Please use this identifier to cite or link to this item:
Title: Artificial Neural Network modeling of spin-transition behavior in two-dimensional molecular magnet: The learning by experiences analysis
Authors: Wimalin Laosiritaworn
Yongyut Laosiritaworn
Authors: Wimalin Laosiritaworn
Yongyut Laosiritaworn
Keywords: Chemistry;Materials Science
Issue Date: 1-Apr-2013
Abstract: In this work, the spin-transition behavior in molecular magnet was investigated via Monte Carlo simulation on Ising model with mechano-elastic interaction extension. The initial spin-arrangement took hexagonal lattice structure in two dimensions, where spin molecules situated on the hexagonal lattice points were allowed to move under spring-type elastic interaction potential. Metropolis algorithm was used to update the spin configurations and thermal hysteresis loops were recorded to extract the hysteresis properties, such as period-average magnetization, loop area, loop width and height, as functions of parameters associated to magnetic and elastic interaction in the Hamiltonian. From the Monte Carlo results, the dependence of the hysteresis loop characteristic on magnitude of energy differences and number of available states between the low spin state and the high spin state was evident. The occurrence of the cooperative effect was notable, in agreement with previous experimental investigation, when the range of Hamiltonian parameters used is appropriate. Then all the measured hysteresis characteristic were passed to the Artificial Neural Network modeling to create extensive database of how the thermal hysteresis would respond to the change of molecular magnet Hamiltonian parameters. The scattering plots between the Artificial Neural Network and the real measured results have R-square closed to one which confirms the success of Artificial Neural Network in modeling this thermal hysteresis behavior. One is therefore allowed to use this Artificial Neural Network database as a guideline to design ultra-thin-film molecular magnet application in the future. © 2013 Elsevier Ltd. All rights reserved.
ISSN: 02775387
Appears in Collections:CMUL: Journal Articles

Files in This Item:
There are no files associated with this item.

Items in CMUIR are protected by copyright, with all rights reserved, unless otherwise indicated.