August 29, 2011

Building a more efficient nuclear fission reactor

Current nuclear plants are very inefficient, and burn only a fraction of the available fuel. The startup corporation Terrapower , funded in part by Bill Gates has an innovative plan for a reactor that can burn depleted uranium. Depleted uranium is currently a dangerous waste product produced by conventional lightwater reactors. Enough uranium-238 exists in depleted uranium, uranium deposits, and seawater to meet the earth's energy requirements for centuries. In an interview with Sander Olson (For Nextbigfuture), Terrapower nuclear Engineer Robert Petroski describes how reactors using depleted uranium could potentially play a major role in ameliorating the world's energy problems.

Robert Petroski

Question: Tell us about TerraPower:

TerraPower became an independent company in 2008 to advance the goal of building a traveling wave reactor. All future energy projections clearly indicate that energy use will increase substantially during the next half century. We believe that a new type of nuclear reactor, called the traveling wave reactor (TWR), has the potential to meet this growing need for electricity and to largely solve the world's energy problem.

Question: What is the traveling wave reactor?

The TWR is a nuclear reactor that can use uranium much more efficiently. Current nuclear reactors need to first take uranium and enrich it, which creates a large amount of unused depleted uranium. The TWR is able to burn this depleted uranium, as well as unenriched natural uranium, and is able to burn the fuel further (current light water reactors only have 0.7% fuel efficiency). So the TWR can have more than 10 times as much uranium utilization as conventional reactors.

Question: Are we talking about uranium 235 or uranium 238?

Conventional light water reactors run on mostly uranium 235, which comprises less than 1% of fuel. The TWR needs 235 to get the process started, but then sustains its operation with only uranium 238, which makes up the other 99% of fuel. It does this by converting the uranium 238 into usable plutonium fuel, burning the plutonium immediately in a self-sustaining reaction. Unlike other reactors that convert uranium 238, the TWR is able to directly use the newly created fuel, without needing any reprocessing to extract the plutonium.

Question: How much uranium 238 exists? How much exists within the U.S. and Europe?

A lot of uranium 238 is available in depleted uranium, which is a byproduct of the enrichment process for conventional light water reactor fuel. There are over 750,000 metric tons of this uranium just sitting around in the U.S. That alone could provide all of the earth's electricity needs for about a century. There are also substantial deposits of natural uranium remaining to be mined, and huge amounts of uranium that could be extracted from seawater, though that process is currently cost-prohibitive. Since TWRs use uranium more efficiently, they could make this currently expensive source of uranium become economic.

Question: How long could a traveling wave plant safely operate? How would the safety of such a reactor compare to that of a conventional nuclear plant?

A traveling wave reactor can be designed to run for 40 years without refueling, and should be safer to operate than conventional light-water reactors. TWR reactors use sodium for coolant, which is a liquid metal. In the event of a shutdown, it will naturally cool down by itself. So although a TWR will employ both active and passive cooling mechanisms, the active cooling mechanisms aren’t required for removing decay heat.

Question: What byproducts would a traveling reactor produce?

After the uranium 238 is burned in a TWR, a small volume of spent fuel is left over. This spent fuel could be recycled and used again to start up another TWR, but since the spent fuel is highly radioactive, recycling would be difficult. So the byproducts could either be processed and used again or disposed of in a geologic depository. When I say TWR fuel can be recycled, this is different from what’s usually thought of as “reprocessing,” in which plutonium is chemically separated to make new fuel. TWR fuel can be recycled just by physically recasting it into new fuel, without doing any chemistry.

Question: How much of the fuel would be consumed after recycling?

In the first pass, 20% of the uranium would be consumed. In a second pass, up to 40% could be consumed. In principal, this process could continue until all of the fuel was consumed.

Question: What are the costs associated with reprocessing fuel?

The TWR eliminates the need for reprocessing. Reprocessing plants cost billions of dollars, mainly due to the difficulty of handling highly radioactive waste. One of the advantages of the TWR is that it wouldn’t need a reprocessing plant to burn uranium 238. Another advantage of not doing reprocessing is that you are eliminating a fuel source that could potentially be turned into weapons.

Question: What are the anticipated costs per kilowatt-hour for a traveling wave reactor?

Most of the costs would be simply in building the reactor, and these costs should be comparable to those of conventional nuclear power plants. There are a number of factors that could alter the cost considerably, such as how much experience one has in building a kind of power plant. For example, France has made extensive use of nuclear energy, and now derives 78% of its electricity from nuclear. Its electricity is among the cheapest in Europe. The TWR does have one major advantage: since we will be using sodium instead of water, we can operate the plant at higher temperatures, thereby increasing how efficiently electricity can be made.

Question: I assume that one of the most important cost factors pertain to regulatory issues.

Some of the factors are regulatory, because that can affect how long it takes to build a reactor. There are also issues relating to the location of the plant, such as cost of materials and labor. For a light-water reactor, 10% of the costs are for buying and enriching the uranium. The TWR can reduce those costs significantly.

Question: So there are no regulatory showstoppers?

Getting regulatory approval will be more difficult for a TWR than for a conventional light water plant, mainly because there hasn’t been a lot of prior experience in licensing a sodium-cooled reactor. Like with any nuclear reactor, we will need to demonstrate that the reactor will always operate as it is designed. Despite these hurdles, I am confident that regulatory issues will not be intractable.

Question: How much would it cost to construct a full-scale reactor capable of producing gigawatts of power? How much would it cost to operate?

The ultimate goal of TerraPower is to create a TWR that generates at least one gigawatt electric of power. The size of the commercial plant we are aiming for could generate several gigawatts thermal, which translates to one gigawatt electric. The traveling wave concept works best at large scales, so we don't anticipate building any mini-reactors at this time. Costs have been estimated to be comparable to current light-water reactors.

Question: Are there any fine-grained computer simulations of traveling wave reactors that have shown the viability of this concept?

One of TerraPower's competitive strengths is that we have done extensive computer simulations on the TWR. We have conducted simulations to see if the performance of every aspect of the TWR is what we expect, for example what happens if coolant-flow is lost. All our simulations indicate that although this design needs to be optimized differently from currently technology, it should work according to our expectations.

Question: Assuming sufficient funding and regulatory approval, when is the earliest that a traveling wave reactor could become operational?

We are aiming to get a prototype 500 megawatt electric reactor up and running by 2021. If everything with the prototype goes as planned, it could be followed within five years by a commercial-sized gigawatt reactor.

Question: Does TerraPower derive all of its funding through private sources?

Yes, TerraPower is privately funded. We are a reactor design company, so we don't plan on actually building any reactors ourselves. Rather, we plan on licensing our technology to companies that do build reactors. However, we plan on being involved in the process from design to construction to operation to decommissioning.

Question: From a resource standpoint, could TWR reactors meet all of the earth's energy needs by midcentury?

From a resource standpoint, that is definitely feasible. There is enough uranium on the earth to power all of the earth's energy needs for millennia. With a sufficiently high build rate, TWRs could, within the next several decades, be supplying a substantial portion of the world's electricity requirements.

If you liked this article, please give it a quick review on ycombinator or StumbleUpon. Thanks

blog comments powered by Disqus