Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Inside a container with non-pure helium

  1. Jul 30, 2012 #1
    Is helium miscible with O2, N2 or any of the other gases in air?
    If a container were filled with 95% helium and 5% O2, would the O2 gas settle to the very bottom of the container, leaving very pure helium gas at the top?
     
  2. jcsd
  3. Jul 31, 2012 #2
    The helium won't react with the O2 or N2. The helium would mostly settle to the top of the container, but I'm not sure about the purity. You'd probably get a small amount of O2 and N2 along with any sample you pulled out.
     
  4. Jul 31, 2012 #3

    krd

    User Avatar


    No, they would mix. Think of it this way, the earth's atmospheric gases do not separate into layers according to relative weights of each gas.

    You can remove the oxygen and nitrogen by reducing the temperature and liquefying them. It's a form of distillation.
     
  5. Jul 31, 2012 #4

    mfb

    User Avatar
    2016 Award

    Staff: Mentor

    If you fill your room carefully, you can get unmixed gases for a while. But if you wait long enough (or have enough sources of turbulence), you will get a mixture of those gases, with slightly more helium at the top.
     
  6. Jul 31, 2012 #5
    Actually, the atmosphere is separated into layers according to the relative weights of gasses to some degree. Each layer of the atmosphere (mesosphere, stratosphere etc) has a different composition of gasses. One reason that there's a lot of gas that doesn't take its expected position in the atmosphere is because there are heavier molecules that the gas molecules can react with and attach themselves to. In the case of a container with only helium, nitrogen and oxygen, there are no reactions happening to form heavier molecules, so they should separate into layers. You can demonstrate the ability of fluids to separate into layers with some vegetable oil and water. Put them in a container and mix them together, and they'll separate again into layers, because the oil and water don't react well and they have different densities.
     
    Last edited: Jul 31, 2012
  7. Jul 31, 2012 #6
    I'm not sure that is true? Ethanol and water have very different densities but they mix completely. Liquid mixing is totally different to gaseous mixing because of the intermolecular bonding in liquids.

    I think the gases will partially stratify, but they will certainly not be ordered into perfect layers.
     
  8. Jul 31, 2012 #7
    Vegetable oil isn't ethanol, that's alcohol. I agree that the gases won't make perfect layers like the liquids do because they lack the intermolecular forces, but I can show you a demonstration that allows you to visualize a similar layered separation of gases.

    http://youtu.be/ckaJs_u2U_A
    This is quite a heavy gas, but the ratio of weight between the SF6 in that video, and air molecules is smaller than that between He, and O2, or He and N2. This suggests that the He would separate from the other gases in the container even better than the SF6 in this video separates from the air.
    As for the O2 and N2 mixing or separating, they might be close enough in molecular weight that they don't separate well.
     
  9. Jul 31, 2012 #8

    mfb

    User Avatar
    2016 Award

    Staff: Mentor

    The SF6 does not separate there, it gets injected at the bottom of the container as carefully as possible. If you watch the "sinking" around 3:50 carefully, you can see a boundary layer, where the SF6 concentration gradually increases towards the bottom. Filling more SF6 causes the boat to go down significantly, but only at 3:58 it is fully within the SF6 and sinks quickly.

    The atmosphere has a height of >10km, do not expect the same layers in a small room.
     
  10. Jul 31, 2012 #9

    krd

    User Avatar

    But gases don't behave as fluids. And you will find heavier molecules in the upper atmosphere.

    Gases are more free to move around so they will mix. And there isn't much of anything in the upper atmosphere, all gases are densest near the ground.

    I'm pretty sure thermal equilibrium would also cause the gases to mix.
     
  11. Jul 31, 2012 #10
    But when a volume of helium is released into the open, all of it is pushed up by the buoyancy of the surrounding air. And in that case, only helium rises, no oxygen or nitrogen mixes with the helium and rises with it.
    Wouldn't the same happen in the smaller container? The 95% helium would be displaced upwards by the buoyant force by the 5% of oxygen in the tank.

    And also, since helium rises only up to the point where it is no longer affected by the buoyant forces, does that mean that there is a slight layer of helium in the upper upper atmosphere? I don't think a helium atom would have acquired enough kinetic energy by then just from buoyant forces to have an escape kinetic energy.
     
  12. Jul 31, 2012 #11

    mfb

    User Avatar
    2016 Award

    Staff: Mentor

    If you release it carefully, yes. If not, it mixes.

    Helium atoms have a typical thermal velocity of maybe ~2km/s, and the escape velocity is ~11km/s. e^(-5.5) is still significant, therefore some helium atoms can escape.
     
  13. Jul 31, 2012 #12
    Gases do behave as fluids; Gases are fluids.
     
  14. Aug 1, 2012 #13

    krd

    User Avatar

    Come on, you're really stretching it. They are two different states. They may have characteristics that overlap, but they are not the same thing.

    My understanding of the states

    Plasma: Highly energised particles explosively trying to escape in all directions. Like a gas, but simultaneously unlike a gas.

    Gaseous: Energised particles with a great deal of free movement. They can exhibit pseudo-fluid characteristics, like the laminar flow of smoke from a cigarette. In the laminar flow, the particles can be considered to be weakly bound (they are not in fact bound) At lower temperatures/particle momentum, the molecules will become bound by Van der Waal forces and condense into liquid state.

    Liquid: Lower energy/momentum particles. With freedom of movement, but held together by a stickiness of Van der Waal forces on some particles. (Some particles are stuck together others are free moving - the state is maintained by particles being bound and then knocked free from their bounds by other particles - the state is maintained as long as their is enough momentum in the particles to keep breaking the Van der Waal bonds - for the liquid to become a gas, the VdW bonds must be completely overwhelmed).

    Colloid : A state between liquid and solid. Not exhibited by all materials. Custard is a colloid. Colloids have some of the characteristics of both solids and liquids, but not others.

    Solid : Lowest energy. When the net Van der Waal forces in a liquid overcome the momentum of unbound particles, crystal nucleation of large groups of particles is allowed. If the crystallisation continues, particles are left with little freedom of movement - they are fixed in position and can only vibrate.

    The thing is, all the states are just particles at different net energies - it's all the same thing, except it isn't. It's really how you want to define it. Deuterium Oxide and briny water will sink to the bottom of the ocean, but it does not mean you will not find salt and deuterium near the surface.
     
  15. Aug 1, 2012 #14

    krd

    User Avatar

    No. think about what's happening on an atomic scale. When you release the helium and the atoms start to climb, they will collide with oxygen and nitrogen, giving them momentum to fly off. And even recoil downwards.

    The reason a helium filled balloon is buoyant is due to the Archimedes principle. The net effect of the helium in the balloon is it's less dense then the air. You can treat the balloon as a single particle - the net effect of all the free air collisions on the outside of the balloon is to give it upwards momentum. If it was a heavier than air gas, the net effect would be to give the balloon a downward momentum.
     
  16. Aug 1, 2012 #15

    sophiecentaur

    User Avatar
    Science Advisor
    Gold Member

    It's common for people to confuse the terms 'liquid' and 'fluid'. Another one that you frequently hear is when people talk of 'opaque glass' when they mean 'obscure glass'.

    It 'literally' drives me mad!!!! :eek:
     
  17. Aug 1, 2012 #16

    krd

    User Avatar

    You're better off not to confuse people. A gas can be a fluid because it flows. But when people say fluid they generally mean a liquid. Gases can have fluid characteristics but they are not liquids. But you could put that the other way, liquids share characteristics with gases - but that doesn't make them gases.

    A colloid is a fluid. Because it can flow, but it is neither a solid, a liquid, nor a gas. Generally, referring to a colloid as a fluid, in context where the specific characteristics of the colloid are important, is incorrect - though it may be right in the literal sense, it's the wrong thing to do.

    Plus, playing Gotcha!!! is just being a smart arse. No one calls a gas a fluid outside of specific context, unless they're playing Gotcha!!!.
     
  18. Aug 1, 2012 #17

    sophiecentaur

    User Avatar
    Science Advisor
    Gold Member

    I take your point to some extent but PF is surely supposed to be more than just "people" and it's not the world of 'Pub Chat'. We are surely supposed to be discussing and explaining things in formal terms and trying to avoid confusion. When a word is used here, it should be in its properly defined sense or where would we be in discussing a topic like Electricity, for instance.

    Archimedes' Principle refers to 'a fluid' and it works for hot air balloons as well as for boats. The "specific context" on PF is Physics. It isn't playing 'gotcha' here; it's spelling things out to avoid misconceptions and misunderstandings. I wouldn't pick holes in an instruction over a hospital bed which said "Check fluid intake" because we would all expect the patient to continue to breathe, unmonitored. That's a different world.
     
  19. Aug 1, 2012 #18
    The problem is people are blanketing two separate things under one definition, then using that definition to infer that the two things have the same characteristics. Gases and liquids act very differently- calling them fluids is only useful if you can remember that.
     
  20. Aug 1, 2012 #19

    sophiecentaur

    User Avatar
    Science Advisor
    Gold Member

    If "people" used the relevant terms then there would be no problem. I have just been pointing out that it is not just being 'posy' to ask for the right terms to be used.
     
  21. Aug 1, 2012 #20

    krd

    User Avatar

    Yes it does refer to fluid. But Archimedes would not have been aware of why it was happening. In the context of what is happening, the fluid characteristic is unimportant - the Archimedes principle will work in a gas or a liquid. Flow is not really something you have to think about - more the unbound nature of the particles in gases and liquids, gives a clearer idea. Flow would muddy the water.

    Archimedes was aware that displacement of volume led to buoyancy. But he would have been unaware, that there is a connection with thermodynamics. Fluids and flow, were a real puzzle. Even in early classical thermodynamics, heat was thought of as a fluid.

    Fluid behaviour can be deceptive. A flow - a laminar flow - can look so smooth from a distance. But if you get really close you can see it isn't. And it's the same with the chaotic swirls of turbulence - from a certain distance they look smooth - you get in really close and they're not. The really complicated nature of fluids emerges from something much simpler.
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook