Chronicles of Soviet Nuclear Rocket Engines

4 877 15
Chronicles of Soviet Nuclear Rocket Engines


Decades before Burevestnik


Today, a nuclear engine is still perceived as something unusual and almost exotic. Based on publicly available information, only one Russian nuclear-powered product is actually in serial production—the strategic cruise missile. Rocket "Burevestnik." However, in the 1950s, such projects didn't seem out of the ordinary. Riding the wave of post-war scientific optimism and advances in atomic physics, many engineers believed that the creation of nuclear engines wasn't just a matter of fundamental possibility, but rather a matter of the near future.



The idea of ​​using a different, much more concentrated form of atomic energy for propulsion, rather than chemical energy, arose long before the advent of actual reactors. As early as the late 1920s, Valentin Glushko conducted experiments at the Gas Dynamics Laboratory in Leningrad using the electrical explosion of a metal wire. He was interested in the possibility of generating thrust without a traditional oxidizer. Light metals, primarily lithium, were considered in his experiments.


Valentin Petrovich Glushko

By 1933, this research had led to the creation of a small electrothermal engine. The principle was simple: an electrical pulse converted the working fluid to a high-temperature state and ejected it through a nozzle. For its time, this was a truly groundbreaking idea, but it lacked a key element: a compact power source. Electric generators were too heavy, meaning the design had no practical application.

During those same decades, Konstantin Tsiolkovsky proposed the possibility of harnessing the internal energy of matter for interplanetary travel. Later, after World War II, when nuclear reactors became an engineering reality, this idea took on concrete form. If a reactor can heat the working fluid to extreme temperatures, then theoretically it could replace a chemical combustion chamber and provide a much higher specific impulse.

The fundamental advantage was obvious: a chemical rocket requires both fuel and oxidizer, while a nuclear rocket requires only the working fluid, most often hydrogen. This dramatically improved the energy balance. For comparison, the best oxygen-hydrogen chemical engines achieved a specific impulse of approximately 430–450 seconds, while solid-phase nuclear engines promised 800–900 seconds, and gas-phase engines, up to 1500–2000 seconds.


The American Kiwi reactor, built as part of Project Rover, 1960

The United States was the first to launch large-scale practical research. In 1955, the Rover program was launched, focused primarily on nuclear rocket engines for space missions, while Project Pluto focused on a ramjet nuclear engine for the SLAM ultra-long-range cruise missile. Preparations for hot-fire tests of Kiwi reactors began at the Jackass Flats test site in Nevada. Later, the Americans developed the NERVA series of engines, and in 1969, they achieved a thrust of approximately 25 tons on a test rig with a burn time of over ten minutes—one of the most impressive results in the world. stories nuclear rocket technology.

In the USSR, the signal for similar research was given as early as 1953, when Mstislav Keldysh was tasked with exploring the possibility of using nuclear energy in direct-flow systems. A group led by Vitaly Ievlev was formed at NII-1. It soon became clear that this was not a private project, but a whole new field of science and technology.


Vitaly Mikhailovich Ievlev

On November 22, 1956, the Council of Ministers of the USSR and the Central Committee of the CPSU issued secret decree No. 1529-769 "On the development of ballistic missiles with a nuclear engine." Sergei Korolev was appointed chief designer of the missile, Valentin Glushko and OKB-456 were responsible for the engine, and Alexander Leypunsky and the Obninsk Institute of Physics and Power Engineering were responsible for the reactor section. The Kurchatov Institute of Atomic Energy, TsIAM, TsAGI, VIAM, NII-9, and dozens of manufacturing enterprises also participated in the work.

Three schemes, three levels of difficulty


By the end of the 1950s, three main concepts had emerged.

Type A is a solid-phase engine. A reactor heats hydrogen, which expands and is expelled through a nozzle. This design was considered the most realistic, and it was the one later used by both Soviet and American designers.

Type "B" was a hybrid design in which, after reactor heating, the working fluid was further accelerated or burned in a combustion chamber. Theoretically, this provided a thrust gain, but added complexity to the system.

Type "B" was a gas-phase reactor, in which the nuclear fuel itself was in a gaseous or plasma state. This design promised record-breaking performance, but was extremely complex: it required containing the hot, fissile environment while preventing structural failure and fuel release.

