How do I calculate boiling point of a 3 part mix?

I wouldn't run to my professor until someone confirms, but I am pretty sure they do not average.

so I would expect the lowest boiling point material to "wick off" first, followed by the second lowest boiling point, etc.

This is how distilling is done.

Just for giggles I tried to work this out for you and hit a brick wall before getting out of first gear. It's been too long.

Dang if Chemistry isn't just about the closest thing to black magic we have. Calculating the simplest reactions always seem to involve jumbling around ions and moles and Avacado's number. Then you multiply by light years and subtract the weight of the beaker.

Don't see how you chemists do it!

I'll stick with electronics. Nothing mysterious about that at all.

This link will help you to understand the basics -

http://www.chemguide.co.uk/physical/phaseeqia/idealpd.html

You may want to search literature for actual results from lab measurements that have documented over the years.

Hi,

Thank you very much. It helps.

Job.

This sort of calculation is the task of Chemical Thermodynamics. Frankly, the calculation is very iterative usually starting of with the assumption of the fugacity as equals 1 and then iterating until the changes becomes than 0.001.

Many years ago when I used to make these calculations, I also referred to the book, Chemical Thermodynamics by Van Ness; a copy of it should be in just about every library that is associated with Chemical Engineering, the publisher is McGraw-Hill.

Alternatively look up any book on Distillation Engineering and you should find the computational procedure.

Good luck

Hi,

Thanks; Iwill.

Job

I would also recommend:

Chemical and Engineering Thermodynamics (Sandler, Wiley Series in Chemical Engineering)

Perry's Handbook for Chemical Engineers. I think somewhere around Chapter 13 is the section on Distillation.

Separation Processes (King, McGraw Hill)

Also here is a link that may help:

https://eee.uci.edu/programs/hongchem/RDGbpdistill.pdf

People, homeworks should not even be discussed here! Let these kids hit the wall, that way will wake up and learn about life a bit. They know there is a library out there, but too busy for that...

What made you think that this is a home work?

Hello Paddler,

Thank you verymuch

Job

This seems easy enough for ideal solutions. However I don't think these 3 chemicals when mixed together behave as an ideal solution. Experiment or molecular simulation can verify this.

For non-ideal mixtures, taking the weighted average (Raoults Law) of the individual vapor pressures is inaccurate. Short from using lab data for tertiary mixtures, more rigorous calculations, which utilize activity coefficients (for liquid phase) are necessary. The coefficients correct for the deviations from ideality and can only be derived from experimental data.

This is where chemcad and other simulation softwares come in handy. Given there is experimental data pre-loaded to regress to find the coefficients to fit the VLE curve.

Boil it and stick in thermometer!

Although sticking a thermometer is a cute idea, it will not work with accuracy to do meaningful design work. Unfortunately from the moment it starts boiling the boiling point or bubble point drops, unless it is an azo-trope. Even as it is heating to boiling the concentration is changing and lowering the boiling point (unless you saturate the vapor space with the correct mixture of the bubble point concentrations) and the reading will not be of the mixture but of a partially separated mixture.

Dear Lemmy70,

My question was "How do I calculate"; not "Ditermine". May be I will rephrase the question for you. How I find the ratio of ingradiants in a three part mix of water, ethyline glycol and methanol whose boiling point is 95 degree celcious, and freezing point below 20 degrees? (Glycol in the mix is to lower the freezing point below 20 and obviously I have to add methnol to lower the boiling point of the aqeous glycol mix).

By trial and error and sticking a thermometer, as you suggested, I will be able to determine the ratio of a number of mixes whose boiling point equals 95 and freezing point below 20 degree C. Now suppose that at the time of my experiment, the atmospheric pressure was only 750 torr, what do you think the boiling point should be? So do you think that it is that easy as you think? I do not believe that your answer was "Near good" as rated.Thank you for the suggestion any way.

Job

I still maintain that if you have the vapor pressure curve for each component, all the deviations from theory are taken into account. The partial pressure from each component is the molar concentration times its vapor pressure. Add them up and viola!

The only deviation I can see is if there an component not considered, like the boiling point elevation from salt, or lack of nucleation locations allowing super heating. I I will look at some tables today and try to estimate it the "old, manual way" and compare it to chem cad.

Thank you Paddler;

This is the kind of answer that I am looking for.

Job.

Now you have changed the original question somewhat slightly. In any event as I had written before, the method of calculation is clearly explained in the book [Introduction to] Chemical Thermodynamics by [Smith and] Van Ness.

If my memory does not fail me, the method is convoluted: You need the fugacity of the vapor phase to calculate the activity coefficient of the liquid phase, but you need the activity coefficient which accounts for the non-ideal state of the liquid, to calculate the boiling point (BP) Temperature, then, of course you need the BP Temperature to calculate the fugacity.

So in general, you start off by assuming that the vapour phase is ideal by assuming the fugacity to be equal to 1; and then you begin the iteration, and the fugacity will change because the vapor phase is non-ideal in general. In fact if you digressed a little bit into Quantum Mechanics of dense gases which best describes vapor dynamics, you will realize that the vapor-phase is non-ideal in general.

So you have to refer to the book by Van Ness to find your solution to the problem.

That said now here is an approach to the problem as you have posed it now: There are three(?) equations with the following representations:

• f = n(x_i, T_b, P) where xi are the vector set of liquid concentration molar ratios/fractions, T_b is the boiling point and P is the operating pressure;
• g(a, f, x_i, P) = 0 where a is the activity coefficient, f is fugacity;
• Tb = k(a, x_i, P);
• 1 = ∑x_i
  

Now to solve your problem think of it this way:

Q(xi, a, P, f) = {1,Tb};

keep in mind that Q is a nonlinear function, so you will now use any of the many Numerical Methods of solving nonlinear equations to obtain something like this:

y = Q^-1(1,T_b) where the parenthesis is actually a vector representation - written as best I can. Also keep in mind that this solution of course entails extensive iteration.

where y is a vector set {xi, a, f}. Now for a given boiling point Temperature you can now calculate x_i and accordingly adjust the concentrations. These can be easily developed into a software and embedded in a Microcontroller.

Again you must refer to the book, for a computational procedure, I have written this only from memory, so do double check for a more precise formulation, just in case you need additional equations in the formulation.

The vapours are not ideal and the liquid mixture is not ideal hence the need for the calculation of the activity coefficient.

I hope this helps. Good luck

I was jesting, to be frank, I wouldn't have a clue!

There's also another confounding item to consider.

The moment that the mixture starts to "boil" the relative concentrations in the remaining fluid will then be different and the boiling point of the remaining mixture will be different, the vapour will not be in the same proportions as the original mixture (until all is vapourised).

I remember my cousing making home brew rum. Distilation temperatures varied across a substantial range for initial removal of methanol, then ethanol, then the balance point where the majority of remaining fluid was water.

By the way, are you trying to make a thermal refrigeration cycle, similar to the old amonia/water/hydrogen systems powered by a small kerosene flame?