It was back in 1950 that Arthur C. Clarke looked at electromagnetic methods for getting a payload into space. The concept wasn’t new but Clarke’s paper in JBIS set out to examine what he saw as a practical use of it, an electromagnetic catapult on the lunar surface that could accelerate payloads back to Earth. The system was built around a three-kilometer long electromagnetic launcher that could accelerate payloads at 100 g’s to 2.3 kilometers per second (lunar escape velocity) in a matter of seconds. Gerard O’Neill thought such methods could deliver lunar raw materials to low Earth orbit for delivery to a space manufacturing site.
Clarke’s ideas played naturally into O’Neill’s, for building large space habitats requires vast amounts of raw materials that we’d just as soon not have to lift out of Earth’s gravity well. But Clarke’s thinking wasn’t restricted to near-Earth uses of the technology. He saw no necessary limit to the lengths and accelerations that could be used. Provide enough power to a space-based electromagnetic launcher and it could be configured to send robotic payloads to nearby stars. As with solar sails, we’re accelerating no propellant — only the payload — and that’s magic to any rocket scientist.
By the time E. H. Lemke looked at the idea in 1982, the size of such launchers had become astronomical. Lemke’s accelerator reached 108 kilometers in length, and used a solar collecting array hundreds of kilometers to the side to store the energy needed to boost payloads up to a third of lightspeed. Probes would be flung at 5000 g’s to nearby stars. It’s hard to see how a civilization capable of building on a scale like this (Lemke’s accelerator would be ? of an astronomical unit long!) wouldn’t also have developed smaller and more efficient ways of sending probes on the same journey, but the same could be said for many giant sail concepts.
Clifford Singer’s work at Princeton on ‘pellet stream propulsion’ makes use of both Clarke and Lemke to arrive at a somewhat more adaptable idea. Writing in 1980, Singer proposed using a stream of electromagnetically charged pellets to transfer momentum to a departing interstellar craft. You can think in terms of a lightsail being pushed by a laser or microwave beam, but in this case the beam is composed of macroscopic pellets, each with a mass of several grams, being dispatched at 0.2 c to the starship, where they would impact and impart momentum.
At 105 kilometers in length, Singer’s electromagnetic launcher would be huge, though not on the scale of Lemke’s. In the original work, Singer emphasized that a system like this solved the collimation problem — the spread of the beam over distance — that would be faced by laser-beamed missions. His idea was to use several dozen stations, perhaps deployed by the starship itself, to measure pellet positions and send the needed commands to adjust their course at the launcher. He examined the question of dispersion due to the impact of interstellar dust grains along the route and concluded that pellets of 1 gram and up could be delivered to the starship with accuracy.
Later, Gerald Nordley would take the logical leap of ‘smart’ pellets, with enough technology aboard to adjust their course as required. Nordley’s ‘snowflake’ pellets would be built around nanotechnology and could find their own way to a starship that could itself make small changes in course if required to stay within the stream. And Jordin Kare subsequently produced ‘SailBeam,’ in which Singer’s pellets give way to tiny ‘micro-sails’ that can be accelerated to the departing starship and vaporized upon approach, becoming a plasma that drives the spacecraft.
Image: Jordin Kare’s ‘SailBeam’ concept. Credit: Jordin Kare/Dana G. Andrews.
SailBeam, which was a hybrid system crossing pellet propulsion with lightsails, also offered a major advantage. Kare’s studies demonstrated that you could accelerate a large number of small sails with a far less demanding optical system than required by a huge lightsail. The sails could also be accelerated much closer to their power source, which simplifies the problems of maintenance and sail deployment faced by a large sail under the beam for lengthy periods. Made of diamond film, Kare’s sails would accelerate to close to lightspeed within seconds.
So many concepts, so little time! Digging into interstellar exotica is a refreshing exercise that reminds me how many concepts have been kicked around in the literature. Most are grossly impractical given our level of technology, but the refining of some of them seen in Nordley and Kare’s work shows that we can at least nudge some of these notions in a more practical direction. Practical, that is, for a future with greater energy resources and wider options within the Solar System than we enjoy.
