Getting humans to Mars by means of chemical rockets would take approximately 620 days, most of that time being spent coasting through empty space with the fuel being burned only at the beginning and at the end. The figures could be greatly changed if a nuclear rocket were used, and a recent design from NASA and General Atomics indicates that this can be done with a single reactor which carries out two entirely different tasks at the same time.
In 2025 Kurt Polzin, NASA’s chief engineer in charge of the Space Nuclear Propulsion Project at the Marshall Space Flight Centre, and Robert Schleicher, chief engineer of nuclear technologies at General Atomics, began to work on the idea. The Synchronal Bimodal Nuclear Reactor, or S-BNR, was featured in the September 2026 edition of IEEE Spectrum and offers a solution to the two major engineering problems associated with deep-space travel - namely, the need for high thrust when carrying out large manoeuvres and the requirement for continuous electrical power for all the other needs.
Two loops in one core
Earlier bimodal reactor designs had tried to switch a single group of fuel elements between propulsion mode and power-generation mode using complicated valve systems. Since the valves operated at temperatures above 2,000 K in a highly radiation-intensive environment, they clearly posed a serious reliability problem. The S-BNR eliminates the problem altogether by having two hydraulically independent loops within the same reactor core.
The first stage involves the use of a thermal propulsion system, in which liquid hydrogen flows through high-temperature fuel elements, or HTFEs, these containing uranium pebbles. The hydrogen is heated to a temperature of over 2,700 K and is then expelled via a nozzle, thus producing thrust with a specific impulse that exceeds 900 seconds. To make this clearer, the main engines of the Space Shuttle achieved a specific impulse of about 450 seconds when they used both liquid hydrogen and liquid oxygen. In the case of a nuclear thermal rocket the specific impulse is approximately doubled since it is not necessary to carry an oxidiser. The energy for the hydrogen propellant is obtained from the reactor and not from a chemical reaction.
The second loop works on its own. A mixture of helium and xenon flows through low-temperature fuel elements, or LTFEs, which function at about 1,200 K. The gas, after being heated, drives a turbine that is linked to an electrical generator and thus yields a continuous power output ranging from tens of kilowatts to several megawatts. This electricity can be used to operate life-support systems, scientific instruments, communication equipment, and even electric propulsion thrusters for making fine adjustments to the trajectory during cruise phases.
The two loops have the same reactor core but their fluid circuits are not shared. As a result, the thermal propulsion system can carry out the orbital insertion burns while the power-generation loop at the same time supplies electricity to the rest of the spacecraft. There is no need for valve switching, no need for reconfiguration and no downtime.
The decrease in transit time is considerable. The figures currently used by NASA for a Mars mission powered by chemical propulsion are about 620 days; however, the calculations made by Polzin and Schleicher indicate that the S-BNR could reduce this to 335 days or less. Since the journey is shorter, the crew will be exposed to less radiation, the mass of the supplies will be reduced, and the psychological effects of isolation will also be diminished.
Engineering Deep Dive: Nuclear Thermal Propulsion and Specific Impulse
Specific impulse, expressed in seconds, is the most important performance figure for a rocket engine since it shows how efficiently the engine turns propellant mass into thrust. In a formal sense, it is the amount of thrust produced per unit weight of propellant used each second, and the greater the specific impulse, the more velocity change the rocket gets from each kilogram of fuel.
Chemical rockets are constrained by the energy obtained during combustion. The most energetic practical mixture, liquid hydrogen and liquid oxygen, results in exhaust gases at about 3,500 K and a specific impulse of around 450 seconds. A nuclear thermal rocket overcomes this limitation by not burning fuel but instead passing a propellant (pure hydrogen) through the reactor core where it is heated directly by nuclear fission. The exhaust temperature can go above 2,700 K, and since hydrogen has the lowest molecular mass of all elements (2 g/mol as H₂), the exhaust velocity is very high. This relationship is described by the ideal rocket equation covered in A-level Physics: exhaust velocity is proportional to the square root of temperature divided by molecular mass. The combination of low molecular mass and high temperature leads to a high exhaust velocity, which in turn gives a high specific impulse.
