Topological relics of symmetry breaking: winding numbers and scaling tilts from random vortex–antivortex pairs

I show that random distributions of vortex–antivortex pairs (rather than of individual vortices) lead to scaling of typical winding numbers W trapped inside a loop of circumference C with the square root of that circumference, W∼√C, when the expected winding numbers are large, |W| ≫ 1. Such scaling is consistent with the Kibble–Zurek mechanism (KZM), with 〈W2〉 inversely proportional to ξ-hat , the typical size of the domain that can break symmetry in unison. (The dependence of ξ-hat on quench rate is predicted by KZM from critical exponents of the phase transition.) Thus, according to KZM, the dispersion √ scales as √(C/ ξ-hat ) for large W. By contrast, a distribution of individual vortices with randomly assigned topological charges would result in the dispersion scaling with the square root of the area inside C (i.e., √ ∼ C). Scaling of the dispersion of W as well as of the probability of detection of non-zero W with C and ξ-hat can be also studied for loops so small that non-zero windings are rare. In this case I show that dispersion varies not as 1/√( ξ-hat ), but as 1/ ξ-hat , which results in a doubling of the scaling of dispersion with the quench rate when compared to the large |W| regime. Moreover, the probability of trapping of non-zero W becomes approximately equal to 〈W2〉, and scales as 1/ ξ-hat 2. This quadruples—as compared with √ ≃ √C/ξ-circumflex valid for large W—the exponent in the power law dependence of the frequency of trapping of |W| = 1 on ξ-hat when the probability of |W| > 1 is negligible. This change of the power law exponent by a factor of four—from 1/√( ξ-hat ) for the dispersion of large W to 1/ ξ-hat 2 for the frequency of non-zero W when |W| > 1 is negligibly rare—is of paramount importance for experimental tests of KZM. (paper)