Physics Van 3-site Navigational Menu

Physics Van Navigational Menu

Q & A: pressure from liquids

Learn more physics!

Most recent answer: 12/27/2011
Please forgive my ignorance, but I am having trouble understanding what heat is, and what causes a gas to exert pressure on the walls of a vessel? If the answer is that pressure is the result of the kinetic energy of the sum of all the collisions per unit time of gas molecules impacting the vessel wall, and the higher the temperature of the gas the more frequent and energetic these collision due to increased molecular velocity, then why can't we adiabaticlly condense the gas (using pressure) until it changes to a liquid state with a density equal to either the liquid or solid phase, and in so doing, dampen out these velocities such that when the pressure is released, all the molecules remain in a liquid or solid phase? If gases are compressible because of the empty space between molecules, and liquids and solids are nearly incompressible because of the intimate contact (lack of space) between adjacent molecules, shouldn't compressing a gaseous element or compound into a liquid or solid density force all the molecules into contact with one another and destroy their velocity of motion? Yet this does not happen; if we compress a gas adiabatically to a liquid or solid phase, and then release the pressure, the substance immediately expands back into a gas occupying the original volume. Yet if we compress it through this same density increase, but then cool it so that we remove heat and lower its temperature to one appropriate for the liquid or solid phase of the particular substance, then the substance remains in said phase and does not expand to its original volume when the pressure is released. If heat is atomic and molecular "vibrations" within an element or compound, rather than velocity of motion, then why cannot we suppress these vibrations with extreme pressure and effectively cause a phase change without removal of heat? Thank you, Dave
- David Gibbel (age 51)
Albany, OR USA
That's not an ignorant question at all. Long maybe, but not ignorant.There are several aspects to it.

Your picture of how gas molecules exert pressure just by bouncing off the walls of a container is quite good. The energy exchange isn't really relevant. What counts is that on each bounce the particle gives some momentum to the wall, as the particle's own momentum changes direction.

Let's take another aspect: why when you start with a liquid or solid in a snug container and then expand the container do some molecules fly off making some gas phase? There are several ways to see this.

One is that, unless things were cooled to absolute zero temperature, things were still rattling around in (for example) the liquid. The distinction you draw between vibrations and velocities isn't so sharp. Anything vibrating has some velocity, alternating directions.

If there's some space open, occasionally a molecule will get enough energy to break loose and fly off into that space. So those loose molecules are the gas. To totally "dampen out those velocities" means to cool all the way to T=0 K. You can't ever quite do that, and it gets hard to even approach it too closely. So your picture of the condensed phases as lacking motion isn't very accurate.

Under extreme pressure, or even under moderate cooling, you can largely suppress the vibrational energy, for quantum mechanical reasons. (For technical types, I mean you can get the temperature below the Debye temperature, so that the energies fall far below equipartition values.) Nonetheless, so long as there's some thermal energy present (and there always is), a few molecules will fly off into the gas phase, if they're given room.

A more abstract way of seeing the same thing is to start with the deepest principle of thermal physics- that nature explores all the accessible states over time. (That's a sort of informal way of stating the second law of thermodynamics, that total entropy gets maximized.) Failure to form the gas would mean that lots of states were staying unoccupied for no particular reason.

Mike W.

(published on 12/27/2011)

Follow-up on this answer.