search

mostafa-razavi / ITIC-paper

0 stars 0 forks source link

Adding new example simulations #1

Closed mostafa-razavi closed 5 years ago

mostafa-razavi commented 6 years ago

We decided to add 2-5,dimethyhexane and 1-propanol to the Example Simulations section

ramess101 commented 6 years ago

@mostafa-razavi

I thought we said 2,5-dimethylhexane and cyclohexane, since 1-propanol is a lot slower in Cassandra.

I also have some branched alkane simulation results already completed (although without the virial calculations) if we want to use those instead. Namely, neopentane, isobutane, isopentane, isohexane, and isooctane.

mostafa-razavi commented 6 years ago

I thought we said 2,5-dimethylhexane and cyclohexane, since 1-propanol is a lot slower in Cassandra.

Yes, now I remember.

I also have some branched alkane simulation results already completed (although without the virial calculations) if we want to use those instead. Namely, neopentane, isobutane, isopentane, isohexane, and isooctane.

Nice. Are these GROMACS simulations or Cassandra?

ramess101 commented 6 years ago

Are these GROMACS simulations or Cassandra?

These are GROMACS simulations for the TraPPE and Potoff potentials. I did not simulate the lowest density points for the virial. So it might not be as useful after all. Let me know if you can make use of them.

mostafa-razavi commented 6 years ago

@ramess101 I'll definitely use them. Running low density simulations are not very time consuming.

ramess101 commented 6 years ago

@mostafa-razavi

OK, I will upload them in a new folder. Also, I have some results for the TAMie potential which we could compare to what Gross reported for the GC-TMMC approach.

ramess101 commented 6 years ago

@mostafa-razavi

I think the most useful compounds that I have for the Potoff Mie and TraPPE potentials would be 2-methylpropane (isobutane) and 2,2,-dimethylpropane (neopentane) because Siepmann has reported values for TraPPE using GEMC, and Potoff has reported GCMC values for both the Potoff Mie potential and the TraPPE model. This would be a very useful comparison of GEMC, GCMC, and ITIC.

Potoff has also reported values for the Potoff Mie and TraPPE potentials of 2-methylbutane (isopentane), 2-methylpentane (isohexane), and 2,2,4-trimethylpentane (isooctane). I have results for each of these compounds and force fields as well.

As for TAMie, I cannot find any literature values for the branched compounds that I simulated with TAMie. The only two alkanes that I simulated with TAMie are ethane and n-octane, for which we do have literature values.

But, we probably don't need to analyze all of these, since the information is likely redundant and we don't want to have to run the low density simulations for each of these compounds and potentials.

I think it would be good enough to compare the GEMC, GCMC, ITIC results for isobutane and/or neopentane with the TraPPE model.

The only reason to also include the Potoff Mie potential results for these compounds would be to show how well the Potoff potential performs at predicting low Psat. This is not really the purpose of the study, but could demonstrate the power of ITIC at providing additional low Tsat information.

ramess101 commented 6 years ago

@mostafa-razavi

OK, I have created a new folder that contains my Gromacs results for isobutane and neopentane using the TraPPE model. I have also provided the GEMC and GCMC values reported in the literature for both compounds with the TraPPE model. GCMC are from the recent branched alkane study of Potoff while GEMC are from the TraPPE website. Note that isobutane GEMC are from 1999 and neopentane has values from 1999 and 2015.

Also, note that I have not performed replicate simulations for bootstrapping. If we are sure that we are going to include these compounds, I will run a few replicates for that purpose.