CO2 removal optimisation

[seb07150]

Hello All,

I'm a fairly new process engineer working at a gas processing plant with a CO2 removal train which uses BASF's aMDEA amine solution for removal. I am working on a project which is looking at our available capacity for CO2 removal and if this can be increased to accommodate additional throughput. One of the options being looked at is to optimise the performance of the train by removing bottlenecks in the process. The way I see it, the bottlenecks are:

Maximum gas velocity through the column
Mass capacity of the amine regeneration heaters limiting CO2 removal capacity
Max amine velocity in piping

I have looked into increasing the amine velocity but due to integrity of the pipes this is not possible. The amine solution has also been replenished so it matches BASF recommended spec. Without physically altering the column or the regen heaters, I'm at a bit of a loss as to other options to explore (aside from bringing another train online).

Does anyone have experience of this type of problem and could point me in the right direction?

Thanks

 

[thorium1]

perhaps you can post a sketch with the relevant numbers like flows etc

 

[seb07150]

Apologies for the delay in getting back to you. A link to a basic sketch with the unit operations is given below. The black lines represent process gas, orange represent rich amine, green are lean amine. Up until this point the only option I've been able to look at is adding in a molecular sieve upstream of the CO2 removal train to partially remove some of the CO2 from the gas and reduce the concentration going into the absorber, however this proved uneconomical.

http://postimage.org/image/jyacy58x1/7ce402c0/

 

[zdas04]

This problem assumes that the original designers missed some crucial element that can be retrofitted. It happens, I suppose, but not all that often. I've looked at several dozen of these "just a bit more capacity" problems over the years and I've yet to find a major blunder in the original design that I could exploit (the blunders I have found did not lend themselves to exploitation, only replacement). I've been able to reduce some turbulence here and there by changing valve technology (e.g., replacing globe valves that are always fully open with ball valves) and gotten a few thousandths of a percent improvement here and there. Probably nothing that ever paid for itself.

One thing that sometimes works is to change the boundary conditions. For example, I worked on a project to improve the performance of an evaporation pond. The only way to improve it was to add heat. It was sitting next to a large compressor station that was heat limited 5 months out of the year (i.e., they had to run at less than 80% capacity in the spring and summer because they couldn't get rid of enough heat). We retrofit shell and tube heat exchangers to pull the various streams on their way to the fin-fan coolers. We took the water out of the new heat exchangers and ran it through a final heat exchanger that took the heat out of the exhaust stack. Then we sprayed the hot water over the evaporation pond and got a significant improvement in evaporation (on the order of 6 fold). Original designers of the compressor station didn't factor in the new heat sink so we were able to get the compressor station throughput from 80% in the summer to 120% year round (by lowering the interstage temperatures significantly).

I don't know if you have any of these "changeable boundary conditions" or not, but if you don't then my experience says that you've got a tough task.

David Simpson, PE
MuleShoe Engineering

"Belief" is the acceptance of an hypotheses in the absence of data.
"Prejudice" is having an opinion not supported by the preponderance of the data.
"Knowledge" is only found through the accumulation and analysis of data.

 

[poli60]

Hi Paul.
What is the CO2 content in the feed? (better if you give the entire composition)
What is the target CO2 content in the sweet gas?
In my experience, the normal ways to face this kind of problem are the following two:
- increase the amine concentration (but it seems you say that BASF has confirmed the current one);
- adopt a more active amine (DGA?).

Regards

 

[seb07150]

Thanks zdas04 and poli60 for your replies,

I'll explore the boundary conditions suggestion as it is something I haven't looked at as of yet.

As for the composition:
N2: 11.9mol%
CO2: 5.36%
C1: 76.29%
C2: 4.20%
C3: 1.33%
C4: 0.60%
C5+: 0.32%

The target CO2 content in the sweet gas is ideally 50ppm as this was the original design spec for the downstream nitrogen rejection unit, however the vendor recently confirmed that the unit can actually handle 100ppm, so there is a bit of a buffer if the 50ppm target is exceeded.

I'll have a look into other types of amine, particularly DGA.

Thanks again

 

[Montemayor]

Paul:

Face the real facts of this situation:
[ul]
[li]You have a proprietary, licensed Acid Gas removal process. You cannot play around with this process without invalidating your license and performance warranties - if indeed, you have any. You can't increase the acid gas loading in the Rich solution and you can't strip beyond the stated Lean solution limits, so you are #$%@** as far as doing anything with your solution to increase capacity.[/li]
[li]All you are really left with is increasing the solution flow rate. But to do that sucessfully, you have to increase aMDEA solution velocities which you admit you can't do. You also would have to have available excess heat transfer capacity in your reboiler, aMDEA Heat Exchanger, aMDEA solution cooler, and LP acid gas cooler-condenser - which I seriously doubt any contractor would have left you with.[/li]
[li]Therefore, you my friend are stuck with what you either bought or inherited. You need a new acid gas unit to operate in parallel with the existing one.[/li]
[/ul]
By the way, the title of this thread has nothing to do with the topic. This is a plant upgrading problem, not an "Optimization" application. Your unit was optimized to perform at the rate you originally bought it for. You now want to get more production from it; this is not optimization.

 

[poli60]

Paul,
I'm currently evaluating a case similar to yours.
If you want, I kindly ask you to contact me (see my profile) to share some views/ideas.
Unfortunately, there are some confidentiality issues that, for the time being, I can't disclose on the site.
I'm very sorry for that.
Regards

 

[seb07150]

Montemayor: Thanks for the honest answer, it is something that is inherited and now looks like the only option on the table is indeed another acid gas unit.

Poli60 - thanks, I will contact you via email.

Regards,
Paul

 

[Aggie87]

Assuming that to treat 285 MMSCFD of this gas with aMDEA to 50 ppm CO2 uses all of your available amine circulation and reboiler duty, then switching to a sweet gas spec of 100 ppm CO2 only increases your capacity to 293 MMSCFD. No appreciable gains are available from adjusting amine blend recipe either.