Scientists working with the ALICE detector at the Large Hadron Collider report measurements of the spin alignment, or polarization, of Lambda (Λ) and anti-Lambda (Λ̄) particles along the beam direction in lead–lead collisions at a center‑of‑mass energy of 5.36 TeV per nucleon pair. The polarization shows clear sine-shaped variations with angle, and for the first time at the LHC the team observes a polarization pattern tied to the third‑order event plane — a reference direction that comes from how the collision debris flows outward.

The analysis used a large 2023 data set collected in LHC Run 3: roughly 5 billion collisions, corresponding to an integrated luminosity of about 3.2 nb⁻¹. The measurement relied on ALICE subdetectors installed or upgraded for Run 3, in particular the Fast Interaction Trigger (FIT) to find events and estimate collision geometry, and the Inner Tracking System (ITS) and Time Projection Chamber (TPC) to follow charged particles. Λ hyperons were reconstructed through their typical weak decay into a proton and a pion (Λ → p π⁻ and Λ̄ → p̄ π⁺). The team applied standard track and topological cuts — for example, limits on the distance between decay points and the collision point, and on how well the reconstructed momentum points back to the collision — to reduce background.

The basic idea behind the measurement is that a swirling or vortical motion of the quark–gluon plasma (QGP) created in the collision can align particle spins, a process called spin–vorticity coupling. The event planes are angles that reflect the direction of anisotropic flow — the tendency of matter to stream out more in some directions than others — and different “orders” (second, third, etc.) pick out different flow patterns. The polarization along the beam is extracted from how the decay protons are emitted in the Λ rest frame. Because the weak decay violates parity, the proton direction is correlated with the Λ spin with a known decay parameter (αΛ = 0.750 ± 0.010 for Λ and αΛ̄ = −0.758 ± 0.012 for Λ̄). The analysis corrects for detector acceptance and for the limited precision in determining event planes (the event‑plane resolution at 25% centrality is reported as 0.76). Signal and background are separated by fitting the invariant mass and polarization distributions simultaneously; the background polarization is found to be consistent with zero.