Joule Thompson coefficient for Natural Gas

[fefelvn]

I am trying to locate the Joule Thompson coefficient for natural gas or the Beattie-Bridgeman constants for naturaal gas so that I can calculate it. Does anyone know of a good resource?

 

[25362]

If this may help, I happened to see that Perry VI has the approximate inversion curves with reduced coordinates for all gases, and that for methane. See tables 3-150 and 3-158.

 

[25362]

There is an article from Google:

http://www.ipt.ntnu.no/~jsg/studenter/prosjekt/Azad2005.pdf

that might be of help.

 

[davefitz]

I thought most people use the approximation of 7F temp drop for every 100 psid of pressure drop.

 

[UmeshMathur]

fefelvn:

Natural gas is generally a mixture of several components, mostly methane but also containing ethane, propane and at least some butanes. There would be some computational error if you assume that constants for pure methane apply to such a mixture.

To answer your specific query, I found Beattie-Bridgeman constants for methane in Professor S. M. Walas' excellent book: "Phase Equilibria in Chemical Engineering" (Butterworth, 1985) on page 598:

A0=2.2769, a=0.01855, B0=0.05587, b=-0.01587, C=128300. Here, the units are P[atm], V[liter/gmol], T[K], and R=0.08206[liter.atm/gmol/K]

For the other hydrocarbons, you would need to back-calculate the Beattie-Bridgeman constants from published P-V-T data.

However, I would suggest that you look instead at using a cubic equation of state (Peng-Robinson or Soave-Redlich-Kwong). Either of these can be used to do the computation of the J-T coefficient accurately for light hydrocarbon systems. Both options require pure-component properties (Tc, Pc, omega, and ideal gas enthalpy coefficients) plus readily available binary interaction parameters. When dealing with a single phase, all binary interaction parameters can be set to zero without significant loss of accuracy (these are normally used only in phase equilibrium computations).

Finally, most modern process simulators do isenthalpic flash calculations very efficeintly, so you can compute the required J-T coefficients very quickly using any of them.

 

[Rocco8]

I use the process simulator ChemCad from Chemstations. It also predicts hydrate formation.

 

[MortenA]

davefitz

The one you mention (equal to 1/2 deg C/bar) is valid for "normal" pressures and temperatures. But JTC may be other values - it may even be negative (that equal a temperature increase across a valve)!

Best regards

Morten