Experimental physicists design new technology for the CERN Collider

The Large Hadron Collider is the largest machine on Earth and one of the most complex scientific instruments ever built. It uses powerful electromagnets to propel beams of charged particles at nearly the speed of light and manipulates these beams into controlled collisions that create showers of billions of tiny particles. Most of these particles are not particularly remarkable, but some can reveal the underlying physical properties of our universe.

Operated by the European Organization for Nuclear Research (CERN), the Large Hadron Collider consists of two 27-kilometre circular tubes buried deep underground along the border between Switzerland and France. Powerful compressors remove air from these tubes and beams of particles are propelled in opposite directions through them. The tubes are lined with over 1,200 large magnets that keep the particles centered inside, so they don’t collide with the machine itself.

Along the circuit there are 16 radio frequency cavities – metal chambers that maximize resonance to create a powerful electromagnetic field. This field oscillates 400 million times per second, which separates the particle beams into many bunches. As the particles pass through each radio frequency cavity, their electromagnetic force accelerates the particles to ever greater speeds until they reach their maximum speed – 99.999999% of the speed of light.

Welding and assembly of the superconducting crab cavities of the HL_LHC

Finally, another set of magnets focuses these bunches of particles, directing them to collide into one of CERN’s four main detectors. This results in a rain of particles and a lot of radiation. Sensors in CERN’s detectors must be sensitive enough to detect subatomic particles, and the chips that process this data must be able to record more than a billion particle interactions per second. And all of this has to happen in an environment where radiation levels approach those at the heart of a nuclear reactor.

Experimental physicists at Carleton are validating new sensors and readout chips that will be used in the internal tracker of CERN’s largest detector: ATLAS. The Higgs boson was first observed in ATLAS in 2012, and the facility is being upgraded as part of the High-Luminosity Large Hadron Collider project. Scheduled to be completed in 2027, the facility upgrade will significantly improve the performance of the Large Hadron Collider – and enable experiments aimed at demonstrating the existence of dark matter and other dimensions.

In particle accelerators, luminosity is a measure of how many particles can pass through a particular space in a given amount of time. More particles means more collisions to observe and study. The new Large High-Luminosity Hadron Collider will increase the luminosity of the accelerator by an order of magnitude – and so will increase the number of particle collisions it can generate.

High-luminosity upgrade kicks off with installation of two HL-LHC connecting cryostats

To accomplish all of this, significant hardware updates will be required.

“We need to be able to detect individual elementary particles as a single electron,” says Thomas Koffas, associate professor of experimental particle physics at Carleton University.

“The new sensors are so sensitive that if you breathe on them, they will most likely be damaged.

“But in the ATLAS Inner Tracker, they’ll be exposed to full-throttle radiation. There’s nothing in front of them and thousands of particles will hit each sensor with every collision. We want to be able to catch them all. To see what they are, and decide if we care about a particular collision, or let it go and wait for the next one.

The Inner Tracker has an area of ​​about 200 square meters, and about three-quarters of that will be covered with sensors measuring about 10 centimeters by 10 centimeters. That’s extremely large for a sensor. Most are only a few millimeters in diameter.

“Maintaining electrical performance over such a large area was one of the main challenges,” Koffas explains.

“The sensors must be able to withstand at least half a kilovolt without failing. The larger the surface area of ​​a semiconductor, the more difficult it is to achieve this.

Masters student at Carleton, Robert Hunter, Professor Dag Gillberg and Professor Thomas Koffas

From left to right: Robert Hunter, Master’s student at Carleton, Professor Dag Gillberg and Professor Thomas Koffas (Photo: Justin Tang)

The R&D of the project was led by the optoelectronics and microelectronics team at CERN. Carleton joined the initiative in 2014 and contributed to the stereo ring geometry of the sensor silicon wafer design. Due to the irregular shape of the ATLAS Inner Tracker, eight different sensor shapes are required. To correct the irregularities, the researchers had to incorporate rotation angles into the designs. Final prototypes were approved in 2019 and the first sensors were shipped this spring to Hammamatsu Photonics in Japan.

Inside ATLAS, each sensor will transmit data to application-specific embedded chips (ASICs) that record what they have detected. These chips were custom designed for this application by CERN’s Microelectronics Department, in collaboration with Carleton and Rutherford Appleton Laboratories in Oxford, UK. The ASIC chips are manufactured in Vermont by Global Foundries, and over 300,000 will be installed during the upgrade. Each of them must be able to handle around 640 megabytes of data in the brief moment a particle shower occurs. The stakes are high. If a sensor or chip fails during an experiment, data will be lost. This could prevent a major discovery.

To ensure that all chips and sensors meet rigorous performance standards, each sensor and chip will be individually tested. Carleton is the lead ASIC chip beta tester and will test about a quarter of the sensors. To meet the requirements of the project, Carleton physicists are teaming up with the Department of Electronics and DA-Integrated, a local microelectronics testing company and the only company to date to have demonstrated the ability to test the chips. DA-Integrated was awarded a start-up contract and invited to participate in a tendering process – the first time a Canadian company has been invited to do so.

To avoid damage to the sensors, testing should take place in purified air free of dust and moisture. The electrical performance of the sensors will be tested in a clean room of the Carleton University Microfabrication Facility in the Mackenzie Building, while mechanical performance and a visual inspection will take place at FANSSI Nanofabrication Facility at the Minto Center for Advanced Studies in Engineering.

A silicon tracker being worked on in the ATLAS SR1 cleaning room

The chip test will take place at Integrated AD facility in Stittsville, just outside of Ottawa. There, the processing power of each chip will be validated using a suite of tests developed by experimental physicists from Carleton and the University of Oxford to test prototypes during the R&D process.

“There’s a wafer with over 400 chips on it, and a machine tests each chip in sequence. Within seconds it runs several hundred tests to make sure it’s fully operational,” says Dag Gillberg, associate professor of physics working on the project.

“If it fails a test, the chip is removed and will not be sent to CERN.”

It is essential that each component is up to the task.

“We only have one hit. Once we get started, they’ll stay in the detector for 12 years,” says Gillberg.

“We won’t be able to repair it after a year, if it’s damaged. That’s why we have to be so careful. We have to make sure everything works perfectly. »

Comments are closed.