Going a little off topic today. But with this audience, I bet you are more likely to be interested than if this were a blog about politics.
Quantum Mechanics has interested me for a long time. If it doesn't interest you, that's probably because you haven't heard about it. Experiments in Quantum Mechanics have results that are so bizarre, so counter intuitive, that if you really understand how they work, you will scratch your head and never see the world around you the same way again. At least, that's the reaction I had since I first learned about the Double Slit Experiment.
If you know what I'm talking about already -- you know what I mean by bizarre. Even if you don't know what I'm talking about, you have probably heard at some point that light has a strange property about it -- sometimes it behaves as if it were a wave (a 'wave' of what? don't ask), and other times it behaves as a particle.
To visualize the difference, think of ripples in a pond. You splash your hands down into the pond at the same time, thrusting them forward as if to send a message to the shore at the other end. This produces two sets of ripples ('waves') that propagate outward toward the shore at the other end. Waves are interesting in that they can interfere with each other. If a crest of one wave meets with the trough of another, they cancel each other out -- the water level at that point would be as if there were no waves at all. So think about the water level when these two ripples of yours reach the shore line. The water level will form an interference pattern. At any given point along the shore line, the water level is affected by both waves you originally caused. There is no information about why each point is at the level that it is -- you don't know how much of each wave was added to the result (a+b=10, what does a equal?).
Now, the pond is frozen, and so instead of splashing your hands down into it, you slide a hockey puck toward the shore at the other end, one in each hand. All you'll get in this case is two hockey pucks at the other end.
So you can easily see the important difference between the two. As a wave, if you slash your hands continuously, the shore line will build up an increasingly strong interference pattern with each passing wave (imagine each time a wave impacted the shore, it built up some of the sand). As a particle, all you'll get is a bunch of hockey pucks in two spots.
The double slit experiment is as follows. Photons are emitted (the 'wave') through one of two slits (our two 'hands') in a barrier -- which slit they go through is 50/50 -- then they impact on a photo-sensitive screen (our 'shore') that shows where they landed. Normally, this produces an interference pattern. This already is a little counter intuitive... you might expect to see two bars of light (our, 'a bunch of hockey pucks in two spots'), not a wavy pattern. But Quantum Mechanics tells us this occurs because the photons emitted from the source are waves, and so the wave from one slit interferes with the other. Indeed if you close one slit, you see a single bar. Open it and close the other slit, and you see another bar a little bit shifted over. Open them both -- bam, interference pattern. Ok -- a little bit weird, but we can live with that... keep reading, it gets a whole lot stranger.
Now, change the photon emitter so that it only emits a single photon at a time. With each pulse we will wait until the photon hits the screen before emitting another, that way nothing could possibly "interfere" with the photon's path, and we should see what we first expected -- two bars of light on the screen. What happens is this -- each photon goes through, individually creating detection spots on the screen (dots), just as if each photon were going through one slot or the other and hitting the screen. But... as we send more and more photons, an interference pattern emerges! It seems that in fact, the single photon as a wave is passing through both slits at the same time and then interfering with itself. "What the @#$@" would be an appropriate reaction here. Keep reading...
Now we place a device in the two slits -- say, just before them. This device tells us whether the photon is passing through it, but without deflecting it or affecting it's path to the screen. This way we can detect which slit the photon is passing through with each iteration. Ok... one photon, two photons, three... the detectors register the slots each pass through... but the interference pattern vanishes! Two bars of light form on the detection screen. All we have added to the system is a detector, and it behaves differently.
Ok -- perhaps we have somehow poisoned the experiment with the detectors. Perhaps they are affecting the photons path in a way we do not understand. So, lets move the detectors after the slits. That way, the photon will go through one slit or the other, just like it would have without the detectors, and then we will detect which one it went through, having confidence that we aren't influencing which one it chooses. Still no interference pattern!
Ok ok already -- perhaps our detector is bogus. Instead, lets reset the setup so we get the interference pattern again -- just two slits and the photo detecting screen. Now what we'll do is this -- after each photon passes through a slit, but before it hits the screen -- the screen on which it will produce the interference pattern -- we will very quickly replace the screen with two telescopes. One is focused on one slit, and the other on the other slit, such that if you were looking into the telescope and the photo went straight through the slit it were focused on, you would see a flash of light. You might wonder how in the world we could be quick enough to do this, light is pretty fast after all (~300,000 kilometers per second). But assume we can -- just make the distance great enough and the apparatus fast enough, it's not a problem. The setup would be just as described here, or just look at the first diagram.
Now there is no way we could be affecting the experiment via our observation, right? All we are doing is changing how we observe the photons. What do you think we get... if it were interference pattern, we should fail to see some of the photons, because they would land somewhere outside the telescope's detection area. Nope -- every photon is accounted for. We in fact no longer see an interference pattern.
So what you see here is the fundamentally strange property of quantum mechanics. It seems that particles can occupy "waves of possibilities". The particle goes through both slits at the same time because it is equally likely it went through one slit or the other, and which it went through does not matter. It doesn't matter to anything in the whole universe -- and since it doesn't have to choose, it simply doesn't. The particle in a sense doesn't really exist except as a field of possibility. Only when we try to observe the particle is it "forced" to make a choice, and so it "collapses" into an actual particle in a particular location, no longer a field of possibility. And so it behaves like a hockey puck.
Think about it -- if we observe the particle to have come from the left slot, why was it not the right slot? They both had an equal opportunity in the matter. SO WHY THE LEFT SLOT? Why indeed. It could be true randomness in nature. Or maybe it's an stranger than that. Maybe in a parallel universe, the particle went through the right slot, and is merely our consciousness which has "chosen" which universe to be in.
It's really elegant -- Quantum Mechanics is in a way the last frontier of science. It's the study of the smallest of the small -- the very essence of the universe. It is the science of all sciences (possibly). It is like the ruler of the science kingdom, because it affects everything. It's as science as you can get. Then there's the understanding of consciousness. In itself that field is like a final frontier, but it's more on a philosophical level. Understanding consciousness is wrought with religious undertones and ancient philosophers. And yet here we are -- discovering that it's possible that these two seemingly polar opposites are actually talking about the same thing.
It's good to finally put it on paper in my own words, although there are plenty of articles out there that probably do a better job than I could. What really prompted me to write this, is that this experiment has been taken to an even stranger level: The Delayed Choice Quantum Eraser. Basically, it takes the double slit experiment and delays the "collapse" of the wave into a particle until AFTER THE PHOTONS HAVE ALREADY BEEN DETECTED. It adds a time component to the experiment. Essentially, it's the double slit experiment, the one that results in an interference pattern, except carefully constructed in such a way that it's possible to observe which slit the particle went through after it would normally have already impacted the detector and caused an interference pattern. And guess what, there's no interference pattern.
Want to try it yourself? Scientific American has an excellent Do-It-Yourself-Quantum-Eraser that illustrates the double slit effect and the eraser effect. It isn't delayed choice, though -- for that I'm afraid you need more than a laser pointer :) The linked wikipedia articles are also very good and link to even more detailed explanations.
Other good reads on the subject: