Fusion runways remind me of the propulsion methods using pellets that have been suggested over the years in the literature. Before the runway concept emerged, the idea of firing pellets at a departing spacecraft was developed by Clifford Singer. Aware of the limitations of chemical propulsion, Singer first studied charged particle beams but quickly realized that the spread of the beam as it travels bedevils the concept. A stream of macro-pellets, each several grams in size, would offer a better collimated ‘beam’ that would vaporize to create a hot plasma thrust when it reaches the spacecraft.
Even a macro-pellet stream does ‘bloom’ over time – i.e., it loses its tight coherency because of collisions with interstellar dust grains – but Singer was able to show through papers in The Journal of the British Interplanetary Society that particles over one gram in weight would be sufficiently massive to minimize this. In any case, collimation could also be ensured by electromagnetic fields sustained by facilities along the route that would measure the particles’ trajectory and adjust it.
Image: Clifford Singer, whose work on pellet propulsion in the late 1970s has led to interesting hybrid concepts involving on-board intelligence and autonomy. Credit: University of Illinois.
Well, it was a big concept. Not only did Singer figure out that it would take a series of these ‘facilities’ spaced 340 AU apart to keep the beam tight (he saw them as being deployed by the spacecraft itself as it accelerated), but it would also take an accelerator 105 kilometers long somewhere in the outer Solar System. That sounds crazy, but pushing concepts forward often means working out what the physics will allow and thus putting the problem into sharper definition. I’ve mentioned before in these pages that we have such a particle accelerator in the form of Jupiter’s magnetic field, which is fully 20,000 times stronger than Earth’s.
We don’t have to build Jupiter, and Mason Peck (Cornell University) has explored how we could use its magnetic field to accelerate thousands of ‘sprites’ – chip-sized spacecraft – to thousands of kilometers per second. Greg Matloff has always said how easy it is to overlook interstellar concepts that are ‘obvious’ once suggested, but it takes that first person to suggest them. Going from Singer’s pellets to Peck’s sprites is a natural progression. Sometimes nature steps in where engineering flinches.
The Singer concept is germane here because the question of fusion runways depends in part upon whether we can lead our departing starship along so precise a trajectory that it will intercept the fuel pellets placed along its route. Gerald Nordley would expand upon Singer’s ideas to produce a particle stream enlivened with artificial intelligence, allowing course correction and ‘awareness’ at the pellet level. Now we have a pellet that is in a sense both propellant and payload, highlighting the options that miniaturization and the growth of AI have provided the interstellar theorist.
Image: Pushing pellets to a starship, where the resulting plasma is mirrored as thrust. Nordley talks about nanotech-enabled pellets in the shape of snowflakes capable of carrying their own sensors and thrusters, tiny craft that can home in on the starship’s beacon. Problems with beam collimation thus vanish and there is no need for spacecraft maneuvering to stay under power. Credit: Gerald Nordley.
Jordin Kare’s contributions in this realm were striking. A physicist and aerospace engineer, Kare spent years at Lawrence Livermore National Laboratory working on early laser propulsion concepts and, in the 1980s, laser-launch to orbit, which caught the attention of scientists working in the Strategic Defense Initiative. He would go on to become a spacecraft design consultant whose work for the NASA Institute for Advanced Concepts (as it was then called) analyzed laser sail concepts and the best methods for launching such sails using various laser array designs.
Kare saw ‘smart pellets’ in a different light than previous researchers, thinking that the way to accelerate a sail was to miniaturize it and bring it up to a percentage of c close to the beamer. This notion reminds me of the Breakthrough Starshot sail concept, where the meter-class sails are blasted up to 20 percent of lightspeed within minutes by a vast laser array. But Kare would have nothing to do with meter-class sails. His notion was to make the sails tiny, craft them out of artificial diamond (he drew this idea from Robert Forward) and use them not as payload but as propulsion. His ‘SailBeam’, then, is a stream of sails that, like Singer’s pellets, would be vaporized for propulsion as they arrived at a departing interstellar probe.
Kare was a character, to put it mildly. Brilliant at what he studied, he was also a poet well known for his ‘filksongs,’ the science fiction fandom name for SF-inspired poetry, which he would perform at conventions. His sense of humor was as infectious as his optimism. Thus his DIHYAN, a space launch concept involving reusable rockets (if he could only see SpaceX’s boosters returning after launch!). DIHYAN, in typical Kare fashion, stood for “Do I Have Your Attention Now?” Kare’s role in the consideration of macro-scale matter sent for propulsion is secure in the interstellar literature.
And by the way, when I write about Kare, I’m always the recipient of email from well-meaning people who tell me that I’ve misspelled his name. But no, ‘Jordin’ is correct.
