th e o th e r c h a r a c t e r i s t i c e x c ite d
s ta te
life tim e ,
of
a to m -fie ld
of
h o w e v e r, (a )
te s t
ra d ia tio n -m a tte r
in te ra c tio n a
s in g le
s in g le
h o w e v e r, m a trix
a re u s u a lly
p h o to n i s
e v o lu tio n
in a
a tte n tio n th e
not
tim e s h o r t e r
tim e s o f t h e s y s te m ,
e x p e rim e n ta lly th e in te ra c tio n .
su ch a s th e
I t w as t h e r e f o r e
fu n d a m e n ta l t h e o r i e s
T h ese
th e o rie s
and b a s ic e f f e c t s .
a to m
d i s a p p e a r a n c e a n d q u a n tu m
(b )
th e o s c illa to r y
and th e
c a v ity
m o d e,
o f a s in g le
e n e rg y exchange and
re v iv a l o f o p tic a l n u ta tio n
i n a s i n g l e a to m b y a r e s o n a n t f i e l d .
p re d ic t,
T hese in c lu d e th e
t h e s p o n ta n e o u s e m is s io n r a t e
a to m i n a r e s o n a n t c a v i t y , b e tw e e n a
fie ld
of
t h e t i m e o f f l i g h t o f t h e a to m t h r o u g h
som e i n t e r e s t i n g
m o d ific a tio n o f
a to m i n t e r
in te re s t:
t h e c a v i t y a n d t h e c a v i t y m ode d a m p in g t i m e . n o t p o s s ib le to
s tu d y in g
A t t h a t tim e ,
a c a d e m ic
fie ld
an
fo r
e le c tro m a g n e tic a
e le m e n ts d e s c r ib in g th e r a d ia tio n - a to m to o s m a ll,
io n .
tw o -le v e l
a f t e r t h e m a s e r w as i n v e n t e d .
p r o b le m
m aser and th e s tu d y s to re d
s y s te m
s in g le
re c e iv e d
s tu d y
f o l l o w i n g tw o e x p e r i m e n t s
o f th e re so n a n c e flu o re s c e n c e o f a s in g le The
T h is h a s opened
am ong th e m t h e d e t a i l e d
(c )
th e
in d u c e d
11
H W A L TH E R
The s i t u a t i o n
c o n c e rn in g
th e
b a s ic e f f e c ts has d r a s tic a lly s in c e
e x p e rim e n ta l t e s t i n g changed
th e l a s t
fe w y e a r s
f r e q u e n c y - t u n a b l e l a s e r s now a l l o w p o p u l a t i o n o f h i g h l y
e x c i t e d a to m ic s t a t e s
c h a ra c te riz e d
num ber n o f t h e v a le n c e
e le c tro n .
c a lle d
s in c e
R y d b e rg
s ta te s
d e s c r ib e d by th e ra d ia tio n -a to m s ta te s
h ig h
m a in q u a n tu m
T h ese s t a t e s
by
a re g e n e ra lly
th e ir
fo r
th re e
r a t e s b e tw e e n
F in a lly ,
th e s e tr a n s it i o n s
la rg e to
have
and G a lla s ,
The s t r o n g c o u p lin g o f
e v o lu tio n
p rin c ip le :
le v e l;
th e
o s c illa tin g
see
to
H a ro c h e and
w ith
th e re fo re
in c re a s in g
in te rm s o f th e n
e x c ite d
th e c la s s ic a l
e le c tro n
fre q u e n c y to co rresp o n d s
(1 9 8 5 ).
re so n a n t ra d ia tio n
can b e u n d e rs to o d
tra n s itio n
a to m
s till tim e s .
lo n g l i f e t i m e s w ith
re v ie w s
a to m s
th e h ig h ly
(th e
s c a le as
c a v itie s th a t a re
For
R y d b erg
fre q u e n c y o f
i d e n t i c a l w ith th e
th e s e
fie ld
L e u c h s, W a lth e r an d F ig g e r
b e tw e e n n e i g h b o u r i n g l e v e l s c o rresp o n d en ce
F irs tly ,
in
i n t h e m i l l i m e t r e w ave
re la tiv e ly
s p o n ta n e o u s d e c a y .
(1 9 8 5 )
e ffe c ts
th e r a d ia tio n a re
be
h ig h ly e x c ite d
e n s u r e r a t h e r lo n g i n t e r a c t i o n
R y d b e rg s t a t e s
re s p e c t to R aim o n d
T he
can
n e ig h b o u rin g l e v e l s
w h ic h a l l o w s l o w - o r d e r m ode
s u ffic ie n tly
le v e ls
re a so n s.
a r e v e r y s t r o n g l y c o u p le d t o
in d u c e d t r a n s i t i o n re g io n ,
e n e rg y
f o r o b s e r v i n g t h e q u a n tu m
c o u p lin g
S e c o n d ly ,
a
s im p le R y d b erg f o r m u la .
a to m s a r e v e r y s u i t a b l e
n4) .
in
o f th e s e
th e to
w ith th e re s o n a n c e fre q u e n c y .
becom es
n e ig h b o u rin g
a la rg e
(T h e d i p o l e
d ip o le m om ent
i s v e r y l a r g e s i n c e t h e a t o m i c r a d i u s s c a l e s a s n 2 .) In
o r d e r to u n d e rs ta n d th e
e m is s io n r a t e in
m o d ific a tio n
in an e x te r n a l c a v ity ,
q u a n tu m e l e c t r o d y n a m i c s
d e n s ity
of
tra n s itio n
th is
ra te
is
fre e
c o n tin u u m o f
sp ace, m o d es
but is
t h e s p o n ta n e o u s re m e m b e r t h a t
d e te rm in e d
m o d es o f t h e e l e c t r o m a g n e t i c f i e l d fre q u e n c y C O q .
a t th e
a to m ic
fre q u e n c y .
in a re s o n a n t changed
in to
c a v ity a
I f t h e a to m i s in s te a d ,
s p e c tru m
is
ra d ia te d
e n e rg y
d is s ip a tio n
a t a w e ll-d e fin e d
w ith in
th e
c a v ity ,
th e
of d is c re te
m o d es w i t h o n e o f th e m b e i n g i n r e s o n a n c e w i t h t h e a to m . th e re
by th e
T he v a cu u m d e n s i t y o f m o d es p e r u n i t
v o lu m e d e p e n d s o n t h e s q u a r e o f t h e n o t in
of
we h a v e t o
a
S in c e p h o to n
fre q u e n c y w i l l b e sm e a re d o u t o v e r
12
th e
SIN G L E -A T O M OSC ILLA TO R S
fu ll
s p e c t r a l w id th A ^ c o f
w i d t h a t h a l f maximum A U)c i s
t h e r e s o n a n t m o d e.
re la te d
to
th e
The f u l l
c a v ity
q u a lity
f a c t o r Q = U)c /A tO c# The
s p o n ta n e o u s d e c a y
enhanced in r e l a t i o n g iv e n by th e r a t i o
ra te to
o f t h e a to m
th a t
in
in th e c a v ity
fre e
space ^
o f t h e c o r r e s p o n d i n g m ode
/c
by a
is
fa c to r
d e n s itie s
(Vc i s
t h e v o lu m e o f t h e c a v i t y ) : = f c (w0)/?f(w 0) = 27rQ/Vcu03 = QX03/4tt2Vc .
1 ch f
For
lo w -o rd e r
c a v itie s
in
vc
A q 3; th e
s p o n ta n e o u s
in c r e a s e d by a f a c t o r o f Q th e decay r a te case
m ic r o w a v e
e m is s io n
ra te
d e c r e a s e s w hen t h e c a v i t y
accept
it,
re g io n
one
i s th u s
ro u g h ly
in a re s o n a n t c a v ity ;
t h e a to m c a n n o t e m i t a
a b le to
th e
p h o to n ,
has
c o n v e rs e ly ,
i s m is tu n e d .
In t h i s
s in c e th e c a v ity
and t h e r e f o r e th e e n e rg y h a s t o
is
not
s ta y w ith
t h e a to m . R e c e n tly ,
q u i t e a few
e x p e rim e n ts h a v e b e e n
c o n d u c te d
w ith
R y d b e r g a to m s t o d e m o n s t r a t e t h e e n h a n c e m e n t a n d i n h i b i t i o n s p o n ta n e o u s
decay
s tru c tu re s .
For
in
th e
e x te rn a l m ost
c a v itie s
or
r e c e n te x p e rim e n t
of
c a v ity -lik e
se e Ih e e t a l .
(1 9 8 7 ) . T h e r e a r e a l s o m o re m ode d e n s i t y :
s u b tle e f f e c ts
ra d ia tio n
t h e a n o m a lo u s m a g n e t i c
c o rre c tio n s
d i p o l e m om ent o f t h e e l e c t r o n
m o d ifie d w ith r e s p e c t t o th e c u la te d u n d er
th e
b o u n d a ry
1987) . The c h a n g e i s
due to th e change
fre e
sp a c e v a lu e
c o n d itio n s
if
ac c u ra cy .
a r e d e te rm in e d by v i r t u a l t r a n s i t i o n s
fo llo w in g ,
a to m m a s e r in te ra c tin g
R o u g h ly
s p e a k in g ,
one can
say th a t and n o t by
a s in th e c a s e o f s p o n ta n e o u s d e c a y . a tte n tio n
i n w h ic h t h e w ith
(B a rto n
j u s t o f t h e o r d e r o f m a g n itu d e o f p r e s e n t
e x p e rim e n ta l
In th e
a re a ls o
th e y a r e c a l
o f a c a v ity
th e s e e f f e c ts
re a l tra n s itio n s
o f th e
s u c h a s t h e Lamb s h i f t a n d
a
is
f o c u s e d on d i s c u s s i n g t h e o n e -
id e a liz e d
s in g le
m ode
case of
a
o f a t w o - l e v e l a to m ra d ia tio n
fie ld
is
13
H W A L TH E R
re a liz e d ;
th e th e o ry o f t h i s
s y s te m w as t r e a t e d
by J a y n e s and
C um m ings (1 9 6 3 ) m any y e a r s a g o . We c o n c e n t r a t e o n t h e d y n a m ic s o f th e a to m -fie ld of
th e
in te ra c tio n
p r e d ic te d by t h i s
fe a tu re s a re e x p lic itly
n a tu re o f
th e
a consequence
e le c tro m a g n e tic
d is c r e te n a tu re
of
c h a ra c te ris tic s
such
fie ld :
th e
th e o ry .
o f th e
Some
q u a n tu m
s ta tis tic a l
th e
p h o to n f i e l d
le a d s
to
as
c o lla p s e and
re v iv a ls
new in
and
d y n a m ic th e
R abi
n u ta tio n . F i r s t we
r e v i e w t h e m a in r e s u l t s
w ith r e s p e c t t o t h e a to m ic
d y n a m ic s .
a to m i n t h e e x c i t e d s t a t e w h ic h a fie ld
o f n p h o to n s .
in th e e x c ite d s t a t e
o f t h e J a y n e s - C u m m in g s m o d e l We c o n s i d e r a t w o - l e v e l
e n t e r s a r e s o n a n t c a v i t y w ith
The p r o b a b i l i t y
Pe n o f t h e a to m t o b e
i s th e n g iv e n by
Pe , n ( t ) = 1/2 { l+ c o s[2 0 (n + l)1/ 2t]}
w h e re
n
is
th e
s in g le -p h o to n
R abi
f l u c t u a t i n g num ber o f p h o to n s i n i t i a l l y th e
q u a n tu m
R abi
s o lu tio n s
needs
p r o b a b i lit y d i s t r i b u t i o n p (n ) at t
freq u en cy .
W ith
a
p r e s e n t in th e c a v ity ,
to be
a v e ra g e d o v e r th e
o f h a v i n g n p h o t o n s i n t h e mode
= 0: oo Pe(t ) = 1/2 2 p(n) { l+ c o s[2 n (n + l)1/ 2 t ]} 4a~0
At a
lo w
th e rm a l
a to m ic - b e a m
flu x ,
th e c a v ity c o n ta in s e s s e n tia lly
p h o to n s an d t h e i r num ber
fo rm in g t o
B o s e -E in s te in
is
s ta tis tic s .
g i v e n b y P^.j1(n ) = n 11^ / ( ^ t h + l ) n + 1 *
a ra n d o m
q u a n tity con
In
case
th is
w ith th e a v e ra g e
p (n )
is
num ber
o f t h e r m a l p h o t o n s b e i n g n t h = [ e x p (h V /k T )
- l ] ” 1 . The d i s t r i
b u tio n
an
of
R abi fre q u e n c ie s
o s c illa tio n
Pe ^ th *
p h o to n s s t o r e d chan g es. t
= 0,
If th e
P o is s o n ia n : a d e p h a s in g
re s u lts
h i g h e r a to m ic - b e a m
in th e c a v ity a co h e re n t
p ro b a b ility p c (n )
in
is p re p a re d
d is trib u tio n
p (n )
s ta tis tic s
in th e c a v ity is
g iv e n
The P o is s o n s p r e a d
R abi o s c i l l a t i o n s ,
f i r s t e x h ib its a c o lla p s e .
f l u x e s t h e num ber o f
in c r e a s e s and t h e i r
fie ld
= e x p ( - n ) n n/ n !
o f th e
a p p a r e n t ra n d o m
T h is i s
a
in n g iv e s
and th e r e f o r e
d e s c rib e d
by
at
Pe ^c ( t )
in th e re s o n a n t
14
SIN G L E -A T O M OSC ILLA TO R S
c a s e by t h e a p p ro x im a te e n v e lo p e p e n d e n t o f th e a v e ra g e n o t h o ld f o r n o te d l a t e r
e x p ( - «fl.2t 2/ 2 )
p h o to n num ber
(th is
n o n re so n a n t e x c i ta t i o n ) .
T he c o l l a p s e
t i m e s f o r w h ic h Pe ^c ( t )
L a te r
VQ' C ( t )
s ta rts
o s c illa tin g
th e n
e x h ib its
a g a in
is
a l.,
re c o rre la tio n s
i n a v e r y c o m p le x
w as a l s o is
a ran g e
(re v iv a ls )
w ay.
and
As h a s b e e n
o c c u r a t tim e s
w i t h TR = 2 TT(n) ^ 2/S 1 l ( E b e r l y
= kTR (k = 1 , 2 , . . . ) ,
1 9 8 0 , N a ro z h n y e t a l . ,
1 9 8 3 ).
in d e
in d e p e n d e n t o f tim e .
sh o w n b y E b e r l y a n d c o - w o r k e r s t h e r e c u r r e n c e s e t a l.,
is
in d e p e n d e n c e d o e s
i n o t h e r w o rk . A f t e r t h e c o l l a p s e t h e r e
o f in te ra c tio n
g iv e n by t
and
1 9 8 1 , Yoo e t a l . ,
B o th c o l l a p s e a n d r e v i v a l s
1 9 8 1 , Yoo e t
in th e c o h e re n t s t a t e
a r e p u r e l y q u a n tu m f e a t u r e s a n d h a v e n o c l a s s i c a l c o u n t e r p a r t . The i n v e r s i o n
a ls o
c o lla p s e s
c h a o tic B o s e -E in s te in f i e l d th e
p h o to n -n u m b e r s p r e a d
s ta te
is
in th e c a se
R a d m o re ,
f a r l a r g e r th a n
o v e rla p and i n t e r f e r e
i r r e g u l a r tim e e v o lu tio n . s e n te d
by an
A
b u t no r e v i v a l s .
c o n s id e re d a s a
c le a r
q u a n tu m
of a
1 9 8 2 ).
H e re
fo r th e c o h e re n t In a d d itio n ,
to p ro d u c e
c l a s s i c a l th e rm a l
e x p o n e n tia l d i s t r i b u t i o n
show s c o l l a p s e ,
It
re v iv e s
a n d t h e c o l l a p s e t i m e i s m uch s h o r t e r .
r e v i v a l s c o m p le te ly
le s s
and
(K n ig h t an d
o f th e
fie ld
th e
a v e ry re p re
in te n s ity
a ls o
T h e re fo re th e r e v iv a ls can be fe a tu re ,
b u t th e c o lla p s e
is
c l e a r - c u t a s a q u a n tu m e f f e c t .
is
in te re s tin g
p ro c e s se s th e
t o m e n ti o n t h a t
in th e
R abi fre q u e n c y tu r n s o u t
c a s e o f tw o -p h o to n
t o b e 2 X l(n + 1 )
ra th e r
t h a n 2 f L ( n + l ) , e n a b l i n g t h e sum s o v e r t h e p h o t o n n u m b e rs i n Pe ( t ) th e
to be c a rrie d
in v e rs io n
seq u en ce
o u t in s im p le
re v iv e s p e r f e c tly
(K n ig h t,
in je c te d
in to
m odel i s a
w ith
fo rm .
a
In t h i s
case
c o m p le te ly p e r io d ic
1 9 8 6 ).
The e x p e r im e n ta l s e tu p u s e d f o r J a y n e s - C u m m in g s
c lo s e d
show n
th e e x p e rim e n ta l t e s t on
F ig .
s u p e rc o n d u c tin g c a v ity w as t h e
The
a to m s a r e
in th e u p p e r s t a t e
63 P 3
of
th e m aser
tra n s itio n .
p o p u la te d
by e x c i t a t i o n w ith fre q u e n c y -d o u b le d l i g h t o f a dye
la s e r.
T h is
1.
o f th e
/ 2
T h e a to m s a r e m o n i t o r e d b y u s i n g f i e l d
d e te c tio n can be
p e rfo rm e d s t a t e - s e l e c t i v e l y
R y d b e rg l e v e l , io n iz a tio n .
T h is
by c h o o s in g t h e
15
H W A L TH E R
p ro p e r f ie ld
s tre n g th .
Liqu id H e li u m Tem perature
F ig .
1
S chem e o f t h e s i n g l e - a t o m m a s e r f o r m e a s u r i n g q u a n tu m
c o lla p s e and r e v i v a l In m ost o f
th e
(R em pe, W a l t h e r ,
e x p e rim e n ts
K le in ,
(M e s c h e d e ,
1 9 8 5 ; R em pe, W a l t h e r a n d K l e i n ,
1987)
1987)
W a lth e r and
th e tr a n s itio n
M u lle r, 63 p ^ / 2
61 d 5y 2 w i t h a f r e q u e n c y o f 2 1 .4 5 6 GHz w as i n v e s t i g a t e d . th e
c a v ity
is
tu n e d
in
re so n a n c e
n u m b e r o f a to m s i n t h e u p p e r s t a t e
to
th is
s p o n ta n e o u s e m is s io n . The tu n i n g o f t h e c a v i t y s q u e e z in g th e c a v ity w ith p i e z o e l e c t r i c a to m s i s v e r y lo w s o c a v ity a t
t i m e o f t h e a to m s w i t h t h e c a v i t y s e le c to r.
t h e e n e r g y e x c h a n g e b e tw e e n in v e s tig a te d .
e le m e n ts .
l e s s th a n u n i t y . fie ld
In t h i s
th e
enhanced
i s p e rfo rm e d by
t h a t th e a v e ra g e num ber o f
a tim e i s u s u a l l y
o f a F iz e a u v e l o c i t y
tra n s itio n ,
d e c r e a s e s o w in g t o
~
When
The f l u x o f a to m s i n t h e
The i n t e r a c t i o n
c a n b e v a r i e d b y m ean s w ay ,
t h e d y n a m ic s o f
t h e a to m a n d c a v i t y
fie ld
can.
be
16
SIN G L E -A T O M OSCILLA TO R S
W ith
v e ry
lo w
e s s e n tia lly q u a n tity
a to m ic - b e a m
th e rm a l
a c c o rd in g
v e lo c ity
to
in te ra c tio n
a to m i c - b e a m
is
a
Pe ( t )
th e p r o b a b ility
c o lla p s e s ;
b o th c o lla p s e
A t h ig h e r
in th e c a v ity
th e p r o b a b ility
d is trib u tio n a fte r
re v iv a ls
and
sp re a d
ch anges.
d is trib u tio n
in n r e s u lt s
in
an d t h e r e f o r e t h e e n v e lo p e
th e in
th e
o f t h e a to m
ra n d o m w a y .
c o lla p s e
i n a v e r y c o m p le x w ay .
and
When
in c r e a s e s and t h e i r s t a t i s t i c s
T h is
a g a in
a ra n d o m
so t h a t t h e num ber o f p h o to n s
case o f a c o h e re n t f ie ld
o s c illa tin g
c o n ta in s
i n t e r a c t i o n v a r i e s w ith
an a p p a re n tly
th e th re s h o ld
P o is s o n ia n .
Pe ( t )
c a v ity
s ta tis tic s .
a fte r
d e p h a s in g o f th e R ab i o s c i l l a t i o n s , of
th e
T h e ir num ber i s
f l u x e s t h e a to m s d e p o s i t e n e r g y
in th e c a v ity
For th e
changed,
s ta te
tim e in
th e m aser re a c h e s s to re d
o n ly .
B o s e -E in s te in
o f t h e a to m s i s
b e in g in th e e x c ite d th e
flu x ,
p h o to n s
PeC ^)
s ta rts
As m e n t i o n e d a b o v e ,
th e c o h e re n t
s ta te
a re
p u re
q u a n tu m f e a t u r e s . The a b o v e -m e n tio n e d m e n ta lly . and r e v iv a l
have
been
o f m e a s u r e m e n ts
(R em pe,
p ro b a b ility
d e m o n s tra te d e x p e r i
c le a rly
show t h e c o l l a p s e
p r e d i c t e d b y t h e J a y n e s -C u m m in g s m o d e l.
show s a s e r i e s m aser
e ffe c ts
The e x p e r im e n ta l r e s u l t s
W a lth e r
Pe ( t )
l a r g e r N and a
th e s in g le -a to m
1 9 8 7 ).
P lo tte d
is
th e
o f f i n d i n g t h e a to m i n t h e u p p e r m a s e r l e v e l
f o r in c r e a s i n g a to m ic f lu x fo r in te ra c tio n
o b ta in e d w ith
and K le in ,
F ig u re 2
tim e s
N.
The
b e tw e e n
50
s tro n g v a r ia tio n a n d 80
r e v i v a l sh o w s u p f o r N
a c tio n tim e s l a r g e r th a n
140
s.
s
o f Pe ( t )
d is a p p e a rs
= 3000 s " 1
fo r
fo r in te r
The a v e r a g e p h o to n num ber in
t h e c a v i t y v a r i e s b e tw e e n 2 . 5 a n d 5 , a b o u t 2 p h o t o n s b e i n g d u e to
th e
b la c k -b o d y
fie ld
in
th e c a v ity
c o rre s p o n d in g
to
a
t e m p e r a t u r e o f 2 . 5 K. T h e re i s
a n o th e r a s p e c t o f th e s in g le -a to m
in te re s tin g : th e
c a v ity .
fo llo w in g .
th e n o n - c la s s ic a l T h is
is
s ta tis tic s b rie fly
o f th e p h o to n s d is c u s s e d
T h e r e a r e tw o a p p r o a c h e s t o t h e
t h e o n e -a to m m a s e r. a p p ro a c h to
p r o b le m
m a s e r w h ic h i s v e r y
F ilip o w ic z e t a l .
d e s c rib e th e d e v ic e .
(1 9 8 6 )
in
in th e
q u a n tu m t h e o r y o f u s e a m ic ro s c o p ic
On t h e o t h e r h a n d ,
L u g ia to e t
H W ALTHER
0s
£ CL
0) ■o
cO cn if)
-t—i— i— |— i— i— i 1— |— i— i— i— i— |— I— r- 30
0.7
T = 2 .5 K
N = 2 0 00 s ‘ 1
^ 0 .6 Q?
|0 ,5
50 "O ® "5 c 60 — ^ co
15 o n o ion trap
endcap ring electrode Mg beam
electron gun
electron beam
i-
10mm
Fig.
3
S c he me of the Paul tr ap u s e d for the e x p e r i m e n t s
(for de t a i l s see D i e d r i c h a n d Walther,
1987)
25
H W ALTHER
CN cn
-20
0
TIME Fig.
4
20
40
(nS )
Re su lt s for the i n t e n s i t y c o r r el at io n.
Antibunching
for a si ng le ion for d i f f e r e n t laser inte ns it ie s, fr om b o t t o m to top
(see D i e d r i c h and Walt he r,
in c r e a s i n g
1987 for details)
26
SIN G L E -A T O M OSCILLA TO R S
b e m e a s u re d f o r n e g a t i v e L . The l a s e r i n t e n s i t y a
to
d
and th e r e f o r e
th e
d e c r e a s e s fro m
a v e r a g e tim e i n t e r v a l
i n w h ic h a
s e c o n d p h o to n f o llo w s a f i r s t one i n c r e a s e s . For
a s in g le
io n
th e
in te n s ity
c o rre la tio n
fu n c tio n
is
g iv e n
by g(2) ( f ) = l - e ' W 4 [cosnt+(3Y/4fi)sinflt] ft2 = Q + A2 - (7/4)2 ,
w h e re
reso n an ce, o rd e r to
check
sq u a re s f i t
th e
to
dependence
on t h e
be
in te n s ity
th e
s ig n a ls o f F ig . c o u ld
b e in g
th e n a tu r a l lin e w id th ,
4 a ls o
e v a lu a te d .
A
th e d e tu n in g . o f IX 2 ,
dependence
d a t a p o i n t s w as in te n s ity
th e R abi fre q u e n c y a t
and
c o u ld
be
q u ite
(D ie d ric h and W a lth e r,
The f l u o r e s c e n c e
a s in g le
in te re s tin g
p ro p e rty :
in
th e
a s m a ll tim e
s to re d
e x p e c te d f o r a P o is s o n ia n d i s t r i b u t i o n ,
1 9 8 7 ).
io n h a s
flu c tu a tio n s o f th e i n t e r v a l 2
rT
(1 -
1*i) A.
ct)
dr.
