The Light-Water Breeder Reactor

User avatar
Kirk Sorensen
Posts: 4068
Joined: Nov 30, 2006 3:30 pm
Location: Alabama

The Light-Water Breeder Reactor

Post by Kirk Sorensen » Jan 19, 2007 8:17 am

An interesting avenue of thorium reactor development came from the same group that originally developed the light-water reactor for the Navy, namely Admiral Rickover and Alvin Radkowsky.

In their search for a long-life reactor core that might satisfy the demands of naval vessels, they considered the use of thorium fuel elements, moderated and cooled by ordinary light-water. To test their ideas, in the late 1970s they modified the Shippingport nuclear reactor (the first civilian power reactor in the United States) with an experimental thorium/U-233 core.

The excellent article on the subject was written by Rod Adams in October of 1995 and is reproduced here with his permission. It is of personal interest to me because reading this article was my first exposure to thorium and its neutronic advantages over uranium.

AEI: Light Water Breeder Reactor: Adapting a Proven System
by Rod Adams

At 12:30 am, on August 26, 1977, the operators at the Shippingport Atomic Power Station began lifting the central modules of the experimental breeder reactor core into the blanket section. At 04:38 am, the reactor reached criticality. During the next five years, the core produced more than 10 billion kilowatt-hours of thermal power - equivalent to about 2.5 billion kilowatt hours of electrical power - with a current retail value of approximately $200 million.

It showed no signs of approaching the end of its useful life. It was obvious from the core performance that the reactor was at least a very efficient converter with a long life core. However, in October, 1982, the reactor was shut down for the final time under budgetary pressures and a desire to conduct the detailed fuel examination needed to determine if breeding had actually occurred.

A report on the experiment was quietly issued in 1987. The core contained approximately 1.3% more fissile material after producing heat for five years than it did before initial operation. Breeding had occurred in a light water reactor system using most of the same equipment as used for conventional reactor plants.

New Fuel Source

Instead of using uranium-plutonium fuel like a liquid metal fast breeder reactor, the light water breeder reactor used uranium-thorium. In a process very similar to the one that produces fissile plutonium from U-238, it is possible to produce a fissile isotope of uranium, U-233, from thorium 232.

The advantage of this combination from a technical point of view is that U-233 produces more neutrons if fissioned by a low energy (thermal) neutron than does U-235. This characteristic means that more excess neutrons are available to convert fertile material. In a carefully designed and constructed reactor, uranium-thorium reactors have enough excess fission neutrons to overcome the parasitic neutron absorptions inherent in a water cooled and moderated reactor.

Recalling a fire analogy, even wet wood can be made to burn if you have enough high quality, carefully arranged dry wood to overcome the heat lost to absorption in the water.

Core Design

In order to minimize neutron absorptions in the water coolant, the designers went through a process they called "squeezing out the water". They redesigned the fuel elements to make the clearances between the fuel pins tighter. Since water is also the means of moving the heat from the reactor to some place where it can be of use, this process has its limits.

The designers chose a seed and blanket core configuration. In this type of reactor, the fissile material is concentrated in the central core region while the fertile material surrounds the central core region including the top and bottom. Most of the neutrons produced in the central core are used to sustain the chain reaction, while most of those than leak out at the boundary are either reflected back into the fissile material or absorbed by fertile material.

The designers also decided to develop a new form of reactivity control that did not depend on putting neutron absorbing control rods into the core. Every neutron lost to a control rod over the life of the core was one less neutron available for converting thorium into uranium.

The seed elements were movable and less than a critical mass if not reflected. If the seed was lowered, more of the neutrons at the boundary would leak out, causing reactivity to drop. If the seed was raised into the blanket region, neutrons would be reflected back into the fissile material by the blanket allowing the core to reach a critical mass.

Why No Follow-up?

The light water breeder reactor was a technical success. It demonstrated a sophisticated way to more effectively use a proven technology and to make better use of natural resources. It even demonstrated a way to significantly reduce the volume of high level nuclear waste per unit of electrical power output.

