Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

Algorithms for the computation of solutions of the Ornstein-Zernike equation

Peplow, A. T. LU orcid ; Beardmore, R. E. and Bresme, F. (2006) In Physical Review E - Statistical, Nonlinear, and Soft Matter Physics 74(4).
Abstract

We introduce a robust and efficient methodology to solve the Ornstein-Zernike integral equation using the pseudoarc length (PAL) continuation method that reformulates the integral equation in an equivalent but nonstandard form. This enables the computation of solutions in regions where the compressibility experiences large changes or where the existence of multiple solutions and so-called branch points prevents Newton’s method from converging. We illustrate the use of the algorithm with a difficult problem that arises in the numerical solution of integral equations, namely the evaluation of the so-called no-solution line of the Ornstein-Zernike hypernetted chain (HNC) integral equation for the Lennard-Jones potential. We are able to... (More)

We introduce a robust and efficient methodology to solve the Ornstein-Zernike integral equation using the pseudoarc length (PAL) continuation method that reformulates the integral equation in an equivalent but nonstandard form. This enables the computation of solutions in regions where the compressibility experiences large changes or where the existence of multiple solutions and so-called branch points prevents Newton’s method from converging. We illustrate the use of the algorithm with a difficult problem that arises in the numerical solution of integral equations, namely the evaluation of the so-called no-solution line of the Ornstein-Zernike hypernetted chain (HNC) integral equation for the Lennard-Jones potential. We are able to use the PAL algorithm to solve the integral equation along this line and to connect physical and nonphysical solution branches (both isotherms and isochores) where appropriate. We also show that PAL continuation can compute solutions within the no-solution region that cannot be computed when Newton and Picard methods are applied directly to the integral equation. While many solutions that we find are new, some correspond to states with negative compressibility and consequently are not physical.

(Less)
Please use this url to cite or link to this publication:
author
; and
publishing date
type
Contribution to journal
publication status
published
subject
in
Physical Review E - Statistical, Nonlinear, and Soft Matter Physics
volume
74
issue
4
article number
046705
publisher
American Physical Society
external identifiers
  • scopus:33750188845
ISSN
1539-3755
DOI
10.1103/PhysRevE.74.046705
language
English
LU publication?
no
additional info
Copyright: Copyright 2008 Elsevier B.V., All rights reserved.
id
731a46c2-5c8c-43a5-8bc5-0a057b8f67d5
date added to LUP
2021-03-08 15:17:17
date last changed
2022-02-01 20:36:57
@article{731a46c2-5c8c-43a5-8bc5-0a057b8f67d5,
  abstract     = {{<p>We introduce a robust and efficient methodology to solve the Ornstein-Zernike integral equation using the pseudoarc length (PAL) continuation method that reformulates the integral equation in an equivalent but nonstandard form. This enables the computation of solutions in regions where the compressibility experiences large changes or where the existence of multiple solutions and so-called branch points prevents Newton’s method from converging. We illustrate the use of the algorithm with a difficult problem that arises in the numerical solution of integral equations, namely the evaluation of the so-called no-solution line of the Ornstein-Zernike hypernetted chain (HNC) integral equation for the Lennard-Jones potential. We are able to use the PAL algorithm to solve the integral equation along this line and to connect physical and nonphysical solution branches (both isotherms and isochores) where appropriate. We also show that PAL continuation can compute solutions within the no-solution region that cannot be computed when Newton and Picard methods are applied directly to the integral equation. While many solutions that we find are new, some correspond to states with negative compressibility and consequently are not physical.</p>}},
  author       = {{Peplow, A. T. and Beardmore, R. E. and Bresme, F.}},
  issn         = {{1539-3755}},
  language     = {{eng}},
  number       = {{4}},
  publisher    = {{American Physical Society}},
  series       = {{Physical Review E - Statistical, Nonlinear, and Soft Matter Physics}},
  title        = {{Algorithms for the computation of solutions of the Ornstein-Zernike equation}},
  url          = {{http://dx.doi.org/10.1103/PhysRevE.74.046705}},
  doi          = {{10.1103/PhysRevE.74.046705}},
  volume       = {{74}},
  year         = {{2006}},
}