CMS experiment at LHC Geoff Hall Imperial College

CMS experiment at LHC Geoff Hall Imperial College London Feb 2009 Geoff Hall 1

Large Hadron Collider n Latest CERN accelerator n n started 2008 very high intensity 15 collisions per year n 10 very high rate n beams cross @ 40 MHz few “interesting” events n ~100 Higgs decays per year Beams n 7 Te. V protons n => 14 Te. V energy n also ions n (eg Pb) Feb 2009 2 (but a small problem occurred - with a big impact) Geoff Hall

Feb 2009 Geoff Hall 3

Experiment by collisions n Colliding beams maximises the energy available to create new particles (compared to shooting at a target) d u n u d u u Hadron collisions are actually between their constituent parts… ~ 1/p ≈ 1/E gluons quarks: both valence and sea (≈ real and virtual) and the particles they exchange (Z, W, …) n n Feb 2009 4 Geoff Hall

What do we actually do? n We design, build and operate the experiments n n n We analyse the data n n LHC was & is enormously challenging so it’s taken a long time… some illustrations of how the experiments are built in the LHC energy range, theory eventually fails so something new must be found for the first time in many years, only experiment can tell us what Now the construction has finished most effort will go into looking at data n Ph. D students and young researchers will be doing most of the work n Snapshot of some typical work in progress Feb 2009 5 Geoff Hall

CMS Compact Muon Solenoid HCAL Muon chambers Tracker ECAL 4 T solenoid Feb 2009 Total weight: 12, 500 t Overall diameter: 15 m Overall length 21. 6 m Magnetic field 4 T 6 Geoff Hall

Design philosophy high p. T lepton and quarks are signatures of possible new physics n n Large solenoidal (4 T) magnet n n Muon detection – penetration n n iron yoke - returns B field, absorbs particles detectors in yoke measure muons Electromagnetic calorimeter – absorb E n good energy resolution for e & g n + Hadronic calorimeter for pions, … n Tracking system – bend in B field n n reconstruct trajectories of most charged particles momentum measurements from bending observe directly many decays complement muon & ECAL measurements Feb 2009 7 Geoff Hall

Muon System Gaseous planar ionisation detectors embedded in iron magnet return yoke to measure particle trajectories Feb 2009 195 k DT channels 210 k CSC channels 162 k RPC channels 8 Geoff Hall

YE+3 Nov 2006 Feb 2009 9 Geoff Hall

YB 0 Feb 2007 Feb 2009 10 Geoff Hall

December 2007 Feb 2009 11 Geoff Hall

YE-1 Jan 2008 Feb 2009 12 Geoff Hall

CMS August 2008 Feb 2009 13 Geoff Hall

First data First LHC Beam (10 Sept) 10 September 2008: beams were steered into collimators and secondary particles detected in CMS before and after September ~ 300 M cosmic ray events recorded 14 T. Virdee CMS Week Dec 08

Machine incident n A superconducting cable connecting magnets and carrying ~9 k. A “quenched” – became resistive - and began to heat up n n in < 1 s the cable failed an arc punctured the helium enclosure, releasing gas at high pressure all the protection systems worked, but the pressure rose higher than expected Since September, impressive diagnosis of what happened…so: improve monitoring repair magnets restart summer 2009 Feb 2009 15 Geoff Hall

Discoveries… n Look at interactions for n n unexpected behaviour n like large energy at large angle to beam (how Rutherford discovered the atomic nucleus) evidence of short-lived particles n visible evidence n Indirect, by peaks in mass spectra Old picture of a charmed particle production and decay in a bubble chamber Feb 2009 16 Geoff Hall

Physics requirements (I) n Mass peak one means of discovery n n eg H => ZZ or ZZ* => 4 l± n n => small s(p. T) typical p. T(µ) ~ 5 -50 Ge. V/c Background suppression n measure lepton charges good geometrical acceptance - 4 leptons background channel t => b => l +n require m(l l ) = m. Z GZ ~ 2. 5 Ge. V n Feb 2009 precise vertex measurement identify b decays, or reduce fraction in data 17 Geoff Hall

Physics requirements (II) n p resolution n n high precision space points n n n detector with small intrinsic smeas well separated particles n n large B and L good time resolution low occupancy => many channels good pattern recognition minimise multiple scattering minimal bremsstrahlung, photon conversions n n material in tracker most precise points close to beam Feb 2009 18 Geoff Hall

What we hope to find at LHC µ+ p Z H Higgs discovery and measurement n µp eg. simplest SM variant n several detectable decay channels n but, ultimately, modest numbers of events are expected at LHC H-> 4µ 30 fb-1 Z µ+ µn plus much possible new physics n eg SUSY, extra dimensions, … Feb 2009 19 Geoff Hall

The Higgs Model n n The Higgs is different ! Higgs is the only scalar particle in the SM n n n Postulated to give rise to mass through spontaneous electroweak symmetry breaking n n n All the matter particles are s=½ fermions All the force carriers are s=1 bosons Also to neutrinos if Dirac particles It would be the first fundamental scalar ever discovered Frankly, almost nothing is known about the Higgs n n n Feb 2009 V( )=µ 2 + + ( + )2 = (v+H)/√ 2 m. H 2=2 v 2=-2µ 2 Nothing is known for the Yukawa-coupling Nothing is known for the Higgs self-coupling Single Higgs? Two Higgs field doublets? Additional singlet? SUSY? MSSM? NMSSM? Extra-dimensions? If the Higgs is discovered, mapping the potential is crucial Slide 20 Geoff Hall

