16 replies to this topic

[J_Leo]

Hello,

 

Has anybody successfully designed Amine Contactors or Strippers based on the absorption equations in books such as GAS PURIFICATION?

 

Generally speaking, the amine system design is usually by the licensers such as UOP and BASF. Can we design these towers by ourselves?

 

If it is very feasible, I would like to take some time to set up calculations to do the work in the future.

 

Thank you.

 

Best regards,

Edited by J_Leo, 28 May 2015 - 03:36 PM.

[J_Leo]

I have used Promax to theoretically design the system. I can easily obtain tower diameter from Promax. I could use 7 theoretical trays for these towers. If I set 20 trays (common practice) for each of the towers, it seems the tower height is straight forward. However I cannot convince myself the towers will work properly as the program predicts.

[Bobby Strain]

You can enlist some of the solvent suppliers, too. Huntsman is one. They probably use Promax or T-Sweet, both Bryan Research products.

Bobby

Edited by Bobby Strain, 28 May 2015 - 06:25 PM.

[ColinR33]

As suggested, you can use one of the chemical suppliers such as Huntsman, Dow(UOP), Brenntag, etc to figure out the required circ rate for you and then sizing the tower is a matter of packing/tray hydraulics and contact time requirements, we do it all the time. We also design towers for generic amines using (As Bobby suggested) ProMax from BR&E, or even Hysys Amines package (you do have to know what you're doing with that one though, it has some limitations compared to ProMax.

[Art Montemayor]

J_Leo:

 

Please be specific.  There is a world of difference on how you should approach the design of an amine absorber or stripper.  It all depends on  WHICH AMINE you are referring to.  The vapor pressure between MEA and aMDEA at the top of the absorber is very significant.  You must be specific.  Generalizing and dealing with the generic “amine” term will only generate general answers and responses.

 

I have designed a lot of “amine” units - most of them have been MEA, which I consider the worse of the lot due to the vapor pressure, degeneration rate, and corrosion problems.  I even locally fabricated and installed some units in developing countries.  They all worked - and for a long time.  I started working with MEA when the first edition of Arthur Kohl and Fred Riesenfeld’s book, “Gas Purification” was published and didn’t buy a copy of it until 3 years later, when I had already experienced in the field most of what they wrote.  The book was later re-born with Kohl and Neilsen as authors - but with still the same, basic information and not too much new or radical discoveries.  Attached find a section of the 5th edition.  Note how the authors avoid any attempt to cite specific design methods, sizing, or detail calculations.  They don’t know how to determine the correct size of the towers, the proper internals for a specific case or the best, proven solution concentration for a specific application.  An example of the current state of stripper design for MEA plants today (as it was over 50 years ago) is that a so-called “reflux” is employed on the stripper tower.  No one in the industry, the countless papers, drawings, operating manuals, and text books on amine units for acid and sour gas removal have ever (to my knowledge) ever explained in detailed, engineering logic as to why a weak MEA aqueous solution is returned back to the stripper to add more size and reboiler duty requirements to the system.  Everyone just says that you have to use a reflux system on the MEA stripper.  It makes no scientific or engineering sense to think that you can “reflux” a top product that is a gas.  Every student in a distillation course knows that reflux is used to establish an equilibrium on the top tray and set the overheads product purity.  This is not what a stripper is supposed to do.  It is supposed to just simply strip out a non-condensable gas out of solution.  And, in an MEA stripper, I have done this with no more than six trays - not the 14 to 26 trays as recommended.

 

So, I’ve answered your specific question.  I can also add that I used the book to calculate the required MEA solution rate.  However, as you can see if you carefully study the book, the design and fabrication of the absorbers and strippers as done by almost all engineering contractor firms is done on an empirical and field proven basis.  This is obvious in most cases to avoid liability issues on performance.  Some of the designs I’ve seen in the field have been ludicrous and exaggerated in design and dimensions.  This is unfortunate because, as I suspect, most - if not all - of the design engineers involved have never built, installed, or fabricated an amine unit.  They have followed past, proven designs and relied on meeting contractual terms.

