rpv stop paper seminar - Texas A&M...
Transcript of rpv stop paper seminar - Texas A&M...
Search for pair productionof scalar top quarks
decaying to a ττττ leptonand a b quark in___
collisions at TeV1.96s =
Autors:M. Chertok, T. Kamon, V. Khotilovich, R. Lander, A. Safonov
CDF Paper Seminar, February 21, 2008
pp
Godparents:T. Liss (chair), C. Neu, K. Osterberg
CDF Note 9038
02/21/2008 CDF Paper Seminar 2
� Process, motivation and existing limits� Data and MC samples� Event selection� Background estimation� Fitting procedure� Cross-section and mass limits� Summary
Outline
02/21/2008 CDF Paper Seminar 3
RPV Stop (and LQ3) Pair Production
� Supersymmetric top quark can be pair-produced at the Tevatron � R-parity separates SM and SUSY particles:
� if violated, stop can decay into tau and a b-jet� for wide range of parameters high is expected
� Set of leading order Feynman diagrams:
q g
g q
g 1 t ~
1 t ~
b
τ
τ
b
g
g
1 t ~
1 t ~
1 t ~
b
τ
τ
b
g
g
1 t ~
1 t ~
b
τ
τ
b
SLBpR 23)1( ++−=
� Identical to the case of 3rd generation scalar leptoquark (LQ3 )� Existing Limits:
� LQ3: m > 99 GeV/c2 (LEP / CDF Run I)� RPV stop: m >122 GeV/c2 (CDF Run I)
( )Br t bτ→ɶ
02/21/2008 CDF Paper Seminar 4
Signature, Run II Improvements
Signature to look for:� e or µµµµ :
from leptonically decaying tau� ττττh : hadronically decaying tau� 2 jets : from b-quarks
Improvements in Run II:� Increased cross-section (~35%)� Increased muon coverage (CMX)� Dedicated tau triggers� More data� Better particle ID� Improved analysis technique
pp
t~
t~
b
b
τ
τ
jet
jet
02/21/2008 CDF Paper Seminar 5
Data and MC Samples
� Dataset: L=322 pb-1
� Lepton+Track Triggers� 8 GeV lepton – electrons & muons in |η| < 1� 5 GeV/c Isolated Track
� Signal: m(stop) = 100 - 180 GeV/c2
� AcceptancePYTHIA (CTEQ5L PDF) + CDF detector simulation
� NLO cross-sectioncalculated with PROSPINO v2 (CTEQ6M)
02/21/2008 CDF Paper Seminar 6
Backgrounds
� Z (� τlτh ) + ≥ 2jets (dominant; use MC corrected with Data)
� tt-bar, WW/WZ/ZZ (small; use MC)
� Z (�ee/µµ) + ≥ 2jets (dominant; use MC corrected with Data)
� QCD (≥ 4jets) (use Data)
� W (�l / �τ � l ) + ≥ 3jets (small, but hard to estimate; use Data)
Jet, e, µµµµ: misidentified as ττττh
02/21/2008 CDF Paper Seminar 7
Event Selection
� Central e or µµµµ : pT(l)>10 GeV/c� Hadronic tau : pT(τ)>15 GeV/c� ID cuts� Background removal:
� DY, cosmics rays, conversions� ST = pT(l) + pT(ττττ) + MET > 110 GeV
� strong suppression of QCD and Z→ττ� cut value is optimized
� Sort remaining events by N jet & mT(l,MET) � Signal dominates in Region A2
( “the blind box” )� mT(l, MET) < 35 GeV/c2
separates signal and W, top and di-boson backgrounds
� Njet≥≥≥≥2 (ET >20 GeV is optimized)separates signal and most of backgrounds (including Z →ττ)
B0A0
B2A2
MT
N “
extr
a”je
ts0
≥≥≥≥2
35 GeV
1
S
02/21/2008 CDF Paper Seminar 8
Validation: Njet=0 Regions A0 and B0
� Use ST >80(better statistics for validation)
� A0 is dominated by Z→ττ� clear tau signal
A0 B0
A0
B0 is mostly W+jets
mT(l
, ME
T)
< 35
GeV
/c2
02/21/2008 CDF Paper Seminar 9
Signal Event Selection: Summary
Total signal event selectionacceptance (in region A2 asa function of m(stop) for:� Electron channel� Muon channel:
CMUP (|η|<0.6)CMX (0.6<|η|<1.0) separately and combined
� Muons (combined) and Electrons give similar acceptance
02/21/2008 CDF Paper Seminar 10
� Use fitting to the Data to find:� Probability Density Function
as a function of signal cross-section
� estimate W+jets background
� “4-region” method:� Assume for W+jets
R≡≡≡≡[N(A2)/N(B2)]/[N(B0)/N(A0)]~1� unreliable rates but reliable ratios� verified with MC
� Estimate all non-W+jets BG� Estimate NB2/NA2, NA0/NA2, NB0/NA2
for signal events� Include systematic uncertainties
(are in range of 7.0%~11.2%)� Fit to Data in all 4 regions
Fit Strategy
B0A0
B2A2
MT
N “
extr
a”je
ts
0
≥≥≥≥2
35 GeV
1
Result of fit:can set limits or discover stop
A1 B1
02/21/2008 CDF Paper Seminar 11
Signal Region
Njet Distribution for mT<35 GeV/c2
� Signal region (Njet ≥ 2):� was looked after
the validation in control regions was completed
� Nobs = 2 (1e+1µµµµ)� NBG = 0.5
-0.43.0+
used in paper
A0 A1 A2
02/21/2008 CDF Paper Seminar 12
Number of events in Different Regions
used in paper� Note: W+jet and other BG’s are not combined
W+jet are estimated in the fit and not independent from number of data evt., other BG’s and expected signal fractions
02/21/2008 CDF Paper Seminar 13
Stop Cross-Section Upper Limits
� No excess observed � set limits
� Theoretical uncertainties on cross-section:� CTEQ6.1M
� ± ½ of QCD renorm./factor. scale
� 95% C.L. upper limit on whereas function of stop mass, calculated for the cases when theoretical uncertainties were or were not considered:
used in paper
21 1( )t tσ β×ɶ ɶ
1( )Br t bβ τ= →ɶ
02/21/2008 CDF Paper Seminar 14
Stop Mass Limits
� Stop mass 95% C.L. limits for the case :� m > 153 GeV/c2 (theoretical uncertainties considered)� m > 156 GeV/c2 (not considering theor. uncertainties)
used in paper
1( ) 1Br t bτ→ =ɶ
02/21/2008 CDF Paper Seminar 15
Summary and Plans� No RPV stop discovery with L=322 pb -1
� Expected BG events in primary signal region� Observed 2 events� Set 95% C.L. upper limit on
as function of stop mass� Stop mass 95% C.L. limit for case:
� m > 153 GeV/c2 (theoretical uncertainties considered)
� Results are fully applicable to Scalar LQ3
� Possible Prospects:� Bigger dataset will need completely new analysis with b-tagging� Combined search of modes
has more potential
� Thanks� Godparents: Tony Liss, Christopher Neu, Kenneth Osterberg� Readers: SPRG, Cantabria, Pisa, IPP (Canada), Fermilab
0.5-0.43.0+
21 1 1( ) ( )t t Br t bσ τ× →ɶ ɶ ɶ
1 1 1& &t eb t b t bµ τ→ → →ɶ ɶ ɶ
1( ) 1Br t bτ→ =ɶ
02/21/2008 CDF Paper Seminar 16
Backup