Jump to content



Featured Articles

Check out the latest featured articles.

File Library

Check out the latest downloads available in the File Library.

New Article

Product Viscosity vs. Shear

Featured File

Vertical Tank Selection

New Blog Entry

Gasoline Blending- posted in Ankur's blog

1

Simulation Of Dense Natural Gas

hysys simulation dense gas envelope

5 replies to this topic
Share this topic:
| More

#1 akar

akar

    Brand New Member

  • Members
  • 8 posts

Posted 06 April 2018 - 11:23 AM

While I was performing HYSYS simulation for dense phase  methane rich gas of following composition I could notice some results that contradicting my understanding as explained below

 

Composition of natural gas

 

Case1:

 

Methane- 0.8970

Ethane- .0950

Nitrogen- 0.080

Pressure: 1900 psig

Temperature-55F

 

Case 2:

Methane- 0.8955

Ethane- .0948

Nitrogen- 0.080

Water- 0.0016

Pressure: 1900 psig

Temperature-55F

 

When simulated

The following properties obtained same for both cases:

 

Critical Pressure- 827.8psig

Critical Temperature- -83.55F

Cricondenbar- 836.3psig

Cricondentherm- -76.4F

 

My questions are:

1.I observed that Even though the properties and phase envelope suggests that its a dense gas operating well above cricondenbar and cricondentherm ,  I found at the operating pressure(1900 psig) gas becomes liquid at -approximately -34F in both cases and in case2 where water present(critical properties are still same! ) the water starts to drop out even at approximately +125F.(by trial and error method)

 

My understanding from the envelope is,  above criconedtherm and cricondenbar the gas should not be  liquidized. Butt using HYSYS((by trialand error method) stil I coud find temperature above cricondentherm where gas becomes liquid which contradicts my genral undrstanding.

Now I dont know where I went wrong in analysing!

 

2.In both cases critical properties are still same! but  I assume that this is because only small change in composition. Am I correct?

I request expert to show some light on this regard

Thanks in advance.

 

 

Methane- 0.8970



#2 serra

serra

    Gold Member

  • Members
  • 292 posts

Posted 06 April 2018 - 12:00 PM

calculated properties depend from model and I am not sure about the model you use,

for example the GERG 2008 model (in Prode Properties) predicts slighty different values

 

mixture 1 (molar fractions) C1 0.897 C2 0.095 N2 0.008

Pc 832.458 psig
Tc -83.413 F
CP 839.634 psig
CT -76.231 F

mixture 2 (molar fractions) C1 0.8955 C2 0.0949 N2 0.008 H2O 0.0016
for this mixture GERG 2008 predicts a separate water phase (vapor-liquid-liquid phase envelope) so I report only critical properties

Pc 840.043 psig
Tc -82.557 F

 

about your last question, for dense phase the software has to identify a state (gas or liquid)  in order to calculate transport properties (viscosity etc.) ,  with a cubic EOS there is only one root and you know density but other properties must be calculated,

Prode Properties gives different options to detect the state based on Gibbs Energy, isothermal compressibility and liquid density, 

probably Aspen Hysys has similar options...



#3 Bobby Strain

Bobby Strain

    Gold Member

  • Members
  • 2,393 posts

Posted 06 April 2018 - 07:51 PM

With water present the fluid will form hydrate at fairly high temperature.

 

Bobby



#4 akar

akar

    Brand New Member

  • Members
  • 8 posts

Posted 07 April 2018 - 03:10 AM

calculated properties depend from model and I am not sure about the model you use,

for example the GERG 2008 model (in Prode Properties) predicts slighty different values

 

mixture 1 (molar fractions) C1 0.897 C2 0.095 N2 0.008

Pc 832.458 psig
Tc -83.413 F
CP 839.634 psig
CT -76.231 F

mixture 2 (molar fractions) C1 0.8955 C2 0.0949 N2 0.008 H2O 0.0016
for this mixture GERG 2008 predicts a separate water phase (vapor-liquid-liquid phase envelope) so I report only critical properties

Pc 840.043 psig
Tc -82.557 F

 

about your last question, for dense phase the software has to identify a state (gas or liquid)  in order to calculate transport properties (viscosity etc.) ,  with a cubic EOS there is only one root and you know density but other properties must be calculated,

Prode Properties gives different options to detect the state based on Gibbs Energy, isothermal compressibility and liquid density, 

probably Aspen Hysys has similar options...

Thank you for your valuable reply.

Since you got different critical properties for two cases, I believe the same value I got  because of the model I used. In Aspen HYSYS, for Oil and gas we normally use EOS  Peng robinson or SRK . I used peng robinson in this case.

Can any body suggest best EOS to be used in the case of dense gas , the gas normally used for lift gas and gas injection?

My intention is to prove the dense gas which above cricondenbar and cricondentherm will not condensate.

 

Thanks 

Abdul



#5 serra

serra

    Gold Member

  • Members
  • 292 posts

Posted 07 April 2018 - 06:13 AM   Best Answer

actually GERG 2008 is probably the most accurate model for natural gas mixtures,

anyway you can adopt a EOS such as Peng Robinson, Soave Redlich Kwong etc.

but take care that with water (your second example) specific models and options can be required...

 

To answer your question, outside  two-phase area (meaning otside bubble line ... dew line)

the software should return a single phase state (liquid, vapor , ...)

depending from the software (I use PRODE PROPERTIES)  you can draw Vapor Liquid Liquid phase diagrams,

hydrate formation curves (as suggested by Bobby etc.) but above cricoP you shouldn't find vapor-liquid

(meaning two phases at equilibria  or xv*fgv=xl*fgl) ,

above cricoP the software can detect a (pseudo) liquid state or a (pseudo) vapor state but it's a single phase, not two phases at equilibria...

 

there are threads at cheresources discussing the point...



#6 akar

akar

    Brand New Member

  • Members
  • 8 posts

Posted 11 April 2018 - 11:57 PM

Thank you all






Similar Topics