The chiral magnetic effect is one of the most dramatic predictions connecting quantum field theory anomalies to laboratory physics. In a quark-gluon plasma produced by heavy-ion collisions, the collision generates an enormous magnetic field — up to 10^18 gauss, the strongest field in the known universe, lasting for femtoseconds. If the collision also produces a chirality imbalance (more left-handed quarks than right-handed, or vice versa), the magnetic field should separate electric charges along its direction. Positive charges move one way along the field; negative charges move the other. The effect would be observable as a charge-dependent correlation in particle production relative to the reaction plane.
The ALICE Collaboration (arXiv 2602.22900, February 2026) measures exactly this observable in lead-lead collisions at 5.02 TeV per nucleon pair at the LHC, using two independent experimental techniques — event shape engineering and participant-spectator plane correlations. Both techniques are designed to separate a genuine chiral magnetic signal from the dominant background: flow-related correlations that mimic the charge separation pattern.
The result is consistent with zero.
The difficulty is not the measurement itself but the background. Elliptic flow — the collective anisotropy of particle emission caused by the almond-shaped overlap region in non-central collisions — generates charge-dependent correlations that look like the chiral magnetic effect. The background is larger than the expected signal. Extracting the signal requires suppressing or varying the background while keeping the potential CME contribution fixed.
Event shape engineering selects events with the same centrality but different elliptic flow magnitudes. If the observed charge separation scales with flow (background), the CME fraction is zero. If it doesn't scale perfectly (some residual at zero flow), that residual is the CME. The ALICE measurement finds the fraction consistent with zero: no residual above the flow-driven background.
The participant-spectator comparison uses a different approach. The participant plane is determined by the collision geometry of the interacting nucleons; the spectator plane is determined by the non-interacting nucleons that fly forward. The CME signal should correlate with the magnetic field direction, which follows the spectator plane more closely than the participant plane. Measuring correlations relative to both planes and comparing them separates the two contributions. Again: consistent with zero CME signal.
The constraints are the tightest to date for LHC energies. They don't prove the chiral magnetic effect doesn't exist — they prove it's not visible in lead-lead collisions at this energy with this sensitivity. The magnetic field is real. The chirality imbalance may be real. But the signal, if it's there, is smaller than the current experimental resolution can detect, buried under a background that mimics it almost perfectly.