Properties of the charmed strange baryonA+ and evidence for the charmed doubly strange baryonT0 at 2.74 GeV/c2

  • Published on
    15-Aug-2016

  • View
    212

  • Download
    0

Transcript

  • Z. Phys. C Particles and Fields 28, 175 185 (1985) Zeitschrift Par t ides fi3r Physik C

    and Fields @ Springer-Verlag 1985

    Properties of the Charmed Strange Baryon A + and Evidence for the Charmed Doubly Strange Baryon T O at 2.74 GeV/c 2.

    S.F. Biagi 6,a, M. Bourquin 3, A.J. Britten 6,b, R.M. Brown 8, H.J. Burckhart 4'c, A.A. Carter 6, Ch. Dor65, P. Extermann 3, M. Gailloud 5, C.N.P. Gee 8, W.M. Gibson 1, J.C. Gordon 8, R.J. Gray 8, P. Igo-Kemenes 4, P. Jacot-Guillarmod 5, W.C. Louis 8,d, T. Modis 3, Ph. Rosselet 5, B.J. Saunders s, P. Schirato 3, H. W. Siebert 4, V.J. Smith 1, K.-P. Streit 2'e, J.J. Thresher 8, S.N. Tovey 7, R. Weill 5

    1 H.H. Wills Physics Laboratory, University, Bristol BS8 1TL, UK 2 CERN, CH-1211 Geneva 23, Switzerland 3 Universit6 de Gen+ve, CH-1211 Gen6ve, Switzerland 4 Physikalisches Institut, Universitiit, D-6900 Heidelberg, Federal Republic of Germany 5 Universit6 de Lausanne, CH-1000 Lausanne, Switzerland 6 Queen Mary College, University, London E1 4NS, UK 7 University of Melbourne, Australia 8 Rutherford Appleton Laboratory, Chilton, Didcot, Oxon OX 110QX, UK

    Received 8 February 1985

    Abstract. Results are presented from experiment WA62, which searched for charmed strange baryon states produced in X -nucleus interactions in the SPS charged hyperon beam at CERN. Properties of the A + (csu) baryon at 2.46 GeV/c 2 are summarized and upper limits are given for decay branching ratios into various channels. Three events observed at 2.74GeV/c 2 in the 3 -K ~+ lr + mass spectrum are interpreted as the first evidence for the T o baryon with quark content css. Results of a search for the A ~ (csd), the isospin partner of the A are present- ed. The results are discussed in the context of cur- rent theoretical understanding, and a comparison with other experiments on hadroproduction of charmed baryons is made.

    1. Introduction

    We report on the searches, carried out at CERN in experiment WA62, for the production of charmed

    " Now at University of Liverpool, UK b Now at Dept. of Nuclear Medicine, Royal Marsden Hospital, London, UK ~ Now at CERN, Geneva, Switzerland a Now at Princeton University, N J, USA e Also at Physikalisches Institut, Universit~it Heidelberg, FRG * Work supported in part by the UK Science and Engineering Research Council, the Swiss National Foundation for Scientific Research and the Bundesministerium fiir Forschung und Techno- logic, FRG

    strange baryons in S -Be collisions. The use of strange baryon projectiles was expected to favour the production of charmed baryons with non-zero strangeness. Such baryons would have zero or posi- tive charge and couple, through Cabibbo-allowed decays, to final states with strangeness -2 or -3 .

    First results have already been published [1, 2]. In [1], evidence was presented for the production of the A + baryon (quark content csu) in the reaction

    S - +BetA + +X.

    This state was observed in the AK-~ + ~z + mass spectrum as a narrow peak at 2.46 GeV/c z. The pos- itive charge and strangeness -2 of this final state, and the narrow width, consistent with the resolution of the apparatus, supported the interpretation of the signal as a Cabibbo-allowed decay of a charmed strange baryon. This conclusion was further strengthened by the measured lifetime of (4.8_118)+2 9 X 10--13S for the events in the mass peak [2].

    In this article we report results from the search for other decay modes of the A and for other charmed strange baryons. Of particular interest is the observation of three events at 2.74 GeV/c 2 in the channel f f -K ~+ ~+, which is interpreted as the first evidence for the charmed baryon T o , with quark content css.