-T
This expression clearly shows that sub-Poissonian statistics necessarily requires a negative function A(r). A trivial example is the monomode N-photon field which has a constant negative A(r) and
= 0.
The sub-Poissonian character was observed experimentally by Short and Mandel (1963 a-b) in the process of resonance fluorescence emitted from one atom. 2°/ The antibunching effect is another purely quantum effect. It is also due to a non classical behaviour of A(r). For classical fields A(r) has a bump at r=0 (Eq. 11), (’bunching effect’). But this property may be inverted, even in the case where A is positive. It is the case for photons spontaneously emitted in atomic beam experiment : the ’intensity’
40
PH O TO D E T EC T IO N A N D PH OTOSTATISTICS
correlation of the light is minimum for equal times tj = t2 (Kimble, Dagenais and Mandel, 1977 ; Cresser et al., 1982). This photon antibunching effect has a simple quantum explanation : just after the emission of a first photon, the atom is in its ground state and then it cannot immediately emit a second photon. A collective antibunching effect is reported in section VI. These two quantum effects (sub Poissonian statistics and antibunching) are distinct as it is clearly explained by Short and Mandel (1983-b, p 673). (b) Moreover one can imagineother sorts of detectors. Mandel (1966),for example, suggested to use the process of
stimulated emission as a basis for detection. The effectof shining thelight on
these spontaneously decaying atoms is that they emit photons rather than absorb them, therefore the operators would occur in anti-normal order in Eq. (17). Different operator orderings are associated with different sorts of experiments. Symmetrical ordering of the operators is also suggested (Glauber, 1970 , p. 72). The problem of ordering field operators and the correspondance between functions of c-numbers and functions of q-numbers has been systematically treated by Agarwal and Wolf (1968 a-b) and Lax (1968). (c) The above derivations suppose ideal photodetectors with broad band sensitivity function S(u>), much broader than the spectral width of the incident light. These expressions need to be modified when the optical field has a substantial bandwidth over which the detector response varies (see Eq. 14). Glauber (1965)indicates that the correlation function entering into the expression for
...tn) has to be convolved by a certain response function of the detector.
This problem was further investigated by Rousseau (1975) and Kimble and Mandel (1984) : a first order perturbation theory shows that the photocounting distribution is still given by Eq.(17), together with
kW
r rt+T
= I I
d t 1
A-
dt"" E
( t 1)
A+ E
*s lar8er than unity inside a cone of half-angle ec * 1.4 k a (N)1/3
(37)
where N is the number of atoms in an absorption length. The spatial correlation function G^(r ,t ; r ’,t) was investigated, it contains
(36)
46
PH O TO D E T EC T IO N A N D PHOTOSTATISTICS
a) N antibunching terms . b) bunching terms > n*Pc) The emission of two photons by atom p is also described by antibunching terms, , npep,
and
like
like a nd
time,
This and
wave
c a n be is
not
vanish
field
mechanical
at
o p t i c s is ze r o
so b e c a u s e
the
a measurement
mechanical
all
waves
one
at o t h e r
description
a
only
for
can
point
in
point.
This
of
inter
the
in
the
of
a free
the only
x^
in
Fig.
on
the p h o t o e l e c
is p r o d u c e d
of
light
the
description, part
form
of
the
at
the
operator
a single-frequency
classical
1 in
rely
annihilation
measurement
the p o s i t i v e - f r e q u e n c y
d i s t i n g u i s h e d by
light
electron
the p h o t o n
consider
as
description
detectors
in w h i c h
If we
the p o i n t a re
same
in d e s c r i b i n g
source
express
unity
that
quantum
of a p h o t o n ,
role
theory. each
of
We
variation
correlation
identically.
the
the
As p r a c t i c a l l y tric
depth
be
0.
effect.
The
expense
be
fact
affected
in
magnitude might
two
angle
| x ^ - x 2 |.
consequency
be
waves
i.e . ,
+
product
at a
up
term,
the
intensity
classical
together
effects
term
the
+ 2 < IA I B > + 2 < I A I R > c o s [2 tt( x 1 - x 2 )/L] . (4)
o f the
the
with
of
obtain
+ < Ig>
A coming
the
we
spacing
corresponding wavelength
because
average
assumption,
in
plays
the
quantum
contribution
t h e n we
f ie l d ,
(Hilbert-space
a
of
c an
E ^ + ^(x ^ )
at
operators
a caret)
E (' + h x i ) = a Ae x p ( i k r A i ) + a g e x p ( i k r B i ) ,
(6 )
55
C K H O NG , Z Y OU AND L M ANDEL
where
and
f i el ds
g e n e r a t e d by
f o l lo ws s ome
that
narrow p
^
1
l
the
the
s ig n a l
l
characteristic
of
state
th e n
|1
(6),
=
t he r e
calculation
the
However
2
o f the
that
For
detector
value A and
degenerate
appropriate
this
expectation
of
the
operator
B correspond parametric
state
signal
positioned
and
is
the
to
down-
two-
idler photons
state
and
the
in Eq.
(7)
is
the
are
field
operators
easily
evaluated
to
(8)
interference
t erm,
just
as
in
show
up
the
classical
intensity.
fourth-order
6x^
within
Ki 6 x i ,
interference
p r o b a b i l i t y p ^ 2 (x ^ , within
the
because
reduces
is no
of
the
,1 >,
together.
probability
that
at
(7)
expectation
modes
process,
P 1 ( x i ) 6 xi
the
the
(1963a,b)]
l
idler
produced
so
v l
suppose
an d
Eq.
on
It t h e n
a photon
[Glauber
If we
always
the
of d e t e c t i n g
s ta t e .
Fock
g i v en b y
acting
B respectively.
given by
l
a factor
given
conversion
operators
= K . 6 x . ,
photon
an d
is
x^ , a n d < > r e p r e s e n t s
for
A and
the p r o b a b i l i t y
l
is
annihilation
sources
range
(x.)fix.
where at
a^ a re p h o t o n
an d
)6x^6x2
effects
detecting
in w h i c h
case
in the
photons
[Glauber
at
x^
joint and
x2
(1963a,b)]
P l 2 (x i > x 2 ) 6 x i 6 x 2 = K 1 K 2 6 x i 6::2
(9)
+ 2 < a J a g a B a A > { 1 + cos [2tt ( x 1 - x 2 ) / L ] } ) . For
the
right
two-photon
vanish
state
a n d we
tEe
first
two
terms
on
the
obtain
P 12 ( x 1 , x 2 ) 6 x 1 6 x 2 = 2 K 1 K 2 6 x 1 6 x 2 { 1 + cos [2 tt( x ^ - x 2 )/L] } . (10) Interference classical 100
%.
This
position
shows
up
again
description,
can
implies
that
completely
another photon
at
but
as in
the rule
a cosine this
case
detection out
modulation, the
of a p h o t o n
the p o s s i b i l i t y
certain positions.
as
visibility at
in
the
n is one
of d e t e c t i n g
56
N O N C LA SSIC A L PH O TO N IN T E R FE R EN C E EFFECTS
It s h o u l d be particle
nature
separate appears of
the
under
noticed
positions
the
extended
which
same
mechanics
o f an
photons
regions
integrate
th the
the
% visibility
c a n n o t be
time.
This
formally
in
Fock
annihilation
is
due
to
the
detected
at
two
particle the
state
operator
nature
change
of
the
state
to
vacuum
state
the
representing
the
of a p h o t o n .
In p r a c t i c e ,
to
at
100
f r o m a s i n g 1e - p h o t o n
action
detection
the
of a p h o t o n ,
in q u a n t u m field
that
detected
A x ’s c e n t e r e d
the
measured
are not
foregoing
joint
at
x^
at and
expression
detection
a point but x^,
so
o v e r Ax.
probability,
over
t h a t we
ought
F r o m E q . (10),
P ^ 2 ^ X 1 ,X2 ^ ’
§ * ven
by x^+Ax/2
P1
2
(' Xl ’ X2 ' 1
x 20 + A x / 2
Xj - A x / 2 s
"A x/ 2
2 K 1 K 2 ( A x ) 2 {1 Due
to
the
r e d u c e d by is
finite the
applicable
to
fa ct o r was
2.4
The
The
the 0.55
Ax,
the
visibility
[s i n (ttAx/L) / (ttAx/L) ] 2 , a n d classical in
the
(11)
(- a * 7 1 ) cos (2 tt( x 1 - x 2 )/L] } . tA x / L
detector width
factor
this
+
P 1 ? ( X1 ’ X? ) d x i d x ?
calculation
experiment
too.
n has
the
same
been factor
In p r a c t i c e
d e s c r i b e d below.
experiment
outline
of
the
interference
experiment
[Ghosh
and
Mandel
Fig. 2 O u t l i n e o f the t w o - p h o t o n i n t e r f e r e n c e e x p e r i m e n t , [ r e p r o d u c e d f r o m G h o s h , R. a n d M a n d e l , L. 1987, P hys. Rev. Lett. 59 1 9 0 3 - 1 9 0 5 ]
57
C K H O N G , Z Y OU AN D L M ANDEL
( 1987)]
is
shown
oscillating crystal, and
the
where
some
A b e a m stop
cause
angle
the
an
T wo
plates,
g lass
a translator, direct
after
and
so
to
the
into
of
plane
amplification
and
a re
inputs
of a t i m e - t o - d i g i t a 1 c o n v e r t e r .
within
a 5 ns
interval
either
to
’s i m u l t a n e o u s ’ s i g n a l
the
accidental
overlap
subtracted
out,
are
treated
of u n c o r r e l a t e d
the
rate
of
R ELA TIVE
as
together
to
x^
pulses.
coincidence
and
x^
The re
mm. on
edge
on
pulses,
start
and
stop
arriving counts,
photons
When
and
an
mounted
pairs
counting
SEPARATION
at
L-0.34
and
coincidence idler
Ml
bandwidth.
whose
the
UV
passing
mm and at
LilO
signal
the
is m a g n i f i e d
Pulse
and
to
after
t ub e s ,
fed
long
mirrors
spacing
incoming photons
shaping,
two
rad/s
Ax-0.14
two p h o t o m u l t i p l i e r
cm
±3.3°
come
fringe
thickness
the
of
while to
laser
half-frequency
3 x l 0 13
t his the
a 1.5
crystal,
filter with in
on
angles
latter,
m from
to m a k e
each
split at
argon-ion
f al l s
idler photons
formed as
collect
them
the
f r o m an
li n e
emerge
1.1
pattern
a lens
n m UV
deflects
signal
i m a g e d by
Light
UV p h o t o n s
interference
interference
2.
which
6^2° in a p l a n e
through
and
Fig. 351.1
idler photons,
beam. M2
or
in
the
or
due to
the
latter, are
provides
a
(x-x')/L
Fig. 3 E x p e r i m e n t a l r e s u l t s s u p e r i m p o s e d on the p r e d i c t i o n s of q u a n t u m t h e o r y g i v e n b y E q . (11) ( s o l i d c u r v e ) , a nd o f the cl as si ca l t heo ry with m a x i m u m m o d u l a t i o n (dashed curve), [re p r o d u c e d f r o m G h o s h , R. a n d M a n d e l , L. 1987, Phys. Rev. Lett. 59^ 1 9 0 3 - 1 9 0 5 ]
58
N O N C LA SSIC A L PH O TO N IN T E R FE R EN C E EFFECTS
measure
of
the
joint p ro bability
P x 2 ^ X l , X 2 ^ :> aP a r t
^rom a scale
factor. Fig.
3 shows
I x^-x^l* by
the
The
E q . (11)
solid with
adjusted dashed
c ur v e
classical same
are
in
found
in
Ax^0.14
Fig.
The
the
values
3.
clearly
favor
MEASUREMENT
3 .1
3 is
L ^ O .34 mm,
a nd
TIME
t wo
size
are
The
the
with
CORRELATION
by
of 0.275 to
t he
factor
data.
prediction
The
the
for
the
agreement predictions
The p r o b a b i l i t i e s
of
P (y 2 ^ 4 .9)-0 . 1 8 a nd
experimental of
scale
the
theoretical
a n d X 2 U ~0.44.
t his
of
given
with
visibility
the
values
of
agreement
maximum
the p r e d i c t i o n
OF
different
a plot
corresponding
P (y 2 ^ 0 . 4 4 ) - 0 . 9 2 , r e s p e c t i v e l y . f ore
three
of y 2 c o r r e s p o n d i n g
values
of x 2 of
for
for b e s t
X 2 -La s s ~4 . 9
values
Fig.
mm and
3 is
measured
to be
obtaining
curve
theory with
L.
the
obtained
arbitrarily
wave
Ax and
between
results
the
results
quantum
BETWEEN
TWO
there
theory.
PHOTONS
Background
T he
t ime
correlation
between
in
the p a r a m e t r i c
by
B u r n h a m and W e i n b e r g
10 n s . The with
depended
and
lation
time,
Later
t im e
between
pulses
by
and
signal
making
use
What
they
tim e
between
In the
next
section,
out b y H o n g
ference
of
et a l . ( 1 9 8 7 ) ,
effects.
t ime,
T he
of d e t e c t i n g
signal
experiments
between
the two p h o the
but
discuss
experiment
a l . (198 5 ) ]
for
time
u se
not
of
not
correlation
so m u c h
intense
the
light
only
depended
yielded
idler
on
the
gener corre
pulses.
experiment
exhibited
a nd
the
intense
of s e c o n d - h a r m o n i c
two
corre
o f the p h o t o - d e t e c t
measured
by
therefore
also
of a b o u t
the
(1986)
which
but
et
p s . Both
intervals
measured
time
of
the p r o c e s s
will
first
produced
an u p p e r b o u n d
idler beams
two p h o t o n s we
100
resolving
m e a s u r e d was
correlation
probability
the
[Friberg
from detection
only
coworkers and
lation
time
was
photons
a resolving
of orde r
the
idler
process
with
resulting
limited by
Abram
of
and
repeated
time
provided
a t i on .
second
were
resolving
therefore
ors.
pulses,
(1970)
on m e a s u r e m e n t
photoelectric tons,
down-conversion
measurement
improved
signal
carried
a subpico
nonclassical measuring
photons
inter
the
produced
joint in
59
C K H O N G , Z Y OU A N D L M ANDEL
B
Fig.
4 The
schematic
o f the
ti m e
correlation
parametric
down-conversion
after
f i e ld s
m i x e d by
splitter,
3.2
are
The
L et
us
pulse
classical
suppose of
a n d Vg
tha t
are
e a c h s o u r c e in shape
g(t)
are
the
combined expressed
as
shown
in
light
Fig.
so
amplitudes,
the
and
to a b a n d w i d t h
that
assumed
field
4 emits at
a very
same
4.
J g 2 (t)dt
to be
falling
on
random
the
g(t)
short
t ime,
t*16 fi-e l d s f r o m s o u r c e s
complex
again
Fig.
duration
corresponding
is n o r m a l i z e d
f i el ds
and
are
constant
envelope
two
be
that
anc*
pulse
d o w n -converted
description
identical
t hat
a beam
the
experiment.
is
A and
B.
the
re a l
s ha l l
assume
= 1.
The ph as es
of the
and
unrelated.
Then
Ago.
detector
We
and
i at
time
t can
as
V .(t)
= v^v.g(t)
V 2 (t)
= l/fivA g(t)
+ i/ffv g ( t + x ) ,
( 12) w h e r e R a nd T are beam splitter, arrival
t i m es
position across
In Sec.
the
the
reflectivity
w i t h R + T = 1, of
the
two p u l s e s
ar gu m e n t has
the
been
crosssection
2 we
resolving to
+ /TVggCt+x),
ignored
times
any
o f the
description
of
of
and
t is
at
ignored, the
transmissivity the
difference
o f the in
the
the b e a m
splitter.
The
because
the p h a s e
difference
mixed b eam
consideration
detectors, the
and
is
constant.
of
the p u l s e
widths
because
they w ere
irrelevant
interference
effects
discussed.
and
But,
60
N O N C LA SSIC A L PH O TO N IN T E R FE R EN C E EFFECTS
in p r a c t i c e , taneous
what
is
m e a s u r e d by
intensity
of
the
resolving width by
time A t ,
6t*l/Aa)
detectors
calculated
which
o f g(t). 1 and
as
is
t he
their
2.
but
its
assumed
Then
2 and
in Sec.
/ < I 1 (t)>dt
field,
a detector
We
+ if< r B>
/ < I 2 (t)> d t = R < l p >
+ T < I b>
integral
to be
much
integrated
from
the
over
instan the
greater
than
intensities
c r o s s -c o r r e 1ation
obtain
= T< 1 p >
is n o t
C(t)
the
measured c a n be
E q . (12)
(13)
and c (T ) = / / < I 1 ( t ) I 2 ( t ' ) > d t d t ’ = TR< I^>
+ ri?
(14)
+ r 2 < I A I B > + ■ff2
- 2 2 7 ? h ( T ) , where
h(x)
It is
clear
in
the
=
f r o m Eqs. (13)
average
no p e r i o d i c x,
is
an
If we
[/g (t ) g (t + t ) d t ]2 £
define
term
that
to
n is
interference
x,
t he
Although
last
te r m ,
effects
show
E q . (14) which
up
exhibits
depends
on
term.
visibility
interference
no
intensity.
variation with
interference
E q . (14)
that
integrated
1.
n in
the
the
usual
constant
always
l ess
term,
way
as
t he
ratio
then
it
follows
t h a n or e q u a l
to
50
%,
of
the
from
for
2 T R h ( t ) < IA Ig>
(15)
0
TR< I2 > + TR< Ig>
The p h y s i c a l previous 5.5 The
quantum we
parametric quencies, only
the
photon
reason behind
example
Although
+
in Sec.
state
represented
the
produced by
the
¥> = /duxf) (u)q /
signal
process
same
as
discussed
in
the
+o), o)q /
a nd
idler
modes
of
as h a v i n g w e l l - d e f i n e d
have
frequencies
linear 2
the
actually
two
the
in
the
description
treated
the p h o t o n s
is
2.
down-conversion
su m of
t his
mechanical
earlier
(T 2 +i?2 ) < IA Ig>
a wide
bandwidth
is w e l l
defined.
down-conversion
process
A gj
the fre
, and
The
two-
c a n be
superposition 2
-a)) |
1
03o /
2
+UJ ’ coo /
2
- co
(16)
61
C K H O NG , Z Y OU AND L M ANDEL
where is
oj0 is
s ome
The
the
frequency
symmetric
f ie l d s
E ^ + ^(t)
on
detectors
by
the
2,
of the p u m p
E^^Ct) are
= /TEf + h t )
in
the
+ i
= i/ffE< + ) (t)
classical
the
t he
at 00 = 0 .
splitter
signal
and
falling
idler
+ ^ (t + t )
time
(17)
+ / T e J + ) (1: + t )
of
of d e t e c t i n g time
the
t,
photons
which
light,
is
is
by bo th
much
detectors
longer
than
any
giv en by
P 1 2 (t ) = K / / < E { ' ) ( t ) E ^ _ ) ( t ' ) E ^ + ) ( t ' ) E { + ) ( t ) > d t d t ’ and
can be
[Hong et P
1
where
(t)
2
readily
al.
C is
fields
description.
resolving
correlation
the b e a m
to
(J>(ojq / 2 + oj ,o)0 / 2 -oj)
is p e a k e d
*
j oi n t p r o b a b i l i t y
within
past
related
*
E ^ + ) (t)
b e a m a nd
function which
expressions,
~
The
and
1 and
following
E^ + h t )
as
weight
calculated
f r o m E q s . (16)
a nd
(18)
(17).
We
find
(1 9 8 7 ) ]
= C [T2 + R 2 another
2rah(x)]
(1 9 )
constant,
/ g ( t ) g ( t + T ) d t / / g 2 (t)dt,
(20)
h(T)
=
g(t)
= / (a>o / 2 + oj ,ojq / 2 - oj) exp ( - iw t) da)// (4) + ]
The last matrix element can be rewritten as
can
be
regarded
as
a
photodetection
This interpretation becomes clearer if we sum over the states ia> with a
weighting function R(a) incorporating the efficiency of detection and work with a single linear polarisation.
dt'
This gives the probability of detecting a photon to be
dt" S ( t " - t 1)
=
S |2 ( t 2 - i x 2 i) «3.
C—2 D3 - n. a). -------C-ib - ruo)_ -------2 2 |*2 | 1 1 1*1 I
(—2 ] 1
exp
1 1 2
( t j - I Xj 1 ))
2
1 - 2 1
C—1 ] 1 IX Kll
(28)
-jmn,
F
N
(ct) 1 ,o 2 , x 1 ,x2)
where
F J m n (a>1
e/ g.1 « 2
**2 )
\ ( 2 ) imn. N^ 1 -“ 2 ) X (oJo ,fa)l ’aJ2 ) £ i ^ n ^ i
[X2] i [X2] j
and the quantum state of the incoming p u m p is given by a coherent state with amplitude
76
PH O TO N S A N D A P P R O X IM A T E L O C A LISA BILITY
(The field could have a transverse profile but this is incorporated through (27).) the nonlinear susceptibility tensor.
On
performing the r integration we
x(2) is
obtain as a
factor in the remaining a) integration (Pike and Sarkar, 1987)
2' '
r
*3 4
n 2 0>2
0
c
[x 2] 3
n l°>l Cx l ] 3]
|~ 2 |
c
J
|* l|
2' rn2“ 2tX2]l ( n l“ l[x l]ll 4
Since
£3
and
c
£j_ are
IX2 1
large
+
c
compared
|* l|
to
J
kQl,
the
dominant
contribution
to
the
co integration is given by
k
o
=
n2 w 2 c
cos0~ + 2
ni^i c
cos#. 1
(29)
where
-----
|x.|
-
cose.
,
i - 1,2
(31) Pill -------I
=
(,- 11N) is i.n #, .
,
.
1
=
1 0 1 , 2
1
The equations (29) and (30) are the phase-matching conditions. valid in the limit £3 ^ , (satisfying oy[+o^
=
£j_ko
°°*
They would be strictly
Given any two values of oq and a)2>
co0), it is possible to find d \
and
an^
^2
satisfying (29) and
say (30).
Mollow (1973) considered this nearly phase-matched situation and approximated (28) by
77
E R PIKE A N D S SA R K A R
G( 0 >2 ) (x x , iC 2 ’ 1
U
0
(JV
0 do3j do)^
^
r
.
(1
2
2 r2
V
,3 d r exp
exp (-
«i
2-1
r3 4
(32)
i“2 (t2_la2 l)
exp (irj(ko -
Vr "k (2 ) j m n / 0>i0)~ 5(0) -03.,- t O xv /J 1 2 o 1 2
*
~
[X2 ] i [ X2 ] j ]
* 1-1
12
I 1- 2 |
1-2 I
(where the two detectors are taken to be in the plane ( * \ ) 2
= (x2)2 = 0)*
The
o) dependence is kept only in the exponent since in the phase-matched situation this is the most rapidly varying term.
Since we
assume
that the frequencies that we
are
concerned with are far away from any crystal resonances, it is satisfactory to consider X(2)jmn to be independent of frequency.
However in the search for photon tails it is
not adequate as done by Mollow to replace the o^c^ in (28) by o>Jg>2 .
In fact in this
example the interesting aspect of the photon tail comes precisely from the After performing the r integration the o3j and
032
integration is
03^0)2
factor.
78
U
O
PH O TO N S A N D A P P R O X IM A T E LO C A LISA B IL ITY
Uu
2'
O do)^ do)^ exp
i
'
r [°
" 4
n2u 2 [x2] 3 c |X2 |
n l Wl CX1 ] 3~| c IXj , j
rI
r n2w2[ x 2 ] 1 ( n i w i t x i ] i i * 42 C |~21 + C 1*11J
exp
exp [ ” ict)2 t 2] exp [ ~ ia,l t l ]
2(33)
5 (a)0 ~a,i “a)2^
with
t. 1
=
t. l
n. l
PM
(i - 1 ,2 )
c
The o)i integration is automatic with the use of the 5-function.
The Fourier theorem
concerning convolutions allows as to the rewrite the o>2 integration as
exp
dr I
b v i]
[ W r]
(3 4 )
12 ( r )
where
I x (r )
do)
I 2 (t)
do) exp
6(o))
6 ( o )q - o ) )
( o )q - o ) ) o )
ex p (-ia n
)
(35)
and
Here
^ ^Uu^-2Vuj - iu r j
(3 6 )
E R PIKE A N D S SA R K A R
U
-
^ 3
=
£ 3
j^n^cos^^ - n ^ c o s f l j J
79
*t2 + n ^ sinfl^ J
+
(3 7 )
and
V
+
2 r
2
|^kQ - njG)QC o s 0 j J
r
*
[n 2 cos02
- n^cos0^J
^ r ,3^'j n^a)Qs i n 0 j |^i2s i n 0 2 + n ^s i n0 j j
(38)
It is easy to show that
2o)
-ir-
to)
sin
Ia (r)
(3 9 )
and
I2(r)
Jw e x p [ * i H e x p [ - 1r ) e x p [ - * i f ]
Clearly I2 is exponentially localised.
(4 0 )
Ij provides an oscillating power-law tail.
In
contrast to the approximately localised photon formed through a generalised imprimitivity which has a photodetection probability falling off as r - ^, the asymptotic behaviour in parametric down-conversion is
TO)
Ot
Sin
—T r2
Moreover this localisation is highly anisotropic.
(4 1 )
In directions perpendicular to those
defined by 0* and 02 there is, in lowest order, no fall-off. approximate
phase
matching allows
only a limited
contribute to the idler and signal photons.
