How helpful can electric propulsion become as we plan missions into the local interstellar medium? We can think about this in terms of the Voyager probes, which remain our only operational craft beyond the heliosphere. Voyager 1 moved beyond the heliopause in 2012, which means 35 years between launch and heliosphere exit. But as Nadim Maraqten (Universität Stuttgart) noted in a presentation at the recent International Astronautical Congress, reaching truly unperturbed interstellar space involves getting to 200 AU. We’d like to move faster than Voyager, but how?
Working with Angelo Genovese (Initiative for Interstellar Studies), Maraqten offers up a useful analysis of electric propulsion, calling it one of the most promising existing propulsion technologies, along with various sail concepts. In fact, the two modes have been coupled in some recent studies, about which more as we proceed. The authors believe that the specific impulse of an EP spacecraft must exceed 5000 seconds to make interstellar precursor missions viable in a timeframe of 25-30 years, acknowledging that this ramps up the power needed to reach the desired delta-v.
Electric propulsion is a method of ionizing a propellant and subsequently accelerating it via electric or magnetic fields or a combination of the two. The promise of these technologies is great, for we can achieve higher exhaust velocities by far with electric methods than through any form of conventional chemical propulsion. We’ve seen that promise fulfilled in missions like DAWN, which in 2015 became the first spacecraft to orbit two destinations beyond Earth, having reached Ceres after previously exploring Vesta. We can use electric methods to reduce propellant mass or achieve, over time, higher velocities. [Addendum: Thanks to several readers who noticed that I had reversed the order of Vesta and Ceres in the DAWN mission above. I’ve fixed the mistake.]
Image: 6 kW Hall thruster in operation at the NASA Jet Propulsion Laboratory
Unlike chemical propulsion, electric concepts have a relatively recent history, having appeared in Robert Goddard’s famous notebooks as early as 1906. In fact, Goddard’s 1917 patent shows us the first example of an electrostatic ion accelerator useful for propulsion, even if he worked at a time when our understanding of ions was incomplete, so that he considered the problem as one of moving electrons instead. Konstantin Tsiolkovsky had also conceived the idea and wrote about it in 1911, this from the man who produced the Tsiolkovsky rocket equation in 1903 (although Robert Goddard would independently derive it in 1912, and so would Hermann Oberth about a decade later).
As Maraqten and Genovese point out, Hermann Oberth wound up devoting an entire chapter (and indeed, the final one) of his 1929 book Wege zur Raumschiffahrt (Ways to Spaceflight) to what he describes as an ‘electric spaceship.’ That caught the attention of Wernher von Braun, and via him Ernst Stuhlinger, who conceived of using these methods rather than chemical propulsion to make von Braun’s idea of an expedition to Mars a reality. It had been von Braun’s idea to use chemical propulsion with a nitric acid/hydrazine propellant, as depicted in a famous series on space exploration that ran in Collier’s from 1952-1954.
But Stuhlinger thought he could bring the mass of the spacecraft down by two-thirds while expelling ions and electrons to achieve far higher exhaust velocity. It was he who introduced the idea of nuclear-electric propulsion, by replacing a power system based on solar energy with a nuclear reactor, thus moving us from SEP (Solar Electric Propulsion) to NEP (Nuclear Electric Propulsion). Let me quote Maraqten and Genovese on this:
Stuhlinger immersed himself in electric propulsion theory, and in 1954 he presented a paper at the 5th International Astronautical Congress in Vienna entitled, “Possibilities of Electrical Space Ship Propulsion”, where he conceived the first Mars expedition using solar-electric propulsion [4]. The spacecraft design he proposed, which he nicknamed the “Sun Ship”, had a cluster of 2000 ion thrusters using caesium or rubidium as propellant. He calculated that the total mass of the “Sun Ship” would be just 280 tons instead of the 820 tons necessary for a chemical-propulsion spaceship for the same Mars mission. In 1955 he published: “Electrical Propulsion System for Space Ships with Nuclear Source” in the Journal of Astronautics, where he replaced the solar-electric power system with a nuclear reactor (Nuclear Electric Propulsion – NEP). In 1964 Stuhlinger published the first systematic analysis of electric propulsion systems: “Ion Propulsion for Space Flight” [3], while the physics of electric propulsion thrusters was first described comprehensively in a book by Robert Jahn in 1968 [5].
