Physics XXVI in Collison, Bzios, Rio de Janeiro, 6-9 July 2006 u
1
Searches for New Phenomena at the Tevatron and at HERA
Arnd Meyer
III. Phys. Inst. A, RWTH Aachen, 52074 Aachen, E-mail: [e-mail] Germany
Recent results on searches for new physics at Run II of the Tevatron and highlights from HERA are reported. The searches cover many dierent nal states and a wide range of models. All analyses have at this point led to negative results, but some interesting anomalies have been found.
1. Introduction
arXiv:hep-ex/0610001 v1 2 Oct 2006
For all of the decits of the standard model (SM) that we know about since many years be it the nonunication of couplings at a high scale, the quadratic divergences in the loop corrections to the Higgs boson mass, or the lack of a decent dark matter candidate a large number of solutions has been proposed. We know that within the standard model, the WL WL scattering amplitude violates the unitarity bound at a center of mass energy 1.7 TeV [1], and one solution to this problem is oered by the Higgs mechanism [2], through the introduction of a massive scalar particle. To successfully address the WL WL scattering amplitude problem, the Higgs boson mass is constrained to mH < 1 TeV, and if fermions acquire their masses through coupling to the Higgs boson, then mH < 200 GeV is required [3]. If the Higgs boson doesnt exist, some other form of new physics must be present at the TeV scale to prevent the WL WL scattering amplitude from violating the unitarity bound. The most popular models of new physics involve without doubt supersymmetry. However, supersymmetry doesnt explain the number of fermion generations, or their mass spectrum and charges. In this talk, recent results from the Tevatron for searches for manifestations of new physics are reported, in the areas of supersymmetry, extra gauge bosons, leptoquarks, large extra dimensions, quark and lepton compositeness, the Higgs sector, and a few signature based searches. In addition, selected highlights from signature based searches at HERA are presented. The Tevatron is a proton-antiproton collider with a collision energy of 1.96 TeV in the center of mass system. It is situated at Fermilab near Chicago. Run IIa has ended in February of 2006 with a dataset corresponding to an integrated luminosity of 1.3 fb1 per experiment. This represents about 10 times the statistics collected in Run I. Run IIb has started in June of 2006 with the goal to reach at least 4, but possibly 8 fb1 by the year 2009. The two experiments D and CDF are by now well understood in their capabilities to detect and identify electrons, photons, muons, taus, jets of light and heavy avours, and missing transverse energy ET . The current account of delivered and recorded luminosity by D is shown in THUPL07
Fig. 1. Only recent results based on an integrated luminosity of at least 0.3 fb1 are reported here. Details can be found on the corresponding experiment web sites [4].
Figure 1: Integrated luminosity delivered by the Tevatron and recorded by D during Run II.
The HERA electron(positron)-proton collider at DESY in Hamburg has been delivering luminosity since 1992, at a center of mass energy of up to 319 GeV. HERA is currently in its Run II (HERA2) as well, which started in 2002 and is expected to be completed in 2007. The total integrated luminosity accumulated by the two colliding beam experiments H1 and ZEUS for the analyses shown here is up to 300 pb1 , with another substantial increase of the data set expected for the entire HERA data set, as can be seen for H1 in Fig. 2.
2. Isolated Leptons at HERA
Soon after turning on HERA, the H1 experiment reported [6] in 1994, after analyzing the rst 4 pb1 of data, the observation of an event with an isolated muon recoiling against a hadronic system, both of high transverse momentum pT . In addition, substantial missing transverse energy was reconstructed. The dominant SM process leading to such a nal state is
2
XXVI Physics in Collison, Bzios, Rio de Janeiro, 6-9 July 2006 u
Status: 15-Sep-2006 H1 Integrated Luminosity / pb
-1
1/07/07
electrons positrons
bined is shown in Fig. 4, with a fair agreement of data and SM expectation.
l+Pmiss events at HERA 1994-2006 (e p, 341 pb ) T
-1
300
Events
HERA-2
200
102
H1 Data (prelim.) All SM Signal
NData = 46 NSM = 43.0 6.0
HERA-1
100
10
1
0 0 500 1000 1500
Days of running
10-1
0
Figure 2: Integrated luminosity accumulated by the H1 detector at HERA [5].
PX (GeV) e and channels T
10
20
30
40
50
60
70
80
photoproduction of W bosons with a subsequent leptonic decay of the W : e+ p e+ W + X e+ + X (Fig. 3), where the positron escapes detection and the neutrino leads to reconstructed ET . However, the total cross section for this reaction is only 40 fb. In the following years, H1 has reported on the observation of more such events, in excess of the SM expectation, but not statistically signicant to claim new physics.
e e ,Z q W q f
Figure 4: The hadronic transverse momentum distribution in the H1 search for isolated lepton events [7]. Electron and muon channels are combined. The SM expectation (histogram) is dominated by processes with genuine isolated leptons and missing ET (Signal), which in turn is dominated by real W production.
X At large hadronic transverse momentum PT > 25 GeV, a total of 18 events are observed compared to a SM prediction of 11.5 1.8. As can be seen from Table 5, the excess is, within statistics, observed in the e+ p data only. The probability for the SM expectation to uctuate to the observed number of events X or more in the high PT domain for all data is 6.7%, compared to 0.15% for the HERA I data (the majority of which is e+ p data). For the e+ p data alone, this probability is 0.03%.
e e ,Z q W q f
f
f
Electron Muon obs./exp.
(Signal contribution) 9 / 3.9 0.6 (84%) 6 / 2.3 0.4 (84%) 2 / 5.1 0.7 (78%) 0 / 3.1 0.5 (74%) 11 / 9.0 1.4 (80%) 6 / 5.4 0.9 (77%)
Figure 3: Main leading order Feynman diagrams for the process ep eW X [8].
Combined obs./exp.
(Signal contribution) 28 / 18.5 2.6 (73%) 15 / 4.6 0.8 (82%) 18 / 24.4 3.4 (68%) 3 / 6.9 1.0 (67%) 46 / 43.0 6.0 (70%) 18 / 11.5 1.8 (71%)
H1 Preliminary
obs./exp.
(Signal contribution)
These analyses have inspired a large number of possible interpretations in terms of physics beyond the standard model to name a few, leptoquarks, excited fermions, supersymmetry with R-parity violation, or single top production via avour changing neutral currents.
1994-2004 e+ p 158 pb1 1998-2006 e p 184 pb1 1994-2006 e p 341 pb1
Full Sample
X PT > 25 GeV
19 / 14.6 2.0 (70%) 9 / 2.3 0.4 (80%) 16 / 19.4 2.7 (65%) 3 / 3.8 0.6 (61%) 35 / 34.0 4.7 (68%) 12 / 6.1 1.1 (66%)
Full Sample
X PT > 25 GeV
Full Sample
X PT > 25 GeV
2.1. High pT Leptons and Missing ET
The H1 collaboration has now updated the analysis with all data collected until the Summer of 2006, corresponding to 341 pb1 of data [7]. The event selection requires as before a high pT isolated electron or muon, and substantial missing transverse momentum. The distribution of the hadronic transverse momentum pX , determined from all reconstructed particles T excluding identied isolated leptons, for all data comTHUPL07
Figure 5: Summary [7] of the H1 results of searches for events with isolated electrons or muons and missing transverse momentum for dierent data sets: e+ p, e p, and all data; e and channel as well as combined; with and without the requirement of large hadronic transverse momentum pX . The number of observed events is T compared to the SM prediction. The signal component of the SM expectation, dominated by real W production, is given as a percentage in parentheses.
ZEUS has analyzed 249 pb1 of data taken during
XXVI Physics in Collison, Bzios, Rio de Janeiro, 6-9 July 2006 u the years 1998 to 2005 with a similar selection [8], and found the rate of such events at high hadronic transverse momentum to be consistent with the SM predictions (Fig. 6). The excess observed by the H1 collaboration is not conrmed. Both experiments have studied dierences in the acceptances and eciencies of the respective analyses, and reached the conclusion that the experiments have comparable sensitivity in the regions where the H1 excess is observed [7, 8]. Unfortunately, taking into account the amount of HERA data still to be analyzed in the future, it seems increasingly unlikely that the mystery of the isolated lepton events can be resolved by additional data alone. where the hadronic system has a transverse momenX tum PT > 25 GeV, three events are observed in the data where the SM expectation is 0.74+0.19 events. 0.16 All three events have been collected in e p collisions, in contrast to the excess observed in the e and channels. The pX distribution is shown in Fig. 7. T
3
Figure 7: The hadronic transverse momentum distribution of + ET events in H1 data [9]. The SM expectation is shown as histogram with uncertainty band. The signal component of the SM expectation, dominated by real W production, is given by the hatched histogram.
Figure 6: Summary of the results of searches for events with isolated electrons (top) or muons (bottom) and missing transverse momentum at HERA, as shown in [8]. The number of observed events is compared to the SM prediction (observed/expected). The signal component of the SM expectation (W production) is given as a percentage in parentheses. Only the H1 results directly comparable to the ZEUS results are quoted in this Table.
The ZEUS collaboration has published the results of an analysis [10] of events containing isolated tau leptons and large missing transverse momentum based on 130 pb1 of HERA I data. For the identication, six observables based on the internal jet structure were exploited to discriminate between hadronic decays and quark- or gluon-induced jets. Three tau candidates were found, while 0.40+0.12 were expected from 0.13 X SM processes. Requring PT > 25 GeV, two candidate events remain, while 0.20 0.05 events are expected from SM processes, about half of which is real W production. Both events occured in e+ p collisions.
2.3. Events with Multiple Leptons 2.2. High pT Taus and Missing ET
In order to shed additional light on the question of isolated lepton production at HERA, both H1 and ZEUS have investigated the production of high pT tau leptons. In the latest H1 analysis [9], data corresponding to an integrated luminosity of 278 pb1 have been used. The leptons are identied by using an identication algorithm based on the search for isolated charged tracks associated to narrow hadronic jets detected in the calorimeters, a typical signature of the one-prong hadronic decay. The 25 events found in the data are in good agreement with the SM expectation of 24.2+4.2 events. In the region 5.8 THUPL07 Both ZEUS [11] and H1 [12] have studied the production of events containing multiple high pT isolated leptons. The dominant SM contribution to these nal states is the two photon process l+ l , which can be accurately predicted. ZEUS [11] has analyzed 296 pb1 of data for the ee and eee nal states, and compared event yields and kinematical distributions, see for example Figs. 8 and 9. Data and SM expectations are found to be in agreement. The nal state has been analyzed in a smaller data set corresponding to 135 pb1 , and again no deviation from the SM prediction has been found. The H1 collaboration has updated and extended their analysis of multi-lepton events [12], using
4
XXVI Physics in Collison, Bzios, Rio de Janeiro, 6-9 July 2006 u
ZEUS
Events 102 10 1 10
-1 ZEUS (prel.) 96-05 (296 pb ) SM QEDC NC
-1
Events
10
Events
3
102 10 1
10
3
102 10 1
ee
10-1 10-2
10-1 10-2
0
20
40
60
80 100 120 140 160 180
0
20
40
60
80 100 120 140 160 180
PT (GeV)
10 10
3
PT (GeV)
2
H1 Data (prelim.) DIS+Compton Pair Production
10-2 0
50
100
150 200 M12(GeV)
Events
10 1 10-1 10-2
Figure 8: In the ZEUS analysis of events with two high pT electrons [11], comparison of the observed invariant mass M12 of the two electrons with the SM expectation. The contributions of the QED Compton and neutral current DIS processes are also shown.
0
20
40
60
80 100 120 140 160 180
PT (GeV)
Figure 10: In the H1 analysis of events with two or three high pT electrons or muons [12], distributions of the scalar sum of the lepton transverse momenta compared to the expectations, separately for e+ p data, e p data, and the entire data set.
2.4. Model Independent Search
The H1 collaboration has previously presented an analysis based on the HERA I data using a model independent approach to search for deviations from the standard model, reporting no signicant ndings. Since the HERA I data consisted mostly of e+ p data, the analysis has recently been updated with the e p data collected in the years 20052006 and corresponding to an integrated luminosity of 159 pb1 . Events are assigned to exclusive classes according to their nal state involving isolated electrons, photons, muons, neutrinos (ET ) and jets with high transverse momenta. The event yields in the dierent classes are shown in Fig. 11 together with the SM expectation. The class has been discarded because it is dominated by poorly reconstructed muons giving rise to large ET . A statistical algorithm is applied to search for deviations from the SM in the distributions of the scalar sum of transverse momenta or the invariant mass of nal state particles, and to quantify their signicance. No signicant deviation has been found.
Figure 9: In the ZEUS analysis of events with three high pT electrons [11], comparison of the observed invariant mass M12 of the two highest pT electrons with the SM expectation. The contributions of the QED Compton and neutral current DIS processes are also shown.
275 pb1 of data. The nal states ee, , e, eee and e have been studied for anomalies, see for example Fig. 10 for the distributions of the scalar sum pT of the lepton transverse momenta for all nal states combined. For pT > 100 GeV, four events are observed with a SM expectation of 1.1 0.2. All four events have been collected in e+ p collisions, and three of the four events are in the eee nal state and have an invariant mass of the two leading electrons of M12 > 100 GeV. Apart from this moderate but interesting disagreement, data and SM expectation are found to be in good agreement. THUPL07
3. Supersymmetry
Supersymmetry or SUSY, a proposed invariance of nature for the interchange of fermionic and bosonic degrees of freedom, has many important features which justify an intensive research at the highest energy accelerators. It allows for the unication of the four known forces, it is the only non-trivial extension of
XXVI Physics in Collison, Bzios, Rio de Janeiro, 6-9 July 2006 u
H1 General Search e- p (159 pb-1)
H1 Data (prelim.) SM
5
Events
105 104 103 102 10 1 10-1
e-j-j j-j- e-e-j e-e-e
j-j e-j -j
e-e e- -
Figure 11: In the generic H1 search for deviations from the SM [13], the observed number of events in the dierent exclusive event classes, as well as the SM expectation with its uncertainty. Only nal states with either at least one data event or a SM expectation greater than one event are shown.
j-j-j-j-
-j- e-j- j- - e-j-j-j j-j-j-
j- e-
e-j-
j-j-j
j-j-j-j
j- e-
10-2
portant SM backgrounds are tt production and gauge boson production either in pairs or accompanied by jets. In models with R-parity conservation, the LSP, in most cases the 0 , is a natural dark matter candi1 date. Nevertheless, R-parity violation (R) is not excluded, and remains an interesting alternative. R would allow single resonant production of SUSY particles, and often even more jets or leptons in the nal state from the B or L violating processes. In this case the theory contains 48 additional unknown Yukawa couplings. It is usually assumed that only one of these couplings is non-zero. At the Tevatron both R-parity conserving and R processes have been studied.
3.1. Supersymmetry with R-Parity Conserved
The searches for SUSY under the assumption of R-parity conservation are presented as follows: searches for charginos and neutralinos, and squarks and gluinos, in mSUGRA and mSUGRA inspired scenarios, which are benchmark processes at the Tevatron. Following this, results for alternative mechanisms of SUSY breaking are discussed, including split SUSY, GMSB, and anomaly mediated SUSY breaking.
3.1.1. Associated Production of Charginos and Neutralinos
the Lorentz-Poincar group, and it provides an elegant e solution to evade the ne tuning problem of the standard model. In SUSY every SM particle has a partner diering in spin S by 1/2. The SUSY partners are assigned an R-parity R = (1)3B+L+2S = 1, where B is the baryon and L is the lepton number of the particle, in contrast to the SM particles of R = +1. A second Higgs doublet has to be introduced, leading to four additional Higgs particles which have R = +1. In the minimal supersymmetric extension of the SM, the MSSM, 105 additional parameters are introduced, corresponding to sparticle masses, mixing angles etc. Since the R = 1 partners have not yet been observed in nature, SUSY cannot be an exact symmetry. Various mechanisms for SUSY breaking have been proposed, each of them requiring a dierent set of new model parameters. Under certain assumptions the number of free parameters can be reduced to managable numbers. In the model that is probably most studied, minimal supergravity or mSUGRA, ve parameters remain: the common scalar and fermion masses at the GUT scale, m0 and m1/2 , the ratio of the vacuum expectation values of the two neutral Higgs elds tan , the trilinear coupling parameter A0 , and the sign of the higgsino mass parameter . In the case of minimal gauge mediated SUSY breaking (mGMSB), the six parameters are , Mm , N5 , Cgrav , tan and the sign of . In most cases R-parity is assumed to be conserved since there are severe limits on B and L violating processes. Then, the SUSY partners are pair produced and the lightest SUSY particle (LSP) is neutral and weakly interacting, and thus escapes detection. Therefore, the basic experimental signature for R-parity conserving SUSY is missing transverse energy and multiple jets and leptons originating from the cascade decay of the heavy R = 1 partners. ImTHUPL07
The dominant production mechanisms of charginos and neutralinos at the Tevatron along with their leptonic decays are depicted in Fig. 12. The golden experimental signature searched for is three leptons, accompanied by ET . For increased acceptance, the third lepton is sometimes identied as an isolated track, or not required to be found at all in the case of two leptons of same charge. The SM backgrounds (Z/ + jets, QCD (multijets), W W/W Z and tt production) are small and well under control already at the preselection stage, as for example seen in Fig. 13.
Figure 12: Dominant production of charginos and neutralinos at the Tevatron and their decays.
Because of the small leptonic branching fractions several nal states need to be combined. In Table I the results in twelve dierent channels studied by the two collaborations are listed. Since the observed number of events is in agreement in every channel with the
6
XXVI Physics in Collison, Bzios, Rio de Janeiro, 6-9 July 2006 u
Data Z ee Z tautau W e nu Y ee tt 2l 2b WW 2l + 2nu WZ 3l + nu ZZ 4l or 2l + 2nu QCD SUSY
CDF Run II Preliminary Events/9 GeV/c 5 4 3 2 1 0 20
M0=100, M1/2=180, tan=5, >0
Events / 1 GeV
104 10
3
D RunII Preliminary 1.1 fb
-1
Data
L dt = 0.7 fb
-1
QCD Diboson W DY mSUGRA point
102 10 1 10-1 10-2 10
-3
0
10
20
30
40
50
60
70
80
90
ET [GeV]
100
30
40
50
60 70 80 90 100 110 Leading lepton P (GeV/c)
t
Figure 13: Distribution of ET at the preselection stage of the eel chargino/neutralino search by D.
Figure 15: Transverse momentum of the leading lepton in the CDF search for charginos and neutralinos in like sign dileptons. D channels Lint [fb1 ] Background Data 1.1 0.32 0.32 0.9 0.33 0.33 0.35 0.61 0.31 0.75 0.75 0.70 0.76 0.67 1.75 0.57 0.31 0.13 1.1 0.4 0.58 0.14 0.36 0.13 0.17 0.05 0.49 0.10 0.13 0.03 0.64 0.18 0.78 0.11 6.8 1.0 0 2 0 1 0 1 0 1 0 1 0 9
predicted background, upper limits on the cross section times branching fraction were derived, as shown for example in the case of D in Fig. 14. The theoretical predictions are for three mSUGRA inspired scenarios for mass relations as indicated in the gure. Lower limits of the chargino mass have been derived for two scenarios with large leptonic branching fractions: m( ) < 140 GeV is excluded when the 1 slepton mass is slightly above the mass of the second neutralino, thus allowing only 3-body decays (denoted 3l-max), and m( ) < 154 GeV is excluded for the 1 case when squarks are heavy and therefore the destructive t-channel contribution is minimal. The CDF analyses have comparable sensitivity. The slight excess in the CDF same sign dilepton channel is worth noting, because it turns out that four of the nine events have a leading lepton with high transverse momentum in excess of 60 GeV, where neither SM background nor the SUSY signal are expected (Fig. 15).
ee + track + track e + track + + / e + track + track CDF channels ee + e/ ee + track + e/ (low pT ) + e/ (high pT ) e + e/ e e , e ,
Lint [fb1 ] Background Data
Table I Three and two lepton nal states studied by the D and CDF collaborations in the search for charginos and neutralinos.
(12) BR(3l) (pb)
0.5
3.1.2. Squarks and Gluinos
D Run II Preliminary, 0.3-1.1 fb
-1
0.4 0.3
M(1)M(2)2M(1); M(l)>M(2) tan=3, >0, no slepton mixing Observed Limit Expected Limit
r ua
0
0
~
0
0.2 0.1
LEP
large-m0 0 100 110 120 130 140 150 160
Chargino Mass (GeV)
Figure 14: Cross section upper limits times branching fraction of chargino and neutralino production measured by D along with theoretical predictions.
D has carried out a generic search [14] for gluinos and squarks requiring a minimum number of jets, Nj , accompanied by substantial ET and HT , the scalar sum of the jet transverse energies, for the following three topologies: (i) Nj = 2 for Mq < Mg , (ii) Nj = 3 for Mq Mg , and (iii) Nj = 4 for Mq > Mg , where Mq and Mg are the mass of the squark and the mass of the gluino, respectively, and the jet multiplicities are chosen corresponding to the decay modes q q 0 and g q q 0 . A general agreement between 1 1 the data corresponding to an integrated luminosity of 310 pb1 and the expected background at all stages of the selection is observed, for the D analyses as well as for the CDF analysis in the 3-jet nal state, as can be seen for example in Fig. 16, where the CDF HT distribution optimized to search for relatively small gluino masses is displayed. The three D analyses
he y av -sq
0
ks
3l -m ax
THUPL07
XXVI Physics in Collison, Bzios, Rio de Janeiro, 6-9 July 2006 u optimized for the three dierent mass hierarchies are combined, and the exclusion region in the (Mq , Mg ) plane as shown in Fig. 17 is obtained. Also shown is the excluded region obtained by CDF using the 3-jet event topology and 371 pb1 of data. The absolute lower mass limits at 95% C.L. are Mq > 325 GeV and Mg > 241 GeV, while for Mq Mg , Mq, < 387 GeV g can be excluded.
CDF Run II Preliminary Events / 60 GeV
Data (L = 371 pb-1) QCD tt + QCD W, Z, WW + tt + QCD mSUGRA
Events/5 GeV
7
ture is two acoplanar b jets and ET . In both analyses at least one of the jets was required to be identied as a b jet using lifetime information. The selection value of the ET and that of the ET of the jets were optimized according to the mass value of the sbottom to be detected. The data did not show any signicant excess over the expected SM background as can be seen for example in Fig. 18, where the ET distribution obtained by CDF is shown. The excluded mass values of the sbottom and the neutralino are shown in Fig. 19, substantially improving on previous limits.
102
10
(M~ ~ M~ ~ 252 GeV/c2) s
g
6
CDF Run II Preliminary, 295 pb 5
-1
Data Mis-Tag QCD (HF) EWK+Top Signal
(M( b1)=160 GeV/c ,
~
2
1
ET > 75 GeV
4 3
200
300
400
500
600
700
2
HT [GeV]
1
M( 1 )=80 GeV/c )
0
2
Figure 16: Distribution of HT obtained by CDF for the 3-jet event topology and a light gluino, compared with the SM background and the expected SUSY signal.
0 0
50
100
150
200 250 Missing ET (GeV)
600
Figure 18: Distribution of ET in events with at least one b-tagged jet observed by CDF in the search for sbottom pair production with b0 , together with the SM b 1 background and the expected signal.
D0 Run II hep-ex/0604029
500
UA2
M
~ q
=
M ~g
M(1) (GeV/c )
400
2
120
CDF Run II Preliminary (295 pb )
Observed Expected limit D Run 2 CDF Run 1 LEP
CDF Run2 : Theoretical uncertainty incl. PDF, and renormalization and factorization scale D Run2 : Theoretical uncertainty incl. renormalization and factorization scale
-1
Mq (GeV/c 2) ~
UA1
CDF Run II 3-jets analysis
300
no mSUGRA solution
100
0
60
100
LEP 1 + 2 ~ m(q) < m( )
0 1
40
600
0 0
100
200
300
400
500
Mg (GeV/c2) ~
20
Figure 17: Excluded regions in the (Mq , Mg ) plane obtained by CDF and D. It should be noted that D has used conservatively a theoretical cross section reduced by its uncertainty when calculating limits.
0 0
20
40
M( ~ b)
60
1
=M
80 100 120 140 160 180 200 220 2 ~
(b)
200
FNAL Run I
80
+M ( 0 )
1
M(b1) (GeV/c )
Third generation squarks may be light due to the large mixing between the scalar partners of the left and right handed quarks. They may be accessible at the Tevatron and are therefore subject of dedicated analyses by the two collaborations. Both D [15] (Lint = 310 pb1 ) and CDF (295 pb1 ) have searched for direct pair production of the lightest sbottom quark 1 , assuming that it deb cays with a branching fraction of 100% into a b quark and the lightest neutralino. The experimental signaTHUPL07
Figure 19: Mass values of the sbottom and neutralino excluded by the D and CDF analyses of sbottom pair production followed by the decay b0 . b 1
If the mass Mt1 of the lightest stop quark t1 satises the relation Mc + M0 < Mt1 < Mb + MW + M0 , where Mc , Mb , M0 and MW are the masses of the 1 c quark, the b quark, the lightest neutralino and the W boson, respectively, its dominant decay mode is t1 c0 , and a similar search strategy can be ap1 plied as for the sbottom analysis outlined above, ex1 1
8
XXVI Physics in Collison, Bzios, Rio de Janeiro, 6-9 July 2006 u cept that jets should satisfy a c-tag instead of a b-tag criterion. Again, data observed by D and by CDF are in agreement with the SM expectation (see for example Fig. 20), and exclusion limits can be set in the mass plane of the lightest stop and the neutralino, extending the previously excluded regions of Mt1 and M0 , as shown in Fig. 21.
1
a certain number of non-isolated tracks instead of an explicit jet reconstruction, thus increasing the sensitivity. In neither channel a signicant signal for the presence of the t1 quark has been observed. Therefore, the two analyses have been combined to exclude masses of the t1 quark and the sneutrino. The mass region excluded by earlier experiments has been signicantly extended, as can be seen in Fig. 22.
Events / 10 GeV
30 25 20 15 10 5 0 50
D Run II Preliminary
Data SM Signal QCD
100
150
200 250 Missing E T (GeV)
Figure 20: Final ET distribution in events with at least one c-tagged jet observed by D in the search for stop pair production with t c0 , together with the SM 1 background and the expected signal for Mt1 = 130 GeV and M 0 = 50 GeV. 1 Figure 22: Excluded regions of the stop and sneutrino masses obtained by D assuming the decay t1 b + l + (l = e, ).
M(1) (GeV/c )
2
100 90 80 70 60 50
CDF Run II Preliminary (295 pb )
CDF,D Run2 Theoretical uncertainty incl. PDF, and renormalization and factorization scale
-1
M ~ (t 1 )= M
(c
Observed Expected limit D Run 2 CDF Run 1 D Run 1 LEP =56o o LEP =0
0
( 0 1)
3.1.3. Split SUSY
)+
M
10 0 50 60 70 80 90
100 110 120 130 140 2 ~
M ~ (t 1 )= M (b
20
Figure 21: Mass values of the lightest stop and neutralino excluded by the D and CDF analyses of stop pair production followed by the decay t c0 . 1
D has also searched for pair production of t1 quarks assuming that they decay into a b quark, a lepton and a sneutrino via a virtual chargino, which may be favorable due to the relatively weak constraint on the sneutrino mass from LEP, M > 43.7 GeV. The nal state is two isolated leptons with opposite charge, two b jets, and ET . For the two leptons, the combinations and e have been analyzed. While in the analysis a b jet is required to be identied, the e analysis has smaller backgrounds and requires THUPL07
)+
M
M(t1) (GeV/c )
(W
30
)+
M ( 0 1)
40
Split SUSY is a relatively new variant of supersymmetry in which the supersymmetric scalars are heavy (possibly GUT scale) compared to the (SUSY) fermions [16]. It avoids much of the ne tuning required to remain consistent with observations while still preserving the favored consequences: a darkmatter candidate is still present in the theory to explain the observed cold dark matter density in the universe, and coupling unication at the GUT scale still occurs. Due to the high masses of the scalars, the gluino decays are suppressed. The gluinos have time to hadronize into R-hadrons, colorless bound states of a gluino and other quarks or gluons. At the Tevatron, R-hadrons could be pair produced copiously through strong interactions. About half of the R-hadrons are expected to be charged, and some charged R-hadrons can become stopped gluinos by losing all of their momentum through ionization and coming to rest in the calorimeters. Their decays may happen after several bunch crossings. D has searched for stopped gluinos in the g g 0 1 decay channel by looking for randomly oriented jets in bunch crossings without an inelastic p collision. p The background consists mainly of cosmic and beam
XXVI Physics in Collison, Bzios, Rio de Janeiro, 6-9 July 2006 u halo muon induced jets where the muon escapes reconstruction. It has been estimated using the data. The energy distribution of the observed jets is shown in Fig. 23, along with that of the estimated background. As can be seen the data do not show any excess above the expected background. The derived cross section upper limits are shown in Fig. 24 as a function of the gluino mass for dierent mass values of the 0 . The theoretical cross section is also shown, 1 from which gluino masses below 300 GeV can be excluded if the mass of the 0 is less than 90 GeV. It is 1 worth noting that the signature of a long-lived gluino could occur in many models of beyond the standard model physics.
3.1.4. Gauge Mediated SUSY Breaking
9
In the mGMSB scenario, the gravitino is the LSP with a mass less than 1 keV, and the next-to-lightest SUSY particle (NLSP) may be the lightest neutralino, which decays to a gravitino and a photon. The lifetime of the NLSP is a priori unknown and depends on the Cgrav parameter of the model. D and CDF have searched for two highly energetic prompt photons accompanied by large ET , caused by the undetected gravitino. As can be seen in Fig. 25, the observed ET distribution is compatible with the SM background. With recent data corresponding to 760 pb1 , D has improved the previous limits obtained from combined measurements of CDF and D, see Fig. 26. For the parameter, determining the eective scale of SUSY breaking, < 88.5 TeV is excluded at 95% C.L. This corresponds to a 0 mass of > 120 GeV and a 1 1 mass of > 220 GeV.
Figure 23: In the D search for stopped gluinos, energy distribution of the selected jets (dots with error bars), of the background (black histogram) dominated by cosmic muon events , and of a gluino of mass 400 GeV decaying into a gluon and a neutralino of 90 GeV mass.
Figure 25: In events with two photons, ET distribution of recent D data (full circles), of the background (histogram), and the fraction of the latter with true ET (open circles).
Figure 24: Expected and observed cross section upper limits from D for stopped gluinos as a function of the gluino mass for neutralino masses of 50 GeV, 90 GeV and 200 GeV. Also shown is the theoretical cross section (red star).
Figure 26: Cross section upper limits (solid blue line) in the D GMSB analysis with two photons and ET . The theoretical LO (NLO) cross section for the GMSB process (solid (dashed) black lines) corresponds to the SUSY Snowmass slope 8, with a messenger mass Mm = 2, the number of messengers N5 = 1, tan = 15, and > 0.
THUPL07
10
XXVI Physics in Collison, Bzios, Rio de Janeiro, 6-9 July 2006 u CDF searched for signs of GMSB assuming long lifetime for the NLSP, i.e. the neutralino, by looking for late photons in the calorimeter, using 570 pb1 of data. Fig. 27 shows that the arrival time distribution of photons accompanied by ET doesnt show a significant excess at positive times where the GMSB signal is expected. Ten events are observed with 7.6 1.9 background events expected. This allows to exclude a region in the plane of the neutralino lifetime versus the neutralino mass, as shown in Fig. 28.
CDF Run II Preliminary, 570 pb
-1
10
2
Events/0.5 ns
10
+ ET + Jet data All Collisions Beam Halo Cosmics GMSB Signal MC
1
chargino and neutralino is expected to be small, less than 150 MeV. Therefore charginos can have a long lifetime, and leave a muon-like signature in the detector. However, due to their mass they move slowly, and the speed signicance, dened as s = (1v)/v , where v is the speed of the chargino in units of the speed of light as measured in the muon system and v is its uncertainty, is shifted towards positive values depending on the mass, as indicated in Fig. 29 for heavy staus. In the D analysis of 390 pb1 of data, two muons with s > 0 were required, and a nal optimized cut was placed in the plane of the dimuon invariant mass and s. The observed number of events is compatible with the expected background, estimated from muon pairs from the Z + decay. An upper limit of the production cross section was derived (Fig. 30) which, by confronting it with the theoretical cross section, excludes wino-like charginos with a mass of less than 174 GeV.
Events/0.1 D Run II Preliminary
10-1
-10 -8 -6 -4 -2 0 2 4 6 8 10 Photon Corrected Time of Arrival (ns)
1600 1400 1200
Figure 27: Photon arrival time distribution in the CDF electromagnetic calorimeter. The expected GMSB signal is for a neutralino mass of 93.6 GeV with a lifetime of 10 ns.
Data muons
1000 800 600 400
100 GeV staus 300 GeV staus
CDF Run II Preliminary, 570 pb
1 lifetime (ns) 30
Predicted exclusion region with a +ET+1jet analysis with EMTiming Observed exclusion region with a +ET+1jet analysis with EMTiming
-1
200 0 -4 -2 0 2 4 6 8 10 12 14
Speed Significance
25
20
15
Figure 29: Speed signicance distributions of muons and of long-lived staus for two dierent masses. The distributions are expected to be similar for charginos of the same mass.
0
10
10
D Run II Preliminary L = 390 pb
-1
5
(p p +- ) (pb) 1 1
0
95% CL Cross Section Limit
70
75
80
85
90
95 100 105 110 2 0 1 mass (GeV/c )
1
NLO Cross Section Prediction
10-2
3.1.5. Anomaly Mediated SUSY Breaking
Many SUSY scenarios contain charged massive long-lived particles expected to traverse the entire detector, for example staus or charginos with long lifetime. D has studied chargino pair production assuming anomaly mediated supersymmetry breaking. In this case the mass dierence of the lightest THUPL07
50
LEP Excluded
100
Figure 28: Excluded mass and lifetime values of the lightest neutralino (NLSP) in the CDF GMSB analysis with late photons.
10-1
150
200
250
300
Gaugino-like Chargino Mass (GeV)
Figure 30: Cross section upper limit obtained by D and theoretical cross section for a gaugino-like chargino as a function of its mass.
XXVI Physics in Collison, Bzios, Rio de Janeiro, 6-9 July 2006 u
11
3.2. R-Parity Violation
While in many analyses R-parity is assumed to be conserved, which leaves the lightest supersymmetric particle (LSP) stable, SUSY does not require R-parity conservation. If R is allowed, the following trilinear and bilinear terms appear in the superpotential [17]: 1 WR = + ijk Li Lj Ek + Li Qj Dk ijk 2 1 + Ui Dj Dk + i Li Hu 2 ijk where L and Q are the lepton and quark SU(2) dou blet superelds, E, U, D denote the singlet elds, and i, j, k refer to the fermion families. The rst two terms imply lepton number violation, while the third term leads to baryon number violation. The coupling strengths are given by the Yukawa coupling constants , and . The last term, i Li Hu , mixes the lepton and the Higgs superelds. The and couplings give rise to nal states with multiple leptons, which provide excellent signatures at the Tevatron. Stringent limits exist on the size of many R couplings [18, 19], in particular for the case of more than one non-zero coupling.
3.2.1. Gaugino Production in Multi Lepton Final States
charged leptons are required to be identied. Three dierent analyses eel, l and ee with l = e, are performed depending on the avors of the leptons in the nal state. All three analyses are optimized separately using SM and signal Monte Carlo simulations. After all cuts, no events remain in the data, while 0.9 0.4, 0.4 0.1, and 1.3 1.8 events are expected from SM processes in the eel, l, and ee analysis, respectively. The SM dominant backgrounds are due to Z/ l+ l and diboson production. Since no evidence for R-SUSY is observed, the analyses are combined and upper limits on the chargino and neutralino pair production cross section are set. Lower bounds on the masses of the lightest neutralino and the lightest chargino are derived in mSUGRA and in an MSSM scenario with heavy sfermions, but assuming no GUT relation between M1 and M2 . The limits as shown in Figs. 32 and 33 are the most restrictive to date. CDF has shown preliminary results with comparable sensitivity.
