NEW NUCLEAR REACTOR DESIGNS

by Ron Hagen

Overview

This article summarizes nuclear reactor designs that are either available or anticipated to become available in the United States by 2030. Criteria for including reactors are: 1) participation or likely participation in the U.S. Nuclear Regulatory Commission's design certification or pre-certification programs; and 2) inclusion under the Generation IV International Forum (GIF) program for longer-term reactor development. The U.S. Department of Energy is among the sponsors of the GIF program. While no detailed technical description of particular reactor designs is included, such descriptions and schematics are available elsewhere and, when practical, some of these are hyperlinked in the text. Reactor vendors who put forward new designs anticipate that their designs will meet commercial market needs including design safety and affordable, competitive construction costs while maintaining the usually low operating costs of today’s commercial nuclear reactors. This paper does not assess such views, though a section does identify public discussion of efforts within the nuclear industry and the U.S. government to improve the industry's competitive position.1

Existing Reactor Designs and Design Categories

There are now 104 fully licensed nuclear power reactors in the United States, though only 103 are now operational.2 Because each of these reactors is fully licensed and meets national safety standards, a potential builder might choose to replicate any of these designs for future construction. This is less likely, however, because existing, operable reactors in the United States were initiated during or before the 1970s. Technology has progressed and any future construction is likely to incorporate more advanced designs intended to better meet today's commercial and safety criteria.

There are possible exceptions to the preceding statement. Four reactors in the United States were partially built and still possess valid construction licenses. These reactors are WNP-1 in Washington State (Energy Northwest), Watt's Bar 2 in Tennessee (Tennessee Valley Authority), and Bellefonte 1 and 2 in Alabama (TVA). Moreover, these construction licenses have been extended approximately to the end of the present decade. Construction on each unit was halted almost two decades ago. Builders of these units, subject to the rules of their licenses, have the right to resume construction on their reactors that were designed during the 1970s or earlier. Whether the construction under these existing designs will resume and whether former designs will be continued remains to be determined, but appears unlikely. The owners of WNP-1 have indicated an intention to forgo their construction license to allow for eventual disassembly and clearance of present facilities.

All existing commercial nuclear reactors operating in the United States fall into two broad categories, pressurized water reactor (PWR) and boiling water reactor (BWR). Because both types of reactors are cooled and moderated3 with ordinary "light" water, the two designs are often grouped collectively as light water reactors (LWR). LWRs generate power through steam turbines similar to those used for most power generated by burning coal or fuel oil. Light water reactors have so far proven to be the most commercially popular reactor design worldwide though there are notable exceptions.4 There are several available websites that discuss existing reactors in the United States. These include http://www.eia.doe.gov/cneaf/nuclear/page/nuc_reactors/reactsum.html. Information on international operating reactors is available at http://www.iaea.org/programmes/a2.

PWRs use nuclear-fission to heat water under pressure within the reactor. This water is then circulated through a heat exchanger (called a "steam generator") where steam is produced to drive an electric generator. The water used as a coolant in the reactor and the water used to provide steam to the electric turbines exist in separate closed loops that involve no substantial discharges to the environment. Of the 104 fully licensed reactors in the United States, 69 are PWRs. Westinghouse, Babcock and Wilcox, and Combustion Engineering designed the designed the nuclear steam supply systems (NSSS) for these reactors. After these reactors were built, Westinghouse and Combustion Engineering nuclear assets were combined with British Nuclear Fuels Limited to form Westinghouse BNFL. The French-German owned firm Framatome ANP has acquired many of Babcock and Wilcox's nuclear technology rights, though portions of the original Babcock and Wilcox firm still exist and possess some nuclear technology rights as well. Other major makers of PWR reactors, including Framatome ANP, Mitsubishi, and Russia’s Atomstroyexport, have not yet sold their reactors in the U.S. A schematic diagram of a PWR can be found at http://www.eia.doe.gov/cneaf/nuclear/page/nuc_reactors/pwr.html.

The remaining 35 operable commercial nuclear reactors in the United States are BWRs. BWRs allow fission-based heat from the reactor core to boil the reactor’s coolant water into the steam that is used to generate electricity. General Electric built all boiling water reactors now operational in the United States. Framatome ANP and Westinghouse BNFL have each designed BWRs. These have not yet been sold in the United States. A schematic diagram of a BWR can be found at http://www.eia.doe.gov/cneaf/nuclear/page/nuc_reactors/bwr.html.

