Bill Gates, Toshiba, Traveling Wave Reactors, Small Reactors
There is a great deal of buzz coursing through the “pipes” on the Internet regarding talks between Toshiba and Bill Gates’s TerraPower, the company that is developing the Traveling Wave Reactor (TWR). That is not terribly surprising – when Bill Gates talks, the web (and the business press) listens. Here are some links to example stories:
- China Post – Gates, Toshiba in early talks on nuclear reactor
- Fox News – Bill Gates Wants a Nuclear Reactor
- BBC – Bill Gates and Toshiba discuss nuclear power venture
- Fast Company – Bill Gates Goes Nuclear With Toshiba’s 4S Reactor
- The Register – Bill Gates goes (mini) nuclear
- Environmental Leader – Bill Gates, Toshiba May Develop Clean Nuclear Technology
- The Independent – Bill Gates company to go nuclear with Toshiba
- Taipai Times – Gates, Toshiba in talks on reactor
There is certainly some truth to what you might have been reading about the Toshiba-Gates discussions. Gates and TerraPower are definitely interested in developing reactors that can burn depleted uranium and used nuclear fuel for a very long time before they need to be refueled. Toshiba definitely is interesting in designing and building reactors of all sizes – from their 1358 MWe ABWR’s to their 1154 MWe Westinghouse AP1000’s (Toshiba owns 70% of Westinghouse), to their 10 MWe sodium cooled 4S reactors that might end up powering small villages in Alaska.
However, many of the versions of the stories linked above have combined information from several different projects to provide an inaccurate description of the TerraPower Traveling Wave Reactor as something that is quite small – some stories have even described it as being small enough to fit into a hot tub. I have it from a very good source that the physics associated with initiating and sustaining a fission wave that can convert depleted uranium or used nuclear fuel into fissile material so that the reaction can continue for decades requires a reactor that will produce at least 700 MWth (300 MWe) and is better with a reactor that will produce at least 1000 MW of thermal energy (430 MWe).
Those numbers indicate that the Traveling Wave Reactors will be 10-20 times as large as the “hot tub sized” Hyperion Power Modules. They can still be significantly smaller than the 1200-1600 MWe reactors that many people who discuss nuclear matters think of as “standard”.
Gates and his TerraPower team will need to work with a power plant designer and equipment vendors that can build the physical machinery that their computer programs design and simulate. Toshiba would be a great choice for helping to find materials that can withstand a long exposure to an intense neutron flux. However, it is important to understand that TWR’s that last for decades will not be “hot tub” sized systems. Systems in that size range will have characteristics more like the Hyperion Power Modules that use a liquid metal coolant, fast neutrons, and last for 5-8 years before needing to be refueled or replaced.
The world needs a full spectrum of nuclear fission power plant sizes that can fit a variety of customer needs; the concept should be understandable to anyone who recognizes that combustion power plants come in sizes ranging from leaf blowers to low speed diesel engines that power ships like the Queen Elizabeth.
“Systems in that size range will have characteristics more like the Hyperion Power Modules that use a liquid metal coolant, fast neutrons, …”
Er … that description could also apply to the reactor that TerraPower plans to design. The main difference, other than size, is that TerraPower’s potential design would be able to go about 10 times as long without replacement — in theory.
Brian – the size is a key system parameter. It is what allows loading all that extra potential fuel and what keeps neutron leakage to levels that do not have a huge effect on conversion ratios.
This is good to know, since a lot of people – myself included – seems to be confusing the Toshiba 4S with the TerraPower. I was under the mistaken impression from some of the stuff I’ve read that Gates and Co. went to Toshiba to mash up the TerraPower concept with the 4S, or something along those lines.
One other note: the TWR – since it’s not shut down once started (or at least that’s the impression that I’m under) – would be an ideal candidate to use lead coolant instead of sodium. Lead coolant has some great characteristics (opacity to gamma with transparency towards fast neutrons), especially when working in a steam plant (lack of reactivity with water). You could even use lead coolant for the primary and dump the IHX if you were willing to use nitrogen/carbon dioxide/helium coolant for your gas turbines.
I can see it now, you hit the SCRAM button, and a light comes on saying ‘are you sure’…
(ya, I know…)
DV82XL – You are most certainly joking with a reference to some famously unfriendly software systems, but just suppose that you have a reactor that really is passively safe.
Why would you want to hit the “scram” button? If you were supplying the main source of power to a remote location, wouldn’t it be a good idea to have a system that made sure that is what the operator wanted to do before initiating that action.
In fact, many current reactors have scram switches that are designed so that it takes a couple of independent (simple, but hard to do without purpose) actions. Designing the switch that way prevents accidental initiation of a “safety” action that can have significant operational and cost consequences.
As I understand it, in order for the TWR reactor to work (and to control reaction rate), a neutron reflector is moved along in sync with the reaction wave. The reactor can be shut down by moving the reflector away from the reaction zone. About restarting it, I really don’t know, but there will be a bit more fissile material available after shutdown than during operation as what was the fertile material in the process of decaying to fissile material completes the chain. Perhaps there are enough stray neutrons around from decaying fission products to get things going again.
Nathan Myrvhold (spelling) is the ‘genius’ hired at Microsoft to develop their ‘really’ advanced technology. He spent billions and produced nothing. Apparently Bill Gates has a blind spot for this guy.
Bill Gates made a bad Ethanol investment and true to form is making another ill advised decision involving energy. This is far from an energy miracle as stated in Bill’s TED speech this year. A particularly long term investment when a 10-40 year time line is included. Fusion energy will far outpace this entry Bill Gates has made into the nuclear energy market place. Dispersed or Portable Generators based on nuclear in the 1 MW range are still sizeable units needing something like barge or rail car transport. I believe the only MAGIC in this deal is Bill Gates name.
It’s hard to pick winners and losers. The fact that research is being done on TWR is only good and the knowledge learned can help all reactor designs. The old reactor designs have a negative perception. The TWR hype will sway the public into confidence in all future reactor designs. The only bad thing is if Gates makes a lot of patents and then fail to licence them out of greed. Often greed over patent rights only keeps a good ideas from being used.
Is there any reason why Gates isn’t looking to fund LFTR development? Seems to me that if he is really interested in supplying reliable energy, using a design that has already been proven, even if in a relatively primitive state, would bring a better ROI than creating something new from scratch.
@ DocForesight,
Yes, this is my question exactly. The LFTR does most of what the TWR does with an opportunity to recover the fission products on line as valuable items for use and sale. Xeon and others are valuable but in the TWR it seems like everything stays inside and is not recoverable.
Hey Rod, can you let me know how I may contact you for nuclear reactor issues?
@Bill – if you scroll down to the bottom of the Atomic Insights blog page, you will see email contact information.
It is unfortunate that Bill Gates has latched onto the TWR. It is a concept that addresses one very minor issue while introducing several potentially large issues.
Imagine you run an airline and someone offers to build an airplane that only needs refueling once in 30 years. Great, you can fly it round the clock for 30 years. But wait, you still need to perform annual maintenance checks, change the tires, brake pads, hydraulic fluid, pack wheel bearings, upgrade instrumentation and control systems replace time limited components etc, so capacity factor will not increase as much as you might expect.
In the reactor world lots of work goes on during refueling outages besides refueling, so maintenance outages will still be required.
On the down side consider these issues.
1
@ Bill Hannahan
If I understand the physics (which I may have missed not being trained in this area) a molten Salt reactor using U235 would breed very pure Plutonium and several other actinides with a long life. While a Thorium / U233 molten salt would not breed much plutonium and fewer actinides.
So, why would we use U235 which needs to be separated in a fairly expensive process rather than U233 which we have available and can be breed easily?
The issue that the TWR addresses is much larger than just avoiding the act of refueling. If successful, it would prove beyond a shadow of a technical doubt that used nuclear fuel is a resource, not a waste product.
By doing the conversion from fertile to fissile and then from fissile to heat without any handling or recycling, the system eliminates a lot of potentially costly steps. It also shows how a reactor can serve as a great and amazingly secure storage location for material that some people think of as an expensive storage burden. Just envision a future situation where there are a few thousand light water reactors supplying the world’s electricity needs. If the TWR has proven its ability to slowly consume the waste products while producing massive quantities of heat, those light water reactors can keep merrily supplying what they supply while some folks are gathering the waste products and building TWR’s to keep the electricity coming for another hundred years or more. What is not to love about that?
I’d like to say that – like Rod – I’m inspired by the potential of the TWR design for consuming used fuel without the use of reprocessing. This a revolutionary giant leap forward, one which the importance of cannot be overstated. I’m inspired by the 20% utilization factor of the energy that’s available in uranium. Truly, this is an invention that has the potential to change everything. The core neutronics are a fantastic advance and this should be explored as a national priority. Mr. Gates and Mr. Myhrvold deserve kudos for bringing this idea forward.
However, I have several questions regarding the non-neutronic choices the reactor designers have made in the spirit of Bill’s questions.
First, one of the PDFs available on the TWR describes safety rods. Like all reactors, I assume that the TWR needs to be shut down at times. All things need to be serviced, especially very expensive things for which replacement is a non-trivial task.The question that I have is that if you scram the TWR, will the act of scramming it mess up the deflagration wave? Can it be restarted from cold shutdown as a non-trivial act?
Second, anything involving large quantities of liquid sodium coolant introduces major maintenance considerations when it is not “plug and play” replaceable, especially if it is large in size. From what I understand, the core becomes off limits due to activated sodium. Superphenix, Monju, and Fermi 1 come to mind as having lengthy shutdowns due to complications from sodium chemistry and sodium activation. Of course, as a counter-example, the Russian BN-350 and the Russian BN-600, both of which apparently worked (in the case of the BN-350), and work (in the case of the BN-600) very well, come to mind. I wonder what sort of strategy TerraPower is pursuing for in-core maintenance and sodium management? Such things – if not very carefully engineered – have the potential to cost lots of money to fix when designed incorrectly in the first instance, from what I understand.
Third, if the reactor is not able to be shut down once started without causing major expense and major headaches, how is energy dissipated when it is not powering something? What will the costs be of such an energy dissipation system? What will the costs be of making that system meet NRC standards and making it fail-safe?
I think that the TWR is a great concept, it definitely needs to be developed – and I look forward to learning a lot more about it.
I am not sure I understand why some people believe that a TWR cannot be shut down just like any other reactor. The concept of a “wave” is just a way of explaining how fissions that produce 2.47 neutrons can use one of the neutrons for causing another fission, one of the neutrons for converting fertile material into fissile material and the rest of the neutrons for leakage and absorption by materials that will not either fission or be converted. Essentially, a TWR is just a reactor where the designers have carefully figured out how to achieve and maintain a conversion ratio of 1.0 – creating a new fuel atom for every fuel atom that is consumed in the act of producing heat.
If you need to stop the reaction, you simply drive in some additional absorbent materials to change the balance so that the chain reaction stops. There are still plenty of materials left in the core that will spontaneously produce neutrons sufficient to start the reaction again once the absorbent materials are removed.
Sure, there may be some considerations associated with transient build up of temporary absorbers like Xe-135, but all you need to do is to ensure that there is enough spare reactivity margin to overcome those effects. They are modest and completely predictable.
Pool type reactors using sodium coolant can be made passively safe. There was an unfortunately timed demonstration of this concept in 1986 with the EBR-II/IFR. The test involved a complete loss of forced cooling without scram – completely boring (which is good). Unfortunately,less than a month after the demonstration, some fools in Ukraine exploded their reactor and set back nuclear energy development for a couple of decades.
