The dependence of the phase diagram on the range of the attractive intermolecular forces

Abstract

A density functional theory of freezing combined with a thermodynamically consistent integral equation method is used to investigate the phase behaviour of systems interacting via the m-n potential with m = 2n, φ(r) = 4ε[(σ/r) 2n − (σ/r) n ] (n = 6, 8, 10 and 12) and rigid C 60 molecules interacting via the Girifalco potential. It is found that the liquid-vapour coexistence region is gradually suppressed as the attractive part of the potential becomes short range with increasing n and the coexistence ceases to occur at n ≈ 11. The m-n potential with n = 11-12 is similar to the Girifalco potential and the two yield similar phase diagrams. It is also found that the phase diagram of C 60 calculated for a truncated potential is qualitatively in agreement with the corresponding Monte Carlo (MC) simulations of Hagen et al, which have predicted nonexistence of the liquid phase in contrast to the molecular dynamics (MD) simulations of Cheng et al. These results suggest the importance of treating the long-range tail of the potential correctly and provide a partial explanation for the discrepancy between the MC and MD simulations.

Strong weak and metastable liquids structural and dynamical aspects of the liquid state

“…In Fig.2 these phase diagrams are presented. We find good agreement with the results obtained by Hasegawa and Ohno [16]. A decrease in the range of attractive interaction leads to a lower critical temperature and a higher critical density.…”

“…The phase diagrams for n --12 and n = 18 are given in Fig.4 which again are in good agreement with the results of Hasegawa and Ohno [16]. Indeed, as expected, the critical temperature drops further and now sinks below the fluid-solid coexistence line.…”

“…Increasing n leads to a shorter ranged potential. Phase diagrams of this system were calculated by Hasegawa and Ohno [16] using an density functional approach. The precise results of such calculations strongly depend on the approximations made and therefore we used simulation methods to calculate the phase diagrams.…”

“…On the other hand a combined integral equation and molecular dynanlics approach by Chang et al [12] led to the conclusion that there is a small stable liquid region. Recent theoretical work [13][14][15][16] has not resolved this issue. Also experimentally this case has not been settled yet [17].…”

Free energy determination of phase coexistence in model C60: A comprehensive Monte Carlo study

“…A tiny liquid pocket has been predicted also in the phase diagram of similar models of higher order fullerenes [3], although it is argued that even a modest reduction of the range of the forces might cause the disappearance of the liquid phase. The emerging borderline nature of the Girifalco model implies that the overall appearance of its phase diagram sensitively depends on the details of the interaction potential; in fact, the initial controversy around the existence of a stable liquid phase for this system [11,13] has been solved by taking into account on one side the full role of the attractive part of the interaction [14], and on the other, a conveniently large simulation sample [15]. More generally, it has been recognized in Ref.…”