Photoproduction of phi(1020) mesons on the proton at large momentum transfer

Abstract

The cross section for $\phi$ meson photoproduction on the proton has been measured for the first time up to a four-momentum transfer -t = 4 GeV^2, using the CLAS detector at the Thomas Jefferson National Accelerator Facility. At low four-momentum transfer, the differential cross section is well described by Pomeron exchange. At large four-momentum transfer, above -t = 1.8 GeV^2, the data support a model where the Pomeron is resolved into its simplest component, two gluons, which may couple to any quark in the proton and in the $\phi$. Comment: 5 pages; 7 figures

Full text

Page 1

arXiv:hep-ex/0006022v1 16 Jun 2000
Photoproduction of φ(1020) Mesons on the Proton at Large Momentum Transfer
E. Anciant,1T. Auger,1G. Audit,1M. Battaglieri,2J.M. Laget,1C. Marchand,1G.S. Adams,22M.J. Amaryan,35
M. Anghinolfi,2D. Armstrong,7B. Asavapibhop,27H. Avakian,17S. Barrow,10K. Beard,15M. Bektasoglu,21
B.L. Berman,12N. Bianchi,17A. Biselli,22S. Boiarinov,14W.J. Briscoe,12W. Brooks,24V.D. Burkert,24
J.R. Calarco,28G. Capitani,17D.S. Carman,20B. Carnahan,5C. Cetina,12P.L. Cole,32A. Coleman,7J. Connelly,12
D. Cords,24P. Corvisiero,2D. Crabb,33H. Crannell,5J. Cummings,22P.V. Degtiarenko,24L.C. Dennis,10
E. De Sanctis,17R. De Vita,2K.S. Dhuga,12C. Djalali,31G.E. Dodge,21D. Doughty,6 24P. Dragovitsch,10
M. Dugger,3S. Dytman,29Y.V. Efremenko,14H. Egiyan,7K.S. Egiyan,35L. Elouadrhiri,6 24L. Farhi,1
R.J. Feuerbach,4J. Ficenec,34T.A. Forest,21A. Freyberger,24H. Funsten,7M. Gai,26M. Gar¸ con,1G.P. Gilfoyle,30
K. Giovanetti,15P. Girard,31K.A. Griffioen,7M. Guidal,13V. Gyurjyan,24D. Heddle,6 24F.W. Hersman,28
K. Hicks,20R.S. Hicks,27M. Holtrop,28C.E. Hyde-Wright,21M.M. Ito,24D. Jenkins,34K. Joo,33
M. Khandaker,19 24D.H. Kim,16W. Kim,16A. Klein,21F.J. Klein,24M. Klusman,22M. Kossov,14L.H. Kramer,11 24
S.E. Kuhn,21D. Lawrence,27A. Longhi,5K. Loukachine,24R. Magahiz,4R.W. Major,30J. Manak,24
S.K. Matthews,5S. McAleer,10J. McCarthy,33K. McCormick,1J.W.C. McNabb,4B.A. Mecking,24M. Mestayer,24
C.A. Meyer,4R. Minehart,33R. Miskimen,27V. Muccifora,17J. Mueller,29L. Murphy,12G.S. Mutchler,23
J. Napolitano,22B. Niczyporuk,24R.A. Niyazov,21A. Opper,20J.T. O’Brien,5S. Philips,12N. Pivnyuk,14
D. Pocanic,33O. Pogorelko,14E. Polli,17B.M. Preedom,31J.W. Price,25L.M. Qin,21B.A. Raue,11 24A.R. Reolon,17
G. Riccardi,10G. Ricco,2M. Ripani,2B.G. Ritchie,3F. Ronchetti,17P. Rossi,17F. Roudot,1D. Rowntree,18
P.D. Rubin,30C.W. Salgado,19 24V. Sapunenko,2R.A. Schumacher,4A. Shafi,12Y.G. Sharabian,35A. Skabelin,18
C. Smith,33E.S. Smith,24D.I. Sober,5S. Stepanyan,35P. Stoler,22M. Taiuti,2S. Taylor,23D. Tedeschi,31
R. Thompson,29M.F. Vineyard,30A. Vlassov,14H. Weller,9L.B. Weinstein,21R. Welsh,7D. Weygand,24
S. Whisnant,31M. Witkowski,22E. Wolin,24A. Yegneswaran,24J. Yun,21B. Zhang,18J. Zhao18
(The CLAS Collaboration)
1CEA Saclay, DAPNIA-SPhN, F91191 Gif-sur-Yvette Cedex, France
2Istituto Nazionale di Fisica Nucleare, Sezione di Genova e Dipartimento di Fisica dell’Universita, 16146 Genova, Italy
3Arizona State University, Department of Physics and Astronomy, Tempe, AZ 85287, USA
4Carnegie Mellon University, Department of Physics, Pittsburgh, PA 15213, USA
5Catholic University of America, Department of Physics, Washington D.C., 20064, USA
6Christopher Newport University, Newport News, VA 23606, USA
7College of William and Mary, Department of Physics, Williamsburg, VA 23187, USA
8Department of Physics and Astronomy, Edinburgh University, Edinburgh EH9 3JZ, United Kingdom
9Duke University, Physics Bldg. TUNL, Durham, NC27706, USA
10Florida State University, Department of Physics, Tallahassee, FL 32306, USA
11Florida International University, Miami, FL 33199, USA
12George Washington University, Department of Physics, Washington D. C., 20052 USA
13Institut de Physique Nucleaire d’Orsay, IN2P3, BP 1, 91406 Orsay, France
14Institute of Theoretical and Experimental Physics, 25 B. Cheremushkinskaya, Moscow, 117259, Russia
15James Madison University, Department of Physics, Harrisonburg, VA 22807, USA
16Kyungpook National University, Department of Physics, Taegu 702-701, South Korea
17Istituto Nazionale di Fisica Nucleare, Laboratori Nazionali di Frascati, P.O. 13, 00044 Frascati, Italy
18M.I.T.-Bates Linear Accelerator, Middleton, MA 01949, USA
19Norfolk State University, Norfolk VA 23504, USA
20Ohio University, Department of Physics, Athens, OH 45701, USA
21Old Dominion University, Department of Physics, Norfolk VA 23529, USA
22Rensselaer Polytechnic Institute, Department of Physics, Troy, NY 12181, USA
23Rice University, Bonner Lab, Box 1892, Houston, TX 77251
24Thomas Jefferson National Accelerator Facility, 12000 Jefferson Avenue, Newport News, VA 23606, USA
25University of California at Los Angeles, Physics Department, 405 Hilgard Ave., Los Angeles, CA 90095-1547, USA
26University of Connecticut, Physics Department, Storrs, CT 06269, USA
27University of Massachusetts, Department of Physics, Amherst, MA 01003, USA
28University of New Hampshire, Department of Physics, Durham, NH 03824, USA
29University of Pittsburgh, Department of Physics, Pittsburgh, PA 15260, USA
30University of Richmond, Department of Physics, Richmond, VA 23173, USA
31University of South Carolina, Department of Physics, Columbia, SC 29208, USA
32University of Texas at El Paso, Department of Physics, El Paso, Texas 79968, USA
33University of Virginia, Department of Physics, Charlottesville, VA 22903, USA
34Virginia Polytechnic and State University, Department of Physics, Blacksburg, VA 24061, USA
35Yerevan Physics Institute, 375036 Yerevan, Armenia

Page 2

The cross section for φ meson photoproduction on the proton has been measured for the first time
up to a four-momentum transfer −t = 4 GeV2, using the CLAS detector at the Thomas Jefferson
National Accelerator Facility. At low four-momentum transfer, the differential cross section is well
described by Pomeron exchange. At large four-momentum transfer, above −t = 1.8 GeV2, the data
support a model where the Pomeron is resolved into its simplest component, two gluons, which may
couple to any quark in the proton and in the φ.
PACS : 13.60.Le, 12.40.Nn, 13.40.Gp
In this paper we report results of the first determina-
tion of the cross section for elastic φ photoproduction on
the proton, up to −t = 4 GeV2, in a kinematical domain
where the Pomeron may be resolved into its simplest 2-
gluon component. Due to the dominant ss component of
the φ, and to the extent that the strangeness component
of the nucleon is small, the exchange of quarks is strongly
suppressed.
The scarce existing experimental data for this reac-
tion [1–5] extend only to a momentum transfer of −t = 1
GeV2and are well described as a purely diffractive pro-
cess involving the exchange of the Pomeron trajectory in
the t channel [6]. At larger t, the small impact parameter
makes it possible for a quark in the vector meson and a
quark in the proton to become close enough to exchange
two gluons which do not have enough time to reinteract
to form a Pomeron. Such a model of the Pomeron as two
non-perturbative gluons [7] matches the Pomeron model
up to −t = 1 GeV2, but predicts a different behavior at
higher t [8].
Large momentum transfers also select configurations in
which the transverse distances between the two quarks in
the vector meson and the three quarks in the proton are
small. In that case, each gluon can couple to different
quarks of the vector meson [8], as depicted in the middle
diagram of Fig. 1, as well as to two different quarks of
the proton [9] (bottom diagrams in Fig. 1). So, elastic φ
photoproduction at large t is a good tool to gain access
to the quark correlation function in the proton [10–12].
Measurements at such large four-momentum trans-
fers are now possible thanks to the continuous beam
of CEBAF at Jefferson Lab. This experiment was per-
formed using the Hall B tagged photon beam. The in-
cident electron beam, with an energy E0 = 4.1 GeV,
impinged upon a gold radiator of 10−4radiation lengths.
The tagging system, which gives a photon-energy res-
olution of 0.1% E0, is described in Ref. [13]. For this
experiment the photons were tagged only in the range
3.3-3.9 GeV. The target cell, a mylar cylinder 6 cm in
diameter and 18 cm long, was filled with liquid hydrogen
at 20.4 K.
The photon flux was determined with a pair spectrom-
eter located downstream of the target. The efficiency
of this pair spectrometer was measured at low intensity
(105γ/s in the entire bremsstrahlung spectrum) by com-
parison with a total absorption counter (a lead-glass de-
tector of 20 radiation lengths). During data taking at
high intensity (6×106tagged γ/s), the number of coinci-
dences, true and accidental, between the pair spectrome-
ter and the tagger was recorded by scalers. The number
of photons lost in the target and along the beamline was
evaluated with a GEANT simulation. The correction is
of the order of 5%. The systematic uncertainty on the
photon flux has been estimated to be 3%.
2 gluons
correlations
Pomeron
FIG. 1.
or of 2 gluons in the photoproduction of the φ.
Diagrams representing the exchange of a Pomeron
The hadrons were detected in CLAS, the CEBAF
Large Acceptance Spectrometer [14]. It consists of a six-
coil superconducting magnet producing a toroidal field.
Three sets of drift chambers allow the determination of
the momenta of the charged particles with polar angles
from 10 to 140 degrees. A complete coverage of scintil-
lators allows the discrimination of particles by a time-of-
flight technique as described in Ref. [15]. As the field in
the magnet was set to bend the positive particles out-
wards, the K−, from the φ → K+K−decay, were identi-
fied by the missing mass of the reaction γp −→ pK+(X).
In Fig. 2, a well-identified K−peak can be seen above
a background which corresponds to a combination of
misidentified particles, the contribution of multi-particle
channels and accidentals between CLAS and the tagger.
The background is eliminated by subtracting the counts
in the sidebands, indicated in the figure, from the main
peak, in each bin in t (determined by the four-momentum

Page 3

of the detected proton). The contribution of the side-
bands to the K+K−mass spectrum is shown in Fig. 3.
Note that it is very small under the φ peak.
Missing mass squared M (X) (GeV )
Missing mass squared M2(X) in the reaction
γp −→ pK+(X).
2
2
Number of counts
00.10.20.30.40.50.6
0
1000
2000
3000
4000
5000
6000
7000
8000
9000
sidebands
K-
FIG. 2.
Number of counts
Invariant mass squared M (K K ) (GeV )
FIG. 3. The K+K−mass spectrum, before the sideband
subtraction.Slashes:lower mass sideband contribution.
Anti-slashes: higher mass sideband contribution.
2
2
-
+
0.811.2 1.41.61.822.22.4
0
2000
4000
6000
8000
10000
sideband sum
In the Dalitz plot (Fig. 4) of invariant masses squared
M2(K+K−) versus M2(pK−), two resonant contribu-
tions to the pK+K−channel can be clearly seen, namely
the pφ and the Λ∗(1520)K+channels. A cut at M2(pK−)
> 2.56 GeV2further suppresses the contribution of the
Λ∗production to the K+K−mass spectrum.
Invariant mass squared M (pK ) (GeV )
Invariant mass squared M2(K+K−) as function
of M2(pK−).
2
2
-
Invariant mass squared M (K K ) (GeV )
2
2
-
+
22.533.544.55
5.5
1
1.5
2
2.5
3
Φ
Λ*(1520)
FIG. 4.
Number of counts
Invariant mass squared M (K K ) (GeV )
FIG. 5. Invariant mass K+K−for selected values of the
four-momentum transfer t (GeV2), after sideband subtrac-
tion. The curves show the continuum obtained from the fits
discussed in the text.
2
2
-
+
0
100
200
300
400
500
0
20
40
60
80
0.911.11.21.31.4
0
5
10
15
20
25
Background:
f + phase space
0
flat
0.7 < -t < 0.9
1.6 < -t < 2.1
2.7 < -t < 3.5
The resulting mass spectra are shown in Fig. 5 for se-
lected bins in t. The peak of the φ(1020) clearly shows
up over a K+K−continuum contribution which must
be subtracted.The φ events are selected by the cut

Page 4

1.0 < M2(K+K−) < 1.1 GeV2. The CLAS acceptance
in the forward direction limits the data set to values of
−t larger than 0.4 GeV2. This experiment extends the
measured range up to −t = 4 GeV2.
The detector efficiency depends on four variables: Eγ,
t, θcm
of the φ). A GEANT simulation program, which takes
into account the entire CLAS setup, was used to calculate
the detector efficiency, taking into account in an iterative
way the experimentally observed variation of the cross
section as a function of these variables. No variations
of the cross-section against Eγ and φcm
This efficiency varies from 0.15 to 0.25. The accuracy
of the simulation has been evaluated to be 5% from a
comparison between the real data and the Monte Carlo
simulation [16] for the channel γp −→ pπ+π−, where the
statistics are very high.
The continuum background has been subtracted as-
suming an isotropic distribution in θcm
ses for its variation against the mass M(K+K−): i) a flat
contribution, and ii) a phase space distribution plus a
contribution of the f0(980) decaying into two kaons (the
mass of the f0is below the two-kaon threshold but be-
cause of its ∼ 60 MeV width, the tail of the Breit-Wigner
can contribute). Its contribution was determined by fit-
ting the K+K−mass spectrum (up to M2(K+K−) = 1.2
GeV2in each bin in t) with two components: the back-
ground itself and a Breit-Wigner describing the φ meson
peak.
The results for the cross section are the average be-
tween the two values obtained according to these two
background hypotheses, with the difference being taken
as an estimate of the systematic uncertainty due to the
subtraction of the K+K−continuum production. The
data are integrated over the full tagging energy range
(3.3 GeV < Eγ< 3.9 GeV).
The cross sections dσ/dt versus t for the φ photopro-
duction are presented in Fig. 6, for eight bins in t. For val-
ues of −t around 1 GeV2, our data are in good agreement
with the most precise published data. The dotted curve
corresponds to Pomeron exchange [10]. The solid curve
corresponds to the exchange of two non-perturbatively
dressed gluons [9,10] that may couple to any quark in
the φ meson and in the proton. It includes quark corre-
lations in the proton, assuming the simplest form of its
wave function [17]: three valence quarks equally sharing
the proton longitudinal momentum. The parameters in
this model are fixed by the analysis of other independent
channels. It also reproduces the data recently recorded
at HERA [21] up to −t = 1 GeV2(see Ref. [10]).
The solid curve gives a good description of the ex-
periment over the entire range of t except for the last
point at −t = 3.9 GeV2. Here, one approaches the kine-
matical limit and u-channel nucleon exchange may con-
tribute [10]. Performing the experiment at higher average
energy (4.5 GeV) would push the u-channel contribution
K+ and φcm
K+ (the decay angles of the K+in the c.m.
K+ were observed.
K+and two hypothe-
to higher values of |t| (6 GeV2) and leave a wider window
to study two-gluon exchange mechanisms.
FIG. 6. The differential φ photoproduction cross-section
versus the four-momentum transfer t (see text for the explana-
tion of the curves). The error bars displayed are the quadratic
sum of statistical and systematic uncertainties which include
3% for normalization, 5% for acceptance and 5-15% for back-
ground subtraction.
The dot-dashed curve includes the u-channel contribu-
tion with the choice gφNN = 3 for the φNN coupling
(the addition of the u-channel amplitude to the domi-
nant t-channel amplitude does not lead to double count-
ing, because the former relies on quark exchange and the
latter relies on gluon exchange). This value comes from
the analysis of nucleon electromagnetic form factors [18]
as well as nucleon-nucleon and hyperon-nucleon scatter-
ing [19]. It is higher than the value gφNN= 1 predicted
from SU(3) mass splitting or ω−φ mixing [20], thus con-
firming evidence for additional OZI-evading processes at
the φNN vertex.
The predictions of two other models are also presented
in Fig. 6. Both treat the gluon exchange in perturbative
QCD (this leads to the characteristic t−7behavior of a
hard scattering) and use a diquark model to take into ac-
count quark correlations in the proton (this fixes the mag-
nitude of the cross section). Berger and Schweiger [11]
(upper dashed curve) use a wave function which leads to a
good accounting of Compton scattering and nucleon form
factors, while Carimalo et al. [12] (lower dashed curve)

Page 5

use a wave function which fits the cross section of the
γγ → pp reaction. Above −t = 2 GeV2our data rule out
the t dependence of these diquark models, demonstrating
that the asymptotic regime is not yet reached. Recently,
a new anomalous Regge trajectory associated with the
f1(1285) meson has been proposed [22]. It reproduces
the HERA [21] data (−t < 1 GeV2), but its momentum
dependence is too steep to reproduce our high t data.
1.6< -t <2.1
180020406080
Θcm
100 120 140 160
0
25
20
40
60
80
100
120
0
5
10
15
20
0
5
10
15
20
2.7< -t <3.5
K+
0.7< -t <0.9
dN/dΘ
FIG. 7. Angular distributions (corrected for acceptance)
dN/dθ of the K+, in the helicity frame, are compared to the
prediction of SCHC.
Figure 7 shows the decay angular distributions of the
φ in the helicity frame [23] for selected bins in t. The
sideband contributions have been subtracted, but not
the K+K−continuum contribution. Up to −t = 2.1
GeV2they follow a sin2θd(cosθ) dependence, in agree-
ment with s-Channel Helicity Conservation (SCHC): a
real photon produces a φ meson with only transverse
components. Above −t = 2.7 GeV2, there is a viola-
tion of SCHC, likely to be associated with the u-channel
exchange and the interference between the φ and the S-
wave K+K−photoproduction amplitudes.
In conclusion, elastic photoproduction of φ mesons
from the proton was measured for the first time up to
−t = 4 GeV2. Below −t ≈ 1 GeV2, they cannot dis-
tinguish between the Pomeron exchange and the 2-gluon
exchange models which both agree with the existing data.
At high t, the predictions of these models differ by more
than an order of magnitude. Above −t ≈ 1.8 GeV2, our
data rule out the diffractive Pomeron and strongly favor
its 2-gluon realization. This opens a window to the study
of the quark correlation function in the proton.
We would like to acknowledge the outstanding efforts
of the staff of the Accelerator and the Physics Divisions at
JLab that made this experiment possible. This work was
supported in part by the French Commissariat ` a l’Energie
Atomique, the Italian Istituto Nazionale di Fisica Nucle-
are, the U.S. Department of Energy and National Sci-
ence Foundation, and the Korea Science and Engineering
Foundation.
[1] R.L. Anderson et al., Phys. Rev. D1, 27 (1970).
[2] H.J. Besch et al., Nucl. Phys. B70, 257 (1974).
[3] H.J. Berends et al., Phys. Lett. 56B, 408 (1975);
H.J. Berends et al., Nucl. Phys. B144, 22 (1978).
[4] D.P. Barber et al., Z. Phys. C12, 1 (1982).
[5] J. Ballam et al., Phys. Rev. D7, 3150 (1973).
[6] A. Donnachie and P.V. Landshoff, Phys. Lett. B185, 403
(1987).
[7] A. Donnachie and P.V. Landshoff, Nucl. Phys. B311, 509
(1989).
[8] J.-M. Laget and R. Mendez-Galain, Nucl. Phys. A581,
397 (1995).
[9] J.M. Laget, Physics and Instrumentation with 6–12 GeV
Beams, edited by S. Dytman, H. Fenker and P. Ross (Jef-
ferson Lab User Production, 1998), p. 57.
[10] J.-M. Laget, Preprint hep-ph/0003213, March 2000.
[11] C.F. Berger and W. Schweiger, Phys. Rev. D61, 114026
(2000).
[12] C. Carimalo, N. Artega-Romero and S. Ong, Eur. Phys.
J. C11, 685 (1999) .
[13] D.I. Sober et al., Nucl. Instr. and Meth. A440, 263
(2000).
[14] W. Brooks, Nucl. Phys. A663-664, 1077 (2000).
[15] E.S. Smith et al., Nucl. Instr. and Meth. A432, 265
(1999).
[16] Th.Auger, Th` ese de l’Universit´ e PARIS 7, April 1999 ;
E. Anciant et al., Measurement of γp → pπ+π−cross-
section in CLAS, CLAS-Analysis 99-002, March 1999.
[17] J.R. Cudell and B.U. Nguyen, Nucl. Phys. B420, 669
(1994).
[18] R.L. Jaffe, Phys. Lett. B229, 275 (1989) .
[19] M.M. Nagels, T.A. Rijken and J.J de Swart, Phys. Rev.
D20, 1633 (1979); Phys. Rev. D15, 2547 (1977); Phys.
Rev. D20, 744 (1975).
[20] P. Jain et al., Phys. Rev. D37, 3252 (1989).
[21] J. Breitweg et al., Preprint hep-ex/9910038, October
1999.
[22] N. Kochelev et al., Phys. Rev. D61, 094008 (2000).
[23] K. Schilling, P. Seyboth and G. Wolf, Nucl. Phys. B15,
397 (1970).