10 replies to this topic

[NikkiC]

Hello!

I am new to the whole amine system, so this might turn out to be a really long discussion for way too many things! But hopefully, in the end, somebody will find this post interesting and helpful.

 

Now, my company is trying to find ways to optimize MDEA units, which serve to remove H2S from many different sources in a refinery, including recycle H2, offgas, LPG, etc. Based on my background research, I've found that the presence of heat stable salts (HSS) is usually a serious problem in the amine systems, and that its concentration can actually affect the absorption in both positive and negative ways. What I'm trying to do right now is to simulate the effect of HSS in HYSYS v8.6 with the new Acid Gas package. 

 

I am starting with the samples provided by HYSYS (more specifically, "Effect of Heat Stable Salt on Acid Gas Cleaning using MDEA"), and try to understand how to construct that simulation properly. Here are my questions:

 

1. It looks like the concentrations of HSS are constant throughout the columns. But I thought they should be the products from certain reactions that occurred inside the absorber (for the most part)? The purge stream in the sample is pure water, so essentially the sample process doesn't remove reactive products from the aqueous phase.

    (a). If I were to simulate a steady state process from a certain plant (assuming that I can get all the information that I need), am I responsible to specify the concentrations of the ions in the lean amine stream, AND assume that they don't accumulate over time?  Is there a better way to do this?

    (b). What if I cannot get lab data for all the ion concentrations??

 

2. How does the makeup unit work?  Does it only allow pure water to be purged? It doesn't make any sense to me. If it's used as a storage or mixing tank, shouldn't the purge stream have the same composition as the outlet?

 

3. How to specify the tray efficiency here? That option disappeared from the parameter tab.

 

4. What's the real difference between the efficiency and advanced rate based models?  Other than that the advanced method is more "rigorous"?

 

5. Generally speaking, how does one quantify foaming in an existing column?

 

Thanks in advance!

[PingPong]

This is not my specialty, and I don't use Hysys, but it seems to me that you better use your brain instead of Hysys (or any other simulator).

 

Start reading about HSS and how they are formed, what components in the sour gas and the make-up water lead to formation of HSS, and what impact of reboiler tube wall temperature is, et cetera.

 

 

It looks like the concentrations of HSS are constant throughout the columns.

HSS is in the amine, and the amine is in all columns from top to bottom, so what else would you expect.

 

Same with foaming: do proper literature research so you understand what it is caused by, and how to minimise it.

 

If your company wants to minimise HSS and foaming, they should determine what causes the HSS and the foaming, by analysing all feed streams entering the amine system, and detailed analysis of the amine. Identify the trouble makers (based on the knowledge you obtained from literature research) and try to eliminate them from the feed streams.

Edited by PingPong, 08 January 2016 - 08:14 AM.

[Art Montemayor]

NikkiC:

 

I agree with PingPong's basic advice.  How do you expect to "simulate" actual and credible behavior of illusive and evasive phenomena like foaming and heat stable salt formation?  trying to quantify foaming and HSS formation is, my opinion, useless.  What is there to gain from knowing that a simulation program predicts foaming at a certain level of surfactant content?  How would you know the instant that the level is reached?  Would you accept a "little" foaming?  And how do you quantify "little"?  The same practical common sense applies to HSS formation.  How much HSS content is tolerable?  and how do you control it at that level?  The methods I've always used in the past when operating amine systems is to simply avoid the situation entirely.  And for the years I spent in such endeavors I never had a problem.  I always reclaimed, filtered, and used adsorbers on my amine solutions.  I didn't tolerate the presence of HSS in my solutions - plain and simple.

Even if you succeed in organizing a simulation program, how do you warrant and verify the credibility of its results?

 

The entities that I would trust with useful or credible aMDEA design and operational information would be the developers of the chemical - such as BASF.  Presumably you are in China and your company makes aMDEA also.  I would presume that you are already researching the design and operations of aMDEA units using not only simulation programs for the basic design, but also pilot plant operations to confirm and verify all computerized predictions and design.  If your company is truly trying to "optimize" the design of an aMDEA unit for H₂S removal, then they must follow the same path and methods used by existing companies in the business.

