This article discusses advanced reactor technologies that are now getting renewed attention after the Fukushima nuclear plant accident. Interest in smaller reactors has been growing in recent years. Some of these designs have advantages over the traditional large light water reactors (LWRs) for certain applications. The smaller designs carry less of an inventory of nuclear material, so there is less material at risk in an accident involving a release. Proponents of small modular reactors (SMRs) point to cost savings due to the factory fabrication and shorter construction times. They have significant advantages for countries with small grids, where a current 1500 MWe reactor would exceed demand and threaten grid stability. Other designs that are getting the most attention at present are small or medium LWR concepts. In addition to their smaller size, these designs differ from current large, light-water designs in that most of them use an “integral” design. Most major reactor components are inside the reactor pressure vessel, thus significantly reducing the threat of a major loss-of-coolant accident.
Energy: Present an overview of advanced reactor technologies.
A true renaissance in nuclear power development seemed to be taking root in the last decade. The United States was seeing the first new orders for reactors in decades, several countries in Asia were building multiple nuclear power plants and planning many more, and even countries in Europe that had plans on the books to phase out nuclear power were beginning to reconsider their prospects.
All that seemed to change in an instant on March 11, 2011. An earthquake and follow-on tsunami, both of historic proportions, shook the coast of Japan, and in addition to taking a horrific toll on human life, the event wreaked havoc on the Fukushima Daiichi nuclear power station. At this writing, the damaged units are still being brought under control, and the full ramifications of the event are not yet known. While to date no member of the public has been injured or killed by the damaged reactors, the consequences of the accident for both existing reactors and for future reactor development are likely to be substantial, both in Japan and elsewhere.
In the subsequent days and weeks, every country in the world with operating or planned nuclear power plants reacted in some way. A few countries ordered immediate shutdowns of some plants or shelved plans for new reactor development. Most countries, however, continued operation of existing plants and activities on planned reactor development, but they also initiated measures such as stress tests to assure the safety of operating plants and reviews of the accident to identify lessons learned and incorporate necessary changes. Notably, countries in Asia that had ambitious nuclear development programs have announced their intention to stay the course, although with a short delay for review of the lessons learned, and have somewhat scaled back their goals—goals that analysts always thought were overly optimistic anyway. Indeed, the United Arab Emirates held a groundbreaking ceremony for its first nuclear power plant just days after the Fukushima accident.
Where the accident may have its greatest effect over the long term is in the choices of technology for new reactors. The Fukushima accident may result in increasing pressure to adopt newer, more advanced designs, as well as smaller reactor designs, rather than to continue building older and larger, but “tried and true” designs.
Various advanced reactor designs have been proposed. Some are modest evolutions of existing technology, while others are next generation or “Generation IV” concepts. To provide an overview of what may be coming, I’d like to focus on concepts that represent significant developments in design over the light-water reactors and pressurized heavy water reactors that are already operating, licensed, or under construction anywhere in the world today.
Most new reactor construction in progress today continues to involve large light-water reactors. But interest in smaller reactors and in non-LWRs has been growing in recent years. Some of these designs may have advantages over the traditional large LWRs for certain applications, and some might be more robust in conditions similar to those in the Fukushima accident. The smaller designs also carry less of an inventory of nuclear material, so there is less material at risk in an accident involving a release.
The definition of a small reactor is not rigid, but generally, reactors under 300 MWe are considered small by today's standards, while reactors in the 300-700 MWe range are considered medium-sized. This class of reactor is often called an SMR, an abbreviation that may stand either for “small modular reactors” or for “small and medium-sized reactors.” While most small reactors are over 100 MWe, a few being proposed are as small as 10-50 MWe. In general, most SMRs are being designed to be factory fabricated in transportable modules to speed construction and lower prices, so the two terms may overlap.
The nuclear industry began with very small reactors, although they were certainly not modular (and would not meet modern standards). The average size of a reactor grew as the industry gained experience, largely because of economies of scale. A return to smaller reactors seems, to some, to jeopardize those economies, but proponents of SMRs point to cost savings due to the factory fabrication and shorter construction times.
