Los Alamos Innovations: Water Boiler Reactor

Los Alamos Innovations: Water Boiler Reactor

History Page Type: 
"What was needed to get vital information for building a bomb was a homogeneous reactor in which the fissionable material was homogeneously distributed in the medium, because that’s the way a bomb functions. It’s a homogeneous medium of only uranium or plutonium." - Gerhart Friedlander, a member of the Water Boiler Group, on the importance of the Water Boiler
"The water boiler was the first measurements of the enriched U-235 in a water solution, which is not exactly related to the eventual bomb physics, but these same types of calculations, which go into the bomb physics, go into this. There was not enough material to do a real mock up so the various people concerned got interested in how can one at least do a secondary check by saying, well, 'The calculational methods, the numbers, the constants that are used will apply someplace else, and we only have to start with a few grams of this material and eventually will never have enough to do the complete mock up. But can’t we get some handle on this?'" - Raemer Schreiber, one of the scientists who helped build the Water Boiler reactor, on the relationship between the reactor and the bomb
The Los Alamos Water Boiler reactor, circa 1944

The first nuclear reactor went critical on December 2, 1942. Ever since, the Chicago Pile-1 reactor has held a special place in the history of nuclear physics. Less than a year later, the X-10 Graphite Reactor in Oak Ridge, TN, a pilot plutonium production reactor based on Enrico Fermi’s pile, began making plutonium to fuel the atomic bomb project.

The third nuclear reactor ever built had a different design, and a different purpose. By harnessing uranium in its liquid form, the Water Boiler reactor helped scientists learn how to best build the atomic bomb.



Even before Fermi, the “Italian Navigator,” guided the Chicago Pile-1 reactor to criticality in 1942, there was interest in liquid-fuel reactors. While men like Fermi investigated reactors made of solid uranium, early experiments at the Cavendish Laboratory in England indicated that a self-sustaining nuclear reaction could be obtained using a slurry of uranium in heavy water—water molecules with the deuterium isotope in place of hydrogen. The scarcity of heavy water therefore prevented many further studies into liquid-fuel reactors. The production of large amounts of heavy water in the United States and Canada in the years that followed renewed interest into homogenous reactors. These experiments indicated that a smaller amount of heavy water would be required than previously thought.

At Los Alamos, Fermi backed the production of this new kind of nuclear reactor. The reactor was proposed by Robert Bacher to complement other investigations into measuring the critical masses of chain reactions. The Theoretical Division initially opposed the project, seeing it as a distraction. While Bacher prevailed, the calculations required for the Water Boiler were extensive and occupied a large amount of the work of the T Division in 1943.

The Water Boiler reactor, built at Los Alamos over the winter of 1943–44, would become the world’s first homogenous liquid-fuel reactor. With newly available enriched uranium samples, the group was able to forego the use of heavy water entirely in their design. The Water Boilers are so named because it appears as though the liquid fuel is boiling. The bubbles, however, are the result of the splitting of water into hydrogen and oxygen gas by the energetic process of fission. This name also helped obscure the true nature of the device.

Under the direction of Don Kerst, inventor of the betatron, a Water Boiler group was assembled in the Experimental Physics Division of Los Alamos in August of 1943. The group included Charles P. Baker, Gerhart Friedlander, Marshall Holloway, L. D. P. King, and Raemer Schreiber. The first Water Boiler reactor, nicknamed LOPO due to its low power output, went critical on May 9, 1944 with Enrico Fermi at the controls. In the years that followed, two more reactors would be built at Los Alamos based on its design.



Just like any nuclear reactor, the Water Boiler reactor initiated, moderated, and controlled a nuclear chain reaction. Specifically, the Water Boiler reactors were homogenous liquid-fuel reactors. While the initiation step is similar—the addition of uranium to reach critical mass—the moderation step and type of fuel distinguished these new reactors from those Fermi had previously helped build.

