Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Production of Acetone process

  1. Mar 11, 2015 #1


    User Avatar
    Gold Member


    I am working on a process design project and was wondering if I could get some help. While this is classwork material, I thought this is too big of a problem to post in the homework section, and would be more suitable in a forum such as this one which has much less traffic. For reference, here is the project description


    So with that said, we have 5 heuristic steps to dealing with planning a process

    level 1: Batch vs. continuous
    level 2: input-output structure
    level 3: Recycle structure
    level 4: Separation structure
    level 5: Preliminary process flow sheet

    1. So the production here is 100,000 tonnes/year, so it is clear this must be a continuous process (rule of thumb is > 1 million lbs/year should be continuous, we far exceed that amount)

    To be continued

    Attached Files:

  2. jcsd
  3. Mar 11, 2015 #2


    User Avatar
    Gold Member

    Here are some useful azeotropic information

    And my lovely onion diagram for process design
    Last edited: Mar 11, 2015
  4. Mar 11, 2015 #3


    User Avatar
    Gold Member

    This is a very rudimentary system I have come up with on paper so far. My thoughts are that of course I will need to react and get my products and by-products. I thought after the reactor it would be fairly easy to separate the hydrogen and propylene due to the large difference in boiling point, so I ran it at the boiling point of acetone to bring that as well. Since it will form a azeotrope with water, I know it won't be pure. After that I use another separator to easily purge off the gases.

    From the bottom stream of the first separator, I want to recover and recycle my isopropyl alcohol (IP) because it is quite costly ($1000/tonne), so I need another separator. However it too forms a couple azeotropes, so I know I can't send it back to the reactor pure (at least for now). Any hints for how to deal with azeotropes?

    ImageUploadedByPhysics Forums1426091244.474127.jpg
    Last edited: Mar 11, 2015
  5. Mar 15, 2015 #4


    User Avatar
    Gold Member


    Here is what myself and my partner have come up with so far. So basically, in this iteration we know hydrogen and propylene are light non-keys, and everything else is a heavy non-key, so they are most easily separated. Just running the flash drum at ambient temperature to minimize heating costs.

    Now for the second separator, I want to separate the acetone (AC) from the diisopropyl ether (DE), and ultimately my goal is to recycle the isopropyl alcohol (IP) back into the reactor while purifying the acetone at 99 wt%. So I run this separator at 57 C, as at this temperature Acetone and water form an azeotrope. I figure the acetone and water will exit the top stream at its azeotropic composition, and DE is a heavy key, so there should be some in the distillate as well. The bottom stream has IP as a heavy non-key, and heavy keys of water and DE.

    The next cascade of columns aims to purify the acetone and recycle the IP. So the bottom column separates the IP and AC, of which those are purged. Hopefully the AC is very low concentration in this stream. The top column (hopefully) will have very high AC concentration.

    Obviously, there is a lot of information needed to be added, such as the fraction of each component in each stream, for which I need to use the T-xy diagram to determine. However, I am not experienced in azeotropic distillation, so this is proving to be a tough problem. Does there exist azeotropes of 3 or more components? Or do I just treat each one as a two component distillation. Designing a process has so many complications that it makes it hard to keep everything straight.

    Attached Files:

    Last edited: Mar 15, 2015
  6. Mar 29, 2015 #5


    User Avatar
    Gold Member

    I'm working on a column in my design. I have managed to purge out most all of the hydrogen, and I have some propylene. Now, with this column I want to separate my acetone into a second column through the top, but for some reason I have a lot of difficulty getting this column to do ANY separation. It seems to just all go straight to the bottom, and the temperature of the top stream is less than -200 C, which makes no sense.


    Here is the stream profile

    Here are my column specifications

    How the heck can I get this acetone to flow up? This is my error message. It seems that some stages are dry

    It is difficult since I cannot upload my Aspen file onto PF, but if anyone has Aspen Plus I can certainly send my file
    Last edited: Mar 29, 2015
  7. May 3, 2015 #6
    Hi Maylis.

    Dealing with azeotropes can be pretty challenging, but I have four ideas (in rough order of priority).

    The first is to have the reactor operate such that greater than 75% acetone is present in the outlet stream. This way, you can run the column/flash drum at 10 atm, and perform the separation from 75%->99% past the azeotrope. Try exploring pressures at above 10 atm also - the location of the azeotrope might go below 75% acetone.

    The second is to consider running the column as a vacuum. There is an azeotrope at 92% acetone at 1 atm, but pressures below 1 atm might push the azeotrope to a higher percent (it might even go away). However, from a design standpoint, having a vacuum column is going to increase it's cost, so I would avoid vacuum columns if possible.

    The third is pressure swing distillation. This would involve operating a first column at 1 atm, and pushing amount of acetone in the outlet stream past 75%. Then, that outlet stream can be fed into a second column at 10 atm. The second column won't have to deal with the second azeotrope, since the percent acetone is past 75%! This would likely be more expensive than either of the preceding options, since two columns would be needed.

    The fourth (and what your professor might not like), is to incorporate some method other than distillation. Other methods would include using membranes or some absorbent material to scavenge the water. I wouldn't design my plant around it, but you could perhaps put it in your report/memo as an option which was considered, but not pursued.

    For your column, under operating specifications, I noticed you have "distillate to feed ratio" selected with a value of 0.01. That is pretty much why everything is going into the bottoms. You are basically forcing Aspen to put only 1% of the feed into the distillate. This is also probably why you have weird **** happening - everything might be turning into a vapor, and the heat of vaporization is lowering the temperature significantly. Actually, I don't even think that is thermodynamically possible. Then again, acetone is quite volatile, so maybe. That's one of the great things about Aspen, thermodynamically impossible stuff is allowed to happen in favor or attempting to satisfy a constraint (sarcasm). I feel your pain.

    But yeah, get rid of the distillate to feed ratio constraint. Instead maybe specify 99% acetone in the distillate.

    Oh and one more thing. Regarding the tradeoffs for running the reactor - increasing temperature leads to less undesired DE, but more undesired PY. In fact, you get more undesired PY than DE, so this would be a negative tradeoff. The benefit is an increased overall reaction rate. This is something I would optimize for AC production. Definitely have a recycle stream going into the reactor, because that will allow you to operate at a lower temperature. I also wouldn't worry about purifying the feed before it goes into the reactor. Although, you could have a separator before the reactor, where everything gets separated. You could try both having all of the reactions before and after separation, but if you don't have infinite time (which you don't) maybe just put the separators after the reactor. Also you need a purge stream to get rid of the DE and PE. They will not leave otherwise.

    Hope this helps, and I am not too late!
  8. May 3, 2015 #7


    User Avatar
    Gold Member

    Well it is a little late, but I will check this out during the summer just for my own enrichment. I can email you my final aspen file to see what we ultimately ended up doing if you are interested.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook