Full text
| |
| Author(s): |
Leonardo Crochik
Total Authors: 1
|
| Document type: | Master's Dissertation |
| Press: | São Paulo. |
| Institution: | Universidade de São Paulo (USP). Instituto de Física (IF/SBI) |
| Defense date: | 2005-06-23 |
| Examining board members: |
Tania Tome Martins de Castro;
Wagner Figueiredo;
Vera Bohomoletz Henriques
|
| Advisor: | Tania Tome Martins de Castro |
| Abstract | |
We studied a model of a gas in contact with two baths of particles. We used a model of a gas in a lattice (Ising model) in which the net is divided in two: the sub-net R1 and the sub-net R2. The system evolves in time according to the competition between two dynamics: one (dynamic A) that realizes the flow of particles from one sub-net to the other, simulating the contact with a heat bath at temperature T while the other one (dynamic B ) removes or put particles in the sub-nets, simulating the contact with particle baths at chemical potentials mu1 and mu2 and temperature T. We studied, using mean-field approximations and Monte Carlo simulations, the phase diagram and the critical properties of the model, getting similar critical behavior to the Ising model in equilibrium, except in a small region of the phase diagram in which there are reentrant phases. We also calculated the entropy production of the model. The study of its critical behavior results in the definition of a new critical exponent zeta related to the divergence of the derivative of the entropy production with respect to the temperature. We obtained zeta =0 (logarithmic divergence). We verified, finally, using in this case mean-field approximations, the limit of validity of two theorems from nonequilibrium thermodynamics: the minimum entropy production theorem and the universal evolution criteria. Regarding the first theorem, we determined in what limits we can consider the model\' s dynamics as ``close\'\' to equilibrium. Regarding the universal evolution criteria, we found situations in which the theorem is apparently violated. We believe that this violation must be consequence of an improper instability element brought by the approximation (of mean-field) used. The investigation of this question was delayed to a next work. (AU) | |