Project Topics
Get Complete Project Material File(s) Now! »
The Chain of Accelerators and the Tevatron
Located near Chicago (USA), the Fermilab hosts many particle physics experiments and accelerators, in particular the Tevatron, a p¯p collider operating since 1988. Several discov- eries have been made at knowlodge such as the γ meson, demonstrating the existence of the bottom quark from the E288 collaboration led by Leon Lederman in 1977 [43], the top quark by the CDF and D0 experiments in 1995 [44, 45] and the tau neutrino in 2000 by the DONUT collaboration [46]. In order to provide high energy collisions, Fermilab benefits from a complex acceleration chain, the most powerful being the Tevatron accelerating beams of protons and antiprotons with an energy of 980 GeV to produce collisions at a center-of-mass energy of 1.96 TeV. The beams are crossing in two interaction points where are located the CDF and D0 detectors.
An aerial view is shown in Figure 2.1 along with a diagram depicting the acceleration chain.
Collisions at the Tevatron occurred during two major periods:
• the ”Run I” when the Tevatron was operating a center-of-mass energy of 1.8 TeV, from 1988 to 1996, collecting 125 pb−1 of data.
• the ”Run II”, starting in 2001 after the Tevatron had its center-of-mass energy in- creased to 1.96 TeV. This period is separated is two phases. The RunIIa period corre- sponds to an integrated luminosity of about 1 fb−1 recorded between March 2001 and March 2006. The RunIIb characterized, among others things, by the installation of an additional layer of detector to the Silicon Microstrip Tracker (SMT) at the closest to the beam pipe (more details in Section 2.2.2). The delivered luminosity was about 12 fb−1 by September 2011, when the Tevatron shut down.
Tracking and Vertexing System
Two majors components are part of the tracking system. As illustrated in Figure 2.11, the Silicon Microstrip Tracker and the Central Fiber Tracker are embedded in a magnetic field generated by the solenoid magnet. Each subpart of the tracking system depicted on the figure will be detailed in this section, Layer 0:
During the 2006 shutdown, an additional layer of the Silicon Microstrip Tracker (detailed in the next Section) has been introduced in the D0 detector [54]. The motivation of adding a new layer at the closest point to the beam pipe are the following:
• recover tracking performances coming from radiation damages.
• better track finding efficiency which is deteriorated at higher instantaneous luminosity.
• impact parameter resolution improvement (see Figure 2.12), impact parameter is the closest distance between the charged track and the primary vertex, impacting b-tagging performances.
The annular space between the beam pipe and the first layer of the SMT has a radius of 22.90 mm, where the detector, cables and support structure have to fit in this gap. This component is built with silicon microstrips sensors, similarly to the SMT. After the addition of the Layer 0, the impact parameter resolution is improved for tracks pT < 5 GeV as shown in Figure 2.12, therefore yielding an overall improvement of 15% in b-tagging performance with respect to the beginning of Run II.
Trigger and Data Acquisition System
The time between two bunch crossing being of 396 ns, the frequency at which events would be recorded by the D0 detector would be around 2.5 MHz. Recording such amount of data is not achievable from a technical point of view, taking into account that some of these events could be detector noise or have a high cross section, which may necessitate to be filtered.
A preliminary selection on objects reconstructed in the detector is imposed in order to reduce the acceptance and the recording rate. This allows to reject for example low pT multijet events which have a very high cross section, and not used in analyses. The trigger system is used for this purpose and is designed in multiple layers (see Figure 2.22), each of them relying on detector information as shown in Figure 2.23.
Data Format and Detector Simulation
The format of the data after passing the 3 levels of the trigger system is the electronic signal coming from all parts of the detector, called also ”raw data”. In order to analyze the data, objects have to be reconstructed. This process is carried by the d0reco1 software. In simulated events obtained from Monte Carlo event generators, objects have to be reconstructed as well, but after detector simulation. This part is taken care by the d0gstar2 software, which is based on the description of the structure and electronics of the D0 detector by Geant3. Hence, simulated events are in the same format as raw data, and reconstructed in a similar manner. In order to reproduce multiple bunch crossing in simulation and detector noise, randomly selected events from data (Zero Bias data) are overlayed to the simulation.
Data and simulation events are stored in ROOT files, with sufficient informations from detector parts. A common format, CAF [65], is used across the D0 collaboration in order to unify software. To facilitate data analysis without dealing with unnecessary big files of data containing irrelevant events for a given analysis, both data and simulation are subdivided in smaller samples. This splitting is performed by a logical OR of several basic object selection and trigger requirements. In the WH analysis, the EMinclusive and MUinclusive skims are used respectively for the electron and muon channels.