In 1957, the Walt Disney television program ‘Mars and Beyond’ (shown in the series ‘Tomorrowland’) featured the fleet of ten nuclear-electric powered spacecraft that Stuhlinger envisioned for the journey. As you can see in the image below, this is an unusual design, a vehicle that became known as an ‘umbrella ship.’ I’ve quoted him before on this, but let me run the passage again. It’s from Stuhlinger’s 1955 paper “Electrical Propulsion System for Space Ships with Nuclear Power Source”:
A propulsion system for space ships is described which produces thrust by expelling ions and electrons instead of combustion gases. Equations are derived from the optimum mass ratio, power, and driving voltage of a ship with given payload, travel time, and initial acceleration. A nuclear reactor provides the primary power for a turbo-electric generator; the electric power then accelerates the ions. Cesium is the best propellant available because of its high atomic mass and its low ionization energy. A space ship with 150 tons payload and an initial acceleration of 0.67 x 10-4 G, traveling to Mars and back in a total travel time of about 2 years, would have a takeoff mass of 730 tons.
Image: Ernst Stuhlinger’s Umbrella Ship, built around ion propulsion. Notice the size of the radiator, which disperses heat from the reactor at the end of the boom. The source for this concept was a Stuhlinger paper called “Electrical Propulsion System for Space Ships with Nuclear Power Source,” which ran in the Journal of the Astronautical Sciences 2, no. Pt. 1 in 1955, pp. 149-152. Credit: Winchell Chung.
While I’ve only talked about Stuhlinger’s work on electric propulsion here, his contribution to space sciences was extensive, ranging from a staging system crucial to Explorer 1 (this involved his pushing a button at the precise time required, hence his nickname as ‘the man with the golden finger’), to his work as director of the Marshall Space Flight Center Science Laboratory, which involved an active role in plans for lunar exploration.
For his contributions to electric propulsion, the Electric Rocket Propulsion Society renamed its award for outstanding achievement as the Stuhlinger Medal after his death. In terms of his visibility to the public, those interested in space advocacy will know about his letter to Sister Mary Jucunda, a nun based in Zambia, which laid out to a profound skeptic the rationale for pursuing missions to far destinations at a time of global crisis.
Image: In the above photo, taken at the Walt Disney Studios in California, Wernher von Braun (right) and Ernst Stuhlinger are shown discussing the technology behind nuclear-electric spaceships designed to undertake the mission to the planet Mars. As a part of the Disney ‘Tomorrowland’ series on the exploration of space, the nuclear-electric vehicles were shown in the program “Mars and Beyond,” which first aired in December 1957. Credit: NASA MSFC.
In the next post, I want to look at the deep space applications that Maraqten and Genovese considered in their IAC presentation.
For more details on Stuhlinger’s Mars ship, see Adam Crowl’s Stuhlinger Mars Ship Paper, and the followup I wrote in these pages back in 2015, Ernst Stuhlinger: Ion Propulsion to Mars. The Maraqten & Genovese paper is “Advanced Electric Propulsion Concepts for Fast Missions to the Outer Solar System and Beyond,” 73rd International Astronautical Congress (IAC), Paris, France, 18-22 September 2022 (available here). Ernst Stuhlinger’s paper on nuclear-electric propulsion is “Electrical Propulsion System for Space Ships with Nuclear Source,” appearing in the Journal of Astronautics Vol. 2, June 1955, p. 149, and available in manuscript form here. For more background on electric propulsion, see Choueiri, E., Y., “A Critical History of Electric Propulsion: The First 50 Years (1906-1956),” Journal of Propulsion and Power, vol. 20, pp. 193-203, 2004.
From the EP paper abstract:
Yes. Coupling an external power source that should be more effective than solar PV is a better way to get the power needed for a high Isp ion engine as it ventures into deep space. This is not unlike the laser thermal rocket post on CD to reduce flight times to Mars. I’m a little surprised that Chang hasn’t suggested something similar for his VASIMR plasma rocket to avoid the multi-MW nuclear reactor needed for fast flight times.
