By Septimus van der Linden, Principal, Brulin Associates
South Africa’s electricity shortage has made headlines around the world since January when a near total collapse of the electricity grid occurred. The situation is improving only slowly and the country’s energy shortage has been declared a national emergency. Spot outages have shut down the nation’s lucrative gold mines and rationing has affected coal production, the major fuel for the country’s electricity generators. Therefore, recent approval from South Africa’s cabinet of the country’s nuclear policy is a welcome and necessary measure for South Africa’s major utility, Eskom. The approved policy will help the country reduce its reliance on large fossil plants by allowing nuclear technology to play a greater role in alleviating the critical power shortage.
Eskom intends to take full advantage of the approval by building a 165 MW closed-cycle gas turbine demonstration plant that will be powered by a high-temperature, helium-cooled pebble bed modular reactor (PBMR). With regulatory clearance, the demonstration plant is on track to begin construction in 2010 and commence operation in 2014. Based upon the demonstration plant’s successful completion and operation, the South African government committed to 20 to 30 of these 165 MW reactors. To expedite future shipments and help reach this commitment, Eskom has already ordered 24 PBMR plants rated at 165 MW. These reactors should begin domestic commercial operation in 2016.
The demonstration plant is being designed by PBMR (Pty) Ltd. in South Africa and is being built at Eskom’s existing Koeberg nuclear plant station near Cape Town. It will be a closed-cycle design rated at 165 MW full load output, with about 41.2 percent efficiency. At partial load output, the efficiency will fall to about 20 percent.
Undoubtedly the cost will escalate considerably, as has been seen with large-scale new build power plants planned for the United States, which are heading for $4,500 to $6,000 per kW or more. Smaller, well-dispersed plants of PBMR-type make sense, even if the original cost estimates are subject to the same commodity price increases as other power plant projects.
A major advantage of this design is that it is small enough for modules to be largely preassembled. In addition, it allows flexibility in matching incremental load growth and can be operational within 24 to 30 months from start of construction—compared with seven or eight years for pressurized water reactor (PWR) plants.
The 165 MW plant is designed around a high temperature 400 MWth reactor core, with helium coolant and a direct Brayton closed-loop helium cycle that powers a single-shaft turbine design developed by Mitsubishi Heavy Industries (MHI) (Figure 1) under a design and development contract. The turbine directly drives high- and low-pressure compressors at 6,000 rpm for the helium flow and also drives the electrical generator through a planetary reduction gear for 50-Hz or 60-Hz power generation.
Planned testing will incorporate the turbine driver coupled into the loop and the reactor for cold flow tests, which are expected to take four to six months to complete. This critical path work will include data gathering, predictable performance characterization and measurements to confirm design team simulations and system controls before starting nuclear operation. This could be modified based on HTF (High Temperature Facility—the helium high-temperature closed-loop component test facilty at Pelindaba, South Africa) and MHI factory tests that will shorten the site test period.
Given that the demonstration plant has not yet been built much less operated, there is a high degree of uncertainty about projected equipment and cost of electricity estimates. Although the increases in commodity prices have impacted power plant costs worldwide, some generalizations can be drawn to make broad comparisons. For instance, the 165 MW PBMR as a small stand-alone system will cost more to build per MW than large centralized 1,000 MW to 1,500 MW nuclear power plants. However, the demo plant is being redesigned for greater modularity and factory preassembly, which should reduce the unit cost of commercial PBMR plants by 15 percent to 20 percent. Furthermore, shorter construction times and earlier deployment are added benefits.
Designers expect the cost of electricity to also be higher. Several 1,000 MW nuclear plants now under construction in China for 2015 start-up are expected to have a cost of electricity of $45 to $55 per MWh vs. $55 to $60 per MWh projected for PBMR generation. But those large nuclear plants also take seven to nine years to build before they start generating revenue vs. three to four years for a 165 MW stand-alone pebble bed plant installation and two to three years for additional units at the same site. The delta differential might decrease depending on overall cycle and execution improvements.
PBMR’s power plant division management believes several unique advantages of PBMR technology and plant design more than mitigate the higher relative cost of electricity compared with 1,000 MW and 1,500 MW nuclear installations. The economic advantages include lower transmission losses (can be sited close to load centers), use of helium cooling (eliminates dependency on water availability), much smaller up-front financing requirements (less risk), faster revenue flow (shorter lead time), and immunity to the cost of unplanned outages.
One outstanding feature of the pebble bed reactor design is that a core meltdown is virtually impossible. If the flow of helium coolant is interrupted or interferes with the safety operation of the fission reaction control rods, the integrated heat loss from the reactor vessel automatically exceeds the decay heat production in a postaccident condition.