This is because the
directionality
In the directions
of
wavevectors
to
or 02 although a range
of frequencies are allowed, the range is not large enough to obtain the weaklv localised
80
PH O TO N S A N D A PP R O X IM A T E LO C A LISA B IL ITY
states of Jauch, Piron and Amrein.
The main reason for this restricted range is due to
the details of the parametric down-conversion experiment.
We are thus still some the laboratory.
way off
Indeed it
down-conversion, which
from producing
would seem
that any
themost localisedphotons possible in experiments, such
as
parametric
have a preferred direction ofwavevector (such as in the input
pump) are not likely to have available through atom-field interactions enough different types of wavevector to
give a localised photon states approaching
the optimal weakly
localised states.
ACKNOWLEDGEMENTS
We would like to thank C. M. Caves, J. G. Rarity and P. R. Tapster for discussions.
REFERENCES Amrein, W.O. 1969, Helv. Phys. Acta 42 149. Burnham, D.C. and Weinberg, D.L. 1970, Phys. Rev. Lett. 25 84. Friberg, S., Hong, C.K. and Mandel, L. 1985, Phys. Rev. Lett. 54 2011. Glauber, R.J. 1963, Phys. Rev. 131 2766. Hillery, M. and Mlodinow, L.D. 1984, Phys. Rev. A30 1860. Jakeman, E. and Walker, J.G. 1985, Opt. Commun. 55 219. Jauch, J.M. and Piron, C. 1967, Helv. Phys. Acta 40 559. Mackey, G.W. 1978, Unitary Group Representations in Physics, Cumming, Reading. Mollow, B.R. 1973, Phys. Rev. A8 2684. Newton, T.D. and Wigner, E.P. 1949, Rev. Mod. Phys. 21 400. Pike, E.R. 1986, Coherence, Co-operation and Fluctuations, Ed. H. Haake, L.M. Narducci and D.F. Walls, Cambridge University Press, Cambridge. Pike, E.R. and Sarkar, S. 1986, Frontiers in Quantum Optics, Ed. E.R. Pike and S. Sarkar, Hilger, Bristol. Pike, E.R. and Sarkar, S. 1987, Phys. Rev. A35 926. Pike, E.R. and Sarkar, S. 1987, Power Law Tails ofSingle Photon States inParametric Down Conversion, RSRE Malvern Preprint. Pike, E.R. and Sarkar, S. 1988, Photon Localisation in Parametric Down Conversion, RSRE Malvern Preprint Rarity, J.G., Tapster, P.R. and Jakeman, E. 1987, Opt. Commun. 62 201. Wightman, A.S. 1962, Rev. Mod. Phys. 34 845.
QUANTUM NOISE REDUCTION ON TWIN LASER BEAMS
E GIACOBINO, C FABRE, S REYNAUD, A HEIDMANN AND R HOROWICZ L a b o r a t o i r e de S p e c t r o s c o p i e H e r tz . ie n n e de l ' E . N . S . U niversity 4,
1.
Pierre
place Ju ssieu
e t M arie C u r ie ,
75252 P a r i s
T12-E01,
Cedex 05,
France.
INTRODUCTION
Among
the
processes
classical"
properties,
turns
to
out
Takahasi
be
1965,
the
a
M ol l o w
et
1 9 7 3 ) . The p r o c e s s
fields
at
frequencies w ith
conservation placed
a
requires
Degenerate threshold
and
( L o u i s e l l e t a l . 1961,
Mollow 1973,
cc>2
by
S toler
at
a non-linear
pump
field
that
coq = ^ + co2 ) .
frequency
1974,
coq
crystal (energy
The c r y s t a l
cavity operating
P aram etric
c a n be
e i t h e r below or
et
a l. al.
A m plifiers
(OPA)
(below
to x o>2 ) , w h i c h h a v e b e e n s h o wn b o t h
(M ilburn e t a l .
1981,
1984,
experim entally
(Wu e t
lig ht.
the n o n - lin e a r
A ctually
interactio n
i n v o l v e s g e n e r a t i o n o f t wo " s i g n a l "
co^
o p e ra tio n w ith
C ollett
"non
m ost o f t h e work ha ve c o n c e n t r a t e d on Q u a s i
O ptical
theoretically
param etric
one
1967,
l i g h t w ith
t h r e s h o l d d e p e n d i n g on t h e pump p o w e r .
Up t o now,
1984,
al.
in a resonant o p tic a l
above o s c i l l a t i o n
generate
optical
prototypic
Gr aha m
irradiated
which
1986)
L u g i a t o e t a l . 1982,
C ollett to y ie ld
interaction
et
al.
Yurke
1985)
squeezed s t a t e s taking place
in
and of the
82
OPA
Q U A N T U M N O ISE R E D U C T IO N ON TW IN L A SE R BEA M S
leads
to
d e a m p lif ic a tio n of
some q u a d r a t u r e out of
both
components of th e e l e c t r o m a g n e t i c
is
the
purpose of
theo retical
obtained ( 0P0)
on t h e
and
this
be
experim ental
results
N on-Degenerate O p tic a l
s h o wn
understood highly
that
features
In different
its
that
going
into
corresponding
give
proof (§
of
the
3 ) . The
representation properly
of th e f i e l d s
the no ise
in the g en eral experim ent
for
case.
In the
of
properties
tw in f i e l d s
are
a sim ple tre a tm e n t in a good p h y s i c a l
The s e c o n d o n e i s
is
in ten d ed to
model i n a f u l l y quantum based
upon
a classical
b u t i n w h ic h vacuum f l u c t u a t i o n s (§
4).
The l a s t the
fifth
our
r e d u c t i o n on t h e d i f f e r e n c e (§ 5 ) .
em its
This
"classical"
approach
s p e c tr a of the v a rio u s m easurable q u a n t i t i e s
and p r e s e n t
applications
one
crystal
be
w ill presen t th ree
one i s
corpuscular
th ird
accounted
we
p h o to n s which p r o v i d e s (§ 2 ) .
can
it
paper.
first
th e problem
have
) . In p a r t i c u l a r
The c o r r e l a t i o n
in th is
we
Param etric O s c illa to r
the n o n -lin e a r
of the
The
that
characteristics
photons.
studied
"corpuscular"
t w i n beams
of
the t h e o r e t i c a l s e c tio n ,
in sig ht
y ields
tw in
approaches.
of
a
fact
the i n t e n s i t i e s
the b a sic
terms
most
from t h e
correlated
between
manner
fields
on
paper to g iv e a review of
(above t h r e s h o l d o p e r a t i o n w ith
w ill
fluctuations
the c a v ity .
It
are
t h e vacuum
we w i l l d i s c u s s
observation
between th e
of
quantum
in ten sities
the
noise
of the
t wo
s e c t i o n w i l l d e a l w i t h new p o s s i b l e
tw in photons
g e n e r a tio n of am plitude
section,
t o s p e c t r o s c o p y and t o
s q u e e z e d l a s e r beams
(§
6
).
the
et al
E G IA C O B I N O
2.
CORPUSCULAR MODEL
We in
83
first
present
which p h o to n s
a "corpuscular"
are considered
m o d e l(R e y n a u d 1987)
as c l a s s i c a l
particles.
We
c o n s i d e r o n ly th e ca se of n o n - d e g e n e r a te e m is s io n in which th e two
ty p es of
s ig n a l photons
th eirfreq u en cies
can be
(o^ ^ coz ) o r
d i s t i n g u i s h e d by e i t h e r
th eir polarizations.
s hown
both ex p e rim e n ta lly
( Bu r nh a m e t a l . 1 9 7 0 ,
19 8 5)
and
( Mo ll ow
Gr a ha m
theoretically
1984,
conversion photons
Hong
et
al.
et
1985)
the n o n -lin e a r c r y s t a l
a l . 1967,
that
in
em its p a ir s
1 e t 2 e a c h t i m e a pump p h o t o n i s
It
has been
F riberg
et al
Mollow 1973,
p a r a m e t r i c down of sim u lta n e o u s
annihilated.
Such a
t w i n p h o t o n g e n e r a t o r can be u s e d as t h e pumping m echanism f o r an
op tical
cavity
resonant at
the s ig n a l
a>2 . Above some pump p o w e r t h r e s h o l d ,
the
yields
i.e .
" t w i n p h o t o n " b e a m s Bi
having
highly
correlated
the
cavity
d ecorrelation delivered equal
at
in the
storage
inten sities
In
A ctually,
the c a v ity
b e a ms
since
the
go o u t
induces
some
: th e numbers of p h o to n s
other
a tim e
lo n g compared to th e
words,
the d iffe re n c e
must r e v e a l
and
b e am s a r e e x p e c t e d t o b e n e a r l y
counted d u rin g
S i (co) o f
spectrum
tim e,
t w i n b e am s
t wo o u t p u t
tim e.
system o s c i l l a t e s
c r y s t a l do n o t n e c e s s a r i l y
s ame
between th e
o n l y when
cavity
the
and
t wo l a s e r - l i k e
in ten sities.
tw in-photons produced in the of
a n d B2
frequencies
m easuring the n o ise
I between
t h e s i g n a l beam
a re d u c tio n of photon n o is e
inside
the
c a v ity bandw ith.
The prediction we
w ill
photons em ission
purpose of t h i s fo r the noise
assume are is
that
randomly denoted
the
section
to give a q u a n t i t a t i v e
s p e c t r u m Sj (co) . I n t h e p r e s e n t m o d e l e m issio n s of
distrib u ted . A.
is
In
T he
addition
the v ario u s mean
pairs
rate
each c o l l i s i o n
of
of p a ir o f any
84
Q U A N T U M N O ISE R E D U C T IO N ON TW IN L A SE R B EA M S
signal
photon
probability
w ith
R for
transm itted
the
coupling
m irror
being r e f l e c t e d ,
is
d e s c r i b e d by a
a probability
T fo r being
( w i t h R + T = 1) .
In
this
corresponding
framework,
to
we c a n e v a l u a t e
the d e te c tio n
the
spectrum
scheme s k e t c h e d i n
fig.
1
(to) :
S, ( 1 At w ith I(w)
and I ( t )
F igure
i At
=1
I(t)
= I 1 (t)
is
in serted
dt
(2)
(3)
non-degenerate
constituted
in
separated
exp(-iw t)
- I 2 (t)
1 : The
(NDOPO)
(1)
an
by
optical
optical a
tw in-photon
ca vity.
and
their
in ten sities
photodetectors
(PD) .
T he
n o is e spectrum Si
In instantaneous
photodetectors in teg ratio n ch aracteristic
beam
; I (to)
tim e
(TPG)
t w i n s i g n a l beams a r e and I 2
Ia
spectrum
expressions,
signal
tim e
The
generator
analyzer
a re measured by (SA) g i v e s
o f the i n t e n s i t y d i f f e r e n c e I = ^ - I
(to)
these
param etric o s c illa to r
At ;
is
I i (t)
in ten sities the F o u rie r
is
much
t h e symbol
and
I 2 (t)
are
longer < >
than
any
2
.
the
m o n i t o r e d by t h e
tra n sfo rm of I ( t )
the
two
; the other
in e q . (1 ) d e s c r ib e s
a
E G IA C O B I N O
"classical"
mean v a l u e
A
p articular
random v a r i a b l e s and
k2
and 2. if
event
thequantum
where
8
(t
is
8
detection the
-
tim e
sim plicity,
(
t
the
roundtrip
signal
on t h e
0
1
t
)
)
D irac
( no tim e
-
8
the
we s u p p o s e t h a t
|I(co )|2is
=
2
contributions the
reflectio n s
(t
1
and t h e n
1
t
to
:
t o be
+ k
;
t
2
t
)
(and
),
1
the
:
(4)
)
the
first
possible
the coupling m i r r o r ) , x is
and
)
inside are
t
of
this
the
the
cavity
sa me f o r
event to
) - e x p ( - i c o ( t Q+ k 2
(for
the
t wo
the F o u rier
t )
)
| 2
(5)
- k 2 ) c o t ) ]J
w eighted and
S i (co)
(Eq.l)
by t h e undergo
2
tim e
b y su mmi ng t h e s e
p r o b a b i l i t y TR respectively
m u l t i p l y i n g by
k
TR
1
k
k9 2
and kg
t h e n u m b e r A At o f p a i r :
00
S (co) = 2 A
The
(t 0
photons
em issions d uring the i n t e g r a t i o n oo
-
on
the spectrum
(Eq.5)
photons
assumed
1
thus
[u l - c o s ( ( k
We now o b t a i n
th e numbers k
event to I{ t) is
contribution
| X (co) | 2 = | e x p (-i co( t Q+ k i t
that
are
function
of
follow ing
measured as d e t e c t i o n r a t e s
reflection
m o d e s ) . The
transform
are
p articu lar
+ k
events.
c o u p lin g m i r r o r of th e p h o to n s
efficiencies
c o n t r i b u t i o n of t h i s
=
can be d e s c r i b e d by t h e
: th e em issio n tim e of a p a i r ,
the i n t e n s i t i e s
I(t)
85
taken over the v a rio u s p o s s ib le
of r e f l e c t i o n s If
et al
summation of
(TR
1
the
) (TR
2
) [1 *• - c o s ( k 1 - 2k
series
)
co t
(6 )
]
appearing in Eq.
6
leads
86
Q U A N T U M N O ISE R E D U C T IO N ON TW IN L A SE R B EA M S
S, (co) = So
-
1
( 7)
T2 + 4 R s i n 2
(?))
w i t h S o = 2A It com pletely to
(8)
clearly
suppressed at
the q u a l i t a t i v e
photon
noise
m ultiples
of
is
peaks,
Sj (co)
corresponding = = 2A)
2 : P r e d i c t e d n o i s e s p e c t r u m S^ transm ission
(co)
: T = 5 0 % (a ) ,
fo r
various values
of
the m irror
1 0 % ( b ) , 2% ( c ) . S 0
is
t h e u s u a l p h o t o n n o i s e . Th e f r e q u e n c y u n i t Q i s
the c a v i t y
free spectral range.
For a h ig h f i n e s s e
cavity
(T = 1
w ill
(co)
is
equations as
(co)
e q . (21).
(co) pi variance :
fluctuation
The d i f f e r e n t i a l
algebraic
ryj
w hile qt
the
:
( 29)
be
directly
< | p ° ut (co)
| 2
g i v e n by t h e o u t p u t
>
( 30)
Pi 4.3
F luctuation
We which
the
(r
r
1
=
spectra
w ill
first
lo sse s are = r , u* =
'l
2
in the balanced case
give
the
u,
s ame f o r
=
2
u,,
the
1
be
pump f l u c t u a t i o n s
fluctuations am plitude
r ’=
^
I t can im m ediatly that
the r e s u l t s
-
8
fluctuations
:
6
a1
a2
T
the 2
in the
t wo s i g n a l
= r
sim ple case
in
fields
).
seen
on E q s .
w ill
not
(27-a)
and(27-b)
bec o u p l e d i n t o
. Thus f o r t h e d i f f e r e n c e
of
the the
1
r = —
(Pt
- P
2
)
( 31 )
\ | 2
(w ith get
sim ilar notations
the
t
for
the
i n p u t a n d o u t p u t f i e l d s ) , we
follow ing equation:
r +
2
t ’ r = \f 2 r r
1 0
+ \[ 2 ^ r ’ ln
( 32 )
97
et al
e g ia c o b in o
r\i
We t h u s o b t a i n
r
for
th e F o u r i e r component r
(co) = -------------2 r ’ + icor
The o u t g o i n g f i e l d
r '"
(cflrr
r o ut
is
easily
( 33)
and
d e r i v e d from
^
(34)
to
1
,
the
s i m p l y g i v e n by t h e v a r i a n c e
(co) = V°r ui
In
(10)
( 33)
:
equal
difference
Sj
:
of
the
^>ut
r
2T'+ico'r
(co)
( 35)
i n p u t vacuum f l u c t u a t i o n s
squeezing
on t h e r c o m p o n e n t i s
of
4 wT' + co2 t 2 (co) = ------------------4 T i 2 + co2 t 2
the p re s e n t case,
are
to
2 v p r>
(co) +
the v a ria n c e s
norm alized
leads
ut
(co) = ------------------- r 2T'+ico'r Since
be a ms
’ 1 " (co)))
( 34)
2 T - 2 T 1- i o x r
:°u1
Sr
1" (co) + r
:
= \[2 t r - r 1"
H e n ce c o m b i n i n g Eqs
are
(co) o f r
and
(co) i s
the
( 36)
t h e mean i n t e n s i t i e s noise
proportional
spectrum to
(co)
on t h e
on t h e
SQ i s
equal
to
absence resu lt can
the
shot
t h e sum o f t h e of
as in
internal
be s e e n t h a t
(37)
n o i s e o n a beam w i t h a t o t a l in ten sities losses
the p rev io u s in ternal
in ten sity
:
4 u , r ' + co2 t 2 S J (co) = Sq ---------------------4 t ’ 2 + co2 t 2 where
tw in
of th e
(p/=0 , T ' = T ) , section
losses
tw in beams. one
(recalling
decrease
gets
inten sity In the t h e same
th at T=2r). I t
t h e amount o f n o i s e
98
Q U A N T U M N OISE R E D U C T IO N ON TW IN L A SE R B EA M S
r e d u c t i o n a t co= 0
by a f a c t o r
:
V* f* — = -----T*
( 38)
p,+ T
Note
that
this
factor
is
e m i t t e d p h o t o n s w hich do n o t r e a c h for
example
scattered
m irror d if f e r e n t
by
the
^ have
losses
through the o u tp u t m irro r.
in Fig. 4 fo r v ario u s v alues
the
shot noise
the
cavity
losses
p r o p o r t i o n of
or em itted
from t h e o u t p u t o n e ) . T h is
a s s o c ia te d w ith
4 :N o i s e
the
the p h o to d e te c to r crystal
losses
Figure
ju st
t o be
im plies
(they are through a that
k e p t much s m a l l e r
The n o i s e
spectrum i s
extra than s ho wn
o f y*.
spectrum o f
the i n t e n s i t y d i f f e r e n c e . S i s 2 r°' f o r a b e am o f i n t e n s i t y Ii + 2I , coc = -----is T
bandw ith.
(fj, = 0) , c u r v e
Curve (b)
(a)
to
corresponds
equal
to no i n t e r n a l
transm ission
and l o s s e s
(p = r) , w h i l e t h e t o t a l l o s s e s r' a r e t h e s a m e i n a) a n d b ) . T he f l u c t u a t i o n between S etting
the :
tw in
spectrum
fields
can
of
the
phase
be c a l c u l a t e d
in
difference t h e sa me w a y .
E G IA C O B IN O
s = —
et al
99
( qx - q 2 )
(39)
vl2 one f i n a l l y
S
(co)
gets
= 1 +
Figure t wo
: 4r r ' ( 40)
Noise spectrum
signal
field s.
Th e z e r o - f r e q u e n c y l i m i t
Schawlow-Townes l i m i t
This
the
fluctu ation
of a
to
the
when co t e n d s
phase d if f u s io n
process.
t o 0,
which i s
One c a n d e d u c e
s p e c t r u m SQ (co) o f t h e b e a t f r e q u e n c y b e t w e e n
two s i g n a l mo de s
fi = co
correspond
the
a s s o c ia te d w ith phase d i f f u s i o n .
expression diverges
characteristic the
o f the b e a t fr e q u e n c y between
-co2 =
\J"2 a
:
100
Q U A N T U M N O ISE R E D U C T IO N ON TW IN L A SE R B EA M S
= > Ti
situ atio n .
transm ission
to a n a l y t i c a l l y
components of
the
solve
am plitude
(co) .
here
(co) but ra th e r Hi restrict ourselves
spectrum
unequal
possible
rvy
(co) a n d S w ill
of
is
the
f o c u s on t h e i r p h y s i c a l to
the d is c u s s io n
difference much s m a l l e r
, Tz ) w h i c h
lengthy expressions
Sx (co) i n than the
corresponds
for
content.
of the n o ise
th e c a s e where signal
cavity
to our experim ental
104
Q U A N T U M NOISE R E D U C T IO N ON TW IN LA SER BEA M S
Figure
8 :
N oise spectrum o f
non-halanced
case.
coefficients
and
(T
= 0, 55 %
twin
b ea m
, T2
For
= 0,45%,
in ten sity
shot noise
pump
in
displayed
losses
transm ission p^
mean
a m p l i t u d e p a r a m e t e r o' i s a
the
the i n t e n s i t y d i f f e r e n c e curves,
are
slig h tly
= 0,48%, values
equal
= 0 ,5 2% )
d iffer
to 1 .1 .
beam.
Curve b)
30dB e x t r a
noise
at
e x p o n e n t i a l l y to ze ro
by
C u r v e a)
l i m i t e d p ump
beam w i t h
the
the
d ifferen t so
15%.
that
the
T h e pump
corresponds
corresponds
to
to the
zero frequency decaying
a t high fre q u e n c y w ith a h a l f width o f
0 . 1 5 coc . T he m a i n c o n s e q u e n c e o f t h e u n b a l a n c e in ten sity in
fluctuations
an i n c r e a s e d n o i s e
We d ifferen t only low
from u2
between
for
t h e mean
t h e pump
case
i n w h i c h p^
a n d r 2 . T h e n we f i n d
frequency p a rt and
now c o u p l e d i n t o
close of th e loss
to
is
that
th resh o ld
spectrum .
t h e pump
Sx (co) , r e s u l t i n g
of
shot noise
a n d 7^
the
(cr&l)
giving
and i n gives
is
the very S i (co) f o r
15% d i f f e r e n c e
t wo b e a m s ,
lim ited
are not very
th e added n o is e
F i g . ( 8 -a)
coefficients
in ten sities
field
that
a t a n y f r e q u e n c y on t h e w h o l e s p e c t r u m .
the
im portant very
cr = 1 . 1 ,
that
consider
are
is
and a s s u m in g
a t any f r e q u e n c y
:
E G IA C O B IN O
one
notices
frequencies. l ow
m oderate
Fig.
(8-b)
frequency,
situation
:
co # 0 .
Of
course,
the
are
may a p p e a r .
increased
into
of
observes if
the
a
the
noise
to
a
sharp r i s e
low
more in
But t h e
frequency
pump
breaks
down,
general
conclusion
rem ains c e r t a i n l y
at
realistic for
in ten sity
approach fo r
and m u l t i p l i c a t i v e that
b e c a u s e o f u n b a l a n c e o n l y i n a. l i m i t e d
frequencies
l ow
the n o ise
our sim ple l i n e a r i z e d
dynamics
at
a c c o u n t e x t r a pump n o i s e
corresponds
too la r g e ,
fluctuations
noise
increase
takes
which
one t h e n
fluctuations
5.
a
105
et al
(co) i s
range of
l ow
valid.
EXPERIMENT
F igure
9 :
acousto-optic beam -splitter
Experim ental s e t-u p . m odulator
;
F
FR : F a r a d a y r o t a t o r :
filter
;
; AO :
PBS : p o l a r i z i n g
; SA : s p e c t r u m a n a l y z e r .
We u s e a t w o - m o d e p a r a m e t r i c
o scillato r
o p e r a t e d above
106
Q U A N T U M NOISE R E D U C T IO N ON TW IN L A SER BEA M S
th resh o ld al.
to g e n e ra te
1987).
The
aim
high i n t e n s i t y of
fluctuation
s p e c t r u m on
tw in
beams.
We w i l l
shot
noise
the the
experim ent in ten sity
show t h a t
lim it
tw i n beams is
(Heidmann e t
to
study
difference
the
between the
s i g n i f i c a n t r e d u c t i o n below th e
is observed
over a
broad
range
of
frequencies.
Figure pu mp e d an
by a
shows t h e
single
external
50kHz).
(9)
mode Ar*
Fabry-Perot
(FR)
(AO)
are used
to o p t i c a l l y
back
reflected
light
then
focused in to
type
II
signal
of c u r v a tu r e ,
a large
finesse the
17mm.
is
is
the
flat
high
and t r a n s m i t s
signal
The f o u r phase
which i s
reflectin g
transm itting
fo r the
pump f i e l d .
from t h e
conditions
fo r both s ig n a l
inserted
discrete
series
of
for
cavity
f o r t h e pump b e a m .
0. 8% o f
the
and i d l e r f i e l d s
condition,
the c a v ity and low f o r
cavity
values.
(energy
resonance
In
in terv als
nanom eters)
around
the
well-known 1973).
high s e n s i t i v i t y
Thus t h e
electronic output m onitored
OPO l e n g t h
feedback
intensity. on t h e
of
the
This
is
so
that
it
delivers
For
this
purpose,
we ak c o u n t e r p r o p a g a t i n g
(Smith
stabilized
a nearly
the
(a f e w
cause of the
OPO t o v i b r a t i o n s
h a s t o be a c t i v e l y
for a
practice,
occurs only in very sm all le n g th values.
The
in frared
are only f u l f i l l e d
length
an
infrared
oscillatio n
these
in
w i t h a 2cm
the
C onsequently,
and i d l e r )
is
a 7mm l o n g ,
OPO o s c i l l a t i o n c o n d i t i o n s
matching
strong
t h e OPO. The pump l i g h t
p a r t of the green l i g h t .
conservation,
laser
The i n p u t m i r r o r ,
highly
a n d i d l e r be a ms a n d
output m irror and
isolate
4
is
of
and an a c o u s t o - o p t i c m o d u l a t o r
KTP ( KTiOPO ) c r y s t a l ,
length
The OPO i s
(re sid u a l frequency j i t t e r
coming from
phase-m atched
set-up.
a t 528nm s t a b i l i z e d on
t h e p a r a m e t r i c me di um,
c a v ity of
radius
ion l a s e r
cavity
A Faraday r o t a t o r
optical
experim ental
OPO
by
constant
output
is
i n f r a r e d b eam
(I R)
lig h t
E G IA C O BIN O
which
is
transm itted
deflected stable
by t h e
reference
stabilized
on
domains,
back through
level.
the
As a
side
corresponding
experim ental
conditions,
of
the
a slight
we
in tensities
side
lig h t).
