Andy Jung had a secret. And he wasn’t the only one.
Jung and many of his colleagues on the CMS experiment were in Traverse City, Michigan, at the 2023 International Workshop on Top Quark Physics. Jung was listening attentively as physicists announced an exciting finding: The ATLAS experiment—the other general purpose discovery experiment at the Large Hadron Collider—had measured entanglement at the LHC.
“I don’t think it happens very often in the career of a scientist that you have something that is exciting and new, and that no one had done before.”
Entanglement is a fundamental property of the quantum world. It occurs when two particles become inextricably linked, such that any measurement of one immediately influences the other—whether the particles are one Planck length or one universe apart.
Since Albert Einstein first proposed entanglement in the 1930s, physicists have studied it in small laboratory experiments using beams of photons and cold atoms. (Some of this work earned the 2022 Nobel Prize in physics.) They also use more complicated systems like superconductors. And entanglement is the basis for quantum computation.
But in the last five years, scientists began to wonder if they could look for entangled particles at the LHC, where beams are 12 orders of magnitude more energetic. At TOP23, a physicist from the ATLAS experiment was announcing that they’d done just that.
Jung remembers squirming in his seat in the audience. The CMS experiment had recently seen the same thing as ATLAS. But the CMS collaboration was still running the nascent measurements through their rigorous internal review process, checking and re-checking to ensure the analysis’ veracity and compatibility with other ongoing analyses on the same topic. Only once all the final checks were completed would CMS announce the measurement. Until then, it had to stay under wraps.
“It was very hard,” Jung says. “That’s part of why it’s exciting. I don’t think it happens very often in the career of a scientist that you have something that is exciting and new, and that no one had done before.”
Entangled paths
In 2019, four years before TOP23, Juan Muñoz de Nova was doing postdoctoral research in condensed matter physics at the Technion in Haifa, Israel. Down the hall was Yoav Afik, at the time a graduate student on ATLAS searching for dark matter at the LHC. The two physicists struck up a conversation in the hallway, then gradually became friends over coffee breaks where’d they’d talk about their lives and research ideas.
During one fateful chat, Muñoz de Nova asked Afik if he thought it was possible to measure entanglement at the LHC. Afik was intrigued. He knew they’d need to find the right high-energy system, and eventually landed on top quarks entangled with their antimatter counterparts, antitop quarks.
The top quark is the most massive elementary particle, first discovered at the Tevatron in the United States in 1995. The Tevatron produced about 10,000 top quarks in its three-decade lifetime, whereas the LHC produces millions every run. Still, they are considered rare: The LHC generates a top quark pair about once a second, even though particles are colliding every 25 nanoseconds.
Unlike the stable photons used in other entanglement experiments, top quarks live for just 5 times 10^-25 seconds before decaying into other particles. The instability carries a benefit. The particles decay so quickly that they don’t experience effects that could cloud the measurement of entanglement. Information about the quark pairs’ spins can be gleaned from the particles into which they decay, leaving a signature of entanglement behind.
Muñoz de Nova and Afik outlined this rationale for studying entanglement in top quarks at the LHC in a preprint posted to arXiv in March 2020. Several journals rejected the paper, Afik says.
High-energy physicists use quantum mechanics in their calculations all the time. The Standard Model is a quantum mechanical theory. Many just didn’t see the appeal of spending time searching for proof of it, Afik says.
“We assume quantum mechanics in our calculations, but the things that we measure in colliders are typically not quantum,” he says. “Our proposal is evidence of quantumness in a collider.”
After publishing the preprint, Afik began campaigning for the project within ATLAS. The idea gained traction, particularly with a group that was studying spin correlations between top and anti-top quarks. In September 2021, the European Physical Journal Plus published the paper, and shortly after Afik got approval to bring Muñoz de Nova on as a part of the analysis.
The preprint had also caught someone else’s attention: Jung, who was doing research on CMS. He had been studying top quarks since 2010, and was getting interested in quantum computing. He worked on a small demonstration of how quantum computers could make it easier to track particles’ paths through detectors. The work got him thinking in the language of quantum mechanics and looking for other applications. And Afik and Muñoz de Nova’s paper was exactly what he needed.
“Quantum mechanics is your standard daily food for thought, but you don’t really think about it that way,” Jung says. “It needed an external trigger, and that was our external trigger to think, ‘Oh yeah, entanglement. You can measure entanglement.’”
Jung brought the idea to his students, including graduate student AJ Wildridge. He was immediately on board. They knew ATLAS would be searching, too. And like any friendly competitor would, they wanted to win.
They had to decide between two datasets: They could use the full set of interactions between top and antitop quarks in the second run of the LHC, or they could use a smaller subset. The first option meant a longer analysis but a higher likelihood of the statistical power needed for a significant result. The second was riskier—less statistical power—but faster. Jung bet his students that the quicker path was the way to beat ATLAS.
At stake? A lager or a plate of fish and chips at 9Irish pub in Purdue.
The moment of truth
After two years of work, Jung, Wildridge, Afik and many of their team members found themselves at the top quark workshop in Michigan. Jung and Wildridge arrived a few days ahead of the Monday opening sessions. The entanglement analysis was running through CMS’s internal review process, but there was still a chance the collaboration would give Jung the green light to present their findings at the conference. Sunday night, rumors spread that the ATLAS team would soon share a result on entanglement.
Undeterred, Jung and Wildridge—joined by Jung’s postdoctoral student Giulia Negro—hunkered down at tables in the hotel hallway, furiously responding to questions from their collaborators on CMS. Wildridge would tilt his laptop screen down as his colleagues on ATLAS walked by, fielding friendly teasing about how his work was going. By the time they were done, they’d have 200 pages of questions and answers.
On Tuesday, Jung and Wildridge got the verdict back from the collaboration: The analysis still needed more work.
On Thursday afternoon, they sat in the audience as Afik gave a talk titled “Spin correlation and entanglement with the ATLAS experiment.” The result showed a stronger signal for entanglement than the team’s models predicted. That difference could indicate new physics, or it could be explained by a gap in the model predictions.
One explanation is a theorized state called toponium, in which a top quark and anti-top quark briefly become bound together as they orbit each other in the chaos of a collision. First proposed in the 1980s, such a maximally entangled bond could inflate the signal of the type of entanglement ATLAS and CMS were hunting.
The ATLAS team hadn’t accounted for the possible existence of this theoretical phenomenon in their analysis. But Jung and Wildridge had. As their colleagues began asking questions about toponium, Jung and Wildridge had to bite their tongues. Jung says he tried to ask questions that could push the conversation forward without giving away what they suspected. The topic raised so many questions that after the presentation, theorist Alexander Mitov gave an impromptu talk on it.
“He was saying, ‘How big of an effect does this have? We don’t know,’” Wildridge says. “And I was like, ‘We do know!’ I really wanted to comment these things.”
Instead, they stayed quiet and went home to finish their work. For the next six months, Jung, Wildridge and their colleagues meticulously reworked the analysis to address all the questions and concerns their colleagues had raised during the internal review process. In March, their updated analysis once again underwent CMS review. This time, they were given the go-ahead. They posted a paper to arXiv, confirming the ATLAS measurement and showing that toponium—if real—could account for a small amount of the discrepancy that the ATLAS team had seen.
The competition between the two experiments had spurred the physicists on, but it was their scientific dialogue with the larger physics community that gave each collaboration added confidence in the research. “ATLAS and CMS were able to measure quantum entanglement in complete different ways but reached the same conclusion,” Afik says. “This is one of the reasons why we have two experiments.”
High energy and quantum physics collide
Jung may owe his students a beer, but their effort was in no other way a loss. The results from the two teams are sparking new connections between high-energy physics and quantum information science, as well as efforts to search for other quantum effects at the LHC. Researchers are also working to confirm the existence of toponium, which could improve models used in many LHC experiments.
In October, about 50 physicists met in Oxford for a first-time workshop to discuss measurements of quantum mechanics at the LHC. As a member of the workshop’s advisory committee, Jung is considering a second gathering in Florence next year.
Regina Demina, professor of physics at the University of Rochester and a CMS member, was among the physicists at the workshop. Demina was part of the discovery of the top quark at the Tevatron, and has been pursuing an observation of top quark entanglement using a different decay channel than Jung’s team. She and her colleagues posted their work to the arXiv in September.
To Demina, the recent observations are just the beginning of a deeper understanding of entanglement the LHC could reveal.
In LHC collisions, the top quarks and antitop quarks appear, become entangled, and decay. That means they present an opportunity to study entanglement as a process, rather than a static state, as they have been studied before. And if scientists can understand where entanglement arises from, they may get hints about how the universe is built.
“Philosophically, quantum mechanics is not really understood at the deep level,” Demina says. “We do not understand why nature behaves this way. Maybe we can learn something about what this quantum behavior means and what actually leads to this.”
The LHC also provides a unique opportunity to study quantum effects at the relativistic scale, where energies are large enough to be governed by the laws of relativity. For nearly a century, for example, physicists have tried to define the inherently quantum property of spin in relativistic terms. The LHC could finally answer that question, Muñoz de Nova says. “High-energy colliders are the only place in the world where we can really do a genuine relativistic experiment.”
It’s a moment for physicists across fields to come together and push to answer and questions in new ways, Jung says. Both as a cooperative effort—and as a competition. “The race is still on.”