Although no LWR projects have been initiated in the United States since the 1970s, the overall performance record of the existing fleet has been reasonably successful. Some 111 LWRs have entered service in the U.S. since 1969.5 Only seven of those have been permanently shut down. The average annual capacity factor for nuclear reactors in the United States has been around 90 percent during the 2000's. Average operating costs, as reported by the Federal Energy Regulatory Commission, are slightly lower for LWRs than for operating coal-fired plants and considerably below operating costs for gas-fired plants.  Fuel costs for LWRs are particularly low.6

There have been attempts to operate additional classes of reactors in the United States, though most examples were prototypes and were not commercial successes. Perhaps the most famous example was the Fort Saint Vrain reactor that operated between 1974 and 1989. It was a high temperature gas-cooled reactor or HTGR. Other HTGRs operated elsewhere, notably in Germany. HTGRs, of which there are many sub-categories, continue to stimulate commercial interest. HTGR designs are promoted by firms in China, South Africa, the United States, the Netherlands, and France.  There is some interest in building commercial HTGRs in several nations including South Africa and China.  Small research prototypes already exist in Japan and China. HTGRs use a gas- helium has been preferred- to generate electricity. In some cases the turbine is run directly by the gas, in other cases steam or alternative hot gases such as nitrogen are produced in a heat exchanger to generate the power. HTGRs are distinguished from other gas-cooled reactors by the higher temperatures attained within the reactor. Such higher temperatures might permit the reactor to be used as an industrial heat source in addition to generating electricity. This improves HTGR’s suitability for commercial hydrogen production. Advocates of HTGR designs hold that HTGRs have high safety, low costs, and a potential to supply power to smaller markets than do LWRs.  HTGRs also are reputed to adapt better to changing load requirements of electricity markets than LWRs.

Additional commercial reactor designs that operate outside of the United States include fast breeder reactors (FBRs), pressurized heavy water reactors (PHWRs), and gas-cooled reactors (GCRs). FBRs have received much research funding but only limited market support.  A "commercial" unit still operates in Russia and prototypes exist elsewhere, notably France, Japan, and India. China also intends to build a prototype FBR while India and Russia are building FBRs that might be described as commercial.  "Breeder" or "fast" reactors have advantages because U-235 is the only naturally occurring uranium isotope that is directly suitable for commercial energy production. U-235 is only 0.7 percent of natural uranium.7 Most natural uranium is the U-238 isotope that is not directly usable as a reactor fuel. During the course of any reactor’s operation a portion of the U-238 in the fuel is converted to plutonium, primarily the useful Pu-239 isotope, which provides a large portion of the energy used in nuclear power production. The bulk of the U-238 content in a commercial reactor is typically not converted to plutonium nor does it contribute significantly to electricity production.  A breeder reactor converts more U-238 to usable fuels than the reactor consumes. Any unused fuel produced by this procedure would have to be "reprocessed" before some of the plutonium and the remaining U-235 and U-238 might again be usable as a reactor fuel. FBRs have, so far, proven to be more expensive to build and operate than LWRs. It is unclear whether this is because most FBRs have been prototypes or if this reflects underlying costs. The plutonium content of the spent and reprocessed fuel also raises concerns over weapons proliferation. Many earliest FBR designs experienced system failures, though some, notably the BN-600 in Russia, have operated reliably over extended periods. Proponents of advanced reactor designs believe that some commercial FBR designs could be deployed prior to other highly advanced, though untested reactor designs.8

PHWRs have been promoted primarily in Canada and India, with additional commercial reactors operating in South Korea, China, Romania, Pakistan, and Argentina. Canadian-designed PHWRs are often called "CANDU" reactors.9 Siemens, ABB (now part of Westinghouse), and Indian firms have also built commercial PHWR reactors. Commercial heavy water reactors now in operation use heavy water as moderators and coolants. No successful effort has been made to license PHWRs in the United States. PHWRs have proven to be popular in several countries because they use less expensive natural (not enriched) uranium fuels and can be built and operated at competitive costs. PHWRs have often been preferred by nations wishing to develop an indigenous fuel cycle without expensive enrichment facilities. The continuous process of refueling PHWRs have raised some proliferation concerns as has the high Pu-239 content of the spent fuel.  PHWRs, like most reactors, can use fuels other than uranium.  Particular interest has been shown in thorium-based fuel cycles.10

The term gas-cooled reactor (GCR) can be used ambiguously. HTGRs, for example, are a subset of GCRs that operate at higher temperatures. As used here, GCRs include "Magnox" reactors designed and built in the United Kingdom since the 1950s and the derivative, advanced gas-cooled reactor (AGR), also operated in the United Kingdom. Similar reactors had been built and operated in France, Sweden, and Japan but have since closed. No GCR design, as defined here, has operated commercially in the United States. Commercial GCRs11 in the United Kingdom have operated longer than any category of commercial reactors anywhere else in the world. Like the PHWRs, the original GCR designs use natural uranium fuels, though newer designs (AGRs) use slightly enriched fuels and are not confined to uranium fuels.12

Other potential designs for commercial reactors abound. They have not been widely or recently considered for commercial applications in the United States. There is some experience with additional concepts elsewhere and at research facilities.

New Designs

1. Certified Designs

In recent years, the Nuclear Regulatory Commission (NRC) has set up a process by which reactor designs might be certified prior to any actual construction plans. The certification process seeks to reduce site development time by resolving design issues prior to construction. Design certification is an optional process and might occur simultaneously with site licensing or construction licensing.  Normally reactor certification is the responsibility of the reactor vendor rather than any utility that might choose to build a new reactor.

