8 replies to this topic

[Art Montemayor]

Noora:

See my response on the Industrial Forum

Art Montemayor

[Art Montemayor]

Noora:

Thanks for responding with the additional explanation. Now the picture takes on more clarity and understanding.

I designed an analogous system in 1971 that was very similar to what you are studying. Perhaps it can help to guide you.

My problem was to vaporize an equilibrium quantity of Furfural liquid into a stream of hot Hydrogen and subsequently feed the resultant vapor mixture to a fixed bed reactor where I produced Furfuryl Alcohol. Prior to this, the operation was being done in a small, packed tower. What I designed and built was similar, but bigger, and is still operating as 2 production trains in Geel, Belgium. I don't know if it has been up-graded or improved, but it worked then and continued to do so for many years. I understand it is still functioning as a showcase plant, 33 years later. It's a source of pride and nostalgia for me.

The problems I found were:
1) The operation is basically one of simultaneous heat and mass transfer; it is very similar to what happens in a cooling water tower; however, in this case, there are little or no mass transfer rates available and the operation had to be piloted - in other words, you need empirical and experienced field data.
2) Although the operation is one of simplicity and involves a minimum of hardware, the results can be very complex and difficult to control and operate. Let me explain further, because this is important: you must use or employ some expanded surface media - such as packing (or trays) - in order to facilitate the intimate contact between the two phases. Countercurrent operation is more efficient, but you may have to use parallel current if you have a heat sensitivity issue with one or both of your phases. The time required for contact is controlled by the mass transfer coeficient and not the heat transfer. Therefore, the diameter and height of the "vaporizing" tower is what you have to identify.
3) The operation becomes complex because of the nature of the fluids: if one of the fluids is heat sensitive and can polymerize (as Furfural did, in my case) you have a serious problem. Additionally, you might form some side reactions. What this translates to is that you have a critical decision on the type of tower internals that you should use. Any impurities in the liquid will precipitate or clog and plug your packing or trays. Maintenance will become a nightmare and operation will be very difficult to control. You must prepare for this or design around it - otherwise, your design will be un-economical and fail.
4) What is going to happen in your proposal is that you will soon start to concentrate impurities in your re-circulated liquid and these will reach a point where they will precipitate and plug up your system - especially any atomizing nozzles that you employ. The worst event is when the packing plugs because you have to stop and evacuate all the packing and clean it. Trays are similar and probably harder to maintain. But both will plug inevitably if allowed to, especially if a process design is not incorporated to keep the liquid fluid clean and homogeneous.

Basically what you are doing is "humidifying" your Hydrogen stream with the liquid - so the operation is one of simultaneous heat & mass transfer. Your text book on this and on the design of cooling water towers will cover the subject. You should design the system under this scope and not have to resort to classical heat exchanger technology - except for the pre-heating of the liquid and Hydrogen.

If you have ever operated or seen the operation of a simple water humidifier, you can appreciate what is happening. A person can never imagine the amount and quality of impurities that abound in water until they see the results of a clogged-up water humidifier that has been fed water with dissolved solids. The same effect will ultimately take place with an industrial humidifier - or "vaporizer" as you call it.

That's about all I can comment on, since you obviously haven't given us all the detailed basic data. But I hope this helps.

Art Montemayor
Spring, TX

[Art Montemayor]

Noora:

I'm familiar with the production of THF. I've produced it from Furfural; for many years it was the only route to obtain NYLON, using the DuPont method.

Your main concern should be the employment of atomizing nozzles. I know, from experience, exactly what you're trying to design around. And I believe it to be ill-advised from a practical point of view and field experience.

You are working on the basis that if you atomize the liquid paricles small enough, you will get almost instantaneous vaporization into the hot, circulating Hydrogen stream. This is theoretically possible. But on a more practical and industrial basis, this is not a workable -nor reliable- method or process due to what I've previously pointed out in my prior posting.

The way this is done industrially is to use a packed tower, allowing the liquid to wet the packing and humidify the circulating hot Hydrogen. Use two towers, one always on standby to replace the other one when it becomes clogged. Believe me, this is the only reliable way. The nozzle idea will quickly become your worst nightmare come true! Nothing plugs up or becomes inoperable faster than a so-called engineered nozzle - especially on industrial fluids.

I agree, there are industrial and successful "spray dryers" or similar apparati out in industry working right now. But if you look carefully, they employ very, very special controls and clean fluids to ensure that the nozzles are kept clean and operable. A lot of this technology is captive and arcane. Unless you possess this kind of proven technology, I wouldn't design around an atomizing nozzle to produce industrial chemicals on a continuous basis.

If you resort to a packed tower, you don't have to worry about critical liquid particle sizes, and other complex parameters. You treat the Unit Operation as a humidifier - just as I've previously explained. You may have to employ empirical values - certainly, the Souders-Brown relationship (or something similar) will come into play at the top section of such a contact tower. I would keep the Hydrogen superficial velocities conservatively low to eliminate physical entrainment and allow the Souders-Brown equation to take effect - remember, all it relies on is the liquid and vapor densities - not particle size! Whatever the particle size turns out to be, will be of nil effect and you can mechanically trap out any liquid droplets that conglomerate in the vapor outlet line and return the liquid back to the vaporizer tower base.

Remember to differentiate between physical liquid particle sizes and the amount of prior liquid that exists as subsequent vapor mixed with the Hydrogen! There is a big difference between the two. Try to keep as much of the physical liquid droplets out of the overhead vapor stream and you will be OK. The downstream, hot reactor will vaporize any small quantity of liquid droplets that find their way there. This is the way it works industrially.

I realize that the packed tower vaporizer unit operation is full of unknown Kga and Kla values and the design method is complex. But the adiabatic spray nozzle method has even more unknowns and is less studied. Both are subject to a lot of empirical and pilot plant data. I'm afraid you have to confront and discuss this reality with your Profs or supervisors to advise and get permission to design on that basis. The theoretical method may look OK, but it doesn't work and it doesn't apply directly and simply. This is just another example of why Engineers are so important and why "pure" scientists are not capable of resolving industrial problems of this genre. It isn't because the scientists don't "know"; it's because they don't have the experience in knowing "how" the ultimate result is achieved in real practice. This is why engineers are expected to be "practical" - the word comes from "practice".

I hope this experience helps. (& thanks for the birthday note; I hope Chris Haslego doesn't get mad with us using this forum for birthday cards.)

Art Montemayor
Spring, TX

[Natural FAL]

Hi. I am also working on the vaporizer.
Still quite new to process calculation, need some help from you guy expert,

Based on above, "time required for contact is controlled by the mass transfer coefficient"
Can you evaluate more on the above? How to calculate the surface area/ retention time of the pall rings packing?

I look through the cooling water tower design, still cannot get the concept & formula right.
How to calculate the heat integration/ balance & mass balance of the vaporizer?

Is the gas flow rate & speed affected the vaporization rate?
Please assist me.

Thanks!
Natfal