You can simulate all your heart desires.  But the bottom line will always be: what is your experience in operating the specific operation and what will you guarantee as the essential equipment to carry out the specified operation and yield the desired product?

 

As you suspect, you had better have second thoughts about how much faith you can put on a stupid computer’s output.  In the end, it’s not the computer who will have to respond to claims of “lack of performance”.  It will have to be you.  If you have not done this before with repeated success, then heed what our esteemed forum members have so expertly replied: leave it to someone who not only has justifiable claim to being an expert, but will also back up the results with legal responsibility.

 

I hope this experience helps out.

 

 Acid Gas Removal Design Criteria.docx   111.38KB   307 downloads

[J_Leo]

Thank you all for your inputs, especially Art.

 

I have worked on two projects with Amine System so far. One of the solvents used is aMDEA (BASF) and the other is MDEA (UOP Ucarsol). They have designed the system  for us. I used Promax to verify their results.

 

What I am trying to do now is to figure out how to do a preliminary design, which may be used in estimating and evaluation. Later on with more experience, I may be able to design the system. I really wish I can take the steps further, not stay at the same place every time.

 

Thank you.

 

Best regards,

Leo

[RockDock]

Our company does this frequently using ProMax. We have a few templates we use, but this is the order we generally use to design the units:

1. Set the flooding to 70% in the absorber and input our standard column internals
2. Choose our solvent (MDEA to slip CO2 and aMDEA or DGA to remove CO2)
3. Set the solvent concentration to 45wt%
4. Set the circulation rate to achieve a rich loading of 0.45
5. Use 1 lb of LP steam per gallon of solvent being circulated for the reboiler portion of the stripper

After this we run a lot of different scenarios using the ProMax Scenario Tool to evaluate different solvents, solvent concentrations and steam ratios to be sure the design is robust. This will ordinarily optimize it by another 10% or so. We then size all the lines, columns, vessels and heat exchangers. It is rare that we use a licensor, but sometimes we do. When we do, it is normally Huntsman since they provide the majority of the solvents we use. They are the best at working with us. We use a lot of their solvents in our designs.

 

Using this approach, we have successfully started up countless units with only minor issues that have less to do with the design than the design basis.

[J_Leo]

Rock,

 

Thank you very much for your inputs.

 

Now I am familiar with the simulation using ProMax and rules of thumbs for MDEA system design. I am not confident to design the towers especially the absorber because I don't have much basis for the actual number of trays or packing height. For example, If I achive the purification goal by using 7 theoretical trays and assume tray efficiency of 33%, the actual tray number would be 21. If I assume tray efficiency of 25%, the number would be 28. Eventually, it comes to tray efficiency determination. Can we really use a fixed efficiency for all projects? I guess there should be some calculation or emperical data to guide the efficiency determination for a specific acid gas removal process.

 

Best regards,

Leo

[RockDock]

I always use a 33% efficiency. I have not departed from that in nearly 20 years of designing these units. The efficiency is based more on experience than any empirical value, I believe. There may be some empirical evidence that I don't know about...

[J_Leo]

Rock,

 

Thank you very much for the design tips. What about packing towers? Can we fix the HETP if we know the packing type?

 

Thank you,

 

Leo

[Philip le Grange]

Hi,

 

Another possible simulator is ProTreat. We use it regularly and find it pretty good.  It uses actual stages based on mass transfer rather than HEPT based ideal stages.

 

Refluxes on amine strippers are used to condense water in the stripped gases. This lowers water make up requirements in the facility and in H2S containing systems lowers the size of downstream treating units. The difference in water loss between 105C and 45C is pretty dramatic.

 

Kind Regards,

Philip le Grange

(www.AmineExperts.com)

Edited by Philip le Grange, 06 January 2016 - 10:22 AM.

[Art Montemayor]

Philip:

 

I don’t understand your explanation for using a return of the stripper overheads condensate back to the stripper top with the intention of condensing water in the overheads vapor stream.  Could you please furnish more details regarding this process design feature?  You can refer to the attached Excel workbook sketch I furnish, using call-outs to indicate the specific locations and process conditions (temperature, pressure) in the streams.

 

The reason I ask for your detailed explanation is that I have designed, built, and operated CO₂ amine strippers without any so-called “reflux” and the units ran just as efficiently or even better than those that used “reflux” – for years.  I was not the first one to do this; the original inventor of the Amine process - the Girdler Corporation - did this also.  This categorically has proven that the traditional “reflux” serves no useful purpose, other than keep that portion of the solution in total recycle inside the system.  In fact, its use puts more load on the Stripper, making it bigger and also imposes a larger steam consumption on the reboiler tube bundle (an energy factor that is a big problem with amine reboilers).  I always returned all recovered condensate from the Stripper Overheads condenser-cooler and returned it back into the circulating system either at the absorber sump or in the suction of the Lean pump.

 

You may have a valid point with regards with H₂S removal requiring special attention, but I can’t see that the water content at the top of the amine stripper will be any different whether you return the condensate to the top of the stripper or somewhere else in the system.  The temperature at the top of the stripper is always kept constant (approx. 220 ^oF in the case of an MEA solution) and it is this temperature (& pressure) that establishes the water content in the gas stream exiting the stripper top section.  It should make no difference if you return cool (45 ^oC) condensate as “reflux”.  The top of the Stripper is still maintained at the prescribed saturation temperature that corresponds to the stripper pressure - approx. 220 ^oF in the case of an MEA solution.  And that is the prime subject I allude to: the introduction of cool condensate liquid at the stripper top necessitates the introduction of more steam into the reboiler steam bundle in order to heat and vaporize the mentioned “reflux”.  I believe that is a factual happening because the reboiler is the only source of system energy to heat and strip out the absorbed acid gases.  In order to fulfill the stripper scope of work, the outlet gases exiting the top of the stripper MUST be at the saturation temperature of the solution used.  Consequently any water vapor rising through the stripper that is condensed by falling cool condensate MUST be replaced by rising saturated steam vapor coming from the reboiler in order to maintain the specified (and required) top saturation temperature.  It is this top section temperature that ensures that the required acid gas loading is “stripped” from the solution within the top section of the stripper and not within the reboiler.

 

You are correct in stating that the difference in water loss between 105 ^oC and 45 ^oC is pretty dramatic.  However, another important point I can make is that there should be NO water loss involved with introducing the recovered condensate in any place within the amine solution circuit.  The only water loss(es) to any amine solution exist at:

• the overhead treated gas exiting the absorber (as water of saturation);
• the acid gas exiting the condensate separator downstream of the stripper overheads condenser-cooler (as water of saturation); and
• the usual leaks, spills, and samplings.

One major contractor designer once tried to justify the use of the “reflux” on an MEA system to me by stating that the returned condensate served as a scrubbing medium to avoid entrainment of rich MEA into the stripper overheads condenser-cooler.  My response was that if the stripper diameter was designed adequately, there should be NO entrainment in the overheads – especially when the proposed reflux was not used.  I never got a response to that statement.

 

One of my favorite process modifications that I was never able to incorporate because of top management budget constraints was to use excess steam to generate a refrigeration glycol solution to sub-cool the captured stripper condensate and introduce it in a cooler-scrubber tower where I could scrub and cool the absorber treated exit gas and thereby cool those gases and recover amine from them.  My calculations estimated a pay-back in a year – which indicated a handsome return on the investment and a benefit to downstream facilities due to reduced amine contamination and loss.  I always saw the condensate from the overheads condenser-cooler as an ideal scrubbing fluid to recover amine and reduce losses.  This is what is depicted in my submitted sketch.

 

Let me know your thoughts.  I’m always interested in learning more and your comments may give me further enlightenment on the subject.

 

 MEA LP CO2 Removal Process.xlsx   48.83KB   160 downloads

 

[Sisca]

Hello,

sorry, I am very new in the field of gas sweetening process.

I wanted to do research on the optimization levels of CO2 and energy savings.

Can you all give input variables that affect the optimization.