Certification Process for New Reactors in the United States
Reactor Design Lead Vendor(s) Design Category Status at NRC
System 80+ Westinghouse BNFL PWR Certified
ABWR GE, Toshiba, Hitachi BWR Certified
AP600 Westinghouse BNFL PWR Certified
AP1000 Westinghouse BNFL PWR Finalizing Certification
ESBWR GE BWR Pre-certification
SWR-1000 Framatome ANP BWR Pre-certification, deferred
ACR700 AECL PHWR/PWR hybrid Pre-certification
PBMR Eskom HTGR Pre-certification, deferred
GT-MHR General Atomic HTGR Pre-certification
IRIS Westinghouse BNFL PWR Pre-certification
EPR Framatome ANP PWR Pre-certification
ACR1000 AECL PHWR/PWR hybrid No application decision
4S Toshiba Sodium-cooled No application decision
Note: Reactor design names are defined in the text. ESBWR, ACR700, EPR and IRIS vendors have indicated intentions to begin certification in the near future.

Any new reactor built in the United States over the next decade or so would most likely use designs either recently certified by the NRC or that will be certified by the NRC in the near future. (Design approval can alternatively coincide with construction and operation licensing, skipping the certification process.)  The re-creation of older designs is popular overseas and cannot be ruled out in the United States. Presently there are three certified new reactor designs in the United States: the System 80+, the Advanced Boiling Water Reactors (ABWR), and the AP600. These designs are sometimes called Advanced Light Water Reactors (ALWR) because they incorporate more advanced safety concepts than the reactors previously offered by vendors. They are also sometimes called Generation III reactors to distinguish them from earlier designs now operating in the U.S. and globally and from later designs now seeking certification which are sometimes called Generation III plus. Design certifications can expire if not supported by a vendor.

System 80+ (Westinghouse BNFL): The System 80+ reactor is a PWR designed by Combustion Engineering (CE) and by CE's successor owners ABB and Westinghouse BNFL. The NRC has certified the System 80+ for the U.S. market, but Westinghouse BNFL no longer actively promotes the design for domestic sale. The System 80+ provides the basis for the APR1400 design that has been developed in Korea for future deployment and possible export. Information on the System 80+ reactor can be found on http://www.nei.org/index.asp?catnum=3&catid=703 and http://www.nuc.berkeley.edu/designs/sys80/sys80.html.

 

ABWR (General Electric, Toshiba, Hitachi): Among the three NRC- certified ALWR designs only the ABWR has been deployed. Three ABWRs operate in Japan, and three are under construction, two in Taiwan and one in Japan. While the ABWR design is usually associated in the United States with General Electric, units now being built in Japan are products of Toshiba and Hitachi. Toshiba, and Hitachi frequently associate with General Electric in possible ABWR projects in the U.S.  There are many variations in ABWR design. The most frequently mentioned capacities are in the 1250-1500 MWe range though smaller and larger designs have been proposed depending on the vendor. Vendors now claim costs for building the ABWR that are low enough that they have attracted some customer interest.  Information on the ABWR can be found at http://www.nei.org/doc.asp?docid=110, and http://www.nuc.berkeley.edu/designs/abwr/abwr.html

 

AP600 (Westinghouse BNFL): The AP600 is a 600 MW PWR certified by the NRC. The AP600, while based on previous PWR designs, has innovative passive safety features that permit a greatly simplified reactor design. Simplification has reduced plant components and should reduce construction costs. The AP600 has been bid overseas but has never been built. Westinghouse has deemphasized the AP600 in favor of the larger, though potentially less expensive (on a kilowatt basis) AP1000 design. Information on the AP600 can be found at http://www.ap600.westinghousenuclear.com/ and at http://www.nei.org/index.asp?catnum=3&catid=704.

The initial ALWR designs as a group have been praised for their improvements in reactor safety and simplicity, but construction costs on a “per kilowatt of capacity” basis might remain a barrier to commercial success in the U.S. The ABWR design however has many variations and continues to be selectively promoted by several vendors. It has been evaluated, along with other designs, for construction at Bellefonte by the Tennessee Valley Authority (TVA).

2. Undergoing Certification

Only one reactor design, the AP1000, is presently undergoing certification with the NRC. This situation could change shortly as additional designs move from "pre-certification" to actual "certification". The certification process is anticipated to begin for several additional designs during 2005 and 2006.  Designs that vendors anticipate submitting for certification during the next two years include the ESBWR, the ACR700, the EPR and IRIS.  The process of certification takes several years and depends heavily on how unique the proposed design is and whether the design is supported by potential vendors and buyers.  NRC hearings have emphasized that new and innovative designs might take more time for certification because of limited NRC staff familiarity with the designs.

AP100013 (Westinghouse BNFL): Quite often when a reactor is named, its name includes digits such as the "1000" in the AP1000. This usually indicates the initial electricity generating capacity of the design, in this case 1000 MWe. Seldom do the digits indicate the present design capacity as the design evolves. The most recent AP1000 design has been bid in China with a 1175 MW-capacity. The AP1000 is an enlargement of the AP600, designed to almost double the reactor's target output without proportionately increasing the total cost of building the reactor. Westinghouse anticipates that operating costs are anticipated to be below the average of reactors now operating in the United States. While Westinghouse BNFL owns rights to several other designs, the AP1000 is the principal product that the company now promotes in the United States for near term construction. The AP1000 is a PWR with innovative, passive safety features and a much simplified design intended to reduce the reactor’s material and construction costs while improving operational safety. One consortium of nine utilities called NuStart Energy promotes the AP1000 in the United States and has informed the NRC that it intends to apply for a combined construction and operating license (COL) for the design.  This is not a commitment to build the design. Westinghouse submitted a bid in early 2005 to build as many as four AP1000s at two sites in China.  Information on the AP1000 can be found at http://www.nei.org/doc.asp?docid=770. Information related to NRC certification for the AP1000 can be found athttp://www.nrc.gov/reactors/new-licensing/design-cert/ap1000.html.

3. Undergoing Pre-Certification

While pre-certification is a technical concept within the NRC regulatory environment, the process can mean many things to potential reactor vendors. Concepts such as the ESBWR, and the ACR700 appear to be much further along toward certification than the other designs.14 The French designed EPR is undergoing construction in Finland and has recently moved to pre-certification.  Pre-certification represents a vendor's intention to proceed toward commercialization in the U.S. and perhaps globally. Pre-certification is a less expensive early stage of the certification process. Actual certification procedures are much more complex.  At an early stage in pre-certification most NRC regulatory costs are borne by the applicant.

ESBWR (Economic Simplified, Boiling Water Reactor) (General Electric): The ESBWR15 is a new simplified BWR design promoted by General Electric and some allied firms. The ESBWR constitutes an evolution and merging of several earlier designs including the ABWR that are now less actively pursued by GE and other vendors beyond the exceptional case of Bellefonte in Alabama. The intent of the new design, which includes new passive safety features, is to cut construction and operating costs significantly from earlier ABWR designs. GE and others are investing heavily in the ESBWR though the design might not be available for deployment for several years.  The ESBWR’s builders however anticipate that the design will be available in time to meet any potential construction targets in the U.S. The nine-utility NuStart Energy group promotes the ESBWR as well as the AP1000 design.  NuStart has informed the NRC that it intends to apply for a COL for the ESBWR in addition to any AP1000 application.  Dominion Resources is also evaluating the ESBWR for its North Anna plant in Virginia but has not declared its COL intentions for the design.  Information related to certification of the ESBWR can be found at http://www.nrc.gov/reactors/new-licensing/design-cert/esbwr.html.

 

Siedewasser Reaktor (SWR-1000) (Framatome ANP): The SWR-1000 is a Framatome ANP design for an advanced BWR. Framatome ANP was created through the merger of the French nuclear vendor Framatome and the nuclear power assets of the German firm Siemens. The SWR-1000 was originally designed by Siemens. Framatome ANP began SWR-1000 pre-certification with the NRC several years ago. The SWR-1000 presently has no U.S. utility sponsor and is no longer being actively promoted by Framatome which now emphasizes its EPR design. Literature on the design notes the reactor's passive safety features. Passive safety also potentially mean lower construction costs though this has not been as heavily promoted by Framatome. Information on the SWR1000 can be found on http://www.de.framatome-anp.com/anp/e/foa/anp/products/s112.htm. Information related to certification of the SWR-1000 can be found at http://www.nrc.gov/reactors/new-licensing/design-cert/swr-1000.html.

 

ACR700 (Atomic Energy of Canada Limited): AECL's "Advanced CANDU Reactor" ACR70016 has been developed over a lengthy period of time and is considered by its vendor to be an evolution from AECL's internationally successful CANDU line of PHWRs. CANDU reactors and their Indian derivatives have been more of a commercial success than any other line of power reactors except the LWRs. One of the innovations in the ACR700, compared to earlier CANDU designs, is that heavy water is used only as a moderator in the reactor. Light water is used as the coolant. Earlier CANDU designs used heavy water both as a moderator and as a coolant. This change makes it debatable whether the ACR700 is a PHWR, a PWR, or a hybrid between the two designs. AECL has aggressively marketed the ACR700 offering low prices, short construction periods, and favorable financial terms. As is the case for most non-LWR reactors, most U.S. utilities, nuclear engineers, and regulators have only limited working familiarity with the design. Interest was initially shown by Dominion Resources regarding possible construction at North Anna (Virginia) as well as by utilities in several international locations, notably in Canada and the United Kingdom. Dominion has recently switched to the ESBWR design for North Anna in anticipation of the slow regulatory approval process for the innovative Canadian-design.  AECL has subsequently slowed its efforts to certify the ACR700 in the United States though the firm still intends to begin the certification process toward the end of 2005.  AECL announcements indicate increased interest in a larger ACR1000 design.  Information on the ACR700 can be found on http://www.aecltechnologies.com/Content/ACR/default.htm and http://www.aecl.ca/index.asp. Information related to certification of the ACR-700 can be found at http://www.nrc.gov/reactors/new-licensing/design-cert/acr-700.html

 

Pebble-bed Modular Reactor  (PBMR) (Eskom): The PBMR, which uses helium as a coolant, is part of the HTGR family of reactors and thus a product of a lengthy history of research, notably in Germany and the United States. More recently the design has been promoted and revised by the South African utility Eskom and its affiliates. Westinghouse BNFL is a minority investor.  Prototype variations of the PBMR are now operating in China and Japan.  Eskom has received administrative approval to build a prototype PBMR in South Africa, but has also been delayed in implementation by judicial rulings regarding the reactor’s potential environmental impact. Certification procedures in the U.S. have slowed, but never have been abandoned. At around 165 MWe the PBMR is one of the smallest reactors now proposed for the commercial market. This is considered a marketing advantage because new small reactors require lower capital investments than larger new units.  Several PBMRs might be built at a single site as local power demand requires. Small size has been viewed as a regulatory disadvantage because most licensing regulations (at least formerly) required separate licenses for each unit at a site. The NRC also does not claim the same familiarity with the design that it has with LWRs.  Fuels used in the PBMR would include more highly enriched uranium than is now used in LWR designs. The PBMR design is considered a possible contender for the U.S. Department of Energy's Next Generation Nuclear Plant (NGNP) program in Idaho. China has also indicated interest in building its own variation of the PBMR. China and South Africa have also discussed cooperation in their efforts.  Details regarding the PBMR design can be found on https://www.pbmr.com/. Information related to certification of the PBMR can be found at http://www.nrc.gov/reactors/new-licensing/design-cert/pbmr.html.

 

Gas-turbine Modular Helium Reactor (GT-MHR) (General Atomic): The GT-MHR is an HTGR design developed primarily by the U.S. firm, General Atomic. The most advanced plans for GT-MHR development relate to building reactors in Russia to assist in the disposal of surplus plutonium supplies. Parallel plans for commercial power reactors would use uranium-based fuels enriched to as high as 19.9 percent U-235 content. This would keep the fuel just below the 20 percent enrichment that defines highly enriched uranium. In initial GT-MHR designs, the conversion of the energy to electricity would involve sending the heated helium coolant directly to a gas turbine. There has been concern regarding untested, though non-nuclear aspects of this generation process. This has led potential sponsors to advocate similar ideas involving less innovative heat transfer mechanisms prior to generating electricity or commercial heat. The U.S. utility, Entergy, has participated in GT-MHR development and promotion and has used the name "Freedom Reactor" for the design. Because coolant temperatures arising from HTGRs are much higher than from LWRs, the design is viewed as an improved commercial heat source. There has been particular attention paid to the design's potential in the production of hydrogen from water. The GT-MHR is considered a potential contender for the US Department of Energy's Next Generation Nuclear Plant (NGNP) program. Information on the GT-MHR can be found on http://www.ga.com/gtmhr/. Information related to certification of the GT-MHR can be found athttp://www.nrc.gov/reactors/new-licensing/design-cert/gt-mhr.html.

 

International Reactor Innovative and Secure (IRIS) (Westinghouse BNFL led consortium): Westinghouse BNFL has promoted the IRIS reactor design as a significant simplification and innovation in PWR technology. The reactor design is smaller than most operating PWRs and would be much simplified. The IRIS reactor includes features intended to avoid loss of coolant accidents. Pre-certification is proceeding. The IRIS reactor may show potential during the next decade.  Certification could precede commercial availability.  IRIS has a targeted 2010 certification completion date. IRIS presently has no utility sponsor in the U.S. Information on the IRIS can be found on http://www.nei.org/index.asp?catnum=3&catid=712. Information related to certification of the IRIS can be found at http://www.nrc.gov/reactors/new-licensing/design-cert/iris.html.

European Pressurized Water Reactor (EPR) (Framatome ANP): Framatome ANP announced in early 2005 that it would market its EPR design in the United States and has recently begun pre-certification.  The EPR is a conventional, though advanced, PWR in which components have been simplified and considerable emphasis is placed on reactor safety. The design is now being built in Finland with a target completion during 2009. The French government also proposes building an additional EPR at Flamanville 3 in France.  Present French policy suggests that additional EPRs might replace additional commercial reactors now operating in France starting in the late 2010s. The EPR was bid in early 2005 in competition to the AP1000 for four reactors at two sites in China.  The proposed size for the EPR has varied considerably over time but might be around 1600 MWe.  Earlier designs were as large as 1750 MWe.  In either case the EPR would be the largest design now under consideration in the United States. Some redesign might occur for the U.S. market. Framatome had earlier indicated that U.S. certification for the EPR would occur after European development proceeded.  This decision has since been made and the U.S. utility Duke Power is evaluating the EPR, along with the AP1000 and ESBWR, for a COL application process that began during 2005.  A formal COL application by Duke would occur several years later though design selection might occur earlier.  Framatome has posted material on the EPR on http://www.framatome-anp.com/servlet/ContentServer?pagename=Framatome-ANP%2Fview&c=rubrique&cid=1049449651371&id=1049449651371.The NRC has not yet posted a status page for the EPR but one might be anticipated on http://www.nrc.gov/reactors/new-licensing/design-cert.html.

4. Anticipated for Possible Pre-Certification

Two designs, the ACR1000 and the 4S have not been formally submitted for pre-certification in the United States. Because of the attention that the designs are now receiving and their potential submission for certification, they are summarized below.

ACR1000 (Atomic Energy of Canada Limited): While AECL has promoted its ACR700 design, an ACR1000 has been designed as well.  If the scale economies attributed by Westinghouse BNFL to its AP series and by GE's its ABWR/ESBWR series are valid, one might anticipate parallel, cost-lowering results for the ACR series. Advertised costs for the ACR700 are already as low as any design proposed in the United States for the near term. Promised construction times, as short as three years, would set modern records for large reactor completion. When Dominion Resources indicated in late 2004 that it was no longer pursuing ACR700 construction at North Anna, AECL stated that while it will continue with ACR700 certification, perhaps in late 2005, more effort would be placed on the 1100+ MWe ACR1000 design. Information on the ACR1000 can be found on http://www.aecl.ca/index.asp?menuid=21&miid=519&layid=3&csid=294.

 

4S (Toshiba): The 4S is a very small molten sodium-cooled reactor designed by Toshiba.  The reactor presently being considered is 10 MWe though larger and smaller versions exist.  The 4S is designed for use in remote locations and to operate for decades without refueling.  This has led to the reactor to be compared with a nuclear “battery”. The use of molten-sodium as a coolant is not particularly new, having been used in many FBR designs.  Sodium-coolants allow for higher reactor temperatures.  Potential fuels are uranium or uranium-plutonium alloys.  When uranium is the likely fuel in the United States, present plans call for 19.9 percent fuel enrichment.  This high level of enrichment is one reason the reactor could be able to operate for extended periods without refueling.  Toward the end of 2004 the town of Galena, Alaska granted initial approval for Toshiba to build a 4S reactor in that remote location.  Original plans called for completion in 2010 though it was acknowledged that this was ambitious.  Galena and Toshiba officials discussed their plans with the NRC in early February 2005.  The NRC indicated that it was not familiar with the 4S design and that design certification (at vendor expense) might be costly and prolonged.  Design certification can be incorporated in the COL process thus it is not clear if a separate design certification will be pursued, if the project continues.  A University of Alaska study of the proposed Galena reactor is available on http://www.iser.uaa.alaska.edu/Publications/Galena_power_draftfinal_15Dec2004.pdf#search='Toshiba%204S'

5. Generation IV (Gen IV) Concepts

The U.S. Department of Energy participates in the Generation IV International Forum (GIF), an association of thirteen nations that seek to develop a new generation of commercial nuclear reactor designs before 2030.  The U.S., Canada, France, Japan and the United Kingdom signed an agreement on February 28, 2005 for additional collaborative research and development of Gen IV systems. Criteria for inclusion of a reactor design for consideration by the initial GIF group include:

       1. Sustainable energy (extended fuel availability, positive environmental impact);
       2. Competitive energy (low costs, short construction times);
       3. Safe and reliable systems (inherent safety features, public confidence in nuclear energy safety); and
       4. Proliferation resistance (does not add unduly to unsecured nuclear material) and physical protection (secure from terrorist attacks).

GIF members agreed during 2002 to concentrate their efforts and funds on six concept designs whose goal is to become commercially viable between 2015 and 2025. There is thus some leeway between the 2030 target for the GIF program implementation and the targets for individual concepts. Individual GIF participant nations are free to pursue any individual technology they choose. The United States intends to pursue each design.

The GIF group, along with the U.S. Department of Energy's Nuclear Energy Research Advisory Committee (NERAC), published "A Technological Roadmap for Generation IV Nuclear Energy Systems" (December 2002) which summarizes plans and designs for Generation IV projects. This is accessible through http://gif.inel.gov/roadmap/pdfs/gen_iv_roadmap.pdf and describes each design in some detail including reactor schematics. Each design is evolutionary; thus while the following descriptions involve comparison to present designs, these analogies should be interpreted with caution. Designs are expected to evolve. Gen IV programs are summarized on http://www.inelgov/initiatives/generation.shtml.

The U.S. Department of Energy and the Idaho National Laboratory are developing a program, the Next Generation Nuclear Plant (NGNP), for implementing the first Gen IV reactor designs, and have initiated discussions with potential private managers of the project.  Potential portions of this program are included in the above discussion of the GT-MHR and PBMR designs above.  The NGNP program anticipates completing the first Gen IV concept by 2020 and possibly earlier.  Project efforts will include the production of hydrogen at the prototype reactor.  While very high temperature gas-cooled reactors appear most likely for eventual consideration, additional U.S.-based Gen IV designs might be submitted to the program managers.

Nuclear Regulatory Commission officials have indicated that present staff at the NRC are not familiar with innovative reactor designs, thus any application for design certification would consume more time than for more evolutionary LWR designs.  Because GIF reactors involve very long term plans, NRC familiarity with designs might evolve before Generation IV reactors are ready for design certification.

Gas-cooled Fast Reactor (GFR): The GFR uses helium coolant directed to a gas turbine generator to produce electricity. This parallels PBMR and original GT-MHR designs. The primary difference from these designs is that the GFR would be a "fast" or breeder reactor. One favored aspect of the design is that it would minimize the production of many undesirable spent fuel waste streams. The reference design size was targeted to be 288 MWe with a deployment target date of 2025. In addition to producing electricity the design might be used as a process heat source in the production of hydrogen.  For further information see http://nuclear.inl.gov/gen4/gfr.shtml

 

Lead-cooled Fast Reactor (LFR): So far, most breeder reactors have used molten metal technologies for their coolants. Many FBRs have used molten sodium, a metal with which there is considerable experience but which has sometimes proven difficult to handle. The LFR uses molten lead or a lead-bismuth alloy as its coolant. Similar designs are being investigated in Russia which is not a GIF participant. Some designs favored under the Generation IV program would result in long periods between refueling, as much as 20 years or more. Target ranges for this reactor would be 50-150 MWe. That would be rather small by historic nuclear standards, but might meet localized market needs. Designs as large as 1200 MWe have been suggested. Initial targeted deployment would be in 2025. Proposed designs would favor electricity production though proponents consider the production of process heat at LFRs as possible. For further information see http://nuclear.inl.gov/gen4/lfr.shtml.  One design in this family of reactors is described on http://www.coe.berkeley.edu/labnotes/1002/reactor.html.

 

Molten Salt Reactor (MSR): The MSR involves a circulating liquid of sodium, zirconium, and uranium fluorides as a reactor fuel though the design could use a wide variety of fuel cycles. The MSR has been presented as providing a comparatively thorough fuel burn, safe operation, and proliferation resistance. The initial reference design would be 1000 MWe with a deployment target date of 2025. Temperatures would not be as hot as for some other advanced reactors, but some process heat potential exists. Versions of the MSR have been around for some time but were never commercially implemented. The MSR was down rated within the Gen IV program during 2003 because it was seen as too distant in the future for inclusion within the Gen IV schedule.  At the same time proponents see some MSR potential for the NGNP program.  For further information see http://nuclear.inl.gov/gen4/msr.shtml.

 

Sodium-cooled Fast Reactor (SFR): Sodium-cooled fast reactors have been the most popular design for breeder reactors. Designs have been proposed under the Department of Energy’s “roadmap" for Generation IV reactors ranging from 150 to 1700 MWe. Elements of the SFR are included in the 4S design proposed by Toshiba for Galena, Alaska.  Molten metal technology is no longer "new" but several early SFR prototypes had difficulty obtaining sustained operation.  The BN-600 in Russia has been regarded as highly reliable. Design supporters believe that the SFR promises superior fuel management characteristics. The original target deployment date of 2015 reflected the considerable research that the design has already received though the design is clearly not as ready for U.S. deployment as LWR designs being evaluated for roughly the same period. The target date seems to be lagging as the VHTR designs gain favor. Prototypes have been built in France, Japan, Germany, the United Kingdom, Russia, India, and the United States starting as early as 1951. Initial deployment would probably focus on electricity due to comparatively low "outlet temperatures" for the design. Sodium-cooled reactors are discussed at http://nuclear.inl.gov/gen4/sfr.shtml and http://wwwnuc.berkeley.edu/~gav/almr/01.intro.html.

 

Supercritical-water-cooled Reactor (SCWR): The SCWR design is to be the next step in LWR development and has been proposed with alternatives that evolve from both the BWR and the PWR. SCWRs would operate at higher temperatures and thermal efficiencies than present LWRs. The reference plant might be 1700 MWe, at the upper end of present LWR designs. The deployment target date was 2025. Some GIF participants favor the SCWR design because it is more familiar to commercial markets than are more innovative concepts. Much of the design research has been in Japan. Designers intend the SCWR to be much less expensive to build than today's LWRs though some of these economies appear to be shared by units now undergoing certification or pre-certification. Operating cost savings are also anticipated. For further information see http://nuclear.inl.gov/gen4/scwr.shtml.

 

Very-high-temperature Reactor (VHTR): The VHTR is an evolution from the HTGR family of reactors but would operate at even higher temperatures than designs now undergoing pre-certification.  Some of the VHTR design standards might be met by modified PBMRs or GT-MHRs.  In contrast with the GFR, the VHTR would not be a breeder reactor, thus it would produce less potentially usable fuel than it consumes. In addition to generating electricity, the design can provide process heat for industrial activities including hydrogen production and desalinization. Deployment is targeted for 2020, earlier than most Generation IV designs. The VHTR is now a favored design in the U.S., where it is the basis for most anticipated submissions for the still-evolving Next Generation Nuclear Plant (NGNP). France also favors the design which is also popular in Asia and South Africa. The VHTR is discussed at http://nuclear.inl.gov/gen4/vhtr.shtml and http://www.nuc.berkeley.edu/designs/mhtgr/mhtgr.html.

Each GIF project involves new or untested reactor design concepts. It would be surprising if each design concept met the program's initial targets or that prototypes would match originally intended standards. The research involved in the program has the potential to contribute to the understanding of alternative types of commercial nuclear power and process heat production even if individual projects fail to meet initial expectations.

6. Outlook

Efficiency Issues

A primary source of doubt regarding the potential of nuclear power, at least in the U.S., has been whether the recent nuclear technology has been too expensive to compete in the commercial marketplace. There have been no orders for new nuclear power plants during the last three decades in the United States and Canada. Finland’s order for a new reactor in 2003 broke a similar extended hiatus in Western Europe, excepting France where orders tailed off later. France now looks likely to follow. Reactor vendors have not ignored the message that their product has recently involved high investment costs and long construction periods. Vendors now seek to position their product with promises of lower prices, shorter construction times, and specified financial arrangements. Most competitors are now offering fixed and historically low prices for at least the nuclear components of their designs.  These promises vary with the price of basic materials such as steel and concrete and as first of a kind engineering costs are allocated or eliminated.  Location, buyer specifications, and regulatory requirements can also alter anticipated costs.

