1 CDF 実験の最新結果 Yuji Takeuchi (Tsukuba) for CDF collaborations 日本物理学会 第 64 回年次大会 Mar. 28 th, 2009.

1

CDF 実験の最新結果

Yuji Takeuchi (Tsukuba)for CDF collaborations

日本物理学会　第 64 回年次大会Mar. 28th, 2009

2

CDFD0

Chicago

TEVATRON

MAIN INJECTOR

Tevatron Run IIStill at the Energy Frontier

Tevatron:• The world's highest-energy particle collider• Proton-antiproton collisions at √s = 1.96 TeV• Tevatron is performing really well:

Reach 6 fb-1 and 7~8 fb-1 expected by the end of 2009.

3

The CDF II Detector

• Silicon vertex detector (1+5+2 layers)• Central drift chamber (8 super layers) • 1.4T solenoid• Good particle identification (K,)• Central/Wall/Plug calorimeters• Scintillator + drift chamber muon detectors

4

CDF Collaboration

~700 physicists from

12 nations and 61 institutionsMcGill Univ.Univ. of Toronto

Argonne National Lab.Baylor Univ.Brandeis Univ.UC DavisUC Los AngelesUC San DiegoUC Santa BarbaraCarnegie Mellon Univ.Univ. of ChicagoDuke Univ.FermilabUniv. of FloridaHarvard Univ.Univ. of IllinoisThe Johns Hopkins Univ.LBNLMITMichigan State Univ.Univ. of MichiganUniv. of New MexicoNorthwestern Univ.The Ohio State Univ.Univ. of PennsylvaniaUniv. of PittsburghPurdue Univ.Univ. of RochesterRockefeller Univ.Rutgers Univ.Texas A&M Univ.Tufts Univ.Wayne State Univ.Univ. of WisconsinYale Univ.

JINR, DubnaITEP, Moscow

Univ. Karlsruhe

Univ. of Geneva

Glasgow Univ.Univ. of LiverpoolUniv. of OxfordUniv. College London

Univ. of Bologna, INFNFrascati, INFNUniv. di Padova, INFNPisa, INFNUniv. di Roma, INFNINFN-TriesteUniv. di Udine

IFAE, BarcelonaCIEMAT, MadridUniv. of Cantabria

LPNHE, Paris

KHCL

KEKOkayama Univ.Osaka City Univ.Univ. of TsukubaWaseda Univ.

Academia Sinica

USA Canada

Russia

Germany

Switzerland

UK

Italy

Spain

France

Korea

Japan

Taiwan

5

• Continually establishing new records! – Peak luminosity

~3.5×1032 cm-2 s-1

– Weekly integrated luminosity78 pb-1/week

– Annual integrated luminosity1.8 fb-1 (US FY 2008)

– Average pbar accumulation rate

• 21×1010 pbar/ hour

• Very stable operation– Maximizing integrated

luminosity– ~12 weeks shutdown in 2009

summer

Tevatron Status

FY02FY03

FY04FY05

FY06FY07

FY08

FY09

6

Integrated Luminosity• Integrated luminosities for each FY

– Extrapolation for FY2009: ~2 fb-1

• Data taking efficiency– Typically ~85%

FY08

FY09

7

FY09 start

9.32 fb-1

7.78 fb-1

Highest Int. Lum

Lowest Int. Lum

FY10 start

time since FY04

Inte

grat

ed lu

min

osity

(fb

-1)

FY08 start

~ 1.8 fb-1 delivered in FY08

FY2010• Fermilab plans to run (Budget approved)

and CDF/DØ are ready

• Another ~2.5 fb-1 ， total ≳9 fb-1

FY2009• Run continues

~2 fb-1, total ~7 fb-1

Shutdown ： ~12 weeks from June

Tevatron Prospects

5

9

0

8

jets

HF

WZ

W Z

WW tt WZ

singletop

ZZ

Higgs

NewPhysics?

Physics Programs at Tevatron

• New SM process– WZ,ZZ, Single-Top, ...

• Top quark physics• Precise measurement

– W mass, ...

• Heavy flavor phyics (bb)50s

• Higgs search• Search for new physics

– First exploration of TeV scale

9

Top Quark Physics

Single top については二つあとの中村氏（筑波）のトーク

10

Top Pair production

Dominant process

10%~20%

11

Top Pair Signature

Dilepton• 2 lepton• 2 bjet• MET

All Hadronic• 6jet(2bjet)

Lepton+Jet• 1lepton• 4jet(2bjet)• MET

-1-1QCD

11 MeV) (200 ~ << GeV 1.5

Top quark decays as a bare quark Information on spin polarization and momentum is directly

transferred to decay products

12

Top Pair Production Cross-section

• Now better than theoretical uncertainties• Luminosity is dominant source of uncertainty

13

new(ttbar)/(Z) in lepton + jet

channel

• The total uncertainty is decreased by 10% by taking the ratio of the top pair to the Z cross sections

+

• Event selection– High PT lepton 3jet, MET

• NN technique– Trained to separate

ttbar from W+jets bkg

14

updateTop Quark Mass

Goal before Run-II started : 3 GeV/c2 with 2 fb-1

0.9% precision!

15

Two Breakthroughs in Top Mass Measurements

Matrix Element Method• Use information on

leptons and jets maximally

in-situ Wjj JES calib.• JES (jet energy scale) calibration

using di-jet invariant mass from W• Incorporate JES into likelihood

function– L(Mt)→L(Mt,JES)

– Turn JES systematics into statistics Reduced with statistics of top events

Observables

Matrix element with a given top mass gives p.d.f. of observables

Likelihood function of top mass for a given set of observables

M ｊｊ→ MW

16

update

Top Mass Measurement in Lepton + Jet channel

World Best. Better than 1%

17

update

Measurement of dσ/dMtt in Lepton + Jet channel

Response matrix

κ/MPl > 0.16 (95% CL)

Unfolding

Reconstructed Mtt

Unfolded dist.

True Mtt vs Recons. Mtt

18

Forward-Backward Asymmetry in ttbar production

PositiveAsymmetry

NegativeAsymmetry

LO: Afb=0NLO: Afb=(5 1.5)%

19

update

Forward-Backward Asymmetry in ttbar production

unfolding

~2 away from NLO prediction~2.8 away from null asymmetry

20

B Physics

21

newY(4140) in B+ → J/ψφK+

• B→J/ψφK is OZI suppressed

Yield =145

m = 1046.3 2.9 (stat) MeV/c2

Width = 11.7 +8.3-5.0 (stat) MeV

Mass = 4143.0 2.9 (stat) 1.2 (syst) MeV/c2 (adding J/ψ mass)

Width = 11.7 +8.3-5.0 (stat) 3.7(syst) MeV

JPC=??+ tentatively name it as Y(4140)

Significance: 3.8σ

22

SM Higgs Search

23

Low mass Higgs region:m

H<135 GeV/c2

H → bb dominant decay. Search for associated W/Z production.

High mass Higgs region:m

H>135 GeV/c2

H → WW dominant decay. Gluon fusion production search (gg → H).

Higgs cross-section and BR

24

updateH WW l+ l-

• 2 opposite sign leptons + MET• WW from spin 0 higgs

– leptons tend to same direction– ∆φll is best discriminant

• Neural network technique– ∆φll, ∆Rll, HT, ...

25

updateH WW, WH WWW*

• Also search for same sign leptons for WH WWW* ll X – WH→WWW (2.7fb-1): Next talk ( 大阪市大　脇坂氏 )

• Combine OS+0j, OS+1j, OS+2j, and SS+1j• 95% CL Limit/SM: 1.33 (1.45 expected)@ MH=165 GeV/c2

26

b

b

l

W* H

W

WHlbb

• Neutral network (NN) technique

– Limit: < 5.2 x SM @MH=115GeV/c2

• Matrix element (ME) + Boosted decision tree (BDT) technique

– Limit: < 5.24 x SM @MH=115GeV/c2

Bkg: W+bb, W+cc, W+qq, ttbar, ...

NN, 1 b-tag

永井氏（筑波）

27

updateWHlbb (Combined)Updated on Nov 7, 2008• Combined result of NN and ME+BDT• Improved by 15%

• 95% CL Limit/SM: 5.6 (4.8 expected) @MH=115 GeV/c2

28

bb

/l

W*/Z* H

W/Z

Other Low Mass Higgs Searches

bb

l

Z* H

Z

l

• 2lepton+2b-jets• Limit/SM: 7.1

@MH=115 GeV/c2

• MET + 2b jets• Limit/SM: 5.6

@MH=115 GeV/c2

had+had+ lep+2 jetslep+2 jets• Limit/SM: 25

@MH=115 GeV/c2

寄田氏（早稲田）

29

CDF Combined Results

WHlbb

WHWWW

Updated on Mar 5, 2009

• 95% CL Limit/SM: 3.64 (3.22 expected) @ MH=115 GeV/c2

• 95% CL Limit/SM: 1.46 (1.61 expected) @ MH=165 GeV/c2

ZHllbb

HWW

VHbbMET

Hττ

update

30

Higgs Exclusion @ Tevatron

CDF+DZero combination Updated on Mar 5, 2009update

31

Projections

• CDF+DZero projections assuming they perform the same (2xCDF)• Possible improvements: (1.5~1.6x sensitivity)

– Acceptance, Analysis method, jet/MET resolutions, ...• Good possibility to see 3 evidence to a Higgs of 165 GeV/c2

– Doesn’t take data already analyzed into account.• Good possibility to see 2 excess to low mass higgs or to exclude

at 95% CL if there is no higgs and with improvements.

32

Constraining the Higgs boson mass

33

Summary

• Tevatron is operating well. Better than ever!• CDF is ready to run FY2010. ∫Ldt > 9 fb-1 is expected.

• CDF is still providing constantly update results as well as new topics.– Top, Bottom, EWK, QCD, Higgs search, BSM search, ...

• CDF+DØ: Higgs exclusion: 160-170GeV/c2 (95%CL)– Indirect measurement:

• このあと CDF 実験のトークを引き続きお楽しみください．– Wh→WWW ： 脇坂氏（大阪市大）– Single top search: 中村氏 ( 筑波大 )

34

Backup

35

Higgs Projections (low mass)

36

Higgs Projections (High mass)

37

S/B

52/48

Var1 > x1 Var1 ≤ x1

Var2 > x2 Var2 ≤ x2

S/B

4/37

S/B

48/11

S/B

2/10

S/B

46/1

What is a Boosted Decision Tree ?

• Weight Wi for each event(= 1 to start)• Purity: P=∑Ws/(∑Ws+ ∑Wb)• Split until minimal #event reached or limit

of purity reachedif P>Pmin “signal leaf”if P<Pmin “background leaf”

• Boosting– if a signal event is on a bkg leaf or if a bkg

event is on a signal leaf weight modified

• Then training re-performed with the new weights

38

39

SM Higgs cross-section

Summer 2008