2 replies to this topic

[mrbabu]

I have to create several tables with values of liquid densities (for generic hydrocarbon mixtures) at different temperature and pressures,

I use PRODE PROPERTIES and Excel, 

the method =StrLD(stream,t,p)

returns directly the calculated liquid density at specified temperature and pressure,

for liquid density PRODE allows to specify different models as Extended PR or SRK or a EOS with Peneloux correction,

which is the best to adopt for hydrocarbon mixtures (C1...C13) in a range of temperature and pressures ? 

It would seem that Extended PR or Extended SRK are more accurate that equivalent versions with Peneloux correction, is that true ?

[PaoloPemi]

for liquid densities with EOS PRODE PROPERTIES allows three options (in addition to std. version which calculates density and volume from EOS),

1) std. version with Peneloux correction (available for both Peng Robinson and Soave Redlich Kwong EOS)

2) extended EOS with 3 parameters calculated to fit vapor pressure, heat capacity and liquid density of each pure component

3) extended EOS with 5 parameters calculated to fit vapor pressure, heat capacity and liquid density of each pure component

 

option 2) is available for SRK, PR, CPA etc. in base version

option 3) is available for SRK, PR, CPA etc. in extended versions

 

option 1) is the less accurate (but better than density calculated from EOS without correction)

option 2) is accurate (generally within 5%) for most hydrocarbons

option 3) is the most accurate

 

in Excel you can calculate the density at specified temperature and pressure by entering the macro

 

=EStrLD(stream,t,p)

 

this macro solves pahse equilibria at specified t and p and then returns the liquid density calculated with selected models.

Edited by PaoloPemi, 27 June 2013 - 01:05 AM.

[mrbabu]

thanks Paolo,

I have evaluated liquid densities as calculated by PRODE PROPERTIES for propane, ethane and some mixture plus water and ammonia,

the extended PR EOS with 5 parameters gives errors less than 2% which is a good result

(compared with Peneloux and other methods)