Unfortunately, the program leaders were not focused on factors that make new innovations successful in the market. The following weaknesses prevented commercial success.
  • There was little effort to promote the technology. Knowledge of the program is rare even within the nuclear industry. There is little chance of an unknown idea - particularly one with as much potential impact as a light water breeder reactor - becoming a new technical standard.
  • The core engineers did not pay enough attention to production difficulties. The assembly of the core modules required a great deal of manual labor including 2,000 precise measurements for each module. This effort implies a high production cost even if raw materials are used more efficiently.
  • There was no effort to develop other uranium-thorium reactors in an effort to help spread the fixed cost of fuel material production.
  • The program was viewed as Admiral Rickover's pet project.

Professional rivalry or ingrained hard feelings against Rickover probably helped seal the fate of the program. By the time the experimental core was shut down, the Secretary of the Navy had already declared his intention to retire Rickover. By the time that the core had been analyzed, Admiral Rickover was dead and many of his strongest political supporters were either retired or dead.

Dead End or New Direction?

It may be that the light water breeder reactor is not a viable alternative to conventional light water reactors. It seems that there is an abundant supply of fissile materials and that the higher costs of core manufacture will not be overcome without a significant automation effort. However, it is useful to know that efficient conversion - even breeding - is not only possible in a thermal reactor without the use of liquid metals, it has been actually demonstrated in a large scale experiment.

User avatar
Posts: 1938
Joined: Nov 30, 2006 9:18 pm
Location: Montreal

Re: The Light-Water Breeder Reactor

Post by jaro » Jan 19, 2007 10:01 am

Rod Adams wrote:In order to minimize neutron absorptions in the water coolant, the designers went through a process they called "squeezing out the water". They redesigned the fuel elements to make the clearances between the fuel pins tighter. Since water is also the means of moving the heat from the reactor to some place where it can be of use, this process has its limits.

Too bad they went to so much trouble to "squeeze out the water": the fuel form used in heavy water CANDU reactors are simple bundles of fuel pins:


User avatar
Posts: 501
Joined: Jan 10, 2007 5:09 pm
Location: Los Altos, California

Re: The Light-Water Breeder Reactor

Post by iain » Jul 30, 2009 1:04 am

If they squeezed out the water, how was the Shippingport reactor moderated?


Posts: 909
Joined: Mar 07, 2007 11:02 am
Location: Ottawa

Re: The Light-Water Breeder Reactor

Post by David » Jul 31, 2009 11:04 am

iain wrote:If they squeezed out the water, how was the Shippingport reactor moderated?

The didn't squeeze it all out of course, but they minimized the water fraction so the spectrum was much more epithermal. U233 if fine in any spectrum so by having less water they lost less neutrons to hydrogen.

David L.

paul wick
Posts: 51
Joined: Sep 01, 2009 1:01 pm

Re: The Light-Water Breeder Reactor

Post by paul wick » Oct 11, 2010 12:09 pm ... rom-israel

from the article:

"Israeli nuclear engineer Eugene Shwageraus is one of those minds. The 37-year-old Ben-Gurion University (BGU) of the Negev lecturer and his research partner, Dr. Michael Todosow of the Brookhaven National Laboratory in New York, received a three-year Energy Independence Partnership Grant last May from the US-Israel Binational Science Foundation to develop a self-sustainable fuel cycle for light water reactors.

And that's where the Israeli scientist's innovation comes in. By taking advantage of proven LWR technology, he and Todosow intend to make a cost-effective light water cooled reactor that will be as efficient as a fast breeder in extracting energy from the fuel."

"The process of development is three years, and at that time we'll choose from among several ways to see which is optimal to combine safety, economics and resource utilization," Shwageraus relates.

Collaborating on alternative, renewable solutions

The goal is a self-sustaining reactor, meaning one that will produce and consume about the same amounts of fuel. This isn't possible with uranium and light water coolant. The better choice is thorium, whose nuclear properties offer considerable flexibility in the reactor core design. Some experts believe that the energy stored in the earth's thorium reserves is greater than what is available from all other fossil and nuclear fuels combined.

