Charm hadroproduction results from Selex
ABSTRACT The SELEX experiment (E781) is 3-stage magnetic spectrometer for a high statistics study of hadroproduction of charm baryons out to large xF using 650 Gev , and p beams. The main features of the spectrometer are: a high precision silicon vertex system, powerful particle identication provided by TRD and RICH, forward s decay spectrometer and 3-stage lead glass photon detector. An experiment overview and spectrometer features are shown. Reconstructed charm states and results on c , D + particles and antiparticles produced by , and p beams at xF > 0:3 and asymmetry for c are presented. 1. Introduction Charm physics explores QCD phenomenology in both perturbative and nonperturbative regimes. Production dynamics studies test leading order (LO) and next to leading order (NLO) perturbative QCD. Charm lifetime measurements test models based on 1=MQ QCD expansions. The present xed target experiments have considerably improved the statistics but many problems remain. All exp...
arXiv:hep-ex/9812031v1 24 Dec 1998
First Charm Hadroproduction Results from SELEX†
The Selex Collaboration
J. Russ3, N. Akchurin17, V. A. Andreev11, A.G. Atamantchouk11, M. Aykac17, M.Y. Balatz8, N.F. Bondar11, A. Bravar22,
M. Chensheng7, P.S. Cooper5, L.J. Dauwe18, G.V. Davidenko8, U. Dersch9, A.G. Dolgolenko8, D. Dreossi22, G.B. Dzyubenko8,
R. Edelstein3, A.M.F. Endler4, J. Engelfried5,13, C. Escobar21,a, I. Eschrich9,b, A.V. Evdokimov8, T. Ferbel19,
I.S. Filimonov10,c, F. Garcia21, M. Gaspero20, S. Gerzon12, I. Giller12, G. Ginther19, V.L. Golovtsov11, Y.M. Goncharenko6,
E. Gottschalk3,5, P. Gouffon21, O.A. Grachov6,d, E. G¨ ulmez2, C. Hammer19, M. Iori20, S.Y. Jun3, A.D. Kamenski8,
H. Kangling7, M. Kaya17, C. Kenney16, J. Kilmer5, V.T. Kim11, L.M. Kochenda11, K. K¨ onigsmann9,e, I. Konorov9,f,
A.A. Kozhevnikov6, A.G. Krivshich11, H. Kr¨ uger9, M.A. Kubantsev8, V.P. Kubarovsky6, A.I. Kulyavtsev6,c, N.P. Kuropatkin11,
V.F. Kurshetsov6, A. Kushnirenko3, S. Kwan5, J. Lach5, A. Lamberto22, L.G. Landsberg6, I. Larin8, E.M. Leikin10,
M. Luksys14, T. Lungov21,g, D. Magarrel17, V.P. Maleev11, D. Mao3,h, S. Masciocchi9,i, P. Mathew3,j, M. Mattson3,
V. Matveev8, E. McCliment17, S.L. McKenna15, M.A. Moinester12, V.V. Molchanov6, A. Morelos13, V.A. Mukhin6, K. Nelson17,
A.V. Nemitkin10, P.V. Neoustroev11, C. Newsom17, A.P. Nilov8, S.B. Nurushev6, A. Ocherashvili12, G. Oleynik5,h, Y. Onel17,
E. Ozel17, S. Ozkorucuklu17, S. Parker16, S. Patrichev11, A. Penzo22, P. Pogodin17, B. Povh9, M. Procario3, V.A. Prutskoi8,
E. Ramberg5, G.F. Rappazzo22, B. V. Razmyslovich11, V. Rud10, P. Schiavon22, V.K. Semyatchkin8, Z. Shuchen7, J. Simon9,
A.I. Sitnikov8, D. Skow5, P. Slattery19, V.J. Smith15,k, M. Srivastava21, V. Steiner12, V. Stepanov11, L. Stutte5, M. Svoiski11,
N.K. Terentyev11,3, G.P. Thomas1, L.N. Uvarov11, A.N. Vasiliev6, D.V. Vavilov6, V.S. Verebryusov8, V.A. Victorov6,
V.E. Vishnyakov8, A.A. Vorobyov11, K. Vorwalter9,l, Z. Wenheng7, J. You3, L. Yunshan7, M. Zhenlin7, L. Zhigang7,
M. Zielinski19, R. Zukanovich Funchal21
1Ball State University, Muncie, IN 47306, U.S.A.
