1998
DOI: 10.1103/physreve.57.2553
|View full text |Cite
|
Sign up to set email alerts
|

Multicanonical methods, molecular dynamics, and Monte Carlo methods: Comparison for Lennard-Jones glasses

Abstract: We applied a multicanonical algorithm to a two-dimensional and a three-dimensional Lennard-Jones system with quasicrystalline and glassy ground states. Focusing on the ability of the algorithm to locate low-lying energy states, we compared the results of the multicanonical simulations with standard Monte Carlo simulated annealing and molecular-dynamics methods. We find slight benefits to using multicanonical sampling in small systems ͑less than 80 particles͒, which disappear with larger systems. This is disapp… Show more

Help me understand this report
View preprint versions

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1

Citation Types

2
30
0

Year Published

1998
1998
2007
2007

Publication Types

Select...
7
1

Relationship

0
8

Authors

Journals

citations
Cited by 42 publications
(32 citation statements)
references
References 50 publications
2
30
0
Order By: Relevance
“…Jumping ahead of ourselves, let us say that the statistics of our model at T , 0:003 is dominated by glassy states. Exactly the same behaviour of the multicanonical algorithm as that found by our study has recently been reported by Bhattacharya and Sethna [20] in the MC study of glassy Lennard -Jones liquids. It was argued that it is the entropy barriers, which trap simulations in multicanonical ensemble in contrast to the potential energy barriers trapping conventional MD simulations.…”
Section: Resultssupporting
confidence: 84%
“…Jumping ahead of ourselves, let us say that the statistics of our model at T , 0:003 is dominated by glassy states. Exactly the same behaviour of the multicanonical algorithm as that found by our study has recently been reported by Bhattacharya and Sethna [20] in the MC study of glassy Lennard -Jones liquids. It was argued that it is the entropy barriers, which trap simulations in multicanonical ensemble in contrast to the potential energy barriers trapping conventional MD simulations.…”
Section: Resultssupporting
confidence: 84%
“…To assess the performance of our algorithm in sampling low lying energy states quantitatively, we compare the lowest energies sampled with the results of a recent study on the optimization properties of traditional simulation algorithms applied to the Lennard-Jones glasses [13]. The authors reported on a lowest energy configuration found for N = 60 at e = −5.38 by cooling down slowly from the liquid phase using standard molecular dynamics.…”
mentioning
confidence: 99%
“…The difference in energy δe = 0.07 corresponds to a factor of 100 in computer time according to the estimate given in refs. [13,10]. For T run ≤ 0.5 around 15% of all states sampled using our method are lower in energy then e = −5.38.…”
mentioning
confidence: 99%
“…[38]), each update of τ -EO a variable is selected based on its rank according to the probability distribution in Eq. (8). When a rank k(≤ n)…”
Section: B Update Probabilities For Extremal Optimizationmentioning
confidence: 99%
“…Examples in the sciences are the determination of ground states for disordered magnets [2,3,4,5,6,7] or of optimal arrangements of atoms in a compound [8] or a polymer [9,10,11]. With the advent of ever faster computers, the exact study of such problems has become feasible [12,13].…”
Section: Introductionmentioning
confidence: 99%