Explosions!

Today I attended a physics colloquium; every Tuesday they bring in someone to talk about physics research or a physics topic. Today it was "Recreating Astrophysical Explosions with Combustible Gases in the Laboratory."

Stars are giant fusion reactors, but they mostly convert hydrogen into helium. The heavier elements come from when a star 'dies' and explodes, in an x-ray burst, a nova, or a supernova. The particular system that today's talk was looking at was accretion driven explosions, which happen in binary star systems. Often, in a binary star system, one star will collapse into a neutron star or a white dwarf-- a highly massive, but very small star-- and the other will burn itself out and expand becoming a red giant. When this happens, material from the now much larger star will be attracted to the smaller one, and will be captured by this. As this happens, the small star, due to its compact state, will no longer act as an ideal gas-- when more matter is added to an ideal gas, temperature will rise and volume will increase, which will then cool it back down. In a degenerate star-- that is, the white dwarf or neutron star-- pressure will increase instead of volume. As more mass is added to the star, the pressure rises, increasing the temperature, which then increases pressure again, which then increases temperature; rather than the star cooling itself, it creates a runaway reaction. In this environment, temperatures are such that nuclear reactions compound, and heavier elements are created.

When a nuclear reaction like this happens, many, many transformations happen; an element is created, and before it can decay back down to a more stable state, it will be bombarded again and instead raised up to a new, heavier element.

It is impossible to create this sort of cascading reaction in a lab; the temperatures and pressures required are only present in these super-compact degenerate stars. Creating something like this on Earth would be catastrophic. What we can do, and what our laboratories are trying to do, is create the individual reactions in the cascade, one at a time.

Here is where the talk got really interesting; as much fun as the theory is, it is fascinating to see how these things are actually done.

The Oak Ridge National Laboratory, where our speaker was from, has an enormously complex setup, and I cannot speak about it in great detail because, to my dismay, I don't understand a lot of what it all does. I'll tell what I can, though. They have a cyclotron that produces radioactive ions, which they then send to a massive tower to accelerate them. The principle is fairly simple. They have a huge electromagnet, the top of which they set to positive. The negatively charged ions accelerate upwards towards it. At the top there is a carbon stripper foil; when the ions pass through it, the density and energy are increased, so they can no longer sustain as many electrons. Thus they are turned from negatively charged ions to positively charged ones. Now positive, the ions accelerate bock down the tower, to be directed into the appropriate mechanism from there-- depending on what sort of ions they are, the facility can direct them to a number of places. Today's talk focused on the Nitrogen 17 to Neon 18 reaction, so that is the only set up that we heard about.

In order to create the reaction they wished to measure, they needed three things: Nitrogen 17, Hydrogen, and a lot of energy. Above I discussed briefly the setup to create the high-energy beam of Nitrogen 17; now I'll talk a bit about the Hydrogen. The ORNL facility uses an extremely low pressure gas chamber, with multiple pumps set up to keep the pressure low. The pumps have to create a constant pressure for the gas window, as it is called, so that their measurements will be consistent. The benefit of having these pumps, the speaker told us, was that it doesn't take a whole lot of time to get the machines up and running; this is aided further by the fact that they are being used to lower pressure, rather than raise it.

So now we have a beam at high energy being focused into a target, of the right elements. This will produce the reactions desired-- though not very often. For every reaction produced, you need as many as 10^12 particles to pass through the window. Now we need to filter out the non-reactions so that we can measure the ones we want to look at. This particular reaction emits a gamma ray along with producing the Neon 18; unfortunately, gamma rays do not have a lot of momentum, so the reactions are barely deflected. However, they are at different enough velocities to be filtered that way. Directly after the gas window in the setup is a recoil separator. This device has two functions: it will re-focus the slightly deflected atoms back to where we want them, and it will only allow atoms traveling at a specific velocity to pass through it. The first is accomplished by magnetic fields that are vertical-- perpendicular to the path of the beam-- and the second by electric fields that are horizontal. The parts of the beam we don't want to look at can be deflected away, and only the one we do want passes through.

After that there is a dipole magnet that deflects the beam at an angle, measuring the momentum of the atoms. Now that we have both velocity and momentum, we can measure the mass, to be sure that we have what we wanted to create. There is also a detector that measures energy loss, which will yield the atomic number, after some calculations.

So! We have created a reaction. Now, what can we learn from it? After plotting the data, we can see the resonance strength-- the bell curve will either have a very steep slope, or a more gentle one, and that will show how sensitive the resonance is. (I'm not going to try to explain resonance at the moment; it has to do with quantum energy states, and I do not feel confident enough in my understanding to say much about it.) The resonance strength, it turns out, is directly proportional to the reaction rate-- and that is a very useful thing to know. This particular reaction, for example, has a half life of about 2 hours. This is long enough that we can measure it, with satellites, and based on those measurements, we can figure out what the star was like before it went nova.

If we can get accurate data for every one of the reactions we can produce in a lab, we can extrapolate the more high energy ones from the data. When we have a thorough understanding of every reaction that takes place in a star's explosion, we can create a more accurate model of a star's life. And as I have mentioned before, models are extremely useful things.