We need to talk about SailBeam at greater length one day soon. Kare saw it as “the most engineering-practical way to get up to a tenth of the speed of light.” It makes sense that a mind so charged with ideas should also come up with a fusion runway that drew on his SailBeam thinking. Following on to the work of Al Jackson, Daniel Whitmire and Greg Matloff, Kare saw that if you could place pellets of deuterium and tritium carefully enough, a vehicle initially moving at several hundreds of kilometers per second would begin encountering them with enough velocity to fire up its engines. He presented the idea at a Jet Propulsion Laboratory workshop in the late 1980s.
We’re talking about an unusual craft, and it’s one that will resonate not only with Johndale Solem’s Medusa, which we’ll examine in the next post, but also with the design shown in the Netflix version of Liu Cixin’s The Three Body Problem. This was not the sleek design familiar from cinema starships but a vehicle shaped more or less like a doughnut, although a cylindrical design was also possible. Each craft would have its own fusion pellet supply, dropping a pellet into the central ‘hole’ as one of the fusion runway pellets was about to be encountered. Kare worked out a runway that would produce fusion explosions at the rate of thirty per second.
Like Gerald Nordley, Kare worried about accuracy, because each of the runway pellets has to make a precise encounter with the pellet offered up by the starship. When I interviewed him many years back, he told me that he envisioned laser pulses guiding ‘smart’ pellets. Figure that you can extract 500 kilometers per second from a close solar pass to get the spacecraft moving outward at sufficient velocity (a very optimistic assumption, relying on materials technologies that are beyond our grasp at the moment, among other things), and you have the fusion runway ahead of you.
Initial velocity is problematic. Kare believed the vehicle would need to be moving at several hundreds of kilometers per second to attain sufficient velocity to begin firing up its main engines as it encountered the runway of fusion pellets. Geoff Landis would tell me he thought the figure was far too low to achieve deuterium/tritium ignition. But if it can be attained, Kare’s calculations produced velocities of 30,000 kilometers per second, fully one-tenth the speed of light. The fusion runway would extend about half a light day in length, and the track would run from near Earth to beyond Pluto’s orbit.
And there you have the Bussard Buzz Bomb, as Kare styled it. The reference is of course to the German V-1, which made a buzzing, staccato sound as it moved through English skies that those who heard it would come to dread, because when the sound stopped, you never knew where it would fall. You can’t hear anything in space, but if you could, Kare told me, his starship design would sound much like the V-1, hence the name.
In my next post, I’ll be talking about Johndale Solem’s Medusa design, which uses nuclear pulse propulsion in combination with a sail in startling ways. Medusa didn’t rely on a fusion runway, but the coupling of this technology with a runway is what started our discussion. The Netflix ‘3 Body Problem’ raised more than a few eyebrows. I’m not the only one surprised to see the wedding of nuclear pulse propulsion, sails and runways in a single design.
Clifford Singer’s key paper is “Interstellar Propulsion Using a Pellet Stream for Momentum Transfer,” JBIS 33 (1980), pp. 107-115. He followed this up with “Questions Concerning Pellet-Stream Propulsion,” JBIS 34 (1981), pp. 117-119. Gerald Nordley’s “Interstellar Probes Propelled by Self-steering Momentum Transfer Particles” (IAA-01-IAA.4.1.05, 52nd International Astronautical Congress, Toulouse, France, 1-5 Oct 2001) offers his take on self-guided pellets. Jordin Kare’s report on SailBeam concepts is “High-Acceleration Micro-Scale Laser Sails for Interstellar Propulsion,” Final Report, NIAC Research Grant #07600-070, revised February 15, 2002 and available here. You might also enjoy my SailBeam: A Conversation with Jordin Kare.
I like it, though it strikes me as one of those approaches you take when you already have a lot of space infrastructure, as compared to the more classic fission/fusion driven generation ship approach.
Do you have any details/citations about what can be achieved with this magnetic field in terms of accelerating particles or macro objects? Do the forces rely on Jupiter’s rotation or more likely, using an electric current to move through the field?
I wrote this a few years back:
“Peck adds that getting the Sprites up to speed might itself take decades, and the journey to the nearest star would still be a matter of several centuries. But 300 years to Alpha Centauri beats any solar-sail-plus-Sundiver-maneuver mission I’ve ever seen, and unlike the admittedly faster beamed lightsail missions (some of Forward’s missions get down to decades), the Sprites take advantage of a form of propulsion that doesn’t require vast infrastructure in space.”
Of course the infrastructure still have to be pretty robust! The last time I discussed this with Mason Peck, it was out at one of the Breakthrough Starshot meetings. Let me check with him on his most recent work on this idea and I’ll pass the citations along.