I’m not worried about the possibility of shutting down the reactor. In fact, considering the guy who is promoting the design and his experience with technology, I’m sure that it will need to be “rebooted” quite frequently, if history is any guide. 😉
I’d say that the largest concern is the possibility of viruses. Now, before you point out that a nuclear reactor cannot possibly carry a virus, let me just say that I can remember a time when we could say the same thing about email.
Thank you, Bill Gates.
A molten salt reactor will breed no purer plutonium than any other reactor that uses U238 as part of the fuel. In molten salt form, Pu is not particularly easy to recover from used fuel salt that is a witch’s brew of carrier salts, fission products and fission product salts, and fissionable and fertile uranium salts. I believe that all of the developed processes for recovering Pu from irradiated fuel use metal or metal oxide chemistry to separate the bred Pu from the fertile U. The uranium recovery process from fuel salts uses additional flourination of the uranium to UF6 to recover pure Uranium and leaves the Pu behind with all the other stuff.
In any case, unless you are trying to make Pu and remove irradiated fresh fuel every month or so, too much weapons grade Pu 239 is converted to weapons useless Pu 240. Used commercial reactor fuel cannot be used for weapons. Even “reactor” grade Pu is useless for weapons.
One of the huge problems with Gates backing this design is that there is still a lot of money that thinks he is a technical genius. There are all sorts of IT people that will tell you that MS ‘solutions’ were shoved down their throats by an upper management that thought anything Gates touched must be gold. If he still can attract that kind of cult following, or worse if he can talk his good buddy Pickens into it, this concept might draw huge interest from the investment community, and turn into a juggernaut.
History can repeat itself.
DV82XL – if Gates was still in his acquisitive mode of world market domination, you might be right. However, I really believe that he has turned a corner and cares deeply about trying to make the world a better place for all humanity. I do not think that there is much of a risk of focusing on a single solution here – I think the investment will merely shine a light onto a technical area where investors will recognize that there is incredible potential for fission based systems that can compete in the power/energy markets. I cannot think that is a bad thing.
David, the Plutonium from the MSR is even less suitable for weapons than that from conventional LWR
The big advantage is that first, it has someone like Bill Gates behind it, and second, it builds momentum for everyone else. Whatever you think of the man, when a big player like Gates steps onto the side of nuclear, everyone on that side gains credibility, cachet, and mindshare; further, everyone has more oomph and doors that might not have been open previously may now be open. Don’t forget – Gates has his own money – so it won’t take away from other projects.
Even if a “beta version” turns out to be a diamond in the rough, shall we say, we’ve got more allies on the side of nuclear energy. Plus, it’s kind of inspiring, too. Whatever one may think of Gates, there is no question that he was an excellent coder (he personally wrote most of the original feature-rich dialects of BASIC for the first microcomputers; it got to the point that any platform that didn’t have MS BASIC was at a very major disadvantage in comparison to the competition – with the exception of the Apple II), a phenomenal businessman, and a very lucky man, with a good sense of timing. Plus, Windows isn’t that bad…at least since they moved to the NT kernel.
An excellent coder?!
Well, I guess if you’re a fan of BASIC, then … well … it’s pointless for me to argue with you.
Don’t forget that Gate’s old company is also responsible for Microsoft Bob (which they were working on while other companies discovered the Internet), and it is also the company that requires nearly a decade today to get a new, decent version of their OS out the door.
Customer complacency and strong-arm tactics with hardware suppliers have been the key to Microsoft’s success, not superior technology, and certainly not Bill Gate’s excellent coding skills.
Given the pedigree of this company, I fully expect that TerraPower is just looking to do enough work to file a series of patents. Then they’ll sit back and hope that somebody else actually builds their reactor before their patents expire.
Why else do you think that they have chosen an unproven design on which almost no work has yet been done? Certainly, they don’t want a design that has already been built and demonstrated. Why would they want lose any of their IP revenue to prior art?