Pebble-bed reactors use 330,000 tennis-ball-sized fuel elements in place of conventional fuel rods.

Pebble-bed reactors use 330,000 tennis-ball-sized fuel elements in place of conventional fuel rods.


temperature) to produce uranium dioxide fuel particles or kernels. The kernels are then put in a chemical-vapor-deposition (CVD) furnace at 1,000°C (1,832°F). Successive carbon and silicon-carbide layers are added to create the Triso microsphere.

It's probably understandable why some people protest the deployment of nuclear power. The safety systems on current reactors don't inspire a lot of confidence. They are characterized by numerous motors and ac power supplies, pumps, and valves. To the uninitiated, the complicated collection of components might smack of Rube Goldberg.

Contrast this with the recently approved Westinghouse AP-600. The 600-MW pressurized light-water reactor (LWR) employs safety systems that are predominately passive. They rely only on natural forces such as gravity, natural circulation, convection, evaporation, and condensation. All in all, the AP-600 contains 35% fewer pumps, 50% fewer valves, 70% less cabling, and 80% less ducting and piping than conventional LWR systems.

Evolutionary improvements like the AP-600, however, probably won't revitalize nuclear energy. What's needed are designs for smaller plants that are economical and can be built quickly. Such concepts fall under the auspices of NERI, the Nuclear Energy Research Initiative program. NERI, — through competitive, peer-reviewed selection processes — sponsors scientific and engineering R&D by universities, national laboratories, and industry. NERI-sponsored work is poised to drastically change the way reactors are designed, built, tested, and deployed. Current reactor research includes advanced light and heavy water, high-temperature gas-cooled fast breeders, and liquid-metal cooled, to name a few.

Many NERI funded projects are part of another R&D program called Generation IV. The goal here is to design reactors that are safer to operate, create less radioactive waste that can't be converted to weapons production, and are modular to allow economical deployment worldwide. Researchers aim to have Generation IV plants on-line between 2020 and 2030, or sooner.

Here are two Generation IV reactor concepts:

Pebble-bed reactors
The pebble-bed reactor is the most viable reactor design option for near-term deployment, say researchers at the Massachusetts Institute of Technology (MIT) and a South African consortium called the Pebble Bed Modular Reactor (Pty) Ltd., headed by Eskom of South Africa with U.K.-based British Nuclear Fuel, Exelon Corp., Chicago, and the IDC (South African Industrial Development Corp.).

Both group's designs are based on a German HTGR reactor built over 20 years ago. And both are radically different from pressurized or boiling-water reactors currently in use in the U.S. Instead of conventional fuel rods, these pebble-bed reactors employ 330,000 tennis-ball-sized fuel elements or pebbles. Each pebble consists of an outer graphite matrix covering and an inner fuel zone. Fuel zones contain about 9 gm of enriched uranium separated into 15,000 particles or kernels called Triso fuel.

Each 0.5-mm-diameter uranium kernel is surrounded by a porous carbon buffer layer that accommodates fission gas release. A protective outer layer of silicon carbide, sandwiched between pyrolitic carbon, contains the fuel and radioactive decay products produced during fission.

The Triso fuel is said to withstand temperatures on the order of 2,000°C, well above the working temperature inside the reactor core.

The pebble-bed reactor houses the fuel pebbles, and 110,000 similarly sized graphite balls that act as neutron reflectors in a central column of pebbles in the core. Helium gas passed over the pebbles removes heat from the chain reaction, raising gas temperature to about 900°C. The heated helium directly or indirectly drives gas-turbine-driven generators to produce electricity.

How helium is used in the secondary power generating system or balance of plant (BOP) differs between the Eskom Pebble Bed Modular Reactor (PBMR) and MIT Modular Pebble Bed Reactor (MPBR). The Eskom design uses a direct Brayton cycle. Here the hot, high-pressure helium exits the reactor, enters the BOP, and directly powers the turbine generating systems. The generator set weighs approximately 28 tons and rotates at 3,000 rpm. It could mount vertically and would be supported entirely by two radial magnetic bearings and an axial magnetic bearing.

A heat exchanger or recuperator improves plant efficiency by heating the helium as it cycles back to the reactor. Ten of these modules ganged together and run by a centralized control center would produce 1,100 MW of power. This pebble-bed park would fit in an area of about the size of three soccer fields.

The use of a closed cycle and a helium-powered gas turbine with magnetic bearings gives the Eskom PBMR a thermal efficiency of approximately 45% compared to 33% for a conventional pressurized-water reactor (PWR).

Contrast this with the MIT MPBR. It uses an indirect Brayton cycle and an intermediate heat exchanger (IHX). The helium in the reactor is an independent closed cycle. Heat transfers from the reactor helium to the closed BOP helium system via the IHX.

According to MIT Professor, Andrew Kadak, the success of the MIT pebble-bed project hinges on the ability to package the reactor, the IHX, and the remainder of the BOP in such a way to let all modules be trucked to the site rather than by barge. The plant must also be easily assembled with minimal tooling and rework and operate in the same size footprint as conventional power plants. The modules can't be longer that 60 ft, nor bigger than 8 × 12 ft. And weight must not exceed 200,000 lb (trailer-truck capacity).

The MIT pebble bed has already undergone a number of design revisions. For example, it originally employed a two-shaft (turbine) design — one high-speed high-pressure (HP) turbine runs all three compressors sets. A second turbine driven by HP turbine exhaust runs the power generator.

Now MIT is evaluating a four-shaft system where one low-speed turbine powers a generator and three separate turbocompressor sets. The change limits shaft power of any one turbine to less than 36 MW to match current state-of-the-art turbomachinery. In addition, reducing the length of each turbocompressor set makes for a simpler plant layout, as each shorter shaft can be positioned in adjacent modules, horizontally or vertically.

To limit the IHX weight, the group plans to split the single IHX into six smaller IHX modules, each with its own containment vessel. This will bring each module to within the 200,000-lb truck limit. Splitting the IHX into smaller modules also eases maintenance. The recuperator will also be split into six modules for easier placement near the IHX modules.

Each identical module will be built in a centralized factory and modules can be interchanged. MIT's modular approach is said to have the best chance of competing with an industry benchmark — a 200-MW natural-gas-fired plant with an efficiency of 50%. The MIT pebble bed should be about 45% efficient. It will run continuous thanks to an online refueling system. Short maintenance outages are expected every five years or so, says Kadak.

The pebble bed reactor will produce about 110 MW or about 10% that of a conventional nuclear or about a fourth that for a fossil fuel plant. And PBMR cost/kW-hr reportedly will be on par or less than existing fossil fuel systems. Modularity will let energy suppliers tailor power plants to meet current and future energy needs. Unlike conventional construction methods, the MIT modules will be factory built, then assembled onsite. This should lower construction costs, improve quality, and speed construction.

Safety by design
Operators of current Generation II LWRs cope with loss of coolant (LOCA) accidents by active, often mechanical, means. Pebble beds, in contrast, are said to be "naturally" safe and cool on their own using only conduction, convection, and radiative heat transfer without the need for emergency core cooling systems to prevent core failure or meltdown. Best of all, pebble bed reactors experiencing LOCA reportedly will release little if any radioactivity to the environment to the extent that no offsite emergency plan would be required.

Besides being safer to operate, pebble bed reactors use fuel more efficiently and can achieve higher fuel utilization (higher burn up) than conventional nuclear plants. Because pebble bed reactors use more of the fuel, once discharged there is less fissile material left that can be extracted for weapons use. This, along with the protective silicon carbide armor surrounding the Triso fuel, makes the pebble bed design more proliferation resistant.

The near term
If the proposed test plant near Capetown, South Africa, passes technical review, Eskom will apply to the National Nuclear Regulator (NNR) of South Africa for an early site permit next year and construction and operating licenses later in 2002 or 2003.

MIT is collaborating with the Idaho National Engineering and Environmental Laboratory (INEEL). MIT plans (within the next five years or so) to build a full-scale research/demonstration reactor at the INEEL. The goal: Show that even with a maximum LOCA the reactor core won't melt. According to Dr. Kadak this so-called "License by Test" approach would help boost public confidence in the pebble bed technology and dramatically reduce licensing costs and accelerate commercialization.

This approach is unlike conventional NRC licensing that approves a plant design on paper before it's built. The INEEL has conducted many similar LOCA tests on small-scale reactors but the MIT MPBR would be the first full-scale reactor to under go such rigorous testing.

Liquid-metal-cooled reactor
Another Generation IV design from The University of Tennessee (UT) is a liquid-metal lead/ bismuth (PbBi)-cooled reactor based on sodium-cooled, fast-breeder reactors developed in the 1950s.

According to Professor Laurence F. Miller, UT proposes a two-loop primary system with no intermediate heat exchanger. Corrosion caused by the liquid metal limits the maximum temperature for PbBi to less than 550 to 600°C, depending on construction materials. This is low for state-of-the-art supercritical secondary systems used in fossil fuel plants, but it is suitable for superheated-steam secondary systems.

The UT PbBi design, is typical of established fossil fueled systems. Miller projects that an overall thermal efficiency of greater than 40% is possible with a core inlet temperature of 300°C and an outlet temperature of 550°C.

Miller says one advantage of PbBi as a coolant is that the reactor can use materials with atomic numbers ranging from 90 to 103 as fuel, which eliminates waste disposal. The design also operates about eight years without refueling or reshuffling of the fuel. And time between refueling is limited by the fuel materials rather than the ability to sustain a chain reaction.