Another economic advantage claimed for SMRs is that they can better fit the variety of markets that exist. Clearly, they have significant advantages for countries with small grids where a current 1,500 MWe reactor would exceed demand and threaten grid stability. A number of such countries are in various stages of considering nuclear power, so there is a healthy prospective market for reactors that are compatible with small grids.
However, even for larger markets, proponents note that small modules have some benefits. First, SMRs can be built incrementally, and thus should be able to match the growth of demand better. In this concept, a “4-pack” or “8-pack” would be planned, but the units would be built one at a time, so the investment at any one time would be only a fraction of the investment required for a traditional large unit. Second, once the full complement of units is in place, scheduled maintenance can be staggered, reducing the need for purchasing electricity from other power plants. A large-scale natural event, however, such as a flood, earthquake, or tsunami, still could conceivably affect all the units on a site.
In addition, the smaller size of SMRs facilitates the use of underground or below-grade siting, which may improve security. The smaller size also lends itself to the incorporation of more passive safety characteristics, which should reduce the vulnerability of the reactors under abnormal conditions. However, at this point, the designs have not undergone full risk assessments. Detailed analyses are needed both to verify claims of enhanced safety and to identify any new risks that might be introduced by the novel design and siting features.
Indeed, it should be remembered that most of the smallscale designs are still in the conceptual design stage, so claims of their benefits cannot be verified. To paraphrase Admiral Hyman Rickover, who directed the development of nuclear propulsion in the U.S. Navy, a “paper reactor” is always simple, cheap, and quickly built, while a “real” reactor is complicated, expensive, and behind schedule.
Any claims to increased safety margins for small reactors also have to be viewed with caution. There are a variety of ways to make large reactors, current and advanced, more robust than the Fukushima reactors proved to be. Many of these provisions are already in place.
Even before the Fukushima accident, interest in advanced reactors both large and small had been growing. Since about the year 2000, several domestic and international efforts have focused on a variety of design concepts.
It is difficult, at this point, to produce a truly comprehensive list of designs under development, and even more difficult to sort out completely the exact stage of development of each concept. Some 50 or 60 innovative designs, most of them small or medium size, have been proposed by organizations in Argentina, China, France, Japan, Russia, South Africa, South Korea, and the United States. Some might be deployable in the near future, while others may take 20 to 30 years. In the past few years, designs that seemed to be on a fast track have stalled, usually for lack of financial support, while other designs have emerged.
Perhaps the best way to look at the range of designs is to group them loosely in several technology categories and to consider the potential pluses and minuses of each group. The traditional way to categorize reactors is according to how each is cooled, with reactors grouped as either light water, gas-cooled, liquid metal, or molten salt designs. Although there are many variations within these groups and any attempt to describe them as a single group will miss some distinctions, the major differences among the groups are the following:
LWRs use water as both the coolant and the moderator (that is, to slow the neutrons down to an energy level conducive to fissioning the uranium atoms in the fuel). The designs that are getting the most attention at present are small or medium LWR concepts. In addition to their smaller size, these designs differ from current large, light-water designs in that most of them use an “integral” design—most major reactor components are inside the reactor pressure vessel, thus significantly reducing the threat of a major loss-of-coolant accident. Light-water SMR designs have some initial advantages because of their similarity to existing large LWRs: Both regulators and operators are familiar with this technology, current regulations have been written based on knowledge about this technology, and it is widely believed that it will initially be easier to license and build light-water SMRs.
The use of a gas coolant such as helium enables operation at high temperatures, resulting in efficiencies approaching 50 percent. High temperatures also make GCRs suitable for some industrial applications, including hydrogen production for use in transportation. Most GCR concepts use graphite as a moderator, but a few are unmoderated and operate as “fast” reactors, in which the neutrons are not slowed down. The GCRs have some potential safety advantages; the lower power density and lack of need for water for cooling provide for passive safety. Designers of these reactors claim they have no core melt potential and will not release radiation in an accident. On the other hand, the larger relative size of GCRs may increase capital costs of the plant and waste volumes.
The use of molten metal (most concepts use sodium, lead, or lead-bismuth) as a coolant allows the reactor to operate as a fast reactor. Most of the design concepts have passive safety features such as natural convection for circulation of the liquid metal coolant. In addition, a loss of coolant flow would lead to higher core temperature, which tends to cause the reactor to shut down. Like the gas-cooled reactors, LMRs operate at a higher temperature than LWRs, and thus are more efficient. They also operate at near-atmospheric pressure, which simplifies the design and improves safety. The fast fuel cycle can be used to “breed” new fuel and to burn spent fuel from LWRs. Thus, they may be able to play a significant role in dealing with used fuel from conventional reactors. However, the use of LMRs would also require a system for reprocessing nuclear fuel, and that has been controversial in the U.S.
Molten salt reactors
In a molten salt reactor, the primary coolant is one or more molten salts, usually fluorides, often with a graphite moderator. The fuel may be in solid rods, as in conventional reactors, or dissolved in the coolant in the form of fluorides of uranium, plutonium, or thorium. Much like liquid metal reactors, MSRs operate at higher temperatures than LWRs and at near atmospheric pressures, and the power generated decreases as the reactor heats up. What's more, MSRs can burn spent fuel from existing light-water reactors. Although there was much experimentation with molten salt reactors in the early days of nuclear technology development (as there was with all the basic design concepts), unlike the other concepts, no commercial molten salt power reactor has been built to date.
Many of the advanced and small reactor concepts are being designed for a much longer core life than current reactors. In the United States, refueling for current large LWRs is usually done on an 18 to 24 month basis, with about a quarter of the fuel changed out at each refueling. By contrast, SMR designs call for core lives that may be as long as 30 years, although this varies considerably by concept. Furthermore, since the designs are modularized, in many cases, the entire core could be removed and replaced when the reactor is refueled, which should provide smaller countries without extensive facilities the ability to use nuclear technology without having to invest in facilities to handle used fuel. A removable core should also improve the proliferation resistance of the fuel cycle, as the used fuel is never separated from the core until it is returned to the manufacturer (or other ultimate handler) for removal, storage, and if appropriate, reprocessing.
Clearly, more development is needed for most of the advanced reactor designs, and many of them will falter for lack of sufficient investment, or later on, for lack of market interest. One indication of the degree of maturity of the designs and seriousness of the designers is active engagement by the U.S. Nuclear Regulatory Commission. While that is not the only relevant measure, it is a real-world milestone that the design has reached a state of development such that NRC can reasonably begin to assess the concept. On that basis, three designs can be singled out as ones on which there are pre-application reviews at NRC: The U.S. Department of Energy's Next Generation Nuclear Plant, NuScale Power's NuScale reactor, and mPower from Babcock & Wilcox.
NuScale and mPower are light-water reactors, while the NGNP is a gas-cooled reactor. NuScale and mPower currently appear to be the most advanced in development, and at least one utility, the Tennessee Valley Authority, has expressed interest in the mPower concept.
Critical aspects of the NGNP, such as size and fuel type, have yet to be confirmed. In addition, as of about the middle of 2011, NRC was interacting with several other reactor designers who are considering applying for design certification in the future. A majority of these other designs are liquid metal reactors.
The Fukushima accident may give new urgency to the work already under way on more advanced reactor designs, and many such designs are in various stages of development. While it is clear that the total number of options that will make it to the marketplace will be much smaller than the number of designs being discussed today, it is possible, perhaps likely, that the future inventory of reactors will come in two basic sizes, the traditional large variety and one or more smaller concepts. Larger reactors may continue to have a cost advantage over small reactors in markets with large grids and large demands. However, the increasing interest of smaller countries with limited grids and limited demands, coupled with other possible benefits of small reactors, may make them superior for some applications.
Given the large body of experience with light water reactors, they appear most likely to be the first advanced small reactors deployed. While the other concepts have some potential benefits, they also require more development. The support for that development will likely depend on whether they really provide a significant safety improvement, and on the perceived need for their other applications, such as hydrogen production or waste reduction potential.