Inertial confinement fusion has mostly been the playground of the US. Since the basic idea is that you need a big-ass laserTM that heats and compresses a nuclear fuel, you need to be rich enough to afford the laser system. Not long ago, China joined that club with a laser that was designed to deliver 100 kilojoules per pulse. Although this is still a generation behind the 1.8MJ at the National Ignition Facility in the US, the researchers in China are focusing on innovative ways to achieve high compression.
The key to inertial confinement fusion is that the laser crushes a pellet of nuclear fuel, increasing the pressure and temperature to the point where fusion can occur. This works if you can get a set of laser beams that illuminate the pellet from all sides at once, delivering an even and clean crushing force.
But, what would happen if, instead of applying the force from the outside, the force was applied from the inside and directed outward?
Everyone hates turbulence
At first glance, the idea of applying the force from the inside out seems both bizarre and counterproductive. But the goal is to overcome some of the shortcomings that have slowed progress in inertial confinement fusion.
Let’s follow the train wreck in slow motion. The laser pulses approach the pellet from many different angles. They all have to hit at the same time, and the intensity profile of each beam has to be smooth and identical. Any failure to achieve this results in a large portion of the fuel squirting out the side and escaping the burn region. Think of it like squeezing an inflated balloon: although you can compress it between your hands, it bulges out through the gaps between your fingers.
Even if you have the timing and intensity profile of the beams close to perfect, you still face problems. The implosion is generated indirectly: a heavy metal casing around the fuel absorbs the radiation, and its exterior burns off. As the material flies away from the casing, the back reaction drives the casing inward, compressing the fuel. This compression is so vigorous that, in the very center of the pellet, the fuel begins to fuse.
The fusion forces the inner material to expand into the material that is still collapsing, and this heats and compresses that material so that fusion can continue. In theory, all the fuel is consumed and lots of energy is released in the form of high-energy neutrons.
But what actually happens is that all the imperfections in the outer casing, the shape of the fuel pellet, and the laser beams spiral out of control. Because of the imperfections, the pressure shockwave that initiates fusion is not perfectly spherical, so fusion may initiate slightly off center and with a non-spherical distribution. The resulting outbound shockwave is even more aspherical than the incoming shockwave. The material responds to this by forming a turbulent flow that absorbs energy, reducing the pressure and temperature.
In the end, the amount of neutrons produced is disappointingly low.
Excuse me, you’ve got your fusion inside out
The problem is that the way the shockwave is created is very finicky, so instabilities that turn into turbulence are almost inevitable. Even if turbulence could be avoided, the instabilities lead to large variations in the efficiency of fusion.
Researchers think that by changing the way the shockwave is applied, these instabilities could be substantially reduced. The basic idea is the same: a spherical heavy-metal casing has a shell of deuterium (hydrogen with a neutron) coated on the inner surface. (Actually, it is a polymer with all the hydrogens replaced with deuterium, but the idea is the same.) Then two holes are drilled in the shell. The laser beams are aimed at the holes so that they hit the shell from the inside.
The key point here is that the outer shell does not provide the force for compression. Instead, the direct and fast evaporation and ionization of the fuel creates a rapidly expanding plasma that collides and stops in the center of the sphere. But stopping means that there is a whole bunch of energy that has to be given up by the ions. The result is that the temperature of the plasma shoots up, allowing fusion to occur. Essentially, directed motion with few collisions is converted into random motion with lots of collisions.
This is very similar to standard inertial confinement fusion, but the important difference is that the outward shockwave is almost certain to be spherical because of the way the plasma stops in the center. This naturally causes the plasma to take on a spherical shape.
In a set of experiments, run at quite low energy (~6kJ per pulse), researchers showed that the number of neutrons per shot scaled as expected with input power. They also showed that the neutron yield was quite stable from shot to shot, which is rather important.
The team also showed that the overall neutron yield still depended on how evenly the laser pulses illuminated the inside of the shell. The scientists think that their results would be improved by switching to an octahedron-shaped target with a hole at the center of each face. That would be close to spherically symmetric and yield even better results than they could achieve with the current design.
They also calculated what they could achieve with more energy per shot. If they did the same thing in the National Ignition Facility, they could achieve 1017 neutrons per shot, assuming deuterium-tritium reactions (tritium is a hydrogen atom with two neutrons). For comparison, the National Ignition Facility folk report about 1014 neutrons per shot. However, the number of neutrons also depends on the amount of fuel available, and that is not so easy to estimate from the papers.
Is it really that much better?
Those estimates should be taken with your yearly salt intake. The problem is not with the estimates themselves, but with the idea that the scaling will continue as the laser energy is increased by about four orders of magnitude. That may be right, but my first thought is that the outer gold casing will burn through, reducing the amount of energy that goes into the plasma.
I also think that as the energy increases, the number of gold ions in the plasma will increase, and those ions will suck energy out of the plasma. For comparison: hydrogen has one electron—once it’s gone, all the energy has to be expressed by the nucleus zipping about and colliding with things. That’s perfect for fusion. But gold has a lot of electrons, and they won’t all be stripped. So, when a gold ion collides with a hydrogen ion, it can absorb some energy from the hydrogen ion by driving an electron into an excited state. The electron then relaxes by radiating some light. The light escapes the plasma, removing energy from the plasma. And the gold ion is ready to repeat the process, siphoning more energy out.
Whether this is important or not depends on how fast fusion occurs. When the laser is incident from the outside, everything happens so fast that the ions from the gold shell have little chance to cool the plasma. However, because the process here is a lot slower, I wonder if there is a chance for the gold to mix with the plasma before the outgoing shockwave arrives to complete the fusion burn.
Physical Review Letters, 2017, DOI: 10.1103/PhysRevLett.118.165001