Fusion!
Ooooohh....
I think this should probably be funded...
Update:
More info...
And a video
Further update:
I thought I'd do a quick-and-dirty explanation of the idea to see if I can get a feel for how the thing's supposed to work.
So. With fusion, the idea is to slam light nuclei together past the barrier of their electric repulsion to the very short distances at which the nuclear forces take can control, reorder the arrangement of neutrons and protons, and release large amounts of energy. The problem is that the electric repulsion of nuclei is very, very, large, and the range at which nuclear forces act is very, very small. This means that your nuclei have to be moving at very high speeds in order to allow them to collide. Now, high speed ions aren't that hard to do. The other problem is that there has to be a high density of high-speed ions as well, otherwise there won't be another ion for them to collide with.
The method that's been worked on for a while is what's called magnetic confinement techniques. You use very very strong magnets in a toroidal arrangement, and accelerate your ions along the ring. The magnets exert a right hand force on the charged ions that keeps them moving in spirals through the ring. You pump ions into the system until you have a high enough density to get ion-ion collisions, and hopefully, fusion events. The problem with these systems is that each ion-ion collision that isn't a fusion event, which is most of them, will bump ions to a "higher" field line, closer to the walls of your toroid. They start making looser and looser spirals, until the hit the walls of your ring and become useless to you. This is a problem with the maintaining critical ion density for sustained fusion.
What Bussard's been working on is a development of an old invention of Farnsworth, called the fusor. It's what's called electric confinement. Instead of using magnetic fields to contain the ions, you use an electric potential to contain the ions. Make a grounded spherical conducting grid. Inside it, put an anode at some very high potential, in the kV or higher range. Then you just drop ions into the potential well, they roll down the potential well and harmonically oscillate. You don't lose ions when they have unproductive collisions, since they're still in the potential well - they'll just come back for another try.
How do you make the potential well? Well, you need a lot of negative charge in the middle to attract the ions at high enough speeds. So you contain electrons in there. In the middle of your spherical grounded grid, you put another grid, and you charge this one to a high potential. Zap your electron gun at your device, and you can capture electrons the same way that you can capture ions with a potential well. Then you use the new potential well created by the electrons to capture ions and make them fuse.
These things actually work - you can build one for <$1000, and measure neutron flux from a D-D reaction. The problem is that the interior grid is being constantly bombarded by high energy ions and tends to, well, melt. After like a half-hour running time. So the improvement that Bussard came up with was to realize that although magnetic confinement is a lousy way to contain ions, it's a really good way to contain electrons. So you throw out your grids and use magnets to create a magnetic field with a low potential at the center of a sphere. Shoot your electrons into the sphere and let them get captured by the field until you have a nice virtual anode in the middle. Then add fusile ions, let them flow downhill at high speed, and you get fusion.
Awesome, eh? There's even more cool stuff. There's lots of different types of fusion reactions. The common one is a D-T reaction - fusion of two different isotopes of hydrogen to form helium, a neutron, and some various other radiation. Neutron radiation is really, really bad. It'll kill you, and it can also cause structural damage to your machine from changing the atomic structure of the structural components. Tokomaks use molten lithium surrounding the fusion chambers to absorb neutrons. Molten lithium is not a happy material.
There's another fusion reaction, though, that doesn't produce neutrons, the p-B11 reaction. Take a proton, or hydrogen, and Boron-11. These are both the most common isotopes of their elements, and, conveniently enough, the U.S. is a huge producer of boron. When a proton and a B-11 nucleus fuse, you get an excited C-12 atom that quickly decays to three alpha particles, one of exact energy, and two varying. No neutrons, no neutron radiation. The problem with the p-B11 reaction is that it takes much higher energies on the ions to reach fusion. When you're trying to reach those speed by pumping heat into the system, you're getting a Maxwellian distribution in which only the right tail of your distribution is up to fusile speeds, and getting that high gets harder and harder the higher you need to go. But with an electric confinement system, you're getting a peaked speed, because all your ions are dropping down the same well. It's not Maxwellian, it's peaked around the energy you want, so all your ions are at fusile speeds. And if you need them going faster, you just up the potential! Easy-peasy compared to the problems with heat injection.
Further than that, the alpha particles come out with an energy on the order of MeV. That's way bigger than the potential you're generating with your virtual anode, so the alpha particles come out of your potential well. If you've, say, set up a voltage gradient outside of your potential well, you can directly convert the kinetic energy of those alpha particles, which contains most of the energy of the fusion reaction, into electric power. No steam, no heat transfer. You're capturing >90% of the produced energy directly as electricity.
If this thing works as described, he's talking about a compact, insanely efficient power generator that burns easily-available fuels of which the U.S. has its own sources, and emits pure helium as a by-product.
Tell me why the hell the government cut funding to this? If we spend in the billions of dollars on the damned war, or on the tokomak reactors that aren't even close to working, you'd think we could spare the $200 million he says he needs to get a net-energy producing generator working. Anyone want to tell me why his design won't work, so I stop being so upset about this? I think I'll write my Representative...