The
(PBS)
which
quantum
regim e.
have
m illiw atts
are
efficiencies
coated.
two
photocurrents
The
subtracted
using
a 180*
difference
is
checked
beam splitter
are
are
w ithin
matched
two at
corresponding analyzer.
200mW o f g r e e n
by
less
of
90%.
b e a ms
are
\xm
and
s t o p p e d by
a polarizing
of
: the
by
a
an tireflectio n and
The n o i s e
then on t h e
spectrum a n a ly z e r
the
detection
im perfections
The
surfaces
analysis.
than 1 % ; the 1 %.
A ll
am plified,
com biner.
m onitored
characteristics
carefully
beam
(80mW
t w i n be a ms
= 1.048
A
are
power
co n n ected to a computer f o r d a ta
pump
interv al,
a n d t h e n f o c u s e d on t wo InGaAs p h o t o d i o d e s
t h e t wo i n f r a r e d
the
for
separated
by
between
behaviour
Above t h r e s h o l d
two c r o s s - p o l a r i z e d
encountered
been
the o s c i l l a t i o n
t r a n s m i t t e d pump beam i s
t w i n b e am s
beam splitter
The
some
a r e d e t e r m i n e d by t h e c o l l i n e a r p h a s e m a t c h i n g
The
resulting
f r e q u e n c i e s . In
w avelengths,
The r e m a i n i n g
filter.
m i s m a t c h b e t w e e n t h e OPO
of
f ew
em ission
X2 = 1 . 0 6 7 p,m, conditions.
a
is
b u t one c an a lw a y s c h o o s e f o r
a b istable
of
t h e OPO l e n g t h
can o b s e rv e a b i s t a b l e
g r e e n l i g h t ) , t h e OPO e m i t s
(with
and n o t
such sm all o s c i l l a t i o n
eigen
length,
right
does not y i e l d
of
cavity
a fu n c tio n of c a v ity
which
a
the
one
m irror
and com pared t o a
consequence,
of
to
and
stabilization
the in p u t
a c o u s to -o p tic m odulator,
frequencies
as
107
et al
of
c h a n n e ls have the p o la riz in g
a m p lifie r voltage gains
overall
common mode r e j e c t i o n
c h a n n e l s h a s b e e n m e a s u re d by m o d u l a t i n g t h e a
frequency
coherent
The r e s u l t
of
of
peak this
10
MHz
reduction
measurement i s
and on
m easuring the
25 d B .
the
spectrum
108
Q U A N T U M N O ISE R E D U C T IO N ON TW IN L A SE R BEA M S
A
key p o i n t f o r
is
the c a l i b r a t i o n
we
have used
the
Ei
rotation
and of
between fields
Ea
the
in
the of
Eg
o f s u c h an e x p e r i m e n t
level.
half-w ave p la te
E2 e m i t t e d
axes and
shot noise
beam splitter
29
the
of
a rotating
polarizing
fields
the r e l i a b i l i t y
(labeled by t h e
p late
are
test,
in f r o n t of
in F i g . 9).
The two
OPO u n d e r g o a p o l a r i z a t i o n where 0 i s
the angle
and o f t h e p o l a r i z e r .
respectively
t h e p o l a r i z i n g beam s p l i t t e r
in serted
A/2
half-w ave p la te ,
the
As a f i r s t
transm itted
The t wo
and r e f l e c t e d
by
:
EA = c o s 2 0 E 1 - s i n 2 0
E2
(50-a)
EB = s i n 2 0
E2
(50-b)
E1 + c o s 2 0
When 0 = 0 ° role
and th e
(modulo 4 5 ’ ),
measured s ig n a l
t w i n beam i n t e n s i t i e s . half-w ave p la te 50% beat
term s
observed gives
between
shot
noise
for
the
signal
Ia - I
(modulo 4 5 * ) ,
tw in
the
t wo
fie ld s is
between th e the
like
varies
an u s u a l the the
mo de s do n o t a p p e a r i n
the
the
measured s ig n a l
f o r a b eam o f i n t e n s i t y
that
system
a b o u t 5 THz,
C onsequently,
level
acts
p l a y s no
experim ental co n d itio n s,
the
One c a n show f r o m E q s . (50)
difference
beam splitter
in our
frequency range.
the
the
When 0 = 2 2 . 5 *
Since
fre q u e n c y between
crossed
is
and p o l a r i z i n g
b eam splitter.
the half-w ave p la te
Ii + I2.
t h e n o i s e p o w e r s p e c t r u m S$ (fi) sinusoidally
as a f u n c t i o n of
the angle 9 :
0 (Q)
= S 1 (Q)
F i g . 10 fixed
c o s 2 40 + S 0 s i n 2 40
shows
the
variation
(51)
of
f r e q u e n c y H / 2 tc = 8MHz. One o b s e r v e s
S$ (O)
recorded
at
a
a stro n g m odulation
E G IA C O B IN O
of
the no ise
noise
level
le v e l w ith the
at 0 ’ is
109
expected p e r i o d i c i t y
checked t h a t
frequencies levels
of
this
test,
(0=22.5*)
t h e YAG
higher the
of
22.5*.
to y ie ld
shot noise
l a s e r was
shot noise
level.
We
lim ited
at
YAG w i t h e q u a l me an i n t e n s i t i e s .
can
assert
that
coincides
w ithin
1%
(dashed l i n e
the
: the
t h a n 2MHz. We h a v e t h e n m e a s u r e d t h e n o i s e
OPO a n d
we
45*
we h a v e u s e d a cw YAG l a s e r
independent c h a r a c te r iz a tio n
have
of
a b o u t 30% l o w e r t h a n t h e o n e a t
As a s e c o n d t e s t , an
et al
From
the upper l e v e l of F ig. w ith
the
shot
noise
22
level
in F i g . 10).
^ )
(a)
A /V W
V
150 pA2/H z (b)
—i---------1---------- 1--------- 1----90° 18 0° Figure
1 0 :a ) V a r i a t i o n
in ten sity
in
terms
level
dashed
the
d i f f e r e n c e S$ (H)
(expressed noise
of
gives
as a f u n c t i o n
o f photodiode
equivalent
lin e
m e a s u r e d n o i s e p o w e r on t h e
to
current noise)
is
5 0 s, w ithout
F i g . 11 the
ratio
0=0*,
to
of
the sh o t n o is e l e v e l
the
input
at
The
t h e same f r e q u e n c y
Scan
t i m e f o r a)
and
vid eo filter.
gives the
; b)
th e whole e l e c t r o n i c n o i s e .
o f a YAG l a s e r h a v i n g t h e s a m e i n t e n s i t y . b)
o f 9 f o r Q/2n=8MHz
the noise re d u c tio n "squeezed"
shot noise
noise
spectrum ,
f a c t o r K{Q)
spectrum , 0 = 22.5*
which i s
recorded at (both s p e c tr a
110
Q U A N T U M N OISE R E D U C T IO N ON TW IN L A SER B EA M S
have
been
clearly
corrected
of
30% ± 5% i s
reduction
In because
electronics
n o i s e ) . The c u r v e i s
below 1 o v er a b ro a d fre q u e n c y r a n g e .
reduction noise
from
is
the
the
rejected
low
large
in
in ten sities
b etter
the
A maximum n o i s e
observed a t a frequency of t h a n 15% f r o m 3 t o
frequency
domain,
13MHz.
the
noise
ex tra noise
o n e a c h beam i s
difference
process.
are not
exactly
in
the
equal,
slight
difference
losses
This
is
a cause fo r a d d itio n a l
into
the measured s i g n a l ,
w h i c h we
in
the
attribute
mean to a
t wo i n f r a r e d b e a m s .
fluctuations
as d is c u s s e d
increases
not com pletely
Moreover,
fo r the
8MHz. The
t o be c o u p l e d b a c k
the
theoretical
§ 4.
1
1 0 M hz I
0 F i g u r e 11 It
is
(9 = 0)
20M hz I
: Experim ental n o ise r e d u c tio n
obtained S(Q),
by recording 3 spectra the
shot
e l e c t r o n i c n o i s e E(co), dashed l in e
At
is
R (f l)
noise
(9
= ( S( Q)
a Lorentzian f i t
high fre q u e n c ies ,
of
R(ft)
=
f a c t o r R(Q) : the
spectrum .
"s q u e e z e d " n o i s e
2 2 , 5 * C)
N(Q)
and th e
- E (co)) / (N (co) - E ( c o ) ) .
The
the exp erim en ta l p o in ts .
is
seen to
go t o 1 : t h e
E G IA C O B IN O
noise
of I i
higher
than
Lorentzian higher 50%,
- Ig rises the fit
than
in agreement
w ith
internal
for frequencies
The d a s h e d l i n e spectrum
The e x t r a p o l a t e d
from t h e e f f e c t o f t h e
6.
bandw idth.
the experim ental
5 MH z ).
which i s
111
back to the sh o t n o ise
cavity
of
et al
the
value value
(for
shows a
frequencies
a t co=0 i s
about
one would e x p e c t
losses.
APPLICATIONS OF TWIN BEAMS
The
OPO
provides
photons. noise
This
In
feature
in se v e ra l
enhance the this
high
sensitivity
of
beams
made o f c o r r e l a t e d
a p p e a r s v e ry p r o m is in g to re d u c e quantum
experim ental
paragraph,
generation
in ten sity
configurations,
of quantum n o i s e
we w i l l
lig h t
o utline
beams
below
and t h e r e f o r e
to
l i m i t e d m easurem ents.
t wo
such a p p l i c a t i o n s
the
shot noise
lim it
:
and
measurement of v ery sm a ll a b s o r p t i o n s .
6.1
R eduction of i n t e n s i t y
n o i s e b e l o w s h o t n o i s e on a s i n g l e
l i g h t beam
As Jakeman a l.
s u g g e s t e d by
et al.
1987),
1986,
on
correction.
Two
F i g . (12).
In
m onitored
by a
such
b e a ms
single possible
both
of
may
b ea m,
m odulator,
t h e s e c o n d one
is in
(Fig.
(S a le h e t a l . 1985,
Yuen 1 9 8 6 ,
be via
used
them,
the
intensity first
to c o r r e c t
( 1 2 -b )),
for
"intensity intensity
s c h e m e s a r e sh own i n
scheme
used to modify th e order
Yamamoto e t
electronic
experim ental
p h o to d io d e . In the
an i n f o r m a t i o n
inten sity In
a
authors
Ha us e t a l . 1 9 8 6 ,
correlated
squeezing"
several
o f beam Bz
is
( F i g . ( 1 2 - a ) ),
t r a n s m i s s i o n o f an t h e pump i n t e n s i t y .
t h e sa me i n f o r m a t i o n i s u s e d
112
Q U A N T U M N OISE R E D U C T IO N ON TW IN L A SE R B EA M S
d irectly
to c o rre c t
The m a i n d i f f e r e n c e loop
process
w ay ,
which i s Let
in
us
as
the
first case
now
way.
t w i n beam 1^
between th e
not the
quantitative in-phase
the
two s c h e m e s i s
one w hic h a c t s in
fluctuations
in
same t e c h n i q u e . that
th ere
is
a
a selfconsistent
the second one.
consider From
by t h e
the
E q s . ( 27)
first
technique
and(28),
(in F o u rie r space)
i n a more
we c a n w r i t e
in
the
the balanced case
:
F i g u r e 1 2 : Two p o s s i b l e s c h e m e s o f i n t e n s i t y inten sity
of
inten sity
b y a c t i n g on an i n t e n s i t y m o d u l a t o r
and i n b)
p ° ut
to c o r r e c t
used
in
a)
to c o r r e c t
t h e i n t e n s i t y o f beam
Bi
(IM)
rvj a p Jn is
(feedback)
feedforw ard) .
sources.
proportional
(52-a)
Let
associated
w ith
a n d it
The e l e c t r o n i c
t o p®u 1
intensity.
(52-b)
the p a rt of flu c tu a tio n s
a n d it
t the
: the
t h e p ump
+ a P J"
\
fluctuations noise
is
B2
= itt + a p J ”
P
where
beam
correction
th e re m a i n in g p a r t due t o o t h e r feedback loop produces
which i s be
the
c o m i n g f r o m t h e pump
a signal
u s e d t o r e a c t o n t h e pump b eam am plitude
am plitude
transm ission
m odulator.
The
pump
factor field
E G IA C O B IN O
am plitude
t r a n s m i t t e d by t h e m o d u l a t o r i s
ot* n = t a n + J^ l - t 2
where
is
because Due
of
to
113
et al
the
in ten sity
(53)
losses
fluctuating
G
field
i n t r o d u c e d by t h e
feedback loop,
fluctuations
t = F
:
ccn n
the n o ise
the
then
the
o f beam 1,
entering
the
system
i n t e n s i t y m odulator.
transm ission
t d e p e n d s on t h e
p°u1 , according
to
:
p j ut
oc 0 where and
t
is
G the
the am plitude effective
sake of s i m p l i c i t y ,
transm ission
loop
f a c to r w ithout feedback
gain fa c to r,
t o be r e a l
assumed h e r e ,
and f r e q u e n c y
for
the
independent.
T h en p °1u 0 + t Sot0 - G*1
The inserted
real
in
eq.
fluctuations
part (52)
^ i
as
the
oc1 otn n1
( 54)
fluctuations value
of
p n0
is
now i n s e r t e d
pump
fluctuations,
which
flu ctuations
in eq.
yields
(54)
the
on beam Bz :
must
t h e n be
t h e pump i n t e n s i t y
t h e OPO, w h i c h g i v e s rvj f o r p ®ut :
| 1i -—t 1 + a ( t p 0lnn + vJN
value
in-phase
such
tc
This
the
\\J^1 l -- tt*0
i m p i n g i n g on
s e l f - c o n s i s t e n t value
p out
of
+
the
follow ing
(55) 1+aG
to give
the e f f e c tiv e
follow ing expression
for
114
Q U A N T U M NOISE R E D U C T IO N ON TW IN LA SE R B EA M S
Ga ° 2
Tt
2 When
1+Ga
to
on
p i U1 " P 2 U1 ' the noise
13
electronic b ea m the
having
in
technique.
B 2 the
fs
:
the
open
intensity
this
loop
feedback c o rre c ted
L
o pe n 1o o p
noise
c o r r e c tio n . So/2
( 56)
expression
feedback loop
fluctuations
of
on
As a r e s u l t ,
in ten sity
co2 t 2 + 4 lct’ = So -------CO2 T 2 + 4 T ' 2
spectrum
is
the
difference.
(co)
In tensity
b ea m
Bz
is
:
( 58)
after
the sh o t n o is e o f a c l a s s i c a l
t h e s a m e mean i n t e n s i t y . Th e me an t r a n s m i s s i o n
in ten sity
losses
beam
the high gain l i m i t ,
s p e c t r u m on t h e
fe e d b a c k
1+Ga
I r\y p ‘ n + vjl-t 2 p " ) 0 0
( t
g ain goes to i n f i n i t y ,
in
on
[V"’]
+
:
As e x p e c t e d , tran sfers
1
th e loop
sim ply reduces
Figure
a
=- I t
the
m odulator OPO.
a)
is
80%
feedback
and
of
t h e r e a r e no i n t e r n a l
t e c h n i q u e , b)
feedforw ard
L et us s t r e s s previously I
(E q.(7)),
+ 1^ = 2 1 i
background the
when
. In
the shot n o ise
corresponds
form ula
reached for
standard
possible
that
for
the
is
due
to
the
to
fact
that,
to
correct
feedback loop a c t u a l l y
u s now c o n s i d e r
c o rre c tin g device.
whe n for
In t h i s
case,
the
the
as l a r g e
as
m onitors
the
fluctuatio ns if
beam
Bz
.
The
transm itted
field
d e v i c e h a s an e x p r e s s i o n s i m i l a r
♦J:1 - t 2 Here th e e x t r a n o i s e
of
t h e quantum
2
in-phase
a m p l i t u d e a*
t o Eq.
term a“ i s
inserted
the unescapable
through t h i s
of
c o u n t e r p a r t of :
a n2
(60)
and f r e q u e n c y i n d e p e n d e n t , beam
B2
after
one f i n d s
that
the v a r ia b le
:
~in
+ Bpo
on
(53):
a c t i n g on beam B2 . T h e n
a r e g i v e n by - Git
on
( 59)
fluctuations
_
flu ctuations
transm itter
+ t a 2 - Gp°1 u 1 + J l - E
again G r e a l
"feed forward"
a"2
transm ission
a 21 = t a
the
in ten sity
a r e u s e d t o a c t on a v a r i a b l e
P2 = t i t
one
th e problem of
Bi
transm itter
( 1 3 - a ) , the
l a r g3---------------------e r t h a n 50%.
adds n o is e
beam
the
tw ice
inten sity
are u n c o rre la ted .
Let
Taking
in Fig.
on I 1 - I 2 i s
beam
a variable
of
i n t e n s i t yJ I i . I t i s t h u s n o i s e on o n e b e am , b u t o n l y
o f beam
fluctuations
beam
therefore
fluctuations Bz , t h e
a
Sq , d e f i n e d a s
a beam o f
factor
Bi
level
displayed
intensity
the noise re d u c tio n
This
(58)
co -*■ oo i s
shot noise
to reduce
115
et al
E G IA C O B IN O
I
( t - G) +
2
1-t
a;
(61)
116
Q U A N T U M NOISE R E D U C T IO N ON TW IN L A SE R B EA M S
In o rd e r Ii
- I2,
like
precise the is
to in
value of
noise
transfer the
previous
the e f f e c t i v e
spectrum of the
g i v e n by
on beam Bz
This
final
to
T2 + 4T' u. S0 ----------------------- + — CO2 T 2 + 4T ' 2
in
the
may
notice
precisely the the
1^ ,
the
in Fig.
obtained
t wo c a s e s ,
50% a t
n o i s e on
choose a
intensity
I2
least
for
the that
required
( 62)
c a n be
feedback
an open lo o p reduce
the
the reaso n given p re v io u s ly .
But
follow ing
is
(13-b),
in
we s e e
_ (1-t2)
differences
to
between th e
t wo
:
(i)
to
:
displayed
form ula
r e d u c tio n of
situations
in
form ula,
the
configuration
one
on
i n which c a se
t o2
t
0
fe e d f o r w a r d
in ten sity
need to
feedfordw ard c o rre c te d
= S
2
noise
we
existing
:
(co)
compared
case,
loop g a in G = t ,
2
SJ
the n o ise
The g a i n fixed
first
in
one.
electronic (ii)
attenuator
value necessary
the
the
factor
close
the n o ise
is
and n e e d s o n l y t o be h i g h
T h i s makes t h e s e c o n d s e t - u p v e r y s e n s i t i v e gain f lu c tu a tio n s .
the noise does not
low ers
second case,
to reduce
entering
m atter
squeezing f a c to r to u n ity
is
in in
the the
the
system v ia
feedback case,
second c ase.
therefore
the v a r ia b le
required
in
whereas
it
A transm ission the
feedforw ard
case. (iii) correcting the
system .
one t a k e s
system
feedback
o scillation,
If
(neglected
system which
into
is
may never
account the in
this
very the
tim e d e la y s
first
easily case
in
the
sim ple a p p ro a c h ), lead
in
the
to
spurious
feedforw ard
117
et al
E G IA C O B IN O
6 . 2 M e a s u r e m e n t o f w ea k a b s o r p t i o n b e y o n d t h e
Intensity reduce
the
principle al.
noise
1 98 6 )
:
an
e(co)
t h e beam
varied
floor
b e a ms
in
absorbing
resonant at
.
are
medi um
a r o u n d cog , a n d t h e
in ten sity
difference
m easurem ents.
having
an
to The
(Fabre e t absorption
a g i v e n f r e q u e n c y coa
( by t e m p e r a t u r e o r a n g l e
lim it
suited
extrem ely sim ple
The p h a s e - m a t c h i n g
co^ v a r i e s
ideally
absorption
of s u c h an e x p e r i m e n t i s
co efficient on
correlated
shot noise
is
in serted
c o n d i t i o n o f t h e OPO i s
tuning)
so t h a t
the
frequency
a b s o r p t i o n d i p i s m e a s u r e d on t h e
1^- I 2 as a f u n c t i o n of fre q u e n c y .
As t h e
b a c k g r o u n d n o i s e p o w e r on s u c h a s i g n a l i s r e d u c e d by a f a c t o r P* — , w ith re s p e c t to s h o t n o i s e , t h i s te c h n iq u e a llo w s us to r ' measure a minimum a b s o r p t i o n c o e f f i c i e n t e (co^ ) s m a l l e r b y a factor
— .
This
method
w ill
be
useful
for
example
m
in ten sity
is
T'
measurements im possible
where
increasing
(biological
samples
But t h e s i m p l e the
s ame k i n d ,
w ill
measured
are in
a
in stan ce).
above,
as o t h e r s
a lw a y s be h a m p ered by t h e u n a v o i d a b l e
then
of low
w hich w i l l p r e s u m a b l y p r e v e n t any h i g h
measurement
techniques
for
laser
technique d e sc rib e d
frequency excess n o ise , sensitivity
the
at
needed,
frequency
zero which
range
frequency. tran sfer
where
M odulation
the s ig n a l
the n o ise
is
t o be
minimum
(Gehrtz e t a l . 1985).
However,
phase
or
f r e q u e n c y co a d d s q u a n t u m n o i s e be
sim ply
techniques
understood
by
am plitude (Yurke a t recalling
m odulation
a
a l . 1987a)
: this
that
m odulation
any
c o u p l e s mode s h a v i n g f r e q u e n c y d i f f e r e n c e s
co. Vacuum f l u c t u a t i o n s
at
equal
can
to
o f e m p t y mode s d i s t a n t b y co f r o m f i l l e d
118
Q U A N T U M NOISE R E D U C T IO N ON TW IN LA SE R BEA M S
m o de s a r e
th en coupled backi n t o
in crease
of
tech n iq u es reduce
fluctuations. such as
the n o ise
squeezed s t a t e s
In
in ten sity
therefore
F igure
One
m ultiple
m odulation
absorption
experim ents
losses
M odulation
signal
It w ithout not
m odulator get _
is
the
to
Yurke e t a l . 1 9 8 7 b ).
as w e ll,
h a v i n g a non z e r o
the m odulator mean v a l u e ,
and
amount o f quantum n o i s e .
scheme
nevertheless
allow ing
itself
but
pulsation, :
let
possible
: this
means su c h
possibility
elaborate
to
transfer
the
(PM) a n d a p o l a r i z i n g
( PB S) .
t h e beam
population
more
t o an
a t g i v e n f r e q u e n c y co w i t h o u t a d d i n g q u a n t u m
adding n o ise
different
rise
squeezing in o rder
n o i s e , by using a p o la r iz a tio n m odulator beam -splitter
giving
needs
frequency
adds a s i g n i f i c a n t
:
then
(G ea-B an aclo ch e 1987,
periodic
14
system ,
in phase-m odulation experim ents w ith the h elp
of
intro d u ces
the
is
the
to m odulate
absorption c o e ffic ie n t o scillating
Stark
etc..
We p r e s e n t
in Fig.
and p o l a r i z i n g follow ing values
insert
on
beam
beam -splitter for
B
I i going out of th e d e v ic e
:
signal
done f o r exam p le by m o d u l a t i n g
as an
us
the
the
Bi
PBS.
sh ift
a
(14)
e (co) , by or le v e l another
polarization
I n s u c h a w ay ,
t wo mean i n t e n s i t i e s
we
I * and
E G IA C O B IN O
et al
119
f I- A = c os^c ot I < 1 1
(63)
I = s i n 2 cot I Vi i Of it
course,
couples
mode
such
a sy stem adds quantum n o i s e ,
the o u tp u t to
having
t h e v a cu u m f l u c t u a t i o n s
a polarization
same f r e q u e n c y .
More p r e c i s e l y
orthogonal
on t h e
t h e beam Bi
the o u tp u t f i e l d s
because input
and th e
ocA a n d otA a r e
g i v e n by :
a A = coscot oc 1
(
ocB = - s i n c o t a i
then
+ s i n c o t ft 1 1 + coscot ft
i
(64)
i
: I A = | a A I2 = c o s 2 cot I + s i n 2 c o t t I 1 1 1 Nl i ^IB = | a B | 2 = s i n 2 cot 1^ - s i n 2 c o t JIT
An one
a b s o r b i n g medium
m onitors
When
there
m odulation
the is
in
no
absorption
c a n c e l s when
the
the
(65)
on t h e the
on beam BA a n d
signal
e x tra noise the
w ithout
m odulation
a d d e d by t h e
sum o f i n t e n s i t i e s
IA i s t h e s ame a s 1 + 1I B 2- I m odulation : s i n c e no p h o t o n i s
process, the
signal
no
noise
is
added.
to
the
At
:
c o s 2 cot e (co )
2co e q u a l
.
IA + IB -
n o i s e on
(66)
hav in g a F o u r i e r component choose
i
ft^
inserted
one m e a s u r e s
f r e q u e n c y co , o n e m e a s u r e s
S = T
then
a b so rp tio n dip
IA 1 +1 I B . T h e r e f o r e the n o i s e on I -I lost
is
ft
(I l /2)
e(co a ) a t
f r e q u e n c y 2co.
f r e q u e n c y o f maximum n o i s e
I f we
reduction,
120 we
Q U A N T U M NOISE R E D U C T IO N ON TW IN L A SE R B EA M S
can
take
full
advantage
of
such a r e d u c tio n
to o b ta in
a b s o r p t i o n measurement beyond s h o t n o i s e . ACKNOWLEDGEMENT This
work
has
been
done
with
the
support
of
EEC
Stimulation
Action Grant Number ST2J0278C.