The original Arthur C. Clarke paper is “Electromagnetic Launching as a Major Contribution to Space-Flight,” JBIS 9 (6) (1950), pp. 261-267. It’s reprinted in Clarke’s Ascent to Orbit: A Scientific Autobiography (New York: John Wiley & Sons, 1984). The Lemke paper is “Magnetic Acceleration of Interstellar Probes,” JBIS 35 (1982), pp. 498-503. Cliff Singer’s first pellet paper is “Interstellar Propulsion Using a Pellet Stream for Momentum Transfer,” JBIS 33 (1980), pp. 107–115. And you can find Jordin Kare’s NIAC report “High-Acceleration Micro-Scale Laser Sails for Interstellar Propulsion,” (Final Report, NIAC Research Grant #07600-070, revised February 15, 2002) on the NIAC site.
centripetal accelerators for relativistic velocities are out of the question since the centrifugal force goes with the square of gamma:
F= r * gamma^2 * omega ^2
where omega is the angular velocity. Even for the orbit radius of Neptune (10^12 meters) where experience over 1000-g for reaching 0.3c
linear accelerators don’t do much better as explained in the papers linked in this article. The only possible way for future interstellar accelerators would be to have huge and heavy disks rotating nearly at the speed of light, generating enough gravitational frame dragging so that higher g-forces can be achieved in smaller distances, because all those g’s won’t be felt by the spaceship (only the tidal forces)
But, but! the disk (presumably made of matter) will tear apart before the tangential speed reaches the speed of sound in the material. So the inner centrifugal force in the material needs to be counterbalanced with other forces. A possibility is a strong radial magnetic field gradient that levitates diamagnetically the disk material against its own centrifugal weight. But for reaching relativistic velocities, we are easily speaking of fields in the hundreds of thousands of Tesla, several orders of magnitude above what superconductors can achieve
There is another way that just occurred to me after writing the comment:
I’ve recently read an article that gives a very interesting insight into how the electromagnetic field can be split into a null component (that propagates with velocity c) and a time-like component (that propagates at velocities below c)
http://arxiv.org/abs/1102.0238
This is very interesting and relevant for the above, because if we can increase and manipulate the non-null components to behave as a disk made of photons, we would not be bound by the same mechanical constraints as with a matter disk
Here is a very advanced idea by physicist John Cramer. Submitted for your approval:
http://www.analogsf.com/2012_05/altview.shtml
There might be a nice economy to be found in accelerating each component of an interstellar probe (or vessel) separately and then having them all assemble themselves into a whole, en route. After they’ve reached cruising speed to where ever. Smaller payloads. We could launch them all in sequence or simultaneously. They’d each have to have some maneuvering fuel. This would be a tricky lot of programming that we’d have to practice in system a lot before we try it . But there might be a nice economy.
I’ve heard about a lot of different possible ways to travel to other star systems. Some are impractical. Some are laughable. But with so many options, it’s hard to believe none of them would work. There’s a way to make this happen, and Alpha Centauri is pretty damn close by. We can do this.
Since we’re talking extreme approaches to interstellar travel, I thought it might be time to mention the book “Time Travel and Warp Drives” (2012) (I did a search but could not find if anyone had written about it, so here goes.) This is a first-rate book concerning the whole business of “exotic matter” — and the wonderful things we could do with it (FTL, time travel, that stuff that Cramer talks about, see link @Thomas Mazanec above) if only we could make and keep the stuff.
The thing about exotic matter is — you recall the saying about how nature abhors a vacuum? — well nature really abhors this exotic stuff, hates it with a passion. Unfortunately, a whole lot of contemporary physics, quantum electrodynamics mostly, follows from that realization. Drawing upon research that has been on going for a third of a century at least, the authors Allen Everett and Thomas Roman show how impossible given our present state of knowledge it is to make the stuff and keep it stable so we can do something with it. The book is endlessly fascinating and frustrating, though I should add the door is far from closed. It’s just that we’ve got so much more to learn. The authors state we really won’t know the answers until we have a solid theory of quantum gravity, one we can take to the bank.
Anyway, get the book. Unreservedly recommended.
