Документ взят из кэша поисковой машины. Адрес оригинального документа : http://lav01.sinp.msu.ru/~vlk/omegaeta.ps
Дата изменения: Thu Dec 21 15:00:37 2000
Дата индексирования: Mon Oct 1 19:43:30 2012
Кодировка: IBM-866

Observation of a New J PC = 1 + Isoscalar State
in the Reaction p ! !n at 18 GeV/c
E852 Collaboration
P. Eugenio, 3# G. S. Adams, 4 T. Adams, 5 Z. Bar-Yam, 3 J. M. Bishop, 5 V. A. Bodyagin, 2 B. B. Brabson, 6
D. S. Brown, 7 N. M. Cason, 5 S. U. Chung, 1 R. R. Crittenden, 6 J. P. Cummings, 3;4 A. I. Demianov, 2 S. Denisov, 8
V. Dorofeev, 8 J. P. Dowd, 3 A. R. Dzierba, 6 A. M. Gribushin, 2 J. Gunter, 6 R. W. Hackenburg, 1 M. Hayek, 3
E. I. Ivanov, 5 I. Kachaev, 8 W. Kern, 3 E. King, 3 O. L. Kodolova, 2 V. L. Korotkikh, 2 M. A. Kostin, 2 J. Kuhn, 4
R. Lindenbusch, 6 V. Lipaev, 8 J. M. LoSecco, 5 J. J. Manak, 5y J. Napolitano, 4 M. Nozar, 4 C. Olchanski, 1
A. I. Ostrovidov, 1;2;3 T. K. Pedlar, 7 A. Popov, 8 D. R. Rust, 6 D. Ryabchikov, 8 A. H. Sanjari, 5 L. I. Sarycheva, 2
E. Scott, 6 K. K. Seth, 7 N. Shenhav, 3 W. D. Shephard, 5 N. B. Sinev, 2 J. A. Smith, 4 P. T. Smith, 6 D. L. Stienike, 5
T. Sulanke, 6 S. A. Taegar, 5 S. Teige, 6 D. R. Thompson, 5 I. N. Vardanyan, 2 D. P. Weygand, 1y D. White, 4
H. J. Willutzki, 1 J. Wise, 7 M. Witkowski, 4 A. A. Yershov, 2 D. Zhao 7
1 Brookhaven National Laboratory, Upton, New York 11973
2 Nuclear Physics Institute, Moscow State University, Moscow, Russia 119899
3 Department of Physics, University of Massachusetts Dartmouth, North Dartmouth, Massachusetts 02747
4 Department of Physics, Rensselaer Polytechnic Institute, Troy, New York 12180
5 Department of Physics, University of Notre Dame, Notre Dame, Indiana 46556
6 Department of Physics, Indiana University, Bloomington, Indiana 47405
7 Department of Physics, Northwestern University, Evanston, Illinois 60208
8 Institute for High Energy Physics, Protvino, Russia 142284
(September 27, 2000)
Results are presented on a partial wave analysis of the ! nal state produced in p interactions
at 18 GeV/c where ! ! + 0 , 0 ! 2 , and ! 2 . We observe the previously unreported
decay mode !(1650) ! ! and a new 1 + meson state h1(1595) with a mass M = 1594(15) +10
60

MeV/c 2 and a width = 384(60) +70
100
MeV/c 2 . The h1(1595) state exhibits resonant-like phase
motion relative to the !(1650).
12.39.Mk, 11.80.Et, 13.25-k, 13.75.Gx COLLAB DRAFT 2.4
INTRODUCTION
Studies of meson spectra via strong decays of
hadrons provide insight regarding QCD at the conne-
ment scale. These studies have led to phenomenological
models such as the constituent quark model. However,
QCD demands a much richer spectrum of meson states
which includes extra states such as hybrids(qqg), mul-
tiquarks (qqq
q), and glueballs (gg or ggg). Experiment
E852 at Brookhaven National Laboratory is an experi-
ment in meson spectroscopy congured to detect both
neutral and charged nal meson states of p collisions
in a search for meson states incompatible with the con-
stituent quark model.
The apparatus was located at the Multi-Particle Spec-
trometer (MPS) of Brookhaven's Alternating Gradient
Synchrotron (AGS). The AGS delivered an 18 GeV/c
beam to a xed liquid hydrogen target at the MPS. The
MPS facility was augmented with additional detectors
designed specically for E852 which consisted of 3 inte-
gral regions: target, charged tracking, and downstream
regions(see Figure 1).
The target region was located in the middle of the MPS
DEA
Beam Veto
3 meters
Array (LGD)
CsI
E852 Plan View
Beam
Lead Glass
TPX 1н3
TCYL
Drift Modules
DM1н6 TDX4
FIG. 1. The plan view of E852 Apparatus located at
the Multi-Particle Spectrometer of Brookhaven's Alternating
Gradient Synchrotron.
1

dipole magnet(1 Tesla) and contained the following ele-
ments: a 30.5 cm long liquid hydrogen target, a four{
layer cylindrical drift{proportional wire chamber, and a
198 block CsI(Ti) barrel veto detector.
The downstream half of the MPS magnet housed the
main components of the charged tracking region. It con-
sisted of 3 proportional wire chambers (TPX1{3) and 6
drift chamber modules(DM1{6) each with seven{layers.
In addition to the tracking chambers, there were two scin-
tillation veto counters (CPVB and CPVC) and a window-
frame lead-scintillator sandwich veto detector (DEA).
The downstream region contained a 3000 element lead{
glass calorimeter (LGD) for detecting and measuring the
energy of the forward going gammas and a large drift
chamber (TDX4) located directly in front of the LGD
for tagging charged particles entering the LGD. More
detailed descriptions and general features of the E852
apparatus, data acquisition, event reconstruction and se-
lection are given in References [1{3].
The ! nal state is of considerable interest because it
has been virtually unexplored, and since it has not been
observed in the decay modes of known mesons. Also,
Close and Page [4], using an extension of the ux-tube
model of Isgur and Paton [5], suggest that the isoscalar
J PC = 1 qqg hybrid should decay dominantly to
and !.
The exclusive ! system has been studied by two
experiments previously. The GAMS Collaboration ob-
served less than 100 ! events produced in p inter-
actions [6] and claimed to observe a narrow (less than
50 MeV/c 2 wide) structure at 1650 MeV/c 2 . Photopro-
duction of nearly 100 ! nal state events was observed
by the Omega Photon Collaboration [7]. They reported
observing a peak in the ! spectrum at a mass of 1610
MeV/c 2 but with a width of 230 MeV/c 2 . Both experi-
ments suered from having too little data and could not
perform a partial wave analysis. We report here the re-
sults of a partial wave analysis of approximately 20000
exclusive ! events.
FEATURES OF THE DATA
During the 1995 AGS data run, a sample of 750 million
triggers was acquired of which 108 million were of a type
designed to enrich the yield from the reaction p !
n + 4 . About 6 million events containing + 4
were fully reconstructed.
The data were kinematically tted to select events con-
sistent with a n + Ї hypothesis. To eliminate bro-
ken and incorrectly hypothesized events, kinematic ts
with condence levels less than 5% were excluded.
There are six ways to combine the 4 's into pairs,
and one problem in interpreting the results of the kine-
matic t is deciding between these combinations for
ambiguous solutions. For example, twenty-nine percent
of the tted n + Ї events were also found to t the
n + Ї Ї hypothesis. Events of this type with a con-
dence level of at least 0:01% for the n + Ї Ї hy-
pothesis were removed from the data sample. A total of
113000 n + Ї events remained for further analysis.
The two prominent resonances seen in the + Ї in-
variant mass distribution (see Figure 2) are consistent
with the well known and !. A Gaussian plus a second
order polynomial t to these peaks results in a measured
mass and width of the of (550:4 0:2MeV /c 2 ) and
23:4 0:2MeV /c 2 and of the ! of (787:5 0:3MeV /c 2 )
and 38:3 0:3MeV /c 2 respectively. The mass values are
in good agreement with the PDG [8] mass values for the
and ! whereas the tted widths are a measure of the
experimental mass resolution of the E852 apparatus. The
! + Ї events have been selected for a study of the
system which is currently in progress.
Events/
10
MeV
GeV
Mass[ ]
+ 0 н
p p p
2
1000
Events/
10
MeV/c
h
w
p p p
+ 0
-
Mass[ ]
2
GeV/c
1.0
0.2 0.6 1.4 1.8
1
2
3
4
FIG. 2. The + Ї eective mass distribution. Two
prominent resonances are seen: the and the !. Events with
an invariant + Ї mass in the region of 750MeV /c 2 <
mass(3) < 830MeV /c 2 were used to select ! events.
Events with an invariant + Ї mass in region of
750MeV /c 2 < mass(3) < 830MeV /c 2 were used to se-
lect ! events. Figure 3 exhibits the J P = 1 nature of
the !. Displayed is the ! decay matrix element squared
( ! ) [9]:
j~ + ~ j 2
3
4 ( 1
9 M 2
3 M 2
) 2
where ~ + and ~ are the three-momentum vectors of
the the + and in the 3 rest-frame. The 3 decay
amplitudes are constructed from the momentum vec-
tors, and due to the overall negative intrinsic parity of
3, a J P = 1 decay amplitude has to be built out of a
2

pseudovector combination ~q [9]:
~q = ~ + ~ = ~ ~ 0 = ~ 0 ~ + (1)
The distribution should rise linearly for omega events,
whereas it would be constant for 3 events distributed
according to phase-space. The 3 background under the
! is considerable, but the decay information of the ! can
be used to weight ! events more highly than non-omega
events (see the section on Background Study).
l w
Events/0.01
200
400
600
800
0
Events/0.01
0.4
0.2 0.6 0.8 1
0
l w
FIG. 3. Distribution of events as a function of the omega
decay matrix element squared, . Pure J P = 1 events would
exhibit a distribution which linearly increases with whereas
phase-space events should be at.
In Figure 4, the ! invariant mass distribution of 19530
! nal state events is shown using 40 MeV/c 2 mass
bins. The invariant mass distribution rises rapidly near
threshold, then increases at a slower rate to a maximum
near 1600 MeV/c 2 . But, in general the mass distribu-
tion shows no clear evidence for structures indicating the
presence of resonant states.
PARTIAL WAVE ANALYSIS
A partial wave analysis (PWA) of the data was carried
out using the BNL PWA program. A general description
of the BNL PWA program is given in Reference [10]. The
methods used in the PWA are based on the formalism
of the Isobar Model [11]. For related material on this
formalism see References [12,13].
The data used in the PWA are shown in Figure 4.
A PWA, taking into account both nucleon spin- ip and
spin-non ip contributions, was performed in fteen 50
MeV/c 2 mass bins from 1320 to 2070 MeV/c 2 .
GeV
Events/
40
MeV
wh
Mass[ ]
2
GeV/c
2
Events/40
MeV/c
1.6 2.0 2.4
1.2
800
1600
1200
400
0
wh
Mass[ ]
FIG. 4. The ! eective mass distribution from the reac-
tion p ! !n (not corrected for acceptance).
An extensive series of tests were performed to judge
the stability of the PWA t. These include a systematic
study of the PWA by varying the:
allowed waves in the t,
t starting parameters,
data selection cuts, and
mass bin size.
In each test, the general features of the PWA t remained
unchanged.
An appropriate minimal set of partial waves needed to
describe the data was determined by performing PWA
ts with varying sets of partial waves. Initially, all par-
tial waves with L < 4 were introduced (L is the relative
orbital angular momentum between the ! and the ).
Partial waves were discarded from the t if their contri-
butions were negligible and if the removal had little eect
on the remaining features and goodness of the t. It was
determined that a minimum set of 11 waves, which in-
cluded a non-interfering isotropic background wave, were
required in order to achieve a reasonable agreement of the
experimental data and the PWA t results (see section
on PWA t quality). A notable feature was that the ex-
otic partial waves for J PC = 0 and 2 + were found
not to be required in the analysis.
Table I lists the partial waves included in the nal
PWA t. Also included was an isotropic, non-omega,
non-interfering background wave. The amplitudes are
expressed in the re ectivity basis which takes into ac-
count parity conservation in the production process by
a transformation of helicity states to eigenstates of the
3

TABLE I. The minimal set of partial waves required in the
PWA of the ! system.
Partial Waves J PC M L
unnatural parity exchange natural parity exchange
2 1 P 1 1 + P
3 0 F 1 + 0 + S
1 + 0 + D
1 + 1 + S
1 + 1 + D
2 0 + P
2 1 + P
3 1 + F
production plane re ection operator [14]. For p interac-
tions, the re ectivity coincides with the naturality of the
exchange particle and amplitudes of dierent re ectivity
= do not interfere.
QUALITY OF THE PWA FITS
The best test of the PWA t is to compare the events
generated according to the PWA to the data itself. For
this step, Monte Carlo(MC) events were subjected to a
simulation of the E852 apparatus and weighted with the
probability obtained from the PWA t. These weighted
MC events should mimic the data for successful PWA
ts. Since the ts were performed in independent mass
bins, the MC events were normalized to the number of ex-
perimentally observed events in the corresponding mass
bin:
NMC X
i
w i = N observed : (2)
Figures 5 and 6 shows a comparison of the weighted
MC events to the data for the: Gottfried-Jackson angles

and helicity
angles
h . The weighted MC events is
in good agreement with the data. A more stringent test
compares the angular moments, H(l; m; L; M) . The
results for this comparison (see Reference [1]) show good
agreement between data and predicted events. Overall,
the comparisons of the distributions demonstrate that
the data is very well described by the PWA t results.
The angular moments are the average value of the product
of the two Wigner D functions H(l; m; L; M) hD L
Mm
D l
m0(
0 )i.
0
-1 1
0.5
-0.5
0
100
300
200
0 3
-3 -1.5 1.5
0
100
300
200
q
Cos( )
Events/
0.125 b)
a)
Events/
0.04
f
FIG. 5. The quality of the t is shown by a comparison
of angular distributions for the data (with error bars) and
PWA t predicted Monte Carlo (dashed histogram): the Got-
tfried-Jackson frame a) Cos(), and b) .
100
0
-1 1
0
200
300
0.5
-0.5
0 3
-3 -1.5 1.5
150
250
350
50
q h
Cos( )
f h
Events/
0.125
Events/
0.04 c)
d)
FIG. 6. The quality of the t is shown by a comparison of
angular distributions for the data (with error bars) and PWA
t predicted Monte Carlo (dashed histogram): the omega he-
licity frame c) Cos(h ), and d) h .
4

PWA RESULTS
Figures 7 and 8 show the acceptance corrected partial
wave intensities. The most signicant partial waves in
the analysis are the 1 + 0 + S and 1 1 + P natural parity
waves. These two waves also exhibit relative phase mo-
tion indicative of resonant states (see Figure 9c). Note
that since the amplitudes which interfere have either
purely real or purely imaginary decay factors there is
an unavoidable but trivial ambiguity in the sign of the
relative phase. The remaining natural parity waves are
small and do not exhibit clear phase motion. The con-
tributions of the two unnatural parity exchange waves
tend to become more important at masses higher than
1700 MeV/c 2 . Their relative phase is not well deter-
mined and exhibits large errors; therefore it cannot be
reliably used to study resonance behavior. Nonetheless,
the peak in the 2 1 P intensity was tted with a Breit-
Wigner resonance shape giving a mass of 18308 MeV/c 2
and a width of 86 31MeV/c 2 . This is interesting since
the quark model predicts J PC = 2 states, yet none
have been observed. A prediction by Godfrey and Is-
gur suggests a mass of 1700 MeV/c 2 for the lowest lying
J PC = 2 qq state [15].
1.4 1.5 1.6 1.7 1.8 1.9 2.0
0
500
1000
1500
2000 Events: 2 1 P
нн н
1.4 1.5 1.6 1.7 1.8 1.9 2.0
0
500
1000
1500
2000 Events: 3 0 F
нн н
1.4 1.5 1.6 1.7 1.8 1.9 2.0
0
500
1000
1500
2000
Events: 1 1 P
нн +
1.4 1.5 1.6 1.7 1.8 1.9 2.0
0
500
1000
1500
2000
2500
Events: 1 0 S
+н +
1.4 1.5 1.6 1.7 1.8 1.9 2.0
0
500
1000
1500
2000 Events: 1 0 D
+н +
1.4 1.5 1.6 1.7 1.8 1.9 2.0
0
500
1000
1500
2000 Events: 1 1 S
+н +
1.4 1.6 1.8 2.0 1.4 1.6 1.8 2.0
1.4 1.6 1.8 2.0 1.4 1.6 1.8 2.0
1.4 1.6 1.8 2.0 1.4 1.6 1.8 2.0
wh
Mass[ ] wh
Mass[ ]
0
500
1000
1500
0
500
1000
1500
0
500
1000
1500
0
500
1000
1500
0
500
1000
1500
0
500
1000
1500
GeV/c 2
GeV/c 2
GeV/c 2
Events/50
MeV/c
2
Events/50
MeV/c
2
Events/50
MeV/c
2
FIG. 7. Acceptance-corrected partial waves intensities for
(J PC M L): 2 1 P; 3 0 F; 1 1 + P; 1 + 0 + S; 1 + 0 + D;
and 1 + 1 + S.
The only previously reported isoscalar J PC = 1 +
states are the h 1 (1170) and h 1 (1380) [8]. Both of these
states have masses much lower than the observed struc-
ture in the 1 + 0 + S wave. Also, the h 1 (1380), believed
to be the ideal mixed ss state, is not expected to be pro-
duced in p interactions due to OZI suppression. For
the isoscalar J PC = 1 , there are two states listed by
the Particle Data Group (PDG) in the 1000-2000 MeV/c 2
range: the !(1650) and the !(1420) [8]. The !(1650),
which has a PDG mass of 164924 MeV/c 2 and a width
of 220 35 MeV/c 2 , is in good agreement with the mass
and width of the observed structure in the 1 1 + P wave.
In order to determine if the 1 + 0 + S and the 1 1 + P
observed structures are consistent with resonance behav-
ior, a mass dependent analysis(MDA) of the PWA re-
sults was performed. The input quantities included the
1 + 0 + S and 1 1 + P partial wave intensities and their
relative phase for each mass bin. The errors were cal-
culated using the error matrix from the PWA t, which
takes into account correlations between the intensities
and relative phases. Relativistic Breit-Wigner forms
were used to parameterize the amplitudes for the two
waves. The parameters of the MDA t included the
Breit-Wigner masses, widths, and intensities. An ad-
ditional constant parameter was included to allow for a
relative constant production phase between the waves. A
series of mass dependent ts were performed for dierent
hypotheses and for the dierent phase solutions.
1.4 1.5 1.6 1.7 1.8 1.9 2.0
0
500
1000
1500
2000 Events: 1 1 D
+н +
1.4 1.5 1.6 1.7 1.8 1.9 2.0
0
500
1000
1500
2000 Events: 2 0 P
нн +
1.4 1.5 1.6 1.7 1.8 1.9 2.0
0
500
1000
1500
2000
Events: 2 1 P
нн +
1.4 1.5 1.6 1.7 1.8 1.9 2.0
0
500
1000
1500
2000
Events: 3 1 F
нн +
1.4 1.5 1.6 1.7 1.8 1.9 2.0
0
2000
4000
6000
8000
Events: All
1.4 1.5 1.6 1.7 1.8 1.9 2.0
0
500
1000
1500
2000
2500
3000
3500
Events: background
1.4 1.6 1.8 2.0 1.4 1.6 1.8 2.0
1.4 1.6 1.8 2.0 1.4 1.6 1.8 2.0
1.4 1.6 1.8 2.0 1.4 1.6 1.8 2.0
wh
Mass[ ] wh
Mass[ ]
0
500
1000
1500
0
500
1000
1500
0
500
1000
1500
0
500
1000
1500
0
500
1000
1500
0
500
1000
1500
GeV/c 2
GeV/c 2
GeV/c 2
Events/50
MeV/c
2
Events/50
MeV/c
2
Events/50
MeV/c
2
FIG. 8. Acceptance-corrected partial waves intensities for
(J PC M L): 1 + 1 + D; 2 0 + P; 2 1 + P; 3 1 + F; allwaves;
and background wave.
5

The primary goal of this analysis was to understand
the structure in the 1 + wave. A t of the 1 in-
tensity to a single Breit-Wigner resulted in a mass of
1700(20) MeV/c 2 and a width of 250(50) MeV/c 2 for
a 2 =(dof) of 0.77. These values are in reasonable agree-
ment with the PDG values of the !(1650), and since the
!(1650) is rather well known (see references [16]), we used
the 1 signal as a \calibration" by identifying it with
the !(1650) and xing the 1 Breit-Wigner parameters
to the PDG values and then tted the for the 1 + pa-
rameters. This identication would constitute the rst
observation of the ! decay mode of the !(1650). Un-
der these assumptions, a simultaneous MDA t to the
1 + intensity and the 1 + =1 relative phase resulted
in a 2 =(dof) of 0.71. The 1 + 0 + S mass and width re-
sulting from this t were 1594(15) +10
60
MeV/c 2 and
384(60) +70
100

MeV/c 2 respectively. The quoted errors
correspond to statistical and systematic uncertainties, re-
spectively. The systematic errors were estimated by t-
ting the PWA results obtained for dierent set of partial
waves which varied from 9 to 22 waves. Figure 9 displays
the results of the mass dependent analysis for this t: 9a,
9b, and 9c show results of this t overlayed on the partial
wave intensities and relative phase. Figure 9d shows the
absolute Breit-Wigner phases for 1 and 1 + and the
relative constant production phase. On the other hand,
xing the 1 parameters to the Breit-Wigner tted val-
ues (mass= 1700 MeV/c 2 and width = 250 MeV/c 2 )
and performing a similar MDA t resulted in a less ade-
quate 2 =(dof) of 1.28 and a 1 + 0 + S mass = 1600(20)
MeV/c 2 and = 455(80) MeV/c 2 .
If the 1 + 0 + S object is interpreted as being caused by
a single resonance, then this state, an h 1 (1595), does not
coincide with any known states. Possible interpretations
for an h 1 (1595) include an h 1 radial excitation, an h 1
hybrid, or a radial-hybrid mixture. In the Godfrey-Isgur
potential model [15], a mass of 1780 MeV/c 2 is predicted
for the 2 3 P 0 h 1 radial excitation. This is approximately
180 MeV/c 2 higher than the value from our analysis.
However this model also predicts a high mass for the
h 1 (1160) and a high mass for the h 1 (1380) (1220 MeV/c 2
and 1470 MeV/c 2 , respectively), therefore one might ex-
pect the radial state to actually lie 100-200 MeV/c 2 below
the Godfrey-Isgur prediction.
A calculation by T. Barnes [17] (also see Reference [18])
using a 3 P 0 model suggests that a 1700 MeV/c 2 h 1 ra-
dial excitation should decay almost equally to 1 + S and
1 + D partial waves y which is not consistent with what
we observe. Alternatively, since the ux-tube model pre-
y The calculation is mass dependent, and one expects the S
to D wave ratio to increase for a lower mass h1 state due to
angular momentum barrier eects.
1.4 1.5 1.6 1.7 1.8 1.9
0
500
1000
1500
Events: 1 1 P
нн +
1.4 1.5 1.6 1.7 1.8 1.9
0
500
1000
1500
2000
2500
Events: 1 0 S
+н +
1.4 1.5 1.6 1.7 1.8 1.9
0.0
0.5
1.0
1.5
2.0
2.5
3.0
Phase: 1 1 P , 1 0 S
нн + +н +
1.4 1.5 1.6 1.7 1.8 1.9
0.0
0.5
1.0
1.5
2.0
2.5
3.0
Phase: BW1, BW2, production
1 +-
1 --
1.4 1.6 1.8 1.4 1.6 1.8
1.4 1.6 1.8 1.4 1.6 1.8
0
1
2
3
0
1
2
3
Events/50
MeV/c
2
Events/50
MeV/c
2
2
GeV/c
2
GeV/c
500
1000
1500
2000
2500
500
1000
1500
radians
radians
wh
Mass[ ]
wh
Mass[ ]
FIG. 9. Mass dependent analysis of the 1 + 0 + S and
1 1 + P partial waves shows that the 2 waves are well de-
scribed by two interfering Breit-Wigner resonances: the 1
state xed to the PDG parameters of the !(1650), which is the
rst observation of !(1650) ! !, and a new 1 + state with
mass = 1594(15) +10
60
MeV/c 2 and = 384(60) +70
100

MeV/c 2 .
6

dicts a J PC = 1 + hybrid near 1900-2000 MeV/c 2 [5],
and exotic states have been reported at masses lower than
the ux-tube model expectations, an h 1 (1595) would be
a candidate for a hybrid or radial-hybrid mixed state.
BACKGROUND STUDY
The distribution can be used to estimate the back-
ground due to non-omega events under the omega sig-
nal. The omega signal-to-background ratio estimated
from Figure 3 is a little larger than 1 to 1. But since
the omega is a spin 1 particle, the PWA uses the omega
angular decay information to weight omega events more
highly than non-omega events.
To better understand the background under the omega
and to see if these events are responsible for the ob-
served structure in the 1 + 0 + S and 1 1 + P partial
waves, events were selected from the sidebands around
the omega region and a partial wave analysis of these
events was performed. As expected, the sidebands dis-
tribution, shown in Figure 10, is at and consistent with
phase-space.
0.4
0.2 0.6 0.8 1
0
l sidebands
200
0
400
600
800
Events/0.01
FIG. 10. Distribution of the sideband events as a function
of the omega decay matrix element squared, . Pure ! events
would exhibit a distribution which linearly increases with
whereas phase-space events w/ould be at.
A PWA of the sideband events was performed in the
same manor as the original PWA of the data. Decay
amplitudes were calculated using the sideband events
whereas the normalization integrals and the list of partial
waves included in the t were identical to those used in
the original PWA. Figure 11 shows results of the sideband
PWA. Note that all partial waves except the background
wave have an omega decay factor in the amplitude. This
makes the background wave more favorable to describe
those events which do not contain omegas. As seen in
Figures 11a and 11b, very few sideband events contribute
to the 1 + 0 + S and 1 1 + P waves demonstrating that
non-omega events are unlikely to be responsible for the
observed resonant-like structures. The sideband events
overwhelmingly contribute to the background intensity
as shown in Figure 11c.
1.4 1.5 1.6 1.7 1.8 1.9 2.0
0
500
1000
1500
2000 Events: 1 1 P w
нн +
1.4 1.5 1.6 1.7 1.8 1.9 2.0
0
500
1000
1500
2000
2500
Events: 1 0 S w
+н +
1.4 1.5 1.6 1.7 1.8 1.9 2.0
0
500
1000
1500
2000
Events: background
b)
c)
1.6
1.4 1.8 2.0
wh
Mass[ ] 2
GeV/c
2
Events/50
MeV/c
a)
0
500
1000
1500
2
Events/50
MeV/c
1.6
1.4 1.8 2.0
wh
Mass[ ] 2
GeV/c
2500
2000
1000
1500
500
0
2
Events/50
MeV/c
1.6
1.4 1.8 2.0
wh
Mass[ ] 2
GeV/c
2000
0
500
1500
1000
FIG. 11. The PWA results for the sidebands events. Shown
are the : a) 1 1 + P intensity, b) 1 + 0 + S intensity, and c)
the background intensity.
CONCLUSIONS
We have collected a high statistics sample of the reac-
tion
p ! n + 0
at 18 GeV/c beam momentum, and performed a partial
wave analysis on the ! system where
7

! ! + 0 0 ! 2 ! 2 :
The ! invariant mass distribution rises rapidly near
threshold then increases slowly to a maximum near 1600
MeV/c 2 . A PWA nds the data dominated by natural
parity exchange waves. The results show 2 signicant
isoscalar J PC M L partial wave intensities: a 1 + 0 + S
intensity peaking near 1600 MeV/c 2 with a width of
about 300 MeV/c 2 , and a 1 1 + P intensity peaking near
1650 MeV/c 2 also with a width of about 200 MeV/c 2 .
In addition, a peak in the unnatural parity exchange
wave 2 1 P exhibits a Breit-Wigner mass of 1830 8
MeV/c 2 and width of 86 31MeV/c 2 . Unfortunately
the phase of this state was not well determined and thus
claim of a 2 state at this mass requires further con-
rmation. Another notable feature was that the exotic
partial waves for J PC = 0 and 2 + were found not to
be required by the data.
