Table Of Content

Status and Commissioning of the CMS Experiment 7 0 O. Buchmu¨ller and F.-P. Schilling (on behalf of the CMS collaboration) ∗ a 0 2 aCERN, CH-1211 Geneva 23, Switzerland n a After a brief overview of the Compact Muon Solenoid (CMS) experiment, the status of construction and J installation isdescribed in thefirst part of thenote. Thesecond part of thedocument isdevoted toa discussion 1 of thegeneral commissioning strategy of theCMS experiment,with aparticular emphasison trigger, calibration 1 and alignment. Aspects of b-physics, as well as examples for early physics with CMS are also presented. CMS will beready for datatakingin time for thefirst collisions in theLarge Hadron Collider (LHC) at CERN in late 1 2007. v 9 1 1. OVERVIEW OF CMS given by a cylinder of length 5.8 m and diameter 0 1 2.6m. Inorderto dealwithhightrackmultiplic- AschematicdrawingofCMSisshowninFig.1. 0 ities, CMSemploys10layersofsiliconmicrostrip The total weight of the apparatus is 12500 tons. 7 detectors, which provide the required granular- 0 The detector, which is cylindrical in shape, has ity and precision. In addition, 3 layers of silicon / lengthanddiameterof21.6mand14.6m,respec- x pixel detectors areplacedclose to the interaction e tively. Theoverallsizeissetbythemuontracking regiontoimprovethemeasurementoftheimpact - systemwhichinturnmakesuseofthereturnflux p parameter of charged particle tracks, as well as of a 13 m long, 5.9 m diameter, 4 Tesla super- e the position of secondary vertices. The electro- h conducting solenoid. This rather high field was magneticcalorimeter(ECAL)usesleadtungstate : chosentofacilitatetheconstructionofacompact v (PbWO4) crystals with coverage in pseudorapid- trackingsystemonitsinteriorwhilealsoallowing i ity up to η < 3.0. The scintillation light is de- X good muon tracking on the exterior. The return | | tected by silicon avalanche photodiodes (APDs) r field saturates 1.5 m of iron containing four in- a in the barrel region and vacuum phototriodes terweavedmuon tracking stations. In the central (VPTs) in the endcap region. A preshower sys- region (pseudorapidity range η < 1.2) the neu- | | tem is installed in front of the endcap ECAL tron induced background,the muon rate and the for π0 rejection. The ECAL is surrounded by residual magnetic fields are all relatively small, a brass/scintillator sampling hadron calorimeter while in the forward regions (1.2 < η < 2.4) | | (HCAL) with coverageup to η <3.0. The scin- all three quantities are relatively high. As a re- | | tillation lightis convertedby wavelength-shifting sult, drift tube (DT) chambersandcathode strip (WLS) fibres embedded in the scintillator tiles chambers (CSC), are used for muon tracking in and channeled to photodetectors via clear fibres. thecentralandforwardregions,respectively. Re- This lightisdetectedby hybridphotodiodesthat sistive plate chambers (RPC) with fast response can provide gain and operate in high axial mag- andgoodtimeresolutionbutcoarserpositionres- netic fields. This central calorimetry is comple- olution are used in both regions for timing and mented by a “tail-catcher”in the barrelregion— redundancy. ensuring that hadronic showersaresampled with The bore of the magnet coil is also large nearly 11 hadronic interaction lengths. Coverage enough to accommodate the inner Tracker and up to a pseudorapidity of 5.0 is provided by an the calorimetry inside. The tracking volume is iron/quartz-fibrecalorimeter. TheCerenkovlight emitted inthe quartzfibresis detectedby photo- ∗Talksgivenatthe11thIntl. ConferenceonB-Physicsat multipliers. The forward calorimeters ensure full HadronMachinesBEAUTY2006,Oxford(UK),Septem- geometric coverage for the measurement of the ber2006 1 2 To achieve this goal, CMS chose a large su- perconducting solenoid with a 4 T field [1]. The 12.9 m long and 5.9 m diameter magnet is operated with a current of 19.5 kA. The overall energy stored in the field corresponds to 2.7 GJ. TheCMSmagnethasbeenassembledonthesur- face and was successfully tested during the re- cent CMS Magnet Test and Cosmic Challenge (MTCC, see section 3). 2.2. The Muon System Each Endcap (ME) of the Muon Detector [2] contains 234 CSCs. All of the 468 endcap CSCs havebeeninstalledonthemagnetyokedisksand are now being commissioned with cosmic rays. Figure 1. Schematic view of the CMS detector. Eachtrapezoidal endcap CSC chamber has 6 gas gaps containing a plane of radial cathode strips anda planeofanode wiresparallelto the longest edgeofthetrapezoidandso,roughlyperpendicu- transverse energy in the event. lar to the strips. The spatial resolution provided by each chamber ranges from 100 µm in Station 2. STATUS OF CMS (SEPTEMBER 1 to roughly150µmin Stations 2 to 4. Wire sig- 2006) nals are fast and are used in the Level-1 Trigger though they have coarser position resolution. In the current CMS Master Schedule the ini- The manufacture of Barrel DT chambers is tial detector will be ready for first collisions in completeandalmostallofthe266DTshavebeen the last quarter of 2007. Installation of the pixel installed. The chambers installed in the barrel TrackerandtheECALendcapsisforeseenduring yokes(YB) areorganizedin 4 stations. EachDT the 2007/2008 winter shutdown, in time for the chamber is piggy-backed by one or two RPCs. first physics run in spring 2008. CMS has made The chambers are staggered from station to sta- much progress over the summer. The detector tionsothatahigh-p muonnearasectorbound- hasbeenclosedforthefirsttime,thesolenoidhas T arycrossesatleast3stations. Thechamberscon- beentesteduptothe designfieldof4T,andcos- sistoftwelveplanesofaluminumdrifttubes;four mics datahavebeentakensimultaneouslyfroma rφ measuring planes are placed above, and four slice ofallthe sub-detectorsusing predominantly below, a group of four z measuring planes. Each the final components. The items staged for de- stationgivesamuonvectorinspacewithapreci- sign luminosity running include the fourth end- sionoflessthan100µminpositionandlessthan cap muon station, RPC chambers at low angles, 1mradindirection. Severalchambersofthebar- several online farm slices and the third layer of rel and endcap muon systems were successfully forward pixel disks. operated during the MTCC in July and August 2.1. The Magnet (see section 3 for more details). Therequiredperformanceofthemuonsystem, and hence the bending power, is defined by the 2.3. The Tracker invariantmassresolutionfornarrowstatesdecay- TheCMSTracker[3]occupiesacylindricalvol- ingintomuonsandbythe unambiguousdetermi- ume of length 5.8 m and diameter 2.6 m. The nationofthechargeformuonswithamomentum outer portion of the Tracker is comprised of 10 of 1TeV.Thisrequiresamomentumresolution layers of silicon microstrip detectors and the in- ≈ of ∆p/p 10% at p=1 TeV. nerportionismadeupof3layersofsiliconpixels. ≈ 3 Silicon provides fine granularity and precision in (400 or 500 crystals) at CERN and in Rome. At allregionsforefficientandpuretrackreconstruc- present122modules outof144havebeen assem- tion even in the very dense track environment of bled. Thirty bare supermodules (SM, each com- high energy jets. The three layers of silicon pixel prising 1700 crystals) have been assembled. The detectors at radii of 4, 7 and 11 cm provide 3D crystal production is now proceeding at the rate spacepointsthatareusedtoseedtheformationof ofabout1250crystalspermonth. ThelastBarrel tracksby the pattern recognition. The 3Dpoints crystal will be delivered by March 2007,allowing also enable measurement of the impact parame- insertion of the last supermodule in May 2007. ters of charged-particletracks with a precisionof The production of Endcap crystals will start im- the order of 20 µm in both the rφ and rz views. mediatelyaftertheendoftheBarrelcrystalspro- Thelatterallowsforprecisereconstructionofdis- duction and is expected to finish by February placed vertices to yield efficient b tagging and 2008. For the Barrel electromagnetic calorime- good separation between heavy and light quark ter, the integration status is the following: Dur- jets. ing last spring, an integration rate of 4 SMs per The CMS Tracker continues to make good month was achieved. Currently 24 SMs are com- progress. Hybrid production was completed at pleted, which is consistent with the general con- theendof2005,andmoduleproductionwascom- struction schedule of CMS. The performance of pletedinthe springof2006. InOctober 2006the theintegratedsupermodulessatisfiesfullythede- forward(+)halfoftheTrackerOuterBarrelTOB sign performance stated in the Technical Design willbecompleted,TrackerInnerBarrelTIB+will Report. After integration, each SM is subjected be readyforintegrationinto TOB+,andTracker to a pre-calibrationand commissioning run of 10 EndCap TEC+ will be delivered to CERN from days with cosmic muons. In addition, eight SMs Aachen,whereitwascompletedinSeptember. In have already been pre-calibrated in an electron November2006thebackward(-)halfofTOBwill beam. Preliminarycomparisonsbetweenthecos- be completed, TIB- will be ready for integration mic and test beam data indicate that the cosmic intoTOB-andTEC-willbe completedreadyfor datayieldaninitialinter-calibrationwithapreci- integration into the Tracker support tube. The sion better than 2%. Finally, two SMs have been qualityofthe Trackersub-detectorsis verygood. successfully tested in the magnet of CMS dur- The number of dead or noisy channels is two per ing the MTCC, showing that the performance is milleandthesignaltonoiseratiois>25:1. The maintained inside CMS and in the 4 T magnetic Pixel Detector continues to make good progress. field. Large pre-series of all off-detector readout Allcomponentsarenowavailableand15%ofthe modules are in hand. Their production and test- modules(BarrelPixels)andPlaquettes(Forward ing will be completed by mid-November 2006. Pixels) have been successfully produced. A pixel sector will be delivered to CERN in December 2.5. Hadronic Calorimeter 2006for integrationbefore installation into CMS All HCAL [5] module types [HB (barrel), HE in September 2007. The full Pixel Detector will (endcap), HO (outer) and HF (forward)], includ- be ready to be installed into CMS in November ingabsorberandoptics,arecompleted. Photode- 2007. tectors and electronics have been installed and a comprehensive calibration of HCAL using Co-60 2.4. Electromagnetic Calorimeter sources has been completed. HF will be the first The ECAL [4] is a hermetic, homogeneous sub-detector to be lowered into the underground calorimeter comprising 61200 lead tungstate experimental cavern. This is expected to occur (PbWO4) crystals mounted in the central bar- in October 2006. The HB and HE are fully in- rel part, closed by 7324 crystals in each of the 2 stalled. The HF has been calibrated to 5%, ∼ endcaps. HE and HB to 4%. The HE and HB have also ∼ About 56500 of the barrel crystals have been been run globally with muons. HCAL Trigger deliveredandarebeingusedtoconstructmodules and DAQ have been tested with cosmics. HCAL 4 thiscombinedMagnetTestandCosmicChallenge was the recording, offline reconstruction and dis- playofcosmicmuonsinthe4subsystemsofCMS (Tracker, ECAL, HCAL and Muon Detector), with the magnet operating at 4 T. The original scope was expanded to include substantial offline as well as online systems ob- jectives. Data transfer to some Tier-1 centers, online event display, quasi-online analysis at the mainCERNsite,andfastofflinedata-checkingat FermilabweresomehighlightedtargetsofMTCC Phase I designed to offer a first hand taste of a “CMS-like” running experience. After the cabling of the detector elements the final closing of the yoke proceeded. It was com- pletedon25Julyallowingthestartofthemagnet test. The power circuit was completed and the testing of the coil progressed in steps: 5, 7.5, 10, 12.5,15,17.5andthen19.12kAtoreachthenom- inal 4 T field on 22 August. At each value, fast Figure2. Characteristiceventdisplayofacosmic discharges were provoked to learn how to tame muon thatleft signaturesinall four activedetec- the dumping of energy inside the cold mass dur- tor elements (Tracker, ECAL, HCAL and Muon ing such a discharge. After running at 3.8 T for Detector). 48 hours for the cosmic challenge, the coil was run at 4 T for 2 hours on the 28 August, and the functionaltestsofthemagnethavebeendeclared slowcontrolsarefully operationalanddataqual- completed with success. ity monitoring(DQM)isnowunderdevelopment Twenty five million cosmic triggered events and partially up and running. wererecordedwiththeprincipalsubdetectorsac- tive,ofwhich15millioneventshavestablefieldof 2.6. Trigger and Data Acquisition 3.8T. Data-takingefficiency reachedover90% Thetrigger[6,7]systemis wellinto production ≥ for extended periods. Several thousand of these withmanycomponentsalreadycompleted. Com- eventscorrespondtothe“4-detector”benchmark, ponents are exercised with other trigger and de- andthewholedatasamplewillprovideusefulun- tectorelectronicssystemsinthe ElectronicsInte- derstanding and calibration of the combined de- grationCenteratCERN.Integrationtestsarerun tector and software performance. A characteris- in which detector primitives are generated and ticeventdisplayofacosmicmuonthatleftsigna- used to feed the trigger and DAQ. Final system turesinallfouractivedetectorelements(Tracker, components were also used in data taking during ECAL, HCAL and Muon Detector) is shown in the MTCC this summer (see section 3). Fig. 2. The shown trajectory corresponds to theresultofthestandalonemuonreconstruction. 3. THE MAGNET TEST AND COSMIC In the inner detector this trajectory coincides CHALLENGE (MTCC) withintheexpecteduncertaintieswiththeECAL, A full-scale test and field-mapping of the CMS HCAL andTrackermeasurementsdemonstrating 4 T solenoidmagnetsystempriorto loweringthe that the propagation of the standalone muon in- major elements has been carried out this sum- formationwassuccessfullycarriedout. Themain mer. In addition to this magnet test also a “cos- conclusions of the successful MTCC are: mic challenge” was performed. The objective of – CMS can be opened and closed on the 5 timescales intended; – Single beams: Single beams give rise to – The magnet workedstably and safely at 4 T; beam halo muon and beam-has events. The – The subdetectors work with the magnet and near-horizontalbeamhalomuonscanbe usedfor each other; Tracker and muon end-cap alignment. Beam-gas – The subdetectors can be integrated with the interactions can be used for various commission- central DAQ, trigger, DCS, DQM etc.; ing tasks. – The commissioning strategy broadly worked; – Colliding beams: LHC collisions will pro- – CMS can work as a world-wide, unified team. vide a cocktail of physics events, depending on the specific luminosity. For commissioning pur- 4. CMS COMMISSIONING STRATEGY poses, muons and electrons from W±,Z0 decays, but also minimum bias and QCD jet events are OnceCMSisfullyinstalledintheunderground useful. cavern,a comprehensive programof commission- Duringthelow-luminosity2007calibrationrun ing has to be carriedoutin orderto optimize the at√s=900GeV,theavailabledatawillbecom- detector performance, and to prepare it for an pletelydominatedbyminimumbiasandQCDjet optimal exploitation of its physics potential. events. At mosta few tens of W±,Z0 events can The commissioning program includes the pre- be expected in this phase. However, the 2008 pi- cisecalibrationoftheECALandHCALcalorime- lotrunwillseesubstantialratesofW±,Z0events ters, the alignment of the tracking and muon de- (Tab. 1). Large samples of high p muons from T tectorsandthedefinitionofanefficientandflexi- these events can thus be accumulated within a bletriggersetup. Oncethedetectorisalignedand short timescale and used e.g. for alignment. calibrated, physics tools such as b-tagging and missing E measurement can be commissioned. T 4.3. CMS Commissioning Strategy The main CMS commissioning goals are: (1) 4.1. LHC Schedule Efficient operation of trigger and DAQ; (2) Recently, the LHC start-up schedule has been Tracker and Muon alignment; (3) Calorimeter revised[8]. Theclosingoftheexperimentswillbe calibration. Once these goals are achieved, the followedbythestart-upandinitialcommissioning commissioning of higher level physics tools and ofthe LHC acceleratorwith singlebeams atE = objects such as b-tagging, jets, missing E can 450 GeV in autumn 2007. In December 2007, T proceed. a 2-3 week “calibration run” will be carried out Provided a sufficient amount of physics, cos- with collisions at √s=900 GeV and low specific luminosities 1029 cm−2s−1. mics and beam-halo/gas events can be accumu- sp L ∼ lated,alotofimportantcommissioningtaskscan Afterthewintershutdown,LHCoperationwill be performedalreadyduringthe 2007calibration be continued with commissioning at 7 TeV and run: Trigger and DAQ can be timed in, syn- a three-stage running scenario: (1) a one-month “pilot physics run” at 1032 cm−2s−1; (2) chronized and their data integrity be checked. sp L ∼ The trigger algorithms can be debugged and im- pushing of the machine parameters in order to increase the specific luminosity to 1033 cm−2s−1; proved. ECAL and HCAL can be calibrated to 2%. Tracksfromcosmic andbeamhalo muons (3) towards the end of 2008 the luminosity could ∼ bepushedtothe nominal2 1033 cm−2s−1 value. aswellascollisiontrackscanbeusedtoalignthe ∗ Tracker to significantly better than 100 µm, and 4.2. Commissioning Data Samples to align the muon chambers. From the CMS commissioning point of view However, important commissioning tasks can threedistinctphasescanbeidentified,whichgive only be carriedoutin2008: ThefinalHCAL and access to complementary data sets: ECAL calibrations, the final Tracker alignment – No beams: Accumulation of large samples of (the pixel detector will only be installed in 2008) cosmic muons, which are very beneficial e.g. for and the commissioning of b-tagging, missing E T Tracker and muon barrel alignment. etc. 6 Table 1 Anticipated rates of W± µ±ν and Z0 µ+µ− events after HLT in 2008. Luminosit→y 103→2 cm−2s−1 2 1033 cm−2s−1 ∗ Time few weeks 6 months 1 day few weeks one year Int. Luminosity 100 pb−1 1 fb−1 1 fb−1 10 fb−1 W± µ±ν 700K 7M 100K 7M 70M Z0 →µ+µ− 100K 1M 20K 1M 10M → 5. MAJOR COMMISSIONING TASKS area of 200 m2. This large number of indepen- ∼ dent silicon sensors and their excellent intrinsic Details on the commissioning procedures as resolution of 10...50 µm make the alignment of well as the physics potential can be found in [9, theCMSTrackeracomplexandchallengingtask. 10]. In the following, a few aspects of the major commissioning tasks are highlighted. 5.2.1. Impact of Misalignment Misalignment will degrade the track parame- 5.1. Trigger and DAQ ter resolution and hence affect the physics per- The CMS trigger consists of the hardware formance of the Tracker, for instance the mass based Level-1 trigger (nominal accept rate resolution of resonances or the b-tagging per- 100kHz)usingcalorimeterandmuonsignals,and formance. To assess the impact of misalign- oftheHighLevelTrigger(HLT),amassivelypar- ment onthe trackingandvertexingperformance, allel processor farm in which the full event infor- Two misalignment scenarios have been imple- mationisavailable. TheHLTreducestherateby