Over the next month or two, I will publish four extracts from the book Plentiful Energy — The story of the Integral Fast Reactor by Chuck Till and Yoon Chang.
Reproduced with permission of the authors, these sections describe and justify some of the key design choices that went into the making the IFR a different — and highly successful — approach to fast neutron reactor technology and its associated fuel recycling.
These excerpts not only provide a fascinating insight into a truly sustainable form nuclear power; they also provide excellent reference material for refuting many of the spurious claims on the internet about IFR by people who don’t understand (or choose to wilfully misrepresent) this critically important technology.
The IFR metal alloy fuel was the single most important development decision. More flows from this than from any other of the choices. It was a controversial choice, as metal fuel had been discarded worldwide in the early sixties and forgotten. Long irradiation times in the reactor are essential, particularly if reprocessing of the fuel is expensive, yet the metal fuel of the 1960s would not withstand any more than moderate irradiation. Ceramic fuel, on the other hand, would. Oxide, a ceramic fuel developed for commercial water-cooled reactors, had been adopted for breeder reactors in every breeder program in the world. It is fully developed and it remains today the de facto reference fuel type for fast reactors elsewhere in the world. It is known. Its advantages and disadvantages in a sodium-cooled fast reactor are well established. Why then was metallic fuel the choice for the IFR?
In reactor operation, reactor safety, fuel recycling, and waste product—indeed, in every important element of a complete fast reactor system—it seemed to us that metallic fuel allowed tangible improvement. Such improvements would lead to cost reduction and to improved economics. Apprehension that the fast reactor and its associated fuel cycle would not be economic had always clouded fast reactor development. Sharp improvements in the economics might be possible if a metal fuel could be made to behave under the temperature and radiation conditions in a fast reactor. Not just any metal fuel, but one that contained the amounts of plutonium needed for reactor operation on recycled fuel. Discoveries at Argonne suggested it might be possible.
Metal fuel allows the highest breeding of any possible fuel. High breeding means fuel supplies can be expanded easily, maintained at a constant level, or decreased at will. Metal fuel and liquid sodium, the coolant, also a metal, do not react at all. Breaches or holes in the fuel cladding, important in oxide, don’t matter greatly with metal fuel; operation can in fact continue with impunity. The mechanisms for fuel cladding failure were now understood too, and very long irradiations had become possible. Heat transfers easily too. Very little heat is stored in the fuel. (Stored heat exacerbates accidents.) Metal couldn’t be easier to fabricate: it’s simple to cast and it’s cheap. The care that must be taken and the many steps needed in oxide fuel fabrication are replaced by a very few simple steps, all amenable to robotic equipment. And spent metal fuel can be processed with much cheaper techniques. Finally, the product fuel remains highly radioactive, a poor choice for weapons in any case, and dangerous to handle except remotely.
Important questions remained—whether uranium alloys that included plutonium could be developed that had a high enough melting point and didn’t harm the fuel cladding, while at the same time retaining the long irradiations now possible for the uranium EBR-II fuel. Early metal fuel had swelled when irradiated—the reason it had been discarded. But the swelling problem had been solved for all-uranium fuel. EBR-II had been operating with fairly long burnup uranium metal fuel for over a decade. Long-lived metal fuel resulted from metal slugs sized smaller in diameter than the cladding that allowed the metal to swell within the cladding. If properly sized, the metal swelled out to the cladding in the first few months of irradiation, and when it did, it exerted very little stress on it. After that, the fuel would continue to operate without any obvious burnup limit nor any further swelling.
Before the metal swelled sufficiently to give a good thermal bond with the cladding the necessary thermal bond was provided by introducing liquid sodium inside the cladding. The compatibility of liquid sodium with uranium metal allows this. As the fuel swells, sodium is displaced into the empty space at the top of the fuel pin, provided to collect fission product gasses. The bond sodium is important. It provides the high conductivity necessary to limit the temperature rise at the fuel surface and therefore the temperature of the fuel itself. The swelling itself, it was found, is caused by the growing pressure of gaseous fission products accumulating in pores which grow in size in the fuel as operation continues. But as swelling goes on, the pores interconnect and release the gasses to the space above. At less than 2 percent burnup the point of maximum swelling is reached, and the interconnections become large enough that sodium enters the pores. This, in turn, has the effect of restoring heat conductivity, which then acts to minimize the fuel temperature rise in the fuel.
The soundness of the basic uranium design had been established by thousands of uranium fuel pins of this design that had been irradiated without failure in EBR-II. But now, metallic uranium-plutonium would need to be designed to accommodate swelling. Would the plutonium content cause swelling behavior different from uranium alloy? And, more worrying, plutonium forms a low-melting-point eutectic (mixture) with iron, below the temperature required for operation. A new alloying element would be necessary to raise the eutectic melting point. Zirconium was known to be helpful in that. Zirconium also suppressed the diffusion of the cladding elements, iron and nickel, into the fuel. Iron and nickel form a lower melting point fuel alloy; worse, they form those alloys in the fuel next to the cladding. Zirconium solves these problems. Ten percent zirconium was chosen as optimal, because higher amounts gave fuel melting points too high for the techniques we intended to use to fabricate the fuel. Ten percent gave fuel with adequate compatibility with the cladding, and a high enough melting point to satisfy operating requirements, and could be fabricated with simple injection-casting techniques.
Thus the fuel would be a U-Pu-10Zr alloy. But would it work? Ten percent burnup, about three years in the reactor, was our criterion for success. We would have one set of tests initially, and everything depended on its success. In the event, the fuel passed 10% with no difficulty. It got close to 20% before it was finally removed from the reactor. There were no failures (such as burst cladding). The very first IFR fuel assemblies ever built exceeded the burnup then possible for oxide fuel in the large programs on oxide development of the previous two decades. Metal fuel which included plutonium had passed the test. All the benefits from its use were indeed possible. The program could then turn to a thorough sequence of experiments and analysis to establish, in detail, its possibilities and limitations.