The microsails are better as they can be accelerated to very high velocities and can be guided by the main beam even after the acceleration phase, just needs a little light to move onto target again if they are made smart The energy released can also be much higher than fusion by the simple kinetic energy of impact at fractions of c velocity and as a more pleasant plasma with no neutrons. And there is no need for a huge onboard laser to vapourise the discs, just slow each alternate one by a few km/s and you get a plasma on impact to be reflected via magnetic fields.
This was Jordin Kare’s concept,
https://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=&ved=2ahUKEwiYk-6WvL2FAxUYU0EAHXtRAscQFnoECBAQAQ&url=https%3A%2F%2Fwww.niac.usra.edu%2Ffiles%2Fstudies%2Ffinal_report%2F597Kare.pdf&usg=AOvVaw0Y-GXfNsMktoNh3AUI9PcR&opi=89978449
The notion of a mass-driver beam seems inelegant, sort of like propelling a rocket with a fire hose, but I suppose the inelegance is the point – this is an accessible method that could provide the sort of short-term solution called for in scenarios like 3 Body Problem. Fission or fusion rail drivers and deuterium pellets strewn like bread crumbs on runway to Pluto are just refinements of the cruder buzz bomb notion.
We are indeed fortunate to have feasible methods and time to work toward refinements.
I particularly like the idea of focusing (and perhaps to some degree aiming) solar wind or coronal mass ejections. That would be quite a fire hose.
Fission is orders of magnitudes easier than fusion needing only grams in well designed systems. I am surprised Winterbergs thoughts are not more widely known in space circles.
https://www.degruyter.com/document/doi/10.1515/zna-2004-0603/html?lang=en
Was unaware of Winterberg’s gram-scale fusion-fission-fusion micro nuke and application for a more efficient Orion-style buzz bomb. Appears relevant for seeding a runway as well, or for electrical power generation.
With modern lasers the explosive shell can be triggered very uniformly allowing the use of much smaller amounts of explosive. The issue i see with the runway concept is the components have to orbit the sun and will quickly go out of alignment unless perhaps they behave as sun hovered statlites.
@Michael
Recall the post about using “statites” to provide a linear escape velocity trajectory. This will provide a decent trackable trajectory for the fuel. How much the other planets will perturb the trajectory, idk.
The frequency of the fuel objects emitted along that trajectory will need to be increased to account for the velocity change of the craft, and this will determine the distributions of the fuel objects in their initial, inner system orbit before deploying their statite solar sails.
Are tiny pellets of frozen deuterium and tritium all that expensive to produce and then propel along a vector?
I’m thinking maybe not, at least as to production – as we apparently generally produce deuterium by separating the deuterium atoms from heavy water that in turn was processed from ordinary “light” water. And while tritium perhaps may be more rare, we nonetheless have terrestrial processes to produce tritium as well.
Those no doubt are not wholly inexpensive processes. But, in comparison to something like helium-3, it’s not like we have to go to the Moon and/or Jupiter to obtain deuterium and tritium in sufficient quantity. We currently produce them here on Earth, through established processes.
If deuterium and tritium are – well, comparatively – cheap enough, then perhaps the collimation issue can be addressed at least in part by overkill rather than precision and next-gen (or next-next . . . gen) engineering.
As in firing a somewhat diffuse swarm of deuterium and tritium pellets along the desired pathway, similar to an aerosol in distribution (albeit “suspended” in a far, far more diffuse gas cloud than a true aerosol on Earth).
The interstellar craft then scoops up the requisite percentage of pellets from within the perhaps swirling/rotating swarm that – individually – happen to intersect the craft’s of course very precise trajectory.
All based on some nice, tight, and precise mathematical projections as to both amount and time interval – similar to how we might calculate how many suspended particulates would be intersected and at what general time intervals by a projectile passing through an aerosol – allowing a calculation as to how much of the swarm will be intersected by an interstellar craft passing through the swarm on a particular vector and at what approximate intervals.
So a random process as to a particular pellet, but perhaps a sufficiently predictable one – as to the overall amount of pellets encountered and the time intervals of the encounters – with regard to an object passing through the swarm as a whole.
Somewhat similar to how a baleen whale doesn’t scoop up the entirety of a krill swarm but nonetheless gets enough krill mass as she passes through a swirling swarm to sate her hunger and fuel her endeavours.
The whale of course never targets any specific krill within the swarm. She just takes a vector through the swarm itself and, being the savvy cetacean that she is, lets math do the rest.
Such a swarm (albeit a far more diffuse one) of micro-pellets more or less along a particular vector – perhaps – then could help provide a fuel source for succeeding flights going sufficiently in the same direction (such as multiple missions to Proxima Centauri, possibly with settlers for a new world), perhaps “restocked” a bit each time.
A highway, as it were, of fuel stock rather than a road.
But even if only one mission was possible per swarm given how things move around in the universe . . .
