a Fermi National Accelerator Laboratory
-
FERMILAB-Pub-77/58-EXP
7100.288
(Submitted to Phys. Rev. Lett. )
OBSERVATION OF A DIMUON RESONANCE AT 9.5 GeV
IN 400 GeV PROTON-NUCLEUS COLLISIONS
S. W. Herb, D. C. Horn, L. M. Lederman,
J. C. Sens, H. D. Snyder, and J. K. Yoh
Columbia University, New York, New York 10027
and
J. A. Appel, B. C. Brown, C. N. Brown
W. R. Innes, K. Ueno, and T. Yamanouchi
Fermi National Accelerator Laboratory, Batavia, Illinois 60510
A. S. Ito, H. Jostlein, D. M. Kaplan,
and R. D. Kephart
State University of New York at Stony Brook
Stony Brook, New York 11794
July 1977
3
Operated by Univarsitles Research Association Inc. under contract with the Energy Research and Development Administration
OBSERVATION OF A
DIMUON
RESONANCE AT 9.5 GeV
IN 400 GeV PROTON-NUCLEUS COLLISIONS
S. W. Iterb
D. C.' Horn, L. M. Lederman,
J. C. Set-s,
f
Ii. D. Snyder, and J. R. Yoh
Columbia University
New York, New York
10027*
and
J. A. Appel,
B. C. Brow", C. N. Brow",
W. R. Innes, K. Ueno, and T. Yamanouchi
Fermi National Accelerator Laboratory
Batavia,
Illinois 60510
and
A. S. Ito,
H. J&tlein, D. M. Kaplan,
and R. D. Kephart
State University of New York at Stonyl;;;T$
Stony Brook, Long Island, New York
ABSTRACT
Dimuon production is studied in 400 GeV
proton-nucleus collisons. A'strong enhancement
is observed at 9.5 GeV mass in a sample of 9,000
dimuon events with a mass. m,,+U- > 5
GeV.
l
Research supported by the.National Science Foundation.
t
Permanent address:
Foundation for Fundamental Research
on Matter,
The Netherlands.
-2-
We have
observed a strong rnlracevci~t at 9.5 GeV in the
mass spec?rum of dimuons produced in 400 GeV proton-nucleus
collisions. Our conclusions are based upon an analysis of 9,000
dimuon events with a reconstructed mass I$+~- greater than 5
GeV corresponding to 1.6 x 1016
protons incident on Cu and Pt
targets.
p + (cu, Pt) + r;' + u- + anything.
The produced muons are analyze& in a double-arm magnetic s&c-
trometer
system
with a mass resolution dm/m (r.m.s.)Z 2%.
The experimental configuration (Fig. 1) is a modification
of an earlier dilepton experiment in
the
Fermilab Proton Center
Laboratory." I*' Narrow targets (-.7 mm) with lengths corres-
ponding to 30% of an interaction length are employed.
Beryllium
(18 interaction lengths) is used as a hadron filter, covering
the 50-95 mr (70-llO"
CM)
horizontal and ?lO mr vertical aperture
in each arm.
The Be is closely packed against steel and tung-
sten which minimize particle leakage from outside the aperture,
especially from the tungsten beam dump located 2.2 m downstream
of the target. Polyethylene (1.5 m) and a 2.2 m steel collimator
complete the shielding: The first 30 cm of beryllium (starting
13 cm downstream of the target center) can be remotely exchanged
for 30 cm of copper.
The spectrometer dipoles deflect vertically, decoupling
the
production angle of each muon from its momentum determina-
tion. At full excitation (1500 A),
the
magnets provide a trans-
-3-
verse momentum kick pt '-1.2 GeV.
In order to maximize the usable
luminosity,
no
detectors are placed upstream of the magnet.
Conventional PWC's and scintillation hodoscopes serve to define
the muon trajectory downstream of the air dipole.
FollowLng
the PWC's is a solid iron magnet (1.8 m long, energized to 20
kG) used to partially refocus the muons vertically and to re-
determine the muon momentum to ?15%. A threshold zeeenko" counter
on each arm also helps prevent possible low momentum
muon
trig-
gers.
The apparatus is arranged symmetrically with respect to
the horizontal median plane in order to detect both v+ and v-
in
each arm.
' The data sets presente,d here are listed in Table I.
LOW
current runs pwduced
-15,000 J/JI and 1000 $' particles which
provide a test of resolution, normalization, and uniformity of
response over various parts of the detector.
Figure 2b
shows
the 1250 A J/Q and ‘L' data.
The yields are in reasonable agree-
ment with ouz earlier measurements.'
High mass data (1250 A and 1500 A) were collected at a rate
of
20 events/hour for mu+u- > 5 GeV using 1.5-3 x lo11 incident
protons per accelerator cycle. The proton intensity is limit-4
by the requirement that the singles rate at any detector plane
not exceed 10
7
counts per sec. The copper section of the hadron
filter has
the effect of lowering the singles rates by a factor
of 2, permitting a corresponding increase in protons on target.
The penalty is an
q
15% worsening of the resolution at 10 GeV
mass.
Figure 2a shows the yield of muon pairs obtained in May
and June, 1977.
-I-
At
the
present stage of
the analysis the
following conclu-
sions may be drawn from
the
data (Fig. 3a):
1)
A
statistically significant
enhancement
is observed at 9.5
GeV W+U- mass.
2)
Excluding
the
8.8-10.6
GeV
region the continuum of p'u-
pairs falls smxathly with mass.
A simple functional form,
&I,=, = Ae-h,
A = 1.89 i .23 x 1O-33 cm2/GeV/nucleon
b = 0.98
+ .02 GeV-l
gives a good fit to the data for 6 GeV < m,,tu- < 12
GeV
(X2
= 21 for 19 degrees of freedom)."'
3)
In the
excluded
mass
region the
continuum
fit predicts 350
events.
The data contain 770 events.
4)
The
observed width of the enhancement is greater than our
apparatus resolution of FWHM = .5 f. .l GeV.
Fitting
the
data minus
the
continuum fit (Fig. 3b) with a simple gaus-
sian of variable width yields
the
parameters:
Mass
= 9.54 + -04 GeV
gMBjy;ol; y-4 ? .3 x
10e3' cm*/GeV/nucleon
+ .09 GeV
2
-
X -
52 for 27 degrees of freedom.'
An
alternative fit with two gaussians whose widths are fixed
at the resolution of the apparatus yields: