University of Florida participation in CMS Experiment at Large Hadron Collider

Faculty (experiment): Acosta, Avery, Konigsberg, Korytov, Mitselmakher, Yelton
Faculty (theory and phenomenology): Field, Matchev, Ramond

The Large Hadron Collider (LHC), now under construction at CERN, will provide proton collisions with 14 TeV energy in center-of-mass frame and luminosity of 1034 cm-2s-1, 7 times more energy and nearly 100 times more luminosity compared to the Tevatron at Fermilab. LHC is expected to produce the first collisions by the end of 2007. Collisions at this energy and luminosity will be sufficient to resolve the puzzle of electroweak symmetry breaking, either by detection of the Higgs particle predicted by the Standard Model or by discovering new interactions. The new energy frontier may also make possible discoveries of new particles predicted by variety of theories with new symmetries, e.g., Supersymmetry that would unify bosons and fermions, and/or extra spatial dimensions.

The Compact Muon Solenoid (CMS) is one of the two major detectors at LHC being built to conduct searches for new high mass particles, such as Higgs boson, supersymmetric particles, etc. The other competing experiment is ATLAS. Contrary to its name, CMS is a huge detector, 15 m high and 25 m long, weighing more than 12,000 tons. It contains subsystems which are designed to identify photons, electrons, muons, and hadrons, all originating from proton-proton collisions right in the center of the detector, and identify their energies. The CMS Detector is designed around its poweful superconducting solenoid which generates a magnetic field of 4 T in the volume of roughly 350 m3 (3 GJ of stored energy). The innermost part of CMS is a silicon-based tracker. Surrounding it is an electromagnetic calorimeter made of lead-tungsten glass crystals (PbWO4), which is itself surrounded with a copper/scintillator hadron calorimeter. The tracker and the calorimeters are compact enough to fit inside the CMS solenoid. Outside the magnet are the large muon detectors, which are inside the return yoke of the magnet. The CMS Collaboration encompasses more than 2000 physicists from 155 institutions from 37 counties (as of 2006). The UF group strongly contributed to the design and construction of the CMS Detector, continues to play leading roles in CMS commissioning and operation, computing and software development, and is very active in CMS physics analysis preparations.

CMS design, construction, and operation. One of the key characteristics of the CMS Experiment is its capability to identify muons and measure their momenta in a wide angular coverage. Muons, like electrons, are not influenced by the strong force---hence they are a clean probe of the hard scattering, isolated from any hadronic debris. Many channels through which new physics may manifest itself at LHC involve muons, so a clear elucidation of these processes must involve muon detection.

  • G. Mitselmakher (at UF since 1995) is the member of the CMS Steering Committee and co-manager of the CMS Muon System. He was also the project manager of the design and construction of the Endcap Muon System (EMU), a $40M project involving about 80 US physicists from 12 universities and Fermilab. The project also included international collaboration with PNPI (St.Petersburg, Russia) and IHEP (Beijing, China). The Endcap Muon System consists of 500 detectors, so-called, Cathode Strip Chambers, covering area of 1000 m2 and detects muons passing through them with a precision of ~100 μm. Unlike many other muon detectors, it is capable of identifying muon track six-hit stubs within a fraction of 1 μs and, with a precision of about 1 mm, can operate in high-rate environment and highly non-uniform magnetic fields. Although the idea of Cathode Strip Chambers was around since 1979 (G. Charpak), it is for the first time that they were used at such a large scale.
  • A. Korytov (at UF since 1996) led the design and construction of the EMU Cathode Strip Chambers ($18M part of the $40M project mentioned above). Korytov also coordinated the effort on developing the program of tests for commissioning chambers with electronics. The High-Bay HEE lab of the UF Physics Department hosted the pilot site for such tests, so-called Final Assembly and System Tests site. Seventy-five of the largest Cathode Strip Chambers were tested at the UF FAST site. Later, six more FAST sites were commissioned at UCLA, PNPI (Russia), IHEP (China), JINR (Russia), and two sites at CERN---all of them were coordinated via UF. Korytov also leads the UF effort on developing analysis and software tools for Endcap Muon System Data Quality Monitoring project. Korytov now is the deputy manager for the Endcap Muon System maintenance and operation.
  • D. Acosta (at UF since 1997) was responsible for the design and construction of a large portion of the Track-Finder Trigger for the Endcap Muon System. In particular, algorithms had to be conceived and implemented to perform track finding in the endcap muon system---a pattern recognition problem of identifying penetrating muons in several detector stations amid a large flux of random coincidences. Specifically, a dedicated processor had to be developed that would compute the momentum of muon candidates within a fraction of 1 μs from track stubs provided by muon chambers. The cost of the project was $1.3M. Acosta is now the deputy manager for commissioning CMS detector performance.

CMS Computing and Software. The enormous amount of data to be recorded by CMS (from Petabytes in the early years to Exobytes in the later years) must be stored, distributed geographically, and analyzed by thousands of CMS collaborators. To cope with the extreme information technology demands of the LHC, a global computational infrastructure, known as the Worldwide LHC Computing Grid (WLCG), is being developed and deployed.  The WLCG is composed of a hierarchy of computing centers which fan-out from the data source:  A Tier-0 Center represents the computing facility located at the CMS site (CERN), Tier-1 Centers represent national computing facilities (such as Fermi National Accelerator Laboratory) providing custodial services for CMS data, Tier-2 Centers represent advanced regional computing facilities enabling data access and analysis for the CMS collaboration, and Tier-3 Institutes represent the traditional computing facilities located at a typical university. 

  • P. Avery (at UF since 1985) is the project lead for the University of Florida CMS Tier-2 Computing Center and is helping to define, develop and deploy the required global cyber-infrastructure supporting CMS and data intensive science in general.  In particular, the University of Florida leads the Grid Physics Network ($11M project), the International Virtual Data Grid Laboratory ($12M project), and participates in the Open Science Grid ($35M project), the UltraLight Networking Project ($2M), and the Data Intensive Science University Network ($10M project).
  • D. Acosta (at UF since 1997) leads the CMS Computing, Software, and Analysis 2006 Challenge, an effort to integrate and commission the CMS information technology infrastructure in preparation for data taking in 2007.

CMS Physics. The UF group leads a number of efforts on preparations for physics analyses with CMS. The first data are expected to start coming in 2008.

  • CMS Physics Technical Design Reports (volumes 1 and 2) contain many chapters and sections describing analyses originating from the UF group. In addition, Acosta was the Editor of the whole CMS Physics Technical Design Report, v.1.
  • Higgs Boson Search in its Decay Mode H→ZZ→4μ (Bartalini, Drozdeskiy, Korytov, Mitselmakher). This is one of the cleanest channels to search for the Standard Model Higgs boson, also known as a golden channel. Our recent suite of studies gives a coherent strategy for searching for the SM Higgs boson in this channel, including various methods of obtaining critical corrections, and estimating systematic errors on them, directly from data. We also outlined a scheme of proper evaluation of statistical significance of an event excess (should it be observed), taking into account our estimates of systematic errors on backgrounds and the fact that we would carry out the search a broad range of possible Higgs boson masses. We can start excluding certain SM Higgs boson mass regions as early as at L~2 fb-1 and a discovery can be possible at L~10 fb-1.
  • Search for Higgs Boson Produced in Vector Boson Fusion Process and Decaying via WW→lνjj (Avery, Pi) Cross section for the SM Higgs Boson production via vector boson fusion is substantially smaller in comparison to the dominant production via gluon-gluon fusion. However, the absence of jets in the central part of the detector helps significantly suppress the tt and Wbb backgrounds, which puts this channel among frontrunners aiming at the early observation of the Higgs boson for the mass range 160-180 GeV/c2.
  • Inclusive Supersymmetry Search with Muons, Jets, and Missing Energy (Acosta, Cavanaugh, Korytov, Matchev, Mitselmakher, Pakhotin, Schmitt, Scurlock). Once over the kinematic threshold, SUSY particles would be produced copiously via strong interactions at the LHC. They will further decay in successive steps to the lightest stable particle, at each decay step producing either jets or leptons. We carried out detailed studies of two channels: single muon plus jets plus missing energy, and two same-sign muons plus jets plus missing energy.
  • Early observation of ZZ events via the 4μ-decay channel (Drozdeskiy, Korytov, Mitselmakher). Observation of the Standard Model ZZ→4μ production is one of the necessary steps towards the Higgs boson discovery in its H→ZZ→4μ decay mode. Also, a detailed analysis of ZZ event properties at larger integrated luminosities will allow for probing anomalous triple gauge boson coupling constants. We show that the first few events can be observed with virtually zero expected background as early as L~0.2 fb-1. The CDF experiment just announced seeing the first ZZ event at Tevatron.
  • Underlying Event (Acosta, Bartalini, Field, Kotov). The physics of the underlying event is poorly understood and its modeling at LHC energies, even after tuning using Tevatron data, remains unreliable. Moreover, underlying event processes can be one of the main sources of uncertainties in searches for new physics and various Standard Model physics measurements. Therefore, we are gearing up for the early analysis of underlying event properties right after LHC turn-on. Our group carried out similar studies at Tevatron (Pythia Tune A, just to name one), which puts us in a good position to do it quickly and efficiently at LHC.
  • Missing Transverse Energy Reconstruction (Avery, Cavanaugh, Schmitt, Scurlock, Pi, Yelton) Missing Transverse Energy (MET) is the most difficult physics object reconstructed from the detector response. On the other hand, MET is one of the most important signatures for various physics analyses, including measurements within the Standard Model or searches for new physics. Our group is responsible for the MET reconstruction software in CMS and is actively involved in developing tools for accurate and reliable MET calibration.


Last time updated: November 10, 2006 (AK)