I am using ASPEN for simulating Methanation of CO2 and i choose LK-Plock EOS. I saw binary parameters were missing for certain pair of components, is NIST data good enough for DATA regression to find missing parameters?
I don't think you should have to alter anything in the equation of state to model Methanation . If you think there is a problem with Lee Kessler Plock try Peng- Robinnson or one of Redlich-Kwong EOS' and see if there is a significant difference. I think they should all work well out of the box for Methanation of CO2
Thank you Mr. Rick Manner for your suggestion.
The reason i'm using LK-Plock is because i'm doing my simulation at high pressure.
Some interaction parameters are missing for like CO and CO2 and also i saw deviation of VLE diagram when compared with NIST Binary VLE data for CO and CO2.
So is it a good idea to do data regression from NIST Binary VLE data to obtain missing binary parameter for CO and CO2?
Pratheek, What are your operating conditions? What reactions are you considering? Just methanation of CO and CO2 with H2 to form CH4 at high pressure? Most H2 plants built before 1980 have a methanators that operates around 500 F and 300 psig. I usually model this type of reactor with an equilibrium model and Peng-Robinson equation of state. Are your conditions significantly different than this?
In fact, models based on cubc equations of state have difficulty to treat systems containing hydrogen.(I never solved the problem satisfactorily)
I suggest to correlate gas density data of binary systems containing hydrogen, at the foreseen process conditions, to determine binary parameters.
It is lkely however that VLE calculations may give inaccurate concentrations of hydrogen in the liquid phase.
H2 behaves more like a salt than a gas. The solubility of H2 in Hydrocrabon increases as temperature increases the American Petroleum Institute (API) has published a nomagraph that works pretty well at predicting solubility of H2 in oil as temperature increases. I have never had any significant problems matching H2 solubility in oil. But I think that is a different situation I'm not sure what the liquid phase consists of in Pratheek's methanation problem.
Mr. Rick Manner in my simulation i am going up to 3000 psi and 600 F that's why I am using LK-Plock EOS suitable for extreme pressures.
Because of my high pressures i am not worried about the liquid phase, But the gas phase is showing little deviation.
if the problem is not VLE, but predicting fugacity coefficients in the gas phase, so as to calculate correctly chemical equilibria, I believe that Hydrogen parameters (mainly omega ) could be adjusted to second virial coefficient data (whiat gives a good measure of the capacity of reproducing hydrogen volumetric properties at all pressures.
Then binary parameters kij can be fitted to volumetric data of binary systems, namely second virial coefficients or gas density data of binary systems.
At 3000 PSI and 600 F methanation should pretty much go to completion unless you are running out of H2 and if you run out of H2 your catalyst will deactivate. What application is this for? Are you concerned about ppb levels of CO?
Typical commercial H2 plant methanators operate at 300 psig and an outlet temperature of 500-600F the spec is usually
Mr. Rick Manner my i am trying to do a thermodynamic analysis on how the presence of different composition CO2 , O2 and H2O will affect the methanation at high pressure.
Even though the concentration of CO in the product is not detectable.
in generaI i want to know whether Data Regression is a good idea for determining missing binary parameters or is there some other way to determine them in an EOS.
I think you will find that Hysys will calculate that a 3000 PSI methanator operating at 600 F outlet temperature will go to completion within the numerical tolerance of the program as long as their is enough H2 in the feed.
There are a few different published values for the steam methane reforming equilibrium. I typically use a regression of Janaf done about 30 years ago. I have found it is more accurate at typical SMR reforming conditions than some of the more widely distributed equations but they aren't different by much. I have attached a spreadsheet with the equilibrium equations that I use.
- Badges
- Science topic