A mass dependent analysis of 1 + 0 + S and 1 0 + P
partial waves shows that the intensities and relative
phase motion are well described by two resonating
states. A 1 1 + P state in good agreement with
the PDG values for the !(1650), and a new h 1 (1595)
state with mass = 1594(15) +10
60
MeV/c 2 and =
384(60) +70
100
MeV/c 2 . It is interesting to note that
Close and Page [4] predict that the isoscalar J PC = 1
hybrid should decay dominantly to states containing a
vector meson such as and ! and not to traditional
ux-tube decays such as b 1 . Possible interpretations for
an h 1 (1595) include an h 1 radial excitation, a h 1 hybrid,
or a h 1 radial-hybrid mixture.
We are grateful for the support of the technical stas
of the MPS, AGS, BNL, and various collaborating in-
stitutions. This research was supported in part by the
US Department of Energy, the National Science Founda-
tion, and the Russian State Committee for Science and
Technology.
# Present Address: Dept. of Physics, Carnegie Mellon Uni-
versity, Pittsburgh, PA, USA.
y Present Address: Physics Dept., Thomas Jeerson Na-
tional Accelerator Facility, Newport News, VA, USA.
Permanent Address: Rafael, Haifa, Israel.
[1] P. Eugenio, Search for New Mesons in the ! and ! Ї
Systems Produced in the Reaction p ! n + 4
at 18 GeV/c Ph.D. thesis, University of Massachusetts
Amherst (1998).
[2] S. Teige et al., Phys. Rev., D59, 12001, (1999)
[3] S.U. Chung et al., Phys. Rev., D60, 92001, (1999)
[4] F. Close and P. Page, Nucl. Phys. B443 (1995) 233.
[5] N. Isgur and J. Paton, Phys. Rev. D31 (1985) 2910.
[6] V.D. Samoilenko et al., Sov. Phys. Dokl. 36 (1991)
473.; Proceeding of the VIth International Conference on
Hadron Spectroscopy Hadron'95 (1996) 439.
[7] M. Atkinson et al. , Z. Phys. C34 (1987) 303.
[8] D.E. Groom et al., Particle Data Group, The European
Physical Journal C15 (2000) 1.
[9] C. Zemach Phys. Rev. 133:B5 (1964) 1201.
[10] J.P. Cummings and D.P. Weygand, Brookhaven Report
BNL-64637 (1997).
[11] D.J. Herndon et al., Phys. Rev. D11 (1975) 3165.
[12] S.U. Chung, CERN Yellow Report 71-8 (1971).
[13] S.U. Chung, Brookhaven Report BNL-QGS-93-05
(1995).
[14] S.U. Chung and T.L. Trueman, Phys. Rev. D11 (1975)
633.
[15] S. Godfrey and N. Isgur, Phys. Rev. D32 (1985) 189.
[16] N. Achasov, Phys Lett. B462 (1999) 365; C. Amlser, Pro-
ceedings of the Eight International Conference on Hadron
SpectroscopyHadron'99 (1999) 87; A.B. Clegg, Z. Phys.
C62 (1994) 455; A. Antonelli, Z. Phys. C56 (1992) 15;
A. Donnachie, Z. Phys. C42 (1989) 663; M. Atkinson,
Phys Lett. B127 (1983) 132.
[17] T. Barnes, Private communication.; T. Barnes, F.E.
Close, P.R. Page, and E.S. Swanson, Phys. Rev. D55
(1997) 4157.
[18] L. Burakovsky and P.R. Page, The 7th Conference on the
Intersections of Particle and Nuclear Physics CIPANP
2000 (2000).
8