    In Sect. 2 we briefly decribe the apparatus, in Sect. 3 the trigger and the event selection and in

  • 176 S.F. Biagi et al.: Properties of the Charmed Strange Baryon A +

    Sect. 4 the Monte Carlo simulation. The production and decay properties of the A + are given in Sect. 5. The results of the search for the A ~ the isospin partner of the A +, are presented in Sect. 6, and Sect. 7 is devoted to the T o analysis. In Sect. 8 an interpretation of the results in terms of some theoretical models is given and a comparison is made with other experiments. The results are sum- marized in Sect. 9.

    2. Apparatus

    The apparatus was designed to accept charmed strange baryons produced in the forward direction in I - -Be collisions and to identify them by studying the effective mass distributions of combinations of particles.

    The experiment was performed in the charged hyperon beam at the CERN SPS, which has been described elsewhere [3]. For this experiment, the beam was tuned to its maximum momentum of 135 GeV/c. A differential Cherenkov counter (DISC) selected 2 x 104 incident 22 in each 1.5 s beam pulse of 1.5 106 particles (mainly ~-). The trajectories of the beam particles were measured in multiwire pro- portional chambers (MWPCs) located upstream and downstream of the DISC (Fig. 1).

    The I - struck an 8 cm long Be target, located downstream of the DISC. Charged particles pro- duced in the forward direction were tracked in a double magnet spectrometer equipped with MWPCs and drift chambers (DCs). In order to resolve am- biguities between closely spaced or overlapping tracks, a large number of chamber planes (46 in total) were installed between the target and the first

    magnet SM1. A minimum momentum of about 14 GeV/c was necessary for particles to traverse both magnets. The two multicell gas Cherenkov counters C1 and C2 downstream of the second magnet SM2 had pion thresholds of 14 GeV/c and 10 GeV/c, re- spectively. The cell structure of these Cherenkov counters was matched by two scintillator hodo- scopes H4 and H5. The additional deflection in SM2 was sufficient to separate positive and negative charged particles at the position of H4. Further details of the apparatus are given in [1, 2].

    3. Trigger and Event Selection

    The primary aim of the trigger was to select final states of strangeness -2 or -3 , with zero or posi- tive charge. Therefore the trigger required among the produced particles, a A and a K - together with at least one more charged particle. This restricted choice was made because the data recording system had a capacity of 250 events per beam spill, which at the full X rate of 20,000 per spill was not sufficient to accept all possible decay channels.

    The trigger requirements were: i) A signal from a X- passing through the

    DISC. ii) Signals from at least two charged particles in

    hodoscope H 1 and at least four charged particles in H2. This requirement was met by events with two charged particles coming from the target and a A---,prc- decay occurring upstream of H2. At least four charged particles were also required in H 3.

    iii) A proton candidate, which had traversed hodoscope H5 on the "positive side" without pro- ducing Cherenkov light in the corresponding cell of C2.

    DC EE EE I

    )"- A AB C

    Target / H2

    SMI : llsM2

    I " / / / / /Z

    I EE

    H3

    ~2

    ~ I'o I's z'o z's 3Lo

    rn

    1.5

    1.0

    0.5

    0

    -0.5

    -1.0

    -1.5

    H5

    35 m

    Fig. 1. Schematic layout of the apparatus. A, B, C, D, E, F=clusters of MWPCs; DC = clusters of drift chambers; SM 1 and SM2 = magnets; C 1, C2 = gas Cherenkov counters; H l -H5 = scintillator hodoscopes; I - =incident hyperon beam

  • S.F. Biagi et al.: Properties of the Charmed Strange Baryon A + 177

    > z \

    .c_

    , , 1 , , , ,

    a) ~ K~

    6 7T- IT*

    2

    o&, & ' s'oo' slo

    r 600 g ~0(

    2o(

    0

    b) ~A

    pl t -

    I I I0 1115 1120

    >o

    \

    u

    300

    200

    100

    1310 1320 1330

    ' ' , i ,

    d) mQ- 30 ~ K-

    ~0 j~

    i

    01 I i i i 1660 1670 1680

    E f fec t ive mass ( HeV/c 2 1

    Fig. 2a d. Reconstructed masses for K~ + ~ , A~r,~-, ~- --,A~- and (2- ~ AK decays. The histograms show the data, the curves the MC predictions. The MC curves have been centred on the experimental distributions

    iv) A K candidate, which had traversed hodo- scope H4 on the "negative side" without producing Cherenkov light in the corresponding cell of C 1.

    In the off-line analysis, charged particle trajecto- ries had to come from the target or from a V ~ downstream of the target. For trajectories passing through SM 1, the particle momenta were calculated. The A and K ~ decays were kinematically identified by computing the p~- and ~+ ~ effective mass. For A candidates, we required the computed effective mass to be within 4MeV/c 2 of the A mass, the width of the effective mass distribution being 1.6 MeV/c 2 (r.m.s.), (Fig. 2b). The background under the A signal was less than 5 %. For K ~ candidates, the effective mass was required to be within 9 MeV/c 2 of the K ~ mass. The width of this effective mass distribution was 3.5 MeV/c 2 (r.m.s,), (Fig. 2a).

    The K - requirement in the trigger was strength- ened off-line by combining information from both Cherenkov counters, taking into account their dif- ferent thresholds. Counter C1 had a pion threshold of 14 GeV/c and reached its maximum efficiency of 97 % at a pion momentum of 30 GeV/c. In order to reject pions a momentum cut at 17 GeV/c was ap- plied to the K - candidates. At this momentum the efficiency of C 1 for detecting pions was 68 %. Count- er C2 had a kaon threshold of 36 GeV/c. Below this momentum it was used to further reduce the pion contamination.

    Charged particles which were not identified as p or K - were generally taken to be pions. Cherenkov identification for these particles was not possible, as more than 90% of them had momenta less than 17 GeV/c or did not pass through SM2.

    Although the trigger was designed to accept (AK-~++X) events it was also sensitive to final states of the type (pK-+2 charged +X) and (pK-K~ where the decay K~ + ~- occurred before H2. These event classes were also considered in the analysis.

    4. The Monte Carlo Simulation

    The acceptance and resolution of the apparatus were determined from the Monte Carlo (MC) simulation of the experiment. This included the geometrical ar- rangement of the detectors, their efficiencies and spatial resolutions. The multiple scattering in the material of the apparatus, and the decays of second- ary particles were taken into account.

    This simulation was able to reproduce the mea- sured effective mass distributions for the decays K~ -, A~p~- , E---+A~ and O-~AK- apart from small tails in the distributions which were more pronounced in the data (Fig. 2). These decays, together with the well known resonances K~ X+(1382) and Y~ Fig. 3, were used to check the absolute calibration of the mass scale by studying the dependence of the reconstructed masses on the calibration constants of the magnetic field and the geometrical alignment. From the un- certainties on these quantities we determined the systematic error in the mass calibration.

    ~. 4000 >

    3000

    g 2000

    1ooo

    >< 2000

    o

    o 1000

    0 800 900 1000

    b) ' + An*' ,%

    o 1~;o I~OO l& 1;oo

    N i

    c) __-_ if 60

    . H

    1530 90

    Ef fect ive Mass (MeV/c z)

    Fig. 3a c. Reconstructed masses for K~ ~ K n +, S+(1382)-~An + and Z~ n + decays. The lines are drawn to guide the eye. The arrows indicate the resonance masses

  • 178 S.F. Biagi et al.: Properties of the Charmed Strange Baryon A +

    J i I , , I ~ i i

    10

    ,"o

    u <

    i i 0 50 100 Longitudinal momentum (6eric)

    Fig.4. Calculated acceptance as function of the A + longitudinal momentum Pr for the decay A + ~AK ~+ ~+

    1211

    >~ :E 8Q

    \

    i i i

    " { t

    ,, + +~++*+,~+

    2000 2500 3000

    Effect ive mass [~eVlc l)

    Fig. 5. The AK-~+ ~+ effective mass distribution. Crosses with error bars are data. The line is a fit of a polynomial of order 3. A Gaussian with a width equal to the experimental resolution and an area equal to the observed number of events was added to the polynomial curve