Production of the Higgs The production crosssection is calculable. It depends on the Higgs mass, and the production mechanisms. NLO The Higgs mass is not known and there are few theoretical constraints on it. Feb 2009 21 Geoff Hall

H -> ZZ(*) ->4 l - golden mode Background: tt, ZZ, llbb (“Zbb”) Selections : - lepton isolation in tracker and calo - lepton impact parameter, , ee vertex - mass windows MZ(*), MH Feb 2009 22 H->ZZ->ee mm Geoff Hall

The luminosity challenge n H ZZ ee, MH= 300 Ge. V for different luminosities in CMS 1032 cm-2 s-1 1033 1034 1035 Full LHC luminosity ~20 interactions/bx Feb 2009 Proposed SLHC luminosity ~300 -400 interactions/bx 23 Geoff Hall

Tracker system n Two main sub-systems: Silicon Strip Tracker and Pixels n n n as many measurement points as possible with the most precise measurements close to the interaction point ionisation in silicon produces small current pulses silicon sub-divided into small measuring elements: strips or pixels ~14 layers, ~210 m 2 of silicon, 9. 3 M channels 3 layers, 1 m 2 pixels, 66 M channels TOB TIB Radiation environment ~10 Mrad ionising ~1014 hadrons. cm-2 TEC TID PD Feb 2009 24 Geoff Hall

Microstrip Tracker Outer barrel n automated module assembly 3. 1 M channels Endcaps 3. 9 M channels Inner barrel 2. 4 M channels Feb 2009 25 Geoff Hall

Electromagnetic Calorimeter Scine Scintillating crystals of heavy material – Pb. WO 4 Light produced by electromagnetic showers Light signal proportional to electron or photon energy Parameter Barrel Endcaps Depth in X 0 25. 8 24. 7 # of crystals 61200 14648 Volume 8. 14 m 3 2. 7 m 3 Xtal mass (t) 67. 4 22. 0 Feb 2009 26 Geoff Hall

Trigger and DAQ systems n n Trigger selects particle interactions that are potentially of interest for physics analysis DAQ collects the data from the detector system, formats and records to permanent storage n First-level trigger: very fast selection using custom digital electronics Higher level trigger: commercial computer farm makes more sophisticated decision, using more complete data, in < 40 -50 ms n Trigger requirements n n n n High efficiency for selecting processes of interest for physics analysis Large reduction of rate from unwanted high-rate processes Decision must be fast Operation should be deadtime free Flexible to adapt to experimental conditions Affordable Feb 2009 27 Geoff Hall

Triggering n Primary physics signatures in the detector are combinations of: n n n Candidates for energetic electron(s) (ECAL) Candidates for µ(s) (muon system) Hadronic jets (ECAL/HCAL) Vital not to reject interesting events Fast Level-1 decision (≈3. 2 µs) in custom hardware n n up to 100 k. Hz with no dead-time Higher level selection in software jet p H Z p Z e+ e- n Tracker not part of L 1 trigger n n Feb 2009 Data volume enormous Technically not possible for LHC 28 Geoff Hall

LHC Trigger Levels Feb 2009 29 Geoff Hall

Snapshot of work in progress Feb 2009 Geoff Hall 30

Supersymmetry n A new symmetry of nature? n n n there is a lightest SUSY state into which others decay it does not interact with ordinary matter n n could therefore be the explanation for dark matter it would not be directly observed in CMS n n each fermion has a boson partner (& vice versa) not yet observed! - therefore likely to be heavy SUSY solves some problems with Higgs mass (in GUTs) the signature would be large missing energy – this relies on good hadron calorimetry but it would wise not to depend on a single technique If SUSY exists, it may show up very early at LHC Feb 2009 31 Geoff Hall

Early SUSY searches with the all-hadronic n-jet channel. Tom Whyntie On behalf of the CMS IC SUSY Group (+ friends)

Overview • Introduction • How can we discover SUSY with CMS? • The dijet search channel • A calo-MET independent SUSY search? • The n-jet search channel • How do we go from n to 2 jets? • A suggested strategy for n-jets • S/B ~7 for LM 1 SUSY? • Conclusions and plans Tom Whyntie IC CMS Meeting, 22 nd October 2008 33

Introduction: SUSY at CMS Goal: discover SUSY at CMS • Early data, L < 1 fb-1; • Minimal understanding of the detector. SUSY parameter space considered: • CMS benchmarks: LM 1 -9 (TDR) • Low mass MSu. Gra SUSY • e. g. LM 1: m 0 = 60 Ge. V, m 1/2 = 250, A 0, tan b = 10, sign( ) = + Tom Whyntie IC CMS Meeting, 22 nd October 2008 34

The Dijet Search Channel Analysis note recently approved: CMS AN-2008/071 (Flaecher, Jones, Rommerskirchen, Stoye) q q q • Two high pt jets LSP + similar • Large missing energy q q q LSP Backgrounds • QCD dijet events • Z nn + jets • tt + jet(s), W + jet(s), etc. Tom Whyntie Missing energy relies on calorimeter – is there a way of just using the jets? Is it possible to formulate a discriminating observable based on jet kinematics? IC CMS Meeting, 22 nd October 2008 35

Results for the Dijet System Tom Whyntie IC CMS Meeting, 22 nd October 2008 36