Research Activity

Pixel Detector: Design, Construction and Operation

The Inner Detector is the first subsystem in ATLAS to see the products of the collisions, so it is compact, highly sensitive and very transparent. It is immersed in a magnetic field parallel to the beam axis. The Inner Detector reconstructs the  trajectories of the electrically-charged particles that cross it, measures their momentum and charge and determines the position of the primary vertices, i.e. where the collisions happen, and secondary, i.e. where some particles raising from the collisions decay.  

The three components of the Inner Detector are: the Pixel Detector, the Semiconductor Tracker (SCT) and the Transition Radiation Tracker (TRT). Meeting the technical specifications for these detectors is a highly challenging task, due to the very intense particle flux they have to record; with LHC operating at design luminosity, at every beam crossing, i.e. every 25 ns, around 40 inelastic proton-proton collisions are produced, generating around 2000 charged particles.

The ATLAS Genova Group played a leading role in the design and the construction of the Pixel Detector currently operating in the ATLAS Experiment. In 2002-2005 one third of the 1744 modules constituting the detector was assembled and tested in the ATLAS Genova Laboratory. In addition, the design of many key ingredients of the acquisition chain (VLSI on-board electronics, detector calibration tools, read-out software interfacing with the rest of the experiment) happened under the responsibility of Genova group members.

In 2014 a new layer (so called IBL) has been added to the current tracker. The technologies used represent a big step forward towards more radiation hard detectors. In particular 3D sensors have been for the first time used in a High-Energy Physics experiment. The ATLAS Genova Group led the construction of the detector and had the responsibility of key parts such as the 3D sensors and the on-detector electrical services. In our laboratory we have built half of the detector.


The Pixel Detector for HL-LHC

The current Inner Detector of the ATLAS experiment will be replaced with a new, all-silicon detector to cope with the tough environment of the High Luminosity LHC (HL-LHC). The instantaneous luminosity of HL-LHC (a factor 5 to 7.5 higher than LHC) will result in harder conditions for the detector: increase in occupancy, bandwidth and radiation damage. The Inner Tracker (ITk) will consist of an inner Pixel and outer Strip detector aiming to provide tracking coverage up to |η|=4: tracking and vertexing performance is expected to be similar or better to the current tracking system in a much harsher environment.

In particular, the new  Pixel Detector which will have a surface of 13 m2, i.e. ~ 7 times larger than the current one. The detector will have five layers of pixel silicon sensor modules in the central region and several ring-shaped layers in the forward region. Italy is responsible to build and deliver at CERN one of the two Pixel Outer End-Cap.

Several are the challenges in the construction of the new Pixel detector in which the Genova group is involved:

  • due to their radiation hardness, 3D sensors are a promising option for the innermost pixel layer while in the other layers planar sensors will be used,
  • The construction of modules for the innermost layer of the detector
  • the construction of the light carbon supports with Ti pipes for the end-cap
  • the development of electrical services with serial powering and cooling with CO2 liquid

The Genova group has links with several Institutes for the construction of the detector: IFAE/CNM (Barcelona, Spain); SLAC, LBNL Berkeley (San Francisco, USA); Rutherford Lab, Liverpool, Manchester (UK); Bonn, DESY-Hambourg (Germany).
Even thanks to the historical importance of the group in the Pixel construction and operation of ATLAS, we are currently steering the HL-LHC tracker project:  Claudia Gemme (INFN) as ITk Project Leader and Paolo Morettini as ITk Pixel Project Leader.


Hardware Track Reconstruction

Not all collisions produced at the LHC provide an interesting physics content. For this reason, the ATLAS Experiment is equipped with a multi-level selection, operating a preliminary real-time data analysis (trigger), to actually save to disk only a small fraction of the data (around 40 in a million).

The current implementation of this real-time selection performs an hardware-based ultra-fast scan of data in the calorimeters (looking for jet candidates) and in the outer muon tracking (looking for muon candidates), which is then followed by a more precise software analysis of data from those detectors, coupled to the reconstruction of the trajectories of charged particles starting from the signals they deposited in the Inner Detector, running on a dedicated commercial CPU farm.

The fact that tracking information is not used right from the start in this process is a strongly limiting factor for all the data analyses heavily based on charged tracks, like the identification of jets containing beauty and charm hadrons (flavour tagging).

To overcome this limitation when operating at the High-Luminosity LHC, the ATLAS Experiment will be equipped with the Hardware Tracking for the Trigger (HTT) system, acting as a massive co-processor for the CPU farm and providing very fast tracking information.

In this context, the ATLAS Genova Group is working on the development of the HTT pattern recognition engine, which has the goal to select and group together all the detector signals produced by the same particle, connecting the dots to reconstruct its trajectory and kinematics.


Identifying Flavour and Structure of Jets

The most abundant objects out of the collisions at hadron colliders are the jets, narrow cones of hadrons and other particles initiated by quarks or gluons. The identification of jets from beauty quarks, commonly referred to as b-tagging, plays a crucial role in high-energy physics experiments for both precise measurements of the Standard Model and for exploring scenarios of new physics. The jets are discriminated as being originated by beauty quarks (b-jets) or charm quarks (c-jets)  instead of light quarks (light-jets) by extracting information based on the displacement of the tracks and on the reconstruction of vertices displaced from the primary interaction.

The ATLAS Genova Group has a long-standing and successful tradition in the development and calibration on real data of the b-tagging algorithms, both when selecting collisions in quasi-real time and when analysing the collisions for physics analyses.

The main current activities focus on improving the b-jet tagging identification for jets with high transverse momentum and on developing a new approach for tagging jets in a boosted environment. The application of modern deep-learning techniques will be the core of this development.


Entering the Era of Higgs Precision Measurements

The Higgs boson, discovered in July 2012 by the ATLAS and CMS experiments, is a very peculiar particle (no charge, no spin, only mass) which, in the Standard Model, is responsible for the origin of the mass. Measuring with high precision the Higgs boson properties is a challenging but very promising research program; exploring this new field of physics will in fact shed light on the ultimate structure of the mass mechanism and of the laws of nature (see https://cerncourier.com/in-it-for-the-long-haul/ for an intriguing perspective on the future of particle physics).

In this context, the ATLAS Genova Group focused on the study of the decay of Higgs boson to pairs of beauty quarks. This decay mode, discovered in July 2018, has still to be fully explored and plays a key role in understanding the Higgs self-coupling and thus its mass.

We contributed to this discovery, studying the “vector -boson fusion” production mode, characterized by a very peculiar decay topology.
Our focus is now shifting to the boosted decay channel where the two beauty quarks, produced in the decay of a high-momentum Higgs boson, are merged in a single reconstructed jet. Developing a deep understanding of jet substructure and of flavour signatures is required, but this challenging task is worth the effort, since it provides the only way to investigate the “gluon fusion” production channel, still completely unknown, despite being the most abundant.null


Beyond the Standard Model: Search for Heavily Ionizing Particles

Many extensions of the Standard Model predict the existence of charged and heavy long-lived

particles. These particles, if produced at the Large Hadron Collider, should be moving at speeds significantly below the speed of light and are therefore identifiable through the measurement of an anomalously large specific energy loss (dE/dx).

The dE/dx measurement is done using the ATLAS pixel detector which provides “hits” that both indicate where the charged particle passed and how much ionization it left in the detector. The simultaneous measurement of the track dE/dx and momentum allows to deduce the particle mass and eventually detect the heavy particle as a mass peak over the smooth expected background.

The ATLAS Genova Group pioneered this largely model-independent search strategy since the beginning of the ATLAS experiment and is now leading the measurement with the full statistics so far accumulated by ATLAS, which represents a 4-fold increase relative to the latest results obtained by the group and published in early 2019. Aim of this new measurement is both to check if the mild excess currently observed around a mass of 600 GeV will be confirmed with the increased statistics and to extend the model-dependent (SUSY gluino) exclusion limit at the highest mass range beyond the currently attained value of about 2 TeV.null


Computing and Distributed Data Analysis

ATLAS records several billion “events” (collisions of high-energy particles) for every year of data-taking, and produces up to 3 times that number of simulated events. The total size of accumulated data amounts, after the first 10 years, to about 200 PB of disk space, plus the archived data on tape. The data are distributed to the computing centres of the collaborating institutes, about 120 all over the world, for calibration, processing and analysis. The ATLAS Distributed Computing (ADC) activity organises all data storage and processing and provides the tools for workflow management and data access to all Collaboration members, using the Grid and Cloud technologies.

The ATLAS Genova Group has a long involvement with the ADC organisation and particularly with the development of tools to ease access to the data for analysis purposes. The EventIndex, the general catalogue of all ATLAS real and simulated events, was designed and developed in Genova and is now under active upgrade, to adapt it to the higher data production rates expected for LHC Run 3 starting in 2021. It is the first project in High-Energy Physics that used BigData technologies from its design phase; R&D work continues in this line.

Genova also hosts a “Tier-3” computing cluster for ATLAS. It provides storage space and computing power for local simulations and analyses, as well as providing, at lower priority, computing power of data generation and processing for the whole Collaboration. It is fully integrated with the ADC system (see above).

This site uses session cookies, necessary for its proper functioning and to manage authentication to online services, intranets and reserved areas. The use of these cookies, which are not stored permanently on the user's computer and are deleted by closing the browser, is strictly limited to the transmission of local session data. Their deactivation could impair the use of part of the online services. more information

The cookie settings on this website are set to "allow cookies" to give you the best browsing experience possible. If you continue to use this website without changing your cookie settings or you click "Accept" below then you are consenting to this.

Close