 

Combating the formation of HSS in amine solutions is nothing new.  It began when R.R. Bottoms invented and patented the amine acid gas process in 1933 - 83 years ago.  So the problem, the study and science of the subject is not a new one.  Although MDEA is a relatively newcomer to the amine family, I am quite sure that companies like BASF have already confronted the best method to remove or control the presence of HSS in their licensed aMDEA H₂S removal process.  Kindly refer to the attached documents.  Perhaps they can be of help.

 BASF Experience with aMDEA CO2 Removal System.pdf   491.05KB   110 downloads

 Heat Stable Salt Effect on Amine Absorber & Stripper.pdf   208.61KB   94 downloads

 HSS Removal from Amine Solutions.pdf   154.18KB   91 downloads

[NikkiC]

Hey PingPong and Art,

Somehow I was really hoping the two of you would reply to this thread. So thank you both very much!

I want to clear one thing up, though, by "optimizing the amine system", I am essentially building a proper model that describes the amine unit, so that it can be used to reduce reboiler duty etc. Unfortunately, our company doesn't manufacture anything, so we rely on the data provided by the plant. I'll also check with the manufacturer for more information! Thanks Art! So Art is absolutely correct, I need to ask myself this question:

 

Even if you succeed in organizing a simulation program, how do you warrant and verify the credibility of its results?

 

To be honest, quantifying foaming (as a result of surfactants) doesn't really concern me - an empirical foaming factor would suffice. I was just kind of curious

 

@PingPong:

HSS is in the amine, and the amine is in all columns from top to bottom, so what else would you expect.

Would you mind explaining this part to me please? They are generated by chemical reactions, so if I don't remove them from the process, they should be accumulating, right? So I guess my question really is, why are the HSS concentrations NOT increasing? I'm not relying on this simulation to produce any results for now; but rather, I'm trying to understand why HYSYS does what it does. 

To the best of my knowledge, HSS can form in the following scenarios (correct me if I'm wrong):

1. Acid in the feed gas, causes formate, sulfate, sulfide, acetate etc. to be generated. And I believe these are almost inevitable? 

2. Presence of oxygen that is dissolved in makeup water, carried in the gas feed from upstream operations or insufficient nitrogen blanket in the amine storage tank etc.

3. CN- from sulfur recovery unit, coker, FCC etc.

4. Cl- carried in the water

5. Corrosion inhibitors or anti-foaming additives, cause nitrites and nitrates to form. 

And all of the above five scenarios essentially increase the concentration of RNH+, and therefore will shift the reaction equilibrium constant. So I do think it's important to quantify them. Or at least quantify the activity of amine. More specifically, in the 2nd article that Art attached, they mentioned that there's essentially an optimum RNH+ concentration that yields the smallest H2S slip. And I'm actually trying to use HYSYS to confirm such an optimum exists, that's another reason that I need to know why the HSS concentrations in HYSYS stay as user specified. 

 

Best,

Nikki

[PingPong]

Would you mind explaining this part to me please? They are generated by chemical reactions, so if I don't remove them from the process, they should be accumulating, right?

Yes, but normally that will go very slowly. If you measure HSS at the top as well as the bottom of the absorber you will not measure much difference.

To the best of my knowledge, HSS can form in the following scenarios (correct me if I'm wrong):

You covered most causes, but you still did not inform us what kind of sour gases your amine unit is processing: from refinery units, or sour natural gas, or CO₂ capture  from syngas, or what?

 

For simulating amine systems one normally uses ProMax or ProTreat. I have no idea how accurate Hysys is for this application.

 

Note also that the 2nd article that you mention deals with the theoretical benefit of partially neutralizing the amine with some acid.

If have also read similar by Bryan Research (ProMax licensor) that used phosphoric acid to partially neutralize the amine.

 

However what I don't understand is why they don't use for example acetic acid, as that would be more realistic. I am not so sure that the HSS in an actual plant is comparable with the pseudo HSS that they create. It seems to me that the RNH⁺ concentration will be different if it was created by a fairly strong acid, or a weak one like acetic acid. In the latter case I would expect that the acetate ion would bind to most of the RNH⁺ and therefor the same molar amount of acetic acid will have much less impact on the RNH⁺ concentration than a stronger acid.

