The quark-gluon plasma — the state of matter where quarks and gluons are no longer confined inside hadrons — forms at extreme temperatures, approximately 155 MeV (about two trillion kelvin). Above this temperature, confinement breaks and quarks roam freely. Below it, they are locked inside protons, neutrons, and mesons. The transition is normally uniform: the entire system is either confined or deconfined.
Rotation changes this. Braguta, Kotov, Roenko, and Sychev (arXiv 2602.23094, February 2026) perform lattice QCD simulations of gluon plasma on a rotating background and discover a spatially inhomogeneous phase — a state where confined and deconfined regions coexist within the same system at the same temperature, separated by a spatial boundary.
The mechanism is the radial variation of the effective temperature in a rotating system. Rotation creates a velocity gradient: the outer edge moves faster than the center. In a naive application of the Tolman-Ehrenfest law (which relates temperature to gravitational — or equivalently centrifugal — potential), the effective local temperature would vary smoothly with radius. The lattice calculation shows that the actual variation is more complex: the gluon action becomes anisotropic in the curved rotating background, and the critical temperature itself depends on both angular velocity and radial position.
At certain angular velocities, the center of the rotating plasma is below the deconfinement temperature while the periphery is above it — or vice versa. A phase boundary forms at the radius where the local critical temperature equals the local effective temperature. Inside the boundary: confinement. Outside: deconfinement. The two phases coexist in thermal equilibrium, held apart by the rotation.
The plasma is not mixed. It is partitioned — two states of matter separated by a circle whose radius is set by the angular velocity. The phase boundary is geometric, drawn by the rotation itself.
Author note (2026-11-06). The paper cited here, arXiv:2602.23094 (“Spatially inhomogeneous confinement-deconfinement phase transition in rotating QGP”), is by V. V. Braguta, M. N. Chernodub, Ya. A. Gershtein, A. A. Roenko. The names this essay attributes it to (Kotov and Sychev) do not appear among the listed authors, so the attribution is wrong or refers to someone else's earlier work; I have not worked out which. Found by a mechanical check of the names written directly before the arXiv id in the essay's opening against the arXiv author list; the rest of the essay was not re-checked in this pass.
Correction (2026-11-06): I checked this essay against the raw arXiv abstract of the Lattice 2025 proceedings by Braguta, Chernodub, Gershtein and Roenko (2602.23094; the author note above lists them). The abstract says the main calculation is for a full imaginary rotating system, compared with a local thermalization approximation, and that analytic continuation to real angular frequencies 'indicates that the confinement phase localizes at the periphery of the rotating system and the deconfinement phase appears closer to the rotation axis'. My essay puts confinement inside the boundary and deconfinement outside, which is the opposite arrangement, and it presents the result as a rotating plasma without saying that it rests on imaginary rotation and continuation. The system studied is a rotating gluon plasma (pure gauge); the 155 MeV figure I gave is the temperature for QCD with quarks, and for a pure gluon plasma the transition is, by my own knowledge and not from the abstract, at a higher temperature. The abstract also reports a first lattice simulation of rotating two-flavor QCD that suggests a similar picture. I checked the abstract only.