Nuclear power in the United States achieved some remarkable successes in its early years and experienced dramatic growth in the late 1960's and early 1970's. While this rapid growth was seen as a measure of the success of the technology, in retrospect it may have been detrimental. As discussed in chapter 5, the size and complexity of reactors increased rapidly and there was little opportunity to apply the experience gained from older plants to the design of newer ones. In addition, the regulatory framework was incomplete when many of the plants were designed. As new regulations were formulated, the designs had to be adjusted retroactively to accommodate to changing criteria. With the rush of construction in the mid-1970's, it was difficult to fully integrate these new requirements into the original designs; hence some portions of the reactor designs emerged as a patchwork of nonintegrated and often ill-understood pieces.

Several changes in design requirements have had far-reaching effects in today's reactors, even though they were not originally expected to have such an impact. For example, new criteria on fire protection in nuclear powerplants have spawned new features and systems to prevent, contain, and mitigate fires. This led to greater separation of safety systems, changes in cable-tray design, requirements for more fire-resistant materials, and changes in civil structures to prevent the spread of fires. Clearly, these modifications can have ramifications for other plant systems. Other regulatory actions concerning seismic design, decay heat removal systems, and protection of safety systems from other equipment failures have also had extensive impacts.

A fresh look at the design of light water reactors (LWRS) could be useful if it more fully integrated the cumulative changes of the past and reexamined the criteria that stimulated those changes. In addition, a new design could incorporate analytical techniques and knowledge that have been acquired since the original designs first were formulated. In fact, it could be beneficial

to investigate designs of alternative reactors that have different and potentially desirable characteristics. It is possible that a new design could improve safety and reliability at an acceptable cost and within a reasonable timeframe.

This provides the basic technical reason for reevaluating current nuclear technology as embodied in LWRs. It is important to question, however, the justification for actual changes to the current system. Are there any indications that the current generation of LWR is less than adequately safe or reliable? The public appears to be increasingly skeptical that nuclear reactors are good neighbors. As discussed in chapter 8, more than half of those polled expressed the belief that reactors are dangerous. The same percentage of the public opposes the construction of new plants. While this is not an absolute measure of the adequacy of today's reactors, it does reflect a growing concern for their safety.

The nuclear utilities also have assessed the current reactors in view of their special needs and interests. While they do not believe that LWRs are seriously flawed, the utilities have expressed a desire for changes that would make plants easier to operate and maintain and less susceptible to economically damaging accidents (13). Some movement has already been initiated within the nuclear industry in response to utility needs. Most of these efforts focus on evolutionary changes to the current designs and thus represent normal development of LWR technology.

The increasing levels of concern for safety among the public and the utilities has contributed to an interest in safety features that are inherent to the design of the reactor rather than systems which rely on equipment and operators to function properly. The emphasis on inherent safety is reflected to some extent in evolutionary designs for LWRS, and to a much greater degree in innovative designs of alternative reactors. In this chapter, LWRs as well as several proposed alternatives will be examined and their relative advantages and disadvantages assessed.

To assess the safety and reliability of current reactors and compare them with alternative designs, it is important to understand the basic principles involved in generating power with nuclear technology. At the center of every nuclear reactor is the core, which is composed of nuclear fuel. Only a few materials, such as uranium and plutonium, are suitable fuels. When a neutron strikes an atom of fuel, it can be absorbed. This could cause the nucleus of the heavy atom to become unstable and split into two lighter atoms known as fission products. When this occurs, energy in the form of heat is released along with two or three neutrons. The neutrons then strike other atoms of fuel and cause additional fissions. With careful design, the fissioning can be made to continue in a process known as a chain reaction.

A chain reaction can be sustained best in uranium fuels if the neutrons are slowed before they strike the fissionable materials. This is done by surrounding the fuel with a material known as a moderator that absorbs some of the energy of the neutrons as they are released from the fission process. Several different materials are suitable as moderators, including ordinary water, heavy water,* and graphite.

