Purchasing new hardware? Read our latest product comparisons

NASA team pushing towards thermal nuclear propulsion systems


January 21, 2013

Artist's rendition of the Nuclear Cryogenic Propulsion Stage (NCPS), a nuclear engine intended primarily for the upper stages of launch vehicles and as primary propulsion for deep space missions (Photo: NASA)

Artist's rendition of the Nuclear Cryogenic Propulsion Stage (NCPS), a nuclear engine intended primarily for the upper stages of launch vehicles and as primary propulsion for deep space missions (Photo: NASA)

Image Gallery (13 images)

Nuclear-powered rocket engines are not new. In the 1960s, both the U.S. and the Soviet Union developed and tested thermal nuclear rockets fitted with flight-worthy components. However, Project Rover and NERVA (Nuclear Engine for Nuclear Rocket Application) programs were defunded in the early 1970s just before test flights were to start. Now, as part of the Advanced Exploration Systems program at NASA, the Nuclear Cryogenic Propulsion Stage team is tackling a three-year project to demonstrate the viability of and to evaluate materials for thermal nuclear propulsion systems for use in future deep space missions.

A thermal nuclear rocket engine uses a nuclear reactor to heat hydrogen to very high temperatures, with the superheated hydrogen expanding and forced through a nozzle to generate thrust. As such, the performance of a thermal nuclear engine is limited by the high temperature strength of solid materials. Unlike solar or nuclear powered ion or plasma engines, thermal nuclear engines perform largely the same job as chemical rockets (produce multi-g acceleration), but do so with roughly half the fuel requirement.

Cross-section of a 1960s NERVA-type thermal nuclear rocket engine (Image: NASA)

As seen above, a set of turbopumps feed liquid hydrogen into the engine. Most of the hydrogen flow enters the top of the engine enclosure, but part is split off to feed a heat exchanger on the walls of the rocket nozzle, thereby cooling the nozzle and preheating the hydrogen. The turbopumps provide the very large pressure required to keep fuel flowing against the back pressure of the engine.

The hydrogen is fed into the reactor core, which has a matrix of longitudinal channels in which the hydrogen is heated to temperatures in excess of 2500° K. The reactor core is surrounded by a neutron reflector, which allows the use of smaller amounts of fissionable material in the reactor. Once the hydrogen is heated by the reactor, it is exhausted through the nozzle, attaining an exhaust velocity as large as 10 km/s or more.

NERVA (Nuclear Engine for Nuclear Rocket Application)

Comparative sizes of various 1960s US nuclear rocket engines (Photo: NASA)

The accomplishments of the NERVA program were considerable. At the very end of the program it had developed nearly flight-ready engines able to operate for periods in excess of an hour while providing maximum thrusts up to 220,000 lb (1000 kN), reactor core temperatures of 2500° K, a specific impulse over 850 seconds (roughly double that of chemical rockets), and a thrust-to-weight ratio of better than 3:1.

An interesting aspect of the NERVA program is how small in physical size they were able to make nuclear engines – reactor cores were about 150 cm (5 ft) in length, and from 50 cm (NRX-A3) to 140 cm (Phoebus-2A) (20 to 55 in) in diameter. Flight tests were upcoming just as the program was cancelled (along with the first U.S. program to put men on Mars).

Nuclear Cryogenic Propulsion Stage

NASA is now taking another look at nuclear engines, primarily for the upper stages of launch vehicles and as primary propulsion for deep space missions. The Nuclear Cryogenic Propulsion Stage (NCPS) is working toward the design and flight validation of a 25,000 lb (100 kN) class thermal nuclear rocket system. As today's materials and production technologies offer considerably better performance than was available 50 years ago, core temperatures and thrust-to-weight ratios are both expected to increase significantly.

The NCPS is being designed on the foundations developed during the NERVA program. It will have a fueled weight of about 40 tons, and will be 7.5 m (25 ft) in diameter and 12.3 m (40 ft) in length. It will be launched without the reactor ever having become critical, so that in the event of a launch failure, the only radioactive contaminant will be the uranium fuel (100-150 kg (220-330 lb) of non-weapons grade enriched uranium).

If the NCPS is being used as an upper stage, it will not be started until successfully reaching space. If being used to propel a deep-space craft, the NCPS will not be started until the craft is ready to leave orbit. The development cost is expected to total about four billion US dollars over a 10-12 year period.

NTREES (Nuclear Thermal Rocket Element Environmental Simulator)

The first step toward designing the next generation of thermal nuclear rocket engines is to test and certify new materials with which the maximum reactor core temperature can be increased. To drive this effort forward, NASA has designed and built a facility, called NTREES (Nuclear Thermal Rocket Element Environmental Simulator), that can safely and thoroughly test a wide range of nuclear fuel elements without danger to the environment.

Cross-section of an experimental fuel rod design for the Nuclear Cryogenic Propulsion Stage (NCPS) (Photo: NASA)

One of the proposed fuel assemblies is made of a cermet (a composite ceramic and metallic material) with a composition of sixty percent uranium dioxide (UO2) and forty percent tungsten metal. It is pierced with about one hundred holes that heat the hydrogen as it passes through the reactor. The holes are about two mm in diameter, and are clad with 90 micron thick tubes of tungsten-rhenium alloy. Each of the hexagonal fuel elements in the assembly is enclosed in a tungsten cladding sleeve 50 microns in thickness.