    The Monte Carlo program was used to deter- mine the acceptance for the various decay modes of the A baryon. For these calculations the A + were generated according to an invariant differential cross section of the form

    d3 (7 E d~(1 --x)ne -bp~' (x = pF/p~,mx)

    where the parameters n and b could be adjusted. The A decay products were generated according to phase space, but sequential decays via resonances could also be introduced. Figure 4 shows the accep- tance of the apparatus for the channel AK- rc + rc + as a function of the A + longitudinal momentum, for a PT distribution with the experimental value b= +1.1 (see Sect. 5.1). The corresponding curves for the other A + and A ~ decay modes discussed further down have similar shapes. For a given PL, variations of b within the experimental limits would change the acceptance by less than 20~o. The K - momentum cut at 17GeV/c reduced the acceptance for A + ~AK- rc + ~z + by 20%.

    5. Investigation of the A +

    5.1. The AK- ~+ 7r + Decay Channe l

    Events were selected if they had a A, a K - meson and at least two additional positive particles (as- sumed to be pions). The tracks of the K - and both ~+ had to intersect the 2;- trajectory in the target region with a closest distance of approach of less than 2mm. These criteria were fulfilled by 4,002 AK- ~+ ~z + combinations from 3,352 events.

    The AK- ~+ ~+ effective mass distribution shows a prominent peak in two bins of 15 MeV/c 2, centred

    at 2,460 MeV/c 2, containing 229 entries (Fig. 5). Very few events produced two entries in the same 30 MeV interval. This happened twice in the peak region and on the average three times in neighbouring 30 MeV intervals. We therefore attributed the double entries in the peak region to the background.

    This peak was interpreted as a Cabibbo-allowed decay of the charmed strange baryon A+. The signal contains 82 events above a background estimated by various methods [1] to be 145_+5 events. The width of the peak is (21+_47)MeV/c 2 (FWHM). The mass resolution of the apparatus was determined with the Monte Carlo program to be (223-2) MeV/c 2 (FWHM), which gives an upper limit on the intrinsic width, F

  • S.F. Biagi et al.: Properties of the Charmed Strange Baryon A 179

    distribution of the distances between the production and decay vertices. The decay vertex was defined by the intersection of the K - and the two ~+ tracks. The production vertex was determined from the in- tersection of the beam track with additional tracks not participating in the AK-~+ ~+ mass combi- nation. Stringent geometrical criteria were applied in this reconstruction (Ref. 2) resulting in a resolution of 6mm (r.m.s.) on the distance between vertices. After this selection a signal of 53 events remained above a background of 59. The value obtained for the lifetime is +2 9 (4.8 l ls)X10-13s where the error on the positive side takes into account the possibility that tracks from the decay of associated D mesons were included in the reconstruction of the produc- tion vertex.

    The invariant cross section for A + production was found to be adequately described by the form

    d3cr E d~OC(1 -x ) "e -@~

    in the region x > 0.6, corresponding to PL > 82 GeV/c, where the signal contained 60 events above a back- ground of 94. This cutoff was chosen to avoid the large acceptance corrections at lower momenta (Fig. 4). From a fit to the data we obtained n= 1.7

    +07 _+0.7 and b= 1.1_0i4) (GeV/c) -2. The product of cross section and branching ratio

    for the observed decay mode A ~AK-~+ ~+ was found to be

    cr.B=(5.3_+2.0)lab/Be nucleus for x>0.6.

    5.2. Other A + Decay Channels

    The number of A + decay channels which could be investigated in this experiment was restricted by the lack of a neutral particle detector and by the trigger requirement of a proton, a K - and at least two additional charged particles. The decay modes

    AK ~z + 7r + (A~prc ), pK K ~ rc + (K~ + rc ), pK -K - rc +~+ and 12-K +~+ (~2 ~AK ) were therefore the only Cabibbo-allowed ones to which the experiment was sensitive. For decay channels with higher charged multiplicities, e.g. with ad- ditional ~+ ~- pairs, the acceptance dropped rapid- ly. Differences between the mass calibrations for the different A decay channels were estimated to be smaller than 10 MeV/c 2. The relative acceptances for the different decay modes were calculated using the Monte Carlo program. No cut on PL was applied to the Monte Carlo or data samples.

    5.2.1. The pK-K ~ rc + Channel. The effective mass distribution for this channel is shown in Fig. 6. The

    p K- K-~ 2~6o

    ' ill tt+tt t t i +;!+ +++ }

    ~ ++ 23oo Lo ' 200 26'oo Effective mass (MeV/c 2)

    Fig. 6. The pK K ~ ~-- effective mass distribution

    two 15MeV/c 2 bins around the A mass together contain 36 events. An estimation of the background level, made by averaging over five channels on each side of the signal bins, yields 45_+3 events. If one shifts the background window on the low-mass side downwards by three bins, to avoid including the bump just above 2,400 MeV/c 2, the linear interpo- lation gives 37_+3 events. The value adopted is 41 _+4 events where the error has been increased to take into account possible non-linearities in the background distribution.

    The acceptance for this decay mode relative to the AK ~+ ~+ mode was 1.33_+0.05, where the er- ror reflects the uncertainty on the shape of the A + momentum spectrum. With the branching ratios

    BR(A-- ,p~-)=0.642 and BR(K ~ --*~+ ~ )=0.5 x 0.686 we obtained

    F(A + ~pK K ~ +)

    F (A + --* AK- Tc + zc +) < 0.08 (90 ~o CL).

    5.2.2. The pK- K - 7c + ~+ Channel. The distribution of the pK- K - ~+ ~z + effective mass, requiring only one of the K - be identified in the Cherenkov coun- ters, is shown in Fig. 7. No signal is visible in the two bins around 2,460MeV/c 2 which together con- tain 20 events. The shape of the background cannot be described by a linear approximation. A poly- nomial fit results in a background estimate of 24 + 10 events, where the error takes into account the systematic uncertainties from the choice of the back- ground window and the order of the polynomial. The relative acceptance for this decay mode was 2.88+0.02. This yields, together with the branching ratio for the A ~ p ~- decay, an upper limit

    F(A ~pK- K - Tc + ~+) < 0.03 (90 ~o CL).

    F(A + --* AK- rc + zc +)

    Requiring the second K - candidate to be identified by the Cherenkov counters reduces the acceptance

  • 180 S.F. Biagi et al.: Properties of the Charmed Strange Baryon A +

    80

    6O

    LO

    20

    ao,W 0

    i I

    pK- K" n'n"

    + ++ 2460 I

    +,+++l + i

    2400 2500

    Effective mass IMeV/c z)

    Fig. 7. The p K K zc ~z effective mass distribution

    N'~

    ) 20

    a) n ~x-T[ +

    I I I f l

    ties arising from reconstruct ion losses for ivery close tracks. Therefore we do not give an upper l imit for the relative branching ratio of this decay mode.

    5.2.4. The AK-V(896)z + and X+(1382) K ~+ Chan- nels. We have looked for K* and X* contr ibut ions to the AK-~+ ~+ final state. F igure8 shows the K-~z + and A~ + effective mass distr ibutions from the events in the A + signal, after a background subtract ion using events in neighbouring bins. As the final state contains two ~+, every event gives two combinat ions in each plot. Also shown are the distr ibutions expected for the four-body phase-space decay A + - -+AK- r t + ~z + (smooth curves) and for the

    three-body phase-space decays A+~AK~ + and A + ~X +(1382) K - 7r +, respectively (dashed curves).

    F rom a compar ison of the data with these ex- pected distr ibutions we obtained a value of 0.30 +0.30 for the K* contr ibut ion and a value of 0 _+0.18 for the X* contr ibut ion to the final state, which yield upper limits of 0.7 and 0.25, respectively, at the 90 ~o confidence level.

    ~o ~1 ?l /~" f l

    >

    r

    4~ I !L

    1200 1400 1600 1800

    Effective mass (NeV/c 2)

    Fig. 8a and b. Effective masses for the events in the A ~- mass peak of Fig. 5 after background subtraction. Smooth lines show the MC prediction for four-body phase-space decay, a (K ~+) effec- tive mass. The dashed curve corresponds to A + --*AK~ + decays, b (Arc ~) effective mass. The dashed curve corresponds to A+ ~X+(1382)K rc + decays

    by a factor of about 6. The observed spectrum con- tains one event in the signal region with an esti- mated background of one, resulting in a similar up- per limit.

