About Topological Defects
Topological defects are singularities in ordered fields that cannot be removed by continuous deformations. In the 2D XY model — spins free to rotate in a plane on a square lattice — vortices appear as points where the spin angle winds by ±2π around the core. The stability of these defects is guaranteed by topology: the winding number is quantised and conserved, so vortices can only be created or destroyed in oppositely-charged pairs.
The Berezinskii-Kosterlitz-Thouless (BKT) transition at T_BKT ≈ 0.89 J/k_B separates two phases: below T_BKT, vortices are bound in tight pairs that cancel each other's long-range field; above T_BKT, entropy frees the pairs and proliferating vortices destroy quasi-long-range order. This topological phase transition — awarded the 2016 Nobel Prize in Physics — has no conventional order parameter and cannot be described by the standard Landau theory of phase transitions.
Frequently Asked Questions
What is a topological defect?
A topological defect is a stable configuration in an ordered field that cannot be continuously unwound to a uniform state. In the 2D XY model, vortices are point-like defects where spins wind by ±2π around the core. The winding number is a topological invariant — it cannot change without creating or destroying a defect-antidefect pair.
What is the 2D XY model?
The 2D XY model consists of classical spins of fixed magnitude placed on a 2D lattice, each spin free to point in any direction in the plane. The energy H = −J Σ cos(θᵢ−θⱼ) favours aligned neighbours. Unlike the 2D Ising model, the XY model has a continuous symmetry (rotation), which leads to qualitatively different physics — including the BKT transition rather than a conventional phase transition.
What is the Berezinskii-Kosterlitz-Thouless (BKT) transition?
The BKT transition is a topological phase transition in 2D systems with continuous symmetry. Below T_BKT ≈ 0.89 J/k_B, vortices exist only as bound pairs (vortex+antivortex); the system has quasi-long-range order with algebraically decaying correlations. Above T_BKT, thermal fluctuations unbind the pairs, creating free vortices that destroy quasi-order. The BKT transition has no conventional order parameter — it was the first example of a topological phase transition (Nobel Prize 2016).
What is a winding number?
The winding number counts how many times the spin direction rotates by 2π as you travel around a closed loop. A +1 vortex has spins winding +360° around its core; a −1 antivortex winds −360°. For a loop not enclosing any defects, the winding number is 0. Winding numbers add algebraically: enclosing a +1 and −1 defect gives winding 0.
Why do vortex-antivortex pairs attract?
Opposite-sign vortices attract via a logarithmic potential V ∝ −(J/π) ln(r/a), analogous to 2D electrostatics where the vortex charge plays the role of electric charge. This attraction leads to vortex-antivortex annihilation: the pair approaches, merges, and disappears, releasing energy into spin waves. Below T_BKT all pairs are bound by this attraction.
What happens at the BKT transition temperature?
At T_BKT, the free energy cost of unbinding a vortex pair becomes comparable to the entropy gain of having free vortices. Above T_BKT, entropy wins and vortices proliferate as free defects. The correlation length diverges in an essential singularity manner as T→T_BKT from above — qualitatively different from ordinary second-order transitions.
Where does the BKT transition occur in real materials?
BKT physics appears in thin helium-4 films (superfluid transition), 2D superconductors, ultracold atomic gases in 2D, and arrays of Josephson junctions. It also describes melting of 2D crystals (KTHNY theory) where dislocations play the role of vortices. The Nobel Prize 2016 was awarded to Kosterlitz, Thouless, and Haldane for these and related discoveries.
What are disclinations vs dislocations?
Disclinations are rotational defects where the spin field has a non-zero winding number around the core — this is what this simulation shows. Dislocations are translational defects in crystal lattices where the lattice has a Burgers vector (partial shift). Both are topological defects but in different order parameters. In 2D crystal melting (KTHNY theory), dislocations first unbind, then disclinations, giving two distinct transitions.
Why does the 2D XY model not have conventional long-range order?
The Mermin-Wagner theorem proves that continuous symmetries cannot be spontaneously broken in 2D at finite temperature (for short-range interactions). Spin waves (Goldstone modes) cost only logarithmic energy at long wavelengths — their entropy always overcomes the energetic ordering tendency. Below T_BKT the system instead has quasi-long-range order: correlations ⟨cos(θᵢ−θⱼ)⟩ ∝ r^{−η(T)} with η depending on temperature.
How is the XY model simulated with Metropolis Monte Carlo?
In the Metropolis algorithm, one spin is chosen at random, and a new angle proposed by adding a uniform random increment. The energy change ΔE from changing this one spin (only involving its four neighbours) is computed. If ΔE < 0, the move is accepted. If ΔE > 0, it is accepted with probability exp(−ΔE/T), implementing detailed balance and ensuring convergence to the Boltzmann distribution at temperature T.
Related Simulations