I need to calculate the solvation free energy of a solute in dimethylacetamide (DMA). Since AMBER does not have the force field for it. I am testing whether I could use GAFF as DMA force field. The density is ok comparing with experimental data. I saw people used heat of vaporization to testify the solvent force field, so I want to do that.
Literatures gave the similar equation to calculate it.
deltHvap = Eintermolecular + RT
I can understand the principle. But I am confused about the unit of each term. AMBER should use kcal/mol as the default energy unit. But the sum of EELEC and VDW as the intermolecular energy is extremely high. At 300K, RT is only 0.6kcal/mol. It doesn't make sense to add the two term simply.
I think I misunderstood some key points. Could you guys give me some help?
A fragment of AMBER output is attached.
You need to divide by the number of molecules. Formally:
dHvap = < Ugas > - < Uliq >/Nmol + RT
There may be different conformational sampling and therefore bonded energy in the gas phase, so typically one does a simulation of one molecule in vacuo to get .
You need to divide by the number of molecules. Formally:
dHvap = < Ugas > - < Uliq >/Nmol + RT
There may be different conformational sampling and therefore bonded energy in the gas phase, so typically one does a simulation of one molecule in vacuo to get .
Yes, as Justin Lemkul said the simulation in vacuo can be essential. Dimethylacetamide is a very polar molecule. Perhaps that the simplified formula is not enough here. I would have considered the gas phase too.
Thanks! Justin had helped me to figure out how to calculate it. And I posted the solution on my blog.
Dear Xiaoquan, I would like to add some extra information: The formula
dHvap= -Upot(liq) + RT can be used only in the case of rigid force fields, excluding any kind of intramolecular interactions. Otherwise, as Justin and Alexey mentioned, you have to perform a simulation in vacuo (no periodic boundary conditions needed) in order to get E(gas) and use the formula
dHvap= E(gas)-E(liq)+ p( V(gas)- V(liq) )
Since the volume of the liquid is negligible in comparison with the volume of the gas and using the assumption that the gas is ideal so that the kinetic energies of a molecule in the gas and liquid phases are identical then you can use the formula:
dHvap= Upot(gas)-Upot(liq) +RT, which is a very reasonable approximation.
(Just keep in mind, as Justin mentioned, that you need to divide the values that you got from your simulation input with the number of molecules!)
- Badges
- Science topic