2Bogazici University, Bebek 80815 Istanbul, Turkey
3Carnegie-Mellon University, Pittsburgh, PA 15213, U.S.A.
4Centro Brasileiro de Pesquisas F´ ısicas, Rio de Janeiro, Brazil
5Fermilab, Batavia, IL 60510, U.S.A.
6Institute for High Energy Physics, Protvino, Russia
7Institute of High Energy Physics, Beijing, PR China
8Institute of Theoretical and Experimental Physics, Moscow, Russia
9Max-Planck-Institut f¨ ur Kernphysik, 69117 Heidelberg, Germany
10Moscow State University, Moscow, Russia
11Petersburg Nuclear Physics Institute, St. Petersburg, Russia
12Tel Aviv University, 69978 Ramat Aviv, Israel
13Universidad Aut´ onoma de San Luis Potos´ ı, San Luis Potos´ ı, Mexico
14Universidade Federal da Para´ ıba, Para´ ıba, Brazil
15University of Bristol, Bristol BS8 1TL, United Kingdom
16University of Hawaii, Honolulu, HI 96822, U.S.A.
17University of Iowa, Iowa City, Iowa 52242, U.S.A.
18University of Michigan-Flint, Flint, MI 48502, U.S.A.
19University of Rochester, Rochester, NY 14627, U.S.A.
20University of Rome ”La Sapienza” and INFN , Rome, Italy
21University of S˜ ao Paulo, S˜ ao Paulo, Brazil
22University of Trieste and INFN, Trieste, Italy
The SELEX experiment (E781) at Fermilab is a 3-stage magnetic spectrometer for the high statistics study of charm hadroproduction
out to large xF using 600 GeV Σ−, p and π beams. The main features of the spectrometer are:
• high precision silicon vertex system
• broad-coverage particle identification with TRD and RICH
• 3-stage lead glass photon detector
Preliminary results on differences in hadroproduction characteristics of charm mesons and Λ+
beams there is a striking asymmetry in the production of baryons compared to antibaryons. Leading particle effects for all incident
hadrons are discussed.
c for xF≥ 0.3 are reported. For baryon
Understanding charm hadroproduction at fixed-target
energies has been a difficult theoretical problem because
of the complexities of renormalization scale, of parton
scale, and of hadronization corrections. The recent re-
view by Frixione, Mangano, Nason, and Ridolfi sum-
marizes the theoretical situation, using data through
19961. More recent data from Fermilab E791 (500 GeV
π−beam) greatly improves the statistical precision on
charm meson production by pions, but E791 has not yet
reported absolute cross sections or compared yields be-
tween charm species. In this first report of the SELEX
hadroproduction results, we compare our pion results at
580 GeV with those from E791 as well as comparing SE-
LEX pion data with our proton data at 550 GeV and Σ−
data at 620 GeV mean momenta. All SELEX data were
taken in the same spectrometer with the same trigger.
We limit this report to data having xF ≥ 0.3, where the
spectrometer acceptance is essentially constant with xF
for all final states.
2 The Experiment
SELEX used the Fermilab Hyperon beam in negative po-
larity to make a mixed beam of Σ and π in roughly equal
numbers. In positive polarity, protons comprised 92%
of the particles, with π+making up the balance. The
beam was run at 0 mrad production. The experiment
aimed especially at understanding charm production in
the forward hemisphere and was built to have good mass
and vertex resolution for charm momenta from 100-500
GeV/c. The spectrometer is shown in Figure 1.
