Determination of tan β at a Future e + e LC. Jack Gunion 1 Tao Han 2 Jing Jiang 3 Steve Mrenna 4 André Sopczak 5

Transcription

1 SUSY02 at DESY, June 2002 Determination of tan β at a Future e + e LC Jack Gunion 1 Tao Han 2 Jing Jiang 3 Steve Mrenna 4 André Sopczak U.C. Davis, 2...University of Wisconsin, 3...Argonne N.Lab 4...Fermilab, 5...Lancaster University

2 Outline Introduction bba bbbb simulation High luminosity: 2000 fb 1 HA bbbb event rate H and A width from HA bbbb New aspect: H + H t b tb Conclusions

3 Introduction Framework: Two-Higgs Doublet Model or MSSM. Production and Decay of A, h, H, and H ±. Considered reactions for TESLA: strong dependence on tan β. Extrapolation bba bbbb for 100 <m A < 200 GeV. Estimate of HA bbbb event rate. Estimate of H and A width determination. Estimate of charged Higgs bosons branching ratio and decay width.

4 b-tagging Experimental potential depends strongly on the b-tagging performance. Hadronic events e + e q q (5 flavors). b-tag efficiency Purity Efficiency: Ratio of simulated bb events after the selection to all simulated bb events. Purity: Ratio of simulated bb events after the selection to all selected qq events.

5 bba Simulation e + γ e Z b A b Simulated Higgs boson mass: 100 GeV Channel bba qq WW ewν tt ZZ eez ha sum (in 1000) After Presel. 73% Simulated ha rate corresponds to twice the luminosity (maximum cross section in general Two-Higgs Doublet Model).

6 Signal and Background Events/ 500 fb signal+bg total bg tt(nγ) ZZ qq(nγ) ha WW 1 Zee Efficiency

7 Interference: bba ha b bb b Expectation before event selection: σ bba σ(e + e bba b bb b) σ ha σ(e + e ha b bb b) σ bba+ha σ(e + e bba, ha b bb b) σ interf = σ bba+ha σ bba σ ha. For m b =4.62 GeV: σ bba =1.83 ± 0.01 fb σ ha =36.85 ± 0.10 fb σ bba+ha =39.23 ± 0.12 fb σ interf =0.55 ± 0.16 fb Positive interference, reduction in statistical error.

8 Interference: bba ha b bb b Expectation after event selection: 100 bba b bb b events 2 ± 1hA b bb b events. Maximum interference magnitude: ( ) Similar ratio of interference to signal 30% before and after event selection. Interference events: background-like: small systematic error. signal-like: large systematic error. Solution: fit signal and background to data for various tan β. Another systematic error is the running b-mass. Higher-order corrections should be very precisely known by the time the LC is constructed.

9 bba Results for 500 fb 1 For tan β = 50 and m A = 100 GeV: tan β/tan β =0.07. tan 2 β/tan 2 β = N signal /N signal = N signal + N background /N signal =0.14. Smaller values of tan β, the sensitivity decreases rapidly. 5σ signal detection for tan β = 35. MSSM: bbh would double the number of signal events and have the same tan β dependence: tan 2 β/tan 2 β 300/ For tan β = 50 and m A = m h = 100 GeV: tan β/tan β =0.04. (For heavier A, bbh will contribute). Experimental challenge: at 10% efficiency, tan β/tan β<0.05 requires ɛ/ɛ < 0.1, thus ɛ <1%.

10 e + e b b b ba Rate Events s = 500 GeV L = 500 fb -1 tanβ = Higgs mass (GeV/c 2 ) At 100 GeV: 1000 events. At 200 GeV: 200 events.

11 bba Results for 2000 fb 1 tanβ/tanβ 0.75 m A = 200 GeV L = 2000 fb -1 bba GeV GeV tanβ

12 HA bbbb EventRate Assume b-tagging purity 80% per bb pair, thus 40% signal efficiency. For bbbb: 16% efficiency. Further reduction: kinematic event selection: final efficiency 10%, and negligible background. Small tan β: constant MSSM cross section and large variation of branching fraction. Typical expected signal rate for 2000 fb 1 and m A = m H = 200 GeV for tan β = 5: 500 events. tan β/tan β 0.01 Nice overlapwith bba results.

13 HandAwidthfrom HA bbbb Assumption: 5 ± 0.5 GeV detector resolution. Reconstruction of mean H and A widths from bb-mass. There are two bb masses per event. Wrong jet-jet pairings: 25%. Available statistics for mass reconstruction: 1.5 HA bbbb rate. Determination of intrinsic Higgs width by convolution with detector resolution. 0.5(Γ H +Γ A )=12.5 ± 0.54 GeV, dominated by error on detector resolution. tan β/tan β<0.02 for tan β = 55.

14 e + e H + H t b tb 500 GeV and 10 fb 1, the charged Higgs can be reconstructed Z. Phys. C65(1995) GeV and 1000 fb 1, high precision reconstruction, M. Battaglia, A. Ferrari, A. Kiiskinen, T.M. Ki, hep-ph/ Sensitivity for 500 GeV and 1000 fb 1, for a 200 GeV charged Higgs boson mass: uncertainty about 0.5 GeV. Very similar uncertainty as for the HA. Similar precision on tan β as from HA production tan β/tan β 0.02 for tan β =5.

15 Further information on tan β tbh ± tbτν J. L. Feng and T. Moroi, Phys. Rev. D 56, 5962 (1997). A width from bba b bb b Scalar taus (E.Boos et.al.)

16 Conclusions Experimental challenges: High luminosity needed. Best possible b-tagging performance. Precision detector resolution measurement. Combined analyses, e.g. Fit of all bbbb processes as function of tan β. Precision determination of signal efficiency. Precision cross section and decay width meas. bba b bb b HA bbbb H + H t b tb TESLA: Precision determination of tan β with different independent complementary methods.