The IFR fuel cycle is the uranium-plutonium cycle. In this, non-fissile uranium-238 is converted slowly and inexorably to fissile plutonium-239 over the life of the fuel. If there is a net gain in usable fuel material, the reactor is a breeder; if not, the reactor is a called a converter (of uranium to plutonium), as are all present commercial reactors. But all reactors convert their uranium fuel to plutonium to some degree. Water reactors convert enough that about half the power the fuel eventually produces comes from the plutonium they have produced and burned in place. A significant amount of the plutonium so created also stays in the spent fuel.
A large and lasting nuclear-powered economy depends on the use of plutonium as the main fuel. The truth about this valuable material is that it is a vitally important asset. Its highly controversial reputation has been built up purposefully from the activists, with little countervailing public awareness of its “whats and whys.” Its very existence is said to be unacceptable. In this way, breeder reactor development was stopped in the U.S. and today continues only fitfully around the world. The fact that present reactors fueled with uranium convert uranium to plutonium very efficiently indeed, creating new plutonium in yearly amounts comparable to the best breeders possible, is lost in the rhetoric. But facts are facts. The principal plutonium-related difference between breeders and converters is that breeders recycle their plutonium fuel, using it up, cycle after cycle, so the amount need not grow. Present reactors leave most of the plutonium they create behind as waste. For efficiency in uranium usage, there is little incentive to recycle it; perhaps a twenty percent increase in uranium utilization is achievable, at a considerable cost to the fuel cycle. (Other reasons, such as waste disposal, may make reprocessing of thermal reactor fuel attractive, but not the cost benefits of plutonium recycling.)
However, it is plutonium that brings the potential for unlimited amounts of electrical power. Plutonium no longer exists in nature except in trace amounts. Its half-life is too short: 24,900 years. The earth’s original endowment decayed away in the far distant past. It has to be created from uranium in the way we just described. Plutonium is a metal. It’s heavy, like uranium or lead. It is chemically toxic, as are all heavy metals if sufficient quantities are ingested, but no more so than the arsenic, say, common in use for many years. It is naturally radioactive, but no more so than radium, an element widely distributed over the earth’s surface. Its principal isotope, Pu-239, emits low-energy radiation easily blocked by a few thousands of an inch of steel, for example, and it is routinely handled in the laboratory jacketed in this way. It is chemically active, so in fine particles it reacts quickly with the oxygen in the air to form plutonium oxide, a very stable ceramic. If this is ingested, either through the lungs or the digestive system, as a rule the ceramic passes on through and the body rids itself of it. A popular slogan by the anti- nuclear organizers is that “a little speck will kill you.” Nonsense—a little speck of the ceramic plutonium oxide will not react further, and will generally pass through the body with little harm.
Plutonium has been routinely handled, in small quantities and large, in laboratories, chemical refineries, and manufacturing facilities around the world for decades. There have been no deaths recorded from its handling in all this time. A study of the wartime Hanford plutonium workers gave the unexpected result that these people on average lived longer than their non-plutonium-exposed cohort group. This was explainable as the likely result of better and more frequent checkups because they were involved in the study, but at the very least there was certainly no shortening of lifespan.
The last point is plutonium’s use for weapons. The very fact that Pu-239 is fissile makes this a possibility, as it does also for the two fissionable isotopes of uranium, U-233 and U-235. But plutonium for a time was exceptional because it could be chemically separated from the uranium that it was bred from, and it did not require the large, expensive diffusion plants necessary for the separation of the fissile U-235 isotope of uranium. But this ease of acquisition argument changed with the development of centrifuges. Now the fissile element U-235 can be separated from bulk uranium with machines. And instead of a stock of irradiated fuel, a chemical process, and facilities for handling, machining, and assembling a delicate implosion device, as one must have for plutonium, for uranium one has a nearly non-radioactive natural uranium feed, centrifuges that can be duplicated to give the number needed, and a nearly non-radioactive product, easily machined and handled, which allows a more simply constructed weapon. Plutonium can no longer be singled out as more susceptible to proliferation of nuclear weapons than uranium. The fact is that uranium now is probably the preferred route to a simple weapon in many of the most worrying national circumstances. Iran’s current actions are a case in point.
Weapon-making is complicated by the presence of radioactivity. Plutonium processed by an IFR-type process remains very radioactive; it must be handled remotely, and delicate fabrication procedures are correspondingly difficult. Uranium is so much easier. This is not to imply that the large and sophisticated weapons laboratories like Los Alamos or Livermore could not use such isotopically impure reactor-grade plutonium; it is sufficient to say they would not choose to do so with much more malleable material available. And the beginner would certainly avoid the remote techniques mandatory for IFR plutonium.
This is the situation: Plutonium, as used in the IFR, cannot be simply demonized and forgotten. It is the means to unlimited electricity. The magnitude of the needs and estimates of the sources that might be able to fill those needs lead to one simple point: Fast reactors only, taking advantage of the breeding properties of plutonium in a fast spectrum, much improved over any uranium isotope, can change in a fundamental way the outlook for energy on the necessary massive scale. Their resource extension properties multiply the amount of usable fuel by a factor of a hundred or so, fully two orders of magnitude. Fine calculations are unnecessary. Demand can be met for many centuries, by a technology that is known today, and whose properties are largely established.
This technology is not speculative, as are fusion, new breakthroughs in solar, or other suggested alternatives. It can be counted on.