As a result, the reactor itself will inherently shut down and eventually cool down naturally. In addition, the peak temperature that can be reached in the core (1,600 C under the most severe conditions) is well below the temperature that may cause damage to the fuel. This is because the radio nuclides, which are the potentially harmful products of the nuclear reaction, are contained by two layers of pyrocarbon and a layer of silicon carbide that are extremely robust at withstanding high temperatures.
The PBMR fuel consists of particles of enriched uranium dioxide coated with silicon carbide and carbon. These particles are then encased in graphite to form a fuel sphere (or pebble) about the size of a billiard ball (Figure 2). A nuclear fission reaction within the silicon carbon particles encased within the fuel spheres generates heat which is emitted into the space between the fuel pebbles in the reactor core.
Heat generated by the fission reaction is removed when the helium gas, which enters the reactor vessel at about 500 C and 9 MPa (mega Pascal) pressure, flows down between the hot fuel spheres and cools them. The gas then leaves the bottom of the vessel at a temperature of about 900 C and powers the closed-cycle power turbine to drive the electric generator, as well as low- and high-pressure compressors.
MHI developed the helium power train, consisting of the helium turbine and compression train. MHI will also manufacture the turbomachinery for the commercial PBMR production units.
In MHI’s horizontal single-shaft turbine design, all the turbo units operate at 6,000 rpm shaft speed, with the power turbine driving low- and high-pressure compressors. The power turbine also drives the electrical generator through a speed-reducing gear that can serve both the 50-Hz and 60-Hz markets. This solution replaces the original concept of separate vertical turbo-compressor and power turbine units.
This speed-reducing gear, designed by Renk to AGMA standards, will be a planetary gear conservatively designed to provide a high reserve margin of performance.
Two dry gas seals at each shaft casing penetration dividing the high- and low-maintenance areas will separate the compressor and power turbine casing.
Another advantage of this power train drive is its ability to use proven oil-lubricated bearings instead of electromagnetic bearings. This means thrust forces are easier to deal with and maintenance is similar to conventional gas turbine equipment.
After exhausting through the turbine, the 500 C helium passes through a series of precoolers, intercoolers and compressors (to repressure the system) before being returned to the reactor for heating. Because the helium working fluid circulates within a closed cycle indirectly heated by the nuclear reactor, turbine and compressor airfoils remain exceedingly clean throughout their operating lives.
Theoretically, however, a remote possibility of plating from AG 110m due to silver neutronic activation at 900 C and higher temperatures exists. As a precaution, Eskom will closely monitor the demonstration plant operation over time to ensure this does not occur. The single-loop helium direct Brayton cycle also has a much higher specific power density factor than equivalent sized open-cycle industrial gas turbines and is therefore much more compact.
PBMR entered into an agreement with Stellenbosch University of South Africa that will enable the university to conduct research and development on behalf of PBMR in areas that will enhance the reactor’s performance and safety. One area is a contribution that will help researchers and designers understand the dynamics of the silver isotopes that could migrate out of the fuel during operation.
For example, the demo plant at a 3.2-to-1 pressure ratio is design rated to flow at 185 kg per second, which works out to 892 kW per kg of flow per second. By contrast, a current technology industrial gas turbine rated for 165 MW power output, operating at an 11.7-to-1 pressure ratio and 528 kg per second air flow, would have a power density factor of 312 kW per kg of flow per second.
Because of direct closed-cycle characteristics, the PBMR plant is able to operate at high efficiency in a 20 percent to 100 percent load range. It can also be idled at very low energy consumption, which will allow the operators to rapidly start it and bring it up to load after an overnight or weekend shutdown. This feature cannot be duplicated by larger units and makes the PBMR an ideal add-on to existing nuclear sites to deal with variability of renewable energy sources such as wind and solar.
Some ramp-up restrictions exist when operating below 50 percent output, but they are relatively minor and the plant can be considered fully load-following above 50 percent power output.
Online refueling is another key feature of the PBMR. The reactor does not need to be taken out of service for refueling. New and recycled fuel is introduced at the top of the reactor while used fuel is removed at the bottom to keep the reactor operating at full power. Fuel pebbles continuously travel through the core from the top to the bottom during normal plant operation.
Once a pebble exits the bottom of the reactor, staff will measure it and test it to ensure that it conforms to physical integrity specifications. Staff will also evaluate the amount of burn-up. Pebbles in good physical condition that have not reached the target burn-up will be returned to the top of the reactor for reintroduction into the core. Rejected pebbles will be pneumatically transported to spent fuel tanks for storage. The aim is to operate the reactor uninterrupted for six years before shutting it down for scheduled maintenance. However, for the demonstration plant, a number of interim shutdowns have been scheduled so the staff can evaluate components and system performance.
The reactor will be shut down by fully inserting the control rods. It will be made critical by withdrawing the control rods. Then, by running the electrical generator as a starting motor, the Brayton cycle (consisting of the reactor, turbine, coolers, recuperators and compressors) can be idled to attain a self-sustaining speed working off the reactor’s residual heat.
