Estimation of Modified UNIFAC (Dortmund) Model Parameters

[mayurchem]

Hello Folk,

I would like to ask you guys one question, which is related to the solving of UNIFAC model in Aspen Plus.

Actually I am trying to find the group-group interation parameters by using the experimental heat of mixing data of different amines.

I would like to describe my problem by taking structural formula of 2-(Methylamino)ethanol. The structural formula of this amine is CH3-NHCH2-CH2-OH, thus i considered the groups as following to find the group-group interation parameters between them,

CH3
CH2NH
CH2
OH
and the second component is water.

Despite the fact that OH and CH2NH are entirely different groups from each other, I am getting error in Aspen Plus that "OH and CH2NH groups are belong to the same main group and thus remove one group from these two groups" and so i dispatched one group from these two and because of this i couldn't be able to find any intractions of one of these two group with the other groups. When i removed OH group then aspen solved this model with very good accuracy but then the question is what about the interation of OH group with other groups.

so i would like to know why such error is appear.

 

[UmeshMathur]

You said:

Despite the fact that OH and CH2NH are entirely different groups from each other, I am getting error in Aspen Plus that "OH and CH2NH groups are belong to the same main group and thus remove one group from these two groups".

This clearly shows some sort of bug in your program that should be referred to Tech Support for the program.

Also, despite this error, you concluded that you have good accuracy. Was this verified against some independent source of data?

Finally, using heat of mixing data for estimating group-group UNIFAC interaction parameters does not ensure accuracy for estimating VLE or LLE using these parameters. This is because the UNIFAC theory is not good enough to permit a universal set of main group interaction parameters for VLE, LLE, and heats of mixing. (For example, there are two entirely different tables for VLE and LLE group-group interations).

 

[mayurchem]

Umeshmathur,

I said when i removed one group, lets say OH (Which is not the proper way), i am getting good fit of my experimental results. Obviously such fit is wrong as i am not considering OH group because of the error.

I think its a bug in the programme.

I am fully agree with your statement that "UNIFAC theory is not good enough to permit a universal set of main group interaction parameters for VLE, LLE, and heats of mixing".


Thanks

 

[mayurchem]

Umeshmathur,

I would like to ask you one question. Structual formula of 2-(Methylamino)ethanol is CH3-NHCH2-CH2-OH, in which i am sure that OH is a primary alcohol. But to check in aspen plus i selected secondary alcohol, and it worked well. So i would like to know that is this solvent contain primary alcohol or secondary. I am confused, because with primary alcohol i am getting the error as i mentioned in my earlier post, but with secondary alcohol it runs perfactly fine.