The heat from the fission process is removed from the core by a continuous stream of fluid called the primary coolant. The reactors examined in this chapter use water or helium as the coolant, although other fluids have been considered. The heat in the coolant can be used directly to produce electricity by driving a turbine-generator, or it can be transferred to another fluid medium and then to a turbine-generator. Both methods have been used effectively in U.S. nuclear powerplants.

There are many possible combinations of fuel, coolant, and moderator that can be used in the design of nuclear reactors. There are advantages and disadvantages associated with the various

*A molecule of light water is made from one atom of oxygen and two atoms of the lightest isotope of hydrogen. By contrast, a molecule of heavy water is made with the isotope of hydrogen called deuterium, which has twice the mass.

materials, and no single combination has emerged as being clearly superior to the others.

Several designs have been developed for producing electricity commercially. The most common reactors are known as light water reactors, which use ordinary water as both coolant and moderator. LWR fuel is slightly enriched uranium, in which the percentage of fissionable material has been increased from its naturally occurring value of 0.7 percent to about 3 or 4 percent. After enrichment, the fuel is shaped into ceramic pellets of uranium dioxide and encased in long, thin fuel rods made of a zirconium alloy.

Another commercially feasible reactor is the heavy water reactor (HWR), which is moderated by heavy water and cooled by ordinary water. The fuel in an HWR is similar in form and composition to LWR fuel, but it need not be enriched to sustain a chain reaction. Another design is the high temperature gas-cooled reactor (HTGR), which uses helium as a coolant and graphite as a moderator. The HTGR can use uranium as a fuel, but it usually is enriched to a greater concentration of fissionable material than found in LWR fuel. The fuel form is very different from LWR and HWR fuel, with the uranium shaped into small coated spheres, mixed with graphite to form fuel rods, and then inserted into hexagonal graphite blocks.

In addition to selecting a fuel, moderator, and coolant, reactor designers also must devise a means to transfer the heat from the core to the turbines. In the United States, two different steam cycles have been developed for LWRS. The pressurized water reactor (PWR) shown in figure 19 maintains its primary coolant under pressure so that it will not boil. The heat from the primary system is transferred to a secondary circuit through a steam generator, and the steam produced there is used to drive a turbine.

The second type of LWR that is in commercial use is the boiling water reactor (BWR), shown in figure 20. It eliminates the secondary coolant circuit found in a PWR. In the BWR, the heat from the core boils the coolant directly, and the steam produced in the core drives the turbine. There

is no need for a heat exchanger, such as a steam generator, or for two coolant loops. In addition, since more energy is carried in steam than in water, the BWR requires less circulation than the PWR.

The two LWRS described above can be used to illustrate another crucial part of reactor design. Since nuclear reactors produce highly radioactive materials as byproducts of the fission process, it is essential that the design incorporates enough safety features to ensure the health and safety of the public. During normal operation, the radioactive materials are safely contained within the fuel rods and pose no threat to the public. The concern is that during an accident the fuel may become overheated to the point that it melts and releases the fission products that accumulate during normal operation.

Safety is designed into a nuclear reactor on several levels. First, every effort is made to pre

vent minor events from developing into major problems. This is accomplished in part by incorporating inherent features into the design to ensure stable and responsive operation. For example, the physics of the core dictates that most reactors will internally slow down the fission process in response to high coolant temperatures, and thus dampen the effects of problems in removing heat from the core. Both PWRS and BWRs have been designed to respond in this way.

Other features, known as engineered safety systems, operate in parallel with, or as a backup to, the inherent physical safety features. They are designed to ensure that the chain reaction is interrupted promptly if there is a problem in the plant and to remove heat from the core even under extreme circumstances. This is necessary because radioactive decay continues to produce heat long after the reactor has been shut down. If decay heat is not removed, the core can overheat to the point of melting. In the event that the shut