A long while ago, I taught a summer camp class called "How to Walk Through Walls." It was a physics class that culminated in the idea that matter is really made of waves of probability, and these waves could in principle transport through barriers — that is, you could in theory walk, metaphorically speaking, through a wall.
Well, you can't, but a team in Finland has proposed observing the effect, called quantum tunneling, in a mechanical system. That's something no one's done before, and if they succeed, it will be really, really cool.
Read the story here.
Nice.
ReplyDeleteTwo technical problems with the article: both common "errors" in explaining this subject, but neither is necessary. The first one concerns the two-well problem: as you say, the particle classically can not go to the top of the barrier. The entire phenomenon of tunnelling occurs because quantum mechanically it *can*, that is, even if the other well is not present. In other words, qm says you have a non-zero (though small) chance of being *in the wall* any way (similar to evanescent waves for light that cannot normally leave a medium because of total internal reflection), no surprise that you can walk through it. Ultimately this arises from the fact that to bound a particle to a region *precisely* needs one to have enough uncertainty in momentum so that it has a finite probability of having enough energy to cross any finite barrier: classically, one may know the energy precisely to be lower than the barrier, in the correctly setup qm problem, one does not know the energy that precisely.
Second problem about why *you* can't cross the wall actually. In addition to saying that the numbers are small because you are big, you have another problem: every time an air molecule or a photon hits you, it pins you for an instant to the same side of the wall as itself; and every air molecule and photon on the other side that you would have to displace after reaching that side pins you to this side for an equal instant. Working out the actual meaning of that "for an instant", this actually drastically reduces the probability (though does not make it zero for any finite interaction rate) of tunnelling from what you would naively calculate. This effect is usually called decoherence, but can also be stated as saying that one can tunnel only when one can do so without the rest of the universe observing the act of tunnelling: observation (i.e., interaction or non-interaction of certain kinds) `classicalizes' things.
Of course, you can tunnel along with all the air molecules from your side while all the air molecules from the other side leave the exact regions of space free for this to happen: this process is not reduced by decoherence, but is much much smaller due to thermodynamic reasons. There are other similar non-decohering mechanisms, but they are all small for similar reasons.
Thanks for the comments! Those are both good points — of course, one of the challenges of writing a (relatively) short news story is trying to cram all of that in under 600 words and have it be readable by a general audience. You'll notice my editor and I didn't succeed, even given what we did write! It's a good reason to have the 'blog — more room to discuss.
ReplyDelete