Fusion – ain’t gonna happen February 28, 2015Posted by Maury Markowitz in fusion.
Tags: fusion power
A while back I wrote a lengthy article on the problems facing the fusion energy concept, and how even if the technical issues were solved, it wouldn’t make a difference…
No matter how hard you try, a fusion reactor will be more expensive than power sources we have now, so no one is going to build one.
The argument was made by, and backed up by, lots and lots of independent studies. I was merely the messenger. The messenger was attacked, of course. This tech has convinced so many people of it’s total awesomeness that logic simply no longer works. Well, maybe this will…
Someone over on Reddit pointed out this document. It’s written for the Directorate-General of the European Parliament. It’s been written by a number of PhDs in physics, including the director of the Max Planck Institute for Plasma Physics and the director of the French National Center for Scientific Research.
Let just cut right to the chase:
It is generally agreed that nuclear fusion can not play a role at least until 2050, 2060 or even later – if at all, as too many uncertainties and problems still must be solved, some of them being “no-go” criteria for the whole concept.
The go/no-go criteria in question is one I touched on in my first post but subsequently removed. Fusion plants run on a mixture of deuterium and tritium. The first we can separate from seawater, expensively. The second is naturally radioactive and decays in about 12 years, so we need to find an artificial source. Currently there are two artificial sources we know about.
The first are fission reactors. The only non-military source is the Darlington reactor up the road from my house. That’s scheduled to shut down in 2025, which means there won’t be any more tritium by the time they move on to the commercial demonstration systems, DEMO and PROTO. That leaves military sources that make tritium for the spark plugs of nuclear bombs. It is highly unlikely any will be released from those sources.
The second is a fusion reactor, which can “breed” new tritium. Of course this assumes you can scrounge up enough from fission sources to get it running, and they need a lot. But assuming that, it’s not at all clear that they can actually make more tritium from the fusion reactor, once its going. There’s a number, the “tritium breeding ratio”, that needs to be greater than 1.15 in order for this to work, because you have to make up what you burn as well as what decays naturally.
Current estimates are between about 0.8 and 1.1.
Worth repeating: if the ratio is less than 1.15, fusion will just not happen. That’s why the paper calls this a no-go criterion. Now given this, don’t you think someone would test this one part before spending billions on billions for a design that we can’t use? You guessed it, the answer is no. Although we have any number of ways to build test systems, not one single one has ever been built.
And that’s assuming we even need or want this power. As I pointed out in the other article, fusion power would be ridiculously expensive no matter what the technology you use. Sure, you can argue that it provides baseload and, say, wind doesn’t. But wind does provide baseload if you build three times as much as you need at peak, and then build lots of batteries. And that will still be cheaper than fusion. But don’t take my word for it, take the experts:
With respect to these expectations it must be questioned whether the additional spending of billions of Euros today for a technology which at best could contribute to energy supply beyond 2050 when there is no need for that technology can still be justified.
And then to cap it all off, they point out that most of the materials needed for a fusion reactor are already scarce, and will probably be far too precious to even consider burning away in a reactor. If you do put them in there, you can’t even recycle them back out because the core will be furiously radioactive for about 300 years.
This is not a technical issue, it doesn’t matter if the reactors work or don’t work, they will still be too expensive to ever consider building. And when you have better, cheaper, solutions right now, there’s not a whole lot of point in spending money to see if we might build something worse in the future, is there? And we’re already fixing the problem:
We have to solve our problems right now not having further time to loose. In 2050 or 2060 when fusion reactors might contribute, there is no need for them as the necessary transition is almost over