|
StatusThe thesis was presented on the 30 August, 2010Approved by NCAA on the 6 October, 2010 Abstract![]() |
The thesis is written in Chisinau in 2009 and consists of introduction, five chapters, conclusion, bibliography of 243 names and 2 annexes.
It contains 71 figures, 11 tables and is carried on 221 pages. The results are published in 45 scientific papers.
Main scientific significance of the thesis is to develop and classify the theory of phase transitions in supercooled liquids based on the concept of cluster, and taken into account the recent discovery of the generation and extinction of crystal nuclei in supercooled liquids at very low temperatures, and the practical value of this work is determined by the importance of understanding the connection between physical properties, microscopic structure of the substance and macroscopic conditions of materials processing, which is vital to produce new materials with advanced technological properties. However, the results allow to formulate the essence of a new research development in the study of complex cluster systems by methods of statistical physics, numerical calculation and computer modeling.
In particular, the generation and extinction of nuclei was found at low temperatures in supercooled liquids of o-benzylphenol, salol and 2,2´- dihydroxybenzophenone, and was brought a new concept of irreversible structural relaxation in supercooled liquids and glasses. In the generalized Szilard model the evolution of cluster structures for an arbitrary number of particles in the system and for different transition rates, as well as for the case of an open system was investigated. Structural relaxation at constant pressure using Morse potential is studied in supercooled liquids of silicon dioxide and beryllium fluoride by molecular dynamics computer simulation method. Role of the intermediate metastable state in the kinetics of phase transitions induced by fluctuations between the initial and final states is elucidated in the framework of a parametric model with a kinetic potential taking into account the asymmetry and the presence of an external field. It has also been proposed and applied to heterogeneous systems with stochastic interactions a new methodology to optimize the distribution of particles in clusters as a function of their total number in the system and the appropriate number of states, as well as the existence of the connection between stochastic modeling and computational ABM models has been proven.