Searches for New Phenomena at the Tevatron

Home , Tevatron

SEARCHES FOR NEW PHENOMENA AT

THE TEVATRON

Marc Paterno

DepartmentofPhysics and Astronomy

University of Ro chester, Ro chester, NY 14627-0171

Representing the CDF and D Collab orations

ABSTRACT

We present an up date on the current status of some new particle

searches at the Tevatron. We rep ort on searches for rst generation

scalar lepto quarks in the eej j and ejj channels, searches for the par-

ticles of sup ergravity in the multijet + missing transverse energy and

dielectron channels, searches for charginos and neutralinos in the sin-

gle photon and diphoton channels, and a search for a light c squark in

an R -parity violating sup ersymmetry mo del.

1997 by Marc Paterno.

1 The Tevatron and Its Detectors

The Tevatron at the Fermi National Accelerator Lab oratory is the highest energy

collider in the world. It collides counter-rotating b eams of protons and antiprotons

at a center-of-momentum energy s of 1:8 TeV. An excellent intro duction to

the Tevatron accelerator complex is found in Ref. 1. The 1992{1996 run of the

Tevatron concluded with delivery of an integrated luminosity in excess of 100 pb

to each of the two general-purp ose collider detectors, CDF and D .

Detailed descriptions of each detector can be found in Refs. 2{4. CDF Col-

lider Detector at Fermilab is a general-purp ose detector, lo cated at the B0 high-

luminosity interaction region of the Tevatron. Closest to the interaction region

is a silicon vertex detector, surrounded by a large wire chamb er, b oth of which

are lo cated inside a 1.4 Tesla sup erconducting solenoid magnet. Surrounding the

tracking system is a lead-scintillator calorimeter and a muon sp ectrometer. D is

a general-purp ose detector, lo cated at the D high-luminosityinteraction region

of the Tevatron. It consists of a nonmagnetic central tracking system, surrounded

by a hermetic uranium{liquid argon sampling calorimeter, and a toroidal muon

sp ectrometer.

2 Why Search?

Searches for new physics at the high-energy frontier test the Standard Mo del SM

by seeking phenomena b eyond it|we are less likely to discover new physics if we

study only standard pro cesses.

The searches we p erform may b e group ed into two broad categories: some are

driven primarily by theory, and some by data. For example, supersymmetry is a

feature often incorp orated in extensions to the SM. Although there has b een little

exp erimental evidence of sup ersymmetry, the theory is of sucient interest that

we consider searching for its existence. Two examples of data-motivated searches

are the current upsurge of interest in lepto quarks and in gauge-mediated sup er-

symmetry mo dels; the observation of an excess of high-momentum transfer events

at HERA spurred interest in lepto quarks, while the observation of a p ossible ee

event with large transverse momentum imbalance often called missing transverse

energy, or E caused an increased interest in gauge-mediated sup ersymmetry. T

Searches deal with some of the most imp ortant theoretical issues in HEP. Some

of the questions addressed by the searches discussed in this pap er are \Why is

the numb er of quark generations equal to the numb er of lepton generations?" and

\Why is the exp ected mass of the Higgs so much less than the Planck mass?"

It is through searches for physics beyond the predictions of the Standard Mo del

that wemay nd the information that allows us to answer these questions. In this

pap er, we present the status of these searches as of the time of the conference.

The reader should note that many of these analyses have b een up dated since that

time.

3 Standard Mo del: Generations

The Standard Mo del has three generations of leptons and an equal number of

generations of quarks. To date, all searches for additional leptons and quarks

have given null results. Measurements of the width of the Z b oson preclude the

existence of another light neutrino sp ecies. Furthermore, to prevent anomalies,

the Standard Mo del requires that there be an equal number of generations of

leptons and quarks. However, the SM contains no explanation of why the number

of quark and lepton generations should b e equal.

One p ossible explanation for this equality could b e the existence of a symmetry

that relates quarks and leptons. Such a symmetry could imply the existence of

a particle that has the prop erties of b oth leptons and quarks: these hyp othetical

6{8

particles are called lepto quarks.

4 Lepto quarks

4.1 A Bit of Theory

Lepto quarks LQ are hyp othetical particles combining prop erties of leptons and

quarks. They are color triplet b osons either scalars or vectors, carrying frac-

tional electric charge, nonzero baryon number B , and lepton number L. Most

mo dels containing lepto quarks contain three generations, corresp onding to the

three generations of leptons and quarks. Mixing between generations would give

rise to avor-changing neutral currents, which are strongly constrained by exp eri-

9{11

mental data ; therefore the lepto quarks of a given generation would decay into

one lepton and one quark of the corresp onding generation; e.g. LQ ! eu . Here

we have used the common convention for lepto quarks LQ and antilepto quarks

LQ .

There can b e several di erenttyp es of lepto quarks in a single generation, each

of which has di erent quantum numb ers, corresp onding to any of the four p ossible

combinations of the charged or neutral lepton, and the u-ord-typ e quark of that

generation. Each di erenttyp e of lepto quark then has well-sp eci ed decay mo des

and branching fractions. Partly for historical reasons, and partly for additional

generality, in exp erimental searches we typically intro duce a parameter , which

is the branching fraction for the lepto quark decay to the charged lepton and a

quark of the corresp onding generation. Since we do not generally identify the

typ e of quark in our detectors, we make no such distinction b etween u- and d-typ e

quarks in the decay pro ducts.

