Abstract
Loop models in two dimensions can be related to models. The low-temperature dense-loops phase of such a model, or of its reformulation using a supergroup as symmetry, can have a Goldstone broken-symmetry phase for . We argue that this phase is generic for when crossings of loops are allowed, and distinct from the model of noncrossing dense loops first studied by Nienhuis [Phys. Rev. Lett. 49, 1062 (1982)]. Our arguments are supported by our numerical results, and by a lattice model solved exactly by Martins et al. [Phys. Rev. Lett. 81, 504 (1998)].
- Received 2 May 2002
DOI:https://doi.org/10.1103/PhysRevLett.90.090601
©2003 American Physical Society