The First Nuclear Reactor in Orbit

At the beginning of the Space Age, a number of new technologies were being examined to support increasingly sophisticated missions then being considered. Among these were a variety of nuclear-based technologies to generate power for a range of spacecraft requiring anywhere from watts to megawatts of electrical power. While other power-generating technologies quickly matured to meet the needs of most mission types, the needs of spacecraft exploring the outer solar system or requiring large amounts of power for advanced electric-based propulsion systems were best met using nuclear-based electricity generation systems. Based on the results of the earliest studies, nuclear reactors had the potential to meet these requirements reliably with the most kilowatts of electric power produced per kilogram of mass.

A “space cruiser” concept from JPL which would have used ion propulsion to reach the outer planets was one of many early spacecraft designs considered that would rely on nuclear reactors for power. (JPL/NASA)

Nuclear Reactors for Space Applications

During the 1950s, a number of studies performed by various companies and government entities in cooperation with the US Atomic Energy Commission (the predecessor for today’s Energy Research & Development Administration and Nuclear Regulatory Commission) examined the question of nuclear power to support various types of space missions. After a joint USAF-AEC committee had set the requirements for space-based nuclear power sources in 1955, a new development program initially called “Pied Piper” was started. Later renamed SNAP (Space Nuclear Auxiliary Power), the goal of the program was to develop compact, lightweight and reliable nuclear-based electric generation hardware for a range of not only space-related missions but terrestrial applications on sea and land as well. The AEC was given responsibility for directing the development of the necessary hardware with the USAF and, eventually, NASA being the procurement agencies.

The nuclear power sources developed under SNAP fell into two broad categories. The first were devices that converted the heat created by the decay of various radioisotopes into electricity. This category includes Radioisotope Thermal Generators or RTGs that have powered a variety of American spacecraft and other space hardware over the last half of a century including the Voyagers, Cassini and the Curiosity Mars rover. RTGs generate power by using solid-state thermoelectric devices to convert the heat created by radioisotopes directly into electricity. The second category are nuclear reactors that used fissile materials like uranium-235 to generate heat in controlled nuclear chain reactions that would then be converted into electricity by a number of different methods. While several reactor-based systems were built and tested on the ground between 1957 and the end of the SNAP program in 1973, the US launched only a single nuclear reactor into orbit as part of a test flight named “Snapshot”.

A SNAP 10A reactor undergoing ground testing. (DOE)

What would become Snapshot was started in May 1960 by the AEC and USAF as a program known as the Space System Abbreviated Development Plan for Nuclear Auxiliary Power Orbital Test. Initially intended to launch four reactors into space, budget constraints and other consideration eventually forced the flight program to be pared down to just a single test flight of a SNAP 10A reactor. The 435-kilogram SNAP 10A payload for the Snapshot mission consisted of a reactor and power conversion unit designed to generate 500 watts of electrical power for up to one year. The SNAP 10A reactor, built by North American Aviation’s Atomic International Division, had 37 fuel-moderator elements that used uranium-235 as the nuclear fuel and zirconium hydride as a moderator to slow the neutrons in order to maintain the chain reaction in the reactor. Each fuel element was 3.2 centimeters in diameter and 32.6 centimeters long with a mass of 1.38 kilograms of which 128 grams was fissile uranium-235.

Diagram illustrating the major components of the SNAP 10A nuclear reactor. (DOE)

Outside of the reactor housing was a reflector assembly consisting of sets of moveable neutron reflectors composed of beryllium. When the SNAP 10A was launched, these reflectors were positioned to allow most of the neutrons generated by the reactor to escape and keep the system in a non-critical, low-power condition. Once safely in orbit, the reflectors would be rotated into position on ground command to reflect neutrons back towards the reactor to initiate a controlled chain reaction. This would make the reactor go critical and allow it to generate heat that would be converted into electricity. In case of a malfunction in the reactor, the outer reflectors were designed to be ejected either automatically or by ground command to shutdown the reactor permanently.

A liquid metallic alloy called NaK was used to transfer heat out of the reactor and cool it. Consisting of 23% sodium and 77% potassium, NaK was highly reactive and would combust if exposed to air. Despite the safety issues, NaK was preferred as a reactor coolant because it remained liquid at temperatures as low as -13° C to as high as 785° C, had a low vapor pressure and was more effective at removing heat than other coolants like water. Its high boiling point also meant it could operate at much higher temperatures increasing the efficiency of any system that converted heat into electrical power.

Schematic showing the flow of NaK coolant from the SNAP 10A reactor through the thermoelectric elements to generate electricity. Click on image to enlarge. (AEC)

The 110-kilogram reactor and its reflector assembly were mounted on top of a conical radiator and power conversion unit whose structure was composed of a heat-resistance titanium alloy. NaK at a temperature of 533° C was pumped out of the reactor and through stainless steel tubes in this structure. The heat from the liquid NaK would be conducted through a series of 40 thermoelectric converter modules built by RCA holding a total of 2,880 doped silicon-germanium pellets that were electrically isolated from the metallic NaK coolant. These pellets would convert about 1.8% of the heat that passed through them into usable electricity with the waste heat dumped into space by the radiator with a total area of 5.8 square meters. While not as efficient as other types of systems that convert heat into electricity, the thermoelectric converter system was much lighter and had fewer moving parts which improved reliability – desirable properties for any space-based system.

Drawing of the major components of the power conversion and radiator of the SNAP 10A as flown on Snapshot. Click on image to enlarge. (USAF)

The conical power conversion structure also included a 98.6-kilogram lithium hydride shield assembly designed to protect spacecraft systems from neutron radiation generated by the reactor. The flux of gamma radiation from the reactor was considered low enough not to require additional shielding for spacecraft systems. At the base of the structure was an instrument module that contained all of the reactor control systems and instrumentation. The entire Snapshot payload was 3.65 meters tall and had a diameter of 1.5 meters at its base.

The Atlas-Agena D SLV-3 used to launch the Snapshot mission. Click on image to enlarge. (USAF)

Snapshot used a version of the Atlas-Agena D rocket known as the SLV-3 (Standard Launch Vehicle-3) as its launch vehicle. Once in orbit, the SNAP 10A remained attached to the Agena D final stage which supplied vital support functions for Snapshot’s nominal 90-day mission. The Agena was designed to perform similar support functions for a wide range of earlier USAF space payloads including the Corona-series reconnaissance satellites. The total in-orbit mass of Snapshot with its Agena stage was about 2,400 kilograms.

In addition to the SNAP 10A, Snapshot also included additional secondary payloads. One of these was a 5-centimeter ion engine built by Electro-Optical Systems. The first tests of ion engines in space, like NASA’s SERT I launched on July 20, 1964, were limited to short suborbital flights in part due to the relatively large power demands of ion engines (see “The First Ion Engine Test in Space”). The substantial amounts of electrical power being generated by the SNAP 10A reactor was ideal for meeting the requirements for a long-term ion engine test in orbit. The ion engine was mounted on the Agena D and drew about 400 watts of electrical power to generate 8.5 millinewtons of thrust using nonradioactive cesium as a propellant. The ion engine was designed to operate for about one hour at a time using batteries that were recharged over the course of 15 hours using about 100 watts of power drawn from the SNAP 10A reactor.

The major components of the Snapshot spacecraft and its Agena D. Click on image to enlarge. (USAF)

A second piggyback payload was the SECOR 4 satellite also known as EGRS 4. Managed by the US Army Corps of Engineers and built by ITT Federal Laboratories, SECOR (Sequential Collation of Range) or EGRS (Electronic and Geodetic Ranging Satellite) was a series of experimental geodetic satellites that in many ways were the forerunners of today’s GPS navigation satellites. SECOR was designed to be used with a system of three ground stations whose positions were precisely known to determine the position of a fourth station. The Type II SECOR satellites like SECOR 4 were a box shape 25 by 30 by 35 centimeters with its sides covered with solar cells. With a mass of 17.6 kilograms, SECOR 4 included a transmitter beacon and a C-101 transponder built by ITT that used nine deployable antennas on the satellite’s exterior. SECOR 4 would be ejected from the Agena D to fly independently once Snapshot had achieved orbit.