Given Tesla (Nikola) was demonstrating radiated power in the 19 century, and microwave-beamed power by the mid-20th century, this idea seems, like solar sails, a technology with a slow gestation.
Can we get a post on the pros and cons of laser vs microwave beaming?
More on this is coming up, as it strikes me as a productive way to handle the power problem.
Hi Paul
The welcome news that Room Temperature MASER’s are now a thing means we might have a microwave competitor to laser power beaming. Though no doubt Jim Benford would argue the option already exists merely using conventional HPMW sources.
Yes, I can hear Jim on that subject now. And I assume he’s right.
Why not direct heating by focusing the laser light into a small magnetically protected chamber. A huge pulse of laser light from earth is focused at a time interval when a pellet of material is loaded into it which causes a temperature rise of millions of degrees and then magnetically directed as exhaust.
Well, I do not know the details of this concept, it would be interesting to know which specific impulse can be obtained. Another important issue is the mass of the power source, this concept seems quite heavy.
In order to make interstellar precursor missions realistic, we need a power source with a specific mass much lower than 1kg/kWe and specific impulses much higher than 10,000 s.
Once you have a plasma, there are other options for accelerating the ions rather than thermal, and you get more control. Opting for thermal requires high precision to focus the beam onto the thrust chamber port. The focusing mirror must be extremely reflective and work even after micrometeoroid damage that will scar the surface. On the plus side, all the energy is used without conversion losses.
If one considers the precision requirements and difficulties of laser fusion at NIF, one can imagine the further difficulties of focusing a laser with a mirror, on a moving spacecraft, at different distances and orientations to the source of the beam. Compare that situation to just having tuned PV arrays (or microwave rectennas) to convert the beam to electricity to power an electric engine as well as recharging batteries/capacitors and powering other ship’s systems. Power conversion lowers energy efficiency and maybe increases mass compared to a mirror, but offers gains in flexibility and robustness.
I am more interested in the technical issues and cost comparisons between using phased microwave and phased laser arrays as power sources for spacecraft and facilities from LEO to the edge of the solar system and beyond. My very limited impression is that lasers are rapidly heading down the cost curve to make these the most cost-competitive systems, but apart from maintaining a tighter beam, IDK the state of the art and where it is headed to understand which approach technically works best and whether the tradeoffs and beam type selection differs under different situations.
If you look at the mirror design of the James webb telescope and replace the secondary mirror with the magnetic chamber and reduce the movable mirrors focal range that would concentrate any laser light to a very high value. That concentration of laser light even if not compressing the pellet would drive the temperature of the pellet to hundreds of thousands if not a few million degrees.
Consider just the engineering issues of the mirror and thrust chamber-exhaust orientation differences between the outward and inward bound flights. Then think of the same orientation problem for a ship on an elliptical orbit constantly changing orientation to tyhe beam source on Earth.
It is far easier to use a beam collector to convert the energy to electricity, ionize the propellant and accelerate it electrically or magnetically. You lose on the convervion efrficiency, but make gains with the far simpler engineering and special designs for the port on the thrust chamgerr that allows the focused beam to enter and heat up the propellant to the temperature, that like a Tokamak, the plasma must be kept away from the chamber walls. Propellant can be ionized at a far lower temperature, something you can do in a microwave oven. Given enough power, one could accelerate the ions to nearly light speed as they do in accelerators. An exhaust velocity of 1% c (3,000 km/s) is equivalent to an Isp of 3,000,000!
Oops. isp = 300,000. Still plenty high.
The problem with laser illumination is the fall off in power at increasing distance. So using the thermal component as much as possible initially to get maximum thrust would be better I would think. And the magnetic thrust chamber can be gimballed for increased control.
The thrust chamber has to be opaque with a small transparent port for the bean to enter and focus to the point to generate the high energy needed for the pellet atoms/ions to gain velocity. The only way this seems possible to me, is to position the mirror at 90 degrees above the thrust chamber so that any “horizontal” gimballing of the thrust chamber does not change the relative orientation of the port. [Not that the design for the beamed thermal approach for the Mars transport outlined in a CD post has that port at about 45 degrees in the rear wall of the thrust chamber which would require the mirror to reorientate, I believe.]
