It strikes me that the following experiment could be a good one for high
schoolers or college students (or even graduate students, with the requisite
complications) -- or me!
===========================
For water on a "clean" and "smooth" silver surface (both terms of which are a
bit tricky to define on small scales), the liquid-solid-air contact angle at 25
C is close enough to pi/2 radians that the surface tension is pretty close to
zero. For materials with air-water-solid contact angles of less than pi/2
radians (that is, surfaces which are hydrophobic), air bubbles which form on
the surfaces are "stuck" (the water is not "pulled" to the bare surface; if it
were, it would replace the bubble, and the bubble would preferentially pop off
of the surface).
For CO2 at STP and water, the situation doesn't change a whole lot.
I mention this because I suspect that nucleation (rather than blowing a bubble
underwater, at a surface) is probably the "gentlest" way of creating a bubble,
so long as the nucleation rate isn't drastic.
I'm thinking that the rate at which bubbles lift off from the horizontal
underwater surfaces (especially hydrophobic ones), and their maximum sizes
before letting go, may have some influence on our discussions about buoyancy.
Rightly, the experimentalists should be calling foul and claiming that surface
defects and impurities are going to have a large effect on the results.
Probably so. Somehow, I'm envisioning a relatively clean, smooth, Teflon
surface (contact angle awfully close to pi radians -- a very hydrophobic
surface) and a slow nucleation rate (so that a bubble, once it forms, can
"clear" water from between it and the surface; in other words, the surrounding
water can, through surface tension, suck its brethren molecules from under the
bubble for very small bubbles). Such a slow rate would also reduce the effects
of viscosities.
As (if!) the bubble reaches 2mm in diameter or more, the effects of surface
tensions will go down, because 2mm is about the capillary length for air-water
systems. At about that length/curvature scale, things tend to fall through water
surfaces if they're more dense than water (water striders' legs are much smaller
in radius than 2mm; paperclips are on that order of size; even coins which can
be made to "float" on tension-dominated surfaces deform the surface so that its
radius of curvature is no more than 2mm). It may also be possible to add
surfactants to the water so that such surface tension effects are minimized.
Would not such an experiment meet our criteria?
*Nominally buoyant gas, mostly uncluttered by issues of thixotropism or
viscosity and "stickiness" of the suction cup edges.
*Gas is (at least for a while) simply present at an underwater surface.
*So long as the amount of water between the bubble and the solid surface is
minimized, will the bubble lift off?
/**************************************
"The four points of the compass be logic, knowledge, wisdom and the unknown.
Some do bow in that final direction. Others advance upon it. To bow before the
one is to lose sight of the three. I may submit to the unknown, but never to the
unknowable." ~~Roger Zelazny, in "Lord of Light"
***************************************/