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

Gravimetric Separation of Air

  1. Oct 3, 2012 #1
    Is it possible to separate out the oxygen from atmospheric air by centrifugal force. I am not asking theoretically but practically. In other words is there any device that will act as a gas centrifuge which can separate air molecules in real time? A continuous process as opposed to a batch method.
    Last edited: Oct 3, 2012
  2. jcsd
  3. Oct 3, 2012 #2


    User Avatar

    Nothing that would be practical, no. The required accelerations are simply too high.
  4. Oct 4, 2012 #3
    cji, I tend to agree with you on conventional approaches but I like to think outside the box. For example, what about a nano-vortex? The laws of physics change with scale, what isn't possible in the macro world might be possible in the micro world. Perhaps we could imagine a MEMS (Micro Electric Mechanical Systems) device that would suck in air and form a microscopic tornado, maybe then the molecules would behave differently?

    The reason I ask is I have this idea about an air breathing rocket propulsion system. In order for it to work you have to somehow get the oxygen out of the atmosphere without all the nitrogen, water, and CO2. The method of fractionating the air would have to be very efficient to serve this purpose. If anyone is interested in this idea you can look at the attachment.

    Remember, it's just a concept. Which isn't even an idea. I suspect the complexity of it is too much toever be reduced to practice. But you can't blame me for thinking about it :^) Physics is fun.

    Attached Files:

  5. Oct 4, 2012 #4


    User Avatar
    2017 Award

    Staff: Mentor

    The relevant quantity for the separation power is the kinetic energy of the outer molecules (from the rotation), multiplied with the relative mass difference, relative to their thermal energy. To get a noticable separation, this value should be of the order of 1. For air at room temperature, this needs a velocity of >1km/s or >3 times the speed of sound. The size of the system does not matter, but 1km/s in a microscopic device does not look feasible.
    Cooling is easier - the different components will condense at different temperatures.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook