4 replies to this topic

[gabo]

Hello,
I have been designing a liquid adsorption system, in which an organic solvent is going to be dryied. The content of water is about 1 to 3 percent weight. The feed is 150 cubic meters per hour, and the selected adsorbent is silica gel. The system will be 2 column, with bed height of 3 meters and diameter of 1.5 meters and one column will be working for about 3 hours while the other is been regenerated with air or nitrogen at a higher temperature than ambient.
I would like to know if somebody can give me some suggestion about the system, rule of thumbs about this kind of systems, and i would like to know if this system dimensions are normal... . I didn´t find any information about industrial adsorption systems.

Thanks,

Gabo

[Art Montemayor]

Gabo:

I've done this one - for both gas drying and liquid drying - so maybe some of my comments might help you.

I don't know just what specifically you are doing because you haven't given details such as what your assignment is, is the design a process one, is it a mechanical design, is it an academic exercise, is it an actual industrial application, how you have decided on your adsorbent, how you've derived your bed dimensions, where you are seeking information on adsorption, why (if you haven't done this before) you are encharged with designing such a unit, what are the fluids involved: process and regeneration, etc., etc. So you see, you are really being very general and yet are expecting specific answers. This is not going to to work according to your rules. If you want specific information to help or assist you, you must furnish specific basic data - as mentioned.

There are many thing you can do -or should be doing - but since it seems you've already done a process design, all that is left is the mechanical part. Since we don't know the regeneration process or method, all that is mentioned will be conjecture:

1. A 6-hr NEMA cycle (3 hr drying + 3 hr regenerating) seems too short and out of the ordinary. An 8-hr cycle is normally the shortest cycle employed in adsorption and its basis is that it involves one man-shift in a 24 hr processing environment. But this is dependent on a lot of other factors: the fluids involved, the process upstream, the process downstream, the schedule, the process conditions, the contingencies required, etc.

2. The manufacturer of the adsorbent should play a major role in determining the actual design sorptive capacity for the adsorbent used (silica gel, in this application). If not, you are working in the dark or relying on your own research and decisions - something that is normally not smart to do since the user of the adsorbent is not usually an expert in how it is made and what its sorptive capacities and life expectancies are.

3. The drying process fluid direction through the bed should always be downward. This avoids any bed movement and attrition with subsequent dusting problems. This is particularly true of Silica Gel, more than any other adsorbent. Activated Alumina is much tougher and durable in this application. But you selected the adsorbent, so you have to live with it. Bed restraints and hold-downs are critical.

4. The vessels should be equipped with adequate draining, venting and purging nozzles in order to regenerate as quickly and efficiently as possible. However, you haven't stated how and with what you are regenerating, so that's all I can comment on.

5. Your L/D =3/1.5 =2 seems to be too small. I usually employ an L/D that ensures sufficient residence time, superficial velocity, and accessability to ensure no channeling takes place and that I can arrest a break-through. I don't know your process design basis, so that's all I can say.

You are wrong about there being a lack of information on industrial adsorption systems. There are text books as well. In my opinion you probably haven't looked enough or in the right places. How is it you were able to come up with a process design? Surely you have some information. As I said, the manufacturers are the usual source for the critical process design information.

I hope these comments help out.

[gabo]

Dear Art,

Sorry about my short writting, here goes my answer, and what i am looking for, answering your questions and comments:

I don't know just what specifically you are doing:
- design of a process, for academic exercise, and is part of an hydrogen peroxide production by anthraquinone method, in which the oxidized working solution is dryied to be reused.
how you have decided on your adsorbent:
from literature since silica gel has high sorptive capacity, large specific surface area, high affinity for water, regenerable, is non toxic, low regeneration temperatures and cheap, etc. Chemical Engineering/Nov 1995, Thermal Engineering, Vol. 4 • No. 1 • June 2005 • p. 3-7.
how you've derived your bed dimensions:
- I am making a preliminary design, and took dimensions assuming abrupt transition, since silica gel has favorable isotherm and I compared it with an example in “Department of the Army - Army Corps of Engineers - DG 1110-1-2, U.S.”, where is designed a liquid phase adsorber which dimension where diameter: 2.4 meters and bed depth: 2.3 meters.
- Using my dimensions the breakthrough curve should be at 3.5 hours, so 3 hours I assumed as cycle time.
where you are seeking information on adsorption:
- Perry’s Chemical Engineering Handbook, Chemical Engineering and Thermal Engineering papers, Coulson and Richardson Chemical Engineering, etc.
why (if you haven't done this before) you are encharged with designing such a unit:
- exercise
what are the fluids involved:
- mixture of xilene and methylciclohexanol 1:1, with 1 to 3 weight percent of water.
process and regeneration:
- Thermal swing adsorption, with heated nitrogen or air regeneration. “Chemical Engineering/Apr 1999.”
So you see, you are really being very general and yet are expecting specific answers:
- I just want somebody who has experience with adsorption give me some thumb rules, and/or an idea about an industrial liquid adsorption unit to compare, nothing else.
This is not going to to work according to your rules:
- I don’t expect nothing work according to my rules in this forum.
Since we don't know the regeneration process or method:
- Heating of the desiccant in the dryer is accomplished with internal heaters (electric or steam or thermic fluid) in conjunction with the use of a dry non-reactive purge gas (like Nitrogen)

You are wrong about there being a lack of information on industrial adsorption systems:
- Sorry, I meant that there is a lack of information on dimensions used in industry for liquid adsorption columns, so I can compare.

I hope these comments help out:
- Yes, and thanks, your comments helped, but still didn’t answer my question of usual superficial velocities, or loading rates, and dimensions or L/D.

Again, thanks :-)

Hi All,

Since i am not a process engineer, I apologize in advance for the questions if they are stupid?

What is purpose of regenerating one column with air or nitrogen? Is it to remove the water collected? Can steam be used?

I am a jr engineer and was in a recent meeting with a vendor....Silica (They were germans...they talked about silica gel). I was just there to take minutes of meeting since I am new to job. I was trying to follow what all the senior guys/vendor reps were talking about but since I have no background in the area it was mostly greek to me.

Can anyone recommended any books to read about adsorption columns? From what I gather it seems like it has many aspects to it.....process, mechanical, I and C. The process was combustible and the vendor was talking about provisions to ensure no explosions would take place (like having provisions for oxygen measurement). It was very interesting to be in the meeting and thats why I would like to read more about it.

I know sizing the column and so on is more propertiary in nature but what about the mechanical aspects (Since I am a Mechanical EIT). Art touched on a few things, like venting and nozzle sizes. Any papers / books that would help me be more familar with the equipment and process.

I would ask the Mechanical Engineer assigned to the equipment but he doesnt seem to be very familar/competenent in this area.

thanks in advance for any replies..............

Kellie

[gabo]

Dear kellie,

Yes, the purpose is to remove the water that was adsorbed in the silica bed (or any other adsorbent). What i design was a adsorption column for drying a liquid organic solvent, so when the bed is about to be all saturated, we change the column for one fresh, and the one that is saturated, is has to be regenerated. In this case, vapor is not the most suitable, since is water vapor, and you want to remove water. So, another gas is used, like hot air, or nitrogen. In this way you can make the water adsorbed to be released by meaning of the high temperature and the adsorption of the water in the gas.

You can read something in Perry's Chemical Engineering Handbook, or search something in the Chemical Engineering Magazine, where you can find a lot of good information. There are a lot of good books where you can find info, and there are some that are specialized in adsorption, but i don´t have acces to them and thats why i came to this forum, but they didn't help me too much.

Well, I posted a reply for Art and he didn't answer... :-( I wrote the papers / books that I have been using for the column design, look up in my previous reply.
About the mechanical design, is the same as any packed column design.

If you want i can send you some info i founded.

Bye