**2. Power plant and reactor design**

EBR-II was a complete power plant along with an attached fuel cycle facility. The reactor containment was centered between the sodium boiler building and the turbine/generator building. The reactor was an SFR which acted as a breeding facility and test bed for liquid metal fast breeder reactors [3]. Along with this, EBR-II produced electricity as part of its overall demonstration. Being a fast neutron spectrum reactor, the neutron chain reaction was driven primarily by fast neutrons. Fast neutrons often invalidate many assumptions commonly assumed for light water reactors. The long neutron mean free path associated with a fast neutron spectrum is indicative that much of the core is coupled, meaning there are relatively few localized reactivity effects. This often helps prevent localized peaking. The long mean free path of neutrons also means that negative reactivity insertion due to control rods in a few sections of the core provides the necessary means to shut down the reactor.

EBR-II was a pool-type SFR, meaning the core, and all supporting structures, were contained in a double walled vessel comprised of 86,000 gallons of primary sodium [4]. Due to this design, leaks in any of the primary system piping would drain into the primary coolant. This would result in a loss of plant efficiency but would not leak primary sodium outside the vessel. This design is unlike loop type reactors (i.e. Fast Flux Test Facility, Monju, SuperPhenix), where a leak in the primary coolant had the potential to cause a sodium fire and release activated sodium and would likely cause prolonged outages for repairs.

From a reactor operating perspective, sodium couples four very important properties: 1) extremely high boiling point (870 C) at atmospheric pressure, 2) outstanding heat transfer properties owing to its metallic nature, 3) relatively high atomic weight compared to neutrons leading to limited neutron moderation, and 4) a low neutron absorption cross section along with a relatively short neutron activation half-life of 15 hrs. These properties allow sodium to be used as an outstanding fast reactor coolant. The most obvious drawback of using sodium metal as a reactor coolant is the fact that it reacts with water and evolves hydrogen in the reaction process. The sodium-water reaction can be violent especially when the evolved hydrogen combines with oxygen. The reaction between sodium and water follows two primary schemes forming sodium hydroxide and sodium oxide as shown in Eqs. (1) and (2). In both reactions, hydrogen is also produced which presents a flammability and explosion hazard. It is important to keep in mind that a leak of high temperature sodium to an air atmosphere will result in dense white smoke which makes leak identification simple.

$$Na + H\_2O \to NaOH + \frac{1}{2}H\_2 \tag{1}$$

$$2\text{Na} + \text{H}\_2\text{O} \to \text{Na}\_2\text{O} + \text{H}\_2\tag{2}$$

The primary coolant arrangement for EBR-II can be seen in **Figure 3**. This highlights the major components associated with the primary coolant. Cold coolant (~370 C) was drawn in via two primary pumps, each of which supplied ~18,000 liters per minute of coolant and was split into a high-pressure and lower pressure inlet plenum at the bottom of the core. Of special note, the two primary coolant pumps were single-stage centrifugal mechanical pumps: a first of their kind for liquid metal coolant at the time. After flowing through the core, hot coolant (~480 C) then flowed into a shared upper plenum with a single outlet (shown as a "Z" in both figures). The hot coolant then entered the heat exchanger and was discharged back into the primary coolant pool. To filter out impurities, a cold-trap system continually filtered primary coolant by reducing the sodium temperature to reduce the solubility limits and precipitate out impurities. Above the sodium was ~12 in. of argon gas providing a protective inert cover for the sodium coolant.

The secondary system extracted heat from the primary system which was then used to drive a Rankine cycle for power generation [4]. The sodium flow rate for the secondary system was 23,000 liters per minute, with an inlet temperature of 310 C and an outlet temperature of 460 C. Transferring heat from the radioactive primary sodium to non-radioactive secondary sodium provided a safety enhancement and the ability to place much of the secondary system in a separate sodium boiler building, which was physically separate from the main reactor building. This separation reduced the time required in containment and reduced the potential for radioactive impurities to cause exposure. The sodium boiler building design incorporated a

*Experimental Breeder Reactor II DOI: http://dx.doi.org/10.5772/intechopen.105800*

**Figure 3.** *Primary coolant system for EBR-II [3].*

sacrificial plastic wall located away from the reactor building. The sacrificial wall would fail in the event of a catastrophic sodium water reaction in the sodium boiler building thereby directing the reaction energy away from the reactor building.

For the Rankine cycle, superheated steam was generated at 450 C with a pressure of 9000 kPa: this powered an off-the-shelf 20 MW turbine generator. The ability to use off the shelf components, helped reduce cost in the secondary system (one of the primary objectives of EBR-II). The secondary system allowed for a steam by-pass to continually dump heat despite any electrical needs. The overall EBR-II heat transfer pathway is shown in **Figure 4**.

In addition to the primary and secondary systems, an auxiliary pump was used to ensure a low-pressure flow rate was always present, despite normal power failure.

**Figure 4.** *EBR-II heat transfer pathway [2].*

The auxiliary pump was attached to a DC battery system, which would last long enough to allow the EBR-II system time to convert from forced cooling to natural circulation. To aid in the natural circulation, two shutdown coolers penetrated the primary coolant tank and allowed for heat removal directly to the atmosphere. The shutdown coolers contained sodium-potassium which extracted heat from the primary system and was exposed to an air-cooled heat exchanger.

The EBR-II core used 637 hexagonal subassemblies that made up the driver, inner blanket, and outer blanket regions. **Figure 5** shows the top of the reactor core prior to the introduction of sodium coolant. The driver region was where a majority of the neutron flux was generated, which meant that a majority of the power was generated in this region. In terms of an equivalent cylinder, EBR-II had a diameter of ~20 in. and a height of ~14 in. Subassemblies were generally broken up into a few major categories: driver, blanket, control, reflector, and experiment [5].

Subassembly types shared many characteristics, the most notable being the outer dimensions which allowed for subassemblies to be moved throughout the core, depending on the specific needs. Each assembly was hexagonal in shape, and had an outside flat-to-flat distance of 5.82 cm with a flow duct wall thickness of 0.10 cm. All subassemblies also had an upper adapter (this allowed for subassemblies to be placed and removed from the core), and a lower adapter. The lower adapters had slightly different configurations to ensure subassemblies were placed in the correct location.

Driver fuel assemblies were comprised of, in general, a lower adapter, fuel pin grid, and upper preassembly. Coolant flowed from the inlet plenum into the lower adapter, through fuel pin grid (where heat was transferred to the coolant), and out the upper preassembly into the outlet plenum. Multiple driver fuel designs were used throughout the lifetime of EBR-II, and as such, a brief description of the MK-II fuel assembly design is given [5]. Since these were used throughout the life of the reactor. Comprised within the fuel pin grid were 91 fuel pins in a hexagonal lattice with a fuel pitch of 0.56 cm. Fuel pins are described further in a later section. Halfworth driver assemblies where nearly identical to driver fuel assemblies, however, half of the fuel pins were replaced with stainless steel pins; this reduced the reactivity of the fuel assembly. Half-worth driver assemblies were typically placed near the center of the core to dampen peaking effects.

**Figure 5.** *EBR-II reactor Core [4].*

#### *Experimental Breeder Reactor II DOI: http://dx.doi.org/10.5772/intechopen.105800*

Blanket assemblies were used throughout the life of EBR-II, where they were initially inserted around the core to breed plutonium. Blanket assemblies contained 19 fuel pins comprised of a fuel slug (outer diameter (OD) 1.1 cm), sodium bond (OD 1.16 cm), and a stainless-steel cladding (OD 1.25 cm). Blanket fuel pins were much larger than their driver counterparts due to the lower power density and a desire to increase the fuel to sodium ratio to promote breeding in the pins. Blanket fuel pins were 1.43 m long.

EBR-II, like many SFRs, used full assembly positions for the safety and control rods (denoted control assemblies from here on). Control assemblies had an inner hexagonal duct (flat-to-flat diameter of 4.90 cm) which contained a fuel region with 61 fuel pins which could be brought into the plane of the driver fuel to add reactivity to the core. Some control rods (designated high worth control rods) had a region comprised of seven B4C pins directly above the fuel, which acted as an additional poison to ensure the reactor could shut down and remain shut down.

Reflector assemblies did not contain a pin grid section, but instead contained stacks of stainless-steel hexagonal blocks. These blocks were used to reflect neutrons back into the core and were typically placed in the periphery.

Experimental assemblies were unique in both design and contents. These assemblies maintained the hexagonal duct but could contain fuel, material, monitor, etc. experiments. Experimental assemblies are described in greater detail in a subsequent section.

Fuel pins consisted of a metallic fuel slug (OD of 0.33 cm), sodium bond (OD 0.38 cm) and stainless-steel cladding (OD 0.44 cm). The total length of the fuel pin was 62.04 cm, where the fuel slug had a length of 34.29 cm. Above the fuel slug was a helium plenum to capture fission product gasses and was often tagged with trace amounts of xenon to allow for the determination of burst fuel pins. Each fuel pin was surrounded by a wire-wrap with a diameter of 0.125 cm and an axial pitch of 15.24 cm. The wire wrap was used to ensure fuel pins did not come in contact with each other and provided additional coolant mixing to encourage heat transfer. Throughout the lifetime of EBR-II, the fuel pins changed slightly in dimensions, however, the dimensions presented provide a reasonable representation of a typical fuel slug. **Figure 6** shows an arrangement of driver fuel pins along with the wire-wrap.

The fuel slugs in Mk-II subassemblies comprised a uranium-fissium alloy (95 wt. % uranium 5 wt. % fissium), meaning that the fuel was metallic in nature, compared with the typical ceramic fuel (uranium-oxide) found in light water reactors. The uranium in the fuel was enriched to between 45 wt. % and 67 wt. % U-235, again in stark contrast to the typical 5 wt. % light water reactor fuel. Fissium was comprised of elements to simulate dominant mid-fuel cycle fission products. The short-highly-enriched fuel for EBR-II created a very short-flat core, which provided multiple inherent safety benefits, described in greater detail later.

One other noteworthy feature of the EBR-II design involved a fuel storage basket located within the primary tank. The fuel storage basket contains 75 indexed storage tubes in three concentric rings. Each tube could accommodate a single fuel assembly. The fuel storage basket was accessed essentially anytime by operators including when the reactor is operating at full power. The fuel storage basket provided great operational flexibility. During reactor operation, spent fuel assemblies stored in the basket could be removed one at a time and transferred out of the reactor facility and delivered to a hot cell facility for storage and disassembly. Fresh fuel and experimental assemblies could also be loaded into the basket during reactor operation. When the reactor was shut down, operators could then quickly move spent fuel assemblies from the core into the fuel storage basket and move fresh fuel from the basket into the core making the refueling outage time as short as possible. Since the driver

**Figure 6.** *Fuel pin arrangement [6].*

region of the core contained roughly 100 assemblies, the 75-assembly fuel storage basket provided ample capacity for staging fresh fuel assemblies as well as holding spent fuel assemblies removed from the core. With the fuel storage basket located within the primary tank, the sodium coolant provides sufficient heat transfer capacity to ensure the spent fuel assemblies are adequately cooled prior to their removal.