The fact that uranium pebble fuel has been selected for use in the HTFEs is the result of many decades of development. The pebble-bed reactor concept involves the use of spherical fuel elements, each of which contains thousands of small uranium kernels surrounded by layers of pyrolytic carbon and silicon carbide. The TRISO (tristructural isotropic) particles are capable of withstanding temperatures in excess of 1,600°C without releasing the radioactive fission products. In the case of the S-BNR, the pebble bed offers a large surface area for transferring heat to the hydrogen propellant, and the coating layers function as small containment vessels for each fuel kernel.
The fact that a helium-xenon mixture is used in the power-generation loop is an interesting engineering decision. Helium has great thermal conductivity and a high specific heat ratio, which makes it an efficient working fluid for Brayton-cycle turbines. Since xenon is a much heavier noble gas, it is added in order to increase the molecular weight of the mixture and thus improve the turbine’s power density by lowering the rotor speed that is needed for a given amount of power output. Because neither gas is chemically reactive, there is no risk of corrosion in the high-temperature loop.
The power loop’s thermodynamic cycle is a closed Brayton cycle, not a Rankine cycle. In a Brayton cycle the working fluid stays in a gaseous state the whole time. The hot gas from the LTFEs expands through a turbine, carries out work, and is then cooled by a radiator before being recompressed and returned through the reactor. Since there is no atmosphere in space to dump heat into, the radiator has to depend entirely on thermal radiation. The Stefan-Boltzmann law applies to radiative heat transfer: the power radiated is proportional to the fourth power of the radiator’s absolute temperature. It therefore follows that the radiator must be large and its design is one of the main mass penalties in any space nuclear power system.
Any students who are familiar with Newton’s third law will be able to identify the basic physical principles behind rocket propulsion. The hot gas which is expelled backwards through the nozzle exerts on the spacecraft an equal and opposite reaction force. The Tsiolkovsky rocket equation, which connects the change in velocity (delta-v) with the exhaust velocity and the ratio of the initial to the final mass, is used in planning missions. Since a nuclear thermal engine has twice the specific impulse of chemical rockets, it can achieve the same delta-v using considerably less propellant, or it can achieve a much greater delta-v when the amount of propellant is the same. In the case of a mission to Mars, this means shorter transit times and a lower launch mass.
The problem that earlier bimodal designs had difficulty solving was the valve system which linked both modes to the same fuel elements. Valves having to function at 2,000 degrees above absolute zero had to seal perfectly against hydrogen under high pressure while withstanding neutron bombardment that causes most materials to become brittle after a time. By using a two-loop system the S-BNR removes this difficulty even though its core becomes a little more complex and includes two kinds of fuel elements. Polzin and Schleicher clearly considered that the passive separation of the fluid circuits was more reliable than active valve-controlled switching, which is a typical engineering compromise between complexity and reliability.
The flying of a nuclear thermal rocket came to an end when the test firings of the NERVA programme concluded in 1973. Regulatory, political, and financial obstacles are still as great as the engineering ones. However, the S-BNR design deals with one of the most persistent technical criticisms of nuclear spacecraft, the difficulty of combining propulsion and power generation, in a way that might one day prove difficult to overlook.
- NASA and General Atomics have designed a Synchronal Bimodal Nuclear Reactor (S-BNR) that combines thrust and electrical power generation in one unit
- Two independent fluid loops share one reactor core, eliminating the unreliable valve systems that plagued earlier bimodal designs
- The thermal propulsion loop heats hydrogen to over 2,700 K, achieving specific impulse above 900 seconds, roughly double that of chemical rockets
- The power-generation loop uses a helium-xenon Brayton cycle to produce kilowatts to megawatts of continuous electricity for onboard systems
- The design could reduce Mars round-trip transit from 620 days to 335 days, cutting crew radiation exposure and consumables mass
