Site Meter

Monday, January 15, 2007

back to physics

Sometimes I fel like I forget how to do physics. There are actually lots of different activities that fall under the description. One break-down of physics is in terms of computational/theoretical/experimental. Lately I've been focussing on the computational and theoretical, which basically means computer programming and math. If you do this too much, you can forget that you are supposed to be talking about the real world.

So, I was given a calculation to do: some gold ions fly along at relativistic speeds through a cold (4.5 K) metal pipe. There's a very high vacuum inside the pipe, but there's still some air floating around. If the gold ion traverses 3 KM of this pipe, then about how much energy will it lose?

This is a side project, so I haven't been able to put too much time into it, but I've been slowly panicking. I forgot how to do physics! But (again slowly) I read a few things and start to find that there aren't that many formulas out there, and I maybe figured out how to do it.

One challenge is to be a practicing physicist and to continue reading Nancy Cartwright. Its funny that this should be the case, but its like you need a certain amount of propaganda to keep going. Cartwright doubts the completeness of physics. She says that it covers much less than we think it should.

How do I think about the problem I just mentioned? Think of the gold ion flying along. Use the ideal gas law to figure out the density of air... let's say its hydrogen molecules- H_2. Now figure out an estimate of the cross section for ionization for H_2. I look it up in the Handbook of Accelerator physics and Engineering and find that it is around 2E-23 m^-2. The cross section is basically how big of a target the process under consideration presents. In this case- for the Gold ion to kick off an electron from the Hydrogen. So now, given the density (2E13 molecules/m^3), you can figure out how many interactions the ion is likely to have in the 3KM. It turns out to be about 5E-5. In other words, in 10^5 times through this pipe (around the accelerator) you'll get about 5 ionizations. Multiply by the average energy loss per ionization (~20 ev?) and you get the energy loss.

Anyway, this is basic physics. I make a variety of assumptions here. Someone else might make slightly different assumptions. Experience gives faith that all these ways will give similar results which are approximately accurate to what happens in the real world. It is this experience, this craft of doing physics that rebels against the picture Nancy Cartwright paints of physics.

But I don't want to say that this implies that Nancy Cartwright is wrong. When you learn a craft, you learn all sorts of things without realizing it. For example, I happen to know that people do calculations like these all the time and they generally give pretty reasonable results. This is one such hidden assumption. Analyze this calculation enough and you can probably make it fit with Cartwright's approach.

But like I said, its hard to practice physics at the same time as holding Cartwright's philosophy. It encourages the kind of doubt that slows you down. I suppose in the same way that it would be hard to be a CEO and hold a Marxist philosophy. You might think: good, reform from inside the system! But in practice, its pretty difficult.

The general question here is to do a philosophical analysis of back of the envelope physics. Such back of the envelope calculations are supposed to be the true test of a good physicist. You should be able to estimate anything within an order of magnitude within a few minutes. That this is possible is the faith of the physicist. If true, even in some approximate sense, it says something interesting about the world. I take it that Cartwright says it is not true. Its only appears true because the questions are rigged ahead of time.

No comments: