Hadron and Nuclear Physics with Electromagnetic Probes
Hadron and Nuclear Physics with Electromagnetic Probes
ThisPageIntolyLfBk This Page Intentionally Left Blank
Hadron and Nuclear
Physics with Electromagnetic
Probes Proceedings of the Second KEK-Tanashi International Symposium Tanashi,Tokyo, October 25-27, 1999
edited by
K. Maruyama Center for Nuclear Study, University ofTokyo Tokyo, Japan
H. Okuno Institute for Nuclear and Particle Physics KEK Tokyo, japan
2000 ELSEVIER Amsterdam - Lausanne - NewYork - Oxford - Shannon - Singapore -Tokyo
ELSEVIER SCIENCE B.V. Sara Burgerhartstraat 25 P.O. Box 21 1, 1000 AE Amsterdam, The Netherlands
02000 Elsevier Science B.V. All rights reserved T h i s w o r k i s protected under copyright
by Elsevier Science, a n d the f o l l o w i n g terms and conditions apply t o its use:
Photocopylng Smplc photocop~rbof \ingle chapter, ,may he !made for perwnal use as allowed by natlonal copyr~ghtlaws Permlcsion o f the Puhlliher and payment of a fee 15 retlu~rcd lor all olher phillrwopy~n~, lncludnrlg r r ~ u l t ~or~ ry\~erna~lc lr cnpylng, copying for advenivng or promollonal porpo\e\. re\alr, and all forms of document dellvery Specla1 rate\ .ire d\a~labletor educa~jonal~nrl~totton* that wl\h ro make photocopies for non-profit cducatlonal classroom u5e.
OX5 IDX, UK: phone: ( 4 4 ) 1865 8.13830, fax ( 4 4 ) 1865 853333. r-nidil. perm~~~~onrOelsev~er.co.uk You may al\o contacr Global Rights dlrectly through Elsev~er'shome page (http://www.clscviernl), Perm~\\~ons nlay he \ought directly from Elwvler Science Global Rlghtr Department. PO Box 800. Oxford b) selecting 'Obtalnlng Perm~rsion\'.
I n thc IISA. urrr, ma) deal pernmr\lon* and mahe payment5 thrnugh the Copyr~gh~ Clearance Center. Inc . 222 Ro\ewuod Dnue. Danvers, MA 01923. USA; phone- (V78) 7SO84X). f i ~ r (9781 750-17-1-1. and ~n ~ h IrI K through Ihe Copyright Llcen\lng Agency Rapld Clearance Serblce (CLARCSI. 90 Tottenham Court Road. London W I P ULP. L K. phone: (4) 201 63 1 5555. fdl; (-1 207 631 5500. Other countries may have a local reprograph~crlghls agellcy for payments. Dcr~vatlreWork* Tdhle, of content, may he reproduced for Internal clrrulatlon. but permlwon of Elsevler Sclrncr 1s requued fnr external resale or dl\tr~but~on of such material. Penn~sstonof the Publisher I\ required for all other dcrlvat~veworks. mclud~ngcompllat~on~ and t n m s i a t t ~ ~ ~ s Elrclrur~lcStorage or l'\age l'crm~ss~onof the Publl'hrr I\ requlred to store or use r l c c t r u n ~ c a l.in) l ~ matecral contamed In t h work, ~ including any chapter or p;ln uf
;I
clldpter
Ehccpr ar autl~ncdaho\c, no pan of this work ma! he reproduced. sored in a re~r~e\nl \y*trm or t~ansmlttedIn any form or by any means. electronic, ntcchanlcal. photocopy~ng.record~npor othrrul\e. wlthout prlor wrltten permlsuon of the Puhllsher. Addreu permlslons reque\t\ to: El\evter Clohal Rlght, Department. at the mall. fax and e-mail ;~ddrr\\ec noted above. Yotjcr Uo r r i p o n \ ~ h l l ~~r l ya\\u~nrdhy the Puhl15hcl-for any q u r y andlor damage to per\on\ or property a\ a matter of producls I ~ i ~ b ~rlepl~gence l~ly. or otherw~he,or frum any uar oi operdllon ul sny rnelhod*. pruducli. ~n\tructlnn\or ~lledrconldlned In the matertal here~n.Bccauie of rapid advance\ In tllr rned~calsclences. In part~cular.lndependenr \er~ficatlono f diagnose\ and drug dosagei \hould he made.
First edition 2000 Library of Congress Cataloging in Publication Data A catalog record from the Library of Congress has been applied for.
ISBN: 0 444 50539 3 '@(Thepaper used in this publication meets the requirements of ANSUNIS0 239.48-1992 (Permanence of Paper). Printed in The Netherlands.
Preface The second KEK-Tanashi International Symposium on "Hadron and Nuclear Physics with Electromagnetic Probes" was held on the Tanashi campus of KEK, High-Energy Accelerator Research Organization from Oct. 25 to Oct. 27, 1999. 'The new KEK was established in April 1997 by merging three laboratories including Institute for Nuclear Study (INS): the University of Tokyo. INS has a long successful history of hosting international symposia and schools on particle and nuclear physics. The new KEK intended to continue its activity by organizing a series of international symposia. This is the second of the new series. The aim of this Symposium was to discuss recent experimental and theoretical developnients of hadron and nuclear physics, where emphases were placed on the hadron and nucleus studies wit,h electron and photon beams. At KEK-Tanashi, the 1.3-GeV Electron Synchrotron had long been operated for these purposes. In rccent years, the main research areas were photonuclear reactions and meson productions by using the first highduty tagged-photon beam and the TAGX spectrorrieter. Although this field is developing quite rapidly, the synchrotron was closed in 1999 after 37-year operation, and these activities will be continued at new facilities. n'e think it is a good time to discuss the present status and future directions of this field at this occasion. The Symposiuni was attended by 85 physicists and 35 talks were presc11tr:d. This book contains the papers presented in thc scientific program of the Symposium. We would like to thank all the speakers and the participants for their stirnulating talks and active discussions. We are also grateful t,o the members of the international advisory coninlittee and the organizing committee for polishing up the program. We appreciate all the staff members, in particular Mrs. I. Yarnamoto and Dr. K. Niki, for smooth organization. Finally, we would like to thank the Inoue Foundation for Science and the Foundation for Accelerator Science for their financial support.
KEK-Tanashi, March 2000 Editors: Koichi Maruyama and Hideki Okuno
ORGANIZATION International Advisory Committee: L.S. Cardman (Jefferson Laboratory, USA) H. Ejiri (Osaka University, Japan) R. Redwine (MIT, USA) B. Schoch (University of Bonn, Germany), T. Walcher (University of Mainz, Germany) J. Wambach (Julich, Germany) W. Weise (Miinchen, Germany) S . Yamada (KEK, Japan) K. Yazalu (University of Tokyo, Japan) President: S. Sugimoto (KEK-Tanashi) Organizing Committee: Y. Akaishi (KEK-Tanashi) M. Asakawa (Nagoya University) I. Endo (Hiroshima University) H. En'yo (Kyoto University) 0. Hashmoto (Tohoku University) J. Kasagi (Tohoku University) T. Kishimoto (Osaka University) K. Maeda (Tohoku University) K. Maruyama (CNS, University of Tokyo, Co-Chair) T. Motoba (Osaka University of Electric-Communication) H. Okuno (KEK-Tanashi, Co-Chair) T. Oshima (Nagoya University) T.-A. Shibata (Tokyo IT) Y. Sumi (Hiroshima International University) K. Tokushuku (KEK-Tanashi)
Host Institute Institute of Particle and Nuclear Studies, KEK
Sponsors Inoue Foundation Foundation for High-Energy Accelerator Science
ThisPageIntolyLfBk
ThisPageIntolyLfBk This Page Intentionally Left Blank
CONTENTS Preface Opening address S. Sugimoto I. INTRODUCTION TO THE SYMPOSIUM Hadrons and nuclear physics with EM probes - from QCD point of view T. Hatsuda 11. MESONS IN NUCLEAR MEDIUM Vector mesons in medium and dileptons in heavy-ion collisions R. Rapp and J. Wambach p0 mesons in the nucleus
K. Maruyama Nuclear production of 4 meson at KEK H. En'yo, J. Chiba, H. Funahashi, H. Hamagaki, M. leiri, M. Ishino, S. Mihara, T. Mipashita, T. Mumkami, R. Muto, M. Naruki, M. Nomachi, K. Ozawa, O. Sasaki, M. Sekimoto, H. D. Sato, T. Tabaru, K. H. Tanaka, S. Yamuda, S. Kokkaichi and Y . Yoshimura (KEK-PSE325 Collaboration) Lepton pair spectroscopy with HADES at GSI J. Friese (HADES Collaboration) 111. NUCLEON RESONANCES IN NUCLEI AND RELATED TOPICS 61 Sll (1535) resonance in nuclei H. Yamazaki, T. Yorita, T. Kznoshita, T. Okuda, H. Matsui, T. Maruyama, J. Kasagi, T. Suda, K. Itoh, T. Miyakawa, H. Okuno, H. Shimizu, H. Y. Yoshida and T. Kinashi Delta in nuclei excited by hadronic processes J. Chiba The physics of AA exitation G. M. Huber
The neutral pion photoproduction on the proton near threshold Il-T. Cheon and M. T. Jeong Photoneutron cross section measurement on 'Be by means of inverse Compton scattering of laser photons H. Utsunomiya, Y. Yonezau~a,H. Akimune, T. Yamagata, M. Ohta, M. Fujishiro, H. Toyokawa and H. Ohgaki Nuclear disintegration induced by virtual photons at heavy-ion colliders I. A. Pshenichnov
IV. STRANGENESS PHYSICS Kaon photoproduction on nuclei K. Maeda Subthreshold and near threshold K+ meson photoproduction on nuclei E. Ya. Paryev Phenomenological aspects of kaon photoproduction on the nucleon T . Mart, S. Sumowidagdo, C. Bennhold and H. Haberzettl Hyperon polarization in Kaon photoproduction from the deuteron H. Yamamura, K. Miyagawa, T. Mart, C. Bennhold, H. Habeerzettl and W. Glockle Physics of associated strangeness production at ELSA E. Paul Retrospect and prospect of hypernuclear physics 0. Hashimoto Spin-orbit splitting of 13*C H. Kohri, S. Ajirnura, R. E. Chrien, P. M. Eugenio, G. Fkanklin, J. Franz, T. Fukuda, L. Gun, H. Hayakawa, P. Khaustov, T. Kzshzmoto, K . Matsuoka, M. May, S. Minami, Y . Miyake, T. Mori, K. Morikubo, J. Nakano, H. Noumi, H. Outa, K. Paschke, P. Pile, B. Quinn, A. Rusek, E. Saji, A. Sakayuchi, R . Sawafta, Y. Shimizu, M. Sumihama, R . Sutter, T . Tamagawa, H. Tamura, K. Tanida, L. Tang and L. Yuan
V. N-N CORRELATIONS AND FEW-BODY PHYSICS Photodisintegration reactions of 3He and *He at TAGX T . Suda
Photonuclear cross sections of three-nuclcon systcms and the role of threenucleon forces G. Orlun,dini, W. Leidernann, V. D. Efros and E. L. Tomsiak
169
Quasi-deuteron picture for %e and 4He photodisintegration S. Hirenzaki, Y . Umemoto and K . Kume Two-nucleon emission experiments at Mainz P. Grabmayr Quark substructure and isobar effects on deuteron form-factors E. Lomon
VI. NUCLEON STRUCTURE STUDIED BY HIGH-ENERGY ELECTRONS The proton and the photon, who is probing whom in electroproduction? 20 1 A . Levy High-Q2 neutral- and charged-current reactions at HERA K. Nagano Spin structure of the nucleon studied by HERMES Y. Sakemi First double polarization measurements on the way to test the GerasimovDrell-Hearn sum rule W. Meyer (GDH- and A2-Collaboration)
239
VII. NEW FACILITIES Few-body physics at Jefferson Laboratory F. W. Hersman (Hall A and C L A S Collaboration) Experimental test of the K-A relative parity - Use of polarized photon beams at high energies Y. Yamaguchz Nuclear physics experiments with 1.2-GeV STB ring at LNS-Tohoku J. Kasagi
265
Laser electron photon facility at Spring-8 T. Hotta, J. K . Ahn, H.Akimune, Y. Asano, W. C. Chang, S. Date, M. Fujiwara, K. Hicks, K. Imai, T. Iwata, T. Ishikawa, H. Kawai, 2. Y. Kim, T. Kzshimoto, N. Kumagai, S. Makino, T . Matsumura, N. Matsuoka, T. Mibe, M. Miyabe, Y. Miyachi, T . Nakano, M. Nomachi, Y. Ohashi, T. Ooba, H.Ookuma, M. Ooshima, C. Rangacharyulu, A. Sakaguchi, T. Sasaki, D. Seki, H. Shimizu, Y. Sugaya, M. Sumihama,
27 1
T. Tooyarna, H. Toyokawa, A. Wakai, C. W. Wang, S. C. Wang, K Yonehara, T. Yorita and M. Yosoi MUSES project at FUKEN RI beam factory T. Katayama, K. Maruyarna and M. Wakasugi Summary of the symposium H. OAuno Symposium program List of participants
Opening Address
Ladies and Gentlemen: On behalf of the organizers, I would like to express our hearty welcome to all of the participants of the 2nd KEK-Tanashi International Symposium, especially to those from abroad. We are very glad to have you all here, in this Tanashi Campus. Tl-le series of Tanashi Symposium was established in 1998 after the merger of three laboratories including Institute for Nuclear Study of the University of Tokyo, which was commonly known as INS-Tokyo. The INS has a long successful history of hosting a medium-sized international symposium every year on a topics concerning nuclear and particle physics. The new Research Organization, KEK-Tanashi also intends to continue such an activity in organizing a scries of intcmational symposia for the research field. This symposiunl is the 2nd of the new series and the 27th of the previous one. As you know, about 44 years ago Tanashi campus was opened for the first interuniversity research institute for nuclear and particle physics. At, the beginning our Electron Synchrotron was built here as the first high-energy accelerator in Japan. The ES played an essential role in creating high energy physics in this country and provided very important steps toward the birth of KEK-PS and Synchrotron Radiation facilities. Since we are going to move to the Tsukuba site within several months to expand our research activities, we have just closed the ES this summer after 37-year successful operation. The wide research field we have developed a.t the ES will bc taken over by new electron facilities of domestic and foreign laboratories. We think it is a good time to discuss the present status and future directions of t,his field at this occasion. So, the topics of this synlposiunl is focused on Hadron and Nuclear Physics with Electromagnetic Probes.
I hope that this symposium will contribute to further development of this field through active and fruitful discussions. In conclusion, I wish all participants a pleasant stay in Tokyo. Thank you for your attention.
Shojiro Sugimoto President of the 2nd KEK-Tanashi International Symposium
ThisPageIntolyLfBk
ThisPageIntolyLfBk This Page Intentionally Left Blank
I. INTRODUCTION TO THE SYMPOSIUM
ThisPageIntolyLfBk
This Page Intentionally Left Blank
Hadron and Nuclear Physics with ElectromagneticProbes K. Maruyama and H. Okuno (Editors) 2000 Elsevier Science B.V. All rights reserved.
'
Hadrons and nuclear physics with EM probes - from QCD point of view "Phys. Dept., Kyoto Univ., Kyoto 606, Japan After a brief introduction of the current problems in QCD, selected topics on the physics of in-medium hadrons are discussed. They are in-medium QCD sum rules, spectral functions in lattice QCD, and critical fluctuation of the chiral order parameter in dense medium. Special emphasis is put on the QCD spectral function which is the key observable in both theoretical and experimental studies. 1. INTRODUCTION
Quantum Chromodynamics (QCD) is the "theory of everything" in the physics of strong interaction. The QCD action can be written as [l]:
where m being the quark mass, G:, is the field strength tensor of the gluon field A:, and D p is the covariant derivative. This action has been well-established by high energy experiments such as the deepinelastic lepton-hadron scattering (DIS) and e+e- annihilation into hadrons. A key ingredient here is the asymptotic freedom of QCD 121 . . which leads to the logarithmic decrease of the QCD running coupling constant CX,(K) as K (renormalization point the typical scale of a given process) increases:
--
where N, and Nfare the number of color and flavor respectively, and AQcD is the QCD scale parameter which takes a value around 200 MeV. Due to the asymptotic freedom, the interaction among quarks and gluons become weak at high energies and the systematic perturbative analyses with appropriate factorization of the hard and the soft part can be ) high chemical potential carried out. Also, at very high temperature ( K -- T >> A Q c ~and (K p >> AQcD), the small running coupling constant as well as the Debye screening allow us to treat the system as non-interacting gas of quarks and gluons in the lowest order [3]. (Even at extremely high T a,nd p, however, the system still may show some interesting phenomena such as the spatial confinement [4]and the color superconductivity 151.1
-
At low energy scale, non-perturbative effects emerge as a result of the strong non-linear gluon self-interactions. Typical examples are the confinement of quarks and gluons and the dynamical breaking of chiral symmetry, both of which are related to the non-trivial QCD vacuum structure. Although it is still difficult to show analytically how these effects appear from the QCD action eq.(l), they are fundamental phenomena for the existence of mesons and baryons, and also of the atomic nuclei. The current physics issues in QCD may be classified into the following three categories: (1) QCD at extreme conditions such as at high temperature or at high baryon density. This is relevant to the physics of ultra-relativistic heavy-ion collisions at CERN and BNL, and also to the physics of neutron star and quark star [6]. (2) Interplay between the hard physics and soft physics in high energy QCD processes. Any physical process is always a combination of the hard scattering part and the soft part: how to factorize them and how to extract information of the non-perturbative effect are theoretical and experimental challenge.
(3) The properties of hadrons in the hotldense medium. Recent experimental developments with hadron-nucleus and nucleus-nucleus collisions give us a new insight into the properties of hadrons inside hot and/or dense matter. It is also a challenging many-body problem in QCD and could give us information on how nuclear matter undergoes a phase transition t o the quark gluon plasma. Since QCD is the theory of everything, we should always start with eq.(l) to attach the above problems. However, there are many ways to represent the same dynamics in different field choices. This is because the quantum amplitude is obtained after the integration over the field variables. For example, consider the full partition function in QCD: =
J
[dQdqdA]exp [iSQco(qlY, A)] .
This can be equivalently rewritten by integrating out the high frequency quarks and gluons above an arbitrary cutoff A, which gives the eq.(5) below. One can then make change of variables from q, Q, A to the hadronic fields (e.g., pions and nucleons), and obtain eq.(6) below.
..I
-
[ d ~ d N . exp L
A
(a,N, . . .; A)] .
Here sZf is the effective action written in terms of the low-energy quarks and gluons, while Sef is the effective action written in terms of the low-energy hadrons. Both effective actions must have explicit A dependence which should be cancelled by the A dependence from the field integration so that the full partition function is independent of A and gives a unique prediction: namely dZ/dA = 0.
-
So far, the argument is exact, but is not practical at all. In the actual application, one sometimes sets A 47r f, = 1.2 GeV (the chiral symmetry breaking scale) and construct a most general S,Hffallowed by symmetry and then determine the parameters in S,Hff by experiments. The chiral perturbation theory is a typical example of this sort [7]. If one has a big parallel computer, one may carry out the integration [dqdqdA] in eq.(5) numerically by taking A large ( E l / a with a being the lattice spacing). This is the lattice QCD numerical simulation which has considerable progress in recent years [8]. QCD sum rules which have great success in reproducing hadronic properties [9] are based on the Wilson's operator product expansion; in a way it makes use of the appropriate window of energy scale where both hadronic description and the quark-gluon description are valid (quark-hadron duality). After this introductory remarks, we will focus on in-medium hadrons in the following and review some recent theoretical developments. 2. IN-MEDIUM HADRONS
The idea of the in-medium hadrons has started in early 80's. In 1982, Pisarski discussed hadron modifications at finite temperature using phenomenological models and pointed out a possibility to observe the mass shift of the pmeson in dilepton measurement [lo]. In 1985, Kunihiro and I have studied the change of the scalar-isoscalar meson "a" near the critical point of chiral phase transition as a typical example of the dynamical critical phenomena in QCD [Ill. Brown and Rho has generalized these ideas and proposed a universal scaling hypothesis of hadron masses, which is, however, not yet proved. Heavy mesons such as J / Q have been also studied in relation to the deconfinement transition at finite T. In 1986, Hashimoto et al. has discussed the mass shift of J / Q and 7, just below the deconfinement transition [13],while Matsui and Satz proposed the disappearance of J / Q as a signature of the formation of the quark-gluon plasma [14]. In go's, it was widely recognized that the spectral functions of hadrons (instead of the ambiguous quantities such as "mass-shift" and "width-broadening") are the relevant quantity to be studied both theoretically and experimentally, although it was already implicit in ref.[11]. New theoretical tools were also developed in 90's such as the in-medium QCD sum rules [15],meson-baryon effective lagrangians [16],quark-meson coupling model [17]and so on. Since the number of references are enormous, I will refer to the Proceedings of the recent Hirschegg meeting on hadrons in dense matter [18]. In the next section, we will first show a direct connection of the spectral function in the vector channel with various experimental observables to illustrate its importance. 3. SPECTRAL FUNCTIONS FOR EM CURRENT
Among the various two point functions in QCD, the time-ordered correlation function of the electro-magnetic current is the most important one from the experimental point of view. We define the imaginary part of this correlation (the spectral function) as
where /target) is a given many-body state and J,, is the electromagnetic current in QCD:
The physical meaning of the spectral function Im(TJJ)ta,,et becomes clear by inserting the complete set between J's. Then the spectral function is a sum of contributions from all the states having non-vanishing matrix element with the state Jltarget). Namely, it probes the excited state of the many-body system which is obtained by acting the external current J onto the target. Now, let us show some example of the direct connection of Im(TJJ)tar,et with observable~. 1. For e+e- annihilation into hadrons (e+e- -+ y* 4 X ) , the cross section is directly proportional to the vacuum matrix element,
where s = q;. > 0 and the "target" is the vacuum. Therefore one is probing the structure of the vacuum through the time-like photon in this experiment. 2. For the deep-inelastic process with proton target (y* is written as
+ N -+ X), the cross section
is the Bjorken variable. Here the "target" is the where Q2 = -q:, < 0 and nucleon. Therefore, we are probing the nucleon structure through the space-like virtual photon.
+
3. For the emission of dileptons from the hot plasma (plasma -+ y* X ) , the emission rate per unit phase-space volume is
with q;, = u2 - q2 > 0. Here the "target" is the thermal distribution of hadrons or quarks and gluons. Thus we are probing the hot plasma through the emission of the time-like photon which eventually decays into dileptons. 4. If one can carry out the following hypothetical experiment ?*(time like) with A being a nucleus, the cross section is written as
which gives an information on the vector meson propagation in matter.
+A
4
X
5. In hadron-hadron collisions associated with a production of time-like photon ( A B + y* X) such as the Drell-Yan process, the cross section is written as
+
+
from which one can probe the the internal structure of the projectile and target and also the production mechanism of the time-like photon. 4. THEORETICAL TOOLS
Now, what kind of theoretical tools can be used to study the spectral functions in QCD? I will pick up two of them in the following; QCD sum rules and lattice QCD. 4.1. QCD sum rules By using the dispersion relation for the two-point function and the operator product expansion at short distance, one can derive a set of sum rules for the spectral function in the medium at finite density and at finite temperature [15]. They have the following general form of the energy weighted sum;
where Cnis the known Wilson coefficients, 0,is the gauge invariant local operators such as
Im(TJJ)pQcDis the known spectral function calculated perturbatively and thus does not depend on the target structure. The left hand side of eq.(14) can be estimated in the medium by the low energy theorem and low temperatureldensity theorem or possibly by the direct lattice QCD simulations. Therefore, one can make some constraints on the exact spectral function Im(TJJ)t,r,,t in the medium through the sum rules eq.(14). This QCD sum rule constraints have been useful for making some predictions of the spectral shift as well as for checking the validity of the effective field theory calculations ~91. 4.2. Lattice QCD The lattice QCD simulations have remarkable progress in recent years for calculating the properties of hadrons as well as the properties of QCD phase transition. In particular, the quenched QCD simulation on the masses of light mesons and baryons agree within 5-10 % with the experimental spectra [20]. However, the lattice QCD had difficulties in accessing the dynamical quantities in the Minkowski space, because measurements on the lattice can only be carried out for discrete points in imaginary time. The analytic continuation from the imaginary time to the real time using the noisy lattice data is highly non-trivial and is even classified as an ill-posed problem. Recently a first attempt to extract spectral functions (SPFs) of hadrons from lattice QCD data by using the maximum entropy method (MEM) has been reported [21]. MEM
is a method which has been successfully applied for similar problems in quantum Monte Carlo simulations in condensed matter physics, and image reconstruction in crystallography and astrophysics [22]. 4.2.1. Basic idea of MEM The Euclidean correlation function D ( r ) of an operator O(T,2)and its spectral decomposition at zero three-momentum read
D(T) =
J (ot(r,Z)O(O,
6))d3x=
1"
K ( r , w)A(w)dw,
(17) where T > 0, w is a real frequency, and A(w) is SPF (or sometimes called the image), which is positive semi-definite. The kernel K ( r , w) is proportional to the Fourier transform of a free boson propagator with mass w: At zero temperature (T = 0) in the continuum limit, K ( r , w) = exp(-TW).
(18)
Monte Carlo simulation provides D(ri) on the discrete set of temporal points 0
< ri/a <
N,. From this data with statistical noise, we need to reconstruct the spectral function A(w) with continuous variable w. This is a typical ill-posed problem, where the number of data is much smaller than the number of degrees of freedom to be reconstructed. This makes the standard likelihood analysis and its variants inapplicable 1231 unless strong assumptions on the spectral shape are made. MEM is a method to circumvent this difficulty through Bayesian statistical inference of the most probable image together with its reliability [22]. MEM is based on the Bayes' theorem: P[XIY] = P[Y IX]P[X]/P[Y], where P[XIY] is the conditional probability of X given Y. The most probable image A(w) for given lattice data D is obtained by maximizing the conditional probability P[AIDH], where H summarizes all the definitions and prior knowledge. The reliability of the obtained result can be checked by the second variation of P[AIDH] with respect to A(w). By the Bayes' theorem, (19)
P[AIDH] E P[DIAH]P[AJH],
where P[DIAH] (P[AIH]) is the likelihood function (the prior probability). For the likelihood function, the standard X 2 is adopted, namely P[D(AH]= Z;'exp(-L) with
ZLis a normalization factor. D(ri) is the lattice data averaged over gauge configurations and DA(ri) is the correlation function defined by the right hand side of (17). C is an N x N covariance matrix of the data with N being the number of temporal points to be used in the MEM analysis. The lattice data have generally strong correlations among different r's, and it is essential to take into account the off-diagonal components of C. Axiomatic construction as well as intuitive "monkey argument" [24] show that, for positive distributions such as SPF, the prior probability can be written with parameters a and m as P [ A ( H a m ]= 2;' exp(aS). Here S is the information entropy,
s=
1"[a(w)
- m(w) - ~
(l:;)]
( wlog ) - dws
Zs is a normalization factor. a is a real and positive parameter and m(w) is a real and positive function called the default model. In the state-of-art MEM [22],the output image Aout is given by a weighted average over A and a:
where A,(w) is obtained by maximizing the " - free-energy"
for a given a . Here we assumed that P[AIDHam] is sharply peaked around A,(w). a dictates the relative weight of the entropy S (which tends to fit A to the default model m) and the likelihood function L (which tends to fit A to the lattice data). Note, however, that a appears only in the intermediate step and is integrated out in the final result. Finding a global maximum of Q in the functional space of A(w), which has typically 1000 degrees of freedom in our case, can be done by utilizing the singular value decomposition (SVD) of the kernel K ( r , w) [25].
4.2.2. MEM with lattice data To apply MEM to actual lattice data, we have done quenched lattice QCD simulations with the plaquette gluon action and the Wilson quark action by the open MILC code with minor modifications [26]. The lattice size is 203 x 24 with P = 6.0, which corresponds to a = 0.0847 fm (a-I = 2.33 GeV), r;, = 0.1571 [27], and the spatial size of the lattice L,a = 1.69 fm. Hopping parameters are chosen to be r; = 0.153, 0.1545, and 0.1557 with NConf= 161 for each r;. For the quark propagator, the Dirichlet (periodic) boundary condition is employed for the temporal (spatial) direction. To calculate the two-point correlation functions, we adopt a point-source at 2 = 0 and a point-sink averaged over the spatial lattice-points. 12 for the Dirichlet boundary condition in the temporal We use data at 1 ri/a direction. We define SPFs for the PS and V channels as
<
<
so that pp,,,(w -+ large) approaches a finite constant as predicted by perturbative QCD. (the upper limit for the w We take Aw = 10 MeV for the w-integration in (17). w, integration) should be comparable to the maximum available momentum on the lattice: Wmas n / a 7.3 GeV.
-
In Fig. 1 (a) and (b), we show the reconstructed images for each K . Here we use the continuum kernel K = exp(-rw). In these figures, we have used m = mow2 with mo = 2.0(0.86) for PS (V) channel motivated by the perturbative estimate of mo We have checked that the result is not sensitive, within the statistical significance of the image, to the variation of mo by factor 5 or 115. The obtained images have a common structure: the low-energy peaks corresponding to .ir and p, and the broad structure in the high-energy region.
Figure 1. Reconstructed image pOut(w)for the PS (a) and V (b) channels. The solid, dashed, and dotted lines are for K = 0.1557, 0.1545, and 0.153, respectively. For the PS (V) channel, mo is taken to be 2.0 (0.86). w,,, is 7.5 GeV in this figure and Fig.2.
1. The mass of the pmeson in the chiral limit extracted from the peaks in Fig.l(b) reads m,a = 0.348(15). This is consistent with m,a = 0.331(22) determined by the asymptotic behavior of D ( T ) on larger lattice [27]. Although our maximum value = 12 marginally covers the asymptotic limit in r, we of the fitting range r,,,/a can extract reasonable masses for 7r and p. The width of 7r and p in Fig.1 is an artifact due to the statistical errors of the lattice data. In fact, in the quenched approximation, there is no room for the pmeson to decay into two pions. 2. As for the second peaks in the PS and V channels, the error analysis shows that their spectral "shape" does not have much statistical significance, although the existence of the non-vanishing spectral strength is significant. Under this reservation, we fit the position of the second peaks and made linear extrapolation to the chiral limit with the results,
These numbers should be compared with the experimental values:
Figure 2. Same with Fig.1 except for the use of the lattice kernel Klat.
One should remark here that, in the standard two-mass fit of D ( 7 ) , the mass of the second resonance is highly sensitive to the lower limit of the fitting range [27]. This is because the contamination from the short distance contributions from T < ?-,in is not under control in such an approach. On the other hand, MEM does not a suffer from this difficulty and can utilize the full information down to ~ , ~ ~=/ 1. Therefore, MEM opens a possibility of systematic study of higher resonances with lattice QCD data.
3. As for the third bumps in Fig.1, the spectral "shape" is shown to be statistically not significant, and they should rather be considered a part of the perturbative continuum instead of a single resonance. Fig.1 also shows that SPF decreases substantially above 6 GeV; MEM automatically detects the existence of the momentum cutoff on the lattice r/a.It is expected that MEM with the data on finer lattices leads to larger ultraviolet cut-offs in the spectra. The height of the asymptotic form of the spectrum at high energy is estimated as N
where Zv is the renormalization factor of the lattice composite operator. Our estimate in eq.(25) is consistent with the high energy part of the spectrum in Fig.l(b) after averaging over w . We made a similar estimate for the PS channel: p,,(w 2: 6GeV) 2.0, which is also consistent with Fig. l ( a ) .
-
In Fig.2(a) and (b), the results using the lattice kernel Klat are shown. Klat is obtained from the free boson propagator on the lattice. It reduces to K = exp(-TW) when a -+ 0. The other parameters and boundary conditions are the same with Fig.l(a,b). The difference of Fig.1 and Fig.2 can be interpreted as a systematic error due to the finiteness of the lattice spacing a. 4.2.3. Error analysis
The statistical significance of the reconstructed image can be studied by the following procedure [22]. Assuming that P[AIDHom] has a Gaussian distribution around the most probable image A, we estimate the error by the covariance of the image,
where is a functional derivative and (.) is an average over a given energy interval. The final error for Amt is obtained by averaging the covariance over o with a weight factor P[alD H m ] . We found small error for the lowest peak, which supports our identification of the peak with p. Although the existence of the non-vanishing spectral strength of the 2nd peak and 3rd bump is statistically significant, their spectral "shape" is either marginal or insignificant. 4.2.4. Summary of SPF in lattice QCD
We have made a first serious attempt to reconstruct SPFs of hadrons from lattice QCD data. We have used MEM, which allows us to study SPFs without making a priori assumption on the spectral shape. The method works well for the mock data and actual lattice data. MEM produces resonance and continuum-like structures in addition to the ground state peaks. The statistical significance of the image can be also analyzed. Better data with finer and larger lattice will produce better images with smaller errors, and our study is a first attempt towards this goal. So far, we have calculated only the spectral functions in the vacuum. Preliminary studies on the lattice at finite temperature are under way and will be reported in the future. 5. PARTIAL CHIRAL RESTORATION IN NUCLEI 5.1. Quark condensate at finite density The direct measure of the partial restoration of chiral symmetry in medium is the chiral condensate ( q q ) . There is indeed an exact theorem in QCD for the behavior of the condensate [28]
where CnN = 45 i 10 MeV is the pion-nucleon sigma term and E(p)/A is the nuclear binding energy per particle with m being the current quark mass. If one makes low density expansion of the right hand, one obtains almost 35 % reduction of the condensate even in the nuclear matter density po = 0.17fm-3. Direct evidence of such partial chiral restoration could be a threshold enhancement of the spectral strength in the scalar-isoscalar channel as was recently pointed out in
ref.1291. A simplified explanation of the idea behind this spectral enhancement is as follows: Suppose we have a typical double-well effective potential of QCD written in terms of the order parameter a = (qq) as
where a is positive in the vacuum but changes sign at the critical point, while b remains positive. The minimum of the effective potential a0 and the curvature at the minimum read
Therefore, as a becomes small, not only the order parameter a0 but also the curvature decrease. This is nothing but the softening of the vibrational mode associated with the 2nd order phase transition [30]. In the real world, the situation is not that simple, since a has a large with decaying into two pions. Nevertheless, there is an interesting possibility that the spectral function just above the two-pion threshold could be enhanced due to the change of ao, which was originally shown in ref.1311 at finite T and later generalized to the case for finite baryon density in ref. [29]. 5.2. Model calculation Let us first describe the general features of the spectral enhancement near the twopion threshold. Consider the propagator of the scalar-isoscalar "a-meson" at rest in the medium :
where m, is the mass of a in the tree-level, and C,(w; p) is the loop corrections in the vacuum as well as in the medium. The corresponding spectral function is given by 1 p, (w) = - - ImD,(w). 'T
Near the two-pion threshold, the imaginary part in the one-loop order reads ImC, a e
( -~2m,)
.J -.43 1-
When chiral symmetry is being restored, m; ("effective mass" of a defined as a zero of the real part of the propagator ReDil(w = m:) = 0) approaches to m,. Therefore, there exists a density p, at which ReDgl(w = 2m,) vanishes even before the complete a-n degeneracy takes place; namely
At this point, the spectral function can be solely represented by the imaginary part of the self-energy;
which shows an enhancement of the spectral function at the 2m, threshold. We remark that this enhancement is generically correlated with the partial restoration of chiral symmetry. To make the argument more quantitative, let us evaluate p,(w) in the SU(2) 1'inear a-model:
where tr is for the flavor index and M = a+i?.?i. Although the model has only a limited number of parameters and is not a precise low energy representation of QCD, we emphasize that it does describe the pion dynamics qualitatively well up to 1GeV as shown by Chan and Haymaker 1331. The coupling constants p2,X and h have been determined in the vacuum to reproduce f, = 93 MeV, m, = 140 MeV as well as the s-wave T-n scattering phase shift in the one-loop order. We parametrize the chiral condensate in nuclear matter (a) as
In the linear density approximation, @(p)= 1 - C-.P (38) Po The plausible value of @(p= po) is 0.7 0.9 [30]. The spectral function together with ReDgl(w) calculated in the linear sigma model are shown in Fig.3 and 4: The characteristic enhancement of the spectral function is seen just above the 2m,. To confirm this threshold enhancement experimentally, measuring 27r0 and 27 in experiments with hadron/photon beams off the heavy nuclear targets are useful. Measuring a 27r0 -, 47 is experimentally feasible 1321, which is free from the p meson background inherent in the T+T- measurement. Measuring a --+ 27 is also interesting because of the small final state interactions, although the branching ratio is small. (One needs also to fight with large background of photons mainly coming from T's.) Nevertheless, if the enhancement is prominent, there is a chance to find the signal. Recently CHAOS collaboration [34] reported the data on the 7r+n* invariant mass distribution M$,* in the reaction A(T+,T+T*)X with the mass number A ranging from 2 to 208: They observed that the yield for M,"+,- near the 2m, threshold is close to zero for A = 2, but increases dramatically with increasing A. They identified that the T+Tpairs in this range of M,"+,- is in the I = J = 0 state. The A dependence of the the invariant mass distribution presented in 1341 near 2m, threshold has a close resemblance to our model calculation in Fig.3 and 4, which suggests that this experiment may already provide a hint about how the partial restoration of chiral symmetry manifest itself at finite density. Further theoretical works are necessary to unlabel the real nature of the threshold enhancement in the CHAOS data and some preliminary studies have been already started 135,181.
-
--+
Figure 3. Spectral function for o and the real part of the inverse propagator for severa1 values of = ( L T ) / with o~ m y k = 550 MeV. In the lower panel, decreases from bottom to top.
Figure 4. Same with Fig.1 for m y k = 750 MeV
6. SUMMARY
QCD is a theory of everything in strong interaction. Although the effective field theory descriptions are always possible at low energies in terms of the effective action S,Hff, one should keep in mind the fundamental relation such as eqs. (4,5,6) which relate the fundamental theory and phenomenology. Among many other observables, the spectral function (SPF) is the simplest but interesting quantity. I can be studied both from the QCD action (QCD sum rules, lattice QCD etc) and from the effective hadronic action. Also, SPFs are either directly or closely related to the experimental data. Namely, SPFs are the observables through which we can directly compare theory and experiment without much intermediate interface. In relation to the physics of chiral symmetry and its restoration in hot/dense matter, SPFs of light vector mesons ( p , w and 4 ) are most relevant since they can be observed by dileptons. SPF of scalar-isoscalar "oncould be also interesting: "a" does not show up clearly in the vacuum because of it huge width decaying into two pions, but it may appear as a soft and narrow collective mode when the chiral symmetry is (partially) restored. In relation to the deconfinement of quarks in hotldense matter, SPFs of heavy mesons such as J/* and Q' are most relevant. Since there are not only numerous recent developments on SPFs in QCD sum rules, in
lattice QCD and in effective field theories, it is quite likely that we may have more solid understanding of the in-medium hadrons in the near future. Experimental inputs and the cooperative works between theorists and experimentalists are, of course, essential for this purpose.
REFERENCES 1. Y. Nambu, in Preludes in Theoretical Physics, in honor of V. F. Weisskopf (North Holland, Amsteldam, 1966). T. Muta, Foundation of Quantum Chromodynamics, (World Scientific, Singapore, 1987). 2. D. J. Gross and F. Wilczek, Phys. Rev. Lett. 30 (1973) 1343. H. D. Politzer, Phys. Rev. Lett. 30 (1973) 1346. 3. J. C. Collins and M. J. Perry, Phys. Rev. Lett. 34 (1975) 1353. 4. C. Borgs, Nucl. Phys. B261 (1985) 455. 5. D. Bailin and A. Love, Phys. Rep. 107 (1984) 325. M. Iwasaki and T. Iwado, Phys. Lett. B350 (1995) 163. M. Alford, K. Rajagopal and F. Wilczek, Phys. Lett. B422 (1998) 247. R. Rapp, T . Schafer, E. V. Shuryak and M. Velkovsky, Phys. Rev. Lett. 81 (1998) 53. See the recent review; T . Schaefer, Color Superconductivity, nucl-th/9911017. 6. Quark Matter '97 Proceedings, Nucl. Phys. A638 (1998) 1. Quark Matter '99 home page: http://www.qm99.to.infn.it/ 7. See e.g. H. Leutwyler, Ann. Phys. 235 (1994) 165. A. Pich, Rept. Prog. Phys. 58 (1995) 563. 8. Lattice 98 Proceedings, Nucl. Phys. B (Proc. Suppl.) 73 (1999). 9. Vacuum Structure and QCD Sum Rules, ed. M. A. Shifman (North-Holland, Amsterdam, 1992). 10. R. D. Pisarski, Phys. Lett. BllO (1982) 155. 11. T . Hatsuda and T. Kunihiro, Phys. Rev. Lett. 55 (1985) 158. 12. G. Brown and M. Rho, Phys. Rev. Lett. 66 (1991) 2720. 13. T. Hashimoto, 0 . Miyamura, K. Hirose, and T. Kanki, Phys. Rev. Lett. 55 (1986) 2123. 14. T. Matsui and H. Satz, Phys. Lett. B178 (1986) 416. 15. T. Hatsuda and S. H. Lee, Phys. Rev. C46 (1992) R34. T . Hatsuda, Y. Koike and S. H. Lee, Nucl. Phys. B394 (1993) 221. S. H. Lee, Phys. Rev. C57 (1998) 927; erratum-ibid. C58 (1998) 3771. 16. C. Gale and J. Kapusta, Nucl. Phys. B357, 65 (1991). M. Asakawa, C.M. KO, P. Levai and X. J. Qiu, Phys. Rev. C46 (1992) 1159. M. Herrmann, B. Riman and W. Norenberg, Z. Phys. A343 (1992) 119. G. Chanfray and P. Schuck, Nucl. Phys. A545 (1992) 271c. 17. K. Saito, K. Tsushima and A. W. Thomas, Phys. Rev. C55 (1997) 2637. 18. Proceedings of Hirschegg 2000 workshop on Hadrons in Dense Matter, to appear. 19. M. Asakawa and C. M. KO, Nucl. Phys. A572 (1994) 732. B. Riman, S. H. Lee, and H. Kim, Nucl. Phys. A653 (1999) 91. F. Klingl and W. Weise, Eur. Phys. J. A4 (1999) 225.
20. 21. 22.
23.
24. 25. 26. 27. 28. 29. 30. 31. 32.
33. 34. 35.
S. Leupold and U. Mosel, Phys. Rev. C58 (1998) 2939. A. Hayashigaki, Prog. Theor. Phys. 101 (1999) 923. F. Klingl, S. Kim, S. H. Lee, P. Morath and W. Weise, Phys. Rev. Lett. 82 (1999) 3396; erratum-ibid. 83 (1999) 4224. S. Aoki et al., Phys. Rev. Lett. 84 (2000) 238. Y. Nakahara, M. Asakawa, and T. Hatsuda, Phys. Rev. D60 (Rapid Comm.) (1999) 091503; hep-lat/9909137. N. Wu, The Maximum Entropy Method, (Springer-Verlag, Berlin, 1997). M. Jarrell and J . E. Gubernatis, Phys. Rep. 269 (1996) 133. W. von der Linden, R. Preuss, and W. Hanke, J . Phys. 8 (1996) 3881. R. N.Silver et al., Phys. Rev. Lett. 65 (1990) 496. D. B. Leinweber, Phys. Rev. D 51 (1995) 6369. D. Makovoz and G. A. Miller, Nucl. Phys. B468 (1996) 293. C. Allton and S. Capitani, Nucl. Phys. B526 (1998) 463. Ph. de Forcrand et al., Nucl. Phys. B (Proc. Suppl.) 63A-C (1998) 460. J. Skilling, in Maximum Entropy and Bayesian Methods, ed. J . Skilling (Kluwer, London, 1989), pp.45-52; S. F. Gull, ibid. pp.53-71. R. K. Bryan, Eur. Biophys. J., 18 (1990) 165. The MILC code ver. 5, http://cliodhna.cop.uop.edu/-hetricklmilc . Y. Iwasaki et al., Phys. Rev. D 53 (1996) 6443. E. G. Drukarev and E. M. Levin, Prog. Paet. Nucl. Phys. A556 (1991) 467. T. Hatsuda, T. Kunihiro and H. Shimizu, Phys. Rev. Lett. 82 (1999) 2840. T . Hatsuda and T. Kunihiro, Phys. Rep. 247 (1994) 222. S. Chiku and T . Hatsuda, Phys. Rev. D57 (1998) R6, ibid. D58 (1998) 076001. H. Shimizu, in Proceedings of XV RCNP Osaka International Symposium, Nuclear Physics Frontiers with Electro-weak Probes (FRONTIER 96), (Osaka, March 7-8, 1996), p.161. L. H. Chan and R. W. Haymaker, Phys. Rev. D7 (1973) 402; ibid. D l 0 (1974) 4170. F. Bonutti et al. (CHAOS Collaboration), Phys. Rev. Lett. 77 (1996) 603; Phys. Rev. C60 (1999) 018201. R. Rapp et al., Phys. Rev. C59 (1999) R1237. Z. Aouissat et al., Phys. Rev. C61 (1999) 012202. M. J. Vicente Vacas and E. Oset, Phys. Rev. C60 (1999) 064621.
ThisPageIntolyLfBk
ThisPageIntolyLfBk This Page Intentionally Left Blank
11. MESONS IN NUCLEAR MEDIUM
ThisPageIntolyLfBk
ThisPageIntolyLfBk This Page Intentionally Left Blank
Hadron and Nuclear Physics with Electromagnetic Probes K. Maruyama and H. Okuno (Editors) 2000 Elsevier Science B.V. All rights reserved.
Vector mesons in medium and dileptons in heavy-ion collisions Ralf Rappa * and Jochen Wambachb "Department of Physics and Astronomy, SUNY Stony Brook, New York 11794-3800, USA bInstitut fur Kernphysik, TU Darmstadt, Schlofigartenstr. 9, D-64289 Darmstadt, Germany Theoretical approaches to assess modifications of vector mesons in the medium, as well as their experimental identification via electromagnetic probes, are discussed. Implications for the nature of chiral symmetry restoration in hot/dense matter are outlined and put into context with the axialvector channel. 1. INTRODUCTION
The investigation of hadron properties inside atomic nuclei constitutes one of the traditional research objectives in nuclear physics. However, in terms of the underlying theory of strong interactions QCD) even the description of the nuclear ground state remains elusive so far. Valuable insights can be expected from a careful study of transition regimes between hadronic and quark-gluon degrees of freedom. E.g., in electron-nucleus scattering experiments the corresponding control variable is the momentum transfer, whereas heavy-ion reactions, performed over a wide range of collision energies, aim at compressing and/or heating normal nuclear matter to witness potential phase transitions into a Quark-Gluon Plasma (QGP). Among the key properties of the low-energy sector of strong interactions is the (approximate) chiral symmetry of the QCD Lagrangian and its spontaneous breaking in the vacuum. This is evident from such important phenomena as the build-up of a chiral condensate and constituent quark mass (Mq N 0.4 GeV), or the large mass splitting 0.5 GeV between 'chiral partners' in the hadron spectrum (such as ~ ( 1 4 0 ) of A M a(400 - 1200))p(770)-a1(1260) or N(940)-N*(1535)). It also indicates that medium modifications of hadron properties can be viewed as precursors of chiral symmetry restoration. In this talk the focus will be on the vector (V) and axialvector (A) channels. The former is special in that it directly couples to the electromagnetic current (i.e., real and virtual photons) at which point it becomes 'immune' to (strong) final state interactions thus providing direct experimental access to in-medium properties of vector mesons, e.g., through photoabsorption/-production on nuclei, or dilepton (e+eV,p'p-) spectra in heavy-ion reactions. The key issue is then to relate the medium effects to mechanisms of chiral
-
*research supported in part by the A,-v.-Humboldt Foundation (Feodor-Lynen program) and U.S. Department of Energy under Grant No. DE-FG02-88ER40388.
restoration. This necessitates the simultaneous consideration of the axialvector channel, which, however, largely has to rely on theoretical analyses. This talk is structured as follows: Sect. 2 is devoted to vector-meson properties in nuclear matter, Sect. 3 contains applications to heavy-ion reactions and Sect. 4 finishes with conclusions. A more complete discussion of the presented topics can be found in a recent review [I]. 2. (AXIAL-) VECTOR MESONS IN COLD NUCLEAR MATTER 2.1. Correlators and duality threshold The general quantity that is common to most theoretical approaches is the currentcurrent correlation function which in the (axial-) vector channel is defined by
For simplicity we will concentrate on the (prevailing) isospin I = 1 (isovector) projections 1 j:==, = -(u rp u - d rp d) with = y p , I?$ = y5yp . (2) 2 At sufficiently high invariant mass both correlators can be described by their (identical) perturbative forms which read (up to as corrections)
( M 2 = qi - p). At low invariant masses the vector correlator is accurately saturated by the (hadronic) p spectral function within the Vector Dominance Model (VDM), i.e.,
with a similar relation involving the al meson in the axialvector channel. The spontaneous breaking of chiral symmetry (SBCS) manifests itself in both the difference of the al and p spectral functions as well as the additional pionic piece in ITA (notice that f, is another order parameter of SBCS). In vacuum the transition from the hadronic to the partonic regime ('duality threshold') is characterized by the onset of perturbative QCD around MdualN 1.5 GeV. In the medium, chiral restoration requires the degeneration of V- and A-correlators over the entire mass range. 2.2. Model-independent results: V-A mixing and sum rules In a dilute gas the prevailing medium effect can be computed via low-density expansions. Using soft pion theorems and current algebra Krippa [2] extended an earlier finitetemperature analysis [3] to the finite-density case to obtain
i.e., the leading density effect is a mere 'mixing' of the vacuum correlators IIOfiV. The 'mixing' parameter
(eN: nucleon density) is determined by the 'long-range' part of the T N sigma term, STN= ~ T ~ ~ : ( N ~ T -2 ~ 20( MeV N)
.
(8)
Chanfray et al. 141 pointed out that anNis in fact governed by the well-known nucleonand delta-hole excitations in the pion cloud of the p (or al) meson which have been thoroughly studied within hadronic models to be discussed in the following section. A naive extrapolation of eq. (7) to the chiral restoration point where E = 112, gives Q, E 2 . 5 ~ 0which , is not unreasonable. Nonetheless, as we will see below, realistic models exhibit substantial medium modifications beyond the mixing effect. Similar in spirit, i.e., combining low-density expansions with chiral constraints, is the so-called master formula approach applied in ref. [5]: chiral Ward identities including the effects of explicit breaking are used to express medium corrections to the correlators through empirically inferred TN, p N (or y N , etc.) scattering amplitudes times density. Resummations to all orders in density cannot be performed either in this framework. Model independent relations which are in principle valid to all orders in density are provided by sum rules. Although typically of little predictive power, their evaluation in model calculations can give valuable insights. One example are the well-known QCD sum rules which have been used to analyze vector-meson spectral functions in refs. [6,7]. It has been found, e.g., that the generic decrease of the quark- and gluon-condensates on the right-hand-side (r.h.s.) is compatible with the phenomenological (left-hand) side if either (i) the vector meson masses decrease (together with small resonance widths), or, (b) both width and mass increase (as found in most microscopic models). Another example of sum rules are the ones derived by Weinberg [8], being generalized to the in-medium case in ref. [9]. The first Weinberg sum rule, e.g., connects the pion decay constant to the integrated difference between the V- and A-correlators:
for arbitrary three-momentum q (here, the pionic piece has been explicitly separated out from nA).We will come back to this relation below. 2.3. Hadronic models a n d experimental constraints Among the most spectacular predictions for the behavior of vector mesons in medium is the Brown-Rho Scaling hypothesis [lo]. By imposing QCD scale invariance on a chiral effective Lagrangian at finite density and applying a mean-field approximation it was conjectured that all light hadron masses (with the exception of the symmetry-protected Goldstone bosons) drop with increasing density following an approximately universal sealing law. The scaling also encompasses the pion decay constant (as well as an appropriate power of the quark condensate) and therefore establishes a direct link to chiral symmetry restoration being realized through the vanishing of all light hadron masses.
More conservative approaches reside on many-body techniques to calculate selfenergy contributions to the vector-meson propagators Dv arising from interactions with surrounding matter particles (nucleons). They are computed from gauge-invariant (vectorcurrent conserving) as well as chirally symmetric Lagrangians. The p propagator, e.g., takes the form
for both transverse and longitudinal polarization states (which in matter, where Lorentzinvariance is lost, differ for q > 0). C,, encodes the medium modifications in the pion cloud (through NN-' and AN-' bubbles, so-called 'pisobars') [ll-13,6,14], and C p stems from direct 'rhosobar' excitations of either S-wave (N(1520)N-l, A(1700)N-1, . . .) or P-wave type (AN-', N(1720)N-', . . .) [15-181. The parameters of the interaction vertices (coupling constants and form factor cutoffs) can be estimated from free decay branching ratios of the involved resonances or more comprehensive scattering data (e.g., .irN -+ pN [19], or y N absorption) which determine the low-density properties of the spectral functions. Additional finite-density constraints can be obtained from the analysis of photoabsorption data on nuclei. Invoking the VDM, the total photoabsorption cross section can be readily related to the imaginary part of the in-medium vector-meson selfenergy in the zero mass limit (needed for the coupling to real photons). An example of such a calculation is displayed in Fig. 1 where a reasonable fit to existing data on various nuclei has been achieved. The low-density limit (represented by the long-dashed line in -10
-
0
300
600 900 q, [MeV]
1.200
1500
0.0
vacuum
0.2
0.4
0.6
0.8
1.0
1.2
M [GeV]
Figure 1. Photoabsorption spectra on nuclei Figure 2. Spin-averaged pmeson spectral with (full line) and without (long-dashed function in cold nuclear matter [14,20]. line) higher order medium effects [18]; short-dashed line: C,,, contribution.
-
Fig. 1) cannot reproduce the disappearance of especially the N(1520) as seen in the data. A selfconsistent calculation to all orders in density [I?], however, generates sufficiently large in-medium widths, on the order of 200-300 MeV (resulting in the full line). Fig. 2 shows the final result for the p spectral function [20] which has been subjected to the aforementioned constraints. The apparent strong broadening is consistent with other
~
~
calculations [6,17]. Similar features, albeit less pronounced, emerge within analogous treatments for w and 4 mesons [6]. Let us now return to the question what these findings might imply for chiral restoration. In a recent work by Kim et al. [21]an effective chiral Lagrangian including al-meson degrees of freedom has been constructed. Medium modifications of the latter are introduced by an 'al-sobar' through N(1900)N-I excitations to represent the chiral partner of the N(1520)N-I state. Pertinent (schematic) two-level models have been employed for both the p and al spectral densities which, in turn, have been inserted into the Weinberg sum rule, eq. (9) (supplemented by perturbative high energy continua). The resulting
Figure 3. Pion decay constant at finite density when evaluated through the Weinberg sum rule using schematic two-level models for in-medium p and al spectral functions [21].
-
density-dependence of the pion decay constant, displayed in Fig. 3, exhibits an appreciable decrease of 30% at QN = eO,which bears some sensitivity on the assumed branching ratio of the N(1900) + Nul decay (or N(1900) Nul coupling constant). However, the mechanism is likely to be robust: due to the low-lying pN(1520)N-I and ~ l - N ( 1 9 0 0 ) N - ~ excitations, accompanied by a broadening of the elementary resonance peaks, the p and al spectral densities increasingly overlap, thus reducing f,. 3. ELECTROMAGNETIC OBSERVABLES IN HEAVY-ION REACTIONS
In central collisions of heavy nuclei at (ultra-) relativistic energies (ranging from pl,b=l200 AGeV in current experiments to +=0.2-10 ATeV at RHIC and LHC) hot and dense hadronic matter is created over extended time periods of about 20 fm/c. Local thermal equilibrium is probably reached within the first fm/c, after which the 'fireball' expands and cools until the strong interactions cease ('thermal freezeout') and the particles stream freely to the detector. Electromagnetic radiation (real and virtual photons) is continuously emitted as it decouples from the strongly interacting matter at the point of creation. The thermal production rate of e+e- pairs per unit 4-volume can be expressed through
I0-'
-----
m -free in-medium an
----- free nn
i
in-medium m
0.0
0.2
0.4
0.6
0.8
1.0
-
1.2
M, [GeVl
Figure 4. e+e- production rates in hot hadronic matter from free 7r+7r- (dashed line), in-medium n-+7r- (full line) [20] and O(&) qq annihilation (dashed-dotted line). the electromagnetic current correlation function (summed over all isospin states I=0,1),
( f B : Bose distribution function; a similar expression holds for photons with M + 0). Fig. 4 shows that the medium effects in the p propagator (including interactions with nucleons as well as thermal pions, kaons, etc.) induce a substantial reshaping of the emission rate (full lines) as compared to free n7r annihilation (dashed line) already at rather moderate temperatures and densities (left panel). In fact, under conditions close to the expected phase boundary (right panel) the p resonance is completely 'melted' and the hadronic dilepton production rate is very reminiscent to the one from a perturbative Quark-Gluon Plasma (dashed-dotted lines in Fig. 4) down to rather low invariant masses of -- 0.5 GeV (as corrections to the partonic rate might improve the agreement at still lower masses). It has been suggested [20] to interpret this as a lowering of the in-medium quarkhadron duality threshold as a consequence of the approach towards chiral restoration. The total thermal yield in a heavy-ion reaction is obtained by a space-time integration of eq. (11) over the density-temperature profile for a given collision system, modeled, e.g., within transport [22] or hydrodynamic [23] simulations. At CERN-SpS energies (160200 AGeV) this 'thermal' component is dominant over (or at least competitive with) final state hadron decays (at low M ) and hard initial processes such as Drell-Yan annihilation (at high M) in the invariant mass range M 2 0.2-2 GeV. A consistent description of the measured data [24-271 is possible once hadronic many-body effectsare included [20,28,29], cf. Figs.5 and 6. However, at this point also the dropping mass scenario [lo] is compatible with the data [30] (cf. dashed curve in Fig. 5). Optimistically one may conclude that strongly interacting matter close to the hadronQGP phase boundary has been observed at the CERN-SpS. Other observables such as hadro-chemistry 1311 or J / Q suppression [32] also support this scenario. Nonetheless, further data are essential to substantiate the present status and resolve the open questions.
h P
<
v
central S+Au 200AGeV
1
r
9
g
lo-"
--. 22
,F 1 0 - ~
-5
Pp lo-'
.
2
"lo-"
1
I
Figure 5. Low-mass dilepton spectra at the Figure 6. Direct photon spectra at the CERN-SpS [25];Dashed-dotted line: parti- CERN-SpS: upper limits as extracted from cle decays after freezeout (hadronic 'cock- the WA80 experiment [24] compared to tail'), other lines: cocktail + .ir.ir annihila- in-medium thermal radiation contributions tion as indicated. (full line) [I].
4. CONCLUSIONS
This talk has focused on medium modifications of vector mesons in connection with chiral symmetry restoration in hotldense matter. In accordance with a variety of empirical information hadronic spectral functions are characterized by the appearance of low-lying excitations as well as a broadening of the resonance structures. A schematic treatment of the al meson on similar footings shows that these features encode an approach towards chiral restoration in nuclear matter as signaled by the decrease of the pion decay constant when evaluating the first Weinberg sum rule. The application of these model calculations to electromagnetic observables as measured in recent heavy-ion experiments at the CERN-SpS leads to a reasonable description of the data from 0 to 2 GeV in invariant mass. The structureless in-medium hadronic dilepton production rates resemble perturbative qQ annihilation in the vicinity of the expected phase boundary indicating that chiral restoration might be realized through a reduction of the quark-hadron duality threshold which in vacuum is located around 1.5 GeV. It would also corroborate the interrelation between temperatureldensity and momentum transfer in the transition from hadronic to partonic degrees of freedom. In the near future further dilepton data will be taken by the PHENIX experiment 1341 at RHIC (advancing to a new energy frontier) as well as the precision experiment HADES [33] at GSI. Thus electromagnetic observables can be expected to continue the progress in our understanding of strong interaction physics.
Acknowledgments It is a pleasure to thank G.E. Brown, E.V. Shuryak and H. Sorge for collaboration and many fruitful discussions.
REFERENCES 1. 2. 3. 4.
R. Rapp and J . Wambach, to appear in Adv. Nucl. Phys. (2000))and hep-ph/9909229. B. Krippa, Phys. Lett. B427 (1998) 13. M. Dey, V.L. Eletsky and B. Ioffe, Phys. Lett. B252 (1990) 620. G. Chanfray, J . Delorme, M. Ericson and M. Rosa-Clot, nucl-th/9809007. 5. J.V. Steele, H. Yamagishi and I. Zahed, Phys. Rev. D56 (1997) 5605. 6. F. Klingl, N. Kaiser and W. Weise, Nucl. Phys. A624 (1997) 527. 7. S. Leupold, W. Peters and U. Mosel, Nucl. Phys. A628 (1998) 311. 8. S. Weinberg, Phys. Rev. Lett. 18 (1967) 507. 9. J.I. Kapusta and E.V. Shuryak, Phys. Rev. D49 (1994) 4694. 10. G.E. Brown and M. Rho, Phys. Rev. Lett. 66 (1991) 2720. 11. M. Herrmann, B. Friman and W. Norenberg, Nucl. Phys. A560 (1993) 411. 12. G. Chanfray and P. Schuck, Nucl. Phys. A555 (1993) 329. 13. M. Asakawa, C.M. KO, P. Lkvai and X.J. Qiu, Phys. Rev. C46 (1992) R1159. 14. M. Urban, M. Buballa, R. Rapp and J . Wambach, Nucl. Phys. A641 (1998) 433. 15. B. F'riman and H.J. Pirner, Nucl. Phys. A617 (1997) 496. 16. R. Rapp, G. Chanfray and J. Wambach, Nucl. Phys. A617 (1997) 472. 17. W. Peters et al., Nucl. Phys. A632 (1998) 109. 18. R. Rapp, M. Urban, M. Buballa and J . Wambach, Phys. Lett. B417 (1998) 1. 19. B. Friman, in Proc. of ACTP Workshop on 'Hadron Properties in Medium' (Seoul, Korea, 27.-31.10.97); and nucl-th/9801053. 20. R. Rapp and J . Wambach, Eur. Phys. J. A6 (1999) 415; R. Rapp, Nucl. Phys. A661 (1999) 33c. 21. Y. Kim, R. Rapp, G.E. Brown and M. Rho, nucl-th/9912061. 22. W. Cassing and E.L. Bratkovskaya, Phys. Rep. 308 (1999) 65. 23. J . Sollfrank et al., Phys. Rev. C55 (1997) 392; C.M. Hung and E.V. Shuryak, Phys. Rev. C56 (1997) 453. 24. R. Albrecht et al., WA80 collaboration, Phys. Rev. Lett. 76 (1996) 3506. 25. G. Agakichiev et al., CERES collaboration, Phys. Lett. B422 (1998) 405; B. Lenkeit, Doctoral Thesis, University of Heidelberg, 1998. 26. A.L.S. Angelis et al. (HELIOS-3 collaboration), Eur. Phys. J. C5 (1998) 63. 27. E. Scomparin et al. (NA50 collaboration), J. Phys. G25 (1999) 235; P. Bordalo et al. (NA50 collaboration), Nucl. Phys. A661 (1999) 538c. 28. R. Rapp and E.V. Shuryak, Phys. Lett. B473 (2000) 13. 29. K. Gallmeister, B. Kampfer and O.P. Pavlenko, Phys. Lett. B473 (2000) 20. 30. G.Q. Li, C.M. KO, G.E. Brown and H. Sorge, Nucl. Phys. A611 (1996) 539; G.Q. Li and C. Gale, Phys. Rev. C58 (1998) 2914. 31. P. Braun-Munzinger and J. Stachel, Nucl. Phys. A638 (1998) 3c. 32. H. Satz, Nucl. Phys. A661 (1999) 104c. 33. J . F'riese, these proceedings. 34. see, e.g., the PHENIX homepage a t http://www.phenix,bnI.gov.
Hadron and Nuclear Physics with ElectromagneticProbes K. Maruyama and H. Okuno (Editors) 2000 Elsevier Science B.V. All rights reserved.
'
meson in the nucleus K. Maruyama
"
"Center for Nuclear Study, University of Tokyo Midori-cho 3-2-1, Tanashi, Tokyo, Japan, 188-0002
meson in nuclear Theoretically predicted modification in the mass spectrum of the medium (3He) has been tested by a tagged-photon experiment at Tokyo. The p0 mesons are identified in the coherent photoproduction process on the nucleus. The mass and width of determined by using the Soding model parameterization are consistent with the vacuum values. While, the T + T - invariant mass spectrum observed in the incoherent process requires a modified-mass contribution in addition to the vacuum-mass pO.
1. INTRODUCTION
For the mass of the hadrons in vacuum, there are several types of theoretical calculations such as the constituent-quark models. the bag models. and the Lattice models. The masses and widths predicted by the theories have been examined by observations, and one believes that the experimental values are the true hadron masses and widths. While, a variety of theories predict the modification in the mass spectrum of the vector meson in nuclear medium. These predictions are due to be tested by experimental observations. Although. many theoretical papers of these topics are available at present [1,13], no direct mass measurement is in hand. A direct way is to determine the mass of the hadron which decays in nuclear medium by constructing the invariant mass from the measurements of the decayed particle momenta. A mass-shifted peak structure is the signal of the modification. This is a direct experimental determination of the p0 meson mass in nuclear medium (3He). The p0 meson (the lightest vector meson) is selected from the hadrons as the easiest candidate to examine the prediction. Two talks in these Proceedings [14,15] are excellent experimental and theoretical reviews to the physics topics. I will make a very brief summary of the theoretical status. Then, I will concentrate on a experimental test of the mass shift of the p0 meson in 3He. The measurement was carried out at the Institute for Nuclear Study. University of Tokyo (INS), using the 1-GeV photon beam from the 1.3GeV Electron Synchrotron (ES) by the TAGX (ES134) collaboration, which is consisted of 25 members from 9 institutions of four countries: Canada, Italy, Korea, and Japan.
