Saturated CO2 issue 1

[npf]

I have one question for you all

We have water saturated CO2 gas (100 % Relative Humidity). This gas contains around 100 ppm of H2S , Temp : 110F Pressure: 150 psig.
It enters in the fixed bed of iron oxide where H2S is removed.But we are seeing problems such as the bed gets exhausted earlier than expected.Also we see some heating of the gas which is unusual.We have no clue what might be happening
These are some of the doubts that I have
Is carbonic acid formed that reduces the reactivity?
Reaction of carbonic acid with iron oxide?
Nature of carbonic acid
Can you provide any suggestions how to overcome this problem

Regards
sp

p.s I saw Art Montemayor addressing this sort of issue.Art can you assist me ?

 

[Montemayor]

sp77:

I am familiar with the Iron Oxide Process from experience in Europe where it is still used in some areas for purifying coal syn gas with up to 1,000 grains H2S/100 Scf gas. The process depends on the following exothermic reaction:
2Fe2O3 + 6H2S = 2Fe2S3 + 6H2O

The process is hardy, robust and flexible - although high in labor costs. A considerable temperature rise occurs in the oxide beds when feed gas containing a high concentration of H2S is treated. The normal way I've seen the exotherm handled is by providing liberal external surface for dissipation of the generated heat by radiation and convection.

The iron oxide is usually revivified "in situ" by addition of air (or O2) in the feed gas:

2Fe2S3 + 3O2 = 2Fe2O3 + 6S

I've never used it on CO2 gas; I've always treated CO2 with liquid solutions for H2S removal. This is considered more efficient - especially in recovering the sulfur in a subsequent Claus process. The iron oxide cycle is described as proceeding most satisfactorily at 100 oF and in an alkaline environment. This would call for attention when applying it to water-saturated CO2 - which is acidic.

What I think is happening in your case is that high H2S "spikes", or content, in the feed CO2 is causing high reaction heat and early bed exhaustion. Are you monitoring your H2S content coming in? I do not believe that the CO2 is causing any reaction or temperature rise; if anything, I would suspect the carbonic acid (hydrated CO2 radical) is retarding the reaction rate because of its acidity. Outside of the acidity and corrosive tendency of the carbonic acid, it should behave in an inert manner. It should not partake in any significant reactions with the iron oxide, the sulfur, or the oxygen.

Art Montemayor
Spring, TX

 

[25362]

I'd like to comment on the subject lacking any direct experience on the process. I think that the presence of moisture in the incoming gas is of importance. The reaction itself, as pointed by Montemayor, produces water and this water keeps the Fe cations sufficiently acidic to avoid interference by absorption of CO₂. The reaction is exothermic and rates may increase very quickly with temperature and by spikes of H₂S as indicated by Montemayor.

If the gas were dry, or not sufficiently saturated, or the reaction warmer, water may be evaporated and this minimum acidity level would tend to disappear. A fact that may lead to bed inactivity, because, as I think, CO₂ would start competing with H₂S for the basic sites.

As I said, I'm trying to analyse the problem, w/o having dealt with it in the past. To Art Montemayor, please tell me if I'm way out of the right track.

 

[Montemayor]

25362:

I think you are right on track with your analysis. This may be a tough nut to crack without direct, hands-on information. As you know, this is a classical, iorganic, stoichiometry reaction that "should" be sweet and simple to carry out and control. However, like all things that appear to be sweet and simple, this one could bite you in the posterior with trade-offs and the idiosyncrasies of carbonic acid's pH.

I've seen the process work - but with town gas, not CO2 and it's characteristic failings. As you've pointed out, although the CO2 is normally considered "inert" and commonplace, it carries some peculiar chemical activity results due to the acidic properties that surround the carbonic acid radical when moisture is present. And the thing I remember about the iron oxide process is that it is dependent on moisture being present. From what I remember of operators' comments, the exotherm heat doesn't scare me, but as you point out, the heat up also upsets the humidity inside the bed(s) and this may start to influence the competition between both acid gases (CO2 & H2S) for the basic sites on the oxide. I may be wrong, but I don't think this is that simple a problem.... I always worry about confronting something that appears to be so simple and straightforward. If we luck out and get someone with hands-on operating experience on this process, this might turn out to be a very interesting thread to follow.

Art Montemayor
Spring, TX

 

[25362]

to Art Montemayor, thanks a lot.

I think we may find here a similarity with treating combustion flue gases to reduce SO₂ in the presence of CO₂. It is known that NaOH although highly effective, "cannot" be used because CO₂ consumes much of this chemical, and it is of common knowledge that CO₂ would be kept out, so to say, by using a chemical that would result in a pH between 4 and 6.
It was this situation that prompted the use of limestone. A use that brought with it other problems and complications.

I also remembered that FeCl₃ solutions are corrosive to metals, and acid because of the cation Fe⁺⁺⁺, while the anion Cl^- is basic. The cation being small and highly charged reacts with six water molecules to form a complex [Fe(H₂O)₆]³⁺ that by itself reacts with water as follows:

[Fe(H₂O)₆]³⁺(aq) + H2O(l)-> Fe(H₂O)₅(OH)²⁺(aq) + H₃O⁺(aq)
with a Ka=4.0*10^-3, quite acidic.

If the amount of the trivalent iron cation left "in solution" would be just that resulting from the solubility product of the formed iron sulfide, this "solution" would be quite acidic. Sufficiently acidic so as to impede the absorption of CO₂.

All these assumed reactions need water to take place. That was the reason of writing my previous message.

As you say, this would be an interesting thread to follow and learn from the participation of experts in the field.

 

[npf]

hello,

Art & 25362
Thanks for your assistance.From this discussion we got some direction on how to solve this issue.I will get back to you soon with latest development in this issue

Again Thanks
Regards
sp77