Saturday, September 10, 2005

I've Got Cavities

I haven't written about science for a while. That's because nowadays it's hard to say anything about the topics I'm learning without writing about work, and we all know that blogging about your workplace is a bad idea.

Also, many of those topics are boring. I don't even want to think about the mess of cables, oscilloscopes, amplifiers, local oscillators, function generators, piezoelectric crystals, etc that I've been training to use, much less explain what they do.

But there's still some stuff worth typing about, I guess. So here's the short version.

When you shine laser light through a one-way mirror onto another mirror, it gets trapped in the "cavity" between the two, reflecting back on itself over and over again. (Over-simplification warning: we don't have true "one way mirrors." But we have things that are close enough.) Now say the further mirror lets one percent of the light pass through it, out of the cavity.

You look at this output. What do you see? Guess.

Okay, the answer is -- not much, unless the wavelength of your light is exactly right. As you adjust the wavelength of your laser (And by the way, this is why you need a laser. Ordinary white light has waves of all different lengths. Even light of one color, say red, still has too big a range of wavelengths, unless it came from a laser)... As you adjust the wavelength of your laser, the output of your cavity goes from really dim to really bright. If you go past the right wavelength, it gets dim again. What's the right wavelength? One that fits in the cavity so that the reflections line up on top of each other, peaks on top of peaks, valleys on top of valleys. Then all those waves add up, so that even one percent of the light inside is really bright. This is called resonance.

So what's the point? Well, number one, if you didn't believe light was a wave before, you will after seeing this. Number two, it gives you intense, trapped light, which is what you want if you're going to do quantum mechanics experiments. It makes everything simpler. Number three, it allows you to measure wavelengths or distances (if you know one, you can measure the other) to like, nanometers. Number four, you need cavities to create laser light too, and it turns out laser light has a surprising number of applications.

So Ken and I are building cavities. For different reasons, which come under the headings of complicated and boring and talking about work. His are four centimeters long and mine is one meter. We're fighting problems like "thermal drift" (things expand and contract as the temperature changes, by enough nanometers to ruin your resonance) and vibration (I had to build a semi-sound proof box) and Ken's case, making everything teeny-tiny, and out of materieals that can go in a vacuum chamber.

This is where all the amplifiers and piezowhatsits come in, so this is where I'll stop.

Anyway, that's the kind of science I do, lately.


Ashi said...

"You look at this output. What do you see? Guess.

Okay, the answer is -- not much, unless the wavelength of your light is exactly right. As you adjust the wavelength of your laser (And by the way, this is why you need a laser."

Is this true? I would imagine that if you had coherent blackbody radiation or any other distribution of wavelengths which included the correct one, that you'd end up with a bright laser coming out. (Maybe my assumptions are wrong?)

Mary said...

Well, you definitely wouldn't end up with laser light, coming out of an empty cavity. The cavity doesn't actually amplify the light you put in. It's more like a filter. It lets almost 100% of light at the resonant wavelength through, and something like 1% of light at any other wavelength. (Assuming you make it with 99% reflecting mirrors)

If you filled the cavity with a "gain medium" then you could actually amplify the light you put in. This is how a laser works. A laser always needs a power source.

It's true that you can put a light with a broad spectrum in to a cavity-with-gain-medium (AKA a laser) and get intense single-frequency light out. All the other frequencies are filtered out, and the resonant one is amplified.

If you put broad spectrum light into an empty cavity, all you will get out is the small component which was at the resonant frequency. Cavities are actually used this way, as filters.

But for most other applications, or the ones I'm working on at least, you want to be putting a known wavelength in, so that you can "lock" the cavity length. As soon as the length drifts, the output light level of your cavity falls, and that's how you know the length has changed. If you were putting in a broad spectrum, then the light level wouldn't fall. One of the other frequencies in the light would still be resonant.