A more radical option was also being studied in parallel: a ramjet nuclear engine. In it, atmospheric air passed through the reactor core and was heated without conventional combustion. But this design almost immediately ran into a fundamental problem: the air passing through the core itself became radioactive. Practical use of such an engine in the atmosphere seemed extremely dangerous.

On June 30, 1958, Resolution No. 711-339 redefined the program. Now, the focus was not only on military applications but also on heavy space launch vehicles with nuclear stages. Korolev envisioned such systems as a means for long-range expeditions to the Moon, Venus, and Mars. OKB-1 created a special division, led by Mikhail Melnikov, to develop nuclear and electric propulsion systems.

By the end of 1959, a preliminary design for a rocket was ready, in which the central reactor block was to be ignited after the system had entered the upper atmosphere. This was an important detail: even then, it was understood that launching a full-fledged nuclear engine near the Earth's surface was extremely risky. A number of designs assumed that chemical stages would place the vehicle on a safe trajectory, and only then would the nuclear power plant be activated.

Gradually, another idea matured at OKB-1: using the reactor not to directly heat the working fluid, but as a source of electricity. In this case, it would power ion or plasma engines. This approach yielded less thrust, but a much higher specific impulse and was better suited for long-duration space flights. Essentially, it was here that the foundations of Soviet nuclear space energy were laid. On June 23, 1960, new decree No. 715-296 established the course for the creation of powerful launch vehicles and spacecraft with nuclear stages. The program included 74 organizations, and their number later exceeded one hundred. It was a project of national significance.


The RD-0410 became the only domestic nuclear rocket engine made of metal.

The main challenges were not only in the reactor physics but also in the materials. The core and fuel supply channels had to withstand temperatures of 2500–3000°C, sudden thermal loads, vibrations, and neutron irradiation. To achieve this, molybdenum, niobium, graphite, beryllium, uranium and zirconium carbides, as well as high-temperature ceramics, were studied. A separate problem was the cracking of fuel elements during repeated startups.

Biological shielding was no less challenging. Early calculations estimated the reactor's mass, including shielding, could reach 20 tons or more. This was especially critical for manned missions: the crew needed to be shielded from neutron and gamma radiation without making the spacecraft unwieldy. This led to design solutions featuring long trusses, with the reactor positioned as far away from the living quarters as possible.

To test this concept, a special complex, "Baikal," was created at the Semipalatinsk test site. It was intended to conduct bench tests of reactor components and associated power systems. However, even preparing the test facilities proved enormously challenging. There was a shortage of electromagnetic pumps for liquid-metal circuits, the technology for pure refractory metals had not been established, and the industry was already overburdened with urgent defense programs.

Soviet projects and real results


In the late 1950s and early 1960s, OKB-456 developed a number of experimental engines: the RD-401, RD-402, RD-404, and RD-405. They differed in the moderator type, propellant, and core layout. Simultaneously, calculations were underway for more complex systems, including the gas-phase RD-600, fueled by hydrogen with added lithium. This project was nearing the ultimate in complexity, with magnetic confinement, nuclear fuel circulation, and extremely intense thermal conditions.


But by 1962, it became clear that the program was disintegrating. Numerous organizations were duplicating each other's work, some projects were lagging, and some areas were too far from achieving practical results. Vasily Mishin reviewed the entire cooperation and proposed cutting redundant staff and concentrating resources on key tasks.

The Cuban Missile Crisis was an additional blow. The military and political situation demanded rapid and comprehensive solutions. Chemical intercontinental missiles could be developed and deployed immediately, while nuclear engines remained a matter of an uncertain future. It was at this point that the program effectively lost its former priority.

However, it didn't disappear without a trace. On the contrary, it gave rise to many real advances in electric propulsion technology. In the 1960s, the USSR actively developed ion and plasma thrusters, which were later used to create attitude control and correction systems for spacecraft. Later, the Soviet school became one of the world's leading experts in the field of stationary plasma thrusters, known today from the SPT series, which are widely used on satellites.

The RD-0410 was the primary practical outcome of the entire Soviet nuclear rocket engine program. It was the project that truly brought the idea of ​​nuclear propulsion to fruition in engineering form. It was a solid-core rocket engine: its core contained a compact reactor that heated liquid hydrogen to extremely high temperatures, after which the superheated gas was expelled through a nozzle, generating thrust.

