8 replies to this topic

[Trotsky]

Hi, I'm new the forums and the reason I'm here is thermodynamics has forced me to succumb to such acts of desperation.

My current thermodynamics teacher has recently assigned a project regarding an "azeotrope-like" composition consisting of dichlorotrifluoroethane, methanol, cyclopentane, and nitromethane along with the journal article explaining its applications and some properties. I've attached the article if you're interested in perusing it.

Anyways, specifically I'd like some input on where or in what literature one might find vapor-liquid equilibrium data for some of these mixtures. But, more importantly, what Gibbs Free Energy model should be used when calculating azeotropes in this four component mixture. The CALPHAD method keeps popping up in my research and I'll check out a book tomorrow over it but I'm not sure if its the right approach. I'm not really looking for any solutions here just a nudge in the right direction so I can focus the research that I have to do.

Any help would be appreciated, thanks.

Attached Files

•  azeotropeproject.pdf   304.14KB   66 downloads

[MrShorty]

Anyways, specifically I'd like some input on where or in what literature one might find vapor-liquid equilibrium data for some of these mixtures.

The first place I look for measured VLE data is the DECHEMA's Chemistry Data Series: Vapor-Liquid Equilibrium Collection. They seem to have a pretty thorough collection of what VLE data is available in the literature.

But, more importantly, what Gibbs Free Energy model should be used when calculating azeotropes in this four component mixture.

I won't give a direct answer for this, in part because I wouldn't make this determination until I had gathered my VLE data. After I have the data in hand, then I could decide which model best fit the data. Among the considerations I would use would be to use a model that will calculate multicomponent VLE from binary parameters only.

Also recognize that, sometimes, this question doesn't necessarily have a "right" or a "wrong" answer. This is the kind of question that comes down to best judgment based on the available data.

[Trotsky]

In this problem you're asked to determine whether the ternary mixture (assuming ternary since nitromethane is negligible) has an actual azeotrope. Now that we have all three binary VLE models how do we go about modeling a ternary system? Is there any good literature on this?

[MrShorty]

The topic should be covered in any phase equilibria text. The texts I've used are Molecular Thermodynamics of Fluid Phase Equilibria by Prausnitz et al., The Properties of Liquids and Gases, also by Prausnitz and Poling and others, and Phase Equilibria in Chemical Engineering by Stanley Walas. Any of these, or similar texts you have access to should be able to explain how to extend the binary calculations to ternary systems.

[Trotsky]

Thanks for the information, it was quite helpful.

We have decided on the Van Laar equation not only for simplicity sake but our instructor seems to be urging us in that direction. Now, we have to prove whether or not this ternary system of ours has an azeotrope at some point. I don't think it matters what method we use to prove or disprove the existence as long as we can support it. So, is there any concise way to approach an azeotrope modeled in a Van Laar ternary system? (Remember we have the data on the 3 binary systems)

Thanks again, for the responses.

[MrShorty]

In general, we aren't fond of doing students' homework for them. At this point, I think I need to know what you know about solving this problem, so I can help you see how to apply what you know to the problem at hand. I prefer this to simply telling you how to do it (I know it's harder, but I think it's important for learning). So, in my best Socratic method:

1) You've chosen the Van Laar equation. That will be fine for our purposes. What quantity will you be calculating using the Van Laar equation?

2) There are many quantities in thermodynamics that are somewhat abstract concepts. How will you relate the quantity calculated using the Van Laar equation to the physically meaningful quantities of temperature, pressure, and composition?

3) How are you defining "azeotrope", and what are some "properties" of azeotropes? While answering this question, try to look for relationships to pressure, temperature, and/or composition, so you can use the result from question 2 as a bridge between 1 and 3.

[Trotsky]

I think I see where you are going. Let me try to respond the best I can.

This ternary Van Laar equation can be utilized to solve for the three activity coefficients that will be used to plot the ternary data. This in turn can be easily related to sum(x_i*gamma_i*Pvap_i)=total pressure. Now, Pvap is naturally dependent on temperature, and either that or the total pressure will be fixed. Since the so called "azeotrope" exists around 29 degrees celsius we can simply fix the vapor pressure and alter total pressures.

To find an azeotrope x1=y1,x2=y2,x3=y3, these are liquid and vapor mole fractions, respectively. Now, for binary systems. There was also a relationship between the activity coefficient, total pressure, and vapor pressure (at low to moderate pressures), however, I doubt this is still the case. I tried graphically approaching this by modeling the residue curve of the ternary system on CHEMCAD to try and locate an azeotrope that way, but to no avail. Which brings me to another problem in graphing, but that can be resolved after I find whether there is a pesky azeotrope. I have an excel with all the values on it, and the ternary van laar equation already set up. I could theoretically play around with this altering the two variables, x1 and x2 (since the 3rd component is simply 1-1stcomp-endcomp), but that seems entirely too burdensome for a trivial azeotrope value.

I know I'm missing something obvious, I just don't know what. But thanks for the direction.

[MrShorty]

Your first relation relates pressure, temperature, and liquid composition. Your definition of an azeotrope relates liquid and vapor composition. How do you "bridge" those two relationships (or, perhaps another way, how can you relate vapor composition and total pressure)?

Your description of an azeotrope is a pretty basic definition. Are there other properties of azeotropes (other than the defining yi=xi) that might help determine if there is an azeotrope?

Why does your proposed "brute force" method seem too burdensome? I wouldn't be surprised if there is a relatively simple relationship that could be derived, but sometimes I find that it takes as much or more effort to derive the simple relationship than it does to do the brute force calculation (especially once you have a spreadsheet set up). It's up to you how you would want to proceed, I might try several calculations to see if it helped me see the simpler relationship.

As I wrote this response, I was reminded of an article I read many years ago. The authors don't describe in great detail the calculational procedures, but they do discuss basic binary and ternary azeotrope concepts. It may help if you can get a copy. It's "Designing Azeotropic Distillation Columns" by Stanislaw Wasylkiewicz, Leo Kobylka, and Marco Satyro in the august 1999 issue of Chemical Engineering.

[Trotsky]

Finished this not long ago and gave a presentation on it, just thought I'd wrap up the thread. I ended up just graphing a multicomponent van laar system using Mathcad and modeling the pressure. It yielded a three dimensional graph with a pressure maximum that one could conclude was a ternary azeotrope, quite simple actually once the binary parameters were established. Just wanted to thank you in particular MrShorty, you, sir, are a gentleman and a scholar. Thanks.