Interactions occurred in a target stack of 5 foils: 2
Cu and 3 C. Total target thickness was 5% of Λint for
protons. Each foil was spaced by 1.5 cm from its neigh-
bors. Decays occurring inside the volume of a target were
rejected in this analysis. Interactions were selected by a
scintillator trigger. The charm trigger was very loose,
requiring only ≥ 4 charged tracks in a forward 10◦cone
and ≥ 2 hits in a hodoscope after the second analyz-
ing magnet. We triggered on about 1/3 of all inelastic
A major innovation in E781 was the use of online
selection criteria to identify reconstructable events. This
experiment uses a RICH counter to identify p, K, or π
after the second analyzing magnet. A computational fil-
ter used only these RICH-identifiable tracks to make a
full vertex reconstruction in the vertex silicon and down-
stream PWCs. It selected events that had evidence for a
secondary vertex. This reduced the data size (and offline
computation time) by a factor of nearly 8 at a cost of
about a factor of 2 in charm written to tape, as normal-
ized from a study of unfiltered K0
sand Λ0decays. Most
H O D 1
H O D 2
H yperon Targetti ng
Ri ng-I m agi ng Cherenkov
cal ori m eter
V ee A
V ee B
V ee C
20 m40 m
10 m30 m
Proton Center Layout
20 µm pitch
VX Si detectors
25 µm pitch
VX Si detectors
u y x
20 µm pitch
x y 1 u v
x y 2 u v
x y 3 u v
x y 4 u v
x v 5 u x
Figure 1: E781 Layout
of the charm loss came from selection cuts that are in-
dependent of charm species or kinematic variables. No
bias is expected from the filter. Filter operation depends
on stable track reconstruction and detector alignment.
These features were monitored online and were extremely
stable throughout the run.
All data reported here result from a preliminary pass
through the data, using a production code optimized for
speed but not efficiency. Final yields will be higher than
these preliminary results. However, our simulations in-
dicate that the inefficiency does not affect the kinematic
features of the results for xF ≥ 0.3. For all final states,
the charm selection required that the primary vertex lie
within the target region and that the secondary vertex
occur before the start of the VX silicon. At our high
energy, this latter cut removed a number of D±events
which can be recovered later.
In this analysis secondary vertices were recon-
structed when the vertex χ2for the ensemble of tracks
was inconsistent with a single primary vertex. All com-
binations of tracks were investigated, and every sec-
ondary vertex candidate was tested against a reconstruc-
tion table that listed acceptable particle identification
tags for a charm candidate, track selection criteria nec-
essary (RICH identification for a proton, for example),
and any other selections, e.g., minimum significance cut
for primary/secondaryvertex separation. Selected events
were written to output files and the essential reconstruc-
tion features for each identified secondary vertex were
saved in a PAW-like output structure for quick pass-II
analysis. All data shown here come from analysis using
this reduced output.
3.1 System performance for charm
Vertex resolution is a critical factor in charm experi-
ments. The primary and secondary longitudinal vertex
resolution for all data in a typical run of the experiment
are shown in Figure 2. The lower plot shows the primary
vertex distribution overlaid on rectangles that represent
the physical placement of the 5 targets. The average rel-
ativistic transformation factor from lab time to proper
time for charm states in these data is 100. This spa-
tial resolution corresponds to about a 20 fs proper time
resolution for lifetime studies.
Another important factor in charm studies at large
xFis having good charm mass resolution at all momenta.
Figure 3 shows that the measured width of the D0→
K−+ π+is about 10 MeV for all xF.
Primary and Secondary Vertex ResolutionPrimary and Secondary Vertex Resolution
Primary vertex positions
Primary vertex positions
Figure 2: Typical Primary and Secondary Vertex Error Distribu-
Finally, we depend on the RICH to give correct iden-
tification of K and p decay prongs. Figure 4 shows the
π/K separation in interaction data for 100 GeV/c tracks,
a typical momentum for prongs from our charm states.
The RICH gives π/K separation up to 165 GeV/c (2σ
4Overall Charm Features at Large xF
Previous high-statistics charm production results from
pions3and protons4have emphasized central production,
Figure 3: D0Mass Resolution versus D0Momentum
although both NA32 and E791 have presented results for
xF ≥ 0.5. SELEX and E769 are the only high energy
experiments reporting results from three different beam
particles with identical systematics. The important fea-
tures of the SELEX data can be seen at a glance in Fig-
ure 5 for the charged states D±, Λ+
spectively by Σ−, π−, and proton beams. The pion data
show comparable particle and antiparticle yields both for
charm mesons and for charm baryons, as reported by
NA32 at lower energy3. It remains a surprising feature
of hadroproduction that one finds significant antibaryon
production from pions even at xF ≥ 0.5. The source of
the antiquark pair which combines with the charmed an-
tiquark has been the subject of considerable theoretical
speculation. The pion provides a u valence quark which
can contribute in some models. No present model gives
an adequate description. There is good agreement for
the D±production asymmetry integrated over xF≥ 0.3
between these preliminary results and the E791 results3.
E791 has not published Λ+
c asymmetry results. Their
observations are consistent with these shown here5.
c, and Λ
The relative efficiencies for each beam particle are
almost the same in this xF region, so that one can quote
the ratio of the cross sections even though we have not
yet determined absolute yields. The normalization be-
tween different incident hadrons depends on the number
of incident beam particles for each data sample and on
the total inelastic cross section for each beam particle.
We use 34 mb for the proton inelastic cross section, 27
σr = 0.156cm
Figure 4: RICH K and π Response at 100 GeV/c
mb for Σ−, and 22 mb for π−to compare yields for dif-
ferent beam particles. For these data the relative yields
of selected charmed states, normalized to pion produc-
tion, are given in Table 1. No errors are included in this
preliminary analysis. Note that this table does not di-
rectly provide information about the relative yields for
the different charmed states.
Relative Charmed Particle Yields for xF ≥ 0.3 versus
Perhaps the most surprising result from this table
is the observation that baryon beams are very effective
charm baryon producers, at least at large xF. Also, for
the states listed here, the Σ−beam has yields comparable
to pions, except for the non-leading case of the D+. We
have not yet compiled the yields for the c-s-q baryons,
where we expect the Σ−beam relative yields will large.
The previous table gave the relative efficacy of each
beam particle for producing a given charm state at large
xF. It does not compare relative yields of the different
charm states for the same beam. As can be seen from
Figure 5, there are strong asymmetries. These are tabu-
lated in Table 2. Again, errors are omitted at this stage
Table 2 shows for both baryon beams there are
striking differences in production asymmetries for charm
Figure 5: Charm and Anticharm mass distributions for Σ−,π−,
and p beams in modes Λ+
c → pK−π+or c.c. and D+→ K−π+π+
Table 2: Charmed Particle Antiparticle Ratios for xF≥ 0.3 versus
baryons compared to the pion beam. For charm mesons,
that is not the case. Baryon beams, which have no va-
lence antiquarks, show strong suppression of antibaryon
production, compared to pions. This feature was not ob-
served by NA27 in 400 GeV pp collisions. They reported
comparable baryon/antibaryon production but had only
a few events, all in the central region. No other pro-
ton data exist for charm baryons. The WA89 results for
charm baryon production by Σ−are consistent with our
The D−and Λ+
care leading hadrons in the sense that
all 3 beam hadrons may contribute at least one valence
quark to the final state. The large difference in the Λ+
asymmetry between the meson beam (largely symmetric)
and the baryon beams (very asymmetric) is a new issue
for charm hadroproduction analysis, which has assumed
that there is a universal baryon/meson fraction for all
The SELEX experiment complements previous charm
hadroproduction experiments by exploring different re-
gions of production phase space and by using different
beams. The early results already show some noteworthy
new features of charm production. Further studies of dif-
ferent states and details of single- and double-differential
charm production distributions are underway and will be
reported at meetings in the fall.