But I am not an expert in this field, and never do amine unit simulations.

 

In any case I don't think your company will allow you to add acid to the amine, no matter what benefit your model will calculate.

 

Remember also that HSS is corrosive and has some contribution to foaming.

Edited by PingPong, 12 January 2016 - 10:44 AM.

[RockDock]

I'll go ahead and add that HSS are not bad to have in the system. I personally like having HSS in my amine because it helps me strip the H2S in the regenerator. There are different recommendations from experts on this. Typically, I want the HSS to be less than 1 wt% of my solvent. Since our company does not use licensed solvents, we generally purge some portion of the amine from time to time to maintain HSS below 1%. That is normally once a year or two or three. HSS should not build up quickly. If they are, something is wrong.

 

As for the Software to use: Hysys is definitely not the one to use. I would only use ProMax or ProTreat as shown above. My preference is ProMax for several reasons.

 

One question you ask is: What if I don't get an ionic analysis of the amine?

 

Then, you cannot model it. In my ProMax model, I use acetic acid, sulfuric acid and several other acids in order to model HSS. It all depends on the ionic analysis I get from the lab. I also use the pH as an indicator. The nice thing about modeling it this way, is the ProMax model will show me my salt concentrations in my ionic analysis.

 

When an acid is added to an amine in the plant, it is generally phosphoric acid. Perhaps that is what the ProMax people were doing in their paper.

[NikkiC]

@PingPong

Thank you for your input! My company focus on refinery gas streams. But we also exclude sulfur recovery units from our analysis, because such units usually operate on their own and we'd rather avoid this dangerous process.

I feel like one of the reasons why people like adding phosphoric acid to the amines is because it has 3 protons/(mol acid), so that it's more effective than stuff like acetic acids. 

 

@RockDock

Based on the information I got from one particular refinery, they have an AmiPur system installed inline and also try to maintain the HSS below 1%wt. But they still keep the AmiPur running even when HSS is lower than 0.5%wt. I doubt if this is a good practice though (?)

 

HSS should not build up quickly. If they are, something is wrong.

 Regarding your comment here, are you modeling the rate of accumulation in a dynamic environment? Do you find it beneficial for operations? How did you establish a baseline? And if HSS does build up more rapidly, how do you troubleshoot this amine system?

[RockDock]

I would not consider it necessary to run the AmiPure system when HSS are below 0.5%wt. I would switch it on when it gets to 0.9-1%wt.

 

I do not model the rate of accumulation, nor would I recommend you to even if a simulator claimed to be able to do so. There are too many factors that contribute to the formation of HSS. I model it based on the lab result so I can see if it helps or hurts my process. HSS help the stripping of the H2S in the reboiler, but hurts the absorption in the absorber. At some point the absorber behavior outweighs the benefits in the stripper. The model helps me know when this happens.

[RockDock]

If HSS build up rapidly, I look at the absorber T and the inlet filters. There must be some contaminants in the gas. I want to minimize the formation of the salts, which is why the absorber T plays a role.

[NikkiC]

@RockDock, Thanks for the information! It really helps!   

 

I have another question in terms of the absorber operations. So I went to a refinery a few days ago, and did some field research of the recycle hydrogen desulfurization unit in a hydrocracking process. I've attached a schematic of the unit. A process engineer from the plant told me that they sometimes observe a rapid liquid buildup on the overhead flash drum (i.e. compressor inlet separator D-111 in the sketch). My question is, what's causing this liquid build up? Is it possible and necessary to include this phenomenon in a simulation? I suspect that it is caused by high vapor velocity in the column (so essentially jet flooding), and that it's also intensified by foaming. Does anyone have an alternative explanation perhaps?

Attached Files

•  HCU Recycle H2 Desulfurization Sketch.jpg   49.38KB   10 downloads

[RockDock]

It may be foaming. I would check the dP of the column during these events. If there is a spike in the dP, it is likely caused by foaming.

 

Alternatively, if the solvent is MEA and there is enough temperature drop between the column and the drum, you may have water and MEA dropping out.