“centripetal accelerators for relativistic velocities are out of the question since the centrifugal force goes with the square of gamma:”
I ran the numbers on it years ago, it seemed feasible to get very sturdy probes up to 2-3% light speed if you used a accelerator around the orbit of Ceres. That’s maybe not the speed you’d really want, but it would be one heck of a “first stage”. The advantage, of course, of a circular accelerator, is that it can target any destination in it’s plane. Reorienting a linear accelerator long enough to get probes up to relativistic speeds could be the work of decades at best.
Hmmm…the microsail idea might even permit near-c flyby starprobe missions, using the microsails themselves *as* the probes. An initial batch of the microsails (without instrumentation and imaging electronics “printed” onto them) could serve as a buffer against impacts for the instrumented microsail probes that would follow behind them, like the collision defense dust cloud that the Daedalus starprobe was to deploy ahead of it. Also:
The microsail probes could be starlight-powered during the stellar encounter, take images via the “distributed swarm” method, and relay their signals to Earth via “bucket brigade,” signaling down a “stream” of the probes extending back to Earth (or relatively close to Earth, in interstellar distance terms). Are such microprobes beyond today’s technological state of the art? You betcha! But they may not always be. If they could be mass-produced using automated equipment, their very high flyby velocities would not pose an insurmountable problem for data collection and imaging, as that would be done by a long stream of the probes. Best of all, their “engine” would remain at home in the solar system, where it could be maintained and upgraded.
I don’t see how the pellet system avoids the collimation problem. Tiny differences in exit angle from the “muzzle” will have the same effect as beam divergence. This approach still has the same problem as beamed sails; keeping the beam operational over long periods or social and political changes.
As will beamed sails, I do think the idea has merit for deep space exploration within the solar system. However I worry that the high cost of the launcher and its single target probe may not be an economically viable solution. I think the idea of exploding a nuke behind a sail makes more sense. In this context, exploding electromagnetically accelerated small nukes behind the sail, so that most of the energy is in the bomb, not its momentum.
The problem with the nuclear bomb option on a sail is that the EMP is very harsh across the whole E’ spectrum, conducting materials will produce currents that would fry electronic components
The pellet system doesn’t solve the *initial* collimation problem, but it does solve the problem of subsequent beam spreading due to interaction with the interstellar medium, and, of course, due to wavelength limited focus precision. (The wavelength of a 1gm pellet is pretty darn small.)
The real payoff is when you have “smart” pellets capable of making minor course corrections to track a beacon on the target. They could potentially hit a target at a distance of lightyears, with even modest aiming precision at the start, an in the face of slight course deviations by the target. Perhaps even picking up speed along the way, if driven by a laser from the home system.
If you built an upside down maglev track around the circumference of the moon, to what speed could you accelerate a payload before the net centripetal force expirienced exceeds 5g? The answer is only about 9.4 km/s.
Alex: Most of these pellet stream concepts, starting with Singer and including Kare’s SailBeam, proposed guidance stations located along the acceleration path within the solar system to adjust each pellet/microsail trajectory with a small laser nudge. Alternatively, pellets could be electrostatically charged and collimated using magnetic fields generated by guidance stations.
in the novel “Threshold” printed by Baen books, the pellet propulsion makes an appearance under the name “Mass beam drive’. a german spacecraft makes use of it for missions beyond mars, in order to avoid carrying overly massive fuel tanks for it’s NERVA drive. a launcher station at earth fires streams of nano-tech devices called ‘fairy dust’, which the german ship can use as propulsion directly, or can ‘catch’ using magnetic fields, then ‘burn’ as fuel in its NERVA drive. at one point the launcher station fires massive amounts out to pre-planned ‘refueling points’ so the ship can make a trip from Ceres to Saturn on a fastest time route.
@James Jason Wentworth: This is an excellent idea, similar to one once expressed a while ago by Caleb Scharf on his blog:
http://lifeunbounded.blogspot.com/2010/08/stepping-stones.html
It is one of my favorite ideas about how interstellar travel might begin.
Great article. I’ve had the pleasure of sitting in on both Gerald and Jordin’s talks on their proposals at the Space Access conference. One minor nit: you seem to have typoed Gerard O’Neill’s name.
Glad you caught that typo — I just fixed it. Thanks.
Eniac wrote:
[ @James Jason Wentworth: This is an excellent idea, similar to one once expressed a while ago by Caleb Scharf on his blog:
http://lifeunbounded.blogspot.com/2010/08/stepping-stones.html ]
Thank you. I also noted your (and Paul’s) comments to Caleb Scharf’s ideas on his blog–as you all wrote, there are indeed numerous scientifically interesting things that they could do while enroute to their destination star(s). The beauty of the “pearl probes” (playing off Mr. Scharf’s “string of pearls” description of them), whether they be tiny like the microsail probes or larger (though still small) conventional spaceframes with subsystems, is that they could be used even within our own solar system to inexpensively investigate the Kuiper Belt and the Oort Cloud. In fact, a prelude of sorts to using such tiny probes to explore deep space occurred just a few days after Sputnik 1 was launched, and it went as follows:
Following a failed attempt involving a captured German V-2 missile, on October 16, 1957 Professor Fritz Zwicky succeeded in firing three artificial meteoroids to well above Earth’s escape velocity (one was found to have achieved 33,000 mph) using shaped explosive charges mounted in the nose of an Aerobee sounding rocket. One or both of the other two are in solar orbit. Professor Zwicky published an article about sending rocket/gun-launched payloads to the Moon, titled “Possible Operations on the Moon,” in the journal “Spaceflight” (Volume 3, Number 5, September 1961, pages 177 – 179), and below are links to articles about the Aerobee-boosted artificial meteoroids experiment:
http://calteches.library.caltech.edu/181/1/zwicky.pdf
http://utenti.multimania.it/paoloulivi/aerobee.html
http://www.slac.stanford.edu/pubs/beamline/31/1/31-1-maurer.pdf
http://www.time.com/time/magazine/article/0,9171,825363,00.html
Dear Paul, as a newcomer to your very exciting website “Centauri Dreams” , I would like to mention that I added my comments to Adam Crowl’s article in your April 2012 issue. Regarding the acceleration of macroscopic bodies to very high velocities one encounters a technology gap: While it is relatively easy to accelerate to relativistic velocities heavy nuclei like lead or uranium, it is very difficult to accelerate even microparticles to a velocity of 1000 km/s, which would require a linear accelerator at least 100km long. And to accelerate superconducting cm size projectiles to 200km/s by a travelling magnetic wave accelerator would require an accelerator at least 30km long. These numbers have come up with the idea of “impact fusion”, which is to ignite a thermonuclear microdetonation by hypervelocity impact.
Great physics, limited engineering. I keep running into 2 repeated conjectures: 1) if you could already do x; then x is irrelevant since capability is already there.
2) if you can do it; why hasn’t it already been done?
I am lucky. In my lifetime the Solar System was 9 planets on a children’s table mat… now it’s a Hubble Archive.
Half the job is getting the abilities to ‘build’ these fantastic machines… the other half is to ‘bounce’ with the unexpected of what we find while getting there.
With enough power, complexity & interest… there are worlds to be discovered. There will be firsts that we can’t even put a name on? Risk & failure will always be around; I think we really need to get out there and try something larger, because it ain’t coming to us by wishing it so.
If you take the principle of the solenoid chronograph, ( used to measure projectile velocity some time ago ) one must realize it is very much related to the principles of these ridiculously huge electromagnetic launchers discussed in this article. However, it makes no difference if we shoot a projectile through a fixed coil, or shoot the coil over a fixed object, the end result is the same. The moving projectile, distorts the electromagnetic field, in order for the solenoid chronograph to work. There is no reason we could not apply the reciprocal of such an electromagnetic launcher, and turn the launcher, inside out, so to say, by putting the field inside the projectile or object we want to accelerate. It has been known for some time electromagnetic field, has momentum, and thus can provide a reactionary force, independent of any “mass ejection”.
Thus, it seems obvious to me we could put the accelerating field inside the probe or ship and get the same exact result as an electromagnetic launcher. Then we are only dealing with a power issue. How to produce the power required to create such an accelerating electromagnetic field, within a ship, and it seems to me there is significant evidence this has been already solved, somewhere.