I plan to do the optimization of simulation results aspen Hysys, if it makes sense.

thanks.

[Art Montemayor]

Sisca:

 

It is very good that you would want to optimize a CO₂ removal process such as the amine system.  Although that is not what you state, I think that is what you mean.  Am I correct?

But before attempting to optimize any process, you must dominate all the details of the process - from the concept, design, and actual fabrication of all the critical equipment used.  And that takes some time in study, practice, and experience.   Some factors that you must take into consideration are as follows (note: this is not all-inclusive):

• The type and composition of the raw gas fed to the absorber;
• The composition of the acid gas removed in the stripper;
• The type of Amine system used;
• The concentration of the solution employed;
• The CO₂ loadings in the lean and rich solutions;
• The type of stripper-reboiler arrangement;
• The type of internals used in both the absorber and the stripper;
• The need to design and employ a reflux ratio on the stripper; (this will affect the reboiler size, steam consumption, and CO₂ cooler-condensate condenser);
• The type of reboiler heat source used;
• The cooling medium for the amine solution and the CO₂ cooler-condenser;
• The type of solution reclaiming or filtration used;
• The utilization or waste of the CO₂ removed;
• The process pressure in the absorber.

Your simulation results do not reflect an optimization of the process.  All that does is give an estimated optimization of process design conditions.  It does not tell you the physical, mechanical results if the process design conditions are implemented.  It does not reveal the capital and operating costs resulting from such a process design.  Therefore, it is not a practical "optimization".  An example would be the awful mistake that many chemical engineering students - and even some university professors - have done in the past by spending time and effort in simulating an MEA process based on using a 20+% wt. solution.  The ultimate, real world result is a rapidly corroded carbon steel system.  An MEA solution system in excess of 20% concentration is very, very corrosive and isn't used in the real world without taking into consideration the ultimate negative results.

 

If you are to attempt an optimized process simulation of all possible amine systems you would be spending a good portion of your life doing this.  Therefore, you must narrow down your scope of work to a specific, well defined process system.

 

If you want to discuss this specific subject any further, please start a specific, new thread based on your topic instead of using this thread.  This topic could take a lot of space and time to deal with in a constructive manner.

[Philip le Grange]

Hi Art,

 

interesting design, think we had our wires crossed I understand what you are talking about now.  When I talked about reflux I was referring to the entire overhead system loop. 

 

It is an interesting point you make about the reflux, I am with the engineering contractor you mention on this ... it serves as a water wash to remove entrained amine droplets from the gas, same as the wash section in the absorber on your design. That and as far as I can tell there is no better location for reintroduction of the condensed water into the system. Any suggestions on where to put it following the absorber water wash shown in your design?

 

I did a study a while back on reflux water temperature impact ... varying reflux return temperature made almost no difference to required steam input to maintain a specific stripper top temperature.  I think most of the reboiler energy goes into vapourising the reflux water again. To this end it does not really make a difference (in terms of energy) where that water is returned to as it will recirculate and still need to be vapourised.  

 

One possible downside of returning the condensed water to any other location would be that there would be different amine strengths in different sections of the system.(one refinery I was on actually did this intentionally so they could run their liquid-liquid treater at a lower strength than the rest of the amine system). I suspect varying strength will generally result larger equipment than necessary in some sections of the plant.  Be interesting to see a cost comparison on different reintroduction points in the plant (paper topic?).

 

Interesting ... have not really thought about this possibility before today.

 

Kind Regards,

Philip le Grange

(www.AmineExperts.com)

[Kvan]

Hello everyone I have question about sweeting gas processes

Q. Why we need high pressure and low temperature in the amine absorber tower thanks for answer my question

[Art Montemayor]

Kvan:

 

This is a basic, student-level query on the Amine process.  It doesn't belong on this thread because it is not part of the thread owner's topic,

 

If you have interest in discussing your specific topic with the Forum members, then please start your own, specific thread in the appropriate Forum. Our members will respond accordingly.

 

I will delete this post of yours found here after you post your own thread.

 

Thank you.