The first phase of the South African PBMR project entails building the 165 MWe (400 MWth) power plant at Koeberg and a fuel plant at Pelindaba near Pretoria. The power plant form the basis for the multimodule electricity plants and also the follow-on process heat plants.
Eskom is building the demonstration plant so that it can evaluate and confirm functional integrity and design performance before placing the reactor into commercial service. The utility will be studying total plant availability, outage management, online maintenance of critical equipment, and ease of achieving the six-year maintenance interval between general overhauls.
The operational demonstration’s scope includes consistent and predictable base load operation, load following, transient characteristics, load rejection, cycle efficiencies, fuel handling and design performance of helium turbine and compressors, gearbox and generator. The reactor’s dynamics will also be monitored to ensure consistent and predictable operation and performance under different operating regimes.
Of equal importance will be the demonstration of key commercial performance parameters such as construction costs, plant availability and efficiency, operating and maintenance costs and midlife upgrades.
Once Eskom terms the demonstration at Koeberg a success, the next step will be to begin the process to commercialize the units and make them available as early as 2016. This will allow Eskom to distribute nuclear generation in small increments around the country close to industrial development areas without having to build extensive transmission lines.
To compete with 1,000 MW and larger nuclear facilities, project engineers are also working on a four-pack PBMR configuration that will provide 660 MW of capacity. This will allow the reactors to step into the role that gas-fired combined-cycle plants now perform, something the large baseload nuclear plants are not capable of doing. Future cycle efficiency improvement is possible by adding a bottoming cycle to recover the “waste” heat from the precooler (90 MWth) and the intercooler (120 MWth) that are incorporated in the helium loop. Organic cascading closed-loop Rankine cycle avoids water use, adding potentially 30 MWe or more.
Applications and Markets
In addition to traditional electricity generation, the pebble bed reactor design is also well suited to diverse process heat applications. Single or multiple units can be easily located at current permitted nuclear sites. They can do fast load following and can be “parked” over the weekend, as well as operate at part load to meet the demand cycle requirements.
Though the reactor core dimensions will remain the same for different process heat applications, the technology can be differentiated into two configurations depending on the reactor outlet temperature:
- An intermediate-temperature gas-cooled reactor (TGR), operating at reactor outlet temperatures up to 750 C.
- A high-temperature gas-cooled reactor (HTGR), operating at reactor outlet temperatures up to 950 C. This is essentially a so-called very high temperature reactor (VHTR), which is generally expected to operate at temperatures in the 900 C+ temperature range.
Intermediate temperatures can be used to generate process steam for cogeneration applications, in-situ oil sands recovery, ethanol applications and refinery and petrochemical applications. High temperatures can be used to efficiently coproduce electricity to reform methane to produce syngas (where the syngas can be used as feedstock to produce hydrogen, ammonia and methanol); and to produce hydrogen and oxygen by decomposing water thermochemically. Hydrogen can be sold as a merchant product or directly supplied to various industrial operations such as coal-to-liquids, coal-to-gas, refineries, petrochemical applications and steel production operations.
Lower temperature waste heat can be used to produce water via desalination processes.
The size of this potential market is illustrated by the fact that the global installed capacity of desalination plants is more than 35 million cubic meters per day and growing by 7 percent a year. There are more than 12,500 desalination plants worldwide, most of which use fossil fuel energy sources.
In South Africa, there is interest in the possible use of PBMR’s technology in petrochemical complexes, notably for the synthetic fuels giant Sasol, to either produce process steam and/or hydrogen to upgrade coal products.
In Canada there is interest from oil sands producers to use PBMR to produce the temperature and associated pressure needed to extract bitumen from oil sands instead of gas-fired plants currently in use.
In the United States, PBMR is a partner in the Westinghouse-led consortium awarded a contract by the U.S. Department of Energy to consider the technology as heat source for producing noncarbon hydrogen. This Next Generation Nuclear Plant (NGNP), which aims to use HTGR technology to produce hydrogen and electricity, is still in its preconceptual phase, but it could result in the construction of a South African-designed PBMR in the United States before the end of the next decade.
Chinese developers are working with PBMR (Pty) Ltd. through Chinergy Co. of Beijing to pursue the development of high-temperature reactor demonstration projects. (See “Powering Up a Growing Nation” on page 21 in Nuclear Power International’s June issue for more detail.)
Attractive applications for nuclear process heat are driven primarily by the opportunity to displace natural gas and other premium fuels and to respond to incentives to reduce CO2 emissions. Even with conservatively low forecasts for growth in long-term gas prices, clear commercial benefit exists in reducing exposure to gas price volatility and rapid increase. Economic assessments of PBMR process heat applications, based on current trends, have confirmed that PBMR is likely to become economically competitive in many markets, especially markets with high premium fuel costs and CO2 emissions.