sign in sign up
<<
Home , Photino

183

SEARCH FOR NEW WITH L3 DETECTOR AT LEP

L3 Collaboration

C.Dionisi Dipartimentodi Fisica, Universitli Roma "La Sapienza" and INFN, Sezione di Roma,di Roma, Italy

ABSTRACT L3 has searched forscalar , scalar and Winos from the reactions e+e- � t:T€', µ:µ:, w+ w- at 91 GeV. No events has been found and the results are 41 GeV 41 = GeV, and -{S44 GeV at 95% confidence level. From the z• invisible widthMe�> a lower limitMµ> of 28.8 GeV on theM;;,; s- > has also been put. For the excited muons search L3 excludes mass less than 45.1 GeV from pair production and less than 87.4 GeV from single production always at 95% confidence level. 184

1. INTRODUCTION The e+e- collider, LEP is ideal for thestudy of the production of new particles, since the initial state consists of pointlike particles and the production mechanism is. purely electroweak. The [1] has been very successful in describing all data concerning electroweak interactions, however, it leaves many fundamental questions unexplained such as the - spectrum, the mass generation, the unnaturalness of the Higgs mechanism and the large number of arbitraryparameters. One has then to look beyond 1the Standard Model. One model is the SupersymmetryModel (SUSY) [2], which predicts many new particles with differing fromthe known particlesby half a unit. A different approach [3] would be imagine to that the , and gauge are all composites with an associated scale A. One possible effect of compositeness is the existence of excited states, of known leptons, L*, L. In this paper I shall review the searches for supersymmetric particles and excited leptons

performed with the L3 detector around the Z0 energies at LEP.

2. THE MINIMAL SUPERSYMMETRIC STANDARD MODEL extends the Standard model with a new type of symmetry between bosons and [2]. Supersymmetric theories predict for each elementary a with the same quantum numbers but with a spin that differs by � fromthat of the particle. Since no such partners have been found up to now, supersymmetry must be badly broken or is not a symmetry of . In table I list the known particles with their supersymrnetric partners in the so-called 1 "minimal scheme". All these s-particles are present in any supersymmetricmodel with the exception of the Goldstino which in supergravity is eaten by the spin �: (the supersymmetric G L partner of the ) to provide its mass as in the standard model the charged Higgs is eaten by the W to provide its mass. Let us now recall here the main points relevant for our presentation and refer thereader to the several excellent reviews on supersymmetric theories [4] for the details. i) there are two spin-zero for each which correspond to the two helicity states of the fennion right and left handed states); (fR h the Higgs structure of the supersymmetric models is richer than that of the standard model ii) with only one physical Higgs. In fact at least two Higgs doublets are necessary in supersymmetry to give masses to both up and down type quarks. In the Higgs spectrum in the case of minimal supersymmetric models: one has 1 charged scalar H±, two neutral scalars (H 1, H2), H2 being by definition the lightest, and a pseudoscalar H3. The

lightest neutral scalar (H2) is always lighter than the Z0 boson [5]. Apart from v'/v (see below) all masses are fixed in terms of one more parameter, for instance MH± ; 185

the weakly interactingfermions ofthe same charge mix to formthe mass eigenstates. Namely iii) the Photino, Zino and neutral fields mix, and the resulting states are called "". The Wino and charged Higgsinos fields mix and they are called "". In the modelwe use we have: charginos: (Wino) + W (heavy Wino) Wh neutralinos: (Zino) + Z Zi, (heavy Zino) (photino) y () H iv) because of the conservation of R-parity (R + 1 for ordinary particles and R -1 for = = supersymmetric ones): a) SUSY particles are always pair produced, b) the lightest SUSY particle (LSP) is absolutely stable and all SUSY particles decay into it. LSP candidates are the photino, the lightest ,or a scalar neutrino. The following "basic assumptions" have been used in the calculations and Monte Carlo simulations of section 3 and 4: i) We have used models of Supergravity Unified Theories with N 1 (SUGRA) to compute the = SUSY cross sections [6], [7]. The mass eigenstates and couplings are dominantly governed by fourparameters: Mo : the common scalar mass at the unification scale, the ratio of the Higgs field vacuum expectation values is responsible for masses v (v of up type quarks and for down type quarks), v' Mv,;: the Wino mass, the Photino mass. ii) PhotinoMy: is the lightest SUSY particleand it escapes detection. iii) In most of the reactions we have taken v' Under this condition all the leptons are v = I. degenerate in mass: Mt Mv . = iv) For the reaction e+e- we have also studied the case with light s-neutrino: < -7 WW }".v I. v) For s-quarks and s-leptons, being (fermion masses)/M0 <<1, the mixing is negligible and we have assumed that and are the mass eigenstates. fR fL

3. SEARCH FOR SCALAR LEPTONS The data presented here forthe SUSY search have been collected at LEP with the L3 detector during the first energy scan in September 1989 and corresponds to 157 nb-1 of integrated luminosity (around 2500 z· hadronic decays ) [8]. The detector and its performancies are described in [9]. 186

The productionof a pair of supersymmetric scalar leptons ( can occur through the s­ e, µ, :t') channel exchange of a Z' or , and also through the t-channel exchange of a photino in the case of scalar production. The sleptons then decay 100% to their associated leptons and photino. L3 identifies scalar muons and electronsby the presence of acoplanar dimuons or dielectrons with large missing PT. The missing PT is defined as the product of the energy of the second most energetic particle and sin <)>,where <)>is the azimuthal angle between the two most energetic particles. Dimuon events are selected according to the followingcriteria: Each of the two muons is reconstructed in at least two sets of chambers in the bending (xy) I. plane and one set in the non-bending (rz) plane. 2. The momentum sum of the two muons is above 15 GeV. 3. The total energy measured in the calorimeters is less than Ge V. 30 Criteria 2 and 3 are used to reject all backgrounds including residual cosmic rays and hadronic events. Dielectronevents selected according to the following criteria: are There are two or three energy clusters in the calorimeters in the angular range 42' 138', I. < e < whose longitudinal and transverse shower distriibutions are consistent with being of electromagnetic origin. 2. The energies ofthe two most energetic clusters are above 5 GeV. 3. The total calorimeter energy is less than 20% of the energy measured in the electromagnetic calorimeter. Criteria 2 and 3 are used to reject hadronic events. The missing PT spectrafor dimuons and dielectronsare shown in Figs. 1 and 2 respectively. The radiative dilepton events, with one or more energetic . observed in our detector have been removed from the data since they clearly cannot be SUSY candidates. There were four such events observed with PT > 4 GeV, while 6 events are expected from Monte Carlo simulation [6]. As seen from Figs. 1 and 2, there are no pure dilepton events in either clistribution with PT greater than 6 GeV. The predicted PT distributions forscalar muons or electrons were calculated using Monte Carlo methods [6]. The resultant spectra for 41 GeV mass scalar muons and 40 GeV mass scalar electrons are shown in Figs. 1 and 2 respectively. We see that the scalar leptons are expected to cover a region in PT between 6 and 26 GeV. No data events are observed in this region of PT. We compute the expected number of dimuon and dielectron events resulting from SUSY particle production by normalizing them to the observed collinear dimuon and dielectron events, N zz due to the Z' decays: N11 187 where and are the accepted dilepton cross sections for scalar lepton pairs and for normal crn cru lepton pairs respectively, computed from Monte Carlo. Since no acoplanar dimuon or dielectron events are observed above PT > 6 GeV, we set lower mass limits of 41 GeV for scalar muons and 41 GeV for scalar electrons at the 95% confidence level. This result is insensitive to the photino mass up to 20 GeV.

