Jump to content



Featured Articles

Check out the latest featured articles.

File Library

Check out the latest downloads available in the File Library.

New Article

Product Viscosity vs. Shear

Featured File

Vertical Tank Selection

New Blog Entry

Low Flow in Pipes- posted in Ankur's blog

Correlating Experimnetal Vle Data


This topic has been archived. This means that you cannot reply to this topic.
1 reply to this topic
Share this topic:
| More

#1 g3wtter

g3wtter

    Junior Member

  • Members
  • 10 posts

Posted 05 January 2017 - 10:37 AM

Hello,

the system is alcohol and water + pH-buffer that is supposed to alter the volatility of the mixture and I need to correlate the experimental data (x1,y1,T measured at constant P).

Question: Can I use binary models like margules, van Laar and NRTL to correlate the data and see if the models represent the experimental values correctly or do I need to treat this system as ternary?

If I need to treat it as ternary I dont know which model to use and how to use it because all the models I found are ment for a mixture of three liquids. I also found the eleyctrolyte-NRTL model but I am fairly certain that I will not find the literature parameter required for the pH-buffer that I am using.

any help would be appreciated
kind regards



#2 MrShorty

MrShorty

    Gold Member

  • ChE Plus Subscriber
  • 517 posts

Posted 05 January 2017 - 05:18 PM

I don't think there is any getting around using an electrolyte model, if you plan to do this rigorously. So, I guess that next question, perhaps, is to understand what you really need to do with this correlation and how rigorous do you want this to be.

 

I think this kind of question is often only answered by trial and error. If you want to try modeling the system as a pseudo-binary, set up your pseudo-binary model, correlate the data, then see how well the resulting model fits the measured data. I would not be surprised to find that, for small ranges of temperature and composition, a pseudo-binary model can be made to fit the data. Or, you might run the correlation and find that the model predicts all kinds of wrong things compared to the data, and decide that it is completely inadequate.






Similar Topics