ATLAS Experiment heavy-ion reaction plane detectors, built at Loomis Lab

Jamie Hendrickson and Siv Schwink
for Illinois Physics Condensate

UIUC Nuclear Physics Laboratory researchers and students develop novel reaction plane detectors and new machine learning algorithms for the Large Hadron Collider

 

The world’s largest particle accelerator, the Large Hadron Collider (LHC) at CERN in Switzerland, smashes tiny particles together at high energy to study the fundamental constituents of matter and the forces that bind them together. Most of the time, the particles that the LHC collides are protons, but each winter for about a month, the beams are switched to heavy-ion particles—usually the nuclei of lead atoms—to enable the heavy-ion experiments that since 2010 have been garnering increased interest from theorists and experimentalists looking for new physics.

“Using machine-learning algorithms to measure the nuclear collision plane has not been done before, and we’re looking forward to testing our instrument and new analysis techniques.”

Illinois Physics Professor Matthias Grosse Perdekamp

Since 2010, CERN scientists have been planning for the next-generation High-Luminosity Large Hadron Collider (HL‑LHC), now slated to come online in 2029. The HL-LHC will usher in a new era in particle physics, enabling precision studies of the LHC’s findings to date, yielding potential new discoveries, and expanding our understanding of the fundamental interactions of matter. But first, major technological upgrades to multiple systems must be designed, built, tested, and installed in one of the most extreme laboratory environments on Earth.

Work on those upgrades is well underway. As part of this international effort, the  UIUC ZDC‑RPD Group in Urbana has designed and constructed novel reaction plane detectors (RPDs)—the first-ever RPDs to be installed in the ATLAS zero-degree calorimeter (ZDC)—for use in heavy-ion collision experiments. The project is led by principal investigator Illinois Physics Professor Matthias Grosse Perdekamp and managed by co-principal investigator, Illinois Physics Research Scientist Riccardo Longo.

Perdekamp explains, “In such experiments, ZDCs and RPDs together let us characterize the geometry of lead-lead collisions. The ZDC measures the degree of                  overlap between two colliding lead ions. The RPD measures the orientation of the two colliding nuclei in relation to the laboratory reference frame.”

Characterizing the collision geometry will enable ATLAS scientists to study the physics of the quark gluon plasma (QGP), the ultra-hot matter present during the first microseconds of the universe following the Big Bang.

Enhancing detector capabilities

The ATLAS RPDs employ a novel design and material, developed by the Perdekamp team, to withstand the extreme radiation inside the collider. The new detectors will employ machine-learning algorithms to achieve a two-dimensional mapping of the collisions’ neutron showers, in a transverse profile. The algorithms were written by Illinois Nuclear, Plasma & Radiological Engineering alumnus Sheng Yang while he was a student researcher in the UIUC ZDC‑RPD Group. If the software meets performance expectations, ATLAS researchers around the world will be able to use nuclear collision geometry information in their studies of the QGP formed in nuclear collisions at the LHC.

Perdekamp comments, “Using machine-learning algorithms to measure the nuclear collision plane has not been done before, and we’re looking forward to testing our instrument and new analysis techniques.”

The performance of the RPDs and algorithms in Run 3 will serve as a pilot program for what the Perdekamp team plans to build for Run 4 and the high-luminosity era.

High luminosity refers to higher particle collision rates and the associated increase in the volume of experimental data collected. It should be noted that for heavy-ion collision the main increase in luminosity will occur in Run 3, with small additional increases for Run 4. 

“Increases to proton beam intensity and to lead-ion beam intensity are separate undertakings,” notes Perdekamp. “Comparing LHC’s Run 2, completed in 2018, with Run 3, the current experimental campaign, the integrated luminosity is increased by a factor of 2 for proton-proton collisions and by a factor of 5 for heavy ion collisions. In Run 4, starting in 2029, the HL‑LHC will again double the proton-proton collision rate, but will increase the heavy-ion collision rate only slightly.”

Where particle physics and
heavy-ion nuclear physics overlap

Some of the most promising experimental approaches to studying new fundamental phenomena in collider experiments are being proposed at the intersection of particle and nuclear physics. The annual switch over to heavy-ion collisions at CERN is rooted in two closely related lines of scientific inquiry.

