0
$\begingroup$

I was looking at why fusion reactors need to get so hot and the internet said that it needed the momentum to overcome the repulsive forces of two atoms. Is heating the atoms the only way to give it enough momentum? Could we spin the two atoms In opposite directions instead and bring them close to let the angular momentum smash the two atoms into each other?

$\endgroup$
  • $\begingroup$ Using an ion accelerator works just fine as well. $\endgroup$ – Jon Custer Jun 24 at 14:47
2
$\begingroup$

Is heating the atoms the only way to give it enough momentum?

Oh no, in fact, this is among the hardest ways.

The very first fusion reactions were produced in a simple particle accelerator that fit in a small lab at Cambridge. This shot deuterium ions into a metal foil that had been infused with deuterium gas. Every once in a while one of the D's from the accelerator would hit one of the D's in the foil, and presto, fusion. This was way back in 1932.

Unfortunately, its trivially easy to demonstrate that the number of missed collisions is many orders of magnitude greater than the number of collisions, and the energy you spent accelerating those ions is thus orders of magnitude higher than what you get back out. In spite of several very clever attempts to get around this, no one's come close to a working accelerator-driven reactor.

Next, lets consider the H-bomb. These work through a combination of factors, but the main one is that they compress a tiny rod of plutonium in the core of the fusion fuel, the "spark plug" to ridiculous pressures. What was once sub-critical immediately goes super-mega-critical and the resulting blip of fission gives off a burst of neutrons.

These neutrons fly off in all directions through the surrounding fusion fuel. This is normally a dry fuel of lithium deuteride. When a neutron hits the lithium ions in the fuel, they turn into tritium. In the super-condensed state of the fuel, the chance that will impact a nearby deuteron is so high that they fuse, release another neutron, and so on.

So right off the bat, one way to get around the heating side is to simply compress the fuel. That heats it too, the fuel in the bomb is the equivalent of millions of degrees (ideal gas law here, nothing more) but it's the compression that's key. What you're trying to do is ensure that everything is so dense that various bits like neutrons and alphas don't have a chance to escape off the surface and carry away your energy. This is non-trivial.

Around 1960 Nuckolls suggested one could do this with a laser instead of a bomb. If you shine lasers all over the outside of a sphere, the "blowoff" compresses the interior, hopefully to the point where you get reactions. 60 years later we have NIF, and while it "works" in that it does indeed create fusion reactions without directly heating the fuel, it failed to reach the magic number called "ignition" that is required (in this design) to output net power.

Which brings us to heating. Recall that the idea behind the bomb is to keep everything inside the fuel. This is called the "confinement time" (which you might see confused with the amount of time a fuel keeps in the reactor, not the same thing at all). As density goes up, energy confinement improves and you need less outright heating to reach your magic number. As density goes down, its not like confinement suddenly disappears, but you do have to add more external heat to make up for the losses.

So something like ITER is at one end of a scale - it has VERY low density, "a very good vacume", and makes up for the constant energy loss through external heaters. On the other end of the scale is NIF, which has no external heating and lots of compressions (100 times the density of lead - think about that). There are any number of designs that live between those two extremes, like MAGLIF.

| cite | improve this answer | |
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.