Gas Solubility calculations with Aspen Plus

[Camarena]

I am running a simulation with Aspen Plus to study gas solubility in alcohols, the gas feed is compose by H2 O2 N2, the operation conditions are P= 90-150 bar (high pressure) and low temperatures (30-70 ºC). I would like to know what property method of those available in Aspen could calculate the gas solubility in alcohol more accurate, because I have tried with several of then (PRWS, RKSMHV2, RKSWS, PSRK, RKSMHV2) and there are big differences between then.

Thanks

 

[hacksaw]

I don't think you will get precise results with generalized correlations. you need correlations built on specific lab data .

 

[phex]

at least you should choose the property method by evaluating against lab results. take the one with the best fit to real data.
thats more or less all advice i'm able to offer...

 

[EGT01]

Unless you can get a response from someone who has specific experience with your system, you may want to try Aspen technical support center.

http://support.aspentech.com/SupportPublic/AllFrameDef.html

Your key info may be the Henry's Law data for your system. I'm not sure if that would vary between the thermodynamic methods you mentioned but how the different methods use that data could vary.

As previously mentioned, it would be a good idea to validate your simulations against either lab data or actual operating data.

 

[pmureiko]

You need to invoke the Henry's law package in the Apsen screen where you choose the overall Thermopackage. From there it will prompt you to review the Henry's law constants.

Can you provide some details about what you are trying to model? I modelled the degassing of hydrogen from alcohol in atmospheric storage tanks for my previous employer and found that hydrogen will accumulate and form an flammable mixture. Also I know of a plant in Italy a few years ago that had an explosion and fatality because of degassing in the storage tank.

 

[sshep]

As per previous guidence you should use an activity coef method (ideal or other) and henry's- none of these EOS methods can be trusted for your system (esp if no RKS-ij params are being used). If you have solubility at 2 temps, regress henry-1/2 yourself. Use the available databank henry's data for lower MW alcohols (butanol, etc) as a reality check for your data. What alcohols?- May have some data applicable for your system.

pmureiko, write me: sean.shepherd@sasol.com