2. PHYSICS MOTIVATION 2.1. Theoretical predictions The Chiral symmetry between quarks, building blocks of matter, and gluons interactions in Quantum Chromodynamics (QCD) is spontaneously broken in vacuum. The symmetry is restored in the high-energy hadron collisions, where the strong coupling constant (a,) decreases to reach an asymptotic small value. This decrease is predicted by QCD, and is experimentally proved to be correct. Many theories predict the partial restoration of the symmetry also in high-temperature as well as in high-matter-density environments [1,13]. Chiral effective Lagrangian theories and QCD-Sum-Rule calculations predict the change in the vacuum expectation values of the antiquark-quark condensates (Gq) depending on the environmental baryon density. The change is due to partial restoration of the Chiral symmetry breaking in such environment. This dictates to mass shift and/or width shift of the vector mesons such as p, w , and 4 even at the normal nuclear-matter density from the values in vacuum. Mass-shift formulae for vector mesons in nuclear medium are given by some model calculations such as: scaling hypothesis in Chiral effective Lagrangian by Brown-Rho [3], QCD sum rules calculation by Hatsuda-Lee [12], VDM calculation by Asakawa et al. [16], hadron rescattering calculation by Rapp et al. [17]. and the one that includes momentum dependence by Lee [la]. All predictions are summarized in the following expression,
mass in the nucleus, mv the p0 mass in vacuum, p* the matter where mv*is the density of environment, p the normal nuclear-matter density, and a a parameter which describes the magnitude of mass shift. In order to determine a , where it is predicted to be cr = 0.0-0.25, a feasible task for experimentalists at present is to determine it at the normal matter density. 2.2. How t o observe Mass shift of the vector meson in nuclear medium might be observed as effects in spacelike region by finding the modification of the meson propagator. The other, which takes place in time-like region, is to observe the decay of the vector meson in nuclear medium. We take this more direct way to determine the magnitude of the mass shift. One has to assume that it is possible to place a vector meson in the nucleus without losing its entity as a hadronic particle. The vector meson decays into particle pairs (e+e-, .rrfn-, K+Kd etc.). The decay takes place inside the nuclear sphere at a certain probability, and the decayed particles carry out the information of the parent mass in nuclear medium. The probability depends on vector-meson type, the energy, and the size of the nucleus as shown in Figure 1. By observing these particles, one can determine the mass, and compare it with the mass in vacuum. Table 1 lists the properties of the three lightest vector mesons p, w , and 4. In order to get a high decay probability inside the nucleus, the p0 meson with the shortest life time is the best solution. The photon is superior to the hadron as the method to produce the
Table 1 The lightest vector mesons. I)
Mass (MeV/c2) Width (MeV/c2) CT (fm) ree/r rhadrons/r
768 151.5 1.3 4.5~ 1.0
W
05
782 8.4 23.4 7~ 0.90
1019 4.4 44.4 3x 0.84
vector meson because of it weaker interaction to produce the vector meson even in the interior of the nucleus.
EP (GeV)
Figure 1. Decay probabilities of the p0 meson inside nuclei are calculated as a function of the energy. P is the velocity. At around 1 GeV, the difference between 3He and Pb, where the numbers are radii of the nuclei, is smaller than those a t higher energies.
M,,
(MeV/c2)
Figure 2. A schematical T + T - invariant mass spectrum, which is a superposition of the pn mass shape in vaccum and that with mass shift of AM. Broadening of the pn mass shape or a two-peak structure depending A M is the signal of the mass modification.
2.3. E x p e r i m e n t a l design (TAGX options) Bertin and Guichon [19] proposed a method of measurement of the vector-meson (VM) mass inside nuclear medium by using the vector meson photoproduction on nuclear targets. The procedure is: (1) produce a stable VM inside the nucleus by the photon, (2) the VM interacts with nuclear medium, (3) the VM decays inside the nucleus into two particles, (4) detect the particles to reconstruct the effective VM mass. No experiment using this method has been carried out up to now.
We have examined the conditions that interfere a prompt experimental test. They are: (1) to reduce escaping probability of the VM from the nucleus, and to give enough time interval interacting with the nuclear medium, the momentum of the VM should be low, (2) to reduce the distortion due to final state interaction (FSI), detect decayed lepton pairs. and (3) to enhance the decayed signals, the nuclear radius should be large. Decay This fact in addition to the small fractions of the VMs to lepton pairs are small photoproduction cross section requires a high-current, and high-duty factor electron beam producing the photons, which had not been available before the operation of the JLAB electron accelerator. This is the reason why the test has been delayed. Our strategy is to change the conditions at the minimum level (TAGX options): (1) use VM+r+.ir- decays which has a high decay fraction instead of e+e- pairs, and (2) to compensate this, use light nuclei instead of heavy nuclei to reduce FSI of the pions suffering before escaping from the nucleus. The decay probability in Figure 1 shows that the use of light nuclei is not bad option at lower energies. TAGX options are to measure the 3He(y, n+.ir-)reaction just at the p0 meson production threshold. 2.4. What is the signal ?
The VM photoproduced inside the nucleus decays both inside and outside. When the mass shift takes place inside the nucleus with a significant magnitude, the resulting mass spectrum is a superposition of the two mass spectra. Figure 2 shows a schematic such .irf n- invariant mass spectrum. In reality. the mass spectrum is integration over nuclear matter density, momentum space, and production points. 3. EXPERIMENT 3.1. Photon beam
ES supplied a tagged-photon beam of energies up to 1.2 GeV with a duty factor of less than 1 %. This small number has not been appropriate for the coincidence measurement such as the present ,oO-mesonproduction. An idea was proposed to extend the duty factor with minor investment by Yoshida et al. [20], and it was successfully realized in 1987. The tagged-photon beam with more than 10%-duty factor had been available, which was the best in the world in the energy region above 1 GeV. Figure 3 shows the accelerators in the world before the operation of MAMI-I1 and CEBAF. 3.2. TAGX spectrometer
A magnetic spectrometer with a large solid angle of .ir sr has been constructed as a general-purpose detector of photoreaction measurements by using the higher-duty factor beam in 1987. This TAGX spectrometer was used for an experiment that realized the design in the preceding section. It was for the measurements of the r+.ir-photoproduction on the 3He nucleus in the E, region close to the ,oO production threshold. No measurement of p0 photoproduction on any nuclei has been carried out so far in the E, region. The experiment used the tagged-photon beam and the large-acceptance TAGX spectrometer [21] at the 1.3-GeV ES. The photons ( 5 x 1 0 ~l / s in intensity) with energies of 0.80-1.12 GeV ( 5 MeV in the energy resolution) were incident on a cryogenic 3He tar-
CEBAFJ
zc.
4 -
0
3 -
W
-
x
2 -
B
0 ELSA :Bonn)
i
Frascati
.,
'87
A
1
-.~ ---
('
0 Tokyo
>O
--
-
MIT-Bates
o
~1
SPrinp-8
10
I
Duty Factor (%)
MAMl Mainz) ~
ai 100
Figure 3. The ES extended its duty factor from 1% up to 20% in 1987 in prior to other electron accelerators in the world. Those accelerators with 100% duty factor with the energies greater than 1 GeV were in operational several years after this ES operation.
get (0.35 g/cm2 in thickness). Four-momenta of photoproduced charged particles were measured simultaneously with TAGX. The methods were time-of-flight measurements for velocity determination and trajectory measurements in the magnetic field for momentumvector determination. The results of event-by-event analysis of the recorded two-chargedpion events such as the reaction point distribution which shows non-target background events and the mass of the spectrometer particles which show a clear peak of pions were used for identification of the 3He (y,.ir+nP) events [22] . Detection efficiencies of TAGX for the observed events were estimated by a detector simula! ! tion calculation, and they were used for correcting for detector biases and acceptances to obtain the .ir+n- invariant mass (MT,) distribution. 3.3. Calibration
The capability of the TAGX spectrometer as charged particle detector has been examined in several experiments, and the performance was well known on this stage [21]. The mass of the meson is reconstructed to be the invariant mass of two pions. Calibration of TAGX was made by looking for the narrow-width K, which decays into two pions. Taking into account the longer life of K, of cr=2.54 cm than that of pO, a cut is applied in the distance between the n+.ir- vertex from the beam (23cm) to enhance the K, contribution. MT, distribution (Fig.5) shows a clear peak. A gaussian fit t o the mass spectrum gives the peak position to be 500.61.4.6 MeV/c2 and the width 19.9f 5.4 MeV/c2, which are consistent with the K, assumption. We obtain absolute mass accuracy of less than 4 MeV/c2, and Albfn,=20MeV/c2, which are much smaller than the p0 width. 3.4. Data analysis
Data accumulation with TAGX was made at Tokyo in 1994 by the TAGX(ES-134) collaboration. The data were analyzed in Tokyo, and then delivered to collaborators in Regina, where another physics analysis has been performed by Kagarlis et al. [23]. In this paper, the results obtained mainly at Tokyo by Yamashita et al. [22] are presented.
Figure 4. A schematical layout of the TAGX spectrometer.
4. RESULTS AND DISCUSSION 4.1. P r o d u c t i o n mechanisms There are two types of p0 production mechanisms on the nucleus, i.e. the coherent production mechanism that leaves the target nucleus untouched, and the incoherent one that breaks the nucleus. In the former mechanism, the vector dominance model predicts that the photon fluctuates into the vector mesons in the nuclear field. The meson and the nucleus make an elastic diffractive scattering. All the nucleons in the nucleus contribute coherently to produce the meson. While the photon produces the meson also by the quasi-free production on the nucleon. The nucleons contribute incoherently. Neither data is available in the threshold region. p0 photoproduction threshold on the free nucleon is 1080 MeV (at the mass peak of 768 MeV/c2), and it is 870 MeV on the 3He nucleus for coherent production. The results obtained in our highest photon energy bin of 1040-1120 MeV, which is just above the threshold of both production mechanisms, are discussed in this section [22,28]. The invariant mass distribution, which has a broad peak structure, can be a superposition of several n+n- production mechanisms. They are (1) coherent (diffractive) and
300 400 500 600 700 800
M,,
(MeV/c2)
Figure 5. A .ir+n- invariant mass for the events with the vertex away from the beam (see text). A peak at around 500 MeV/c2 is the K,.
Table 2 Mass and width for the T + T - mass spectrum in the coherent process are reproduced by the Soding model. Mass (MeV/c2) Width (MeV/c2) x2 /dof No. of free parameters 747517 168+24 0.98 4
(2) incoherent (quasi-free) p0 meson production, and (3) the .ir'.ir- production in addition to (4) 357 production which is excluded from the discussion in the following by rejecting M, where m, is the pion the events whose missing mass (Mx) is higher than the Am, mass and M is the 3He mass. The rest of the events are categorized into two mechanisms according to their missing mass values: one is for the events which are consistent with the coherent (Mx- M within resolution), and the other is consistent with the incoherent.
+
4.2. Coherent production Figure 6 shows the invariant mass distribution of the T + T - system in the coherent kinematical region. The data may be reproduced by the p-wave Breit-Wigner mass formula,
r the full width of the resonance, where M,, is the invariant mass of the ~ + n system, and Mo the resonance mass. A Breit-Wigner fitting was made. The width obtained of 151121 MeV/c2 is consistent with the table value, but the mass peak of 706%9 MeV/c2 is significantly low (cf. Table 1). This cannot be interpreted to be the mass shift, because the shift should associate with the non-mass-shift peak which comes from the p0 decays outside the nucleus. Possible causes of the apparent mass shift are examined. There are several known massskewing effects such as the pw interference, the Ross-Stodolsky mechanism, rescattering, and the Soding model [25]. We used the Soding model which takes into account the
7
lo ytd->pOd
8
M,,
'
1
y 3 ~ e - > p 03 ~ (TAGX) e -----Ai,z
(MeVlc2)
Figure 6. A T + T - invariant mass distribution obtained in the photon energy range 1040-1120 MeV. Thick dashed curve is a fit with the p-wave Breit-Wigner mass formula to the invariant mass distribution. Thick solid curve is a fit with the Soding model, where the solid is for the vacuum-mass p0 and the dotted for the interference.
Figure 7. Cross sections for coherent production. Present data (solid circle) and the data on the deuteron [24] (solid depensquares). Dashed curve shows dence.
interference between the p0 production with the Drell type
T+T-
production.
ADrellis an interference between the Breit-Wigner and the Drell type background. The result of a fit using this model is shown in Fig. 6. It reproduces the present data with parametrization listed in Table 2. The results are consistent with the values of the p0 meson in vacuum, i.e. cr--0. Therefore, we observed the photoproduced p0 meson in the 3He nucleus at around the threshold for the first time. As for the production cross section, the value integrated over the measured mass range is shown in Fig. 7 , which is slightly larger than the total cross section for p0 production on the deuteron measured by Benz et al. [24]. When one assumes a mass number dependence of A2I3, two sets of the data are consistent each other as seen in Fig. 7. 4.3. Incoherent production Figure 8 shows the invariant mass distribution of the T + T - system determined in the incoherent region. The peak position and the width are different from the coherent distribution. An excess in the mass region 600-700 MeV/c2 is not reproduced reasonably well by neither the p-wave Breit-Wigner mass formula nor the Soding model. The difference might be filled with other mechanisms such as .ir+.ir- production through T A channel or the phase space production. Fittings including one or two of these mechanisms do not
improve the reproducibility. This is because no mechanism enhance the 600-700 MeV/c2 region. Another mechanism which might deform the invariant mass spectrum is proposed by Yamazaki-Akaishi [26]. They estimate the effects of the average potential energy of the target nucleon and the correlation with surrounding nucleons. A substructure in the mass region 500-600 MeV/c2 appears when the matter density is much higher than the normal density, that is not the case in this experiment. Saito-Tsushima-Thomas [27] predicts p0 Inass reduction in the 3He nucleus by using the quark-meson coupling model under the mean field approximation. According to the a meson mass reduction they get 7 2 5 ~ 7 3 2MeV/c2 as the pO meson mass, which is not large enough to explain the present low mass enhancement. A fit with the sum of the free ,oO mass shape and a mass-shifted. whose magnitude is a free parameter to be determined by the fitting, Breit-Wigner formula is tried. The results are shown in Fig. 9 and listed in Table 3. The inclusion of the mass-shifted p0 reduces the fitting X2/dof frorn 8.9 to 2.0. Obtained value is a = 0.15h0.01. The value and its error are subject to change when assumed conditions are not satisfied. (1) The width is also modified, (2) the collisional broadening is large, (3) momentum dependence of mass modification is taken into account. and (4) the Breit-Wigner formula does not work. 40 o Coherent Incoherent -Soedlng
30
0
TAGX ........
20
10 .............................
0 400
600
M,,
800
1000
(MeV/c2)
Figure 8. Invariant mass distributions obtained in the incoherent region (solid) and the coherent region (open). The drop higher above 850 MeV/c2 is due to the acceptance cut. A fitting with the Soding model does not work for incoherent data.
M,,
(MeV/c2)
Figure 9. Incoherent process. A fitting with a sum of the ,oO mass shape, a massshifted shape (first peak), and the phase space contribution is successful in reproducing the present data.
5. CONCLUSIONS Theoretically predicted mass modification of the p0 meson in the 3He nucleus has been tested by a tagged-photon experiment at Tokyo. Coherently produced pO mesons observed have the vaccum mass and the vacuum width, and these facts confirm no mass
Table 3 The n+.ir- invariant mass spectrum in the incoherent process is reproduced by a sum of three contributions: (1) pO,(2) mass-shifted pO,and phase space. Numbers without errors are fixed in the fitting. Width (MeV/c2) Relative pOfraction Mass (hleV/c2) (1) 655&8 151 82&14
modification in the coherent production mechanism. Incoherently produced nfn- mesons have the mass spectrum that is consistent with an assumption that the p0 meson mass reduction by 15 %.
REFERENCES 1. 2. 3. 4. 5. 6.
G.E. Brown, Nucl. Phys. A488 (1988) 689. H. Kurasawa and T . Suzuki, Prog. Theor. Phys. 84 (1990) 1030. G.E. Brown and M. Rho, Phys. Rev. Lett. 66 (1991) 2720. G . Chanfray, Z. Aouissat. P. Schuck, and W. Norenberg, Phys. Lett. B256 (1991) 325. M. Lutz, S. Klimt. and W. Weise, Nucl. Phys. A542 (1992) 521. M. Herrmann, B.L. Riman, and W. Norenberg. Z. Phys. A-Hadrons and Nuclei 343 (1992) 119. 7. M. Herrmann et al., Nucl. Phys. A560 (1993) 411. 8. G.E. Brown, M.Rho, and M. Soyeur. Nucl. Phys. A553 (1993) 7 0 5 ~ . 9. M. Asakawa and C.M. KO, Phys. Rev. C48 (1993) R526. 10. H-C. Jean, J . Piekarewicz, and A.G. Williams. Phys. Rev. C49 (1994) 1981. 11. Y. Koike, Phys. Rev. C51 (1995) 1488. 12. T. Hatsuda and S.H. Lee, Phys. Rev. C46 (1992) R34. 13. T. Hatsuda and T . Kunihiro. Phys. Rep. 247 (1994) 221. 14. T. Hatsuda, in these Proceedings. 15. R. Rapp, in these Proceedings. 16. M. Asakawa, C.M. KO, P. Lkvai, and J. Qju. Phys. Rev. C46 (1992) R1159. 17. R. Rapp et al., Xucl. Phys. A617 (1997) 472. 18. S.H. Lee, Phys. Rev. C57 (1988) 927. 19. P.Y. Bertin and P.A.M. Guichon, Phys. Rev. C42 (1990) 1133. 20. K. Yoshida et al., IEEE Trans. Nucl. Sci. NS-32 (1985) 2688. 21. K. Maruyama et al. (TAGX Collaboration), Nucl. Instrum. Methods A376 (1996) 335. 22. H. Yamashita. Ph. D. Thesis (in Japanese), Tokyo University of Agriculture and Technology (1997). 23. M.A. Kagarlis et al., Phys. Rev. C60 (1999) 025203-1. 24. P. Benz et al., Nucl. Phys. B79 (1974) 10. 25. P. Soding, Phys. Lett. B19 (1966) 702. 26. T . Yamazaki and Y. Akaishi, Phys. Lett. B453 (1999) 1. 27. K. Saito, Phys. Rev. C56 (1997) 566. 28. K. Maruyama, Nucl. Phys. A629 (1998) 351c.
Hadron and Nuclear Physics with Electromagnetic Probes K. Maruyama and H. Okuno (Editors) 2000 Elsevier Science B.V. All rights reserved.
'
Nuclear production of
4 meson at KEK
H. En'yo ", J . Chiba b , H. F'unahashia, H. Hamagaki, ', M. Ieirib, M. Ishinoa, S. Miharaa * , T . Miyashitaa, T. Murakamia R. Mutoa, M. Narukia, M. Nomachi d , K. Ozawaa, 0 . Sasakib, M. Sekimotob, H. D. Satoa, T. Tabarua, K. H. Tanakab, S. Yamadaa, S. Yokkaichia and Y. Yoshimuraa t (KEK-PS-E325 collaboration) "Department of Physics, Kyoto University, Kyoto 606-8502, Japan bKEK, 1-1 Oho, Tsukuba, Ibaraki 305-0801 Japan "CNS, University of Tokyo, 3-2-1,Midori-cho, Tanashi-shi, Tokyo, 188-8501, Japan dRCNP, Osaka University,lO-1 Mihogaoka, Ibaraki, Osaka 567-0047, Japan
+
We have observed q5 -+ K+K- decays in 12-GeV p A interaction for three different targets, polyethylene, carbon and lead. The experiment E325 is being performed at the primary beam-line E P l B at the KEK Proton Synchrotron. The data from the first physics run in 1997 are reported. The kinematical region of the observed q5 mesons covers 1 to 3 in /371ab where modification of q5 meson in nuclear matter is expected to be visible. The shape of the invariant mass spectra of the observed q5 mesons are consistent with that observed in free space within the statistics which are presently available. The observed mass-number dependence of the production cross section follows A" (a=0.98f 0.10). No significant p, dependence has been observed on the a parameter. 1. INTRODUCTION
In QCD, hadrons are described as a bound state of confined valence quarks which obtained the effective mass due to the spontaneous breaking of the chiral symmetry. This mechnism determines the properties of hadrons in free space, such as the mass and the decay width. If the vacuum structure was changed from that in free space, those properties should also change. The quark condensate (qq) is the order parameter of the chiral symmetry which is subject to change in hot (T# 0) or dense ( p # 0) matter providing a vacuum structure different from that in free space. Although (qq) is not an observable, the mass of vector mesons can provide a good measure of the mass of valence quarks[l]. An experimental observation of in-medium modifications of vector mesons is thus desirable to study the nature of QCD in the non-perturbative region. *Present Address: ICEPP, University of Tokyo, 7-3-1 Hongo, Tokyo 113-0033, Japan +Present Address: Xaxon Cooperation. 1-3-19, Tanimachi, Chu-ou, Osaka, Japan
In hot matter, (qq) as a function of temperature is expected to decrease rapidly to zero around the critical temperature T, where the QCD phase transition will occur and the chiral symmetry will be restored. The observed excess in lepton pairs by the CERN heavy-ion experiments may already hint a precursor of the symmetry restoration[2]. On the other hand, in dense matter, linear decrease of (qq) is expected theoretically [I]. Several experimental efforts, including this work, have been started to investigate effects of the partially-restored chiral symmetry at normal nuclear-matter density.
Table 1 Experiments to study nuclear density effect on vector mesons lace channel status Published KEK(Tanashi)-ES y + A -, p + X (p -+ T + V ) KEK-PS p A -+ 4 X ( 4 -t K+K-/e+e-) Running (this work) y + A - + $ + X ($-+K+K-) Ready to run Spring-8 GSIIHADES Preparation T A -+ w X (w + e+e-) GSI/FRS Ready to run d A d 3He + A*(bound)
+ + +
+ +
Ref. 131 [4] [51 161 171
Table 1 lists those experiments. The present work, KEK-PS E325, is characterized by the simultaneous measurements of the hadronic and the leptonic decay modes of the $. Based on the QCD sum rule, Hatsuda and Lee predicted[l] that the $-meson mass decreases as shown in Figure 1. According to this model, the mass decrease of 4 meson is in the range of 20 to 40 MeV at normal nuclear density po. Figure 2 shows the expected invariant mass spectra of the q5 for beryllium, copper and lead nuclei, according to the predictions of the mass shift in Fig.1. The Woods-Saxon type nuclear density distribution is employed in the Monte Carlo calculation which treats the $ with P Y ~1. The ~ ~dispersion of mass in media, which is experimentally measurable, is neglected in the calculation. In copper case, about 10% of 4 mesons are expected to decay inside the nucleus when any in-medium modification is absent. And in the invariant mass spectra, they make the second peak on the left side of the main peak which consists of the 4 mesons decayed outside the nucleus. It should be noted that in the K + K - channel, the invariant mass spectra could be deformed due to rescattering and modification of kaons in nucleus. Measurements in the ete- channel are thus more desirable. The decay branching ratios are sensitive to the mass shift of 4 meson and kaon because of the small Q-value, 32MeV, in the K + K - channel. Even if the kaon mass shifted and the q5 mass was unchanged, the effect would result in a change of the branching ratios. If the kaon mass were not modified and the $-meson mass decreased more than 32 MeV , the decay to K + K - would be forbidden in nucleus. Recently, Klingl et a1.[8] calculated the in-medium modification of 4 meson being at rest in nucleus. They obtained the mass shift of $ meson being consistent with the results of Hatsuda and Lee, with significant broadening of the width as large as 45 MeV/c2, 10 times larger than in free space. Their calculations was done for the 4 at rest, and the
-
Invariant Mass (GeV)
Figure 1. Hatsuda and Lee's predictions for the density dependence of the mass of $ meson. Dashed lines are the threshold of the K K channels. The two solid lines represent the range of the strange-quark condensate expected from the experimental observation (Y=~(SS)/((UG)(dl)).
Invariant Mass (GeV)
Figure 2. Expected signals for 3 different targets and for the different assumptions of the strangeness condensate as of Fig.1.
+
dispersion, i.e. the width change as a function of the momentum of 4 meson is yet to be worked out. This calculation, however, implies a significant increase of the decay rate in nucleus for $ mesons in flight and a modification of the invariant mass shape in the $ + K+K- channel. Experimentally a detection of 4 mesons at rest is nearly impossible since the daughter kaons decay immediately. Nevert,heless with the slower 4 mesons, the larger decay probability inside a nucleus and the lager nuclear matter effect are expected. It should be noted that the production of 4 meson with a nuclear target is not well studied so far. The cross section of particle productions with a nuclear target normally follows the relation, a(A) = a(1) x Aa, where A is the nuclear mass number. In the production of the J/+ and the Drell-Yan lepton-pairs, a is close to unity in high energy, as measured in the p+p- pairs by Binkley et al. in 300-GeV/c n A interaction[9]. To produce those particles it requires partonic interactions like g g -+ J / + g or q q --+ y*,i.e. the production is supposed to be perturbative and all the nucleons in a nucleus contribute equally to the production. In the production of n,p and w mesons a is almost 2/3, which corresponds to the projective surface area of nucleus. Those particles, consisting of light u(G)and d(d) quarks, are produced in the fragmentation, and the yield is supposed to be governed by the first collision of an incident proton whose mean free path in nucleus is much shorter than a typical nuclear radius. On the production of 4 meson Binkley et al. showed a to be 0.66 f0.03, close to 213, although their identification of the 4 meson was not perfect due to the poor mass resolution. Bailey et al. measured the mass number dependence of the $ production in the K+K- channel in 120-GeV/c p + A
+
+ +
+
interaction and obtained cr = 0.86 f 0.02 [lo]. Aleev et al. measured 4 -+ K + K - decays in 70-GeV n A interaction and obtained cr = 0.81 f 0.06 [ l l ] . All the 4 productions mentioned above were measured in the kinematical region of 0 5 XF I 0.3 and 0 5 p~ 5 1 GeV/c. No measurement of the target mass dependence of the 4 meson production had been performed in the region of XF 5 0 , close to the target rapidity. The simultaneous measurements of 4 -+ K + K - and 4 -+ e f e- channels being planned in the present experiment can cancel the effect of unknown production mechanism by taking the ratio of the A-dependence between the both channels, and the media effect on the produced 4 mesons can be extracted. It is, however, very important and interesting to study of the nuclear dependence of the production itself, to understand how the nature of the $ reveals in nuclear interaction. The experiment E325 has been designed to explore the physics discussed above. The spectrometer newly built for the experiment is located at the primary beam line of KEK 12-GeV Proton Synchrotron. The special emphases are put on the detection of slowly moving $ mesons (Pyla65 2 ) which have a larger probability to decay inside nucleus, and the simultaneous measurements of 4 -+ K + K - and 4 -, e+e- decays in the same apparatus. It should be noted that high quality primary beam on thin targets is needed to suppress the background from y-conversion in the e+e- channel. We started the construction of the spectrometer on June 1996. The engineering run was performed on November 1996, and the first physics run was executed in June 1997, focusing on the 4 -+ K + K - decay channel. The results from the run in 1997 are reported in this manuscript.
+
2. EXPERIMENT E325 AT KEK-PS
Figure 3 and 4 show a schematic layout of the experiment. Beam protons were delivered to three targets ( carbon, lead, and polyethylene ) placed in-line at the center of the spectrometer magnet. The spectrometer has two arms for kaon and electron detection, sharing the magnet and the tracking devices as common. The kaon arms cover from 112" to f54" horizontally and f6" vertically, where the horizontal angle was measured from the beam line and the vertical angle was measured from the horizontal plane. The electron arms have a much larger acceptance covering from f12" to 690" horizontally and f22" vertically. The trajectories and momenta of K+K- pairs and e+e- pairs from $-meson decays were determined with the Cylindrical Drift Chamber (CDC) and the Barrel Drift Chambers (BDC). Kaons were identified with the Time-Of-Flight (TOF) method using the Start Timing Counters (STC) and the Forward TOF Counters (FTOF) in the off-line analysis. The Aerogel cerenkov counters (AC) and the Hodoscope Counters (HC) were used with FTOF for the on-line trigger to select kaons. Electron identification was performed with the segmented Gas ~ e r e n k o vcounters (FrontGC / RearGC) and the lead glass EM Calorimeters (SideEMC / RearEMC). Details of the experiment can be found elsewhere[4]. To evaluate the achieved tracking performance, A -+ pn- decays were studied. In Figure 5, the invariant mass spectrum of pn- pairs are plotted. We required the decay points are more than 20 mm apart from the target. The clear resonance peak of A shows
.
Rear GC
magnet return yoke
Aerogel Cerenkov
T target chamber
I Figure 3. Top view of the E325 setup
near EMC
T"pl-
1
KDC Hodoscope
1 .
Figure 4. Side view of the E325 setup
the mass resolution of 2.2 MeV/c2. By rescaling this mass resolution by taking account of the differences in the Coulomb scattering effects and in the momentum distributions of protons, pions, and kaons, we evaluated the mass resolution of the q!~ + K+K- invariant mass to be 1.2 MeV/c2, which is good enough for the present analysis. The peak value of the Gaussian fit is 1115.11 MeV/c2, which is consistent with the mass of A in the literature[l2], 1115.68 MeV/c2.
Figure 5. The p r - invariant mass spectra. The line is the best fit result of a Gaussian with a linear background. A is seen at 1115.11 MeV/c2 with a resolution of 2.2 MeV/c2.
3. RESULTS AND DISCUSSION
The invariant mass spectra of the K + K - pairs are shown in Figure 6 for three different targets, polyethylene, carbon and lead. Clear resonance peaks corresponding to 4 + K + K - decays are observed. To evaluate the shape of the background, the event-mixing method was applied. We have assumed that the background is due to an uncorrelated pair of particles so that the two tracks from different events should have the same distribution as the background shape. The hatched areas in the Figure 6 are the shape obtained by the event mixing.
pigure 6 . ~h~ invariant K + K - spectra. In (A) we include all the three targets. The spectra (B);(C)and (D) are torresponding to polyethylene, carbon and lead targets respectively. The hatched areas are the backgrounds evaluated by the event-mixing method.
Figure 7. The resonance position (A) and width (B) of the observed 4 mesons. The line is drawn at the values in free space.
The spectra were fit with the sum of the Breit-Wigner function and the background shape. The resonance positions and the widths are plotted in Figure 7. The resonance positions obtained by the present work are consistent with the known value of 1019.4 MeV/c2. The widths of 4 meson from the polyethylene and the carbon target are also consistent with the natural width of 4 meson, 4.43f0.05 MeV/c2[12]. For the lead target case, the width looks larger but the present statistical significance is not enough to conclude the difference. We analyzed the mass number dependence of the production of 4 meson using the standard parametrization, a(A) = a ( A = 1) x A" , where A is the atomic mass number
45
of the target nucleus. The results are plotted in Figure 8 together with the best fit line.
r - u(Pb)/o(C) rctio n-0.98 0.10
*
a=0.97
* 0.13
1
Figure 8. Mass number dependence of the cross section. The solid line is obtained by the ratio of the lead to the carbon data. The dashed line is the best fit over the three points. The data point for hydrogen is obtained by the CH2-C subtraction. The obtained a parameter is plotted in Figure 9 together with other available data[lO, 111 as a function of XF of observed $ mesons. In spite of the large difference in incident energies and in kinematical coverages, the obtained a is consistent with those of other experiments. Table 2 summarized those a parameters. The region of the transverse momentum p~ is higher in the present experiment than in others. One has to consider that the Cronin effect[l3],characterized by the increase of a as a function of p ~ can , be the origin of the large a value obtained in the present experiment. In the upper part of the Figure 10 we plot, the p~ dependence of a parameter of the present measurement, together with the a measurements in the 4.3-GeV/cp+A -+ T++X interactions[l4]. In the pion measurements, the increase of a above p~ = 0.6 is attributed to the Cronin effect. We did not observe any increase of a in the $-meson production in the present data. The origin of the Cronin effect is not well understood so far. One of possible explanations is the change of the production mechanism from non-perturbative interaction to perturbative one. If we assume the cy value is determined with the Q2 involved in the interaction, the plot of p~ dependence of a mentioned above should be re-plotted as a function of transverse mass, m r . as is shown in the lower part of the Figure 10. The a value of the present data, then, is reasonably understood together with the cr increase seen in the pion data.
Table 2 Comparison of the a parameters and the kinematic regions.
+
4 (GeV)
12GeV p Pb/C(present work) 120GeV/c p Ta/Be[lO] 100GeV/c p Be/p[lO]
+ +
5.1 15.1 14.2
a 0.9850.10 0.86k0.02 0.9650.04
PT (GeV/c) XF 0.4< p~ <1.5 -0.8< XF <0.3 O< p~ <1 O< XF <0.3 O< p~ <1 0.11< XF <0.24
"' E
+ +
1 2 GeV 2 + A s x (our dcto) 0 43GeV/cp+A+n+X
02
F . . '02
, , 0 4
. 0 6
I 08
1
~ ' 2
1.4
rn, (GeV/cZ)
Figure 9. The XF dependence of the a: parameter
~ of Figure 10. The p~ and r n dependence the a: parameter.
The fact that the a parameter is close to unity at this energy region fi = 5.1 GeV) is surprising. It is generally believed that incident energy of 12GeV is dissipated rapidly and the nucleons in the back half of the nucleus can not contributed to the particle production. The present results suggests that the production mechanism of 4 mesons is similar to that of J/$ at higher energy, which is visible because the production from the fragmentation is largely suppressed due to the significant heavy mass of the sS and the OZI suppression. The result that the a: is close to unity suggests that the q5 mesons are generated at anywhere in the lead nuclei. And the average pylabof the observed C#J mesons from the lead target is 1.8. The in-medium decay probability can then be deduced with an assumption of in-medium width of d meson. If we take the width 45 MeV from the theoretical
I
calculation of Klingl et a1.[8], The in-medium decay probability becomes as large as 40%, and the resonance peak shape should deformed consequently. The present observation is not consistent with such a picture. As was explained in the Section 1, their calculation was made for the q5 at rest, and the decay width is subject to change as the momentum of the q5 increases. Their calculation should be considered as the maximum broadening of the q5 meson in nuclear media, because the major mechanisms of the broadening is attributed to the K--nucleon interactions in nucleus and the interactions gets weaker as the kaon momentum increases. Further data accumulation and data analysis of the present experiment will enable us to determine the in-medium decay rate and in-medium decay width simultaniousely. 4. CONCLUSION AND OUTLOOK
The present experiment is the first attempt to measure the production and the decay of $ mesons in the kinematical region where the effect of nuclear media should play an important role. The experiment is on going at KEK with protons of 12-GeV incident energy, which is the lowest& amongst the $-meson measurements in the hadron induced nuclear reactions. The data from the first physics run in 1997 were reported. The K + K - invariant mass spectra for three different targets, carbon, polyethylene and lead, have been obtained. The widths of the resonance are 4.9h2.8 MeV for carbon and 11.5f 5.0 for lead, for the 4 mesons with /3yLabin the region from 1 to 3. The widths as well as the resonance positions are consistent with those in free space, and any significant resonance-shape modification, which is expected from the in-medium modification of hadrons, was not seen in the statistical accuracy which is presently available. The mass number dependence of the production cross sections in the $ + K + K channel has been also obtained using carbon and lead targets. The parameter a! is found to be 0.98 f 0.10, very close to unity. This value is consistent with the ones measured at higher incident energies from 70 to 120 GeV using hadron induced nuclear reaction. The present measurement has firstly covered the region of XF < 0 closer to the target rapidity, while other measurements covered only XF > 0. No significant p~ dependence has been observed on the a! parameter. In other words, the steepness of the A-dependence of the $ production at this energy is not likely from the Cronin effect. We have the data being analyzed from the runs in 1998 and 1999. Further data accumulation is scheduled in 2000. The better statistics in the q5 + K + K - decay mode as well as the new information from the $ + e+e- decay mode will soon be available.
REFERENCES
1. T. Hatsuda and S. H. Lee, Phys. Rev. C46 (1992) R24. 2. G. Agakichiev et al., Phys. Lett. B422 (1998) 405. 3. K. Maruyama, proceedings of this symposium. G.J. Lolos et al. (TAGX Collaboration),Phys. Rev. Lett. 80 (1998) 241-244. 4. E325 proposal, http://www.pn.scphys.kyoto-u.ac.jp/- enyo/e325/phi-root.ps.Z). S. Yokkaichi, Memoir of The Faculty of Science, Kyoto University, (doctor thesis), in print.
5. T . Kinashi, K. Takanashi, M. Fujiwara, T. Hotta, T. Nakano, In Hyogo 1997, Exciting physics with new accelerator facilities 55-62. 6. W. Schoen et al.( HADES Collaboration ), Acta Phys. Polon. B27:2959-2963,1996. 7. GSI/SIS proposal S214, Search for bound q- and w- nuclear states using the recoilless (d:3H e ) reaction. 8. F . Klingl, T. Wass: W. Weise, Phys. Lett. B 431 (1998) 254. 9. M.Binkley et al., Phys. Rev. Lett. 37 (1976) 571. 10. C. Daum et al., Z. Phys. C 18 (1983) 1. R. Bailey et al., Z. Phys. C 22 (1984) 125. 11. A. N. Aleev et al., JINR, Dl-90-168(1990), Dubna. H 12. Europ. Phys. J. C3 (1998) 376. 13. J . W. Cronin et al., Phys. Rev. D l 1 (1975) 3105. 14. M.Ono et al., Phy. Lett. B84 (1979) 514.
Hadron and Nuclear Physics with Electromagnetic Probes K. Maruyama and H. Okuno (Editors) 2000 Elsevier Science B.V. All rights reserved.
"
Lepton pair spectroscopy with HADES at GSI J. Friesea, for the HADES collaboration "Physik Department E12, Technische Universitat Miinchen, D-85748 Garching, Germany The High Acceptance DiElectron Spectrometer HADES is presently set up at GSI, Darmstadt, for the systematic study of medium modifications of hadron properties. Short lived hadrons will be produced in nuclear matter from normal up to medium baryon densities utilizing pion, proton and heavy ion induced reactions. Electron pair spectroscopy will be used to measure resonance widths and effective masses of vector mesons as well as transition formfactors of neutral mesons and baryon resonances in the low mass region M 5 1 GeV/c2. The spectrometer is designed for an invariant mass resolution AMtn,/Min, 21 1% (g),a large signal to background ratio, and a large acceptance Rpi, 2: 40%. 1. INTRODUCTION
Over the past two decades the properties of strongly interacting particles embedded in hot and dense matter has increasingly attracted both, experimental and theoretical efforts (for recent reviews see [I-31). The availability of relativistic and ultrarelativistic heavy ion beams allows in a unique way to prepare and explore such states of matter in earthbound laboratories. The long range goal of such studies is to verify experimentally a state of matter, in which chiral symmetry, one of the fundamental symmetries in nature, is restored. Under normal conditions this symmetry is broken and the quark condensate < q, q >, the order parameter of the chiral phase transition, has a non-zero vacuum value. Various approaches within the theory of strong interaction (QCD) predict a reduction of the quark condensate with increasing temperatures and particle densities finally leading to at least a partial restoration of that symmetry [4-121. One expected consequence of this phenomenon is the modification of hadron properties (masses, decay widths, etc.) which can already become significant even for hadrons embedded in groundstate nuclear matter. However, the propagation of hadrons in the medium leads to additional modifications in the particle spectral functions and screens the fundamental QCD effects. Experimental verification of these modifications is therefore one of the main guidelines in nowadays heavy ion physics. The light and short lived vector mesons p, w , and q5 are particularly attractive for such investigations, as they may be abundantly produced in relativistic nuclear collisions and decay at least partly within the hot and dense reaction volume. The w meson is of particular interest, since a modification of its narrow decay width [13,14] might be observed already in cold nuclei when the w is produced in T - induced reactions under adjusted
kinematical conditions [15]. Experimentally, the invariant mass reconstruction and the particle spectral functions can be obtained preferably from lepton pair decay spectroscopy, as electrons and myons experience only very little rescattering in the hadronic medium. Up to now, lepton pair spectroscopy has been performed for light and heavy collision systems at incident energies of 1 - 4AGeV [16,17] and at 160 - 200AGeV [18-231. All experiments observe in the low invariant mass range 200 MeV/c2 < M,,, < 600 MeV/c2 a significant excess of yield as compared to current theoretical calculations [24]. However, the limited invariant mass resolution and background separation has prevented so far from unambiguous conclusions with respect to medium modifications. To improve and extend the existing experimental evidence the new High Acceptance Di-Electron Spectrometer HADES [25,26] is presently being set up at GSI, Darmstadt. This second generation experiment aims at in-medium meson spectroscopy in the medium temperature and density domain with unprecedented invariant mass resolution and signal to background discrimination. After a short motivation for our choice of the GSI-SIS energy regime we give a short description of the spectrometer and report the present status of the experiment. Finally we present simulated dilepton spectra for two examples of reactions and summarize the physics programm for the first runs. 2. L E P T O N P A I R S P E C T R O S C O P Y I N T H E SIS E N E R G Y R E G I M E
In central heavy ion collisions of 1 - 2 AGeV incident energy a hadronic fireball is formed, with temperatures up to about T = 50 - 90 MeV and densities ranging up to 3 times normal nuclear matter densities. At this energy density the system is still in a "pretransition" phase, i.e. far beyond the one needed for the deconfinement and chiral phase transition. Nevertheless, these conditions are advantageous for studies of medium modifications, since in contrast to much higher energies, where the fireball expands violently and experiences rapid changes in temperature and density, the formation of the reaction volume lasts for as long as 10 fm/c. This transient state is already accompanied by a significant reduction of the quark condensate down to as much as 21 30% of its vacuum value [27]. Furthermore, only low lying resonances ( A , ..) and light mesons (n, v, p, w ) are produced and propagate in a baryonic environment with small pion content. In this way all background due to pion induced secondary reactions is minimized. Electron pairs are not only produced in purely binary vector meson decays but come also from 3-body no, A, 7, ...- Dalitz decays and from the nucleon-nucleon Bremsstrahlung. To disentangle these continuum type of contributions from the measured spectra and to allow for a sound interpretation of the anticipated experimental findings in heavy ion collisions, additional knowledge on the fundamental electromagnetic properties of all involved hadrons is needed. These properties can be be gained from exclusive measurements of elementary reactions in n- and nucleon induced reactions on cold nuclei. The importance of this aspect may be seen already from the fact that the interpretation of the DLS yield [16] as arising from an enhanced 17- Dalitz decay would be inconsistent with results obtained by the TAPS collaboration for r]- production in C C and Ca Ca collisions [28]. The GSI accelerator facility offers a unique variety of projectile beams ranging from pions to Uranium ions and allows for a systematic study of lepton pair production with various projectile target combinations in a broad rapidity and transverse momenta regime.
+
+
The available projectile energies can be chosen such that certain reactions channels (multi pion, higher resonances, etc.) are either below threshold or are populated less abundantly as compared to higher energies. Consequently, TO -+ y~ and 7 --t yy decays with subsequent pair conversion occur less frequently and hence limit the combinatorial e+ebackground such that a large signal / noise ratio > 1 can be obtained, in particular in the mass region of the p- and w meson. Therefore it is expected that systematic precision measurements with a mass resolution comparable to the natural resonance width of the wmeson (F = 8MeV/c2) might allow to disentangle the underlying QCD phenomena from the background processes and perhaps verify a partial restoration of chiral symmetry. 3. THE HADES SPECTROMETER 3.1. General properties The design of the spectrometer is governed by the high-multiplicity environment of heavy ion collisions, the production cross sections close to threshold, and the small branching ratio for the dielectron decay channel due to the electromagnetic coupling constant. Au collisions will produce an e+eFor example, only one out of lo5 - lo6 central Au pair from i.e. a p meson decay. For this reason, the key features of the new instrument are:
+
A large acceptance
E
40% combined with a rate capability
A resolution for invariant mass reconstruction AMi,,/Mi,,
Y
A signal to background ratio > 1 for invariant masses up to M
E
lo6 s-'
1% ( a ) .
-- 1GeV/c2
High granularity for multi hit capability and redundant track reconstruction. The spectrometer (see left panel in Fig. 1) is azimuthally segmented into 6 sectors and 0 85'. Four planes of Multi-wire Drift Chambers (MDC) tocovers polar angles 18' gether with a 6-coil superconducting magnet form the magnetic spectrometer for charged particle momentum measurement. A RICH (Ring Imaging Cherenkov) counter surrounding the target is used for electron identification [29]. A set of electromagnetic PreShower detectors (at polar angles below 45O) and a Time-of-Flight (TOF) wall provide additional lepton identification and a multiplicity signal to trigger on central collisions. The areal mass density of all detector materials is optimized as to limit the effective radiation length pd/Xo E and hence to keep multiple scattering and combinatorial background from external conversion low.
< <
3.2. The magnet and the "low mass" drift chambers The toroidal magnet is formed by 6 superconducting coils producing a maximum field of 3.5 Tesla with a current density of 120 A/mm2 per coil. For a magnetic field region of 0.5 m length the transverse momentum kick for leptons is less than 100MeV/c. Under these conditions the required momentum resolution can only be obtained by a position resolution of 11 100 p m and minimum multiple scattering in the detector materials. The tracking system consists of 24 trapezoidal multi cell drift chamber modules arranged in a frustum like geometry to form 4 tracking planes of increasing size. Each module
MDC
-
Beam
-.
Figure 1. The new dielectron spectrometer HADES at GSI. Leftpanel: Schematic cross section showing the main components and detectors. The system is designed with a sixfold symmetry around the beam axis. The magnet coils are projected into the plane. Right panel: Upstream view of the present spectrometer setup with the superconducting toroid in the center. Two sectors are equipped with detectors. Four driftchambers and the RICH are mounted inside the magnet. The outer detectors (PreShower and TOF) are pulled back to the service position.
consists of six sense-field wire layers separated by cathode wire layers and oriented at different angles with drift cell sizes increasing from 5 x 5 mm2 (plane 1) to 14 x 10 mm2 (plane 4). The use of 20 pm sense wires, bare aluminum cathode and field wires ( 80 resp. 100 p m ), and a He-iC4Hlo (60140) gas mixture keeps the radiation length and multiple scattering on a tolerable level. The intrinsic position resolution of a full size chamber (plane 2) was measured with 2.1 GeV/c protons and determined to < 70 pm(a) [30] utilizing both, a self tracking method in subsequent cells and external silicon p-strip trackers. This spatial resolution ensures the required precision in momentum measurement of about Aplp 5 1.5% ( a ) . 3.3. Detectors for electron identification The RICH consists of a gaseous decafluorobutan (C4FIo)radiator around a segmented target, a spherical VUV mirror, and a fast gaseous photon detector separated by a CaF2 window from the radiator gas. The Cherenkov photons are reflected and focused by the mirror t o rings of roughly constant diameter (E 5 cm for all polar angles) onto a multiwire proportional chamber (MWPC) operated with pure methane and a solid CsI photon converter covering the cathode pad plane [31]. For the given geometry about 12 to 17
Cherenkov photons will be detected for each incident lepton, corresponding to a figure of merit No = 108. The Time of Flight (TOF) wall is designed to discriminate electrons from pions up to 0.5 GeV/c and from protons up to 2 GeV/c. Simultaneously the device will be used as a multiplicity trigger array for central heavy ion collisions. The whole system consists of altogether 1148 scintillators arranged in an outer (# > 45O, 648 sc.) and an inner (0 < 45'; 500 sc.) wall. The detector is calibrated with a UV nitrogen laser coupled to the scintillators by means of quartz fibres. The start detector will be placed in-beam close to the target and will consist either of a diamond strip detector (for heavy ions, a 2i 50ps) or a scintillating fiber hodoscope (for hadrons). A time resolution loops 5 at,f 5 160ps has been measured with a full size detector [32],corresponding to an electron identification efficiency of 96% in the low momentum region at large polar angles. With ~ has been obtained. Including the modified prototype modules a resolution 0 ~ =072ps contribution from the start detector this would lead to OTOF N 90ps for the whole system. For particles of high momenta emitted at polar angles 19 5 45' the electron - pion separation is improved by an additional electromagnetic PreShower detector. The detector consists of two lead converters enclosed by three layers of wire chambers with pad readout operated in a limited self-quenching streamer mode. A comparison of signals from the preand the two post-converter layers allows for separation of showering electrons and nonshowering hadrons. The granularity (6192 pads per layer) limits the double hit probability to < 2% for central Au Au collisions at 1 AGeV. With a full size detector module a detection efficiency of 82% for single electron showers has been obtained from 800 MeV electron beam measurements. Investigations with proton beams in the momentum range of interest ( 5 2 GeVlc) lead to a hadron / electron misidentification probability of < 10%. Only for lower momenta (200 MeV/c) the hadron / electron misidentification probability reaches li 25%, while the pure electron identification efficiency is still 40%.
+
3.4. Trigger and data aquisition system The trigger system [33]is organised in three levels. The first level trigger is obtained by a fast multiplicity signal from the TOF wall and selects central collisions. From these the second level selects di-electron candidates with a minimum opening angle by matching the positions of recognised rings in the RICH with those in the TOF and PreShower detectors. Finally the third level suppresses background by making use of an online tracking analysis from driftchamber signals. With a beam intensity of lo8 s-l and a 1% interaction target about lo5 s-' events with lepton content are expected. Therefore the signal processing and data acquisition is laid out to accept the signals from all 73,000 detector channels every 10 ,us. The primary data rate of 4 Gbytels is reduced to a maximum of about 4 Mbytels for Au Au collisions by online hit pattern and data analyis performed in hardware image processing units. The data acquisition system [34] is based on a switched ATM network connecting the VME crates for detector readout and the 2nd level trigger processor with the eventbuilder workstation and an optional 3rd level trigger computing stage. Subevent data of each detector are collected for a large number of 2nd level triggers before they are sent asynchronously to the eventbuilder. The system allows for a sustained transfer rate of 14 Mbytels.
+
3.5. Status of installation and first results Although the spectrometer is still in the set-up phase, many components have been already installed (see right panel in Fig. 1) and tested in commissioning runs. The superconducting magnet has reached full field (B,,, = 3.8 T ) at a current of 3500 A and the field mapping measurements showed good agreement (< 1%) with the calculations. The cave, beamline, and mainframe assembly is finished and two of the six sectors are meanwhile equipped with detectors including the new readout electronics, part of the trigger system, and the ATM network for data acquisition. The integration of the remaining detectors and components is presently under way and, after further commissioning, first physics runs are planned for autumn 2000.
+
Ti at E = 1.75 Figure 2. Response of the RICH in central heavy ion collisions Ar AGeV. Left panel: Ring image measured for one electron or positron from external pair conversion in the target or radiator gas. A coincidence track signal is found in the drift chamber. Note the low background. Right panel: Distribution of ring centers found in one of the six sectors of the RICH for about 2 . lo5 central collisions.
The two sector setup has been operated successfully in two commissioning runs with heavy ion reactions C+C at E = 1.2 AGeV and Ar Ti at E = 1.75 AGeV. All detectors showed stable operation at beam intensities up to lo7 s-' and a performance close to or even beyond the design values. The available trigger modules and the ATM data acquisition system worked reliably and allowed for first data taking. The segmented (8 strips) CVD diamond detector used as in-beam T O F start detector withstood beam intensities up to lo8 s-l with a time resolution a = 50 ps. For the T O F detectors typical time resolutions of a = 100 - 150 ps were obtained. The drift chambers showed routinely a noise level of < 40mV (peak-to-peak) while signals from minimum ionizing particles saturate the amplifiers at 11 400 mV. The PreShower detector was read out successfully utilizing the new image processing unit. The electron identification capability
+
+
of the RICH is illustrated in Fig. 2. In a preliminary analysis of 2 . lo5 central Ar Ti collisions at E = 1.75 AGeV about 2500 candidates for Cherenkov rings were found, in reasonable agreement with simulation results. The corresponding electrons and positrons come mostly from external pair conversion of y-s in the target and the radiator gas. In almost all events the ring candidates (left panel of Fig. 2 ) were not obscured by electronic noise or signals from charged particle background. The extracted ring centers (right panel of Fig. 2 ) show a rather homogeneous distribution declining with increasing polar angle as expected from simulations. A more quantitative analysis of the available data is presently under way. 4. HIGH QUALITY INVARIANT MASS SPECTRA
The expected quality of dielectron spectra has been studied by detailed Monte Carlo simulations [35,36] anticipating the performance of detectors and trigger as discussed in the preceding chapters. The following two examples may give an impression of the capabilities of HADES and the perspectives for meson spectroscopy. 4.1. Exclusive q- Dalitz decay studies in p p collisions The low mass (100 M e v / c 2 < Mt,, < 500 M e V / c 2 ) range of dielectron spectra in nuclear collisions is characterised by comparatively large production cross sections and is dominated by contributions from nucleon - nucleon Bremsstrahlung, the no- and 7-Dalitz decays and - less pronounced - the A- Dalitz decay. In p p reactions the electron background from secondary sources can be kept at minimum. Since the excellent time resolution of the TOF wall allows also for n - p discrimination. both, the leptons and the protons can be identified and detected simultaneously. For these events the reconstruction of a Dalitz decay is kinematically complete without requiring an additional photon detector. Applying a cut on the corresponding particle mass in the p p missing mass distribution suppresses the contribution from background sources considerably. Measurements at incident energies below and slightly above the qproduction threshold will allow to study the continuum distributions above the no- mass. The production of q- mesons in p p reactions has been studied extensively [37-391 and the inclusive cross sections are well known. The two published measurements of the qDalitz decay are based on a sample of 80 counts [40]and on a sample of 2366 counts [41], the latter corresponding to a hypothetical background free yield of 106 signal pairs. For the A- Dalitz decay experimental cross sections may be deduced from the DLS data at energies below the q- production threshold Epp < 1.27GeV [17]. In the simulation the q production is assumed to be mediated by the N*(1535) resonance : pp -+ pN*(1535) -+ ppq -+ ppe+e-(7). For the Dalitz decays the y goes undetected. Considerable background of electron pairs will come mostly from pp --+ ppno -+ p p y y with subsequent photon conversion and the no- Dalitz decay pp -+ ppno -+ pp,l)ye+e-. In the q mass region also pp -+ pp2n0 -+ pp7e+eV ye+e- reactions will contribute. The expected electron pair invariant mass distribution for the 7 Dalitz decay is shown in Fig. 3. The combinatorial background from external two photon conversion (no-+ y y ) is not included but is estimated to be small. Once measured, the observed electron pair spectrum for the 7-Dalitz decay branch can be used to calibrate the spectrometer acceptance, to check the contributions from other sources (A- Dalitz, combinatorial background, etc.), and to
-.
___-_L-
Au+Au 1 AGeV
I
10
1
2n0 M, cut 3
0
,.
-
0 '
~-
,
O.?
.I
,, 0.3
. . ! . .~,-.-A Me+e-[G~VC-'1 0 4
e'e- Mass
0.5
0.6
0.7
(cev)
-
Figure 3. Simulated invariant mass spectra of e+e- pairs from light and heavy collision systems. Left panel: Yield of e+e- pairs from lo6 7- Dalitz decays (line) expected for cx 100h of pp collisions at E, = 2 GeV. The small background from pp27ro (dotted) reactions is further reduced to the dashed curve by requiring the missing mass of the Au collisions ppq- reaction. Right panel: Relative yield of e+e- pairs per central Au at E, = 1AGeV. The contributions from the various sources are indicated. Medium modifications are not taken into account. Note the low combinatorial background at high invariant masses.
+
deduce the electromagnetic formfactor of the 7 meson to a better precision
+
4.2. Electron positron pairs in Au Au collisions Central collisions of the heaviest ions are expected to produce the largest space-time volume of compressed nuclear matter. Simultaneously they are the most challenging systems to be investigated. The invariant mass spectrum of lepton pairs produced in central Au Au collisions at E = 1AGeV was obtained from a GEANT simulation and is shown in the right panel of Fig. 3. The events have been generated in a BUU transport code [42]without taking any medium modifications into account. The lepton pairs originate from decaying sources with Maxwellian momentum distribution in a fireball of about 65 MeV temperature. The different contributions from continuum sources (pnBremsstrahlung and Dalitz decays) and from purely 2 particle resonance decays form a cocktail which shows a significant structure in the region around the masses of free w and 4 mesons. It becomes obvious that the good suppression of combinatorial background and the high mass resolution should allow to detect any mass shift or resonance broadening, in particular of the w. A detailed and quantitative analysis will be carried out by systematic investigations as function of transverse momentum and rapidity and by further exclusive
+
meas~irementsof the Dalitz decay channels. The latter cxperiments are planned for the future utilizing coir~cidcricemeasurements with the photon spectrometer TAPS and/or the GSI large area neutron detector LAND. 5. CONCLUSIONS AND OUTLOOK
HADES is a high rate, high acceptance apparatus for electron pair spectroscopy with < 1% invariant mass resolution and a fast and selective multi level trigger scheme. Presently, two out of six sectors of the spectrometer are completed and first commissior~ingruns have been performed. Results obtained with this setup as well a,s with full size detector prototypes bcfore dcmonstratc the feasibility of the cxpcriment and in particular the efficient and redundant lepton identification. The spectrorrleter will be completed in 2001, with first physics runs beginning a,lrcatly in 2000. Thc physics program is broad and includes the study of lepton pair emission in relativistic hcavy ion collisions, dielect,ron productiori in elemcntary reactions and experiments aimed at studying the structure of hadrons. Particular attention will be paid to the decay properties of vector mesons with respect to rnass shifts and resonance broadening. The shape of thc invariant mass spectrum will give access to electromagnetic transition form fact,ors of neutral mesons. The long-range aim is to better describe the dynamical propcrties of hot and comprcssed hadronic matter, an environmerlt in which partial restoration of chiral symmetry is expected to show up and information about the origin of hadron masses may be obtained. The experimental program of HADES will focus first on pp collisions and light heavy ion systems to verify the yield enhancements found in the DLS measurements. When finally the designed invariant resolution becon~esavailable, dedica,ted investigations of thc w meson decay are foreseen utilizing both, pion and heavy ion induced reactions. The series of experinients planned will exploit the full range of primary and secondary bearns available at GSI, including pions and heavy ions up to uranium. Future combinations with an clectromagnetic calorimeter and/or a neutron detector for missing mass measurements will allow more exclusive measuremerits arid - for a long tirne - guarant,ec substantial scientific output.
ACKNOWLEDGEMENTS The collaboration gratefully acknowledges the support by EC (HTM, TMR program), ASCR, BMBF, DAAD, DFG, DLR (WTZ), GSI, IN2P3, INFN, INTAS, NATO, as well as the local support in all collaborating institutions. REFERENCES C.M. KO, V. Koch and G.Q. Li, Annu. Rev. Nucl. Part. Sci. 47 (1997) 505 W. Cassing and E.L. Bratkovskaya, Phys. Rep. 308 (1999) 65 R. Rapp and J. Wambach, hep-ph/9909229 G.E. Brown and M. Rho, Phys. Rev. Lett. 66 (1991) 2720 T. Hatsuda and S.H. Lee, Phys. Rev. C 46 (1992) R34 6. M. Herrmann, B. Friman and W. Norenberg, Nucl. Phys. A560 (1993) 411 7 . C.M. Shakin and W.-D. Sun, Phys. Rev. C49 (1994) 1185 8. J. Engels, F. Karsch and K. Redlich, Nucl. Phys. B435 (1995) 295 9. K . Saito. k. Tsushima and A.W. Thomas, Nucl. Phys. A609 (1996) 339 10. F. Klingl and W. Weise, Nucl. Pllys. A606 (1996) 329
1. 2. 3. 4. 5.
11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42.
Y. Koike and A. Hayashigaki, Prog. Theor. Phys. 98 (1997) 631 R . Rapp, G. Chanfray and 3. Wambach, Nucl. Phys. A617 (1997) 472 F . Klingl and W. Weise, hep-ph/9802211 M. Effenberger et al., Phys.Rev. C60 (1999) W. Schon et al., Acta Phys. Pol. B27 Vol. 11 (1996) R . J . Porter et al., Phys. Rev. Lett. 79 (1997) 1229 W.K. Wilson et al., Phys. Rev. C57(4) (1998) 1865 G. Agakichiev et al., Phys. Rev. Lett. 75 (1995)1272 T h . Ullrich e t al.. Nucl. Phys. A610 (1996) 317c, and A. Drees, Nucl. Phys. A610 (1996) 5 3 6 ~ G. Agakichiev et al., Phys. Lett. B422 (1998) 405 T . Akesson et al., Z. Phys. C68 (1995) 47 M.Masera: Nucl. Phys. A590 (1995) 93c M.C. Abreu e t al.. Nucl. Phys. A590 (1995) 117c E.L. Bratkovskaya, W . Cassing, R. Rapp, J. Wambach, Nucl. Phys. A634 (1998) 168 HADES, Technical Proposal: GSI 1994 J . Friese et al., Nucl. Phys. A 654, (1999) 1017c B. Friman, W. Norenberg and V.D. Toneev, Eur. Phys. J . A3 (1998) 165 R . Holzmann et al., Phys. Rev. C56 (1997) R2920 K. Zeitelhack et al., Nucl. Instr. Meth. A 433 (1999) 201 C. Garabatos et al., Nucl. Instr. Meth., A 412 (1998) 38 J . Friese et al., Nucl. Instr. Meth. A 433 (1999) 86 Agodi et al., IEEE Trans. Nucl. Sci., 45, No.3 (1998) 665 M. Petri et al., Proc. ofXXXVIInt. Wint. Meet. Nucl.Phys., Bormio, Italy (1998), 622 M. Miinch et al., Proc. ofXXXVIIInt. Wint. Meet. Nucl.Phys., Bormio, Italy (1999), 509 R. Schicker et al., Nucl. Instr. Meth., A 380 (1996) 586 H. Neumann et al., Acta Phys. Slov. 44 (1994) 195 H. Calkn et al., Phys. Lett. B366 (1996) 39 A.M. Bergdolt et al., Phys. Rev. D48 (1993) R2969 E . Chiavassa et al., Phys. Lett. B322 (1994) 270 M.R. Jane et al., Phys. Lett. 59B (1975) 103 and erratum Phys. Lett. 73B (1978) 503 G. Agakichiev et al., Eur. Phys. J. C4 (1998) 231 G. Wolf et al., Nucl. Phys. A517 (1990) 614
111. NUCLEON RESONANCES IN NUCLEI AND RELATED TOPICS
ThisPageIntolyLfBk
ThisPageIntolyLfBk This Page Intentionally Left Blank
Hadron and Nuclear Physics with Electromagnetic Probes K. Maruyama and H. Okuno (Editors) 2000 Elsevier Science B.V. All rights reserved.
'
Sll (1535) resonance in nuclei H. Yamazakia * , T . Yoritaa t : T . Kinoshitaa, T . Okudaa, H. Matsuia, T . Maruyamaa J. Kasagia, T . sudab 5 , K. Itohc, T. Miyakawac, H. Okunod, H. Shimizue t H. Y. Yoshidae and T . Kinashie
,
"Laboratory of Nuclear Science, Tohoku University, Sendai 982-0826, Japan bDepartment of Physics, Tohoku University, Sendai 980-8577, Japan 'Department of Applied Physics, Tokyo University of Agriculture and Technology, Koganei, Tokyo 184-8588, Japan dKEK-Tanashi, Midori-cho, Tanashi, Tokyo 188-0002, Japan eDepartment of Physics, Yamagata University, Yamagata 990-8560, Japan In order to study the S11(1535) resonance in the nuclear medium, total cross sections of the (7,Q ) reactions on C, A1 and Cu have been measured for photon energies between 680 and 1000 MeV. A broad resonance structure due to the excitation and decay of the ,311 resonance in the nucleus has been clearly observed for the first time. The apparent energy and width of the resonance are 900 and 300 MeV, respectively. Model calculations based on the quantum molecular dynamics (QMD) have been performed. The Q photoproduction on nuclei can be basically understood as the quasi-free reaction with the reasonable medium modifications. 1. INTRODUCTION
The property . of nucleon resonance in nuclear medium is one of the most important issues in the nuclear physics. Recent measurements of the total photoabsorption cross section have shown clear evidence of nuclear medium effects on nucleon resonances /1,2]. . . The DI3(1520) and FI5(l68O) resonances are not observed in the photoabsorption cross section on nuclei, while the A resonance is clearly seen. The behavior of the A resonance which corresponds up to 500 MeV of incident photon energy can be described well by the A-hole model. For higher energy region from 600 MeV to 1000 MeV, proposed was an interpretation based on the collision broadening of the resonance. However, this attempt needs very large broadening, which are hardly justified in terms of NN* -+ NN cross section. Hirata et al. [3] proposed that disappearance of nucleon resonance is caused by the cooperative effect of the collision broadening of A and N*, the pion distortion *Corresponding author e-mail
[email protected],Tel +81 22 743 3433, Fax +81 22 743 3401. +Present address Research Center for Nuclear Study, Osaka University, Ibaraki 567-0047, Japan %Presentaddress College of Bioresource Sciences, Nihon University, Fujisawa, 252-8510, Japan §Present address RIKEN, Wako 351-0198, Japan
and interference in the two-pion photoproduction processes in the nuclear medium. Their calculation can reproduce the experimental disappearance of D13 and F15 by using modest collision broadening of about 80 MeV in width. Thus. measurements other than the total absorption are highly desirable. In the (7,T ) reactions, however, it is hard to determine the resonance properties in nuclei, because the D13 and FI5 resonances mainly decay to T-N channel and are overlapping with the tail of the A resonance strongly. We have aimed at the Sll(1535) resonance, which strongly couples with IKq and decays to the Nq channel with the branching ratio of 45 55 %. There is no nucleon resonance which has such a large branching ratio to 7-N decay around this energy region. Because of this unique property, one can excite only the Sll resonance by using ( y , q ) reactions below 1.0 GeV of incident photon energy. Thus we have measured the total (y, 7) cross sections on C, A1 and Cu.
Figure 1. Schematic drawing of the experimental setup. Tagged photon beam bombarded targets and produced the q mesons. The charged particles accompanied with the photons were removed by the sweep magnet which was placed at 1 m upstream of the target. Two decaying photons from the 77 mesons were detected by the two pure CsI calorimeters.
2. EXPERIMENT
The experiment was performed using the 72 beam line in the 1.3 GeV electron synchrotron of KEK-Tanashi (former INS). The schematic drawing of the experimental setup is shown in Figure. 1. The 72 beam line supplied a tagged photon beam with its intensity of 9 x 105/sec for 680 5 E, 5 1000 MeV. The tagging efficiency was measured once a day. and its typical value was about 70 80 %. Tagged photons bombarded the C, A1 and Cu targets and produced 7 mesons. which were identified by the traditional invariant mass analysis of two decaying photons. Photons were detected by the two sets of pure CsI calorimeters with plastic scintillators for charged particle veto. which were placed in front of the calorimeters. Each calorimeter consists of 29 pure CsI crystals with 13.5 radiation length in thickness. The energy of the incident photon was deduced from the sum of the energies deposited in the 7 crystals. The energy weighted average of the center position of the 7 crystals was regarded as the incident position of the photon. The energy resolution of the pure CsI scintillator was 3 % for photons of 1 GeV and the angular resolution of the photon emission direction was about 5". The invariant mass resolution of two photons is about 6 % at 7 mass region, which is consistent with the energy and angular resolution ) were measured for of the calorimeters. Differential cross sections of the ( 7 , ~ reaction the polar angles from 0" to 90" in the laboratory system with several opening angles of two CsI calorimeters. Detection efficiency was deduced from the GEANT3 Monte Carlo simulation [4]. The total cross section of q photoproduciton on nuclei was deduced by integrating the differential cross section.
