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


Cryogenic Air Separation To Produce Argon, Oxygen, Nitrogen

cryogenic air seperation argon oxygen nitrogen distillation sieves adsorption column

3 replies to this topic
Share this topic:
| More

#1 chemwim


    Brand New Member

  • Members
  • 3 posts

Posted 17 April 2018 - 09:22 AM



I am currently in the process of designing an air separation plant for the production of argon, oxygen and nitrogen by cryogenic distillation. I have up to this point managed to design the plant using approximate methods and this has allowed me to create an Aspen simulation using the equilibrium model. The process is as follows:


Air Feed = 100kg/s 




Adsorption Columns (molecular sieves) for removal of impurities

Air Booster Compressor

Heat Exchanger Network to achieve cryogenic state

HP Column

LP Column

Crude Argon Column

Pure Argon Column


Purpose of Plant: Obtain Argon with 99.9996% purity, Oxygen with 99.995% purity, Nitrogen with 99.999% purity from a 100kg/s air feed.


Then, I did some optimization on Aspen by performing sensitivity analysis tests and manual optimization and obtained some results. I have attached an Excel sheet with the specs of the columns and streams up to this point.


Currently, all the columns are designed to be trayed columns, the HP and LP Column have not yet been joined to form a double column as I wanted to start basic and work towards a more complex design. As you can see, the reflux ratios are considerably large as are the tray numbers which in turn makes the column heights unreasonable. The purities have not been reached, due to the approximate methods used at this point but are getting there.


My next steps are to make the LP and crude column a packed column, introduce the reflux stream into the LP Column and begin a Rate-Based Aspen model. The scale at which this process can be improved is huge and the purpose of my project is to design a viable plant, to a third-year student level.


My questions:

1) Which impurities should the design of the molecular sieves focus on and how do you account for competitive adsorption when calculating the adsorbent required?

2) What is the purpose of this recycle stream from the Crude Column into the LP Column? (better conversion?)

3) How else can I reduce the reflux ratios for the HP, LP, Crude and Pure Argon Column? What should I be looking at to vary? What is a sensible reflux ratio for cryogenic distillation columns (I know standard is under 2, is this applicable here as well)?  

4) What shall I be looking at to improve the purities of the products, towards their pure grade equivalent?

5) I know the crude argon column and LP column should use packing, what type of packing is typically used, and is there any sources for the design procedure associated with these?

6) For the HP and Pure column, what are the typical tray types being used? single pass trays, double pass trays, sieve trays etc.

7) I read that there is a way to improve argon purity by running the crude argon through a catalytic combustion reactor and reacting it with hydrogen to remove the oxygen but is the flow rate that this process operating at meet the criteria? How is the packed crude argon column capable of avoiding this?

8) I might attempt this, although seems a little complicated but how do you go about combining the HP and LP column?


Thank you to anyone for your time and the answer to any of these questions would be of great help. Any tips on how to build on the basic design would be greatly appreciated!


Attached Files

#2 Art Montemayor

Art Montemayor

    Gold Member

  • Admin
  • 5,774 posts

Posted 17 April 2018 - 02:59 PM



What you are “designing” is a very large air separation plant.  What have you calculated or estimated as your storage size requirements?

On preliminary review of your very basic data, some observations and comments are:

  • Why are you using six decimal places of accuracy in your air feed flow rate and only 3 places for your individual component flows?  Which, if any, are credible accuracy figures?  Or are you just being lazy and not paying attention to your data?
  • Your adsorption system is going to be huge.  Have you estimated the approximate sizing?  This may affect the type of pre-purification steps you may have to take.  For example, you may have to apply refrigeration to the feed air to reduce the water load on the adsorbent and reduce the bed size.
  • I don’t see any reflux calculations or flow rates - or even the number of trays in each of the proposed columns.  Did you forget to include them?

Answers or comments to your questions are:

  1. You don’t furnish a detailed flow diagram, so we can’t know what your process involves or looks like.  I don’t know what you mean by “competitive” adsorption.  You design the adsorption system in accordance with the type and size of system that you select.  The regeneration portion may affect how you design it.  The adsorbent selected will also have a bearing as will the regeneration temperatures.
  2. Again, no flow diagram, so I don’t know which or what crude or LP column you refer to.
  3. Without your calculations for reflux ratios and the methodology or algorithm employed, I don’t know how you can reduce the reflux ratio.  How do you know the “standard” is under 2??
  4. When you are designing the process, there is no “improving” the purity of the product.  You design for that purity and if you don’t get it, then your design is bad.  I don’t understand your thinking or logic here.
  5. In order to give us a degree of confidence, how do you “know” that the crude argon column uses packing?  Any packing employed is subject to availability, compatibility, and design characteristics that comply with the design needs.
  6. which columns are you referring to?  Please, once again, BE SPECIFIC in describing your process.  We are engineers; we are not trained to guess, but to be correct!
  7. What you describe is the old argon process; you may be able to improve argon purity, but that depends on what you are starting with.  Refer to the attached information I submit.
  8. Again, BE SPECIFIC in identifying what you are referring to.....  which Columns???

I am submitting your workbook as Revision1.  If you are going to continue with this thread and submit additional information and calculations, use this same workbook with subsequent Revision numbers in order to have some logical, engineering organization in the exchange of information and data.


I hope these comments help you get organized in your data and communications.

Attached Files

#3 chemwim


    Brand New Member

  • Members
  • 3 posts

Posted 19 April 2018 - 05:49 AM

Dear Ant,


Thank you very much for your quick response. I was expecting a longer response time and currently in a very busy period with final exams over the next two weeks.


I have taken your points on board and will return with a more detailed description of the current design in two weeks as I would prefer to be as detailed and specific as I can be and not rush the reply. I have most of what you are asking for but will need to summarise it for it to be clear.


Thank you again and speak to you soon.

#4 ashry3x


    Brand New Member

  • Members
  • 1 posts

Posted 23 April 2018 - 01:16 PM

I work on a science project & I need this circuit to work in dynamic mode 

can any one help me 

Similar Topics