Dear Boris Sedunov,

Firstly, thermodynamic results of real gases can be used to fit the parameters of the interaction potentials that later will be used in Molecular Dynamics simulation of systems at high densities  (at least for very simple molecules). Taking into account the computational effort, it is in principle easier to look for the interaction parameters that give the experimental values of the virial coefficients of real gases using alternative procedures based on the statistical mechanics. (See the book of Hansen and McDonald, Theory of simple liquids).

On the other hand, Molecular Dynamics (MD) (or Metropolis Monte Carlo) techniques are really powerful when dealing with condensed matter (say high density). At these conditions, it might occur that the interaction parameters that give the thermodynamic behaviour of real gases do not describe accurately the, say, thermodynamics of the liquid phases. Therefore, very often the interaction potentials are parametrized using physical properties of the condensed phases.

So, from my point of view, the ability of MD to reproduce the thermodynamic results for real gases depends on the parameters of the interaction potentials, and it is not difficult to tune those parameters to get a very good match between simulation and experiment. So, there is not real new "science" in doing this (it can be seen just as a technicality to build up or probe models that will be used later for other purposes).

Finally, there is a subtle technical point, for low (or perhaps very low) densities, MD can be inefficient due to a very short "number of collisions"  between particles,

best regards,

Noé

Dear Boris,

Iam not sure that I understand your question. There are many, many simulations aimed at real gases and liquids, some even with the aim of providing data where there are no experiments, yet.

The problem are the interaction potentials (“force fields”). With Lennard-Jones potentials, there are some limitations. But there exist “global simulations”, i.e., simulations using ab initio potentials (including 3-body potentials), which have been used in simulations and compared with experimental data.

 Dear Dr. Noé G. Almarza,

Thanks for your comprehensive answer! I totally agree with you.

What I do is the extraction of molecular and atomic interaction parameters from precise experimental data contained in electronic databases. I came to a number of no obvious results that are not taken yet into account in MD or Monte Carlo simulations.

For example, I have discovered the growth of the interaction forces at low temperatures, near the triple point. This result tells that the Lennard-Jones potential does not work for Noble gases at low temperatures because of dimers' freezing with orientation of interatomic forces instead of their spherical symmetry, supposed in the LD model. This effect is obvious for molecular interactions, but for atoms nobody takes it into account. The atoms, like molecules, are not round balls, they have electron orbits with their own symmetry. But I came to this conclusion not from ab-initio consideration, but from a careful analysis of precise thermophysical data from the NIST Database for Fluid Systems.

I am looking for a partner to show that the quantum considerations may confirm my conslusions about asymmetry of atomic interaction potential.

Dear Boris,

in the absence of external fields, atoms are “round balls”. The presence of other atoms or a crystal lattice disturbs the electron distributions to some extent. This is described by interaction potentials.

Pair potentials are known to be insufficient at high densities; one needs at least three-body potentials (the Axilrod–Teller potential is a good example).  But then quantitative simulations of noble-gas solids, but also ionic crystals are possible without the need to assume dimers.

Dear Henk Smid,

Thank you for your response.

What as to CO2 real gas, it seems to be an unusual no polar gas with extra large bond energy of dimers, as compared to other no polar gases of a similar mass. I wonder, if there is an explanation of unusual properties of this widely used in practice gas.

I have published the book: Discovering the Cluster World, where I describe the methods to extract properties of clusters from precise experimental thermophysical properties. Basing on the knowledge of clusters it is possible to calculate the thermophysical properties.

https://www.ljubljuknigi.ru/bookprice_offer_1992ef5ff3064d1760f906ce605269de1f1c0821?auth_token=d3d3LmxhcC1wdWJsaXNoaW5nLmNvbTphNWU2ZDdiN2ViNzVjOGI0ODRmMzIyNTFhYzIwMjlhOQ%3D%3D&locale=ru¤cy=EUR

Dear Boris,

CO2 has a strong electrical quadrupole moment. This explains some of its peculiar properties.

Dear Ulrich,

I have discovered it myself in 2012.

Do you know some paper or a book about the quadruple moment in CO2?

Do you know other molecules with this property?

Is it possible to compute the quadruple moment in CO2 from ab initio approach?

Thanks for your information,

Boris