Lepto quarks, if they exist and have suciently small mass, should be pair-

pro duced at the Tevatron through the strong interaction. Searches for lepto quarks

at the Tevatron are thus not sensitive to the strength of the LQ-`-q coupling,

except that it not be so small that the LQ escap es the detector b efore decaying.

The interest in lepto quarks was increased by the observation of an excess of

2 + 12,13

high-momentum transfer high Q events in e p collisions at HERA. If it

is not due to a statistical uctuation, this excess could most easily be explained

as the result of the pro duction and decay of rst-generation lepto quarks with a

mass of approximately 200 GeV/c . Accordingly, most attention has b een paid to

searches for rst-generation lepto quarks, LQ .

4.2 First-Generation Lepto quark Searches at the Tevatron

The signature for the pro duction and decay of lepto quark pairs dep ends on the

branching fraction to charged leptons, . For = 1, the signature is two isolated

leptons plus two jets; for = 0 the signature is two jets plus large missing

transverse energy. For intermediate values of , three nal states are p ossible:

``j j , jj, and `jj. To obtain the b est p ossible limits, one must combine the

results for searches in all channels.

Both CDF and D have preliminary results from searches for rst-generation

lepto quarks, in the eej j and ejj channels. We present these results in the fol-

lowing sections.

4.2.1 CDF

CDF has searched for rst-generation lepto quarks in the eej j channel in a data

sample corresp onding to an integrated luminosity of 110 pb . The ma jor back-

grounds to this channel were the pair pro duction of t quarks in the dielectron

channel tt ! be be and electroweak pro duction of dielectron pairs Drell-

e e

Yan and Z b oson ! ee in asso ciation with jets.

Initial selection consisted of the requirement of two electrons with E >

25 GeV, and two jets, with E > 30 GeV and E > 15 GeV; the invari-

T 1 T 2

ant mass m of the dielectron pair was required to be away from the Z mass

2 2

m < 76 GeV/c or m > 106 GeV/c ; and the separate sums of elec-

ee ee

tron and jet transverse energies were each required to be greater than 70 GeV

j j

e e

1 2

2 1

> 70 GeV and E + E > 70 GeV. In the data, 12 events were ob- + E E

T T T T

served to satisfy these requirements. Final event selection was based on requiring

events to have electron-jet invariant masses consistent with the decay of a pair

of ob jects with the same mass. The two p ossible pairs of electron-jet invariant

masses were calculated, and the pairs with the smallest di erence in mass were

designated as M and M . Events were then required to satisfy M < 0:2M ,

ej ej LQ

1 2

where M = jM M j, and M is the mass of the hyp othetical lepto quark;

ej ej LQ

1 2

2 2

discrete values of M ranging from 140 GeV/c to 240 GeV/c were used in this

calculation. This selection requirement is illustrated in Fig. 1, which also shows

the 12 events passing the penultimate selection requirement, as well as the dis-

tribution for a sample of simulated lepto quark events, with M = 200 GeV/c .

The nal eciency forlepto quarks of this mass was 25.

In the nal sample, no event has an invariant mass greater than 140 GeV/c .

Three events pass all selection criteria, compared with an exp ected 5:8 2:2

events from SM sources. In the absence of evidence of a signal, this result was

used to set an upp er limit at the 95 con dence level on the cross section for

rst-generation lepto quark pro duction, shown in Fig. 2. Interpreted as a lower

limit on the mass of the rst-generation lepto quark, this result excludes rst-

generation lepto quarks with = 1, and mass less than 213 GeV/c at the 95

con dence level.

The CDF Collab oration also has preliminary results from a search for lepto-

quarks in the ejj channel, using a data sample with an integrated luminosityof

110 pb . This search required one electron with E > 20 GeV; E > 35 GeV;

T T ) 300 2 (GeV/c

ej2 250 M

200

150

100

0 0 50 100 150 200 250 300 2

Mej1 (GeV/c )

Fig. 1. The distribution of m and m for the data dark dots, and for a

ej ej

1 2

simulated sample of lepto quark events m = 200 GeV/c , from the CDF eej j

analysis. The region b etween the solid lines is the signal region.

no second lepton; at least two jets, one with E > 30 GeV and a second with

E > 15 GeV, and with the sum of all jets' transverse energies greater than

80 GeV; no b-tag for any jet; and transverse mass M e > 120 GeV. After

these requirements, two events were observed in the data, with a total back-

ground prediction of 3:2 0:8events 2:4 0:7events from tt pro duction, and

0:8 0:4 events from W +2 jet pro duction. The nal selection was based on

reconstructing the invariant masses of the two lepton-jet systems. Both combina-

tions of -jet transverse mass and e-jet invariant mass were calculated, and the

pairing that gave the smallest di erence was selected. To select signal events, a sig-

nal region in the M jversus M plane was de ned; the optimal region varied

T ej

for di erent hyp othetical lepto quark masses. For M = 180 GeV/c , the signal

2 2

region was de ned as 90

T ej

One event remained in this signal region. The nal eciency for lepto quarks with

a mass of 180 GeV/c was 16. In the absence of a signi cant signal, this result

Fig. 2. The cross section limit for the pro duction of rst-generation lepto quarks

from the CDF exp eriment search in the eej j channel. For a branching fraction

to charged leptons = 1, the resulting lower limit on the lepto quark mass is

213 GeV/c , at the 95 con dence level.

was used with no background subtraction to exclude lepto quarks with =0:5

and mass less than 180 GeV/c at the 95 con dence level.