The beam focus and pellet stream must be perfectly aligned for this to approach. Any beam power must be used immediately rather than stored, or energy and mass throttled. Conversely, with a PV to electric system, energy can be applied as needed, and even stored so that intermittent bursts of thrust can be configured, with burst frequency or propellant mass controlled to manage thrust requirements, with e same power system used for controlling the magnetic/electric accelerators and spacecraft power system requirements.
Mirrors for solar thermal are fine as the thrust chamber is simply an absorber of energy to heat the propellant, but the Isp is far lower and therefore the propellant mass is much higher. [Having said that, solar thermal seems like a simple, cheap, way to transport ice around the system, consuming a fraction of the ice cargo as propellant.]
The chamber is opaque around the waist to protect the magnetic field generator or coils and has two openings. The magnetic field is powerfully pinched effectively closing plasma off to the mirror side and a weaker one toward the exhaust side allowing plasma out. There is also a possibility of using a two coil cylinder design on its side with the exhaust bent at 90 degrees out the back. The vector or gimbaling of the thrust can be done by adjusting the magnetic field strengths and/or moving the whole mechanical unit. The laser can still be shone on the mirrors even if there is no pellet loaded as it will just exit the hollow chamber. Countering the torque generated could be a issue though. The mirrors would still work using sunlight but would much less powerful but act over a longer time period and would need a proper chamber as the exhaust would not be heavily ionised. And yes water would make a good propellant around the solar system but the laser frequency would have to be considered. In effect it would just be a powerful booster enabling high earth exit velocities perhaps aiding a sundiver maneuver.
Indeed, which is why Breakthrough Starshot is using the immense power of the laser array to accelerate the tiny beamed sails quickly, rather than over a longer time and distance.
If your argument is to go for high thrust to consume the propellant , then fine, but you can do the same thing by multiplying the number of electric engines and live with the mass penalty.
I wonder if there is a simple way to make the thrust chamber work like a solar thermal system, where the laser power is just absorbed and actively transferred to the propellant, a little like the way rocket engines circulate the cryogenic propellants through the bell and thrust chamber to keep the walls from melting and preheat the fluids before combustion. Can this provide a high enough energy flux to heat the propellant to the plasma temperatures required?
Another way is to use a lens system, the probe is thrown into a large elliptical path and as it approaches the earth again the laser is set upon it. Could be a great way to get to the solar focal line quickly.
These spacecraft were presented on television in 1957 by Disney with their program titled Mars and Beyond:
https://www.youtube.com/watch?v=3wIXZsbjIxA
The companion comic book online, which includes the Umbrella ship flotilla mission:
http://atomic-surgery.blogspot.com/2011/11/mars-and-beyond-disney-four-color.html
Winchell Chung’s Atomic Rockets page on the Mars Umbrella Ship here:
http://www.projectrho.com/public_html/rocket/realdesigns3.php#id–Mars_Umbrella_Ship
Lots of details and diagrams!
Thanks Lawrence
Some good links there
I’d love to see someone who can break down and work through this paper: https://www.nature.com/articles/s41467-022-33497-1 It’s not nuclear, but according to the phys.org blurb ( https://phys.org/news/2022-10-heat-proof-chaotic-carbides-revolutionize-aerospace.html ) “The new materials are hard enough to stir molten steel and can withstand temperatures above 7,000 degrees Fahrenheit.”
The ears perk up at such things – at face value, that hull sounds plausible to build a solar LANDER out of, provided you pick a shady sunspot for your runway and have a way to repel from the local magnetic field. Apart from any role in heat shielding though, I have to wonder: what kind of energy storage are they talking about? Could a balloon of such material in interplanetary space store enough plasma inside itself to compete with Stuhlinger’s nuclear idea? (I have absolutely no idea, not even to an order of magnitude, but… my ears are perked)
Can anyone provide examples and where this would create new performance opportunities?
Are there no speculative ideas, or is this really unknown application territory?
China’s power
https://twitter.com/CNSpaceflight/status/1486798962963070979
Ion Mars Ships envisaged
https://forum.nasaspaceflight.com/index.php?topic=53526.0
Thank you for a nice presentation Paul.
And a nice nod to Goddard and Tsiolkovsky, both persons one might think had been born at least 50 or perhaps 100 years to early. Then again, they did live in the first age where ideas started to travel far and wide.
Stuhlinger’s ship have been depicted in many ways, none the less for the unusual design which triggered those who liked futuristic designs. But at the end of the day it was simply sound engineering. When there’s a strong power source, there’s also waste heat. And hence the disk, where the individual blades could be stacked and so be stowed in a practical way for transport and launch.
Hall thrusters are obsolete today for manned space travel considering we have VASIMR which is much faster considering if we have a powerful, space worthy nuclear reactor like one in a submarine, we might as well have VASIMR. I like the fact the Hall thrusters need less fuel, but one would still have to rotate the passenger cabin due to the long time in space and deteriorating effects of zero gravity on bone mass and muscle mass.
Disadvantages
According to Ad Astra as of 2015, the VX-200 engine requires 200 kW electrical power to produce 5 N of thrust, or 40 kW/N.[6] In contrast, the conventional NEXT ion thruster produces 0.327 N with only 7.7 kW, or 24 kW/N.[6] Electrically speaking, NEXT is almost twice as efficient, and successfully completed a 48,000 hours (5.5 years) test in December 2009.[8][9]
New problems also emerge with VASIMR, such as interaction with strong magnetic fields and thermal management. The inefficiency with which VASIMR operates generates substantial waste heat that needs to be channeled away without creating thermal overload and thermal stress. The superconducting electromagnets necessary to contain hot plasma generate tesla-range magnetic fields[10] that can cause problems with other onboard devices and produce unwanted torque by interaction with the magnetosphere. To counter this latter effect, two thruster units can be packaged with magnetic fields oriented in opposite directions, making a net zero-torque magnetic quadrupole.[11]
In order to conduct an imagined crewed trip to Mars in 39 days,[31] the VASIMR would require an electrical power level far beyond anything currently possible.
On top of that, any power generation technology will produce waste heat. The necessary 200 megawatt reactor “with a power-to-mass density of 1,000 watts per kilogram” (Díaz quote) would require extremely efficient radiators to avoid the need for “football-field sized radiators” (Zubrin quote).[32]
There is always an energy|thrust tradeoff. The more energy you deliver, the higher the Isp, the lower the propellant consumption rate, the higher the thrust for the same propellant flow rate, and the more propellant efficient the system. The multi-GW phased laser beams of the proposed Breakthrough Starshot could deliver the sort of power that a very high Isp engine would need. The question is how lightweight the PV array can be to make this approach worthwhile.
What I haven’t seen is the see of a very large, thin-film, parachute mirror or large flat Fresnel lens to concentrate light onto PV arrays to increase light intensity and power output. What sort of power-to-mass ratio could this offer, especially coupled with a powerful beam for really deep space missions? It is analogous to the aerial density of solar sails – the lower the density, the higher the performance.
IIRC, one of Zubrin’s criticisms of a fast VASIMR ship is that if something goes wrong, the ship might continue to head out into space killing the crew. OTOH, a Hohmann transfer orbit means that any problems and the ship returns to Earth’s orbit with the potential of a crew rescue. IMO, Hohmann orbits are particularly suitable for cargo, making them equivalent to oceangoing cargo ships. Until we have very cheap, lightweight power supplies, slow and efficient is the way to go if you have time. For the equivalent of air travel speeds in space, we need the equivalent of transitioning from steam to lightweight, high power, IC engines, whether props or jets. Fusion or anti-matter would be one approach.
One can carry along a conventional rocket engine which uses liquid or solid fuel if VASIMR breaks down which easily solves that problem. Also, one can use more than one VAZIMR engine. Three engines I recall is the design used by AdAstra. Nuclear fusion propulsion has a high specific impulse, but we don’t have that technology yet. Consequently, a Plasma rocket without needing fusion, I always thought was really a brilliant idea and cutting edge and it is already developed. I have to think at some time we will use them it’s just a matter of time. Zubrin’s idea was to use gas core nuclear thermal rocket which is much more fuel inefficient than VASIMR.
