By Andy Tomaswick
May 27, 2025
Nuclear thermal propulsion (NTP) is a promising potential alternative propulsion technology for decades. Chemical rockets have started to reach their maximum theoretical efficiency, and their developers have changed concentration to make them cheaper rather than more effective. The NTP should respond to this by offering a strong push and a specific impulse. The Draco de la NASA program, the standard of NTP systems, provides a specific impulse of approximately 900 seconds, about double a traditional chemical rocket, but half that of most of the ion propellants. To further increase this number, researchers from the University of Alabama in Huntsville and Ohio State University have worked on a new NTP configuration called Centrifugal Nuclear Rocket (CNTR) which almost promises to double the specific impulse of traditional NTP systems while retaining similar levels. However, the system has engineering challenges to overcome, and a new outgoing document in Acta Astronautica describes certain progressive progress to make this improved engine a reality.
The greatest difference between the CNT and traditional NTP projects is the fact that CNTR uses liquid uranium for fuel, rather than the solid type. To keep its fuel fluid, a CNTR system must run it quickly in a centrifuge. Once the uranium is melted, the CNTR Bubble hydrogen through it and expels it from a nozzle for a thrust reaction. With this system, the researchers believe that they could obtain a specific impulse of around 1500 seconds, almost double what a traditional NTP engine would have, while only having less push.
This seems to be a significant improvement because the use of these rockets could considerably increase the “Delta-V” achievable. But, as with all important engineering challenges, there is a catch. Ten of them, in fact. The document mentions ten engineering challenges which retain the development of the system, ranging from the development of a coating which can manage liquid uranium and all the different types of propellants at high temperatures to manage the transient vibrations of the system. Throughout their work on the article, which is the fourth in a series describing the development of this engine, the authors have decided to focus on four specific challenges, one of which could have the most significant impact on the specific engine impetus – the vaporization of uranium.
Fraser discusses the Draco project, which hopes to test a nuclear rocket by 2027.
First of all, they approached the neutronics of the system, or how the different elements created by the nuclear reaction have an impact on the operation of the engine. In this iteration, the researchers added the ERBIUM-167 to their model to manage the temperature inside the system, which makes it more stable. However, they also noted that the nuclear fission process could be “poisoned” by xenon and the Samarium created as part of it if they are not effectively removed from the system. They will continue to develop models and simulations to find the best way to eliminate these elements without exacerbating the possibility of losing more uranium.
The bubbles are the second challenge – the hydrogen bubbles which are necessary to overheat the fuel, then eject it are difficult to model. The researchers turned to two experimental configurations, “Ant Farm” (static) and “Blender II” (rotation), to try to understand how hydrogen bubbles move through the liquid fuel. Blender II includes an x -ray to examine bubble dynamics in the prototype systems they have developed that use Galinstan, a gallium alloy, to act as an equivalent of uranium and nitrogen to act like hydrogen. Bubble cartography remains a challenge mathematically, however.
An area where modeling was better understood was the integration of the engine. The researchers used a genetic algorithm to maximize the output of the engine given the different parameters, resulting in a specific impulse under ideal conditions of 1512 seconds. However, it would require additional centrifugal and higher rotation rates than the original design.
This video deals with CNTR design details, compared to other NTP concepts.
Credit – Channe YouTube Space Eager Space
But all these pale challenges compared to the big – how to prevent uranium fuel from fleeing in the nozzle. If too much uranium (rather than hydrogen) is lost in thrust, it could slow down the nuclear reaction, reducing the specific potential impetus of the engine by two thirds. The authors believe that they can use a technique called Diélectrophoresis (DEP) to isolate vaporized uranium molecules and redirect them in centrifugal, but even with a recovery rate of 99%, there could be massive impacts on the viability of the engine.
As with all good research projects, more research is needed. Researchers recognize in the article that the CNTR is not yet ready for a complete prototype and will still require several modeling and optimization cycles before it is ready. The loss of uranium will draw special attention with the next series of research, with a bench test planned to test the DEP solution. While researchers continue to work on it and the need for high / high level impulse propulsion systems continuously grow, the CNT will be hopeful, to maintain funding until it is finally materialized.
Learn more:
D Thomas et al – Return to the challenges of the engineering of the centrifuge nuclear thermal rocket
UT – Are nuclear propulsion systems the future of spatial exploration?
UT – New design of nuclear rockets to send missions to March in just 45 days
UT – NASA invests in a new concept of nuclear rocket for the future of spatial exploration and astrophysics