Nuclear, NERVA, and NASA's Grand Tour
Nuclear, NERVA, and NASA's Grand Tour
Can Nuclear Thermal Engines Help Open the Solar System?
Nuclear Propulsion
The Key to Interplanetary Travel
The goal of a rocket engine is to take advantage of Newton’s third law of motion: for every action, there is an equal and opposite reaction. By pushing mass (propellant) away from the vehicle, the engine creates axial thrust in the opposite direction. The faster that an engine can expel the most mass possible, the more thrust it will generate. To do this, liquid engines use complicated chemical reactions and mechanical systems to increase the pressure in the combustion chamber and the efficiency of the propellant combustion. The measurement of this fuel efficiency is known as specific impulse, or Isp. Measured in seconds (s), Isp is essentially equivalent to the miles per gallon of a car. It is the objective of every rocket engine designer to improve Isp, as each second of added performance is additional payload that the rocket can carry to orbit. Isp also serves as a useful benchmark to compare the efficiency of different engine cycles, fuels, and architectures.
As you can see in the chart above, different engine styles are able to achieve different local maximums with respect to specific impulse. The standard solid rocket motors top out around 290 seconds, liquid rockets range anywhere from ~300 seconds to ~450 seconds depending on the propellants and engine cycles. Electric ion thrusters commonly found on satellites can get into the thousands of seconds and air breathing propulsion such as turbojets and turbofans can get upwards of 10,000 seconds. Despite their seemingly superior performance, these engines come with caveats that make them unsuitable to power anything as energy-dependent as a rocket. Ion thrusters are limited by their power and fuel molecular density which limits their thrust values to orders of magnitude lower than chemical rockets. Air breathing propulsion is limited by…well…air, and cannot operate once the atmosphere loses the required oxygen density (~50,000 feet for most turbofans). While decades of research have gone into optimizing solid and liquid chemical rocket engines (as well as ion engines and air breathers), there is still a largely unexplored category that could potentially break the current rules of space travel - nuclear.
There are two main forms of nuclear powered spacecraft engines, nuclear electric propulsion (NEP) and nuclear thermal propulsion (NTP). Nuclear electric engines are similar to the ion thrusters current flying on satellites that use solar panels to generate electric fields that speed up particles of fuels, such as xenon. The nuclear electric engine utilizes the same principles but would generate the electricity via miniature nuclear reactors on board the satellite. This advancement would significantly increase the performance of satellite propulsion by improving energy efficiency, particularly for deep space missions, but the engine itself would largely remain the same as current ion thrusters.
Nuclear thermal is a whole different ball game.
A nuclear thermal engine is more akin to today’s liquid rocket engines which use complex turbomachinery to bring a fuel and an oxidizer to high pressures before igniting them to create an extremely energetic chemical reaction that expels the hot propellants at high speeds to create thrust. Similar to the chemical rocket engine, a nuclear thermal rocket engine looks to expel hot propellant (typically hydrogen) but instead of relying on combustion, it utilizes the heat of a nuclear reactor to bring the fuel to incredibly hot temperatures before it’s pushed through the engine to generate thrust. By using a nuclear reactor, you avoid the headaches caused by the complexities of combustion physics and gain efficiency in the process. Compared to a chemical rocket’s Isp between 300-450 seconds, nuclear thermal engines can reach upwards of 950 seconds while maintaining the levels of thrust necessary to propel a standard launch vehicle.
This all sounds great! So why don’t we see Falcon 9 running on a cluster of nuclear powered Merlin engines? Well, there’s a complex and, frankly, frustrating history that explains the hold up in development for these types of engines, which could potentially unlock the key to interplanetary human spaceflight.
From Atomic Bombs to Atomic Engines
In 1946, less than one year after the United States dropped the world’s first nuclear weapons on Hiroshima and Nagasaki, scientists at Los Alamos Laboratory (home of the Manhattan Project) published the first known paper that references the use of atomic energy for spaceflight. Around the same time, sparked by the observed power of the atomic bombs, researchers in the United Kingdom and China were similarly drawn to the idea of long distance space missions powered by nuclear reactions. As with most aerospace innovation, conflict helped expedite technological development, and with the growing threat of a Cold War with Russia, the United States began to investigate ways to improve the intercontinental ballistic missile (ICBM) fleet.
One way was to install a nuclear powered upper stage on the Atlas fleet of ICBMs which would allow for heavier weapons payloads and longer duration flights (most notably, the ability to launch to all parts of the Soviet Union). By January 1957, scientists from Los Alamos, Livermore, and the National Advisory Committee for Aeronautics (NACA), were able to apply political pressure to secure $100 million in funding from the Atomic Energy Commission (AEC) to begin research on a nuclear powered ICBM.
However, in parallel to this research, nuclear weapons miniaturizing and the Atlas missile program’s liquid rocket engines were performing better than expected. It became clear that a nuclear-powered ICBM wasn’t totally necessary and the nuclear rocket research was consolidated at Los Alamos and reassigned to spaceflight missions under the name Project Rover.
In October of 1957, the Russians launched Sputnik 1 and further intensified the United States’ interest in space access. One year later, on October 1, 1958, NASA was officially formed and Project Rover was one of the first responsibilities officially transferred to NASA’s control. The project became a joint mission between NASA and AEC, organized under the newly formed Space Nuclear Propulsion Office (SNPO). The office set out to prove the baseline capabilities of nuclear rocket engines and improve on the reliability of hydrogen-based nuclear reactors.
By 1961, Los Alamos, SNPO, and NASA had developed the first testbed engine, known as Kiwi (since it was never meant to fly). Built in just 2 years, Kiwi went on to run a series of tests between 1959 and 1964 at Jackass Flats test site in the Nevada desert. Throughout these efforts, the project demonstrated hydrogen as a fuel for nuclear propulsion, advanced material compatibilities, proved that nuclear engines could be clustered from a single reactor, and ran hotfire tests at full power levels of over 1000 MW and 825 seconds of specific impulse. During this time, funding for Project Rover ran upwards of $500 million under the Kennedy administration.
NERVA
In 1961, building on the early successes of the Kiwi demonstrator, SNPO began conceptualization of the Nuclear Engine for Rocket Vehicle Applications, or NERVA. Where Kiwi was never meant to resemble a real engine, NERVA was set to be a full flight prototype for a nuclear thermal rocket. A proposal was sent to industry providers and eight contractors submitted bids: Aerojet, Douglas, Martin, Lockheed, North American, Rocketdyne, Thiokol and Westinghouse. Aerojet and Westinghouse were selected to design the NERVA engine, with Westinghouse focused on the reactor and Aerojet on the integrated system.
SNPO settled on a 75,000 lbf, hydrogen-fueled engine with a highly enriched uranium reactor. It was essential to the Kennedy administration that NERVA not take away from Aerojet’s and Westinghouse’s critical support of the Apollo project but by 1963, over 1100 people were working on NERVA. Within three years, by 1964, Aerojet and Westinghouse ran the first tests on NERVA and flawlessly reached full power levels, unsupported by any external power sources (as had been the case with Kiwi). A series of successful tests followed in subsequent years and NERVA showed the ability to restart a nuclear engine, power balances for turbines, control systems for long duration operations, and startup transients. In terms of an R&D program, NERVA was a complete success.
With the success of Kiwi and NERVA, NASA planned for a series of tests and long term missions utilizing the NERVA engine. The first proposed program consisted of Reactor In-flight Test (RIFT) flights throughout the 1960’s and 1970’s. RIFT looked to place the NERVA engine on a modified Saturn V upper stage to demonstrate launch capabilities for future space missions. 42 total tests of the system, including 6 test flights, were scheduled and NASA estimated that a nuclear-thermal enabled Saturn vehicle could bring launch costs down to $700/kg to LEO! Additional plans for NERVA included supporting a Mars mission by 1978, a lunar base by 1981, and a “Grand Tour” of Jupiter, Saturn, Uranus, and Neptune between 1976 and 1980 when the planets were at their most favorable alignment (which only occurs once every 174 years). With a nuclear powered shuttle and NERVA propelled space tugs, the travel times of these missions was a fraction of what it would take with traditional propulsion methods.
Unfortunately, reality got in the way of NASA’s big vision. NERVA and SNPO struggled to work around the Partial Nuclear Test Ban Treaty and the National Environmental Policy Act which looked unfavorably on nuclear testing. With large budgets and intense political scrutiny, NERVA struggled to scrape by the NASA budget cuts in 1969 that closed the Saturn V production line. Finally, with additional pressures from the Vietnam War and a struggling job market, President Nixon abruptly cancelled the NERVA project in 1973, despite bipartisan support in Congress. After 17 years and $1.4 billion spent, the U.S. never launched a nuclear thermal engine.
What’s Next - NewSpace NERVA?
Despite NERVA’s shortfall, the dream of nuclear thermal propulsion continued for the last five decades. In 1983, as part of the Strategic Defense Initiative (Star Wars Program), President Ronald Raegan asked the Depart of Defense and industry partners to explore various methods to protect the United States from nuclear warheads. Part of the program included investigating a 100 KW nuclear thermal rocket engine with performance greater than 1000 seconds of Isp. Nicknamed Project Timberwind, the engine program was taken over by the Space Nuclear Thermal Propulsion (SNTP) program at Air Force Phillips Laboratory. After roughly 7 years of design and $200 million spent, the program was shuttered in 1994 when NASA and SNTP determined that performance of the engine did not surpass NERVA from years earlier and did not offer a strategic advantage to the US missile defense initiatives.
Following Project Timberwind, research on nuclear thermal propulsion progressed at NASA and at the university level, but had no substantial government backing until 2020 when DARPA announced their “demonstration rocket for agile cislunar operations”, or DRACO, program. The goal of DRACO is to lean on industry to reinvigorate the idea of a nuclear thermal powered spacecraft for cislunar and interplanetary missions. The first phase of the program was posted in April 2021 and awarded $22 million to General Atomics to develop a nuclear reactor and roughly $2 million to both Blue Origin and Lockheed Martin to mature nuclear powered spacecraft designs. The second phase of DRACO was announced on May 4th of this year and released an open call to industry to design, develop, fabricate and assemble a nuclear thermal rocket engine for a flight test by 2026.
NASA and the Department of Energy also awarded $5 million in 2021 to BWX Technologies, General Atomics, and Ultra Safe Nuclear Corporation to further their design strategies and requirements definition for space nuclear reactors. The FY2021 NASA budget allocated an additional $110 million towards nuclear thermal propulsion with $80 million of that dedicated to a flight test program. It is clear that, once again, the eyes of government leaders are turning towards the promise of rapid maneuverability and long duration space missions enabled by NTP.
As technology across the space industry matures and trips to Mars and beyond become more feasible, nuclear thermal propulsion will likely continue to garner attention. Capable of reducing travel time to Mars by more than 50% (6-9 months to 2-3 months), NTP can have a cascading effects for the health of astronauts who may struggle with radiation and isolation during long duration space missions. For uncrewed flights, nuclear thermal propulsion may offer the ability to increase payload sizes, decrease transfer times, and enable longer duration science missions. Regardless of its end use case, it seems apparent that this technology will be a focus for NASA, the DoD, and commercial providers in the coming decade. NERVA exhibited the idea of a high performing, safe, reliable nuclear rocket engine but was plagued by geopolitics and cost. Though fear of fallout from a spacecraft failure continues to linger in the backdrop of many nuclear-based space programs, with a much more mature private space industry and (hopefully) a more open dialogue surrounding nuclear technologies, it seems possible that DRACO or another NTP project will take flight before 2030.
Bonus: Diagrams and report from Harold Finger who led the NERVA Program
Bonus: Awesome “Grand Tour” artwork from JPL