4. SEARCH FOR WINOS Winos can be pair produced in e+e· annihilation through the reaction:

which takes place via s-channel and Z° exchanges and via the t-channel exchange. y v From the above analysis we conclude that the scalar lepton masses must be above 41 GeV. It is natural to assume the scalar quarks cannot be lighter than the scalar leptons. Winos with mass less than 41 GeV could thereforedecay into an ordinary lepton accompanied by a scalar neutrino:

w--dv,

or if Mv> Mw, the also could decay into an ordinary lepton, a neutrino, and a photino: W w--Ltvy,

where we have used a branching ratio of 11% for each lepton flavor. In either case there would be a large amount of missing energy, due to the scalar neutrino or photino which escapes undetected. The signature of Wino pair production can be either acoplanar dimuon, dielectron, or ­ electron [6] events. The first two channels are already ruled out by the above analysis (see below). The muon-electron sample is selected according to the followingcriteria: 1. One and only one muon is reconstructedin at least two sets of chambers in the xy plane and one set in the rz plane. There must be no other detected muons. 2. The momentum of the muon is above 5 GeV. 3. The total energy measured in the hadron calorimeters is less than 10 GeV. 4. There are one or two clusters, whose longitudinal and transverse shower distributions are consistent with being of electromagnetic origin. 5. The energy of the most energetic cluster is above 5 GeV. Criteria 2 and 3 are used to reject hadronic events. 18B

A total of 9 eµ events are observed in our data sample while 8 are expected from pair 't production. There are no events with a P-rgreater than 1 GeV. For the case of the two body decay, we assume a scalar neutrino mass of 10 GeV. Normalizing to the observed dimuon and dielectron events, we expect a total of 96 acoplanar ee, eµ and µµ events with P-r 6 GeV fora Wino mass of 42 GeV. In the case of the three body Wino > decay, we expect 7.4 events for a Wino mass of 42 GeV and a photino mass of 10 GeV. For a Wino mass of GeV, we expect events for the case of two body decay, and 5 events for the 44 64 case of three body decay. We therefore are able to set a lower limit of 44 GeV for the Wino mass. This result is again insensitiveto the photino mass up to 20 Ge V.

5. CONSTRAINT ON S- From a simultaneous model independent to the Z0 hadron and lepton line shapes L3 obtain fit the partial width (invisible width) for Z0 decays into non interacting particles like neutrinos to f'inv be f = 537±48 MeV [10]. inv A deviation of from the SM predictions can be a s:igne of New beyond the f'invisible SM [11]. As a of fact the determination of L'i.ftNV (Cnvisible - rfn� l ), where f = = isib e �� isible

3xl 66.5 = 499.5 MeV, gives the opportunity to test if any new neutrino like particle is coupled to the Z:. The one-sided 95% C.L. L3 upper limit on L'i.fINv. obtained by varying the = X;in 2.69 x2 + which corresponds to 1.64 sigma, is L'i.fINv 116 MeV. < If we assume that L'i.f INV is 100% due to a new s-neutrino family, being the s-neutrino the spin zero supersymmetric partner of the neutrino, and if we remember that each s-neutrino family contributes a width

4m2� -- 1 f(Z ---> vv) = 2 ·f(Z ---> VV) 1 - v [ M� /2 where f(Z0 --->w) = 166.5 MeV, with three degenerate s-neutrinos weJ find 28.8 GeV at 95% mv > confidence level.

6. SEARCH FOR EXCITED MUONS In this analysis an excited muon is assumed to have spin � and to decay into an ordinary muon and photon with a 100% branching ratio. If the mass of an excited! muon is lighter than half of the a 189

z• mass it can be pair-produced. The z· is assumed to couple to a spin t excited lepton pair in the same way as to ordinary lepton pairs. The single production of an excited muon is also possible through the coupling µ*µz·. In this case, the search can reach µ* mass regions close to the z· mass. To this purpose the processes e+e- µ+µ-yyand e+e- µ+µ-y have been studied [12]. The � � integrated luminosity for this analysis correspond to about 16000z• hadronic decays. We select the 2µ events by using the following cuts:

(1) There are two and only two muon tracks which have the momentum

1.0 > �> 0.04

(2) The displacement of the origin of muons from the interaction point must be

R 50 mm; Z 250 mm < <

(3) The two muons have a scintillator timingdiffere nce less than 3 ns. (4) The number of clusters (except the two from muons) is less than 4. (5) The energy deposited in the Hadron calorimeter is less than GeV. 20 There totally 478 events which pass these cuts. are The standard µµy events are obtained by applying one further cut:

(6) The sum of the two muon momenta is greater than 0.5-VS. The total acceptance for KORALZ events including the inefficiency of cuts is 58.0±3.2%. This yields finally 434 standard dimuon events which will be used for normalization. The µ+µ- yevents are obtained by applying two furthercuts: (7) There is at least one photon which has energy greater than 1 GeV. The angle between this photon and the closest muon must be larger than 5°. (8) The acollinear angle of the two muons is greater than 5°. There are 7 events which passed our µ* cuts. No two photon event was found andwe exclude excited muons with mass less than 45.1 GeV at the 95% confidence level. For the single production we found 7 events whereas from QED background we expect 8.2±2.5 events. By comparing bin by bin (5 GeV /bin) the data with the background, the upper limit of the cross

section, and then the coupling V/mµ* as a function of mµ• was obtained. The result is shown in "J... Fig. 9 together with those from other experiments [13]. The vertical cut-offs are due to the lower limits of mµ• extracted from the pair production. The beam energy and the momentum cut for muons limited our investigation range for mµ• up to 87.4 Ge V. 190

7. CONCLUSIONS In fewmonths of data talcingat Z0 pole L3 has found no evidence of supersymmetric particles nor of excited muons. Furthermore, from the measurement of the Z0 invisible width a lower limit of 28.8 GeV at 95% has been put on the s-neutrino mass.

ACKNOWLEDGEMENTS

I thank the organizers fortheir kind invitation and warm hospitality. I am grateful X. Tata to for collaborations, helpful discussions and clarifying comments on the subject of supersymmetry.

REFERENCES

[l] S.L. Glashow, Nucl. Phys. 22 (1961) 579; S. Weinberg, Phys. rev. Lett. 19 (1967) 1264; A. Salam, Theory, Ed. N. Svartholm, Stockholm, "Almquist and Wiksell" (1968), 367.

[2] Y.A. Golfand and E.P. Likhtman, JETP Lett. 13 (1971) 323; D.V. Volkhov and V.P. Akulov, Phys. Lett. B46 (1973) 109; J. Wess and B. Zumino, Nucl. Phys. B70 (1974) 39; P. Fayet and S. Ferrara, Phys. Rep. 32 (1977) 249; A. Salam and J. Strathdee, Fortschr. Phys. 26 (1978) 57.

[3] R. Kaul, Rev. Mod. Phys. 55 (198 1) 449; Farhi and L. Susskind, Phys. Rep. 74 (1981) 277. E. [4] P. Nath, R. Arnowitt and A. Chamseddine, "Applied N=l Supergravity" ICTP Series in Theoretical Physics Vol. 1, World Scientific, Singapore 1984; H.P. Nilles, Phys. Rep. 110 (1984) l; H.E. Haber and G.L. Kane, Phys. Rep. 117 (1985) 75; Ellis, Lectures presented at the 28th Scottish Universities Summer School in Physics, Edimburg,J. Scotland 1985, CERN prep. TH-4255.

[5] K. Inoue, A. Kakuto, H. Komatsu and S. Takeshita, Prog. Tu,wr. Phys. 67 (1982) 1889; H. Nilles and M. Nusbaumer, Phys. Lett. 145B (1984) 73; P. Majunder and P. Roy, Phys. Rev. 030 (1984) 2432.

[6] M. Chen, C. Dionisi, M. Martinez and X. Tata, Phys. Rep. (1988) 20 1 and references therein. 159

[7] X. Tata, UH-5 11-673-89 and references therein; R. Barbieri et al. "Z0• Physics at LEPl" CERN 89-08 Vol. 2 pag. 119.

[8] B. Adeva et al., Phys. Lett. B233 (1989) 530. 191

[9] L3 Collabortaion, B. Adeva et al., to be published in Nuclear Instrumentsand Methods.

[JO] B. Adeva et al., Phys. Lett. B238 (1990) 122.

[11] H. Baer et al., FSU-HEP-891 104, CERN-TH 5582/89.

[12] Y. Wang, L3 internalreport.

[13] S.K. Kim et AMY Collaboration, Phys. Lett. 223B (1989) 476; I. Adachi et al., TOPAZ Collaboration, Phys. Lett. 228B (1989) 553; K. Abe et al., VENUS Collaboration, Phys. Lett. 213B (1988) 400; H.-J. Behrend et al., CELLO Collaboration, Phys. Lett. 168B (1986) 420; CELLO Collaboration, EPS Conf., Uppsala, June 25 - July 1, 1987 and Int. Symp. on Lepton and Photon Interactions, Hamburg July 27-31, 1987; P. Janot, Thesis Orsay LAL 87- 31 (1987), unpublished; D. Decamp et al., ALEPH Collaboration, CERN-EP/89-167. 192

Table 1 Minimal particle content

Spin 0 2I 1 l2 2

Maner LL , LR l Multiplicity qL , QR q

--I!: I!: + H± w±, wi; w± Field - y y Multiplicity H3o z· z, zh

H�, H� if

G o 193

0.5 100

> > 80 e+e--+ µ+ µ- 0.4 " " <.? <.? LO LO / c:i c:i "' "' 60 0.3 c c " " > > w w e+e-� µ+ µ:- l'+:i I >·:i 40 0.2 :i l+µ�µ+ 'i L y 0.1

0 0 4 8 12 GeV16 20 24 28 Pr

Fig. 1 - Left scale: the missing PT spectrum forthe reaction e+e- � z• � µ+µ- . The observed events represented by the darkhistogram. Right scale: the Monte Carlo predictionfor are the missing PT spectrum in dimuon events from the reaction e+e- � z• � µ + µ- � µ+µ- y y, for a mass of the scalar muon of 41 GeV and a photino mass of 10 GeV. The arrow indicates the PT 6 GeV cut used in the analysis. >

100 0.5

80 e +e� - e+e- 0.4 > > " " ./ 0 0 LO LO c:i 60 0.3 c:i "' .l!l c + -- c "' e e - -+e e "' > � > UJ UJ e"'i 0.2 . "' 40 e+4y ,+ ., + L "' '"' 20 l 0.1

0 0 4 8 12 GeV16 20 24 28 Pr Fig. 2 - Left scale: the missing PT spectrumfor the reaction e+e- � e+e-. The observed events are represented by the dark histogram. Right scale: the Monte Carlo prediction for the missing PT spectrum in dielectron events from the reaction e+e- � + - � e+e- e e y y, for a mass of the scalar electronof 41 GeV and a photino mass of 10 GeV. The arrow indicates the PT 6 Ge V cut used in the analysis. > 194

L3 ALEPH AMY CELLO TOPAZ I I I -2 I I I , I I , ,

,' , ,, .... _.,

,/ /\ ,�, ,,.,_,,.. _.,,. , �\.:. \_, \ ... I I I I I I -) I 1 0

20 30 40 50 60 70 80 90

rn"· (GeV/C' )

Fig. 3 - The upper limit of the coupling Y/mµ• as a function of mµ•. The excluded region is J.. above and left the curves.