MDEA Absorber using Hydroxy Ethyl Piperazine (HEP) as promoter- HYSYS simulation problems

[chengg29]

Hi all,

I have recently simulated an AGRU design on HYSYS Ver. 8.8 using Acid Gas package as a thermodynamic package and using the Acid Gas template provided as part of HYSYS Ver 8.8 simulator package. This AGRU plant design utilizes HEP, instead of Piperazine, as a promoter to MDEA solvent in the Absorber column. However if I replace Piperazine with HEP in the simulation, HYSYS is unable to initialize the thermodynamic package and "crashes out". Nevertheless I used Piperazine, instead of HEP in the simulation while keeping all other plant design parameters the same and obtained a converged solution with the desired results.
I also did one simulation from "scratch" for the same plant design (HYSYS Acid GAS template is "ready-made" with all the unit operations in-built including a property table). I used Glycol Package as a thermodynbamic package for this instance of simulation since the NRTL package threw up a warning stating that the binary coefficients for some of the components was unknown and hence it was setting them zero for those compoenents.

My queries are:

a) Is it acceptable, design-wise, to substitute HEP with Piperazine, assuming that both these compounds are related to each other / derivatives, and accept the results of the simulation for the concerned plant design exercise?

b) Would it be appropriate to use the Glycol Package as the thermodynamic package when simulating the AGRU in case the Acid Gas package template is not used in HYSYS Ver. 8.8?


Many thanks in advance

 

[georgeverghese]

When you look at the chemical structure of N-hydroxy ethyl piperazine, attachment of an acidic species such as CO2 to the electron pair on one of the base N sites is hindered by the ethanol group, so you can expect both pKb to be lower than for piperazine and reaction rate to be slower also. Morever, the hydroxy group brings back all the undesirable side reactions of the more traditional amines DEA and MDEA ie formation of higher mol weight oxo compounds and other heat stable compounds.

From my experience with glycol package simulations with the NRTL package / EOS on Pro II - Simsci, you have to remove all the heavy hydrocarbon species in the feed gas, including pseudo components, and replace with an equivalant amount of hexane. Hydrocarbons heavier than n-hexane and pseudo components are not supported in the NRTL package for glycol simulations in Pro II. (Also check which aromatic species is supported). Check the simulator reference manuals for Hysis for similar limitations that may be causing these run time error messages. Replacement of these heavy HCs' with hexane in the feedgas slate should be okay , since it doesnt affect water content in the feedgas; though I cannot tell if co absorption of hexane into TEG is any different to that for other alkanes. BTEX co absorption for aromatics heavier than xylenes may be different, but then the composition contribution of these is probably very small.

 

[chengg29]

Thanks for your response

Whatever little I know about HEP is that it has a Hydrogen atom attached to it as the primary substituent which makes it easy for the CO2 molecule to be reacted to form the bicaronate ion. Correspondingly the dissociation constsnt should be comparable to that of the generic Piperazine molecule.

I take note of you have said about your experience on using NRTL on Pro II and pseudo-components. However the gas composition that I am using for the simulation does not have any pseudo components (at least whatever I received from my sub-surafce team does not have it). I could not find much assistance on the runtime errors in Hysys manuals. This, I reckon, leaves me with the only option to search the literature for vapor pressure / molar volume data for HEP and use regression techniques to generate BIP's.

 

[georgeverghese]

Rather the opposite, the H atom in HEP hinders the attachment of the CO2 molecule to the donor electron pair on nitrogen. This will mean slower kinetics, and lower equilibrium pKb also. The same effect is seen when comparing the reactivity of DEA and MDEA for CO2.

 

[chengg29]

Hmmmm....interesting. Then I am wondering why the AGRU plant's licensor in our NGL facility suggested HEP as a promoting agent to the MDEA solution that he proposed for removal of CO2 and H2S i.e. bringing their content down to ppm levels in the treated gas. I will try to find out why.

Thanks Georgeverghese

 

[georgeverghese]

The activated MDEA that BASF marketed first some 20years ago uses piperazine as a promoter (and is in common use now in many gas processing plants), and you can see why they chose this rather than some other chemical like HEP - there is no steric hindrance to the lone electron pairs on the 2 cyclic nitrogen atoms for electrophiles such as CO2 or H2S.