On the Top Mass Reconstruction Using Leptons

I discuss the feasibility of measuring the top quark mass by the using of final states with leptons and J/ψ at hadron colliders. I also investigate the impact of matrix-element corrections to the HERWIG simulation of top decays. Top quark physics is presently one of the most interesting fields of investigation in both theoretical and experimental analyses. In the next Tevatron RUN II and, ultimately, at the LHC, a large amount of tt̄ pairs will be produced, which will allow precision measurements of the top quark properties and especially of its mass. In tt̄ events at hadron colliders, the final states are classified according to the decay modes of the two W bosons produced in the decays t → bW . At the Tevatron RUN I, the top mass was determined by means of the lepton + jets and the dilepton modes by both CDF and DØ collaborations, CDF also considering the all-hadron mode. The estimated average value for the top mass reads [1]: mt = 173.8± 3.5± 3.9 GeV. In this talk I discuss the method of reconstructing the top mass by means of final states where the two produced W bosons decay leptonically, i.e. W → lν, l being either an electron or a muon, and the B meson, coming from the hadronization of the b quark, decays into a state containing a J/ψ, eventually decaying into a μμ pair. According to the LHC experimentalists [2], this is is a favourite channel, with the estimated systematic error being no larger than 1 GeV. In three years of high luminosity L = 3× 10 pb, about 3× 10 final states with well-identified leptons and J/ψs are foreseen. The idea is that one should relate either the invariant mass mJ/ψl or mμl to the top mass by the using of Monte Carlo simulations. The suggested channel presents some advantages which make it quite suitable: the backgrounds are small and can be suppressed by setting cuts on the transverse momentum and rapidity of the final-state leptons; the effects of the initialand finalstate radiation are negligible and can be easily taken into account; the spectra W → lν and B → J/ψ are well known. ‡ Talk given at the UK Phenomenology Workshop on Collider Physics, Durham, U. K., 19-24 September 1999. On the Top Mass Reconstruction Using Leptons 2 Figure 1. Invariant mass of the B-lepton system at the Tevatron (a) and at the LHC (b) for mt = 175 GeV, according to HERWIG 6.0 (dotted) and 6.1 (solid). Since the proposed method crucially relies on the Monte Carlo model used to simulate top production and decay, I wish to investigate the effect of matrix-element corrections to top decays, recently implemented [3] in the HERWIG algorithm [4]. Being the B → J/ψ spectrum well known, I shall consider the spectra of mBl. In HERWIG 5.9, the latest public version, a few bugs were found for the purpose of top decays. These bugs are corrected in the intermediate version 6.0, HERWIG 6.1 being the new version, including also the improved treatment of top decays. The Tevatron statistics will be too low to detect J/ψs via top quarks, nevertheless it is worthwhile performing the analysis even at the Tevatron in order to check the consistency of the method which claims to estimate the top mass just via its decay, independently of the production mechanism, which is mainly qq̄ → tt̄ at the Tevatron and gg → tt̄ at the LHC. In Fig. 1 the invariant mass spectra given by HERWIG 6.0 and 6.1 are plotted at both the machines for mt = 175 GeV; in tables 1 and 2 one can find the average values < mBl > and the statistical errors on the obtained discrepancies for different values of the top mass. We observe a systematic shift of about 800 MeV towards lower values of < mBl > after the inclusion of matrix-element corrections. Also, the results at the Tevatron and at the LHC are the same just within the range of 100 MeV. If we try to fit the relation between < mBl > and mt into a straight line, by means of the least square method, we shall find for the LHC: 6.1 : < mBl > = 0.459 mt + 8.683 GeV , ǫ = 3.41× 10 −3 ; (1) 6.0 : < mBl > = 0.448 mt + 11.262 GeV , ǫ = 7.03× 10 −4 ; (2) where ǫ is the mean square deviation. The straight lines are plotted in Fig. 2: from the values of the slopes, it follows that the discrepancies on < mBl > result in an impact of about ∆mt ≈ 1.8 GeV on the top mass. More details on this analysis will be found in [5], where also jet distributions in tt̄ events will be considered. On the Top Mass Reconstruction Using Leptons 3 mt < mBl > 6.1 < mBl > 6.0 < mBl > 6.0 − < mBl > 6.1 171 GeV 87.14 GeV 87.98 GeV (0.872± 0.042) GeV 173 GeV 88.17 GeV 88.86 GeV (0.732± 0.040) GeV 175 GeV 88.94 GeV 89.84 GeV (0.930± 0.038) GeV 177 GeV 89.92 GeV 90.66 GeV (0.775± 0.043) GeV 179 GeV 90.71 GeV 91.53 GeV (0.874± 0.044) GeV Table 1. Results at the Tevatron for different values of mt. mt < mBl > 6.1 < mBl > 6.0 < mBl > 6.0 − < mBl > 6.1 171 GeV 87.11 GeV 87.94 GeV (0.859± 0.042) GeV 173 GeV 88.14 GeV 88.85 GeV (0.745± 0.041) GeV 175 GeV 89.06 GeV 89.79 GeV (0.769± 0.043) GeV 177 GeV 89.92 GeV 90.64 GeV (0.764± 0.043) GeV 179 GeV 90.81 GeV 91.53 GeV (0.771± 0.044) GeV Table 2. As in table 1, but for the LHC. Figure 2. < mBl > as a function of mt at the LHC after a fit into a straight line, according to HERWIG 6.1 (solid line) and 6.0 (dotted).