

 14 Apr 1995

EUROPEAN ORGANIZATION FOR NUCLEAR RESEARCH

CERN-PPE/95-46

April 5, 1995

Status of Higgs Hunting at LEP --

Five Years of Progress

Andr'e Sopczak PPE Division, CERN, CH-1211 Geneva 23

\Lambda

Abstract New results from general searches for the Higgs boson of the Minimal Standard Model (MSM), and for neutral and charged Higgs bosons of nonminimal Higgs models are reviewed from the four LEP experiments at CERN: ALEPH, DELPHI, L3, and OPAL. Much progress has been made due to the analysis of new data sets. A total of about 13 million hadronic Z decays are recorded from 1989 to 1994. The Higgs boson discovery potential for LEP2 is presented.

Presented at the IX Int. Workshop on High Energy Physics, Moscow 1994.

To be published in the proceedings.

\Lambda E-mail: andre@cernvm.cern.ch

Status of Higgs Hunting at LEP -- Five Years of Progress Andr'e Sopczaka

\Lambda

aPPE Division, CERN, CH-1211 Geneva 23

New results from general searches for the Higgs boson of the Minimal Standard Model (MSM), and for neutral and charged Higgs bosons of non-minimal Higgs models are reviewed from the four LEP experiments at CERN: ALEPH, DELPHI, L3, and OPAL. Much progress has been made due to the analysis of new data sets. A total of about 13 million hadronic Z decays are recorded from 1989 to 1994. The Higgs boson discovery potential for LEP2 is presented.

1. Introduction

On August 14, 1989, the first Z boson has been registered at the Large Electron Positron collider (LEP). During fall 1994 up to 30,000 hadronic Z bosons are produced per day and experiment corresponding to about twice the design luminosity. The integrated luminosity delivered to each LEP experiment is shown in Fig 1. The large data set allows to pursue one of the most challenging quests of experimental particle physics: the search for Higgs particles [1]. The experimental evidence of Higgs bosons would be crucial to understand the mechanisms of the SU(2) \Theta U(1) symmetry breaking and the mass generation in gauge theories.

In 1995, almost a doubling of data is anticipated after the successful tests of running LEP with 4x4 bunches. In 1996, LEP2 will operate with a center-of-mass energy above the W+W

\Gamma

threshold. In addition to the larger kinematic reach, LEP2 will also result in an improved signal to background ratio for the Higgs boson search.

The Higgs mass is a free parameter in the MSM [2]. Current precision measurements of the Z-lineshape do not reveal a favored Higgs mass range, as illustrated in Fig. 2 (from [3]). The theoretical framework is reviewed, for example, in [4]. This paper reviews the search for the MSM Higgs (Sec. 2), and the search for non-minimal Higgs bosons (Sec. 3). Interpretations are summarized in the two-doublet Higgs model (Sec. 4) and in the Minimal Supersymmetric Standard Model (MSSM) [5] (Sec. 5). The physics potential of LEP2 is addressed (Sec. 6). This report updates [6].

\Lambda e-mail: andre@cernvm.cern.ch

LEPOPERATION 1990 - 1994

De live

red In teg rat ed Lu

mi no siti es (p

b-1

)

0 10 20 30 40 50 60

13 37 61 85 109 133 157 181 205

1990

7.6 pb-1

1991 17.3 pb-1

28.6 pb-1

40 pb-1

64.45 pb-1

1993 1994

1992

Number of Operation days 95 086/lb Figure 1. Integrated luminosities seen by each LEP experiment.

Figure 2. Comparison of Z-lineshape measurements with top and Higgs mass variations in the MSM.

1

2. MSM Higgs Search

The expected Higgs boson event rate [7] for the bremsstrahlung process [8] is known to better than 1% including radiative corrections [9]. The expected number of Higgs boson events per 1 million hadronic Z decays is shown in Fig. 3.

\Gamma \Gamma e\Gamma @@e +

Z0 \Gamma \Gamma

Z0

?

@@ H0

Figure 3. MSM Higgs production rate as a function of the Higgs boson mass.

The Higgs decay mode determines the Higgs signature in the detectors. Higgs bosons with low masses decay into e+e

\Gamma and _+_\Gamma pairs, for intermediate masses they decay into light hadrons and o/ +o/

\Gamma pairs, and for high masses they decay

predominantly into a bb quark. The possible decay modes are shown in Fig. 4 (from [10]).

H mass (GeV) \Gamma (GeV)

Figure 4. MSM Higgs decay branching ratios.

2.1. Very Low-Mass Higgs Bosons

For mH ! 2m_ the Higgs boson has a decay length such that it does not decay at the primary interaction point. Two signatures can be distinguished, a) the Higgs decays outside the detector, and b) the Higgs decays inside the detector material, leaving a `V' signature. Figure 5 (from [10]) shows the decay length.

10 10 10 10

10 10 10

0 -3

-6 -9

-2 0 2

10-4

10-7 10-10 10-13

M (GeV)H c (cm

)

t \Gamma (G

eV)

y1086Yanis

a) b) Figure 5. MSM Higgs decay length.

Searches for these signatures have been performed by all LEP experiments, and no indication of a signal has been observed. An example of the number of expected Higgs events is given in Fig. 6 (from[11]).

Figure 6. DELPHI: Number of expected Higgs events in the very low-mass region.

2

2.2. Low-Mass Higgs Bosons

Various different final states are expected as illustrated in Fig. 7. No indication of a Higgs signal in any channel has been found, and the mass region below 4 GeV is excluded at 99% CL [12-15].

Figure 7. Diagrammatic view of a low-mass Higgs signal.

2.3. Intermediate-Mass Higgs Bosons

Mono-jets are expected in this mass region between about 4 and 15 GeV. Such mono-jets, as illustrated in Fig. 8, have not been observed and the mass region is excluded at 99% CL [12-15].

Figure 8. Diagrammatic view of an intermediatemass Higgs signal.

2.4. High-Mass Higgs Bosons

In this mass region the muon, electron, and neutrino channels are most important due to their distinct signatures (Z0 ! Z0?H0 ! qqH0 is not used due to large QCD background). Typical Higgs signatures are illustrated in Fig. 9.

Figure 9. Diagrammatic view of a high-mass Higgs signal.

Figure 10 (from[15]) shows a Z0?H0 ! _+_

\Gamma qq

candidate event which has passed all of the selection criteria, and Fig. 11 (from[14]) shows a Z0?H0 ! e+e

\Gamma qq candidate.

Figure 10. OPAL:Higgs candidate mH=61.2GeV.



L3



Hadron

Calorimeter

BGO

TEC



Jet 2

Jet 1



e

-

e

+

Event Nr. 4701

Run Nr. 390102

Figure 11. L3: Higgs candidate mH = 67:6 GeV shown in the plane perpendicular to the beam line.

3

Table 1 lists the Higgs candidates [12-15] with mH ? 30 GeV. The most precise measurement of the mass corresponding to the Higgs mass is calculated from the e+e

\Gamma and _+_\Gamma pairs (recoiling

mass).

Table 1 MSM Higgs Candidates.

Experim. Event Type Year Mass(GeV) ALEPH _+_

\Gamma qq 93 51:4 \Sigma 0:5

_+_

\Gamma qq 94 49:7 \Sigma 0:5

OPAL _+_

\Gamma qq 93 61:2 \Sigma 1:0

L3 e+e

\Gamma qq 91 31:4 \Sigma 1:5

e+e

\Gamma qq 92 67:6 \Sigma 0:7

_+_

\Gamma qq 91 70:4 \Sigma 0:7

_+_

\Gamma qq 93 74:0 \Sigma 0:7

DELPHI e+e

\Gamma qq 91 35:9 \Sigma 5:0

_+_

\Gamma qq 93 75:0 \Sigma 0:7

The origin of the candidate events is well understood. They are a result of 4-fermion background. Their production graphs are shown in Fig. 12. Annihilation (a) and conversion (d) processes are most important after all Higgs boson selection cuts are applied.

e+ eg

a)

ee+

q q-

Annihilation

Z, g

e- ee+ e+

q

q-

g

g

b) Multiperipheralfl

(2-photon)

q q-g e- ee+ e+

g

c) Bremsstrahlung

e+

ee+ eq

q-

g Z, g Z,

d) Conversion

y1096Sopczak Figure 12. Feynman graphs of 4-fermion background reactions.

The spectrum of the recoiling mass, corresponding to the Higgs mass, is shown in Fig. 13 (from [12]) before a cut on this variable is applied. Two out of the three events of Fig. 13 (in the mass region above 50 GeV) are rejected since the jets are not likely to be b-flavored as expected from a Higgs decay. The simulated 4-fermion spectrum is in full accordance with the data. About nine 4- fermion events with recoiling mass ? 50 GeV are expected from all four LEP experiments, while six events have been observed. It is remarkable that about two * _*qq background events are expected while none has been observed. Table 2 summarizes the Higgs mass limits in the MSM [12-15].

1059 jpm rd Cut

Monte Carlo Aleph Data(Inc. 1993)

0 4 8 12

0-20 20 40 60 80

Recoil Mass (GeV/c )2

Eve nts pe r 10

Ge V/c 2

Figure 13. ALEPH: 4-fermion simulation and data.

Table 2 MSM Higgs boson mass limits from ALEPH, DELPHI, L3, and OPAL

A D L3 O Prel. Prel. Data Sample 89-94 90-92 90-94 90-93 Z0 ! q_q \Theta 106 3.6 1.6 3.1 1.9 Mass Limit 95%CL (GeV) 62.9 58.3 60.1 56.9

4

The number of expected Higgs events are shown in Fig. 14 (from[14]) for the _+_

\Gamma , e+e\Gamma

and * _* channels. The limit is set using Poisson statistics. In a mass region without a Higgs candidate the 95% CL limit is set where the sum of the expected events is 3.

0.1

1 10

All channels H0nn H0e+e-+H0u+u95 % C.L. Line 60.1 GeV

Expected Events

MH (GeV)

PRELIMINARY L3 10-1

1 10

50 52.5 55 57.5 60 62.5 65 67.5 70

Figure 14. Prelim. L3: Expected events in the _+_

\Gamma ,

e+e

\Gamma ,

and * _* channels, and the 95% CL line.

2.5. Combined Limit and Prospects

The number of expected events is given by each LEP experiment [12-15], and shown in Fig. 15 for combined data corresponding to a total of 10.2 million hadronic Z decays. In good approximation, a combined Higgs mass limit can be set by the summation of the number of expected Higgs events. The calculation of the 95% CL limit takes the background events into account and corrects for up to 25% reduction due to tighter selection cuts with increasing statistics. Owing to the new results, the combined Higgs mass limit is significantly increased compared to the value reported a year ago (63.5 GeV [6]). The combined mass limit is 65.1 GeV. Figure 15 shows that with larger statistics the reduction of 4-fermion background will be crucial to increase the sensitivity mass range. This can be achieved with enhanced microvertex b-quark tagging.

The evolution of the published Higgs mass limits is shown in Fig. 16. The sensitivity can be extrapolated assuming 50% efficiency in the _+_

\Gamma ,

e+e

\Gamma and * _* channels. With about 20 million

hadronic Z decays a sensitivity of 65 to 70 GeV could be obtained, depending on additional candidate events. The combined LEP limit also lies below the extrapolated line, since all experiments have tuned the events selection on their own maximal visible Higgs mass. One should note that combined mass limits vary by about 1 GeV comparing other statistical methods [16-18].

Higgs mass (GeV) Number of expected Higgs eventsALEPHL3

OPAL DELPHI

sumcorr. sum

65.1 GeV65.1 GeV 3

10-1

1 10 10 2 10 3

0 10 20 30 40 50 60 70 Figure 15. Combined MSM Higgs mass limit from results of ALEPH, DELPHI, L3 and OPAL.

Number of hadronic Z decays (x1000) Higgs mass limit (GeV)

LEP ALEPH L3 OPAL DELPHI 10

20 30 40 50 60 70

10 10 2 10 3 10 4 Figure 16. Higgs mass limits and extrapolation of sensitivity.

5

3. Non-Minimal Higgs Boson Search

There are three classes of searches for non-minimal Higgs bosons: a) searches for Higgs bremsstrahlung with reduced production rates compared to the MSM prediction, b) neutral Higgs pair-production, and c) charged Higgs pair-production. The production graphs are shown in Fig. 17.

Z0 \Gamma \Gamma \Gamma

Z0

\Lambda

@@

@ h0

a)

Z0 \Gamma \Gamma \Gamma

h0

@@

@ A0

b)

Z0 \Gamma \Gamma \Gamma

H+

@@

@ H

\Gamma

c)

Figure 17. Non-minimal Higgs production. 3.1. Invisible Higgs Boson Search

Supersymmetric models with broken R-parity or possible h0 ! O/0O/0 decays where O/0 is the lightest Supersymmetric particle predict invisible Higgs decays. Such invisible Higgs bosons can be searched in bremsstrahlung production (Fig. 17a) in analogy to the MSM Higgs boson. The e+e

\Gamma ,

_+_

\Gamma and qq channels are important. Larger

sensitivities are expected compared to the MSM search, since the Z0 ! qq channel gives a clean signature for the invisible Higgs while it could not be used in the MSM due to large QCD background. One invisible 60 GeV candidate event, shown in Fig. 18 (from [19]), is compatible with the expected rate from the 4-fermion background.

Run=15238 Event=4802ALEPH

ECAL

HCAL

HCAL ECAL TPC

TPC ITC

Figure 2

Figure 18. ALEPH: Invisible Higgs candidate in the e+e

\Gamma

channel.

3.2. Z0 ! h0A0 ! b_bb_b Search

In this channel 4-jet events are expected. A good hadronic mass resolution would allow reconstruction of both Higgs masses, as shown in Fig. 19 (from [20]) and thus the combinatorial background from Z0 ! hadrons (Fig. 20) can be

reduced. Many 4-jet events pass the event-shape and invariant mass selection cuts. A further event selection is based on the fact that Higgs events produce b-flavored jets. These jets can be selected by semileptonic b-decays, as shown in Fig. 21 (from [21]). A more efficient method uses the fact that B mesons are formed. B mesons have a long lifetime (o/B = 1:5 ps) which gives a larger number of detectable secondary vertices. All LEP experiments are equipped with microvertex detectors. These detectors allow the tagging of b-flavored jets with secondary vertices. Figure 22 (from [22]) shows a bbbb candidate.

All mass combinations up to the kinematic production threshold are scanned. Typically, about 20 data events remain which is in agreement with the QCD background expectations. The limits on \Gamma (Z0 ! h0A0)=\Gamma (Z0 ! q_q) vary with mh and mA. These limits are of the order of 10

\Gamma 3 to

10

\Gamma 4 [19, 22, 20, 23]. An example of branching

ratio limits (L3 preliminary) is given in Fig. 25.

Figure 19. L3: a) Simulated Higgs masses; b) Mass-O/2 for data, qq and signal simulations.

6

+fl gfl gfl

gflqfl qfl

gfl gfl

gfl

qfl

qfl e e o"" g gfl-fl e e o"" fl+fl -fl

_flqfl qfl qfl_fl _fl

qfl_flqfl

qfl_fl qfl_flqfl qfl_flqfl

Figure 20. QCD Feynman graphs leading to 4-jet events.

0 10 20 30p (GeV)0 800 1600 2400

Events / GeV

uData MC: u,d,s,c,bMC: b o"" c o"" u

MC: b o"" u

0 10 20 30p (GeV)0 800 1600 2400

Events / GeV

0 1 2 3 4 5p

t (GeV)

0

800 1600 2400

Events / 0.25 GeV

uData MC: u,d,s,c,bMC: b o"" c o"" u

MC: b o"" u

0 1 2 3 4 5p

t (GeV)

0

800 1600 2400

Events / 0.25 GeV

0 10 20 30p (GeV)0 400 800 1200

Events / GeV

eData MC: u,d,s,c,bMC: b o"" c o"" e

MC: b o"" e

0 10 20 30p (GeV)0 400 800 1200

Events / GeV

0 1 2 3 4 5p

t (GeV)

0

400 800

Events / 0.25 GeV

eData MC: u,d,s,c,bMC: b o"" c o"" e

MC: b o"" e

0 1 2 3 4 5p

t (GeV)

0

400 800

Events / 0.25 GeV Figure 21. L3: b-jet tagging with semileptonic bdecays. Good b-jet purity is achieved for events with large transverse lepton momentum.

Run 30426 event 7698 Delphi Vertex Detector 26/Apr/92 01:06

0.0 cm 0.5 cm

Figure 22. DELPHI: b-jet tagging using a microvertex detector. Central beam position and secondary vertices are marked.

3.3. Z0 ! h0A0 ! o/ +o/

\Gamma b_b Search

In this channel a o/ -pair recoiling to a jet system is expected. The invariant mass of the o/ -pair can be reconstructed using kinematic constraints. Figure 23 (from [20]) shows a simulated Higgs signal in comparison with data and background simulation. Branching ratio limits (L3 preliminary) are given in Fig. 25.

Figure 23. L3: Example of o/ o/ bb selection. 3.4. Z0 ! h0A0 ! o/ +o/

\Gamma o/ +o/ \Gamma Search

The four o/ signature has been searched for and no signal has been observed. The most important background originates from Z0 ! o/ +o/

\Gamma events.

This background can be largely suppressed by requiring exactly two tracks in one hemisphere as expected from one-prong h0 ! o/ +o/

\Gamma decays, as

shown in Fig. 24 (from [20]).

Figure 24. L3: Example of o/ o/ o/ o/ selection. 3.5. h0 ! A0A0 Search

The h0 ! A0A0 decay can be dominant if kinematically allowed. No indication of a Higgs has been observed and limits are set, for example, on six o/ 's or six b's of about 10

\Gamma 3 [19, 22, 20, 23].

7

Figure 25. Preliminary L3: 95% CL limits on \Gamma (Z0 ! h0A0)=\Gamma (Z0 ! q_q) as function of mh and mA. 3.6. Z0 ! H+H

\Gamma ! c_s_cs Search

Compared to the bbbb channel very similar signatures are expected. Furthermore, harder kinematic constraints can be applied, and a charged Higgs mass resolution better than 1 GeV is expected. However, as a consequence that no btagging can be applied in the cscs channel, more irreducible background events remain as shown in Fig. 26 (from [20]).

3.7. Z0 ! H+H

\Gamma ! cso/ * Search

Event shape selection cuts and the requirement of an isolated o/ lead to a good background rejection. After all selection cuts, a Higgs signal would be clearly visible in the reconstructed jet-jet invariant mass distribution, as shown in Fig. 27 (from [20]).

8

Figure 26. L3: Data, simulated background and 42 GeV cscs Higgs boson signal.

Figure 27. L3: Data, simulated background, 30 and 44 GeV cso/ * Higgs boson signal.

3.8. Z0 ! H+H

\Gamma ! o/ +*o/ \Gamma _* Search

Figure 28 (from [20]) shows a good separation of simulated Higgs signal and Z0 ! o/ +o/

\Gamma background.

Figure 28. L3: Data, simulated background and 44 GeV o/ *o/ * Higgs boson signal.

4. Interpretation in the 2-Doublet Model

Production rates for Higgs boson bremsstrahlung and neutral Higgs boson pair-production are complementary. Therefore, a Higgs boson cannot escape detection if it is kinematically accessible. The search for Higgs bremsstrahlung in the MSM Higgs decay channels with reduced production rates is particularly important. The experimental results set limits on the parameters of the general two-doublet Higgs model.

4.1. Non-Minimal Neutral Higgs Bosons

The combined LEP limit from Higgs boson bremsstrahlung searches of Fig. 15 can be interpreted as a limit on the parameter sin2(fi \Gamma ff) of the two-doublet Higgs model, shown in Fig. 29.

Figure 29. L3. Limit on sin2(fi \Gamma ff).

The value cos2(fi \Gamma ff) can be constrained by precision Z-lineshape [24] measurements owing to the large production rate of Z0 ! h0A0. Any non-minimal MSM contribution to the Z-width larger than 23 MeV is excluded at 95% CL [25]. This value results from a comparison of the measurement and theoretical prediction taking into account the dominating uncertainties in the top quark mass, the MSM Higgs boson mass and the strong coupling constant. The production rate of Z0 ! h0A0 depends on the Higgs boson masses and the cos2(fi \Gamma ff) value. A limit on cos2(fi \Gamma ff) is shown in Fig. 30 (from [25]). As a consequence, the combination of sine and cosine limits excludes a large region in the (mh,mA) parameter space, as presented in Fig. 31.

9

Figure 30. Limit on cos2(fi \Gamma ff). Figure 31. L3: Limits on (mh,mA). 4.2. Non-Minimal Charged Higgs Bosons

In the two-doublet Higgs model the charged Higgs boson production rate is only a function of the charged Higgs mass [26]. The number of expected events is shown in Fig. 32 for 1 million hadronic Z decays.

Figure 33 (from [22, 20]) shows two recent results of 95% CL mass limits on charged Higgs bosons as a function of its hadronic (leptonic) branching ratio obtained from the search in the cscs, cso/ *, and o/ *o/ * decay channels. Results with lower statistics are reported in [27, 28].

Figure 32. Charged Higgs production rates.

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9

1

28 30 32 34 36 38 40 42 44 46

Charged Higgs mass GeV/c2

Br(H +5

hadrons)

excluded

by

DELPHI

(b)

(a)

Figure 33. DELPHI and preliminary L3: Charged Higgs mass limits from direct searches. Upper plot: (a) cscs and (b) o/ *o/ *-analyses.

10

5. Interpretation in the MSSM

The MSSM [5] Higgs boson production rates and decay branching ratios are functions of the Higgs boson masses. When the important radiative corrections to the tree-level calculations are included, the production rates and decay branching fractions will also depend on a large number of unknown parameters of the Supersymmetric model. The effect of radiative corrections is illustrated in Fig. 34 (from [29]). The regions are shown where more than 250 Z0 ! h0A0 events per 1 million hadronic Z decays are expected for a) no radiative corrections, up to d) large radiative corrections (mt = 200 GeV and m~t = 1 TeV). Compared to the tree-level calculations (Fig. 34a), the (mh,mA) parameter space is largely extended.

0 20 40 60 20

40

60

h m (G eV)

A

y94224Sopczak m (GeV)

20 40 60

20

40

60

a) b)

c)

d)

Figure 34. Regions with large Z0 ! h0A0 production depending on the amount of radiative corrections in the MSSM.

So far all LEP experiments have interpreted their results as a function of top and stop masses only. Figure 35 (from [19, 22, 20, 23]) shows the MSSM results of the four LEP experiments for independent variation over top and stop masses (except DELPHI, which has fixed top and stop masses). An analysis with larger theoretical precision [30] has revealed a new unexcluded mass region as shown in Fig 36 (from [30]) marked with

thick contour lines. This plot can directly be compared with the L3 result of Fig. 35. The effects of Supersymmetric particles on Higgs boson cross sections and branching ratios are significant. A detailed discussion is given in these proceedings.

DELPHIflALEPHfl

L3fl OPALfl 0fl 20fl 40fl 60fl 80fl 20fl 40fl 60fl 80fl0fl

20fl 40fl 60fl 80fl

20fl 40fl 60fl 80fl m (GeV)flhfl

m (G eV)

fl

Afl

Excludedfl Allowedfl Not allowed in the MSSMfl

y1095Sopczakfl

Figure 35. ALEPH, DELPHI, L3, OPAL: MSSM results. The dark region is excluded, the hatched region allowed, and the light region not allowed by the theory.

Figure 36. MSSM results with full one-loop radiative corrections.

11

6. Prospects of Higgs Searches at LEP2

The physics potential for minimal and nonminimal Higgs searches has been studied for center-of-mass energies of 175, 190, and 210 GeV. According to the planning for LEP2 [31], it will be possible to obtain a center-of-mass energy of about 175 GeV. This corresponds to the installation of 196 approved cavities with an acceleration gradient of 6 MV/m. At a later stage, the installation of 256 cavities would increase the centerof-mass energy to 190 GeV. The installation of 384 cavities would further increase the energy to 210 GeV. The ultimate energy limit of the LEP programme, of about 240 GeV, which is set by the maximal bending power of the magnets, could eventually be achieved with additional or better performing cavities. The aim is to reach the W+W

\Gamma threshold in 1996 [32].

The method of search developed at LEP1 will be fully applicable at LEP2; in addition, new techniques for b-tagging and invariant jet mass reconstructions will be important to cope with the e+e

\Gamma ! W+W\Gamma background production. Figure 37 shows diagrams of background reactions and their expected cross sections forp

s = 190 GeV. All processes have been simulated with PYTHIA [33], except e+e

\Gamma f+f\Gamma which

has been simulated with DIAG36 [34]. A fast, but realistic, detector simulation has been performed. Details of the simulations and the event selections are given in [35] for ps = 175, 190, and 210 GeV2.

e+ g

e-

e

Z/ *

gZ/ *

e+

e-

n

W+

W-

e+ e-

* Z e

g e+

e-

f

f_ gZ/ *

e+ e-

n n

ZWW

e+

e- n

W

g W-e

+ e+

e-

e- Z

e+g g

g

f

f_

e+ e+

e- e

-

0.01 0.1 1 10 1000.015 .9

2 1.1 3.3 18 25 48 137 (pb)

M >30 GeVf f

Figure 37. LEP2 background reactions and cross sections for ps = 190 GeV. 2Consistent with the energies of the Higgs and New Particle LEP2 working group, which studies more details.

Unlike at LEP1 the 4-jet channel (H0Z0 ! bbqq) can also be used in the MSM Higgs search at LEP2 due to the much-suppressed background from hadronic Z decays. New sources of 4-jet background will arise from W+W

\Gamma ! qqqq and

Z0Z0 ! qqqq.

In addition, W+W

\Gamma decays will lead to the

same final states as expected from charged Higgs decays. Branching ratios are listed in Table 3.

Table 3 WW and ZZ decay branching ratios.

WW Decay BR(%) ZZ Decay BR(%) o/ +*o/

\Gamma _* 1.1 bbbb 2.3

qq o/ * 14 ccbb 4.0 qqqq 47 cccc 1.7

qqqq 49

6.1. MSM Higgs Boson

All LEP experiments can obtain approximately the same sensitivity for the MSM Higgs boson. A selection sensitivity (minimum theoretically predicted cross section to observe a signal) of about 0.05 to 0:15 pb with L = 500 pb

\Gamma 1 for

a 3oe effect of signal to background ratio (oe = signal=pbackground) can be achieved over the Higgs mass range from 70 to 120 GeV depending onp

s [36]. The sensitivity in the mass range between 90 and 110 GeV is slightly weaker due to the irreducible background from e+e

\Gamma ! Z0Z0 events.

12

More generally, any bremsstrahlung-produced Higgs boson in a non-minimal SM with decay branching ratios similar to those expected for the MSM Higgs boson would be discovered if

oe(e+e

\Gamma ! h0Z0) * 0:2 pb: (1)

In the MSM, the expected Higgs boson cross section is well known as a function of its mass, and its discovery limit at LEP2 can be expressed in good approximation as a function of ps:

mlimitHMSM = ps \Gamma mZ (\Sigma 5 GeV); (2) where the positive sign is valid for a center-ofmass energy near the W+W

\Gamma threshold and the

negative sign for a center-of-mass energy around 210 GeV. The cross section for the MSM Higgs boson as a function of the center-of-mass energy is shown in Fig. 38 (from [35]), see also [37].

The minimum luminosity needed for a Higgs boson discovery as a function of the Higgs boson mass is shown in Fig. 39 (from [38]) for ps = 175 GeV. A Higgs boson with a mass of about 83 GeV would be detectable with a 5oe effect forL

= 500 pb

\Gamma 1.

center-of-mass energy (GeV) s(e +e-

5 HZ) (pb)

H 60GeV

80GeV

100GeV

120GeVsensitivity 0.2

0.4 0.6 0.8

1 1.2 1.4

80 100 120 140 160 180 200 220 240 260 Figure 38. MSM Higgs cross sections and experimental sensitivity.

Figure 40 (from [35]) shows the L3 limit [20] and an extension of the sin2(fi \Gamma ff) sensitivity range as a function of the Higgs mass at ps = 210 GeV with a detection sensitivity of 0.2 pb. A sufficient overlap with the current limits is achieved when a selection sensitivity of 0:2 pb can be maintained also for a 30 GeV Higgs boson. A similar extension has been reported in Ref. [39].

50 60 70 80 90 1000 100 200 300 400 500 600

O"s = 175 GeV

Discovery limit

mH (GeV/c2)

Lm

in ( pb -1)

y94235 Figure 39. Minimum luminosity needed to discover the MSM Higgs bosons with a 5oe effect.

Figure 40. Limit on sin2(fi \Gamma ff) from L3 and sensitivity extension for LEP2 with ps = 210 GeV and L = 500 pb

\Gamma 1.

6.2. Non-Minimal Neutral Higgs Bosons

Already in the first phase of LEP2, a significant increase of the experimentally accessible mass parameter space compared to LEP1 for a discovery of non-minimal Higgs bosons will be possible as shown in Fig. 41. Figure 42 (from [35]) illustrates the effect of b-jet tagging in the e+e

\Gamma ! h0A0 ! b_bb_b search. The simulated effects of b-tagging on signal efficiency and background rejection are listed in Table 4 (from [35]) for an example of mh = 60 GeV and mA = 100 GeV at ps = 210 GeV, applying a simple b-tagging algorithm [40, 41].

Table 4 b-tagging efficiency and background rejection.

Eff. (in %) Rejection Power (in %)

bbbb qq flZ0 W+W

\Gamma Z0Z0

60 31 36 105 11

13

h mass (GeV) A mass (GeV)

LEP1

175GeV 190GeV

210GeV

kinematic limits LEP2

Z5hA Searches

0 20 40 60 80 100 120 140 160 180

0 20 40 60 80 100 120 140 160 180 Figure 41. Kinematically accessible regions for Higgs boson pair-production at LEP1 and LEP2.

reconstructed invariant mass (GeV) events

ZZ

qq WW MC Signal

WITHOUT B-TAGGING

reconstructed invariant mass (GeV) events WITH B-TAGGING

0 20 40 60 80 100 120 140 160

0 20 40 60 80 100 120 140 160 180 200

0 5 10 15 20 25

0 20 40 60 80 100 120 140 160 180 200 Figure 42. Simulated Higgs bosons and background: b-tagging is required for signal sensitivity (ps = 210 GeV, L = 500 pb

\Gamma 1).

Higgs boson mass resolutions of about 10% and a 3oe detection sensitivity of 0.12 pb have been obtained in this simulation. The sensitivities vary strongly as a function of (mh,mA) [35]. Based on the experience acquired at LEP1, larger sensitivities are expected for the o/ o/ bb and o/ o/ o/ o/ channels.

6.3. On a Decisive Test of the MSSM

The upper mass on mh is shown in Fig. 43

(from [42]) as a function of various parameters of the Supersymmetric model and for two top masses. For a top mass of 180 GeV 3 the upper bound on mh is 137 GeV.

Owing to the complementary character of Higgs bremsstrahlung and Higgs pair-production, a decisive test of the MSSM will require simultaneous searches. The parameter regions in which a Higgs signal can be discovered for ps = 210 GeV and L = 500 pb

\Gamma 1 are shown in Fig. 44

(from [30]). Four regions can be distinguished for mt = 175 GeV and tan fi * 0:5:

(A) The sensitivity region.

(B) The region, where sensitivity depends on

the choice of Supersymmetric parameters. (C) The non-sensitivity region. (D) The region not allowed in the MSSM. A substantial region (B) reflects a dependence of the discovery potential on the choice of Supersymmetric parameters.

6.4. Non-Minimal Charged Higgs Bosons

A discovery of a charged Higgs boson would

be unambiguous evidence of physics beyond the MSM, and even beyond the MSSM if mH\Sigma ! mZ. The charged Higgs production rate [26] for L = 500 pb

\Gamma 1 is illustrated in Fig. 45 (from [35]) forp

s = 175, 190 and 210 GeV.

In the cscs channel a mass resolution of about 1 GeV can be obtained, as shown in Fig. 46 (from [35]). In addition to the selection for a cso/ * signal at LEP1, the reconstruction of the invariant mass of the o/ * system can also be used to discriminate against W+W

\Gamma ! c_so/ \Gamma * background, as shown in Fig. 47 (from [35]). In the o/ *o/ * channel, leptonic W+W

\Gamma decays can largely

be rejected by the reconstruction of the visible o/ energies, as shown in Fig. 48 (from [35]).

3Recently the CDF Collaboration reported evidence for the top quark with mt = 174 \Sigma 10+13\Gamma 12 [43].

14

Figure 43. Upper h0 mass bounds for various values of Supersymmetric parameters in the MSSM.

Figure 44. Accessible MSSM (mh,mA) regions.

H- mass (GeV) events 175 GeV190 GeV

210 GeV

0 50 100 150 200 250 300 350 400 450 500

0 20 40 60 80 100 Figure 45. Number of charged Higgs bosons expected for ps = 175; 190, and 210 GeV.

Figure 49 (from [35]) shows the combined reach of the three search channels. A signal would be visible up to mH\Sigma ss 70 GeV. A large total luminosity is crucial for a significant extension of the charged Higgs boson discovery potential beyond the LEP1 limit due to the small variation of the event rate with the center-of-mass energy and the rather small number of expected events at LEP2.

15

reconstructed invariant mass (GeV) events

reconstructed invariant mass (GeV) events

Mean=60.1 GeV Sigma=0.96 GeV

Mean=70.0 GeV

Sigma=1.0 GeV

reconstructed invariant mass (GeV)

events

Mean=60.1 GeV Sigma=0.96 GeV

Mean=70.0 GeV

Sigma=1.0 GeV

reconstructed invariant mass (GeV)

events

Mean=60.1 GeV Sigma=0.96 GeV

Mean=70.0 GeV

Sigma=1.0 GeV

reconstructed invariant mass (GeV)

events

Mean=60.1 GeV Sigma=0.96 GeV

Mean=70.0 GeV

Sigma=1.0 GeV

0 5 10 15 20 25

30 40 50 60 70 80 90 100 110 120 130 Figure 46. Simulated 60 and 70 GeV charged Higgs bosons and background in the cscs channel.

reconstructed inv. mass tn (GeV) reconstructed inv. mass cs (GeV)

Cut

Cut

MC 60 GeV H+HWW Background

0 20 40 60 80 100 120

0 20 40 60 80 100 120 Figure 47. Reconstruction of Mcs and Mo/* .

visible energy t1 (GeV) visible energy t2 (GeV)

MC 70 GeV H+HWW BackgroundCut

0 5 10 15 20 25 30 35 40

0 10 20 30 40 50 60 Figure 48. Final selection in the o/ *o/ * channel.

H- mass (GeV) Br(H - o""tn

)

100pb-1

200pb-1

500pb-1 0

0.2 0.4 0.6 0.8

1

40 50 60 70 80 Figure 49. Sensitivity regions for ps ss 200 GeV and L = 100, 200, and 500 pb

\Gamma 1.

16

7. Conclusions

The search for the Higgs boson of the MSM has exceeded expectations. The pre-LEP expectations for the sensitivity range of LEP1 were about 30 GeV [4], while a Higgs mass larger than 60 GeV has already been excluded by individual LEP experiments. The combined limit from the four LEP experiments on the MSM Higgs boson mass is 65:1 GeV at 95% CL. At LEP1, the MSM Higgs boson sensitivity approaches its saturation. A sensitivity increase up to about 70 GeV can be expected with 20 million hadronic Z decays and stronger rejection of 4-fermion events.

In the two-doublet Higgs model, searches for neutral and charged Higgs bosons lead to various limits on their production rates. Charged Higgs bosons are excluded independently of the decay mode up to the kinematic reach of LEP1 of about 45 GeV. Mass limits and limits on sin2(fi \Gamma ff) and cos2(fi \Gamma ff) are obtained. Additional LEP1 data will be important to establish higher sensitivities for Higgs bremsstrahlung production and neutral Higgs pair-production, since both production rates are unpredicted. In the MSSM, LEP1 has almost covered the kinematically accessible parameter mass region and excluded it.

The prospects of the Higgs search at LEP2 will predominantly depend on the achievable centerof-mass energy and the total integrated luminosity. The MSM Higgs boson reach will be from 80 to 110 GeV for about ps = 170 to 210 GeV. Already in the first phase of LEP2, a significant increase of the mass parameter space compared to LEP1 for a discovery of non-minimal Higgs bosons will be possible, while the mass range for a discovery of the MSM Higgs boson will increase only by 10 to 15 GeV. With a large center-ofmass energy almost the entire allowed (mh,mA) parameter space of the MSSM will be accessible. A decisive test of the MSSM depends on the values of top mass and Supersymmetric parameters. The sensitivity mass range for a charged Higgs boson will be about 70 GeV for L = 500 pb

\Gamma 1

depending largely on the total integrated luminosity.

8. Acknowledgements

I would like to thank my fellow Higgs hunters for many fruitful discussions and the organizers of the conference for the warm hospitality.

REFERENCES 1 P.W. Higgs, Phys. Lett. 12 (1964) 132;

P.W. Higgs, Phys. Rev. Lett. 13 (1964) 508; P.W. Higgs, Phys. Rev. 145 (1966) 1156; F. Englert and R. Brout, Phys. Rev. Lett. 13 (1964) 321; G.S. Guralnik, C.S. Hagen and T.W.B. Kibble, Phys. Rev. Lett. 13 (1964) 585. 2 S.L. Glashow, Nucl. Phys. 22 (1961) 579;

A. Salam, Phys. Rev. 127 (1962) 331; A. Salam, in: "Elementary Particle Theory", ed. N. Svartholm (Stockholm, 1968), p. 361; S. Weinberg, Phys. Rev. Lett. 19 (1967) 1264. 3 G. Altarelli, R. Barbieri and S. Jadach,

Nucl. Phys. B 369 (1992) 3 and Nucl. Phys. B 376 (1992) 444; G. Altarelli, R. Barbieri and F. Caravaglios, Preprints CERN, TH/6770-93 and TH/6859- 93 (1993); J. Mnich, private communications. 4 S. Dawson, J.F. Gunion, H.E. Haber and

G.L. Kane, "The Physics of the Higgs Bosons: Higgs Hunter's Guide" Addison Wesley, Menlo Park, (1989). 5 Y.A. Golfand and E.P. Likhtman, JETP Lett.

13 (1971) 323; D.V. Volkhov and V.P. Akulov, Phys. Lett. B 46 (1973) 109; J. Wess and B. Zumino, Nucl. Phys. B 70 (1974) 39; P. Fayet and S. Ferrara, Phys. Rep. 32 (1977) 249; A. Salam and J. Strathdee, Fortschr. Phys. 26 (1978) 57; H.P. Nilles, Phys. Rep. 110 (1984) 1. 6 A.Sopczak, Preprint CERN, PPE/94-73

(1994) Proc. Int. Lisbon Fall School 1993, Nucl. Phys. B Proc. Suppl. C 37 (1995) p. 168. 7 F. A. Berends and R. Kleiss, Nucl. Phys. B

260 (1985) 32. 8 B.L. Ioffe and V.A. Khoze, Preprint LINP,

Leningrad, Nov. 1976;

17

J.D. Bjorken, Proc. of the 1976 SLAC Summer Institute on Particle Physics, Stanford, ed. M.C. Zipf (SLAC, Stanford, 1977) p. 1. 9 B.A. Kniehl, Z. Phys. C 55 (1992) 605,

A. Denner et al., Z. Phys. C 56 (1992) 261. 10 P.J. Franzini et al. in "Z Physics at LEP 1",

eds. G. Altarelli, R. Kleiss and C. Verzegnassi, CERN Report CERN-89-08, Vol. 2 (1989) p. 59. 11 DELPHI Collaboration, P. Abreu et al.,

Nucl. Phys. B 342 (1990) 1. 12 ALEPH Collaboration, D. Buskulic et al.,

Phys. Lett. B 313 (1993) 299; Physics notes 94-28, 94-29, 94-30, and 95-22. 13 DELPHI Collaboration, P. Abreu et al.,

Nucl. Phys. B 373 (1992) 3; Nucl. Phys. B 421 (1994) 3. 14 L3 Collaboration, Phys. Lett. B 303 (1993)

391, and preliminary results. 15 OPAL Collaboration, M.Z. Akrawy et al.,

Phys. Lett. B 253 (1991) 511; Phys. Lett. B 327 (1994) 397. 16 F. Richard, Preprint LAL, 94-50 (1994) Talk

at the 27th International Conference on HighEnergy Physics - ICHEP 94, Glasgow, Scotland, UK, 20 - 27 July 1994. 17 P. Janot, Preprint LAL, 94-54 (1994) Talk

at the 16th International Conference on Neutrino Physics and Astrophysics, Eilat, Israel, 29 May - 3 Jun 1994. 18 F. Grivaz, Preprint LAL, 92-23 (1992), erratum 92-45. 19 ALEPH Collaboration, D. Buskulic et al.,

Phys. Lett. B 313 (1993) 312. 20 L3 Collaboration, O. Adriani et al., Z. Phys.

C 57 (1993) 355, and preliminary results. 21 L3 Collaboration, B. Adeva et al., Phys. Rep.

236 (1993) 1. 22 DELPHI Collaboration, P. Abreu et al.,

Z. Phys. C 64 (1994) 183; Preprint CERN, PPE/94-218 (1994), in press Phys. Lett. B. 23 OPAL Collaboration, P.D. Acton et al.,

Phys. Lett. B 265 (1991) 475; Z. Phys. C 64 (1994) 1. 24 LEP Experiments, Preprint CERN, PPE/94-

187 (1994). 25 A. Sopczak, Preprint CERN PPE 94-188

(1994), in press Mod. Phys. Lett. A.

26 S. Komamiya, Phys. Rev. D 38 (1988) 2158. 27 ALEPH Collaboration, D. Decamp et al.,

Phys. Lett. B 241 (1990) 623. 28 OPAL Collaborations, M.Z. Akrawy et al.,

Phys. Lett. B 242 (1990) 299. 29 A. Sopczak, "Search for Non-minimal Higgs

Bosons from Z0 Decays with the L3 Experiment at LEP", Ph.D. Thesis, U. of Cal., San Diego, 1992. 30 J. Rosiek and A. Sopczak, Phys. Lett. B 341

(1995) 419. 31 C. Rubbia, Proc. LP-HEP 1991, Geneva,

(World Scientific, Singapore, 1991) p.441. 32 C. Wyss, LEPC Meeting, CERN, LEPC/94-

6, February 1994. 33 T. Sj"ostrand, Preprint CERN-TH., 6488/92

(1992). 34 F. A. Berends, P. H. Daverveldt and R. Kleiss,

Nucl. Phys. B 253 (1985) 441. 35 A. Sopczak, Int. J. Mod. Phys. A9 (1994)

1747. 36 D. Treille, Review talk given at the LEPC

Meeting, CERN, November 3, 1992. 37 E. Gross, B. Kniehl and G. Wolf, Preprint

DESY, 94-035 (1994). 38 P. Janot, Proc. XIII Moriond Workshop,

1992, France (Ed. Frontieres, 1993) p. 317. 39 A. Lopez-Fernandez, J.C. Romao,

F. de Campos and J.W.F. Valle, Phys. Lett. B 312 (1993) 240. 40 B. Zhou, `Study of b-tagging for LEP200',

Internal Report, Boston University, Oct. 29, 1992. 41 J. Alcaraz et al., Preprint CERN, PPE/93-28

(1993). 42 J. Rosiek, A. Sopczak, P. Chankowski, and

St. Pokorski, Preprint CERN, PPE/93-198 (1993). 43 CDF Collaboration, F. Abe et al., Phys. Rev.

D 50 (1994) 2966; S. Abachi et al., FERMILAB, PUB-94/354-E.

18

