Abstract
dc:descriptionThroughout the present thesis, different physical systems are investigated. These systems are controlled by phenomena such as crystal growth, stress energy, diffusion, phase transitions. Despite their diverse nature, all these systems are studied by means of percolation theory and its variations. Most of the effort invested on this thesis, was carried out by computational modeling, but whenever it was feasible, analytical and experimental results were also taken into account. The first chapter is introductory to percolation theory and phase transitions. At first, percolation is studied from a theoretical aspect and basic notions are explained. Examples of applications to different disciplines are presented along with the corresponding phase transitions that take place. Moreover, variations of the standard percolation theory are introduced, such as rigidity percolation, especially since they are implemented in the study of the physical systems, which are investigated in the present thesis. Then, computational methods for the simulation of percolation are presented. A short introduction to Monte Carlo methods is followed by a more detailed presentation of algorithms, such as the Hoshen-Kopelman one and Depth First Search. These algorithms are essential for deriving key physical properties, which help us describe the systems under investigation. At each one of the next four chapters, a different physical problem is dealt with. In the 2nd chapter, the main concepts are crystal growth, stress energy and rigidity percolation. It is extremely encouraging that the results of the simulations which took place coincided with the experimental ones, thus providing a valid confirmation of our research. In the 3rd chapter, the diffusion of a tracer particle is monitored within a grid of floppy and rigid regions. In the fourth chapter, a system of growing islands is investigated. Staring from a random seed, the initial site grows along its periphery and this growth stops when it touches a neighboring growing site. According to the results of the simulations and the ones derived from the implementation of the scaling theory, it has been made clear that this problem belongs to a new universality class, which is different from that of the simple percolation problem. In chapter 5, the problem studied is not quite relevant with percolation theory, since the notion of negative diffusion was studied. This was an extra problem during the preparation of this thesis, which aimed at offering to the author of this thesis an insight to different physical problems. The philosophy of chapter 6 is different from the one of the previous chapters. Instead of using simulations in order to study physical systems, a simple percolation model is used in order to investigate the robustness, the efficiency and the speed of a parallel Hoshen-Kopelman algorithm. Message Passing Interface (MPI) was engaged for the implementation of the parallel program. The thesis resulted in five publications in peer reviewed magazines with international referees.
Degree
thesis:*- Grantor dc:publisher
- Aristotle University Of Thessaloniki (AUTH)
- Year dc:date
- 2011
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Τσακίρης, Νικόλαος-Παναγιώτης
- Tsakiris, Nikolaos-Panagiotis
Subjects
dc:subject × 12Rights
- Language dc:language
- gre
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/27031