The designers created a compact reactor capable of stable operation under extremely high heat fluxes, selected materials that could withstand high temperatures, vibrations, and neutron irradiation, and ensured a reliable supply of hydrogen through the core without damaging its structure. Safety was no less challenging: testing such an engine required specialized infrastructure and exceptional care. But the RD-0410 remained a mere episode in Soviet technical history.

Why the nuclear missile didn't take off


The reasons were systemic.

First, complexity. The nuclear engine turned out to be more than just a new product, a component that required a revolution in reactor design, materials science, cooling systems, radiation protection, and ground infrastructure.

Secondly, there was the danger. Any accident at launch or during rig testing threatened radioactive contamination. Even if the engine was ignited in space, the actual launch of the reactor into orbit remained a sensitive task.

Third, economics. Chemical engines were inferior in theoretical efficiency, but superior in price, mass production, and technological readiness. For the military, this was the decisive argument.

Fourth, a shift in priorities. After the moon race and the shift in interest from ultra-expensive interplanetary programs to more practical applications, political support for nuclear space propulsion has weakened.

Although the nuclear rocket engine never became a mass-produced reality, the program left a significant legacy. It accelerated the development of electric propulsion, gave impetus to new materials and technologies for welding refractory metals, and strengthened cooperation between the nuclear industry and rocket and space design bureaus.

Moreover, it laid the intellectual groundwork for later Soviet and Russian space power systems. As early as the 1970s and 1980s, the USSR launched the Buk and Topaz series of reactor power systems into space, used on radar reconnaissance satellites. These weren't nuclear rocket engines in the strict sense, but they demonstrated that compact space nuclear power had moved from the realm of theory to practice.

Today, as humanity once again considers manned missions to Mars and beyond, the idea of ​​a nuclear rocket engine is experiencing a renaissance. Remarkably, modern projects—both Russian and foreign—are largely based on the very groundwork laid in secret design bureaus and research institutes six decades ago. What nearly became a dead end in technological progress has become one of the cornerstones of future space exploration. And this is perhaps the best reward for those who once—in the era of great hopes and the harsh realities of the Cold War—believed that the atom was capable not only of destruction but also of carrying humanity to the stars.
15 comments
Information
Dear reader, to leave comments on the publication, you must sign in.
  1. +2
    31 May 2026 07: 34
    Nuclear rocket engines were revolutionary in their time, but more powerful engines now exist. A trip to Mars, which currently takes almost a year, could be reduced to one or two months in the future. Just a few years ago, such claims would have been dismissed as science fiction. However, Rosatom's Troitsk Institute has already unveiled a laboratory prototype of a next-generation engine powered by a magnetoplasma accelerator, and its characteristics are currently being discussed not only by space experts but also by those who have long followed the development of promising technologies. Many experts believe that plasma engines could be the next big step in space exploration after chemical rockets. The fact is that traditional engines have effectively reached the limits of their capabilities, as increasing thrust requires a huge amount of fuel, which automatically increases the rocket's weight and complicates launch.

    The particle flow velocity that the new engine is capable of is approximately 100 kilometers per second, which is almost twenty times higher than the best modern chemical rocket engines. The Russian specialists' development is interesting because it uses a completely different operating principle. While traditional rockets obtain thrust through the combustion of fuel, the magnetoplasma engine uses electromagnetic fields that accelerate charged plasma particles to colossal speeds. Simply put, the engine doesn't "push" the ship in the traditional explosive manner, but creates a directed flow of ultra-fast particles. Hydrogen ions are the primary working fluid in the project, and the system itself is designed to convert electrical energy into plasma motion as efficiently as possible. This is what makes it possible to achieve such a high flow velocity and a significant increase in specific impulse compared to traditional technologies.
    Particularly interesting is the fact that the development already exists beyond computer calculations. Rosatom reported that a full-scale laboratory prototype of the engine has been created, operating in pulsed-periodic mode, with an average power of approximately 300 kW. For a scientific project of this scale, this is an extremely significant achievement, as it is no longer just a theoretical idea, but a real engineering system.


    https://dzen.ru/a/afxELGZUSGn7fQq8
    1. +3
      31 May 2026 08: 51
      A flight to Mars, which currently takes almost a year, could be reduced to one or two months in the future.
      Let me ask an amateur question: what acceleration will the vehicle have to achieve when accelerating and braking?
      1. -1
        31 May 2026 09: 19
        The spacecraft will be launched using a traditional method. According to the researchers, the propulsion system will be capable of accelerating particles (charged electrons and protons) to speeds of 100 km/s. The engine operates in a pulsed-periodic mode and produces approximately 300 kW of power. A service life of over 2400 hours has previously been demonstrated, which is sufficient for a Mars transport mission.

        According to the scientist, the spacecraft will be launched into orbit using the traditional method—launch vehicles equipped with chemical engines. The plasma units, in turn, will be activated after reaching the reference orbit. They could also function as part of a space tug—a ship designed to transport cargo between the orbits of planets in the Solar System.
        "A plasma rocket engine is a type of electric engine. It's based on two electrodes. Charged particles are passed between them, and at the same time, a high voltage is applied to the electrodes. The resulting current creates a magnetic field, which pushes the particles out of the engine. This directs the plasma's movement and creates thrust," explained Yegor Biryulin, a junior researcher at the Rosatom Research Institute in Troitsk. To generate propulsion, the engine uses electrical energy and hydrogen as the working fluid, explained Yegor Biryulin. The onboard nuclear reactor will serve as the power source. The advantage of hydrogen is that its atoms are light and have a high flow rate. This allows significant speeds to be achieved without a large consumption of working fluid. Furthermore, hydrogen is the most abundant element in the universe. Therefore, its reserves can be easily replenished in the future. "Another positive feature of the new system is that the plasma in the proposed mechanism does not require significant heating." As a result, engine components and assemblies are not subject to thermal overload, and the electrical energy used for its operation is converted almost entirely into motion. This makes the system more efficient than similar systems, the specialist added.


        https://iz.ru/1834706/andrei-korsunov/plazmennoe-serdce-rossiiskii-dvigatel-dostavit-na-mars-za-odin-dva-mesaca
        1. +3
          31 May 2026 19: 14
          Quote from Orange Bigg
          Its capacity is about 300 kW.

          Uncle, you don't even understand what you're copypasta'ing.
          300 kWh = 3,600,000 J * 300 energy, i.e. ugh 1,08 * 10 ^ 9 J = 1,08 GJ.
          To launch 22 tons of PN RN into LEO, a proton spends 3*10^11 Joules, i.e. 3,000 GJ of energy.
          3 (Karl 3☝️) orders of magnitude more
          1. With 22 tons of payload, you have nothing to do near Mars.
          2. You need to not only reach 25-8=17 km/s, but then slow down from 25 km/s to 3,6 km/s at Mars.
          A Ek=m*V^2/2💁
          I haven't even taken into account the influence of Earth and Mars yet.
          🥱 Rosatom will give you even worse. Let them at least launch one power unit at the Akkuyu Nuclear Power Plant by the end of 2026 (the planned launch was for April 2025... It was. It wasn't even cold).
          And there are 4 blocks
      2. +1
        31 May 2026 09: 29
        Well, that's a strange question. We simply divide the entire flight cycle into two stages. You can accelerate smoothly during the first half of the flight, and decelerate smoothly during the second half.
        This still means the ship operates for 2-3 weeks in the on mode in each of the two stages. Over three continuous weeks, even with low acceleration, it's easy to gain or lose a decent speed.
        1. +1
          31 May 2026 10: 43
          We simply divide the entire flight cycle into two stages.
          Even an orbital calculations amateur like me understands this, but I just wanted to hear the actual numbers—how many FEs are there? Is two months to Mars a manned mission or not?
          1. 0
            31 May 2026 16: 27
            Quote: Popandos
            We simply divide the entire flight cycle into two stages.
            Even an orbital calculations amateur like me understands this, but I just wanted to hear the actual numbers—how many FEs are there? Is two months to Mars a manned mission or not?
            I read some material on this topic; there were no exact figures, but I would draw attention to the following nuance:
            Half of the time, the rocket accelerates toward Mars, and half of the time, it slows down—that is, the ship experiences gravity the entire time it flies, and it must not exceed the force of gravity on Earth, otherwise the astronauts will be uncomfortable during the flight.
          2. +2
            31 May 2026 22: 41
            Quote: Popandos
            I just wanted to hear the real numbers of how many housing units there are?

            1/2 of the way to Mars is approximately 75 million km.
            Comrade is thirsty
            Quote: Mustachioed Kok
            so the ship operates in the on mode for 2-3 weeks in each of the two stages

            S=V0*t+ a*t^2/2
            3 weeks is t= 2,721,600 sec
            V0 = 8000 m / s
            All that is required is 0,0143719 m/s^2

            To reach Mars you need a ship of at least 100 tons.
            It's not clear where the mustachioed cook will get the energy and mass for this.
            And this is only to roll from 8 km/s to ~25 km/s
            But then you have to brake from 25 km/s to 3,6 km/s.
            This will require almost 4 and 2 times more energy and mass.
            1. +2
              1 June 2026 13: 03
              This will require almost 4 and 2 times more energy and mass.

              Oh, such propaganda, ruined by mathematics (((
        2. 0
          31 May 2026 16: 33
          Quote: Mustache Cock
          You can accelerate smoothly during the first half of the flight, and brake smoothly during the second half.
          It still turns out that the ship operates in the on mode for 2-3 weeks in each of the two stages.

          At the same time, the effect of weightlessness for the crew can be avoided good
    2. +2
      31 May 2026 09: 28
      And how will you feed it?
      Even the article states that, in addition to nuclear engines, there's also the development of nuclear power plants that power ship systems. This is where the "compact nuclear reactor + ion/magnetoplasma engines" combination comes into play. Our TEM is precisely designed to do just that. It's just that we've heard little news about it in the last few years.
      1. 0
        31 May 2026 15: 57
        Quote: Mustachioed Kok
        And how will you feed it?
        And also to cool and remove the charge (the emission of hydrogen ions will give the ship a huge negative charge, which will actively attract charged particles from the solar wind, which will cause degradation of the equipment).
    3. 0
      31 May 2026 18: 32
      Quote from Orange Bigg
      The nuclear rocket engine was a revolutionary solution in its time, but now there are more powerful engines.

      Can I have a sample?
      If so, then power is dE/dt = “energy velocity” or for rd:
      N=F*V*cos (angle between the velocity vector and F)
      F=force, thrust
      V= velocity of the body.
      1. Rocket on the launch pad: colossal thrust, V=0
      2. Electric rocket engines: power of millions of Newtons, speed... Hmm, it will become acceptable in a year.
      In RD the indicator Iud (efficiency) and thrust (acceleration)
      Quote from Orange Bigg
      The fact is that traditional engines have actually reached the limit of their capabilities, since increasing thrust requires a huge amount of fuel.

      But how else?
      F(jet engine thrust) = m'*Ve+Snozzle (Pnozzle cross-external environment).
      Where m'-second mass flow rate of the working fluid in kg/s
      Ve is the outflow velocity of mercury in m/s
      P-pressure
    4. 0
      4 June 2026 15: 52
      A magnetoplasma accelerator cannot operate on its own, as it requires a colossal amount of electrical energy to create magnetic fields and generate plasma. Solar panels far from the Sun (for example, near Mars or Jupiter) are incapable of producing such power. Therefore, Rosatom is creating a symbiosis of these technologies—a space nuclear-electric power system (NEP): A nuclear reactor onboard the spacecraft acts solely as a "space power plant" (generating hundreds of kilowatts or megawatts of electricity). A magnetoplasma engine captures this electrical energy and converts it into a high-speed plasma flow, creating constant acceleration. Bottom line: The nuclear engine provides enormous instantaneous force (needed for rapid maneuvers or takeoff), while the magnetoplasma engine, coupled with the onboard reactor, is an ideal "slow-moving marathon runner" for long interplanetary flights, which, due to constant operation, ultimately develops tremendous speed and delivers cargo faster. Comparing them directly is incorrect, since they pursue different goals, and in Rosatom's actual projects, these technologies work together, complementing each other.
  2. Owl
    0
    31 May 2026 09: 34
    Nuclear rocket engines are needed for flights into deep space (to neighboring planets and beyond); using a ramjet nuclear engine for a cruise missile is a show-off, a show-off example; the main thing is to detect such a cruise missile, and then it can be shot down with an air defense/missile defense missile with a "special warhead."