First, heavy-ion collisions are the only way to produce the quark-gluon plasma (QGP) in a laboratory setting, recreating the conditions of the early universe. The QGP is theorized to have existed in the first 10–20 microseconds after the Big Bang. As the universe expanded and cooled, quarks and gluons condensed into protons, neutrons, and other hadrons, the building blocks of nuclear matter as we know it. 

The QGP cannot be directly observed. Its properties are studied through the experimental signatures it leaves behind in the aftermath of heavy-ion collisions. When heavy nuclei collide, the hundreds of protons and neutrons that make up the nuclei release a large portion of their energy into a tiny volume, momentarily liberating quarks and gluons from their bound states in protons and neutrons to form a hot, dense plasma of free quarks and gluons. Just as quickly as the plasma forms, it expands, cools, and condenses into composite particles and anti-particles, bound quark states including protons, pions and kaons, which zoom out in all directions from the collision point.

Second, heavy-ion runs allow scientists to test at the highest temperatures and densities ever achieved in a laboratory setting the predictions of quantum chromodynamics (QCD)—the theory of the strong force that binds quarks and gluons together into protons and neutrons inside the nuclei of atoms. Outside of the accelerator laboratory environment, quarks and gluons cannot be pried apart—the further apart the particles, the more tightly the strong force binds, like a Chinese finger trap.
Longo notes, “Our group is most interested in the QCD-related questions enabled by data collected within the heavy-ion program. But our new RPDs will enable new measurements of particle flow, providing insights into how the QGP expands after its formation in heavy-ion collisions, and graduate students in our group do plan to carry out QGP analyses as well, once the RPDs are operative.” 

Design challenges

The ATLAS RPD has been developed in the context of a cooperative effort between the ATLAS and CMS ZDC groups, in the first-ever joint instrumentation project between the otherwise fiercely competing collaborations. To minimize systematic uncertainties resulting from collision-geometry characterization in lead-lead collisions, the two experiments will use identical ZDCs in Run 4. From the ATLAS side. the institutions involved in the Joint Zero-Degree Calorimeter Project (JZCaP) are UIUC, Ben–Gurion University of the Negev, Columbia University, and Brookhaven National Laboratory. CMS is participating with groups from Kansas University and University of Maryland. JZCaP scientists meet weekly to coordinate instrumentation design, Monte Carlo simulations, test-beam measurements, software development, and detector construction. 

It has not yet been determined whether CMS will use the same RPD design or the openly shared machine-learning software developed in Urbana. The decision will likely wait until the new innovations are deployed in Run 3 and proven successful.

According to Longo, the most significant RPD design and construction challenges faced by the Urbana team included finding a radiation-hard material that could withstand the harsh radiation environment, to maintain stable performance over the running period; constructing a detector that could easily be installed and connected using available remote-handling technologies at CERN, to reduce radiation exposure to personnel; designing a detector capable of providing insight on the distribution of non-interacting particles, because the reaction plane of heavy-ion collision is deduced from “spectator” neutrons; and accomplishing all of this within strict size limitations, because two out of three dimensions are on the order of a few centimeters. 

The ATLAS Run-3 RPDs each comprise 256 fused silica-core optical fibers of four different lengths, grouped into 16 channels in a novel pan-flute-shaped design. 

First deployment of the new ATLAS RPDs, originally slated for November 2022, is now scheduled for the end of 2023. CERN cut short its 2022 season in response to the European energy crisis. The Urbana team took advantage of the early shutdown to ensure a smooth installation in 2023. 

Longo notes, “We have studied in great detail the integration of our RPD with the machine and with ATLAS Data Acquisition System. However, when operating a new system for the first time, there are usually unexpected issues that must be quickly resolved during operations. We decided to take advantage of the additional time provided by the first heavy-ion run getting postponed until next year to install the RPDs and fire up the ATLAS trigger and data-aquisition system, in order to check for any integration or commissioning issues. Graduate student Matthew Hoppesch travelled with me to CERN to complete this full system test. Graduate students Mason Housenga and Chad Lantz and undergraduate students Yi Liu and Maya Vira supported the dry-run data-taking remotely, from Urbana.”

It will not be possible to overhaul the Run-3 RPDs and ZDC for Run 4, because the beam and magnet configuration in the tunnel will change significantly, reducing the horizontal space available for the detectors by a factor of 2. For this reason, new ZDCs and RPDs will be constructed for both ATLAS and CMS. Illinois Physics is leading this $2.5 million Department of Energy–funded upgrade.

Training next-generation scientists

Over the past two years. Longo has overseen and mentored 7 graduate students and 25 undergraduate students in the UIUC ZDC‑RPD Group in Urbana, during the R&D, building, and testing of the new RPDs. “I believe that a crucial part of our role as researchers is to help train the next generation of scientists,” he shares.

Physics graduate students Matthew Hoppesch, Mason Housenga, Chad Lantz, Aric Tate, and class of 2023 undergraduate Farah Mohammed Rafee all had the chance to visit CERN over the past two summers and to take part in testing the RPD prototypes. Daniel MacLean, an Illinois Physics alumnus, now an engineering physicist in the accelerator division at Fermi National Laboratory, served as an academic hourly in the group for over two years.

“Gradually involving young scientists in a project like this and in operations at CERN is of incredible value,” Longo asserts. “It allows them to get in touch with a vibrant international scientific environment, fueling their interests. And it gives them hands-on experience of the applications of the research that they carried out in our lab at UIUC.”

Perdekamp adds, “I couldn’t be prouder of the hard and ingenious work of our students, technicians, and postdocs who successfully carried out this challenging project during the pandemic and deployed the two detectors successfully at CERN.” 

This research is funded by the National Science Foundation under Grant Nos. NSF PHY 18-12377 and NSF PHY 21-11046. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the scientists and do not necessarily reflect the views of the National Science Foundation.

 

Illinois Physics alumnus Daniel MacLean (BS, 2019) worked as an hourly technician and research assistant in the ATLAS ZDC-RPD group after graduation, under Longo’s supervision. He is now an engineering physicist at Fermi National Laboratory.

 

MacLean recalls, “I began working for the ATLAS ZDC group in September 2019. My primary task was the mechanical and electrical design of the ATLAS Reaction Plane Detectors. I also oversaw the manufacturing of the detector components, did most of the detector assembly, and, in parallel, managed our group’s laboratory facility in Loomis. 

“The first stage was a single prototype detector as a proof-of-concept, which we completed in 2021 and tested in the Super Proton Synchrotron at CERN that September. Using that experience, we improved many aspects of the detector during my next task, which was to redesign and build the two identical (mirror-image) ATLAS RPDs for Run 3 in the LHC. These were completed and commissioned at CERN by Dr. Longo and I in the Spring of 2022.

“I have since moved on from the group and now work in the Accelerator Division at Fermi National Accelerator Laboratory as an engineering physicist, specifically at the FAST/IOTA Facility. I can say with no doubt that the three years I worked with Dr. Longo in the ATLAS ZDC-RPD Group were the most important of my life in terms of learning and honing the technical skills I now use every day at my new job. It solidified my desire to work in the field of particle accelerators, which I would not have been able to do without the experience I gained as part of the ATLAS ZDC-RPD Group. It is difficult to overstate how lucky I was to have the opportunity to do that job; it has completely altered the trajectory of my life in the most positive way possible.”

 

CERN and the European Energy Crisis

Located astride the Franco-Swiss border near Geneva, CERN has actively been making sustainability efforts to decrease its energy consumption over the past decade and has already succeeded in reducing the laboratory’s overall energy consumption by 10 percent. 

In light of the current global energy supply and cost crisis that has been escalated by Russia’s war on Ukraine, CERN announced on September 30, 2022 that it would take steps to significantly reduce its energy consumption into 2023. This included moving the 2022 year-end-technical stop (YETS) up by two weeks to November 28, and reducing the operation of the accelerator complex by 20 percent in 2023. Additional measures are being taken to optimize the buildings’ energy usage in accordance with the EU’s recent emergency regulation for its member states to reduce electricity usage by 5 percent during the peak hours of the upcoming winter. 

“This choice was made by CERN as a part of its social responsibility,” Illinois Physics Research Scientist Riccardo Longo explains, “so as not to induce additional energy cost increases for people residing in France. It was agreed that this reduction in beam time should be borne equivalently by proton-proton and heavy-ion programs, so, after several discussions among the CERN experiments, it was decided to cancel the 2022 heavy-ion run and to slightly extend the heavy-ion runs in 2023, 2024, and 2025. On the current schedule, our new detector, built for this fall’s heavy-ion run, will first be used in fall 2023.”