    5.2.3. The O-K + rt + Channe l . The phase space for this decay channel is much smaller than for those discussed above. No A + signal was found in the f2 K + ~+ effective mass distr ibution. Because the A + mass is very close to the threshold of this chan- nel, the acceptance calculat ion had large uncertain-

    6. Search for A o Decays

    In this section we describe the search for the A ~ (quark content csd), the isospin partner of the A +. Their masses are expected to be equal to within a few MeV/c 2.

    The A ~ decay most similar to the observed decay A+- -+AK- r t+rc +, would be A~ n+7c ~ Due to the lack of y detectors, this decay could not be detected in our experiment. However the apparatus was sensitive to the decay A~ n+. The distri- but ion of the AK-7r + effective mass is shown in Fig. 9, which contains 22,884 combinat ions from 19,086 events. The selection criteria for the events in

    <

    ~E m \

    .s

    i i

    AKR +

    i i

    6OO

    +~+ 2460

    ~,+ ' ++ ,

    0 ,+L I I I I J I I J ) I I 1800 2000 2200 2400 2600 2800

    Effective mass (HeV/c 2)

    Fig. 9. The AK r~ + effective mass distribution

  • S.F. Biagi et al.: Properties of the Charmed Strange Baryon A + 181

    this figure were the same as those used for the AK- ~+ re + effective mass distr ibution (Fig. 5) except for the number of positive pions. There are 582 combinat ions in the two mass bins centred at 2,460 MeV/c z, coming from 580 different events. The line represents a polynomial fit of order 2. No signal is visible at 2,460 MeV/c 2.

    To determine an upper l imit on the number of events from A~ ~z + decay, we took into ac- count that the A ~ mass is expected to be 4 MeV/c 2 larger than the A + mass [4]. Differences in the mass cal ibrations for A~ AK - rc + and A + --, AK- rc + 7~ + decays were estimated to be smaller than 5 MeV/c 2. We obtained an upper l imit of 46 events (90 ~o CL). If x>0.6 is required, as in the determinat ion of the A + product ion cross section, the limit is 39 events.

    The appl icat ion of the vertex criteria, used for the determinat ion of the A + lifetime, reduced the background level by approximately a factor of 2. Even under these condit ions no A ~ signal was seen.

    Other Cabibbo- favoured decay channels of the A ~ to which the apparatus was sensitive, were

    f2 -K + ( (2 - - -+AK- ) , pK - K - rc + and pK- K ~ (K~ + re-). In none of these channels was an A ~ signal observed.

    Assuming the same x and Pr dependence for A ~ product ion as observed in A + product ion (cf. Sect. 5.1), we derived an upper l imit on the product of cross section and branching ratio for the reaction X +Be~A~ A~ - rc + for x>0.6 :

    o-. B < 3.8 ~tb/Be nucleus (90 ~0 CL).

    This number does not change significantly if n and b are varied within the experimental errors found for A + production. It may be expected, however, that in X-N collisions A ~ are produced with higher mo- menta than A because the A ~ can contain two quarks from the incident S - , but the A + only one. If, for example, we use n= 1 for A ~ product ion and the observed value n=l .7 for A + product ion, the upper limit decreases to 3.4 gb/Be nucleus. This l imit also is insensitive to changes of n and b within the experimental errors, provided the difference n(A ~ -n (A +) is kept constant.

    7. Evidence for the T~ Baryon In this section a study of final states with strange- ness S = -3 and charge Q = 0 is presented.

    As described in [1] the S = -2 assignment of the A + signal in the AK-7t + rc + effective mass distr ibu- tion was checked by computing the effective mass under various assumptions for the mass assignments to the charged tracks. The events in the A + signal

    1o

    r I 1300 14t,0

    Effective mass (MeV/r

    n/l 15~80 - -

    Fig. 10. The An effective mass distribution for events in Fig. 5 with a A 7r- decay vertex downstream of the target

    > nn L

    l] I]

    i 17 ,rtHITrl , _ , rl 2400 2?00 3000 33O0 3600

    Effective mass (MeV/r

    Fig. l l a d. The E-K ~+~+ effective mass distributions a All events, b Events for which the interaction vertex was reconstruct- ed. e Events which were compatible with the decay K*-, K-~z +. d Events which were compatible with the decay E*-*E ~z +

    bins did not show significant structure in any other mass combinat ion. This demonstrated that the A + was not a "ref lect ion" of another state.

    In cases where the A resulted from the decay E- - -+ATr- , its observat ion in associat ion with a K - would indicate an S = -3 final state. Using the same sample of events that yielded the A + signal, a search was made for events in which an addit ional negative track intersected the A line of flight downstream of the target with a closest distance of approach of less than 1.5mm. Figure l0 shows the effective Are- mass distr ibut ion obtained when this extra track was interpreted as a ~ . A clear signal of 20events is visible within +7 MeV/c 2 of the E - mass. None of these events contr ibuted a AK-~z + ~+ mass com- binat ion to the A + peak.

    The E -K -~+ rc + effective mass was calculated for these 20 events, which yielded 26 such com- binat ions (Fig. l la ) . These events were subjected to

  • 182 S.F. Biagi et al.: Properties of the Charmed Strange Baryon A +

    0 t I 2~oo 270o ~'ooo 3'?oo 3600

    Eff~five Mass IMeV/r 2)

    Fig. 12. The ~-K ~+ ~z + effective mass distribution for T O can- didates. The four events are numbered to indicate their com- binatorial entries into this plot. Hatched entries have negative A z

    a detailed analysis. Figure 11b shows the mass distri- bution for events which had additional tracks in the chambers B, C and D1 thereby permitting the re- construction of the production vertex. This require- ment was suggested by the observation that the ratio of signal to background for the A + -+AK- rc + rc + channel increased from 82/147 to 53/59 (see Sect. 5.1) when the same condition was imposed. This sample contained 15 combinations from 10 events.

    The events shown in Fig. 11c were required to have a K-rc + combination with an effective mass

    compatible with a K~ i.e. 0 .85

  • S.F. Biagi et al.: Properties of the Charmed Strange Baryon A +

    Table2. Comparison of mass differences predicted by various models with experimental results (units are MeV/c 2)

    Authors [6] Am(A~ +, A +) Am(A +, T ~ A m(A +, T ~

    De Rujula et al. (1975) 220 480 260 Fuchs and Scadron (1979) 110 550 440 K6rner et al. (1979) 210 470 260 Maltman and Isgur (1980) 220 470 250 Sakharov (1980) 235 500 265 Vaisenberg (1982) 110 470 360 Richard and Taxil (1983) 180 380 200 This experiment 180_+15 460_+20 280_+10

    decay length. Furthermore it can be expected that in X-N collisions T o are produced with a lower rate than A because of the additional strange quark contained in the T ~ In 200 GeV/c p N collisions, for example, the ratio of S and S - produced at x =0.66 is 20 [3]. Therefore the observation of three T o decays, as compared to 53A decays is not unreasonable.

    The experiment was sensitive to three other Cabibbo-allowed T o decay channels: ~2 zt +, f2- 7z + 7r- ~r + and AK- K - n + zr +. None of these ef- fective mass distributions showed a T o signal.

    8. Discussion

    In this section we discuss the results for the A + and T o masses, the A lifetime and the A production and compare them with results from other cxperi- ments and with theoretical predictions. The only other charmed baryons observed up to now are the A~ +, 2;~ ++, and Z~ +. Only nine X C events have been observed (8Z ++, 1Z~ +) [5]. The A + c, on the other hand, has been observed in many experiments and in various decay modes [5]. However, even here our knowledge is still poor; for example, the sum of the measured branching ratios is less than 10% and experiments disagree on the value of the mass.

    In the present experiment, the known A~ + decay modes (Art +Tz +~- , pK ~ Tr 7r +, pK-n +, [5]) were suppressed by the trigger which required at least four charged particles, including a K - candidate. For the possible sequential decay Z~ -, AT--*pK zr +, the acceptance was very small.

