Experience of EW and BSM physics at HERA and lessons for EIC Elisabetta Gallo (DESY and UHH) (A personal recollection and selection of topics) Electroweak and BSM physics at the EIC, 6/5/2020 Elisabetta Gallo (DESY) 1 HERA (1992-2007) First F2 presented by H1 at Diffractive events Durham 1993 discovered by ZEUS, DESY seminar 1993 • HERA experiments were built for high Q2 physics, F from H1 but adapted very well to lower Q2 low x L • With HERA II a stronger electroweak program coulD be starteD anD searches boosteD • At the enD of HERA II we went back to low-x physics with FL Elisabetta Gallo (DESY) 2 Increasing the luminosity (i.e. HERA II) • Access to higher Q2 • Access to valence quark distribution at higher x in NC, xF3 • Electron vs positron running (endless discussions) • More precise measurements of charged current • Polarized beams • Combination H1-ZEUS data (it started for xF3) • LHC was getting closer -> PDFs for LHC, more exchange, HERAPDFs • More exotic searches • Isolated leptons plus missing energy and large hadronic p Triggered by observation in T H1, no specific- related search • Multilepton events • Leptoquarks • FCNC • Contact interactions Elisabetta Gallo (DESY) 3 NC/CC at high Q2 Elisabetta Gallo (DESY) 4 Few words about ZEUS and H1 combinations • Unless specified the plots I am going to show are based on the latest combined data • And compared to HERAPDF2.0: based on data taken 1994-2007, 2927 points combined to 1307, 21 HERA I data samples, 20 HERA II data samples NC+CC EPJ C75 (2015) 580 • xFitter, open-source tool originally started • NC: 0.045< Q2< 50000 to fit PDFs at HERA GeV2 -7 ZEUS and H1 combination • Now used also for LHC 6 X 10 < x < 0.65 • CC: 200< Q2< 50000 is the main dataset for experiments (EIC?) PDFs fitters GeV2 1.3 X 10-2 < x < 0.40 Elisabetta Gallo (DESY) 5 Polarization at HERA II • EleCtrons/positrons get naturally transverse polarized (Solokov-Ternov effeCt) • Spin rotators to Change in long. polarization in the straight seCtion Before the experiments • Polarization time ~ ½ h • Polarization defined as: • Polarization 30-40% • Spin flip every 2-3 months • Measured by three deviCes Tpol,Lpol and Cavity • Original aim was a preCision of 1%, i.e. not ContriButing signifiCantly to syst. errors Elisabetta Gallo (DESY) 6 Polarization at HERA II • Compton scattering cross section depends on polarization • Circularly polarized laser beam off the electron/positron beam • Asymmetry measured from scattered gamma in special calorimeters • TPOL: measured transverse polarization in straight West section, Sci-Tungsten e.m. calorimeter, +/- 1.9 % precision • LPOL: measured longitudinal polarization in-between the HERMES spin rotators, compact Cherenkov crystal calorimeter, +/- 3.6% precision • Note luminosity uncertainty: 1.8% ZEUS, 2.0% H1 Elisabetta Gallo (DESY) 7 Charged Current events at HERA • Kinematics of reconstructed from hadronic system • Hadronic calorimeter electron proton resolution crucial • Both electron and positron running crucial • e+ p suppressed by (1-y) • Give information on u,d valence density separately Elisabetta Gallo (DESY) 8 CC cross section Combined data Combined u-dominated d-dominated Earlier result • Data important input to PDFs, i.e. u- valence and d-valence separation • Complemented by W-charge asymmetry at the LHC these days, which is however at lower x Elisabetta Gallo (DESY) 9 CC polarized cross section H1 Collaboration 120 [pb] tot CC e p → νX σ 100 H1 HERA I H1 HERA II + 80 e p → νX • Textbook plot: the charged current H1 HERA I H1 HERA II cross section goes to zero for right- 60 handed electrons, as predicted by H1PDF 2012 the SM 40 Linear Fit JHEP09 (2012) 061 20 Q2 > 400 GeV2 y < 0.9 0 +I II HERA -100 -50 0 50 100 Pe [%] Elisabetta Gallo (DESY) 10 Neutral Current at high Q2 • Need electron identification at high angle, in the forward region, so optimized algorithm • At high Q2 the cross section (here expressed as reduced cross section) cannot neglect the xF3 term Elisabetta Gallo (DESY) 11 NC+CC cross sections 2 2 2. • Textbook plot: at high Q ~ MZ , MW become of the same strength • Here shown with the QCD preDiction with the HERAPDF2.0 fit NC • In NC gamma-Z interference and Z- exchange visible at very high Q2 CC Elisabetta Gallo (DESY) 12 NC cross section • Effect of Z-gamma interference clearly visible at high Q2 + - • Measuring e and e one can extract xF3, directly sensitive to the valence quark distribution +0.078 • Integral= 1.790±0.078 (stat) -0.100 ~ 5/3 as predicted Elisabetta Gallo (DESY) 13 QCD fits – HERAPDF2.0 PDFs extracted from these data H1 and ZEUS H1 and ZEUS 2 v 2 v 1 µ = 10 GeV 2 f µ2 = 10 GeV 0.5 f xu HERAPDF2.0 NLO xd 1.8 HERAPDF2.0 NLO 3 MMHT2014 NLO 0.8 MMHT2014 NLO CT10 NLO 0.4 1.6 CT10 NLO (68% CL) NNPDF3.0 NLO NNPDF3.0 NLO 0.6 0.3 1.4 2 1.2 0.4 0.2 • 14 parameters vv vv 1 xu / (HERAPDF2.0) xd / (HERAPDF2.0) 1 0.2 0.1 fit 0.8 0 0 • 10 -4 10 -3 10 -2 10 -1 1 10 -4 10 -3 10 -2 10 -1 1 10 -4 10 -3 10 -2 10 -1 1 10 -4 10 -3 10 -2 10 -1 1 Starting scale x x x x 2 2 1.4 1.3 xg xS 4.5 Q =1.9 GeV 20 4 1.2 • HQs from RT 3.5 15 1.2 3 1.1 VFNS 2.5 10 2 1 1 • 5 orders of 1.5 xg / xg(HERAPDF2.0) xS / xS(HERAPDF2.0) 5 1 0.9 0.8 magnitude in 0.5 2 0 0 0.8 -4 -3 -2 -1 -4 -3 -2 -1 10 -4 10 -3 10 -2 10 -1 1 10 -4 10 -3 10 -2 10 -1 1 x,Q fit 10 10 10 10 x 1 10 10 10 10 x 1 x x • Chi2/NDF=1357 /1131 Elisabetta Gallo (DESY) 14 Polarized neutral current cross sections • EFFect of polarization visible at very high Q2 • Direct observation of parity-violation in NC • Exploiting the polarization the u,d electroweak couplings to the Z can be determined, in quite a competitive way • Polarized F2 constrains the vector couplings JHEP 09 (2012) 061 • Unpolarized xF3 constrains the axial couplings • Special QCD fits with EW parameters free Elisabetta Gallo (DESY) 15 Combined QCD+EW fit ZEUS-EW-Z ZEUS ZEUS u d v v ZEUS-EW-Z NLO (HERA I+II) ZEUS-EW-Z NLO (HERA I+II) LEP+SLC LEP+SLC 1 CDF 1 CDF D0 D0 H1 (HERA I) H1 (HERA I) Standard Model Standard Model 0.5 0.5 type 0 - 0 typecouplings - up down couplings −0.5 −0.5 Phys. Rev. D 93 (2016)Phys. Rev. 092202 −1 −1 −1 −0.5 0 0.5 1 −1 −0.5 0 0.5 1 a u ad Very competitive 13+4 PDF fit ZEUS-EW-Z measurement - at least to constrain the Z to u,d compared to Tevatron - and couplings can constrain the sign Elisabetta Gallo (DESY) 16 Combined QCD+EW ZEUS-EW-S 2 • Dependence on sin !" ∶ ZEUS eff W 0.242 θ • in the Z propagator 2 ZEUS-EW-S all data sin 0.24 binned data • In the GF coupling in CC • through the vector coupling of Z 0.238 0.236 to the quarks Qw(Cs) 0.234 NuTeV E158 • 13+1 parameter fit CMS 0.232 ATLAS D0 CDF LEP+SLC 0.23 LHCb PDG14 0.228 10−3 10−2 10−1 1 10 102 103 Q (GeV) D 93 92016)Phys. Rev. 092202 • Precision not very high W-boson mass ALEPH but other text-book ATLAS CDF plot D0 DELPHI L3 • Similar H1 fit with extraction OPAL H1 Eur.Phys.J.C78 (2018), Eur.Phys.J.C78 of W mass in the t-channel 777 (unique measurement) PDG 2017 H1 80.2 80.3 80.4 80.5 80.6 m [GeV] Elisabetta Gallo (DESY) W 17 Very very high-x • Let’s suppose to have a very high Q2 electron (reconstructed in the detector) and very high x, so that the jet disappears in the fwd region • Reconstruct Q2 from the electron and integrate cross section in x- from edge up to x=1 DATA/THEORY A. Caldwell https://indico.desy.de/indico/event/10523 Phys. Rev. D 89 (2014)Phys. Rev. 072207 Uncertainties in PDFs at high-x still very high, these data at x>0.6 Elisabetta Gallo (DESY) 18 Lessons learned • Precision of scattered electron, hadronic jets is crucial in the whole kinematic range • Combination H1-ZEUS was fundamental for precision PDFs and electroweak measurements in particular (first one in 2006 for xF3 !) • We did not invest enough probably in the polarization measurement – at least at the beginning - its uncertainty as important as the luminosity uncertainty • Very high x still very unknown region (crucial for searches at the LHC) • Interaction with theory worked well at HERA, also crucial • New ideas many years after HERA end (EW fits from 2016) • Strong attention to PDFs from LHC community Elisabetta Gallo (DESY) 19 Exotics Elisabetta Gallo (DESY) 20 Leptoquarks • Classical search at HERA, a resonance in x expected • Early possible signal ~ 200- 220 GeV observed with 1996 data by both H1 and ZEUS (at high y), not confirmed later with more statistics ZEUS • λ BRW model used to classify 1 leptoquarks (still used now) L S1/2 10-1 • Competitive limits set at that ZEUS e±p (498 pb-1) H1 e±p time, now LHC taking over ATLAS pair prod.