Also carried by Snapshot were seven other minor experiments that remained attached to the spacecraft in orbit. Six of these were supplied by the Air Force Cambridge Research Laboratories (AFCRL) to measure micrometeoroids and electric fields in space. The seventh was a technology experiment designed to characterize the stability of the surfaces of various materials in space for the Air Force Space Systems Division (AFSSD).

The Snapshot Mission

After extensive testing, the primary payload of the Snapshot mission, SNAP 10A reactor number FS-4, was shipped to Vandenberg Air Force Base in California on February 18, 1965. After checkout, it was subsequently mated to SLV-3 number 7401 on April 2. Snapshot was launched at 1:24 PM PST on April 3, 1965 from PALC 2-4 at Vandenberg (today known as SLC-4E which now supports west coast Falcon launches for SpaceX). A nominal performance by the Atlas and the first burn of the Agena successfully placed Snapshot into a temporary 160 by 1,300-kilometer transfer orbit. An additional ten-second burn by the Agena D at apogee placed Snapshot into a long-lived 1,287 by 1,306-kilometer polar orbit inclined 90.3° to the equator with an estimated lifetime of around 4,000 years.

The launch of Snapshot from Vandenberg’s PALC 2-4 on April 3, 1965. (USAF)

Shortly after Snapshot reached its final orbit, the SECOR 4 satellite was ejected into an independent 1,266 by 1,324-kilometer orbit. Unfortunately, SECOR’s transponder failed to operate rendering the satellite unusable. After the reactor systems had been checked out, the AEC authorized reactor startup which began 3 hours and 48 minutes after launch during the second revolution. As the reactor heated up, the insulating heat shield covering the radiator that kept the NaK coolant from freezing during the early phases of the mission was ejected during the sixth revolution. Criticality was reached 10 hours and 13 minutes after launch and the reactor went to full power 2 hours and 15 minutes later. Electrical power generation peaked at 600 watts but eventually stabilized at 530 watts.

As the reactor was being tested, the ion engine was warmed up and started its first test firing. Unfortunately, during its one hour of operation, which was ended automatically by an onboard timer during the ninth revolution, the ion engine produced excessive electromagnetic interference (EMI) which affected Snapshot’s horizon attitude sensor and caused the spacecraft to slew out of its proper orientation. Subsequent ground testing showed that the EMI was the result of an unexpectedly large number of high voltage breakdowns. Because of the nature of the problem and with no solution immediately at hand, no other attempts were made to run the ion engine during the rest of the Snapshot flight. The secondary experiment payloads were switched on during revolution 23 to absorb some of the power being generated by the reactor that would have otherwise gone into recharging the ion engine system’s batteries. One of the two AFCRL micrometeoroid detectors failed to operate and part of the AFSSD experiment malfunctioned after a couple of weeks of operation but otherwise the secondary experiments returned useful data.

Artist’s depiction of Snapshot in its high polar orbit. (AEC)

As ground controllers put the SNAP 10A reactor through its paces and stabilized its power output, the active attitude control of the Agena D was turned off several days after launch before its limited supply of gas was depleted. Snapshot then used passive gravity gradient stabilization with the reactor pointed away from the Earth to maintain its orientation. All was going well with the mission until May 16, 1965 when Snapshot failed to contact ground controllers via the tracking station in Hawaii as scheduled during revolution 555. After 40 hours of attempts, contact with Snapshot finally began to be reestablished. It was quickly determined that the spacecraft systems were operating on battery power alone. By May 18 it was determined that the reactor had automatically ejected its neutron reflectors as planned when a failure in the bus voltage regulator on the Agena had been detected. This permanently shutdown the reactor after only 43 days of operation effectively ending the Snapshot mission.

Mission Aftermath

Despite the early shutdown of the reactor, the objectives of the Snapshot mission were largely met. Much had been learned from the operation of the SNAP 10A reactor in orbit resulting in improvements in future SNAP-series reactors. Unfortunately for the advocates of space-based nuclear reactors, Snapshot would prove to be the only time the US would fly a nuclear reactor in space. Ground testing continued until the end of the SNAP program in 1973 but there was little interest in actually attempting another flight for a number of budget, policy and safety reasons.

An artist’s depiction of one of today’s many advanced outer planet exploration concepts that would rely on nuclear power. (LANL)

Even though there was a resurgence in interest in space-based nuclear reactors starting in the 1980s in part to support elements of the Strategic Defense Initiative (popularly known as “Star Wars”) and later to power advanced space missions to the outer planets, the only nuclear power sources flown on American spacecraft to date have been the simpler and safer RTGs. Only the Soviet Union has flown nuclear reactors in space operationally between 1967 and 1988 to power 33 of their RORSATs (Radar Ocean Reconnaissance Satellites) designed to monitor the movements of American and NATO ships at sea. Several incidences with these satellites, including a wayward reactor coming down in Canada’s Northwest Territories in 1978, combined with the economic decline of the Soviet Union finally halted their use.

While Snapshot is in a high orbit that should keep it safely away from the Earth for about another 4,000 years as its nuclear fuel decays, it has still created unanticipated problems for the space community. Starting in late November of 1979, it was noticed that it was shedding debris of some sort. Over the course of six years, about 50 trackable objects had been created as a result of a half a dozen shedding events. While some have theorized that a collision with a piece of space debris is the cause, given the low probability of such an event at an altitude of 1,300 kilometers and the low relative speed of the debris, it is much more likely that some unknown internal malfunction is the culprit. The radar properties of the debris make it unlikely that the objects are from a NaK coolant leak but the possibility that the debris is radioactive can not be excluded. While not as dramatic as the debris shedding events that some Soviet reactors have experienced in their storage orbits (sometimes the result of NaK coolant leakage), the breakup of Snapshot has unfortunately added to the growing problem of space debris in high altitude Earth orbit.

Follow Drew Ex Machina on Facebook.

Related Video

Here is a documentary produced by Atomic International for the US Atomic Energy Commission from 1965 about the Snapshot mission:

Here is another Atomic International-produced documentary from before the launch of Snapshot detailing the development of the SNAP-10A reactor:

Related Reading

“The First Ion Engine Test in Space”, Drew Ex Machina, July 20, 2014 [Post]

“When the Nuclear Age Met the Space Age: The Beginnings of Nuclear Rocket Propulsion Development”, Drew Ex Machina, February 24, 2021 [Post]

General References

William R. Corliss, Nuclear Reactors for Space Power, US Atomic Energy Commission, 1966

Robert H. Nichols, Geodetic SECOR Satellite, AD/A-002 625, Army Engineer Topographic Laboratories, June 1974

David S.F. Portree and Joseph P. Loftus, Jr., Orbital Debris: A Chronology, NASA/TP-1999-208856, January 1999

Glen Schmidt, “SNAP Overview”, Keynote address for opening dinner of 2011 Nuclear and Emerging Technologies for Space Conference (Albuquerque, NM), February 7, 2011

James S. Sovey, Vincent K. Rawlin and Michael J. Patterson, “Ion Propulsion Development Projects in the U.S.: Space Electric Rocket Test I to Deep Space 1”, Journal of Propulsion and Power, Vol. 17, No. 3, pp. 517-526, May-June 2001

Susan S. Voss, SNAP Reactor Overview, AFWL-TN-84-14, Air Force Weapons Laboratory, August 1984

“SNAP 10A Shut-Down”, Flight International, p. 890, June 3, 1965

“Snapshot”, TRW Space Log, Vol. 5, No. 2, pp. 48-50, Summer 1965