Jump to content



Featured Articles

Check out the latest featured articles.

File Library

Check out the latest downloads available in the File Library.

New Article

Product Viscosity vs. Shear

Featured File

Vertical Tank Selection

New Blog Entry

Low Flow in Pipes- posted in Ankur's blog

Dew Point Of Acetic Acid


This topic has been archived. This means that you cannot reply to this topic.
18 replies to this topic
Share this topic:
| More

#1 M.Basaar

M.Basaar

    Junior Member

  • Members
  • 17 posts

Posted 03 April 2014 - 09:15 PM

Hi

 

I have this problem. I need to calculate the dew point of acetic acid in a mixture of the components as listed below in the photo attached,Attached File  222.jpg   29.74KB   10 downloads @ 1 bar and 35 degC. acetic acid is assumed 300-400ppm..make any assumption if needed.. please help on this

 

 



#2 M.Basaar

M.Basaar

    Junior Member

  • Members
  • 17 posts

Posted 03 April 2014 - 11:03 PM

any hints or guiding tips would be highly appreciated



#3 MrShorty

MrShorty

    Gold Member

  • ChE Plus Subscriber
  • 517 posts

Posted 04 April 2014 - 10:11 AM

Interesting problem. 1st thing I'd probably suggest is to clarify terms: What do you mean by "dew point of acetic acid"? Considering that acetic acid and water have almost the same boiling point, I doubt there is a point where acetic acid will condense out by itself, if that is what you mean. Second: Almost all the VLE calculations I do are done on a mole fraction basis -- it is not clear to me what units the concentrations are given in. Third: you indicate an acetic acid concentration of 300-400 ppm, your picture indicates 3100 ppm. The exact value is likely not important to describing the overall algorithm, but it will give different results.

 

The next big consideration will be choosing a thermodynamic model. It would be nice to know where you are at in your studies and what this exercise is supposed to be teaching you before making any recommendation. By far the simplest (computationally) approach will be to assume the gas/vapor is an ideal gas and the condensate is an ideal solution (Raoult's law). Assuming an ideal gas when organic acids are involved is not always a good assumption, even at low pressures. Perhaps for the educational purposes behind the problem, it doesn't matter as long as you state your assumptions, I don't know. Ideal condensate may not be a good assumption either, especially considering the presence of the xylene with water.

 

So, I would start by reviewing what dew point means. Then I suggest you review the various VLE modeling equations you know and choose one that will be suitable based on what you think you should be learning from the problem. Let us know if and how we can help you in this.



#4 M.Basaar

M.Basaar

    Junior Member

  • Members
  • 17 posts

Posted 17 April 2014 - 03:22 AM

@MrShorty 

 

Thank you

 

the compositions above are in %wt

 

is there any spreadsheet calculator or clear step by step way to calculate the dew point of the mixture?



#5 MrShorty

MrShorty

    Gold Member

  • ChE Plus Subscriber
  • 517 posts

Posted 17 April 2014 - 10:14 AM

Any VLE textbook should describe algorithms that are used in calculating dew points. If you will explain where you are at in your studies and what this exercise is supposed to teach you, we will be better able to give a specific answer.

 

As I see it, most VLE calculations (in particular gamma-phi type models) can be approached as "corrections" to Raoult's law. Most of the time, when I would build my own spreadsheets for these kind of calculations, I would start by developing the algorithm around Raoult's law. Then, when I had that algorithm working to my satisfaction, it was relatively easy to add the activity and/or fugacity coefficients from there. I have also found that bubble point is much easier to calculate than dew point. Most of my dew point spreadsheets/algorithms start from a bubble point calculation, then incorporate an algorithm to adjust T and/or x to match the given P and y.

 

A quick flow chart for bubble point using Raoult's law:

 

1) Specify T and x (mole fraction)

2) From T, calculate pure component vapor pressure (P0) for each component.

3) From x, calculate partial pressure Pi for each component from Raoult's law: Pi=xi*P0i

4) Total pressure is the sum of the partial pressures: P=sum(Pi)

5) Vapor composition (y mole fraction) is the partial pressure divided by the total pressure yi=Pi/P



#6 M.Basaar

M.Basaar

    Junior Member

  • Members
  • 17 posts

Posted 28 April 2014 - 02:21 AM

Dear MrShorty

 

I am an intern and we are designing a sample conditioning system.

 

I started a spreadsheet with compositions and other parameters. I got stuck while finding the vapour pressure for oxygen and nitroge. for example, most of websites give the vapour pressure at temperatures < -140 degC where my stream is +35 degC. is there any assumption to proceed or any way to find the vapour pressure value at this temperature.

 

Thank you

 

 

A quick flow chart for bubble point using Raoult's law:

 

1) Specify T and x (mole fraction)

2) From T, calculate pure component vapor pressure (P0) for each component.

3) From x, calculate partial pressure Pi for each component from Raoult's law: Pi=xi*P0i

4) Total pressure is the sum of the partial pressures: P=sum(Pi)

5) Vapor composition (y mole fraction) is the partial pressure divided by the total pressure yi=Pi/P



#7 breizh

breizh

    Gold Member

  • Admin
  • 6,292 posts

Posted 28 April 2014 - 06:50 AM

http://www.cheric.or...rop/cmpsrch.php

 

From this data base , you should find the vapor pressure of your components at different temperatures !

Scroll down the page to get access to the data and graph.

 

http://www2.questcon...rmo/dewbub.html

 

 

 

Hope this helps

 

Breizh


Edited by breizh, 28 April 2014 - 07:21 PM.


#8 MrShorty

MrShorty

    Gold Member

  • ChE Plus Subscriber
  • 517 posts

Posted 28 April 2014 - 09:16 AM

I got stuck while finding the vapour pressure for oxygen and nitroge. for example, most of websites give the vapour pressure at temperatures < -140 degC where my stream is +35 degC. is there any assumption to proceed or any way to find the vapour pressure value at this temperature.

 

Probably because 35 C is so far removed from the critical point of these gases that we really cannot calculate anything that could be thought of as "vapor pressure". Historically, VLE calculations (especially at low pressures) involving light gases (like nitrogen) have been performed using Henry's law -- Pi=xi*Hi where Hi is the Henry's constant. Calculations are analogous to the case of Raoult's law. The real challenge could be looking up the Henry's constant in a mixture of p-xylene, acetic acid, and water.

 

If you recognize that, at the dew point, there will be essentially no gases dissolved in the condensate, then we can make the simplifying assumption that we can ignore the amount of dissolved gas in the condensate. I would probably even go so far as to say that, at 1 bar total pressure, the amount of dissolved gas will be negligible -- especially where we are at the stage of learning how to do these kind of simple estimations using very idealized models. I would probably save the "how do I handle the inert gases" part of the calculation until we are closer to the more complex and rigorous models that will come after we have an understanding of the basic algorithms.


Edited by MrShorty, 28 April 2014 - 09:16 AM.


#9 M.Basaar

M.Basaar

    Junior Member

  • Members
  • 17 posts

Posted 28 April 2014 - 08:31 PM

 

If you recognize that, at the dew point, there will be essentially no gases dissolved in the condensate, then we can make the simplifying assumption that we can ignore the amount of dissolved gas in the condensate. I would probably even go so far as to say that, at 1 bar total pressure, the amount of dissolved gas will be negligible -- especially where we are at the stage of learning how to do these kind of simple estimations using very idealized models. I would probably save the "how do I handle the inert gases" part of the calculation until we are closer to the more complex and rigorous models that will come after we have an understanding of the basic algorithms.

 

 

Thanks MrShorty for the explanative reply. I just want to confirm my understanding. Do you mean that I should neglect oxygen and nitrogen while calculating my dew point?

or should I use the maximum vapour pressure value which is at -115 degC while my temperature is 35 degC?


Edited by M.Basaar, 28 April 2014 - 08:36 PM.


#10 xavio

xavio

    Gold Member

  • Members
  • 102 posts

Posted 28 April 2014 - 09:55 PM

Hi M. Basaar,

 

I think your question is not clear.

You haven't answer what the purpose of your calculation is.

 

I see that you have inert gas mixture (nitrogen) with minor amount of hydrocarbons.

The mixture is currently at 1 bar and 35degC, and you want to cool it down to a temperature at which acetic acid is condensed out for the first time.

You want to know the temperature, am I right?

My guessing is, according to your gas composition, the temperature will be well below 0degC.

 

It is analogous to condensing water moisture content from air, or hydrocarbon content from natural gas.

 

First off, paraxylene is heavier than acetic acid, probably the heaviest in the mixture, so it will condense earlier (at higher temp) than acetic acid. Most probably, however, you will not be able to pinpoint the temp which acetic acid begins to condense, as the condensed liquid will be multicomponent mixture. So, what you will get is the temp where first droplet of liquid is observed, i.e. the dew point.

 

To theoretically predict the temperature, you will need to use Equation of State (ex: PR or SRK), which is more conveniently performed by simulator.

I don't think Henry's Law is appicable for this calculation.

What MrShorty suggested is bubble point calculation, but what you need is dew point calculation.

See here to learn the difference:

http://www.mycheme.c...ideal-mixtures/

 

If you have the resources, you may want to do it experimentally.

The instrument is something like this:

 

"Do you mean that I should neglect oxygen and nitrogen while calculating my dew point?"

I'm not sure if it works, but you may try.

Neglect the supercritical gas partial pressures (nitrogen and oxygen), and instead take summation of partial pressures of hydrocarbons as system total pressure.

Iterate the temperature until summation of liquid molar fraction equals to one.

Please update us with your result so that we can comment if it is reasonable.

 

Good luck.

 

xavio



#11 M.Basaar

M.Basaar

    Junior Member

  • Members
  • 17 posts

Posted 29 April 2014 - 03:42 AM

Hi Mr. Xavio,

 

Many thanks for your help.

 

I have mentioned the purpose before. We want to get rid of acids and hydrocarbons to prepare oxygen for analysis.

 

I have been refering to the post you referred me to. I followed the steps and made a spreadsheet. I got a result which seems unreasonable.

I Excluded oxygen and nitrogen from my calculations.

 

I got SUM(Xi)=1 at T=410 using Ptotal=SUM(Pi)=2908mmHg which is calculated using Antoine equation.

I got even higher T when I used Ptotal= pressure given in the process datasheet 1900mmhg.

 

 

 

That is really confusing.

 

 

my file is kind of messy because i tried different approaches. Hoever, You can simply see the columns in yellow highlight for the temperature guessing and equivalent mole fraction summation

 

Attached Files



#12 MrShorty

MrShorty

    Gold Member

  • ChE Plus Subscriber
  • 517 posts

Posted 29 April 2014 - 10:10 AM

I haven't been able to go through your spreadsheet in detail (as you note, it can be kind of hard to follow). First thing I noted in your spreadsheet: you claim your Antoine equation parameters are for log/exponent base 10 (except for water, which looks like base e). If I followed your spreadsheet, you used the exp() function, which is for base e. I may have missed a conversion from base 10 to base e. In any event, I would first make sure you are calculating the pure component vapor pressure correctly.

 

2nd thing I noted is that it almost looks like your expression for Raoult's law is incorrect. It looks like you are using something=y*P0, which does not look like Raoult's law to me. It seems to me that, before we can really worry about finding dew point, we need to first get our expression for Raoult's law correct.

 

Step 1 is to get the vapor pressures correct. Assume a T (100 C might be a good first choice, just so we can make sure we are calculating the vapor pressure of water correctly). It might be worth putting the normal boiling point of each component in to make sure that you can get 760 mmHg for the pure component vapor pressure of each component at its boiling point.

 

Step 2 is to guess a liquid composition (ignoring dissolved N2/O2), and calculate the partial pressures of each component from Raoult's law (Pi=xi*P0i).

 

Step 3 is to calculate the bubble point pressure at our guess for T and x (Pbub=sum(Pi)). the vapor composition at our assumed T and x will then be the partial pressures divided by the bubble point pressure. yi=Pi/Pbub

 

Let's see if we can get that far.



#13 xavio

xavio

    Gold Member

  • Members
  • 102 posts

Posted 01 May 2014 - 09:57 PM

M. Basaar,

 

Your spreadsheet is truly hard to follow.

How is the total pressure is 2.5bar now? Didn't you mention 1 bar?

With the inerts neglected, the total pressure for calculation purpose should be less than 0.005 bar.

MrShorty has given you very good advice, check all inputs and known data before proceeding.

 

By the way, how will you get rid of the hydrocarbons? What tools/equipments do have now?

To what extent you want to reduce the hydrocarbon content?

 

I'm afraid that such a low hydrocarbon content cannot be removed by condensation.

You should consider adsorption.

 

Good luck.

 

xavio



#14 M.Basaar

M.Basaar

    Junior Member

  • Members
  • 17 posts

Posted 02 May 2014 - 04:07 AM

Dear Mr. Shorty & Mr.Xavio

 

Appologies for the untidiness of the file. I was not sure how to do it thus it was messy.

 

The formula Xi=Yi* Ptotal / Pi , I got it from the post you recommended http://www.mycheme.c...ideal-mixtures/

 

I figured out i mistakenly inversed Log10(x) as exp(x) instead of 10^x

 

I am thankful to you and I am really advancing in understanding the thing. My issue is limited access to internet (only at work) yet many jobs to do.

 

and yeah, regarding techniques getting rid of HC, we are using activated carbon and CaCO3 besides cooling.

 

 

I apologize again for that file. I myself gave up on it.

 

 

 

 

 

I would like your consultance. Now when I divide the total pressure by the vapour pressure. What calue of total pressure should I use? the one in my process datasheet or the sum of the individual vapour pressure.??

 

 

I have another problem if your times allows. This time I am trying to make it neat and correct my mistakes. But my result is still unsatisfactory to the rule sum(Xi)=1

 

Check the attached file at your convenience. 

 

Many many thanks to you guys.

 

Best wishes

Attached Files



#15 breizh

breizh

    Gold Member

  • Admin
  • 6,292 posts

Posted 02 May 2014 - 06:19 AM

Basaar,

 

You seem confused by Pt = Ptotal = 1 bar or 760 mmHg = Sum( Pi )   >>> xi =yi*Pt/Pi0

 

review your last  column and then do iterations on temperature till Sum (xi)=1.

 

Hope this helps.

 

Breizh


Edited by breizh, 02 May 2014 - 11:34 PM.


#16 MrShorty

MrShorty

    Gold Member

  • ChE Plus Subscriber
  • 517 posts

Posted 02 May 2014 - 11:27 AM

The formula Xi=Yi* Ptotal / Pi , I got it from the post you recommended
Can you see how this was derived from Raoult's law + Dalton's law? This works well when we are assuming an ideal liquid solution (Raoult's law), but when we get to considering non-idealities -- when we add in the activity coefficients -- you will need to understand where this equation came from so you can figure out how to add the activity coefficients to it.

 

I figured out i mistakenly inversed Log10(x) as exp(x) instead of 10^x
Good clarification, though it looks to me like there is something wrong with the Antoine equation for H2S -- it looks like these coefficients go with a natural log equation and the C parameter is for T in K. There is so little H2S, that it probably has very little effect on the dew point, so it might be easiest, for now, to ignore it.

 

Another thought on vapor pressure. With the range of vapor pressures in this hydrocarbon mixture, I would probably use something other than the Antoine equation. The Antoine equation is a rather limited vapor pressure equation (as you can see from the Tmin and Tmax for each compound). I would probably use an equation (like the Riedel or Wagner) that can represent the vapor pressures up to the critical point. That will likely mean that you will have a good vapor pressures at the dew point for each component (except methane and nitrogen, which will likely still be supercritical). It might also be worth noting that, if the dew point ends up being below 0, there will be some question of what to use for the vapor pressure of water below its freezing point.

 

I would like your consultance. Now when I divide the total pressure by the vapour pressure. What calue of total pressure should I use? the one in my process datasheet or the sum of the individual vapour pressure.??
There are usually four main quantities in any VLE calculation: P, T, x, y. Mathematically, we have to be given at least two of these quantities to calculate the other two. In the case of a dew point calculation, we are given P and y, and we attempt to calculate x and T. I don't see anywhere in the spreadsheet where you list the given P. The sum of the individual pure component vapor pressures does not have much meaning, so the total P should be what is given to you in the data sheet.

 

With these corrections, I calculated a dew point at about -30 C (assuming 1 atm total pressure), and the condensate is mostly water with significant hydrocarbons. The condensate will almost certainly be two phases (a water/ice phase and a hydrocarbon phase).



#17 xavio

xavio

    Gold Member

  • Members
  • 102 posts

Posted 05 May 2014 - 01:36 AM

Hello Basaar,

 

I think you should make yourself clear about: total pressure, partial pressure, saturated vapor pressure.

Your calculation breaks down because you don't understand the definition of those pressures.

 

Back to your project, why bother to calculate the theoretical condensation temp if you have the adsorption beds?

Just cool down to the lowest temp possible and proceed with the adsoprtion for final purification.

You then analyze the outcome whether it is acceptable or not.

 

Good luck.

 

xavio



#18 M.Basaar

M.Basaar

    Junior Member

  • Members
  • 17 posts

Posted 06 May 2014 - 03:11 AM

Basaar,

 

You seem confused by Pt = Ptotal = 1 bar or 760 mmHg = Sum( Pi )   >>> xi =yi*Pt/Pi0

 

review your last  column and then do iterations on temperature till Sum (xi)=1.

 

Hope this helps.

 

Breizh

 

________________________

________________________

 

 

 

I figured out i mistakenly inversed Log10(x) as exp(x) instead of 10^x
Good clarification, though it looks to me like there is something wrong with the Antoine equation for H2S -- it looks like these coefficients go with a natural log equation and the C parameter is for T in K. There is so little H2S, that it probably has very little effect on the dew point, so it might be easiest, for now, to ignore it.

 

I would like your consultance. Now when I divide the total pressure by the vapour pressure. What calue of total pressure should I use? the one in my process datasheet or the sum of the individual vapour pressure.??
There are usually four main quantities in any VLE calculation: P, T, x, y. Mathematically, we have to be given at least two of these quantities to calculate the other two. In the case of a dew point calculation, we are given P and y, and we attempt to calculate x and T. I don't see anywhere in the spreadsheet where you list the given P. The sum of the individual pure component vapor pressures does not have much meaning, so the total P should be what is given to you in the data sheet.

 

With these corrections, I calculated a dew point at about -30 C (assuming 1 atm total pressure), and the condensate is mostly water with significant hydrocarbons. The condensate will almost certainly be two phases (a water/ice phase and a hydrocarbon phase).

 

 

Mr. Breizh & Mr. Shorty,

 

This is very enlightening. Thank you so much.

 

I am really glad with your patience and assistance and I finally got an answer for my second problem. I will redo my first problem again.

 

for my second problem , the pressure is 75 bar.g.= 56254.6262 mmHg and I managed to get dew point of 58 degC. Do you think it is logical answer considering my high pressure?

 

 

Appreciate your effort :)



#19 MrShorty

MrShorty

    Gold Member

  • ChE Plus Subscriber
  • 517 posts

Posted 06 May 2014 - 10:56 AM

That does seem reasonable -- considering all of the simplifying assumptions we have made (vapor is an ideal gas, liquid is an ideal solution, hypothetical pure component vapor pressures for the supercritical components are extrapolated from our favorite vapor pressure equation). At 75 barg (87 bara), our assumption of an ideal gas is almost certainly wrong. We know that our condensate will have both water and hydrocarbons, so our assumption of an ideal liquid solution is also almost certainly wrong. Extrapolating vapor pressure equations may or may not be good, especially for something as far removed from its critical point as methane.

 

At this point, you can decide to be satisfied with that solution, or we can decide how we want to "correct" for the non-idealities that almost certainly exist in a system like this.






Similar Topics