    8.1. Masses

    Much theoretical work on charmed baryons has been devoted to the calculation of their masses. Most authors used the then current value of the A~ + mass to fix their mass scale. It is therefore more meaningful to consider the mass differences between A[, A and T o instead of the absolute values. We

    W 9 / /~ I

    / I

    S

    183

    tl u

    W* i I

    E S

    c}

    u~

    Fig. 13a-e. Diagrams contributing to the Cabibbo-favoured de- cays of charmed particles, a Spectator diagram, b Exchange dia- gram. e Annihilation diagram

    use the 1984 world average of the At + mass, m = (2,282 _+ 3) MeV/c 2 [5] to calculate mass differ- ences Arn(A~+,A +) and Am(A~+,r~ It should be kept in mind, however, that the experimental situa- tion with respect to the Ac + mass is not yet satisfac- tory.

    In Table 2 various predictions for the mass differ- ences between A[ , A + and T O are compiled [6]. Three of these predictions [6a, 6c, 6d] are in agree- ment with the experimental values for both A and T o"

    8.2. A Lifetime

    The measured A + lifetime +2 9 3 s (4.8_ li8) X 10 -1 is of the magnitude expected for charmed particles [7]. The lifetime of the A~ + is +o 7 3 (2.2_o14) x 10 -1 s [8] based on 19 events observed in three different ex- periments.

    The weak decay of charmed particles is com- monly classified into spectator and non-spectator processes (Fig. 13). If the spectator processes were dominant, one would expect the lifetimes of all singly charmed hadrons to differ only by phase space cor- rections. The A~ + decay can occur via the exchange process c d~us , whereas for the Cabibbo-allowed decays of the A and the T o only spectator pro- cesses are possible. Therefore the lifetime of the A is expected to be larger than the lifetime of the A7, two theoretical estimates for the ratio of the A and A~ + lifetimes being 2.5 [6c] and 4 [9]. This tendency is supported by the measured values. In the case of

  • 184 S.F. Biagi et al.: Properties of the Charmed Strange Baryon A +

    the T o the decay lengths of the events suggest a lifetime larger than the A~ + lifetime, in agreement with the above arguments.

    8.3. A + Production

    The A + was observed in the reaction 2;-+Be--*A + +X via its decay into AK-~+ ~r +. In the region x>0.6, where the experiment was sensitive, the pro- duction cross section times branching ratio, a. B, for this process was measured to be

    or. B = (5.3 2.0) btb/Be nucleus

    A reliable estimate of the total cross section for the production of A + in 2; -nucleon reactions is not possible with the present knowledge of charmed bary- on production and decay. Such an extrapolation would be very model-dependent and the result could vary by more than one order of magnitude. The three main problems are:

    i) The cross section has to be extrapolated from the x region where the experiment was sensitive to the whole region - l _A + +X 10 >0.5

    --+pK~ To+ rc 10_+4 ~A~+~+Tz 2.3-t-_1.1

    NA I l p+Be~A,++D+X 17 >0.1 __

  • S.F. Biagi et al.: Properties of the Charmed Strange Baryon A + 185

    The following results were obtained for the pro- duction of the A + in the reaction 27 + Be ~ A + + X,

    A + ~AK- ~+ ~+, at 1 /s=16GeV:

    d 3 ry 1.7+0.7 E d~-Oc(1 -x ) xe (11 +o~

    or- B = (5.3 _+ 2.0) pb/Be nucleus for x > 0.6.

    The mass and lifetime of the A + were measured to be

    m A + = (2,460 + 15) MeV/c 2 - - +2.9 3 "CA+--(4.8 a.s) x l0 -1 s.

    Upper limits at the 90% confidence level were found for the branching ratios of the following decay modes:

    F [A + ~ AK ~ (896) rc + ]

    F(A + ~AK- ~z + rc +)

    F [A+- - .X+(1 ,382)K ~+3

    F(A + ~AK ~+zc +)

    F(A + --* p K - K - rc + rc +)

    F (A + --* AK- ~+ rc +)

    F (A + ~pK- K ~ ~z +)

    F(A + ~AK- ~z + Tc +)