Ms. Mossallanejad:

You ask a simple but difficult question. It is not a simple task to establish the "goodness" of a force field. It depends on the details of what you are trying to accomplish. OPLSaa is a good force field, as are others like CHARMM and AMBER. You will find proponents for all. Your question hints at conducting simulations with ammonium nitrate ions in the solution but you will find that it is generally difficult to find places where you can directly compare simulation results with experiment.

If you are thinking, for example, to titrate the ionic content in a series of simulations, that will not work. In a simulation box 100 A (10 nm) on a side, you enclose a volume of 10^-21 L. A single ion in the box is a concentration of 0.6 mM; concentrations in the nM range would require a simulation volume a million times greater, which likely will exceed your computational capabilities. You must use some kind of free energy methods to establish something like a pH dependence that is easy to measure in the lab. Such methods are quite limited in their precision.

On the other hand, something like a radial distribution function is easy to compute and, potentially, can be compared with the experimentally determined structure factor, although such comparisons are not simple.

So, what exactly are you trying to accomplish with your simulations? How does the ionic content matter? If the ions are not directly interacting with substrates, then any charged entity will probably be sufficient and the veracity of the OPLSaa representation of ammonium and nitrate will not be particularly important to the results.

Mark

Just a short answer: proofing of any forcefields used in simulations can't be done by one or two simple calculations.

I might be wrong here but I believe that you cannot break or form any bonds in MD. So Mark has a point here, what are you trying to do?

Thanks all for your replies. I want to calculated the interfacial tension (IFT) and wanted to use OPLSaa but was not sure if this force field is good enough. As Mark said finding experimental results to compare with is difficult.

Ms. Mosallanejad:

I believe that calculating something like interfacial tension will be quite difficult, regardless of force field choice. MD really does not have the precision for absolute calculations; it works better as a comparison tool. It is difficult, but possible, to obtain relative rates of reactions, for example. (One can use alchemical transformations as a subterfuge.) If you want to ask if the interfacial tension of an ammonium nitrate solution is higher than some other nitrate solution then you could use the aforementioned alchemical transformations to estimate the differences in free energies. If you want to say something definitive, you might have some hope if the difference in free energies is 100 kJ/mol or so. Then you can say with some authority that the real tension will be larger. If the difference is only 10 kJ/mol or thereabouts, then that is within the general precision of the methodology and you cannot make any strong statement about the relative tensions.

Mark

Hi Mark,

Thanks so much for your suggestions.

Comparison of New and Old Thermodynamics

1. Logic of the Second Law of Thermodynamics: Subjectivism, Logical Jump, Interdisciplinary Argumentation.

2. New thermodynamics pursues universality, two theoretical cornerstones:

2.1 Boltzmann formula: ro=A*exp(-Mgh/RT) - Isotope centrifugal separation experiments show that it is suitable for gases and liquids.

2.2. Hydrostatic equilibrium: applicable to gases and liquids.

3. The second and third sonic virial coefficients of R143a derived from the new thermodynamics are in agreement with the experimental results.

3.1. The third velocity Virial coefficient derived is in agreement with the experimental data, which shows that the theory is still correct when the critical density is reached.

4. See Appendix Pictures and Documents for details.