REFERENCES
Bjork,
G.
a n d Ya ma moto,
Burnham,
D . C.
and
Y.
1988 P h y s .
Weinberg,
D.L.
A 31_ 1 2 5 .
Rev.
1 97 0 P h y s .
Rev.
L ett.
25
84. C ollett, Fabre,
M.J.
C.,
and G a r d i n e r ,
G iacobino,
SPI E 70 ECOOSA186 Friberg,
S .,
Hong,
E.,
C.W.
Reynaud,
(Florence) C.K.
1 98 4 P h y s . S.,
Rev.
A 30. 1 3 8 6 .
D e b u i s s c h e r t , 1 98 6 T.
489.
and Mandel,
L.
1985 P h y s .
Rev.
L ett.
54 2 0 1 1 . G ardiner,
C.W.
1 98 3 " H a n d b o o k o f S t o c h a s t i c M e t h o d s "
(Springer V e r la g ) . Gea-Banacloche,
J.
a n d L e u c h s , G.
1987 J .
Opt.
S o c . Am. B 4
and W h i t t a k e r ,
E.A.
1985 J .
166 7 . G ehrtz, Soc.
M ., B jorklund,
R.
V erlag, Graham,
C.
Jakeman, Lane,
vol.
1 984 P h y s .
A.,
Rev.
G.
and Mandel, E.
A .S .,
L.
and R a r i t y , Reid,
be p u b l i s h e d .
Y.
Horowicz,
a n d Camy,
i n Modern P h y s i c s
(Springer
66.
a n d Ya ma mot o,
Heidmann, C.
1 97 3 S p r i n g e r T r a c t s
B erlin)
R.
H a u s , H.
Hon g,
Opt.
Am. B 2 1 5 1 0 .
Graham,
Fabre,
G.C.
M.D.
L ett.
52, 1 1 7 .
198 6 P h y s . R .J.,
Reynaud,
198 7 P h y s . 1985 P h y s. J.G .
Rev.
Rev. Rev.
A 3_4 2 7 0 . S.,
L ett.
G iacobino, 59 2 5 5 5 .
A 31_ 2 4 0 9 .
1 98 6 O p t i c s Comm.
and W a lls,
D.F.
E .,
59^ 2 1 9 .
198 8 P h y s .
Rev.
A to
L ouisell,
W.H.,
Y ariv,
A.
121
et al
E G IA C O B IN O
and Siegm ann,
A. E . 1961 P h y s .
Rev.
124 1 6 4 6 . Lugiato,
L . A. and S t r i n i ,
M ilburn,
G.
Mollow,
B.R.
and W a l l s ,
G.
D.F.
1 98 2 O p t i c s Comm. 198 1 O p t i c s Comm.
and G la u b e r ,
R .J.
1967 P h y s .
4JL 6 7 . 39^ 4 0 1 .
Rev.
160 1 07 6 a n d
1097. M o ll o w B . R .
1 973 P h y s .
Reynaud,
1987 E u r o p h y s i c s L e t t .
S.
Reynaud,
S.,
Fabre,
Rev.
C.
A 8^ 2 6 8 4 . 4 427.
and G i a c o b i n o ,
E.
1 987 J .
Opt.
Soc.
Am. B 4 1 5 2 0 . Saleh,
B.
and T e i c h ,
Sargent,
M.,
M. 1 98 5 O p t i c s Comm.
Scully,
M .,
A d d ison-W esley, Reading Savage,
C.M.
Slusher,
R .E.,
Potasek, Smith,
and W a l l s ,
R.G.
Handbook,
D.F.
1987 J .
P.,
Rev.
197 3 " O p t i c a l A recchi,
W. 19 7 4
Laser
Physics
(USA).
G rangier,
M. 1987 P h y s .
Lamb,
5^2 4 2 9 .
La
L ett.
Opt.
Porta,
Soc.
Am. B 4 1 5 1 4 .
A.,
Y u r k e , B.
59i 2 5 6 6 .
Param etric O s c illa to r s " ,
F.T.
and
and
Schultz-D ubois
in Laser
E.
Ed.
North
H o l l a n d Amsterdam. S t o l e r , D. T akahasi, Wu,
L.A .,
R e v.
1974 P h y s . H.
Yamamot o,
1 9 6 5 Adv.
Kimble,
L ett.
Rev.
L ett.
Comm.
H .J.,
33 1 3 9 7 .
System 1 227.
H all,
J.L .
and
Wu,
L.
1986 P h y s .
57 2 5 2 0 .
Y.,
Machida,
S.
and N i l s s o n ,
O.
1 986 P h y s .
Rev.
Kitagawa,
M.
A
34 4 0 2 5 . Yamamoto, Bjork,
Y .,
Machida,
G.1987 Soc.
Imoto,
N .,
and
Am. B 4 1 6 4 5 .
Yu e n ,
H .P .
1986 P h y s .
Yurke,
B.
1 984 P h y s .
Yurke,
B.
and W h i t t a k e r ,
Yurke,
B.,
G rangier,
Am. B 4 1 6 7 7 .
S.
Rev. Rev.
P.
L ett.
56 2 1 7 6 .
A 29 4 0 8 . E.A.
1987a O pt.
and S l u s h e r ,
R.E.
L ett.
12 2 3 6 .
1987b J .
Opt.
Soc.
NONCLASSICAL EFFECTS IN PARAMETRIC DOWNCONVERSION
J G R A R IT Y A N D P R T A PST E R
1
IN T R O D U C T IO N
T he m ixing of three electrom agnetic waves w ithin a m edium w ith a second order refractive non-linearity can lead to several different para m etric processes. The sem iclassical theory of these m ixing effects is well reviewed by Yariv (1976). The sem iclassical m ethods hide the funda m ental quantum nature of the interaction. In param etric downconversion the quantum effects arise because two photons are created from one pum p photon. T hese photons satisfy the necessary conditions of m om en tum (phase) and energy conservation w ithin the non-linear m edium . In non-degenerate downconversion separable photons (either by angle or by polarisation) are created. Burnham and W einberg (1970) dem on strated that the two photons are created nearly sim ultaneously, using photon counting coincidence techniques, while M ollow (1973) developed a quantum m echanical treatm ent allowing the two photon coincidence detection probability to be calculated. M ollow suggested that the tim e separation of the two photons would be related to the coherence tim e of the illum inating radiation. This has since been disproved by more recent coincidence experim ents with tim e resolutions of a hundred picoseconds (Friberg et al 1985). U sing non-degenerate type I downconversion two identical trains of photons can be selected by apertures satisfying the phase m atching con ditions. D etection of photons in one train can be used to m odify the quantum statistics of the other train to produce antibunched an d /o r sub-poissonian light (section 2). We report here on two experim ents of this type. The first em ploys single photon detection techniques and a fast optical shutter to create antibunched and slightly sub-poissonian light. T he second uses analogue detection and feedback to produce sub© British Crown Copyright 1988
J G R A R IT Y A N D P R T A PSTER
123
poissonian light at picowatt power levels with a Fano factor of 0.78 over a lim ited bandwidth. T his, to our knowledge, is the lowest post-detection Fano factor yet reported. Sub-poissonian light is potentially useful in finite dose transm ission m easurem ents and the high degree of tem po ral coincidence of the downconverted photon pairs can be exploited in absolute m easurem ent of detector quantum efficiency (section 3). Up until recently, m easurem ent of the degree of tem poral coincidence has been apparatus lim ited. R ecent theory (Fearn and Loudon, 1987;Ou et al, 1987; Prasad et al, 1987) and experim ent (Hong and M andel, 1987; 1988) has shown that recom bining indistinguishable photon pairs in a beam splitter leads to a fourth order interference effect which allows a direct m easure of the photon overlap. Such experim ents allow m ea surem ent of sub-picosecond tim e delays at the single photon level. We show here (section 4) that the detailed shape of the effect can be re lated to the fourier transform of the photon bandw idth determ ined by the phase m atching uncertainty and aperture size. Experim ental results are presented which confirm this theory. The m inim um photon length is lim ited by the bandwidth of the crystal non-linearity coupled with the detailed physics of the param etric downconversion process (Pike and Sarkar, 1988).
2
A N T I B U N C H I N G A N D S U B - P O IS S O N IA N L I G H T SOURCES
2.1 Phase m atching conditions for non-degenerate param etric downconversion For non-degenerate param etric downconversion (Yariv, 1976) a non linear uniaxial crystal is used. The crystal is aligned with its optic axis at an angle (f) with an incident short wavelength laser beam . A bout 10~8 of the beam is converted into pairs of longer w avelength photons which appear in a cone surrounding the laser beam . The photon pairs satisfy the conditions of energy and wave vector conservation in the crys—* tal. Given an incident photon of wave vector ko and angular frequency — a — * cjo, and downconverted photons with wave vectors fci, &2 and angular frequencies these conditions can be w ritten as ko = k\ + &2
(1)
LOq — U)\ -f-
(2)
124
N O N C LA SSIC A L EFFECTS IN PA R A M E T R IC D O W N C O N V E R SIO N
The geom etry illustrated in figure 1 depicts type I phase m atching, where the incom ing beam is polarised in the plane of the crystal optic axis, and the dow nconverted light is polarised perpendicular to this. In this case equation 1 can be rew ritten (jJqtiq = ujitii cos 0 i + oo2n 2 cos 62 (jJitii sin 0i — 002*12 sin 02
(3)
where n\ and n 2 are the ordinary refractive indices of the crystal at frequencies 00\ and o;2, and n 0 is given by n 0 = n e sin + n 0 cos (j)
(4)
where n 0 is the ordinary and n e is the extraordinary refractive index of the crystal at A s a consequence of equations 2-4 the crystal must be negatively uniaxial (rce < n 0) and the cone angle is m axim ised by choosing (j) = 90°. U sing these equations and the refractive index data for deuterated P otassium D ihydrogen P hosph ate (KD*P) (Yariv, 1976), illum inated w ith 413nm light at this angle, the dow nconverted light at 826nm is em itted into a cone of half angle 9.8°. T his is internal to the crystal, and corresponds to 14.3° externally. A broad spectrum of light is em itted at angles around this.
ooq.
Tw o near identical photon stream s can be selected from the cone by placing apertures at opposite ends of the cone base, at angles satisfying
J G R A R IT Y A N D P R T A PSTER
125
the phase m atching conditions (equations 2-4). A more detailed discus sion of the coincidence properties of the photon pairs will be given in section 4.2 2.2 A ntibunching experim ents G iven two identical trains of photons, one can use the detection of a photon in one train (trigger channel) to gate detection of photons in the other train. P u ttin g a dead tim e between gate open events can, in principle, produce an antibunched light source. After each photon there will exist a reduced probability of seeing another for a period equal to this dead tim e. If the photodetection rate in the triggering channel is much higher than the inverse dead tim e (ie we overdrive the system ) the gate open events will becom e regularly spaced. T his leads to a reduction in the noise of the gated beam to below that of Poisson light. The experim ent can be more easily understood by referring to the schem atic diagram shown in figure 2. It is a m odified version of the original apparatus used by Walker and Jakem an (1984; 1985) where the shutter was placed before the non-linear crystal. Splitting of incident UV photons takes place w ithin the non-linear crystal and the pairs of angu larly resolved photons so produced are observed by (photon counting) trigger and signal detectors. At the trigger detector an electronic dead tim e td is introduced so that detection events in this channel are at least td apart. W hen each of these events are registered an optical shutter in the other, signal channel, is opened for a short tim e t. An optical delay is included before the shutter to com pensate for electronic and shutter response tim es, etc. Thus only partners of events actually registered in the trigger channel should be detected in the signal channel. In princi ple this overcom es the problem of random partnerless events arising in the signal channel due to the low efficiency of the trigger channel detec tor, which considerably reduced the the non-classical effects observed in the earlier experim ent (Walker and Jakem an, 1985; 1986). There is, in practice, always som e chance of registering further events in the signal channel during the finite open tim e f. Given N shutter openings per sam ple tim e T with f < T , ^ and assum ing initially that the signal channel detector is 100% efficient we can write the number of counts in the signal channel in a typical sam ple
126
N O N C LA SSIC A L EFFECTS IN PA R A M ET R IC D O W N C O N V E R SIO N
tim e as n = N (1 + m)
(5)
where m is drawn from a Poisson distribution of m ean (1 — rji)rt. Here r is photon pair arrival rate w ithout losses and r)i is the trigger detec tor efficiency. U sing equation 5 we can express the norm alised second factorial m om ent of the photocount n in the form n (2) _
< n ( n - 1) > < n >2 < N ( N - 1) 1) > < m > ( 2 + < m >) > w fi , < m > 1+ — — \2 < N >2 ( 1 + < m > ) 2 + < N > (1 + < m > ) 2
( 6) angular brackets denoting ensem ble averages w ith sam ple tim e T. This quantity is not affected by the finite quantum efficiency of the signal de tector. T he reduction in m easured signal variance over the conventional Poisson variance,the Fano factor F , is however dependent on signal channel quantum efficiency. B y definition F = (Var n ) / < n > = 1 + < n > (n ^ — 1)
(7)
where < n > = rj2 < N > (1 + < m > ) and is given by eq. 7. As photon counting detector efficiencies are low we do not exp ect a large reduction in Fano factor in this experim ent. On the other hand the second factorial m om ent is m inim ised by choosing T < t q when the first term in eq. 6 vanishes. From the theory of deadtim es (M uller, 1974) < JV > = J i r 3-
l + J?irT
(8)
v ’
we can w rite in this case (Jakem an and Jefferson, 1986) (2) = 1 + n r r n I, ________ 1______ ] W T 1 + (1 - m ) 2r t j
(9)
The first term in eq. 9 is a bunching term as T < tq and is reduced by overdriving the trigger channel (rjirrD > 1). However the second term shows an antibunching effect when rji is large and rt the number of photon pairs arriving in the gate tim e is sm all. Hence a short gate tim e and high trigger channel detection efficiency are desirable. It m ust also be noted that any losses or m isalignm ent of the phase m atched apertures will reduce the effective efficiencies rji and ?/2 appearing in eqs. 7 and 9
127
J G R A R ITY A N D P R T A PSTER
TRIGGER
CHANNEL
Figure 2: Block diagram of the photon antibunching apparatus. These effective quantum efficiencies can be estim ated by m easuring the photodetection event coincidences between trigger and signal chan nels C C = tfiq2r (10) and dividing by the mean count rate in each channel nj = Vjr
j = 1 ,2
(11)
To avoid detector afterpulsing effects one can estim ate n ^ from the cross-correlation function g ^ ( r ) from two detectors view ing the signal channel via a beam splitter. It has been shown that J 2)(t] = < "i(0) M r ) > 21 < ni >< n 2 >
i
=
P0)
The accuracy of such a m easurem ent will be lim ited by the excess noise in the detector amplifier and correlations from m acroscopic laser fluctu ations occurring in the m easurem ent bandwidth. The broadband nature of the downconverted cone m eans that in prin ciple detector quantum efficiencies can be m easured over a wide band w idth using a variable m onochrom ator and linked goniom eter to preserve the phase m atching angles. 3.2 Reduced quantum noise in transm ission m easurem ents Changes in the intensity due to scattering,absorption,reflection or re fraction by a sam ple are fundam ental to m icroscopy and im aging tech niques generally and the absorption spectrom eter is a standard piece of laboratory equipm ent finding uses in many scientific disciplines. In the case, often occurring in bio-m edical m easurem ents, where the spec imen can be dam aged or optically changed on illum ination there is an upper lim it to the intensity and exposure tim e. Under this dose lim it or pow er/bandw idth constraint (Yuen 1988) the accuracy of the m ea surem ent can be improved by reducing the number uncertainty in the incident beam . However the loss in the sam ple itself will lim it the noise reduction achievable. In a conventional transm ission m easurem ent only one channel is used. The m ean flux of photons from the source r (or the m ean detected flux r' = rjr) can be m easured to arbitrary accuracy before introducing the sam ple into the apparatus. Introducing a sam ple with absorption coefficient 0 < a < 1 into the beam for a finite tim e T and counting (or integrating the photocurrent in an analogue apparatus) the number s of transm itted photons with a detector of quantum efficiency r] we can
138
N O N C LA SSICA L EFFECTS IN PA R A M ET R IC D O W N C O N V ER SIO N
e s tim a te d a from a = 1 —s / r T
(31)
var(d) = < a 2 > —< a > 2— —— r'T
(32)
with m easurem ent variance
given that s is a Poisson random variable. A m easurem ent m ade using a correlated pair source such as our para m etric downconversion source (w ithout feedback) would involve placing the sam ple in one channel of the apparatus and m easuring direct, n and tran sm itted, s counts in sam ple tim e T . An estim ate of a can be obtained from ~ S f o(33) o\ a- — n------n where for sim plicity we assum e equal detector quantum efficiencies rj. n and s are Poisson random variables with correlated noise
< n s > — < n > < s > = < A n A s > = rj2{ 1 — a)
(34)
hence to order < n > ~ 2 (Jakeman and Rarity, 1986), = ( l - ( 2, - l ) ( l - q ) ) ( l - « l
r'T This estim ator is slightly biased but the bias is always less than the error of m easurem ent. We can define a variance ratio (or Fano factor) relating the conventional (C ) to the correlated pair (P ) m easurem ent as R - ^ 7^ 1 = 1v a r (a c )
(2*7 - 1)(1 -
a)
(36)
There is no reduction in variance possible until rj is greater than 0.5. U sing an estim ator based on partial use of the pair count
(1 - k) < n > + k n - s a = ----:-------- —------------------- :------(1 — k) < n > + k n
.
. [6 i )
and choosing k ~ rj( 1 — a) one can obtain a result which is always sub-poissonian with R = 1 - rj2 ( l - a) (38) In high sen sitivity spectroscopy techniques where the absorption is extrem ely sm all the stab ility of the source at low frequency can be the lim it to sensitivity. Frequency m odulation techniques are thus used to
J G R A R ITY A N D P R T A PSTER
139
move the signal out of the noise dom inated range and differential in tensity m easurem ents (across a spectral feature) are made which can, with care, lead to shot noise lim ited m easurem ent of absorption (Wong and Hall, 1985; Gehrtz et al1985). In these situations both the attenu ated and unattenuated beam s have to be estim ated from a lim ited tim e m easurem ent and the estim ator for a is sim ilar to eq. 33. However in a conventional apparatus the correlation (eq. 34) is zero and var(a) =
(39)
When compared with the photon pair subtractive estim ator (eq. 35) this leads to an improvem ent for all detector efficiencies with
w
When a is sm all this result becom es R ~ 1 — r;
(41)
which is equivalent to the result of G iacobino and Fabre (1988). Using our apparatus as a source of correlated photons in a differential absorp tion m easurem ent of this type would reduce the variance by a factor of 2 below the shot noise level. In the general case of sam ple illum ination using a sub-poissonian source, either the param etric source described above or a direct source such as a high efficiency light em itting diode (Tapster et al, 1987) or sem iconductor laser (M achida et al, 1987), the reduction factor is given by R = 1 — 7/(1 — a ) ( l — F )
(42)
where F is the source predetection Fano factor. Other estim ators can be based on coincidence m easurem ents (Jake man and Rarity,1986) as in the detector quantum efficiency m easure ment discussed above. There the detector can be thought of as an ideal unit efficiency device behind an absorber with a = 1 — r/. In all absorption m easurem ents the noise reduction is always lim ited by R > a
(43)
hence the technique is lim ited to highly transm issive sam ples and all optical losses (eg container windows, filters etc.) have to be m inim ised.
140
4
NO N C LA SSIC A L EFFECTS IN PA R A M E T R IC D O W N C O N V E R SIO N
P H O T O N L O C A L I S A T IO N A T A B E A M S P L I T T E R
4.1 Introduction Several sim ple quantum m echanical m odels of beam splitters have re cently been published (Fearn and Loudon, 1987;Ou et al, 1987; Prasad et al, 1987). Such a beam splitter appears to interact w ith single beam s of classical and non-classical light in the sam e way as m acroscopic m od els. One sim ple result is the binom ial partition noise im parted on an N -photon state due sim ply to the random selection of output direction for each input photon (Brendel et al,1988). The beam sp litter is, how ever, a four port device and a less intuitive result is obtained when indistinguishable N -states appear sim ultaneously at both input ports of the beam splitter. The sim plest case of this type occurs when one-photon states are sim ultaneously input into each port of a 5 0 /5 0 beam splitter (Fearn and Loudon, 1987). The theory predicts that each output will contain either two photons or none. A naive (particle) theory would pre dict a m ixture of tw o’s and on e’s. The required | i , i > photon state can be selected in non-degenerate param etric downconversion experim ents by photon counting coincidence detection (M ollow, 1973). The disap pearance of this coincidence after a beam splitter would be expected if the theory is correct. In a more rigorous theory this effect will occur over a range of path length differences due to the finite length of the photons which can be related, by fourier transform , to their bandw idth. This in turn can be related to the geom etry of the param etric downconversion apparatus and the detailed phase m atching conditions at the crystal. We have constructed a w hite light interferom eter (section 4.4) behind a dow nconversion crystal which ensures that photons from the sam e spatial m ode, and w ithin the sam e bandw idth arrive at the two inputs of a beam sp litter. We dem onstrate that the effect only occurs if the difference of the arrival tim es is w ithin the coherence tim e or inverse bandw idth of the dow nconverted photons. As these photons have a bandw idth of a few nanom etres, this tim e is short (less than lOOfs). 4.2 C oincidence properties of param etric photon pairs T he two photon detection probability P 12, that of d etectin g photons at tim es ti, t2 at points X i , X 2 from a source centred at the origin can
141
J G R A R IT Y A N D P R T A PSTER
be obtained from P n oc< A [ - \ t u
>
(44)
where T W (f) is the positive frequency part of the electric field operator (which can also be described as a photon annihilation operator) at tim e t. For a spontaneous param etric downconversion source (or in the general low flux lim it) M ollow showed that P n oc|< i i +)(ti, Y i ) 4 +)(*2, * 2 ) > |2 A/ | \
—*
A/ I
(45) \
— *
is a good approxim ation. The product A \ } ( t i , X i ) A \ {t 2 , X 2) can be thought of as a two photon wavepacket. For brevity in the ensuing discussion we define the shorthand
(46)
For a finite size crystal evaluation of g12 involves a coherent sum m ation over the illum inated region of all the conjugate frequencies contributing to the two photon wavepacket (M ollow, 1973)
9u oc J JduJidoj2oJico2x ^ { wo, u)i , lj2)8( u)q —
uq —
to2) (47)
x J d h e ' 1 ^ i(*i “ X i / C) +
--->Pn )-
Q
or T*Q,
the
is a linear functional on
In the natural co-ordinate basis
These
are
the
familiar momenta.
Conventionally the co-ordinate basis of the 1-forms is written as (dq*,dq^,...,dqN ), and so a general 1-form on T Q is
155
S SA R K A R
This one form can also be regarded as a one form on the manifold T*Q. dynamics of points on T * Q we need a two form on T*Q.
For the
This special form is known
as a symplectic form a)2 and is written as an exterior derivative
IN
-d
'j
IN
^ Pjdq1
-
Li=l
(9)
^ d q 1 A dp. i=l
(This two form is a linear functional of any pair of tangent vectors at a point of T*Q.) A n y physical observable can be regarded as a smooth mapping from T * Q to the set of all real numbers. 0(q,p).
Consequently an observable O
R, the
can be written as a function
Since it is customary to write the basis of tangent vectors of T * Q as
r 3
3
V
3
3
3
’ 9 ? ........... 9? ’ ^
3 i
’ ^2 ’ " ’ ’ ^
we can associate the tangent vector field O'
3_
I. H ri to an observable O.
30
3 i
3q*
Given any two observables
3o„l 3o„ i-i
( 10)
o i 3p J 3q *1
and O
2
we can easily show that
90, 90 . ( 11) 9q‘
q1
This is just the familiar Poisson bracket; it is thus closely related to the existence of a certain symplectic form on the state space T*Q.
The usual process of going from the
classical theory to the quantum theory is by replacing the Poisson bracket with (i/fi) times the commutator.
The above discussion indicates to us space
of
real
classical
solutions
equivalently, A(x,t) in the cavity.
that we should seek a symplectic structure to
(Moore,
1972)
of the
electromagnetic
the
fields,
or,
In order to have the field kinetic energy bounded
we
will require that A(t,x) is in the function space Hl(I), and 3/3t A(t,x) is in the function space l2(I).
Here I is the closed interval (x 1 o
functions which
is square integrable on
I
and
^ x ^ b(t)}.
is the* set of
H* (I) is the set of functions
whose
156
M O V IN G M IR R O R S A N D N O N C LA SSIC A L LIGH T
x derivative is in L^(I).
The elements of the cotangent bundle are
{A(t,x)
This
is rather
approach
familiar
since
to electromagnetism
,
A,
^
and
A(t,x)j
E
are
canonically
in the radiation gauge.
conjugate
Given
two
in the
standard
classical solutions
A(l)(t,x) and Ap)(t,x) we can introduce the symplectic form
b(t) dx | A ^ ( t , x )
A^Ct.x)
-
A^(t,x)]
A ^ 1 ^ (t ,x)|
( 12)
Owing to (3) and (7) this form is independent of t. set of
real
classical solutions
um (t,x) and
It is always possible to choose a
vn(t,x) such
that
we
have
a
form
of
'orthogonality' relations
(14)
ml n
Such a solution forms a basis for the cotangent bundle. basis will be given presently.)
A(t,x)
where an and
0n
(An explicit construction of the
A n y solution A(t,x) can be written as
=
y (a v (t,x) - P u (t,x)) L n n n n n
(15)
are real numbers.
The un and vn m a y also be regarded as observables since they induce maps into R from the cotangent bundle into R, eg
S SA R K A R
u :
{*•
n
m
157
-
iv
€
( 16 ) v : n
(A> I f )
— + (A|Vn}
W e can then associate operators pn and qn , in the quantum case, with these observables. In particular (Moore, 1972; Sarkar, 1988)
u
> p
n
n (17)
v
n
-- > q
n
and
L n
(we have
mJ
nI
chosen units so that t
=
m
1.)
nm
If R( ) is a function which
is twice
differentiable and invertible and satisfies (Moore, 1972)
R(t-b(t))
=
R(t+b(t)) - 2
(19)
then it is easy to verify that
u (t,x)
=
— --- r (cos(n7rR(t-x) ) - cos (n7rR( t+x) ) ) (2 nx ) 2
(20a)
v (t,x) n
=
— --- 7 (sin(nxR(t+x) ) - s in(nirR( t -x) ) ) (2 nx)
(20b)
satisfy (13) and (14). written as
In terms of the functions in (20) the field operator A(t,x) can be
158
M O V IN G M IR R O R S A N D N O N C LA SSIC A L LIGHT
A(t,x)
=
I
(pnvn (t,x) - qnun (t,x)}
(21)
n
Fr o m (13) and (14) we deduce that
pm
-
(A um)
(22a)
qm
-
-(A v j
(22b)
If we consider n o w a join of two motions bCQ(t) and b^)(t) at t = t', then A
can be
written as
A ( t , x)
I
,X^I
for
1< l'
^23‘>
^ {p n ^ v n ^ ( I ,x ) - q^2 ) i / 2) ( t , x ) j
for
t > t'
(2 4 )
{pn ^ vn1') ^ ,X') " qn ^ Un ^ ^
and
A (t , x)
=
F r o m the orthogonality relations at t = t' we find
p ,
I
m
I n
i< »
-
(u ( 1 ) |u ( 2 ) )
m
I n
If we had another join to a different motion b(3)(t) at t = t" (> have identical relations with 2
3 and 1
2.
linear relation between the (p^,q3) and (p*,ql).
t’) then we would
Clearly we can obtain in this way a W h e n such a relation is converted into
one between creation and annihilation operators, we find that the destruction operator associated with motion
3
involves both destruction and creation operators for motion
1.
159
S SA R K A R
Consequently if we consider a sequence of motions: no motion, motion and no motion, then, even if there were no photons initially, in the final stationary cavity photons would have been generated.
The physical reason for these photons as mentioned earlier are
accelerating charges in the mirrors; this indicates that motion has to be relativistic in order to get large effects.
Our analysis, hopefully, will indicate the magnitude of these
effects together with qualitative features concerning the nature of the fluctuations of the photon states produced.
The important quantities that need to be calculated are
, (i)I v (i+l)v) (v m I n
for i = 1,2.
and.
, (u (i)I v (i+l)N ) m I n
Let us explicitly categorise the motions:
=■ b ( 2 ) ( t )
-
b( 3 ) (t)
0 < t ^ t
=
b ( 2 ) (t ) o
-
b
o
Moreover b(^)(o) = L, b(2)(o) = 0 and l5(3)(t0) = 0. all t.)
t
o
o
(27)
< t
(Clearly we require b(t) ^
0
for
For b(l)(t) and b(3)(t) it is easy to verify that
R (1 ) (S)
-
|
(28)
R (2)( 0
=
- t J -----
(29)
and
b( ’(‘o)
Consequently we have
160
M O V IN G M IR R O R S A N D N O N C LA SSIC A L LIGH T
..(!)/. U (t,X )
1
=
r
(2nx)
COS
nx(t-x) -------- :---------
1
___ COS
-
nx(t+x)) -------- ----------
L
L
J
(30) (!)/«. \ (t,x)
=
I f . n7r (t+x) . n7r(t-x)] ------ j j^sin — ^ - sin — ^ ----I (2 nx)*
and
“ ■*> -
7(27n-7r) T
H
--
o
R ^o l ) (31)
(3).
Vn
The
.
( t ' X)
important
I
"
f
- I
7 7
.
fnx(t+x))
l Sln l - E
(2 n 7r)
o
missing ingredient is an
indicated a method to calculate this when specify b(^)(t) in detail.
.
fnx(t-x)D
J - Sln [—
expression 16 P)(t) 1
o
JJ
for r P)(£). <
o
o
J
>
(69)
0
0 we can arrange this and so have squeezing for some
values of 6 .
A particularly simple form for b(2)(t) is simple harmonic motion
b
( 2)
( t )
=
CQ
+
with c0 + C| = L and c0 , cj > 0.
b
o
For ojt0
b(t ) - L b(o)
o
=
=
1
(1 — cosoJt ) (L + c.cosoJt )
o
but for a full
Consequently, in this case, we can consists of half a cycle of sinusoidal The squeezing parameter (Walls and
gj
N o w for this motion
2 CJ c.
7r this is non-zero,
(70)
C O SG Jt
1
period
(71)
o
(ie u)tQ =
2%)
this is zero.
only have a squeezed photon state if the motion motion.
For complete cycles there is no squeezing.
Reid, 1986) is easily calculated to be
c 1 (L - c 1)
gj(L-2 c
T 2”n 2 47r
(72)
^)
Although taking the vacuum state to be the initial state is somewhat intriguing, the case when
there
are some
photons in the initial state m a y
be
easier to realise.
For
definiteness we consider the state initially to be
n' > I o
=
K O1>+) ° h
nQ is the mo de number while n^ is the number in the mode. photons in the state generated is calculated as before and we find
The me an number of
170
M OVING M IRRORS A N D N O N C LA SSIC A L LIGHT
< n ' | a < 3)V
3 > | n ’>
36n
n'
o
~z (1-5 nno )
\6 7r4 / (n-n )
nn o
o
n .I oJ
d
o
-
(74)
So if iIq is large enough the number of photons created in modes with m o de number (n (* n0) is proportionately large.
In some sense the presence initially of photons is
stimulating the generation of ne w ones (but not at the same m o de number).
I o
n'Xn' I o o
fl + 2n' L o
5
(77)
V n nJ
For m o d e numbers other than nQ the m i ni mu m uncertainty condition is satisfied.
From
(76) we see that the discussion of squeezing for photons in these m o de numbers goes through as before.
For
all mechanical
mirrors
the
treatment
we
have
given
should
be
adequate.
If
non-mechanical mirrors can be designed to have relativistic motion then corrections will be necessary although our hope is that qualitative features will remain unchanged. check this a non-perturbative calculation of R.P) is needed. in (70) we can take
To
For the case of the motion
171
S SA R K A R
ncot
(78)
Substituting in (19) we obtain
iucjt V I
rn
, .. u-n r e t
J
n-u
(nojc. ) - e 1
There is one such relation for each integer u. such an expansion m a y be convergent.
° J
n-u
(-nucjl 1J
= 26
uo
(79)
Preliminary investigations suggest that
However the calculation of the me an number of
photons, etc, will still be difficult.
W e conclude by giving some estimates of the orders of magnitude of the effects that we are predicting.
For the case of a full cycle of periodic motion
, ( 3) + (3)
< o | a ^ ' ' a ^ ' |o>
1
1
=
(bv~ y (o)L)2
^
m=l
1
^ I n T 2
l(m +l)3
(1 -m)3J
(80)
where r = L/c^ and e = c^o/c.
(Here we have changed units and reintroduced c the
speed of light which we took to be 1 before.)
state, n^
~ 9 x 10^, then T
is large and the average number of
photons produced per second is 6 x 10^, a substantial increase. experimentally detectable.
If we choose for
Such an effect m a y be
For the case when the initial state is a vacuum the squeezing
parameter is (6 x 10“^)/n^.
For an initial state inQ> this result continues to hold for
m o d e numbers n different from nQ .
4.
We
CONCLUSIONS
have shown that in general motion of a mirror (from rest to eventual rest) will
create photons, even if the initial state is a vacuum. from
the
vacuum
squeezed. Without en h an ced
are
super-Poissonian,
Unless the mirror motion
In particular the photons generated
non-thermal,
going to extremely fast motions the amount if
Experiments
the
initial
checking
state these
is
a
and
in
certain
circumstances
is effectively relativisitic the effects are small.
p ure
predictions
n um ber would
be
of photon state
with
important
generation is mu ch a in
large
population.
deepening
our
understanding of quantum field theory in non-trivial geometries.
ACKNOWLEDGEMENTS
I a m grateful to J.G. Rarity, J.S. Satchell and P.R. Tapster for discussions.
REFERENCES Arnold, V.I. 1978, Mathematical Methods of Classical Mechanics, Springer-Verlag, N e w York. Caves, C.M. 1980, Phys. Rev. Lett. 45 75. Caves, C.M. 1981, Phys. Rev. D23 1693. Loudon, R. 1981, Phys. Rev. Lett. 47 815-818. Moore, G.T. 1970, J. Math. Phys. U 2679-1691. Sarkar, S. 1988, J. Phys. A21 971-980. Stoler, D. 1970, Phys. Rev. DI 3217. Yuen, H.P. 1976, Phys. Rev. A13 2226. Walls, D.F. and Reid, M.D. 1986, Frontiers in Quantum Optics, Ed. E.R. Pike and S. Sarkar, Hilger, Bristol, 72-105.
PROPAGATION OF NONCLASSICAL LIGHT
I ABRAM
1.
INTRODUCTION C la s s ic a l o p tic s d e sc rib e s q u ite s u c c e s s fu lly th e
g a tio n
of
la se r lig h t,
s p a r e n t m e d iu m . ra d ia tio n ,
b o th in f r e e - s p a c e and i n s i d e a t r a n
The r e a s o n i s t h a t i n
th e e le c tr ic
be
tre a te d
c la ssic a lly
T h e s am e h o l d s a l s o
alw ay s
d e sc rib e d
be
as
a
and th u s t h e i r
f o r th e rm a l
sta tistic a l
o f th e e le c tro m a g n e tic f i e l d .
e q u a tio n s
and
tre a t
On t h e o t h e r h a n d ,
lig h t
c la s s ic a l o p tic s
in
sta te ,
M a x w e ll
e le c tro m a g
of th e
in d u ced
f i e l d w ith a p o la riz a tio n .
of
q u a n tu m
th is
p ro b lem
th e
e le c tric
con
o p tic s can ad d re ss ade
is
not
fie ld
T he f a i l u r e
su rp risin g : in
such as a s in g le -p h o to n s t a t e
of T he
a n o n c la ssic a l or
does n o t c h a r a c t e r i z e th e p h o to n s t a t i s t i c s .
a squeezed T hus,
its
t h e M a x w e ll e q u a t i o n s c a n n o t g i v e a n y i n f o r m a t i o n
on
use
in
th e
e v o lu tio n
of
th e
p h o to n s t a t i s t i c s
th ro u g h l i n e a r o r n o n lin e a r t i o n a l q u an tu m o p t i c s t h a t t h i s th e o ry m odes.
can
n e ith e r c la s s ic a l o p tic s nor th e
fo rm u la tio n
e x p e c ta tio n v a lu e of sta te ,
w hich
p ro p a g a tin g
q u a te ly th e p ro p a g a tio n o f n o n c la s s ic a l l i g h t .
ra d ia tio n
in
p ro p a g a tio n
c la ssic a l
th e in te r a c tio n
p h e n o m e n o lo g ic a lly th ro u g h th e
v e n tio n a l
of
p e rm it th e c a lc u la tio n o f b o th th e s p a t i a l p r o g r e s
fie ld ,
m edium
sta te
m ix tu re o f c o h e re n t
The
s io n and th e te m p o ra l e v o lu tio n o f a n e tic
co herent
t h r o u g h t h e m a c r o s c o p i c M a x w e ll
e q u a tio n s. sta te s
a
a n d m a g n e t i c f i e l d s may b e w r i t t e n
term s o f t h e i r e x p e c ta tio n v a lu e s , may
p ro p a
is
m ed ia. le ss
as l i g h t p ro p ag ates
The i n a d e q u a c y o f c o n v e n
o b v io u s,
and l i e s
in th e
fact
i s b a s e d on t h e H a m i l t o n i a n o f t h e r a d i a t i o n
I t th u s a d d re s s e s o n ly th e tim e - e v o lu tio n o f th e
fie ld
174
PR O P A G A TIO N OF N O N C LA SSIC A L LIGHT
p ro p a g a tin g in
(m o stly )
em pty s p a c e an d i n t e r a c t i n g w i t h i n d i
v id u a l m a te ria l p o in ts p o la riz a b le e n titie s
th a t cause
t h e r a d i a t i o n m odes. how ever,
cannot
e m itte rs,
(n o n lin e a r)
A t h e o r y b a s e d on a
address
p ro g re ssio n of th e co p ic ,
re p re se n tin g
fie ld
ab sorbers
in te ra c tio n s m odal
among
H a m ilto n ia n ,
d i r e c t l y t h e p ro b lem o f t h e s p a t i a l o p e r a to r s p ro p a g a tin g th ro u g h m acros
t r a n s l a t i o n a l l y - i n v a r i a n t , p o l a r i z a b l e m ed ia,
c an n o t produce
or
sp a tia l
d iffe re n tia l
e q u a tio n s
sin c e i t
w h ich
co u ld
a c c o u n t f o r t h e c h a n g e i n mode s t r u c t u r e a n d i n t h e e x c i t a t i o n o f e a c h m ode,
a s d i f f e r e n t m ed ia a r e i n t r o d u c e d
a l i g h t beam .
For t h i s
reason,
one o f t h e m ost b a s i c p ro b lem s
of c la s s i c a l p ro p a g a tiv e o p tic s ,
th e
sp a tia l
p ro g re ssio n
l i g h t t h r o u g h a t r a n s p a r e n t r e f r a c t i v e m e d iu m , tio n
and
re fra c tio n
v a c u u m /d ie le c tric
th a t
o ccurs
in te rfa c e ,
as
in te ra c tio n
te rm
th a t,
w hen
lig h t
cro sses T here
fie ld
due to
is
fre e -fie ld
can d e s c rib e th e m o d ific a tio n o f th e s p a t i a l
sio n of th e
a
a t a l l w ith in
o p tic s:
added to th e
of
and th e r e f l e c
can n o t be t r e a t e d
t h e c o n v e n t i o n a l f o r m a l i s m o f q u a n tu m to n ia n ,
in th e p a th of
no
H am il
progres
re fra c tio n . d e sc rib e d
by
s e c o n d - q u a n t i z a t i o n o p e r a t o r s a c t i n g o n to t h e vacuum s t a t e
N o n c la ssic a l s t a t e s
|0>
of th e ra d ia tio n v a lu e s
of
C le a rly ,
th e
of lig h t
fie ld ,
and
e le c tric
are
th e
u su a lly
c o rre sp o n d in g
e x p e c ta tio n
and m a g n e tic f i e l d s a r e o f t e n
th e d e s c r ip tio n o f p ro p a g a tio n o f n o n c la s s ic a l
re q u ire s
zero. lig h t
a d i r e c t q u a n tu m m e c h a n i c a l t r e a t m e n t o f t h e s p a t i a l
p ro g re ssio n of th e
fie ld
d e v e lo p e d
t h e f o r m a l i s m o f P r o p a g a t i v e Q u an tu m O p t i c s
(A bram , th e
w ith in
1987;
1 9 8 8 ), w h ich c a n a d d r e s s t h e p ro b le m
in d e p e n d e n tly o f i t s
In th e
of
(re fra c tiv e ) in te rfa c e th e
tem p o ral e v o lu tio n .
fo llo w in g S e c tio n s ,
p ro p a g a tiv e
p a rtic u la rly
of
Such a t r e a t m e n t h a s b een
q u an tu m m e c h a n i c a l d e s c r i p t i o n o f t h e s p a t i a l p r o g r e s s i o n
of lig h t
of
A bram ,
o p e ra to rs.
q u a n tu m
o p tic s
th e tre a tm e n t of m edium
(S ectio n 4 ). p h y sic a l
(S e c tio n
(S ectio n 2 ),
p ro p a g a tio n 3),
In o rd e r to
id e a s,
g e o m e try w hich g r e a t l y
we o u t l i n e t h e
we
use
and
fe a tu re s
a n d e x a m i n e m o re
th ro u g h acro ss
fa c ilita te a
m ain
th e
p lan e-w av e
s im p l i f i e s th e m a th e m a tic a l
a
lin e a r
a re fra c tiv e d isc u ssio n p ro p a g a tio n tre a tm e n t.
175
I ABRAM
We t h u s c o n s i d e r e l e c t r o m a g n e t i c p l a n e w a v e s p r o p a g a t i n g a l o n g th e + z -a x is
i n a r e f r a c t i v e m e d iu m , w i t h t h e e l e c t r i c
fie ld
and th e m a t e r i a l p o l a r i z a t i o n P b o th a lo n g th e x - a x i s , m ag n etic f i e l d a x is.
We
H and th e m a g n e tic in d u c tio n B
c o n s i d e r n o n m a g n e tic m a t e r i a l s ,
t h i s g e o m e t r y we c a n t r e a t a l l d ire c tio n Q ua n tu m
b e in g
a lw ay s
m e c h a n ic al
c o rre sp o n d in g
q u a n titie s
im p lic it.
o p e ra to rs
c la ssic a l
are
We
a lo n g
as
by
th e
+Y”
sc a la rs,
In
th e ir
G au ssian u n i t s .
d istin g u ish e d
q u a n titie s
and th e
s o t h a t B = H.
use
E
from
th e c a re t
th e
( A) a n d we
t a k e H = 1.
2.
BASIC CONSIDERATIONS OF PROPAGATIVE QUANTUM OPTICS T he d e s c r i p t i o n o f l i g h t p r o p a g a t i o n r e q u i r e s
tio n
of
le a st)
th e v a r ia tio n of th e e le c tro m a g n etic f ie ld
tw o d i m e n s i o n s :
d in a te .
tim e p lu s
C o n v en tio n al
m ech an ical th e o r ie s )
q u a n tu m
(at le a st) o p tic s
fie ld ,
(lik e
m o tio n
of
m ost
th e
q u a n tu m
H am ilto n ian
th e f ie ld
(i.e .
th e
t h a t can be o b ta in e d
equa
from t h e
H e is e n b e r g e q u a t i o n ) , th r o u g h w hich t h e t i m e - e v o l u t i o n o f norm al
m odes o f t h e
p l e t e quan tu m re q u ire s
an
fie ld
m e c h a n ic al e x te n sio n
can be c a lc u la te d . d e sc rip tio n
of
sp a tia l
o f q uantum o p t i c s t o
d i f f e r e n t i a l eq u a tio n s.
q u i t e e a s i l y by u s e o f t h e tro m a g n e tic f i e l d 6,
momentum
C le a rly ,
lig h t
c a lc u la tio n s of th e s p a tia l p ro g re ssio n o f th e d ire c t
(at
and th u s w ith in t h i s th e o r y o n ly
te m p o r a l d i f f e r e n t i a l e q u a t i o n s c an be s e t up of
a lo n g
one s p a t i a l c o o r
g iv e s a c e n tr a l r o le to
o f th e e le c tro m a g n etic tio n s
c o n sid e ra
th e
a com
p ro p a g a tio n
in c lu d e e x p l i c i t fie ld ,
th ro u g h
T h is can be a c h ie v e d
o p e ra to r
of
th e
e le c
s i n c e by d e f i n i t i o n (2 . 1)
or eq u iv a le n tly
6 = -i w here
'V
is
[M ]
any w a v e fu n c tio n and 0 any o p e r a t o r .
(2 .2)
176
P R O P A G A T IO N OF N O N C LA SSIC A L LIGHT
In c a l c u l a t i n g th e H a m ilto n ia n th e
fie ld
is
tre a te d
q u a n tu m
and
momentum
m e c h a n ic a lly th ro u g h seco n d -
q u a n tiz a tio n o p e ra to rs.
On t h e
in tro d u c e d
th ro u g h
in d u c e d p o l a r i z a t i o n ,
th ro u g h i t s
o p tic a l s u s c e p tib ility
p h en o m en o lo g ical
its
tre a tm e n t
o th e r
of
hand,
^
th e
=
th e
m ed iu m
P /E .
m a te ria l
T h is
q u a si-
m edium p e r m i t s fie ld
from t h e
d e s c rip tio n of th e fie ld -m a tte r in te ra c tio n .
o p tic a l s u s c e p tib ility
is
or e q u iv a le n tly
s e p a r a t i o n o f t h e p ro b le m o f p r o p a g a tio n o f t h e m ic ro sc o p ic
o p e ra to rs,
can be c a lc u la te d b e fo re h a n d
T he
( f o r exam
p l e th ro u g h th e d ia g ra m m a tic te c h n iq u e s d e v e lo p e d in n o n lin e a r o p t i c s b y Yee a n d G u s t a f s o n c h a ra c te ristic p a g a tio n . tre a te d
(1978))
Thus, w ith in t h i s in
a
a
approach,
lig h t
p ro p a g a tio n
is
m anner c o m p le te ly a n a lo g o u s t o th e m ac ro sc o p ic
M axw ell e q u a t i o n s , o p tic s.
and th e n u se d sim p ly a s
c o n s t a n t o f t h e m edium i n t h e t r e a t m e n t o f p r o
on w h ic h
T h is s i m i l a r i t y
is
based
c l a s s i c a l p ro p a g a tiv e
su g g e sts t h a t th e r e s u l t s
c a l o p t i c s may s e r v e a s a g u i d e i n
c h o o sin g
th e
of c la s s i a p p ro p ria te
q u a n t u m m e c h a n i c a l e n e r g y a n d momentum o p e r a t o r s t h a t d e s c r i b e l i g h t p ro p a g a tio n p ro p e rly . From c l a s s i c a l wave
e n te rs
o p tic s,
we know t h a t w hen a h a r m o n i c l i g h t
a tra n sp a re n t
e v o lu tio n does n o t change: sa m e
frequency.
n e tic a ll
fie ld
in th e
fie ld
of
re m a in s
E q u iv a le n tly ,
p ro p a g a tin g
f i e l d , and i s
consequence m a te ria l
re fra c tiv e
it
energy
m e d iu m ,
its
tim e -h a rm o n ic
at
tim e th e
th e energy o f th e e le c tro m a g
th ro u g h
d iffe re n t
m ed ia,
i n d e p e n d e n t o f t h e m e d iu m . c o n se rv a tio n ,
sin c e
a
re m a in s T h is i s
a
tra n sp a re n t
d o e s n o t e x t r a c t any e n e r g y from t h e e l e c t r o m a g n e t i c
p ro p a g a tin g th ro u g h i t .
On t h e o t h e r
hand,
th e
e le c
t r o m a g n e t i c e n e r g y - d e n s i t y , g iv e n by
U
is as
=
o TT
(E 2
+
H2
+
4TTPE)
l a r g e r i n s i d e t h e r e f r a c t i v e m edium ( t h a n i n it
e le c tric tio n s,
in c lu d e s th e e f f e c ts
of th e
free
(2 .3 ) space),
in d u ced p o l a r i z a t i o n .
For
a n d m a g n e t i c f i e l d s c o n s i s t e n t w i t h t h e M a x w e ll e q u a th e
e n e rg y -d e n sity
i n d e x o f t h e m ed iu m
is p ro p o rtio n a l to th e r e f r a c tiv e
(n = J 1 + 4irp() , a n d i s
th u s r e la te d to th e
177
I ABRAM
fre e -fie ld
e n e rg y by u. R
n u
w here t h e s u b s c r i p t s R and o
(2 .4 )
o
refer
to
th e
d e n sity
r e f r a c t i v e m edium a n d i n e m p ty s p a c e r e s p e c t i v e l y .
in
th e
T h is means
t h a t t h e quan tu m m e c h a n ic a l d e s c r i p t i o n o f
lig h t
m ust
in d e p e n d e n t o f th e
be
based
on
a
m edium o f p r o p a g a t i o n , e ig e n v a lu e s
are
H a m ilto n ia n t h a t i s
b u t on a n e n e r g y - d e n s i t y o p e r a t o r w hose
p ro p o rtio n a l to th e r e f r a c tiv e
th e H a m ilto n ia n and e n e r g y - d e n s ity o p e r a to r s each
o th e r
p ro p a g a tio n
in d ex .
are
S in ce
re la te d
to
by i n t e g r a t i o n o v e r t h e v o lu m e V o f t h e c a v i t y o f
q u a n tiz a tio n ,
Ot = V u th is
(2 .5 )
im p lie s t h a t e n e rg y -c o n s e rv a tio n upon p ro p a g a tio n can
in tro d u c e d
in
q u a n tu m o p t i c s t h r o u g h t h e a p p r o p r i a t e d e f i n i
tio n of th e c a v ity of q u a n tiz a tio n , A t t h e s am e t i m e , th e
p a th
of
be
as w ill be seen l a t e r .
i n t r o d u c t i o n o f a r e f r a c t i v e m edium
a l i g h t beam , m o d i f i e s i t s
in
s p a tia l p ro g re ssio n :
t h e w a v e l e n g t h o f a h a r m o n ic wave i s
re d u c e d by th e r e f r a c t i v e
in d e x
t h e momentum ( w a v e v e c t o r )
o f t h e m e d iu m .
E q u iv a le n tly ,
o f a h a r m o n i c w a v e d e p e n d s o n t h e m edium o f p r o p a g a t i o n a n d i s p ro p o rtio n a l to th e re f r a c tiv e
in d ex .
