Absorption and stabilization of non-condensable gases at the top of the stabilizer (high H2S content)
Thread Content
This post was last edited by qhhqhh on 2016-7-12 at 14:49. Hello everyone: I developed my own absorption and stabilization Aspen model, and after running it, I found that non-condensable gas appeared at the top of the stabilizer (separation tank: 1.1 MPa, 40 degrees Celsius). Looking at the composition of this non-condensable gas: approximately 30% is H2S, less than 1% is C2, and the remainder is mainly liquefied gas (the values above are in terms of molar fractions); at this time, the temperature at the bottom of the separation tower is 125 degrees Celsius. To prevent non-condensable gases from escaping at the top of the stabilizer tower, it is necessary to raise the temperature at the bottom of the separation tower to around 143 degrees; however, this will result in a significant increase in energy consumption across all components as well as in the volume of material that needs to be processed. Personally, it seems a bit odd; clearly, when the temperature at the bottom of the reactor is 125, the C2 content at the top of the stabilizer is already very low, mainly due to the high level of H2S. In the original process, the non-condensable gas from the top of the stabilizer was recycled together with the dry gas from the top of the reabsorption tower for dry gas desulfurization; obviously, the situation of having non-condensable gas at the top of the stabilizer as described above is not appropriate. I wonder if, in actual production processes, it happens that the non-condensable gas at the top of the stabilizer consists mainly of liquefied gas and H2S How was it handled? Is it to increase energy consumption in order to raise the temperature at the bottom of the distillation column? Note: The molar fraction of H2S in the feed gas is relatively high, at around 12%.
Please share the calculation process and results for everyone to analyze.
I have encountered similar problems as well. You can try using a different property-based method, such as SRK or PR, and set H2S as a Henry component to see if there is a significant difference. If the problem cannot be resolved, it might be that the binary interaction parameters for H2S are not accurate enough. I hope this can help you.
This post was last edited by qhhqhh on 2016-7-15 at 17:03. Thank you; since the company computers cannot handle sliding verification and thus prevent access to the website, I am responding only now. I also used SRK, but later a senior suggested using the RK-SOAVE property method. If the temperature is insufficient, non-condensable gases with H2S and liquefied gas as their main components will form at the top of the stabilizer. After attempting the calculation using RK-SOAVE, a bottom temperature of around 139 degrees seems appropriate for the desorption tower. The separation efficiency of each tower is much better than that achieved with the Grayson method I used initially (much less absorbent is required). Did you resolve that situation by using a method of changing the properties? Could you roughly describe the problems you encountered at that time and the solutions you used?:P
It’s a case I’ve been working on for a long time; from what I remember, various physical modeling methods were tried. For components with high H2S content, simulating the desorption tower does indeed lead to the problems you mentioned: high temperature at the bottom of the tower and high load. Personally, I think the H2S content might be too high, so certain physical property methods need to be adjusted.
I learned *aspen on my own; I’m not familiar with how to modify the material property methods. Isn’t it only when the physical property method does not match reality that corrections need to be made? What I mean is: perhaps a high level of H2S requires a higher bottom temperature and greater load; if we make changes, won’t that end up being incorrect? And how do I fix it? Thank you.