Further analysis will extend the xFcoverage down to
about 0.1, to enhance overlap with other experiments and
to increase statistics. Also, other charm baryon states are
being analyzed and results will be reported later.
We are indebted to the technical staff at Fermilab and our
home institutions, especially B. C. LaVoy, D. Northacker,
F. Pearsall, and J. Zimmer , for invaluable technical
support. This project was supported in part by Bun-
desministerium f¨ ur Bildung, Wissenschaft, Forschung
und Technologie, Consejo Nacional de Ciencia y Tec-
nolog´ ıa (CONACyT), Conselho Nacional de Desenvolvi-
mento Cient´ ıfico e Tecnol´ ogico, Fondo de Apoyo a la In-
vestigaci´ on (UASLP), Funda¸ c˜ ao de Amparo ` a Pesquisa
do Estado de S˜ ao Paulo (FAPESP), the Israel Science
Foundation founded by the Israel Academy of Sciences
and Humanities, Istituto Nazionale de Fisica Nucle-
are (INFN), the International Science Foundation (ISF),
the National Science Foundation,NATO, the Russian
Academy of Science, the Russian Ministry of Science and
Technology, the Turkish Scientific and Technological Re-
search Board (T¨UB˙ITAK), the U.S. Department of En-
ergy, and the U.S.-Israel Binational Science Foundation
1. S. Frixione, M. Mangano, P. Nason, and G. Ridolfi,
“Heavy Quark Production” in “Heavy Flavours II”,
A. J. Buras and M. Lindner, eds., World Scientific
Publishing Co, Singapore (1997); see also preprint
2. Fermilab-Pub-98/299-E, submitted to Nucl. Instr.
3. NA32:230GeV; S. Barlag, et al., Z. Phys.
C39(1988)451; ibid C49(1991)555; Phys. Lett.
E769: 250GeV; G.A. Alves, et al., Phys. Rev. Lett.
69(1992)3147; ibid 77(1996)2388;2392
WA92:330GeV; M.I. Adamovich, Phys. Lett.
Lett. B411(1997)230; ibid B403(1997)185; ibid
4. E769: 250GeV; G.A. Alves, et al., Phys. Rev. Lett.
NA27:400GeV; M. Aguilar-Benitez, et al., Z.
5. Simon Kwan, private communication
6. M.I. Adamovich, et al., submitted to Eur. Phys. J.
C; preprint hep-ex/9803021
500GeV;E.M. Aitala,et al.,Phys.
†Invited talk presented at the 1998 International Conference on
High Energy Physics, Vancouver, B.C., Canada, July, 1998.
aPresent address: Instituto de Fisica da Universidade Estadual de
Campinas, UNICAMP, SP, Brazil.
bNow at Imperial College, London SW7 2BZ, U.K.
dPresent address: Dept. of Physics, Wayne State University, De-
troit, MI 48201, U.S.A.
ePresent address: Universit¨ at Freiburg, 79104 Freiburg, Germany
fPresent address: Physik-Department, Technische Universit¨ at
M¨ unchen, 85748 Garching, Germany
gCurrent Address: Instituto de Fisica Teorica da Universidade Es-
tadual Paulista, S˜ ao Paulo, Brazil
hPresent address: Lucent Technologies,Naperville, IL
iNow at Max-Planck-Institut f¨ ur Physik, M¨ unchen, Germany
jPresent address: Motorola Inc., Schaumburg, IL
kGenerous support of Carnegie-Mellon University is gratefully
lPresent address: Deutsche Bank AG, 65760 Eschborn, Germany