Simple H2S & Water Absorption
Simulate chemical processes using advanced thermodynamic models
Brought to you by:
danwbr
Menu
▾
▴
Hello!
I am having trouble with my very simple absorption simulation of H2S, carried by an inert gas, with water in a counter-current flow. I am just learning to use DWSIM so I am probably missing something obvious. I keep getting error messages such as INDEX WAS OUTSIDE THE BOUNDS OF THE ARRAY and FLASH CALCULATION ERROR. I have tried a lot of different algorithms and every feasible property package.
Could it be that I need to use the electrolyte PP to simulate this absorption? How would one do so with a gas inlet stream?
I will attach my files into the comment section below.
Thank you!
Here is the attachment, one of many attempts to converge this absorption column.
I am getting a bit further, but the mol frac on my top stream is way greater than 1. Please help !
There was a problem with a connection, and also the inlet streams were defined incorrectly. It calculates now but, in order to get some gas in the water, you'll have to raise the pressure a lot in the column.
Last edit: Daniel Medeiros 2015-02-06
Thank you so much, the gas won't scrub at these conditions, which is what I was hoping to prove!
Could you explain how you corrected the inlet streams please?
Much appreciation,
Catherine
Also, the water should be cooling the gas, not the gas heating the water, in this situation. Do you know how I might be able to fix that?
The bottoms in this simulation comes out at 366 K which is what the gas-in Temperature is equal to. With such a high liquid to gas ratio, the bottoms should come out at their starting Temperature, at or around 288 K. Could you please advise?
Tried a lot of things but it seems that this problem is not solvable, not by the newton methods that I use in the solvers. You could try using the ChemSep column model which has more solving options, maybe it works.
If you installed ChemSep with DWSIM, it should be available as a selectable option when you add a CAPE-OPEN Unit Operation.
I don't know but it seems we're missing something here... it seems too obvious, but maybe the models in DWSIM cannot predict the heat effects correctly and try to increase or decrease the temperature in the column too much, way beyond the limits.
ChemSep worked just fine. :-)
Awesome! This is really great. Thank you so much for helping.
Yes, it seems like the heat gradient was lost on the absorption model.. but the ChemSep is working really well.
Hi Catherine,
I've fixed my solver and now it gives the same results as ChemSep's. Can you try running this file with Parallel calculations disabled to see if it works? Anyway, it should work normally in the next release in a few days.
Thanks,
Daniel
Thanks Daniel,
I tried this file with the parallel calcs off, but I keep getting the weird heating as before. When does the next release come out?
Oh right, I forgot to tell you that I've changed the temperature initialization too...
The new build should be out in a few days.
Hey Daniel,
New build has been working really great, thank you!
Everything was going smoothly, however, until I added NaOH to the cooling water mix. Now I am getting liquid as a result in both tops and bottoms, and the heating calcs went crazy as well.
Do you mind taking a look at my simulation with NaOH?
I building the SIM with the CHemSep module as well, I will let you know how it goes.
Thanks again so much,
Catherine
And here is my attempt with the Cape-Open unit. It seems like the unit has ignored the addition of caustic and other gases?
So, I figured out my problem with CAPE-OPEN. I don't have it. Duh.
How can I get it?
Is there an easy way to install the CAPE-OPEN property manager?
I found this code...
https://raw.githubusercontent.com/DanWBR/dwsim3/master/DWSIM/Objects/PropertyPackages/CAPE-OPEN/CAPEOPENPropertyPackageManager.vb
I also downloaded the CO process standards v 1.0, however I do not know how to integrate them into DWSIM..
Please help!
Sorry to keep bugging. I found it in the tutorials ^^^ and I figured it out. THanks!
Hi Catherine,
CAPE-OPEN is not something you have to install, it is an interface definition. You can have some components installed which will communicate with DWSIM using CAPE-OPEN interfaces.
NaOH won't really work because it is an electrolyte, I doubt that Peng-Robinson is able to handle it nor ChemSep has a model to handle it too...
You could try using the Extended UNIQUAC model in DWSIM but I don't know if it will work in a absorption column operation.
Thank you for your answer, I have really learned a lot about software exploring these problems! I will attempt the Extended Uniquac & let you know how it goes..
hello Sir,
good day. i think i have the same problem above there the same error exist, i really need help i need it badly for my design.
how fix this problem?
thank you so much.
hello Sir,
good day. i think i have the same problem above there the same error exist, i really need help i need it badly for my design.
how fix this problem?
thank you so much.
Hi Joseph,
can you check if this problem is solvable, do you have an example in a book or in another simulator?
Hi Joseph,
your component "Nitrogen Oxide" has a very low boiling point and a very low critical temperature. This means that this component is a supercritical gas to be absorbed in a liquid phase of NO2. For this kind of problem your selected setup is a little wrong:
First: the liquid stream at the top is to hot now, therefore you have a vapour but the absorber column needs a liquid here.
Second: you need to enter the gas phase at the bottom stage, but you are entering one stage above that.
Third: You selected "NRTL" as property package. This kind of PP would need a Henry coefficient to calculate solubility of supercritical gases in a liquid correctely. The actual version of DWSIM is not supporting this Henry calculation. But the next release already was prepared for this kind of calculations. You will have to wait for that, sorry!
As solubility of the gas in the liquid will be very low you will have to apply a large amount to a little gas flow not vica verca as you do now!
As an alternative you may try an EOS model like PengRobinson. This would give at least a some kind of result but you have to consider the calculated solubility of NO to be correct.
This is just an intermediat result from my first quick investigations. Even regarding all that points from above i still encountered problems with the absorber column. As you only use two components you also should get the same right just by mixing both streams in a flash vessel. I am going to check if this problem will be calculated with our next release version of DWSIM without problems.
regards
Gregor
I've tried a lot of different setups last night and the column temperature profile just doesn't converge, that's why I would like to see it running in another simulator to see what happens... or even running it with a ChemSep column which I just noticed that I forgot to do... :-)
ChemSep doesn't work either. I think that this problem is not solvable the way it is setup, even if the temperature of the solvent is lowered a little bit to form a liquid phase.
Greg gave me the hint and I think we've found out why it wasn't converging.
You should put a small amount (mf 0.0001) of Nitrogen dioxide in the solvent stream and also a small amount of Nitric oxide in the gas stream. Make sure that the solvent stream is all-liquid.
The column solvers work by using newton's method which must get the correct temperature dependence of the mixture enthalpy. By inputting a small amount of each compound on each stream you help the solver on finding the right convergence "path".
This probably happens because the mixture behaves very differently than both pure compounds in terms of enthalpy-temperature dependence.