If you want to find accurate paramethers for a cubic equation of state in a specific temperature you have to find specific parameters for that compound and for that temperature. So, don't use the original "a" and "b" parameters for PR EOS, but find that parameters using the equilibrium properties. Summary you have to find specific a and b, with the equality of fugacity for that temperature.

Hope this is helpful.

@Pierantozzi

thank you for your attention. I believe that the predictive equations of state are better than EOS with optimized parameters for numerical methods

thanks a lot  

“I believe that the predictive equations of state are better than EOS with optimized parameters for numerical methods …”

Probably not, because optimized parameters are optimal. The question is (a) which properties do you want to represent well and which not (a simple cubic EOS cannot match all), and (b) do you have experimental data to which the EOS can be fitted. If not, a predictive method is evidently required. You can fit your own parameter set on the ThermoC website.

Molecular dynamics (or Monte Carlo simulation techniques) can be used to generate pseudo-experimental data if no experimental data are available—if you have a reasonably good interaction potential (“force field”) and a fast computer.

http://thermoc.uni-koeln.de/index.html

@Ulrich Deiters

 Thank you so much for all the information, it will be very useful. I believe that the evolution of science is only possible if we cooperated in this manner, discussions and debates are of paramount importance for the scientific development

regards