Calculating Dew Point of a Specific Gas in a Mixture 1

[ME27272727]

I was trying to get a quick ballpark on what the cooling load and entering water temperature to a condenser, and quickly realized there is a lot of effort that would have to be put forth to calculate this load. 1500 CFM mixture at 95°F, made up of CO2, ethanol, ethyl acetate, air, and acetaldehyde. Acetaldehyde is the target gas that needs to be condensed. It makes up 0.5% of the mixture, and 90% of it needs to be condensed. With it being such a small percentage of the mixture, I would think an extremely low condenser temperature would be required, much lower than the boiling point of 68°F at 1 ATM. The process engineers will be doing these calcs and getting back to me, but I was trying to get a rough idea of cooling load and temp before then, and was curious to know the correct way to come up with these numbers. Thanks.

 

[zdas04]

This is the way that every cryogenic plant in Oil & Gas works. If the boiling point of your product is 68F, then any temperature under 68F will result in condensation, eventually. The closer you keep the vessel to 68F, the longer "eventually" is. I doubt that you are actually keeping the vessel at atmospheric pressure, so you need to find a boiling point vs. pressure table for your product and find a temperature that is a compromise between residence time, boiling point, and the boiling point of the other products. The colder the vessel, the faster the condensation. The smaller the vessel, the shorter the residence time.

[bold]David Simpson, PE[/bold]
MuleShoe Engineering

In questions of science, the authority of a thousand is not worth the humble reasoning of a single individual. Galileo Galilei, Italian Physicist

 

[jari001]

You can calculate the dew point via an iterative script and maybe the Henry's Law constant for acetaldehyde, but your process engineers will likely do this with some software or an excel sheet. Ethanol and ethyl acetate have boiling points higher than 95°F so this should be a two-phase stream? Is your mixture an azeotrope that has a significant amount of these higher boilers (wrt. acetaldehyde) into the vapor phase? I'm assuming you have a single vapor phase because I always think gas when I see CFMs, correct me if I'm wrong.

My scrap work would be something like this:
Assume: no liquid phase, isothermal, well mixed
*Think about assumptions over coffee*
*Open thermo textbook*
Convert 7.5 CFM (0.5% of 1500 CFM) to mass flow at 95°F
Use heat of vaporization and mass flow to determine the minimum heat duty
*Use heat capacity to determine further cooling requirements*
Find out what your utilities are running at and approximate the heat transfer needs that David mentioned

 

[ME27272727]

I'm not sure I agree with your statement that some condensation will occur below 68 degrees. Well, some may occur, but at that partial pressure, the rate of evaporation would be much higher, essentially amounting to no condensation. In my mind, that would be like an air handler heating coil at 160 degrees condensing because it is below 212.

This is exhaust from a tank being vented to atmosphere, kept around atmospheric pressure. 0.5% of 14.7 PSI gives us a partial pressure of 0.0735 psi for acetaldehyde. The lowest vapor pressure I found on the substance was 2.7 psi at 0°F. If I followed the trend, it looked like I needed to get well below zero to start condensing.

Am I correct in this line of thinking?

 

[don1980]

Condensing a mixture is vastly different from condensing a pure component. There's just no simple way to look at a mixture like this and estimate the necessary coolant temperature. The alcohol and aldehyde will make this mixture behave nonideally, and the CO2 and air will greatly reduce the dew point. A low coolant temperature is going to be necessary to condense out a high percentage of these hydrocarbons

Mixtures condense over a range of temperatures, rather than at a single temperature, because the vapor composition progressively changes as the stream passes from the condenser inlet to the condenser outlet. The vapor's dew point at the inlet is relatively high, but it decreases as the mixture passes through the condenser. The extent to which it decreases depends on the mixture composition. The presence of even a very small amount (1%-2%) of "noncondensable gas" (e.g. air, CO2) has a disproportionately high impact. You need to run a process simulation (e.g. Aspen), using a proven physical property set, to estimate the necessary coolant temperature. Without that, you're just shooting in the dark, unless you can reference an existing condenser that is condensing a very similar stream.

 

[ME27272727]

Thank you Don. The manufacturer and process guys will be doing those calcs. I just wanted to see if I could shoot a dart at a condensing temperature while waiting on them. Things usually much simpler in my world of steam / water.

 

[Latexman]

In this application, "quick ballpark" [≠] "the correct way to come up with these numbers". Best to wait on your process folks.

Good luck,
Latexman

To a ChE, the glass is always full - 1/2 air and 1/2 water.

 

[apetri]

a problem, with that mixture, could be the identification of the correct parameters for the thermo model used for VLE,
nowadays many different models, say, for example, EOSs with complex mixing rules which surely can do the work, are available in simulators and even in tools as Excel (i.e. prode),
but an experienced engineer is required to tune the model in order to obtain accurate results...

 

[Latexman]

Thanks apetri for that information, and the Prode advertisement.

Good luck,
Latexman

To a ChE, the glass is always full - 1/2 air and 1/2 water.

 

[apetri]

of course that's a possibility, if one don't have aspen as previously suggested...

 

[Latexman]

Touche, mio amico.

Good luck,
Latexman

To a ChE, the glass is always full - 1/2 air and 1/2 water.

 

[apetri]

no problem,
anyway we know little about the problem (no detailed information given) and I am afraid there is no way to suggest a "correct way to come up with these numbers" as requested by ME27272727

 

[Latexman]

Agreed, it would take a reply that fills a tome to cover all the possible unknowns.

Good luck,
Latexman

To a ChE, the glass is always full - 1/2 air and 1/2 water.

 

[georgeverghese]

If you are trying to get 90% recovery of acetaldehyde in a single stage flash, you'll need a much lower operating temp than if you were to run a column with an overheads condensor and a reboiler. With more desk engineering effort, an extraction solvent with good selectivity for acetaldehyde would reduce your utility cooling and heating loads considerably. An azeotropic column is usually much more troublesome to operate.

A non ideal mix like this is a hornets nest when it comes to getting reliable thermo physical data - there could be all kinds of wierd azeotropic phases - this is most likely a specialist field many seasoned general skills process engineers will fear to take responsibility for.