The Augustine committee’s report last year on the U.S. Constellation program for manned space exploration exposed its problems — it was mired in cost overruns and behind schedule — heralding the collapse of the 2004 Vision for Space Exploration, termination of the space shuttle and a dim future for the international space station, major accomplishments of the U.S. space program. The report did not identify the root cause of this misery, nor offer any remedy.

The root cause is mass, driven by the inadequate performance of chemical rockets (liquid oxygen and hydrogen) and the consequent costs of new, heavy-lift freight rockets and ferry flights to low Earth orbit (LEO), mostly propellant for the Mars mission engines.

This situation can be dramatically remedied by exploiting the higher performance of nuclear thermal rockets for transit to Mars. They demand much lower propellant mass and consequently enable cost-effective utilization of the space shuttle (the Shuttle-C design, which replaces the orbiter with a cargo carrier) to ferry cargo to LEO, avoiding the need for the Ares 5 Earth-to-LEO transporter.

Nuclear propulsion was very successfully accomplished 50 years ago in the joint NASA/U.S. Atomic Energy Commission Rover/NERVA (Nuclear Engine for Rocket Vehicle Application) program. By the late 1960s, graphite composite fuel elements delivering 925 seconds Isp (specific impulse, a measure of rocket engine performance) were proved in nuclear test rockets Kiwi B, Phoebus 1 and Phoebus 2, and Pewee, producing thrust levels from 12,000 pounds to 210,000 pounds. The rockets were declared ready to begin flight qualification.

Recapturing the graphite composite fuel element production capability now provides a powerful head-start for cost-effective revival of the nuclear rocket. Test-verified next-generation multi-carbide fuel elements, capable of even higher temperatures (about 3,100 kelvins), could enable a graded hybrid core delivering over 1,000 seconds Isp, compared with about 465 seconds Isp for the best liquid oxygen and hydrogen rockets. 

Review commissions and many workers have conducted mission mass studies with nuclear propulsion, and have unanimously found reductions to one-third or less of the mass required by chemical propulsion. The reduced mass under the nuclear option would save billions of dollars compared with the best chemical propulsion. A necessary change in propulsion technology is thrust upon us because chemical rockets cannot meet the technical and practical demands of manned Mars missions.

The high performance of the nuclear rocket, over twice that of the best chemical rockets, is a decisive factor that could revive the Constellation program.

The many ferry flights of propellant stores to LEO, demanded for each mission by chemical propulsion to Mars, are greatly diminished by utilizing the much-higher-performing nuclear rockets for Mars transport propulsion. This reduction in ferry flights enables utilization of lower-cost, more-reliable and safer shuttle-derived transporters from Earth to LEO. Shuttle-C (side-mount) derivatives offer 100 metric tons of cargo capacity to LEO with proven launch systems, operations and existing personnel; there are no huge, risky new Ares 5 heavy-lifters costing billions to develop and later billions to launch, no change in ground facilities and operations, and no retraining of personnel. On-orbit logistics would be performed by staff on the international space station. This approach would save many billions of dollars versus chemical rockets, even at the inaugural mission, and preserve America’s leadership and stand-out accomplishments in space.

Further motivation is provided by the actions of our friendly competitors in Russia. The Soviet satellite launch in October 1957, followed by the launch of a cosmonaut to orbit in April 1961, were a major stimulus to the U.S. Apollo and Rover/NERVA programs. Rover/NERVA set out to develop and qualify a nuclear rocket, which culminated in a prototype flight point design by 1969. The Mark 9 liquid hydrogen turbo-pump built by Rocketdyne for Rover nuclear rockets led to the J-2 (and J-2X) turbo-pumps of Apollo and Constellation. Our motivation to establish pre-eminence in space technology has not changed.

Since achieving the great success with nuclear rockets through the 1960s, repeated examinations of the U.S. space program by presidential commissions, while recognizing benefits of nuclear propulsion, took no action to recapture the nuclear rocket. When the government shut down nuclear rocket development in 1972, the Russians were amazed at us, and incredulous. The Russian nuclear propulsion work had never risen to our extensive full-scale testing of nuclear rocket engines, but they continued work to develop viable nuclear rocket engine designs with refractory metal-fuel elements. At present, the Russians, with top political backing from President DmitryMedvedev, are taking the initiative to build the nuclear propulsion module for their next Mars spacecraft.

At the same time, the Russians have reneged on their long-standing deal to sell plutonium-238 (Pu-238) to NASA. This surprise capped a refusal by Congress to provide $30 million for the United States to resume Pu-238 manufacture. Fortunately, President Barack Obama recently renewed the Pu-238 funding request. Human exploration of space is not likely to proceed without radioisotope power fueled by Pu-238.

While public fear of radiation must be addressed in this context, the risk is greatly overblown. The developments and testing of nuclear rockets throughout the 1960s were, in fact, remarkably safe. No notable radiation injury occurred at the Nuclear Rocket Development Station in Nevada over a decade of nuclear rocket operations. The United States’ extensive experience with nuclear materials in space and ground fission power systems ensures that nuclear rocket development would not present a public health hazard. Nuclear-test rocket exhaust management can include: discharging the exhaust into a deep hole able to absorb it into surrounding alluvium; cooling the exhaust through two stages of heat exchangers, eventually trapping all radioactive debris in an auxiliary container, with the cleaned hydrogen eliminated via a burn stack, a method proposed by Rocketdyne test stand engineers; and installing a three-stage exhaust cooling system that reliquifies the cleaned hydrogen for return to the propellant run tank and recycling.

A number of other merits of nuclear propulsion need to be mentioned:

  • Reduced flight times, with lower crew space exposure or more massive, lower-cost unmanned cargo flights.
  • More flexibility to choose low-time higher-energy transits, or possibly rescue mission considerations.
  • Scalability. The Rover/NERVA experience clearly demonstrated this, with nuclear rocket engine core diameters of 50.8 centimeters,  88.9 centimeters and 139.7 centimeters all performing pretty much as intended. This benefits final flight design decisions.
  • Proven operation of clustered reactors. This validates design work at NASA’s Glenn Research Center focused on a cluster of three nuclear rockets designed to deliver mission power as well as thrust, and provide engine-out capability.
  • Radiation effects on the hydrogen propellant demonstrated turbo-pump capability for 30 percent mixed-phase flow, and revealed no shielding issues for engine subsystems. The critical questions were asked and answered in Rover/NERVA, and there were no show-stoppers.

The nuclear rocket application also saves funds because it would pick up from where the work was stopped after a $1.5 billion investment, to recapture the composite and multi-carbide fuel, the core design and controls, the well-proven testing facilities and protocols. Obvious options for a new Nuclear Rocket Development Station include the old station at the Nevada Test Site, the Idaho National Lab and NASA’s White Sands Test Facility in New Mexico.

With Constellation at this crucial nexus, serious consideration of a new direction to nuclear propulsion is warranted.


V. Gunn is a retired Rocketdyne Rover/NERVA lead engineer. Ernest Y. Robinson is a retired nuclear engineer from Aerospace Corp. who worked in the Pluto Nuclear Ramjet Program at the Lawrence Radiation Laboratory at the University of California, Berkeley.