First measurement of forward – backward Asymmetry in bb Production at CDF on behalf of CDF collaboration P. Bartoš Comenius University
Motivation ➢ Noticeable A FB seen in tt production ➢ Deviation from the predictions is CDF, L+J analysis, parton level observed: → does it appear only at large M qq ? ➢ Measuring of bb asymmetry does not probe the large M qq → it probes almost the same SM theory computation needed for predicting A FB in tt ➢ new physics in leptophobic scenarios could appear at relatively low mass, 05/02/12 P. Bartoš 2
Source of the asymmetry ➢ SM: LO strong interaction processes Fig. 1 q b b and g g b q : no A FB b ➢ sources of A FB : radiative corrections involving either virtual or real gluon emission: ✔ Interference of (a)+(b), (c)+(d) in Fig.1 ✔ Interference of flavor excitation in quark+gluon processes (Fig.2) Fig. 2 contribution from EW production q ,Z b q b processes: ➢ No asymmetry in gluon fusion interactions 05/02/12 P. Bartoš 3
Theoretical expectations ➢ b quark production at Tevatron: main source gluon fusion (no A FB ) ➢ The expected SM global A FB from qq and qg (qg) reactions is small due to a big symmetric gluon contribution ➢ However for some selection criteria (which increase qq→bb fraction) sizable asymmetry is predicted : Rodrigo, Kűhn 1999 Phys. Rev. D 59 054017 → black dash-dotted line is differential s = 400GeV A FB for ( Θ * - b quark production angle in qq rest frame) 05/02/12 P. Bartoš 4
Events selection → dijet events with soft µ inside one of jets: ν µ µ − ✔ p T ( µ ) > 10 GeV, | η ( µ ) | < 0.6 → muon jet (MJ) : E T (MJ) > 20 GeV b → away jet (AJ): ✔ E T (AJ) > 15 GeV, | η (AJ) | < 1.0 ✔ AJ back-to-back to MJ: | ∆Φ |>2.8 ∣ p T AJ − p T MJ ∣ max{ p T AJ , p T MJ } 0.6 ✔ → both jets b-tagged by Secondary Vertex tagging algorithm 05/02/12 P. Bartoš 5
Secondary vertex b-tagging This technique takes advantage of the long lifetime of a b-hadrons => identify jets from the bottom quark hadronization through the presence of a decay vertex displaced from the primary interaction. Secondary vertex: → vertex displaced from the primary vertex → contains at least 2 tracks which passed a quality criteria 05/02/12 P. Bartoš 6
Methodology A FB = N y b 0 − N y b 0 ✔ Integrated asymmetry is defined: N y b 0 N y b 0 ( y b – rapidity of the b-quark in qq rest frame) ✔ Assuming CP conservation ( ) we can express A FB N − y b = N y b y b = y b − y using Lorenz invariant : b y b = Q ⋅ y AJ − y MJ In our case: * Q( µ ) – charge of the muon, AJ – away jet, MJ – muon jet A FB = N y b 0 − N y b 0 N y b 0 N y b 0 ✔ The study is performed as a function of M bb and | ∆ y b | 05/02/12 P. Bartoš 7
A FB corrections for non bb events → b-tagging procedure is not 100% pure → mistag (assign the non-b jet as b-jet) → rate depends on jet E T and η → correction for non bb events are needed → let's denote reconstructed asymmetry as: reco − N 2 reco reco = N y b 0 reco = N 1 reco , where N 1 A FB reco = N y b 0 reco N 2 N 1 N 2 reco b = A FB b non- b b = 0 A FB , if we assume that A FB f b b To obtain fraction of bb – need to know b fraction of muon and away jets! 05/02/12 P. Bartoš 8
b-fraction on away jet side (M bb [35, 75] GeV) → We fit the data distribution of secondary vertex mass, M vtx , by MC templates for b and non-b jets: M vtx > 1 GeV is used to b template suppress mistags filled part - used for fit Result of the fit: non-b template fill part - used for fit 05/02/12 P. Bartoš 9
b-fraction on away jet side (M bb [75, 95] GeV) → We fit the data distribution of secondary vertex mass, M vtx , by MC templates for b and non-b jets: M vtx > 1 GeV is used to b template suppress mistags filled part - used for fit Result of the fit: non-b template fill part - used for fit 05/02/12 P. Bartoš 10
b-fraction on away jet side (M bb [95, 130] GeV) → We fit the data distribution of secondary vertex mass, M vtx , by MC templates for b and non-b jets: M vtx > 1 GeV is used to b template suppress mistags filled part - used for fit Result of the fit: non-b template fill part - used for fit 05/02/12 P. Bartoš 11
b-fraction on away jet side (M bb > 130 GeV) → We fit the data distribution of secondary vertex mass, M vtx , by MC templates for b and non-b jets: M vtx > 1 GeV is used to b template suppress mistags filled part - used for fit Result of the fit: non-b template fill part - used for fit 05/02/12 P. Bartoš 12
b-fraction distributions for away jet side M bb dependence | ∆ y b | dependence We see small decreasing tendency, what is expected due to No dependence is observed the b-jet mis-tagging (as is expected) => distribution is fitted by constant function 05/02/12 P. Bartoš 13
b-fraction on muon jet side (M bb [35, 75] GeV) → We fit the data distribution of muon p T,rel by MC templates for (b→ µ ) and (b→c→ µ + c→ µ ): Result of the fit: 05/02/12 P. Bartoš 14
b-fraction on muon jet side (M bb [75, 95] GeV) → We fit the data distribution of muon p T,rel by MC templates for (b→ µ ) and (b→c→ µ + c→ µ ): Result of the fit: 05/02/12 P. Bartoš 15
b-fraction on muon jet side (M bb [95, 130] GeV) → We fit the data distribution of muon p T,rel by MC templates for (b→ µ ) and (b→c→ µ + c→ µ ): Result of the fit: 05/02/12 P. Bartoš 16
b-fraction on muon jet side (M bb > 130 GeV) → We fit the data distribution of muon p T,rel by MC templates for (b→ µ ) and (b→c→ µ + c→ µ ): Result of the fit: 05/02/12 P. Bartoš 17
Corrections of b-fraction on muon jet side p T,rel > 1 GeV is used to suppress mistags → need to obtain b-fraction for this cut: → we calculate absolute fractions of b and c for the cut: ∞ Af b = f b ⋅ ∫ b templ dp T ,rel 1 ∞ ∫ b templ dp T ,rel Af b 0 f b p T ,rel 1 = ∞ Af b Af c ⋅ ∫ c templ dp T ,rel 1 Af c = 1 − f b ∞ ∫ 0 c templ dp T ,rel → to get the true fraction of b-jets on muon jet side we also need to include the cascade decays (b→c→ µ + and b→c→ µ – ) → the fraction of cascade decays is obtained from MC and scaled to PDG2011 true MJ = f b MJ , p T ,rel 1 f b MJ , p T ,rel 1 f b 1 − f seccondaries 05/02/12 P. Bartoš 18
b-fraction distributions for muon jet side f b MJ , p T ,rel 1 M bb dependence | ∆ y b | dependence We see decreasing tendency, No dependence is observed what is expected due to the b-jet (as is expected) mis-tagging => distribution is fitted by constant function 05/02/12 P. Bartoš 19
bb fraction → we obtain as follows: 1) if all non- b on muon jet side are paired with non- b on away jet side → we can set upper limit = f b (AJ) 2) if all non- b on muon jet side are paired with b on away jet side true (MJ) ] → we can set lower limit = f b (AJ) – [ 1 - f b 3) bb-fraction is obtained as the average of the above limits * MC does not contain certain contributions existing in the data. 05/02/12 P. Bartoš 20
Fraction of secondary decays and mixing f mix – fraction of B 0 -B 0 mixing events f secOS – fraction of events with secondary decays ( b→c→ µ + ) … obtained from no-mixing events f secmix = f secOS . 1 − f mix 1 − f secOS .f mix fractions are obtained from MC => scaled to PDG 2011 values increasing tendency comes from increasing of f secOS with M bb , what is No dependence is observed caused by muon p T cut 05/02/12 P. Bartoš 21
f secmix corrections of A FB → cascade decays and B 0 -B 0 mixing change sign of the µ → ∆ y b has opposite sign => this dilute our calculations → another correction is needed → (For now) let's assume that we have only true bb events: b b = N 1 true . 1 − f secmix N 2 true .f secmix b b N 1 b = N b y b 0 N 1 where b b = N 2 true . 1 − f secmix N 1 true .f secmix b b N 2 b = N b y b 0 N 2 b b A FB reco A FB true = A FB true = A FB Total correction for 1 − 2.f secmix b . 1 − 2.f secmix f b reconstructed A FB 05/02/12 P. Bartoš 22
Expected sensitivity to bb asymmetry (I) 1 true = reco b . 1 − 2.f secmix ⋅ A FB A FB f b M bb dependence |∆ y b | dependence 05/02/12 P. Bartoš 23
Expected sensitivity to bb asymmetry (II) M dijet dependence: M bb (GeV) A FB (in %) 35 - 75 X.YZ ± 0.96 (stat) ± 0.05 (syst) 75 - 95 X.YZ ± 1.15 (stat) ± 0.11 (syst) 95 - 130 X.YZ ± 1.57 (stat) ± 0.1 (syst) X.YZ ± 2.56 (stat) ± 0.68 (syst) > 130 ∆ y b dependence: |∆ y b | A FB (in %) 0.00 - 0.50 X.YZ ± 0.82 (stat) ± 0.07 (syst) 0.50 - 1.00 X.YZ ± 0.99 (stat) ± 0.14 (syst) X.YZ ± 1.48 (stat) ± 0.12 (syst) > 1.00 A FB = X.YZ ± 0.62(stat) ± 0.10(syst) Integrated A FB : Systematics which comes from the obtaining of bb-fraction: 05/02/12 P. Bartoš 24
Recommend
More recommend
