ATLAS

ATLAS is one of the largest and most complex particle detectors ever built, one of the world’s leading tools in the search for beyond-Standard Model physics and a key TRIUMF international collaboration.

Based at CERN as part of the Large Hadron Collider (LHC), ATLAS is one of the largest scientific collaborations in history. As part of the 10-institution ATLAS-Canada collaboration (TRIUMF and nine universities), TRIUMF scientists, engineers and technicians have provided critical expertise to all aspects of ATLAS’ success.

This includes detector design, construction and installation, accelerator science, high-performance scientific computing (See TRIUMF’s ATLAS Tier-1 Computing Centre), data analysis, theory and large-scale scientific project management.

TRIUMF’s involvement in ATLAS is an outgrowth of the lab’s contribution to the LHC. From 1995 to 2005, TRIUMF managed Canada’s $41.million contribution to the LHC’s construction. Leveraging TRIUMF’s accelerator expertise, Canada made important in-kind contributions, including to the injection kickers, twin-aperture quadrupole magnets, and injector synchrotrons.

ATLAS is renowned for the historic 2012 discovery of the Higgs boson, the final predicted piece in the puzzle of the Standard Model (SM). TRIUMF staff and technology played an essential role in the Higgs boson discovery, including building parts of the detectors that recorded it, the intensive Higgs boson data analysis, and the high-performance computers and software that provided data reduction and modeling.

However, ATLAS is also designed to explore beyond-SM physics, including searches for dark matter and supersymmetric particles, and completely new and unimagined physics. To this end, the global ATLAS science team is in the midst of a decade-long series of staged upgrades to ATLAS to super-charge it for the High Luminosity LHC (HL-LHC) era, when the collider will provide multiple times the design beam intensity opening the way for postulated beyond-SM physics discoveries.

The upgrades will increase ATLAS’ radiation hardness and the detector’s overall resolution and speed, and the TRIUMF-ATLAS group is making major contributions to three of the planned major upgrades.

ATLAS involves more than 3000 scientists from about 182 institutions in 38 countries, with more than 1200 doctoral students involved in detector development, data collection and analysis.

The TRIUMF ATLAS scientific group is a Canadian keystone for the participation of over one hundred Canadian researchers, graduate and undergraduate students at nine ATLAS-Canada participating universities. The TRIUMF ATLAS group includes TRIUMF researchers, emeriti, and affiliated scientists; co-appointed researchers based at Canadian universities; post-doctoral research associates based at TRIUMF and CERN; and graduate students from various universities who are supervised by TRIUMF research scientists.

Beyond its historic physics discoveries, ATLAS is driving and inspiring a broad array of related science and technological advance, including in nuclear and medical research. In particular, ATLAS has come to define the era of big data and its scientists and data analysis techniques are fuelling big data applications from e-commerce to fintech and e-health. TRIUMF’s participation is a key conduit for bringing these ancillary benefits to Canada.

The LHC, the world’s most powerful particle accelerator, is a 27 km ring of superconducting magnets with a number of accelerating structures to accelerate two beams of protons in opposite directions to 99.999999% the speed of light. The beams inside the LHC are made to collide at four detector locations around the accelerator ring – CMS, ALICE, LHCb, and ATLAS.

Almost half as long as a football field and weighing 7,000 tonnes (about the mass of the Eiffel Tower), ATLAS is the largest volume detector ever constructed for a particle collider. Located in a cavern 100 m below ground, the cylindrical ATLAS detector, 46 m long and 25 in diameter, consists of six different detecting subsystems wrapped concentrically in layers around the collision point to record the trajectory, momentum, and energy of particles, allowing some of them to be individually identified as distinct particles, from muons to electrons.

The proton-proton collisions at the centre of the ATLAS detector produce collision debris in the form of new particles which fly out in all directions. Presently, more than a billion particle interactions take place in the ATLAS detector every second, but only one-in-a-million proton-on-proton collisions triggers the data collection system as interesting physics and thus recorded for further analysis.