During the NERVA program it was found that similar fuel elements suffered from reduction of the UO2 into uranium metal that migrated within the cermet, causing structural faults. To stabilize the UO2, five to ten percent of gadolinium oxide was added to the uranium dioxide in the cermet. However, gadolinium is a good neutron absorber, and its use as a stabilizer forces an increase in the size of the reactor. Previous testing used a maximum temperature of 2,750° K, short of the 3000+° K design temperature for the NCPS.

The NTREES primary chamber and diagnostic equipment (Photo: NASA)

The NTREES facility is designed to test fuel elements and materials in hot flowing hydrogen, reaching pressures up to 1,000 pounds per square inch and temperatures of nearly 5,000° F (2,760° C) – conditions that simulate space-based nuclear propulsion systems to provide baseline data critical to the research team. NTREES is an ideal system in which to study the segregation phenomena described above as well as a host of other potential problems under realistic conditions.

"The information we gain using this test facility will permit engineers to design rugged, efficient fuel elements and nuclear propulsion systems," says NASA researcher Bill Emrich, who manages the NTREES facility at Marshall. "It's our hope that it will enable us to develop a reliable, cost-effective nuclear rocket engine in the not-too-distant future."

The video below shows a short fragment of a NERVA operational test.

Source: NASA, Idaho National Laboratory(PDF)

About the Author
Brian Dodson From an early age Brian wanted to become a scientist. He did, earning a Ph.D. in physics and embarking on an R&D career which has recently broken the 40th anniversary. What he didn't expect was that along the way he would become a patent agent, a rocket scientist, a gourmet cook, a biotech entrepreneur, an opera tenor and a science writer. All articles by Brian Dodson


Maryland, USA

The problem is that nuclear rockets of this type are only good for pushing things into orbit. Ion engines are much more efficient for deep space applications. A nuclear aerospace plane* could theoretically work but North Korea is just about the only country that could fly it because their leadership doesn't care about public opinion.

*The design I envision would be larger than an A380 but shaped like an Avro Vulcan. If possible I would use distilled water as reaction mass and steam turbine powered ducted fans for takeoff and to clime to in excess of 50 000 feet (15 240 meter).


Finally someone heard what engineers have been telling congress for 40 years. Chemical rockets will never truly "open" space.

Nextbigfuture covered these as a side discussion of the Orion project. http://nextbigfuture.com/2007/07/gaseous-core-nuclear-design-liberty.html

The liberty ship they describe is using the next generation of engine the NERVA engineers were planning after proving the solid cores described here, gaseous core vortex engines. While the solid cores "merely" double ISP the gaseous cores were expected to perform more than twice that efficiently. The important line is about using a vehicle the size of the Saturn V it could "...conservatively deliver 1000 tons to low earth orbit" with a reusable vehicle.

I also agree that with the fud and misinformation that has been incorporated into the mind of the general public it would be very difficult to get one of them off the ground as all of the people who are convinced there is no such thing as safe nuclear anything fought it tooth and nail despite the proven track record of the US nuclear industry.


These plans specifically refer to use after launch, the public would freak if they heard of plans for using anything nuclear in the atmosphere and that isn't going to change anytime soon. There are some advantages to assembling such a craft in orbit anyway.

On the topic of orbital construction (of both nuclear rockets and space stations and such) current rockets are designed to carry their payload in a dense region on top so anything big has to be put in a big rocket, assembled from smaller parts or expanded/unfolded. While these are good solutions, being able to bring up long building materials (beams and girders effectively but not necessarily the steel ones we are familiar with) would allow more strength and simplicity for many projects. Would a small cluster of long rockets surrounding a core cargo area be as feasible as the current put-it-all-on-top model?

Snake Oil Baron

To clarify my "V-ger!" post: I was confusing two different sci-fi plots involving Voyager probes: The probe renamed V'ger by the creators of a gigantic energy-based alien spaceship in Star Trek: The Motion Picture: http://www.startrek.com/database_article/vger

and the atomic-powered Voyager in the 12th episode of Space: 1999: http://en.wikipedia.org/wiki/Voyager's_Return

Maryland, USA

Everything we need to move into the solar system and colonize it has already been invented, most of it tested, some of that extensively tested - then tucked away in some back closet and abandoned.

If not for the determination of people like Senator William Proxmire to stop it, the near space environment today would look like the science fiction of the 1960's depicted the early 21st century.

Thank you, luddites, for keeping the human race "down on the farm".

Gregg Eshelman

My Father worked on NERVA and was there in NV for the test. According to him, the only reason why the rocket ran for only an hour was because of running out of fuel. He also told me that the engine was designed to be used in short bursts and have a lifespan of only 20 minutes. They were all pleased with the results but he is still very disappointed to this day NERVA was cancelled. I told him about this article yesterday and the nearly 80 year old man feels a little vindicated and hopes this comes to fruition. NERVA was my dad's all time favorite project he worked on. PS: He had a good time living in Vegas for a few months!

Post a Comment

Login with your Gizmag account:

Related Articles
Looking for something? Search our articles