The current design for the VASIMR is to work under 200 kW, but if we indeed would go for 200 megawatt, extremely large radiators would be needed.
An interesting little back of envelope calculation for the the larger ion engines would be that 1000 such units powered by that reactor 200 MW reactor would give ½ a ton of propulsion 0,509 to be exact.
No one would build that way of course, but it’s one amazing amount of continuous power and show as a comparison that Hall thrusters certainly is a contending system. What’s needed is further development into units that can handle even higher power, and develop a power source for such use.
The problem with Zubrin is he is in competition with Ad Astra, so he does not have an unbiased view of VASIMR. His opinion is not factual, but only false The Saturn five rocket could launch 140 tons into orbit. The average weight of the reactor core of a nuclear submarine is 110 tons with a 165 Megawatts of power output. Wikipedia source. VASIMR needs only 10 to 20 megawatts to get to Mars in 39 days. I can’t imagine why we can’t scale down a nuclear reactor power planet which is lighter and smaller for spacecraft like Hall thrusters, VASIMR, ion power etc.
The problem is we have so much money invested in old technology, because of conservativism and wanting everything now. I will admit I am biased against the technology of the past like all types of nuclear thermal rocket’s including Zubrin’s gas core design which he promotes.
Zubrin is definitely in the camp of “my way or the highway”. Is he in favor of a gas core nuclear rocket or the salt water nuclear rocket (a very dangerous device, IIRC).
Hi Alex
I think just ISRU for return propellant and a reactor to make it at destination. He hasn’t really spoken in favour of any nuclear propulsion – reactors are best deployed to make propellant at the destination. I’ve done the maths and he’s not wrong. NTR’s really do suck.
Adam
NUCLEAR ROCKET USING INDIGENOUS MARTIAN FUEL
NIMF
My Nuclear Salt Water Rocket!
Looks like advocacy to me.
Hi Alex
Zubrin has also spoken out against NTR’s for a variety of reasons, but in the context of the first Mars missions. In that respect he sees them as an impediment to the first Mars missions, a technological distraction to what can be done differently using power reactors and ISRU.
Hi Geoffrey
The 39 day to Mars VASIMR needed 200 MW electrical power and is explicitly stated to need a gas core nuclear reactor to get the required performance by their original journal paper. Please don’t create a new truism.
Hi, Charlie.
Your observations about NEXT and VASIMR caught my attention.
First, it took me a while to figure out I was reading your comparison upside down, so to speak. But then, I wondered what else might be considered. Two things that come to mind are propellant types and specific impulse. The question of mission power sources and radiators were already touched on.
From what came readily to hand it appears that the tests with NEXT were conducted with xenon. Other possibilities could have been argon, neon… I don’t know what the track record is for other inerts. In the past mercury and cesium have been considered. But for VASIMR, the selection, last I heard, was argon. That’s plentiful in comparison to xenon. Also, VASIMR is ( or should be ) tunable.
You can trade between thrust level and specific impulse.
Saying that, NEXT might be very well suited for a number of targets or missions, if you don’t need vast quantities of propellant because the vehicle is as small as a cube sat.
But matching up power plants and VASIMR can be an interesting exercise if you are trading power plants and mission times, say for flight to Mars. About a dozen years ago, had an opportunity to participate in such an exercise and it was fun. Principally for Mars.
Beside the varying launch window opportunities and times of flight associated with them, there is also the issue of human exposure in long duration flights. Consequently, the potential power sources and radiators were definitely at issue.
To summarize on that last point, a submarine nuclear reactor in space would seem to solve our power available problems to first order – but it’s just too bad that space does not have the same thermal properties as our terrestrial oceans when it comes to getting rid of excess or radiated heat. VASIMR thermal is what it is.
But perhaps when it comes to spaceflight power sources, crash program work should be focused more on “cool power” sources.
A couple of fusion concepts do fall in that category; e.g. Robert Bussard’s (et alia) polywell reactor looked promising at the time.