Table of contents :
Introduction
1 The Standard Model and the Higgs boson
1.1 The Standard Model
1.1.1 The elementary constituents of matter
1.1.2 The fundamental forces
1.2 The Quantum Electrodynamics Field Theory
1.3 The Quantum Chromodynamics Field Theory
1.4 The Electroweak Sector
1.5 The Higgs Mechanism
1.5.1 The Scalar Higgs Field
1.5.2 Mass Generation for the Standard Model Particles
1.6 Higgs Searches
1.6.1 Theoretical Constraints
1.6.2 Experimental Constraints
1.7 Conclusion
2 The Tevatron and the D0 Detector
2.1 The Chain of Accelerators and the Tevatron
2.1.1 Proton Beam Production
2.1.2 The Main Injector and Recycler
2.1.3 Antiproton Beam Production
2.1.4 The Tevatron
2.2 The D0 Detector
2.2.1 Coordinate System
2.2.2 Tracking and Vertexing System
2.2.3 Preshower Detectors
2.2.4 Calorimeter
2.2.5 Muon System
2.2.6 Luminosity Monitor
2.2.7 Trigger and Data Acquisition System
2.2.8 Data Format and Detector Simulation
3 Reconstruction and Identification of Leptons, Jets, and 6ET
3.1 Tracks
3.2 Leptons
3.2.1 Electron
3.2.2 Muon
3.3 Primary Vertex
3.4 Jets
3.4.1 Jet Reconstruction
3.4.2 Jet Identification and Vertex Confirmation
3.4.3 Jet Energy Scale
3.4.4 Jet Shifting, Smearing and Removal
3.4.5 Summary of the Jet Stratety
3.5 Missing Transverse Energy
4 Tagging of b-quark Jets
4.1 b-jets Properties
4.2 b Jet Identification Algorithm Prerequisites
4.2.1 Taggability
4.2.2 V 0 rejection
4.3 b-jet Identification Algorithms
4.4 MVAbl Algorithm
4.4.1 MVAbl Efficiency
5 Monte Carlo Used in the WH Analysis and Event Preselection
5.1 Overview
5.2 Foreword on the Analysis Work Flow
5.3 Data and Monte Carlo used in the WH analysis
5.3.1 Data Samples
5.3.2 Monte Carlo Samples and Generators
5.3.3 Event Trigger
5.4 Event Preselection
5.4.1 Primary Vertex Selection
5.4.2 Lepton Selection
5.4.3 Missing ET Selection
5.4.4 Jet Selection
5.4.5 Triangular Cut
5.4.6 Vetoes
6 Treatment of the Background and Result of Event Selection
6.1 Reweighting of W+jets and Z+jets Samples
6.2 Multijet Background
6.2.1 Multijet Background Modeling Strategy
6.2.2 Lepton Fake Rates
6.3 Multijet and V+jet background normalization
6.4 b-tagging
6.4.1 Taggability
6.4.2 b-tagging MC Corrections
6.4.3 b-tagged Event Distributions
6.5 Event Selection Result
7 Multivariate Signal Discriminants and Validation Through Diboson
7.1 Boosted Decision Trees
7.1.1 Variable Selection
7.1.2 Training and Optimization
7.2 Rebinning
7.3 Performance
7.3.1 Multivariate Multijet Discriminators
7.3.2 Final WH → ℓνb¯b MVA Analysis
7.4 WZ and ZZ Production with Z → b¯b
7.4.1 Diboson MVA
7.4.2 Result
8 Systematic Uncertainties
8.1 Jet Energy Scale (JES)
8.2 Jet Resolution (JSSR) and Jet ID (EFF)
8.3 Vertex Confirmed Jet (VCJet)
8.4 Lepton-ID
8.4.1 EM-ID
8.4.2 μ ID
8.5 ALPGEN
8.6 Taggability (TAG)
8.7 B-ID
8.8 Trigger
8.9 QCD
8.10 Cross Section Uncertainties
8.11 Parton Density Functions (PDF)
8.12 Plots for Systematic Variations
9 Results on the Higgs Boson Search
9.1 The CLS Method
9.1.1 Sensitivity Estimator
9.1.2 Handling of Systematic Uncertainties and Profile Likelihood Ratio .
9.1.3 Limit Calculation
9.1.4 Systematic uncertainties : sources and treatment
9.2 Limits Obtained in the WH Analysis
9.3 Combinations
9.3.1 Combined search for the Higgs boson with the D0 experiment
9.3.2 Combined Higgs Boson Results from CDF and D0
10 Summary