Posts: 2292
Joined: Apr 19, 2008 1:06 am

Re: The Light-Water Breeder Reactor

Post by jagdish » Oct 12, 2010 9:24 am

I have myself given the matter some thought and I think the requirement is a liquid fuel, which will permit Xe to move out. Uranyl phosphate is required for fissile part in solution and thorium as solid internal and external blanket producing U233 to compensate the fissile burnt out. X will be escaping with steam and removed during condensation . Some samarium may be precipitated as phosphate salt. Initial uranyl phosphate could be U233 or U235.

Posts: 68
Joined: Mar 16, 2010 1:48 am
Location: Guangdong, China

Re: The Light-Water Breeder Reactor

Post by longwei1221 » Oct 13, 2010 8:55 am

this is the first time I learn some detailed information about shipping port reactor and thorium-based fuel used for breeding
what a pity!

Posts: 1
Joined: Nov 17, 2009 5:43 am

Re: The Light-Water Breeder Reactor

Post by NNNG » Nov 05, 2010 1:39 am


It is obvious that nuclear technology can be improved dramatically, however to me it looks as if an advanced reactor like the LFTR or IFR is unlikely for at least a decade. I understand that water cooled Th/U233 thermal breeders are possible, however I am not sure of all the details of the fuel cycle. In particular, it seems like reactors like the AHWR can provide many of the advantages of LFTR (except higher thermal efficiency, safety, cost, let's forget about them for a moment) while being easier and quicker to develop.

1. Could a LWBR combined with reprocessing increase the utilization of the fuel by a factor of well over 100? Is the final waste only fission products as it would be with LFTR & IFR?
2. Would a LWBR be able to consume existing nuclear waste, leaving only fission products?
3. If I recall correctly, the advantage with the LFTR is that the Protactinium and Xenon can be continuously removed from the reactor - what fuel cycle advantages does this have over a LWBR?

Thanks in advance.

User avatar
Charles Barton
Posts: 251
Joined: Jun 24, 2007 10:43 am
Location: Dallas, TX

Re: The Light-Water Breeder Reactor

Post by Charles Barton » Nov 05, 2010 6:42 am

Another major flaw of the Shippingport reactor was its low power output. Wgat shippingport demonstrated was that thorium breeding was possible with light water reactors, but if breeding was the goal, energy production would be very limited.

User avatar
Kirk Sorensen
Posts: 4068
Joined: Nov 30, 2006 3:30 pm
Location: Alabama

Re: The Light-Water Breeder Reactor

Post by Kirk Sorensen » Feb 17, 2012 9:53 am

LWBR: A successful demonstration completed
by W. J. Babyak, L. B. Freeman, and H. F. Raab, Jr.

(W. J. Babyak and L. B. Freeman are with the Bettis Atomic Power Laboratory in West Mifflin, Pa.; H. F. Raab. Jr., is with the Department of Energy's Office of Naval Reactors in Washington. D.C.)

September 30, 1987, marked the successful completion of a major nuclear reactor development by the Office of Naval Reactors (ONR) of the Department of Energy. On that date, all technical work was finished in support of the Light Water Breeder Reactor (LWBR). LWBR was designed and built by the Bettis Atomic Power Laboratory, which is operated by Westinghouse Electric Corporation under the direction of ONR. The LWBR was run by Duquesne Light Company for five years at the Shippingport Atomic Power Station in Shippingport, Pennsylvania sending power into the utility's commercial grid. All major goals were accomplished: LWBR achieved breeding in an existing light-water-cooled nuclear power plant, employed the uranium-233/thorium oxide fuel system, and reliably generated electricity.

LWBR history

LWBR represents a major milestone in the development of nuclear power for civilian use. Construction and operation of the original Shippingport plant, which began service in 1957 as the first large-scale central station pressurized water reactor (PWR), provided the foundation for today's civilian PWR industry. From 1977 to 1982, this same facility was operated with LWBR; the core alone is what made the reactor a breeder. With its proven ability to breed and also produce electricity while installed in a previously used plant, LWBR is a viable alternative for use in other water reactor plants, increasing the nation's potential energy resources. Instead of the present low-conversion reactors that convert less than one percent of mined uranium into energy, the choice can be a breeder reactor with its advantage of much more efficient fuel utilization, i.e., conversion of about 50 percent of plentiful thorium into energy, along with the use of conventional PWR technology.
LWBR was designed as a seed-blanket reactor core (Figure 1) with Zircaloy-clad tubular fuel elements loaded with oxide fuel pellets. The plant, rated at 236 MWt and 60 MWe (net), produced roughly 2,129,000 MWh (gross) with an 86 percent availability factor and a 65 percent capacity factor. The factors would have been higher, but for the planned shutdowns for testing and deliberate swing load operation to confirm the versatility of the LWBR design. Though originally designed to last for 18,000 effective full power hours, LWBR accumulated 29,047 EFPH before the core was shut down, disassembled, and examined to determine if breeding occurred as predicted and if the core fuel and structure adequately withstood the stresses of the extended operating environment.

LWBR accomplishments

Breeding achieved. The proof-of-breeding experiment was designed as a highly precise, unbiased comparison of measured initial and end-of-life core loadings. A total of 524 whole fuel rods was statistically sampled as representative of the 17,290 in the core; this was the most comprehensive measurement of core breeding ever performed. Each of the rods in the sample was assayed non-destructively for total fissile content through the use of a specially designed irradiated fuel assay gauge. Seventeen of these rods were subsequently dissolved and chemically analyzed with high accuracy by Argonne National Laboratory to provide an independent standard against which the nondestructive results could be compared. This approach produced a best-estimate core fissile content at end-of-life for direct comparison with the known loading at beginning of life. The result was a measured fissile inventory ratio (FIR) of 1.0139, compared to a calculated prediction of 1.0135. The 95-percent-confidence lower bound of this measurement was 1.0115.
The expended core fuel distribution was not as well-predicted as the total fuel content. The measured fissile content of the reflector blanket was about 4 percent greater than predicted (Figure 2). This higher than predicted loading in the reflector blanket is believed to be mainly due to inaccuracy in calculation of the neutron leakage into the reflector blanket. In a large (more than 500-MWe) LWBR, the reflector blanket itself and the inaccuracy of predicted neutron leakage would be less important. LWBR, with its reflector blanket, simulated the neutron leakage characteristics of a large reactor.

Core integrity retained. During operation, LWBR showed no signs of distress; in particular, there was no indication of fuel element failure. After operation was terminated, the expended core components were subjected to a variety of examinations to uncover any unexpected condition in the fuel assemblies, fuel rods, grids, and structural components, All data indicate that the core materials performed well. The fuel assemblies and fuel rods experienced no excessive distortion; there was no evidence of any rods touching their neighbors; and there was no significant cladding wear. Detailed nondestructive and destructive evaluations were performed on selected fuel rods. Nondestructive examinations included visual inspection, diameter and length measurements, oxide film thickness, local wear characterization, cladding defect inspection, and neutron radiography. The results demonstrated that: a) the fuel and structural design procedures were conservative; b) the fuel operated at relatively low temperature, as predicted; c) fuel pellet-to-cladding interaction was not significant; d) seed cladding remained freestanding, as designed; e) blanket cladding-to-pellet gap closed down during reactor operation, as designed; and f) fuel pellets generally remained intact, with minor cracking observed only in the highest burnup regions.

LWBR worth

The success of the LWBR program is highly relevant to the world's energy potential. The results demonstrated that U-233/Th cores can be designed and built, can be operated in existing light-water reactor plants to produce electricity, and can breed enough fissile fuel to overcome modest losses in reprocessing and refabrication. For the United States in particular, this means that the plentiful domestic supply of thorium, a material with no other significant use, can become an important resource. This resource can provide about 50 times as much energy as the domestic supply of uranium used in current LWRs. The light-water breeder thus has an energy potential that could meet the entire electrical needs of the United States for centuries.


This article provides an overview of the Light Water Breeder Reactor Program. Details may be found in numerous technical memoranda, journal articles, and symposia transactions. Some key references include the following:

R. Atherton, Ed., "Water Cooled Breeder Program Summary Report," WAPD-TM-1600, Bettis Atomic Power Laboratory (Oct. 1987).