3.2.2. Long-Lived Neutral Particles in Dimuon Final States
NLO, 95%CL [pb]
The charginos and neutralinos are produced in pairs or associated and with R-parity conserved. The produced sparticles (cascade) decay to the lightest neutralino 0 . Under the assumption of a single non1 zero LLE coupling, the neutralino decays into two charged leptons and one neutrino by violating Rparity (Fig. 31). The nal state therefore contains at least four charged leptons and two neutrinos which lead to missing transverse energy in the detector. Both D and CDF have searched for this signature, assuming a suciently large LLE coupling leading to a prompt decay of the lightest neutrlino, i.e. the cor responding LLE-coupling (121 , 122 , or 133 ) has to > 0.01. In the D analysis [20], based be larger than on 360 pb1 of data, for best acceptance only three
Figure 31: Two examples of R-decays of the lightest neutralino via the LLE couplings 1jk . In each decay, two charged leptons and one neutrino are produced.
In this D analysis [21], a small LLE coupling 122 is assumed, leading to a long neutralino 0 lifetime 1 and consequently to a displaced dimuon vertex. The nal state is of particular interest due to an anomaly reported by the NuTeV collaboration; in 2000, NuTeV reported [22] on a search for heavy neutral leptons decaying to , amongst other nal states. In the dimuon channel, they observed three events while only 0.07 events were expected. Because of the asymmetry in the muon momenta, and the absence of a signal in other channels, NuTeV argued that the signal was unlikely to be due to neutral heavy leptons. To shed light on the origin of these events, D has searched for pairs of oppositely charged isolated muons originating from a common vertex located be M 1 [GeV]
300
2 1
150
200
-1
250
D, 360 pb
NLO (SUSY) expected limits 121 observed 122 observed 133 observed
10-1
m0 =1 TeV tan =5, A 0=0, >0 210 60 80 100
-2
120
0 M 1 [GeV]
140
160
Figure 32: Cross section limits from D for three dierent LLE couplings compared to the mSUGRA cross section prediction.
THUPL07
12
XXVI Physics in Collison, Bzios, Rio de Janeiro, 6-9 July 2006 u tween 5 and 20 cm radially displaced from the beamline. In this region, a good calibration using KS mesons is possible. No events have been found in 380 pb1 of data, with an estimated background of 0.8 1.6 events. The result is interpreted as cross section upper limit on the production times branching fraction of a neutral long-lived particle decaying into + + X as a function of its lifetime, as shown in Fig. 34. To compare with the NuTeV result, the momentum of the hypothetical new particles in the neutrino beam was converted to the Tevatron center of mass frame. While the result is somewhat dependent on the assumptions made regarding the decay of the long-lived particle, this result excludes an interpretation of the NuTeV excess of dimuon events in a large class of models.
3.2.3. Resonant Second Generation Slepton Production
0 0 (pp NLLNLL) BF(NLL + -+X) (pb)
109 108 107 106 105 104 103 102 10 1 10-1 10-2 -13 10
NuTeV 99% Exclusion
9% Pr ef er
r
ed
Re
gi
on
D
Nu
Te
V
9
0
L = 380 25 pb
-1
D 95% Exclusion D 99% Exclusion
10-11
10-9
10-7
10-5
10-3 10-1 1 Lifetime (s)
Figure 34: Cross section upper limits times branching fraction for neutral long-lived particles decaying to muon pairs as a function of their lifetime.
The LQD coupling oers the opportunity to produce sleptons in p collisions as resonances. For a p non-zero coupling 211 this is either a smuon or a muon sneutrino. The slepton cascade decays into the lightest neutralino 0 and associated leptons. The 1 neutralino decays via the same R-parity violating coupling into a 2nd generation lepton and two jets. 211 The cross section is proportional to ( )2 , so that 211 limits on this coupling can be derived. D has recently published [23] the results of a search for resonant second generation slepton produc tion. The three channels (i) 0 , (ii) 1 0 , and (iii) resulting in dimuon 2,3,4 1,2 and multijet nal states are analyzed separately. For the further discrimination of the signal and the SM background, the analysis makes use of the possibility to reconstruct the neutralino and the slepton masses: using the leading muon and the two leading jets one
M [GeV] 1
would be able to reconstruct the lightest neutralino, and a peak in the invariant mass of the two muons and all jets would indicate the presence of the slepton. The selection criteria are optimized depending on the slepton and neutralino mass, and for all 117 mass combinations being probed, the data corresponding to 380 pb1 show no excess with respect to the SM expectation. In the absence of an excess in the data, cross section limits on resonant slepton production were set. The results are interpreted within the mSUGRA framework with tan = 5, < 0 and A0 = 0, and an exclusion contour as a function of is derived af211 ter combination of all three channels, as shown in Fig. 35. Lower limits for the slepton mass of 210, 340 and 363 GeV are obtained for values of 0.04, 211 0.06 0.10, respectively, a signicant improvement and compared to previous results.
260
0.06 0.08 0.1
220
is LSP 1
180
140 Exclusion domains 100
obs. 121 obs. 122 obs. 133
0.12 0.14 0.16
LEP excluded
20 60 100
exp. 121 exp. 122 exp. 133
0.18 0.2
140
180
M 0 [GeV]
1
220
Figure 33: Exclusion contours for the three LLE couplings 121 , 122 , 133 in the m(0 )m( ) plane for 1 1 an MSSM scenario without mass unication of M1 and M2 .
Figure 35: Cross section upper limits for the slepton mass versus the neutralino mass obtained in the D search for R SUSY with a non-zero coupling. 211
THUPL07
D, 360 pb
-1
0.04
0.02
XXVI Physics in Collison, Bzios, Rio de Janeiro, 6-9 July 2006 u
3.2.4. Resonant Sneutrino Production
13
Here, a is produced with a Yukawa cou 311 pling which subsequently decays to an oppositely charged electron-muon pair via a non-zero 132 coupling (Fig. 36). If such a process existed, a peak in the electron-muon invariant mass would be seen. CDF searched for such electron-muon pairs in 344 pb1 of data, but has seen no indication for an enhancement in their mass distribution (Fig. 37). Therefore exclusion limits as a function of the mass and the two Yukawa couplings are derived as shown for example in Fig. 38.
0.16 0.14 0.12 0.1 132 : 0.050 0.040 0.06 0.04 0.02 0.05 0.1 0.15 0.2 0.25 0.3 2 Me (TeV/c ) 0.35 0.4 0.45 0.030 0.020 0.010
CDF Run 2 Preliminary, 344 pb
-1
311
0.08
Figure 38: Excluded regions in the plane versus the 311 mass of the , obtained by CDF.
Figure 36: Production and decay of a tau sneutrino involving two dierent R couplings.
branching fraction as a function of the stop mass, together with the theoretical expectation, is displayed in Fig. 39. A lower limit of 155 GeV for the stop mass has been obtained, assuming 100% branching fraction for the decay t1 b + .
CDF Run II Preliminary (322 pb )
CDF Run I, LEP (from 3rd Gen LeptoQuark Search)
-1
( pp t1 t1 ) Br 2( t1 b ) pb
NLO ( pp t1 t1 ) 2 + 2 PDF scale
~~
Br( t1 b ) = 100%
~
102
~
10
95% C.L. upper limit: Observed Expected ()
CDF Run 2 Preliminary, 344 pb
-1
Data
Events/ 5 GeV/c2
10
Mass = 200 GeV/c 132 = 0.05 311=0.16 Z tt Diboson
2
~~
e
1
1
-1
m > 155 GeV/c 2
100 110 120
t
130
140
2
150
160
170
Fakes
m~ GeV/c
10
10-2
50
100
150
200
2
250
300
350
Me (GeV/c )
Figure 39: Upper limits of cross section times branching fraction as a function of the stop mass obtained by CDF assuming the decay t1 b + . Also shown is the theoretical expectation.
Figure 37: Invariant mass distribution of the electron-muon pairs in the CDF search for resonant tau sneutrino production.
3.3. Search for MSSM Higgs
In models of electroweak symmetry breaking with two Higgs doublets, such as the MSSM, there are ve physical Higgs bosons: two neutral CP -even scalars, h and H, a neutral CP -odd state, A, and two charged states, H . Neutral Higgs bosons are generically denoted as . At tree level, the Higgs sector of the MSSM is fully specied by two parameters, generally chosen to be MA , the mass of the CP -odd Higgs boson, and tan . At large tan , the coupling of the neutral Higgs bosons to down-type quarks and charged leptons is strongly enhanced, leading to sizeable cross sections. In this region, the A boson is nearly degener-
3.2.5. Stop Pair Production
CDF has searched for pair production of stop quarks where both of them decay promptly into a b quark and a via the R-parity violating non-zero coupling, and one of the taus decays leptonically, 333 the other one into hadrons. In 322 pb1 of data, two events have been found with an isolated electron (or muon), a hadronic tau decay and two additional jets, whereas 2.26+0.46 SM background events are ex0.22 pected. The derived upper limit of cross section times THUPL07
14
XXVI Physics in Collison, Bzios, Rio de Janeiro, 6-9 July 2006 u ate in mass with the h or the H boson. The dominant decay modes are b ( 90%) and + b ( 8%), and the most promising search channels are considered to be either in association with b quarks, b( b b( or + . b) b b),
4. New Gauge Bosons
A possible way of resolving the inherent problems of the standard model is by extending the gauge sector of the theory. New heavy gauge bosons are predicted in many extensions of the standard model. For example, in little Higgs models, the quadratically divergent radiative corrections to the Higgs mass are canceled individually, leading to the appearance of partners of the W and Z bosons at the TeV scale. In grand unied theories heavy partners of the electroweak bosons generally appear; the left-right symmetric model is a SO(10) GUT extension of the SM, postulating the existence of a right-handed version of the weak interaction as well as an additional Z boson.Finally, the sequential standard model, where the couplings to quarks and leptons are as in the SM, may not be gauge invariant, but it serves as a good benchmark for comparisons of results. In the search for Z bosons, the latest results are from CDF analyzing dielectron resonances using 819 pb1 of data. The invariant mass spectrum of dielectron events used in this search is shown in Fig. 41. No signicant excess is seen at any mass value leading to a limit on the mass of a sequential, standard model-like Z of m(Z ) > 850 GeV at 95% C.L. CDF also searches for pair production of doubly charged Higgs bosons in the lepton avor violating modes H ++ e+ + and H ++ + + (and charge conjugates). At least three leptons are required in the nal states so that the backgrounds are very small, and no events are observed in approximately 350 pb1 of data. Interpreted in a left-right symmetric model, the corresponding mass limit on doubly charged Higgs bosons coupling to left-handed particles is m(H ++ ) > 114(112) GeV at 95% C.L. in the e ( ) channel.
Di-Electron Invariant Mass Spectrum
Nr Events / 5 GeV/c 2
105
4
CDF has published results of a search for neutral supersymmetric Higgs bosons in the decay to + [24], based on 310 pb1 of data, and D published [25] results both in the + channel as well as for the decay into b in association with b quarks, leading to b nal states with three or four b jets, using integrated luminosities of 325 pb1 and 260 pb1 , respectively. In the + channel, both experiments require one of the taus to decay leptonically, and the other one hadronically; in addition, D uses the e channel. The main discriminating variable in the search is the visible mass Mvis , calculated from the partially reconstructed tau decays. The dominant background is from Z/ + decays. Searching for a MSSM Higgs in the decay into b in b association with b quarks, the signal would show up in the dijet invariant mass distribution of the two leading jets. To suppress the dominant multijet background, D requires at least three jets with b quarks identied by a secondary vertex algorithm. In neither of the analyses a signicant access has been found, and exclusion limits are derived in the (MA , tan ) plane. All results are summarized for one of the benchmark scenarios [26], the no-mixing scenario, in Fig. 40. With increasing amounts of data, these analyses impose more and more stringent constraints on the MSSM Higgs sector.
tan
100 90 80 70 60 50 40 30 20 10 0 80
-1 D 260-325 pb -1 CDF 310 pb
no-mixing
D <0 D >0 D <0 D >0 CDF <0 CDF >0
CDF Run II Preliminary
Data Drell-Yan Jet Background EWK+ Background
10
103
2
10
L dt = 819 pb
-1
LEP 2 (>0)
100 120 140
LEP 2 (<0)
160 180 200
10
1 10-1 50 100 150 200 250 300 350 400 450 500
MA [GeV]
Figure 40: Excluded region in the (MA , tan ) plane for the no-mixing scenario [26] for = 0.2 TeV and = +0.2 TeV. The regions labeled D < 0 and D > 0 refer to the combination of the D + and multijet analyses.
Di-Electron Mass (GeV/c )
2
Figure 41: Dielectron invariant mass spectrum in the CDF search for electron-positron resonances.
In the search for singly charged gauge bosons, both D and CDF looked for a SM-like W decaying to an
THUPL07
XXVI Physics in Collison, Bzios, Rio de Janeiro, 6-9 July 2006 u electron and a neutrino. The best limit, obtained from a study of the transverse mass spectrum in 900 pb1 of data as shown in Fig. 42, is from D and requires m(W ) > 965 GeV at 95% C.L. (Fig. 43). of the order of the unication scale, in some models they can be relatively light. Experimentally, it is customary to consider one leptoquark per generation. These are assumed to be very short-lived and decay to a quark and a lepton. The branching fraction to a charged lepton and a quark is then denoted as . At hadron colliders, leptoquarks can be pair-produced through the strong interaction, or are singly produced. In the latter case the production cross section depends on the (unknown) quark-lepton coupling, which is generally taken to be of the same order of magnitude as the ne structure constant. Only the three most recent results from leptoquark searches at the Tevatron are discussed here. D has searched for scalar leptoquarks decaying to a quark and a neutrino ( = 0) in the jets plus missing tranverse energy topology in 310 pb1 of data [27]. Experimentally this is a dicult analysis which suers from substantial QCD dijet background due to mismeasured jets. To mitigate this, D requires exactly two acoplanar jets. The ensuing missing transverse energy distribution, before nal analysis cuts, is shown in Fig. 44. The background from QCD dijet events, dominant at low missing transverse energy, is extrapolated to higher values using two dierent tting functions as shown in the inset. The dominant non-QCD standard model background is Z boson plus jets production with the Z decaying to a pair of neutrinos. No excess is observed, so D sets a limit on the leptoquark mass of MLQ > 136 GeV at 95% C.L. (Fig. 45).
