Transferable Potentials for Phase Equilibria−Coarse-Grain Description for Linear Alkanes

Abstract

Coarse-grain potentials allow one to extend molecular simulations to length and time scales beyond those accesssible to atomistic representations of the interacting system. Since the coarse-grain potentials remove a large number of interaction sites and, hence, a large number of degrees of freedom, it is generally assumed that coarse-grain potentials are not transferable to different systems or state points (temperature and pressure). Here we apply lessons learned from the parametrization of transferable atomistic potentials to develop a systematic procedure for the parametrization of transferable coarse-grain potentials. In particular, we apply an iterative Boltzmann optimization for the determination of the bonded interactions for coarse-grain beads belonging to the same molecule and separated by one or two coarse-grain bonds and parametrize the nonbonded interactions by fitting to the vapor-liquid coexistence curves computed for selected molecules represented by the TraPPE-UA (transferable potentials for phase equilibria-united atom) force field. This approach is tested here for linear alkanes where parameters for C(3)H(7) end segments and for C(3)H(6) middle segments of the TraPPE-CG (transferable potentials for phase equilibria-coarse grain) force field are determined and it is shown that these parameters yield quite accurate vapor-liquid equilibria for neat n-hexane to n-triacontane and for the binary mixture of n-hexane and n-hexatriacontane. In addition, the position of the first peak in various radial distribution functions and the coordination number for the first solvation shell are well reproduced by the TraPPE-CG force field, but the first peaks are too high and narrow.

Derivation of coarse-grained potentials via multistate iterative Boltzmann inversion

“…5,[7][8][9][10][11] While generic, non-specific CG models have been widely applied, [12][13][14][15][16] providing important information regarding trends and design rules, it is often necessary to use CG models specifically mapped to the system of interest to provide a direct one-to-one correspondence with experiment. While several "transferable" CG forcefields, such as TraPPE-CG 17 and MARTINI, 18 have been developed, akin to forcefield development at the atomistic level, [19][20][21][22][23] the development of new CG forcefields is still often necessary. This is often required since the available forcefields may be lacking the necessary molecular species/groupings or may not have been optimized for the properties of interest.…”

“…This second point is of particular consequence, since, for example, a forcefield optimized to match phase behavior may not appropriately capture subtle structural features. 17 Generally speaking, direct structural correspondence is needed to accurately transition between different a) Author to whom correspondence should be addressed. Electronic mail:…”

SAFT-γ force field for the simulation of molecular fluids: 8. Hetero-segmented coarse-grained models of perfluoroalkylalkanes assessed with new vapour–liquid interfacial tension data

“…In addition to the higher resolution (atomistic and united-atom) force fields mentioned earlier, CG models have also been developed for n-alkanes [60][61][62][63] and for perfluoro-n-alkanes [64,65]; we are not, however, aware of an integrated CG model which can deal quantitatively with the subtleties of having the two chemical moieties fused on the PFAA molecule. For this purpose, the SAFT-γ Mie force fields developed in our current study are assessed by comparing the simulated and experimental surface tension data of PFAAs which are not used to parameterize the model; the comparison thus serves as a validation of the molecular models.…”

SAFT-γ Force Field for the Simulation of Molecular Fluids. 1. A Single-Site Coarse Grained Model of Carbon Dioxide

“…30 Different functional forms are used for the non-bonded interactions in these CG force fields. In the MARTINI force field the non-bonded interactions are parameterized with a simple LJ potential.…”