Non-homogeneous quantum Markov states and quantum Markov fields

The program relative to the investigation of quantum Markov states for general one-dimensional spin models is carried on, following a strategy developed in the last years. In such a way, the emerging structure is fully clarified. This analysis is a starting point for the solution of the basic (still open) problem concerning the construction of a satisfactory theory of quantum Markov fields, i.e. quantum Markov processes with multi-dimensional indices.     

On the Structure of KMS States of Disordered Systems

We study the set of KMS states of spin systems with random interactions. This is done in the framework of operator algebras by investigating Connes and Borchers –invariants of W*–dynamical systems. In the case of KMS states exhibiting a property of invariance with respect to the spatial translations, some interesting properties emerge naturally. Such a situation covers KMS states obtained by infinite–volume limits of finite–volume Gibbs states, that is the quenched disorder. This analysis can be considered as a step towards fully understanding the very complicated structure of the set of temperature states of quantum spin glasses, and its connection with the breakdown of the symmetry for replicas.     

Fourier Analysis for Type III Representations of the Noncommutative Torus

Journal of Fourier Analysis and Applications - For the noncommutative 2-torus, we define and study Fourier transforms arising from representations of states with central supports in the bidual,...     

Entangled Markov chains

Motivated by the problem of finding a satisfactory quantum generalization of the classical random walks, we construct a new class of quantum Markov chains which are at the same time purely generated and uniquely determined by a corresponding classical Markov chain. We argue that this construction yields as a corollary, a solution to the problem of constructing quantum analogues of classical random walks which are “entangled” in a sense specified in the paper.The formula giving the joint correlations of these quantum chains is obtained from the corresponding classical formula by replacing the usual matrix multiplication by Schur multiplication.The connection between Schur multiplication and entanglement is clarified by showing that these quantum chains are the limits of vector states whose amplitudes, in a given basis (e.g. the computational basis of quantum information), are complex square roots of the joint probabilities of the corresponding classical chains. In particular, when restricted to the projectors on this basis, the quantum chain reduces to the classical one. In this sense we speak of entangled lifting, to the quantum case, of a classical Markov chain. Since random walks are particular Markov chains, our general construction also gives a solution to the problem that motivated our study.In view of possible applications to quantum statistical mechanics too, we prove that the ergodic type of an entangled Markov chain with finite state space (thus excluding random walks) is completely determined by the corresponding ergodic type of the underlying classical chain. Mathematics Subject Classification (2000) Primary 46L53, 60J99; Secondary 46L60, 60G50, 62B10     

On the local implementations of gauge symmetries in local quantum theory

Under the general assumptions of quantum field theory in terms of local algebras of field operators fulfilling the split property, we prove that any two local coveriant implementations of the gauge group (or, in the case of a connected and simply connected Lie gauge group, any two choices of local current algebras) relative to a pair of double cones ₁, ₂, are related by a unitary equivalence induced by a unitary in the algebra of observables localized in ₂ which commutes with all fields localized in ₁, where ₁ is any double cone contained in the interior of ₁, and ₂ any double cone containing ₂ in its interior.     

Optimal strategy to model the electrodialytic recovery of a strong electrolyte

In this work, mostly Nernst–Planck derived relationships were used to simulate the electrodialytic recovery of a strong electrolyte, namely sodium chloride. To this end, it was set up a five-step experimental procedure consisting of zero-current leaching, osmosis, and dialysis, electro-osmosis, desalination, current–voltage and validation tests. The contribution of leaching and solute diffusion across the electro-membranes was found to be negligible with respect to the electro-migration. On the contrary, solvent diffusion tended to be important as the solute concentration difference at the membrane sides increased or current density was reduced. The electro-osmosis and desalination tests yielded the water and solute transport numbers.

By performing several limiting current tests at different solute concentrations and feed flow rates using anionic or cationic membranes, it was possible to determine simultaneously the limiting current intensity, the ratio of the differences between the counter-ion transport numbers in the anion- and cation-exchange membranes and solution, the overall resistance of the electro-membranes, the effective membrane surface area, and the solute mass transfer coefficient.

All these process and design parameters allowed the time course of the solute concentration in the concentrating (C) and diluting (D) compartments, as well as the voltage applied to the electrodes, to be reconstructed quite accurately without any further correction factors. The capability of the above parameters to simulate the performance of the electrodialysis (ED) unit was checked by resorting to a few validation tests, that were performed in quite different operating conditions from those used in the training tests, that is by filling tank C with a low feed volume with a low solute concentration and applying a constant current intensity to magnify the effect of electro-osmosis or by changing the current intensity step-wisely to simulate the continuous-mode operation of a multistage ED unit. Finally, a parameter sensitivity analysis made the different contribution of the process and design parameters to be assessed, thus yielding a straightforward procedure for designing or optimising accurately ED desalination units up to a final salt concentration of about 1.7 kmol m⁻³.