As I’ve made clear before, I remain skeptical that carbon emissions pose a significant marginal threat of climate change. The likely climate sensitivity to CO2 is substantially less than the natural variability over human timescales. Seeing as how temperature trends over geologic time scales are currently downward, I don’t think it’s worth wasting much effort on CO2 reductions.
However, let’s assume for a moment that I’m wrong. What should we do? I don’t think we can actually decrease our energy usage very much and support our civilization. So we have to find non-petroleum energy sources. Biofuel technology doesn’t look very good at the moment. Scaling will require major land use changes that I contend are probably a net negative environmental impact. The cost-benefit for solar does look better, especially in certain geographic areas. But it seems to me the only massively scalable solution at our current level of technology is nuclear fission.
From an engineering perspective, this looks like a pretty obvious conclusion. The problem is that standard light water reactor (LWR) technologies have some serious drawbacks: (1) they don’t scale down well so you need to build big installations, (2) you can’t easily turn them on and off quickly so they typically only supply your base load, and (3) they generate a substantial amount of radioactive waste that we still haven’t figured out what to do with long term. So even if we can overcome the ideological resistance to nuclear power, there are also some serious fundamental economic issues to overcome as well.
Enter the thorium fuel cycle, from which you can build a Liquid Fluoride Thorium Reactor. It essentially solves all these problems and they are simpler to manufacture because you don’t need giant pressure vessels. They can scale down to 5MW. If you’re willing to run them at only slightly lower efficiency, you can use them for load following and peak reserve roles. They generate about .1% of the long term radioactive waste as standard LWRs. It’s also incredibly difficult to reprocess fuel into weapons-grade material.
The coolest part is you can mass produce the components and then assemble them on site so we could actually get these babies up and running quickly. For details, see the Energy from Thorium blog and site (I recommend starting with the former and moving on to the latter’s discussion forums and document repository if you want gory details).
Run your Tesla on clean, reliable thorium power!
There is also a technology that allows current light water reactors to utilize the Thorium cycle. There is a “seed and blanket” fuel assembly that uses Plutonium from spent reactor fuel to drive the reaction in a Thorium containing blanket that surrounds it. The blanket remains in the reactor producing power for up to 12 years while the seed is replaced every three. During this time, 2/3rds of the Plutonium is burned down. The Blanket produces less waste by mass (not being exchanged as often) and the seed represents a decrease in Plutonium mass that is 3 times more efficient than MOX. This fuel design is also significantly proliferation resistant relative to standard light water reactor fuel. A prototype ran for 6 years in the Shippingport nuclear reactor and was further developed by a company called Thorium Power Ltd to use as a means of rapid Plutonium disposal. It acts as a drop in replacement for current fuel assemblies.
Good point, Ken. While burning down plutonium in existing reactors using the thorium cycle doesn’t help our energy problem, it does help our proliferation and waste management problem.
Along these lines, I should note that manufacturing a lot of LFTRs would actually _consume_ a lot of our existing high level waste because you need that waste to start the thorium cycle.
There is a graphic tutorial about the benefits and technologies of the liquid fluoride thorium reactor at Aim High!, http://rethinkingnuclearpower.googlepages.com/aimhigh
Kudos for the recognition that if you believe in global warming, you have to think seriously about nuclear power.
Issue (2) peak vs. base power generation, is true with other forms of generation as well. Potential energy storage (pump water uphill to a reservoir and recapture via hydroelectic) is one widely deployed solution.
Pluggable electric cars recharging overnight can be a very good “smoothing” solutions, although of course they have their own issues.
My first job was stochastic modeling of electric utility systems so I’m rather familiar with the devilishly tricky problem of peak loads. While there are a variety of load smoothing strategies, none of them are probabilistically sufficient on peak summer days. So you always end up with a bunch of gas turbines and oil fired plants. It would be nice if we could replace some of them with smaller LFTRs.
Thorium fuel cycle link is mis directed to the LWR page