T h is i n t u r n means t h a t
t h e a p p r o p r i a t e momentum o p e r a t o r t h a t d e s c r i b e s p ro g re ssio n m ed iu m , to
th e
of
and i t s
th e
fie ld
e ig e n v a lu e s
re fra c tiv e
m a te ria l
m e d iu m ,
(th e w av ev ecto rs)
in d ex .
a r e s e v e r a l w ays o f
"m ech an ical"
d e fin in g because
of
momentum.
th e above re q u ire m e n ts i s th e and
corresponds
th e
fie ld
th e
tio n a ry sity
is
m edium
momentum
d iffe re n t
so -c a lle d to
a
w ays
in sid e
a
th a t th e
"e le c tro m a g n e tic "
M in k o w sk i
19 8 5 ) .
a v e c to r d i r e c t e d a lo n g th e
momentum,
p s e u d o - m o m e n tu m
in v a ria n c e of th e f i e l d
(P e ie rls,
are p ro p o rtio n a l
The d e f i n i t i o n t h a t c o n fo rm s t o
e sse n tia lly
d e sc rib e s th e s p a tia l
sp a tia l
In c l a s s i c a l e le c tro d y n a m ic s th e r e
" t o t a l 11 momentum may b e p a r t i t i o n e d b e t w e e n and
th e
q u an tu m m e c h a n i c a l l y d e p e n d s on t h e
in sid e
a
th a t sta
T h e M in k o w s k i momentum d e n z -a x is,
w hose
v a lu e
is
178
P R O P A G A T IO N OF N O N C LA SSIC A L LIGHT
g iv e n by
(2 .6) w here D =
is th e e le c tr ic
E +
4-rcP =
(1
+
4 rtjx)
(2 .7 )
E
d isp lacem en t.
We t u r n now t o t h e c a v i t y o f q u a n t i z a t i o n . is
u su a lly
tak en
i n t e g r a l number o f in te re st,
to
be
fin ite
w av elen g th s
of
obeys
T h is
th e
ra d ia tio n
stu d y th e
can
cover
fie ld
a ll
in i n f i n i t e
of
(in fin ite )
space,
in tro d u c e d in
th e
box,
p e rio d ic
boundary
fille d
T hus,
th u s
in
th re e
In o rd e r to
su ffic ie n t
to
space,
th e d im en sio n s
co m b in ed w i t h e q .
c a v ity of q u a n tiz a tio n , its
and
th e
p la n e s
th e
on
p e rio d ic -b o u n d a ry
is
V o n
p e rio d ic -b o u n d a ry
th e n
of
w i t h a r e f r a c t i v e m e d iu m ,
re q u ire
th a t
in d ex ,
c o n d i t i o n s a r e o b ey ed g e t c l o s e r by
T h is e q u a tio n ,
m e d iu m .
is
space.
When a t r a n s p a r e n t r e f r a c
t h e v o lu m e
R
te n ts,
th e
p e rio d ic -b o u n d a ry c a v ity change s in c e a l l w a v e le n g th s
t h e sa m e f a c t o r . c a v ity
it
(in fin ite )
becom e s h o r t e r by t h e r e f r a c t i v e w h ich
of
mode s p a c i n g i s b e y o n d
At o p p o site s id e s o f
ex am in e one " r e p r e s e n t a t i v e u n i t " . th e
an
p e r io d ic boundary c o n d itio n s and th u s th e c a v ity
t i v e m e n d iu m i s of
is
waves
c o n s t i t u t e s a " r e p r e s e n t a t i v e u n i t 11 w h o s e r e p e t i t i o n d im e n sio n s
c a v ity
b o x , w hose l e n g t h
b u t l a r g e enough so t h a t i t s
e x p e rim e n ta l r e s o lu tio n . fie ld
a
(2 . 8)
(2 .4 )
in d ic a te s
c o n d itio n s
in d e p e n d e n tly
of
its
energy c o n te n ts a re a ls o
th a t
if
we
be obeyed by th e m a te ria l
con
in d e p e n d e n t o f th e
That is , (2 .9 )
and th u s t h e H a m ilto n ia n tro m a g n e tic
fie ld
(and t h e t i m e - e v o l u t i o n )
q u a n tiz e d
in
such
a c a v ity
of th e e le c is
t h e sam e,
179
I ABRAM
w h e th e r
th e
c a v ity
r e f r a c t i v e m ed ia. tiv e
o p tic s
e m p ty
or
c o n ta in s
(one
or
m o re )
The e n e r g y c o n s e r v a t i o n f e a t u r e o f p r o p a g a
is th u s e q u iv a le n t
p e rio d ic -b o u n d a ry tio n
is
to
th e
p re se rv a tio n
of
th e
c o n d itio n s used fo r th e c a v ity o f q u a n tiz a
i n quan tu m o p t i c s . S im ila rly ,
th e
M in k o w s k i
p e rio d ic -b o u n d a ry c a v ity
momentum
co n ta in e d
in
th e
is, V
G = VRg R = T hus, th e
(1 +
E B = n Vo g o
a c c o rd in g to t h i s d e f i n i t i o n ,
( 2 - 10)
t h e momentum
c a v ity a re p ro p o rtio n a l to th e r e f r a c tiv e
c o n te n ts
in d ex ,
t h e y d e p e n d on t h e m a t e r i a l c o n t e n t s o f t h e c a v i t y way a s t h e w a v e v e c t o r .
of
and th u s
i n t h e sam e
T h i s i m p l i e s t h a t t h e q u an tu m m e c h a n i
c a l o p e r a t o r t h a t c o r r e s p o n d s t o t h e M i n k o w s k i momentum s h o u l d d e sc rib e
s u c c e s s fu lly th e s p a tia l p ro g re ssio n of th e
fie ld
in
a n y m e d iu m .
3.
PROPAGATION IN A REFRACTIVE MEDIUM T he d i s c u s s i o n o f S e c t i o n 2 may e a s i l y b e a p p l i e d t o
q u a n tiz e d fie ld
e le c tro m a g n e tic
o p e ra to rs
(A b ram ,
fie ld .
The e l e c t r i c
q u a n tiz e d in f re e space
may
be
th e
and m a g n e tic w ritte n
as
1 98 7)
£ (z ,t)
=
)
( 3 . la )
(£j+6_j)
(3 -lb )
ej
and
~iSj|~v Ij j w here
is th e
S j = +1
is
( t o w a r d s + z)
th e
j
freq u en cy o f th e j - t h sig n
of
and n e g a tiv e
j , fo r a
h a rm o n ic p la n e wave,
p o sitiv e
fo r
b ac k w a rd -g o in g
and
a fo rw a rd -g o in g (to w ard s
-z)
180
PR O P A G A T IO N OF N O N C LA SSIC A L LIGHT
wave.
The
th e j - t h
o p e ra to rs
mode o f t h e
re la tio n s. c re a tio n
They
d iffe r
im p lic it
d e fin e d in eq s.
and fo llo w
in
in
(3 .1 ),
6 t(6 j).
Bose
th a t
co m m u tatio n
th e
sp a tia l
The
are
not
a sso c ia te d
w ith
a
sin g le
in d ic e s,
wave,
fie ld .
and
and do n o t
They a r e
sim p ly
sin c e th ey co rresp o n d to th e
sp a tia l
and m a g n e tic f i e l d s
The f r e e - s p a c e e l e c t r i c us
e^ a n d fij
a r e com posed o f a l i n e a r c o m b in a tio n o f
v i d u a l co m p o n en ts o f t h e
p e rm it
and
a re u n d er
o p e ra to rs
and a n n i h i l a t i o n o p e r a to r s o f o p p o s ite
a m a th e m a tic a l co n v e n ie n c e,
(3 .1 )
a p h o to n in
h o w e v e r from t h e t r a d i t i o n a l p h o to n
o p e ra to rs,
re p r e s e n t p h y s ic a l o b se rv a b le s o f th e
e le c tric
(a n n ih ila te )
o s c illa tio n s of th e ele c tro m a g n etic f i e l d
sto o d to be
th u s
fre e -fie ld ,
(a n n ih ila tio n )
te m p o ra l
c re a tio n
c re a te
F o u rie r
ex p a n sio n
in d i
of
th e
re sp e c tiv e ly . and
to convert a l l
m ag n etic
fie ld
o p e ra to rs
c la s s ic a l o b se rv a b le s o f th e
e le c tro m a g n e tic f i e l d
i n t o t h e i r q u an tu m m e c h a n i c a l e q u i v a l e n t
o p e ra to rs,
f r e e - s p a c e a n d i n s i d e a m a t e r i a l m e d iu m .
T h is
b o th
in
c o n v e r s io n can be done by sim p ly r e p la c in g
and m a g n e tic f i e l d s
th e
e le c tric
in th e c la s s ic a l d e f in itio n o f th e o b serv
a b le by th e c o rre s p o n d in g o p e r a to r s o f e q s. In p a r tic u la r ,
th e
energy
d e n sity
(3 .1 ). o p e ra to r
in sid e
a
l i n e a r m edium i s
u = jA
( £ 2 + ft2 + 4n^fe2 ) =
= si
j®-j +
(3 .2 a )
+ 4^ gj®-j)
4 4 w h e r e we h a v e a s s u m e d t h a t t h e e n e r g y - d e n s i t y u n ifo rm ly
in
space,
term s t h a t o s c i l l a t e B ose
c re a tio n
a n d t h u s we e l i m i n a t e d sp a tia lly .
In term s
and a n n i h i l a t i o n o p e r a t o r s ,
(3,2b) is
d istrib u te d
f r o m t h e sum , a l l
of
th e
fre e -fie ld
each in d iv id u a l j -
c o m p o n e n t o f t h e e n e r g y - d e n s i t y o p e r a t o r may b e w r i t t e n
aj = ^
as,
{6j 6j + 6- j 6-j " 2*2(6j " 6- j )(6-j " 6j )} (3-3)
181
I ABRAM
T hus,
in s id e a lin e a r d i e l e c t r i c th e e n e rg y -d e n sity o p e ra to r u
i s n o t d i a g o n a l w hen e x p r e s s e d i n te r n s o f f r e e - f i e l d a n d i n c l u d e s a c o u p l i n g b e t w e e n t h e tw o m e m b e rs o f
o p e ra to rs each
p a ir
o f c o u n t e r - p r o p a g a t i n g m odes.
T h is c o u p lin g c o n s i s t s o f te rm s
t h a t do n o t c o n s e r v e e n e r g y t o
first-o rd e r,
and
are
of
th e
form
2^ ( 6 j A j T h e s e i n t e r a c t i o n t e r m s may b e d e n sity o p e ra to r
becom es
r e la te d to th e of
a
w hen
and
th e
energy-
ex p ressed
in
T he r e f r a c t e d - w a v e o p e r a t o r s
th e ^ j/^ j
f r e e - s p a c e o p e r a t o r s & j / £ j t h r o u g h a Bogo-
liu b o v tr a n s f o r m a tio n . a p p lic a tio n
(3 .4 )
e lim in a te d ,
d ia g o n a l
re fra c ted -w a v e b a s is s e t. are
+ 6 j6 _ j)
T h is tr a n s f o r m a tio n c o rre s p o n d s to th e
u n ita ry o p e ra to r
(sim ila r to th e
"squeeze"
o p e r a t o r u s e d f o r exam ple i n t h e p ro b le m o f p a r a m e t r i c g e n e r a tio n
(S to le r,
0
1970; S t o l e r ,
1 9 7 1 ; Y uen,
1976)
of th e
form
= e x p { J ^ ( 6 _ .6 _ _ .- 6 t6 * ..) } = i
(3 .5 a )
= e x p ( ^ ( 6 . 6__.-£*&*_.) )
(3 .5 b )
i w here
\ w ith
n b e in g th e
tra n sfo rm a tio n tio n )
I n ( 1 + 4 irpj) = |
re fra c tiv e
in d e x
In(n) of
th e
(3 .6 ) m e d iu m .
r e la te s th e re fra c ted -w a v e c re a tio n
o p e r a t o r s fit(fi_.)
T h is
(a n n ih ila
to th e c o rre sp o n d in g f r e e - f i e l d
o pera
t o r s th ro u g h
f i t = ft 1 f i t = 6 1
fit
=
6 fit
ft = c o s h j
fit
- sin h ^
fi_j
ft = c o s h ^ £>j - sin h 'jj fi*j
ft”"1 = c o s h ^
fit
+ s i n h j j fi_^
(3 .7 a )
(3 .7 b )
(3 .7 c )
182
P R O P A G A T IO N OF N O N C LA SSIC A L LIGHT
( 3 . 7d)
The r e l a t i o n s h i p b e tw e e n t h e w ave
o p e ra to rs
th e r e f r a c tiv e
(eqs. in d e x ,
3 .7 )
fre e -fie ld
and
re fra c te d -
may a l s o b e e x p r e s s e d i n t e r m s o f
sin c e (3 .8 a ) (3 .8 b )
In se rtin g to r
e q s . (3 .7 )
in to eq.
(3 .3 ),
th e energy d e n s ity
op era
i n t h e r e f r a c t e d - w a v e b a s i s s e t may b e o b t a i n e d a f t e r som e
a lg e b ra as
(3 .9 )
The
e n e rg y -d e n sity
re fra c te d -w a v e
o p e ra to r
b a sis s e t,
is
th u s
and a l l
its
d iag o n al
in
th e
e ig e n v a lu e s a re r e la te d
to th e
fre e -fie ld
e n e r g y - d e n s ity e ig e n v a lu e s by th e r e f r a c t i v e
in d ex
n.
is
o b ta in e d
T h is
in c la s s i c a l
M ax w ell e q u a t i o n s in
free-sp ace.
o p e ra to r
e x a c tly
in
o p tic s
fo r th e
Thus,
t h e sa m e (eq. 2 .4 ) ,
fie ld
o p tic s
w h en
as
th a t
so lv in g
th e
i n s i d e a r e f r a c t i v e m ed iu m a n d
d ia g o n a liz a tio n
q u an tu m
re la tio n sh ip
is
of
th e
e n e rg y -d e n sity
e q u iv a le n t to th e s o lu tio n of
t h e M a x w e ll e q u a t i o n s i n c l a s s i c a l o p t i c s . The p r o p o r t i o n a l i t y o f t h e e n e r g y - d e n s i t y e i g e n v a l u e s th e r e f r a c tiv e
in d ex im p lie s t h a t in o r d e r t o have e n e rg y con
s e r v a t i o n up o n p r o p a g a t i o n i n quan tu m o p t i c s in
order
c a v ity
to
(or e q u iv a le n tly ,
to p re s e rv e th e p e rio d ic -b o u n d a ry c o n d itio n s o f th e
of q u a n tiz a tio n ),
q u a n tiz a tio n c a v ity ,
is
th e
v o lu m e
re la te d to th e
of th e
re fra c te d -w a v e
fre e -fie ld
q u a n tiz a tio n
v o lu m e by
V as d isc u sse d
in S e c tio n 2.
V o n
W ith t h i s
(3 .1 0 ) q u a n t i z a t i o n v o lu m e,
th e
183
I ABRAM
r e fra c te d -w a v e H a m ilto n ia n i s
(3 .1 1 ) and i s th e
id e n tic a l
i n f o r m a n d w i t h t h e sa m e e i g e n f r e q u e n c i e s a s
f r e e - s p a c e H am ilto n ian . T he e q s .
o p e ra to rs
(3 .7 )
p e rm it
r e la tin g th e f r e e - f i e ld us
to
re fra c ted -w a v e b a s is s e t. e le c tric
convert
T hus,
and m a g n e tic f i e l d
a ll
and r e f r a c te d - w a v e o p e ra to rs
su b stitu tin g
in eqs
to
th e
(3 .1 )
th e
o p e ra to rs can be e x p re sse d in th e
re fra c ted -w a v e b a s is s e t as
(3 .1 2 a )
and
(3 .1 2 b ) j T he r e f r a c t e d - w a v e momentum may b e o b t a i n e d
in a
sim ila r
fa sh io n as
(3 .1 3 )
J w h e r e t h e j - t h momentum e i g n e n v a l u e
(3 .1 4 )
K. = D j i s t h e w a v e v e c to r o f t h e j - t h r e f r a c t e d w ave. P ro p a g a tio n o f th e e le c tro m a g n e tic m ed iu m ,
fie ld
in sid e a
i s c o m p le te ly d e s c rib e d th ro u g h th e H a m ilto n ia n
a n d momentum ( 3 . 1 3 )
o p e ra to rs,
lin e a r (3 .1 1 )
w hich p e r m i t t h e c a l c u l a t i o n
of
184
PR O P A G A T IO N OF N O N C LA SSIC A L LIGHT
th e
te m p o ra l
to rs
and a l l
e v o lu tio n and s p a t i a l p r o g r e s s io n o f a l l sta te s
tro m a g n e tic
( c l a s s i c a l and n o n c l a s s i c a l )
fie ld .
In p a r tic u la r ,
e v o lu tio n o f any f i e l d tio n .
F o r ex am p le,
o p e ra to r,
eq.
(3 .1 1 )
opera
of th e
e le c
g iv e s th e tim e -
th ro u g h th e H e ise n b e rg
equa
fo r th e a n n ih ila tio n o p e ra to r of th e j - t h
mode we h a v e ,
3 k,
A
= i [ # , k j] = In te g ra tin g
t h i s e q u a tio n o f m o tio n ,
(3.15a)
we o b t a i n
atim e -h a rm o n ic
e v o l u t i o n a t f r e q u e n c y UK f o r t h e r e f r a c t e d w a v e s ,
= e 1^ t
k. (t) id e n tic a l to
th a t
S im ila rly ,
th e
th ro u g h eq.
(2 .2 )
sp a tia l Thus,
of
= k .(0 )e
th e
sp a tia l
c o rre sp o n d in g
^
(3 .1 5 b )
fre e -fie ld
p ro g re ssio n of th e
fie ld
is
m odes. o b ta in e d
w h ich c a n be c o n s i d e r e d a s a H e i s e n b e r g - l i k e
e q u a tio n
of
fo r th e j - t h
m o tio n
i n v o l i n g t h e momentum o p e r a t o r .
p l a n e w a v e i n s i d e a l i n e a r m e d iu m , ^ §. = -i
In te g ra tin g
t h i s e q u a t i o n we o b t a i n ,
th e o s c illa to r y
(z)
fa c to r
of
Thus, a its
lin e a r
n
freq u en cy
(w a v e v e c to r)
o p tic s,
(3 .1 6 b ) ren o rm a liz ed
by
w ith r e s p e c t to th e c o rre sp o n d in g f r e e - f i e l d
as in c la s s ic a l o p tic s , d ie le c tric ,
its
w hen a l i g h t be a m t r a v e r s e s
fre q u e n c y rem ain s unchanged, w h ile th e
re fra c tiv e
of
th e
The t i m e - e v o l u t i o n o f t h e r e f r a c t e d w av es i s
th u s
i n d e p e n d e n t o f t h e m edium o f p r o p a g a t i o n : rem a in s
in c l a s s i c a l
a s e x p e c t e d from c l a s s i c a l c o n s i d e r a t i o n s .
w a v e v e c to r i s m o d ifie d by
d ie le c tric .
as
(3 .1 6 a )
= e " i 6 z k j e i 6 z = § j e lK j Z
how ever w ith a s p a t i a l a
= iK jlj
s p a t i a l p r o g r e s s i o n o f a r e f r a c t e d wave,
|
w a v e v e c to r,
[6 ,k j]
tim e -h a rm o n ic
at
th e
in d ex
a t i m e - h a r m o n i c w ave
s am e f r e q u e n c y .
On t h e o t h e r
185
I ABRAM
hand,
its
s p a tia l p ro g re ssio n i s
quency o f o s c i l l a t i o n
a lte re d ,
as i t s
(o r e q u iv a le n tly i t s
s p a tia l
fre
phase v e lo c ity )
are
changed by r e f r a c t i o n .
4 . REFLECTION IN PROPAGATIVE QUANTUM OPTICS In c l a s s i c a l o p tic s , in te rfa c e
b e tw een
w hen a l i g h t beam i s
em pty
space
b e t w e e n tw o r e f r a c t i v e m e d i a ) is
tra n sm itte d
forw ard
and
it
w aves,
in c id e n t,
a r e f r a c t i v e m edium ( o r
sp lits
in to
th e
r e f l e c t e d b a c k i n t o e m p ty s p a c e .
i n t o tw o
beam s:
m edium w h i l e t h e o t h e r i s
t a n g e n t i a l c o m p o n en ts o f t h e e l e c t r i c
in s u r e s t h a t th e
and m a g n e tic f i e l d s
d isc o n tin u ity
in r e f r a c tiv e
c o rre sp o n d s t o an a b ru p t momentum
o p e ra to rs
change
o p tic s,
in d ex a c r o s s th e
of
th e
upon p r o p a g a tio n ,
energy
"sudden ap p ro x im a tio n " a t th e
i n c i d e n t on a r e f r a c t i v e c o rre sp o n d in g
g e ts
face
of th e e le c tr ic
Thus,
of
re la te s,
th ro u g h
(in c id e n t) to rs. to rs,
on th e
f o r e x am p le i n
re fra c ted -w a v e
eq u a tio n s)
th e e le c tr ic
th e o p e ra to rs
re la te d
and
e ith e r eq.
o p e ra to rs
and u s in g e q s.
(3 .7 )
th a t
th e
acro ss th e fie ld
th e
th e
con in te r
to
th e
o p e ra to rs.
in te rfa c e (3 .5 )
are w h ich
fo rw a rd -g o in g
t h e m edium t o t h e (re fle c te d )
th e
A lte rn a tiv e ly ,
tra n sfo rm a tio n
(3 .7 a ),
in
of
o n to
co rresp o n d s
m ag n etic sid e
u n ita ry
and b a c k w a rd -g o in g
W ritin g eqs
th e
fo rw a rd -g o in g
fre e -fie ld
opera
i n t e r m s o f t h e i n c i d e n t w ave o p e r a
(3 .8 ),
we o b t a i n t h e o p e r a t o r
o f th e F re s n e l fo rm u la s o f c l a s s i c a l o p t i c s , re fle c tio n
A f r e e - f i e l d w av e
assu m in g
and m a g n e tic f i e l d s
( i m p o s e d b y t h e M a x w e ll
c o n tin u ity
and in v o k in g
p ro je c te d
w a v e s o f t h e r e f r a c t i v e m e d iu m .
t h e p r o j e c t i o n may b e c a l c u l a t e d b y tin u ity
or
a n d t h u s may b e t r e a t e d
in te rfa c e :
in te rfa c e
th e
in te rfa c e
d e n sity
by e x am in in g t h e s p a t i a l p r o g r e s s i o n o f t h e f i e l d th e
are
in te rfa c e .
W i t h i n t h e f o r m a lis m o f p r o p a g a t i v e q u an tu m abrupt
one
The c o e x i s t e n c e o f t h e t h r e e
t r a n s m i t t e d and r e f l e c t e d ,
c o n tin u o u s a c ro s s th e
i n c i d e n t on t h e
and r e f r a c t i o n o f l i g h t :
e q u iv a le n t
th a t d e sc rib e th e
186
PR O P A G A T IO N OF N O N C L A SSIC A L LIGH T
j
n+ 1
j
n+ 1
A+ _ 2 / n a + j n+1 j
(4 .1 b )
-j
n-1 a n+1 - j
(4 .1 c )
A _ 2Jn a n - 1 a+ j n+1 j " n+1 -j
( 4 . Id)
T h e s e e q u a t i o n s d e s c r i b e q u an tu m m e c h a n i c a l l y t h e re fle c tio n
and
n a tu ra lly th e c ie n ts
as
re fra c tio n
fa m ilia r
at
th e
in te rfa c e ,
tra n sm issio n
re su ltin g d ire c tly
tio n th a t re la te s
an
and
re fle c tio n
i n c i d e n t on t h e
(4 .1 )
c o e ffi
and r e f r a c t e d w aves. p la n e
wave
1963)
Io is
of
from t h e B o g o liu b o v t r a n s f o r m a
fre e -fie ld
When a q u a s i - c l a s s i c a l w a v e p a c k e t o f t h e j - t h (G la u b e r,
p rocess
and g iv e q u ite
and
> = e
in te rfa c e ,
sp lits
in to
-1 it
-1 |o >
(4 .2 )
i s p r o je c te d a c c o rd in g to eq.
a t r a n s m i t t e d and a r e f l e c t e d
q u a si-
c l a s s i c a l w av ep ack ets:
- i * j ^ S ( f t + §+) = e
+6 + ) e
|0 >
w hose e x p e c t a t i o n v a l u e s o f t h e e l e c t r i c
et
-
K?T EI
"t - S I are th u s
HI
and m a g n e tic f i e l d s
ER -
S IT EI
“r -
-
(4 .4 a ,b )
S IT " l
i d e n t i c a l to th e c o rre sp o n d in g c l a s s i c a l sa tisfy
e le c tric
th e
c o n tin u ity
co n d itio n
(4 .3 )
fo r
(4 .5 a ,b ) e x p re ssio n s
and
th e ta n g e n tia l
and m a g n e tic f i e l d s .
The quan tu m m e c h a n ic a l F r e s n e l f o r m u la s
(4 .1 )
can
a lso
I ABRAM
d e s c rib e th e b e h a v io r o f re fra c tiv e
in te rfa c e .
187
n o n c la ssic a l
Thus,
lig h t
in c id e n t
a sin g le -p h o to n s t a t e
f r e e - f i e l d w ave, w h ich ca n b e d e s c r i b e d a s an th e e le c tr ic
fie ld
on
of th e j - t h
o sc illa tio n
w i l l be tra n sfo rm e d a t a d i e l e c t r i c (4 .1 ),
< v s j )|0 > + S i
to g iv e an o s c i l l a t i o n r e f r a c t e d w ave, Thus, F resn el
(4.6) in te rfa c e ,
a c co rd in g
to
as
M
j-th
of
o f t h e form
(6j + fit) |0>
eqs.
a
w ith in
fo rm u la s
i o>
of a sin g le -p h o to n s t a t e
o r in th e - j th is
(4 .1 )
(re fle c te d )
fo rm alism ,
th e
e i t h e r in
th e
f r e e - f i e l d wave.
q u an tu m
m ech an ical
p e rm it to d e s c rib e b e a m s p littin g
m anner c o m p le te ly a n a lo g o u s t o t h a t o f c l a s s i c a l o p t i c s ,
in a w h ile
a t t h e sa m e t i m e t h e y a c c o u n t f o r t h e q u a n t u m s t a t i s t i c a l p r o p e rtie s
o f b o th th e r e f l e c t e d
and th e r e f r a c t e d w aves.