mented [12,13],which are supposed to mimic the three orders of magnitude to 100 Hz which are conditions for different data taking conditions, sent to the Tier-0 centre for prompt reconstruc- namely the First Data Taking Scenario and the tion. Long Term Scenario (Fig. 3). Anefficientoperationofthetriggerisensuredif bothECALandHCALarecalibratedtothelevel of 2%, the muon detector is aligned to 500 µm, 5.2.2. Alignment The alignment strategy for the CMS Tracker and the silicon Tracker is aligned to 20 µm (for HLT b-tagging). Most of these requirements can foresees that in addition to the knowledge of the modulepositionsfrommeasurementsatconstruc- be met already during the 2008 pilot run. Trigger tables [7] for low and high luminosity tiontime,thealignmentwillproceedbymeansof runninghavebeenpreparedandarecurrentlybe- aLaserAlignmentSystem(LAS)andtrack-based alignment. The LAS uses infrared laser beams ing refined. In addition, trigger scenarios for the 2007and2008calibrationandpilotrunsarebeing and operates globally on the larger Tracker com- prepared,withafocusoncommissioningtriggers. posite structures. It cannot determine the po- sitions of individual modules. The goal of the The data reconstructed at the Tier-0 are split in (10) streams [11], according to their physics LAS is to provide measurements of the Tracker O substructures at the level of 100 µm as well as content. Particularly important for commissioning purposes is the express stream, which is de- monitoring of possible structure movements at fined to have a fast turn-around time. Its pur- the level of 10 µm. Track-basedalignmentrepresentsamajorchal- pose is to provide fast feedback on any possible interesting high mass signals,but also to provide lenge at CMS because the number of degrees of data sets used for alignment and calibration. freedom to be determined with a precision of 10 µm is (100,000). Three different algo- ∼ O 5.2. Tracker Alignment rithmsforalignmentwithtracksareusedinCMS: The CMS Silicon Tracker consists of 15000 – HIP algorithm: Minimization of a local χ2 ∼ silicon strip and pixel sensors covering an active function on each sensor [14]. Correlations be- 7 )/p(pmTTm(pT)/pT vspslo hendorgfr,e-t t-cpetter Tamrm li =g aan llm1iigge0nnnmm0tee nGntteV/c 6m]x [ µ----12344321000000000000000000 2 4 6 8 10 N(Alignables) 1234500000000000-40A0fterIIII- ====I3 0115I0t0e0rat-i2o0n0s:-100 0 100 20EMREMR0nnMMeettaaSSrrnnii3 ee 0 ss0 1177 ....779911226655400449900 Iteration 6x [µm] 10-1 6m]y [µ---1234321000000000000000 N(Alignables) 123450000000000 EMREMRnnMMeettaaSSrrnnii ee ss 66-- 22..7788..225544005533 -4000 2 4 6 8 10 0-400 -300 -200 -100 0 100 200 300 400 Iteration 6y [µm] 10-20 0.5 1 1.5 2 2d.5 6m]z [µ----12344321000000000000000000 2 4 6 8 10 N(Alignables) 111110246824680000000000-400 -300 -200 -100 0 100 20EMREMR0nnMMeettaaSSrrnnii3 ee 0 ss0 2299 33..7711..221145500330110 Iteration 6z [µm] Figure3. Transversemomentumresolutioninthe Figure 4. Alignment of the CMS Pixel detector CMS Tracker with and without misalignment. using the HIP algorithm. tweendifferentsensorsarenotexplicitlyincluded, 5.3. Muon Alignment but taken care of implicitly using iterations. No TheCMSMuonsystemconsistsof790individ- inversions of large matrices are involved (Fig. 4). ual chambers with an intrinsic resolution in the – Millepede-II: A global linear least-squares range 75...100 µm. Excellent alignment of the algorithm which takes into account correlations Muon system is particularly important to ensure among parameters. For N alignment parame- efficient muontriggeringand good trackmomen- ters the solution requires the inversionof a NxN tum resolution at large momenta. matrix. A new version, Millepede-II, was devel- For optimal performance of the Muon spec- oped [15] which offers additional solution meth- trometer over the entire momentum range up ods and is expected to be scalable to the full to 1 TeV, the different muon chambers must CMS Tracker alignment problem within reason- be aligned with respect to each other and to able CPU time. the Tracker to within 100 µm. To control mis- –Kalmanfilter: Amethodforglobalalignment alignment during commissioning and to moni- derivedfromtheKalmanfilter. Itisiterativeand tor further displacements during operation,CMS avoidsinversionsoflargematrices[16]. Thealign- will combine measurements from an optical- mentparameterupdatecanbe extendedto those mechanicalsystemwiththeresultsoftrackbased elements that have significant correlations with alignment [17]. the ones in the current track. The current alignment strategy foresees that 5.4. Calorimeter Calibration cosmics and beam halo muons can be used to Precise calibration of the ECAL and HCAL carryoutaninitialalignmentofthestripTracker calorimeters is a key ingredient for precise mea- in 2007, which could be improved using high p surementsofphotons,electrons,hadrons,jetsand T collisiontrackswhenavailable. Whenlargersam- missing E . Certain physics channels, such as T ples of muons fromW±,Z0 decays become avail- H0 γγ, impose very tight requirements such → able in 2008, a standalone alignment of the by astheECALcalibrationbeingknowntothelevel then installed pixel detector can be performed, of 0.5%. On the other hand typical SUSY sig- followed by the precise alignment of the strip natures involve final states with jets and missing Tracker, using the pixel detector as a reference E ,measuredusingtheHCAL.Theknowledgeof T system. the energyscalefor b-jets iscrucialfortopquark 8 %) tential for early physics with CMS [10]. Here, n ( 3 only two examples can be highlighted: o si eci2.5 6.1. Top Physics Pr Since the top quark pair production cross sec- n o 2 tion is very large at the LHC ( 830 pb), top ati physics is an early physics topic∼for CMS. Cross br ali1.5 5 fb-1 sectionandmassmeasurementswillbepossiblein C all major decay channels. For =1 fb−1, 700 L ∼ 1 events are expected in the dilepton channel. A cross section measurement at the 10% level, as 0.5 well as a mass measurement to ∆m 4.2 GeV t ∼ will be possible [19], the systematic error being 0 dominated by the b-jet energy scale uncertainty. 1100 110022 110033 HHLLTT eevveennttss ppeerr ccrryyssttaall At = 10 fb−1, the top mass can be measured L to ∆m 1.2 GeV in the semileptonic chan- t ∼ nel [20], provided the b-jet energy scale is known to 1.5%. These examples illustrate the impor- Figure 5. ECAL calibration precision. ∼ tance of the commissioning of physics tools such asb-tagging(alignment)andjetenergyscale(calibration). mass measurements. 6.2. High Mass Dileptons HCAL will be pre-calibratedto the levelof4% Resonancesathighmassindileptonfinalstates using a radioactive source. This calibration will are very interesting for early discoveries since be improved upon using minimum bias events they could potentially show up at luminosities for HCAL uniformity, E/p from high pT isolated as low as a few 100 pb−1. High mass dileptons tracks extrapolated from the Tracker, and di-jet (where M 1 TeV) are predicted by various ll balance for regions not covered by the Tracker newphysicss∼cenarios. Simulationsshow[10]that acceptance [9]. a resonance in the di-muon mass spectrum could For ECAL, a pre-calibration using test beam be discovered within a few weeks of data taking. and light-yield measurements as well as cosmics However, for a measurement of the mass of the will be performed to a precision of around 2%. resonance and to separate it from the continuum During the 2007 calibration run, it is envisaged background,averygoodalignmentoftheTracker tosetupadedicatedminimumbiaseventstream and muon detectors is crucial. with a bandwidth of 1 kHz. Using phi symmetry aswellasπ0/η0 