I think this should probably be funded...
Update:
More info...
And a video
Further update:
I thought I'd do a quick-and-dirty explanation of the idea to see if I can get a feel for how the thing's supposed to work.
So. With fusion, the idea is to slam light nuclei together past the barrier of their electric repulsion to the very short distances at which the nuclear forces take can control, reorder the arrangement of neutrons and protons, and release large amounts of energy. The problem is that the electric repulsion of nuclei is very, very, large, and the range at which nuclear forces act is very, very small. This means that your nuclei have to be moving at very high speeds in order to allow them to collide. Now, high speed ions aren't that hard to do. The other problem is that there has to be a high density of high-speed ions as well, otherwise there won't be another ion for them to collide with.
The method that's been worked on for a while is what's called magnetic confinement techniques. You use very very strong magnets in a toroidal arrangement, and accelerate your ions along the ring. The magnets exert a right hand force on the charged ions that keeps them moving in spirals through the ring. You pump ions into the system until you have a high enough density to get ion-ion collisions, and hopefully, fusion events. The problem with these systems is that each ion-ion collision that isn't a fusion event, which is most of them, will bump ions to a "higher" field line, closer to the walls of your toroid. They start making looser and looser spirals, until the hit the walls of your ring and become useless to you. This is a problem with the maintaining critical ion density for sustained fusion.
What Bussard's been working on is a development of an old invention of Farnsworth, called the fusor. It's what's called electric confinement. Instead of using magnetic fields to contain the ions, you use an electric potential to contain the ions. Make a grounded spherical conducting grid. Inside it, put an anode at some very high potential, in the kV or higher range. Then you just drop ions into the potential well, they roll down the potential well and harmonically oscillate. You don't lose ions when they have unproductive collisions, since they're still in the potential well - they'll just come back for another try.
How do you make the potential well? Well, you need a lot of negative charge in the middle to attract the ions at high enough speeds. So you contain electrons in there. In the middle of your spherical grounded grid, you put another grid, and you charge this one to a high potential. Zap your electron gun at your device, and you can capture electrons the same way that you can capture ions with a potential well. Then you use the new potential well created by the electrons to capture ions and make them fuse.
These things actually work - you can build one for <$1000, and measure neutron flux from a D-D reaction. The problem is that the interior grid is being constantly bombarded by high energy ions and tends to, well, melt. After like a half-hour running time. So the improvement that Bussard came up with was to realize that although magnetic confinement is a lousy way to contain ions, it's a really good way to contain electrons. So you throw out your grids and use magnets to create a magnetic field with a low potential at the center of a sphere. Shoot your electrons into the sphere and let them get captured by the field until you have a nice virtual anode in the middle. Then add fusile ions, let them flow downhill at high speed, and you get fusion.
Awesome, eh? There's even more cool stuff. There's lots of different types of fusion reactions. The common one is a D-T reaction - fusion of two different isotopes of hydrogen to form helium, a neutron, and some various other radiation. Neutron radiation is really, really bad. It'll kill you, and it can also cause structural damage to your machine from changing the atomic structure of the structural components. Tokomaks use molten lithium surrounding the fusion chambers to absorb neutrons. Molten lithium is not a happy material.
There's another fusion reaction, though, that doesn't produce neutrons, the p-B11 reaction. Take a proton, or hydrogen, and Boron-11. These are both the most common isotopes of their elements, and, conveniently enough, the U.S. is a huge producer of boron. When a proton and a B-11 nucleus fuse, you get an excited C-12 atom that quickly decays to three alpha particles, one of exact energy, and two varying. No neutrons, no neutron radiation. The problem with the p-B11 reaction is that it takes much higher energies on the ions to reach fusion. When you're trying to reach those speed by pumping heat into the system, you're getting a Maxwellian distribution in which only the right tail of your distribution is up to fusile speeds, and getting that high gets harder and harder the higher you need to go. But with an electric confinement system, you're getting a peaked speed, because all your ions are dropping down the same well. It's not Maxwellian, it's peaked around the energy you want, so all your ions are at fusile speeds. And if you need them going faster, you just up the potential! Easy-peasy compared to the problems with heat injection.
Further than that, the alpha particles come out with an energy on the order of MeV. That's way bigger than the potential you're generating with your virtual anode, so the alpha particles come out of your potential well. If you've, say, set up a voltage gradient outside of your potential well, you can directly convert the kinetic energy of those alpha particles, which contains most of the energy of the fusion reaction, into electric power. No steam, no heat transfer. You're capturing >90% of the produced energy directly as electricity.
If this thing works as described, he's talking about a compact, insanely efficient power generator that burns easily-available fuels of which the U.S. has its own sources, and emits pure helium as a by-product.
Tell me why the hell the government cut funding to this? If we spend in the billions of dollars on the damned war, or on the tokomak reactors that aren't even close to working, you'd think we could spare the $200 million he says he needs to get a net-energy producing generator working. Anyone want to tell me why his design won't work, so I stop being so upset about this? I think I'll write my Representative...