10 replies to this topic

[polipo]

Hi there,

I wonder that "is it necessary to make the temperature of the feed to a distillation column in between the temperature of condenser and reboiler?"

Or it is make more sense to keep the net absolute temperature of heater/cooler (before entering dist column), condenser and reboiler as lowest as possible?


Thanks for the answer.

[Zauberberg]

Looking from the enthalpy balance point of view, more heat that is supplied with the feed results in unloading the reboiler, but also in loading the overhead condenser as more reflux is required to maintain the column top temperature than in the case of colder feed. Additionally, this means increased vapor-liquid traffic in the column above the feed point, and reduced vapor-liquid traffic below the feed point. In many cases, this results in capital cost saving, i.e. the tower diameter usually can be lower as it is usually governed by the point of highest vapor load which is the bottom tray.

The best is to introduce the feed (at given temperature) on that particular tray in the tower having the same (or closest) temperature, as the compositions of feed and internal reflux are similar as well. During design stage, one needs to evaluate all the effects of feed preheating: reboiler duty, condenser duty, tower diameter etc.

There is no point in cooling the distillation tower feed (if it comes hot already) as you have to reboil it inside the tower again.

[Art Montemayor]

Polipo:

This is a basic distillation question and you should go back to the basics in process design - the McCabe-Thiele Diagram used for a stage-by-stage design of a distillation column.

If you do this you will have to apply the principles of the famous "q" line. It has been many years since I dealt with this subject, but I still remember the importance of identifying the orientation of the q line. Common sense tells you that if you have cold liquid feed, you will require more reboiler heat and stages to heat it. Common sense also tells you that if your feed is introduced as a saturated liquid, it will immediately start to distill within the feed tray because it is on the verge of receiving latent heat. Therefore, saturated liquid feed is desirable.

You don't fix your feed temperature because of your condenser or reboiler temperatures. You do it the other way around. Your basic query is mixed up or you are not communicating correctly. I hope I have understood your query and answered it.

[polipo]

Well, thank you guys.

But my case is that the feed has higher temperature than the reboiler. If the feed temperature should be in between condenser and reboiler, then I need the cooler. From my trial and error with simulation, I got that the feed temperature which is higher than reboiler requires less net absolute heat. Herein, net absolute heat means absolute heat from condenser, reboiler and cooler.

[Art Montemayor]

Polipo:

You are missing my most important point. Forget about the simulator and concentrate on common sense. This is not “rocket science”. Of course you are going to have “less net absolute heat” if you have the feed temperature higher than the reboiler temperature! You don’t need a simulator to tell you that. Just look at the q-line position. That tells you exactly what you are doing. The simulator tells you nothing about what you are doing; it is YOU that is supposed to tell the simulator what to do. You have your thinking cap on backwards.

Both Zauber and I are confused with your “cooler”. Why don’t you simply supply some of your basic effort and time in the shape of a process diagram to show us what you are trying to put together. I suspect that you have pre-heated or over-heated feed and what you should be looking at is a flash distillation in a flash drum before introducing the saturated liquid into the feed tray. But I can’t recommend that if you don’t describe for us what you are trying to distill and how.

By not heeding what we are telling you and insisting on cooling the feed you are simply wasting useful energy that could be used to contribute to an efficient separation – if you know what you are doing and you start using your own distillation knowledge instead of thinking that the simulator is going to do your thinking and decision-making for you. Use your magnificent human brain to study and plan the solution algorithm and let the stupid computer and simulator do the boring and tedious number crunching that it loves to do. That is real engineering. For your pleasure I am attaching some reading matter.

Attached Files

•  McCabeThiele Method.pdf   87.85KB   215 downloads

[Zauberberg]

Polipo,

Art is correct. We are wasting time here discussing about general issues in distillation, whereas your case needs to be presented in all its aspects before we can make any progress. What Art is suggesting to you, is to spend a fairly small amount of time in preparing the basic PFD including stream data and product quality requirements.

There's plenty of options depending on what you need to separate, and what quality of separation is required. Pre-flash followed by distillation, or utilizing feed stream as a heating medium in the column side-reboiler, or absorber-stripper configuration - all these might be what you need. But this is impossible to tell without knowing the very basic data and the scope of your problem.

[polipo]

Dear all,

Attached is my PFD. My feed is being compressed first, followed by cooler because the temperature is shooting too high. The need of being compress is that the pressure of the dist column is around 12 bar while the condition of raw feed is 1 atm, 25 Celsius. Once gas is compressed, its temperature will be higher. That's why i have to cool it down.

Thanks

Attached Files

•  PFD.JPG   12.37KB   48 downloads

Edited by polipo, 02 December 2010 - 08:40 PM.

[Art Montemayor]

Polipo:

Why are we playing games hiding basic data? Is your basic data Proprietary Information that must remain a secret and therefore, you can’t reveal to us what you are doing?

If you don’t want to (or don’t care to) give us your basic data, simply tell us. We will respect your right to keep it a secret. However, don’t expect us to break a leg trying to help you out.

How can you be feeding a compressed GAS to a distillation column?

Tell us the composition, temperatures and pressures of your feed stream – from the “compressor” to the distillation column.control feed valve. Tell us your overheads composition, temperature and pressure as well as your bottoms composition, temperature and pressure.

Giving us a generalized Hysys flow sketch doesn’t give us any basic data to start to answer your specific query. We need ALL your basic data. We need YOUR HELP in order to help you.

[polipo]

Sorry you guys for missing information.
I just want some guide so that i can think further by myself. Anyway, here is all my data.

After compresser, the stream has the pressure of 29.38 bar, temperature of 157 Celsius with the composition as followings (kmol/hr):

ethylene 3669.276
propylene 7.339
n-butane 92.111
i-butane 23.028
1-butene 293.542
cis-2-butene 97.236
trans-2-butene 121.086
isobutylene 502.691
1,3-butadiene 703.957
methyl acetylene 0.003
vinyl acetylene 5.927
ethyl acetylene 1.903
n-pentane 7.135

Condition of the distillation column is as followed:

Top P = 415 psia, T = -15.23 Celsius
Bottom P = 425 psia, T = 124.0 Celsius

Overhead composition is ethylene 99 mol% and propylene 1 mol%.
Bottom composition: ethylene 1 mol% propylene 99 mol%.
These are for 2 key components. Is that enough, or you need whole composition?

These conditions were obtained by following the method of Seider et al.
I wanna separate ethylene from propylene.

Edited by polipo, 03 December 2010 - 12:30 AM.

[Zauberberg]

Polipo,

You can't have the overhead and the bottom product compositions fixed in the way you propose. Where are the other components from the feed stream, apart from Ethylene and Propylene? And certainly you can't have 99% Propylene in the bottoms when there is equal amount of C5 and even higher content of C4 components in the feed stream. Something is conceptually wrong in the design you are proposing. You'll definitely need more than one column in order to achieve the desired separation into Ethylene (1), Propylene (2), and C4+ (3) products.

I would look at the feed stream conditions downstream of the compressor cooler, and try to evaluate the following:

Separate vapor and liquid in a preflash drum (I can only guess the vapor fraction of feed downstream of the cooler because the temperature is not known), then strip the residual C2= from liquid feed while recovering C3+ from the vapor stream (something similar was suggested by Art), probably in an absorber-stripper configuration. This is to be followed with a De-C3= tower for making high purity Propylene, assuming that such purity can be achieved considering other components present in the feed (you can evaluate this by using simulator).

Of course, this is only a general suggestion since other data are missing, and I believe that's the reason for Art being reluctant to guide you further. If I find time, I'll try to build the model representing your case and will come up with additional suggestions.

[polipo]

Sorry. My info was incomplete.

Actually, the composition of the overhead and bottoms was percentage of feed, i.e., 99 mol% means 99 mol% in feed comes out at the top.