Molecular Sieve Dehydration of Ethanol 1

[beadle]

I was wondering if anyone could help me with my design of the mentioned sieves. Having sized and designed two sieves in parallel for a 100,000te p.a. ethanol plant, and with no small scale lab to model their operation, I need to know for sure if they will dry the ethanol sufficiently. Input is 95%(by mass) ethanol, output is 99.7%(by mass) ethanol.
Conditions are 150oC and 2.7bar, and operating cycle lasts 10 mins. Are there any ways of checking? The isotherms I have modelled this on are minimalistic.
Regeneration is with CO2 at 200oC and 2.2bar.
How should I scrub excess CO2 from the bed after regeneration before putting it back on line?
Also what material is used to build these things? I presume some kind of stainless steel?

 

[TD2K]

Your contact temperature seems high (150C). Molecular sieves typically absorb water at much lower temperatures as the isotherms are quite temperature dependent. Granted, you don't need super dry ethanol. We dried liquid hydrocarbons very successfully with mol sieves but were removing ppm of water from the incoming liquid, not wt% of water.

how are you going to handle a 10 minute switching time? Typically for the liquid filled driers I've seen, the off-line drier is drained, you then are going to use a sidestream of a dry (assume) C02 gas to regenerate it (typically, after you reach the target temperature, the bed is held at that temperature for some period of time), the bed is cooled down, likely with the same C02 stream and then refilled with liquid to be switched back into service. Seems tight for a 10 minute cycle just to heat the mass of the bed up to 200C, much less all the other steps.

Molecular seive vessels are typically carbon steel. In your case with wet C02 from the regeneration, corrosion might be an issue. You might want to look at SS cladding but I'm not much of a metalurgist, perhaps someone else will comment on this.

Some C02 will likely adsorb on the molecular sieve during regeneration but I don't have the data to estimate how much. Is CO2 in your ethanol product stream a problem?

Have you talked to the molecular sieve vendors? They typically provide engineering support or at least would be able to provide you with some references.

Sorry, I know this is more questions than answers for you.

 

[beadle]

Thank you for the thoughts.
I have the inlet temperature at 150oC to maintain vapour phase. Sorry, I should have mentioned the adsorption is to be carried out in the vapour phase. The Heat of adsorption will take the bed temperature up to 200oC while drying, ready for regeneration. A knock-out drum is going to separate the water and small amount of ethanol from the CO2 in the regen cycle.
I was still worried about turn-around time. Obviously I can't simply switch from adsorbing to regen instantly.
Do you know any names of zeolite vendors?

 

[huthmyerjohn]

Dear Beadle
I am currently working on a similar problem in Sunbury (BP). I have designed it to reduce the concentration of of water in the entrainment channel from the KG triple x plant.
I also have an invested intrest in the continual growth of the molecular sieve in the fine separations industry. For this reason I would be greatly interested in any work you have done and would be more than happy to assist you further in any way.
Please phone me on 07879442408 and I will phone you back (its a company mobile and so will cost me nothing)
Yours
Dr Hans Fran Anderson

 

[TD2K]

The two major molecular sieve vendors that I'm familar with are UOP and Grace. I'm not sure if Grace is still in the business, my information is a few years old. Years and years ago, we also looked at a molecular sieve from Japan, Toya Soda if I remember correctly, nice product but too expensive with the exchange rate at that time. There's also one vendor out of Switzerland whose name escapes me. I'm sure this isn't all the manufacturers out there.

Must be drinking too much ethanol in the liquid phase to have assumed that this was a liquid adsorption process you were looking at (a rather hot liquid adsorption, that should have triggered a question in my mind and a quick look at vapor pressures).

I'll be surprised if this is going to work as conventional molecular sieve adsorption/regeneration process if you are going to be adsorping water with bed temperatures going from 150C to 200C and regenerating, even if the CO2 stream is bone dry, at 200C. Our molecular sieve beds were regenerated at about this temperature but the adsorption temperature was about 15C and you needed that difference in temperatures to drive the isotherms to where you wanted them. Granted, we were trying to dry cracked gas down to the ppmv range for ethylene recovery but your operating conditions don't seem like a good match.

You might want to ask about pressure swing adsorption where the adsorption/regeneration is driven by more by pressure changes and not so much by temperature but I'm not sure your pressure during adsorption is enough to make this work either and you'll almost certainly need a lower pressure regeneration stream. The 'good' news is that at these temperatures, CO2 adsorption during regeneration is also going to be low though I don't know how low.

Why did you pick this technology and what others did you consider? Membranes come to mind though I don't know what the diffusion of water/ethanol is like and your temperatures seem too high for most membranes I've read about (though I don't profess to be very knowledgeable on them).

 

[pioneert]

Hello friends,

I am working on a project for my University for ethanol dehydration by Molecular sieves Technology. I have studied a lot of litratures regarding the same, but i haave not found any practical datas of operation of Molecular sieves Reactor. Also, if i am having 92% (wt.%) ethanol and rest 8% water, what quantity of molecular sieves is required. What will be the operating parameters for the same?
I am also not convinced by the data i have recieved from one of the vendors of Molecular sieves that their loading capacity is 0.7-2%. That means the Quantity of molecular sieves required will be much more. Also, what is the estimated time cycle of the adsorption and desorption. please help me with this problem.

 

[wms]

Ethanol industry sop is vapor phase w/ pressure swing. Also regeneration via etoh vapor.

 

[siddharth1]

I have just graduated as a chemical engineer and now i am working for a small firm. They have asked me to design Molecular Sieve Ethanol Dehydrator. I know the technology, but don't have a clue to design it. Can any one gide me or give some referances so I can get on with my work.
Thanks for your help.
Sid.

 

[siddharth1]

I have just graduated as a chemical engineer and now i am working for a small firm. They have asked me to design Molecular Sieve Ethanol Dehydrator. I know the technology, but don't have a clue to design it. Can any one gide me or give some referances so I can get on with my work.
Thanks for your help.

Sid.

 

[MarkkraM]

Good old university design project ... ethanol drying is one of the most common tasks given. Due to the azeotrope formed by ethonal and water, conventional distillation can only go so far (about 95% purity). How do you then take things further????

The lure of the molecular sive may seem to good to be true until you realise how much of the stuff you will need to process 100,000tpa of 95% ethanol i.e. removing 5,000 tpa of water! Typically molecular sieve processes are reserved for removing ppm of water (One of the beds at my work is designed for removing 50ppm of water from hexene at an average rate of about 800kg/hr. Furthermore, it takes about 12hrs minimum to regeneration this bed. I'm guessing with the size you will need this could take longer. Why do you need to be in the vapour phase? Universal Oil Products are a good contact for these beds if you're still interested.

Membranes technology is somthing worth looking into. Also, isn't there an azeotropic distillation technique that can be modelled on ASPEN or similar packages? I shall go through my old uni notes soon and see how my group tackled this common problem.

Regards,
Mark

 

[DeltaCascade]

This is all well established technology, particularly with all the relatively small scale fuel alcohol plants out there (N America, brazil, etc).... about 10 million usg ethanol/yr capacity or greater, i believe.

Competing processes are azeotropic distillation and pervaporation membranes. Azeotropic distillation uses more energy, but at lower capital cost. We used this process in a fuel ethanol plant where i worked 15 years ago. In this process rectifying column overheads (96 vol%) are sent to a dehydration column using azeotropic distillation. This dehy column simply adds, say, cyclohexane, to break the azeotrope ... pure ethanol (>99.6%) comes out the bottom of the dehy column, ethanol-water-cyclohexane comes out the top, is decanted and ethanol-water returned to rectifying column and cyclohexane back to dehydration column. Ethanol product typically >99%. Perhaps better than mol sieves in this regard.

Yup, sounds like a school assignment. Hit the library!
Hopefully, someone knows of some good web resources for chemical engineering design details on these technologies.

Good luck.

 

[Velterop]

For pervaporation membranes that can handle temperatures up to 250 degrees C, I refer to the web page:

http://www.pervatech.nl.

They have ceramic pervaporation membranes suitable for pervaporation or vapor permeation. Dehydration of ethanol at these temperatures is not a problem for these type of membranes.

 

[DesertKid]

Other suppliers of mol sieves are Zeochem and AtoFina.
If you do a Google search on the Internet there are some useful sites as well.

 

[pkv61]

Liquid phase drying of alcohal will be good option and esy to perform, since requirement is only 97.5% ethanol i feel liquid phase will be easy to perform usiing 3A Molecular sieve.

 

[bexcelant]

Sepcor, Inc also sells molecular sieves.

http://www.sepcor.com