View Single Post
Old 02-13-2014, 07:41 AM   #2
High Inquisitor
High Score: 3,994
Perspicacity's Avatar
Join Date: Nov 2007
Location: Elsewhere
Gender: Male
Posts: 555
DLP Supporter Donor Star
High Score: 3,994
TL;DR: It's headline-grabbing hype and oversells what is an otherwise very nice scientific result. But that's really all it is: a bit of good science, not a landmark achievement or major milestone.

The somewhat longer version:

In inertial confinement fusion (ICF), you take a big bank of lasers, whose amplifiers take up the space of about three football fields, and you focus the laser energy (an amount needed to bench press a brachiosaurus five times) into the ends of a cylinder about a centimeter long and made out of a mixture of gold and uranium called a hohlraum.

The 192 laser beams hit the inner walls of the hohlraum and heat it up. Hot things glow. The hotter a thing is, the higher the energy of the radiation it gives off. The universe glows, a bit of leftover from the Big Bang, but in the microwaves. People glow in the infrared, which is how night-vision goggles work. The surface of the sun, at 5800 K temperature, glows in the visible range. The inner walls of the hohlraum wall glow after the laser hits them, but as x-rays. So you can think of the hohlraum as a fancy converter of radiation from visible light (the NIF lasers are blue, at about 351 nm wavelength) to x-rays, acting as a very hot oven. This conversion loses quite a lot of energy, however, 80% or more of the incident laser energy.

The x-rays from this x-ray oven are absorbed in a plastic or diamond capsule and they compress it very symmetrically (though not quite symmetric enough, as we'll see in a moment). Here's a picture of a capsule:

The laser energy is absorbed in the capsule and causes the outer part to boil off. Momentum is conserved, so to make stuff boil off the outside of the capsule, the thing has to be pushed inward, rather like a rocket engine, though think of rocket engines strapped all around the capsule each firing simultaneously and causing thing to compress. This squishing happens until the deuterium-tritium "fuel" in the core goes from cryogenic conditions cold enough to freeze hydrogen to roughly 20 times the density of lead and 100 million Kelvin, at which point fusion reactions should start to go. This requires shrinking the radius of the capsule by a factor of around 35.

If you get everything to work just so, and if our models for how thermonuclear burn works are accurate, then you create the conditions for thermonuclear fusion to happen in the "hot spot" at the center of the capsule. If you get the hot spot fuel burning fast enough, the alpha particles (helium nuclei) that form as a bi-product of the deuterium + tritium -> alpha (3.5 MeV) + neutron (14.1 MeV) reaction can heat the surrounding fuel, bringing it up to fusion burning temperatures as well, "igniting" it in the manner of kindling in your fireplace igniting the nearby logs. And ignition is the real milestone, which hasn't happened (not even close).

So why did NIF fail to ignite after all these years? For a long time, we didn't know. That's what this campaign was to find out. It turns out that squishing things in radius by a factor of 35 is incredibly hard to do. Try this experiment yourself sometime: take a balloon and try to squeeze the balloon to half its volume--you have enough strength in your body to do so, but you'll find the balloon bulging out between your fingers before you can. NIF tries to do the same thing, but with more of a factor-of-10000 compression in volume. When squishing stuff, whether a balloon with your hands or a NIF capsule, you tend to find "low mode asymmetries," a fancy way of saying that stuff doesn't like being squeezed and it squirts out the places that aren't squeezing as much. To really squish stuff well, you need exquisite symmetry.

In the NIF point design (variants of which folks have been trying to get to ignite for about three years now), they do this squeezing with four exquisitely timed shock waves that arrive at the center of the fuel at the same instant. Even the slightest errors in timing or symmetry get amplified during compression, making the final assembly anything but a nice, symmetric hot spot.

The "high foot" campaign being reported in the article on was designed to figure out why NIF failed to ignite and to test the hypothesis that it was these low mode asymmetries that were the cause. The idea was to make compressed fuel with fewer shocks--not four but three (or perhaps even two) shock waves, each of which is stronger than the four shocks in the point design, but leading to less overall compression and an implosion that is far less error prone.