4.2.2 D

The D exp eriment has also p erformed searches for rst-generation lepto quarks,

in b oth the eej j and ejj channels.

The eej j channel search was conducted using a data sample corresp onding to

an integrated luminosity of 123 pb . In this search, initial data selection required

two electron candidates electromagnetic calorimeter clusters with E > 20 GeV,

2 2

+ between each and two jets with E > 15 GeV. The distance R =

electron and jet was required to satisfy R > 0:7, and the invariant mass m of

the dielectron system was required to be away from the Z mass p eak: m <

2 2

82 GeV/c or m > 100 GeV/c . Finally, at least one electron candidate was

required to have a matching track in the central tracking volume. A total of

101 events passed these initial requirements, with a background exp ectation of

93 14 events from all SM sources 67 13 events from Drell-Yan ee + 2

jet pro duction, 24 4 events from four-jet pro duction and double mistaken

electron identi cation, and 1:8 0:7events from tt pro duction.

Final event selection was based on a transverse energy variable S , de ned as

the sum of the transverse energies of b oth electrons, and all jets with E > 15 GeV.

The selection threshold for this variable was set to give an exp ected background

near 0.4 events; the nal threshold was S > 350 GeV. The distribution of S

T T

for the 101 event initial selection compared very well with the exp ectations from

SM sources, as shown in Fig. 3. No event had S > 320 GeV. The nal signal

eciency varied from 16 for lepto quarks with a mass of 160 GeV/c , increasing

to 36 for lepto quarks with a mass of 250 GeV/c .

30 DØ Preliminary Pred. Background 25 Data

Events/25 GeV 20

0 100 150 200 250 300 350 400

ST (GeV)

Fig. 3. The distribution of the transverse energy variable S de ned in the text

for the D rst-generation lepto quark search in the eej j channel. The solid line

shows the prediction from Standard Mo del sources, and the dots show the data.

Since no signal for lepto quark pair pro duction was observed, this result was

used to determine an upp er limit at the 95 con dence level on the cross section

for lepto quark pro duction of 0.09 pb for lepto quark masses of 180 GeV/c , improv-

ing slightly with increasing mass to 0.07 pb at a lepto quark mass of 250 GeV/c ,

as shown in Fig. 4. When compared with the predicted lepto quark pro duction

cross section, this led to the exclusion at the 95 con dence level of lepto quarks

with = 1 and mass less than 225 GeV/c .

0.5

0.45

0.4

0.35 * Cross Section (pb)

2 0.3 β 0.25

0.2

0.15

0.1

0.05

0 180 190 200 210 220 230 240 250

LQ1 Mass (GeV/c2)

Fig. 4. The upp er limit at the 95 con dence level on the cross section for the

pro duction of rst-generation lepto quarks, from the D exp eriment search in the

eej j channel. For a branching fraction to charged leptons = 1, the resulting

lower limit on the lepto quark mass is 225 GeV/c , at the 95 con dence level.

In the ejj channel, D also has preliminary results, based on a data sample

with an integrated luminosity of 103 pb . The basic event selection required one

> 40 GeV and electron with E > 25 GeV, two jets with E > 25 GeV, E

T T

and each jet satisfying > 0:25 and j j > 0:25. with between the E

An additional requirement was S > 170 GeV, where S was de ned as the sum

T T

of the transverse of the electron plus the transverse energies of those jets with

E > 15 GeV. Finally,events with isolated muons were excluded, since muons are

not part of the rst-generation lepto quark signature. This preliminary selection

yielded 32 events, with an exp ected total of 30 5 events from all SM sources

20 4events from W + 2 jet pro duction, 9:0 2:7events from tt pro duction,

and 1:1 0:4 events from three-jet pro duction with misidenti cation of one jet

as an electron. Events were required to have an electron-E transverse mass

greater than 100 GeV/c . One event in the data passes this requirement; this

event is also a t-quark candidate. Standard Mo del background predictions were

0:5 0:3 events from W + 2 jet pro duction, 0:4 0:4 0:2 events from

three-jet pro duction and electron misidenti cation, and 1:5 0:5events from tt

pro duction, for a total background exp ectation of 2:5 0:6events.

The nal selection requirement is based on a variable M = jM M j=M ,

ej LQ LQ

where M is the hyp othesized lepto quark mass, and M is the electron-jet in-

LQ ej

variant mass; for each M , the jet which gave the smallest M was chosen.

Based on Monte Carlo studies of signal eciency and background exp ectations, a

requirementofM < 0:2was applied to the data; no events passed. The exp ected

background from all SM sources was less than 0.4 events.

Having observed no signal for lepto quark pro duction, this result was used to

exclude at the 95 con dence level lepto quarks with = 0:5 and with mass

b elow 158 GeV/c . In combination with the eej j channel search, lepto quarks with

mass b elow 195 GeV/c are excluded, for =0:5.

4.3 Lepto quark Summary

Together, these analyses exclude the lepto quark interpretation of the excess of high

Q events observed by the H1 and ZEUS Collab orations. A nal resolution to the

cause of this excess|new physics of another variety or statistical uctuation|will

havetowait for the additional data now b eing collected by the HERA exp eriments.

5 Sup ersymmetry

Sup ersymmetry SUSY is a symmetry that relates b osons and fermions. In

a SUSY mo del, each particle has a partner with the same quantum numb ers,

but with a spin that di ers by one half. Sup ersymmetry is interesting for many

reasons: it can help solve the quadratic divergence in the Higgs sector of the SM;

it can help resolve the hierarchy problem in grand uni ed GUT mo dels; it raises

the value of the scale for grand uni cation to a level at which predicted proton

16{21

decay rates are consistent with exp erimental b ounds.

Because no known particle has the appropriate qualities to b e the sup er partner of another, the minimal SUSY extension of the SM doubles the particle sp ectrum.

In addition, SUSY mo dels require two Higgs doublets, resulting in a total of four

Higgs b osons, three neutral and one charged.

The particle sp ectrum of the minimal sup ersymmetry extension of the Stan-

dard Mo del MSSM thus includes

quarks q spin and their partners, squarks q~ spin 0;

leptons ` spin and their partners, sleptons ` spin 0;

gluons g spin 1 and their partners, gluinos g~ spin ;

charged Higgs H spin 0 and charged weak vector b oson W spin 1 and

their partners, the mixed chargino states ~ and ~ spin ;

1 2

0 0 0

the neutral Higgs particles h , A , and H spin 0, the photon and the Z

0 0 0

b oson spin 1, and their partners, the mixed neutralino states ~ , ~ , ~ ,

1 2 3

and ~ spin .

By convention, the chargino and neutralino mass eigenstates are numb ered in

order of increasing mass.

Since no known particle has the correct mass and quantum numb ers to b e the

sup er partner of any other known particle, SUSY if it exists at all in nature

must be a broken symmetry. In most mo dels, sup ersymmetry breaking is put

explicitly into a \hidden sector" of very massive particles, and interactions be-

tween these particles and the particles of the low-energy sp ectrum communicate

the sup ersymmetry breaking to the low-energy sp ectrum. The p opular choices

for the messenger interaction are gravity, in sup ergravity SUGRA mo dels, or

gauge interactions, in gauge-mediated sup ersymmetry breaking GMSB mo dels.

SUGRA and GMSB mo dels each lead to distinctive and di erent predictions.

The intro duction of SUSY into the SM Lagrangian suggests the identi cation

of a new multiplicative quantum numb er, called R -parityR . For the particles

of the SM, including the Higgs b osons, R = +1; for their partners, R = 1.

P P

If R is conserved, then R = 1 particles sparticles must be pro duced in

P P

pairs, and the decay of each sparticle must include one sparticle; the result is that

the lightest sparticle the LSP must be stable. In order to avoid cosmological

constraints, in R -conserving mo dels the LSP must carry neither electric charge

nor color.

In SUGRA mo dels, the natural choice is the lightest neutralino, ~ ; in GMSB

mo dels, the natural choice is the gravitino, G, which is the partner to the graviton.

Mo dels in which R is not conserved must be constructed to guarantee stability

against proton decay; this is generally done by including R -violating terms which

pro duce either lepton numb er violation or baryon numb er violation, but not b oth.

Such mo dels predict a phenomenology signi cantly di erent from R -conserving

mo dels.

One can see that a wide variety of mo dels involving sup ersymmetry can be

de ned. The two Tevatron collab orations have conducted searches for evidence

of each of the typ es of mo dels listed ab ove; in the following sections, we present

some of these searches.

5.1 Sup ergravity Searches

The choice of gravity as the mediator of sup ersymmetry breaking leads to the

0 0

identi cation of the lightest neutralino ~ as the LSP. Since the ~ interacts

1 1

with normal matter only through the exchange or pro duction of presumably

heavy sparticles, the cross section for interaction in a collider detector is very

small. The characteristic signature of sup ergravity SUGRA mo dels is therefore

E resulting from the escap e of the pair of ~ s.

The most interesting SUGRA mo dels are those which are also grand uni ed

GUT mo dels. In such mo dels, all the gauge interactions are uni ed at some mass

scale M . At this scale, all the gauginos gluinos, charginos, and neutralinos

share a common mass, denoted by m . It is customary to make an additional

1=2

assumption of scalar uni cation: At the scale M , all the scalar partners of the

quarks and leptons share a common mass denoted by m . Furthermore, the

trilinear couplings between Higgs elds, leptons, and quarks are assumed to be

sp eci ed by a common coupling constant, A . Finally, to completely sp ecify a

mo del, two additional parameters must b e set. One must assign a value to tan ,

the ratio of the vacuum exp ectation values of the two Higgs doublets, and also

sp ecify the sign of , a parameter in the Higgsino mixing matrix, for which the

magnitude is set by requiring the correct mass of the Z b oson. We shall refer

to the SUGRA-GUT mo del de ned by these parameters as minimal low-energy

supergravity, or MLES.

After the sp eci cation of these ve parameters, the masses and mixings of

all particles are calculable, as are decay branching fractions and pro duction cross sections.

5.1.1 SUGRA Searches at D

For much of MLES parameter space accessible at the energy of the Tevatron,

the dominant sparticle pro duction mechanism would be pair pro duction of the

strongly interacting sparticles: pp ! g~g~, q~q~, g~q~, and q~q .

There are many decay mo des available for b oth squarks and gluinos, the details

of which dep end up on the details of the particle mass sp ectrum. This leads to a

rich phenomenology, of whichwe are just b eginning to take advantage. For a wide

range of parameter values, the most common decay mo des of squarks and gluinos

lead to nal states consisting only of quarks and the LSP. For much of MLES

0 0

parameter space, the \direct decays" q~ ! q ~ or g~ ! q q ~ are not common, and

1 1

instead decays often pro ceed through a cascade of sparticles, including charginos

and neutralinos. For example, one common squark decay could b eq ~ ! q g~ followed

0 0 0 00 0

0 00

by g~ ! q q ~ followed by ~ ! q q ~ ; the nal result in this case would be

2 2 1

0 00 0

0 00

q~ ! q q q q q ~ . The signature for such pro duction and decay is the observation

of multiple jets and E . Although the number of quarks app earing in the decay

may be large, most often they are not all individually distinguishable as jets. To

keep reasonable eciency for the detection of such events, typically one requires

a more mo dest number of jets.

The D Collab oration has p erformed a search for squarks and gluinos, based

on an integrated luminosityof79pb , using the signature of three jets and large

23,24

E . This approach di ers from earlier analyses in that this search uses the

constraints of MLES to determine the particle sp ectrum, rather than more ad hoc

assumptions. The ma jor backgrounds included the SM sources of high momentum

neutrino pro duction, in asso ciation with jets: tt pair pro duction, W and Z b oson

pro duction in asso ciation with jets, and WW and WZ asso ciated pro duction. Also

signi cantwas the instrumental background of SM multijet pro duction, with E

as a result of mismeasurement, either of one or more jet energies or of the vertex

lo cation.

> 75 GeV; at least three jets with E > Initial event selection required E

25 GeV, one of which was required to have E > 115 GeV, and all of which

; no isolated were required to be isolated in azimuth from the direction of the E

jets

electrons or muons with E > 15 GeV; and to have H = E > 100 GeV,

T T 2

where the sum is taken over all jets with E > 25 GeV, but do es not include

the leading E jet. To reduce the background from multiple interactions, the T

central jet with greatest E was required to be consistent with emanation from

the primary interaction vertex. A total of 9:3 3:5events was exp ected from all

SM sources 3:1 1:1 from tt pro duction, 2:7 1:0 from W and Z pro duction,

and 3:5 2:6 from SM multijet pro duction; 15 events were observed in the data.

When one considers di erent regions in the m ;m parameter plane, one

0 1=2

nds that for low m and large m , the squark pair pro duction dominates, and

1=2

for large m and small m , gluino pair pro duction dominates. Squark pro duction

1=2

tends to pro duce fewer jets, each with larger E , while gluino pro duction tends

to pro duce more jets, with somewhat lesser E . The amountofE pro duced also

varies, b ecause of the changing mass of the ~ , and b ecause of the changes in the

cascade decays. For these reasons, nal selection criteria for this analysis were

determined by varying the E and H thresholds applied to di erent regions in

T 2

the m ;m parameter space. Thresholds were chosen to maximize the gure of

1=2

merit S= B , where S is the exp ected number of events from the signal pro cesses,

and B is the total uncertainty in the background prediction, including b oth

statistical and systematic uncertainty. The nal E thresholds ranged from 75

to 100 GeV, and the H thresholds from 100 to 160 GeV. For none of the nal

T 2

selection criteria was there an observation of a signi cant excess of events b eyond

SM predictions. Therefore, this result was used to determine a 95 C.L. exclusion

contour in the m ;m plane; the exclusion contour is shown in Fig. 5.

0 1=2

In addition to the multijet + E search, D has also p erformed a search

for sup ersymmetry in the dielectron + jets + E channel. This search takes

advantage of the smaller SM backgrounds for dielectron pro duction. In a data

sample corresp onding to an integrated luminosity of 90 pb , this search lo oked

for events with exactly two electrons with E > 15 and j j < 1:2or1:4 < j j < 2:5,

two or more jets with E > 20 GeV and j j < 2:5, and E > 25 GeV. Events

with 79

event also had E > 40 GeV. The calculated background exp ectation from SM

sources was 2:9 1:3 events; the ma jor contributions were from tt pro duction

1.1 events, Z ! ee 1.0 events and multijet events in which two jets were

misidenti ed as electrons 0.4 events. A total of two events was observed in the

data, consistent with SM exp ectations. In the absence of a signal, this result was

used to determine an exclusion contour in the m ;m plane; the contour is

0 1=2

shown along with that of the D multijet + E search in Fig. 5. T 140 400 D¯ Preliminary (GeV)

4 120 0 1/2 0

M 300 100 3 0 0 80

200 60 2 0 0 40

1 100 0 20 0

0 0 100 200 300 400

M0 (GeV)

Fig. 5. The exclusion contour for the D sup ergravity searches. The region b elow

the dark line is excluded at the 95 C.L. by the multijet + E analysis, and the

region b elow the light line is excluded at the 95 C.L. by the dielectron + jets

analysis. The nearly horizontal lines are contours of constant gluino mass, + E

and the elliptical lines are contours of constant squark mass; the lab els show the

relevant particle's mass in GeV/c .

The most striking feature of the exclusion contour from the dielectron channel

analysis is the dip in the region near m = 90 GeV/c . This dip o ccurs b ecause of

the complicated relationships b etween the sparticle masses in MLES. For most of

the m ;m parameter space, the ma jor source of dielectron events is the decay

1=2

0 0 0

~ ! ee ~ , where the ~ is either pro duced in the decay of a squark or gluino,

2 1 2

or is pro duced directly, in asso ciation with a chargino or other neutralino. In the

region of the dip, however, the mass of the ~ is less than that of the ~ , which

op ens the decay channel ~ ! ~, and e ectively turns o the three-b o dy decay.

This causes the drop in sensitivity. For still lower values of m , the e~ b ecomes

suciently light to app ear in decays, and thus some sensitivity to dielectron decays

re-app ears.

In summary, the two D analyses exclude gluinos with mass b elow 185 GeV/c ,

and exclude gluinos and squarks with equal masses b elow 267 GeV/c .

5.2 SUSY Searches with Photons

The observation by CDF of one event containing two photons, two electromag-

netic clusters one of which is well-identi ed as an electron and large E caused

an upsurge of interest in those SUSY mo dels that predict nal states containing

pairs of photons, since such an event is dicult to explain within the Standard

Mo del. The SUSY mo dels that predict such events include two broad classes:

mo dels with gauge-mediated sup ersymmetry breaking, and mo dels with radiative

26{31

decay of neutralinos. In the GMSB mo dels, the source of the photons is the

0 0 0

decay ~ ! G; in the radiative decay mo dels, the source is the decay ~ ! ~ .

1 2 1

In b oth cases, the nal state contains one or more photons and E . Both CDF

and D have conducted searches for events consistent with such mo dels.

5.2.1 Search for Scalar Top with Radiative Neutralino Decay

CDF has p erformed an analysis searching for the pro duction and decayofcharginos

and neutralinos in a mo del which includes a light scalar top quark t ; this mo del

event. These mo dels sp ecify m = was prop osed to explain the CDF ee E

2 2 2 2

40 GeV/c , m = 70 GeV/c , m = 60 GeV/c , m = 225 GeV/c , and

~ g~

m = 220 GeV/c for q 6= t.

Based on such a mo del, CDF has searched for the pro duction and decay ~ ~ !

0 0 0

! c ~ b ~ , where i =1 or 2. The analysis considered b oth direct t b ~

1 1 1

pro duction and the cascade decays of gluinos and squarks. Initial event selection

required a photon with E > 25 GeV, a displaced vertex indicating the decayof

a b-quark observed in the Silicon Vertex Detector, and E > 20 GeV; 98 events

meeting these criteria were observed in the data. A total of 77 23 20 events

was exp ected from all SM sources, including pro duction of a b quark in asso ciation

with a jet that is misidenti ed as a photon 60, photon + bb or cc 27, and

direct photon pro duction with an erroneous displaced vertex identi cation 13.

Final event selection consisted of raising the E threshold to 40 GeV. Two

events remained in the data; a background calculation has not b een completed.

Since no signi cantnumberofevents was observed, this result was used without

background subtraction to set an upp er limit at the 95 con dence level of

6.4 events on number of events exp ected from the ~ ~ signature. The suggested

mo del predicts 6.7 events in 85 pb , and is thus ruled out at slightly more than

the 95 con dence level, as is any similar mo del with a lighter t .

5.2.2 Search for E

Using a data sample with an integrated luminosity of 106 pb , D has p erformed

an inclusive search for events with two photons and signi cant E . This search

neither required nor veto ed extra jets or leptons, and is thus sensitive to mo dels

with radiative neutralino decay and to GMSB mo dels. The SM backgrounds to

this nal state include direct photon pro duction, with one jet misidenti ed as a

photon, and mismeasured E ; W ! e pro duction in asso ciation with a single

jet, in which the electron track is missed resulting in its misidenti cation as a

photon and one jet is misidenti ed as a photon; and tt ! eej j + E events, in

which b oth electrons are misidenti ed as photons.

Initial event selection required two electromagnetic clusters in the calorimeter,

the rst with E > 20 GeV, and the second with E > 12 GeV, b oth in the range

T T

j j < 1:1 or 1:5 < j j < 2:0; 229 events were observed in the data. This sample

distribution for these included events with electrons as well as photons. The E

events agreed well with SM exp ectations, as is seen in Fig. 6.

The nal event selection consisted of photon veri cation requiring no track

and no signi cant number of tracking chamber hits in the vicinityof the electro-

magnetic clusters and the requirement E > 25 GeV. Two events were observed

in the data, with an exp ected 2:3 0:9 from all SM sources. This result was used

Fig. 6. The E sp ectrum for events passing the initial event selection in the D

diphoton + E analysis, compared with the exp ectation from Standard Mo del

sources. The dots represent the data, and the histogram represents the SM ex-

p ectation.

to set an upp er limit at the 95 con dence level on the cross section for chargino

and neutralino pro duction, in b oth GMSB and radiative neutralino decay mo dels.

For the GMSB comparison, the mo del used was less restrictive than MLES.

This mo del assumes electroweak uni cation but do es not assume grand uni cation.

To describ e the phenomenology of the chargino and neutralino sectors, it requires

the sp eci cation of three parameters: M a mass asso ciated with the SU 2

2 L

gaugino; this value also determines M , the mass asso ciated with the U 1 gaug-

ino; and two of the same parameters as in MLES mo dels, and tan . Unlike

MLES mo dels, b oth the magnitude and sign of are free. With the assumption

that the sleptons, squarks, and gluinos are heavy enough not to o ccur in the decay

of the ~ and ~ , these parameters suce to describ e the phenomenology of the

0 0

~ , ~ , and ~ .

2 1

For GMSB mo dels, this limit was expressed as an exclusion contour in M ;

parameter space; the excluded region is shown in Fig. 7. This analysis excludes

2 0 2

~ with mass less than 156 GeV/c , and ~ with mass less than 79 GeV/c ,

1 1

b oth at the 95 con dence level. The excluded region includes the entire region

o ccupied by the GMSB mo del prop osed to explain the CDF event. For radiative

neutralino decay mo dels, the result can be expressed as a cross section limit for

0 0

~ ~

~ pro duction, e~e pro duction, or ~ ~ pro duction. The cross section limits are

2 2

also shown in Fig. 7.

400

350 (pb) DØ bounds tanβ=2 σ LEP bounds ν~ν~ 300 10 Ellis et al. ~~

95% CL ee 250 χ~0χ~0 2 2

200 (GeV)

2 1 M 150

100

-1 50 Excluded 10

0 10 20 30~ 40~ 50 60 -750 -500 -250 0 250 500 750 m(χ0)-m(χ0) (GeV/c2)

µ (GeV) 2 1

Fig. 7. Left: The exclusion contour in the M ; parameter plane for gauge-

mediated sup ersymmetry breaking mo dels, from the D diphoton + E analysis.

The region b elow the upp ermost lines is excluded at the 95 con dence level.

Also shown are the regions excluded by the LEP exp eriments, as well as the

region o ccupied by the GMSB mo del prop osed to explain the ee E observed by

CDF. Right: The upp er limit at the 95 con dence level on the cross section

for various sparticle pair pro duction mechanisms in radiative neutralino decay

0 0

mo dels, as a function of the mass di erence between ~ and ~ , resulting from

2 1

analysis. the D diphoton + E

5.3 R-Parity Violating SUSY Searches

2 +

The excess of high-Q events in e p collisions observed by the H1 and ZEUS

Collab orations as mentioned in Sec. 4.1 could b e an indication of a squark reso-

/ / /

R R R

+ +

nance interacting via R -violating coupling e d ! c~ ! e d, where ! indicates

P L

32,33

a pro cess that do es not conserve R . Such a mo del violates lepton number

conservation, but not baryon numb er conservation, and therefore, do es not predict

proton decay. In such mo dels, the charm squark c~ may also have R -conserving

L P

decays, whichmay b e predominant. Additionally, b ecause of the presence of R -

violating interactions, the lightest sparticle need not be stable, and thus, it may

carry electric charge or color. In the references noted, two MSSM variants have

b een prop osed to explain the HERA excess. Both mo dels share a similar particle

sp ectrum, with the di erence b etween the two b eing the identi cation of the LSP:

either c~ or ~ .

2 2 2 2

Each mo del sp eci es tan = 2, m > m , m = m m + m

c s

c~ s~ c~

~ L L L

m cos 2 , and all u- and d-typ e squarks degenerate, with mass m .

Such a mo del predicts signatures which should b e observable at the Tevatron.

If the LSP is c~ , then one such signal is gluino pair pro duction and decay pp !

+ +

g~g~ ! cc~ c c~ ! ce dce d, and the charge conjugate. The Ma jorana nature

L L

of the gluino allows decay pro ducts with same-sign dileptons 50 of the time.

If the LSP is ~ , then the interesting signal is c~ pro duction: pp ! c~ c~ !

L L L

0 0 0 0

c ~ c ~ ! cqq e cqq e . In this pro cess, it is the Ma jorana nature of the

1 1

neutralino that allows the same-sign dilepton signature. In contrast with R -parity

conserving sup ersymmetry signals, these signatures do not contain E . The cross

section to pro duce such events from SM sources is very small.

Using a data sample corresp onding to an integrated luminosity of 105 pb ,

CDF has searched for events with like-sign electrons and two jets, consistent with

the mo dels ab ove. Initial event selection required two electrons with E > 15 GeV

and two or more jets with E > 15 GeV, with j e j < 1:1 and j e j < 2:5.

T 1 2

The requirement that the distance b etween the p oints at which the electron tracks

intercepted the b eam axis b e less than 10 cm was used to ensure that b oth leptons

originated from the same interaction, and the E signi cance requirement was

used to veto events with E resulting from the mismeasurement of jets.

The SM contributions to this signature come from heavy quark pro duction and

subsequent cascade decays. Preliminary calculations of tt and b=c-quark pro duc-

tion rates have b een p erformed, yielding a predicted 1:2 0:2 statistical error

signi cance of only opp osite-sign ee events. One such event is observed with E

signi cance cut at 5:0. At the 10:1; this event is lost to the sample due to the E

95 con dence level, fewer than 0:09 events with like-sign dileptons are exp ected

from tt pro duction and decay. More dicult to estimate is the exp ected num-

ber of events from b- and c-quark pro duction, either through direct pro duction or

from gluon splitting in initial- or nal-state radiation. A preliminary upp er limit

of 9:5events was exp ected, at the 95 con dence level.

This result was used to set upp er limits on the cross section for b othg ~g~ ! ` ` +

2j andc ~ c~ ! ` ` +2j. The regions of MSSM parameter space excluded in the

L L

context of the two mo dels are shown in Fig. 8.

∼ ∼ ± ± ∼ ∼_ ∼ _ ∼ ± ± → R/ c c → c χ0 c χ0 → e e + X R/ p gg e e + X p L L 1 1 ) ) 2 2 ) (GeV/c ) (GeV/c 0 1 ∼ ∼ χ ∼ 0 ) + M(c)

M(q χ 1 M(

∼ ) = M( L M(c

∼ Br(c → ed)=1.0 ∼ ± ) L χ 1 ∼ ∼ ) = M( 2 L M(cL) = 200 GeV/c Excluded at 95% C.L.M(c tan β = 2 Choudhury and Raychaudhuri Br to LSee = 1/8 Phys. Rev. D56, 1778 (1997).

∼ 2 ∼ 2 M(c ) (GeV/c )

M(g) (GeV/c ) L

Fig. 8. Left: The exclusion contour in the m ;m mass plane from the CDF

g~ q~

like-sign dilepton analysis, in the framework of the g~g~ ! ` ` +2j mo del. The

shaded region is excluded at the 95 con dence level. Right: The exclusion

contour in the m ;m mass plane from the CDF like-sign dilepton analysis,

in the framework of the c~ c~ ! ` ` +2j mo del. The shaded region is excluded

L L

at the 95 con dence level.

6 Conclusion

The Standard Mo del remains a successful description of elementary particle physics.

Thus far we have observed no evidence of phenomena beyond the predictions of

the SM. Nonetheless, theories which go beyond the SM have attractive features

that make their investigation imp ortant. The current data sample 100 pb ,

for each of the two exp eriments has provided excellent opp ortunities to investi-

gate many such theories, and has provided a fertile base for analyses which prob e

such theories to a degree previously unachievable.

In the year 2000, the CDF and D exp eriments will b egin a new round of data

collection, with a Tevatron upgraded to s =2 TeV, delivering b eam to the two

upgraded detectors. The next run will deliver an integrated luminosity of 2{4 fb

to each detector, a factor of 20 greater than the previous run. Both exp eriments

are currently preparing for this Run I I, which will provide an excellent opp ortunity

to observe the remaining undiscovered memb er of the SM particle sp ectrum|the

Higgs b oson|or, p erhaps, something unexp ected.

References

[1] W. J. Thompson, \Intro duction to Colliding Beams at Fermilab,"

FERMILAB-TM-1909, 1994.

[2] F. Ab e et al. CDF Collab oration, Nucl. Instrum. Metho ds A 271, 387

1988.

[3] F. Ab e et al. CDF Collab oration, Phys. Rev. D 50, 2966 1994.

[4] S. Abachi et al. D Collab oration, Nucl. Instrum. Metho ds A 338, 185

1994.

[5] The LEP Collab orations, the LEP Electroweak Working Group, and the SLD

Heavy Flavour Group, CERN-PPE/97-154.

[6] J. Pati and A. Salam, Phys. Rev. D 10, 275 1974.

[7] E. Eichten et al.,Phys. Rev. Lett. 50, 811 1983.

[8] J. L. Hewett and T. G. Rizzo, Phys. Rep. 183, 193 1989.

[9] M. Leurer, Phys. Rev. D 49, 333 1994.

[10] M. Derrick et al.,Z. Physik C 73, 613 1997.

[11] S. Aid et al. H1 Collab oration, Phys. Lett. B 369, 173 1996.

[12] C. Adlo et al. H1 Collab oration, Z. Phys. C 74, 191 1997.

[13] J. Breitweg et al. ZEUS Collab oration, Z. Phys. C 74, 207 1997.

[14] N. Amos et al., Proceedings of the International Conference on Computing

in High Energy Physics, 95, 215 1996.

[15] Yu. A. Golfand and E. P. Likhtman, JETP Lett. 13, 452 1971; D. Volkov

and V. P. Akulov, JETP Lett. 16, 438 1972.

[16] J. Wess and B. Zumino, Nucl. Phys. B 78,1 1974.

[17] H. Georgi and S. L. Glashow, Phys. Rev. Lett. 32, 438 1974.

[18] E. Gildener, Phys. Rev. D 14, 1667 1976.

[19] S. Dimop oulos and H. Georgi, Nucl. Phys. B 193, 150 1981.

[20] S. Dimop oulos et al.,Phys. Rev. D 24, 1681 1981.

[21] N. Sakai, Z. Phys. C 11, 153 1981.

[22] J. Ellis et al., Nucl. Phys. B 238, 453 1984.

[23] S. Abachi et al. D Collab oration, Phys. Rev. Lett. 75, 618 1995.

[24] F. Ab e et al. CDF Collab oration, Phys. Rev. D Rapid Communications,

56, R1357 1997.

[25] D. Toback, FERMILAB-Conf-96/240-E 1996.

[26] H. Hab er et al.,Phys. Lett. B 160, 297 1985.

[27] S. Dimop oulos et al.,Phys. Rev. Lett. 76, 3494 1996.

[28] S. Ambrosanio et al.,Phys. Rev. Lett. 76, 3498 1996.

[29] G. Kane and J. Wells, Phys. Rev. Lett. 76, 4458 1996.

[30] K. Babu et al.,Phys. Rev. Lett. 77, 3070 1996.

[31] J. Lop ez and D. Nanop oulos, Mo d. Phys. Lett. A 10, 2473 1996.

[32] D. Choudhury and S. Raychaudhuri, Phys. Lett. B 401, 54 1997.

[33] D. Choudhury and S. Raychaudhuri, Phys. Rev. D 56, 1778 1997.