ramess101
commented
6 years ago Are there limitations in this method ? What is the lower reduced temperature allowing safe initialization by Packmol and subsequent simulation by Cassandra ? Is it the reason why the ITIC method is not applied below a reduced temperature of 0.45 ? Does Packmol accept molecules of more complex shape (branched, cyclic, polycyclic) ?
I feel like these questions are totally irrelevant. Of course there are limitations for Packmol. But there are also very well founded methods for overcoming them. For example, starting at a lower density and slowly compressing the system. Initializing an NVT system at a given density is not something we really need to discuss, in my opinion. Packmol is not an integral part of ITIC, and so I would ignore this comment.
It seems that the ITIC method relies on building initial configurations in the same way as GEMC relies on insertion moves in dense liquid phases. It is possible that in the long range, this is also a fundamental limitation in the proposed method for coexistence curve calculation.
No, not at all. Building an initial configuration is completely different than performing GEMC in dense liquid phases. Building an initial configuration is essential for every type of simulation and assumed to be feasible for all reasonable systems of interest.
6.2. in section D the symbols used for enthalpy H and dimensionless enthalpy Hdep= H/RT are very close and this may be confusing. It would be clearer using lower case symbol (h instead of H) for this dimensionless property.
I think the present version is not quite as confusing. But we might want to consider this, or something like it.
@mostafa-razavi
The method presented is also depending on the initialization of NVT simulations of dense liquid phases. This is a delicate procedure because molecular overlaps are generally found. For this purpose the authors use the Packmol package. No detail is given about the efficiency of Packmol to initialize dense liquid phases. It is likely that this is an important limitation preventing the authors from reaching low temperatures (see remark 6.1).
6.1 The initialization of NVT simulations of the liquid phase is a delicate task that he authors achieve with the Packmol software. Are there limitations in this method ? What is the lower reduced temperature allowing safe initialization by Packmol and subsequent simulation by Cassandra ? Is it the reason why the ITIC method is not applied below a reduced temperature of 0.45 ? Does Packmol accept molecules of more complex shape (branched, cyclic, polycyclic) ? It seems that the ITIC method relies on building initial configurations in the same way as GEMC relies on insertion moves in dense liquid phases. It is possible that in the long range, this is also a fundamental limitation in the proposed method for coexistence curve calculation.
6.2. in section D the symbols used for enthalpy H and dimensionless enthalpy Hdep= H/RT are very close and this may be confusing. It would be clearer using lower case symbol (h instead of H) for this dimensionless property.