-
3. RESULTS
In Figure. 2, the total q photoproduction cross sections are shown as a function of incident photon energy. Open circles, closed circles and squares correspond to the total cross section of the (7.7) reaction on C, A1 and Cu, respectively. It should be noticed that almost whole structure of the S11(1535) photoexcitation in nuclei has been shown for the first time. The reaction cross sections on C, A1 and Cu have similar dependence on incident photon energy. The cross section shows the maximum value at around 900 MeV of photon energy; this value is much higher than expected in the simple quasi free picture. The width of the resonance structure is about 300 MeV, which is also wider than expected. In order to clarify the cause of the phenomena, we have performed the calculation based on the semi-classical model, Quantum Molecular Dynamics(QMD). QMD was originally developed to describe the high energy heavy ion reactions [6]. We have modified the QMD code to describe the photon induced 7 production. At the first step, one nucleon absorbs a photon and becomes an SI1 resonance. Two set of the resonance parameter of the Sll(1535) resonance were used in the QMD calculations. One parameter set is 1540 MeV for the resonance energy, 150 MeV for the width at the resonance pole and 0.55 for the branching ratio of the 7-N channel. These values were determined so as to reproduce the H(y, 7) cross section in reference [7-91. The other is 1544 MeV for the resonance energy, 212 MeV for the width at the resonance pole, which were obtained by B. Krusche et al. by using recent measurement of the H(y, q) cross section at Maintz and Bonn [lo]. The results of the two parameterizations are plotted in Figure 3 with the experimental data. A solid line corresponds to the one with r = 212 and a dotted line with r = 150.
Figure 2. Total cross sections of ( y , q ) reactions on C, A1 and Cu with the QMD calculations (solid line) for the resonance width of 212 MeV, and calculation (dotted line) on C for the width of 150 MeV. Open circles, closed circles and squares shows the results on C, A1 and Cu, respectively.
The results of the QMD calculation are plotted in Figure. 2. In the calculations, the Fermi motion, the Pauli blocking and the 7 absorption in a nucleus have been considered; the 7 absorption cross section was deduced from the detailed balance analysis of the i7-p -+ qn cross section. The collision broadening of SI1(1535) in a nucleus have been also considered; the cross section given in ref. [ll]is used. A dotted line is the result on C with the resonance width of 150 MeV. Solid lines are the results on C, A1 and Cu with 212 MeV. As can be seen in Figure 2, the former parameterization underestimates the measured cross section on C target by about 20 p b above 800 MeV of incident photon energy, but the latter can reproduce the experimental results well on all targets. 4. SUMMARY AND OUTLOOK
The present measurements show that the 7 photoproduciton on nuclei can be basically understood as the quasi-free reaction with the reasonable nuclear medium modifications. They include the Fermi motion, the Pauli blocking, the 7 absorption in nuclei and the N-N* collision broadening. It was found that the resonance parameter with F = 212 MeV
V
B. Krusche et al. M. Wilhelm S. Homma et al.
Figure 3. Total cross section of H(y, 7). Two parameterizations are shown with a solid, dotted line for the resonance width of 212 and 150 MeV, respectively.
can reproduce the experimental data well. Recently Armstrong et al. [12] reported precise measurements of S1l resonance parameters based on the p(e,e'p)q experiment at TJNAF; b, = 0.55, M R = 1532 and FR = 154 f 20 MeV. As shown in Figure. 2, this parameterization cannot reproduce present data with the quasi-free picture. More precise measurements of the elementary processes are highly desirable to reduce the uncertainty of the resonance parameters of s11.
REFERENCES 1. N. Bianchi et al., Phys. Lett. B299 (1993) 219; ibid., B309 (1993) 5 2. M.Anghinolfi et al., Phys. Rev. C47 (1993) R922 3. M. Hirata et al., Phys. Rev. Lett. 80(1998)5068 4. CERN Program Library Long Writeup W5013: Geant Detector Description and Simulation Tool, 1993. 5. M. Robig-Landau et al., Phys. Lett. B373 (1996) 45 6. K. Niita et al., Phys. Rev. C52 (1995) 2620 7. B. Krusche et al., Phys. Rev. Lett.74 (1995) 3736 8. M. Wilhelm, Ph.D. Thesis, Bonn, BN-IR-93-43; B.H. Schoch, Prog. Part. Nucl. Phys. Vol. 34 (1995) 43
9. S.Homma et al., J. Phys. Soc. Jpn. 57 (1988) 828 10. B. Krushe et al., Phys. Lett. B397 (1997) 171 11. M. Effenberger et al., Nucl. Phys. A613 (1997) 353 12. C.S. Armstrong et al., Phys. Rev. D60 (1999) 052004
Hadron and Nuclear Physics with ElectromagneticProbes K. Maruyama and H. Okuno (Editors) 2000 Elsevier Science B.V. All rights reserved.
"
Delta in nuclei excited by hadronic processes Junsei Chibaa "IPNS, The High Energy Accelerator Research Organization (KEK) 1-1 Oho, Tsukuba, Ibaraki, Japan 305-0801 The A's in nuclei excited by proton induced reaction is presented. The experiment was carried out at the National Laboratory for High Energy Physics (KEK) about 10 years ago, but the data are still highest quality in this field. Charged particles emitted in the central region were measured with a coincidence of a forward going neutron in (p, n) reactions at 1.5 GeV/c in the A excitation region. The A peak in the neutron spectrum in coincidence with p.ir+shows no shift, as in the (p,p'AO) reactions at 4 GeV/c. On the other hand, the peak in the nuetron spectra in coincidence with other particles, especially with a T + , shifts toward higher momentum. The coincidence data set stringent constraints on theoretical models introduced to interpret the peak shift in inclusive spectra. 1. INTRODUCTION
It is one of the most interesting questions in nuclear physics whether basic properties of particles are modified in nuclear matter compare with those in free space. The A is the first excited state of a nucleon and plays an important role in the dynamics of the excitation of nuclear matter. In 1980's, a shift of the peak position in the momentum spectra of outgoing particles (or in the excitation energy of the A produced in nuclei) in the ( 3 H e ,t ) reactions were reported [I]. It is very interesting and exciting if the shift is caused by a change of the resonance mass of the A, because it would be the first evidence that particles are modified in nuclear medium. However, magnitude of the peak shift depends on incident energy [2] which implies that the peak shift could not be attributed to a simple modification of mass of A in nulcear matter. Such peak shift, though its magnutude is very small, had been seen also in (p,n) reactions at 800 MeV [3,4]. In those days, we carried out coincidence measurements in (p,pl) reactions at the incident momenta of 3 and 4 GeV/c. In this experiment, we found that the A peak in the momentum distribution of forward protons in p.ir-events (a proton and a .ir-are emitted in the target region in coincidence with a forward proton) is consistent with the mass and width of A in free space [5]. It means the A mass does not change when we see the decay products of A, while the mass seems to be modified in inclusive momentum distribution. This was a puzzle. In order to investigate A production in nuclei for further details, we have carried out an experiment to measure both inclusive and coincidence spectra for (p, n) reactions at 1.5 GeV/c. The reaction channel was chosen because (1)the cross section for A productions
is maximum at around 1.5 GeV/c, (2) the contribution of background processes is smaller in (p, n) reactions than in (p,pl) due to the dominance of the pp -+ nA'+ channel, and (3) inclusive data exist at 1.46 GeV/c [6]. 2. EXPERIMENT
Experiments were performed at the 7r2 beam line of the 12-GeV Proton Synchrotron (PS) at the National Laboratory for High Energy Physics (KEK, now named the High Energy Accelerator Research Organization) using a large-acceptance spectrometer called FANCY for coincidence particle detection. This is a cylindrical spectrometer [7] composed of a solenoid (3kG magnetic field), a cylindrical drift chamber (jet-chamber type) and a cylindrical hodoscope. By measuring pulse heights from both ends of resistive sense wires, we are able to reconstruct tracks in three dimensional space. A momentum resolution o p / p 0.1 . p (in GeV/c) was achieved. Particles with momenta up to 800 MeV/c were well identified by the energy loss (dE/dx) measurements. The spectromenter covered a wide range of polar angles between 25 and 105 degrees (with the requirement of a hodoscope hit) or between 12 and 141 degrees (without this requirement). The forward neutron detector was made of a stack of plastic scintillators whose dimensions are 3 cm (thickness) x 20 cm (width) x 100-150 cm (height). Fifty scintillators were assembled into a 10 x 5 array to form a 200cm-wide by 15cm-thick detector. The detector was surrounded by veto scintillators with a thichness of 1 cm to reject charged particles. The flight path of the neutrons was about 10 meters. The neutron detector covered polar angles between 0 and 10 degrees, but only the data between 0 and 6 degrees were analyzed. A schematic drawing of the detector system is shown in Fig. 1.
Hodoscope Solenoid
Neutron Counter
Dipole
beam -C
tArget
Veto
Figure 1. A plan view of the detector system for the El73 experiment. An unseparated beam with an intensity of (1 - 2) x lo5 particles per 0.5-sec beam spill were delivered from the internal target of the PS to the experimental target. Target
materials used were carbon and polyethylene (CH2) with thicknesses of 0.87 and 0.94 g/cm2 respectively. H-target data were deduced by subtraction. The beam momentum measured by time of flight over a 12m - path was 1.49 GeV/c with a momentum spead of 0.02 GeV/c. Event triggers were generated by a coincidence of a beam proton and a hit in the neutron detector. A beam-veto scintillator with a dimension of 15 x 15 cm was placed at 1.3 m downstream of the target in order to reduce the trigger rates for accidental coincidences. Since the cylindrical hodoscope was not used in the trigger logic, the acceptance of the cylindrical spectrometer was very large (between 12 and 141 degrees, covering a solid angle of 88 % of 4n). Tracking efficiecy was more than 95 % in the above angular range. Tracking is possible for particles with momenta as low as 50 MeV/c, but low momentum particles could not be detected due to energy loss in the target material. Cutoff momenta for pions and protons were effectively 75 and 220 MeV/c, respectively. The detection efficiency of the neutron detector was calculated by a Monte Carlo program in which the detector geometry including veto counters was realistically treated. The efficiency was (5.5 f 1) % in the momentum range between 0.8 and 1.4 GeV/c. It slightly decreases monotonically as neutron momentum inreases because events in which a high momentum particle was emitted were rejected by the veto counter. Intrinsic T O F resolution of each counter was 200 psec (rms) for charged particles. Although long-term time drift was calibrated and corrected using laser pulses, overall T O F resolution deteriorated to 350 psec in summed data obtained for a data-taking period of 5 months. The corresponding momentum resolution was 15 MeV/c at 1.0 GeV/c and 50 MeV/c at 1.5 GeV/c. 3. RESULTS
All events were classified into 6 types according to the combination of particles detected in the cylindrical spectrometer(CDC); none in CDC (hereafter called none-event), only a n+(n-event), only a proton (p-event), a proton and a n+ (pn-event), 2 protons, and others. All types except none-events were almost free from background because of the excellent detection capability of the CDC. However, the cross sections for none-events must be deduced by subtracting data for target-out runs. The target inlout ratio was about 2:l for none-events. Inclusive cross sections were thus obtained as a sum of all 6 event types. Using the calculated efficiency mentioned above, inclusive cross sections agreed well with previous measurements at LAMPF [6]. Fig. 2 shows doubly differential cross sections for C(p, n) reactions, together with fractions of each event type in the A excitation region. We'd like to emphasize that our detector acceptance is so large that at least one charged particle is detected in 80 % of all (p,n) events. The fraction of 71.- and pn-events changes rather rapidly as a function of neutron momentum, while the fraction of other event types stays almost constant. The fractions of pn-, T-, none- and p-events in the A region for H target were about 65 %, 20 %, 10 % and 5 %, respectively, which agree well with a Monte Carlo calculation in which the cutoff for low momentum particles was taken into account. Fig. 3(a) shows the differential cross sections for (p, npnf) reactions on C and H targets. The peak in the spectrum for the C target shifts slightly (-50 MeV/c) toward
C
800
Carbon (inclusive)
1000
1200
i/
1400
Neutron Momentum (MeV/c)
-
Figure 2. Doubly differencial cross sections of (p,n) reactions at 1.5 GeV/c on C at 6' = 0 6 degrees. Solid line showed cross sections of the same reactions at 1.46 GeV/c at 4 degrees with a shift of -35 MeV/c to account for the difference of incident momentum. Calculation of quasifree A(with mass of 1200 MeV and width of 160 MeV) production is shown in dashed line. Above picture shows the fraction of each event type.
the low momentum side with significant broadening compared to the H target. But the spectra were well reproduced as shown in the Figure (solid line) if we account for the binding energy and the Fermi motion of nucleons in nuclei. The model used in the calculation was exactly the same as that in Ref. [5]. The width (aF)of the Fermi motion expressed in a gaussian form was 120 MeV/c. The best match to the experimental data was achieved with a binding energy of 40 MeV, while it was 25 MeV for (p,p'AO) reactions at 4 GeV/c. This model is, in some sense, self inconsistent because the binding energy and the Fermi motion are treated independently. In order to overcome this weakness, we also calculate the momentum spectra based on the model introduced by Gugelot [8] in which the two quantities are treated in a consistent manner. If we used a binding energy of 25 MeV in our model, the two model agree with each other in terms of quasifree two-body kinematics. Therefore the difference in the binding energy in (p, nA++) and (p,p'AO) reactions seems not to be caused by a trivial kinematic effect but by some as yet unknown dynamical effect. The ratio of cross sections O ~ ~ , , ~ + + / O ~ ~ , , ~ +is+ 1.2 f 0.1 which agrees well with the (P,~'AO)data at 4 GeV/c [5] where the target mass dependence of the cross section was discussed based on an intranuclear cascade model. While the forward momentum spectrum is well understood with a simple quasifree picture, an anomaly was found in the distribution of the invariant mass of a proton and a r+(MP,+). Fig. 3 (b) shows Mp,+ distributions
0 800
1000 1200 1400 Pn (Neutron Momentum, MeVIc)
. ..,
1100 1150 1200 1250 1300 M +(px+Invariant Mass, MeV)
1350
pn
Figure 3. (a) Momentum spectra of forward neutrons for PT-eventson C and H targets. Lines show our model calculation with the mass and width of Abe the same as those in free space (M=1232 MeV and r=115 MeV). Binding energy (Q) and Fermi motion (OF) are Q=40 MeV, crF=120 MeV/c (solid line), Q=O. ffF=O (dashed line; for H target), and Q=O, aF=120 MeV/c (dotted line), respectively. (b) Distributions of pn+ invariant mass for C and H targets. Breit-Wigner functions with M=1225 MeV and r=115 MeV (solid). M=1207 MeV and r=190 MeV (dashed) are also shown.
for C and H targets. The distribution for the H target was well expressed by a BreitWigner function with standard values for the mass (1232 MeV) and width (115 MeV) of the A. The distribution for the C target significantly differs from that for the H target. Fitting to a Breit-Wigner function yields a mass of 1207 i 7MeV and width of 190k25MeV. Using these values for A's in nuclei, the momentum distribution of forward neutrons from quasifree A production was calculated and is shown in Fig. . Agreement with the experimental data is fair. Those observations suggest two possible scenarios to understand quasifree A production: (a) A's in nuclei differ from A's in free space as shown by the Mp, distribution, and some additional momentum is transferred to a target nucleus for pr-events, or (b) A's in nuclei don't differ from free A's but rescatterings and final state interactions of a proton and a T+ decaying from a A result in a distorted M,, distribution. More studies are necessary to make clear conclusions. Fig. 4 shows momentum spectra of forward neutrons for T-, 2p- and pn-events along with results of a simple model calculation. In this model. a A with M = 1207 MeV and
r = 190 MeV is produced in a quasifree process.
When the momentum of a proton decaying from the A is less than 400 MeV in the target rest frame, the proton is captured by the target nucleus and only a pion comes out from the collision (coherent pion process). The probability for such events depends strongly on the mass of A, and therefore momentum spectrum for T-events differs significantly from that for pr-events as shown in the figure. Angular distributions of pions are also well reproduced by this model. Considering all experimental data obtained so far in hadronic processes, the following scenario seems to be most reasonable, although more sophisticated theoretical calculation is necessary to make a final conclusion: The mass of A in nuclei is effectively smaller than that in free space as had been already seen in r A total cross sections [lo]. Fate of a A depends strongly on its mass; probability of so-called coherent pion process increases when its mass is smaller, while probability of quasifree (decay) process increases when its mass is larger. Probability of absorption process is almost independent of the A mass because the cross section of N A -t N N shouldn't depend on the mass.
l
'
l
'
l
'
l
'
-
pPf
c Pq f n*r%f
@t*qGsa 3,grg @I@@
@
9icSB m i
99
800
9
n+
-
!ti
' ~@@,PI ' ~@
a I
'
.
HH
-
l
a pn+PP-
.p+C+n+(pn+, PP, x+)
.
I
1000
,
I
4,
.
1200
1400
Neutron Momentum (MeVIc)
Figure 4. Momentum spectra of forward neutrons for T-, 2p- and PT-events on C target. Now we qualitatively understand the behaviour of A in nuclei produced in hadronic processes. There are two big questions left; (1) what is the origin of the A mass shift? and (2) why A's produced in hadronic processes differ from those produced in electromagnetic processes? In addition to these questions, we should carefully examine the coherent pion process by experiments with high energy resolution. An experiment to reveal the coherent pion process was carried out at LAMPF [ll]but still clear answer was not yet obtained. 4. SUMMARY
In summary, we have carried out first coincidence and inclusive measurements of (p, n) reactions at 1.5 GeV/c on H and C targets. All (p,n) events were classified into 6 types according to particles emitted in the target region and fractions of each type were determined as a function of neutron momentum. Two types, PT-eventand T-event,
showed larger changes of fractions. For pn-events, monentum spectra of forward neutrons were well understood by a quasifree picture, but the p~ invariant mass distribution differs significantly from that in free A decays. The fate of a A in a nucleus seems to depend strongly on its mass. Probabilities of quasifree (decay), coherent pion, and absorption processes are qualitatively understood.
I would like to thank all collaborators of the KEK PS experiment El33 and E173; K. Nakai, T . Kobayashi, T . Nagae, K. Tokushuku, H. Sano, M. Sekimoto, I. Arai, H. Sakamoto, A. Manabe, K. Aoki, H. Nunokawa, M. Tanaka, M. Nimomiya, M. Tomizawa, H. Kitayama, iV.Kato, T. Baker, C. Glashausser, R. Ransome, G. Edwards and G. Kumbartzki. REFERENCES 1. D. Contardo et al., Phys. Lett. 168B(1986) 331. 2. V.G. Ableev et al., J E T P Lett. 40(1984) 763. 3. B.E. Bonner et al., Phys. Rev. C 18(1978) 1418.
4. C.W. Bjork et al., Phys. Lett. 63B(1976) 31. 5. T . Nagae et al., Phys. Lett. lglB(1987) 31. 6. R.G. Jeppesen, Ph.D. thesis, Univ. of Colorado, 1986 (unpublished). 7. K. Ichimaru et al., Nucl. Instr. Meth. A237(1985) 559. 8. P.C. Gugelot, Phys. Rev. C 30(1984) 654. 9. J. Chiba et al., Phys. Rev. Lett. 67(1991) 1982. 10. A.S. Carrol et al., Phys. Rev. C 14(1976) 635. 11. R. Gilman et al., Nucl. Phys. A577(1994) 227c.
ThisPageIntolyLfBk
ThisPageIntolyLfBk This Page Intentionally Left Blank
Hadron and Nuclear Physics with ElectromagneticProbes K. Maruyama and H. Okuno (Editors) 2000 Elsevier Science B.V. All rights reserved.
'
The physics of AA excitation G.M. Hubera "Department of Physics, University of Regina, Regina, SK S4S OA2 Canada The physics of double delta excitation has been of interest since the early 19801s,when Brown. Toki, Weise and Wirzba [I] pointed out that in the simplest SU(3) consituent quark coupling model A N -+ AA is expected to be much stronger than A N + N N . Since it is generally believed that the A N + NN process plays a major (if not the dominant) role in pion and photoabsorption processes in the resonance region, this claim could have a major implication on our understanding of both processes. In the simplest case of a free deuteron target, A A excitation will require an incident energy of about 600 MeV to be placed completely on shell. On heavier nuclear targets, the process is helped by the high Fermi momentum tail, and so could be proportionately more important at lower energies than it would be on deuterium. The task, clearly, is to find a unique signature which would indicate the presence of this exotic, and otherwise unexpectedly important process. The first detailed calculations involving the A N -+ AA process were done for n absorption. motivated by the greater activity at the meson factories at that time. Arguments were made [I] that A A excitation following n absorption should have a rather unique kinematic signature, in which both A's internally de-excite, leading to the emission of four energetic nucleons via the mechanism shown in figure l a ) . For the sake of clarity, it should be pointed out that diagrams involving the excitation of two sequential A's, on two different nucleons. are not considered to be double delta excitation; the latter process requires both A's to be present simultaneously in the nucleus. In order for two nucleons to reach A excitation, it is clear that they must be at close range, and so the reaction is potentially sensitive to the importance of short range correlations (SRC) in the nuclear environment. The smallest nucleus for which this unique four nucleon emission signature can be studied is 4He, and so a detailed calculation was performed by Schwesinger, Wirzba and Brown [2]. This calculation not only involved A N + AA diagrams of the type shown in figure l a ) . but also virtual n + np decay diagrams such as in figure l b ) . In this case, the n + pn coupling is calculated from p + 27r decay. It is found that both processes play major roles in the 4He(n,4N) reaction, with the two diagrams having significant interference at lower energies. Unfortunately, a nonrelativistic approximation was used for the nucleon degrees of freedom, so the validity of the calculation may be increasingly doubtful above 600 MeV. Nonetheless, a AA cross section of 12 mb at 400 MeV would account for 30% of the measured pion absorption total cross section on 4 H e at that energy 131. . This work result,ed in a number of interesting findings. The calculation is especially ~
Figure 1. Pion absorption through formation of a A A state which subsequently decays, leading to emission of four energetic nucleons. a) the A N -+ A A mechanism. b) absorption through formation of a virtual p~ pair.
sensitive to the treatment of SRC, since it involves four nucleons close together. Their mutual short range repulsion was treated with a range of llm,, where m, = 730 MeV, and led to a reduction in the A A --+ 4 N cross section. In addition, because the two nucleons associated with the A's are initially in a relative S-state, the intermediate A A is dominantly in a I = 2 or a S = 2 state. The I = 2 states can only be reached by a tensor force acting on A N + A A , while the S = 2 states have three open channels reached by the tensor force, and only one by spin-spin forces. A further enhancement of the tensor contribution comes from the fact that the momenta transferred in A N + A A can be arbitrarily small, and so cancellations between T and p tensor contributions are less severe than in A N + N N transitions. Thus, it appears that the major contribution to the A A mechanism comes from the tensor force, while this force can be usually safely disregarded in the A N -, N N transition. To date, relatively little progress has been made in identifying the role of the A A -+ 4 N process in pion absorption. Experimentally, this is because of the high pion energy required, and the tremendous increase in the complexity of the pion absorption reaction at these energies, which mask any possible signature. Further theoretical investigation has also lowered the expectations somewhat. For example, a 1991 paper by Abbas [4] showed that inclusion of proper nucleon deformation in the N -+ A transition leads to a suppression of ( k ) 2 by 27% compared to the earlier SU(4) calculations by Brown, et al. These complications contrast with the early optimistic words of Schiffer: " T h e A A m e c h a n i s m almost has t o be present, its predominance is n o t a m a t t e r of ad hoc assumptions but of Clebsch-Gordan coeficients in a simple SU(4) quark model." [5]. The only significant experimental evidence for the AA mechanism using pion beam has come from studies of (T, 27r) at PSI. Rahav et al. [6] compared 12C(r-, 2 ~ and ~ ) 12C(r-, T-T+) yields at T, = 292 MeV using the CALLIOPE spectrometer, and found that there was a proportionally higher number of 2 r - events than anticipated from quasifree 27r production. If we compare the fundamental reactions, we expect a ratio
- Full calculation -- AN->AA (fig la)
Figure 2. Total cross section of the 4He(7r,4N) process versus total energy of the incoming pion, calculated by Schwesinger, Wirzba, and Brown [2].
while the observed ratio is
so there are nearly twice as many 27r- events as expected. However, according to a spinisospin Clebsch-Gordan coefficient calculation, the AA mechanism is expected to yield a ratio of 0.59, and so provide a more balanced ratio between the two charge states. Assuming that only quasifree and AA mechanisms contribute to the observed yield ratio, A (0.30 f 0.14)aQF. they concluded that ~ A = The clearest evidence of AA excitation has come from photoabsorption studies using the TAGX spectrometer at INS (Tokyo). Asai et al. [7] measured the d(y, r + r - p ) n reaction using a tagged photon beam with energy between 570 and 850 MeV, and employed Monte Carlo simulations of a number of competing r+7r-production mechanisms: Quasifree yp
-+
A++7rP + 7r+7r-p1 with the neutron as a spectator
Double delta ypn
-+
A++A-
-t
7r+r-pn
Four body phase space ypn -+ n+7r-pn, which could mimic a number of overlapping, higher mass resonances not individually resolved by the experiment. This analysis yielded acceptable fits to the experiment kinematic observables. Events with missing momentum < 250 MeV/c were entirely consistent with the quasifree mechanism
Asai et al. (TAGX) Wada et al. (Bonn) - Figs. 4 a)+b)+c) Figs. 4 a)+b)
Figure 3. A++A- cross sections obtained from the d(y, n+n-)n reaction at TAGX [7] and SAPHIR [B], and compared with the calculations of Gomez-Tejedor, Oset and Toki [9]both with and without the contribution of intermediate N*.
(i.e. the neutron was a spectator), while events with p,,,, > 250 MeV/c were best fit with a combination of AA and phase space contributions. The same analysis technique was also applied to d(y, 7r+7r-p)n data obtained by the SAPHIR collaboration at Bonn using tagged photon beam 750 5 E, 5 1280 MeV [B]. Although the greater contribution of N* resonances at this higher energy requires further analysis before final results can be published. the preliminary results are quite consistent with the TAGX experiment, and are clearly of higher quality. The conclusion of both experiments is that A++A- excitation is observed, but that A'AO is too weak to be identified. This is expected on the basis of isospin. It is also seen that the AA mechanism increases in prominence quite rapidly with increasing photon beam energy. attaining nearly 40% of the quasifree cross section by E, = 710 MeV, which is similar to the ratio quoted earlier by the (T, 27r) experiment at lower energy. The results from TAGX prompted a detailed calculation by Gomez-Tejedor, Oset and Toki [9]. Like the earlier pion absorption calculations, this model's foundation is A N -+ AA transitions of types shown in figures 4a) and 4b). In addition, the contribution of the N&,, resonance was taken into account via diagrams similar to that shown in figure 4c). The result of the calculation is shown in figure 3. There is a constructive interference between the A and N* diagrams below E, z 820 MeV and a destructive interference above this energy, which is due to the change of sign of the real part of the N* propagator. Although the importance of the N* diagram is small, the low energy experimental cross sections are better reproduced with its addition. The model acceptably describes the photoproduction data up to E, = 800 MeV, but deviates from the data in a steadily worse manner at the higher energies. This discrepancy was anticipated, because diagrams involving the virtual yp meson exchange current, such
Figure 4. a).b),c) sample diagrams included in the model of the yd + A++A- reaction by Gomez-Tejedor, Oset and Toki [9].d) diagrams including intermediate vector mesons, such as this one, were not included.
as in figure 4d), were not included. If we compare with the result of Schwesinger et al., we see that the inclusion of p diagrams indeed cannot be neglected at these energies. There is also the question of the deuteron D-state, which was neglected in these calculations, but also expected to play a more important contribution as the energy is increased. More recently, these studies were extended to lower energy on 3He by the TAGX collaboration [lo]. Inclusive 3He(:i, n+n-) data were obtained from 380 to 700 MeV incident photon energy. In the case of the lowest energy, this is only 50 MeV above the absolute n+n- detection threshold. The detection of additional protons, which would have made the experiment more exclusive in nature, was precluded by proton rangeout in the target, and the 300 MeV/c proton momentum detection threshold. The results are shown in figure 5, and are generally consistent with the deuteron data, after multiplication by 1.5. However, we also see that the data are in excess of the Gomez-Tejedor et al. prediction (after multiplication by the same factor), a difference which cannot be explained by the x 25 MeV difference in reaction threshold between 2H and 3He. Analysis of higher energy 3He(y. nf T-p) data from TAGX is still underway. In conclusion, we have seen that AA mechanisms do indeed play an identifiable role in n and 7 induced reactions above x 500 MeV, although the importance of their role has been tempered somewhat from the first optimistic expectations. Nonetheless, it would appear that the study of this mechanism can play a role in our better understanding of nuclear dynamics, as it touches on a large number of topics which have sustained interest in the nuclear community. These issues include the role that repulsive SRC play in setting the strength of the A N + A A process, the interference of N * and p r contributions, and the A N tensor force.
1
a Ref. 7 (d*1.5) o Ref. 8 (d*1.5) w Ref. 10 ( 3 ~ e ) - d*1.5 Ref. 9 with N'
1 T
Figure 5. 3He(y, AAp) cross sections obtained at TAGX [lo], and compared with the data and calculations on deuterium target (after multiplication by 1.5).
The author wishes to thank INS and the TAGX collaboration for their role as hosts and collaborators over the last ten years. It was a very enjoyable and valuable experience. This work was supported in part by the Natural Sciences and Engineering Research Council of Canada (NSERC).
REFERENCES G.E. Brown, H. Toki, W. Weise, A. Wirzba, Phys. Lett. 118B (1982) 39. B. Schwesinger, A. Wirzba, G.E. Brown, Phys. Lett. 132B (1983) 269. D. Ashery, J.P. Schiffer, Ann. Rev. Nucl. Part. Sci. 36 (1986) 207. A. Abbas, J . Phys. G 17 (1991) L181. J.P. Schiffer, Comrn. Nucl. Part. Phys. 14 (1985) 15. A. Rahav, et al., Phys. Rev. Lett. 66 (1991) 1279. M . Asai, et al., Z. Phys. 344 A (1993) 335. Y. Wada, representing the SAPHIR collaboration, Particles and Nuclei International Conference, Williamsburg, Virginia, 1996 (unpublished). 9. J.A. Gornez-Tejedor, E. Oset, H. Toki, Phys. Lett. 346B (1995) 240. 10. D.G. Watts, et al., Phys. Rev. C 55 (1997) 1832.
1. 2. 3. 4. 5. 6. 7. 8.
Hadron and Nuclear Physics with ElectromagneticProbes K. Maruyama and H. Okuno (Editors) 2000 Elsevier Science B.V. All rights reserved.
"
The neutral pion photoproduction on the proton near threshold 11-T. Cheona and M. T . ~ e o n g ~ "Department of Physics, Yonsei University, Seoul 120-749, Korea bDepartment of Physics, Dongshin University, Naju 520-714, Korea , have been calculated by taking the vector The Eo+and MI+ amplitudes for ~ ( yirO)p mesons into account in the pole model of t-channel. A conclusion is led that the rescattering process below the n + threshold plays a key role in this process. Our results are Eo+= -1.23 x 10-3mG, MI+ = 7.56 x 10-3m;$kq and MI- = -2.15 x 10-~m;$kq at the n+ threshold. The PCAC hypothesis is well restored in our frame of calculation.
1. INTRODUCTION The neutral pion photoproduction on the proton near threshold drew tremendous attention of physicists in the context of a possibility that the low energy theorem(LET) might be broken down. Since the LET is based on the partially conserved axial current (PCAC) hypothesis [I], a break of the LET implies a break of PCAC. For such a reason, the results caused a sensation [2,3]. In 1994, we pointed out [4] that the pion p-wave contribution should be properly substracted from the total cross section to extract the electric dipole amplitude near threshold and that the experimental results could well be reproduced consistently with the LET if vector meson contributions were taken into account. Fuchs et al. [5] performed again the experiment to measure the total and diffenrential cross sections near threshold with improved accuracy. Their results are in good agreement with those calculated by the chiral perturbation theory (ChPT) [6]. However, the ChPT cannot make an exact numerical prediction for Eo+amplitude so far, because the calculation of the loop diagrams converge too slowly. Even in this situation, a key question is left whether the charge exchange process, yp -+ n + n -+ irop, can contribute to Eo+even below the ir+ threshold because of mass differences between nn+ and nop. Such a process is expressed as [7]
where a(n.ir+ + p r o ) = 0.172f m is the pion-nucleon scattering length, Eo+(v nn+)= 26.6 x 10-3m;1 is larger than Eo+(yp -+ pnO) by one order of magnetitude and q = 0.190 i fm-' is the ir+ momentum below the ir+ threshold. This imaginary momentum makes Eo+real, i.e. -+
Hence, a question arises whether rescattering of the pion carrying an imaginary momentum below the threshold can contribute to the process, i.e. the real part of amplitude. Notice that the propagation particle is carrying a real momentum in usual rescattering processes. It is very interesting to analyse carefully the experimental results to find a definite answer to the question. 2. CALCULATION
The LET prediction of the electric dipole amplitudes for yp is given as
-+
reaction a t threshold
where p = m,o/M, f2/47r = 0.08, e2/47r = 11137, K, = 1.79, mro = 135MeV and M = 938.3MeV. The amplitude E& given in eq.(2) is the result of on-shell calculation. On the basis of quark model, we estimated the off-shell contribution to be +0.15 x 10-~m$. Therefore, the rescattering amplitude is E+ : = -0.72 x 10-~m;:. Magnetic dipole amplitudes were calculated in the framework of the BL theory [8]. The -: = -6.31 x l~-~rn;:k~ Born term has been estimated as M E = 3.13 x l~-~rn;:kg and M . The A contribtions calculated in both s- and u-channels are M& = 4.17 x l~-~rn;;kg and Mf_= 1.84 x 10-~m;:k~. Let us evaluate the anomaly term [9] due to the vector mesons using the pole model in the t- channel of the dispersion theory [lo]. Although the vector coupling term has been considered by several authors [8,10,11],tensor coupling term was usually neglected. Since it is expected to play an important role to reproduce the experimental results, we take it into account in our calculation. Vector meson exchange is described by
where E ~ , , ~is antisymmetric pseudotensor of rank 4, i p is the photon polarization vector and t = (p2 - p1)2. The pion photoproduction amplitude in the 7rN center-of-mass frame is given in the general form
F = i ( o . e ) F l + ( 0 . 6 ) ~[El. x k ] ~ ~ + i ( a . & ) ( E l . ~ ) F ~ + i ( u ~ ~ ) ( ~ l ~ $ ) (5) F~ for the energy up to the (33)- resonance. Here, k and are the unit vectors of the photon and pion momenta. Each amplitude, F,, can be connected to the invariant amplitudes Ai, by a set of linear equations as
+
+
where ($ = m,+/(8nWs), Xv = eg,,,/m,o(m:o mb), C = m: (mi/2W,) and D, = (m E,)lI2 with El = (m2 + k2)lI2 and E2 = (m2 q2)1/2. The coupling constants : = 0.1 f 0.5, in the N N V vertex are enpirically known as [12] gLT/4n = 8.1 f 1.5, gT gPv/4n = 0.55 & 0.06 and g i T / 4 ~= 20.5 2.1, which induce g,v = 10.09, g , ~= 1.12, = 0.1 and gpT/gpV = 6.1. The Qpv - 2.63 and gPT = 16.05. Here, we find g,,/g,v coupling constants, g,, and g,,,, are evaluated with the decay widths [13],I'(w -+ irOy)= (0.717 f 0.042)MeV and r ( p 0 -+ noy) = (0.118 f 0.030)MeV. The results are g,, = 10.313 and g,, = f 0.130. Their ratio is lg,,,/g,,,j = 2.4, which is smaller than the S U ( 3 ) result of 3. Notice that the data for I'(po -t r O y )has rather a large error. The data ir-y) = (0.067f 0.007)MeV for p -+ ir-y decay shows a better precision [13],i.e. r ( p which implies g,, = f0.0983. Their value gives the ratio Ig,,,/g,,,l = 3.18 being very close to the SU(3) result. The signs of these coupling constants are undetermined at this stage. To determine their signs, let us examine the deuteron magnetic moment. The experimental value is pd = 0.843nm which is in deficit of 0.014nm. This deficit might be caused by neglect of contributions from the ynp and ynw currents [13,14]. Our estimates of those contributions are Ap; = f0.019nm and Ap: = 10.006nm. To compensate the above deficit in pd, we must have a positive sign for w contribution at least, i.e. g,, = f0.313. On the other hand, if Ap: is positive we have Apd = +0.025nm, which is too large compared to the deficit Apd = +0.014nml while the negatve value of Apd gives a reasonable result Apd = +0.013nm. Thus, we conclude g,, to be negative, i.e. g,,, = -0.0983. With these values, we find the A E L amplitude at threshold as
+
+
+
-+
3. RESULTS
The s-wave amplitude for the neutral pion photoproduction at threshold yields
Table 1 Values of the electric and magnetic dipole amplitudes, Eo+and MI+, at threshold. The experimental result(I1) for the magnetic amplitudes is available for MI = MI+ - MI-. The values of the Eo+ and MI+ amplitudes are in units of 10-3m;i and 10-3m;tkq, respectively. Experimental results I and I1 are taken from refs. 3 and 5, respectively. Exp. (11) Born A(1232) (p,w ) Rescatt. Sum Exp. (I) Eo+ -2.27 0 1.76 0.72 -1.23 -1.32 f 0.11 -1.31 f 0.08 MI+ +3.13 +4.17 0.26 0 +7.56 +7.41 MI =
The amplitude Eo+ near threshold without taking account of the rescattering process, yp -+ ~ + -+n TOP, mentioned in the beginning is shown by a dot-dashed curve in Fig.1. It is obvious that the dot-dashed curve does not agree with the experimental result. The amplitude Eo+at threshold without taking account of the rescattering procces is is shown by the dotted curve in Fig.1, and it yields
Figure 1. Electromagnetic dipole amplitudes, EO+ and MI. The dot-dashed and dotted curves are the Eo+amplitude obtained without and with the rescattering below the .ir+ threshold. The dashed curve represents the MI amplitude. Experimental results are taken from ref. 5.
Figure 2. Total cross sections. The solid and dotted curves are obtained without and with the cusp effect due to the rescattering, respectively. The dashed curve represents the s-wave contribution with the rescattering. Experimental data are taken from refs. 2, 3, 5.
Contributions of the vector mesons to the MI+ amplitudes can be evaluated with Eqs. ( 7 ) and (8). The results are = 0.26 x 10-~rn;$kq and MK = 2.32 x 10-~rn;$kq. Thus, the total values for the magnetic dipole amplitudes are found as
ML
+
We define Ml = 3E1+ MI+ - MI-. This result is in Fig. 1 with a dashed curve. Figure 2 shows the total cross section calculated with these electromagnetic amplitude near threshold. The solid and dotted curves are obtained without and with the rescattering process, respectively. j R o m these figures, one may confirm the existence of the rescattering process below the T + threshold. The dashed curve in Fig. 2 is the s-wave
I
20
-
,
1
1
I
= 146.1 MeV
Er = 145.3 MeV
E,.
ET=147.8 MeV
E,z148.9 MeV
I
,
~r=1~6.~'?4,4,~./
E, =149.8J&aiJ ,- \
Q
$40-
PION ANGLE
,B,
(7
Figure 3. Differential cross sections. The solid and dotted cures are obtained without and with the rescattering. The dashed curves are calculated with the amplitude given in eq. (14). Experimental data are taken from ref.[5].
contribution with the cusp effect due to the rsscattering process. Our results are consistent with the recent experimental data obtained by Fuchs et al. 151. Figure 3 shows the differential cross sections with recent experimental data [5], which improved accuracy and shifted globally downward compared with the previous results [2]. The solid curve shows the results without the rescattering contribution, while the dotted curve represents the results with the rescattering contribut'ion below the n+ threshold. We can see that the dotted curves agree with the experimental results [5] much better than the solid curves. Since the rescattering discussed here appears only below the .ir+ threshold, we do not have any dotted curves for the cases of E, = 151.7MeV and 152.5MeV. Although we used g , ~= 16.05 evaluated from the empirical value gPT/g,v = 6.1 in the present calculation, our previous analysis used rather small value for the tensor coupling of rho meson, i.e. g p =~ 9.73 evaluated from g p ~ l g P V= 3.7 based on the SU(3) vector dominance. Since AE& = AE,", AE{+ = 0.933 x 10-~rn%in our previous calculation, the amplitude ~ h h was ,
+
This value is somewhat close to that given in eq.(12). Therefore, it may be supposed that
if g p = ~ 9.73 is used, the experimental results can be explained without the rescattering process below the .irf threshold. Indeed, the differential cross sections up to the .ir+ threshold energy can reasonably be reproduced. The results are shown by dashed curves. However, the improved data at E, = 151.7MeV and 152.5MeV can not be reproduced, as is seen in Fig. 3. Agreement with the data of Eo+shown in Fig.2 could not be satisfied either. 4. C O N C L U S I O N
z , = 7.56 x 10-~rn;ikq and MI- = -2.15 x Our results are Eo+= -1.23 x ~ O - ~ r n MI+ 10-3rn;i kg at the .ir+ threshold. We are led to a conclusion that the rescattering of the pion carrying an imaginary momentum below the .ir+ threshold can play important roles as a physically meaningful process. Our present investigation leads to a conclusion that the recent experimental results are consistent with the LET and the PCAC hypothesis remains valid. REFERENCES 1. P. de Baenst, Nucl. Phys. B24, 633 (1970). 2. R. Beck et al., Phys. Rev. Lett. 65, 1841 (1990). 3. J. C. Bergstrom et al., Phys. Rev. C53, R1052(1996). 4. B. G. Yu, 11-T. Cheon and M. T . Jeong, J . Phys. Soc. Japan 63, 78 (1994). 5. M. Fuchs et al., Phys. Lett. B368, 20 (1996). 6. V. Bernard, N. Kaiser and U. Meissner, Z. Phys. C70, 483 (1996). 7. A. N. Kamal, Phys. Rev. Lett. 63, 2346(1989). 8. I. Blomqvist and J . Laget, Nucl. Phys. A280, 405 (1977). 9. J. Wess and B. Zumino, Phys. Lett. 3 7 B , 95 (1971). 10. F. A. Berends, A. Donnachie and D. L. Weaver, Nucl. Phys. B 4 , 103 (1967). 11. T. Schafer and W . Weise, Nucl. Phys. A531, 520 (1991). 12. T. Ericson and W. Weise, "Pion and Nuclei", Clarendon Press, Oxford 1988. 13. R. J . Adler, Phys. Rev. 141, 1499(1966). 14. M.Chemtob, E. J. Moniz and M. Rho, Phys. Rev. C10, 344 (1974).
Hadron and Nuclear Physics with Electromagnetic Probes K. Maruyama and H. Okuno (Editors) 2000 Elsevier Science B.V. All rights reserved.
'
Photoneutron cross section measurement on 'Be by means of inverse Compton scattering of laser photons H. Utsunomiya, Y. Yonezawa, H. Akimune, T. Yamagata and M. Ohtaa M. Fujishiro H. Toyokawa, and H. Ohgakic
" Department
of Physics, Konan University, Okamoto 8-9-1, Higashinada, Kobe 658-8501, Japan
bOsaka Prefecture University, Gakuencho 1-2, Sakai, Osaka 599-8570, Japan 'Quantum Radiation Div., Electrotechnical Laboratory, Umezono 1-1-4, Tsukuba, Ibaraki 305-8568, Japan Photoneutron cross sections were measured for 'Be in the energy range from 1.8 to 4.0 MeV using a laser-electron MeV photon source. These cross sections are relevant to the reaction rate of the first step of a-process, a ( a n , y)gBe, in a neutrino-driven wind of core-collapse supernovae. Results for the 1/2+, 512-, 112-, and 5/2+ states are presented in comparison with the data taken with other photon sources. 1. INTRODUCTION
The neutrino-driven wind formed in core-collapse supernovae may be an ideal site for the r-process [I]. In this rapidly expanding environment, n- and a-rich material is first processed into 9Be through a(a n, y)gBe by bridging mass gaps A = 5 and 8, which is followed by the 'Be(a, n)12C. This sequence of 12C production proceeds more efficiently than the triple a process. The subsequent reaction network is considered to produce seed nuclei beyond j6Fe for r-process. 'Be is a loosely-bound nuclear system consisting of two a ' s and a neutron. Non of any two constituents alone can form a bound system. The lifetimes of 5He and 'Be are 1.1 x sec and 0.97 x 10-l6 sec, respectively. Three adjacent thresholds exist at 1.573 MeV for a a n, at 1.665 MeV for 8Be n and at 2.467 MeV for 5He a . In view of the lifetimes, the synthesis of 9Be is considered to proceed more dominantly through 8Be than through jHe in two successive binary processes. The ternary process is likely to play a minor role. Resonances near these thresholds determine the rate of the 'Be synthesis. Resonance parameters of these states can be obtained by photodisintegrating 'Be. Previously, photoneutron cross sections for 'Be were measured near thresholds using two kinds of real photon sources, Bremsstrahlung and radioactive isotopes. Bremsstrahlung measurements with a photon difference method lacked great accuracy due to poor energy resolution [2]. Measurements with radioactive isotopes (24Na,"A1, 38C1,j8C0, 65Ni,88Y, lo5Ru, lZ4Sb,142,144Prand 206Bi)were limited below 2.75 MeV [3-81. Low-lying states in
+ +
+
+
'Be were also investigated in electron scattering [9,10]. However, virtual photons populated the first excited state (1/2+) only very weakly. No single measurements provided data for all resonance states near thresholds (1/2+, 512-, 112-, 512'). 2. EXPERIMENTAL METHOD 2.1. y-ray production and neutron detection Inverse Compton scattering of laser photons incident on relativistic electrons in the storage ring TERAS of the Electrotechnical Laboratory (ETL) produces quasi-monochromatic y rays in the energy range from 1 to 40 MeV [ l l ] . This new MeV photon source, LEMPS (laser-electron MeV photon source), is available as a pencil-like beam typically with a flux of lo4 photons/sec/mm2 and energy resolution of a few %. The beam properties were previously investigated with an anti-Compton spectrometer [12]. A Nd:YLF laser (A = 1053 nm) was used. The laser system was operated at 1 kHz and produced 100 % linearly polarized light. Unpolarized light was also produced by passing polarized light through an optical element called depolarizer. Electron energy was varied from 314 MeV to 475 MeV to produce y rays in the energy range of 1.8 MeV - 4.0 MeV. y rays were collimated into 2 mm in diameter with a 20 cm thick P b block at 5.5 m from a head-on collision point between electrons and laser photons. y-ray energy was measured with a 155 cm3 pure-Ge detector of coaxial type. A full energy peak was energy-calibrated with natural sources of 40K and 208T1as well as a standard source of 60Co. A 4 cm thick 9Be rod of 99.5 % enrichment (2.5 cm in diameter) was irradiated. Neutrons were measured with four 60 cmHg BF3 counters (MITSUBISHI ND8534-60: 2.5 cm in diameter x 24.1 cm in effective length) embedded in a 30 cm polyethylene cube. The BF3 counters were located, two each vertically and horizontally, in a concentric ring at 7.5 cm from the beam axis. Neutron moderation time in the polyethylene was measured in l ms time range with an ORTEC 566 TAC module. Fig. 1 shows a moderation time distribution. Background neutrons, a time-independent component, made a minor contribution to the neutron measurement. Incident photons were detected directly with a BGO detector (2" in diameter x 6" in length) placed at the end of the beam line. Pile-up spectra were obtained for the photon beam pulsed at 1 kHz. The average number of photons per beam pulse was determined from the pile-up spectra with use of a single photon spectrum that was separately taken with a DC beam. The total number of photons was obtained from the frequency (IkHz) and the data acquisition time. Independent measurements were carried out with a DC beam, in which foreground neutrons and background neutrons were measured before and after each measurement. The results of the pulsed- and DC-beam measurements agreed to each other within the accuracy of 5 %. 2.2. Neutron detection efficiency A 2.16 mCi 24Na(half-life 15 hours) was made by irradiating a 50 mg sample of NaOH for 15 minutes with thermal neutrons at the Research Reactor Institute of Kyoto University. It was mixed with 2 cm3 D 2 0 in an aluminum container (2 cm in diameter x 4 cm in length). The 24Na beta-decay to the 4+ state in 24Mg with a 99.94 % branching ratio, producing 2754 keV y rays in transition to the 2+ transition. These y rays disintegrated deuterons and produced 265 keV neutrons. Contributions of other states in 24Mg
0
200
400
600
800
1000
Channel Number
Figure 1. Moderation time distribution for neutrons in a polyethylene cube. Full scale is 1 ms. The time constant for moderation is about 140 p sec. Background neutrons that arrived at BF3 counters time-independently are clearly distinguished from photoneutrons of interest.
to neutron production were neglected because of small branching ratios. The absolute intensity of the 265 keV neutron source was determined with the water bath method [13]. The source was placed at the center of a 400 liter water bath. Neutron flux was measured from 2.5 cm to 17.5 cm in 2.5 cm steps with a 1.5 cm3 BF3 counter (MITSUBISHI ND8522-60). The efficiency calibration of the small counter for thermal neutrons had separately been done with the standard graphite pile at ETL [14]. Background neutrons were measured with the same counter covered by a 5 mm thick cadmium. Time elapsed during the measurement was corrected for. The source intensity at the time of production was determined to be 2.83 x lo3 counts/sec/4~. The neutron source was placed at the center of the neutron detector. A neutron counting was performed five times, each for 300 seconds, with two bare BF3 counters and two Cdcovered counters. The contribution of background neutrons was very small. The detection efficiency thus determined for 265 keV neutrons was 1.67 & 0.09 % per BF3 counter. The uncertainty stemmed from 1 % systematic error in neutron counting, 5 % in calibration of the small BF3 counter and 1.7 % in determination of the source intensity. Fig. 2 shows the total efficiency of the neutron detector as a function of neutron energy. The efficiency was also measured with a 252Cfsource whose average neutron energy is 2.35 MeV. The energy dependence of the efficiency was calculated with a Monte Carlo code MCNP [15]. 3. RESULTS
Cross sections were obtained with a formula
8
/
'.N~OH
+ D20
Monte Carlo Cal. MCNP
~
Figure 2. Neutron detection efficiency as a function of neutron energy. The efficiency was experimentally determined with sources of 24NaOH D 2 0 and 252Cf. The neutron energy dependence of the efficiency was calculated with the MCNP Monte Carlo code.
+
where Pin is the number of neutrons detected, N, is the number of y rays incident on the BGO detector, Nt is the number of target nuclei per cm2, and E,(E,) is the neutron detection efficiency. A correction factor, f,was needed for the measurement with a thick target.
where p is the total y-attenuation coefficient in 9Be, and t is the length of the 'Be target material. The attenuation coefficient was taken from [16]. Fig. 3 shows measured photoneutron cross sections as a function of y energy. Note that results for unpolarized y rays (open circles) agree with those for polarized ones (solid circles) within statistical uncertainties. The present measurement fully covered the excitation-energy range of three resonances (512-, 112-. 5/2+), while it covered only a high-energy tail of the first excited state (1/2+). This is because as approaching to the 1.665-MeV threshold, it became difficult to accurately evaluate what fraction of the y-ray flux lay above the neutron threshold. A Monte Carlo calculation of the asymmetric energy spread of LEMPS may help deduce the necessary y flux. This will be done in the future. Currently, a least-square fit to the data is in progress to deduce resonance parameters (7-decay widths etc.) towards an evaluation of the relevant reaction rate.
Present results (polarized) Present results(unoolarized) Fujishiro et al. (i982) RI BHarnermesh et ai.(1953) B.Russell et al. (1948) RI J.H.Gibbons et a1.(1959) RI W.John et a1.(1962) RI ..... Jakobson et al. (1961) Bremss. o
1.5
2.0
2.5
1
3.0
3.5
4.0
1 4.5
E (MeV) 7
Figure 3. Photoneutron cross sections for gBe. For comparison, results of the preceding measurements with Bremsstrahlung and radioactive isotopes are also shown.
4. ACKNOWLEDGMENT
We are indebted to Mr. K. Okamoto (Research Reactor Institute of Kyoto University) for his technical assistance in the irradiation of NaOH with thermal neutrons, to Mr. Y. Ogawa (Kinki University) for his help in running the MCNP code, and to Dr. K. Kudo (Electrotechnical Laboratory) for his cooperation in the detector calibration with the standard graphite pile. This work was supported by the Japan Private School Promotion Foundation.
REFERENCES 1. S.E. Woosley and R.D. Hoffman, Astrophys. J . 395 (1992) 202. 2. M.K. Jakobson, Phys. Rev. 123 (1961) 229. 3. B. Russel e t al., Phys. Rev. 73 (1948) 545. 4. B. Hamermesh and C. Kimball, Phys. Rev. 90 (1953) 1063. 5. J.H. Gibbons et al., Phys. Rev. 114 (1959) 1319. 6. W. John and J.M. Prosser, Phys. Rev. 163 (1967) 958. 7. M. Fujishiro et al., Can. J . Phys. 60 (1982) 1672. 8. M. Fujishiro et al., Can. J . Phys. 61 (1983) 1579. 9. H.-G. Clerc et al., Nucl. Phys. A120 (1968) 441. 10. G. Kuechler et al., Z. Phys. A326 (1987) 447.
11. H. Toyokawa et al., Nucl. Instr. and Methods in Phys. Res. A422 (1999) 95. 12. H. Ohgaki et al., IEEE Tran. Nucl. Sci. 38 (1991) 386. 13. Experimental nuclear physics, Vol. 2, edited by E. Segre (John, Wiley & Son, Inc., New York, NY 1978) p. 455. 14. Y. Inoue et al., Bull. Electrotech. Lab. 31 (1967) 60. 15. J.F. Briesmeister, MCNP, A General Monte Carlo N-Particle Transport Code, Version 4B, Las Alamos National Laboratory, 1997. 16. Engineering Compendium on Radiation Schielding, Vol. 1, edited by R.G. Jaeger, E.P. Blizard, A.B. Chilton, M. Grotenhuis, A. Honig, Th. A. Jaeger, and H.H. Eisenlohr (Springer-Verlag New York Inc. 1968) p. 173.
Hadron and Nuclear Physics with Electromagnetic Probes K. Maruyama and H. Okuno (Editors) 2000 Elsevier Science B.V. All rights reserved.
'
Nuclear disintegration induced by virtual photons at heavy-ion colliders I.A. Pshenichnova * t "Institute for Nuclear Research, Russian Academy of Science, 117312 Moscow, Russia A model of electromagnetic interaction of relativistic heavy ions which makes it possible to calculate the inclusive and exclusive characteristics of such interactions is presented. The multistep model includes (1) the single and double virtual photon absorption by a nucleus, (2) intranuclear cascades of produced hadrons and (3) statistical decay of excited residual nuclei described by evaporation, fission and multifragmentation mechanisms. Components of the model were verified by comparison of calculation results with experimental data obtained with monoenergetic photons and heavy ions at KEK-Tanashi Synchrotron and CERN SPS accelerator, respectively. 1. DESCRIPTION OF THE MODEL
The relativistic Coulomb excitation of nuclei is a well-known phenomenon [I]. It is expected to play an important role in collisions of ultrarelativistic heavy nuclei at new colliders, RHIC and LHC [2,3]. The process may be described in the framework of the Weizsacker-Williams method of equivalent photons. According to this method the impact of Lorentz-boosted Coulomb field of a nucleus on the collision partner may be treated as the absorption of equivalent photons. Due to the coherent nature of photon emission by the nucleus, such photons are almost real, Q2 5 1/R2, R is the nuclear radius. Therefore photonuclear reaction data obtained in experiments with real monoenergetic photons and appropriate theoretical models describing photonuclear reactions may be used to predict the properties of electromagnetic dissociation process. Since an ultrarelativistic nucleus moving at the impact parameter b with the velocity p = V / C % 1, (y >> 1) spends a short time At near the collision partner, the virtual photon spectrum contains all the frequencies up to the maximum energy E,ma" l / A t y / R :
where a is the fine structure constant, KO and K1 are the modified Bessel functions and x = E,b/(yp). The mean number of photons absorbed by the collision partner of mass A, is defined by:
*Supported by INTAS fellowship YSF-98-86 te-mail:
[email protected]
-
where OA,(E-() is the appropriate total photoabsorption cross section obtained starting from the photoneutron threshold at Eyin 7 MeV. Assuming the Poison distribution for the multiphoton absorption with the mean multiplicity m(b),Eq.(I), the integral cross sections are calculated for the first- and second-order processes [4] for a particular dissociation channel i :
with the following single and double photon spectral functions:
/ b d b e - m ( b ) ~ (b),~ l , N ( ~ ) ( EE2)~ , DC,
~ ( " (=4I n)
=n
bmm
/m
b d b e - m c b ) ~ (b)~N(E2, l, b).
bmtn
The values f:')(El) and f , ( 2 ) ( ~ 1 ,E2)defined as the branching ratios for the considered disintegration channel i in the single and double photon absorption, respectively, have to be calculated by our multistep model. Similar expressions may be written for the differential distributions of produced particles on the rapidity, transverse momentum and other variables. In order to obtain f:') and f,(2)several mechanisms are included in a new Relativistic ELectromagnetic DISsociation (RELDIS) code for the Monte Carlo simulation, namely, the intranuclear cascade of fast particles produced after the photon absorption on a nucleon or nuclear pair [5], and the evaporation of nucleons and lightest fragments, binary fission or multifragmentation [6] at a later stage of interaction. The multifragmentation process dominates when the excitation energy of residual nucleus, E*, exceeds 3 - 4 MeV/nucleon and is described by the statistical multifragmentation model (SMM) [6]. For fissile nuclei the evaporation may take place before or after the fission. The competition of evaporation and fission is also described with the SMM. The lightest fragments may be also created via coalescence of fast nucleons into dl t , 3He or 4He [7]. The decay of highly excited light residual nuclei with A 16 is treated by the Fermi break-up mechanism 161. Other details of the calculation scheme may be found in Refs. [4,5,8].
<
2. EMISSION OF NUCLEONS AND LIGHTEST FRAGMENTS
Depending on the photon energy, E,, and mass number, A,, different processes take place in the nuclear photoabsorption. Due to the contributions of several mechanisms: y N -+ 7rN, y N -+ 27rN and y(np) + np the calculated double differential spectra d2a/dRdP of pions and protons have complex shapes above the pion production threshold [5]. The spectra of fast particles n+, n-, 7 and p predicted by the model were compared 1000 MeV including in Ref. [5] with available sets of experimental data at 140 E,
<
<
the data obtained with KEK-Tanashi, the 1.3-GeV Electron Synchrotron. A satisfactory description of the spectra was obtained. Fast hadrons produced after the photon absorption initiate a cascade of subsequent collisions with the intranuclear nucleons leading to the heating of a residual nucleus. Later the nucleus undergoes de-excitation by means of the emission of nucleons and fragments. Because of a low Coulomb barrier in light nuclei the rates of proton and neutron emission are comparable. Fragment spectra in the photoabsorption on a carbon nucleus are given in Fig. 1. The low energy part of the deuteron spectra is explained by the explosive Fermi break-up while the high energy part is attributed to the coalescence mechanism. Since the main part of fast nucleons is emitted in the forward direction, the coalescence contribution dominates at small angles. On the contrary, the distribution of deuterons from Fermi break-up is nearly isotropic. Because of the option to accelerate also light oxygen ions at RHIC, the
y(380 MeV)
+C
+d+X
y(580 MeV)
+C
+d+X
1 1 1 1 1 1 1 ' ' ~ 1 1 1
Figure 1. Deuteron emission in photoabsorption on carbon. Points: KEK-Tanashi data [9]. Solid histograms: results of the intranuclear cascade calculations with coalescence and Fermi break-up. Contributions from coalescence mechanism are shown by the dashed, dotted and dash-dotted histograms for different values of coalescence parameter [7]:po = 90,129 and 200 MeV/c, respectively.
calculations of the electromagnetic dissociation contribution for such ions should take into account both of the considered mechanisms.
The photoabsorption scenario is very different for heavy nuclei, gold and lead. In this case neutrons are the most abundant particles produced in the electromagnetic collisions of ultrarelativistic nuclei [4]. When a heavy nucleus absorbs photons in the Giant Resonance region, 6 5 E, 5 30 MeV, the evaporation model may be used with the assumption E* = E, which leads to the emission of one or two neutrons. On the other side of the equivalent photon spectrum, when E, reaches the value of several GeV, the multiple pion photoproduction on intranuclear nucleons becomes the main absorption mechanism. In this case up to 95% of the photon energy is released in the fast particles on average. Nevertheless, the remaining energy deposited in the residual nucleus is sufficient for evaporating many neutrons or even multifragmentation [8]. As it was found in Ref. [4]for PbPb collisions at SPS (158A GeV beam) and LHC (2.75A+2.75A TeV beams) the mean neutron multiplicities are 4.2 and 8.8, respectively. The same value for AuAu collisions at RHIC (100A + lOOA GeV) is equal to 7.2. The neutron multiplicity distributions are shown in Fig. 2. They are strongly peaked at the single neutron emission channel due to GDR decay, while there is a long tail of the multiple neutron emission originating from high excitations. These results provide important information for designing large-rapidity detectors and zero-degree calorimeters at RHIC and LHC.
Ll"
" I "
" I ' " ' I " " I U "
W
Figure 2. RELDIS predictions for multiplicity distributions of neutrons in the electromagnetic dissociation of P b nuclei at LHC and SPS energies (solid and dotted lines, respectively) and Au nuclei at RHIC energies (dashed lines).
3. CHARGE CHANGING REACTIONS AT SPS
CERN SPS accelerator currently supplies the highest energy available for heavy ion studies. A detailed understanding of fragmentation mechanisms at SPS energies provides an important point for the extrapolation of the theoretical and experimental results to RHIC and LHC energies.
<
<
Partial break-up charge changing cross sections (-7 AZ 1) for 158A GeV 208Pb ions interacting with various targets, from H to Pb, were measured recently [lo]. Although the contributions to fragmentation due to the electromagnetic and nuclear forces were not separated in the experiment [lo], one can use RELDIS code to evaluate the role of the electromagnetic dissociation for each particular dissociation mode. As it is shown in Fig. 3, the main part of charge changing cross sections in PbPb collisions may be explained by the electromagnetic interaction.
158A GeV '-Pb on '07pb
Figure 3. RELDIS results for fragmentation charge changing cross sections of 208Pbions at SPS. Points: experimental data [lo].
Figure 4. RELDIS predictions for mass distributions of isotopes with charges Z= 75,76, ...,82 produced in the electromagnetic dissociation of 208Pbions at SPS.
Calculations show that multiple neutron emission also plays a role iri each of the partial charge changing channels with AZ = -6, ..., -1 for PbPb collisions, see Fig. 4. The lost of each proton is accompanied with a high probability by the multiple neutron emission. For example, up to 20 neutrons may be emitted in AZ = -2 fragmentation channel, and the emission of 8-10 neutrons is the most probable process.
This is evident from the fact that due to a high Coulomb barrier of P b nucleus, photoemission of protons takes place well above the GDR region, particularly in the region of quasideuteron absorption, where the excitation energy, E* 20 - 50 MeV, is sufficient to evaporate many neutrons. On the contrary, the mass distribution for AZ = 0 channel is completely different since in this case In and 2 n emission in GDR region strongly dominates. However, a long tail of the multiple neutron emission also exists in this channel. Our expectation of the intense neutron emission in the charge changing electromagnetic , collisions of heavy nuclei is very different from a simple picture assumed in Ref. [ l l ]where the rates of neutron and proton emission above the photoproton threshold were determined by Ar/Z ratio of the nucleus which undergoes fragmentation. A detailed understanding of the fragmentation mechanism may be obtained by measurements of fragment masses and detection of neutrons produced in collisions of ultrarelativistic heavy ions.
-
4. CONCLUSIONS
A variety of fragmentation mechanisms takes place in the electromagnetic dissociation of ultrarelativistic heavy ions, namely the coalescence, Fermi break-up, evaporation, fission and multifragmentation. A partial or complete disintegration of the colliding nuclei is possible despite the absence of geometrical overlap of the nuclear densities. Comparison with CERN SPS data on PbPb collisions demonstrates the dominance of electromagnetic dissociation in the partial charge-changing fragmentation channels with AZ = -6, ..., -1. RELDIS model predicts also a very intense neutron emission in such channels. The author gratefully acknowledges the fruitful collaboration with A.S. Botvina, J.P. Bondorf, A.S. Iljinov and 1.N. Mishustin on the subjects of this talk. Discussions with A.B. Kurepin, G. Giacomelli, M. Giorgini, L. Patrizii and P. Serra are greatly appreciated. The author is very indebted to the Organizing Committee of the KEK-Tanashi Symposium for the kind hospitality and financial support.
REFERENCES 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11.
C.A.Bertulani and G.Baur, Phys. Rep. 163, 299 (1988). G.Baur, K.Hencken, D.Trautmann, J . Phys. G24, 1657 (1998). F.Krauss, M.Greiner and G.Soff, Prog. Part. Nucl. Phys. 39, 503 (1997). I.A.Pshenichnov, I.N.Mishustin, J.P.Bondorf et al., Phys. Rev. C 60, 044901 (1999). A.S.Iljinov, I.A.Pshenichnov, N.Bianchi et al., Nucl. Phys. A616, 575 (1997). J.P.Bondorf, A.S.Botvina, A.S.Iljinov et al., Phys. Rep. 257, 133 (1995). A.S.Sudov, A.S.Botvina, A.S.Iljinov, Nucl. Phys. A554, 223 (1993). I.A.Pshenichnov, I.N.Mishustin, J.P.Bondorf et al., Phys. Rev. C 57, 1920 (1998). K.Baba, I.Endo, H.Fukuma et al., Nucl.Phys. A444, 578 (1986). H.Dekhissi, G.Giacomelli, M.Giorgini et al., Nucl.Phys. A, 1999, in print. S.E.Hirzebruch, E.Becker, G.Hiintrup et al., Phys. Rev. C 51, 2085 (1995).
IV. STRANGENESS PHYSICS
ThisPageIntolyLfBk
ThisPageIntolyLfBk This Page Intentionally Left Blank
Hadron and Nuclear Physics with ElectromagneticProbes K. Maruyama and H. Okuno (Editors) 2000 Elsevier Science B.V. All rights reserved.
"
Kaon photoproduction on nuclei
" Physics Department, Graduate School of Science, Tohoku University, Sendai 9808587, Japan The photoproductions of a A through the 12C(y,K + ) and 3He(y,KO) reactions were studied near the threshold region. The experimental data were compared with a numerical calculation in terms of the quasifree A photoproduction. The cross sections were also discussed by comparing with the different model predictions. We discuss the use of polarized photon beams for the (y, K) reactions. It enables us to study the elementary kaon photoproduction amplitudes in detail. 1. INTRODUCTION
Photonuclear reactions associated with a strangeness degree of freedom, such as (y, KO) and (7,K + ) reactions, leave S = -1 in nuclei converting a nucleon into a hyperon. A hyperon production in a nucleus provides a unique opportunity to study fundamental information on the nucleon-hyperon and the nucleus-hyperon interactions.[l] They can be studied by use of the quasifree (QF) process and the hyper-nucleus spectroscopy. Until now, these investigations have been done using hadoronic probes. In particular, the ( T + , K + ) reactions have been used extensively to study the production and decay of A-hypernuclei. [2] The A-producing (7,K + ) and (y, KO) reactions are believed to be complementary to the hadoronic reaction.[3] The dominance of the spin-flip amplitudes in (y, K ) near threshold region enables us to study the spin dependent behavior of a A in nuclear potential. Since the mean free path of y in nuclear medium is much longer than any hadrons, we can probe A in nuclei with less distortion. In spite of the importance of (y, K ) measurements on nuclear targets, suitable y beams have not been available. High duty and high intensity y beams in GeV-region are required to study the A photoproductions because of the small (7,K ) cross sections and the requirements of the photon tagging. In this article, we describe the 12C(y,K + ) and 3He(y,KO) data near the threshold region. These measurements are the first observation of the K + and KO photoproduction on nuclei using the tagged photon beam. The experimental data will be introduced and compared with theoretical interpretations. Finally, we notice the importance of the detailed study of the elementary kaon production amplitudes, which can be investigated by using the polarized photon beams. They can be done at the Laser-Electron-Photon facility at SPring8.141 They will give us a complete data set for the elementary (7,K ) reaction amplitudes.
2. TAGX EXPERIMENTS 2.1. K+ photoproduction at TAGX 2.5
I
I
I
,
1
Inclus~vcI2C(y.K+)Cross Section
.
8
I
. I GcV
+-1.0-1
50
0 0.7
0.8
0.9
1.0
I 1
1.2
E., (GcV)
Figure 1. Differential cross section for 12C(y,K + ) reaction. TAGX data were compared with the elementary cross section and the numerical estimates in terms of the quasifree process.
11.2
1
1
114
115
11.6
11.7
Missing Mass ( G C V I C ~ )
Figure 2. Missing mass distribution for 12C(y,K t ) reaction in the photon energy range of 1.0 - 1.1 GeV. Lines represent the results of the Monte Carlo calculation. The arrow indicates the mass of the i2B ground states.
The 12C(-y,K + ) reaction was carried out using the TAGX spectrometer[5] at the 1.3GeV Tokyo Electron Synchrotron Laboratory.[6] The incident photon energy range was E, = 0.78 - 1.1 GeV. The charged particle events were momentum analyzed from the hits of Si-Strip Counters and Cylindrical Drift Chambers in the magnetic field. The timing and triggering counters were inner- and outer-scintillation-counter hodoscopes, and a time-offlight wall. They covered the angular ranges of 10 to 40 degrees on the left-hand side and 0 to 30 degrees on the right-hand side. A typical momentum resolution was 6% for protons of 0.4 GeV/c. Figure 1 shows the obtained differential cross section. Solid squares show the TAGX data compared with an average of the elementary cross sections multiplied by a factor of four. Thick curves show the Monte Carlo calculations, in which the QF process is assumed. The proton momentum distribution were taken from the 12C(e,e'p) data[7]. The dot-dashed curves represent the contributions of two protons in the p shell and four protons in the s shell, respectively. The thick solid line shows the sum of these two contributions. We can understand that the nuclear Fermi motion causes the K+ yield to rise more slowly near the threshold than the elementary process. Figure 2 is the missing mass distribution for the kaon events in the energy interval of 1.0 - 1.1 GeV. The curves show the QF calculation. We can see a broad peak structure centered a t 11.55 GeV/c2. It is reproduced well by the sum of the kaons from p and s
shells through the QF process. We find an excess of the data above the QF calculation in the missing mass region below 11.4 GeV/c2. The bound A states are expected in this region. The integrated differential cross section over this region is 0.21 f O.OSpb/sr. The effective proton number can be interpreted as a number of the participating nucleons in the reaction process. Since a proton changes to a A, the effective nucleon number is replaced by an effective proton number in (y, K + ) process. Our result of effective proton number is 4.2 f 0.6. It suggest the (y, K + ) is a useful tool to probe the nuclear interior with less distortion. Theoretical investigations of 12C(?.K + ) have been done in terms of the quasi-free process leading to an unbound A with a three body model,[8,9] and an intranuclear cascade model[lO]. Lee et al. [8] predicted both the inclusive and exclusive 12C(y,K + ) cross sections by using the recently developed amplitudes[ll] of yp -t K+A and wave functions from the relativistic mean-field theory[12]. They obtained O.19pb/sr for the sum of the cross sections for all bound A states. which is close to the experimental value 0.21 f0.05pb/sr. The inclusive cross section in the energy region up to E, = 1.1GeV was calculated based on the three body model. Their numerical estimates are shown in Figure 1. The thin-dotted line is the calculated cross section of the exclusive 12C(y,K+)i2B. The thin-dashed curve is the cross section of the QF 12C(y,K + ) X process. The thin-solid curve is the sum of these two contributions. The QF A production dominates the (y, K') process in the threshold region. A reduction factor 2.2 was used to fit the data. It is mainly due to the medium effects on y, K + and A. When we use the experimental value of the total y N and K + N cross section. the reduction factor become R 1.6. It suggest that the nuclear medium effects on the A propagation in nuclei must be significant. Another numerical approach is an intranuclear cascade model. [lo] It describes the y A reactions in the framework of a two-step cascade process. The first step is a rapid process. The intranuclear cascade develops through the binary collisions. The second step is the decay process of the excited nucleus. They include both the meson production channel and multicollisional intranuclear cascade process. The general behavior of the cascade calculation is in good agreement with TAGX data. The subthreshold cross section can be explained by Fermi motion and effective mass of bound nucleons.
-
-
+
2.2. KO photoproduction at TAGX The KO photoproduction on3He were measured using the TAGX spectrometer. The invariant mass spectrum constructed from two pion events is shown in Figure 3, where the photon energy range is above the reaction threshold for the elementary KO photoproduction and the vertex cut was employed outside of the target cell. We can identify the KO signature at -- 0.5 GeV/c2 in the invariant mass spectrum. In order to estimate the experimental cross section, the TAGX Monte Carlo was used to determine the detection efficiency.[5] We obtained the cross section 0 = 0.69pb for Ko-short photoproduction on 3He. Therefore the g , , ~ o becomes to about 1.4pb, which is similar to a,,K+ at E, 1 GeV. The differential cross section is shown in Figure 4. Although the K+ cross sections are strongly peaked in forward direction, the TAGX data does not show the forward peaking. It is understood by a lack of the t-channel diagram in the elementary KO photoproduction. The curves in Figure 4 are the numerical calculation using a computer code provided
-
Target (Full - Empty)
3
: 3 ~ e ( y , ~ 0TAGX ) DATA : E,= I . I GeV M=499.38 MeVlc2
h
L
25 tn
V
500
1000
MASS (MeV)
Figure 3. Invariant mass spectrum constructed from two pion events. The shadow indicates the KO events a t M -0.5GeV/c2.
21 -
:
-
-
AS1 C4 ....... AW4
.
0 0
60
1 20
180
ec.~.
Figure 4. Differential cross section for 3He(y,KO). Lines show the numerical estimates with different models.
by Sotona.[l3] Input parameters for the production amplitudes are obtained from K + photoproduction models.[l4-161 It is reported that the predicted (y, K f ) cross section using these different models show rather similar behavior in the 1 GeV energy region. In Figure 4, different model predictions denoted by AS1[14], C4[15] and AW4[16] are compared with experimental data. The numerical values of the (y, KO) cross sections are strongly model dependent in contrast to the K + photoproduction. 3. KAON PHOTOPROCUCTION WITH POLARIZED PHOTON BEAMS
In the A producing kaon photoproduction, a spin 0 kaon is photoproduced on a spin 112 nucleon leading to a spin 112 A. There are eight possible spin states in the system. Therefore the elementary kaon photoproduction process can be expressed with four independent amplitudes. The defined experimental observables in (y, K ) are differential cross sections, three single polarization asymmetries of P(recoil), C(beam) and T(target), and twelve double polarization asymmetries. It is sufficient to measure eight observables among them t o complete the data sets for the analysis of elementary N(y, K)A processes. The Laser-Electron-Photon facility a t Spring8 provides almost 100% polarized photon beams.[4] The photon beams are tagged in an enegy range from 1.5 to 2.4 GeV with the energy spread of 15 MeV. The average intensity is 107/sec which is limited by the light source operation. A large acceptance charged particle spectrometer will be completed in 2000. In this condition, we have a chance to measure the elementary A producing kaon photoproduction observables, doldS2, P, C and T in an energy range of E, = 1.5 - 2.4 GeV. It must be the most important information to study the kaon photoproduction on nuclei.
4. S U M M A R Y We measured the differential cross sections of the 12C(y,K + ) and 3He(y,KO) reaction in the threshold region at 1.3-GeV Tokyo electron synchrotron laboratory. The 12C(y,K + ) cross sections are compared with numerical estimates and theoretical interpretations. We found that the accurate study of the elementary (y, K ) process and nuclear medium effects of y, kaons, A propagation must be important for the future investigation of kaon photoproduction on nuclei. We noted that the measurements of the spin observables in (y, K ) is very effective to understand the kaon photoproduction. They can be done at new Laser-Backward-Compton facility in the near future. They will give us important information to study the A-nucleus interaction and the structure of the A-hypernuclei. REFERENCES 1. S.S. Hsiao and S. R. Cotanch, Phys. Rev. C28, (1983) 1668, M. Sotona, et al. A547 (1992) 63c, and C. Bennhold, Nucl. Phys. 547 (1992) 79c. 2. O.Hashimoto, S.Ajimura, K.Aoki, H.Bhang, T.Endo, Y.Fujii, 0 .Hotchi, E.Hungerford, J.H.Kim, Y.D.Kim, T.Kishimoto, K.Koshino, K.Kubota, K.Maeda, T.Nagae, H.Noumi, Y.Ohta, K.Omata, H.Outa, H.Park, Y.Saito, T.Saito, Y.Sato, M.Sekimoto, T.Shibata, T.Takahashi, T.Tamagawa, H.Tamura, L.Tang, H.Tanita, M.Youn, Nucl.Phys. A629, (1998) 405c and references therein. 3. C. B. Dover, Nucle. Phys. A547 (1992) 27c. 4. T.Nakano, H.Ejiri, M.Fujiwara, T.Hotta, K.Takanashi, H.Toki, S.Hasegawa, T.Iwata, K.Okamoto, T.Murakami, J.Tamii, K.Imai, K.Maeda, K.Maruyama, S.Date, M.M.Obuti, Y.Ohashi, H.Ohkuma, N.Kumagai, Nucl.Phys. A629, (1998) 559c 5. TAGX Collaboration: Nucl. Instrm and Methods A376 (1996) 335. 6. K.Maeda, et all Nuclear Physics A577 (1994) 227c, and H.Yamazaki,, et al, Physical Review C 52, R1157 (1995). 7. J . Mougey, M. Bernheim, A. Bussikre, A. Gillebert, Phan Xuan Ha, M Priou, D. Royer, I. Sick and G. J. Wagner, Nucl. Phys. A262, 461 (1976). 8. T.S. Lee, Z.Y. Ma, B. Saghai and H. Toki, Phys. Rev. C58 (1998) 1558. 9. E. Ya. Paryev, private communication. 10. S. de Pina, E. C. de Oliveira, E. L. Medeiros, S. B. Duarte and M. Gonalves, Physics Letters B434 (1998) 1. 11. J . C. David, C Fayad, G. H. Lamot and B. Saghai, Phys. Rev. C53 (1996) 2613. 12. Z. Ma, J. Speth, S. Krewald, B. Chen, and A. Ruber, Nulc. Phys. A608 (1996) 305. 13. M. Sotona, K. Itonaga, T. Motoba, 0 . Richter and J . iofka, Nucl. Phys. A547, 63c (1992), and M. Sotona and S. Frullani, Prog. Theor. Phys. 1 1 7 (1994) 151. 14. R. A. Adelseck and B. Saghi, Phys. Rev. C42 (1990) 108. 15. R. Williams, Ch. R. Ji and S. R. Cotanch, Phys. Rev. C46 (1992) 1617. 16. R. A. Adelseck and L.E. Wright, Phys. Rev. C38 (1988) 1965.
ThisPageIntolyLfBk
ThisPageIntolyLfBk This Page Intentionally Left Blank
Hadron and Nuclear Physics with Electromagnetic Probes K. Maruyama and H. Okuno (Editors) 2000 Elsevier Science B.V. All rights reserved.
"
Subthreshold and near threshold K f meson photoproduction on nuclei E.Ya. Paryev
"
"Institute for Nuclear Research, Russian Academy of Sciences, Moscow 117312, Russia The inclusive K + meson production in photon-induced reactions in the near threshold and subthreshold energy regimes is analyzed with respect to the one-step ( y N 4 K + Y , Y = A, C) incoherent production processes on the basis of an appropriate new folding model, which takes properly into account the struck target nucleon removal energy and internal momentum distribution (nucleon spectral function). Comparison of the model calculations of the K + differential cross sections for the reaction y C12 in the threshold region with the existing experimental data is given. The predictions for the K + total and differential cross sections from yC12 and yPb208 reactions at subthreshold and near threshold energies are provided.
+
An extensive investigations of the production of K + mesons in proton-nucleus reactions [I-91 at incident energies lower than the free nucleon-nucleon threshold have been carried out over the last years. There are essentially less extensive investigations [lo, 111 of the inclusive subthreshold kaon production in pion-nucleus reactions. Finally, the electromagnetic production of K + mesons on nuclei in the threshold region has up to now received very little consideration [12], probably, because of a lack of suitable facilities and associated detectors. The inclusive (y, K+) measurements on nuclear targets in the threshold region are planned to be conducted in the near future at the Continuous Electron Beam Accelerator Facility (CEBAF) [13, 141 using the tagged photon beam in the CLAS (the CEBAF Large Acceptance Spectrometer) detector system. It is clear that, in order to analyze the results of such measurements, a relevant formalism has to be developed. The main goal of the present work is to extend the spectral function approach [6, 91 that has been employed for the description of the measured total [l]and differential [4, 71 kaon production cross sections from pA collisions in the near threshold and subthreshold energy regimes to K+-producing electromagnetic processes as well as to study the role played by nucleon-nucleon correlations in these processes. In this paper we present the predictions for the K + total and differential cross sections from y C12 and y P b 2 0 8 reactions in the threshold energy region obtained in the framework of the first collision model [6, 151 based on nucleon spectral function and compare part of them with the available data. An incident photon can produce a K + directly in the first inelastic y N collision due to nucleon Fermi motion. Since we are interested in the bombarding energy region up to approximately 1.4 GeV, we have taken into account the following elementary processes which have the lowest free production thresholds (respectively, 0.911, 1.046 and
+
+
1.052 GeV):
y+p+ K++A,
In our method the K+ production cross section for yA reactions from primary reaction channels (1)-(3) can be expressed [6, 151 as the respective integral of the free inclusive elementary K+ production cross sections [15] and nucleon spectral function P(pt,E ) over the struck target nucleon momentum pt and removal energy E . The nucleon spectral function P(pt,E) represents the probability to find in the nucleus a nucleon with momentum pt and removal (binding) energy E. The specific expressions for the high momentum and high removal energy part (correlated part) of the nucleon spectral function as well as for the single-particle (uncorrelated) part of the one given in [6, 151 were used in our calculations of K+ production in yC and yPb collisions.
Figure 1. Differential cross sections for K+ production in y+C12 reactions in the angular domain lo0 O K + 2 40° in the lab system as functions of the laboratory energy of the photon.
<
Figure 1 shows a comparison of the calculated differential cross sections for the production of K+ mesons at the laboratory angles of 10' 5 OK+ 5 40' from primary yN + K+Y channels with the experimental data [12] for y C12 + K+ X reaction at the various bombarding energies. The solid and dot-dashed lines represent our calculations with the total nucleon spectral function for primary production processes (1)-(3) and (2), (3), respectively. The dashed line denotes the same as solid line, but it is supposed that the total nucleon spectral function is replaced by its correlated part. The arrow indicates the threshold for the reaction yp -t K+A occuring on a free proton. One can see that:
+
+
1) the contributions to the Kf production from the primary reaction channels (1) and (2), (3) with A and C particles in the final states are comparable at bombarding energies E, 2 1.2 GeV, whereas at lower incident energies the primary production process (1) is essentially more important than (2) and (3); 2) the kaon yield from the one-step Kt production mechanism is entirely governed by the single-particle part of the nucleon spectral function at all considered beam energies (0.8 GeV 4 E, 4 1.3 GeV), what makes difficult to extract the information on the high momentum and high removal energy components within the C12 target nucleus from the first kaon photoproduction experiment [12];
3) while our calculations for the one-step reaction channels (1)-(3) reproduce reasonably well the experimental data [12]in the energy region of interest, they nevertheless provide different energy dependence of the excitation function as compared to the data, what might be due to possible in-medium modifications of the elementary y N -+ K+Y reactions discarded in the present work.
Figure 2. Double differential cross sections for the production of K+ mesons at an angle of 10' in the interaction of 0.8, 0.9 and 1.3 GeV photons with the C12 nuclei as functions of kaon momentum.
Figure 2 presents the results of our calculations for the double differential cross sections for the production of K+ mesons at an angle of 10' in the interaction of photons with energies of 0.8, 0.9 and 1.3 GeV with C12 nuclei. The solid and dashed lines are our calculations for kaon production processes (1)-(3) with the use of total nucleon spectral function and its correlated part, respectively, and correspond from the top to the bottom 1.3, 0.9 and 0.8 GeV incident energy. It is seen that the main contribution to
the K+ production both at subthreshold and above the free yN threshold beam energies considered here comes from the use of the uncorrelated part of the nucleon spectral function in the calculation of the corresponding momentum-energy-averaged differential cross sections for kaon production, what makes rather difficult to extract the information on the correlated part of the n~icleonspectral function even through analysis of the experimental double differential cross sections for K+ production at adopted photon energies (E, L 0.8 GeV).
tL-li,
K'0 1
G(CeV1
Figure 3. Total cross section for K+ production in y C12 reactions as a function of the laboratory energy of the photon.
+
Figure 4. Total cross section for K+ production in y Pb208reactions as a function of the laboratory energy of the photon.
+
Finally, Figures 3 and 4 present the results of our calculations for the total cross sections for K+ production in y C12 and y + Pb208reactions, respectively. The arrows indicate the thresholds for the reactions yp -+ K+Co,yp --, K+A occuring on a free proton and the absolute production threshold. The rest of the notation is identical to that in Figure 1. It can be seen that in these cases the kaon yield from the one-step K+ production mechanism is almost completely determined by the correlated part of the nucleon spectral function only in the vicinities of the absolute reaction thresholds (at bombarding energies of E, 5 0.75 GeV).This conclusion is in line with our findings inferred above (cf. Figures 1 and 2) from the analysis of differential and double differential kaon production cross sections. The values of the total kaon production cross sections in the far subthreshold region (E, 5 0.75 GeV) are very small (in the range of 0.1-10 nb), but one should expect to measure these values at the CEBAF [13, 141. As in the preceding cases, the direct K+ production processes (2), (3) play a minor role in kaon production in y A interactions at beam energies of E, < 1.2 GeV. Thus, our results demonstrate that it is difficult to distinguish the contribution to K+ production in yA reactions from the high momentum and high removal energy part
+
of the nucleon spectral function which is generated by ground-state two-nucleon shortrange correlations inside the target nucleus with the measurements of the K+ total and differential cross sections at subthreshold photon energies. REFERENCES 1. V.P.Koptev, et al., ZhETF. 94 (1988) 1. 2. ASibirtsev and M.Buscher, 2.Phys. A347 (1994) 191. 3. W.Cassing, et al., 2.Phys. A349 (1994) 77. 4. M.Debowski, et al., 2.Phys. A356 (1996) 313. 5. ASibirtsev, et al., 2. Phys. A358 (1997) 357. 6. S.V.Efremov and E.Ya.Paryev, Eur. Phys. J. A1 (1998) 99. 7. A.V.Akindinov, et al., APH N.S., Heavy Ion Physics 4 (1996) 325. 8. A.Badala, et al., Phys. Rev. Lett. 80 (1998) 4863. 9. E.Ya.Paryev, Eur. Phys. J. A5 (1999) 307. 10. S.V.Efremov and E.Ya.Paryev, 2.Phys. A351 (1995) 447. 11. S.V.Efremov and E.Ya.Paryev, Z. Phys. A354 (1996) 219. 12. H.Yamazaki, et al., Phys. Rev. C52 (1995) R1157. 13. CEBAF experiment 91-014 ( C . E . Hyde-Wright spokesman). 14. B.A.Mecking, Nucl. Phys. A639 (1998) 559c. 15. E.Ya.Paryev, Preprint INR-lOO4/99, Moscow (1999); EPJA, in the press.
ThisPageIntolyLfBk
ThisPageIntolyLfBk This Page Intentionally Left Blank
Hadron and Nuclear Physics with Electromagnetic Probes K. Maruyama and H. Okuno (Editors) 2000 Elsevier Science B.V. All rights reserved.
"
Phenomenological aspects of kaon photoproduction on the nucleon T. Marta* , S. Sumowidagdoa*, C. Bennholdbt and H. Haberzettlbt "Jurusan Fisika, FMIPA, Universitas Indonesia, Depok 16424, Indonesia bCenter for Nuclear Studies, Department of Physics, The George Washington University, Washington, D.C. 20052, USA Using an isobar model which can reproduce the existing experimental data of kaon photoproduction on the nucleon we investigate some related phenomenological aspects, i.e. the hadronic form factors, contributions of kaon-hyperon final states to the anomalous magnetic moment of the nucleon, and missing nucleon resonances. By reggeizing the appropriate propagators we extend the model to the higher photon energy regime. 1. INTRODUCTION
A wealth of new high-statistics data on elementary kaon photo- and electroproduction has recently become available in three isospin channels. Along with some new progress in the theoretical side this has made the field of kaon electromagnetic production to be of considerable interest. New models, span from chiral perturbation theory to the relatively simple isobar approach. have been proposed in the recent years as the SAPHIR collaboration made their precise data publicly available. Because the error-bars are sufficiently small, an interesting structure can be resolved in K'A total cross section. This leads to a critical question, as to whether the structure comes from less known resonances or other reaction channels start to open at the corresponding energy. In this paper we discuss some phenomenological aspects, which can be investigated by means of the isobar model. The model has been constructed by including three states that have been found to have significant decay widths into KA and K C channels, the 5 1 1 (1650), Pll (1710), and P13(1720) resonances, to fit all elementary data by adjusting some free parameters. which are known as coupling constants.
2. THE MODEL AND SOME PHENOMENOLOGICAL ASPECTS 2.1. Hadronic form factors Previous analyses of kaon photoproduction have never included a form factor at the hadronic vertex. However, since most of the present isobaric models diverge at higher energies, the need for such hadronic form factors has been known for a long time. Furthermore, it has been demonstrated that models which give a good description of the (7,K + ) data can give unrealistically large predictions for the (y, KO) channels [I]. It 'Supported in part by the University Research for Graduate Education (URGE) grant. +Supported in part by US DOE with grant no. DE-FG02-95ER-40907
is well known that incorporating a hadronic form factor helps alleviate this divergence and, simultaneously, leads to a problem with gauge invariance, since not every diagram in the Born terms retains gauge invariance by i t ~ e l f .The ~ question of gauge invariance is actually one of the central issues in dynamical descriptions of how photons interact with hadronic systems. While there is usually no problem at the tree-level with bare, pointlike particles, the problem becomes very complicated once the electromagnetic interaction is consistently incorporated within the full complexity of a strongly-interacting hadronic system. In the previous work [2] we have studied the influence of hadronic form factors on kaon production, by multiplying the whole amplitude with an overall, monopole, form factor F ( t ) , i.e.
where the subscripts B and R refer to the Born and resonance terms, to simulate the average effect of the fact that nucleons are not point-like. In spite of the success to suppress the divergence and to avoid the problem of gauge invariance, this ad hoc fashion does not have any microscopic foundation. In order to restore gauge invariance properly, one needs to construct additional current contributions beyond the usual Feynman diagrams to cancel the gauge-violating terms. One of the most widely used methods is due to Ohta [3]. For kaon photoproduction off the nucleon, Ohta's prescription amounts to dropping all strong-interaction form factors for all gauge-violating electric current contributions in Born terms. Symbolically, this may be written as
The recipe, however, does not completely solve the problem of divergence, since the electric terms do not have suppression and, therefore, could violently increase as a function of the coupling constants. As shown in Ref. [4], even at the coupling constants values accepted by the SU(3) symmetry, Ohta's recipe already yields very large x2. On the other hand, Haberzettl [5] has put forward a comprehensive treatment of gauge invariance in meson photoproduction. This includes a prescription for restoring gauge invariance in situations when one cannot handle the full complexity of the problem and therefore must resort to some approximations. In our language, this method can be translated as
+
+ +
with P(s,t, u)= a l F ( s ) a 2 F ( t ) + a3F(u) and al a2 a3 = 1. Clearly, Haberzettl's method removes the Ohta's problem by an additional form factor in the electric terms. By fitting to the kaon photoproduction data we found that the method proposed by Haberzettl t o be superior rather than the Ohta's, since the former can provide a reasonable description of the data using values for the leading couplings constants close to the SU(3) prediction. Such couplings cannot be accommodated in Ohta's method due to the absence of a hadronic form factor in the electric current contribution. 3 ~ i n c ethe resonance terms are individually gauge invariant, the discussion will be limited t o the Born terms.
Table 1 Numerical values for the contribution of kaon-hyperon final states to the square of anomalous magnetic moments of proton and neutron. Column ( 1 ) is obtained from Eq. ( 5 ) ,while column ( 2 ) is evaluated by using Eq. ( 6 ) . Experimentally, K: = 3.214 and K: = 3.660.
n;(K) Channel yp-+K+A yp K+CO y p --+ K°C+ Tot a1 -+
(1) -0.026 -0.024 -0.013 -0.063
&i(K) (2) 0.044 0.030 0.031 0.105
Channel y n -+ K O A y n -+ K+Cy n -+ K°CO Tot a1
(1) 0.075 -0.025 -0.019 0.031
(2) 0.110 0.050 0.031 0.191
2.2. The anomalous magnetic moment of the nucleon One of the important ground state properties of the nucleon is the anomalous magnetic moment, which exists as a direct consequence of its internal structure. More than 30 years ago Gerasimov, and independently Drell and Hearn, proposed that this ground state property is related to the nucleon's resonance spectra by a sum rule which was then called the Gerasimov-Drell-Hearn (GDH) sum rule [6]. In the limit of photon point, the sum rule may be written as
where g3/2 and 0112 denote the cross sections for possible combinations of the nucleon and photon spins. Experiment with polarized beam and target has been performed at MAMI with photon energy up to 850 MeV and data are being analyzed [7].Using higher photon energies, experiments have been planned at ELSA and JLab. For practical purpose, instead of Eq. ( 4 ) we use
where OTT, denotes the cross section with polarized real photon and target. In terms of polarization observables this cross section corresponds to the double polarization E [8]. Since there are no data available for OTTI, previous work [9]approximated Eq. ( 5 ) with
to estimate the upper bound of contributions, where OT represents the total cross section. To calculate Eqs. ( 5 ) and ( 6 ) we use our elementary operator with urn, = 2.2 GeV. The result is shown in Table 1. Our calculation yields values of K ; ( K ) = -0.063 and n : ( K ) = 0.031, or I n P ( K ) j / n p5 0.14 and n , ( K ) / n , 5 0.094. This shows that the kaonhyperon final states contributions to the proton's and neutron's magnetic moment are very small. An interesting feature is that our calculation yields a negative contribution , is obviously consistent for the K ; ( K ) and a positive contribution for the K ~ ( K )which with the result of Karliner's work [ l o ] .
Figure 1. Total cross section for K f A photoproduction on the proton. The dashed line shows the model
3.0
without the D13(1960) resonance, while the solid line is obtained by including the D13(1960) state. The new
52
5.
V
,2.0
-
-
0
0
SAPHIR data [12] are denoted by the solid squares, old data are shown by the open circles.
1.0 -
1.6
1.7
1.8
1.9
2.0
2.1
2.2
W (GeV)
2.3. Investigation of missing resonances A brief inspection to the particle data book reveals that less than 40% of the predicted nucleon resonances are observed in .irN + .irN scattering experiments. Quark model studies have suggested that those "missing" resonances may couple strongly to other channels, such as the KA and K C channels [ l l ] . Interestingly, the new SAPHIR total cross section data [12] for the p(y, K+)A channel, shown in Fig. 1, indicate for the first time a structure around W = 1900 MeV. Using the current isobar model we investigate this structure. As shown in Fig. 1, our previous model cannot reproduce the total cross section. Although a structure in total cross section data does not immediately imply a new resonance, the energy region around 1900 MeV represents a challenge not only because of possible broad, overlapping resonances, but also because there are additional production thresholds nearby, such as photoproduction of q', K *A, and KA* final states, which can all lead to structure in the K+A cross section through final-state interaction. Here, we limit ourselves only to the possibility that this structure is in fact due to one of the missing or poorly known resonances. The constituent quark model of Capstick and Roberts [ll]predicts many new states around 1900 MeV. However, only a few of them have been calculated to have a significant KA decay width [ l l ] . These are the S11(1945),P11(1975),P13(1950),and D13(1960) states. We have performed fits for each of these possible states, allowing the fit to determine the mass, width and coupling constants of the resonance. We found that all four states can reproduce the structure at W around 1900 MeV, while reducing the x2, but only for the D13(1960) state we found a remarkable agreement, up to the sign, between the quark model prediction and our extracted result [13]. The result is shown in Fig. 1, where without this resonance the model shows only one peak near threshold, while inclusion of the new resonance leads to a second peak at W slightly below 1900 MeV, in accordance with the new SAPHIR data. The difference between the two calculations is much smaller for the differential cross sections. The largest effects are found in the photon asymmetry. Therefore, we would suggest that measuring this observable is well suited to shed more light on the contribution of this state in kaon photoproduction.
Figure 2. Total cross section for kaon photoproduction on the proton. The dashed line shows the isobar model with hadronic form factors, but without reggeization. The solid line is obtained by reggeizing the K and K* propagators in the model. For K+ photoproduction, notation for the data is as in Fig. 1, for KO production old (open circles) and preliminary data (solid circles) are shown [14]. 2.1
2.2
W (GeV)
3. EXTENSION TO HIGHER ENERGIES
Extending the model t o the higher energy regime requires a non-trivial task, since the Born terms increase rapidly as a function of energy. As shown in Fig. 2, even the hadronic form factors are unable to suppress the cross sections for the energy region above 2 GeV as demanded by the data. Especially in the case of K°C+ production, where the predicted cross section starts to monotonically increase at this point. However, in order to explore the higher-lying nucleon resonances or to account for higher energies contributions to the GDH integral, an isobar model which also properly work at higher photon energies would be demanded. In Ref. [14] it has been shown that the contributions from the t-channel resonances are responsible for the divergence of the cross section, thus indicating that the Regge propagatlor should be used instead of the usual Feynman propagator. While a proper reggeization of the model is considerably complicated and the study is still underway, we investigate here only the qualitative effects of using Regge propagators in the model. Following Ref. [15], we multiply the Feynman propagators l / ( t - m i . ) of the K*(892)
and K1(1270) resonances in the operator with a factor of P R ~(t ~ m$.), ~ ~ where . PRegge indicates the Regge propagator given in Ref. 1151. For the K * intermediate state it has the form
where a ( t ) = a0 +a' t denotes the corresponding trajectory. Equation (7) clearly reduces to the Feynman propagator in the limit of t -+ mg., thus approximating the low energy behavior of the amplitude. The model is then refitted to kaon photoproduction data and the result is shown in Fig. 2, where we compare the isobar model with and without reggeization. Obviously, Regge propagators strongly suppress the cross section at high energies and, therefore, yield a better explanation of data at this energy regime. For the K°Cf process, the use of Regge propagators seems to give more flexibility in reproducing the cross section data. This cannot be achieved without reggeization, since the high energy behavior of both t-channel resonances is less controllable by the hadronic form factors. However, since the data for the K°C+ channel shown in Fig. 2 are still preliminary [16],we have to wait before any further conclusion can be drawn. In future we will include the high energy data in the fit and investigate the model in the transition between medium and high energy regions.
REFERENCES 1. T. Mart, C. Bennhold, and C.E. Hyde-Wright, Phys. Rev. C 51 (1995) R1074. 2. C. Bennhold, T. Mart, and D. Kusno in Proceedings of the CEBAF/INT Workshop on N* Physics, Seattle, USA, 1996 (World Scientific, Singapore, 1997), p.166. 3. K. Ohta, Phys. Rev. C 40 (1989) 1335. 4. H. Haberzettl, C. Bennhold, T. Mart, and T. Feuster, Phys. Rev. C 58 (1998) R40. 5. H. Haberzettl, Phys. Rev. C 56 (1997) 2041. 6. S.B. Gerasimov, Sov. J. Nucl. Phys. 2 (1966) 430; S.D. Drell and A.C. Hearn, Phys. Rev. Lett. 16 (1966) 908. 7. See W. Meyer, these proceedings. 8. R.L. Workman and R.A. Arndt, Phys. Rev. D 45 (1992) 1789. 9. D. Drechsel, Prog. Part. Nucl. Phys. 34 (1995) 181; H.-W. Hammer, D. Drechsel, and T. Mart, nucl-th/9701008; S. Sumowidagdo and T. Mart, Phys. Rev. C 60 (1999) 028201. 10. I. Karliner, Phys. Rev. D 7 (1973) 2717. 11. S. Capstick and W. Roberts, Phys. Rev. D 58 (1998) 074011. Tran et al., Phys. Lett. B 445 (1998) 20. 12. SAPHIR Collaboration: M.Q. 13. T . Mart and C. Bennhold, Phys. Rev. C 61 (2000) 012201(R). 14. F.X. Lee, T. Mart, C. Bennhold, H. Haberzettl, and L.E. Wright, nucl-th/9907119 and references therein. 15. M. Guidal, J.M. Laget, and M.Vanderhaeghen, Nucl. Phys. A 627 (1997) 645. 16. By the time of finishing this paper we realized that the final version of SAPHIR data for the y p + K°C+ channel has been published in: S. Goers et al., Phys. Lett. B 464 (1999) 331.
+
Hadron and Nuclear Physics with ElectromagneticProbes K. Maruyama and H. Okuno (Editors) 2000 Elsevier Science B.V. All rights reserved.
'
Hyperon polarization in Kaon photoproduction from the deuteron H. Yamamura", K. Miyagawa ", T. Mart W . Glockle
b,
C. Bennhold ', H. Haberzettl 'and
"Department of Applied Physics, Okayama University of Science, 1-1 Ridai-cho, Okayama 700, Japan bJurusan, Fisika, FMIPA, Universitas Indonesia. Depok 16424, Indonesia 'Center for Nuclear Studies, Department of Physcs, The George Washington University, Washington, D.C. 20052 dInstitut fur Theoretische Physik 11, Ruhr-Universitat Bochum, D-44780 Bochum, Germany
+
+
+
We analyze the reaction y d -+ K + A(C) N using the Nijmegen soft core hyperonnucleon interactions NSC97f and NSC89 and a recently updated production operator for the y N 4 K + A(E) processes. Significant effects of the Y N final state interaction are found near both the K + A N and K + E N thresholds.
+
+
1. INTRODUCTION
Since hyperon-nucleon scattering experiments are difficult to perform, hyperon producY N appear as tion processes such as y d -t K + Y N and e + d -+ e' K t natural candidates that allow exploring the Y N interaction. One can obtain the information of the Y N interaction by analyzing the correlated Y N final states. An inclusive d(e, e l K + ) Y N experiment has already been performed in Hall C at TJNAF, while the data for d(y. K + Y ) N are being analyzed in Hall B. Recently, we found that various meson-theoretical Y N interactions generate S-matrix poles around the AN and C N thresholds[l]. The pole near the E N threshold is related to the strength and the property of the AN - C N coupling and causes enhancements in the AN elastic total cross sections. The hope is that the pole structure of the Y N t matrix will have visible effects in such production processes mention above. In this paper, we study the inclusive d(y, K + ) Y N and exclusive d(y, K + Y ) N processes for OK = 0' and predict various observables including polarization observables.
+
+ +
+
+ +
2. FORMALISM
The reaction processes y
TiIQd>=
1Utj . t y i ' ~ >, j
+d
7-
K+
+ A(C) + N are expressed by the operator Tias
i, j = Ah[, EN,
= Figure 1. Inclusive d ( y ,K + ) cross section as a function of lab momentum p ~ for + 0" and the photon lab energy E, = 1.3 GeV. The two thresholds K f A N and K + C N are indicated by the arrows. The results around the K + C N threshold are enlarged in (b).
where the operator t y i describes the elementary processes y+ N + K++A(C), and j9,j > represents the deuteron state which is generated by the Nijmegen 93 N N interaction [2]. The operator Uijcorresponds to the Y N final state interaction processes, and is represented as
where V,, is Y N interaction including AN - C N coupling. We ignore the K + meson interaction with the nucleon and hyperon in the final states. From Eqs.(l) and (2), one can deduce the coupled set of integral equations for Ti,
We solve this set (3) after partial-wave decomposition in momentum space. The three elementary process y + p -+ K+ + A(Co) and y + n --+ K + + C- are properly incorporated in the driving term in Eq.(3). Equation (3) is solved on isospin bases A N and E N , but the resulting amplitudes are transformed into those on the particle bases An, Con and C - p by which the inclusive d(y, K + ) , exclusive d(y, K + Y ) cross sections and hyperon polarizations are calculated. For details, we refer the reader to ref.[3]. 3. RESULTS
At present, we calculate the observables only for the K+ meson scattered to 0 degree = 0"). The Nijmegen soft-core Y N interactions NSC97f[4] and NSC89[5] and a
(OK+
+
+
+
recently updated production operator[6] for the y N -+ K+ A(C) N processes are used. Figure l ( a ) shows the inclusive cross sections which sum up the contributions of the K'An, K + C O nand K+C-p final states. The solid and dashed lines are the predictions of the NSC97f and NSC89 Y N interactions, respectively. The dotted line shows the results of the plane wave impulse approximation (PWIA). The arrows indicate the two thresholds K + A N (pK = 977.30 MeV/c) and K + C N (pK = 869.14 MeV/c). The two pronounced peaks around p~ =945 and 809 MeV/c are due to the quasi-free processes of A and C, where one of the nucleon in the deuteron is spectator and has zero momentum in the laboratory system. Significant FSI effects are found around the K + A N and K f C N thresholds. The cross section is increased up to 86% by FSI near the K + A N threshold. Around the K f C N threshold, as shown Fig.l(b), the strengths and shapes of the enhancements by the NSC97f and NSC89 are quite different.
Figure 2. (a) Exclusive d ( y , K+A) cross section and (b) A recoil polarization with incoming polarized photon as a function of the A scattering angle in the An c.m. system. The photon lab energy is E, = 1.3 GeV. The outgoing kaon lab momentum and angle are p ~ =+ 870 MeV/c and 8 ~ =+ 0" respectively.
Figure 2(a) illustrates the exclusive d(y, K f A ) cross sections just below the K f C N threshold ( p K = 870 MeV/c). The FSI effects are seen both at very forward and at large angles. The PWIA cross sections are basically zero at backward angles, while the FSI calculations still show some strength. Figure 2(b) demonstrates the A recoil polarizations with incoming polarized photon. The A recoil polarizations in PWIA are almost one. This is because the incoming photon is polarized along the z axis, but the target deuteron is unpolarized and the outgoing K + meson carries no spin and angular momentum in this case (OK = 0"). However, the final state interactions cause the large deviations from one, and the prediction by NSC97f is quite different from that of NSC89.
- --
0.08 -
1
,
,
,
's
hN .-.,
-- NSC89
8
-.----------
-
0 0
I
\
u
0.02
,
- - - PWIA - NSC97f '
-
(b)
(a)
0.1
, 90
[degl
- , - -- , - . . 180
o
0
-
l
'
'
90
0,-,
'
'
[degl
Figure 3. (a) Exclusive d(y. K+C-) cross section and (b) C- recoil polarization with incoming polarized photon as a function of the C- scattering angle in the C-p c.m. system. The photon lab energy is E, = 1.3 GeV. The outgoing kaon lab momentum and angle are PK+= 865 MeV/c and Q K + = 0' respectively.
Figure 4. Inclusive d(y, K + ) cross section in the PWIA as a function of lab momentum p ~ + for Q K + = 0' and photon lab energy E, = 1.3 GeV. The solid curve shows the prediction by the present version[6] of the production operator for y N -t K+ + A @ ) , while the dashed line corresponds to the prediction by an old version[7].
+
The exclusive results to the K+C-p final states just above this threshold (pK =865 MeV/c) are shown in Fig.3. The prominent FSI effects are seen both in the cross sections and in the double polarization observable. Finally, we briefly discuss the production operator for 7 N -+ K+ A(C) processes. In Fig.4, the inclusive cross sections in PWIA in which an old version[7] of the operator is used are compared to those with the present version[6]. The latter has been improved in the fitting to the data of y + N -+ K+ + A(C) including the new SAPHIR data. The difference between the predictions by the two versions are quite large as in Fig.4, which suggests this reaction d(y, K + ) Y N is another promising candidate for investigating the operator.
+
+
REFERENCES 1. K. Miyagawa, H. Yamamura, Phys. Rev. C60 (1999) 024003; nucl-th/9904002. 2. V. G. J. Stoks, R. A. M. Klomp, C. P. F . Terheggen, and J. J. de Swart, Phys. Rev. C49 (1994) 2950. 3. H. Yamamura, K. Miyagawa, T . Mart, C. Bennhold, W. Glockle, Phys. Rev. C61 (1999) 014001; nucl-th/9907029. 4. Th. A. Rijken, V. G. J. Stoks, and Y. Yamamoto, Phys. Rev. C59 (1999) 21. 5. P. M. M. Maessen, T h . A. Rijken, and J . J. de Swart: Phys. Rev. C40 (1989) 2226. 6. C. Bennhold, T. Mart, A. Waluyo, H. Haberzettl, G. Penner, T . Feuster, and U. Mosel, in Proceedings of the Workshop on Electron-Nucleus Scattering, Elba, Italy, 1998, edited by 0.Benhar, A. Fabrocini, and R. Schiavilla (Edizioni ETS, Pisa, 1999), p. 149; nucl-th/9901066. 7. C. Bennhold, T . Mart, and D. Kusno, in Proceedings of the CEBAF/INT Workshop on N* Physics, Seattle, USA, 1996, edited by T.-S. H. Lee and W. Roberts (World Scientific, Singapore,1997), p.166
ThisPageIntolyLfBk
ThisPageIntolyLfBk This Page Intentionally Left Blank
Hadron and Nuclear Physics with ElectromagneticProbes K. Maruyama and H. Okuno (Editors) 2000 Elsevier Science B.V. All rights reserved.
'
Physics of associated strangeness production at ELSA Ewald Paul Physikalisches Institut, Universitat Bonn, Germany 1. INTRODUCTION
Associated strangeness production has been measured by the SAPHIR Collaboration [I] with the SAPHIR detector [2] at the Electron Stretcher and Accelerator ELSA in Bonn. Results will be reported from about 6% of the data on tape [3,4]. 2. THE STRANGE PARTICLE PROGRAMME AT SAPHIR
The SAPHIR experiment is devoted to measuring photoproduction with a beam of real photons at energies following a bremsstrahlung spectrum from electrons up to 3 GeV from ELSA. The strange particle reactions measured are essentially those with two-body final states:
Results from these reactions [3,4] are presented and discussed in this talk. Data analyses are in progress for the reactions:
The measurements with SAPHIR allow events to be reconstructed and identified over the whole angular space of produced charged particles. The results consist of total and differential cross sections, and transverse polarization of the hyperons. They are compared to models in three distinct parts. In the first part we consider isobar models. They describe the data at the level of hadronic degrees of freedom by fitting a superposition of Born terms and resonant contributions to the data. Parameters which are common with other reactions are determined in a coupled-channel approach. In the second part, the question whether hyperon polarizations also have a more general origin than stemming from interference of specific amplitudes is discussed. This is suggested since A and C states are polarized in similar manner for various beams and over
orders of magnitude in energy [5]. Moreover hyperons indicate the polarization of the s quark within the framework of the static quark model. If the s-quark carries polarization this implies that the polarizations of A and C states are related to each other. This is what is observed in general. The third part is concerned with chiral model calculations based on an effective chiral Lagrangian and on chiral perturbation theory. Such perturbative calculations describe pion photoproduction processes near threshold. When going to kaon-hyperon final states the mass of the kaon, which is considerably larger than the mass of the pion, implies new questions concerning the perturbative calculations directly as well as the handling of chiral symmetry breaking in general.
,
\;:,
drift chambers ,
:, ,
P
,
Figure 1. Topological and kinematical reconstruction of an event of the type yp
+
K°C'
3. THE EXPERIMENT
The ELSA accelerator provides electron beams in the energy range 0.5 to 3.5 GeV with nearly 100% duty cycle. Such a beam was used to produce bremsstrahlung photons which were tagged one-by-one in a tagging system for SAPHIR. The SAPHIR detector comprises a large magnet with a target inside filled with liquid hydrogen or deuterium and surrounded by a cylindrical drift chamber. Downstream and to the sides are planar drift chambers and large scintillator hodoscopes. Further downstream follows an electromagnetic calorimeter, which was not used for the strange particle data analyses. Details of the experimental setup are given elsewhere [2]. The SAPHIR drift chamber system is well suited to measure tracks of charged particles in the full angular range. Pions, kaons and protons are identified by time-of-flight measurements in the scintillator hodoscopes.
The measurements of the primary photon (in the tagging system) and of the charged particles in the central drift chamber are sufficient to reconstruct and identify complete events [6]. As an example we consider a measured event of reaction (3) which has been reconstructed and identified successfully (fig. 1). Three tracks were measured: two are the decay pions from K: -, T + T - decay; the other is the proton from C+ -+ p r o decay. The measurements of the pions yield the reconstruction of the K: at the decay point, the 3-momentum and line of flight in space. The 4-momenta of primary photon and Kfwere used to calculate (in a fit procedure) the 4-momentum of the C+. Then primary vertex and C+ decay vertex were determined simultaneously in an iterative procedure, where the spatial event topology is tested by moving the primary vertex along the K: line of flight. Finally the event kinematics was tested in fits at all vertices and overall. With this procedure the total acceptance of reactions (3) was close to 10%. The contamination by misidentified events from other reactions was negligible.
4 s [GeVI A: SAPHIR 1.5
0:ABBHHM
0.5
0
I
1
'
---..,
-----...-.--. ,L 1.2
1.4
1.6
1.8
E, [GeVI
Figure 2. Total cross sections.
4. EXPERIMENTAL RESULTS
Total reaction cross sections were measured over the photon energy range from threshold to 2 GeV for reactions (1) and (2) and 1.5 GeV for reaction (3), respectively. They are shown in fig. 2 in comparison with the only previous measurement in the energy range carried out with a bubble chamber 30 years ago [7]. The most striking observation is the difference in the rise at threshold: it is very steep for the K A reaction (1)and moderate for the KC reactions (2) and (3). The cusp effect where the KC reactions are opened is clearly visible. At larger energies the cross sections are similar for K+A and K+CO.The cross section for the K°C+ reaction (3) is about 112 of those of the other reactions. Differential cross sections are considered as a function of the production angle of K+ in yp center-of-mass system. They are determined in various energy bins for the reactions (1) and (2), as shown in fig. 3. The full line corresponds to a fit to Legendre polynomials for angular momenta up to 3 according to [3]:
The coefficients a. to a3 are shown in fig. 4. The K+A reaction (1) has a large s-wave contribution near threshold. In addition a p-wave and an s-p interference term give a substantial contribution. The KC cusp effect is visible. In fits to isobar models (shown below) the rise of the s-wave is caused by a significant resonance production of 5 1 (1650). The p-wave contribution is related to a superposition of the two resonances P1~(1710)and P13 (1720). The coefficients of the K+COreaction (2) show large s-, p- and d-wave contributions which peak around 1.4 GeV. Isobar model calculations identify S31(1900) and P31(1910) in this region. When extending the Legendre fits to higher waves (not shown) the data indicate also some f-wave contribution which may originate from known resonances [3]. The differential cross sections of reaction (3) are compared to reaction (2) in some wider energy bins in fig. 5. The main observation is that the distributions differ in the resonance region of reaction (2). Measurements of the transverse polarizations of A, C0 and C+ are shown in fig. 6. The polarization parameters are given as a function of the K+ production angle and for the reactions (1) and (2) in three energy bins. Two observations are made: 1. The angular shapes of A and C0 are nearly independent of energy and described well by a fit allowing angular momenta up to 1 (which is consistent with the Legendre fits shown above).
2. The polarizations of A and C0 have in general opposite signs. The polarization of C+ in reaction (3) was measured for the first time by SAPHIR. However, statistical errors are still large meaning conclusions about shape and sign cannot be drawn.
Differential cross sections: y p
-+ K + h
Fig. 3: Differential cross sections for y p fit explained in the text.
Differential cross sections: y p
-+ K f h
and y p
-+
K+c'.
-+ K+CO
The full line corresponds to a
-
w
-0 2
-02 E, IGeVI
E, [GeVI
E, IGeVl
E, IGeVI
15
15 2 E, [GeVI
E, [GeVI
E, [GeVI
06
-0 2 2
E, [GeVI
Figure 4. Coefficients of Legendre Polynomials.
cos(0.)
:0 4 m
203
A: 7 + p + K 0 + T '
1 250
L E, <
1 350 GeV
m: 7+p+K*+XD
Figure 5. Differential cross sections for yp
-4
K°C+ and yp
+
K f xO.
5. COMPARISON WITH MODELS 5.1 Isobar models Descriptions of strange particle reactions on the basis of isobar model calculations have been tried for a long time. Recently a real break-through has been obtained by including hadronic form factors (also for the kaons) along with a proper gauge prescription [8-101. The model calculations were fit to the data in a coupled-channel approach. The fit results yield a suppression of Born terms, so that mainly resonant contributions as they
A and C0 polarization
C' polarization 1.050 L E,
< ' 550 GeV
-0.5
-1
-1
-05
0
0 5
cos(0,)
Figure 6. Measurements of hyperon polarizations.
Total cross sections
3.0
Figure 7. Comparisons with isobar mlode1 calculations.
are discussed above describe the data. Comparisons with the data are shown in two examples in fig. 7. The full line corresponds to the most recent calculations [lo], the dashed and dotted line to previous ones. Total and differential cross sections prefer in general the full lines. This is particularly obvious for the K°C+ data of reaction (3). Neither calculation successfully describes the C0 polarization measurements [lo] (not shown). 5.2 Origin of hyperon polarizations It has long been argued that hyperon polarizations might be caused by dynamics when constituent quarks form hyperon states. In static SU(6) the polarization of A is identical to that of the s-quark while it is of opposite sign for the C states. This is what is observed, even in the change of signs between forward and backward production. A dynamical model to generate such a correlation was introduce by DeGrand and Miettienen Ill]. They assumed static SU(6) and quark recombination and introduced an empirical rule: accelerated quarks from the target are negatively polarized. Quarks from the projectile slow down and are positively polarized. For a photon beam a consistent picture can be obtained, if the photon acts as a vector meson according to the Vector Meson Dominance Model. Corresponding diagrams are shown in fig. 8. The rule is in qualitative agreement with most of the world data and with photoproduction data, but it fails to describe the kinematics, in particular the p~ distributions of the produced hyperons. A Japanese group has proposed a special spin-orbit potential which allows the correct p~ distributions to be calculated [12].
Figure 8. Quark recombination for y p
-4
K+A
5.3 Calculations with an effective chiral lagrangian. Kaiser and Weise defined an effective chiral Langrangian density without introducing any explicit resonance [13]. They carried out calculations in a coupled-channel approach. The calculations describe the total cross sections of all three reactions well over a considerable energy range (fig. 9). Calculations of differential cross sections and polarizations are not available so far. 5.4 Calculations in chiral perturbation theory Such calculations were carried out by Steiniger and Meissner [14]. They fixed lowenergy constants of the theory and the axial radius of the proton using SAPHIR data of reaction (1). Their calculations are compared to SAPHIR data in fig. 10. Total cross sections close to threshold and the shapes of the hyperon polarization are described fairly well, whereas differential cross sections in particular of y p -+ K + h are not well described.
Since there is no explicit resonance production in yp -+ K+COclose to threshold, it might be promising to start from a fit of the parameters to K+COdata instead of from K+A data. Otherwise it is mandatory to test current chiral perturbation calculations first of all as close as possible to threshold. Statistical errors of SAPHIR data are limiting at present the possibility to restrict the measurements to the threshold regions.
Figure 9. Comparisons with coupled-channel analysis with chiral Lagrangian function. The thin lines indicate s- and p-wave contributions, the fat lines their sums. 6. SUMMARY
Photoproduction of kaon-hyperon states is a rich field for testing physics close to threshold since measurements yield hyperon polarizations in addition to cross sections which constrain studies of dynamics on both hadronic and quark levels. A useful extension would be to carry out experiments with polarized photons- on polarized protons. Isobar models in multichannel analyses approach a consistent description of the available data, but they are less successful in describing the polarizations. -
Hyperon polarizations could be a more general phenomenon caused by dynamics of quarks when hyperons are formed.
c
o
Fig. 10 Comparisons with chiral perturbation calculations.
I
1.050 S E, < I .250 CaV - 1.1 5 0 GeV
E,,..
T
Studies on chiral symmetry with kaons and hyperons are at the beginning. More precise data at threshold are wanted. 7. OUTLOOK
The present evaluation comprises 6% of the SAPHIR data on tape. The forthcoming analyses will provide higher precision for the three considered reactions at both the statistical and systematic level. New results from other reactions, in particular with excited kaon and hyperon states, are expected. The SAPHIR experiment finished data taking and the detector has been dismounted in 1999. Beyond SAPHIR the strange particle program will be continued by the CLAS experiment at Jefferson Lab, and later also at Spring8 in Osaka and at MAMI C in Mainz.
REFERENCES 1. The SAPHIR-Collaboration: J . Barthl, C. ~ e n n h o l d ~W. , ~ r a u n l ,J. ~ r n s t ~ , K.H. Glanderl, S. Goers', J. Hannappell, N. ~ o ~ e nF.' , ~ l e i n lE. , Klempt2, A. KOzela3, R. Lawall', J . Link2, D. Menzel, W. Neuerburgl, M. Ostrickl, E. paull, H. van Pee2, R. P1Mzke2, I. Schuldayl, W.J. Schwillel, J . Smyrski3, B. Wiegersl, F.W. Wielandl, J . Wifikirchenl lPhysikalisches Institut, Universitat Bonn, 2~nstitutfiir Strahlen- und Kernphysik, Universitat Bonn, 3Jagellonian University, Krakow, 4Dept. of Physics, The George Washington University, Washington DC 2. W.J. Schwille et al., Nucl. Instr. Meth. A 344 (1994) 470. 3. M.Q. Tran et al., Phys. Lett. B445 (1998) 20. 4. S. Goers et al., Phys. Lett. B464 (1999) 331. 5. see e.g. E. Paul, Conf. Proc. Vol. 44, The ELFE Project, SIF Bologna, 1993, p. 379. 6. S. Goers, BN-IR-99-09 (1999), Thesis, Bonn. 7. ABBHHM Coll. (R. Erbe et al.), Phys. Rev. (1969) 188. 8. H. Haberzettl et al., Phys. Rev. C 58 (1998) R 40 and Phys. Rev. C 56 (1997) 2041. 9. C. Bennhold et al., Proc. of the CEBAFIINT Workshop on N* Physics, Seattle, USA, 1996, Singapore 1997, p. 166. 10. C. Bennhold et al., Proc. of the Workshop on Electron-Nucleon Scattering, Elba, Italy 1998, Edizione ETS, Pisa 1999, p. 149 and nucl-th/9901066. 11. T.A. DeGrand and H.I. Miettienen, Phys. Rev. D23 (1981) 1227; Phys. Rev. D24 (1981) 2419; Phys. Rev. D31 (1985) 661(e); T.A. Degrand, Phys. Rev. D38 (1988) 403. 12. Y. Yamamoto et al., Prog. Theor. Phys. 98 (1997) 95; K.-I. Kubo et al., ibid. 101 (1999) 615; K.-I. Kubo and Y. Kitsukawa, Tokyo Metropolitan Univ. preprint, 1999. 13. N. Kaiser, T. Waas and W. Weise, Nucl. Phys. A612 (1997) 297 and private communication from N. Kaiser. 14. S. Steininger and U.-G. Meissner, Phys. Lett B391 (1997) 446.
ThisPageIntolyLfBk
ThisPageIntolyLfBk This Page Intentionally Left Blank
Hadron and Nuclear Physics with Electromagnetic Probes K. Maruyarna and H. Okuno (Editors) 2000 Elsevier Science B.V. All rights reserved.
"
Retrospect and prospect of A hypernuclear physics Osamu Hashimoto
"
aDepartment of Physics, Tohoku University, Sendai 980-8578 E-mail:
[email protected] Recent progress and development of hypernuclear physics in the S=-1 regime was reviewed. Wide variety of spectroscopy as a tool to investigate A hypernuclear structure and AN interaction is discussed referring recent spectroscopic data, which have been obtained through the (T+, K+) reaction using a large acceptance high resolution spectrometer ( SKS ) a t KEK 1 2 GeV PS. Very recent result of hypernuclear y ray spectroscopy is also presented, where precision information on the spin dependent AN interaction can be extracted. New high resolution reaction spectroscopy by the (e,elK+) reaction underway at Jlab is described, too. Future prospect of hypernuclear physics is discussed based on these recent development.
1. INTRODUCTION The A hypernuclei provide unique opportunities in the investigation of hadronic manybody system with strangeness -1 and new aspects of the strong and weak interaction in nuclei involving the strangeness degree of freedom. Recent progress demonstrated the significance of A hypernuclear spectroscopy and established new experimental pathways towards these studies in a quantitative way. There are several aspects that hypernuclear investigation plays key roles in shedding lights on strangeness nuclear physics. Firstly, since a hyperon carries a strangeness degree of freedom with a strange quark, it behaves differently from nucleons in a nuclear system. Being free from Pauli exclusion principle, a hyperon can be a probe to study deeply bound states. It is also pointed out that investigation of A hypernuclear structure may answer the fundamental question whether a A hyperon deep inside a nucleus keeps its identity as a baryon or not.[l] Secondly, unique structures of hadronic many-body system will manifest themselves with the new quantum number, strangeness, since the interaction between a nucleon and a hyperon has new characteristics different from those of the nucleonnucleon interaction. Thirdly, weak interaction in nuclear medium can be well investigated through hypernuclear weak decay. The weak decay of a A hyperon in a nucleus shows unique features compared with those in free space. In particular, nonmesonic weak decay, A N --+ N N , is one of the best place to study baryon-baryon weak interaction in nuclei. It involves large momentum transfer and is dominant over the mesonic weak decay, A -, NT, in heavier A hypernuclei. Recent data on the nonmesonic weak decay poses a puzzle cencerning its decay widths and it is under intensive investigation both experimentally
and theoretically, though we will not go into the detail in the present paper. It can be said that A hypernuclear investigation has already undergone three stages of development since the first discovery of the hypernucleus in 1953. The first stage may be called as "emulsion era", when it was established that the A potential depth is about 213 of the nucleon case through the measurements of the binding energies of light A hypernuclei. The second stage began in early 70's when counter experiments became possible with K- beams at CERN and soon later at BNL. The reaction used for the hypernuclear investigation that time was mostly the (K-, n-) reaction. It was claimed that the spinorbit splittings in pshell hypernuclei are much smaller than the nucleon case. Progress of the first and second stages was reviewed in various literatures.[2]The third stage started in mid 80's at BNL utilizing the ( n f , K f ) reaction to populate A hypernuclear bound states, but it took sometime before new experimental facilities became fully available. Since late 80's until now, the superconducting kaon spectrometer (SKS) a t KEK 12GeV PS played a key role in exploring A hypernuclear spectroscopy by the (n+,K f ) reaction. Weak decay of A hypernuclei is also being studied with far better quality than ever obtained. Furthermore, new experimental tools are recently developed, such as hypernuclear y-ray spectroscopy and also electromagnetic production of strangeness. These will be also the subject of the present paper. 2. A HYPERNUCLEAR SPECTROSCOPY
A A hypernuclear state is formed in the reaction as a nucleon-hole A-particle state. As schematically shown in Fig.1 for a 208Pbtarget as an example, A binding energies are usually much larger than those for a proton or a neutron. Therefore, the hypernuclear states in which a A hyperon is bound above the p-shell orbital are mostly nucleon unbound and decay by emitting nucleons. However, the spreading widths of such high-lying states
neutron
Narrow spreading widths
X
< a few 100 keV Likar,Rosina,Povh Bando, Motoba, Yamamoto
207AT1
Bn - - - -_ --. I I I I
A
neutron 207APb
2Os,,Pb
Weak decay nonmesonic mesonic
Figure 1. A hypernuclear states and their decay modes.
are calculated to be narrower than a few 100 keV even they are excited above the nucleon emission thresholds, which is in contrast to the case of an ordinary nucleus.[3,4] When the hypernucleus deexcites below the particle emission threshold, y decays take place. Eventually, the A hypernucleus undergoes weak decays from its ground state. For those states that emit y-rays, precision 7-ray spectroscopy can be carried out as will be described later. Reaction spectroscopy will be applied to all the states regardless of the excitation energy. In addition to the narrowness of the spreading widths, perturbation by the presence of a A hyperon in the nuclear system is modelate because of the weakness of the interaction between a nucleon and a A hyperon. Therefore, information on A hypernuclear structure can be related to the AN interaction rather straightforwardly. Thus, the A hypernuclear spectroscopy is indispensable for the investigation of the AN interaction. Such characteristics are quite unique features realized in A hypernuclei and are hardly expected in other hypernuclei such as C and E hypernuclei. In this regards, the spectroscopy of A hypernuclei is in a unique position in the strangeness nuclear physics. The YN, YY effective interactions are constructed through G-matrix calculation, starting from phenomenological interactions in free space. Analytical functions of the effective potentials can be given in the form of three range gaussian, VAN(T)= C ( a z b,kf czk~)exp(-r2/bT), [5] It is also possible to express the effective AN interaction as, VAN= V ( r ) Vn(r)cA ON VA(r)L. OA VN(r)L . a N VTSAN,without reffering to the elementary interaction. Integration of the radial dependent potentials can be performed with appropriate wave functions. For the p-shell A hypernuclei, these integrals were assumed to be constant and designated as A, SA,SN and T.[6]Wide variety of hypernuclear properties, such as level structures, splittings due to the spin-dependent interaction, reaction cross sections etc, are then calculated. These calculations are reasonably reliable partly because the AN interaction is much weaker than the NN interaction and antisymmetrization with nucleons need not be considered. Therefore. once the information on the hypernuclear structure are obtained experimentally, we can trace the procedure backward and can investigate the free and effective hyperon-nucleon interactions. Hypernuclear spectroscopic data are directly compared and the spin-spin, spin-orbit and tensor interaction between a A hyperon and a nucleon are readily investigated. It is one of the most important aspect of A hypernuclear investigation, particularly because it is not easy to directly study the hyperon-nucleon interaction through hyperon scattering experiments. For these studies, high quality spectroscopic data which were recently accumulated provide valuable information on the An' interaction, being combined with these theoretical efforts.
+
+
+
+
+
+
3. THE ( n + ,K+) REACTION SPECTROSCOPY AND THE SKS SPECTROM-
ETER The spectroscopy by the (n+, K+) reaction was first applied to a carbon target at BNL, and later to heavier ones both at BNL and KEK.[7-101 The hypernuclear cross sections are often about 2 orders of magnitude smaller than those by the (K-, .ir-) reaction. However, the small cross sectios are well compensated by high intensity pion beams at 1.06 GeV/c where the cross section of the elementary process reaches a maximum. The reaction has
Table 1 Hypernuclear physics experiments with the SKS spectrometer E140a (Hashimoto,Tohoku) E278 (Kishimoto, Osaka) E307 (Bhang, Seoul) E336 (Hashimoto,Tohoku) E369 (Nagae, KEK) E419 (Tamura, Tohoku) E438 (Noumi, KEK)
Systematic spectroscopy of A hypernuclei ~ O B , fi2C, i8Si: :9Y, fiS9La, io8Pb Nonmesonic weak decay of polarized :He Lifetimes and weak decay widths of light and medium - heavy A hypernuclei Light A hypernuclear spectroscopy i L i , :Be, i2C, fi3C, i60
[9,10] [13] [I41 1151
Spectroscopy of igy
[I61
y ray spectroscopy of i L i
[171
C hypernnuclei by the (T-,K+) reaction
fall-winter 1999
Nonmesonic weak decay of :He
winter in 2000
an advantage to favorably excite spin-stretched states due to its large momentum transfer and is sutable to populate hypernuclear bound states. With an intension to take full advantage of the ( T + , K f ) reaction for the hypernuclear spectroscopy, a superconducting kaon spectrometer(SKS) was constructed at KEK 12 GeV PS. [11,12] The SKS spectrometer has good resolution of 1.5-2MeV FWHM and simultaneously covers a large solid angle of 100 msr, accepting about 60 % of i2C ground-state yield in the (T+,K+) reaction.[9] The large acceptance of the SKS spectrometer also allows us to perform efficient coincidence experiments such as study of hypernuclear weak decay and y ray spectroscopy. A series of experiments for the investigation of hypernuclei using the SKS spectrometer has been carried out in the past years and they are listed in Table 1. Among them, the four experiments, E140a, E336, E369 and E419 were for the reaction spectroscopy of A hypernuclei and studied A hypernuclear structure and the AN interaction through the measurement of high quality spectroscopic data. The E140a, the first of the series of experiments, intended to investigate the nature of a A hyperon deeply bound to heavy nuclei as well as light ones. The binding energies of a A hyperon bound to a wide range of hypernuclei, up to i9Y, i3'La and io8Pb were deduced from the excitation energy spectra. It was revealed that the single-particle nature of a A hyperon persists even in heavy nuclei.[lO] Much improved statistics and resolution were obtained for i9Y in the most recent E369 experiment.[16] In the E336 experiment which was carried out with the upgraded SKS spectrometer, the p-shell A hypernuclei as listed in Table 1 were investigated with high-quality spectra. It is noted that the yield rate of the i2C ground state is about 5 e v e n t ~ / ~ / c m ~ /pions, l O ~ that is, about 1000 i 2 C ground states are observed for a lg/cm2 thick target by the SKS spectrometer per day. In the i2C spectrum of the KEK E336 experiment, two satellite peaks(#2 and #3,3') were observed as first reported by E140a near the two prominent peaks (#1 and #4)(Fig.2). The #1 and #4 peaks are originated from configurations of vpi/2 @ A w 2
and up$ @ Ap1/2,3/2,respectively. The satellite peaks carry about 1/10 cross sections and were interpreted as originated from an sl,2 A hyperon coupled to excited states of the llC core. Although these excitation energies are expected to be close to those of the core nuclei in the weak coupling limit, the excitation energy of the #3 peak, 6.9 MeV, is considerably greater than the corresponding core excitation energy of 4.8 MeV. It was discussed that the deviation of the excitation energies carries information on the AN spin-spin interaction.[9] Although it is qualitatively so, the data offered serious problem to theoretical interpretation and there are arguments that the peak is not necessarily due only to a simple llC(4.8 MeV)xsA configuration but could have some contribution from intershell configuration mixing.[l9,20] Further investigation on new excitation mode of the A hypernucleus is required.
Excitation energy [MeV]
Preliminary
Figure 2. High statistics excitation spectrum of i2C obtained by 12C(n+,K+)i2Creaction with the SKS spectrometer system. The spectrum has statistics 5 times better than the previous E140a experiment. The high-quality angular distributions were for the first time obtained by fitting the excitation spectra in 2 degree bins and are also shown in Fig. 2. In the figure, the DWIA calculation by Itonaga is compared with the data by solid lines.[21] The overall agreement was observed and it will imply that the present DWIA calculation are reliable in predicting the absolute cross sections by the ( T + , K+) reaction. The i2C spectrum demonstrates a good example of ( T + , K+) reaction spectroscopy and the excitation function of the other A hypernuclei listed in Table 1 were also measured. A part of the result will be found in ref.1151 Here, it is only emphasized that comparison of spectra by the (K-, T-) and (n+, K+) reactions provides valuable information on the A hypernuclear level structure, since they populate different states of spin partners etc.
4. HYPERNUCLEAR 7 RAY SPECTROSCOPY
As described in the introduction, the y ray spectroscopy offers the best energy resolution for those y decaying hypernuclear states. However, it has been thought difficult to perform y ray spectroscopy for A hypernuclei, since the hypernuclear yield is very much limited due to small cross section and/or low intensity of kaon and pion beams. The y ray spectroscopy with Ge detectors has been thought even more difficult than those with NaI scintillators since the photo-peak efficiency is usually low compared to NaI detectors and also it is not easy to operate Ge detectors under the environment that the background associated with secondary beams is severe. Although hypernuclear y rays were detected by P b glasses and KaI scintillators, no experiments were successful in detecting hypernuclear y rays by Ge detectors. The KEK E419 experiment(Spokesperson:H.Tamura) was carried out by installing the Hyperball system which consists of 14 Ge detectors(60 %) in the target region of the SKS spectrometer system. The overall photopeak detection efficiency of about 2.5 % for 1 MeV y rays was achieved with the solid angle of 15 %. The BGO counters surrounding each Ge detector were used to suppress background coming from TO'S and Compton y rays. Two hypernuclear y rays from LLi, which were excited through 7Li(7r+,K+) reaction, were observed as shown in Fig.3. One is the M1 transition between 3/2+ -+1/2+ and
, reaction.[l7] A Figure 3. y ray spectrum measured by Hyperball in the 7 L i ( ~ +K+)iLi relevant decay scheme of the 6Li and LLi together with calculated cross sections is shown on the right.
*
the energy was determined 691.7 f O.G(stat) l.O(sys) keV. The y energy were directly compared with the recent calculation based on a cluster model and the magnitude of spin-spin interaction was discussed.[l7] The 2050 keV peak which corresponds to the 512' + 1/2+ E2 transition has a broader peak shape due to the Doppler effect of the recoiling hypernclei. After line shape fitting taking into account the slow down process f 0.7(sys) of LLi, the lifetime of the 5/2+ state was derived to be T = 5.8 !:::(stat) ps.[18] It was found that the charge distribution of LLi shrinks by 20 % compared with
that of 6Li due to the presence of a A hyperon based on the B(E2) value of the transition which is strongly dependent on the charge distribution of the hypernucleus. The shrinkage is consistent with theoretical predictions that claimed a glue like role of A hyperon by Motoba[22] and also by a recent cluster calculation.[23] Hypernuclear y-ray spectroscopy by the Hyperball was carried out recently also for the 'Be(K-, ~ - ) i B ereaction, moving the hyperball system to the D line of BNL AGS. It is expected to provide precision information on the spin-orbit interaction through the splitting of the 1s partner levels, 3/2+ and 5/2+, which were not resolved in the previous BNL experiment using NaI detectors.[24] It is emphasized that the hypernuclear y ray spectroscopy with Ge detectors became possible with the large solid-angle high efficiency Ge arrays that can handle large background. It is planned to extend the measurement to other light hypernuclei with an intension to derive magnitudes of spin dependent interaction qualitatively. An effort t o double the efficiency of the hyperball is under way at the sametime. The y ray spectroscopy with use of NaI scintillator array was very recently applied in order to observe the y transitions from A p 3 / ~and pl/2 states to the ground state of i3C. The two y rays about 10 MeV were clearly detected and the energy difference will give the magnitude of spin-orbit interaction. This experiment is the second example that demonstrates the powerfulness of the ray spectroscopy.[25] A/
5. HYPERNUCLEAR STUDY WITH ELECTRON BEAMS
I Jlab Hall C HNSS I Side View SOS Spectrometer Resolution 5 x Solld angle 6 msr Splitter
Electron Beam
D
(1.645 MeVIc)
Target ( SSD + Hodoscope )
f ENGE S~lit- ole S~ectrometer ~csolut~2 o n lo4 -
x
1 0-
m
Figure 4. Hypernuclear Spectrometer system in Jlab Hall C. The experiment is scheduled in the spring of 2000. Strangeness production in the elementary processes with electromagnetic probes has
been studied for various reaction channels until now. However, there are few investigations on the strangeness production with the nuclear target. Only recently, quasifree A production on a carbon target by the tagged photon beam at Tanashi-ES was reported as presented in this symposium,[26] though spectroscpic study for the hypernuclear bound states by the tagged photon beam would be difficult since the beam intensity is low. On the otherhand. an electron beam in the GeV region is a unique tool for the hypernuclear study. In contrast to reactions with r , K meson beams which have been mostly used for the hypernuclear experiments, the (e,e'Kf) reaction has large spin-flip amplitude even at 0 degrees. Thus, it excites both spin flip and spin non-flip hypernuclear states simultaneoulsly. In addition, contrary to the ( T + ,K+) and (K-, r - ) reactions, the reaction converts a proton to a A hyperon and produces A hypernuclei which are not accessible by such reactions. Isospin multiplet states will be investigated by comparison of these reaction spectra. Particularly in the light mass region, the (e,e'Kf) reaction tends t o populate neutron rich A hypernuclei. For example, the (e,e'Kf) reaction on the 7Li target make it possible to study a ;He hypernucleus, which would be the neutron ha110 A hypernucleus. From the experimental point of view, the (e,e'Kf) reaction has a potential power that will realize hypernuclear reaction spectroscopy with sub-MeV energy resolution. Even a few 100 keV resolution, which is comparable to that of NaI y detectors, can be achieved when an appropriate spectrometer system is available. Such high resolution becomes possible because the primary electron beam has excellent beam emittance. A small beam spot size of the order of 0.1 mm at the target and high beam current are easily obtained. Thus, we can employ small and thin targets, even enriched isotopes. There are efforts for the A hypernuclear spectroscopy at Hall A and Hall C of Jefferson Laboratory. In Hall C where the experiment is scheduled to run in the spring of 2000, the experimental setup was designed to carry out in the "0 degree tagging" configuration. [27] Since the virtual photon flux is very much forward peaked( nearly at zero degrees) and kaon angular distributions are also forward peaked, hypernuclear production will be most efficiently detected by measuring scattered electrons associated with virtual photons and kaons both at 0 degrees. In the peresent setup in Hall C, a splitter magnet is installed immediately downstream of the target as schematically shown in Fig.4. Scattered electrons and positive kaons are bent to the opposite directions and guided to the ENGE spectrometer that tags electrons from 230-340 MeV/c and to the SOS spectrometer(6 msr) that detects kaons around 1 GeV/c when the beam momentum is 1.645 GeV/c. Although it is known that the (e,elK+) reaction is invaluable to the hypernuclear study, it has become feasible only recently at Jefferson Laboratory where high intensity CW electron beams are accelerated. In addition to a good resolution kaon spectrometer, it is also required to have a scattered electron spectrometer that has good energy resolution and can detect high rate electrons which are associated with bremstrahlung from the target. The Hall C setup with a silicon strip detector array that covers the 70 cm focal plane of the electron spectrometer works well under such envronment. Experiment for the p-shell A hypernuclear spectroscopy(E89-09: L. Tang and E. Hungerford) is in progress with targets of %i. 7Li, 9Be. 1°B. l l B and I2C. It is also planned to extend the investigation by the (e,e'K+) reaction to a heavier target(E97-08:O. Hashimoto and L. Tang) and explore the possibility to carry out high-resolution hypernuclear spectroscopy of heavier A hypernuclei with electron beams.
6. SUMMARY AND FUTURE PROSPECTS
Through the recent progress, we have learned that the A hypernuclear spectroscopy has great significance for the investigation of the A hypernuclear structure and AN interaction. Now, quantative data, which were beyond the dream some years ago, can be obtained taking advantages of new experimental opportunities. In particular, the ( T + , K+) reaction spectroscopy with the SKS spectrometer has established the value of the A hypernuclear spectroscopy with its good resolution and high detection efficiency. Hypernuclear y ray spectroscopy with the Ge detector array proved its unparalleled energy resolution, giving crucial information on the spin-dependent AN interaction.[l7] A new spectroscopic study of A hypernuclei with the (e,elK+)recation is in progress a t Jefferson Laboratory, which is expected t o reveal structure of neutron rich A hypernuclei.
Table 2 Comparison of A hyperon production reactions A2 = 0 A 2 = -1 comment neutron to A proton to A (n+,K + ) in-flight (K-, T - ) stopped (K-, T-) (e, e'KO) (P,P'KO) (P, KO)
(T-
, KO)
stretched,high spin
in-flight ( K - , TO) substitutional at low momentum stopped (K-, TO) large yield, via atomic states (e, elK+) (p,plK+) (P, K + )
spin flip,unnatural parity, virtual (y,K) virtual (T,K) very large momentum transfer
In the near future, A hypenuclear spectroscopy will be explored with wide variety of reactions as listed in Table 2. Hypernuclear states and hypernuclei t o be studied will be expanded greatly. Among them, the (e,e'K+) reaction will play a significant role and will reveal new aspects of A hypernuclear structure with its sub-MeV or even 2-300 keV(FWHM) energy resolution. We expect the y ray spectroscopic studies will be further extended with much higher efficiency of a new Ge detector array and will provide key information on the spin dependent parts of the AN interaction. The reaction spectroscopy, which allows us access t o the hypernuclear states above the nucleon emission threshold requires higher energy resolution keeping high detection efficiency, while the yray spectroscopy needs much larger detection efficiency. Although the present paper deals mostly with spectroscopy of A hypernuclei, investigation of C hypernuclei which had important progress in the last few years will be further explpored, for example, by the (T-,K+) reaction. Beyond that, it is eagerly waited that hypernuclear spectroscopy in the S=-2 regime become possible when high intensity kaon beams are available in planned acccelerator facilities such as the 50 GeV PS of JHF. The author is thankful for the useful discussions with the collabor~torsof KEK-PS E140a, E336, E369, E419 and Jlab E89-09, E97-08, particularly with Professors H.
Tamura, L. Tang and E. Hungerford. Discussions with Prof. T . Motoba, K. Itonaga, D.J. Millener and M.Sotona on the theoretical aspects are also greatly appreciated. REFERENCES
1. T . Yamazaki, Proceedings of the KEK international workshop on Nuclear Physics in GeV Region, KEK, 1984, KEK Report 84-20, p.3. 2. B. Povh, Ann. Rev. Nucl. Part. Sci. 28 1 (1978). R. Chrien and B. Gibson, Ann. Rev. Nucl. Part. Sci. 39 113 (1989). 3. H. Band6. T . Motoba and Y. Yamamoto, Phys. Rev. C31 265 (1985). 4. A. Likar, M. Rosina and B. Povh, Z. Phys. A324 35 (1986). 5. Y. Yamamoto et. al., Prog. Theor. Phys. Supplement No. 118 361 (1994). 6. D.J. Millener, A. Gal, C.B. Dover and R.H. Dalitz, Phys. Rev. 31 499 (1985). 7. C. Milner et. al., Phys. Rev. Lett. 54 1237 (1985). P. H. Pile et al., Phys. Rev. Lett. 66 2585 (1991). 8. M. Akei et al., Nucl. Phys. A534 478 (1991). 9. T . Hasegawa, et. al., Phys. Rev. Lett. 74 (1995) 224. 10. T . Hasegawa, et. al., Phys. Rev. C53 1210 (1996). 11. 0.Hashimoto, et. al. I1 Nuovo Cimento 102 679 (1989). 12. T . Fukuda, et. al., Nucl. Instr. Meth. A361 485 (1995). 13. S. Ajimura, et. al., Phys. Rev. Lett. 80 3471 (1998). 14. H. Bhang, et. al., Phys. Rev. Lett. 81 4321 (1998). H. Park, et. al., to be published in Phys. Rev.(2000) 15. 0. Hashimoto, et. al., Nucl. Phys. A639 93c (1998). 16. T. Nagae, Proceedings of the workshop "Strangeness uclear Physics(Seoul99)" edited by I.T. Cheon, S.W. Hong and T . Motoba(Wor1d Scientific, Sigapore, 2000) to be published. 17. H. Tamura, Proceedings of the workshop "Strangeness uclear Physics(Seoul99)" edited by I.T. Cheon, S.W. Hong and T. Motoba(Wor1d Scientific, Sigapore, 2000) to be published ; H. Tamura et. al., submitted to Pys. Rev. Lett. 2000 ; 18. H. Tanida, Doctor thesis, University of Tokyo, 2000. 19. A. Gal, Proceedings of the 23rd INS symposium, p.23, 1995, University Academy Press, Eds. S. Sugimoto and 0. Hashimoto. 20. T . Motoba, Nucl. Phys. A639 135c (1998). 21. K. Itonaga, et. al., Prog. Theor. Phys. 84 291 (1990). and Private communication 22. T . Motoba, H. Bando and H. Takaki, Z. Phys. 8 189 (1983) 23. E. Hiyama, et. al., Nucl. Phys. A639 173c (1998). 24. M. May et. al., Phys.Rev.Lett. 51 2085 (1983). 25. T. Kishimoto, Private communication 1999. 26. K. Maeda, this proceedings 27. E. Hungerford, Prog. Theor. Phys. Suppl. 117 135 (1994).
Hadron and Nuclear Physics with ElectromagneticProbes K. Maruyama and H. Okuno (Editors) 2000 Elsevier Science B.V. All rights reserved.
'
Spin-orbit splitting of i3C H. Kohria, S. Ajimuraa, R. E. Chrienb, P. M. Eugenioc, G. Franklinc, J. Franzd, T . Fukudae, L. Ganf, H. Hayakawaa, P. Khaustovc, T . Kishimotoa, K. Matsuokaa, M. Mayb, S. Minamia, Y. Miyakea, T . Moria, K. Morikuboa, J. Nakanog, H. Noumie, ~ ,Sajia, A. Sakaguchia, H. Outae, K. Paschkec, P. Pileb, B. Quinnc, A. ~ u s e k E. R. Sawaftah, Y. Shimizua, M. Sumihamaa, R. Sutterb, T. Tamagawag, H. Tamurai, K. Tanidag , L. Tangf and L. Yuanf "Osaka University, Machikaneyama 1-1, Toyonaka 560-0043, Japan bBrookhaven National Laboratory, Upton, New York 11973, USA "Carnegie Mellon University, Pittsburgh, PA 15213, USA dUniversity of Freiburg, Hermann-Herder-Str.3, D79104 Freiburg, Germany "High Energy Accelerator Research Organization, Tsukuba, Ibaragi, 305-0801, Japan 'Hampton University, Hampton, VA 23668, USA guniversity of Tokyo, Tokyo 113-0033, Japan hNorth Carolina A T State University, Greensboro, NC 27411, USA 'Tohoku University, Sendai 980-8578, Japan The y-rays from the 112- and 312- doublet states at --I1 MeV to the ground state in i3C were successfully measured by using 72 NaI detectors to obtain the spin-orbit splitting energy with high precision. The ( K - , T - ) reaction on an enriched 13C target was used a t the Brookhaven Alternate Gradient Synchrotron. The energy resolution for the 11 MeV y-rays is 340 keV FWHM which is much better than those obtained by using magnetic 2 FWHM). The obtained small splitting energy of the 112- and spectrometers ( ~ MeV 312- doublet states in i3C is about 150 keV (preliminary).
1. I N T R O D U C T I O N
Since AN scattering data are extremely limited due to experimental difficulties, the (r+, K f ) reaction data have clarified the gross structure of the A-nucleus interaction [1][2]. The phenomenological analyses with the use of Woods-saxon potentials indicated that the 8 of that dominant central part of the A-nucleus interaction was roughly 213 ( ~ 2 MeV) of the nucleon. While, the spin dependent parts were found to be surprisingly small. For example, the spin-orbit interaction was roughly one order of magnitude smaller than that of the nucleon [3]. It can be naively explained that the strange quark contributes little to the nuclear force. The small A-nucleus spin-orbit interaction has been one of the most interesting subjects in the hypernuclear physics for more than 20 years. Experimentally, the first indication of the small A-nucleus spin-orbit interaction was given by the 160(K-, r-)i60experiment performed at CERN [3]. The splitting of observed peaks with the configurations of [(p1/2),'(p1/2)A]o+ and [(p3/2)i1(p3/2)~]0+ was about 6 MeV. The splitting is very close to the splitting of the neutron pll2 and P3/2 hole states in 150(6.176 MeV). The splitting of pstate A is estimated to be less than 0.3 MeV. In the 13C(K-, r-)i3C experiment at BNL, the A splitting energy in i3C was obtained as 0.36 f 0.3 MeV [4]. The gBe(K-, r- y);Be experiment at BNL also indicated the small A spin-orbit splitting [5]. Observed single y-ray peak suggests that the excited doublet states of [gBe(2+)@(s1/2)A]3/2+,512+ are almost degenerate. The splitting energy has to be less than 0.1 MeV. On the other hand, a large splitting was indicated by the "Y(T+, K + ) i 9 Yexperiment at KEK [6]. The splitting energy of a few MeV for f-state A was observed. They performed a new experiment using the same reaction in order to confirm the splitting. The re-analysis of the emulsion experimental data collected by the European K - collaboration also indicated a relatively large splitting [7]. The observed splitting of two states with the configurations of [(p1/2),1(p1/2)A] and [(pl/2),'(p3/2)A] in i60was 1.56 f 0.12 MeV. There are also discrepancies between theoretical predictions on the spin-orbit splitting. 0.96 MeV for i3C [8]. One boson exchange (OBE) models predict splittings of 0.75 The calculations are performed in the framework of microscopic 3 a A model by using Nijmegen model D, F and NSC97f. On the other hand, quark models predict almost zero splittings for the A-nucleus spin-orbit interaction [9][10]. A new experiment with high resolution and high statistics is necessary to conclude these discrepancies described above. We performed the experiment with an electromagnetic probe in 1998. Our experiment has great advantages over previous experiments.
-
+
2. E X P E R I M E N T
The 13C(K-, r- 7)i3C experiment was performed at BNL-AGS-DGLINE to measure the spin-orbit splitting of i3C with high precision. The K - beam momentum was 0.93 GeV/c, and scattered n- particles at 0 to 16 degrees were detected by 48D48 spectrometer. The (T-, p - , e-)/K- ratio of the beam line was about 0.3. Typical beam intensity was 5 x lo4 K-slspill(4.5 s). The 13C benzene target was an active target of 6.0 cm wide, 1.5 cm high and 12.0 cm thick. The active target is a useful device to detect the weak decay of i3C and to suppress the background due to the in-flight decay of K - . In the (K-, T-) experiment, the in-flight decay of K - is a main background. Especially, we measure
y-rays, therefore the K - -t .ir-.iro decay producing two y-rays must be suppressed. In the off-line analysis, the active target played an important role in identifying the 13C(K-, T-)i3C reaction and in reducing background. 72 NaI detectors of 6.4 cm x 6.4 cm x 30.5 cm were located at 10.5 cm from the target center to measure the y-rays from the 112- and 312- doublet states at --I1 MeV to the ground state in i3C. Configurations of By virtue of the 0+ spin of the two states are mainly [12C(O+)@(p1/2,p3/2)A]l/2-,3j2-. the 12C core, a purer spin-orbit splitting can be observed in i3C. The energy calibration of the NaI detectors in the high energy region is particularly important. We used the 8.999 MeV y-rays from the 581%(n.y) reaction because the energy is close to 11 MeV. The segmentation of the NaI detector is also important for the energy correction of the Doppler shift, and it is necessary to endure high counting rate near the target. Typical energy resolution of the NaI detector is expected to be 340 keV FWHM for the detection of the 11 MeV y-ray. It is roughly one order of magnitude better than those obtained by using the magnetic spectrometers (-2 MeV FWHM). The better energy resolution enables us to measure energies of the doublet states with precision of a few tens keV. According to theoretical calculation on the cross-sections by T . Motoba et al., the 112- state is dominantly excited at the forward scattering angles. while the 312- state is dominantly excited a t the larger angles of -13"[11]. The angular distribution of the (K-, T-) reaction enables us to separate the doublet states, which is one of the most important advantages. Data were collected with 2.1 x lo9 K-s on the target. and the data collected stably with 1.4 x lo9 K-s were analyzed. 3. PRELIMINALY RESULT 3.1. Quasi-free region In the quasi-free region of the 13C(K-, T-) reaction, two peaks are dominantly observed. One peak is y-rays from the 2+ state at 4.439 MeV to the ground state in 12C. Fig.l(a) shows an energy spectrum of y-rays in the quasi-free region, where the T- scattering angles of 0 t o 7 degrees are selected. A fitting to the histogram in the region of 3.0 to 5.6 MeV is performed with the function of two gaussians added to the linear function. The higher gaussian is for the full energy peak, and the lower gaussian is for the single escape peak. The measured energy of the full energy peak is 4.444 rrt 0.008 MeV, and the energy resolution is 220 4~ 18 keV FWHM. The energy is almost the same as 4.439 MeV within the statistical error. Assuming the relation, AE cx for the energy dependence, an estimated energy resolution for the 11 MeV y-ray is 340 keV FWHM. The energy resolution is the same as what we planned to achieve before the experiment. Observed peak structures at 2.2 MeV and 6.2 MeV are thought to correspond to y-rays from the 7/2+ state at 2.164 MeV to the ground state in 27Si and y-rays from the 312state at 6.176 MeV to the ground state in 150. The 28Si and 160nuclei are the most dominant compositions of the target cell made of quartz with total thickness of 8 mm. The other peak is y-rays from the 1+ state ( T = l ) at 15.110 MeV to the ground state in 12C. Fig.l(b) shows an energy spectrum of y-rays in the quasi-free region, where the Tscattering angles of 0 to 16 degrees are selected. A clear peak at 15 MeV is dominantly observed. The amount of the background around 15 MeV is much less than that around 4.4 MeV.
&
2
4
6 8 y-ray Energy (MeV)
15 20 y-ray Energy (MeV)
Figure 1. Energy spectra of y-rays in the quasi-free region at the scattering angles of 0 to 7 degrees (a) and at the scattering angles of 0 to 16 degrees (b).
3.2. Bound region An energy spectrum of y-rays in the A bound region of i 3 C is shown in Fig.2, where the .ir- scattering angles of 0 to 9 degrees (upper) and 9 to 16 degrees (lower) are selected. Theoretically, estimated population ratios between the 112- and 312- states are about 8:2 in the upper spectrum and 2:8 in the lower spectrum. The observed peaks in the both spectra are located closely to each other. Strengths of the single escape peaks are about 50 % of the full energy peaks, which is consistent with the results of the Monte Carlo simulation. Fittings are performed with the function of two gaussians added to the exponential. The higher gaussian is for the full energy peak, and the lower gaussian is for the single escape peak. Position of the lower gaussian is fixed at 0.511 MeV lower than that of the higher gaussian. Widths of the gaussians are also fixed to the estimated width of 340 keV FWHM. Statistical errors are about 20 keV for the means of the higher gaussians. The errors are almost the same as that we expected. The energy difference of the two peaks is -90 keV (preliminary). The splitting of the 112- and 312- doublet states is estimated to be -150 keV by considering the mixing of the two states. In the present preliminary result, systematic errors have not been estimated, yet. A systematic error in the energy calibration of the NaI detector is thought to be small in obtaining the splitting energy because same energy calibration is used for all the scattering angles. Therefore, systematic errors in the fittings and in the energy correction of the Doppler shift are expected to mainly contribute to the splitting. For the estimation of the systematic error in the fittings, various functions such as a function more similar to the energy spectrum obtained with the simulation should be applied. The simulation will also help to estimate the systematic error in the energy correction of the Doppler shift.
O
t ' ~ ' ' ' ' b ' ' "' ~' I ~" " 1 1 2 ~ " ' 1 1 3 " ~ ' ~ 1 4 " " 1 5 " ' ~ y-ray Energy (MeV)
Figure 2. Energy spectra of y-rays in the A bound region. The energy correction of the Doppler shift is performed. The scattering angles of 0" to 9" in the upper spectrum and 9" to 16" in the lower spectrum are selected.
REFERENCES P. H. Pile et al., Phys. Rev. Lett. 66 (1991) 2585. T . Hasegawa et al., Phys. Rev. Lett. 74 (1995) 224. W. Bruckner et al., Phys. Lett. 79B (1978) 157. M.May et al., Phys. Rev. Lett. 47 (1981) 1106. M. May et al., Phys. Rev. Lett. 51 (1983) 2085. T . Nagae et al., Int. INS symp. on Nuclear and Particle Physics with meson beams in the 1 GeV/c region, eds. S. Sugimoto and 0. Hashimoto (Universal Academic Press, Tokyo, 1995) 175. 7. R. H. Dalitz et al., Nucl. Phys. A 625 (1997) 71. 8. E. Hiyama et al., to be published. 9. 0. Morimatsu et al., Nucl. Phys. A420 (1984) 573. 10. H. J. Pirner et al., Phys. Lett. 114B (1982) 308. 11. T. Motoba et al, private communication.
1. 2. 3. 4. 5. 6.
ThisPageIntolyLfBk
ThisPageIntolyLfBk This Page Intentionally Left Blank
V. N-N CORRELATIONS AND FEW-BODY PHYSICS
ThisPageIntolyLfBk
ThisPageIntolyLfBk This Page Intentionally Left Blank
Hadron and Nuclear Physics with ElectromagneticProbes K. Maruyama and H. Okuno (Editors) 2000 Elsevier Science B.V. All rights reserved.
"
Photodisintegration reactions of 3 ~ and e 4 ~ ate TAGX
"RI Beam Science, RIKEN 1-2 Hirosawa, Wako, Saitama, JAPAN
1. INTRODUCTION
The measurement of the photodisintegration cross sections for few-nucleon systems, such as 3He and 4He, has been one of the main experimental programs of the TAGX project since the end of 1980's. This program aimed at investigating reaction mechanisms of the nuclear photodisintegration in the intermediate photon energy ( E , 2 100 MeV). In a series of this TAGX program, the measurements of the photodisintegration reaction cross section for 3He and 4He covering a wide kinematical region have been attempted in a kinematically complete way, which enabled one to explore the reaction mechanisms in detail. The dominant reaction mechanism has been revealed to be photon absorption by two nucleons (2N) in a nucleus. In the TAGX program, different photon absorption modes contributing the photodisintegration processes have been separately identified, and we have succeeded to extract the 2N cross section over a wide E, range for the first time. By comparing the 2 N cross sections with microscopic theories, one can study the twonucleon system in a nucleus. In the case of 3He photodisintegration, the cross sections for photon absorption by a np- and pp-pair (a2N(,,),gaN(,,)) have been separately determined, and the analyzing power for 2N(pp) has been also measured. For 4He, the cross section for 2N(pn) has been extracted, and compared with that of 3He. In this paper, we present most of the experimental results obtained by the 3He and 4He photodisintegration measurements using the TAGX. One may call the 2 N process 'photodisintegration' of a two-nucleon system in nucleus. As the wave function of the deuteron and the nucleon-nucleon force have been extensively studied from the deuteron photodisintegration [I], we may study the wave function of a nucleon pair and the nucleon-nucleon forces in nuclear medium from 'photodisintegration' of the two-nucleon system in nuclei. A building block of a nucleus is considered to be successive nucleon-nucleon collisions inside the nucleus, but our understanding on the two-nucleon collisions in nuclear medium is surprisingly poor. The best model to describe nuclei is the mean-field theory based on the independent-single-particle model, in which a nucleon is assumed to move independently in a nuclear-mean-field potential. It accounts for various properties of nuclei and nuclear reaction well, but this model can not describe how two nucleons scatter each other in a nucleus. The measurements of the 'photodisintegration' of a nucleon pair inside a
nucleus is one of the best way to study this two-nucleon aspect of the nuclear physics. The deuteron is a bound two-nucleon system, whose quantum state is the spin-triplet state, "1. A spin-singlet nucleon pair, 'So, can be also studied when one looks at a nucleon pair inside a nucleus. For example, a proton-proton pair can be assumed to be 'SOsince the isospin is always one. The nucleon-nucleon potential for the 'So-pair becomes repulsive when the distance of two nucleons becomes less than 0.5 fm. It is well known that this repulsive force plays an essential role for many nuclear properties such as the saturation effects, although the origin has not been yet fully understood. One can investigate this using the photon absorption process by two protons in nuclei. The iso-scaler nature of the nucleon-nucleon force ensures the similarity of the wave function of the np- and pp-pairs in a nucleus. The interaction with the electromagnetic field, however, is quite different due to the differences of the contributing electromagnetic currents in the 2N(pn) and 2N(pp) process. In the case of 2N(pn), the meson-exchange currents (MEC) can contribute in addition to normal one-body currents; the convection and spin currents. In the case of 2N(pp), however, no charged meson can be exchanged. Therefore only the one-body current plays a role. This difference of the electromagnetic currents gives significant difference in the absolute magnitude and E,-dependence of the cross section for 2N(pn) and 2N(pp). 2. EXPERIMENTAL DETAILS 2.1. The TAGX spectrometer The TAGX spectrometer was placed in the tagged photon beam line at the 1.3 GeV electron synchrotron of the Institute for Nuclear Study (INS), University of Tokyo. (The TAGX spectrometer has been already moved from INS to LNS, Sendai in the fall of 1999.) TAGX is a large-acceptance magnetic-spectrometer having about r s r for charged particle and 0.8 sr for neutron. Detailed description can be found elsewhere [2]. The trajectories of charge particles photo-produced from the 3He or 4He targets, were traced by straw drift chambers (SDC) and a set of central drift chambers (CDC) placed in the dipole magnetic field. The velocity was measured by the time-of-flight (TOF) method using two set of hodoscopes, Inner Hodoscope (IH) and Outer Hodoscope (OH). The flight time between IH and OH have been measured. Particle identification for charge particles was done using the momentum and the velocity. The energy of the neutron was determined by the time-of-flight method. Since we are interested in photon absorption by two nucleons, two energetic nucleons were detected in coincidence, (y, N N ) . The coincidence signal strobed the photon tagger to determine the photon energy. Due to finite momentum threshold of the TAGX (300 MeV/c for proton and 100 MeV/c for neutron), two different triggers were employed for the data acquisition to make the TAGX spectrometer have enough acceptance for different photon absorption processes such as by two- and three-nucleons. The first one was 'np'-trigger for detecting a np pair, and another was 'ppl-trigger for detecting two protons. 2.2. Production of polarized tagged photons and a polarimeter
In addition to the measurement the photodisintegration cross section for 3He and *He, the analyzing power of the 3He(y',pp)n reaction has been measured to study the 2N(pp)
Diamond Crystal Radiator Tagging Hodoscope A
Tagging Hodoscope B
Electron Beam Profile Monitor
n l u I I I I I IIUIII
Converter
Drift Chamber
Trigger Counter
Photon Beam Profile Monitor
Channeling Radiation Detector
Ill
Veto Counter
TAGX Spectrometer Inner Hodoscope
Helium Target Outer Hodoscope
&
U
Photon Beam Profile Monitor Lead Glass
Figure 1. Experimental setup for the 3He(y',pp)n experiment
process. To do this, the photon tagging system has been modified and a polarimeter has been installed in the tagged photon beam line. We describe this new facility briefly. Figure 1 shows the experimental setup for the analyzing power measurement of the 3He(y',pp)n reaction. Polarized tagged photons were produced by the coherent bremsstrahlung of a 1.16-GeV electron beam in a diamond crystal of 1.1-mm thickness installed at the radiator position. The crystal was mounted on a three-axis goniometer and was placed at the radiator position of the tagger. The axis was set so that the coherent spike appears at E, = 290 MeV as shown in Fig. 2. 40
80 Coherent Bremsstrahlung
-
0
200 400 600 800 1000 1200 Ey (MeV)
Figure 2. Intensity and polarization of the coherent bremsstrahlung for E, = 1.16 GeV. Open circles show the measured photon energy spectrum. The results of a theoretical calculation for the intensity and the polarization are shown together by the solid- and dashed-line. Recoiled electrons were detected by the photon tagging system, which covered the photon energy ranges of 740 - 1060 (Tagging Hodoscope A) and 200 - 600 MeV (Tagging Hodoscope B) with AE, = f5 and flOMeV resolution, respectively. The photon tagging detectors for E, = 200 - 600 MeV were newly equipped for the polarized photon project. In the Fig. 2, open circles show the photon energy spectrum measured by the tagging counters plotted with the result of a theoretical calculation which takes the experimental conditions into account. The theory reproduces the measured energy spectrum well, and the maximum polarization is expected to be about 35% at the coherent spike position. The photon polarization has been determined by two ways. One is by a pair polarimeter, which detects an electron-positron pair produced in a converter. The correlation between the momentum vectors of a pair has been shown to be sensitive to the photon polarization [3]. The pair polarimeter consisted of an A1 converter, a pair magnet, drift chambers and plastic scintillators as shown in Fig. 1. Another was to use the quasi-free negative pion photoproduction process, y' +' n' -+ p n - , whose analyzing power has been known to be large [4]. During the 3He(y',pp)n measurement, two charged particles were detected in coincidence, which enabled one to accumulate 3He(y',pn-)pp events simultaneously. The photon polarization was also
+
0
Pair polarimeter
4
Figure 3. Photon polarization determined by the pair polarimeter (open circles) and by the 3He(y,p ~ - ) p preaction (solid circles).
determined by using the asymmetry of the 3He(y',pn-)pp events. The photon polarization determined by the pair polarimeter and the 3He(y',pn-)pp reaction are plotted together in Fig. 3, and the theoretical calculation reproduces the data well. 3. DISCUSSIONS 3.1. The 2N process 3.1.1. 3 H e In the 3He photodisintegration process, three different photon absorption mechanisms may be considered. They are photon absorption by one nucleon ( I N ) , by two nucleons (2N) and by three nucleons (3N). In the case of I N , the other two nucleons in 3He are spectators of the reaction, one nucleon remains as a spectator in 2N, and no spectator exists in 3N. Considering the facts that the energetic two nucleons were detected in the 3He(y,N N ) measurements and a spectator nucleon was hardly detected due to the TAGX detection threshold, only the 2N and 3 N processes may contribute to the 3He(y,N N ) events. Since one nucleon is a spectator in 2N, whereas not in 3N, they can be identified by looking at the undetected nucleon distribution of the 3He(y,N N ) N events. Figure 4 shows typical missing momentum distributions of the 3He(y,pn)p and 3He(y,pp)n events for E, = 280 MeV [6,7]. The results of model calculations for (a) 2N(pn), (b) 2N(pp) and (c) 3 N taking the detector responses of TAGX into account by the Monte Carlo method are shown together. In the 2N(pp)(2N(pn)) model, a pair of protons (a proton-neutron pair) in 3He absorbs a photon, leaving neutron (proton) as a spectator, whose momentum distribution is sampled from the single-nucleon momentum distribution determined by the 3He(e,e'p)d measurement [5]. In the case of the 3 N model, the 3He nucleus is assumed to be disintegrated into three nucleons whose momenta distribute according to the three-body phase space.
Figure 4. The undetected-nucleon momentum distributions of the 3He(y,pn)p and (y,pp)n events for E, = 280 MeV. The dashed- and dotted-lines show the result of model calculations for 2N (a,b) and 3 N (c). The solid lines are the sum of 2N and 3N.
The sum of 2N(pn)(2N(pp))and 3N, whose strengths are determined by the fitting to the data, reproduces the missing momentum distribution of the 3He(y,pn)p(3He(y,pp)n) events fairly well. 3.1.2. 4 H e There are two different final states in the 4He(y,N N ) events, they are three-body and four-body breakup channels. a three-body breakup : y -t4 a four-body breakup : y +4
He -, pnd
He
-+
pnpn
In the case of 4He(y,pn)X,both three-body and four-body final states contribute, i.e. X = d, np, whereas the 4He(y,pp) always feeds the four-body final state. The 4He(y,pn)d events, where the deuteron remains as a spectator in the final state, could be considered as the 2N(pn) process. Since the energy resolution of the TAGX system was not good enough to resolve the three- and four-body states, the 4He(y,pn)d yield was determined from the 4He(y,pn) events by subtracting the four-body contribution estimated from the 4He(y,pp)nn yield by correcting for the detection efficiency [8]. Figure 5 shows the missing mass (Mx) spectrum of the 4He(y,pn) and 4 ~ e ( y , p p ) events. The results of the calculated Mx distribution of the three-body and four-body breakup channels are plotted together, which are corrected for the detection efficiency of the TAGX spectrometer. Figure 6 shows the missing momentum distribution of the extracted 4He(y,pn)d events, that corresponds to the undetected deuteron momentum distribution, Pd. During the 4He(y,pp) measurement, several 4He(y,pd) events were detected and they are identified
Figure 5. Missing mass distribution of the 4He(y,pn) events. Solid- and open-circles show , events. Dottedthe 4He(y,pn) events and the detection-efficiency-corrected 4 ~ e ( ypp) and dashed-lines show the results of model calculations for the three-body and four-body breakup channels.
Figure 6. Undetected-deuteron momentum distribution of the 4He(y,pn)d events. The Pd distribution determined from the 4He(y,pd)nevents are also plotted by open circles.
as the 4He(y,pd)nevents. One can consider that the spectator deuterons were detected in the three-body breakup channel, y +4 He -t pnd. Open circles in the figure show the deuteron momentum distribution of the detected 4He(y,pd)n events after correcting for the detection efficiency. The agreement in absolute magnitude is satisfactory, although the comparison with the extracted 4He(y,pn)d events is possible only higher Pd region. This fact justifies our procedure to extract the 4He(y,pn)devents from 4He(y,pn) events. 3.2. Total cross section The strength of the photon absorption process; 2N(pn), 2N(pp) for 3 ~ and e 2N(pn) for 4He, was extracted from the data as shown in Fig. 4 and Fig. 5, then the differential and total cross sections were separately determined over an entire A-resonance region [6-81. We will discuss the total cross section of 2N (pn) for 3He and 4He, and the total cross
section of 2N(pp) for 3He. 3.2.1. 2N(pn) Figure 7, 8 shows the total cross section of the 2N(pn) process for 3He and 4He. In the figures, the deuteron photodisintegration cross section ( ~ 7 , ~is) plotted together, which is scaled in magnitude with factors of 1.24 and 6.0 for 3He and 4He, respectively.
0 100
200
300
400
500
EY (MeV) Figure 7. The 3He(y,pn)p,p cross section. The solid line is the deuteron photodisintegration cross section multiplied by 1.24.
0 100
200
300
400
500
EY (MeV) Figure 8. The 4He(y,pn)dsp cross section. The solid line is the deuteron photodisintegration cross section multiplied by 6.0.
The similarity of the E, dependence of the 2N(pn) cross section and that of the deuteron cross section over such a wide E, range is remarkable for both 3He and 4He. How can we understand the scaling factor of 1.24 and 6.0 for 3He and 4He 7 P. Wilhelm, J. Niskanen and H. Arenhovel have discussed the difference of the np wave-function in 3He from that of the deuteron [9]. The 2N(pn) cross section for 3He can be expressed in terms of the cross section for the pairs taking into account the spin-isospin weights spin-singlet ( I S o ) and spin-triplet (3S1) as,
If one neglects the ISo cross section whose contribution is hindered by the spin-isospin weight of 116, one obtains,
Then one can compare ass, with a,d. Using the relation of (2) and the fact that a z ~ ( p ) scales with that of a,d, one obtains,
Figure 9. The 2N(pn) cross section compared with the calculation. The deuteron photodisintegration cross section is also shown together [9].
It is concluded that the 'photodisintegration' cross section of the spin-triplet np-pair is about 2.5 times larger than u,d. This enhancement is due to the compact wave function of the np-system in 3He [9]. This was the first example to demonstrate quantitatively that the two-nucleon photodisintegration process is a good tool to investigate the two-nucleon wave function in nuclei. ~ ~They ) accounted The Mainz group has performed a microscopic calculation of a 2 ~ ( [9]. for both the 3S1- and lSo-pair, whose wave functions reproduce the two-nucleon correlation function of 3He obtained by Faddeev method. The results are shown in Fig. 9 with the 2N(pn) cross section for 3He. In the figure, the dashed- and dotted-line represent the calculated photodisintegration cross section of the 3Sl-and '5'0-pair, respectively. The 3S1 cross section shows a bump centered around 250 MeV coming from the A-excitation in the intermediate state. This is due to the M I transition, which is the leading transition multiple in this energy region. It is not, however, the case for the I S o cross section because of the lack of the Aexcitation contribution in the intermediate state. In addition to the smallness of the 'So cross section, the contribution of the ISo-pair is hindered further by the spin-isospin weights. The solid line is the sum of them; U ~ N ( ~ Although ~ ) . there is some disagreement in E, dependence at lower energy region, the theory reasonably reproduces both the E, dependence and the absolute magnitude of the measured cross section. 3.2.2. 2N(pp) The smallness of the 2N(pp) cross section may be due to the wave function of a twoproton system in 3He. Since one can reasonably assume that two protons in 3He are in
100
200
300
400
500
Ey(MeV) Figure 10. The 2N(pp) cross section compared with the microscopic calculation [lo,111.
spin-singlet state ('So), the dominant M 1 transition leading the 1+final state is forbidden due to the anti-symmetrization of the wave function . Since the pp-system has no electric dipole moment, the El transition is also suppressed, and only the higher-order spin-flip El transition is allowed. The cross section decreases monotonically as a function of E,, and this may be the result from the lack of the M 1 transition. The Mainz group has performed a microscopic calculation also for az~(,) [10,11]. A realistic two-nucleon wave function for the initial state of the pp-system is used and a final state wave function is obtained by a coupledchannel calculation in order to account for the A - N configuration in addition to a pure pp-state. In Figure 10, the results of the calculation are presented together with the TAGX data, which takes into account all possible transition multipoles upto L = 3, i.e. El, E2, E 3 and M2. (In the figures, the E 3 cross section is not presented since it is so small.) The theory reproduces the magnitude of the small 2N(pp) cross section. It is not, however, the case for the E, dependence especially at higher energy than 300 MeV, where the theory predicts that the El and M2 multipoles become important. In order to determine the leading transition multiple around 300 MeV, the analyzing power of the 2N(pp) processes has been proposed [ l l ] . We have measured the analyzing power of the 3He(q,pp)n reaction and the data analysis is underway. In this report some preliminary results will be presented. The analyzing power of 2N(pp) has been measured in the energy range of E, =240 320 MeV. Figure 11 shows the proton angular distribution for the 3He(y',pp)nevents for P m 5 150 MeV/c where the 2N(pp) process dominates, and P m > 150 MeV/c region where the 3 N process dominates. The open and solid circles in these figures denotes 3He(y',pp)n events by vertically and horizontally polarized photons. One may notice that the proton angular distributions by the vertically and horizontally polarized photons are almost identical at higher P m region, but not the case for lower P m region. Three nucleons are likely to be involved at the higher P m region, therefore no asymmetry is expected to be observed.
Figure 11. The proton angular distribution of the 3He(y',pp)n events for P, 5 150 and P, > 150 MeV/c. Solid and open circles are for the events due to vertically and horizontally polarized'photons.
Ey = 240 MeV
0
40
80 Q,(O)
120
Ey = 320 MeV
Ey = 280 MeV
160 0
40
SO Q,(O)
120
160 0
40
80
120
160
Q,(0)
Figure 12. Preliminary results of the analyzing power of 3He(y',pp)n) for P, I150 MeV/c. The curves are labeled by the ratio of the transition strength of El and E 2 : 0.2(solid),0.5(long-dashed) ,2(dashed) and oo(dotted).
Figure 12 shows preliminary results of the analyzing power, C, which is defined as,
for E, = 240, 280 and 320 MeV in the y +' pp' center-of-mass system. In these figures, the relative strength of the El and E2 excitations are varied and plotted together as lines. The E1/E2 ratio is determined for three different photon energies by fitting, are plotted in Fig. 13 with the prediction of the Mainz group [ l l ] . At this moment, data is still preliminary and no definite conclusion can be drawn. The systematical asymmetries are now carefully being estimated.
TAGX -P.Wlhelm
et at.
0 200
300
400
EY(MeV) Figure 13. The preliminary E1/E2 ratio compared with a theory 1111.
TAGX
(yd->pn) * 1.24
W Mainz
A Kolb et al.
a(yd->pn) * 6.0
Bennan et al.
0
100
200
300
400
EY (MeV) Figure 14. The 3He(y,pn)p cross section. The data for lower E, are plotted together. The solid line shows the deuteron cross section with a scaling factor of 1.24.
Lebdev
A Ukraina
-
Torino
0
100
200
300
400
EY (MeV) Figure 15. The 4He(y,pn)dcross section. The data for lower E, are plotted together. The solid line shows the deuteron cross section with a scaling factor of 6.0.
3.3. Comparison with lower E, data There are several data sets of 3He and 4He photodisintegration at lower E, region, and some of them determined the 2N(pn) cross section. In Fig. 14, the 2N(pn) cross section for 3He by TAGX is plotted together with lower E, data [12,13]. The deuteron photodisintegration cross section with a scaling factor of 1.24 is also plotted. The similarity of the E, dependence of the 2N(pn) cross section to that of the deuteron cross section over such a wide E, range from threshold to 500 MeV is remarkable. We can conclude that the photon absorption cross section of the np-pair in 3He scales to that of the deuteron photodisintegration over a wide E, range. Figure 15. shows 4He(y,pn)dcross section with lower E, data [14-181. In the figure, the scaled deuteron photodisintegration cross section with a factor of 6.0 is plotted together. It is our surprising to find that there is a big discrepancy or jump in magnitude around E, = 140 MeV, which is not observed in the 3He case. To confirm the absolute magnitude of the
cross section by TAGX, we deduce the inclusive 4He(y,p) cross section by integrating over the neutron momentum vector of the 2N(pn) events. Figure 16 shows the 4He(y,p) cross section obtained in this way. In the same figure, the 4He(y,p)X cross section measured by Homma et al. [19] are plotted together. The bump observed in Homma's data at around 700 MeV/c was assigned due to the 2N(pn) process. Good agreement of the deduced (y,p) cross section of TAGX with that of Homma's at around 700 MeV/c region confirm that the absolute magnitude of the TAGX cross sections is correct.
Figure 16. The double differential cross section for 4He(y,p). Open circles show the inclusive 4He(y,p)Xresults by Homma et al. [19]. Solid circles show the differential cross section for 4He(y,p)nd determined from the TAGX 4He(y,pn)dcross section integrated over the neutron momentum vector.
Therefore the measurements of the 2N(pn) cross section covering the photon energy of 140 MeV is strongly desired to understand what is going on at this photon energy. 4. CONCLUSIONS
In this paper, we present the results of the photodisintegration experiments for 3He and 4He measured with the TAGX spectrometer. Large acceptance for both proton and neutron of TAGX enabled us to cover a wide kinematical range, and to perform the measurements in a kinematically complete way. This was essential to identify the photon absorption process by two nucleons in the nucleus. The total cross section for photon absorption by the np-pair was determined for 3He and 4He, and the E, dependence of the cross section was found to be quite similar to that of the deuteron photodisintegration. A microscopic calculation succeeded to reproduce the cross section over a wide E, range. Photon absorption by two protons in 3He was also observed and the cross section was extracted. The cross section was found to be about two order smaller than that of the np cross section, and to have different E, dependence. A microscopic calculation using a realistic wave function succeeded to reproduce the magnitude of the cross section, but
failed to account for the E., dependence. The analyzing power for the photon absorption by two protons was measured to study the photon absorption by two protons in detail, and the preliminary results are presented.
REFERENCES 1. H.Arenhove1 and M. Sanzone, Photodisintegration of Deuteron, Few Body Systems Suppl.3 (Springer-Verlag, Wien, 1991) 2. Maruyama et al. (TAGX Collab.): Nucl. Instrum. Methods A376 (1996) 335. 3. M.Kobayashi and K.Kondo, Nucl. Instrum. Methods, A376(1996)335. 4. M.Vanderhaeghen et al., Nucl. Phys. A595 (1995) 219. 5. E.Jans et al., Nucl. Phys. A475 (1987) 687. 6. T.Emura et al. (TAGX Collab.) : Phys. Rev. C 49 (1994) R597. S.Endo, Ph.D. thesis, Journ. Sci. Hiroshima Univ. A57, 1993. 7. T.Emuar et al. (TAGX Collab.):Phys. Rev. Lett. 73 (1994) 404. T.Emura et al. (TAGX Collab.) : Phys. Rev. Lett. 74 (1995) 1035. 8. K.Maruyama et al. (TAGX Collaboration) : Phys. Lett. B393(1997)295. K.Niki, Ph.D. thesis, Journ. Sci. Hiroshima Univ. A55, 1991. 9. P. Wilhelm, J.A. Niskanen and H. Arenhovel, Phys. Lett. B335 (1994) 109. 10. P. Wilhelm, J.A. Niskanen and H. Arenhovel, Phys. Rev. Lett. 74 (1995) 1034. 11. P. Wilhelm, J.A. Niskanen and H. Arenhovel, Phys. Rev. C51 (1995) 2841. 12. B.L. Berman S.C. Fultz and P.F. Yergin, Phys. Rev. C10 (1974) 2221. 13. V.N. Fetisov, A.N. Gorbunov and A. T. Varfolomeev, Nucl. Phys. 71 (1965) 305. 14. A.N. Gorbunov, Sov. J . Nucl. Phys. lO(1969) 268 15. Yu. M. Arkatov et al., Sov. J. Nucl. Phys. 11(1970)639. 16. F.Balestra et al., Nuovo. Cim. 49(1979)575. 17. F.Balestra et al., Nuovo. Cim. 38(1977)145. 18. S.M. Doran et all Nucl. Phys. A559(1993)347. 19. S. Homma et al., Phys. Rev. C36 (1987) 1623.
Hadron and Nuclear Physics with ElectromagneticProbes K. Maruyama and H. Okuno (Editors) 2000 Elsevier Science B.V. All rights reserved.
'
Photonuclear cross sections of three-nucleon systems and the role of t hree-nucleon forces G. Orlandini, W. Leidemann," V.D. Efrosb and E.L. Tomusiakc "Department of Physics, University of Trento, 1-38050 Povo (Trento) Italy bRussian Research Centre " Kurchatov Institute", Kurchatov Square, 1, 123182 Moscow, Russia 'Department of Physics and Engineering Physics, University of Saskatchewan, Saskatoon, Canada S7N OW0
After a brief introduction about the Lorentz integral transform method (LITM) its application to the total photodisintegration of 3H and 3He is presented. In particular the role played by three-nucleon forces is discussed. The results for the three-body systems are analyzed in view of clarifying the missing experimental strength puzzle in the 4He giant dipole resonance.
1. INTRODUCTION
One of the important issues in the study of few body systems by means of the electromagnetic interaction is the role of the three-body force. In particular one would like to understand which other observables, besides the binding energies of 3H and 3He, show sizable effects which can be ascribed uniquely to it and learn more about its origin and mechanism. In addressing this issue, however, a big difficulty is represented by the scarce knowledge of the nuclear systems in their final states. While the few body techniques have reached a very high degree of accuracy in the treatment of bound states, much is still to do for continuum states. These states are very difficult to treat exactly, and the kinds of approximations one is often forced to introduce can mask the effects of interest. It is certainly problematic to disentangle the presumably not large effects of the threebody force from spurious effects introduced by any approximation, be it systematic or statistical (numerical). In order to circumvent the obstacle represented by the continuum states a few years ago the LITM was proposed, a method which allows to calculate inclusive cross sections using bound state techniques [I]. In the following a sketch of the LITM will be given.
2. THE LORENTZ INTEGRAL TRANSFORM METHOD
The inclusive response function of a system to an external probe has the form
where 0 is a transition operator which characterizes the process under consideration, 90 and Q f are the ground state wave function and a complete set of final state wave functions, respectively, while w denotes the excitation energy. These response functions are the essential ingredients of any electromagnetic inclusive cross section. The main idea of the LITM is to avoid calculating R(w) directly, since this implies the knowledge of the final state wave functions. Instead, one calculates an integral transform of it and then arrives at R(w) by inversion. In particular in the case of a Lorentz kernel the integral transform of R(w) is given by
With the help of closure one can show that
where Eo is the ground state energy of the Hamiltonian H, and equation
G is determined by the
The method then proceeds in two steps. First Eq. (4) is solved. This can be done using bound state techniques. That is the essential advantage of the LITM. In fact, because of Eqs. (2)and (3) the norm of exists and thus, contrary to continuum wave functions, \jir vanishes at large distances like a bound state wave function. U I ) , calculated as the norm of The second step is the inversion of the transform @(aR, 3, in order to obtain R(w). For this purpose is calculated for a set of positive UR values with a fixed 01 > 0. Details of the inversion procedure are discussed in Ref.[2]. 3. THE PHOTODISINTEGRATION CROSS SECTION
The total photoabsorption cross section for energies below pion threshold is computed by using the unretarded dipole operator
This form of the operator already includes the effects of exchange currents due to the Siegert's theorem. It is known that as far as the total cross section is concerned corrections to the unretarded dipole operator are small in the energy range studied here. We consider this approximation adequate for investigating the effects of NN and NNN forces to the
process. The total photoabsorption cross section is then given in terms of the dipole response function R by
where
As shown in the previous section the use of the LITM to calculate R(w) implies that one has to solve the following Schrodinger-like equation
This has been done within the Correlated Hyperspherical Harmonic (CHH) expansion method for a1 = 20 MeV and for a large number of OR points between 0 and 150 MeV. The inversion of the Lorentz integral transform has been done using the "Regularization method" as described in Ref. [2]. We present the results obtained for 3H and 3He using the Argonne AV14 potential [3] and the Bonn-B r-space potential [4]. As NNN potentials we include the Urbana-VIII (UrbVIII) [5] and Tucson-Melbourne (TM) [6] NNN models. 4. RESULTS AND DISCUSSION
Before presenting our results a few words should be said about the experimental and theoretical situation on the photodisintegration cross sections of the three-nucleon systems. About 13 sets of experimental data exist from the period 1964-81 for energies below 50 MeV. In particular in 1981 Faul, Berman, Meyer and Olson [7] were comparing twoand three-body break up cross sections of 3H and 3He. At that time the most accurate theoretical results were the Faddeev calculations of Gibson and Lehman [8] and Barbour and Phillips [9], however, in the conclusions the authors of Ref.[7] were judging the situation as follows: More theoretical work will be needed before the experimental results for the photodisintegration of three-body nuclei can test quantitatively the charge symmetry of the nuclear interaction [This was the main issue at that time, which led to comparative studies of the two mirror nuclei]. The present theoretical models are still rudimentary in that they do not include simultaneously a consistent treatment of the final state interactions, the tensor force in the triplet interaction, and the Coulomb repulsion between the protons in 3 H e . In the results we are going to show all those drawbacks are cured. In particular the LITM and the use of the CHH approach to solve Eq. (8) have allowed us to include the full final state interaction and the Coulomb effects. In Figs. 1 and 2 we show the total photodisintegration cross section of 3Hin the peak and tail regions. One notices a sizable effect due to the three-body interaction. In Fig. 1 only the NNN force Urbana VIII results are shown. The effects of the Tucson-Melbourne force are quite similar. The lowering of the peak due to the three-nucleon interaction is also independent on the NN potential. The use of the Bonn potential does not change the results.
Figure 1. Total photodisintegration cross section of 3 H in the peak region.
A quenching of the peak due to the NNN interaction could in principle be expected because of the following reason. For 3 H the so called Bremsstrahlung sum rule is valid
where ( r i ( 3 H ) ) is the triton charge radius.
0.00
1
60
80
100
120
140
a,[MeV]
Figure 2. Total photodisintegration cross section of 3H in the tail region. Since the NNN interaction gives additional binding to the system it is also responsible for its shrinking in size. Because of the validity of the sum rule a smaller charge radius will presumably reflect in a lower peek. In fact this is the region which gives more contribution to the sum rule, due to the inverse energy weight.
Besides lowering the peak the N N N interaction has also an effect on the tail of the cross section which is enhanced. This effect is shown in Fig. 2. For comparison, the results obtained in Ref.[10] for the Malfliet-Tjon potential (MT) [ll],and in Ref.[2] for the TRSB potential of Ref.[l2] are also shown. The MT potential is purely central, contrary to the TRSB which has a tensor component, but a very soft core compared to AV14. So one can say that below pion threshold a combination of short range and tensor correlations enhance the cross section considerably. Three body forces further amplify this effect. The solution of Eq.(8) is obtained for the two isospin channels T=1/2 and T=3/2 separately. This implies that the contributions of the two channels to the cross section can be analyzed separately. In particular in Fig. 3 we show a,(w,) for the T=3/2 channel is. Only three-body break up reactions contribute to this cross section. It is interesting to notice that there is very little dependence on the NN interaction, while the three-body force is responsible for a sizable enhancement of the cross cross section in the region below pion threshold. This effect seems to be a pure three-body effect with no relation to the "binding effect" discussed above in the peak region. This result is very interesting because it suggests the possibility of an experimental investigation of three-body force effects in photoreactions. Considering that an already rather large sensitivity is found in an inclusive reaction one can hope to find even larger effects in exclusive three-body break up reactions. In this respect an extension of the LITM to exclusive reactions is needed. Successful investigations in this direction have already started [13,14].
\
. ..... ..
..
V14
.'. . \
- - - - Bonn
..
....
- V14+UVIII
0.035
80
90
100
110 w, [MeV]
120
130
I
140
Figure 3. Total photodisintegration cross section of 3H in the T=3/2 channel.
In Fig. 4 the results for 3He are presented and compared to experimental data. The comparison is good, even if the uncertainties in the experimental results are large. More precise data are needed, especially if one wants to detect the effects of the three-body force which are of the same magnitude as those in Fig.1. We would like to discuss here briefly the case of 4He. In Ref. [16] the theoretical results
a, [MeV]
Figure 4. Total photodisintegration cross section of 3He. Full circles correspond to experimental data from Ref.[l5]. The two dotted lines represent lower and upper bounds of the data of Ref. [7]
obtained with the LITM and semirealistic central interactions largely overestimated the experimental data. From the study of three-body nuclei we have learnt that the use of realistic forces enhances the cross section in the tail, but very little effect is found in the peak region. There, on the contrary, the quenching effect of the three-body force seems to be sizable and related, via the charge radius, to the effects the NNN-force has on the binding energy. Such effects are even larger in 4He. There the three-body force contributes with 4 MeV to the binding energy. However, the Bremsstrahlung sum rule in 4He is not only connected to the charge radius, but has an additional term proportional to the mean square proton-proton distance [17]. So the argument of binding effects does not apply as simply. Nevertheless it would be of extreme interest to investigate the role of the three-body force in the resonance region of the 4He photodisintegration cross section. Though it requires a somewhat larger computational effort than it does for three-body systems the LITM can be applied for realistic interactions also in this nucleus. On the other hand one has to remember that modern accurate experimental data for both two-, three- and four- break up channel are also missing. So also an experimental effort in few body systems photoreactions below pion threshold is desirable in order to open a new realm of investigations of the properties of the three-body forces.
Acknowledgements This work has been partly supported by Istituto Nazionale di Fisica Nucleare, Minister0 dell'universiti e della Ricerca Scientifica and the NATO grant N. 971504.
REFERENCES 1. V. D. Efros, W. Leidemann and G. Orlandini, Phys. Lett. B 338 (1994) 130. 2. V.D. Efros, W. Leidemann, and G. Orlandini, Few-Body Syst. 26, (1999) 251.
3. 4. 5. 6.
R.B. Wiringa, R.A. Smith, and T.L. Ainsworth, Phys. Rev. C29, (1984) 1207. R. Machleidt, Adv. Nucl. Phys. 19, (1989) 189. R.B. Wiringa, Phys. Rev. C43, (1991) 1585. S.A. Coon, M.D. Scadron, P.C. McNamee, B.R. Barrett, D.W.E. Blatt and B.H.J. McKellar, Nucl. Phys. A317, (1979) 242 ; S.A. Coon and W. Glockle, Phys. Rev. C23, (1981) 1790; S.A. Coon (private communication). 7. D. D. Faul, B. L. Berman, P. Meyer and D. Olson, Phys. Rev. C24, (1981) 849. 8. B.F. Gibson and D.R. Lehman, Phys. Rev. C11, (1975) 29; C13, (1976) 477. 9. I.M. Barbour and A.C. Phillips Phys. Rev. Lett. 19, (1967) 1388; Phys. Rev. C 1 , (1970) 165. 10. V.D. Efros, W. Leidemann and G. Orlandini, Phys. Lett. B408, (1997) 1. 11. R.A. Malfliet and J. Tjon, Nucl. Phys. A127, (1969) 161. 12. R. De Tourreil, B. Rouben and D. W. L. Sprung, Nucl. Phys. A242, (1975) 465 ; J. C6te, R. De Tourreil, B. Rouben and D. W. L. Sprung, Nucl. Phys. -4273, (1976) 269. 13. V.D. Efros: Sov. J. Nucl. Phys. 41, 949 (1985) 14. A. Lapiana, Tesi di Laurea, Universita' di Trento, 1999; A. Lapiana and W. Leidemann, in preparation. 15. V. N. Fetisov, A. N. Gorbunov and A. T. Varfolomeev, Nucl. Phys. A71, (1965) 305. 16. V.D. Efros, W. Leidemann, G. Orlandini, Phys. Rev. Lett. 78, 4015 (1997). 17. See e.g. G. Orlandini and M.Traini, Rep. Prog. Phys. 5 4 (1991) 257.
ThisPageIntolyLfBk
ThisPageIntolyLfBk This Page Intentionally Left Blank
Hadron and Nuclear Physics with ElectromagneticProbes K. Maruyama and H. Okuno (Editors) 2000 Elsevier Science B.V. All rights reserved.