Concerns regarding construction costs for new nuclear power plants contrast sharply with the comparatively low cost of operating commercial reactor designs. Overall operating costs for nuclear power plants, as reported to the Federal Energy Regulatory Commission (FERC), have been roughly the same as (most recently slightly less than) operating costs for coal-fired plants for about two decades. Such operating costs are considerably below the costs of operating most natural gas-fired generation units even when natural gas prices are relatively low. Moreover, the fuel cost component of operating a nuclear power plant is particularly low. This operating cost advantage has given existing nuclear power units a favored position in the provision of base load electric power. Nuclear plant designers hope to take advantage of such low operating costs in positioning their new designs.  Whether they will succeed has not yet been demonstrated.  Discussions of estimates of the capital and operating cost of new power generation units can be found in the "Issues in Focus" section of the  Annual Energy Outlook 2004 and in the Electricity Module of the Assumptions for the Assumptions for the Annual Energy Outlook 2005.

The following publications summarize efforts and procedures to make new nuclear power plants commercially attractive.

  1. "Strategies for competitive nuclear power plants (TECDOC-1123)" International Atomic Energy Agency (November 1999).
  2. "A Roadmap to Deploy New Nuclear Power Plants in the United States by 2010,": http://www.nuclear.gov/NucPwr2010/NucPwr2010_PI.html.
  3.  Scully Capital, "Final Draft, Business Case for Nuclear Power Plants, Bringing Public and Private Resources Together for Nuclear Energy" (July 2002): http://www.nuclear.gov/home/bc/businesscase.html
  4. "A Technology Roadmap for Generation IV Nuclear Energy Systems (GIF-002-00)": http://gif.inel.gov/roadmap/pdfs/gen_iv_roadmap.pdf

 1A large number of reactor designs are excluded from the discussion. These include reactors promoted overseas by nations such as Russia, India, Argentina, Korea, Canada, and China, as well as numerous smaller or even portable reactors (other than the 4S) that are being examined worldwide, including in the United States. Also excluded is the International Atomic Energy Agency's International Project on Innovative Nuclear Reactors and Fuel Programs (INPRO) that covers territory similar to the GIF program in addition to other promising designs. GIF designs have been more heavily promoted within the United States.
 2The one that is not operational, Brown's Ferry 1, has been shut down since 1985, but has not given up its operating license. The plant's owner-operator, the Tennessee Valley Authority, intends to restart the reactor in mid-2007.
3The terms "cooled" and "moderated" are important because they define reactor categories. Cooling in a reactor refers to the process and medium by which heat is transferred from the reactor core to the steam supply cycle of the nuclear power plant. "Moderating" is a concept unique to nuclear power. A moderator controls the rate of the nuclear power reaction and thus the amount of heat that is generated. In a light water reactor ordinary water serves both functions. Light water contains the same isotopes of hydrogen and oxygen as naturally occurring water. Heavy water contains a different, heavier isotope of hydrogen known as deuterium. Beyond the point that these conditions define reactor types, this will not matter in the discussion of existing reactors. It does matter for the group that will be discussed under "Generation IV" reactors.
4Exceptions include Canada, the United Kingdom, India, and part of Russia's industry.
5 Prior to 1969, some smaller commercial reactors were placed in service. All have been retired.
6 This is based on Utility Data Institute/Resource Data International compilations of FERC Form 1 data.
7The discussion here does not directly address "enrichment" the process by which the U-235 content of nuclear fuel is increased.
8This latter statement is based on "A Technology Roadmap for Generation IV Nuclear Energy Systems".  This publication is a major source of Gen IV discussions in the text.
9Candu is a contraction of the term "Canadium deuterium". Canada has an interesting and unique nuclear power history which is covered by the book, Atomic Energy of Canada Limited, Canada Enters the Nuclear Age.
10Inspectors of nuclear power plants have a preference for plants such as the LWRs that are refueled in batches rather than the continuous fueling of PHWRs. Batch refueling allows the fate of spent fuel to be more easily monitored and occurs at intervals of one to two years.
11 Not the AGRs.
12 Most designs of PHWRs also use natural uranium fuels. However, variations in fuel type are possible at any PHWR with plutonium and thorium fuel content subject to particular interest and experimentation.
13 “AP” is sometimes taken to mean “Advanced Passive”.
14 This sentence is a good example of the acronyms that overwhelm the nuclear steam supply system (NSSS) industry. Several of these acronyms no longer have any meaning in "words" while others have only limited actual meaning. They are defined below when possible.
15 The term ESBWR is now called the "Economic Simplified Boiling Water Reactor". Definitions of the initials have changed overtime and occasionally been denied.
16 ACR is usually read to mean "Advanced CANDU Reactor".

Contact:
Ron Hagen: ronald.hagen@eia.doe.gov
(202) 287-1917