ATLAS is a scientific and engineering marvel in the way it is constructed and upgraded. It was built, and is being upgraded, through a globally distributed collaboration that involves thousands of individual components. These components are designed, constructed, tested and assembled at various national particle physics laboratories, universities and industries in 38 countries, and then shipped to CERN for assembling into final ATLAS structures.

From the design and construction of original ATLAS components to planned upgrades, TRIUMF researchers, detector engineers and technicians have played, and continue to play, key roles.

The ATLAS collaboration is divided into six sub-groups, each responsible for one of the six key detector sub-components. TRIUMF significantly contributes to three of these: the Liquid Argon Calorimeters; the Muon Detectors; and the Computing Systems. TRIUMF brings its detector expertise to both calorimetry (measuring particle energy) and tracking (measuring particle trajectories to determine their charge and momentum).

TRIUMF’s ATLAS participation is tightly linked to the ATLAS experimental timeline. The ATLAS detector operates on a schedule of experimental runs, and long shutdowns (LS), when maintenance and major upgrades are performed. The first run was from November 2009 to 2012 (during which the data for the discovery of the Higgs’ boson was collected), followed by a LS1 from 2013 to spring 2015. Run 2 will end in autumn 2018, with LS2 scheduled from 2019-2020. Run 3 will extend from 2021-2023, and LS3 from 2023-2025, after which ATLAS will begin its High Luminosity era, when it will operate at several times the design intensity, enormously boosting collision rates and associated data collection.

In successive runs, ATLAS is using increasingly intense beams (higher luminosity) in order to generate more very rare high energy collisions. In Run 1, the number of collisions per bunch crossing (when the two proton beams collide) was about ten. In Run 2, it’s 40 on average; Run 3 is predicted to be about 50, and in the HL era it will be about 200. This increasing beam intensity produces both data and radiation, and these changes are driving most of the planned ATLAS upgrades.

TRIUMF is significantly involved in three of the major ATLAS upgrades planned for LS2 and LS3: Digitizing the trigger for the Liquid Argon Hadronic End-cap Calorimeters; the new Muon Small Wheel detector; and the replacement of the present Inner Detector with a new Inner Tracker.

In collaboration with 14 national ATLAS teams and the ATLAS-Canada collaboration, TRIUMF-ATLAS researchers, engineers and technicians helped design, build, test, and assemble two of ATLAS’ original Liquid Argon Hadronic Endcap Calorimeter (HEC) wheels. (For a detailed description, see pp. 158-159 here).

A calorimeter is a detector that measures a particle’s energy. At either end of ATLAS’ 360-degree Liquid Argon Hadronic Calorimeter system is a HEC, used to measure the energy of hadrons (particles composed of quarks, such as pions) emerging at low angle (less than 25º) from proton-proton collisions.

Each HEC is a massive, integrated detector weighing 250 tons. At slightly more than four metres in diameter, it’s made up of two wheels, (0.8m and 1.0m thick) each composed of 32 wedge-shaped modules arranged to form the wheel shape. Each module is a layered design of dozens of 25 or 50 mm thick layers of copper, interspersed with 8.5 mm gaps filled with liquid argon and read-out electronics, all cryogenically cooled to -186ºC (87 K).

When high energy hadrons from a proton collision traverse the copper, they produce a hadronic shower in which the collision debris’ initial energy is transformed into lower energy hadrons, including protons and neutrons ejected from copper nuclei. These charged particles subsequently strip electrons from argon atoms, leaving a trail of electron-argon ion pairs. A strong electric field causes the free electrons to rapidly drift to the positive side, producing a distinctive electric signal that’s collected by HEC’s front-end electronics. Based on the shape and combinations of these electric signals, ATLAS scientists can help identify the particles involved.

TRIUMF worked closely with Canadian university-based ATLAS team members, including a number of cross-appointments, on the design, construction and assembly of HEC components. With University of Alberta researchers, TRIUMF designed radiation-hard front-end electronics for the HEC and supplied and contributed to the machining of the copper plates. TRIUMF participated in the design and engineering of the liquid argon cryostat feedthrough project at the University of Victoria. TRIUMF-ATLAS members also worked with ATLAS-team members at Carleton University and the University of Toronto to produce the tungsten components for two of the ATLAS forward calorimeters, located inside the central bore of the HECs.

The TRIUMF-built components operated at 99.6% efficiency during Run 1.

The Phase 1 and 2 upgrades to the HECs are a staged program to improve the speed, resolution and radiation hardness of the HEC detector trigger and front-end electronics. In Phase 1, the key upgrade will be the digitizing of the primary trigger data, providing faster, higher-resolution raw data for trigger determination.

The original HEC trigger system is based on electronics hardware that analyzes analog signals–electronic pulses with a distinctive shape and time structure–from a sub-region of the detector to determine if the signal represents interesting physics. The speed and resolution of this trigger system is essential since only 1-in-40,000 collision events is eventually recorded. Moreover, in coming Runs, the trigger system will have to handle increasingly higher rates of data.

To prepare for this, the Phase 1 upgrade involves the addition of a new Trigger Digitizer Board to the HEC front-end electronics. This will enable raw data to be digitally compressed and thus provide the trigger with higher-resolution information for keep/reject decision making.

 The digital electronics upgrades must be accommodated and integrated within the existing electronics crate. To help facilitate this, TRIUMF-ATLAS, and ATLAS-team members from the University of Victoria have designed, prototyped and built a new HEC front-end-crate Baseplane.

The HEC Baseplane is a highly customized, multi-layered circuit board with a plethora of connectors that integrates the new Trigger Digitizer Board within the existing electronics and enables the higher rates of digital data transfer. The upgrade will maintain the original analog functionality to ensure a smooth transition and redundancy during commissioning of the digital system, and the Baseplane coordinates this digital-analog routing and integration.

The new front-end Baseplanes have gone through a successful, rigorous step-wise process of pre-production and acceptance tests, characterization tests to confirm their high-resolution performance, and compatibility tests at CERN, and are now in full production. TRIUMF and University of Victoria researchers are collaborating on the quality assurance testing of the final baseplanes and upgraded trigger baseplanes will be installed during ATLAS’ LS2 from 2019 to 2020.

During LS 3 (2023-2025), in preparation for ATLAS’ high-luminosity era, the HEC front-end electronics will be completely upgraded to a more radiation-resistant fully digital trigger system, and the TRIUMF-ATLAS team is presently designing, and will construct, key components of this upgrade.

Building on the digital triggering system installed during the Phase 1 upgrade, the Phase 2 upgrade will involve replacing the front-end board, which receives the initial signal from the detector, with radiation-hard electronics that provides digital signal compression in order to handle the higher collision rates ATLAS will experience in upcoming runs.

During the LHC’s high-luminosity phase, there will be about 200 proton-proton collisions per bunch crossing, ten times the average number in Run 1. Under these conditions, the current ATLAS trigger system wouldn’t be able to handle the pile-up conditions creating the equivalent of a data traffic jam in which some data will be discarded in order to clear the build-up. Similarly, the front-end electronics will have reached its radiation tolerance by this point. Thus, HEC’s entire front-end and back-end readout electronics will be replaced during LS3.

As part of the international ATLAS HEC team involved in this Phase 2 upgrade, the TRIUMF-ATLAS team will contribute to the design, prototyping and production of the new Front-End ASIC-based analog chip for the new HEC Front-End Board.

The main Phase 1 ATLAS upgrade project for the ATLAS Muon Detector System is the complete replacement of the existing forward (nearest the beamline) Small Wheel (SW) muon detectors with New Small Wheels (NSWs), to be installed during the 2018-2019 LS.

Muons–heavier sister particles of electrons with half-lives of two micro-seconds–are a key product of LHC proton-proton particle collisions, and a key trigger of an interesting one. Muons interact minimally with ATLAS’ calorimeters and are instead detected and tracked by a massive, multi-stage Muon Detector System bookending the ATLAS calorimeters and Inner Detector.

However, ATLAS’ toroid magnets become irradiated by the bath of neutrons that result from beam interactions with the detector, the irradiated magnet metal radioactively decays, emitting particles that mimic muons. Already, about 90% of the detected muons are such “fake muons” – other particles originating from radioactive decays – rather than muons from proton-proton collisions.

The original muon SW is in part designed to weed out these fake muons. If a particle is detected in the trigger chambers of the muon “Big Wheels” beyond the toroid magnets but its trajectory can’t be linked back to a detection in the SW, this indicates that it didn’t come from the proton-proton collision and it’s rejected. However, the original SW doesn’t have a high enough resolution to provide for efficient removal of the fake muon signals. This situation will be greatly exacerbated during the LHC’s upcoming high luminosity era when the fake muon signals will be so numerous that they would swamp the real muons, forcing most of the good muon events to be thrown away for lack of trigger bandwidth.

TRIUMF-ATLAS Contribution

In collaboration with five Canadian universities in the ATLAS-Canada team, TRIUMF is making a major contribution to the NSWs, building one-quarter of the upgraded small-strip Thin Gap Chambers (sTGCs), gas wire detectors that are one of the NSWs’ two key detector technologies. The remaining sTGCs are being built in China, Russia, Israel, and Chile.

The NSWs will be capable of simultaneous precision tracking and high-speed triggering in order to both weed out fake muons and capture the trajectories of interesting ones. In coordination with the other muon detectors, the NSWs’ triggers will provide location resolution of about 10 microns, about the diameter of a human hair. This will provide the overall muon trigger system with reconstructed track segments of good angular resolution, clearly indicating whether or not the triggered muons originated from the collision point.

Each of the two NSWs, one at each end of ATLAS, is a roughly circular detector nine metres in diameter, composed of eight large, and eight small segments, staggered one in front of the other resembling the petals of a flower.

The NSW uses two detector technologies: Micromegas as the primary precision tracker and sTGCs as the primary trigger. Each of the 16 NSW sectors is composed of two layers each of Micromegas, the inner layers, and two layers of sTGCs, sandwiching the Micromegas. Each sTGC pizza-shaped segment is composed of three sections, termed ‘quadruplets’ because each of these sections consists of four precisely arranged layers of sTGC chambers.

The ATLAS-Canada collaboration is constructing a total of 54 quadruplets (including six spares), 32 of the larger NSW segments, and 16 of the smaller.

To achieve this, the ATLAS-Canada collaboration has created a customized, small-scale industrial supply chain, with the TRIUMF-ATLAS team playing key roles in quality control, half-gap assembly, and final wedge assembly at CERN. TRIUMF’s role begins with providing quality control for the custom manufacturing of the sTGC’s base cathode boards at Triangle Labs in Nevada. There, on-site TRIUMF detector group engineers have helped overcome significant production hurdles. There is also a constant on-site presence of TRIUMF technicians to perform quality control.

Next, the custom cathode boards arrive at TRIUMF for manufacturing and assembly into sTGC “half-gaps”. In collaboration with ATLAS members from Simon Fraser University and University of Victoria, the TRIUMF-ATLAS team has built two custom facilities for this: a custom spray booth for the spray application of a resistive graphite onto one side of the cathode boards; and a specialized assembly room with three very precisely flat large granite tables with vacuum systems for sucking the boards down on them and applying even pressure during gluing.

Once a cathode board is graphite coated on one side, the TRIUMF-ATLAS technicians build it into a half-gap, one side of a small Thin Gap Chamber (i.e. two half-gaps are combined to create a chamber).