"
Quasi-deuteron picture for 3He and 4He photodisintegration S. Hirenzaki, Y. Umemoto and K. Kume Physics Department, Nara Women's University, Nara 630-8506, Japan We have developed a phenomenological quasi-deuteron model to describe photodisintegration reactions of 3He and 4He at intermediate photon energies and compared the results with experimental data obtained by the TAGX group in which contributions from photon absorption by pn pairs were selectively observed. The data can be reproduced reasonably well. However there still remain certain discrepancies between calculated results and data. 1. INTRODUCTION
Photodisintegration reactions of light nuclei have been studied for a long time both theoretically and experimentally. Experimental results for these reactions are naively expected to be reproduced precisely by theoretical works, since the reactions are induced by the well-known electromagnetic interaction and it is believed that the structure of few-nucleon systems can be calculated reliably. Despite of this expectation, there still remain certain discrepancies between experimental data and theoretical results. We are interested in the origins of these discrepancies. The photodisintegration of nuclei was studied theoretically using the quasi-deuteron model by Levinger [I] and later by Futami and Miyazima [2]. They expressed the photodisintegration cross section of nuclei using those of deuteron with a certain factor, the so called Levinger factor, which accounts essentially for two effects, the relative wavefunction of the pn pair and the effective number of deuterons in the target nucleus. Recently, the Levinger lactor was updated and determined for many nuclei using current data of the rms radius [3]. This enables us to obtain information regarding the relative wavefunction of pn pair. The accuracy of the model was also investigated for a wide range of momentum transfer values for trinucleon systems [4]. These results reproduce the experimental data qualitatively well, indicating that photon absorption by a pn pair is important in photonuclear reactions. Since 1987, new experimental data for photodisintegration reactions of light nuclei have been obtained by the TAGX group [5]. They have obtained kinematically complete data p 4He(y,pn)dfor the first time [6,7]. They have strived to separate whole of 3 ~ e ( y , p n )and events into two- and multi-nucleon absorption processes using the momentum ordering method [8]. We are very interested in the da.ta, since the contributions from only photon absorption by pn pairs are selectively observed. It should be possible to precisely calculate these contributions in the quasi-deuteron picture. Another interesting point is that the data are completely exclusive which do not include any extra particles like pions in the
final state. Thus, we think it extremely important to investigate this data to know whether it can be understood theoretically. To this time, the 3He(y,pn)preaction has been studied theoretically by Wilhelm et al. [9,10]. They considered both triplet (i.e. deuteron channel) and singlet configurations for initial pn pairs in the nucleus, and included photon interaction with nucleonic, mesic, and A currents. They calculated the cross sections and spin observables and found that the data for total cross sections deviate from the theoretical results at lower energies ( E , 5 200 MeV). Since their results for differential cross sections are not in a form that could be compared with the data directly, there exist no comparisons between data and theory for angular dependent observables [lo]. The 4He(y,pn)ddata [7] indicate that the total cross section has a strong energy dependence around E, = 150 MeV. This has not been investigated theoretically. Theoretical results by Tezuka [ll]are consistent with the data for higher energies, but there are no theoretical calculations which can be compared to the data over the entire incident photon energy region. Their results include the attenuation factor to account for the distortion effects, and the final nuclear state is not specified as a deuteron. and In this paper we attempt to understand the reactions 3He(y,pn)pspectator 4He(y,pn)dspeCtator theoretically. For this purpose we investigate both 3He and 4He photodisintegraion reactions within the same theoretical framework and try to reproduce all data for both reactions simultaneously. We calculate all observables in suitable forms which can be compared to the data directly. Our theoretical studies are reported in Ref. [12] in detail. 2. FORMALISM
In this section we describe our phenomenological model which is applied to both 3He(y,pn)pspectator and 4He(y,pn)d,p,tator reactions. Since we would like to calculate all observables in suitable forms to compare directly with the data, we evaluate the finalstate three-body phase space exactly. The cross section for the A(y,pn)B reaction can be written as
where A indicates the target nucleus, 3He and/or 4He, and B is the spectator particle in the final-state, psp,tator and/or dsPectator.The quantity s corresponds to the Mandelstam variable in the initial y + A system. We integrate the final-state phase space appropriately for each experimental result. In ( I ) , A(. . .) is the Kallen function, which is introduced to properly normalize the cross section by the initial photon flux, defined as
In our model, we consider only the deuteron channel for the initial pn pair in the nucleus and express the square of the amplitude of the photon absorption by the nucleon pair,
using the differential cross section
1 T(7pn -+ pn) 12=
(g)
( g )of d(7,p)n reaction as
+
Ep En 1 MpMnMd 1 Pp 1 '
X ' ~ ~ ( S , O~,: ) ( 2 r ) '
where all kinematical variables are evaluated in the center-of-mass frame. We use a data [13], which reproduce the d(y,p)n cross section phenomenological fit of the well in the wide energy region E, = 20-440 MeV. In order to calculate the observables for He target cases, we postulate the quasi-deuteron picture and describe the square of the y absorption amplitude by the He nucleus as
(g)
where we evaluate / T(ypn -+ pn) l 2 using Mandelstam's s and t variables for the two nucleon system that absorbs the photon. II, is the relative wave function between the pn pair and the spectator particle B in the He nucleus, which is assumed to take a Gaussian form,
where pf represents the Fermi momentum between the pn pair and the spectator particle B and is roughly estimated to be 146 MeV for 3He and 205 MeV for 4He from the data of the charge radii. p~ is evaluated in the laboratory frame. Here we would like to describe some features of the present model. The advantages of our model are that we can calculate any cross sections that can be compared to the data directly for both 3He and 4He target cases and that we can calculate observables in the entire energy region of interest using experimental information regarding the deuteron photodisintegration reactions. However, we include neither the effect of the deuteron spectroscopic factors nor the effects of the compact wave function of the pn pair in the target He. Both effects are expected to be insensitive to the incident photon energy and to change the overall normalization of the cross sections. Thus, we are free to introduce a constant factor into each reaction as a parameter to normalize the absolute value of the cross sections when it is necessary. In the model we do not include contributions from the singlet pn pair in the initial state, which were evaluated in Refs. [9]and [lo] and total cross shown to be approximately 50 pb at E, = 100 MeV for the 3He(y,pn)pspectator section. This contribution decreases monotonically as a function of E, and is known to give a minor contribution to the photodisintegration reactions at higher energies. We should also mention here that the final state interaction is partly included in the model from the beginning, since we have used the fit to the data of the deuteron case. 3. NUMERICAL RESULTS
In this section we compare our calculated total cross sections with experimental data for reaction taken by the TAGX the 3He(y,pn)pspectat0,reaction and the 4He(y,pn)dspeCtator group [6-81. Other numerical results are found in Ref. [12].
We show the total cross sections of the 3He photodisintegration by two body (pnpair) photon absorption processes in Fig. 1 as a function of the incident photon energy. The experimental total cross sections are obtained by integrating the differential cross section data over the observed kinematical region. The contribution from the unobserved kinematical region is corrected for by extrapolation using the two-nucleon absorption model [14]. Theoretical results only include the deuteron channel of the initial pn pair in the target. As we can see from the figure, the data show the peak at E, = 225 MeV, which seems to be due to the A excitation, while the calculated results predict this peak at 275 MeV. This difference could be explained by inclusion of the siglet configuration of the pn pair in the initial state: which was evaluated by Wilhelm et a1 [9,10] and shown to be larger for lower photon energies. However, the energy dependence of the theoretical results will be caused to deviate more greatly from the data by including the singlet contribution a t E, = 125 - 205 MeV. In this energy region, the experimental values increase monotonically with photon energy, while the theoretical results may become flat or even possess the opposite energy dependence by including the singlet contribution. Our results are consistent with those calculated by Wilhelm et a1 [9,10] for the initial pn pair in the deuteron channel.
.
0
100
200
300
400
EY (MeV)
Figure 1. Total cross section of the 3He(y,pn)p,,,,t,t, reaction as a function of incident photon energy. The solid curve indicates the result with pf=150 MeV. Data are taken from ref. [6].
Total cross sections of the 4He(y,pn)d,,,t,t,, reaction were measured a t several photon energies by the TAGX group [7] and found to have a strong energy dependence around E, = 150 MeV. This energy dependence is almost like a 'discontinuity' to the lower energy data taken by different groups [15-181. We show the calculated results in Fig.2 with varing pf over a wide range to investigate the possibility of reproducing the energy dependence by changing the wavefunction of 4He. We also multiplied the calculated cross sections
by a factor to reproduce the A peak. We find that our results do not reproduce the experimental energy dependence. It is difficult to reproduce the observed 'discontinuity'.
EY (MeV)
Figure 2. Total cross section of the 4He(y,pn)d,,,,,t,, reaction as a function of incident photon energy. The lines indicate calculated result with pf = 100 MeV (dashed curve), 200 MeV (solid curve), 300 MeV (dotted curve), and 400 MeV (dash-dotted curve), respectively. calculated results are normalized to reproduce the experimental peak height in the A energy region. The normalization factors are 1.45 for pf = 100 MeV, 1.53 for pf = 200 MeV, 1.67 for pf = 300 MeV, and 1.88 for pf = 400 MeV. Data are taken from refs. [7] (solid circles), [15] (squares), [16] (crosses), [17] (open circles), and [18] (triangles).
We would like to mention here the recent work by Efros et al, 1191 in which the total 4He photodisintegration cross section is studied in the giant resonance energy region. In addition to the total 4He photodisintegration cross section, they have shown separately the contributions due to the two-body decay channels, 4He(y,n)3He and 4He(y,p)3H. We can estimate the absolute value of the cross section in our case (3-body decay) by subtracting the 2-body decay contribution from the total. It is approximately 1 mb at E, 40 MeV, which is even larger than our largest result (- 500 pb). If this is correct, the older data at lower energies may include large errors.
-
4. SUMMARY
In this paper, we have investigated the photodisintegration reactions 3He(y,pn)pspectator and 4He(y,pn)d,p,,ta,o, which have been observed by the TAGX group. We have investigated both reactions using a single theoretical model and compared the calculated results with experimental data. In the model, we have assumed the quasideuteron mechanism to describe the photon absorption amplitude using the experimental information of deuteron photodisintegration. We have introduced the relative wave func-
tion between the pn pair and the spectator particle in the target He and treated the phase space integration carefully in order to calculate the observables in a suitable coordinate system to compare with the data. Using the present theoretical model, we can calculate observables in entire energy region of interest and can compare with the experimenal data directly. The final state interacton is partly included in the model since we have used the fit to the data of deuteron photodisintegration. Our model is found to reproduce the gross features of all the existing data, but certain descrepancies remain. These descrepancies are (i) the energy dependence of the total cross section for 3He(y,pn)p,,,,tat,, and (ii) the step-like change of the total cross section 150 MeV. We found that these discrepancies of 4He(y,pn)d,,,tat,, reaction around E, are difficult to reproduce with the present phenomenological model. Thus we belive that these discrepancies involve important information on the essential differences between the photodisintegration reactions of 3He and 4He and that of the deuteron and should be treated carefully. For the 4He target case, the step-like change could be due to the inclusion of large error for some parts of experimental data. For further studies, we need to develop a microscopic model based on accurate wave functions of the initial and final states and photon interaction with hadronic currents in order to understand these reactions more deeply. We think it is very important to calculate the observables in a suitable form using the microscopic model to compare with data. We would like to thank the TAGX group for stimulating discussions. N
REFERENCES 1. J. S. Levinger, Phys. Rev. 8 4 (1951) 43. 2. Y. Futami and T. Miyazima, Prog. Theor. Phys. 46 (1971) 802. 3. 0. A. P. Tavares and M. L. Terranova, J. of Phys. G 1 8 (1992) 521. 4. B. Gangopadhyay and J. S. Levinger, J. of Phys.Gl8 (1992) 1933. 5. K. Maruyama et al., Nucl. Inst. Meth., A376 (1996) 335. 6. T . Emura et al., Phys. Rev. C 4 9 (1994) 597. 7. K. Maruyama et al., Phys. Lett. B393 (1997) 295. 8. S. Endo., J. Sci. Hiroshima Univ., 57A (1993) 1. 9. P. Wilhelm, J. A. Niskanen, and H. Arenhovel, Phys. Lett. B335 (1994) 109. 10. J. A. Niskanen, P. Wilhelm, and H. Arenhovel, Nucl. Phys. A586 (1995) 693. 11. H. Tezuka, J. Phys. Soc. Jpn. 5 7 (1988) 3766. 12. Y. Umemoto, S. Hirenzaki and K. Kume, Prog. Theor. Phys. 1 0 1 (1999) 627. 13. P. Rossi et al., Phys. Rev. C 4 0 (1989) 2412. 14. S. Endo, private communication. 15. S. M. Doran et al, Nucl. Phys. A559 (1993) 347. 16. Yu. A. Arkatov et al, JETP Lett. 9 (1969), 278; Sov. J. Nucl. Phys. 1 0 (1970) 639. 17. A. N. Gorbunov and V. M. Spridinov, Sov. Phys. JETP 3 4 (1958) 600; A. N. Gorbunov, Sov. J. Nucl. Phys. 1 0 (1969) 268; Proc. P. N. Lebedev Phys. Inst. 7 1 (1974)l. 18. F. Balestra et al, Nuovo Cim. 49A (1979) 575; F. Balestra et all Nuovo Cim. 38A (1977) 145. 19. V. D. Efros, W. Leidemann, and G. Orlandini, Phys. Rev. Lett. 78 (1997) 4015.
Hadron and Nuclear Physics with Electromagnetic Probes K. Maruya~naand H. Okuno (Editors) 2000 Elsevier Science B.V. All rights reserved.
"
Two-nucleon emission experiments at Mainz Peter Grabmayr
a
"Physikalisches Institut der Universitat Tubingen, Auf der Morgenstelle 14, D-72076 Tiibingen, Germany e-mail address:
[email protected] Two-nucleon emission experiments performed with real and virtual photons at the cw-electron accelerator MAMI at Mainz are discussed. We have studied the reaction mechanisms of photon absorption. Evidence for effects due to nucleon-nucleon correlations at short internucleon distances is presented. The need of a combined analysis of high resolution (y,NN) and (e,elNN) experiments is emphasised. 1. INTRODUCTION
The mean field concept to describe properties of nuclear states has been rather successful in the energy range around the Fermi surface which represents the classical low-energy, low-momentum regime. Only more recent one-particle transfer and knockout experiments [1,2] revealed that about 25% of the strength was missing. Immediately this was traced back to the principal worries stated already at the time of birth of the shell model more than 50 years ago, namely, that the high momentum components apparent at short internucleon distances spoil the mean-field concept and that they shift strength high up into the continuum where the typical spectroscopy cannot pursued anymore. The strong repulsion observed at small relative distances is well known in free nucleon-nucleon (NN) scattering; to a certain degree this short-range feature survives in atomic nuclei despite the creation of an average potential contributed by all nucleons and the supportive Pauli principle. Thus one observes deviations from the mean field behaviour which are the so-called nucleon-nucleon correlations. There are several kinds of trivial correlations as e.g. due to Pauli principle or conservation of centre-of-momentum which are included in the "free" two-nucleon density PNN:
---
f 1 P N N ( ~ I~. 2 =) [ P N ( ~ I. )P N ( ~ - 2 )- P N ( ~ I I. ~P N~( ~) Z I ~ I ) I free
+
e.g.Pau1i-correlations
The dynamical correlations Cdynare in the centre of interest and in particular those which can be connected with short inter-nucleon distances: f Cdyn(r1, r2) = PNN - PNN dynamic genuine free
The long ranged correlations have been incorporated into specific mean field approaches (see. e.g. ref. [3]). However, the treatment of the short ranged correlations is more intricate. One of the strongest argument for short ranged effects in the nuclear medium still arises from theoretical investigations of nuclear binding. A recent compilation by H. Miither [4] demonstrates the need of NN correlations for binding of nuclear matter at the empirical saturation point using realistic NN potentials. Simple Hartree-Fock calculation yield binding only with effective potentials. While for different modern NN potentials about the same binding energy per nucleon is extracted from Brueckner-Hartee-Fock calculations different contributions of the n exchange have been found. This clearly indicates different off-shell behaviour of these potentials which fit the phase shifts for NN scattering with equal quality. The tensor part of these potentials is poorly defined, however tensor type correlations play a crucial part in the calculations for the np channel [4].
Figure 1. The two-nucleon density for two 1pl12 nucleons (adapted from ref. [4]). Recent investigations aim towards modern techniques to produce two-hole spectral functions for real nuclei [4,5,9]. Here, a result based on the exp(S)-method for 160is given [4]. The correlated wave function @ is derived from the uncorrelated one Q, through:
Iq >= exp
C S, IQ, (n:l
A
)
>
with the operator
where the creation and annihilation operators produce sets of n particle-n hole states. Limiting the calculations to n=2 the correlated wave functions read $llVl>
>
$21~1~2
=
ivl>+sl~vl>
=
d$iI~l $ 1 1 ~ 2
>
> + s 2 1 ~ 1 ~>2
The two-body density for two protons in the lpl12 orbit of
160
is given in Fig. 1. The x, z-plane is selected for y=O where the second particle is fixed at (x=0, y=O, z=2 fm). On the left side of Fig. 1 the operator s2=0 and therefore
the two-body density is just the product of two one-body densities (irrespective of the location of the second particle). For s2#0 however, a clear depletion in the two-body density can be observed at the position of the second particle. Similar calculations for the T=O np pair reveals not only an expected increase of strength by a factor of about 4 but also its concentration at short internucleon distances despite larger meson exchange contributions. As one-nucleon removal is not an adequate tool one has to investigate the two-nucleon removal experiments. This experimental method for an access towards correlations had been suggested already in the 50-ies, in particular pion and photon induced reactions. Both probes are understood to be absorbed dominantly on nucleon pairs at relative close proximity [lo]. However, beam quality, energy resolution and background reactions, as e.g. the notorious final-state interaction, prevented serious progress. However, the new generation of experimental techniques for cw accelerators and large area, high resolution detectors on one side and new developments of nuclear structure and reaction models permit a renewed approach towards NN correlations. The aim of the present experimental series is a combined analysis of (y,np), (y,pp), (e,elnp) and (e,elpp) to exploit the advantages of each of the reactions. 2. EXPERIMENTS AT MAINZ WITH REAL AND VIRTUAL PHOTONS
Figure 2. Experimental setup with tagger, proton hodoscope Pip [ll] and neutron timeof-flight array [12]. Three kinematical regions are indicated.
Li
Figure 3. Layout of the three spectrometer facility [13] at Mainz.
The experiments of the PiP/TOF group are performed at the cw electron accelerator MAMI [14] at Mainz using the Glasgow tagging spectrometer [15]. Electron beams of 855 MeV impinge on 4 pm thin Ni radiator or on 100 pm thick diamond crystal in case polarised photons are needed. The photon beam is collimated in order to be well defined (diameter of approx. 1,5 cm) on the nuclear targets which consist either of liquid helium within a kapton can or of sheets of graphite or lithium. The photon flux is calibrated to free running scalers within the tagger focal plane by so-called 'tagging efficiency' mea5 f have been obtained. The target is surrounded surements. Typical values of ~ ~ = 4 2% by a ring of 1 mm thick plastic counters, which act as start detectors for charged particles and as veto detectors for neutrons to be selected from TOF[12]. Charged particles are
detected by the fivefold layered hodoscope Pip [l11. Protons are discriminated against deuterons and pions via the A E - E technique. The energy of these particles is extracted exclusively from the pulse height after proper correction for quenching and energy losses in air and dead layers. The neutron energy is obtained from the time-of-flight measured by T O F with an average FWHM of about 0.7 ns. The charged particles in T O F can be selected by appropriate cuts on pulse height vs. time-of-flight. An overall energy resolution (Em)of 6 MeV has been obtained. The angular resolution of 3' in polar and of 9" in azimuthal direction can be translated into a momentum resolution of 35 MeV/c. The 160(e,e'pp)14Cexperiments with virtual photons employ the three spectrometer facility [13] with high beam currents ( 1 4 0 PA). A waterfall target [16] was used and a so-called super-parallel kinematics was selected [17], which suppressed A currents. 3. REACTION MECHANISMS O F T H E (y,NN) REACTIONS
The three observables of importance are the missing energy, the missing momentum and the relative momentum between the two nucleons of the pair. The missing momentum &, is defined as the difference between incoming photon momentum and the outgoing + nucleon momenta p', and p',: = k - fi,, - p'p. In plane wave impulse approximation (PWIA) pm can be interpreted as the momentum P of the pair before the interaction. It governs the gross features of the angular and energy dependence of the cross sections. The missing energy Em = k -T, -Tp -TR is defined as the difference of kinetic energies of particles in the in- and outgoing reaction channels. In particular, TR is the kinetic energy of the recoiling ( A - 2) system calculated relativistically from p,. Finally, the relative momentum is defined as p', = ;(& - p',). Note, that there is no modelfree possibility to get a handle of the relative momentum 6 in the initial state.
Figure 4. The (y,np) reaction for three photon energy ranges is compared for the targets 12C and 4He.
Figure 5. The (y,pp) reaction for three photon energy ranges is compared for the targets 12C and 4He .
An overall picture of the competing reaction mechanisms is obtained best by comparison [18,19,7]of the data to the predictions of the Valencia model [20]. For three photon energy bins the missing energy spectra for the (y,np) reaction on 12C and 4He are plotted. The 12C data are well described by the calculations. For 4He a qualitative agreement can be stated. It is observed that even at highest photon energies the yield at low Em arises from the two-body photon absorption process. The yield at higher photon energies ( e l 5 0 MeV) for 4He is reduced relative to that for 12C due the reduced influence of final state interactions. The description of the (y.pp) reactions is less satisfying. These findings permit the concentration on low excitation energies Ex= Em - Q of the residual nuclei. In contrast to the Valencia model, state-of-the-art calculations [21,22]permit a proper treatment of the two-nucleon knockout reaction including the competing processes. Effects of the NN correlations are expected to be recognised through excess of high momentum components in angular distributions different from the quasideuteron prediction or in general in unexpected variation of cross sections, particular when final states with specific quantum numbers can be selected because of a sufficiently high energy resolution.
o
-
0
20
l
40
I l I I I 60 en ioa 12a i4a 160 Neutron Anglo (dog)
I
im
/ . . .1.). . .I 0. . . . . . . .3 l..
0
11)
d. .d.d Iso 7m
>?o
I.0
1.0
Upl9
Figure 6. Angular distribution of the 12C(y,np) reaction at E,=(135&20) MeV.
Ex <50 MeV and at
Next, we focus on the angular distribution of the 12C(y,np)reaction at E, <50 MeV and at E,=(135120) MeV [23]. With three detector settings a very large part of the relevant phase space is covered (see Fig. 6, left). The dominant back-to-back emission of neutron and proton is clearly observed. In order to obtain the angular distribution shown in the right of Fig. 6 the counts have been summed as indicated by the small white rectangles along the ridge. The thin solid (dot-dashed) line indicates the acceptance-integrated experimental cross section for photodisintegration of the deuteron with (without) considering Fermi motion of the np pair within the carbon nucleus. The failure to describe the data is a clear indication of medium modification of the pair and of existence of additional processes beyond the simple quasideuteron picture. Proper calculations [22] including T -
and p meson exchange currents can account for shape and magnitude of the data.
150<Ey<250MeV
(,;)'He q0.75 n
a
1
Li(y,np)He
6 ~(55.3) i '11 (51.4)
150<Ey<250 MeV
1 4
missing m o m e n t u m [MeV/cl
Figure 7. Missing momentum distributions in 6,7Liand 4He.
Figure 8. Experimental recoil momentum distributions from the 3 choices of kinematics in the photon energy range 300< E, <400 MeV normalised by the prediction of the 2N model from ref. [6].
The missing momentum distributions give a clear evidence (within the limitation of PWIA) on internal momenta of the pair before interaction. Fig. 7 shows the distributions for the (y,np) reaction on the two isotopes 6>7Liand 4He. For the ground state transitions the emission from the d and t cluster is assumed, whereas he notation excited state indicates that at least one of the two nucleons comes from the inner s-shell. However, the comparison with the 4He missing momentum distribution suggests that there is no shell mixing at all. A connection between the centroids and the binding energy can easily be seen in Fig. 7. The numbers in brackets indicate the integrated strengths and we find another indication of failure of the simple quasideuteron picture in the similarity of those two numbers for the ground state transition which is in contrast to the number of available pairs. This is another hint that the quantum number of the pair plays an important role in the photon absorption process. Fig. 8 shows the momentum distribution for the 12C(y,np)reaction normalised to the prediction of the 2N model of [ 6 ] . Presently the rise a high pair momenta is interpreted as a further hint towards NN correlations. A prediction (solid line) of correlated wave functions in LOA including tensor correlations by Sarra and Orlandini [9] reproduces the trend correctly. The increase at high recoil momentum can explained due to nucleonnucleon correlations involving one ejected nucleon and one which is remaining within the ( A - 2) residual nucleus. As FSI can not be excluded in this broad data sample it will be important to show the recoil momentum distribution for individual states.
4. HIGH RESOLUTION TWO-NUCLEON EMISSION EXPERIMENTS
Despite the new insights obtained with the experiments described above, it turned out that the averaging over all states of a single shell model orbital does not pin down the reaction models sufficiently as ambiguities of the model with respect to nuclear structure or reaction amplitude may not be resolved. Therefore, we presently aim for two-nucleon emission experiments which resolve final states. In general only a few partial waves contribute to a transition to a single final state. The cross sections are predicted to exhibit distinctive distributions depending on the spin and isospin quantum numbers of the final states. The nucleus 160is the best candidate from the nuclear structure point of view both from experimental and from theoretical considerations. High resolution (e,elpp) experiments and the search for short range correlations of scalar type are reported elsewhere [17]. As shown in Fig. 9 the two model calculations by Giusti and Ryckebusch are quite good but do not give sufficiently good results which are needed to select the best potential for the NN correlations. The np channel must be investigated when searching for the tensor type correlations as they are not present in the pp channel. Despite the strong contribu-
l^(T=OI; Ex=3.9S MeV i.
z.
0.2
e
-0.1 -41 . .IW
missing memmm/McV/c
Figure 9. Missing energy spectrum for the 160(e,e'pp)14Creaction and the missing momentum distribution for the ground state transition (from Ref. [17]).
-
-
0 1M 200 3W 4W .1W mlasing momenNm (MeV1
0 IW ZW 3W missing momentum (MeV)
4W
Figure 10. Predicted momentum distributions and nucleon polarisations for the four lowest states in the 160(e,e'np)14Nreaction [22].
tions of meson exchange currents in the np channel there is a very specific sensitivity to N N correlations as predicted by J. Ryckebusch [22]. The final states of the four sets in
Fig. 10 correspond to the first four states of Table 1. The spins of the states are explained by the dominant configurations of pl12 and p3/2 removal; which in turn fixes the angular momentum and thus the partial waves. The different contributions are reflected in the shapes of the missing momentum distributions. The NN correlations (Fig. 10, dashed lines) can make up to 50% of the cross sections; they affect also the shapes. First (e,elnp) experiments on 3He and 160are in preparation.
Table 1 Properties of low-lying states in 14N. Coefficients of fractional pared to measured cross sections [24]. Ex/MeV J;. T configuration L cfp 0.0 ( ~ 1 / 2 ) - ~ 2 0.21 1+,0 0 -0.19 2.3 (p1/2r2 O+, 1 3.95 1+,0 (pl12)1(p312)1 0 -0.19 7.03 2+, 0 ( p l ~ 2 ) 1 ( p 3 1 2 ) 12 0.28 11.05 ( P ~ / z ) - ~ 2 0.33 3+,0
parentage (cfp) are com-
3 [-$I 1.4 f 0.5 -
*
1.7 0.5 1.6 f 0.4 -
Table 1 and Fig. 11 present results from a low energy, high resolution 160(y,np)14N experiment [24] performed at E,=72 MeV at MAXlab in Lund. Unfortunately, no missing momentum distribution could be extracted in this experiment in order to verify the angular momentum transfer. However, the obtained cross sections give a nice example of the sensitivity of the photoabsorption processes t o nuclear structure. Despite the same coefficient of fractional parentage predicted by simple shell model arguments the cross section for the first excited state is less than 5% of its neighbours. This state has isospin T = l and is the isobaric analogue state to the ground states of 14C and 140. In preparation of the new high resolution experiments also the 160(y,pp)14Creaction will be investigated in order to complement, together with the (e,elpp) data of ref. [17], the fourfold data set of real and virtual photoabsorption each with pp and np emission from 160for a combined analysis. Employing solid state HP-Ge detectors with large solid angles (up to 0.7 sr) and of 25 mm thickness, a pilot run of 5 hours clearly showed the feasibility. In Fig. 12 the ground state transition is well separated from levels at Ex ~6 MeV; as the resolution is mainly governed by the tagger resolution which will be improved by a "tagger microscope", also a separation of the excited states is expected in future (y,pp) experiments. 5 . Summary
The present status in search of NN correlations in nuclei has been reported. The new generation of high resolution experiments on 160with real and virtual photons will permit a stringent state-by-state comparison of measured momentum distributions with various model calculations. It is expected that the correlation functions from different potentials
5
10
15
20
25
30
Misslng Energy (MeV)
35
40
excitation energy / MeV
Figure 11. High resolution 160(r,np)14N Figure 12. 5 hour pilot run for the high resolution 160(y,pp)14C experiment [8]. experiment [24]. can be selected and an understanding of the in-medium properties of nucleon pairs will be gained. Acknowledgement
I would like to acknowledge useful discussions with J. Ryckebusch and H. Miither. I am indebted to all members of the PiP/TOF group for the strong support of our common interest.
REFERENCES 1. 2. 3. 4.
P. Grabmayr, Prog. Part. Nucl. Physics 29 (1992) 251 P.K.A. de Witt Huberts, J . Phys. G, Nucl. Part. Phys 16 (1990) 507 P. Ring and P. Schuck, "The Nuclear Many-Body Problem", Springer, 1980 H. Miither, Proc. 4thWorkshop on "Electromagnetically Induced Two-Hadron Emission", Granada, May 26-29,1999; eds. P. Grabmayr and A. Lallena; p. 1 5. W. Dickhoff, ibid, p. 226; A. Fabroccini, ibid, p. 206 6. D. Watts et al., ibid, p. 182; D. Watts et al., submitted to Phys. Rev. 7. T. Hehl, et al., ibid, p. 70 8. P. Grabmayr et al., ibid, p. 331 9. G. Orlandini and R. Sarra, znd Workshop on "Electromagnetically Induced TwoHadron Emission", Gent, May 1997, p. 1 10. K. Gottfried, Nucl. Phys. 5 (1958) 557 11. I.J.D. MacGregor et al., Nucl. Instr. Methods A382 (1996) 479 12. P. Grabmayr et al., Nucl. Instr. Methods A402 (1998) 85 13. K.I. Blomqvist et al., Nucl. Inst. and Meth. A403 (1998) 263 14. H. Herminghaus et al., Proc. Lin. Accelerator Conference, Albuquerque, N M 1990 15. I. Anthony et al., Nucl. Instr. Methods A301 (1991) 230; S.J. Hall et al., Nucl. Instr. Methods A368 (1996) 698
16. N.Voegler and J. Friedrich, Nucl. Instr. and Meth. 198 (1982) 293 17. G. Rosner et al., contrib. to Int. School on Nuclear Physics, Erice, 1999, Prog. Part. Nucl. Phys. 44 (2 000) 18. T. Lamparter et. al., Z. Phys. A355 (1996) 1 19. P. Grabmayr, in "correlations and clustering phenomena in subatomic physics", ed. O.Scholten, NATO AS1 B (1997) 53 20. R.C. Carrasco et al., Nucl. Phys. A570 (1994) 701 21. C. Giusti, zbzd, p. 148; Nucl. Phys. A571 (1994) 694 22. J. Ryckebusch, ibid, p. 19; Phys. Lett. B383 (1996) 1; private comm. 23. T.T.-H. Yau et al., Eur. Phys. J. A1 (1998) 241 24. L. Isaksson et. al., Phys. Rev. Lett. 83 (1999) 3146
Hadron and Nuclear Physics with Electromagnetic Probes K. Maruyama and H. Okuno (Editors) 2000 Elsevier Science B.V. All rights reserved.
"
Quark substructure and isobar effects on deuteron form-factors "Center for Theoretical Physics and Laboratory for Nuclear Science Massachusetts Institute of Technology Cambridge, MA 02139 Elastic ed scattering, with deuteron polarization, up to high momentum transfer provides detailed information on the deuteron wave function. This determines the range dependence of the orbital and spin components of the one- and two-body currents, restricting contributions of isobar and meson-exchange currents and of quark/gluon degrees of freedom, as well as the nucleon component. The R-matrix boundary condition model combines all these effects, predicting nucleon-nucleon reactions and the deuteron formfactors simultaneously. A brief description of the model is followed by a comparison of its results with data, emphasizing the restrictions placed on the model by ed elastic form-factors. 1. INTRODUCTION
There is now elastic electron-deuteron scattering data determining the electric forml ] , magnetic form-factor B(q2) up to 2.8 ( G e V 1 ~ ) ~ [ 2 ] , factor A(q2) up to 6 ( G e V l ~ ) ~ [the and the tensor-polarization form-factor t2,3(q2)up to 1.8 ( G e V l ~ ) ~ [ 3These ] . data restrict orbital and spin components of the deuteron wave function at scales as small as 0.2 fm, at which distance quark degrees-of-freedom (d.o.f.),isobar components and meson-exchange currents all have a significant role. The R-matrix boundary condition method[4,5] provides a hybrid quark/gluon and hadron model. incorporating all the above contributions. Only a few of the parameters are not predetermined by data independent of the nucleon-nucleon (NN) interaction and symmetry requirements. The remaining few are almost all determined by NN scattering data. Essentially one parameter is free to determine the behavior of the three independent elastic electron-deuteron form-factors (edff) over the large range of momentum-transfers, q. and LILI(~D~)in the deuteron, This parameter determines the relative amount of ALI(~D~) which profoundly affects the q dependence of the spin and convective currents[6]. The N N scattering is not sensitive to the ratio, but only to the sum. Following a review of the R-matrix method and its application to the NN system, three specific models for the I = 0, JP = I+ sector, of different levels of completeness, will be "This work is supported in part by funds provided by the U.S. Department of Energy (D.O.E.) under cooperative research agreement #DF-FC02-94ER40818 and in part by KEK funds for Foreign Visiting Scientists. Present address: Theory Group, KEK-Tanashi branch, Tokyo
compared with the N N scattering and edff data. From these one can extrapolate to a model that represents all the data. 2. THE R-MATRIX BOUNDARY CONDITION MODEL
At high momentum-transfer (short range) the running coupling constant of QCD is small, permitting a perturbative description in terms of current quarks and gluons (asymptotic freedom). At low momentum-transfer (long range) nonperturbative effects produce clustering into color singlet hadrons (confinement). The transition between these extremes has been shown to occur over a small range of the running coupling constant[7], and therefore over a short distance. The R-matrix method[8] is well suited to this situation in which two regions, each well represented by a different approximate Hamiltonian, have their wave functions connected by a boundary condition at the separating surface. For the QCD application, in which confinement requires the quark wave function to be small at the transition boundary, the suitable form of the R-matrix equation is[4,5]
with
in which $, is the exterior wave function for hadron-pair channel a , W is the total energy, the poles Wi are the energies of a complete set of internal states vanishing at ro, and the residues ,ohp are given by absolute squares of the internal wave function derivatives at the boundaries. Thus the fOp(W)are meromorphic functions with real poles of positive residue. The residues can be expressed as
where the fractional parentage coefficients are geometric coefficients expressed in terms of Clebsch-Gordon coefficients of the spin/flavor/color space of the given quark configuration. Therefore only the f,Oo, representing the effective constant of distant poles and the pole at infinity, are free parameters. The hadrons at r > ro interact via hadron exchange potentials, as given by known hadron masses and coupling constants fixed by independent experiments or symmetry conditions. The separation radius, ro, must be within the range of asymptotic freedom (5 0.85 of the equilibrium radius of the interior bag model) because of the sensitivity of p& to the derivatives at ro of the internal wave function. Eq. (1) also requires that ro satisfy $ 2 ( r 0 , Wz(rO)) = 0. Using the low energy data, the latter condition fixes the value of ro with some precision, giving a value consistent with the first condition for the Cloudy bag model, but not for the MIT bag mode1[4,5], ruling out the latter for the multi-hadron domain. The Cloudy bag model determines ro = 1.05 fm. The model then gives a good detailed fit to NN data for TLab5 0.8 GeV[9] and also is consistent with some evidence for
the lowest I = 1 exotic resonance, J P = 0+[10] at 2.70 GeV. These resonances, produced correspond to multi-quark configurations other than the minimal near the f-poles at Wi qq and q3 configurations. The lowest in the NN system is the I = 0, JP = 1+ at 2.63 GeV. 3. DEUTERON PREDICTIONS 3.1. Interior wave function The first f-pole in the EN system, corresponding to the [q(ls;)I6 configuration, is in the I = 0, J P = 1+ state, 0.76 GeV above the deuteron mass. As the width of the exotic resonance is only 0.03 GeV, the fap are nearly constant. It has been shown[ll] that the interior wave function vanishes for constant f , and the actual probability of being in the interior has been estimated to be 5 0.004[6]. This implies a large "hole" in the deuteron wave function, which has long been noted as a property of the edff and is also embodied in the effective "hard core" of NN scattering. In our model this apparent repulsion arises from the rapid change in the effective d.0.f. at ro. However at the energy W, there is "matching" of the interior and exterior wave functions, resulting in a substantial interior probability and a resonance. The fact that ro is 2-3 times larger than the cores of repulsive core models is compensated by the discontinuous increase of wave functions at ro, so that the experimental effective range parameters can be produced by both types of core effect. However the models differ at higher NN scattering energies for large q edff. 3.2. Exterior wave function In the I = 0, JP = 1+ system, the NN (3S1)and NN (3D1) states are coupled to each other and to isobar channels by meson exchange potentials as well as the f-matrix. AA(3Dl) Because of their low threshold mass and strong tensor coupling, the and AA(3D1) channels are most important and are included in all our models. NN (3Sl) and AA (3S1)states have nonvanishing 6 with the [q(ls2)l6quark configuration, channel also has a low threshold, contributing to the f-pole residue. The NN*(1440)(~s~) and with its large width is of next importance to the deuteron. But over the energy range which includes the first exotics. other channels are also of some importance and the NSll(1535) ( l PI),NSll (1650)( l Pl) and (1620)(3 PI) are included in our recent work. These channels modify the best choice of f$ for the lower threshold channels, but have negligible components in the deuteron. 3.3. Determining the f 0 for three models The f i p are sharply restricted by fitting the N N scattering data, which in the deuteron 6 and ~ 7 ( ~ D 1 ) sector requires a fit to the np(3S1 -3 Dl) phase parameters 6 and ~7(~S1), and €1 for TLab5 0.8 GeV. For the models C1 and D1[12],which have only N N and AA channels, the lower energy behavior of the 6's and €1 determine the NN sector fap, while the energy dependence for 0.4 GeV < TLab< 0.8 GeV fixes those coupling the NN and LIA(~S~)sectors, and the NN to the sum of AA(3D1) and AA('D1) sectors. The last two have the same threshold behavior, affecting the elastic scattering in the same way. Only detailed A-production data could separate them, so the ratio is free to adjust to the edff. The magnet,ic form factor q-dependence is very sensitive to the ratio because of the opposite spin and convection currents of these states[6].
The model C' did not consider quark configurations[l3]. Without f-poles the best choice of separation radius was ro = 0.74 fm. Model D' included the lowest f-pole with ro = 1.05 fm as discussed above. As shown previously[l2], C' and D' give equivalent fits to the 6's while case D' is a better fit to €1. 12.5
-4
5
10.0
M
7.5
*
5.0
IW
2.5
-5
0.0
--
0
20
-10
8
-
B
m
z
When the other isobar channels are included, the NN*(1440), because of its larger width, modi-3 fies the energy dependence and in6 creases the inelasticity. This reduces the required coupling to the AA channels. This model E results in a better fit to 51,6(3D1)and to , the 7j's (Fig. 1).
0
-20
w
-20
Fw
-30 1 00
1.00
0 98
0.98
Figure 1. The phase parameters for
3 0 96
0.96
2 I = 0, J~ = 1+, n p scattering. Model E (solid curues); SAID SPOO
V]
D
60.94
0.94
0.92
0.92
0 90
o
250
500
TL ( M e V )
750
o
250
500
750
O
phase parameters (dashed curves); Bugg 1990-1991 phase parameters (squares).
TL ( M e V )
3.4. The EDFF predictions In previous work[l2], the edff for models C' and D' were calculated with the nonrelativistic, coupled channel impulse approximation (IA) and the meson-exchange current (MEC) terms of i7, p, and w , "pair" corrections and the p r y term to first relativistic order. For the IA the isobar form factors are assumed proportional to the nucleon electromagnetic form-factors. In all cases the MEC corrections to the isobar channels are neglected. Both Hohler et a1.[14] (HO) and Gari-Kriimpelmann[l5] (GK) nucleon form factors were used. The results (Figs. 3-6 of [12]) are seen to be a good to A(q2) for the HO choice and to B(q2) for the GK choice. This is not inconsistent as A(q2) is dominated by the nucleon electric form-factor and B(q2) by the nucleon magnetic form-factor. The t20(q2) predictions were consistent with the very low q experimental results available at the time. Here we present, versus the extended data range of the edff, the results of models C', D' and E where the first order relativistic correction has been added to the impulse approximation and the second order relativistic corrections have been included in the MEC[16,17]. For model E the ratio of AA(7D1) to AA(3D1) coupling to the N N sector was guided by the C' and D' model ratios. It has not yet been optimized to the data.
Figure 2. A ( q 2 ) : Data points are de- Figure 3. t z o ( q ) :Data points described in scribed in Ref.[l]. Model C' ( H O ) (solid Ref. [3]. Model C' (solid line); model D' line); model (2' ( G K ) (dash-dash);model (dash-dot); model E (long dashes). The D' ( H O ) (dash-dot);rmdel D' ( G K ) (dot- dependence on nucleon emff ( H O or G K ) dot); model E ( H O ) (long dashes); model is negligible. E ( G K ) (dash-dot-dot). Also the balance of AA and N N * ( 1 4 4 0 ) coupling to N N and the value of the N*(1440) magnetic moment, unknown from independent data, have not been varied. Table 1 shows the results of model E for the static properties of the deuteron. For models C' and D' the results are as stated in [12]except for a small relativistic change in Qdeut
.
Table 1 Static deuteron properties of Model E B E ( M e V ) PD (%) PAS(%) Pas(%) PA,(%) Model E 2.2247 5.21 .006 3.24 1.79 EX^ e 2.2246
PN*(%)a Q(f m2) 0.71 .273 .286
PD (p?)
360 .857
"The higher mass channels have neglibible probability.
A ( q 2 ) is shown in Fig. 2. It is seen that model E with either choice of nucleon form) ~ , is only large enough for factors fits the data reasonably well for q2 < 2.5 ( G ~ V / C but larger q2 when the G K choice is made. Models C' and D' on the other hand are better with the HO choice for q2 < 2.5 ( G e V / c 2 ) ,but at 6 ( G e V / c 2 )only model C with G K is large enough. t20(q2) is shown in Fig. 3. For the momentum transfer range there is negligible sensitivity to the choice of nucleon form-factors, as these cancel in the ratio of quadrupole to
monopole electric amplitudes which dominates t z o The result is however very sensitive to the model used. The simple constant f-matrix model, C', gives a good fit over the whole range of q. Model D' puts the maximum of tzo at much too small a momentum transfer. This is related to the large amplitude ofthe L = 2, AA states in this model. Model E, with intermediate A A components, has the maximum of tzo between that of models C' and Dl. The B(q2) for models C' and D' is similar to that of [12] with minima a t slightly smaller q. However the minimum of B(q2) for model E is at much too small a value (q2 = 1.3 (GeV/c)'). 4. C O N C L U S I O N S
The R-matrix boundary condition model E, with all the relevant isobar channels, reproduces the np(3Sl -3 Dl) scattering phases well up to T L 5 ~0.8~GeV. It also reproduces very well the static properties of the deuteron and A(q2) for q2 _< 6 ( G ~ V / C ) ~It. does not fit t20(q2)as well as the simpler model C' (although it is better than model D'), and ) too small a value. To correct the full model (E) has the first minimum of ~ ( q at~ much for B(q2),the ratio of the AA(7D1) to A A ( ~ D couplings ~) to the NN sector needs to be varied. That, and perhaps a further substitution of N*(1440) coupling for A A coupling to the NN sector may also correct the fit to the maximum of t20(q2). REFERENCES 1. L.C. Alexa et al., Phys. Rev. Lett. 82 (1999) 1374, and references therein. 2. R.G. Arnold et al., Phys. Rev. Lett. 58 (1987) 1723, and references therein. 3. D. Abbot et al., arXive:nucl-ex/0001006, and references therein. 4. P. LaFrance and E.L. Lomon, Phys. Rev. D34 (1986) 1341. 5. P. Gonzdez, P. LaFrance and E.L. Lomon, Phys. Rev. D35 (1987) 2142. 6. W.P. Sitarski, P.G. Blunden and E.L. Lomon, Phys. Rev. C36 (1987) 2479. 7. M. Creutz, Phys. Rev. Lett. 45 (1980) 313. 8. E.P. Wigner and L. Eisenbud, Phys. Rev. 72 (1947) 29. 9. P. LaFrance, E.L. Lomon and M. Aw, nucl-th/9306026. 10. J. Ball et al., Phys. Lett. 320 (1994) 206. 11. H. Feshbach and E.L. Lomon, Ann. Phys. (NY) 29 (1964) 19. 12. P.G. Blunden, W.R. Greenberg and E.L. Lomon, Phys. Rev. C40 (1989) 1541. 13. E.L. Lomon and H. Feshbach, Ann. Phys. (NY) 48 (1968) 94. 14. G. Hohler, E. Piertarinen and I. Sabba-Stefanescu, Nucl. Phys. B144 (1976) 505. 15. M. Gari and W. Kriimpelmann, Phys. Lett. B173 (1986) 10. 16. P.G. Blunden, private communication. The formulas of Blunden and Ian Towner are used here. 17. H. Arenhovel, F. Ritz and T . Wilbois, to be published Phys. Rev. C. This work includes boost corrections not included in the work of P.G. Blunden and I. Towner.
VI. NUCLEON STRUCTURE STUDIED BY HIGH-ENERGY ELECTRONS
ThisPageIntolyLfBk
ThisPageIntolyLfBk This Page Intentionally Left Blank
Hadron and Nuclear Physics with Electromagnetic Probes K. Maruyama and H. Okuno (Editors) 2000 Elsevier Science B.V. All rights reserved.
"
The proton and the photon, who is probing whom in electroproduction? Aharon Levy* School of Physics and Astronomy Raymond and Beverly Sackler Faculty of Exact Sciences Tel Aviv University, Tel Aviv, Israel The latest results on the structure of the proton and the photon as seen at HERA are reviewed while discussing the question posed in the title of the talk. 1. INTRODUCTION
The HERA collider, where 27.5 GeV electrons collide with 920 GeV protons, is considered a natural extension of Rutherford's experiment and the process of deep inelastic ep scattering (DIS) is interpreted as a reaction in which a virtual photon, radiated by the incoming electron, probes the structure of the proton. In this talk I would like to discuss this interpretation and ask the question of who is probing whom [I]. The structure of the talk will be the following: it will start with posing the problem, after which our knowledge about the structure of the proton as seen at HERA [2] will be presented followed by a description of our present understanding of the structure of the photon as seen at HERA and at LEP [3,4]. Next, an answer to the question posed in the title will be suggested and the talk will be concluded by some remarks about the nature of the interaction between the virtual photon and the proton [5]. 2. THE QUESTION
- WHO IS PROBING WHOM?
2.1. The process of DIS The process of DIS is usually represented by the diagram shown in figure 1. If the lepton does not change its identity during the scattering process, the reaction is labeled neutral current (NC), as either a virtual photon or a Z0 boson can be exchanged. When the identity of the lepton changes in the process, the reaction is called charged current (CC) and a charged Wi boson is exchanged. During this talk we will discuss only NC processes. Using the four vectors as indicated in the figure, one can define the usual DIS variables: Q2 = -q2, the 'virtuality' of the exchanged boson, x = Q 2 / ( 2 P . q), the fraction of the proton momentum carried by the interacting parton, y = (P . q ) / ( P 9k), the inelasticity, and W2 = (q + P)2,the boson-proton center of mass energy squared. *This work was partially supported by the German-Israel Foundation(GIF), by the U.S.-Israel Binational Foundation (BSF) and by the Israel Science Foundation (ISF). The financial support of my visit to Japan by JSPS is highly appreciated.
proton remnant
Figure 1. A diagram describing the process of deep inelastic scattering (DIS). The four vector of the incoming and outgoing leptons are k and k', that of the exchanged boson is q, and that of the incoming proton is P. The four momentum of the struck quark is xP.
The interpretation of the diagram describing a NC event is the following. The electron beam is a source of photons with virtuality Q2. These virtual photons 'look' at the proton. Any 'observed' structure belongs to the proton. How can we be sure that we are indeed measuring the structure of the proton? Virtual photons have no structure. Is that always true? We know that real photons have structure; we even measure the photon structure function Fz [4]. Let us discuss this point further in the next subsections. 2.2. The fluctuating photon How is it possible that the photon, which is the gauge particle mediating the electromagnetic interactions. has a hadronic structure? Ioffe's argument [6]: the photon can fluctuate into qq pairs just like it fluctuates into e+e- pairs (see figure 2). If the fluctuation time. defined in the proton rest frame as tf Y (2E,)/miq, is much larger than the interaction time, ttnt E r,, the photon builds up structure in the interaction. Here, E, is the energy of the fluctuating photon, m,,- is the mass into which it fluctuates, and r, is the radius of the proton. The hadronic structure of the photon, built during the
remnant photon SYrrent jet
e
Figure 2. Fluctuation of a photon into (a) an e+e- pair, (b) a qq pair.
Figure 3. A diagram describing a DIS process on a quasi-real photon using the reaction e+e- -+ e+ePX.
interaction, can be studied by measuring the photon structure function F; in a DIS type of experiment where a quasi-real photon is probed by a virtual photon, both of which are emitted in e+e- collisions, as described in figure 3. This diagram is very similar to that in DIS on a proton target (figure 1).
2.3. Structure of virtual photons? Does a virtual photon also fluctuate and acquire a hadronic structure? The fluctuation time of a photon with virtuality Q2 is given by tf E (2E,)/(rniq + Q2),and thus at very high Q2 one does not expect the condition tf >> ttnt to hold. However at very large photon energies. or at very low x, the fluctuation time is independent of Q2: tf N 1/(2rnpx).where m, is the proton mass, and thus even highly virtual photons can acquire structure. For instance, at HERA presently W -- 200 - 300 GeV. and since x = Q2/(Q2 W2), x can be as low as 0.01 even for Q2 = 1000 GeV2. In this case, the fluctuation time will be very large compared to the interaction time and the highly virtual photon will acquire a hadronic structure. How do we interprate the DIS diagram of figure 1 in this case? Whose structure do we measure? Do we measure the structure of the proton, from the viewpoint of the proton infinite momentum frame, or do we measure the structure of the virtual photon, from the proton rest frame view? UTho is probing whom? When asked this question. Bjorken answered [7] that physics can not be frame dependent and therefore it doesn't matter: we can say that we measure the structure of the proton or we can say that we study the structure of the virtual photon. I will try to convince you at the end of my talk that this answer makes sense.
+
3. THE STRUCTURE OF THE PROTON
In this section we will refrain from discussing the question posed above and will accept the interpretation of measuring the structure of the proton via the DIS diagram in figure 1. We present below information about the structure of the proton as seen from the DIS studies at HERA. 3.1. HERA With the advent of the HERA ep collider the kinematic plane of x-Q2 has been extended by 2 orders of magnitude in both variables from the existing fixed target DIS experiments, as depicted in figure 4. The DIS cross section for ep --, can be written (for Q2 << M i ) as [8],
ex
In the quark-parton model (QPM), the proton structure function Fz is only a function of x and can be expressed as a sum of parton densities, and is related to Fl through the Callan-Gross relation [9], F2(x) =
eSxqi(x) = 2xF1,
where e, is the electric charge of quark i and the index i runs over all the quark flavours. Note that in Quantum Chromodynamics (QCD), the Callan-Gross relation is violated, and the structure function is a function of x and Q2,
where the longitudinal structure function FL contributes in an important way only at large y.
ZEUS 1996+97
/
ZEUS 1994
'-i ZEIJS SVX 1995 [~I ZI<11SRPC 1995 ZEUS RIT 1997
Figure 4. The x-Q2 kinematic plane of some of the fixed target and of the HERA collider DIS experiments.
EMS
:1
' RCDhlS
- CCFR SIAC
17;NMC
The motivation for measuring F2(x,Q2) can be summarized as follows: (a) test the validity of perturbative QCD (PQCD) calculations, (b) decompose the proton into quarks and gluons, and (c) search for proton substructure. 3.2. QCD evolution - scaling violation Quarks radiate gluons; gluons split and produce more gluons at low x and also qq pairs at low x. This QCD evolution chain is usually described in leading order by splitting functions Ptj, as shown in figure 5. This procedure leads to scaling violation in the
g
Figure 5. Splitting functions Pijin leading order, describing the splitting of parton j
following way: there is an increase of F2 with Q2 at low x and a decrease a t high x. Scaling holds a t about x=0.1. The data follows this prediction of QCD, as can be seen in figure 6. 3.3. Overview of F2
The fixed target experiments provided information at relatively high x and thus enabled the study of the behaviour of valence quarks. The first HERA results showed a surprisingly
HI 97 preliminary
HI 97 prellmlnary k-
NME
-QCDPlt pr.4
i L
I*
18
Q&V2
to
Figure 6. Comparison of the scaling violation behaviour of F2with the results of a nextto-leading order DGLAP evolution equation.
strong rise of F2 as x decreases. An example of such a rise is given in figure 7 where F2 increases as x decreases, for a fixed value of Q2 = 15 GeV2. This increase is the result
4
"
NUC
1 Figure 7. The proton structure function F2,as function of x, at Q2 = 15 GeV2, for HERA and some fixed target data.
of the rising gluon density at low x. Note the good agreement between both HERA experiments, HI and ZEUS, and also between HERA and the fixed target data. 3.4. Evolution of F2 The measurements of F2 as function of x and Q2 can be used to obtain information about the parton densities in the proton. This is done by using the pQCD DGLAP evolution equations. One can not calculate everything from first principles but needs as input from the experiment the parton densities at a scale Qi, usually taken as a few GeV2, above which pQCD is believed to be applicable.
There are several groups which perform QCD fits, the most notable are MRST [lo] and CTEQ [ll]. They parameterize the x dependence of the parton densities at Qi in the form.
The free parameters like 771 and
772
are adjusted to fit the data for Q2 > Qi. An example
Figure 8. F2 as function of x, for fixed Q2 values (in GeV2) as indicated in the figure, for the HERA '94 data together with some fixed target data. The curves are the result of a NLO QCD fit.
-
ZEUS 94 HVX ZEUS 94
svx
z
H? 94 (0' in pore". r h a r different)
of such an evolution study can be seen in figure 8 where the F2 data are presented as function of x for fixed Q2 values. The increase of F2 with decreasing x is seen over the whole range of measured Q2 values. The pQCD fits give a good description of the data down to surprisingly low Q2 values. The resulting parton densities from the MRST parameterization at Q2 = 20 GeV2 are shown in figure 9. One sees the dominance of the u valence quark at high x and the sharp rise of the sea quarks at low x. In particular, the gluon density at low x rises very sharply and has a value of more than 20 gluons per unit of rapidity at x -- lop4. In figure 10 one sees the extracted gluon density by the H1 experiment [12] at three different Q2 values. The density of the gluons at a given low z increases strongly with Q2. 3.5. Rise of F2 with decreasing x The rate of the rise of F2with decreasing x is Q2 dependent. This can be clearly seen in figure 11 where F 2 is plotted as a function of x for three Q2 values. The rate of rise decreases as Q2 gets smaller. What can we say about the rate of rise? To what can one
MRST partons
Q' = 20 G~V'
-
HI prel.
X
Figure 9. Parton density distributions, as function of x, of the MRST global QCD fit, at a scale of Q2 = 20 GeV2.
Figure 10. The gluon density distribution, as function of x, at Q2 = 5, 20 and 200 GeV2.
compare it? The proton structure function F2 is related to the total y*p cross section gtot(-Y*p),
where the approximate sign holds for low x. Since we have a better feeling for the behaviour of the total cross section with energy, we plot in figure 12 the F2 data converted to atot(y*p)as function of W2 for fixed values of Q2. For comparison we plot also the total yp cross section. One sees that the shallow W behaviour of the total yp cross section changes to a steeper behaviour as Q2 increases. The curves are the results of the ALLM97 [13] parameterization (see below) which gives a good description of the transition seen in the data. 3.6. The transition region The data presented above show a clear change of the W dependence with Q'. At Q2=0 the processes are dominantly non-perturbative and the resulting reactions are usually named as 'soft' physics. This domain is well described in the Regge picture. As Q2 increases, the exchanged photon is expected to shrink and one expects pQCD to take
I
Figure 11. The proton structure function F2. as function of x, for three Q2 values. The higher the Q2, the steeper the distributions.
' ' """I
'
" " I
'
' """I
'
'
"""I
'
"""'I
' ' 11
Figure 12. The y*p total cross section cr(y*p) as function of W 2 for fixed values of Q2, including the total photoproduction cross section (Q2=0). The curves are the results of the ALLM97 parameterization.
over. The reactions are said to be 'hard'. Where does the transition from soft to hard physics take place? Is it a smooth or abrupt one? In the following we describe two parameterizations, one fully based on the Regge picture while the other combines the Regge approach with a QCD motivated one. 3.7. Example of two parameterizations Donnachie and Landshoff (DL) [14] succeeded to describe all existing hadron-proton total cross section data in a simple Regge picture by using a combination of a Pomeron and a Reggeon exchange, the former rising slowly while the latter decreasing with energy,
where s is the square of the total center of mass energy. The two numerical parameter, related to the intercepts a ( 0 ) of the Pomeron and Reggeon trajectories, respectively, are the result of fitting this simple expression to all available data, some of which are shown in the first two plots in figure 13. These parameters give also a good description of the total yp cross section data which were not used in the fit and are also shown on the right hand side of the figure. Donnachie and Landshoff wanted to extend this picture also to virtual photons [15] (for Q2 < l o GeV2), keeping the power of W2. which is related to
pp
21 7Oim'"+S6 OEs-""
41 ( ~ e v )
\/a
WIGeVl
(GeV)
Figure 13. The total cross section data of pp, pp, ~ * and p yp as function of the center of mass energy. The different lines are the results of parameterizations to these data (see text).
the Pomeron intercept, fixed with Q2. Their motivation was to see what is the expected contribution from non-perturbative physics, or soft physics as we called it above, at higher Q2.
The other example is that of Abramowicz, Levin, Levy, Maor (ALLM) [16], which was updated by Abramowicz and Levy (ALLM97) [13]. This parameterization uses a Regge motivated approach at low x together with a QCD motivated one at high x to parameterize the whole (x, Q2) phase space, fitting all existing F2data. This parameterization uses a so-called interplay of soft and hard physics (see [17]) . The two parameterizations are compared [18] to the low Q2 HERA data together with that of the fixed target E665 experiment in figure 14. Here one sees again how the cross as Q2 increases. section changes from a (W2)0O8 behaviour at very low Q2 to a (W2)0.2-0,4 The simple DL parameterization as implemented by ZEUS (ZEUSREGGE in the figure) fails to describe the data above Q2 1 GeV2. ALLM97 describes the data well in the whole region. DL98 [19], which adds to the soft Pomeron an additional hard Pomeron, can also describe the data, but loses the simplicity of the original DL one. One can quantify the change in the rate of increase by using the parameter A. Since .tot (w2)"(0)-' this implies that F2 x - ~ .The fitted value of X as function of Q2 is shown in figure 15 for the ZEUS [20] (upper) and the H1 [21] (lower) experiments. One sees a clear increase of X with Q2 which cannot be reproduced by the simple Regge picture but needs an approach in which there is the interplay of soft and hard physics [17].
-
-
-
3.8. What have we learned about the structure of the proton? Let us summarize what we have learned so far about the structure of the proton.
The density of partons in the proton increases with decreasing x. a The rate of increase is Q2 dependent; at high Q2 the increase follows the expectations
from the pQCD hard physics while at low Q2 the rate is described by the soft physics behaviour expected by the Regge phenomenology.
-
Though there seems to be a transition in the region of Q2 1-2 GeV2, there is an interplay between the soft and hard physics in both regions.
ZEUS 1997 (Preliminary) A
A
ZEUS+H194-95 0 E665 ZEUS+Hl ZEUS ~ ~ I ' y 9 9 7 ZEUS REGGE 97
... DL
--- DL98 ALLM97
6
0.45
0.35
i
ZEUS+E665 DATA
r
11 4
I
H1 Preliminary
I
t
, , ~ - - i u ~ _ - i L u l i l 1 10 lo2
Q'/ G ~ V '
Figure 14. The y*p total cross section, as function of W 2 ,in bins of Q2 for HERA data and that of E665, compared with some para~neterizations.
Figure 15. The Q2 dependence of the rameter A, determined from a fit of form F2 x-A at fixed Q2 values.
4. THE STRUCTURE OF THE PHOTON
In this part we will describe wha,t is ~resentlyknown about the structure of the photon, both from e+e- experiments as well as from HERA. 4.1. Photon structure from e+eThe hadronic structure function of the photon, F , , was memured in e+e- collisions which can be interpreted as depicted in figure 3. A highly virtual y* with large Q2 probes a quasi-real y with P2z 0. The measurements of F: showed a different behaviour than that of the proton structure function. From the Q2 dependence, shown in figure 16 [4], one sees positive scaling violation for all x. This different behaviour can be understood as coming from an additional split,ting to the ones present in the prot,on case (see figure 5). In the photon case, the photon can split into a qq pa,ir' y -+ qq. The contribution resulting from this splitting, called the 'box diagram', causes positive scaling violation for all x. In addition, and again
Z
['"'I
'
' """I
'
' """I
'
'
"'"'I
'
""T
Figure 16. The photon structure function F,Y, as function of Q2, for average x values as given in the figure. The curves are the expectations of different parameterization~of parton distributions in the photon.
Figure 17. F;, as function of x, for fixed Q2 as given in the figure. The data points have been taken from the numerical tables in [4].
contrary to the proton case, it also causes the photon structure function to be large for high x values, as can be seen in figure 17 where F; is plotted as function of x for fixed Q2 values. From this figure one can also see that there exist very little data in the low x region. 4.2. Photon structure from HERA At HERA, the structure of the photon can be studied by selecting events in which the exchanged photon is quasi-real and the probe is provided by a large transverse momentum parton from the proton. The probed photon can participate in the process in two ways. In one. the interaction takes place before it fluctuates into a qq pair and thus the whole of the photon participates in the interaction. Such a process is called a 'direct' photon interaction. In the other case, the photon first fluctuates into partons and only one of these partons participates in the interaction while the rest continue as the photon remnant. This process is said to be a 'resolved' photon interaction. An example of leading order diagrams describing dijet photoproduction for the two processes is shown in figure 18. If one defines a variable x, as the fraction of the photon momentum taking part in a 1 in the direct case, while x, << 1 in the resolved photon dijet process, we expect x, interaction. These two processes are clearly seen in figure 19 where the x, distribution shows a two peak structure, one coming from the direct photon and the other from the resolved photon interactions [22]. One way of obtaining information about F,Y from HERA is to measure the dijet phoN
ZEUS 1994
i l 0 1 0 2 0 3 0 4 0 5 0 6 0 7 0 8 0.9 1
O0
x OBS
Figure 19. The xGbS distribution, as obtained from photoproduction of dijet events. The shaded area are the expectations of the distribution of this variable from the generation of direct photon events. The dotted vertical line is the border of an operational definition of direct and resolved photon events.
Figure 18. Examples of leading order QCD (a) 'direct' and (b) 'resolved' dijet production diagrams.
toproduction as function of x, and to subtract the contribution coming from the direct photon reactions. This is shown in figure 20, where the measurements are presented at ZEUS 1996/1997 PRELIMINARY
91* F. ~ ~ ~ , I + C . Vf
&w
Fr:->170ev
f
:
C->ZICN
E
Figure 20. Distributions of for different thresholds on the highest transverse energy jet. The open histogram is the prediction of the MC, and the shaded part is the direct photon component of the MC.
Figure 21. Comparison of the photon gluon density, determined from di-jet photoproduction events taken in 1996, with earlier measurements of HI. The curves are the expectations of different parameterizations.
fixed values of the hard scale. which is taken as the highest transverse energy jet [23]. One can go one step further by assuming leading order QCD and Monte Carlo (MC) models to extract the effective parton densities in the photon. An example of the extracted gluon density in the photon [24] is shown in figure 21. The gluon density increases with decreasing x, a similar behaviour to that of the gluon density in the proton. The data have
the potential of differentiating between different parameterization of the parton densities in the photon, as can be seen in the same figure. 4.3. Virtual photons at HERA
One can study the structure of virtual photons in a similar way as described above. In this case, the Q2 of the virtual photon has to be much smaller than the transverse energy squared of the jet, EZ, which provides the hard scale of the probe. Such a study [25] is presented in figure 22, where the dijet cross section is plotted as function of x, for different regions in Q2 and EZ. One sees a clear excess over the expectation of direct photon reactions, indicating that virtual photons also have a resolved part. This fact can
HtRWG(LC)/DR
-i E R W S
.
LC / D G w=O 1
H1 Data
- - - - i E R w G I L c I / D G o=o 2
ZEUS PRELIMINARY 1995
5 8 '" u
:
D
V
e
.
Gi 4
ZEUS 1-6.
20
$
.....
,o
...
?
-g_ 88
Iv W
K, J.Mnder
Hewig $?bin
.o
9
b a "
$ 5 .
2.5
a 5 3
",
2
,., ,.
... O
05
.O
05
,
F-... 05
,
..............
x"!
Figure 22. The differential dijet cross section, as function of xj,"t" for different scales as indicated in the figure. The open histogram is the prediction of the MC and the shaded part is the direct photon component of the MC.
- 1 . 0
0
0.5
1
1.5
2
2.5
3
3.5
4
a*4.5
Figure 23. The ratio of the resolved to direct photon component, as function of the virtuality Q2 (in GeV2) of the photon.
also been seen in figure 23 where the ratio of resolved to direct photon interactions is plotted as function of the virtuality Q2 of the probed photon [26]. One sees that although the ratio decreases with Q2, it remains non-zero even at relatively high Q2 values. 4.4. Virtual photons at LEP The study of the structure of virtual photons in e+e- reactions was dormant for more than 15 years following the measurement done by the PLUTO collaboration [27]. Recently, however. the L3 collaboration at LEP [28]measured the structure function of photons with a virtuality of 3.7 GeV2, using as probes photons with a virtuality of 120 GeV2. In the same experiment. the structure function of real photons was also measured. Both results
a) < Q ~ >= 120 G E V ~
, =o
2-
DATA QPM ...... GRV
-
t)
....
2 X w
2
1-
~3
AGF LRSN
-/&+=??+ +---.+ ........
...
.......
'.'3
,_.-.
,,rr
0
I
0
0.2
0.4
I
I
0.6
0.8
O 1
1
PLUTO
X
b) = 120 G E V ~ , = 3.7 G E V ~
2-
t)
-
2
QPM
21
-.-- GRS
X w
....
F ;
b,
GRS F2
DATA QPM
,
ILa' ..........
0
0
I
I
I
0.2
0.4
0.6
018
X
Figure 24. The effective photon structure function from L3 for (a) a quasi-rea1 photon target and (b) a 3.7 GeV2 photon target. Both photons were probed at a scale of 120 GeV2.
Figure 25. The dependence of the effective photon structure function on the mass P2 of the probed photon,
can be seen in figure 24 and within errors the structure function of the virtual photons is of the same order of magnitude as that of the real ones. The effective structure function is also presented as function of the virtuality of the probed photon P2 in figure 25 and show very little dependence on PZ up to values of 6 GeV2 [27,28].
-
4.5. What have we learned about the structure of the photon? Let us summarize what we have learned so far about the structure of the photon.
At HERA one can see clear signals of the 2-component structure of quasi-real photons, a direct and a resolved part. Virtual photons can also have a resolved part at low x and fluctuate into qq pairs. a
Structure of virtual photons has been seen also at LEP.
5. THE ANSWER
Following the two sections on the structure of the proton and the photon, let us remind ourselves again what our original question was. At low x we have seen that a y* can have structure. Does it still probe the proton in an ep DIS experiment or does one of the partons of the proton probe the structure of the y*?
The answer is just as Bjorken said: at low x it does not matter. Both interpretation are correct. The emphasis is however 'at low x'. At low x the structure functions of the proton and of the photon can be related through Gribov factorization [29]. By measuring one, the other can be obtained from it through a simple relation. This can be seen as follows. Gribov showed [29] that the yy,yp and pp total cross sections can be related by Regge factorization as follows:
This relation can be extended [I] to the case where one photon is real and the other is virtual,
or to the case where both photons are virtual,
Since at low x one has 0 ;;. 9 p 2 ,one gets the following relations between the proton structure function F:, the structure function of a real photon, F2, and that of a virtual photon, F;" :
and
The relation given in equation (10) has been used 1301 to 'produce' F; 'data' from well measured F; data in the region of x <0.01, where the Gribov factorization is expected to hold. The results are plotted in figure 26 together with direct measurements of F;. Since no direct measurements exist in the very low x region for Q2 > 4 GeV2, it is difficult to test the relation. However both data sets have been used for a global QCD leading order and higher order fits [31] to obtain parton distributions in the photon. Clearly there is a , data for such a study. need of more precise direct FY In any case, our answer to the question would be that at low x the virtual photon and the proton probe the structure of each other. In fact, what one probes is the structure of the interaction. At high x, the virtual photon can be assumed to be structureless and it studies the structure of the proton.
