Stratification of gases (can it be?)

[CostasV]

Hello everybody!

This is a discussion that me and a friend of mine have about the effect of stratification in gases. So the question is : What is the effect of gravity in a closed volume of a mixture of gases such as methane= 95%, ethane=3%, propane=0.5% and butane=0.1% and the rest heavier HC. The pressure of the closed (and undisturbed) volume is 100 bar and the temperature is 20 deg C. This volume is kept undisturbed for almost 4-5 years.

I think that gases are not stratificated at all while my friend thinks that gas do stratificate and the more time passes the more they are stratificated.

Note 1: we are talking about gases well above the liquification point

I would like to know if there would be remarkable relative ((xupper-xlower)/xlower) difference (more than 1%) in the concentration of each gas regarding the lower and the upper level of the volume and if this difference is caused by gravity.

Note 2: the sample volume is 5 ml while the total (closed) volume is 50 l.

Thanks for any contribution!

Costas

 

[jte]

Costas-

Thats what a refinery's all about. I don't know about the specifics (I'm a Mechanical Engineer) as far as the liquid C3-C5's stratifying. Post the question to the Petroleum Refining Engineering (under Petroleum Engineers) forum and you'll get plenty of input from ChemE's.

jt

 

[zdas04]

The effect of gravity is so small in a gas-filled vessel and the gases are so energetic that over time the gas in a "static" vessel will tend towards more homogeneity rather than less.

There are cases where gases of different velocities will retain their character for long distances (e.g. if you blow a sonic stream into a static volume, there will not be any mixing until the gas slows below 0.6M and mixing will be limited till the gas slows to laminar rates). If you remove bulk-velocity effects, there will not be any stratification.

The gas-on-gas separation you see in gas-processing plants are some major actions applied to the gas. Soaking it in water then contacting Amine, glycol dehydradation, removing heat to condense gases. None of the processes on the gas side rely on gravity to separate gases.

David Simpson, PE
MuleShoe Engineering

http://www.muleshoe-eng.com

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[25362]

At first sight I'd tend to agree with zdas04, because non-polar gases wouldn't tend to stick together due to attractive London forces unless they are very near their condensation (dew) point.

In fact, when a gas such as H2S (with rotten eggs odour) is released in a room, quick diffusion would enable to smell it throughout the room in a very short time. True, the molecular weights of H2S and air are similar.

It is also assumed that no effusion through the pores of the container walls over the years would change the composition of the mixture depleting the amount of methane, since pressure stays constant.

Notwithstanding, I've read some years ago that particles (i.e., molecules) of heavier and larger sizes submerged in a sea of lighter (in this case methane) molecules would tend to be displaced by multiple collisions towards the wall of the container, over a long time, to reach a state of equilibrium maximizing the entropy of the system.

As if in a ballroom, those who do not dance collect at the periphery to enable dancing couples to perform on the floor.

Whether gravity enhances the said effect, or not, would depend on the geometry of the container, wouldn't it ?

I don't know of any method to quantify the results, except probably by careful sampling and measuring concentrations.

 

[CostasV]

Hello again all

zdas04 and 25362- thank you for your replies

I understand that you agree with me that there would be no stratification effects between gases of different density.

Then, we should have to give a documented answer to the question : why the light gas detectors (CH4, H2) are located on the higher levels of the rooms, and the heavy gas detectors (LPG in gaseous form, CO2) are located in the lower levels?

In my opinion, the answer is : Because, at the first moments of the gas eneting the room, around the entrance/leak point the concentration of the gas is near 100% and the volume of the gas entering behaves like a lighter (or heavier) object, so it goes up (or down). But as time passes particles of the gas are distributed (altough I need to clarify if the distribution is equal to all the available space) in all the available space (as this is by definition what characterises the gas). So as time passes the mixture will be more homogenous and the stratification that was present for the initial time will be reduced.

Please correct me if I am wrong.

Costas

 

[unclesyd]

When we used to mix our own welding gases and some in house calibration standards for instruments group the standard practice was always to set the cylinder in a heater to mix the gases. The heating time per cylinder was fairly long. I don't know where this cylinder heater can from but it worked.

Anecdotal:
About 1995 the analytical department was having trouble with some of their in house calibration gases and I suggested that the bottles should be heated to mix the gases. Damn near thrown out of the room. One of the PhD's present went to the lab and setup a cylinder and directed a heat lamp near the bottom of the cylinder, this was essentially the way our old cylinder heater worked. The lamp was left on over the weekend and on Monday the gas was checked and I was told that the variations in analysis of the gas dropped essentially to zero. Some time later I was told that variability of the mixed gases was beyond the precision of the analytical instrument.

 

[iainuts]

Costas: you said: "Then, we should have to give a documented answer to the question : why the light gas detectors (CH4, H2) are located on the higher levels of the rooms, and the heavy gas detectors (LPG in gaseous form, CO2) are located in the lower levels?

In my opinion, the answer is : Because, at the first moments of the gas eneting the room, around the entrance/leak point the concentration of the gas is near 100% and the volume of the gas entering behaves like a lighter (or heavier) object, so it goes up (or down). But as time passes particles of the gas are distributed ..."

That's absolutely correct. The forces on a helium filled balloon help to visualize this mechanism. The forces on a balloon are the pressure on the bottom pushing up, versus the pressure on the top pushing down. (air pushing on the sides can be neglected as it cancels) Because air has a very slight difference in pressure between the top and bottom of the balloon, calculated by rho*g*h, the pressure difference pushes the baloon up if it is lighter than the pressure difference. Note that this model is mathematically identical to the more conventional displacement of mass models where we calculate the force by comparing the difference in weight between the baloon and displaced air.

Going back to your example of a gas coming out of an opening, it doesn't diffuse immediately, so the concentration of lighter molecules acts similar to a balloon. The 'skin' of the baloon is simply the fact that molecule collisions on the periphery of the gas cloud prevent air from forcing the cloud to diffuse immediately. Thus there's a fuzzy or hazy 'skin' around the cloud of lighter gas that acts similar to the rubber balloon.

Also, I agree with the other conclusion, that once the gas is diffused, there is little if any tendancy for lighter molecules to rise. I'm not aware of any mechanism that will separate gas molecules like this (ie: well above liquifaction temp).

 

[25362]

No comments on the gas segregation due to "entropic" forces I mentioned reading elsewhere, that seem to contradict Fick's laws of diffusion ?

 

[CostasV]

25362 – I am searching in google to find something relative to “ballroom” phenomenon (keywords ‘gas-in-gas’, ‘solubility”), but I haven’t found anything yet.

If this actually happens, then, for example, a pyramid container would concentrate more of the heavier gas in the bottom, where a cylinder (or any other upside-down symmetrical shape) not.

But does this ‘ballroom’ phenomenon happens?

 

[25362]

I think I read about that effect some years ago in the Science magazine.

 

[25362]

The article I mentioned didn't refer precisely to gas mixtures, it seems I extrapolated its conclusions to a mix of gases with widely different molecular sizes.

It is titled: Gentle Forces of Entropy Bridges Disciplines and appeared as research news on Physics, on p. 1849 of Science Vol 279, 20 March 1998, by David Kestenbaum. Worth reading.