Simulation of Nonuniform Polymer Fluids


Principal Investigator

Clifford Woodward

School of Chemistry,

University of New South Wales

Australian Defence Force Academy

Many systems of natural and industrial importance contain a mixture of large colloidal particles and polymers. It is found that the polymer molecules can have a profound influence on the way the colloidal particles interact. This study aims to develop theoretical methods to study interactions between colloidal particles, as mediated by polymer molecules. We use Monte Carlo simulations in order to obtain the forces between planar surfaces, which confine a polymer solution. Our method incorporates a novel ensemble which allows us to simulate under conditions appropriate to an open system (open to the rest of the solution), without using impractical particle exchanges. The results of this work will have significance across a broad spectrum of natural and technological phenomena.    



Adam Czezowski

School of Chemistry,

University of New South Wales

Australian Defence Force Academy





g87 - PC


What are the results to date and the future of the work?

A working program has been developed which has allowed us to complete a fairly comprehensive set of production runs. These have concentrated on the problem of polymer melts at several concentrations, and for several different types of attractive and repulsive particle interactions. We have been able to determine the free energy of interaction between the surfaces as a function of their separation to high accuracy.

A new means of perturbatively determining free energy differences, has now been successfully implemented. This has allowed for more rapid computation of interaction free energies (sometimes by up to a factor of four). This has and will allow us to explore many more systems than we first envisaged.

A new ensemble, allowing for the presence of both solvent and polymer molecules has just recently been completed. This ensemble allows us to keep the chemical potential of both polymer and solvent fixed. Thus we are now able to simulate equilibrium with a solution, wherein solvent particles are accounted for explicitly.

Density functional theory has been applied to the systems that we have already simulated, in

- Appendix B



an attempt to evaluate the accuracy of that theory. The advantage of density functional theory is that, computationally, it is a cheap calculational tool. Thus far, the density functional results have shown some promise but more work is required in that area to improve its performance.

The results so far form the basis of two scientific publications, which are currently under preparation. Ultimately they will be published as part of Mr Czezowski's PhD thesis.

What computational techniques are used?

Our research makes use of Monte Carlo simulation methods in order to generate particle configurations, consistent with an appropriate thermodynamic ensemble. The computer programs we use are written in FORTRAN 77 and FORTRAN 90.

We also use a density functional theory developed by one of us (CW). Numerical solutions to this equation require repeated matrix vector in an iterative loop. The program is written in FORTRAN 90.

Appendix B -