A billion particle collisions a second: How large particle colliders work | Don Lincoln
Don Lincoln, a particle physicist at Fermilab and CERN, explains how particle accelerators work by converting kinetic energy into mass via Einstein's E=mc², and how detectors like CMS filter a billion collisions per second down to roughly a thousand recordable events. He contrasts Fermilab's Tevatron with CERN's Large Hadron Collider, highlighting the LHC's superior energy and collision rate. The conversation also covers antimatter production, the discovery of the top quark, and the sophisticated trigger systems used to identify rare, scientifically interesting collisions.
Summary
The conversation opens with Don Lincoln providing historical context on the Higgs boson, which was theoretically predicted in 1964 and became practically useful in 1967, with scientists eventually building accelerators powerful enough to detect it. He describes Fermilab's Tevatron, located outside Chicago, as the machine that discovered the top quark in 1995 by colliding protons and antiprotons at near-light speeds.
Lincoln then gives a foundational explanation of why particle accelerators are scientifically productive, centering on Einstein's E=mc². He explains that energy and matter are equivalent and interchangeable — two particles smashed together at high speed can stop, and their kinetic energy converts into new mass, creating particle-antiparticle pairs. This principle, predicted around 1928, is experimentally confirmed and routine. He notes that antimatter electrons were discovered in 1932 and antimatter protons in 1955 at Berkeley's Bevatron.
On antimatter production, Lincoln explains that it is extremely costly and inefficient — at Fermilab, roughly 100,000 protons had to be smashed to produce a single antiproton. He distinguishes between 'point-like' particles like electrons, which are easier to work with, and composite particles like protons, which behave like 'garbage cans full of stuff,' making precise energy-tuning harder. He also notes that higher collision energies yield more antiprotons, following a threshold-based production model.
Lincoln then compares Fermilab and CERN. While Fermilab stopped antiproton production in 2011 and shifted focus to neutrino physics, CERN's Large Hadron Collider operates at about seven times the energy and a hundred times more collisions per second than the Tevatron. This difference is illustrated vividly: the 1995 top quark discovery paper — which Lincoln co-authored — contained only 38 candidate events after months of data collection, roughly half of which were background noise. Today, the LHC produces a top quark every second, and top quarks are now considered background noise that researchers try to filter out.
The conversation then turns to how detectors manage the enormous data rate. At the LHC, approximately one billion collisions occur per second across about 40 million beam-crossing moments, with roughly 20 collisions per crossing. The two main detectors — CMS (which Lincoln works on) and ATLAS — are described in extraordinary physical terms: CMS is 70 feet long, 50 feet tall, and weighs 14,000 tons, while ATLAS is 150 feet long and weighs 7,000 tons. These detectors function as cameras, taking snapshots 40 million times per second.
To handle this data volume, a multi-stage trigger system is used. Fast electronics first reduce 40 million potential snapshots per second to about 100,000 interesting candidates. These are passed to commercial processors running optimized analysis code, which further reduce the set to about 1,000 collisions per second for permanent recording. These recorded events are then handed to graduate students for deeper analysis to find the rare handful that may represent new physics. Lincoln closes with admiration for the engineers and scientists who built this infrastructure — handling petabytes of data flowing around the world seamlessly.
Key Insights
- Lincoln explains that particle accelerators exploit E=mc² by smashing two particles together so their kinetic energies cancel, and the resulting stored energy converts into new particle-antiparticle pairs — a principle experimentally confirmed since the antimatter electron's discovery in 1932.
- Lincoln states that at Fermilab, producing a single antiproton required smashing 100,000 protons into a target, illustrating the extreme inefficiency and cost of antimatter production.
- Lincoln recounts that his 1995 co-authored paper on the top quark discovery contained only 38 candidate events after months of data collection, roughly half of which were background — contrasting sharply with the LHC, which now produces a top quark every second.
- Lincoln describes the LHC's multi-stage trigger system: fast electronics reduce 40 million beam crossings per second to 100,000 interesting candidates, which commercial processors then narrow to approximately 1,000 collisions per second actually recorded for analysis.
- Lincoln notes that the CMS detector — described as the 'small one' — is 70 feet long, 50 feet tall, and weighs 14,000 tons, while the competing ATLAS detector is 150 feet long and weighs 7,000 tons, underscoring the physical scale required to capture rare high-energy collision events.
Topics
Full transcript available for MurmurCast members
Sign Up to Access