Characterization of wedge-local observables in integrable models with bound states
Karim Shedid Attifa
Abstract
Integrable quantum models (IQMs) are a class of interacting quantum field theories in 1+1 dimensions in which the Hilbert space can be described as a Fock space with generalized statistics. Because of this Fock space description, any observables can be expanded in the creation- and annihilation operators with the coefficients of the expansions being the connected matrix elements. One example of such an observable in IQMs without bound states is the so-called wedge-local field (a free field analog with the interacting creation- and annihilation operators) which is wedge-local in the sense that it strongly commutes with its reflection when smeared with a function localized in the left wedge. In this thesis, we study wedge-local observables in integrable quantum models and the effect of bound states on the properties of the connected matrix elements of these observables. This allows us to shine light on the connection between the localization of an observable and the functional analytic properties of its connected matrix elements in models with critical phenomena, in this case models which exhibit bound states. In scalar integrable quantum models with bound states (e.g. the Bullough-Dodd model), we proof that the connected matrix elements of quadratic forms which satisfy a wedge-locality condition are boundary values of meromorphic functions, so-called form factors, which satisfy a number of properties. Conversely, we show that any expansion in such form factors defines a quadratic form which satisfies the wedge-locality condition. In contrast to models without bound states, such an expansion does not terminate due to a recursion relation between the form factors and the wedge-local field, a wedge-local operator which creates one-particle states from the vacuum, does not fall into the class of the characterization theorem. Further, we conjecture a candidate for the wedge-local field in the Z(4)-Ising model. This model is interesting because it has a ”higher-order” bound state structure and no established wedge-local field is known in such models. However, the conjectured candidate comes with some difficulties. Most notably, it is ill-defined as an unbounded operator.