On balance, the creation (and possibly restocking) of such an overabundant deuterium/tritium pellet aerosol-like cloud or swarm might be less daunting from engineering and expense perspectives than attempts to tightly collimate a stream of precisely just enough pellets to propel an interstellar mission (and/or to self-steer those pellets via next-gen – or next-next . . . gen – AI).
We might not be able to employ such an overkill strategy with a fuel having an abundance on Earth something like helium-3, given how much energy would be needed to obtain a rare (on Earth) element like helium-3 in sufficient abundance.
But given the – relative – abundance and/or producibility on Earth of deuterium and tritium, at least in comparison, such a low-tech strategy perhaps might reduce the overall engineering challenge a bit.
Well, over and above all the other challenges, such as having a viable fusion drive technology in the first place.
Tritium is very expensive and only has a half life of about 12 years so will decay of time reducing the effectiveness of the design.
To reduce the requirements of loosing focus, I think it’s better to use high acceleration for the starting phase. For unmanned probes, even 20G could be plausible. Also using a traveling propulsion path as long as possible. For example, from transneptunian, to close to Sun, to the interstellar path, crossing the solar system double time instead a just escape path.
Instead of a only pellet launcher, multiple stations across the solar system could launch with higher focus to the ship with higher precision.
The problem is that this trick only works for an acceleration phase, and because launcher orbits, that platforms will be probably one use only or it will take a lot of time and energy to reposition them in a new useful place for new ships.
One issue not addressed with D/T pellets is that while the deuterium is stable, the tritium has a 1/2 life of a little more than 12 years. This means that the runway must be limited in length if using a common pellet source as the potential energy of the pellet declines with time. The pellets sent on their way first will be decaying. Whether this is important for the accelerating phase, or whether this can be easily compensated in either size of the pellet or D/T ratio is a potential engineering issue. We generally assume fuel is homogenous in rockets. For fusion engines with onboard D/T fuel, perhaps manufacturing the tritium as needed from stable lithium-6 is a better solution. OTOH, tritium decays to helium-3, so perhaps that is the better route to fusion, rather than mining it from the moon.
Tritium decay seems like an ignored issue for fusion pellet runways which becomes important for any runway where the time to reach the runway end is long. Is this an argument for the fusion bomb runway rather than a pellet stream?
Much of this work was done prior to the advent of ‘metamaterials’. One can envision pellets with metamaterial reflecting skirts about their ‘waist’ such that (for example) a specific polarization is retroreflected and other polarizations are reflected off axis in specific directions. A reasonably powerful LIDAR on the starship can then both find the upcoming pellets AND correct their trajectories. (Admittedly, this is highly speculative, but so were metamaterials until they weren’t…)
Back in the 1990s, I had independently proposed Jordin Kare’s Bussard Buzz Bomb concept on rec.arts.sf.science, not aware of him proposing the same thing elsewhere on USENET. But I couldn’t figure out whether or not it could sustain fusion.
Later, I was investigating a similar concept using kinetic impact to initiate fusion rather than magnetic compression. In this concept, the starship places some propellant in the way of the fusion pellet to implode the pellet rather than sacrificing some of the starship’s speed to implode it via magnetic compression.
I still didn’t know how to estimate fusion yield, so I simply made it a variable between 0% and 100% fusion yield.
What shocked me was that the drive still worked with 0% yield. A pure kinetic impact powered pellet ramjet produced thrust and acceleration. That led me to study relativistic kinetic impact powered ramjet/rockets for some years.
Unlike Kare’s “bat out of hell” sailbeam of one sail accelerated at a time within 1km, I preferred the use of a long range XFEL to accelerate a swarm of sails over a long distance. I still feel this is a less challenging approach. But it still requires an incredible amount of total energy.
Isaac K, Jordins concept allows for great variation in velocities of the micro discs and they should be able to handle enormous accelerations. As for impact fusion firstlights idea of fusion could work either in runway or impact mode. Their concept uses a specially designed amplifier to create a spherical implosion to cause fusion.
Yes, I’m aware of Kare’s extremely high acceleration concepts for SailBeam. I have no criticism of his analysis.
I just felt like studying an alternative approach, using things I was more comfortable with and had already been studying for other reasons.
I had long been fascinated by the “grey sail” concept, which handles modest accelerations (more like 3 gees) by glowing white hot rather than attempting high reflectivity. And I had been studying long range XFEL concepts with Luke Campbell for some time, initially as concepts for interplanetary laser weaponry.
I suppose if the velcoity is high enough and the oncoming gases funneled to a pinch point via very powerful magnetic fields the lawson criteria will be met, but I am thinking it’s going to have to be very fast indeed !
There may be a way around the tritium problem at higher velocities, tritium and deutrium will fuse from around 500 to 1000 kms. Helium 3 maybe around 10 times greater but achievable over time for no neutrons. Now if Firstlight get the amplifier to work it reduces the velocities down by a factor of ten making all configurations more workable.