Comparison of Helium to Ammonia molecule size 4

[59133]

If I had a leak in a sealed vessel that was found using a Helium mass spectrometer leak checker and the leak rate was 2x10 to the -8 scale, what would the comparison be of the leak rate if the gas was Ammonia? Can the molecule size dictate the leak rate and be directly compaired accordingly?

 

[Rich2001]

At a leak rate less than 10^-7 cc/s the flow mode is moclecular. With molecular flow, low molecular weight increases gas flow.

There is a relationship between molecular size, MW, pressure and leakage rate.

 

[kenvlach]

You can estimate from molecular weight using Graham’s Law of Diffusion.
“The rates at which gases diffuse under the same conditions of temperature and pressure are inversely proportional to the square root of their densities:

r₁/r₂ = (&[ignore]rho[/ignore];₂/&[ignore]rho[/ignore];₁)^1/2

Since &[ignore]rho[/ignore]; = MP/RT for an ideal gas, it follows that

r₁/r₂ = (M₂/M₁)^1/2

from Lange’s Handbook of Chemistry, 14th Edn. p. 5.157 (1992).

Substituting, r_He/r_NH3 = (17/4)^1/2 = 2.06

For equal pressure, the NH₃ leak rate is estimated as 9.7 x 10^-9 cc/s

I would not rely heavily upon estimates from this rule, especially if you have permeation through a gasket rather than a leak. Gas permeability data for polymers and rubber (ibid. pp. 10.66-10.69) indicate orders of magnitude deviations can occur in either direction. Polarity plays a major role (even more so than size or weight), as H₂O permeates through cellophane, cellulose nitrate, Nylon 6, silicone rubber and PVC much faster than do He, N₂ or O₂. Agreement with Graham’s Law is better for gas permeation through PTFE, polytrifluoroethylene and polyvinylidene chloride.

 

[59133]

Thanks kenvlach for the info.
More specific to what I am looking for is this. The vessel is an all welded assembly that will have Ammonia internally. We check the vessels integrety with a Helium mass spectrometer leak checker. Is there a chart that shows the direct comparison of Helium to Ammonia that states if the leak rate calibration is 1.5 x 10 to the minus 7 with Helium, what would the leak rate be with this calibration number if it were Ammonia. The issue is that how long will it take a vessel charged with Ammonia to lose its charge with this calibrated leak rate? Unfortunately I have little to no experience or knowledge with Molecular science.
Thanks again and looking forward to your response.

 

[59133]

There is one more piece of information I neglected to add. The ammonia being in a welded vessel at ambient room temp and the delta P from inside the vessel to the outside, there is a positive pressure inside. How much of an effect does this have on the leak rate comparison?

 

[kenvlach]

For an orifice-type calibrated leak device (not a gasketed vacuum system) at flow rates low enough to be in the molecular flow regime (as pointed out by Rich2001; see flow ranges at

http://www.amgas.com/ldrefpage.htm),

Graham’s Law of Diffusion appears to give a valid comparison. As supporting evidence, Footnote 5 from a table at

http://www.vacuuminst.com/sub/conversionchart.pdf

states “Generally helium flow through small leaks, 1 x 10^-6 std-cc/sec or less, is 2.73 times the air leakage rate.”

Leakage rates are normally standardized for a 1 atm pressure differential. If leaking into a hard vacuum, I expect flow through the orifice directly proportional to the pressure. The instantaneous flow and remaining mass can be calculated from initial conditions via some simple calculus. If leaking into ambient air, counter diffusion of air would need to be modeled when the pressure nears ambient. Perhaps a porosity model of diffusion based upon the geometry of the orifice device??

 

[59133]

kenvlach,
Thank you once again for the tremendous help!
Have a great weekend!