Saturday, February 6, 2016

Clean(er) Nuclear Power?

Thorium as a better nuclear power system.

The US has made progress reducing carbon emissions, mostly by substituting fracked gas for coal. This is not a long-term solution however, only cutting carbon emissions to half, instead of to zero. The climate needs more help faster, if the biosphere is to survive the anthropocene.

Renewable power sources have become economically viable, but not quite system-viable, since they are intermittent. Progress on grid-scale power storage has been very slow. The best power storage system remains gravitational water pumping/storage, which is available in relatively few places. In the absence of a good power storage and intermittancy management solutions, solar and wind energy simply can not be relied on for more than about 25% of grid energy. While most of the carbon emissions problem is one of public policy and communal action, this is an example of a remaining technical hurdle.

While I hope those issues are solved, the rest of the world, especially the developed world, keeps turning to coal, which is a planetary disaster. We need something better, in operational, economic, and planetary terms. One solution that has been floating around the margins has been nuclear power from thorium. We are familiar with nuclear power from enriched uranium- the pressurized water reactors that have been chugging away for decades both on the power grid and in aircraft carriers and submarines. (Related reactors) These reactors require quite a bit of uranium 235 (using the same systems as bomb-grade enrichment but to lower levels, and generate quite a bit of extremely long-lived waste, especially trans-uranic waste products like plutonium. Indeed, one of the reasons they were selected was that they generate plenty of bomb-making material.

On the other hand, there is another method of nuclear power generation, using thorium. Thorium is not fissile itself, and is four times as abundant as uranium, and doesn't need to be enriched. Indeed, there are virtually limitless supplies. Once placed in a running reactor and bombarded by neutrons, thorium breeds uranium 233 which is fissile, and generates power. The decay series of this isotope is far more favorable than uranium 235, in terms of making far fewer transuranic products, (fewer bombs), and allowing better recycling. Indeed, the ultimate amounts of long-term waste from a thorium reactor are about a thousandth of that from a U235 reactor. The reactor design is very hot, but unpressurized, and comes with safety features that significantly outstrip U235 reactors. (The gung-ho nerd-tube version.)

Since the main issues of conventional nuclear plants are safety and waste disposal, this is all very good news. How does it all work? A demonstration plant was built and run by the US in the 1960's, but was shut down because it did not provide a path to bomb making materials. This seems now a bit short-sighted, yet the basic principles were demonstrated. The main features are that the fissile fuel exists as a liquid rather than a solid, at rather high temperature (600˚C to 700˚C, which leads to higher electricity generation efficiency), and needs an extra circulating loop of another fluid to transfer heat to turbines. This second fluid has typically been lithium/fluoride/berylium mixture, (the same as the fuel, only without the thorium/uranium), of which berylium is particularly toxic. But apparently the berylium could be left out, so this might not be a necessary element of the design.

The fuel cycle is that unenriched thorium 232 is either included in the liquid fuel, or blanketed around the outside of the reactor. It captures neutrons which transmute it to protactinium 233, which decays in a matter of weeks to U233, whose half-life is 160,000 years. U233, the actual fissile fuel, is then burned to completion in the system, leading to various smaller waste products including noble gasses, cesium, strontium, and other elements whose longest half-lives are on the order of 30 years, (though quite toxic and hard to handle). Very little of the waste is up-converted to transuranic elements like plutonium and neptunium that are especially long-lived. Indeed so little is produced that such elements can be added right back to the fuel, along with recycled U233, to be destroyed by further fission. This assumes an on-site method of pyroprocessing or distillation that can separate the burnt waste from the fissile material to be recycled, to the tune of about 200 kg of fuel elements reprocessed per day. The overall cycle leads to far more efficient use of the fuel, however, hundreds-fold less waste, and particularly less long-lived waste relative to the conventional U235 reactor cycle.

A second significant problem is how to start these reactors, given the need to generate the U233. The Oak Ridge experimental reactor was seeded with U235, for instance. Starting up a fleet of such reactors from scratch would require quite a bit of our stockpiles of enriched uranium. And while melt-down is impossible, given the already melted state of the fuel and the ease of diluting it or spreading it to a non-critical state, it presents significant shielding and handling problems, like any nuclear reactor. Here is a point-by-point critique.

Time scales of the budgets of various elements in a thorium/U233 reactor. If initiated with transuranics like plutonium, (yellow), they would burn off to a low steady state by about 50 years. That level is a small fraction of the total fuel mass, and never needs to be disposed of as long as the reactor is running and its fuel is thoroughly reprocessed. Th=thorium, U=uranium, Pu=plutonium, Np=neptunium, Pa=protactinium, Cm=curium, Am=americium.

The thorium reactor is extremely complex- there is no doubt about that. It is unlikely to be made into a plug-and-play system that could be transported to remote or undeveloped areas, or run with little expertise. It would take a decade or more of serious research to make practical utility scale designs and plants. Not only is there a great deal of high-temperature, high-radiation plumbing, but it needs onsite fuel reprocessing of materials that are extreme gamma emitters, and thus can not be handled directly, even with gloves. Additionally, parts of the reactor like key graphite neutron reflectors & heat exchangers would need to be replaced or refurbished every few years.

The current light water pressurized U235 reactor technology is also complex, and has proven itself to be highly problematic. It generates large amounts of unimaginably long-lived waste, demands large amounts of hard-to-enrich fuel, and has been demonstrated to be unsafe in the short and long terms, requiring heroic levels of engineering for utility use. This is what broke the back of current reactor technology around the world, making it economically inviable, despite the fact that a large fleet of plants are being run ever more efficiently (with due respect to the Fukushima and other disasters like Three Mile Island and the San Onofre plant in California.)

Compared with the tens of billions of dollars that have been spent on fusion power research, the thorium system is significantly more realistic, addressing the need for baseline power that is extremely pressing. Perhaps renewable power and power storage will be solved and take the place of fossil fuels. I certainly hope so- we need to get to zero carbon as soon as possible.