Search for MSSM neutral Higgs h 0A boson
Search for MSSM neutral Higgs, h 0/A, boson decaying in muon pair in mass region < 130 Ge. V: Methods & Sample Halina Bilokon, Vitaliano Chiarella, Simonetta Gentile, Giovanni Nicoletti Laboratori Nazionali di Frascati, Università di Roma La Sapienza, INFN Halina Bilokon, Vitaliano Chiarella, Simonetta Gentile, Giovanni Nicoletti Higgs Meeting 14 December 2005, CERN.
Outline Ø Motivation bb h 0/A m+ m- studies (25/05/2005) S. G. presentation, ATLAS-PHYS-2003 -015, June 2003. Methods Ø Software & Generators Ø • Simulation, Reconstruction, Analysis. Ø Sample • Signal, Background Ø Reconstraction features • muon reconstruction • b-tag Halina Bilokon, Vitaliano Chiarella, Simonetta Gentile, Giovanni Nicoletti Higgs Meeting 14 December 2005, CERN.
MSSM Constrained: Two Higgs doublets →h 0, H, A, H+, H Two neutral Higgs bosons A, CP-odd, and h 0 the lightest CP-even. Large loop corrections depend on SUSY parameters: Msusy, M 2, m, tan b, A, mgluino A= 61/2 Msusy , |m|<< Msusy for maximal top mixing Halina Bilokon, Vitaliano Chiarella, Simonetta Gentile, Giovanni Nicoletti Higgs Meeting 14 December 2005, CERN.
Mh max scenario It is designed to yield the maximum value of mh 0 can reach 135 Ge. V A negative search of h 0 boson in such scenario implies the exclusion of the model • Msusy =1000 Ge. V • m = -200 • M 2 =200 Ge. V • Xt =At- m cotb =2 Msusy • Mgluino= 0. 8 Msusy • MA =4 -1000 Ge. V • tanb =0. 4 -50 Pythia jargon: Top trilinear coupling RMSS(16)=2440. Bottom & tau trilinear coupling default value Halina Bilokon, Vitaliano Chiarella, Simonetta Gentile, Giovanni Nicoletti Higgs Meeting 14 December 2005, CERN.
Motivation The h 0 Supersymmetric Higgs Boson discovery in the channel bb h 0/A m+ m- in the region of high tan b ~ 15 -50 and mh 0 ~ 100 -130 Ge. V. In a large region of interesting parameter space A and h 0 are indistinguishable m. A –mh 0 ~ 0. 1 - 3 Ge. V ( mass range 95 -130 Ge. V) GA- Gh 0 ~ 0. 1 - 2 Ge. V (mass range 95 -130 Ge. V) s. A ~ sh 0 0. 1 - 0. 2 pb ØSearch for supersymmetric neutral Higgs h in the decay m+ m- in ATLAS detector at LHC. Simonetta Gentile, Mercedes Paniccia, Paolo Violini, ATLAS-PHYS-2003015, June 2003. Halina Bilokon, Vitaliano Chiarella, Simonetta Gentile, Giovanni Nicoletti Higgs Meeting 14 December 2005, CERN.
MA and Mh below 130 Ge. V Pythia 6. 226 The mass MA , Mh boson are very close The width GA Gh The cross section: s. A sh Halina Bilokon, Vitaliano Chiarella, Simonetta Gentile, Giovanni Nicoletti Higgs Meeting 14 December 2005, CERN.
The width: Gh, GA Pythia 6. 226 Halina Bilokon, Vitaliano Chiarella, Simonetta Gentile, Giovanni Nicoletti Higgs Meeting 14 December 2005, CERN.
The cross sections: sh s. A Pythia 6. 226 Halina Bilokon, Vitaliano Chiarella, Simonetta Gentile, Giovanni Nicoletti Higgs Meeting 14 December 2005, CERN.
Signal & background proces § h 0 → m+ m- and bb A → m+ m- and bb Signal Pythia 6. 226 s~ 0. 2 - 0. 01 pb (even lower) Z/g*→ m+ m- and bb s ~ 22. 8 pb (Z/g*→ e+ e- and bb for bg subtraction) tt → W+ W- bb → bb mn mn s ~ 5. 70 pb m+ s*br( Z → and bb) (Pythia 6. 226) Acer. MC (v. 2. 3) interfaced with Pythia 6. 2 (hep/ph 0405247). § § m- s(tt) *br(t→ b. W)*br(W → mn)*br(t → b. W)* br(W → mn) Missing energy in the event § ZZ →m+ m- and bb : s*br( Z → m+ m-) s*br( Z→ bb) Same order of magnitude of signal. Reduced by kinematical cuts. s ~ 0. 13 pb Halina Bilokon, Vitaliano Chiarella, Simonetta Gentile, Giovanni Nicoletti Higgs Meeting 14 December 2005, CERN.
Signal & background h 0 bb→ m- m+bb Zbb → m-m+ bb tt missing energy Halina Bilokon, Vitaliano Chiarella, Simonetta Gentile, Giovanni Nicoletti Higgs Meeting 14 December 2005, CERN. Final state 2 b-jets and 2 m ( or 3 -4)
h 0 production Halina Bilokon, Vitaliano Chiarella, Simonetta Gentile, Giovanni Nicoletti Higgs Meeting 14 December 2005, CERN.
Background Subtraction Method Precise Knowledge of background is crucial Experimental method proposed based on Z→ m+ m- and Z→ e+ e Relying on experimental data Br( h 0→ e+ e-) neglegible Different Inner Bremhstrahlung Halina Bilokon, Vitaliano Chiarella, Simonetta Gentile, Giovanni Nicoletti Higgs Meeting 14 December 2005, CERN.
Data taking scenario: Sample • Signal h 0 →m+ m- and bb : 8 points tan b[15, 50] 13 point of mass: m. A [95, 130] • Signal A → m+ m- and bb : 8 points tan b[15, 50] 13 point of mass: m. A [95, 127] • bb. Z → bbm+ m. New • tt → W+W-bb →bb mnmn • ZZ → m+ m- bb Test point: tanb =45 , mh 0 = 109. 25 Ge. V, Gh 0 = 3. 31 Ge. V, sh 0 = 0. 20 pb tanb =45 , m. A = 110. 30 Ge. V , GA = 4. 28 Ge. V, s. A = 0. 23 pb Halina Bilokon, Vitaliano Chiarella, Simonetta Gentile, Giovanni Nicoletti Higgs Meeting 14 December 2005, CERN.
Software Release & Sample Generation Athena 9. 0. 4 - Pythia (v. 6. 226) - Acer (v. 2. 3) Simulation, Digitalization, Reconstruction Athena 10. 0. 1 Analysis based on CBNT (root –ntuples). Ø Final study in Lint =30 fb-1 scenario. Ø Aim : signal MC x Lint x 10 : background MC x Lint x (5 -10). Ø Sample of used events in this presentation Ø Technical problem: since 11 October most of dig/recostruction jobs fail for Database problem. Ø Halina Bilokon, Vitaliano Chiarella, Simonetta Gentile, Giovanni Nicoletti Higgs Meeting 14 December 2005, CERN.
Software Release & Sample process L=30 fb-1 gen h 0 up to ~ Up to 15 103 each ~ 150 103 each point up to ~ 15 103 each Up to ~ 15 103 each point Zbb 680000 ttbb ZZ A simul recon This presentation Test point ~800. 000 ~100. 000 6000 Test point ~800. 000 ~100. 000 7000 3. 4 106 ~3. 4106 340. 000 170000 1. 8 106 1. 3 106 ~ 1. 0 106 170000 3800 1. 0 105 ~ 1. 0 105 3800 75000 ~ 50000 Zee Halina Bilokon, Vitaliano Chiarella, Simonetta Gentile, Giovanni Nicoletti Higgs Meeting 14 December 2005, CERN. back ground studies
Reconstruction performances The two critical point of this analysis are: Muon reconstruction: efficiency & mass resolution ( from a student work Enrico Pomarico) b-tag They deserve a preliminary study. Halina Bilokon, Vitaliano Chiarella, Simonetta Gentile, Giovanni Nicoletti Higgs Meeting 14 December 2005, CERN.
Combined muon reconstruction p. T f Halina Bilokon, Vitaliano Chiarella, Simonetta Gentile, Giovanni Nicoletti Higgs Meeting 14 December 2005, CERN.
Mass resolution combined reconstructed muons ATHENA (vs. 10. 0. 1 ) Gaussian fit: Mrec - Mgen= - 0. 82 0. 0 ± 3 Ge. V s. RES = 2. 35 0. 0 ± 3 Ge. V Author: Enrico Pomarico Halina Bilokon, Vitaliano Chiarella, Simonetta Gentile, Giovanni Nicoletti bachelor student Higgs Meeting 14 December 2005, CERN.
b-tagging The b-jet tag is an crucial part of this work few algorithms are available in ATLAS 3 D based on transverse (2 D) and longitudinal (+1 D) impact parameter SV 2 3 D impact parameter + information on secondary vertex (as energy, fraction tracks, mass) • Efficiency of b-jet tagging εb • Rejection of light jets (udsc) Rj Thanks to Vadim Kostyukhin Rozanov Halina Bilokon, Vitaliano Chiarella, Simonetta Gentile, Giovanni Nicoletti Higgs Meeting 14 December 2005, CERN.
b-tag on bbh 0 Pt>15 Ge. V |η| < 2. 5 SV 2 • Efficiency of b-jet tagging εb: 49% SV 2 , 55% 3 D 3 D Weight cut 1 Halina Bilokon, Vitaliano Chiarella, Simonetta Gentile, Giovanni Nicoletti Higgs Meeting 14 December 2005, CERN. • Rejection of light jets (udsc) Rj : 71 SV 2, 27 3 D
b-tag on bbh 0 Pt>15 Ge. V |η| < 2. 5 3 D SV 2 3 D • Efficiency of b-jet tagging εb: 50% SV 2 , 54% 3 D Weight cut 1 Halina Bilokon, Vitaliano Chiarella, Simonetta Gentile, Giovanni Nicoletti Higgs Meeting 14 December 2005, CERN. • Rejection of light jets (udsc) Rj : 58 SV 2, 26 3 D
btag -performance • R =0. 7 weight =1 Algorithm eb Rj SV 2 49% 71 3 D 55% 27 • R =0. 4 weight =1 Alghoritm eb Rj SV 2 50% 58 3 D 54% 26 • R =0. 7 weight =2 • R =0. 4 weight =2 Alghoritm eb Rj Alghoritm SV 2 46% 219 eb Rj SV 2 46% 200 3 D 49% 52 3 D 49% 50 Halina Bilokon, Vitaliano Chiarella, Simonetta Gentile, Giovanni Nicoletti Higgs Meeting 14 December 2005, CERN.
b-tag • No major improvement in changing cone width ( R=0. 7, R=0. 4); natural choice R =0. 7 (default value ATHENA (vs. 10. 0. 1 ). • A study vs. weight cut w = 2 brings to better Rj for light quarks at price of b-tag efficiency. • The study has been done on h 0 bb sample and Zbb with similar conclusions. • Warning: these efficiency have been evaluated “à la mode “of b-tagging group, for comparison performance no b-quark in a cone R=0. 6 around light quark jet p. T>15 Ge. V |h|< 2. 5 , at least, 1 track in jet • SV 2 algorithm is more promising: good b tag efficiency: eb, and better rejection factor for light quarks : Rj Provisional choice SV 2 alghoritm (cut weight 1) Halina Bilokon, Vitaliano Chiarella, Simonetta Gentile, Giovanni Nicoletti Higgs Meeting 14 December 2005, CERN.
b-tagging at different P WH(400), 2 D-alg, 60% eff. Vadim Kostyukhin 25/10/2005 ATL-PHYS-2004 -006, “b-tagging with DC 1 data” “Bad” regions for b-tagging: 1. 2. 3. P <40 Ge. V P >250 Ge. V | |>2 Halina Bilokon, Vitaliano Chiarella, Simonetta Gentile, Giovanni Nicoletti Higgs Meeting 14 December 2005, CERN.
Example: h 0 bb system The h 0 Supersymmetric Higgs Boson discovery in the channel bb h 0 m+ m- in the region of high tan b ~ 20 and mh 0 ~ 110 Ge. V. P of labeled b-jets Vadim Kostyukhin 25/10/2005 P of labeled u, d, s-jets Me. V Halina Bilokon, Vitaliano Chiarella, Simonetta Gentile, Giovanni Nicoletti Higgs Meeting 14 December 2005, CERN. Me. V
h bb system Vadim Kostyukhin 0 Number of “good” tracks in jet 25/10/2005 b-jet u, d, s-jet P >60 Ge. V P <20 Ge. V Halina Bilokon, Vitaliano Chiarella, Simonetta Gentile, Giovanni Nicoletti Higgs Meeting 14 December 2005, CERN.
h 0 bb system light jet rejection Jet-bquark distance, jets marked as u, d-jets, P <30 Ge. V Still problem with labeling in CALO jets reconstruction. Safe jet-bquark distance for labeling is 0. 5!!! Default in Rec. Ex. Common will be changed. Rejection Ntrack_in_jet>0, 10<P <20 Ge. V, Wgt. SV 2>2, Bq. Dst>0. 5 180 (210 with Ntr=0 jets) Vadim Kostyukhin 25/10/2005 to limit occasional coincidence between light jet and b-quark directions. Halina Bilokon, Vitaliano Chiarella, Simonetta Gentile, Giovanni Nicoletti Higgs Meeting 14 December 2005, CERN.
Further study on b-tag 1. 2. 3. 4. B-tagging performance (contrary to TDR expectation 100 at 60% everywhere) is NOT UNIFORM. Regions with very low performance are: 1. P >250 Ge. V 2. P <40 Ge. V 3. | |>2 One should be careful when applying the “real” b-tagging because b-tagging efficiency curve modifies the P and distributions of jets and thus changes the kinematical properties of events. For the moment there is no any easy way to recuperate efficiency of btagging in “bad” regions. More work is needed. Vadim Kostyukhin 25/10/2005 Halina Bilokon, Vitaliano Chiarella, Simonetta Gentile, Giovanni Nicoletti Higgs Meeting 14 December 2005, CERN.
Conclusions MSSM h 0/A Higgs bosons have been generated, simulated and recontructed in data taking scenario. Ten times the statistics is aimed for 8 points tan b[15, 50] 13 point of mass: m. A [95, 130 ] Muon recontruction efficiency & mass resolution at Z peak have been studied Preliminary b-tag studies have been performed • Efficiency of b-jet tagging εb: 49% SV 2 algorithm Weight cut 1 • Rejection of light jets (udsc) Rj : 71 SV 2 algorithm Halina Bilokon, Vitaliano Chiarella, Simonetta Gentile, Giovanni Nicoletti Higgs Meeting 14 December 2005, CERN.
- Slides: 29