After graphite-coating the cathode board and hand-polishing the graphite to a uniform resistivity, TRIUMF-ATLAS technicians glue 1.4 mm-thick insulating precision frames to the edges of the cathode board to hold the grille of gold-plated tungsten high-voltage wires, and a serpentine network of wire supports that guide the gas in the chamber. This half gap assembly requires exacting standards: each half-gap must be constructed with less than 40-millionths-of-a-metre variance from its nominal thickness between them to ensure precision overlap once assembled into quadruplets.

The TRIUMF-built half-gaps are shipped to Carleton University, where ATLAS team members for wiring and assemblage into quadruplets. Then, they are shipped to McGill University ATLAS team members for testing, before shipping to CERN. Finally, TRIUMF-ATLAS members there play an important role as part of the CERN-based team assembling the quadruplets into the sectors for mounting on the NSW metal frame.

The Inner Detector (ID) is a cylindrical with endcap discs immediately surrounding the LHC proton-proton collision site. As such, the ID experiences intense radiation and, over time, radiation damage. During LS3, prior to the even higher radiation levels of ATLAS’ High Luminosity era, the ID will be completely replaced by a new, all-silicon Inner Tracker (ITk). The ITk will provide both significantly higher detector resolution and more radiation tolerant sensors. The ITk upgrade, now entering the construction phase, includes extensive TRIUMF-ATLAS involvement in construction and final design.

This upgrade is a major international collaboration, involving 20 countries, with strong participation from the ATLAS-Canada group (TRIUMF and seven Canadian universities).  Collaboratively, ATLAS-Canada members are providing 1500 of the approximately 7,000 sophisticated silicon strip modules that will form the ITk endcaps. The assembly of sensors and hybrids into completed silicon strip modules will be performed equally at TRIUMF (500 modules), Simon Fraser University (500 modules) and the University of Toronto (500 modules). Each module consists of a silicon strip sensor. Each module consists of a silicon strip sensor (about 10×10 cm²) with radiation-resistant, customized read-out electronics (ASICs and power boards) attached.

Each sensor consists of two or four rows of strips, about 25 mm long and only 80 micrometers wide. Particles crossing the sensor leave signals in the strips they cross which can be detected by the electronics. Knowing which strip was crossed allows the trajectory of the particles to be very precisely determined.

TRIUMF will mount the Canadian modules on support structures called petals, which provide electrical, mechanical and cooling services. The petals will be shipped to Europe for mounting into the two ITk Strip Endcaps: each endcap will have six wheels of 32 petals.

TRIUMF contributions: Microelectronics Clean Room

The SFU and TRIUMF-ATLAS team helped create a dedicated microelectronics clean room for the assembly, prototyping, testing and eventual full-scale manufacturing, scheduled to begin in 2020, of the ITk silicon tracking modules.

The ASICs are mounted on bare hybrids to form an assembled hybrid by ATLAS-Canada’s industrial partner Celestica. The ASICs are binary readout chips built in 130 nm CMOS technology, known as ABC130.

At TRIUMF, the hybrids are glued directly to the silicon sensors forming complete modules. This requires highly precise, microscopic wire connections as each ABC130 has 256 binary readout channels, each of which is directly wire-bonded to the sensor strip. The clean room contains a fully automated, high-speed wire bonding machine that creates thousands of four-layered bonds with two-micron precision for each module. The lab includes metrology analysis equipment to verify the micron-level precision and also equipment for testing the electric integrity of the newly wired circuits. Quality control steps are distributed across all ATLAS-Canada ITk sites to enable parallel testing to expedite this major detector construction project.

The ITK end-caps are formed from modules mounted on a series of lightweight carbon support structures called petals and arranged like the petals of a flower. The TRIUMF-ATLAS team is currently providing detector design expertise on the final layout of the silicon sensors and petal assembly to maximize the ITk’s performance. ATLAS members from SFU and TRIUMF are leading the development of fully automatic mounting of modules onto petals using a robotic gantry system.