A new observable to measure the top-quark mass at hadron colliders. Simone Alioli LBNL & UC Berkeley Seattle, 2 July 2013 EF Snowmass Meeting based on arXiv:1303.6415 SA, P . Fernandez, J. Fuster, A. Irles, S. Moch, P . Uwer, M. Vos
Motivations for precise m t measurements ◮ Fundamental parameter of SM Lagrangian ◮ The top sector might play a role in EWSB 80.60 experimental errors 68% CL: ◮ Important parameter in SM (and MSSM) LEP2/Tevatron: today LHC: future fits M h = 123 .. 127 GeV ILC/GigaZ 80.50 MSSM M W [GeV] 80.40 M H = 123 GeV SM M H = 127 GeV MSSM, M h = 123..127 GeV 80.30 SM, MSSM Heinemeyer, Hollik, Stockinger, Weiglein, Zeune ’12 168 170 172 174 176 178 m t [GeV] ◮ Crucial for vacuum (meta-)stability of SM at NNLO DeGrassi et al. ’12 Alekhin, Djouadi, Moch ’12 0.10 M h � 125 GeV 0.08 3 Σ bands in 180 M t � 173.1 � 0.7 GeV 10 7 10 10 Higgs quartic coupling Λ � Μ � 0.06 Α s � M Z � � 0.1184 � 0.0007 Pole top mass M t in GeV Instability Instability Meta � stability 0.04 175 0.02 1,2,3 Σ M t � 171.0 GeV 0.00 170 Α s � M Z � � 0.1205 10 12 Stability � 0.02 Α s � M Z � � 0.1163 M t � 175.3 GeV � 0.04 165 10 2 10 4 10 6 10 8 10 10 10 12 10 14 10 16 10 18 10 20 115 120 125 130 135 Higgs mass M h in GeV RGE scale Μ in GeV Simone Alioli | Top mass | EF Snowmass Seattle 07/02/2013 | page 2
Theoretical issues in determination of top-quark mass ◮ Confinement = free quarks not observable = no pole in the S-matrix ◮ Parameters of the theory measured through their influence on hadronic observables: fit O exp ( � x ) with O th ( m t , � x ) and extract m t Which mass are we measuring ? At least NLO required to fix the ren. scheme. ◮ Precise value depends on the m t definition: m pole , m MS , etc. t t Which scheme ? Some show better convergence ( e.g. m MS ), t some ill-defined beyond PT (IR renormalons ∆ m pole ∝ Λ QCD ) t ◮ Color reconnections Simone Alioli | Top mass | EF Snowmass Seattle 07/02/2013 | page 3
(Some) top-quark mass measurements at hadron colliders ◮ Template method - Ideogram method = m pole ✪ m MC (1 ± ∆) , ∆ =? , LO t t � high-precision ◮ Matrix element method ✪ LO only, NLO under develop. � � high-precision ◮ Cross section � theoretically clean, NLO, finite Γ t ✪ reduced sensitivity, threshold eff. included ◮ J/ψ method � NLO, small sensitivity to JES unc. and top reco. ✪ finite Γ t , very-high statistics required ◮ Dilepton-specific � NLO, JES unc., top reco., finite Γ t ✪ reduced sensitivity, high statistics required ◮ Kinematic endpoint � NLO?, small sensitivity to top reco. ✪ JES, finite Γ t ? Simone Alioli | Top mass | EF Snowmass Seattle 07/02/2013 | page 4
Current results Tevatron LHC ◮ Dedicated studies of top-quark mass wrt event kinematics show small dependence ⇒ mismodelling is small at current precision. CMS-TOP-12-029 ATLAS-PHYS-PUB-2013-005 Simone Alioli | Top mass | EF Snowmass Seattle 07/02/2013 | page 5
New ∗ proposal : top-quark mass from jet rates ∗ cfr. mb from 3-jets rate @ LEP [Bilenky et al. ’95] ◮ Study t ¯ t + 1 -jet events : large rate at the LHC ( � 30% ), NLO and NLO+PS available ◮ Experimentally accessible, errors reduced through normalization factor dσ t ¯ 1 ρ s = 2 m 0 R ( m pole t + 1-jet ( m pole m 0 = 170 GeV , ρ s ) = , ρ s ) , , √ s t ¯ t t σ t ¯ dρ s t + 1-jet tj ◮ Theoretically well defined, calculable at NLO, small uncertainties and small NP corrections, R ( m MS , ρ s ) also possible. Low ρ S control region s t ¯ tj reco. t 3.5 NLO 3 LO 2.5 ) s ρ , pole 2 t (m 1.5 R 1 0.5 0 1.5 ratio 1 0.5 0.2 0.3 0.4 0.5 0.6 0.7 0.8 ρ s Simone Alioli | Top mass | EF Snowmass Seattle 07/02/2013 | page 6
Sensitivity on m t |R ( m pole , ρ s ) − R ( m pole + η ∆ m pole , ρ s ) | ◮ Linear approximation S ( ρ s ) = � t t t 2 | ∆ |R ( m pole , ρ s ) η = ± 1 t t t +1Jet pole 25.5 0.15 ∆ m = 10 GeV t pole m = 5 GeV ∆ t t t ] ) -1 ρ [GeV pole ( ∆ m = 10 GeV 0.1 17 t S pole × m = 5 GeV ∆ ) t pole ρ ( t m � ∆ m pole � S � � � ≈ m pole � � ∆ R � � � S × t t � � R m pole 0.05 8.5 � � t 0 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 ρ ≈ − 5∆ m pole ◮ Up to five times more sensitive than total xsec, ∆ σ t σ m pole t Simone Alioli | Top mass | EF Snowmass Seattle 07/02/2013 | page 7
Experimental viability study ◮ Event selection (lepton+jets): 1. one lepton ( ℓ = e, µ ) with p T > 25 GeV and | η | < 2.5; 2. missing E T > 30 GeV 3. M W > 35 GeV T 4. ≥ 3 jets within | η | < 2 . 5 , hardest with p T > 50 GeV, other two p T > 25 GeV; 5. two additional identified b -jets 6. two light jets inv. mass compatible with m W within 20% 7. two reconstructed top-jet system masses within 20% ◮ Background contamination kept at the 5-10% level : : QCD (1,2) , single top and W + jets (3,4,5) ◮ Preliminary study with no detector-specific tools. Room for improvement when done by experimental collaborations. ◮ Dilepton channel also possible, but reduced statistics. Simone Alioli | Top mass | EF Snowmass Seattle 07/02/2013 | page 8
Theoretical uncertainties ∆ R µ / R ( m pole ∆ R PDF / R ( m pole ,ρ s ) ,ρ s ) ◮ Scale and PDF uncertainties: t , t S ( ρ s ) S ( ρ s ) ◮ Scale unc. ∆ m pole PDF unc. ∆ m pole ≈ 0 . 2 GeV ≈ 0 . 5 GeV t t Simone Alioli | Top mass | EF Snowmass Seattle 07/02/2013 | page 9
Theoretical uncertainties ◮ Impact of higher-orders and parton showers: calculate R NLO and extract top-quark mass that would fit the distribution from generated events 180 POWHEG t t at NLO + Pythia8 POWHEG t t +1Jet at NLO + Pythia8 175 t t +1Jet NLO (Pert. Calc.) [GeV] 170 pole t m 165 160 0.2 0.3 0.4 0.5 0.6 0.7 0.8 ρ s ◮ POWHEG+Pythia vs. MC@NLO+Herwig gives ∆ m pole ≈ 0 . 2 GeV t Simone Alioli | Top mass | EF Snowmass Seattle 07/02/2013 | page 10
Theoretical uncertainties ◮ Colour reconnection effects: different CR models in Pythia6 vs. Pythia8 ◮ Switching CR on/off very conservative estimate: ∆ m pole ≤ 0 . 4 GeV t Simone Alioli | Top mass | EF Snowmass Seattle 07/02/2013 | page 11
Experimental uncertainties ◮ Jet Energy Scale uncertainty ± 3% results in ∆ m pole ≈ 0 . 8 − 1 . 0 GeV t 3.5 pole pole m =170 GeV Nominal m =160 GeV t t 3 pole pole m =170 GeV JES 3% ± m =180 GeV t t 2.5 ) s ρ , pole 2 t (m 1.5 R 1 0.5 0 3 ratio 2 1 0 0.2 0.3 0.4 0.5 0.6 0.7 0.8 ρ s Simone Alioli | Top mass | EF Snowmass Seattle 07/02/2013 | page 12
Experimental uncertainties ◮ Mass independent unfolding: associated unc. ∆ m pole ≈ 0 . 3 GeV (stat.) t 3.5 pole m = 160 GeV 3 t True Unfolded 2.5 ) s ρ , pole 2 t (m 1.5 R 1 0.5 0 1.05 ratio 1 0.95 0.2 0.3 0.4 0.5 0.6 0.7 0.8 ρ s ◮ Assuming final efficiency ≈ 1% and 5 fb − 1 collected luminosity, expected error is ≈ 1 . 4 GeV stat. in the ρ s > 0 . 65 bin. ◮ Extrapolated to 20 fb − 1 ∆ m pole ≈ 0 . 7 GeV t Simone Alioli | Top mass | EF Snowmass Seattle 07/02/2013 | page 13
Conclusions ◮ Top-quark physics is precision physics at the LHC ◮ Current precision O (1 GeV ) already impressive. ◮ Theoretical interpretation not so well under control. ◮ At least NLO needed to fix renormalization scheme. ◮ Several methods availables for NLO top mass, important to take advantage of all of them. ◮ Observable proposed here complements existing approaches. NLO top-quark mass definition, theoretical and experimental uncertainties evaluated at O (1 GeV ) or below Outlook: ◮ Analysis presented here being performed by ATLAS group in Valencia. ◮ LHC upgrade top-factory: 300 fb − 1 at 13 TeV will produce ≈ 50M ttbar events in the lepton+jet channel, 10M events in the dilepton channel, 15M single tops ◮ Extreme precision in e + e − : threshold scans at LC will reach O (0 . 1 GeV ) with dedicated run, theoretically very clean (N 3 LO and NNLL). Thank you for your attention! Simone Alioli | Top mass | EF Snowmass Seattle 07/02/2013 | page 14
BACKUP Simone Alioli | Top mass | EF Snowmass Seattle 07/02/2013 | page 15
Scale and PDF variations 3.5 CTEQ6.6 3 MSTW2008nlo90cl 2.5 ) s ρ , pole 2 t (m 1.5 R 1 0.5 0 1.5 ratio 1 0.5 0.2 0.3 0.4 0.5 0.6 0.7 0.8 ρ s Simone Alioli | Top mass | EF Snowmass Seattle 07/02/2013 | page 16
Impact of NLO fixed-order corrections 180 LO NLO 175 [GeV] 170 pole t m 165 160 0.2 0.3 0.4 0.5 0.6 0.7 0.8 ρ s Simone Alioli | Top mass | EF Snowmass Seattle 07/02/2013 | page 17
Recommend
More recommend