I
"
'
!
"
I
l
'
"
1
Figure 26. The photon structure function, as function of x, for fixed Q2 values as indicated in the figure. The full points are direct measurements, and the open triangles are those obtained from F; through the Gribov factorization relation. The full line is the result of a higher order fit and the dashed line is that of a leading order parameterization.
6. DISCUSSION
- T H E STRUCTURE
OF T H E INTERACTION
We concluded in the last section that at low x one studies the structure of the interaction. Let us discuss this point more clearly. We saw that in case of the proton at low x, the density of the partons increases with decreasing x. Where are the partons located? In the proton rest frame, Bjorken x is directly related to the space coordinate of the parton. The distance 1 in the direction of the exchanged photon is given by [6],
Therefore partons with x >0.1 are in the interior of the proton, while all partons with x <0.1 have no direct relation to the structure of the proton. The low x partons describe the properties of the y*p interaction. How can we describe a y*p interaction at low x? It occurs in two steps: first the virtual photon fluctuates into a qtj pair and then the configuration of this pair determines if the interaction is 'soft' or 'hard' [32]. The soft process is the result of a large spatial configuration in which the photon fluctuates into an asymmetric small kT qtj pair. The hard nature of the interaction is obtain when the fluctuation is into a small configuration of a symmetric q q pair with large kT. The two configurations are shown in figure 27. At Q 2 x 0 the asymmetric configuration is dominant and the large color forces produce a hadronic component which interacts with the proton and leads to hadronic non-perturbative soft physics. The symmetric component contributes very little; the high k~ configuration is screened by color transparency (CT). At higher Q2 the contribution of the symmetric small configuration gets bigger. Each one still contributes little because of CT, but the phase space for such configurations increases. Nevertheless, the asymmetric large configuration
p-
Figure 27. Fluctuation of the photon into a qq pair in (a) an asymmetric small kT configuration, (b) a symmetric large kT configuration.
aT>
p-remnant
47
current jet
Figure 28. A DIS diagram, as seen from the point of view of a photon fluctuating into an asymmetric pair, in which the fast quark becomes the current jet.
is also still contributing and thus both soft and hard components are present. Another way to see this interplay is by looking a t the diagram in figure 28. In a simple QPM picture of DIS, the fast quark from the asymmetric configuration becomes the current jet while the slow quark interacts with the proton in a soft process. Thus the DIS process looks in the y*p frame just like the Q2=0 case. This brings the interplay of soft and hard processes.
7. CONCLUSION In DIS experiments at low x one studies the 'structure' of the y'p interaction. In order to study the interior structure of the proton, one needs to measure the high x high Q2 region. This will be done at HERA after the high luminosity upgrade.
I would like to thank Professors K. Maruyama and H. Okuno for organizing a pleasant and lively Symposium. Special thanks are due to Professor K. Tokushuku and his group for being wonderful hosts during my visit in Japan. Finally I would like to thank Professor H. Abramowicz for helpful discussions.
REFERENCES 1. A. Levy, Phys. Lett. B404 (1997) 369. 2. For a recent review see H. Abramowicz and A. Caldwell, Rev. Mod. Phys. 71 (1999) 1275. 3. J. Butterworth, Structure of the photon to appear in the proceedings of Lepton-Photon 99; [hep-ex/9912030]. 4. R. Nisius and references therein, The photon structure from deep inelastic electronphoton scattering, [hep-ex/9912049].
5. A. Levy, The structure of the Troika: proton, photon and Pomeron, as seen at HERA, Proceedings of the XXIth Workshop on the Fundamental Problems of High Energy Physics and Field Theory, (Eds. V. Petrov and I. Filimonova), p. 7, Protvino 1998, [hep-ph/9811462]. 6. B.L. Ioffe, Phys. Lett. B30 (1969) 123; B.L.Ioffe, V.A. Khoze and L.N. Lipatov, Hard Processes, (Korth-Holland, 1984). 7. J.D. Bjorken, in response to the question asked at DIS94, Eilat, February 1994. 8. See e.g. E. Leader and E. Predazzi, An Introduction to Gauge Theories and the New Physics, (Cambridge University Press, 1982). 9. C.G. Callan and D.J. Gross, Phys. Rev. Lett. 22 (1969) 156. 10. A.D. Martin, R.G. Roberts, W.J. Stirling and R.S. Thorne, Europ. Phys. J. C4 (1998) 463. 11. CTEQ Collab., H.L. Lai et al., Phys. Rev. D55 (1997) 1280. 12. M. Klein, Structure Functions in Deep Inelastic Lepton-Nucleon Scattering, to appear in the proceedings of Lepton-Photon 99, [hep-ex/0001059]. 13. H. Abramowicz and A. Levy, The ALLM parameterization of atot(y*p): an update, DESY 97-251, [hep-ph/9712415]. 14. A. Donnachie and P.V. Landshoff, Phys. Lett. B296 (1992) 227. 15. A. Donnachie and P.V. Landshoff, Zeit. Phys. C61 (1994) 139. 16. H. Abramowicz, E. Levin, A. Levy and U. Maor, Phys. Lett. B269 (1991) 465. 17. H. Abramowicz, L. Frankfurt and M. Strikman, Surveys in High Energy Phys. 11 (1997) 51. 18. C. Amelung for the ZEUS Collab., proceedings of DIS99, Nucl. Phys. B(Proc.Suppl.)79 (1999) 176. 19. A. Donnachie and P.V. Landshoff, Phys. Lett. B437 (1998) 408. 20. ZEUS Collab., J. Breitweg et al., Europ. Phys. J. C7 (1999) 609. 21. J. Zacek for the H1 Collab., proceeding of DIS99, Nucl. Phys. B(Proc.Suppl.)79 (1999) 86. 22. ZEUS Collab., M. Derrick et al., Europ. Phys. J. C1 (1998) 109. 23. ZEUS Collab., paper 540 submitted to HEP99, Tampere, July 1999. 24. J. Cvach for the H1 Collab., proceeding of DIS99, Nucl. Phys. B(Proc.Suppl.)79 (1999) 501. 25. H1 Collab., C. Adloff et al., DESY 98-205, to be published in Europ. Phys. J. C, [hep-ex/9812024]. 26. D. Kcira for the ZEUS Collab., proceedings of Photon99, Freiburg, May 1999. 27. PLUTO Collab., Ch. Berger et al., Phys. Lett. B142 (1984) 119. 28. F.C Erne for the L3 Collab., proceedings of Photon99, Freiburg, May 1999. 29. V.N. Gribov, L.Ya. Pomeranchuk, Phys. Rev. Lett. 8 (1962) 343. 30. H. Abramowicz, E. Gurvich and A. Levy, Phys. Lett. B420 (1998) 104. 31. H. Abramowicz, E. Gurvich and A. Levy, Next to leading order parton distributions in the photon from y*y and y*p scattering, proceedings of ICHEP98, p.885, Vancouver, July 1998. 32. J.D. Bjorken, Final state hadrons in deep inelastic processes and colliding beams, proceedings of the International Symposium on Electron and Photon Interactions at High Energies, p. 281, Cornell, 1971.
Hadron and Nuclear Physics with Electromagnetic Probes K. Maruyama and H. Okuno (Editors) 2000 Elsevier Science B.V. All rights reserved.
'
High-Q2 neutral- and charged-current reactions at HERA Kunihiro Nagano
a
*
aDESY, Notkestrasse 85, 22603 Hamburg, Germany. b~EK-Tanashi,Midori-cho 3-2-1, Tanashi, 188-0002 Tokyo, Japan. The H1 and ZEUS experiments at HERA measured e+p deep inelastic scattering cross sections at high-Q2 both for neutral-current and charged-current interactions using data taken during the 1994- 1997 running periods. Preliminary results of e-p deep inelastic scattering cross sections from the data taken during the recent 1998- 1999 running period are also presented. 1. DEEP INELASTIC SCATTERING AT HERA
Deep inelastic scattering (DIS) of leptons on nuclei has been the key for our understanding of the structure of the nucleon. HERA at DESY is a unique facility of colliding electrons (or positrons) with protons. The large center-of-mass energy, s, of about 300 GeV allows an extension of the explorable kinematic phase space by two orders of magnitude in Q2, the negative four-momentum transfer squared, compared with that covered by the previous fixed-target experiments. The maximum Q2 reaches almost lo5 GeV2, which corresponds to a spatial resolution of about 10-l6 cm. During the running in 1994-1997, HERA collided an e+-beam of 27.5 GeV and a p-beam of 820 GeV. The two collider experiments, H1 and ZEUS, collected 35.6 and 47.7 pb-' of luminosity during this period, respectively. Both the neutral-current (NC) DIS interaction, e+p -+ e+X, and the charged-current (CC) DIS interaction, e'p -+F X , have been studied using these data at high-Q2, typically Q2 > 200 GeV2 [I-31. From 1998 to April 1999, HERA provided an e--beam of 27.5 GeV and a p-beam of 920 GeV. Both H1 and ZEUS collected about 16 pb-l of luminosity during this period. Preliminary e-p -, e-X (NC) and e-p -+ vX (CC) cross sections [4-71 are also presented in this report. The NC DIS cross sections can be written as
where a is the electromagnetic fine structure constant, x is the Bjorken scaling variable, y is the inelasticity parameter and Y+= 1 1(1- y ) 2 . The kinematic variables are related Q 2 = xys. The helicity dependence of the electroweak interactions is mostly contained *The author is financially supported by t h e Japan Society for the Promotion of Science (JSPS).
in the kinematic factors Y*.The dependences on the quark structure of the proton and on the quark coupling to the Z0 boson as well as that on the mass of the Z0 boson are absorbed in the structure functions, F2NC, F t C and F,NC. The structure function F t C is dominated by a parity-violating term and contributes oppositely in e-p and e+p collisions. The evolution of the structure functions are evaluated using the next-to-leading order (NLO) DGLAP evolution equation of QCD. The longitudinal structure function FFC is not zero at NLO QCD and is only relevant at high y, where it contributes to the cross section by up to about 10%. The CC DIS cross sections can be written as
where GF is the Fermi constant, and Mw is the mass of the W* boson. At leading-order QCD, the structure function terms 4ZC can be written as
where d(x, Q2) is, for example, the parton density function (PDF) for the d-quark. The flavor-selecting nature of CC interaction is clearly seen; only down-type quarks (antiquarks) and up-type antiquarks (quarks) participate at leading order in e f p (e-p) CC DIS. The kinematic suppression factor (1 - y ) 2 is multiplied to quarks (antiquarks) in e+p (e-p) collisions. 2. NEUTRAL-CURRENT DIS CROSS SECTION
NC DIS events were required to have an isolated scattered electron with an energy greater than 11 (10) GeV in H1 (ZEUS) analysis. The selection efficiency is on average 80% for ZEUS. The main source of background is due to photoproduction, i.e. yp interaction with photon virtuality nearly equal to zero. The background contamination was estimated to be less than 1% in overall kinematic region, and was statistically subtracted. The resolutions of the kinematic variables are typically 5% for Q2 and 10% for x for ZEUS. The systematic error for the double differential cross section measurement by H1 is about 8%. The luminosity error is 1.5% (1.6%) for H I (ZEUS) and is not included in the systematic errors. 2.1. Single differential cross section da/dQ2 Figure 1 shows the e+p NC cross section da/dQ2 measured by ZEUS together with the preliminary e-p NC cross section, compared to the Standard Model (SM) predictions evaluated with the CTEQ 4D [8] PDF. The efp cross section was measured up to values of Q2=43000 GeV2. A good agreement was observed between data and SM predictions over six orders of magnitude as Q2 varies by two orders of magnitude, although a small excess at the highest Q2 was observed in the e+p cross section. H1 also observed a similar excess at the highest Q2 15000 GeV2 in efp cross section. The difference between e-p
ZEUS NC N -
10
%
Pn
a
1
CTEQ4D NLO e-p E,=820 GeV
N
Q
$ 10
.1
b P
10
.2
1994-97 e'p DATA CTEQ4D NLO e'p E,=820 GeV
I
Figure 1. The NC DIS cross section da/dQ2 measured by ZEUS both for e+p [I] and e-p [4] compared to the SM predictions evaluated with the CTEQ 4D PDF.
and e+p cross sections is clearly visible at large Q2 3000 GeV2. The figure also presents the expected SM e-p cross section with a proton beam energy of 820 GeV, resulting in a small change. The measurement demonstrates that the Z0 interference affects the cross section oppositely in e-p and e+p. 2.2. Double differential cross section d2a/dxdQ2 Figure 2 shows the reduced double differential cross section, CNC, which is defined as
measured by H1 both for e+p and e-p. The cross sections are displayed as functions of Q2 for fixed values of x ,compared to the SM prediction evaluated with the H1 NLO QCD fit [3]. A good agreement was observed with an exception at x = 0.40 in e+p data, where the measured cross section exceeds the prediction at higher Q2 10000 GeV2. This was already observed in the 1994- 1996 data [9]. The SM prediction overshoots the measured cross section at x = 0.65, where only BCDMS [lo] provides low Q2 data used in the NLO fit.
i,
1
2
' ' "
I
l o 4F
I
"
x=0,08 (x9000) 10 3Fx=0. 13 (~3000)
' ' " ' "
' ' '
BCDMS (x 0.97) o NMC --- Hl e'p QCD Fit
--" -
A
' '
'
HI e-p preliminary H1 etp
"
-
-
I
7 $
.-!-5
-.~,#hif-rry
3
Figure 2. The reduced NC DIS cross section, eNc,measured by H1 both for e+p and e-p compared to the SM predictions (solid lines for e+p and dashed lines for e-p) evaluated with the HI NLO QCD fit [ 6 ] .
3. CHARGED-CURRENT DIS CROSS SECTION
The principal signature of CC DIS events is a large missing transverse momentum, to the undetected final-state neutrino. The gT measured in the calorimeter was required to exceed 12 GeV. The overall selection efficiency is 80% and exceeds 90% in the high-Q2 region as estimated by ZEUS. The main source of background is due t o photoproduction with high transverse energy flow. The background contamination was estimated to be less than 2.5% in general, but is as large as about 10% in the lower-Q2 region of Q2 < 400 GeV2 in the ZEUS analysis. The final sample consists of about 1100 events for ZEUS. The resolutions of the reconstructed kinematic variables are typically 20% for Q2 and 8-15% for x achieved by ZEUS. The dominant systematic error is the energy-scale uncertainty of the calorimeter, resulting in a uncertainty of the measured
f i ; due
-1
Charged Current
.
H 1 e-p ds=3zo G prelimnary
~ V
- Standard Model (HI e'p QCD Fit)
Figure 3. The CC DIS cross section du/dQ2 measured by H1 both for e+p and e-p compared to the SM predictions evaluated with the H1 NLO QCD fit [6].
cross sections of typically 10%) but as much as 25% at the highest Q2 and 20% at the highest x. The total systematic error for the double differential cross section measurement of H1 is about 15%. 3.1. Single differential cross section da/dQ2 Figure 3 shows the e+p CC DIS cross section da/dQ2 measured by H1 together with the preliminary e-p NC cross section, compared to the SM predictions evaluated with the H1 NLO QCD fit. The e+p cross section was measured up to Q2 15000 GeV2. The measured cross sections are consistent with the SM predictions. ZEUS also observed a good agreement between data and SM in da/dQ2 both for e f p and e-p. Figure 4(a) shows the ZEUS measurement of the e+p CC cross section daldx for Q2 > 200 GeV2 compared to the SM prediction with the CTEQ 4D PDF. The plot (b) shows the cross section as a ratio divided by the SM prediction with the CTEQ 4D PDF. Additionally, the SM prediction evaluated with the ZEUS NLO QCD fit [ll]is also presented in the figure. The associated hatched error band represents the uncertainty of the SM prediction arising from PDF uncertainty as estimated from the ZEUS NLO QCD fit. This fit was performed to lower-Q2 data both from fixed-target and HERA experiments, not including the high-Q2 data presented in this report. At x 0.3, the data lie above the SM prediction with the CTEQ 4D PDF. The e+p CC cross section is dominated in the high-x region by the contribution from the d-quark, whose density function is poorly constrained from existing experimental data. This is reflected in Figure 4 in the large error band at high x. A possibility of a larger d / u ratio than currently assumed has been discussed in recent years. A modification of the CTEQ 4D PDF according to the prescription [12]: d/u -+ d/u O.lx(x 1) yields doldx, shown in the figure with the
+
+
ZEUS CC 1994-97 ZEUS 94-97 etp CC SM with CTEQ4D
a
ZEUS Preliminary 1998-99
' "
staq
Figure 4. The e+p CC DIS cross section daldx (defined for Q2 > 200 GeV2) measured by ZEUS [2] compared to the SM predictions evaluated with the CTEQ 4D PDF,
I}
h t . syst
Figure 5. The e-p CC DIS cross section daldx (defined for Q2 > 1000 GeV2) measured by ZEUS [5] compared to the SM predictions evaluated with the CTEQ 4D PDF.
dashed line, close to the ZEUS NLO QCD fit. For comparison, the prediction with the MRST [13] P D F is also shown in the figure. Figure 5 shows the preliminary e-p CC cross section daldx defined for Q 2 > 1000 GeV2 measured by ZEUS. A good agreement was observed between data and the SM prediction. It is worth while to note that the e-p CC cross section is dominated by the contribution from the u-quark at high x.
Figure 6. The reduced double differential CC cross section Zrcc measured by H1 as a function of x for fixed values of Q2, compared to the SM predictions with the H1 NLO QCD fit [6].
which is defined as
measured by H I both for e+p and e-p. The cross sections are displayed as functions of x for fixed values of Q2, compared to the SM prediction evaluated with the H I NLO QCD fit. The reduced cross section depends directly on the quark density distributions at leading-order QCD (see Eqs. (5) and (6)). A good agreement between data and the SM prediction was observed both in e+p and e-p; the measurement clearly demonstrates the flavor-selecting nature of the charged-current interaction. 4. HELICITY DEPENDENCE OF NC AND CC INTERACTIONS
In the approximate Bjorken scaling region of x x 0.1, the helicity dependence of structure functions can be separated from the dependence on parton densities. Figure 7 presents the H1 measurement of the structure function terms, q5NC and $CC, as functions of (1 - Y)' at fixed x = 0.13 both for e+p and e-p. The kinematic factor (1 - Y ) ~
x=O. 13
clcctrou
positron
H1 NC e-p preliminary H1 CC e'p preliminary
- H1 e'p
QCD Fit
.---y exchange only
Figure 7. The structure function terms for NC and CC, HI [7].
$NC
and $",
measured by
is related to the lepton scattering angle in the lepton-quark center-of-mass frame, 8*, as: cos4$ = (1 - y)2. In the upper plots, are presented together with SM predictions evaluated with the H1 NLO QCD fit. Also, SM predictions calculated only from y exchange terms are shown in the plots for illustration; the difference between and '$! is due to Z0 exchange. The lower plots show $2'. The difference in the intercepts (at (1 - y)' = 0) can be understood at leading order QCD stemming from the difference between densities of up-type quarks and of up-type antiquarks. And the difference in the slopes can also be understood at leading order QCD as due to the difference between densities of down-type quarks and down-type antiquarks (see Eqs. (5) and (6)).
$zC
$yC
5.
Mw determination
Two electroweak parameters, Mw and GF, enter the CC cross section; GF determines the absolute magnitude of the cross section, and Mw determines the shape of the cross
ZEUS 1994-97
Figure 8. The ZEUS
Mw
and GF fit to the e+p CC DIS cross section da/dQ2 [2].
section. A chi-squared fit to the ZEUS measured e+p CC da/dQ2, treating G F and Mw as free parameters, yielded +0.026 ( s y ~ t+0.016 G F = 1.171f0.034 tat.)-^,^^^ . ) - ~ ,(PDF) ~,, x G~v-~, (9)
Mw
=
80.8+::! (stat.)::::
(syst.)+i:' (PDF) GeV,
(10)
respectively. The central values are obtained with the CTEQ 4D PDF, and the P D F errors quoted are evaluated from the ZEUS NLO QCD fit. The obtained value of G F is in agreement with the value obtained from the muon-decay experiment, implying the universality of the CC interaction over a wide range of Q2. The obtained value of Mw is in agreement with the value obtained from direct measurements at LEP and Tevatron in the time-like region, demonstrating a consistency of the SM assumptions in complementary measurements. Figure 8 displays the fit result as the triangle with the 70% confidence level contour, which is determined only from statistical errors.
Two more fits were performed with more restrictive theoretical assumptions. First, a "propagator-mass" of the exchanged W + boson was extracted by fixing GF as that obtained from the muon-decay experiment [14]. The fit yielded Mw = 8l.4?$:~(stat.)k2.0(syst.)~~:~(~~F) GeV, which is also shown in Figure 8 as the solid point along the GF = 1.16639 x GeV-2 line. H1 performed a similar fit to their CC double differential cross section and obtained Mw = 80.9zk3.3(stat.)i1.7(syst.) GeVf 3.7(theo.) GeV. The SM uncertainty (quoted as "(theo.)" above) was evaluated by varying the assumptions for the H1 NLO QCD fit. Finally, a "Standard Model fit" was performed by ZEUS using the SM relation between GF and Mw, which includes the radiative corrections, namely contribution from t-quark, Z0 boson and Higgs boson. This constraint is also shown in the figure as the heavy solid line. The fit yielded Mw = 80.50?~:~$(stat.)~::$(syst.)f O,~~(PDF)+:::;(AM~, AMH,AMz) GeV. The result is indicated in the figure as the small star, while the large star along with the SM constraint line shows the position of the minimum x2. A great sensitivity to Mw was obtained within the SM framework. REFERENCES 1. ZEUS Collab., J. Breitweg et al. , DESY-99-056, accepted by Eur. Phys. J. 2. ZEUS Collab., J. Breitweg et al. , DESY-99-059, accepted by Eur. Phys. J. 3. H1 Collab., C. Adloff et al. , DESY-99-107, submitted to Eur. Phys. J. 4. ZEUS Collab., contributed paper #549 to EPS'99 conference, Tampere, Finland. 5. ZEUS Collab., contributed paper #558 to EPS'99 conference, Tampere, Finland. 6. HI Collab., contributed paper #157b to EPS'99 conference, Tampere, Finland. 7. M. Erdmann, a talk given at DESY seminar '99. 8. H.L. Lai et al. , Phys. Rev. D 55(1997) 1280. 9. HI Collab., C. Adloff et al. , 2.Phys. C 74(1997) 191. 10. BCDMS Collab., A.C. Benvenuti et al. , Phys. Lett. B 364(1995) 107. 11. M. Botje, DESY 99-038, NIKHEF-99-011. 12. U.K. Yang and A. Bodek, Phys. Rev. Lett. 82(1999) 2467. 13. A.D. Martin et al. , Eur. Phys. J. C4(1998) 463. 14. Particle Data Group, C. Caso et al. , Eur. Phys. J. C3(1998) 1.
Hadron and Nuclear Physics with ElectromagneticProbes K. Maruyama and H. Okuno (Editors) 2000 Elsevier Science B.V. All rights reserved.
Spin structure of the nucleon studied by
HERMES
Yasuhiro Sakemia* on behalf of the HERMES collaboration "Department of Physics, Tokyo Institute of Technology, 2-12-1 Ookayama, Meguro-ku, Tokyo 152-8551, Japan Spin asymmetries of semi-inclusive cross sections for the production of positively and negatively charged hadrons have been measured in deep-inelastic scattering of polarized positrons on polarized hydrogen and 3He targets in the kinematic range 0.023 < x < 0.6 and 1 GeV2 < Q2 < 10 GeV2 . Polarized quark distributions are extracted as a function of x for up and down quarks. The up quark polarization is positive and the down quark polarization is negative in the measured range. The polarization of the sea is compatible with zero. A measurement of the spin asymmetry in the photoproduction of high-pT hadron pairs is also presented. For h+h- pairs with pkl > 1.5 GeV/c and pk2 > 1.0 GeV/c the measured asymmetry is negative, in contrast to the positive asymmetries typically measured in deep inelastic scattering from protons, and can be interpreted to arise from a positive gluon polarization.
.
1. INTRODUCTION
The HERMES experiment studies the spin structure of the nucleon using semi-inclusive polarized deep inelastic scattering (DIS). In the Quark-Parton Model (QPM), the spin of the nucleon may be decomposed as:
where AX, AG, L,4, and LZGare the contributions from quark and gluon spin and quark and gluon orbital angular momentum. AC can be further divided among the quark flavors, and between valence and sea contributions:
Previous inclusive experiments precisely measured the polarized structure function gl (x), which can be interpreted in the QPM as the sum over all the flavors of polarized parton distribution functions (PDFs) Aq E (qf (x) - qP(x)):
*E-mail address: [email protected]
The function, qf(-), represents the distribution of quarks of flavor q with spin parallel (anti-parallel) to the nucleon spin, and e, is the quark charge in units of the elementary charge. Here, x is the Bjorken scaling variable x = Q2/2Mv which is determined by the scattered lepton's kinematics, where Q2 and v are the negative squared four-momentum and energy of the exchanged virtual photon, and M is the nucleon mass. By measuring gl(x) from various targets and integrating over x, it has been found that AC is only about a third of the nucleon spin, and that the strange sea seems negatively polarized. In the semi-inclusive measurements, hadrons are measured in coincidence with the scattered lepton. Hadron productions in DIS are described with the absorption of a virtual photon by a point-like quark and the subsequent fragmentation into a hadronic final state. Measuring asymmetries in semi-inclusive reactions yields further insights into the individual contributions of the quark flavors to AX. In these reactions, one also detects the properties of the fragments of the nucleon breakup after the initial scattering, which are correlated to the flavor of the struck quark in the reaction. One way to measure AG(x) is by isolation of the leading order QCD photon gluon fusion diagram. In deep inelastic scattering, one can do this via charm production or dijet production. Jet separation is not possible at most fixed target experiments due to the center of mass energy not being high enough. However it is possible to use highp~ hadrons instead of jets. A few phenomenological studies of the sensitivity of spindependent high-pT hadron photoproduction and low-Q2 electroproduction to the polarized gluon distribution AG have been conducted [1][2][3]. This paper reports on the extraction of polarized quark distribution from data taken by the HERMES experiment using the 27.5 GeV beam of longitudinally polarized positrons in the HERA storage ring at DESY, and on the polarized gluon distribution extracted from the spin asymmetry of high-pT hadrons. 2. THE HERMES EXPERIMENT
The HERMES experiment was designed to optimize semi-inclusive DIS measurements [4]. The experiment is located at the DESY laboratory in Hamburg, Germany, where a 27.5 GeV positron (or electron) beam is circulated in the HERA ring. The positron beam is naturally polarized transverse to the beam direction due to an asymmetry in the emission of synchrotron radiation, and longitudinal polarization at the HERMES interaction point is provided by a pair of matched spin rotators. The averaged beam polarization during the run is 55 % with a relative systematic uncertainty in the measurement of 4.0 % (3.4 %) for the 3He(H) data. The HERA beam passes through the center of a 40 cm long cylindrical storage cell containing polarized 3He or 'H.In 1995, the 3He target atoms became polarized through metastability-exchange optical pumping [5]. Polarization is 46 % with a fractional uncertainty of 5 % . In 1996 and 1997, an Atomic Beam Source (ABS) used permanent sextupole magnets and radio-frequency units to select desired hyperfine states of 'H.A Breit-Rabi polarimeter (BRP) monitors the proton polarization, which is 86 % during the run. The HERMES detector is an open-geometry forward spectrometer. A series of tracking chambers placed before and after a dipole magnet determine momenta and directions
of those particles. A threshold ~erenkovdetector, a Transition Radiation Detector, a Preshower Counter, and an Electromagnetic Calorimeter allow separation of electrons from hadrons. The ~ e r e n k o vfurther provides identification of pions above particle momenta of 5.6 and 3.8 GeV/c in 1995 and 1996-7. The data from the polarized 3He and polarized 'H targets is used to extract quark polarizations, and the polarized 'H target data is used to study the gluon polarization, respectively. 3. FLAVOR DECOMPOSITION OF THE POLARIZED QUARK DISTRI-
BUTIONS 3.1. Spin asymmetries of semi-inclusive cross sections In polarized deep-inelastic scattering, one can construct a cross-section asymmetry All of parallel and anti-parallel orientations of the nucleon spin to the lepton beam spin :
Under the assumption that g2(x) is zero, the measured double-spin asymmetry Ail is directly related to the asymmetry of the virtual photoabsorption in nucleon, Al(x), and to the structure function ratio, gl(x)/Fl(x):
where D. y.and q are kinematic factors. For this discussion, the index h of the asymmetry and structure function includes both inclusive positron data and coincident positive and negative hadrons and identified pions (h+, h-, T + , and T-). HERMES has measured asymmetries for inclusive and semi-inclusive h+ and h- reactions from 1995, 1996, and 1997 running. These asymmetries are shown in Figure 1. Inclusive DIS data are selected from events with cuts on Q2 > 1 GeV2, y = v / E < 0.85, and the final hadronic state mass, W2 > 4 GeV2 . The sizes of the inclusive event samples satisfying these requirements were 2.2 x lo6 and 2.3 x lo6 for 3He and 'H respectively. The analyzed semi-inclusive events form a subsample having a detected charged hadron and W2 > 10 GeV2. A cut on the hadron energy fraction, z > 0.2, and the Feynman variable. xf > 0.1, also selects nucleon fragments from the current region. The uncertainties in beam and target polarization measurement uncertainties, uncertainties in R(x, Q2), experimental yield fluctuation in 1995 are the dominant contributions to the systematic uncertainty of the measured asymmetries. 3.2. Interpretation of t h e asymmetries In the QPM and under the assumption of factorization, one sees that the photoabsorption asymmetry probes the helicity difference of the parton distributions Aq(x):
The asymmetries are assumed to be Q2 independent as the experimental values show no significant Q2 dependence. The fragmentation functions DPh(z)are spin independent and represent the probability that a struck quark q fragments into a hadron type h.
Figure 1. Results for the virtual photon asymmetries from HERMES 1995,1996, and 1997 data. The top row shows the asymmetries on the 'H target; the bottom row shows the 3He target. The columns are the inclusive, semi-inclusive h f , and semi-inclusive hasymmetries from left to right. The inclusive asymmetries are compared with the SLAC El43 [6] and El54 [7] measurements. For the semi-inclusive asymmetries with 'H target, a comparison with the SMC [8] results are shown.
This expression can be rearranged by defining a quark purity denoted by Pqh(x,z)for each hadron type h as
P," (x, 2) =
s;;
eiq(x) dzD,h(z) e$ql(x) dzD?/ (2) '
c~~ s;:
Pqhrepresents the probability that a detected hadron h is originated from a struck quark of flavor q in the nucleon. Using these purities, the asymmetries in equation 6 may be rewritten as
The asymmetries, purities, and quark polarizations may be grouped into matrices, and these equations can be solved using matrix algebra,
The quark polarizations are obtained by least square minimization techniques. The quark purities relevant for the HERMES experiment have been estimated with a Monte Carlo simulation of unpolarized DIS. The simulation model is based on LUND string fragmentation [9], CTEQ low Q2 unpolarized parton distributions [lo]. After nuclear corrections 3He asymmetries are related to quark distributions in the proton. In solving equation 9, one would like to separate the spin contributions of six quark flavors: u, dl s , fi, l a n d 3. In fact, statistics and limited sensitivity to the sea quark flavors require a selection of the model on average sea polarization. A simple assumption of flavor independent sea polarization is used here: Aqs - Ads 4s ds
Au,
Ad
Afi
As - As -
s
S
3.3. Extracted polarized quark distributions Using the six experimental asymmetries Ape,Aph+,Aph-, AsHee,AsHeh+,AsHeh- shown in Figure 1, three quark polarizations have been extracted: (Au Aii)/(u ii)(x) , the total up flavor polarization; (Ad + Ad)/(d + d)(x) , the total down flavor polarization; and (Aqs/qs)(x), the average polarization of the sea quarks. After multiplying by x and the unpolarized PDF parameterization, Figure 2 shows the results from this experiment in nine X B ~bins. The extraction has been repeated with GRV parton distributions [ll],with the independent fragmentation model, and with different parameterization for the fit of the LUND string model to HERMES multiplicities; a systematic uncertainty is assigned from the variation of the resulting quark polarizations. For x > 0.3, it is assumed that sea polarization does not contribute significantly to the measured asymmetries; the sea polarization, %(x), is set to zero with a small systematic uncertainty. In Figure 2, the parameterizations of De Florian et al. (0.1 < AG < 0.8, LO) [12] Gehrmann and Stirling (Gluon A, LO) [14], and Gliick et al. (Standard, LO) [13] are provided for comparison. A correction of (1+R) to the De Florian and Gliick parameterization~is necessary for consistency. These parameterizations agree with the HERMES results. Figure 3 compares the extracted valence quark polarization with the similar
+
+
extraction by the SMC experiment. Note that the HERMES definition of the flavor independence of sea polarization differs slightly from the model chosen by SMC; it has been verified that the results are insensitive to the slight differences between the models. Within statistical and systematic uncertainties, the results of both experiments agree, though high precision HERMES proton data is more sensitive to Au, and Afi [15]. Moments of the quark distributions in the measured x range are determined from the area under the measured points. For comparison to previous measurements, the distributions are extrapolated to low x by constant fits to the data; the low x extrapolation is quoted but no value is quoted for the uncertainty due to theoretical ambiguities. The resulting integrals in the measured region are listed in Table 1. The integrals are compared to predictions from Reference 16, which are corrected to fourth order in QCD; these predictions have been extracted from inclusive data assuming SU(3) flavor symmetry. The HERMES values are slightly smaller in magnitude than these predictions.
+
Au Aii Ad+Ad As+As
measured region predictions low-x total integral 0.51 f 0.02 f 0.03 0.04 0.57 f 0.02 f 0.03 0.66 f 0.03 -0.22% 0 . 0 6 f 0.05 -0.03 -0.25% 0 . 0 6 f 0.05 -0.35% 0.03 - 0 . O l f 0 . 0 3 f 0.04 0.00 - 0 . 0 1 f 0.03% 0.04 - 0 . 0 8 f 0.02
Table.1: The first moments of the extracted polarized parton distributions. 4. SPIN ASYMMETRY IN THE PHOTOPRODUCTION OF HADRON PAIRS 4.1. Analysis The cross section asymmetry A l l is given by
where N+ (N-) is the number of hadron pairs observed when the target spin is parallel (anti-~arallel)to the beam spin direction. The luminosities for each target spin state are L' and L,*, the latter being weighted by the product of the beam and target polarization values for each spin state. To identify the hadron pairs of interest, it was required that at least one positive hadron h' and at least one negative hadron h- be observed in the spectrometer. Each of these two hadrons were required to have a momentum greater than 4.5 GeV/c and transverse momentum p~ greater than 0.5 GeV/c. Here, p~ is defined with respect to the positron beam axis. It was required that the invariant mass of the two hadron system (assuming the hadrons to be pions) be greater than 1.0 GeV/c2. This effectively reduced the background arising from diffractive processes, specifically the decays of vector mesons. In order to include statistics from leptoproduction at Q2 x 0, it was not required that the scattered positron necessarily be observed in the spectrometer. The measured All is shown in Figure.4 . A negative asymmetry is observed at high transverse momenta of the negative hadrons. It should be noted that small and positive
Figure 2. Polarized parton distributions for up and down quarks, x(Au AG) and x(Ad + Ad), compared to the parameterizations of world data of g7'P(x). The data points have been evolved to Q2 = 2.5 Gev2 for the comparison.
+
Figure 3. The valence separation of parton distributions at Q2 = 2.5 GeV2 for up x(Auv), down x(Adv) and sea quarks x(Aq,) compared to results from the SMC experiment. The SMC data points have been extrapolated to Q2 = 2.5 GeV2 for the comparison.
asymmetries are expected for DIS from polarized proton target and that diffractive processes are expected to give zero asymmetry. A negative asymmetry is thus an indication for a different contribution to the asymmetry. 4.2. Interpretation The result for the asymmetry is interpreted in a model including contributions from vector meson dominance processes (VMD) and the direct photon processes. The direct photon processes are modelled using the two LO QCD processes: the QCD Compton effect (QCDC) and photon gluon fusion (PGF). The contribution from deep-inelastic scattering has been evaluated to be small. The asymmetries of these processes are related to the measured asymmetry in the following way:
where D is the virtual photon depolarization factor and Ai and fi are, respectively, the asymmetry and fraction of events arising from process i . Under the assumption of a zero spin asymmetry for VMD processes, the first term vanishes. For the small region of the phase selected by the present analysis, the asymmetries for QCDC and PGF are given by:
~ -1 The subprocess asymmetries 6's are directly calculable in LO QCD. They are G p G = (exactly, for real photons and massless participants), and GQcoc x 0.5 (for the average kinematics of this analysis). The polarized quark distributions Aq(x) are relatively well known from inclusive and semi-inclusive polarized DIS measurements. The contributions of the various processes were determined using the PYTHIA Monte Carlo generator [17]. The PYTHIA parameters were chosen according to Ref. [la] while the JETSET fragmentation parameters were tuned to semi-inclusive deep-inelastic scattering data from HERMES. In the region of interest ( pkl > 1.5 GeV/c and pk2 > 0.8 GeV/c ), the model was found to describe the shape of various distributions relatively well, though the normalization of the cross section was found to be underestimated. In this model, the cross section at high p~ is dominated by the photon gluon fusion, so this data was used to extract AG/G. The Monte Carlo showed no leading particle effects (as the background from QCDC was small), despite the fact that the data show a tendency to more negative All for pk- > 1.5 GeV/c , as opposed to large &+. The measured asymmetry (now averaged over the two plots from Figure.4) is compared with predictions from the Monte Carlo with different assumptions for AG/G in Figure.5. The best fit value for AG/G in this model is AG/G = 0.41 f O.lG(stat.) f 0.04(syst.) at an average XG of 0.17 [19].
5. CONCLUSIONS AND OUTLOOK
The first three years of HERMES data taking have yielded high statistics polarized deep-inelastic scattering data on 3He and 'H targets. The polarized measurements yield inclusive and semi-inclusive charged hadron asymmetries, which have been presented as a function of x. In the framework of the Quark-Parton Model these asymmetries are simultaneously fitted to obtain helicity distributions for up, down, and sea quarks. A measurement of the spin asymmetry All in quasi-real photoproduction of high p~ hadron pairs is also presented. When interpreted in a LO QCD model implemented in the PYTHIA Monte Carlo generator, a value of the gluon polarization in the nucleon AG/G is extracted. Further sensitivity to sea polarizations is gained by identification of the different coincident hadron measurements. Identification of pions from hadrons by the HERMES threshold ~ e r e n k o vcounter, may provide better sensitivity to As in the quark polarization extraction [20]. At present HERMES is taking a new data set with high statistics on the deuterium target, which will significantly improve the precision on the d-quark helicity distribution. In 1998, the Threshold ~ e r e n k o vcounter was replaced with a Ring
,
b
o.2
2 1 3
pr>1.5GeV/c
0 ...*......... .+................................................... -0.2 [ -0.4 -0.6 F -0.8 0 . 5
0.75
1
1.25
1.5
1.75
0.6
,
P y > 1 : 5 5
AG/G=-14 I
$ 0.4
-
0.2 : 2
-. -.-._._._._._._._._.-.-.-.-.-.-.-~-.-.j
-t-
4
O
-.:.
.tr
AG/G=O
I
-. -.-.-.-.-.,. . . . . .;i -----a
-0.4
-
-0.6 -0.4 -0.6 -0.8 1 0.5
-
0.75
1
1.25
1.5
1.75 2 ph;*(Ge V/c)
Figure 4. The measured asymmetry: the top plot is A l las a function of pTh- for pk+ ;t 1.5 GeV/c and the bottom plot is Ail as a function of p ~ for~ pk-+ > 1.5 GeV/c. The rightmost point in each plot is identical. Only the statistical error is shown.
- - - GSA ( 0.4)
. . . GSB ( - 0.3) 1 -.-.-. GSC (
-0.8
-
-1
-
I
0.5
0.75
I
1
I
1.25
I
I
I
1.5 &.75 2 PT (GeV/c)
Figure 5. Allcompared with different possibilities for AG/G. Upper to lower solid black curves are for AG/G = -1,0, and 1, respectively. Dashed, dotted, and dotdashed curves are for LO fits of Ref.[l4]
Imaging ~ e r e n k o vcounter (RICH). This new detector will identify T , K, and p particles over nearly all momenta accepted by the spectrometer. As kaon asymmetries provide direct sensitivity to strange quark polarization in the nucleon, a determination of sea quark polarization with much higher precision is expected after the new run.
REFERENCES 1. M. Fontannaz, D. Schiff, and B. Pire, Z. Phys. C 8(1981) 349 2. A. Bravar, D. von Harrach, and A. Kotzinian, Phys. Lett. B421 (1998) 349 3. A. Afanasev, C.E. Carlson, and C. Wahlquist, Phys. Rev. D58 (1998) 054007 4. K. Ackerstaff et al, Nucl. Instrum. Methods A 417(1998) 230 5. D.De Schepper et al, Nucl. Instrum. Methods A 419(1998) 16 6. K. Abe et al, Phys. Rev. D58 (1998) 112003 7. K. Abe et al, Phys. Rev. Lett. 79 (1997) 26 8. B. Adeva et al, Phys. Rev. Lett. D58 (1998) 112001 9. B. Anderson et al, Phys. Rept,. 97(1983) 33 10. H.L. Lai et al, Phys. Rev. Dl26 (1997) 1280 11. M. Gliick et al, Z. Phys. C67(1995) 433 12. D.de Florian et al, Phys. Rev. D57 (1998) 5803 13. M. Gliick et al, Phys. Rev. D53 (1996) 4775
14. T . Gehrmann and W.J. Stirling, Phys. Rev. D53 (1996) 6100 15. K. Ackerstaff et al. Phys. Lett. B464 (1999) 123 16. J . Ellis and M. Karliner. Invited Lectures at the International School of Nucleon Spin Structure. Erice, August 1995, hep-ph/9601280. 17. T . Sjostrand, Comp. Phys. Comm. 82 (1994) 74. The Monte Carlo results were obtained using PYTHIA version 5.724 and JETSET version 7.410. 18. G.A. Schuler and T . Sjostrand, Nucl. Phys. B407 (1993) 539 19. A. Airapetian et all Phys. Rev. Lett. (in press), hep-ex/9907020 20. L.L. Frankfurt et al, Phys. Lett. B230 (1989) 141
Hadron and Nuclear Physics with ElectromagneticProbes K. Maruyama and H. Okuno (Editors) 2000 Elsevier Science B.V. All rights reserved.
"
First double polarization measurements on the way t o test the Gerasimov-Drell-Hearn sum rule W. Meyera for the GDH- and A2-Collaborationsb "Inst. fur Experimentalphysik, Ruhr-Universitat Bochum, D-44801 Bochum, Germany bInst. fur Experimentalphysik, Ruhr-Universitiit Bochum, D-44801 Bochum, Germany Physikalisches Institut, Universitat Bonn, D-59115 Bonn, Germany Physikalisches Institut, Universitat Erlangen-Nurnberg, D-91058 Erlangen, Germany Labo voor Subatomaire en Stralingsfysica. B-9000 Gent, Belgium Department of Physics & Astronomy, University of Glasgow, U.K. 11. Physikalisches Institut, Universitat Gottingen, D-97079 Gottingen, Germany Department of Physics, University of Lund, Lund, Sweden Institut fiir Kernphysik, Universitat Mainz, D-55099 Mainz, Germany Faculty of Engineering. Miyazaki University, Miyazaki, Japan INR, Academy of Science. Moscow, Russia Department of Physics, Nagoya University, Chikusa-ku, Nagoya, Japan CIRSE, Nagoya University. Chikusa-ku, Nagoya, Japan INFN Sezione di Pavia, 1-27100 Pavia. Italy Dept. of Nuclear and Theor. Physics, University of Pavia, 1-27100 Pavia, Italy CEA Saclay, DSM/DAPNIA/SPhN, F-91191 Gif-sur-Yvette Cedex, France Physikalisches Institut, Universitat Tiibingen, D-72076 Tubingen, Germany The helicity dependence of photon induced reactions on the proton has been measured in the energy range from 200 up to 800 MeV for the first time. The experiments have been performed at the Mninz accelerator facility MAMI. A 47r detector system, a circularly polarized tagged photon beam and a newly developed polarized solid target have been used. The horizontal 3 H e / 4 H e dilution refrigerator includes a thin superconducting coil to maintain the polarization of the protons in the frozen-spin mode longitudinal to the incoming photon beam. The data obtained provide new information for multipole analyses of pion photoproduction and give contributions to the Gerasimov-Drell-Hearn (GDH) sum rule and the forward spin polarizability yo.
1. INTRODUCTION In recent years, the improvements in polarized beam and target techniques have opened new possibilities to investigate the polarization degrees of freedom of the nucleon. In the study of the spin dependent reactions on the nucleon: there is a special interest in the experimental verification of the Gerasimov-Drell-Hearn (GDH) sum rule. This sum rule, derived in the sixties by Gerasimov [I] and independently by Drell and Hearn [ 2 ] ,relates
static properties of the nucleon, the anomalous magnetic moment K and the mass m, to the difference of the total photoabsorption cross sections for circularly polarized photons on longitudinally polarized nucleons. It is written as:
where 0 3 / 2 and a l p are the photoabsorption cross sections for the total helicity states 312 and 112 respectively (parallel or antiparallel photon-nucleon spin configuration) and v is the photon energy. The left-hand side of eq. 1 is called GDH-integral. In a similar way, the so called forward spin polarizability -yo can also be obtained:
The GDH sum rule is based on basic physics principles (Lorentz and gauge invariance. unitarity, causality) applied to the forward Compton scattering amplitude. The only questionable assumption made is that this amplitude becomes spin independent at infinite photon energies. Due to its fundamental character, this prediction provides an excellent test of nucleon spin structure. Due to the technical difficulties associated to this kind of experiments, the GDH sum rule has not been verified previously. Without any direct experimental results, some theoretical evaluations on the GDH integral were made using multipole analyses of (mainly unpolarized) single pion photoproduction data, pion scattering data, and a crude model for double pion photoproduction [3-61. All models underestimate substantially the sum rule prediction (-204pb) for the proton. The measurements have been started at MAMI (Mainz) covering the energy range from the pion threshold up to 800 MeV and will be continued at ELSA (Bonn) up to a photon energy of 3.0 GeV [7].In this report the experimental set up at Mainz will be explained with the special focus on the polarized solid target. Its latest technical development was decisive for the performance of the double polarization measurement in combination with the demand of a 4?i particle detection. The first results on the helicity dependence of the total cross section of the yp -+ NT channels below 450 MeV will be published soon [8]. 2. EXPERIMENTAL SETUP 2.1. Circularly polarized photons The experiment was carried out at the Glasgow-Mainz tagged photon facility of the MAMI accelerator in Mainz. Circularly polarized photons were produced by bremsstrahlung of longitudinally polarized elec,trons. A strained GaAs photocathode routinely delivered electrons with a degree of polarization of about 75 % [9]. The electron polarization was monitored throughout the experiment with a precision of f3 % by means of a Moller polarimeter. The polarization direction was flipped randomly every few seconds to minimize systematic effects. The photon polarization was evaluated according to the formula of ref. [lo]. The photon energy was determined by the tagging spectrometer which analyses the momenta of the electrons that radiated the bremsstrahlung photon. This detection system is able to t,a,gphotons in t,he range from 50 to 800 MeV with a resolution of about 2 MeV [ I l l . The tagging efficiency was continuously monitored throughout the experiment by a e+e- pair detector, that was installed downstream of the main hadron detector. The
pair detector was regularly calibrated using a 100 % efficient lead glass detector. Using this technique, the photon flux could be determined with a systematic uncertainly of 2 % 1121. 2.2. Longitudinally polarized protons The low intensity tagged photon beam demands a solid polarized target to reach reasonable luminosities. A butanol (C4H90H)frozen spin target [13] provided the polarized polarizable protons. The target length of 2 cm gave a total thickness of Nt = 0.9. protons per em2, from which a luminosity results of about 1029sec-1. - Novel frozen spin technique A polarized solid state target uses the principle of Dynamic Nuclear Polarization (DNP) in which a target material - butanol (C4HSOH)- is doped with a dilute assembly of paramagnetic radicals by chemical techniques. The material is cooled to about 300 mK in a magnetic field of 2.5 Tesla and then irradiated by microwaves near to the electron spin resonance frequency to drive the hyperfine transitions, which allow the nucleon spin to be aligned parallel or antiparallel to the applied magnetic field. For the cooling of the target material a horizontal 3He/4He dilution refrigerator was built, the outer dimensions of which were determined by the inner dimensions of the DAPHNE detector and by the warm bore of the polarizing magnet. Two precooling stages, a separator and an evaporator, and several counter current heat exchangers provide the precooling and the e mixture. The dilution unit consisting of the still, the liquefaction of the 3 ~ e / 4 H gas final heat exchanger and the mixing chamber permits cooling to temperatures below 50 mK. The cryogenic components of the refrigerator are placed symmetrically around the beamline tubing, which is the central part of the refrigerator and serves also as an internal isolation vacuum to reduce the thermal heat load. For the same purpose the refrigerator is also enclosed by an isolation vacuum and surrounded by two heat shields. For the DNP over the target process a magnetic field of 2.5 Tesla with a homogeneity of < area is required. This can only be provided by a huge magnet that surrounds the target completely, resulting in a reduction of angular acceptance. To increase the acceptance to a totally open geometry for the scattered particles the 'frozen spin' technique with an internal holding coil is used. In this technique the nucleon polarization is preserved after the polarization process at a temperature of about 50-60 mK and a magnetic field of 0.4 Tesla with a relaxation time of more than one week. This magnetic field is provided by a superconducting, internal holding coil that is integrated in the refrigerator. It consists of four layers of NbTi wire wound on a 300 pm copper layer, which results in a total thickness of 700 pm, such that a significant loss of scattered particles occurs only close to the threshold. As a result of its compact design (length 120 mm, diameter 44 mm) the coil has a weak outer fringe field, so that the influence on the detectors located close to the target area is very small in comparison to that of the outer holding coils formerly used. The coil is cooled to a temperature below 1.2 K by a thermal contact to the liquid helium of the still. At this temperature a maximum field of 0.48 Tesla at 12 A was achieved. The homogeneity of the magnetic field over the target area (< 5 . allows for the first time an online polarization measurement during 'frozen spin mode' (data taking), so that the systematic error of the polarization measurement is considerably lowered. A maximum polarization close to 90 % and relaxation times in the frozen spin mode of about 200 h
have been regularly achieved. The degree of the proton polarization has been determined by the NMR technique with a relative precision of less than 2 %. 2.3. 4n particle detection Photoemitted hadrons were registered in a detector system covering almost the full solid angle. This system was based on the large angular acceptance (21" 5 6 5 159'; 0' 5 q5 5 360") detector DAPHNE [14] complemented by additional detectors to increase the acceptance into the forward angular region. DAPHNE is a charged particle tracking detector consisting of 3 coaxial multi wire proportional chambers surrounded by segmented plastic scintillator layers and by a scintillator - absorber sandwich, allowing the detection of neutral pions with a reasonable efficiency. The additional forward detec16"), a aerogel Cerenkov tors were the silicon microstrip detector MIDAS [15] (7" 1 19 I counter to suppress electromagnetic background and the annular ring detector STAR 1161 followed by a forward scintillator-lead sandwich counter (2" 5 6 5").
<
Missing Energyy p + no on Botand (p identhMJ
2,
150
Im
so
o
50
rm
rsa
2m
mlsslng energy (MeV]
Figure 1. Experimental asymmetry for the -+ nopreaction in the A-resonance.
yp
The identification methods for hadrons have been described in detail previously [12,17] and only the characteristics of the data taking from the butanol target will be given here. Using such target, the background contribution of the reactions produced on C and 0 nuclei could not be separated event by event from the polarized H contribution. However, this background, coming from spinless nuclei, is not polarization dependent and cancels when the difference between events in the 312 and 1/2 states is taken. As an example, fig. 1 shows this difference in Em,,, (missing energy) obtained from events with a photoemitted proton in the A region. Emis,is the difference between the measured proton kinetic energy
and the evaluated proton kinetic energy under the assumption that the proton originated from a 71.' production process on hydrogen. Missing energy distributions are shown for both helicity states and for their difference. The region outside the peak at Emiss = 0 corresponds t o quasi free reactions on C and 0 nuclei and has a yield consistent with zero in the difference spectrum. 3. RESULTS AND OUTLOOK
Data have been taken at two different electron energy settings of 800 MeV and 525 MeV, to maximize the photon polarization over the entire energy range. The single pion and yp -. nr+ as well as the double pion channels, photoproduction reactions yp + which open at 450 MeV will be analysed separately. Up to now data below 450 MeV, where the two single pion channels represent the total photoabsorption cross section, will be published [8]. The extraction of the total cross section difference allows the direct determination of the contribution in the energy range investigated to the GDH integral and the forward spin polarizability. The multipole analyses SAID [18] and HDT 1191 overestimate the experimental results by about 10 - 20 %.
REFERENCES S.B. Gerasimov, Sov. J. Nucl. Phys.2 (1966) 430 S.D. Drell, A.C. Hearn, Phys. Rev. Lett. 16 (1966) 908 1. Karliner, Phys. Rev. D 7 (1973) 2717 R.L. Workman, R.A. Arndt, Phys. Rev. D 45 (1992) 1789 V. Burkert, Z. Li, Phys. Rev. D 47 (1993) 46 A.M. Sandorfi et al., Phys. Rev. D 50 (1994) R6681 J. Ahrens et al. Mainz MAMI Proposal A2/2-95 and G. Anton et al., Bonn ELSA Proposal 1992 8. J. Ahrens et al., to be published 9. K. Aulenbacher et al., Nucl. Instr. Meth, A 391 (1997) 498 10. H. Olsen and L.C. Maximon, Phys. Rev. 114 (1959) 887 11. I. Anthony et al., Nucl. Instr. Meth. A301 (991) 230, S.J. Hall et al., Nucl. Instr. Meth. A368 (1996) 698 12. M. MacCormick et al., Phys. Rev. C55 (1996) 41 13. C. Bradtke et al., Nucl. Instr. Meth. A 346 (1999) 430 14. G. Audit et al., Nucl. Instr. Meth A 801 (1991) 473 15. S. Altieri et al., INFN Report TC-98/30 (1998) 16. M. Sauer et al., Nuvl. Instr. Meth. A 378 (1996) 143 17. R. Crawford et al., Nucl. Phys. A 603 (1996) 303 18. R.A. Arndt et al., Phys. Rev. C 53 (1996) 430 (solution SM99K) 19. 0 . Hanstein et al.,Nucl. Phys. A 632 (1999) 561
1. 2. 3. 4. 5. 6. 7.
ThisPageIntolyLfBk
ThisPageIntolyLfBk This Page Intentionally Left Blank
VII. NEW FACILITIES
ThisPageIntolyLfBk
ThisPageIntolyLfBk This Page Intentionally Left Blank
Hadron and Nuclear Physics with ElectromagneticProbes K. Maruyama and H. Okuno (Editors) 2000 Elsevier Science B.V. All rights reserved.
Few-body physics a t Jefferson Laboratory F. W. Hersman" and the Hall A and CLAS Collaborations "Department of Physics, University of New Hampshire, Durham, NH 03824 Studies of few-body nuclei have been recently completed or are now underway using the latest experimental facilities at the Jefferson Laboratory. I discuss the results on deuteron and proton elastic scattering, and the measurements using polarized 3He with the spectrometers. I then outline what we can expect from the recently completed measurements on light nuclei with CLAS. 1. ACCELERATOR STATUS
The Continuous Electron Beam Accelerator at the Jefferson Laboratory consists of two main linear accelerators with five paths of recirculation. It can accelerate up to 200 pA to 4 GeV with emittance better than lop9 meter: meeting all basic design specifications. Three separate beams can be delivered to three experimental end stations. Beams to separate halls may have five orders of magnitude difference in intensity. Accelerator availability is independent of the number of beams being delivered. Some capabilities exceed the original design specification. Beam energy of 5.5 GeV was delivered for physics. Also parity-quality polarized beam is now routinely used for experiments: either 40% at 100 pA or 75% at greater than 40 pA has been provided. 2. DEUTERON ELASTIC SCATTERING IN HALL A
The high momentum transfer now achievable at Jefferson Laboratory for precision coincidence measurements allows a stringent test of nuclear structure theories. The deuteron form-factors have proven to be particularly revealing when examining the extent of agreement between experiment and theory and the character of the breakdown. The impulse approximation underestimates existing elastic scattering data. Relativistic corrections and inclusion of meson-exchange effects can improve agreement, depending on the method of their treatment. The Hall A Collaboration used two High Resolution Spectrometers (HRS) for an experiment on elastic scattering from the deuteron, led by G. Petratos.[l] To suppress backgrounds and separate elastic from inelastic scattering, scattered electrons and recoil deuterons were detected in coincidence. A cryogenic target system with three cooling loops, each capable of 700 watts, was developed to exploit the high luminosity capability of the accelerator. Beam rastering allowed currents up to 120 pA while keeping beaminduced density changes below 2.5%. The luminosity of 4.0 x ~ m - ~ s -isl the highest
" JLab Hall C ---7
+m C IIummel & Tjon
-
RIA
Y
R I A + KZC Van Orden at a1
---
Hummsl & Tjon
Figure 1. The deuteron elastic structure function A(Q2)compared with calculations.
ever. Data on the proton and deuteron were measured with and without collimators and compared with existing proton data to confirm normalization. Figure 1 shows the experimental results. Values for the cross section were converted to A(Q2) assuming the contribution from B(Q2) is negligible. The data are in agreement with the existing SLAC data in the range of the overlap. The estimate of systematic uncertainty is *5.9%, dominated by the uncertainty in the absolute acceptance. Recent theoretical calculations using Relativistic Impulse Approximation (RIA) prescriptions are compa,red with the measurements in the recent paper.[2] The calculations differ in the form of the relativistic equation, the treatment of off-shell behavior, and the prescription for treating the p r y meson exchange. The predictions diverge for Q2 > 3.5 GeV/c2. The one which most closely goes through the high Q2 points is RIA from Van Orden, including MEC. The high and low curves in this region are from Hummel and Tjon without (low) and with (high) MEC. The deuteron magnetic form factor, corresponding to measurements of B(Q2), further determines the deuteron's structure. Precision data were taken over part of the Q2 range of existing data, however extending the data on B(Q2) to higher Q2 awaits improvements in the hall. 3. DEUTERON TENSOR POLARIZATION
Separation of the deuteron monopole from the quadrupole form factor requires knowledge of a third scattering observable, t z o The world's data on tzo was extended in momentum transfer in a double scattering experiment performed in 1997 in Hall C.[3] The three required developments, the high power cryotarget, deuteron magnetic channel, and the deuteron polarimeter POLDER, were undertaken by a J?rench/US/Swiss/Armenian collaboration. During four months of data taking the experiment received integrated beam of 320 coulombs. New data plotted in Figure 2 are compared with existing data and with theoretical calculations. Within the range of overlap, the new data, taken together, appear systematically higher than earlier data from MIT-Bates. The relativistic calculation of Van Orden, e t all and the non-relativistic calculation with relativistic corrections of Wiringa, et all provide better descriptions of the data than the non-relativistic calculation of Wiringa, et al. Results are nearing publication. [4] 4. PROTON POLARIMETRY
The polarized electron beam can be combined with hadronic polarization, either using polarized targets or second-scatteing polarimetry, to measure double polarization observa b l e ~ .In a Hall A experiment to determine the proton electric form factor,[5] polarized electrons were scattered from hydrogen followed by a determination of the recoil proton polarization using the focal plane polarimeter in the hadron HRS. Since the cross section is essentially the sum of the squares of the electric and magnetic form factors, the sensitivity of Rosenbluth separations to separate the proton electric form factor from the dominant magnetic decreases markedly at high momentum transfers. In contrast, the polarimeter asymmetry is proportional to the product of the two form factors, and therfore linear in the electric form factor.
D e u t e r o n tensor p o l a r ~ z a t ~ of, n
~ 05
t
+JLabIPOLDER VEPP-3 BatedAHEAD A NIKHEF NRIA - v18 (Wtnnga et al ) NRIA + RC + MEC - v18 CIA - relat~v~st~c (Van OrdenI et al.) CQM (Buchmann et al ) PQCD (Brodsky et al )
-
i
D e u t e r o n c h a r g e m o n o p o l e f o r m factor
J
Figure 2. The tensor polarization of the deuteron measured in the recoil polarization following elastic scattering is compared with previous data and recent calculations (upper). The t20 data translated to a measurement of the monopole charge form factor using the precision measurements of observables A(Q2) and B(Q2).
Cambndge (1971) (1971) ODESY (1973) D SLAC (1989) *SLAC (1993) O SLAC (1994) I Bonn
Figure 3. Double-polarization results on electron-proton elastic scattering plotted as the Sachs electric (above) and normalized magnetic (below) form factors divided by the dipole form factor
The proton polarimeter installed in the hadron HRS consists of two front straw chambers and a scintillator plane to determine the incoming trajectory and time, a carbon analyzer with thicknes up to 60 cm, and two rear straw chambers and scintillator plane. Data are corrected event-by-event for precession through the spectrometer fields. Data for the electric form factor are plotted in figure 3 as a ratio to the dipole form factor and compared with the existing world data.[6] The data confirm the steeper slope of the form factor indicated by the SLAC data of 1994 and exclude the SLAC data from 1970 and 1989, which tend upward. Extensions of this measurement to higher momentum transfer are planned. 5. EXPERIMENTS WITH A POLARIZED 3He TARGET The high quality polarized beam at Jefferson Lab can be further exploited for nucleon form factor measurements and few body physics by using a polarized 3He target. To a good approximation, the spin of the 3He nucleus can be attributed to the spin of the neutron. Consequently asymmetries measured in quasifree knockout reactions can be interpreted as measurements of the neutron form factors.[7] Small D and S' components of the wave function can be extracted by measuring asymmetries in coincidence proton quasifree knockout.[8] Integrated inclusive asymmetries can be compared to the DrellHearn-Gerasimov sum rule. [9] The polarized 3He target is a two-cell system. A warm polarizing cell with is illuminated on the D l resonance of the vaporized rubidium by several diode lasers. The helium becomes polarized through spin exchange, and diffuses into the colder region, which serves as a target. Thin glass windows intercepted by the electron beam are not within the acceptance of the spectrometers. The target operates routinely with 35% polarization or above with 12 pA rastered beam. Two experiments, a measurement of inclusive quasielastic asymmetries to extract the magnetic form factor of the neutron and a more complete inclusive measurement to compare with the GDH sum rule, were completed during the running period late 1998 and early 1999. While these experiments are under analysis, planning for the next experiments is underway. 6. FEW-BODY PHYSICS PROGRAM AT CLAS
Few-body electrodisintegration and photodisintegration experiments are being investigated in all three experimental end stations. CLAS (CEBAF Large Acceptance Spectrometer) provides for global surveys while the other end stations concentrate on particular kinematics with high luminosity. Six experiments were proposed in 1989 to study nuclear multihadron reactions and interpret them from various physics perspectives. [lo-151 Since many of these measurements were able to use the same targets and beam energies, a merged plan emphasizing the two isotopes of helium and carbon, with additional time on iron was developed and approved in 1991. Later, a motivation for extending the measured observables to include polarized beam asymmetries was advanced and approved.[l6] The strength, shape, and longitudinal-transverse character of the quasielastic, dip and A-isobar regions show deviations from what would be expected from one-body mecha-
nisms. CLAS will map out the separate multihadron emission channels throughout this kinematic regime and study their nuclear mass dependence. Likewise, backward particle production has been extensively studied in inclusive reactions, particularly with hadronic beams, in a regime where a one-body mechanism is forbidden. With CLAS we will more fully characterize the identities and kinematic dependences of backward-going particles in electron induced reactions, determine the momentum and energy transferred with coincident electron detection, and determine the extent of two-particle (proton and pion) correlations in emission kinematics. In addition, these reactions will be characterized with polarized electron beam, allowing us to survey the polarized beam asymmetry and associated fifth response-function. This extensive survey of nuclear reaction phenomenology is designed to shed light on the relevant underlying mechanisms and guide appropriate methods for describing them. Possible mechanisms. can be subdivided into initial state structures, complex many-body currents, and final state interactions. Each of these mechanisms may dominate in particularly favorable kinematic regions. By surveying a very large range of kinematics simultaneously, CLAS can characterize the phenomena over a broad range, and hopefully reveal these mechanisms. The additional information obtained by measuring polarized beam asymmetries deserves special discussion. The polarized beam asymmetries vanish for coplanar coincidence reactions, and also vanish for direct reactions. This observable will be nonzero only for multistep reactions such as those dominated by final state interactions, and only for nonplanar kinematics.
