Abstract
The effects of various boundary conditions on the Monte Carlo method have been studied through an analysis of the two-dimensional Ising model using a 30×30 system. The following boundary conditions were used: mean field, free edges, periodic, and correlated. The results confirm the reliablility of periodic boundary conditions but the correlated boundary conditions are comparable and in some instances yield results (the critical temperature, for example) that are in better agreement with the exact values.
Similar content being viewed by others
References
K. Binder (ed.),Monte-Carlo Methods in Statistical Mechanics (Springer-Verlag, Berlin, 1979).
C. Kalle and V. Winkelmann (to appear inJ. Stat. Phys.).
M. Born and Th. von Kárrnán,Phys. Z. 13:297 (1912).
D. P. Landau,Phys. Rev. B 13:2997 (1976).
D. W. Heermann and D. Stauffer,Z. Phys. 40B:133 (1980).
H. Müller-Krunbhaar and K. Binder,Z. Phys. 254:269 (1972).
N. Jan and D. Pink, 46 Stat. Mech. Meeting, Rutgers University (1981); and to be published.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Jan, N., Steinitz, M.O. Comparison of different boundary conditions for Monte Carlo simulations of Ising models. J Stat Phys 30, 37–44 (1983). https://doi.org/10.1007/BF01010867
Received:
Revised:
Issue Date:
DOI: https://doi.org/10.1007/BF01010867