J. Belle and R. M. Berman, Eds., Thorium Dioxide: Properties and Nuclear Applications, DOE/ NE-0060, Government Printing Office. Washington, D.C. (1984).

W. A. Budd. Ed., "Shippingport Operations with the Light Water Breeder Reactor Core," WAPD-TM-1542, Bettis Atomic Power Laboratory (Mar. 1986)).

D. R. Connors et al., "Design of the Shippingport Light Water Breeder Reactor." WAPD-TM-1208, Bettis Atomic Power Laboratory (Jan. 1979).

V. V. DeGeorge and I. Goldberg, "The Fabrication and Loading of Fuel Rods for the Light Water Breeder Reactor," WAPD-TM-1278, Bettis Atomic Power Laboratory (Mar. 1986).

D. A. Gorscak et al., "End-of-Life Nondestructive Examination of Light Water Breeder Reactor Fuel Rods," WAPD-TM-1605, Bettis Atomic Power Laboratory (Oct. 1987).

H. C. Hecker and L. B. Freeman. "Design Features of the Light Water Breeder Reactor (LWBR) Which Improve Fuel Utilization in Light Water Reactors," WAPD-TM-1409, Bettis Atomic Power Laboratory (Aug. 1981).

H. C. Hecker, "Nuclear Analysis and Performance of the Light Water Breeder Reactor (LWBR) Core Power Operation at Shippingport," WAPD-TM-1421, Bettis Atomic Power Laboratory (Apr. 1984); also see Trans. Ant. Nucl. Soc., 44. 551 (1983).

W. K. Sarber, "Reactor Physics Experimental Program for the Light Water Breeder Reactor (LWBR) at Shippingport,'" WAPD-TM-1455. Bettis Atomic Power Laboratory (Dec. 1981, and addendum, Dec. 1983): also see Trans. Am. Mid. Soc., 44, 552-553 (1983).
W. C. Schick et al., "Proof-of-Breeding in the Light Water Breeder Reactor," WAPD-TM-1612, Bettis Atomic Power Laboratory (Sept. 1987); also see Trans. Am. Nucl. Soc., 55, 607 (1987).

G. Tessler et al., "Nondestructive Assay of Spent Fuel Rods from a Light Water Breeder Reactor," WAPD-TM-1614, Bettis Atomic Power Laboratory (Sept. 1987).

K. H. Vogel, Ed., "Advanced Water Breeder Applications Work-A Summary," WAPD-TM-1532. Bettis Atomic Power Laboratory (Oct. 1982).

Posts: 2292
Joined: Apr 19, 2008 1:06 am

Re: The Light-Water Breeder Reactor

Post by jagdish » Feb 17, 2012 8:14 pm

As a LWBR with low power was proved, why has a higher power version, perhaps with more neutron efficient heavy water never been attempted?
Another interesting idea would be a fast spectrum NaF-ZnF4 cooled U-Pu driver surrounded by a sub-critical Thermal thorium based blanket utilizing the excess neutron flux of the fast driver
. The blanket could be light water moderated and cooled.

Posts: 1336
Joined: Jun 05, 2011 6:59 pm
Location: NoOPWA

Re: The Light-Water Breeder Reactor

Post by KitemanSA » Feb 18, 2012 8:46 am

jagdish wrote:As a LWBR with low power was proved, why has a higher power version, perhaps with more neutron efficient heavy water never been attempted?
The power rating of the core didn't change, did it? It started at 60MWe with a standard design core and remained at 60MWe with the ~isobreeder core. So why bother?
DRJ : Engineer - NAVSEA : (Retired)

Posts: 2292
Joined: Apr 19, 2008 1:06 am

Re: The Light-Water Breeder Reactor

Post by jagdish » Feb 18, 2012 9:06 am

With an adequate power rating, it could have been a niche design like the CANDU PHWR. Now its feasibility is still doubted. The first thorium fuel design, the Indian AHWR, has still not started construction after many years. It is not a breeder.

Post Reply