γγ decays,thecalibrationcan 7. ASPECTS OF B-PHYSICS → beimprovedquickly. TheultimateECALcalibra- tion precisionwillbe reachedfrom the 2008pilot 7.1. b-Tagging Performance physics runonwards,using the E/p method with Various b-tagging algorithms have been imple- high momentum electrons from W,Z decays (see mented in the CMS reconstruction software and Fig. 5). Studies using full simulation have shown their performance evaluated in [9]. Both lifetime that a precisionof 0.5%canbe achievedin the aswellassoftlepton(e± andµ±)tagshavebeen barrel using 5 fb∼−1 of data [18]. studied. The lifetime-tag based algorithms are: ∼ – Track counting: Robust algorithm which counts the number of tracks in a jet with impact 6. EARLY PHYSICS WITH CMS parameter above a given threshold; Provided the commissioning tasks can be be –Probability: Calculatesthe probabilitythata performed successfully, there is an exciting po- set of tracks originate from the primary vertex; 9 y 1 LHC 2007 calibration run and 2008 pilot physics c n runmustbeusedfordetectorcommissioning. Im- e fici 10-1 portant examples are the trigger and DAQ sys- f tem, Tracker and Muon alignment and calorime- e t ter calibration. The commissioning of physics e b-j10-2 toolsreliesonthesuccessofthesetasks. Thereis n an exciting programfor early physics with CMS. o n 10-3 REFERENCES 1. CMS Coll., “Magnet Technical Design Re- 10-4 port”, CERN/LHCC 1997-010 (1997). 2. CMSColl.,“MuonTechnicalDesignReport”, CERN/LHCC 1997-032 (1997). 10-5 00..11 00..22 00..33 00..44 00..55 00..66 00..77 00..88 00..99 11 3. CMS Coll., “Tracker Technical Design Re- bb--jjeett eeffffiicciieennccyy port”, CERN/LHCC 1998-006 (1998); Ad- dendum CERN/LHCC 2000-016 (2000). 4. CMS Coll., “ECAL Technical Design Re- port”, CERN/LHCC 1997-033 (1997). Figure 6. B-tagging performance for c (top), g 5. CMS Coll., “HCAL Technical Design Re- (middle) and uds (bottom) jets. port”, CERN/LHCC 1997-031 (1997). 6. CMSColl.,“L1TriggerTechnicalDesignRe- port”, CERN/LHCC 2000-038 (2000). – Combined secondary vertex tag: Recon- 7. CMS Coll., “Data Acquisition and High- structs the secondary vertex of the b hadron Level Trigger Technical Design Report”, decay and combines several discriminating vari- CERN/LHCC 2002-026 (2002). ables [21] (Fig. 6). 8. http://lhc-commissioning.web.cern.ch/lhc- All lifetime based algorithms require that the commissioning/ pixel detector has been aligned with tracks. 9. CMSColl.,“PhysicsTechnicalDesignReport Vol. 1: Detector Performance and Software”, 7.2. Sensitivity to Bs0 → µ+µ− CERN/LHCC 2006-001 (2006). AnewstudyoftheCMSsensitivitytothe rare 10. CMSColl.,“PhysicsTechnicalDesignReport decay Bs0 → µ+µ− has been performed [22]. It Vol. 2: Physics Performance”, CERN/LHCC usesadedicatedHLTtrigger(acceptrate1.7Hz) 2006-021 (2006). and a cut-based offline analysis. For a signal 11. D. Acosta et al., CMS Note 2006/095. efficiency of 2%, a background rejection of 12. I. Belotelov et al., CMS Note 2006/008. 2.6 10−7isob∼tained. For =10fb−1 thiscorre- 13. P. Vanlaer et al., CMS Note 2006/029. ∗ L sponds to 6.1 (13.8) signal (background) events. 14. V. Karimaki et al., CMS Note 2006/018. The corresponding 90% C.L. upper limit on the 15. P. Schleper et al., CMS Note 2006/011. branchingfractionisB(Bs0 →µ+µ−)<1.4∗10−8, 16. R. Fruehwirth et al., CMS Note 2006/022. including systematic errors. 17. A. Calderon et al., CMS Note 2006/016. 18. L. Agostino et al., CMS Note 2006/021. 8. CONCLUSIONS 19. M. Davids et al., CMS Note 2006/077. 20. J. D’ Hondt et al., CMS Note 2006/066. The construction and installation of the CMS 21. C. Weiser, CMS Note 2006/014. detector is making very good progress, demon- 22. C. Eggel et al., CMS CR 2006/071. stratedforinstancebythesuccessfulMagnetTest andCosmic Challengecarriedout in the summer of 2006. Following the installation of CMS, the