5. CONCLUSIONS I n q u an tu m o p t i c s a g e n e r a l s t a t e fie ld
c o n sists
fie ld For
of
a
fu n c tio n
of th e
e le c tro m a g n e tic
of th e e le c tr ic
and m a g n e tic
o p e r a t o r s a p p l i e d o n to t h e e l e c t r o m a g n e t i c vacuum s t a t e . sta te s
th a t
m ag n etic f i e l d correponds
to
have a c l a s s i c a l a n a lo g u e ,
o p e ra to rs p re se n t th e
an
th e e le c tr ic
e x p e c ta tio n
co m p lex o s c i l l a t i n g
c la ssic a l
p ro p a g a tio n o f such s t a t e s th ro u g h a p o la r iz a b le th u s
be
d e sc rib e d
v a lu e
and th a t
fie ld .
T he
m e d iu m ,
may
t h r o u g h t h e m a c r o s c o p i c M axw ell e q u a t i o n s
w h ich c a n b e u s e d t o c a l c u l a t e b o th t h e tim e e v o l u t i o n and t h e s p a tia l p ro g re ssio n of th e se e x p e c ta tio n v a lu e s. s i c a l l i g h t on t h e o t h e r h a n d , e le c tric
and
th e s t a t e g a tio n
of
m ag n etic
of th e f ie ld . such s t a t e s ,
For n o n c la s
th e e x p e c ta tio n v a lu e s
of
th e
f i e l d s do n o t c h a r a c t e r i z e c o m p le te ly T hus, it
in o rd e r to d e s c rib e th e p ro p a
i s n e c e s s a ry to c a l c u l a t e b o th th e
188
PR O P A G A T IO N OF N O N C L A SSIC A L LIGHT
te m p o ra l e v o lu tio n and th e s p a t i a l p r o g r e s s io n
of
th e
fie ld
o p e ra to rs. P r o p a g a t i v e q u an tu m o p t i c s tio n a l
th eo ry
of
is
an e x te n s io n o f th e conven
quan tu m o p t i c s t h a t r e l i e s
o p e ra to r o f th e e le c tro m a g n e tic f i e l d , th e
sp a tia l
p ro g re ssio n
of lig h t,
o n t h e momentum
fo r th e c a lc u la tio n
i n a d d i t i o n t o t h e H a m il
to n ia n t h a t d e s c r ib e s th e te m p o ra l e v o lu tio n . th is
th eo ry t r e a t s
fie ld
p h e n o m e n o lo g ic a lly
T h e m edium
th ro u g h
its
c o n ta in s,
at
te n ts.
The p r e s e r v a t i o n o f
its
m e d ia , as
is
fie ld
th e o ry
M a x w e ll
o p e ra to r is
p e rio d ic -b o u n d a ry
th a t as
lig h t
p ro p a g a te s
eq u a tio n s:
o p t i c s and re d u c e s t o
it
by
an a lo g y
w ith
g iv e s r i s e
fie ld ,
th e
(w ith
re n o rm a liz e d w av ev ecto r)
A t t h e sam e
th e o ry
c la ssic a l
p ropaga
th e
o p tic s.
T he
th e
tem
s p a tia l p rogres in
th e
m edium
an d t o a r e f l e c t e d wave a t t h e c o e ffic ie n ts,
a re g iv e n w ith in t h i s
d i r e c t l y by th e tr a n s f o r m a tio n t h a t
d ia g o n a liz e s
fam
fo rm a lism
th e
energy
The d i r e c t c o r r e s p o n d a n c e o f t h e r e s u l t s
t h e o r y w ith t h e w ell-k n o w n r e s u l t s o f c l a s s i c a l
o p tic s d em o n stra te s th e v a l i d i t y a p p ly t h i s
equa
g iv e s
affect
b o t h t o a r e f r a c t e d w ave
c la s s ic a l o p tic s,
M ax w ell
(c la ssic a l)
The t r a n s m i s s i o n and r e f l e c t i o n
d e n sity o p e ra to r.
d iffe re n t
i s t h e d i r e c t q u an tu m
b u t changes i t s
it
from
in con
in th e a p p ro p ria te lim it.
sio n :
in te rfa c e .
c o n d itio n s
d ia g o n a liz a tio n o f th e energy d e n s ity
e q u iv a le n t to th e s o lu tio n of
e x p e c te d
m a te ria l con
in c l a s s i c a l o p tic s .
i n d u c e d p o l a r i z a t i o n o f t h e m edium d o e s n o t
us to
is
i n c o r p o r a t e s a q u an tu m m e c h a n i c a l v e r s i o n o f
p o ra l e v o lu tio n of th e
th is
A key
fie ld
th ro u g h
When a p p l i e d t o a l i n e a r m e d iu m , t h i s
ilia r
q u a si-
th e e le c tro m a g n e tic energy i s
T h u s , p r o p a g a t i v e q u an tu m o p t i c s
re su lts
th e
of its
m e c h a n ic al e q u iv a le n t o f th e c o n v e n tio n a l tiv e
in tro d u c e d th a t
irre sp e c tiv e
is th e case a lso
th is
tio n s.
b o u n d a rie s
in s u re s
serv ed in th e
th e
is
i n s u c h a w ay t h a t p e r i o d i c c o n d i t i o n s a r e
obeyed
tim e ,
way,
i n a c a v i t y w hose p h y s i c a l d i m e n s i o n s d e p e n d on t h e
m e d iu m i t
m e d ia ,
th is
in d u ced p o l a r i z a t i o n .
f e a t u r e o f p r o p a g a t i v e q u an tu m o p t i c s
a ll
In
th e s p a t i a l and te m p o ra l c o o r d in a te s o f th e
o n t h e sa m e f o o t i n g .
q u a n tiz e d
of
of th is
approach,
of
(lin e a r)
and p e r m its
fo rm a lism t o th e d e s c r i p t i o n o f t h e e v o lu tio n
189
I ABRAM
o f t h e q u an tu m s t a t i s t i c s
of n o n c la ssic a l s ta te s
o f l i g h t upon
t h e i r p r o p a g a tio n i n l i n e a r o r n o n l i n e a r m ed ia.
REFERENCES A bram ,
I.
1987,
P h y s. Rev. A 35,
A b ra m ,
I.
1988,
P h y s . R e v . A,
G lau b er,
R .J.
P e ie rls,
R.
Ed. F.
1963,
4661
in p re s s
P h y s. Rev.
131,
2766
1985, H ig h lig h ts o f C o n d e n se d -M a tte r P h y s ic s , B a ssa n i,
A m s te r d a m , p p .
F . Fumi a n d M .P . T o s i ,
N o rth -H o lla n d ,
237-255
S to le r,
D.
1970, Phys.
Rev. D 1,
3217
S to le r,
D.
1971, Phys.
Rev. D 4,
1 925
Y uen, H .P .
1976, P hys.
Rev. A 13,
Y e e , T .K a n d G u s t a f s o n ,
T .K .
1978,
2 22 6 Phys.
Rev. A 18,
1597
MODELS FOR PHASE-INSENSITIVE QUANTUM AMPLIFIERS
G L M A N D ER, R LOUDON AND T J SH EPH ERD
1.
INTRODUCTION
S tu d y o f t h e p r o p e r t i e s
of lin e a r o p tic a l a m p lifie rs
i s m o ti
v a t e d by b o th ac a d e m ic and t e c h n o l o g i c a l i n t e r e s t i n t h e f i e l d . In pure s c ie n c e ,
e x p e rim e n ta l t e s t s
a t e m easu rin g a p p a r a tu s , s t a g e o f w hich i s in to
th e
o f t h e o r i e s m ust in c o r p o r
l a s t e s s e n t i a l q u a n tu m -m e c h a n ica l
a h ig h -g a in a m p lifie r to b o o st th e
o n e w h i c h may b e r e g a r d e d a s c l a s s i c a l f o r m e a s u r e m e n t
p u rp o ses.
In a p p lie d s c ie n c e ,
a m p lif ie r th e o ry
t o t h e co m m u n icatio n s i n d u s t r y w h e re , to r
sig n a l
is
f o r e x am p le,
of re le v a n c e se m ico n d u c
l a s e r a m p l i f i e r s may h a v e u s e a s r e c e i v e r p r e - a m p l i f i e r s
and n o n - r e g e n e r a t i v e r e p e a t e r s
in o p tic a l
The e x i s t e n c e o f s q u e e z e d l i g h t ,
and i t s
fib re
lin k s.
p o te n tia l use to c a rry
o p t i c a l i n f o r m a t i o n w ith an im proved s i g n a l - t o - n o i s e r a t i o re la tiv e
to
co h eren t l i g h t,
degree of g a in t h a t i s sq u e e z in g , is
p o ssib le
a n e e d t o know t h e
in t h e sy stem b e f o r e t h e
and a l s o o t h e r quan tu m p r o p e r t i e s o f l i g h t s u c h a s
a n tib u n c h in g , la tio n s
h as le d to
are
lo st.
A d ete rm in in g f a c to r in th e s e c a lc u
t h e am ount o f n e c e s s a r i l y - a d d e d q u an tu m n o i s e i n h e r
e n t to t h e sy ste m .
A lth o u g h t h e r e
e x is t p h a se -se n sitiv e
lin e a r
a m p l i f i e r s t h a t add no n o i s e t o one q u a d r a t u r e p h a s e o f t h e fie ld ,
f o r e x am p le, d e g e n e r a te p a r a m e tr ic a m p l i f i e r s ,
in
g e n e r a l t h e s i g n a l - t o - n o i s e r a t i o w i l l be d e g r a d e d by t h e a m p lific a tio n p ro cess,
a n d q u a n t u m s t a t i s t i c a l p r o p e r t i e s may
b e w ashed o u t by t h e ra n d o m iz in g e f f e c t o f t h e n o i s e f i e l d . We c o n c e n t r a t e o u r a t t e n t i o n h e r e on r e v i e w i n g a n d e x t e n d i n g t h e w o r k on t h e p a r t i c u l a r c l a s s o f l i n e a r a m p l i f i e r s know n a s in v e rte d p o p u la tio n a m p lifie rs .
We d i s c u s s f i r s t
th e
191
G L M A N D E R , R L O U D O N A N D T J SH E P H E R D
f u n d a m e n ta l r e q u i r e m e n t s dem anded by q u an tu m m e c h a n ic s from any t h e o r y o f l i n e a r a m p l i f i e r s ,
and th e n d e s c r i b e t h e b a s i c
model f o r t h e a to m ic a m p l i f i e r i n a c lo s e d c a v i t y . t h e e x te n s io n o f t h i s m odel t o a llo w e x p l i c i t lin g of th e in te r n a l c a v ity fin a lly tig a te fie ld 2.
fie ld
We c o n s i d e r
in c o h e re n t coup
to e x te rn a l f i e l d s ,
an open sy ste m w ith c o h e r e n t f i e l d
c o u p lin g .
and t r e a t We i n v e s
i n p a r t i c u l a r t h e c o n d i t i o n s u n d e r w h ich t h e a m p l i f i e d r e t a i n s q u an tu m p r o p e r t i e s
such as sq u e e z in g .
FUNDAMENTAL REQUIREMENTS OF LINEAR AMPLIFIER THEORY
S i n c e no s y s t e m c a n b e t r u l y
iso la te d
from t h e r e s t o f t h e
u n i v e r s e , b e i n g s u r r o u n d e d a t t h e v e r y l e a s t by a vacuum f i e l d , t h e r e w i l l alw ay s be f l u c t u a t i o n s due t o
its
and d i s s i p a t i o n
in te r a c ti o n w ith th e c o n ta in in g
en v iro n m en t.
L ax
(1 9 6 6 )
su g g e ste d t h a t th e r e s e r v o i r c o u ld be
e l i m i n a t e d from t h e c a l c u l a t i o n s , tio n s
p r o v i d e d t h a t t h e m ean e q u a
o f m o tio n i n c o r p o r a t e t h e r e s e r v o i r - i n d u c e d
sh ifts
and d i s s i p a t i o n , source is
t o add
c le a rly
N a iv e ly , andi d e a l l y ,
o u tp u t f i e l d
out
one
T he e f f e c t o f t h i s
and f o r c e s an a m p l i f i e r
o f n o i s e t o anys i g n a l i t m ig h t w ish t o w r i t e ,
d e s c r i b e d by t h e a n n i h i l a t i o n
d e r i v e d from an i n p u t f i e l d a
added.
irre m o v a b le ,
a c e r t a i n minim um a m o u n t
cesses.
pro f o r an
o p e r a t o r a Qut and
d e s c r i b e d by a i n ;
.= u a . m
(2 .1 )
s o t h a t t h e mean p h o t o n n u m b e r a m p l i f i c a t i o n = IuI 2 < n . > out 1 1 m
and t h e power g a in
(2 .2 )
(lo ss)
c u s s io n ab o v e, and a l s o
is
G = |u |2.
H ow ever, fro m t h e d i s
from t h e u n i t a r i t y
re q u ire m e n ts of
q u a n t u m m e c h a n i c s , we m u s t w r i t e , a
out
,= ua. m
w here F i s of th e
+F
(2 .3 )
t h e n o i s e o p e r a t o r , w hich c an be e x p r e s s e d
i n t e r n a l a m p l i f i e r m odes.
in term s
The i n p u t and o u t p u t f i e l d s
th e n obey t h e n e c e s s a r y co m m u tatio n r e l a t i o n s ,
v iz
192
M O D ELS FO R P H A SE -IN SE N SIT IV E Q U A N T U M A M PLIFIERS
[a . , a + .] L o u t' out if
= [a. ,a + ] = 1 L m ' m J
(2 .4 )
th e n o is e o p e ra to r s a t i s f i e s [ F , F +] = 1 - G
(2 .5 )
Now e x p e c t a t i o n v a l u e s o f h e r m i t e a n s q u a r e s o f o p e r a t o r s m u s t be r e a l and n o n - n e g a t i v e ,
so t h a t
_> 0 >_ 0 and
(5)
and
(6)
(2 .6 ) t o g e t h e r g iv e t h e fu n d a m e n ta l th e o re m f o r
p h a se -in se n sitiv e
lin e a r am p lifie rs
i n C a v e s 1 (1 9 8 2 )
form ,
>_ G - 1 T h is i s
(2 .7 )
t h e m a th e m a tic a l e x p r e s s i o n o f t h e f a c t t h a t su c h an
a m p lify in g d e v ic e m ust n e c e s s a r i l y d eg rad e th e s i g n a l : added n o is e o n ly v a n is h e s f o r G = 1, I f th e boson f i e l d s
carry
i.e .
th e
a p a s s i v e com p o n en t.
co h eren t in fo rm a tio n th en th e expec
t a t i o n v a l u e s < a j_n > ^ < a o u t > a r e n o n “ v an :*-s h ^ n 9 • D e f i n i n g t h e i r f l u c t u a t i o n s as t h e s y m m e tric a l c o r r e l a t i o n f u n c t i o n s , f o r any boson o p e r a to r b < | Ab | 2 > = ^ < b ^ b + bb^> - w hich g i v e s (3)
< | Ab | 2 > >_ %.
(2 .8 )
From t h e b a s i c a m p l i f i e r e q u a t i o n
th e n < IAa | 2 > = G < |A a . I 2 > + < I A F l2 > 1 o u t1 1 m 1 1 1
U sin g t h e e q u i v a l e n t n o i s e f a c t o r o f C aves A
= > %(i - h G
-
“'
(2 .9 ) (1982),
(2 .io)
G
we h a v e , < | Aa J 2 > > G(h + h) 1 o u t1 — T h i s show s e x p l i c i t l y tio n
o p e ra to rs,
fo r sy m m e tric a lly o rd e re d f l u c tu a
th e a m p lific a tio n
t o one h a l f p h o to n a t in p u t flu c tu a tio n o p e ra to rs
th a t,
(2 .1 1 )
is
th e
p ro c e ss adds n o is e e q u iv a le n t
a m p l i f i e r i n p u t , w h e r e t h e m inim um
a ls o one h a l f p h o to n .
If th e flu c tu a tio n
a re n o rm ally o rd e re d , th e n th e e q u iv a le n t n o is e a t
193
G L M A N D E R , R L O U D O N A N D T J SH E P H E R D
th e a m p lif ie r in p u t i s
one p h o to n .
Any v i a b l e t h e o r y o f l i n e a r a m p l i f i e r s m u s t s a t i s f y m e n ta l th eo rem .
th e funda
U s u a l l y one w orks w i t h t h e q u an tu m d y n a m ic s
of th e a m p lify in g d e g re e s o f freedom ,
so t h a t u se o f t h e e f f e c
t i v e q uantum n o i s e s o u r c e s can be a v o i d e d .
T h is a ls o has th e
a d v a n t a g e t h a t t h e a m p l i f i e r p a r a m e t e r s may b e r e l a t e d d i r e c t l y t o t h e p h y s i c a l m e c h a n is m o f t h e p r o c e s s . in v e rte d p o p u la tio n a m p lif ie r s ,
F or e x am p le,
th e g a in i s
in term s o f th e a to m ic i n v e r s i o n .
d ire c tly
in
e x p re ssib le
The a m p l i f i e r a d d e d n o i s e
may a l s o b e e x p r e s s e d i n t e r m s o f t h e a t o m i c p r o p e r t i e s . p re c ise ly th is
sh o u ld be i n t r i n s i c a l l y
irre v e rsib le
When t h e a to m d e c a y s ,
subsequent lo ss
co m p risin g th e a m p lif ie r .
t h e r e m a i n i n g a to m s f o r m a p h o t o n s i n k .
I f a p h o to n c o h e r e n t w ith t h e f i e l d b ed by a n o t h e r a to m ,
That a m p lific a tio n
can b e s e e n by c o n s i d e r i n g
a n y p a r t i c u l a r a to m i n t h e c o l l e c t i o n
its
is
n o i s e t h a t b r e a k s t h e t i m e s y m m e tr y i n h e r e n t i n
t h e b a s i c e q u a t i o n s o f q uantum m e c h a n ic s .
random , t h i s
It
is
e m itte d and th e n a b s o r
i t may b e r e r a d i a t e d from t h e beam .
process is
c le a rly
sp o n ta n e o u s ly , w ith
S p o n tan eo u s e m is s io n b e in g
not re v e rsib le :
a tte m p t t o ru n t h e a m p l i f i e r backw ards
in d e e d any
(G lau b er,
1986), t h a t i s ,
t o a t t e n u a t e t h e beam , w o u ld o n l y r e s u l t i n t h e f u r t h e r d e g ra d a tio n of th e s ig n a l. 3.
THE CLOSED SYSTEM
T h i s m o d e l w as p r o p o s e d i n
1963 b y G o r d o n , W a l k e r a n d L o u i s e l l
(1 9 6 3 ) , a n d h a s s i n c e b e e n g r e a t l y
used in in v e s tig a tio n of
t h e e f f e c t o f o p t i c a l a m p l i f i c a t i o n on t h e p r o p e r t i e s o f q u a n tum l i g h t f i e l d s . its
We w i l l t h e r e f o r e
l o o k i n some d e t a i l
at
m ain f e a t u r e s .
3 .1
M odel
The s y s t e m i s
com posed o f a s i n g l e
r a d i a t i o n mode i n a l o s s l e s s
c a v i t y c o u p le d t o a l a r g e num ber o f n e a r l y (F ig .
1).
E a c h a to m i s
in d e p e n d e n t atom s
assum ed t o have a f i n i t e
s e t o f e q u a lly spaced energy l e v e l s ,
and i t
is
supposed t h a t
t h e r e a r e s u f f i c i e n t l y many a to m s w i t h t r a n s i t i o n c lo se to th e f i e l d
or in fin ite fre q u e n c ies
f r e q u e n c y t h a t t h e a to m d e n s i t y o f s t a t e s
may b e t a k e n a s c o n t i n u o u s .
O n ly a t o m s w i t h t r a n s i t i o n
194
M O D ELS FO R P H A SE -IN SE N SIT IV E Q U A N T U M AM PLIFIERS
fre q u e n c ies n ear th e f ie ld to
th e f ie ld .
F u rth e r,
co u p led to a h e a t b a th , b u tio n .
fre q u e n c y c o u p le w ith any s t r e n g t h
t h e ato m s a r e t a k e n t o b e w e a k ly so t h a t t h e y h a v e a B o ltzm an n d i s t r i
The s t a t i s t i c a l d i s t r i b u t i o n
by v i r t u e o f t h e i r
o f t h e ato m s i s
l a r g e n u m b e r a n d w e ak c o u p l i n g t o
to be c o n s ta n t o v e r th e i n t e r a c t i o n serv e to ensure l i n e a r i t y ,
i.e .
tim e .
ta k e n , th e f ie ld ,
T hese a ssu m p tio n s
n o n - s a tu r a tio n o f th e a m p lif ie r .
Closed Cavity N 2 - level atoms 3 .2 If
N, + N2 = N
A n a ly sis a (t)
is
and a j (t)
th e a n n ih ila tio n o p e ra to r o f th e is
th e lo w e rin g o p e r a to r f o r th e
H a m ilto n ia n o f th e
in te rn a l jth
fie ld ,
a to m t h e n t h e
s y s te m h a s t h e u s u a l form i n t h e r o t a t i n g -
w ave a p p r o x i m a t i o n ,
- +*
N
w here
is
z
N
+ h u I ^oo”i . o"i . + u7 3 3 j= i j =l
H = ftoo o a a
*+*
th e ato m ic i n v e r s i o n o p e r a t o r ,
a to m - f ie ld co u p lin g c o n s ta n t.
- -* +
fiKH. (ai . a + a “i. a ) and
is
(3 .1 ) th e
The r e l e v a n t c o m m u ta tio n r e l a
t i o n s g iv e th e H e ise n b e rg e q u a tio n s o f m o tio n f o r th e
fie ld
and a to m ic o p e r a t o r s as da d T" t = -loo o a
i
3
da . 1
-■-- v *
dt
w here i t
=
is
.
*3 :
- l o o . a .
7
k
• 3
.a . 3 3
, .
+
-z j
j
assum ed t h a t th e p o p u l a t i o n i n v e r s i o n
m a in ta in e d in o rd e r to o b ta in in te re ste d
( 3.2)
i K . < a . > a
in
is
lin e a riz e d e q u a tio n s.
th e e v o lu tio n o f th e
cavity
e x te rn a lly We a r e
mode, and t h e
195
G L M A N D E R , R L O U D O N A N D T J SH E P H E R D
H e i s e n b e r g e q u a t i o n may b e s o l v e d i n t h e W i g n e r - W e i s s t t o p f a p p ro x im atio n to g iv e th e r e s u l t a (t)
= u (t)a (0 )
Il v^. V . ( t )) S a. j= l 1 J
(3 .3 )
+ (N2- N 1 ) y Lt ]
(3 .4 )
+ +
w ith u (t ) = e x p [ - i a > Qt
YL = 5 = P ( G - l )
j
3
3
(3 .5 )
3
w h e r e now t h e g a i n G i s d e f i n e d by G = | u (t ) | 2 = e x p [ 2 (N2 ~N1 ) y Lt ] and P i s
th e p o p u la tio n
(3 .6 )
fa c to r
N2
p =T v v
(3-7)
The r e s u l t o b t a i n e d i s a ( t = 0)
tim e -v a ry in g ,
a s t h e i n p u t mode a n d a ( t )
so one c h o o s e s
to id e n tify
a s t h e o u t p u t mode
o f th e
sy ste m , a in = a(0) ^out = a (t)
(3 -8>
The i n p u t - o u t p u t r e l a t i o n sta n d a rd r e s u l t
(2 .3 )
t h a t th e sy stem
w i l l o n ly
isfa c to ry
(3 .3 )
th e n ta k e s th e
form o f t h e
fo r p h a se -in se n sitiv e a m p lifie rs.
c o n d itio n th a t
N o tic e
show g a i n u n d e r t h e p h y s i c a l l y
sa t
> N ^.
U n i t a r i t y i s o b e y e d o n l y on t h e a v e r a g e ,
J a tx.o m s = 1 c o n s is te n t w ith th e
lin e a riz in g
(3 .9 ) a p p ro x im a tio n .
196
M O D ELS FO R P H A SE -IN SE N SIT IV E Q U A N T U M A M PLIFIERS
T h e p r o p e r t i e s o f t h e o u t p u t f i e l d may b e e v a l u a t e d i n t e r m s of th o se o f th e
in p u t u sin g th e fo llo w in g r e l a t i o n ,
:r -
, r+ s
< [aof u t J] [La o u .t]J =
r F q=0
(J') O. t k .a 7 e ^ l - J - 3 -------------j oo . — co + i r 3 0
Wi th r = %( y j + y 2 ) -
(5.5)
(n2 - n 1 ) y l
as th e in v erse c a v ity
life tim e ,
a n d yt g i v e n b y L
(14).
R ead in g
f r o m l e f t t o r i g h t , we h a v e t h e o u t p u t f i e l d e x p r e s s e d i n t e r m s of re fle c te d
in p u t f i e l d ,
(sp o n ta n e o u s e m issio n )
a m p lifie d in p u t f ie ld s
and th e n o i s e
field.
As w i t h t h e f i r s t m o d e l , u n i t a r i t y
is
o b e y e d on t h e a v e r a g e ,