This campaign, designed to test a reason for failure, succeeded in getting a much more symmetric hot spot, releasing quite a lot more fusion energy than before, and even showing evidence of sufficient symmetry of fuel compression that the alpha particles from the burnt DT plasma in the hot spot indeed coupled into the surrounding fuel. It's a noteworthy landmark, proving that the original design principles were right for the most part, but didn't take into account the sources of asymmetry properly (whose origins we still don't understand). It was conceived along the lines of if you try and fail, it's far worse not to know why you failed than to know. The latter lets you go back and try to fix things. Since a single shot on the NIF costs hundreds of thousands of dollars, you can't afford to "shoot in the dark" to find out why failure happened.

So why do I say this is oversold? Because the high foot path is not one that realistically leads to fusion ignition. The fuel doesn't get compressed enough for that. The only way anyone knows to get the kind of compression is to back off on the very things that made the high-foot shots more symmetric. ("High foot," incidentally, is a descriptive term referring to the size of the initial laser pulse sent into the hohlraum--the "point design" used a low foot, meaning a very weak first shock wave, but also unpredictable behavior in the plasma, probably seeding some bad asymmetries.) Omar himself, who led this study, will admit that this design is probably not on the path to ignition.

You can't look at a huge, generational project and not factor in the people element. Those at LLNL, the Lab who has the laser, are getting beat up for overselling how easy they thought fusion ignition would be and for largely shutting out the rest of the scientific community. They need a feel-good story and they need it yesterday to stave off impending layoffs of scientists and engineers. And Nature, the journal where they published, is notable in that they are both the cachet journal for big, splashy results, and a commercial outlet that coordinates press blitzes and hypes the hell out of results. This is part of the journal's business model which is a bit cringeworthy as a scientist, though something that LLNL is happy to capitalize on in this era of depressing and devastating budget cuts.

A technical reason it's oversold is that yes, the energy put into the hot spot fuel is overcome by fusion energy produced. But this artificial metric is so far from practical fusion power as to be laughable. You're not counting the energy put into the rest of fuel and capsule ablator, which is greater by about a factor of at least 10. You're not counting the additional factor of 10 or so energy loss between energy put into the hohlraum walls and re-radiated as x-rays absorbed in the fuel. You're not counting that the lasers themselves are only about 10% efficient. That's a lot of factors of ten for something that's unlikely to ever ignite. While ignition leads to a big multiplier, even ignition wouldn't be at engineering break-even. Honestly, to imagine that this is some kind of major milestone on the path to fusion energy is to be optimistic to the point of self-delusion.

Don't get me wrong: it's great that we now have a better sense of low mode asymmetry being the culprit for failure. But what can we actually do about it? The only sure-fire way of getting to ignition that anyone can think of would be to build a bigger laser so one can recover margin in the design elsewhere. (If you can compress more fuel, you need to compress it by less.) This just ain't going to happen--it's a political impossibility in the best of budgetary times and it certainly won't happen in anything like the present climate, where Congress hasn't seen any science and engineering R&D that they didn't think, "Hey, let's cut that."

Lost in this ocean of recent hype is that NIF's main mission has never been and will never be fusion power production; that was all window dressing. NIF's mission is validating our understanding of the conditions inside nuclear weapons. The NNSA (the part of the Dept. of Energy that manages our nuclear weapons) has zero interest in advancing fusion energy. It's not their job. The part of DOE and the only part of the government that does care about fusion energy (the Fusion Energy Science Program of the Office of Science) funds primarily magnetic fusion, not inertial fusion, and their budget cuts have been so horrendous over the last couple of decades that the U.S. fusion energy program is depleted beyond any sort of sustainable unit and is merely withering on the vine as the last researchers in the field retire.

If fusion energy ever does happens, it won't be the U.S. leading the charge.

In short, I see no way that this result impacts fusion energy in any meaningful sense beyond maybe stirring up a bit of public enthusiasm. But you can only play that game for so long before the public starts asking, "Hey, where's our flying cars?"

Last edited by Perspicacity; 02-13-2014 at 09:34 AM. Reason: typo fix
Perspicacity is offline   Reply With Quote
Thumbs Up 38 thumbs up