A diagram of the LOPO reactor designThe liquid fuel was made by dissolving uranyl sulfate enriched with uranium-235 in water. The water in the mixture also served as the moderator of the nuclear reaction, slowing down the neutrons to speeds at which fission could occur. The neutrons released during fission travel at speeds around one-tenth of the speed of light, and slowing them down increases the likelihood that they will strike nearby uranium atoms and therefore propagate the chain reaction.

This mixture distinguishes a homogenous reactor from heterogeneous reactors that use separate components for the fuel and moderator—for instance, the Chicago Pile-1 and X-10 graphite reactors used graphite blocks to slow down neutrons from the uranium slugs. The water also makes Water Boiler reactors self-controlling—as the reaction accelerates and heats up the water, the water becomes less dense and less able to slow down neutrons, decreasing the speed of the reaction. Another advantage of the liquid fuel was that it dispensed of the need for dissolving and reprocessing solid uranium fuel in order to extract fission products.

Since it was the first reactor of its kind, the Water Boiler group was faced with many unknowns, including the choice of compound, purification of the fuel, potential corrosion of the fuel vessel, decontamination, and methods of analysis. The group proceeded with caution, and decided to scale back their proposed design in favor of a simple reactor that would operate at very low power. Robert F. Christy of the Theoretical Division calculated the size of the area that would be contaminated if the reactor were to explode, and an isolated location, later named Omega, was chosen away from the town as the site of the reactor building.

There the scientists and engineers got to work solving the design challenges associated with liquid-fuel reactors. The fuel vessel had to be arc welded, as the uranium salt would corrode solder. Uranyl sulfate was chosen for the LOPO reactor, because it absorbed fewer neutrons than other uranium oxide compounds and dissolved easily in water. Meanwhile, the members of the Theoretical Division began estimating the reactor’s critical mass.

This fuel was held in a stainless steel sphere one foot in diameter. Around this fuel vessel was a beryllium oxide reflector, which directed neutrons back into the reaction sphere. The fuel could be added and removed to the reactor, which allowed for continual extraction of fission products. The reactor was designed to have a power output of essentially zero, and so no shielding or external cooling system was required. This drastically simplified the design of the reactor.

In this liquid form, the mass of uranium required to reach criticality was much lower. Only 565 grams of uranium-235 were used in LOPO when it went critical in May 1944. This was still, however, the country’s entire supply of enriched uranium at the time.



The reactor was used to perform calculations related to the construction of the atomic bombs. Estimating the critical mass of the Water Boiler gave the Theoretical Division one of their first real tests of the accuracy of their critical-mass calculations. Another experiment, proposed by Bruno Rossi, determined a quantity called the prompt period, which is related to the length of time after fission the neutrons are emitted, the character of the neutrons, and the character of the core. This value could be used to predict the efficiency of the bombs.

The aqueous homogeneous reactor at Oak Ridge National LaboratoryBy developing new methods of calculating such quantities and testing them against the Water Boiler experiments, the Theoretical Division strengthened their knowledge in applying calculations to new systems. The Water Boiler was also used for other theoretical experiments, including those on cooling and shielding; the effects of temperature changes on criticality; and the effects of movement among the control rods. The Theoretical Division also interpreted fluctuations in the reactor’s operation while the Experimental Physics Division used neutrons produced by the reactor to measure neutron capture and scattering and to test potential materials for the bombs.

The biggest success of the LOPO reactor, however, was as a test for the new reactor concept. The LOPO reactor was dismantled after a couple of months of measurements to make way for a higher-power version, aptly called HYPO, short for “high power.” Completed in December 1944, HYPO had a maximum power output of 5.5 kilowatts and was used as a source of neutrons for a number of experiments at Los Alamos.