7. T H E CEBAF LARGE ACCEPTANCE SPECTROMETER The CEBAF Large Acceptance Spectrometer (CLAS) was designed from the beginning to meet the goals of the nuclear multihadron experimental program. The accelerator provides a high quality electron beam with up to 5.5 GeV energy and polarization around 70%. Hall B can either deliver these electrons to the experiment, or use a radiator and photon tagger to produce a pure, tagged photon beam. CLAS combines magnetic analysis of charged particles, efficient detection of neutrals, and large acceptance. The angular range for charged particles extends from 8" to 130" with angular resolution approximately 1 mr. Momenta above 0.2 GeV/c can be extracted with resolution exceeding 1%for most kinematics. Photons greater than 0.05 GeV are registered with resolution around 9%. Particle identification separates kaons to 1.5 GeV/c and separates pions from protons to 3 GeV/c. This combination is ideal for separating the inclusive response into separate channels, for characterizing multipole components in angular distributions, and for surveying extensive kinematic regions. CLAS consists of a six coil superconducting toroidal magnet with a low field Moller sweeper, multiwire drift chambers, Cerenkov detectors, time-of-flight scintillators, electromagnetic shower calorimeters, a flexible trigger, and a high-rate data acquisition system. 8. ELECTRON PROTON REACTIONS
In order to highlight CLAS capabilities I show some results from electron reactions on a hydrogen target which are in an advanced stage of analysis. The data were acquired for
E = 4.045 GeV, I, = 3375A
Figure 4. Elastic and inelastic scattering of 4.045 GeV electrons from protons in CLAS, showing the yield as a function of final hadronic mass W (left), and the acceptance of CLAS as a function of W and Q2 (right).
run "elan in March 1998. The incident beam energy was 4.045 GeV with a current of a few nA on a 4.0 cm target of liquid hydrogen with thin aluminum windows. By combining the energy loss and scattering angle of scattered electrons as in Fig. 4, one can plot the yield as a function of W , the final hadronic mass, and Q2, the four-momentum transfer. The elastic scattering peak, the A-resonance, and the second and third resonance regions are clearly visible. The r.m.s. resolution in W is 13 MeV. The overall timing performance can be extracted from the time difference measured in the time-of-flight scintillators extrapolated to the projected track time origin and the beam pulse. (The beam pulse has a width of only a few picoseconds.) This timing distribution, including all detectors, was rms 160 ps. (Fig. 5) This excellent performance is vital for separating particle identities. Note that the kaon signal to noise is one to one, averaged over all observed kaon momenta. Additional software improvements can
E = 4.045 GeV, I, = 3375A
0
1
2
3
4 M, GeV
Figure 5. The rms time resolution, referenced to the beam synchronization pulse, averaged over all detectors was 160 ps (left), provides excellent particle identification (right).
increase this ratio. Electron proton coincidence events can be combined to form missing mass, revealing the identity of the decayed meson (regardless of whether that meson was also detected). Combining the missing mass with the final hadronic mass W allows an association between the various resonances and their mesonic decay channels. Small channels contributing to the electron proton coincidence response can be enhanced by requiring additional coincidences. The electron proton missing mass squared can be plotted including only those events in which an additional two hard photons were detected in the calorimeters. For this selection of events, the no,the 77 and the w meson become more cleanly visible. (Fig. 6) Likewise in a selection of events that required detecting four photons, the 77' meson becomes visible with the 7 . These techniques of using multiple coincidences to highlight particular processes and channels have obvious relevance for multihadron coincidences on nuclei.
E = 4.045 GeV, I, = 3375A
Figure 6. Missing mass spectra for electron-proton coincidences, selected on events that also included two (left) or four (right) detected photons, reveal neutral meson decay channels.
9. NUCLEAR MULTIHADRON REACTION RESULTS FROM CLAS
The program of experiments described above was performed at CLAS during April and May 1999, too late for including results for this talk. I will discuss this run at the end. Nevertheless, results for nuclear multihadron reactions were obtained during run "elan due to the need for subtraction of the background from the aluminum end windows. I will discuss what we have learned. The beam energy for these measurements was 4.045 GeV with 20 nA on a 15 pm aluminum target (window). The torus current was 3375 A (nominal). We acquired a half million events with the electron trigger. We first examine inclusive electron yields as a function of the four momentum transfer Q2 and the energy loss w . It becomes clear from this plot (Fig. 7, upper left panel) tliat the kinematic range that is accessed with this beam energy and CLAS is extensive. Along
Figure 7. The yield of Electron proton coincidences from aluminum plotted according to the electron momentum transfer and energy loss, and coincidence variables missing energy and missing momentum.
the left diagonal is low-energy nuclear breakup and quasifree nucleon knockout. The other two edges of phase space at the bottom and top-right (at low and high Q2) are determined by the minimum and maximum angles for electron detection. Most of these events have coincident protons associated with them. In order to make the plots most useful, we must choose appropriate kinematic variables that reveal a particular process. Already we find that these simple plots include a combination of overlapping information: the missing energy and missing momentum are dominated by events where the struck nucleon (either proton or neutron) is not the one used to form the Em and Pm variable. These events overwhelm what we might call "interesting" events, events with significant initial state momentum or large final state interactions. While the reaction is defined by five kinematic variables, missing energy and missing momentum can only be interpreted as the initial momentum and binding energy if the struck nucleon is the one that is detected. For the purpose of characterizing processes in the nucleus that give rise to large initial
P cut
""l
c ! 3 0c C r b
Figure 8. The component of the outgoing proton momentum perpendicular to the momentum transfer is used as a kinematic variable to highlight the importance of high-momentum components and final state interactions.
momenta or large final state interactions, we plot the yield as a function of the transverse momentum, that is the component of the detected momentum (or missing momentum) perpendicular to the momentum transfer. (Fig. 8) Only higher order processes can lead to large values of this kinematic variable. We plot the (e,elp) coincidence yield as a function of the p,perp. The yield peaks around 250 MeV/c and shows strengths out to 1 GeV/c for all observed values of missing energy. A further exploration of the subset of events with high perpendicular momentum components might reveal the process responsible for this exotic type of event. We can attempt to refine our examination by limiting the physics that contributes by using kinematic cuts. We define a set of cuts on electron kinematics associated with the quasielastic, dip, and A regions. We look at coincidence variables of missing energy and missing momentum for the A region. (Fig. 9) While this selection may ultimately be very useful, there are still too many contributions to the Em and p, plots. We have been plotting (e,elp) kinematic variables with various cuts. CLAS detects
s.5
:-
71/' 3
7.5
1.5
,
. ,, I
" A i l : I
Fr
:-
F.
Figure 9. Electron variables are selected to include only events in the A excitation region.
additional particles. We next plot the same (e,elp) kinematic variables, but selected on the type of detected third particle. In particular we can look for a pion or proton in addition to the original proton and electron. These can be examined in each kinematic region, quasifree, dip and delta. Finally we can also form cuts on low, intermediate and high momentum transfer Q2. (Fig. 10) The plots I have shown represent one-half million events obtained last year on aluminum. Last spring our collaboration finished our six week run. We studied four nuclei 3He, 4He, 12C and iron using three beam energies 1.1,2.2 and 4.4 GeV. We obtained over two billion triggers, with 750 million each on carbon and 4He, and 530 million on 3He. Over one billion triggers were recorded at 4.4 GeV beam energy. The data reduction will be performed over the next six months. Comparisons with theory will begin early next year. We urge our theorist colleagues to become involved in this program. New infrastructure is being developed for performing data analysis and comparisons with theory, including our new 100 processor CPU farm in New Hampshire. We look forward, with optimism, to a period of rapid progress in our understanding of multihadron emission from nuclei.
-"
--
Figure 10. Missing energy for e,p coincidences is plotted for bins in
Q 2 and
x.
I'd like to acknowledge John Calarco, Riad Suleiman, Betsy Beise, and Mark Jones for the figures in the discussion on the deuteron and proton form factors, and Stepan Stepanyan, Larry Weinstein, and Rustam Niyazov for the figures from CLAS. This presentation represents the efforts of the Hall A Collaboration, the tzo Collaboration, and the CLAS Collaboration. Nuclear physics research at the University of New Hampshire is supported by the US Department of Energy under grant DE-FG02-88ER40410.
REFERENCES 1. E91-026 Measurement of the Electric and Magnetic Structure Functions, G. Petratos, J. Gomez. 2. L. C. Alexa, et al., Phys. Rev. Lett. 82, 1374 (1999) 3. E94-018 Measurement of the Deuteron Tensor Polarization at Large Momentum Transfers in D(e,eld) Scattering, S. Kox, E. Beise. 4. D. Abbott, et al., ISN report ISN-2000-01 (ISN-Grenoble), and to be published. 5. E93-027 Electric form Factor of the Proton by Recoil Polarization, C. Perdrisat, V.
Punjabi. 6. M. K. Jones, et al., LANL preprint nucl-ex/9910005, and to be published. 7. E95-001 Precise Measurements of the Inclusive Spin-dependent Quasi-elastic Transverse Asymmetry AT from 3He(e,e') at Low Q2, H. Gao. 8. E94-023 Measurement of Small Components of the 3He Wave Function using 3He(e,e'p) in Hall A, F. W. Hersman. 9. E94-010 Measurement of the ~ e t r o n ( ~ HSpin e ) Structure Function at Low Q; a connection between the BJ and DHG sum rules, A. E. Meziani, G. Cates. 10. E89-015 Study of Coincidence Reactions in the Dip and A-Resonance Regions, H. Baghaei. 11. E89-017 Electroexcitation of the A(1232) in Nuclei, D. Branford (originally R. Sealock). 12. E89-027 Coincidence Reaction Studies with the LAS, L. Weinstein, W. Bertozzi, W. Boeglin. 13. E89-031 Study of Multi-Nucleon Knockout with the CEBAF Large Acceptance Spectrometer, R. Miskimen, F. W. Hersman (originally also with J. Lightbody). 14. E89-032 Study of Local Properties of Nuclear Matter in Electron-Nucleus and PhotonNucleus Interactions with Backward Particle Production Using the CLAS Detector, V. Gavrilov, G. Leksin. 15. E89-036 Study of Short Range Properties of Nuclear Matter in Electron-Nucleus and Photon-Nucleus Interactions with Backward Particle Production Using the CLAS Detector, K. Egiyan. 16. E98-104 Measurement of the Polarized Electron Beam Asymmetries in Exclusive Reactions on Nuclei with CLAS, M. Holtrop, F.W. Hersman.
ThisPageIntolyLfBk
ThisPageIntolyLfBk This Page Intentionally Left Blank
Hadron and Nuclear Physics with ElectromagneticProbes K. Maruyama and H. Okuno (Editors) 2000 Elsevier Science B.V. All rights reserved.
'
Experimental test of the K-A relative parity -Use of polarized photon beams at high energiesYoshio Yamaguchia a
c/o KEK-Tanashi, Tanashi, Tokyo 188-8501, Japan Photo K production from helium +;He+KO
(or --+;H+K+)
can be used to test experimentally the K-A relative parity. 1. INTRODUCTION
Polarized photon beams are quite useful in nuclear and hadron physics. In the past electron Bremsstrahlung with the final electron tagging was used to obtain polarized photons, however, whose intensity is not good enough. Now availability of high energy electron beams of good quality and powerful lasers made it possible to produce intense polarized photons using the inverse Compton scattering [I]. Such facilities are constructed (or under construction) at several labs in the world, including Spring 8, and expected soon to uncover many interesting physics outcomes in nuclear and hadron physics. Detailed information of photo-kaon production is one of many others, and expected to contribute complete phase shift analyses of K-N systems at low energies. 2. EXPERIMENTAL TEST OF K-A RELATIVE PARITY
In view of the great success of the Standard Model, no body wonders that K-meson is pseudoscalar. Nevertheless it would be nice to have a direct experimental test of the K-A,relative parity. Notice that the odd parity of no was directly proved long ago by observing correlation of two plans spanned by (e+eP) in no -+ (e+e-) (e+e-). To this end let us consider photo-kaon production from the nucleon
+
whose matrix element in the CM system is given by (NR approximation) f [ k x e ] p + g ( e . p ) k for 0- kaon f ( e . p ) k + g [ k x e]p for 0+ kaon where k, and p are momenta of photon and kaon and e is the polarization vector of photon, f and g depend on k = /kj and (k . p). The differential cross section of (1) for
linearly polarized photon on unpolarized proton target is do
- oc
dR
If
+ cc 1 f I2k21e.pI2 + k2Igl2 for Of
I21e. [p x k]I2 k2Igl2 for 0- kaon kaon
Clear difference between (3a) and (3b) is useful to establish experimentally K-A relative parity, even though sizable relativistic corrections are to be added to the matrix element (2). The situation becomes clearer when the helium target is used [2] (where the relativistic corrections are smaller as compared to the nucleon target). If one can distinguish between two processes
y + 4 ~ +e i ~ e(ground state O+)
+ KO
(4)
and
y + 4 ~ -+:He* e (excited state 1+,1 MeV) + K O ,
(5)
(4) offers better case to test the K-A relative parity. The differential cross section of (4) for the linearly polarized (monoenergetic) incident photon is given by do dR
- oc
/ e . [k x p ] 1 2 ( ~ ) 2for 0-,
oc le. p 1 2 ( ~ ) 2
for 0+,
where [3]
-
k and p are momenta of y and K in CM system of (4). The region, slightly above the threshold, say k 1 GeV, may be preferable to test (6a) and (6b). REFERENCES 1. R.H. Milburn, Phys. Rev. Letters 10 (1963) 75. 2. Y. Yamaguchi, Soryushiron-Kenkyu (in Japanese) 25 (1962) 447. 3. The wave functions for 4He and :He have been kindly supplied by Prof. Y. Akaishi.
Hadron and Nuclear Physics with Electromagnetic Probes K. Maruyama and H. Okuno (Editors) 2000 Elsevier Science B.V. All rights reserved.
"
Nuclear physics experiments with 1.2-Gel? STB ring a t LXS-Tohoku
"Laboratory of Nuclear Science, Tohoku University, Mikamine, Sendai 982-0826, Japan The Laboratory of Nuclear Science at Tohoku University has recently completed construction of the stretcher-hooster ring (STB ring). The STB ring is now producing a continuous electron beam with energies up to 250 MeV for external target stations. and 1s about ready to deliver a tagged photon beam with energies up to 1.1 GeV. We describe the present esperiii~entalprograms for nuclear physics with the STB ring.
1. INTRODUCTION The Laboratory of Xuclear Science (LNS) at Tohoku University has contributed to iluclear science research with electron beams in the last three decades. It has had a 300-hIeV electron linear accelerat,or since 1967. A 150-MeV pulse-beam stretcher was constructed in 1981 as a prototype ring for a GeV stretcher ring proposed in late 70's. However, a11 original proposal of the GeV electron accelerator complex for nuclear physics Ivas not accept,ed, and, t,hen, the project was forced to scale down; the 1.2-GeV stretcherl)oost,er ring (STB ring) was constructed. The STB ring is expected to work not only for nl~clea,rphysics research but also as an injector for a 1.8 GeV SR ring, which is planned for a future project. Tlle STB ring has two functions; the pulse-beam st,retcher ring and the boost and storage ring. In the stretcher mode operation, the ring accepts a pulsed beam from the 300-;\1eV Linac (300 pps) and produces a continuous beam (about 80 % of duty factor). The extracted beam is delivered to target stations for low energy nuclear experiments. In t'he boost and storage mode operat,ion, electrons are accelerated up to 1.2 Ge\' to be stored in the ring. They can be used for internal target experiments including high-energy tagged phot,on production. The comissioning of the STB ring was started in November, 1997, and the acceleration "11 t,o 1.0 Ge\' was succeeded in May) 1998. After the inspection on radiation safety the stret,cher mode ~ p e r a t ~ i ohad n been developed. At present a 200-MeV conrinuous beam in the stretcl~ermode operation is in use for various experiments and a 1.2-GeV tagged photon 11eam is being developed.
2. EXPERIMENTS WITH LOW ENERGY BEAMS The experimental area of LNS is shown in Fig. 1. The low energy pulsed heam from the 300-1Ie\; linac is used for the production of various radioisotopes at the station 1, where electrons with energies up to 60 Ale\' are available. As is shown in Fig. 1, the STB ring occupies Illore than half area of the esperin~ent,alroom no. 2. Two target stations are
preparecl around the ring for 1.2-Gel' experiments; 1.2-GeV tagged photon and internal target experiment. In t h e stretcher mode operation, the low energy continuous beam (150 E, 5 250 MeV) is e x t r a d e d from t h e ring and is transported t o t h e LDM station: where a large dipole nlagnetic spectrogra,ph can be used to analyze t h e scattered electrons. T h e other station for low energy experilllent,s is the low-energy tagged photon station located in t h e esperilnental roo111 no. 1.
<
Figure 1. 1.2-Gel7 Stretcher-Booster ring and experimental stations a t Laboratory of Suclear Science.
The following is a list of the approved nuclear physics experiments with low energy stretclled l ~ e a m s .Experiments (1) and ( 3 ) have been already carried out. ( 1 ) Study of quasifree elastic process in the '2C(e,e'p) reaction. ( 2 ) Stucly of 1.4s with (e,e'p) reactions on heavy nuclei. ( : 3 ) Stud!. of I\iAIR and IVQR in the "Si(e,e'n) reaction. ( A ) Study of the 10B(e,e'n) reaction. ( 5 ) Collective nlotion in heavy nuclei via (e,e'fission) reactions. ( 6 ) Ifechanism of sub-threshold pion production In (e,.;ro) reactions. ( 7 ) Search for escit,ation of two-phonon GR in t h e 40Ca("in) reaction. ( 8 ) Stucly of photo-absorption mechanism in t h e 12C(y.n) and "C(y,np) reaction. Here. I discuss t h e possibility t o excite the two-phonon giant resonance in a photoabsorption process; Experiment (7) proposed by Terasawa[l]. Evidence for t h e twophonon GDR has been found firstly in pion double charge exchange reaction at LAhIPF
(Lo5 Alamoi) [2]. and in heavy ion reactions at SIS[3] and GA?;IL[4]. However, up to the plesent, no attempts have been reported to observe the two-phonon GR states via
photo-absorption.
- yvvg)
-
Two phonon GDR
Two phonon
Two phonon GDR
GDR
GDR
x-
r DGDR
0 . .- ;2 /
DGDR
?
/
GDR 1;9%//
(x
+)
(El)
0+,2+
I
GQR or GDR
/
I O+
To
(a)
( Z', Z- )
reaction
(b) Heavy ion reaction
To
To-1
(c) Photo-absorption
Figure 2. Two-phonon GR states excited in various reactions: (a) pion double charge exchange reaction. (b) heavy ion reaction, and (c) photon absorption reaction.
I11 t'he pion double charge exchange reaction, the two-phonon states can be excited througll two-st,ep particle-hole excitations. For example, the ( T + , T - ) reaction can be T ~ - ~ ( reaction. T ~ , As is shown in regarded as the two-step z T o j ~ fT, ~ ) ~ ~ ~ 7rTT-)zt2T0-2 Fig. 2(a), the first step reaction excites the intemediate states of the AT=l particle-hole (13rot'on-~leutron)excitat,ion corresponding to the GDR of the To-1 nucleus. The second step is the succesive excitation of another particle-hole (proton-neuteron) excitation, and, thus, the t~-o-phononGDR (GDRgGDR) can be excited in the To-2 nucleus. In heavy ion reactions. the strong Coulomb field between the projectile and the target nuclei provides many virtual photons. Then the target nucleus can absorb, for example, two photons si~nult~aneously, and a two-pnonon GR state is excited, as shown in Fig. 3(b). In t'he photo-absorption process, however; only one photon must take part in exciting a. two-phonon state. Thus. t,he interesting question is "can such an excitation occur? X possihle excitation mechanism of the two-phonon states may be a coupling of a photon with the meson exchange current (MEC). The MEC is known to play an important role in photo-absorption process, especially for excitation energies higher than the GDR "
escit,ation. In Fig. 3 ( c ) the escitation mechanism is shematically shown. The incident plioton is co~~plecl with IIEC' which connects the double particle-hole excitat,ions. As the electric dipole absorption should be the main process, the two-phonon states should be I - . A T = l state: a 1- phonon coupled with a 2+ phonon. The 1- phonon corresponds t>o tlie isovector giant dipole resonance (I\'GDR): ancl the 2+ phonon to the isovector giant' quatlrul>ole resonance ( I i 7 Q G R ) .An isoscaler mode is not excited since t h e MEC acts as a T-T+ (or T + T - ) operator. Thus. one of the two phonons of this I V G D R z I V G Q R 1state is t h e proton-particle and neutron-hole type while the other neutron-particle and roton on-hole tj.pe. The 1- tu-o-phonon G R escitecl by photo-absorption can be regarded with the C;R of the nest nucleus. as the pliono11 escitation cou~~lecl 1.Suzulii[5] predicted the photo absorption cross sect,ion to escite t h e t>wo-phononGR to be as large as 2.5% of that of the single GDR escitation in "Ca. One may expect to oljspr~.ethe t,\vo-phonon G R in the 'OCa(? , n ) reaction by measuring neutrons ~vhich escape from t h e two-phonon state to one-phonon stat,e.
3. EXPERIMENTS WITH 1.2 GEV CIRCULATING BEAM For esperinle~ltswith GeV electrons, one of the main interests is t'he behavior of t h e nucleon escitation in nuclei. As presented by Yamazaki in this symposium[6], we have studied S l l resonance in nuclear medium at the ES facility of t h e former IXS. \Ve are ) on various lluclear to estend the measurements of ( - , . I / ) ancl ( ~ , p 7 /reactions targets wit11 nlucli Ijetter statistics, in order to see whether t h e resonance property does cliange in nuclear nletliuill or not.
TAGX
SCISSORS
Figure .3. 1.2-Gel7 tagged photon beam course and target station.
In Fig. 3, we show the experimental area for 1.2-GeV tagged photon experiments. The photon tagging systelll is now being in~t~alled.The system consists of a radiator placetl at t h e entrance of the bending magnet (BM4) of the S T B ring and 100 plastic scintillat,ors placed in a snlall room in the bellcling maganet. The radiator made of a thin carbon foil can be moved ancl set at the beam position very quickly: t h e rac1iat)or is removed from the beam position when t,he electron beam is injected and accelerated. The electrons emitting photons are detected with one of t,he plastic scintillators (5 x 5 x 20 m m ) . Each scirltillator is connected with a special light guide which couples t o 3-m long light f i b ~ r . Then, phot,omult,ipliers can b e placed a,t a distance from t h e bending magnet. T h e target position of the ta,gged photon is &out 7.5 m from the radiator. There. two large detector syst,ems will be placed: one is the TAGX sytem, which has been working very well in the ES facility at INS for these 15 years. T h e other is called SCISSORS (Sendai CsI Scintillation Systenl On Radiat,ion Search): a large scintillation detector system: corlsisting of 200 pure CsI crystals. The former system will be used maily for charged part,icle detection, while the latter for y-ray detection. As sllown in Fig. 3, they can be placed at the target position of the tagged photon beam. The 1.2-GeV tagged phot,on system is now being examined together with the improveinent of t,he quality of the 1.2-GeV electron beam circulating the STB ring. The tagged photo11 esperirnent is expected to start in llay, 2000, and, then, LNS will serve not only for lo~r-energynuclear physics but also for GeV energy nuclear physics.
REFERENCES I. ' I Terasalva, . private coininunicat~ion. 2. S . Mordechai et al., Phys. Rev. Lett. 60 (1988) 408. 3 . J . Ritnlan et al.. Phys. Rev. Lett. 70 (1993) 533; R. schmit et al., Phys. Rev. Lett. 70 (1993) 1767. 4. K. Frascaria, Nucl. Phys. A569 (1994) l l l c . .5. T . Suzulci, private communication. 6 H. Yamazaki, paper in this proceedings; T . I'orita et al., Phys. Lett. B in press.
ThisPageIntolyLfBk
ThisPageIntolyLfBk This Page Intentionally Left Blank
Hadron and Nuclear Physics with Electromagnetic Probes K. Maruyama and H. Okuno (Editors) 2000 Elsevier Science B.V. All rights reserved.
'
Laser electron photon facility at Spring-8 T . Hottaa* , J.K. Ahna, H. Akimuneb, Y. Asanoc, W.C. Changd, S. DatBe, M. Fujiwaraalf, K. Hicksg, K. lmaii, T . Iwatah, T . lshikawai, H. Kawai", Z.Y. KimJ, T . Kishimotom, N. Kumagaie, S. MakinoP, T. Matsumuraklc, N. Matsuokaa, T . Mibea, M. Miyabei, Y. Miyachih, T . Nakanoa, M. Nomachia, Y. Ohashie, T. Oobao, '~~ Sakaguchim, , T. Sasaki', D. Sekii, H. Ookumae, M. Ooshimaf, C. R a n g a ~ h a r ~ u l uA. H. Shimizua, Y. Sugayaf, M. S ~ m i h a m a ~T). ~~ ,o o ~ a mH.a ~Toyokawae, , A. Wakai", C.W. Wangd, S.C. wangd K. Yoneharab, T . Yoritaa, and M. Yosoia "Research Center for Nuclear Physics, Osaka University,lO-1 Mihogaoka, Ibaraki, Osaka 567-0047, Japan bDepartment of Physics, Konan University, Kobe, Hyogo 658-8501, Japan 'Japan Atomic Energy Research Institute, Mikazuki, Hyogo 679-5143, Japan dInstitute of Physics, Academia Sinica, Taipei 11529, Taiwan "Japan Synchrotron Radiation Research Institute, Mikazuki, Hyogo 679-5143, Japan fAdvanced Science Research Center, Japan Atomic Energy Research Institute, Tokai, Ibaraki 319-1195, Japan gDepartment of Physics, Ohio University, Athens, OH 45701, USA hDepartment of Physics, Nagoya University, Nagoya 464-8602, Japan 'Department of Physics, Kyoto University, Kyoto 606-8502, Japan JDepartment of Physics, Seoul National University, Seoul 151-742, Korea kDepartment of Physics, Yamagata University, Yamagata, Yamagata 990-8560, Japan 'Department of Physics, University of Saskatchewan, Saskatoon, S7N 5E2, Canada "Department of Physics, Osaka University, Toyonaka, Osaka 560-0043, Japan "Center for Integrated Research in Science and Engineering, Nagoya University, Nagoya 464-8603, Japan "Department of Physics, Chiba University, Inage, Chiba 263-8522, Japan PWakayama Medical College, Wakayama, Wakayama 641-0012, Japan At Spring-8, we built a 2.4 GeV photon beamline with 350 nm laser photon backscattering off the circulating 8 GeV electron beam. A detector system, optimized to carry out the photoproduction of meson near threshold, is nearly completed.
+
1. LASER ELECTRON PHOTON FACILITY AT SPRING-8
High energy photon beams produced by laser-induced backward Compton scattering off the circulating electrons (laser electron photon) is utilized for nuclear physics studies at various synchrotron radiation facilities in the world [1,2]. Spring-8, the world's highest energy third-generation synchrotron radiation facility, enables us to produce the highest *Corresponding author. E-mail: [email protected]
Experimental Hutch
0
10
20rn
Figure 1. The Laser Electron Photon facility at Spring-8 (LEPS)
energy laser electron photon beam. We have constructed the laser electron photon (LEP) facility a t BL33LEP, one of the 61 beamlines (Fig. 1). At Spring-8, 8 GeV electron beam is circulating with I,, = 100 mA. The laser beam is injected from the Laser Hutch to the straight section. The Compton scattering of the laser photon with an 8 GeV electron produces GeV photons which can be used for the experiments of quark nuclear physics. The maximum energy of the LEP is determined by the electron energy and the laser wavelength. The LEPS facility produces the maximum photon energy of 2.4 GeV with 350 nm Ar laser. At present, Spring-8 is the only facility providing LEP beams above yp -+ q5p threshold. When shorter wavelength (200 nm) laser is used in near future, the maximum photon energy becomes higher than 3 GeV. One of the advantages of LEP beam is its flat intensity distribution for the Compton photons. Thus, experiments do not suffer from high intensity low energy background, a common problem with Bremsstrahlung beams. Another feature is high polarization of photons. If laser photons are 100 % polarized, the LEP is also polarized at the maximum energy. The polarization of the photon decreases a t lower photon energy. However, by changing the laser wavelength, highly polarized photons in wide energy range can be obtained. The position and polarization of the laser can be monitored on both sides of the interaction region; at the Laser Hutch and the end of the beamline. When the electron energy is high, LEPs are emitted in a narrow cone in the electron beam direction. At the LEPS facility, photons in the energy region of interest ( E , 2 1.5 GeV) are within the scattered angle less than 0.15 mrad, which result in beam size of about 1 cm at the target point in the Experimental Hutch. The energy of the LEP is determined by measuring recoil electron with a tagging system. The bending magnet of the storage ring is used for analyzing the momentum of recoil electron. The tagging system is a position detector located inside the ring at the downstream end (along the electron beam) of the bending magnet. It consists of two layers of 500 pm thick silicon strip detectors with 100 pm pitch and two layers of plastic scintillator array. Photons of 1.5 to 3.5 GeV can be tagged by measuring recoil electron
with the momentum range between 6.5 and 4.5 GeV. Simulation results show that the photon energy resolution is expected to be around 15 MeV, and it is mainly determined by energy and angular spread of incident electrons. 2. BEAM COMMISSIONING
The detailed design of the beamline started in 1997. In 1998, some accelerator components were modified to inject a laser beam against the electron, to extract a LEP, and to utilize the tagging system. The laser injection system was completed by the first quarter of 1999. The first LEP beam at Spring-8 was produced on July lst, 1999. A PWO calorimeter was used for the measurement of the beam energy. Fig. 2 shows the measured energy spectrum of the beam. The tagging system was also tested. Fig. 3 shows
0
0.5
1
1.5
2
2.5
3
3.5
4
E, (GeV)
Figure 2. The energy spectrum of the LEP measured by PWO calorimeter.
0
5
lo
15 20 25 position (rnrn)
30
35
Figure 3. The energy of the LEP measured by PWO versus the hit position of recoil electron at the tagging counter.
the correspondence between the energy measured by the PWO calorimeter and the hit position of recoil electron at the tagging counter. While this measurement shows that there is one to one correspondence between the hit position and the photon energy, the resolution of PWO is not good enough to estimate the tagger energy resolution. We plan to measure the e+e- pair conversion of the LEPS beam with the magnetic spectrometer and the detector system to get an estimate of the tagger resolution. As the first stage of the laser system, The UV multi-line photon of Ar laser (-350 nm) was used to produce the LEPs. The integrated intensity at that time was about 2 x lo6 photons/sec with 5 W laser power and 100 mA electron beam current, which is about 5 times lower than we expected. We have found some problems on the reflectivity of the mirror and we expect the intensity reaches 1 x lo7 photons/sec after the beam commissioning.
3. PHYSICS
At the LEPS, many experiments have been proposed and discussed. One of the experiment to be done at the first stage is 4 photoproduction. The photoproduction of vector mesons (VMs) are described by the process that y fluctuates into VM which is well known as the vector meson dominance. Then the VM is scattered diffractively by Pomeron exchange [3]. In the framework of Regge theory, Pomeron was introduced to describe the universal rise of hadronic cross section at high energies. Nowadays Pomeron exchange can be understood as multi-gluon exchange process [4]. In the 4 photoproduction, the meson exchange process is strongly suppressed by OZI rule and the cross section from threshold to HERA energy region is mainly due to Pomeron exchange [ 5 ] . It means that the gluon exchange process at low energy can be studied only by 4 photoproduction. As an example, it is suggested that precise measurement of unpolarized cross section will clarify O+ glueball contribution [6].By using polarized photon and/or polarized target, we can study the contributions of different processes, including sS knockout from a nucleon, in terms of helicity amplitudes [7]. 4. DETECTORS
Figure 4. The LEPS detector.
The LEPS detector setup is shown in Fig. 4. The detector is designed for the measurement of the q5 photoproduction in the forward direction. It consists of silicon strip vertex detectors (SVTX), multi-wire drift chambers (MWDC), a dipole magnet, and a time-of-flight (TOF) wall.
The dipole magnet has 135 cm wide and 55 cm high opening and the length of the pole along the beam is 60 cm. The field strength is 1 T at the center. The SVTX consists of 2 planes (x and y) of single-sided silicon strip detectors (SSDs). The thickness of each SSD is 300 pm and the strip pitch is 120 pm. With the SVTX we measure the positions and energy loss for particle identification. A MWDC (DC1) with 5 planes (x, x', y, y', u) is located upstream and a pair of MWDC (DC2 and DC3) are downstream of the magnet. The sensitive area is 80 cm wide x 30 cm high for the DC1 and 200 cm wide x 80 cm for the DC2 and DC3. Each of the DC2 and DC3 has 5 planes; x, x', y, y' and u (for the DC2) or v (for the DC3). The TOF wall is used to measure times of flight of particles from the target. The TOF wall consists of 40 plastic scintillator bars of dimensions: 4 cm thick x 12 cm wide x 2 m long. The timing resolution of better than a = 100 psec has been achieved. The TOF start signal with a = 12 psec is made from an RF signal of 8 GeV storage ring. As a standard setup, the TOF wall is located at 3 m from the dipole magnet and a flight length of a charged particle is about 4 m. In this case, K/n separation of up to 2 GeVlc momentum is expected at better than 4a level. Each detector components has been tested separately by using the LEP beam and Bremsstrahlung y from the ring. An integrated test of the detector has been just started. For other experiments, a photon calorimeter, made up of 252 lead scintillating fiber detectors of 14 radiation lengths, covers from 30 - 100" in laboratory system. Also, a superconducting solenoid magnet is being built for the future development of polarized target. 5. S U M M A R Y The laser electron photon beam with the maximum energy of 2.4 GeV has been successfully produced at a newly developed beamline, BL33LEP at Spring-8. The beam intensity will be increased up to 1 x l o 7 /sec after the beam commissioning. The detector system has been constructed and the tests are underway. The first test experiment of $ photoproduction is planned for early 2000. REFERENCES 1. A. M. Sandorfi, J. LeVine, C. E. Thorn, G. Giordano and G. Matone, IEEE Trans. Nucl. Sci. 30, 3083 (1983) 2. C. Schaerf, Nucl. Phys. News 2 (1992) No. 1 7-8 3. T . H. Bauer, R. D. Spital, D. R. Yennie and F. M.Pipkin, Rev. Mod. Phys. 50, 261 (1978) 4. A. Donnachie and P. V. Landshoff, Phys. Lett. B296, 227 (1992) 5. M. Derrick et al. [ZEUS Collaboration], Phys. Lett. B377, 259 (1996) 6. T. Nakano and H. Toki, Exciting Physics with New Accelerator Facilities, World Scientific, 1997 7. A. I. Titov, Y. Oh and S. N. Yang, Phys. Rev. Lett. 79, 1634 (1997)
ThisPageIntolyLfBk
ThisPageIntolyLfBk This Page Intentionally Left Blank
Hadron and Nuclear Physics with Electromagnetic Probes K. Maruyama and H. Okuno (Editors) 2000 Elsevier Science B.V. All rights reserved.
'
MUSES project at RIKEN RI beam factory T. Katayama
a)b,
K. Maruyama
a
and M. Wakasugi
"Center for Nuclear Study, Graduate School of Science, University of Tokyo 2-1, Hirosawa, Wako-shi, Saitama 351-01, Japan bRIKEN (The Institute of Physical and Chemical Research) 2-1, Hirosawa, Wako-shi, Saitama 351-01, Japan We are proposing to construct an accelerator system of storage ring and collider at RIKEN Radio Isotope Beam Factory (RIBF). This accelerator complex is named as MUSES(Mu1ti Use Experimental Storage rings). MUSES consists of several rings. One is an Accumulator Cooler Ring (ACR) where an electron cooling device and a stochastic cooling device will be installed to cool down the RI beams. The accumulated and cooled RI beams will be transported to the Double Storage Ring (DSR). The DSR is a new type of collider where the collision of RI beams with electron beams is planned to study the electromagnetic structure of unstable nuclei. In the present paper, the outline of the MUSES project will be described as well as plans of two typical experiments. 1. OUTLINE OF MUSES PROJECT
The Radioactive Isotope Beam Factory (RIBF) is an extension of present heavy ion accelerator facility at RIKEN [I]. The construction of the RIBF is separated into two phases. The first phase is scheduled from 1997 to 2002. In this construction phase, an intermediate ring cyclotron (IRC, Knumber=950), a super-conducting ring cyclotron PLAN VlEVl
OF RI-BEAM FACTORY
Figure 1. Plan view of RIBF and MUSES
(SRC, K=2500), RI beam separators (Big RIPS) and an experimental hall will be completed. The second phase is scheduled from 2001 to 2008, when the MUSES (Multi-Use Experimental Storage rings) project will be completed. The MUSES is an accelerators complex consisting of RI beam separator (RIPS-M), an accumulator cooler ring (ACR), a booster synchrotron ring (BSR), a 300-MeV electron linac (e-linac) and double storage rings (DSR)[2]. Figure 1 shows a plan view of the RIBF. Heavy ion beams from the RRC (Riken Ring Cyclotron K=540) are boosted up to 400A MeV for light ions and lOOA MeV for heavy ions by IRC and SRC. With this heavy ion beam, we can produce RI beams for all elements using projectile-fragmentation process. Details of the SRC and the IRC are described elsewhere [3, 41. 1.1. RI beam separator At the downstream of the SRC, we will construct two RI beam separators. The separator (Big RIPS) provides RI beams for the experimental halls, and another (RIPS-M) for the MUSES system. The primary beam is supplied for three separators with time sharing technique (see Fig. 3). A pulsed beam with the peak beam intensity of 100 (the particles pA (ppA) is supplied for RIPS-M, and the maximum duty factor is beam duration of 30 psec and the interval of 30 msec). DC beams with the intensity of 1 ppA are supplied for the Big RIPS. About 3000 radioactive isotopes including about 1000 new isotopes can be used for experiments with those separators. The RIPS-M used for the MUSES has a momentum acceptance of f 2.5 % and an angular acceptance of f 10 mrad. The momentum spread of the RI beams from the RIPS-M is expected to be f 0.5 %. Since this value is too large from the point of view of cooling time in the ACR, we will place debuncher at 80-m downstream of the RIPS-M. The momentum spread is reduced to 0.15 % with the debuncher system. The maximum RF voltage for the debuncher is required to be 4.23 MV.
+
1.2. Accumulator cooler ring (ACR) Figure 2 shows schematic view of the MUSES system. The RI beams are injected into the ACR by means of a multi-turn injection method (about 30 turns per one injection).
Figure 2. Schematic view of MUSES accelerator system
la1 Primary Beam from SRC
Lbl Injection into ACR
Id1 Time Chart of MUSES Syatam
MuItI-turn lnjrllon
N uri
Figure 3. Time chart of
MUSES operation.
The transverse horizontal acceptance of the ACR is 125 nmm.mrad and the momentum acceptance is k 2 %. Injected RI beam is stacked by controlling the RF voltage and the frequency in the ACR. After RF stacking process, the beam is cooled down in both transverse and longitudinal directions by combination of a stochastic cooling and an electron cooling methods. A cycle of the injection, i.e. the multi-turn injection, the RF stacking and the cooling, is repeated until the number of stored particles reaches to the equilibrium number which depends on the intrinsic lifetime of the RI and the space charge limit. This cycle is shown in Fig. 3(b). If we do not need the cooling, only the stacking process takes about 30 msec. This is why the maximum duty factor of primary beam for the RIPS-M is lop3. Figure 4 shows typical results of analytical calculation of the stochastic cooling time vs. number of stored particle, assuming the pickup impedance of 100 R, the temperature of the pre-amplifier of 20 Kelvin, the system band width of 2 GHz, and the output power
01
---
;. . . . . . ..: """* I
'0'
i 3 9 m. f '"$Ot *:. . ,
-"
OOOli
"""'
1000
""""
10'
loa
""" """"' lo6 10'
' ' .
'.
toe
Ll__j lo9
/
"""'
.
Number of particle
Figure 4. Cooling time of stochastic cooling.
10.
,g
;
i
.
,
:
: ',
;:
,
. .
,.
.-
.
,
-
lo8
109
...........
.. ,i'
-;,y, '
,
,K
:"8mlio'
OOOOl 1000
......
.............. /..*
-
'
''"
los
"' lo6
""'
to'
""""
'I
Number of particle
10'~
of main amplifier of 10 kW. Since the number of radioactive isotopes for one injection cycle is expected to be less than l o 7 particles, we can find that the cooling times in both directions are less than 100 msec. On the other hand, typical results of the electron cooling simulations shown in Fig. 5 tell us that the cooling time is roughly less than 1 sec. In this case, the electron beam temperature is 20 meV and 0.05 meV in transverse and longitudinal directions, respectively, and the electron beam current is 4 A. The stochastic cooling method is suitable to be used as pre-cooling for injected hot beams, which has large momentum spread and large emittance, because this is, in principle, the feedback system. The electron cooling method is effective for the pre-cooled beams. The combination of the stochastic cooling and the electron cooling makes the cooling time shorter than that for the case of only the electron cooling. The total cooling time is expected to be, roughly speaking, less than 1 sec for all RI beam.
Figure 5. Simulation results of electron cooling in ACR
The ACR is not only an accumulation/cooling ring but also an experimental ring. We provide an electron cooler, Schottky devices, four dispersive positions in arc section (the maximum dispersion of 4.52 m), two achromatic straight sections where the internal targets can be placed, so that the ACR can be used for various experiments. 1.3. Booster synchrotron ring (BSR) The accumulated/cooled RI beam is extracted from the ACR and injected into the BSR to boost up to the required energy, and the beam is immediately transported to the DSR as shown in the time chart of Fig. 3(c). The BSR has a circumference of 179.7 m, the maximum magnetic rigidity of 14.6 Tm, a repetition rate of 1 Hz and the acceleration time of 0.3 sec. The ion beams can be accelerated up to 1.4 GeV for proton and 0.8A GeV for uranium at maximum. Relatively wide range of RF frequency of 25-53 MHz is required to boost up to the maximum energy. Two kinds of extraction methods are provided, which are a fast (one turn) extraction and a slow extraction using 113 resonance technique. The fast extraction is for transporting the beams to the DSR, and the slow extraction is used for experiments at the experimental halls. The BSR can accept not
only ion beams coming from the ACR but also an electron beam. The electron beam can be accelerated from 300 MeV up to the required energy. 1.4. 300-MeV electron linac Depending on the way of use of the electron beam at the DSR, either the single bunch or the full bunch operation mode is chosen in the BSR. Corresponding to that, the operation mode of the e-linac is also changed to the short-pulse mode (1-nsec pulse length, 1Ampere peak current) or the long-pulse mode (5-pet pulse length, 100-mA peak current). The e-linac is driven by the RF frequency of 2856 MHz and the length is about 30 m including a SW type of pre-buncher, a T W type of buncher, and 5 constant gradient type of acceleration tube.
1.5. Double storage ring (DSR) As shown in Fig. 2, the DSR is a new type of experimental storage ring that consists of vertically stacked two rings, which are called e-ring and Ion-ring, respectively. It has a circumference of 269.5 m and two colliding points in long straight sections, which are called the colliding section and the merging section, respectively. The colliding section is prepared for the collision experiments with the crossing angle of 20 mrad. The RIelectron collision experiment, which is described later, is planned at this colliding section. The betatron function of the RI beams and the electron beam at the colliding point are designed to be 10 cm and 2 cm, respectively. On the other hand, the merging section with the crossing angle of 175 mrad is for the ion-ion merging experiments. In this section, we can make low energy collision experiments such as a fusion reaction. In this straight section, the RI-X-ray colliding section is also provided. An undulator is installed as a source of high brilliant X ray in this section as describing later. The DSR has different operation modes corresponding to different types of collision experiments. They are the colliding mode and the merging mode for the RI (ion) beams. For the electron beam, we will have the small emittance operation mode required to produce high brilliant X ray, and the large emittance operation mode required to get larger luminosity for the RIelectron collision experiment. According to requirements for the small emittance mode, a double bend achromatic (DBA) lattice is adopted in the arc section and the emittance of order of lo-' mrad is presently designed. On the other hand, the emittance for the mrad. Details of design of the DSR is large emittance mode is designed to be about described in Refs. [5, 61. 2. EXPERIMENS AT DSR
2.1. RI-electron collision experiment One of unique experiments at the DSR is the RI-electron collision [7].We are interested in the elastic scattering (e,el)from which we can determine the nuclear charge distribution of RI's. Physical interests are the proton skin structure, the neutron skinlhalo structure, difference in collective structure between protons and neutrons etc. in RI's which have unbalanced numbers of protons and neutrons. This experiment allows us to make systematic study on these problems. This kind of study, which has been performed only for light elements, can be extended to heavier elements. In our estimation, required range of momentum transfer for electron scattering is q=0.5
Figure 6. Expected yield of the electron scattering experiments
- 2 fm-I
to determine the charge distribution around the nuclear surface. This corresponds to the scattering angle from 10 deg. to 60 deg. in the laboratory frame for the case of electron beam energy of less than 1 GeV and the RI beam energy of less than 1A GeV. The essential point of this experiment is that how large luminosity can we have at the DSR. According to our calculation [a], we can precisely determine the nuclear charge distributions for RI's for the case of the luminosity of more than ~ m - ~ s - lthat , corresponds to isotopes having the life time of more than 1 min. Figure 6 shows examples of the expected yield of the electron scattering for one-week beam time. Here we assume the electron beam current in the DSR is 500 mA. 2.2. RI-X-ray collision experiment The purpose of this experiment is to determine the mean square nuclear charge radii and the electromagnetic moment by means of isotope shift measurements in the 2s2P (so called D l transition) atomic transition of the Li-like RI ions [7, 9, 101. We provide an undulator and an X-ray spectrometer as a monochromatic X-ray source placed near merging section in the e-ring as shown in Fig. 7. Output X-ray from the spectrometer is injected again into the Ion-ring and collides with the circulating Li-like RI ions at the detector position. Requirements on the X ray for this experiment are as follows. The X-ray energy is 30 - 800 eV to excite the D l transition of Z>40 elements, and the
. . .. .. .. .. . ."..
UUU
' U E *
.
L
~
-
~
UUV..-
a()@y -;
:
b ! -
-
c! -
.
\
\
X-ray D ~ t e c t o r
I!
,,,,
Merging section I,
b - ~ - t .--"" !
-
,),,
.f
,I
0
.l-"ml
m m SOP..
,,,",,, .-'L. - ~ ~ i - ' c f !
e 7
, 1.
X-ray Spectrometer
e
..,' ...
(10
\
Ion Ring .f
Undulator
Figure 7. Experimental setup for RI-X-ray collision experiment.
energy resolution is about AEx/Ex=10-4. The X ray intensity should be at least 1012 photons/sec/O.Ol % b.w. at the RI-X-ray colliding section. If we satisfy these conditions, we can measure the isotope shift for quite small number of RI beam stored in the DSR, even if only one ion is in the DSR. To get the required specifications of the X ray, we have to store 500-mA electron beam in the small emittance operation mode of the DSR. The specifications for the small emittance mode is the same like the third-generation synchrotron-light source. In such machine, instability of the electron beam is always big problem especially at lower energy. The instability is caused by the ring-broadband impedance and the narrow-band impedance at the high-Q cavities. We are now investigating the instability and designing the vacuum tube and cavities of the DSR.
3. CONCLUDING REMARKS In this paper, the outline of the MUSES accelerator system and typical experiments proposed at the DSR are described. The key issue of the e-RI collision experiment is the available luminosity. It is found that the luminosity can be reached up to ~m-~sec-l for the nuclei of which the lifetime is 1 min. To get such high luminosity, the important factor of the accelerator aspect is the cooling-stacking of the RI beam in the ACR, optimized lattice structure of the DSR, synchronous collision at the colliding point, and the beam-beam effects. Optimization on these problems are now in progress. For another experiment, RI-X-ray collision, the most important thing is how to get stable and largecurrent of the electron beam under the low-emittance operation mode in the DSR. The special design is needed for the components of the vacuum tube and cavities, and effective feedback system has to be installed in the DSR. The phase 1 of the RIBF project is under the construction and the phase 2, MUSES project, is expected to start the construction from 2001.
REFERENCES 1. Y. Yano et al., Proc. of PAC97 (1998) 930. 2. T. Katayama et al, Nucl. Phys. A626 (1997) 545c. 3. A. Goto et al., Proc. of 16th Cyclotron and Their Applications, to be published. 4. T. Kawaguchi et al., Proc. of PAC97 (1998) 3419. 5. N. Inabe et al., Proc. of PAC97 (1998) 1400. 6. N. Inabe et al., Proc. of EPAC98 (1998) 897. 7. I. Tanihata, Nucl. Phys. A588 (1995) 253c. 8. Y. Batygin et al., RIKEN-AF-AC-10 (1998). 9. M. Wakasugi et al., Proc. of EPAC96 (1997) 611. 10. M. Wakasugi et al., Proc. of EPAC98 (1998) 1017.
ThisPageIntolyLfBk
ThisPageIntolyLfBk This Page Intentionally Left Blank
Hadron and Nuclear Physics with Electromagnetic Probes K. Maruyama and H. Okuno (Editors) 2000 Elsevier Science B.V.All rights reserved.
'
Summary of the symposium "KEK-Tanashi, Tanashi-shi, Tokyo 188: Japan First of all, on behalf of the organizing committee of the symposium, I would like to express our sincere thanks to all the speakers and participants for their stimulating talks and lively discussions, which made this meeting so fruitful. 1. AIM OF THE SYMPOSIUM
The aim of this symposium, which we have intended to achieve, is to discuss recent experimental and theoretical developments of hadron and nuclear physics and future directions of these fields. Especially, emphases are placed on the hadron and nucleus studies with electron and photon beams. The reason why we have chosen these topics is closely related to the history of our 1.3GeV electron synchrotron at Tanashi. The operation of the synchrotron was terminated in June this year after 37-years service for particle and nuclear physics. The main research areas in recent days at the electron synchrotron were studies of photonuclear reactions by using a high-duty tagged-photon beam and a large-acceptance magnetic spectrometer TAGX. Physics topics were the study of meson and baryon properties in nuclear medium and nucleon-nucleon correlations in nucleus. In this symposium, we intended to summarize these experimental results from our electron synchrotron together with recent experimental and theoretical developments of hadron and nuclear physics. Also we wanted to look forward the future of these fields which will be opened by new electron accelerator facilities around the world. 2. HISTORY OF THE KEK-TANASHI ELECTRON SYNCHROTRON
First, I want to introduce our electron synchrotron briefly. Figure 1 shows a picture of our electron synchrotron at early 60's. The diameter of the ring is about 10 m and the ring is composed of eight magnets. This synchrotron is a combined-function-type strong-focussing machine, which is very compact and easy to operate. The Tanashi electron synchrotron was constructed in 1961 as a first high- energy accelerator in our country. The experiment started in 1966 at the maximum energy of 1.3 GeV. In those days. a quark model was proposed by Gell-Mann and Zweig in order to explain the mass spectra of baryon and meson resonances. Therefore, the main experimental topics were the study of electromagnetic structure of nucleon resonances through meson photoproductions. Especially, in 1970's and 1980's, the polarization measurements such as the polarized-beam asymmetry, the polarized target asymmetry and the recoil-proton
Figure 1. KEK-Tanashi 1.3-GeV electron synchrotron.
polarization were extensively measured for each single-pion production process. Through the partial wave analysis, photo-coupling parameters of nucleon resonances were analyzed, which contributed to establish the quark structure of nucleon resonances. In addition, it should be mentioned that this electron synchrotron played a critically important role in the pioneering work of synchrotron radiation physics in our country in 1960-1970. Studies of elementary processes of meson photoproduction on nucleon ended after the main stream of Japanese high-energy community moved to the experiments at the KEK 12-GeV proton synchrotron at Tsukuba. Since then physics interests were shifted to the study of nucleus. The experiments of inelastic scattering of electrons, pion production on nucleus and photodisintegration of nucleus were made sporadically in 1970's and 1980's. Some interesting results were obtained as for the shell structure of nuclei, nucleon-nucleon correlation inside nucleus: the meson exchange current, etc.. However, accuracies of the experimental results were limited due to the small duty factor of the accelerator. Also experiments on these processes were limited to the inclusive measurements, where only a single particle in the final state was detected.
3. HIGH-DUTY TAGGED PHOTON BEAM AND TAGX SPECTROMETER In order to overcome these difficulties, we have developed a high-duty tagged-photon beam and a large-acceptance magnetic spectrometer called TAGX. These fascilities were intended to be used primarily for the detailed study of photonuclear reactions in the
Bremsstrahlung Beam
Tagglng System
y 3 Area TAG, Spectrometer
y 2 Area
Figure 2. Layout of the accelerator and experimental apparatus.
1-GeV region. The floor plan of the synchrotron 1990's is shown in Fig.2. Since the main magnets of our electron synchrotron was excited with a sinusoidal wave form at a repetition rate of 21 Hz, the beam spill was limited to be about 1 msec which corresponded to the duty factor of 2%. This was a standard for this type of the synchrotron. So, at various laboratories, experimentalists wished to improve the duty factor by introducing a beam stretcher ring, a microtron and a superconducting linear accelerator. In our case, we applied a very simple method to excite the all beam transport magnets to follow the wave form of the synchrotron magnet. As a results, we could obtain the duty factor of 20% . This is quite optimum from the viewpoint of the data aquisition rate when we use a large-acceptance spectrometer like TAGX, where a large number of detector components are used. I do not go into details of TAGX because speakers on the
TAGX experiments already presented the detailed description of the TAGX spectrometer. The TAGX experiments concentrated on the study of light nuclei starting from deuterium, 3He and 4He, to 12C. The physics topics are categolized as 1. Photodisintegration of light nuclei, 2. Pion production from nuclei, 3. K+ photoproduction, 4. p0 mass modification in nuclear medium, where the results of these experiments were presented a t this symposium. These experiments were the first pioneering work which used a large-acceptance spectrometer for the photon beam. 4. TOPICS AT THE SYMPOSIUM
These TAGX experiments in mind, we organized the symposium topics as follows; 1. Mesons in Nuclear Medium, 2. Nucleon Resonances in Nuclei, 3. Strangeness Physics, 4. NN Correlations and Few-body Physics, 5. Nucleon Structure Studied by High-energy Electrons, 6. New Facilities. As for the topics 1-4, the TAGX collaboration presented their results. We wanted to discuss these results together with recent experimental and theoretical developments. The topics 5 is a study of nucleon structure studied by high-energy electrons, in which our KEK-Tanashi group led by S. Yamada is working at HERA. A deep understanding of the proton structure including a spin structure is a key issue in QCD. Therefore it was very interesting to hear a beautiful review talk by A. Levy on the recent understanding of proton structure learned from high-energy collider experiments. Since the energy range covered by this symposium is very wide, from MeV t o 100 GeV, it is beyond my ability to summarize all these physics topics. So I just want to say a few words on the limited items, Mesons in Kuclear Medium. At the opening session of this symposium, T . Hatsuda made a nice introductory talk on hadron and nuclear physics from a QCD point of view. In his summary, he claimed that QCD is a theory of everything in strong interactions, from soft, no-perturbative QCD to hard, perterbative QCD. He pointed out that three physics issues which should be investigated are; 1. QCD at extreme conditions (high temperature and high density), 2. Interplay between hard and soft QCD, 3. In-medium hadrons. In order to study hadrons in nuclear matter, the spectral function is a direct link between the theory and experiments. The particles to be looked at are vector mesons p, w and 4, and a scaler meson a. After Hatsuda's talk, R. Rapp described vector mesons in medium and dileptons in heavy-ion collisions from the theorist point. One example of calculation shows the expected spectral function for p, w and $. Then the question is how t o measure these
spectral functions in real experimental conditions. At this symposium, the status of three experiments are presented; 1. KEK-Tanashi ES: yA+pO+A* (p0-+7r+n-), 2. KEK-PS: P A + ~ A *, 3. GSI-HADES: e+e- pair spectroscopy, w in nuclei. These experiments are still in a preliminary stage or in progress. Although there are some hints on mass modification, more detailed experimental study is necessary for detailed comparison with QCD theories. 5. NEW FACILITIES AND FUTURE DIRECTIONS
Although our electron synchrotron has left from the hadron and nuclear physics arena, various new electron facilities came in with a unique beam characteristics. Today, W. Hersman told us the recent activities at Jefferson Lab. Yesterday, Mainz activities and Bonn ELSA activities were introduced by P. Grabmyer and E. Paul. These facilities are characterized by the continuous electron beam of 100% duty factor. The physics objectives which our electron synchrotron has pursued will be investigated more deeply and more extensively at these facilities. In our country, also two new electron accelerator facilities came into operation this year as learned from this afternoon talk by J. Kasagi and T. Hotta. These are the 1.2-GeV stretcher/booster ring at Tohoku and Laser-backscattered photon beam facility at Spring8. We are very glad to hear that the experiments started just at the time of our electron synchrotron's retirement. In addition to these, the symposium attendee heard with a great interest from T . Katayama (M.Wakasugi) about the RIKEN project of the unstable nuclear beam facility. This may open new frontiers on the study of unstable nuclei with electromagnetic probes. At KEK-Tsukuba, the hadron and nuclear physics experiments have long been investigated at the 12-GeV proton synchrotron. Results are presented by H. En'yo for 4 production, J. Chiba for A resonance in nuclei and 0. Hashimoto for the spectroscopy of hypernuclei. These studies by use of the hadron beam are complementary to the studies with the electromagnetic probes. A proposed JHF, Japan Hadron Facility, which is presented by our IPNS Director S. Yamada, is expected to deliver a good quality of pion, kaon and antiproton beams in future. I believe that these new facilities will play a key role in understanding the true nature of hadron and nucleus in the next century. At the end of the symposium, I would like to thank all the speakers and participants, especially participants from abroad, for their enthusiastic contribution to the success of the symposium. I hope everybody have spent a fruitful three days at Tanashi surrounded by a Japanese cultural environment. Please bring back nice memories on Tanashi symposium to your home. Also I would like to thank the advisory committee and organizing committee members, and symposium secretaries for their help in organizing this symposium. Without their sincere help, this symposium could not be realized.
ThisPageIntolyLfBk
ThisPageIntolyLfBk This Page Intentionally Left Blank
Symposium Program Oct. 25 (Mon.) Registration (8:30 - 9:20) Opening Session (9:20 - 1050) Chairperson; A. Masaike (Fukui IT), Scientific secretary; K. Oyama (Tokyo) S. Sugimoto (KEK-Tanashi) Opening address S. Yamada (KEK) Future projects of KEK T. Hatsuda (Kyoto) An introductory talk to the symposium
- Coffee Break (1050 - 11:10) Session I (Mesons in Nuclear Medium; 11:lO - 12:30, 14:OO - 15:20) Chairperson; K. Yazaki (Tokyo Woman's Christian), Scientific secretary; K. Oyama (Tokyo) (30+10) R. Rapp (SUNY) Vector mesons in nuclear medium K. Maruyama (CNS, Tokyo) p0 meson in the nucleus (30+10) - Lunch (12:30 - 14:OO) -
H. En'yo (Kyoto) J. Friese (TU Munchen)
$ production at KEK HADES at GSI
Session I1 (Nucleon Resonances in Nuclei and Related Topics; 15:20 - 16:10, 16:40 - 18:20) Chairperson; K. Nakai (Science U Tokyo), Scientific secretary; Y. Umemoto (Nara Women's) H. Yamazaki (Tohoku) S, ,(1535) resonance in nuclei (20+5) J. Chiba (KEK) Delta in nuclei excited by hadronic processes (20+5) - Coffee Break (16:lO - 16:40) -
G. Huber (Regina)
I.T. Cheon (Yonsei) H. Utsunomiya (Konan) I.A. Pshenichnov (INR)
Physics of double Delta excitation in the deuteron and 3 ~ e (20+5) Vector meson contribution to pion photoproduction on the nucleon (20+5) Photoneutron cross section measurement on ' ~ by e inverse Compton scattering of laser photons
(20+5)
Nuclear disintegration induced by virtual photons at heavy-ion colliders
(20+5)
- Reception (18:30 - 20:OO) -
Oct. 26 (Tue.) Session I11 (Strangeness Physics; 9:00 - 10:20, 10:40 - 12:20 ) Chairperson; T. Motoba (Osaka E-C), Scientific secretary; M. Yamaguchi (Ehime) K. Maeda (Tohoku) K' photoproduction on nuclei E.Ya. Paryev (INR) Comment T. Mart (Indonesia) Theoretical aspects of strangeness electroproduction H. Yamamura (Okayama US)
Hyperon polarization in kaon photoproduction from the deuteron
(20+5) (10) (20+5) (15+5)
- Coffee Break (10:20 - 10:40) E. Paul (Bonn) 0 . Hashimoto (Tohoku) H. Kohri (Osaka)
Physics of associated strangeness production experiments (30+ 10) at ELSA Retrospect and prospect of hypernuclear physics (30+10) Spin-orbit splitting of ' ',c (15+5)
Session I V (N-N Correlations and Few-body Physics; 1350 - 15:55, 16:25 - 17:30) Chairperson; T. Suzuki (Fukui), Scientific secretary; N. Naka (Ehime) T. Suda (RIKEN) e 4 ~at eTAGX Photodisintegration reactions of ' ~ and (SO+ 10) G. Orlandini (Trento)
S. Hirenzaki (Nara Women's)
Photonuclear cross sections of three-nucleon systems and the role of three-nucleon forces (30+10) e Quasi-deuteron picture for %Ie and 4 ~Photodisintegration (20+5)
- Coffee Break (15:55 - 16:25) P. Grabmayr (Tiibingen)
Two-nucleon emission experiments at Mainz searching (30+10) for NN correlations
E.L. Lomon (MIT)
Quark substructure and isobar effects on deuteron form factors (20+5) - Banquet (19:OO - 21:OO) -
Oct. 27 (Wed.) Session V (Nucleon Structure Studied by High-Energy Electrons; 9:00 - 10:40, 11:OO - 12:05) Chairperson; H. Abramowicz (Tel Aviv), Scientific secretary; M. Chiba (Tohoku) A. Levy (Tel Aviv) The proton and the photon: who is probing whom in (50+10) electroproduction ? K. Nagano (KEWDESY) ~ i ~neutralh - and ~ charged-current ~ reactions at HERA (30+10)
- Coffee Break (10:40 - 11:OO) Y. Sakemi (Tokyo IT) W. Meyer (Bochum)
Spin structure of the nucleon studied by HERMES A first measurement of the GDH-sum rule at MAMI
(30+10) (20+5)
- Lunch (12:05 - 13:35) Session VI (New Facilities; 13:35 - 15:55) Chairperson; J.C. Kim (Seoul National), Scientific secretary; Y. Miyake (Osaka) W. Hersman (New Hampshire) Topics of few-body physics at J-Lab Y. Yamaguchi (Tokyo) Use of the polarized photon beams at high energies J. Kasagi (Tohoku) Nuclear physics experiments with 1.2-GeV STB ring at LNS, Tohoku T. Hotta (RCNP, Osaka) LEPS at Spring-8 T. Katayama (TokyoIRIKEN) MUSES Project
(15+5) (15+5) (15+5)
Closing Session (1555 - 16:15) Chairperson; Y. Sumi (Hiroshima Int.), Scientific secretary; Y. Miyake (Osaka) H. Okuno (KEK-Tanashi) Summary talk
(20)
Oct. 28 (Thu.) Tour of KEK-facilities on request
(50+10) (15+5)
ThisPageIntolyLfBk
ThisPageIntolyLfBk This Page Intentionally Left Blank
List o f Partichants Abramowicz, Halina
Tel Aviv University
[email protected]
Akaishi, Yoshinori
KEK
[email protected]
Asakawa, Masayuki
Nagoya University
[email protected]
Cheon, 11-Tong
Yonsei University
itcheon @phya.yonsei.ac.kr
Chiba, Junsei
KEK
[email protected]
Chiba, Masami
Tohoku University
[email protected]
Emura, Tsuneo
Tokyo Univ. of Agriculture and Technology [email protected]
En'yo, Hideto
Kyoto University
[email protected]
Endo, Ichita
Hiroshima University
[email protected]
Endo, Satoru
Hiroshima University
[email protected]
Friese, Juergen
Technische Universitaet Muenchen
[email protected]
Grabmayr, Peter
University of Tuebingen
[email protected]
Hashimoto, Osamu
Tohoku University
[email protected]
Hatsuda, Tetsuo
Kyoto University
hatsuda@ruby,scphys.kyoto-u.acjp
Hersman, Bill
University of New Hampshire
[email protected]
Hirenzaki, Satoru
Nara Women's University
[email protected]
Hotta, Tomoaki
Osaka University
[email protected]
Huber, Garth
University of Regina
[email protected]
Inuzuka, Masahide
Tokyo Metropolitan University
[email protected]
Jeong, Moon Taeg
Dongshin University
[email protected]
Kasagi, Jirohta
Tohoku University
[email protected]
Katayama, Takeshi
University of Tokyo
[email protected]
Kim, Jong Chan
Seoul National University
jckim@phya,snu.ac.kr
Kishimoto, Tadafurni
Osaka University
[email protected],ac.jp
Kohri, Hideki
Osaka University
[email protected],osaka-u.ac.jp
Komatsubara, Takeshi
KEK
[email protected]
Lee, Su Houng
Yonsei University
[email protected]
Levy, Aharon
Tel Aviv University
levy @alzt.tau.ac.il
Lomon, Earle
Massachusetts Institute of Technology
[email protected]
Maeda, Kazushige
Tohoku University
[email protected]
Mart, Terry
University of Indonesia
Maruyarna, Koichi
University of Tokyo
Masaike, Akira
Fukui University of Technology
Meyer, Werner
Ruhr Uni. Bochum
Miyake, Yoichiro
Osaka University
Monmatsu, Osarnu
KEK
Motoba, Toshio
Osaka Electro-Communication University
Murata, Yojiro
Musashino Women University
Muto, Masayuki
KEK
Nagano, Kunihiro
KEWDESY
Naka, Norio
Ehime University
Nakai, Kozi
Science University of Tokyo
Nakayoshi, Kazuo
KEK
Niki, Kazuaki
KEK
Nomura, Toru
KEK
Oka, Makoto
Tokyo Institute of Technology
Okuno, Hideki
KEK
Orlandini, Giuseppina
University of Trento
Oyama, Ken
University of Tokyo
Paryev, Eduard
INR, Russian Academy of Sciences
Paul, Ewald
University of Bonn
Pshenichnov, Igor
INR, Russian Academy of Sciences
Rangacharyulu, Chary
University of Saskatchewan
Rapp, Ralf
SUNY
Sakaguchi, Atsushi
Osaka University
Sakamoto, Koh
Kanazawa University
Sakerni, Yasuhiro
Tokyo Institute of Technology
Sasaki, Atsushi
Akita University
Shibata, Seiichi
Kyoto University
Shibata, Toshiaki
Tokyo Institute of Technology
Suda, Toshimi
Institute of Physical and Chemical Research [email protected]
Sugimoto, Shojiro
KEK
[email protected]
Sumi, Yoshio
Hiroshima International University
[email protected]
Suzuki, Toshio
Fukui University
[email protected]
Tamae, Tadaaki
Tohoku University
[email protected]
Tezuka, Hirokazu
Toyo University
[email protected]
Tokushuku, Katsuo
KEK
[email protected]
Ueda, Tamotsu
Ehime University
[email protected]
Ukai, Kumataro
KEK
ukai@tanashi,kek.jp
Umemoto, Yukiko
Nara Women's University
[email protected]
Utsunomiya, Hiroaki
Konan University
[email protected]
Wada, Yoshichika
Meiji Pharmaceutical University
[email protected]
Yamada, Sakue
KEK
[email protected]
Yamaguchi, Masahiro
Ehime University
[email protected]
Yamaguchi, Yoshio
University of Tokyo
yy [email protected]
Yamamura, Hisahiko
Okayama University of Science
[email protected]
Yamashita, Haruo
Shizuoka Cancer Center
[email protected]
Yamazaki, Hirohito
Tohoku University
[email protected]
Yamazaki, Yuji
KEK
[email protected]
Yazaki, Koichi
Tokyo Woman's Christian University
[email protected]
Yoshida, Katsuhide
Hiroshima University
[email protected]
ThisPageIntolyLfBk
This Page Intentionally Left Blank