Events / 5 GeV 103 (a) 102
Data SM QCD LQ
15
Events / 2.00 [GeV]
106 10
5
Transverse Mass mT
D Run II Preliminary 900 pb-1
Signal m
W
= 500 GeV
We QCD (from Data) W * Z/ e e Z/ WW incl. tt incl. WZ incl. ZZ incl. Data
*
104 103 102 10 1
10-1 10-2
0
100
200
300
400
500
600
700
800
mT [GeV]
Figure 42: Transverse mass spectrum in the D search for W . Also shown is the distribution for a hypothetical W with a mass of 500 GeV.
95% CL Limit
W B(W e )
Theoretical Prediction
10
(incl. NNLO Corrections) obs. Limit 95% CL exp. Limit 95% CL
103 102 10
D L=310 pb
-1
1
Excluded (Run I)
10-1
mW > 965 GeV @ 95% CL
10
40 50 60
70 80 90 100
1
10
-2
D Run II Preliminary 900 pb-1
500 600 700 800 900 1000 1100 1200
50
100
150
200 250 Missing ET (GeV)
W mass [GeV]
Figure 43: Cross section upper limits for the production of a SM-like W obtained by D.
Figure 44: Missing transverse energy distribution in the D search for = 0 leptoquarks before nal cuts. The inset shows the two dierent tting functions used to evaluate the background from mismeasured QCD dijet events.
5. Leptoquarks
Leptoquarks are a natural consequence of the unication of quarks and leptons into a single multiplet, and as such are expected to be gauge bosons as well. While their masses may naturally be expected to be THUPL07
D searched for the pair production of scalar third generation leptoquarks decaying into a tau neutrino and a b quark, leading to a nal state of two b jets and ET , identical to the sbottom search described earlier. Both jets are required to be identied as b jets using a lifetime based algorithm. Further cuts on ET and HT are optimized depending on the leptoquark mass MLQ . For MLQ = 220 GeV, ET > 90 GeV and HT > 190 GeV, one event is selected in the data cor-
16
XXVI Physics in Collison, Bzios, Rio de Janeiro, 6-9 July 2006 u chosen.
D L=310 pb
-1
Cross section (1-) (pb)
2
102
Scalar LQ cross section Observed limit Expected limit
6. Large Extra Dimensions
Models postulating the existence of extra spatial dimensions have been proposed to solve the hierarchy problem posed by the large dierence between the Planck scale Mpl 1016 TeV, at which gravity is expected to become strong, and the scale of electroweak symmetry breaking, 1 TeV. In the original large extra dimensions model of Arkani-Hamed, Dimopoulos and Dvali [28], in which only gravitons propagate in the bulk but all standard model elds are conned to a 3-brane, a tower of Kaluza-Klein excitations of the graviton emerges. The graviton states are too close in mass to be distinguished individually, and the coupling remains small, but the number of accessible states is very large. It is therefore possible to produce gravitons which immediately disappear into bulk space, leading to an excess of events with a high transverse energy jet and large missing transverse energy, the monojet signature: q q gG, qg qG and gg gG, where G is the emitted graviton. The dominant standard model backgrounds are the production of Z or W bosons plus jets, with the Z decaying to a pair of neutrinos or the lepton from the W decay escaping detection. Using 1.1 fb1 of data, CDF has recently updated their published [29] analysis, and set limits on the eective Planck scale between MD > 1.33 TeV and MD > 0.88 TeV for a number of extra dimensions ranging from 2 to 6 (Fig. 47).
1.6
10
1 80
90
100
110
120
130 140 mLQ (GeV)
Figure 45: Cross section upper limits obtained in the search for = 0 scalar leptoquarks by D, together with the theoretical prediction.
responding to an integrated luminosity of 310 pb1 , while 2.6 0.6 events are expected from SM backgrounds. In the absence of an excess in the data, a limit of MLQ > 219 GeV at 95% C.L. is set, assuming a branching fraction B(LQ b) = 1; allowing also decays to t, the limit is reduced to MLQ > 213 GeV (Fig. 46).
B2, pb D0 Run II Preliminary
Signal cross-section, = 1M B2 [ b
PDF 2 () PDF LQ
2 () + 2 () ], B=1 2() + ], B=(1 - 0.5*Fsp)
1
B2 [ b
10-1
MD Lower Limit (TeV)
Observed , MHT,
310pb-1
Observed , MUJET, 367pb-1
213 GeV
1.4 1.2 1 0.8 0.6 2
CDFII (1.1 fb ) D0 (RunI) LEP Combined
CDF II Preliminary
-1
140
160
180
200
220 240 MLQ, GeV
Figure 46: Cross section upper limits by D obtained in the search for = 0 scalar leptoquarks of the third generation, together with the theoretical prediction.
CDF has released results on a search for vector leptoquarks of the third generation in the LQ3 b decay channel. The signature consists of a dijet plus ditau nal state, in which one tau is required to decay leptonically and the other hadronically. The main discriminating variables are the number of jets and an HT -like variable, the scalar sum of the transverse energies of all jets, leptons and ET . This allows CDF to set a limit at MLQ > 294 GeV assuming = 1 and using 322 pb1 of data. Note that this limit is higher than the typical limits on leptoquark masses at the Tevatron due to the model choice of vector leptoquarks, which have a much larger production cross section than the scalar leptoquarks which are usually THUPL07
Number of Extra Dimensions
3
4
5
6
Figure 47: CDF limits on the eective Planck scale MD as a function of the number of extra dimensions, compared to previous results.
In the model by Randall and Sundrum [30] gravity is located on a (3 + 1)-dimensional brane, the Planck brane, that is separated from the standard model brane in a fth dimension with warped metric. In the simplest version of this model gravitons are the only particles that can propagate in the extra dimension. The gravitons appear as towers of KaluzaKlein excitations with masses and widths determined by the parameters of the model. These parameters can
XXVI Physics in Collison, Bzios, Rio de Janeiro, 6-9 July 2006 u be expressed in terms of the mass of the rst excited mode of the graviton, M1 , and the dimensionless cou pling to the standard model elds, k 8/Mpl . If it is light enough, the rst excited graviton mode could be resonantly produced at the Tevatron. It is expected to decay to fermion-antifermion and to diboson pairs. Both CDF and D have recently presented new results in the search for Randall-Sundrum gravitons based on 0.8 1.2 fb1 (CDF) and 1.1 fb1 (D) of data, respectively. Both experiments analyze the e+ e and nal states. The invariant mass spectrum measured by D is shown in Fig. 48, and general agreement between data and the background expectation is observed. Using a sliding mass window, upper cross section limits are derived, which are then translated into lower mass limits for the lowest excited mode of Randall-Sundrum gravitons (Fig. 49). For a coupling parameter k 8/Mpl = 0.1 (0.01), masses M1 < 865 (240) GeV are excluded at the 95% C.L. by the D analysis. The corresponding CDF exclusion limits are M1 < 875 (242) GeV.
17
7. Excited Fermions and Other Resonances
In the search for excited le...