Thermodynamic consistency in dissipative particle dynamics simulations of strongly nonideal liquids and liquid mixtures

Abstract

Dissipative particle dynamics ͑DPD͒ is a mesoscopic simulation method for studying hydrodynamic behavior of complex fluids. Ideally, a mesoscopic model should correctly represent the thermo-and hydrodynamic properties of a real system beyond certain length and time scales. Traditionally defined DPD quite successfully mimics hydrodynamics, but is not flexible enough to accurately describe the thermodynamics of a real system. The so-called ''multibody'' DPD ͑MDPD͒ is a pragmatic extension of the classical DPD that allows one to prescribe the thermodynamic behavior of a system with only a small performance impact. Here we present a practical improvement to the ''multibody'' DPD model and test it on a number of single-component examples. We also generalize MDPD to multicomponent systems, which are an important target of DPD studies. The improved model provides a correction for particle correlations in strongly nonideal systems that were neglected in the original MDPD model. The implications of the coarse-graining procedure on the MDPD are discussed.

Constant-pressure simulations with dissipative particle dynamics

“…Note that we exclude the contribution of the particle i itself from the sum in ͑11͒, in contrast to the original papers by Pagonabarraga and Frenkel. 14,15 Including the self-contribution would not be consistent with the assumption of no particle correlations used by Pagonabarraga and Frenkel ͑see our earlier paper 16 for a more extensive discussion of this issue͒.…”

“…For example, it is impossible to simultaneously map the isothermal compressibility and the pressure. 16 If we succeed in mapping compressibility, the pressure will have an unrealistically high value, or if we map the pressure, the compressibility will be too low.…”

“…To illustrate the operation of MDPD under constant-pressure conditions we take the ͑coarse-grained͒ Lennard-Jones fluid from our earlier paper. 16 …”

“…However, the model has been derived under the assumption of no particle correlations and provides at best qualitative accuracy for strongly nonideal systems, which are of most practical interest. In an earlier paper 16 we have presented a modification of the original MDPD model that features an explicit correction for particle correlations allowing one to reproduce the thermodynamic behavior of a strongly nonideal ͑multicomponent͒ system with quantitative accuracy.…”

“…VI we summarize the main results presented in the paper. In the Appendix we briefly presented an adaptive local-density approximation approach, 16 which allows one to significantly improve the MDPD accuracy and is extensively used throughout this paper.…”