HYPO used a slightly different fuel, a solution of uranyl nitrate, and contained cooling coils within the fuel vessel. A thick concrete shield was built surrounding the reactor, but a hole was left into which samples could be placed for direct neutron irradiation. Many important experiments were run using the neutrons emitted by HYPO, including calculations of the critical radius of radioactive material and measurements of the absorption characteristics of a number of elements. These experiments were critical to the design of the early atomic bombs. In addition, scientists measured the delayed neutron and gamma ray emission of the Water Boiler to calibrate their equipment for measuring the gamma ray and neutron intensities at the Trinity test.

By the 1950s, the HYPO reactor was heavily modified to increase its neutron output to a maximum of 35 kilowatts. Following the pattern, this new “super-powered” reactor was named SUPO. The major modifications included further enrichment of the uranium fuel, replacement of the beryllium oxide reflector with graphite, and the addition of a gas recombination system to fuse the hydrogen and oxygen gas produced during the reaction back into liquid water.

SUPO operated for over twenty years, during which it was used for many experiments related to nuclear weapons. The Health Division also used the reactor for a series of experiments on the effects of neutron, beta, and gamma radiation on live animals. In particular, scientists studied mice, rats, rabbits, and monkeys for life shortening, the loss of reproductive power, and the development of various diseases as a result of irradiation. The reactor’s capabilities and reactivity were also explored.

The low risk, cost, and fuel consumption made the Water Boiler reactor a popular research tool at other sites around the country. A Water Boiler reactor was built at Oak Ridge and went critical in 1952. In 1981, the ARGUS reactor at the Kurchatov Institute in Russia became operational. This reactor, which is similar in design to the original Water Boiler reactors, produces a large number of radioactive isotopes for use in medical procedures around the world.

These Water Boiler reactors were part of a larger series of experimental reactors built at Los Alamos. After the Water Boilers came a fast neutron reactor called “Clementine”; a series of power reactor experiments that used a fuel solution of highly enriched uranium dioxide dissolved in acid; and a reactor fueled by molten plutonium and cooled using molten sodium. With each new reactor model, scientists learned more about reactor design and nuclear energy. Reactor development continues to this day, building on the techniques and methods acquired from these early reactor experiments. 

Related Video: 

Gerhart Friedlander's Interview

In his interview, Gerhart Friedlander discusses the Water Boiler reactor, his first major assignment at Los Alamos. Friedlander also discusses how he won a bet against Enrico Fermi on what fuel the reactor should use.
More Historical Resources: 
  • Anderson, Herbert L. “‘All in Our Time’: Fermi, Szilard and Trinity.” Bulletin of the Atomic Scientists 30, no. 8 (1974): 40-47.
  • Bunker, Merle E. “Early Reactors: From Fermi’s Water Boiler to Novel Power Prototypes.” Los Alamos Science 7 (1983): 124-131.
  • Hawkins, David. Book VIII, Los Alamos Project (Y), Volume 2 – Technical, vol. 35 of Manhattan District History. Los Angeles: Tomash Publishers.
  • Hoddeson, Lillian, Paul W. Henriksen, Roger A. Meade, and Catherine Westfall. Critical Assembly: A Technical History of Los Alamos during the Oppenheimer Years, 1943–1945. Cambridge: Cambridge University Press, 2004.
  • Lane, James A., H. G. MacPherson, and Frank Maslan, eds. Fluid Fuel Reactors. Reading, Mass.: Addison-Wesley Publishing Company, 1958. (Online access from moltensalt.org)
  • Serber, Robert. The Los Alamos Primer: The First Lectures on How to Build an Atomic Bomb, edited with an introduction by Richard Rhodes. Berkeley: University of California Press, 1992.
  • War and Peace in the Nuclear Age. “The Los Alamos Water Boiler Reactor.” WGBH Media Library & Archives. Jan. 1, 1960. http://openvault.wgbh.org/catalog/V_04504A22143A41F0ACE25CF596292CFF.
  • Oral histories of Raemer Schreiber and Gerhart Friedlander on Voices of the Manhattan Project.
  • The World Nuclear Association has a number of resources relating to past, present, and future of nuclear reactors.