≡ Menu

Stars as Stellar Engines

I’ve always loved the idea of an O’Neill space habitat because of the possibility of engineering a huge environment to specification. That notion translates well to worldship ideas — a multi-generational journey would certainly be easier to take in an environment that mimicked, say, a Polynesian island, than aboard something more akin to a giant metal barracks. But best of all is to take your environment with you, which is why the thought of moving entire stars and planets to another location has such appeal when we’re talking on an intergalactic scale.

Adam Crowl reminded us of the possibilities on Monday:

In theory a tight white-dwarf/planet pair can be flung out of the Galactic Core at ~0.05c, which would mean a 2 billion year journey across every 100 million light-years. A white-dwarf habitable zone is good for 8 billion years or so, enough to cross ~400 million light-years. It’d be a ‘starship’ in truth on the Grandest Scale.

Back in November of 1973, Stanley Schmidt’s The Sins of the Fathers began as a three-part serial in Analog, then under the editorship of Ben Bova, who had taken over after the death of John Campbell in 1971. Schmidt would go on to become Analog’s editor himself in 1978, only retiring recently, so that his own tenure at the magazine matched Campbell’s long run. The Sins of the Fathers would be published as a paperback in 1976 with a cover by the brilliant SF artist Richard Powers. Lifeboat Earth would continue the journey in the Berkeley paperback of 1978.


Schmidt’s plan was to make an intergalactic crossing to M31, the Andromeda Galaxy, with the help of alien technologies. The plot involves an explosion in the core of the galaxy on such a scale that planets will be rendered uninhabitable throughout the Milky Way. Fortunately, an alien race called the Kyrra has arrived to help, equipping the Earth with what Schmidt called an ‘induced annihilation’ drive that converts matter to energy without the need for antimatter. With this gigantic rocket nozzle mounted at the South Pole, the Earth is nudged out of its orbit, at which point the Kyrra’s FTL technology (Schmidt calls this the Rao-Chang drive) cuts in.

Of course, the effects of maneuvering the planet in this way are substantial. Schmidt explained them in an article called “How to Move Planet Earth,” which ran in the May, 1976 issue of Analog, after the serialization of his novel was complete. Here’s a bit of this:

Perhaps the most immediately striking of these [effects] is to change the effective ‘up-down’ direction. To a person standing on what had been a level plain (or floor or ocean), the appearance and feel of this is exactly as if the Earth were tilting under his feet. All over the Earth, the ground appears to slope downward to the south. The amount of tilt, and the strength of the effective field, vary with latitude… One of the first globally important consequences of this effective tilting will be a tendency for the oceans and most of the atmosphere to flow ‘downhill’ and concentrate (to such extent as they aren’t blasted or blown away) at and near the South Pole.

And so on. Schmidt works through all the consequences in the article, which recounts his valiant attempt, having plugged in magical alien technologies, to work out their physical effects according to known physics. Surely he was smiling when he wrote: “But these things — the Rao-Chang, induced annihilation, and exhaustless conversion process, together with their logical implications — are the only really new physics I have assumed.” I love that ‘only’! In any case, the journey is a nightmare, with the alien technologies consuming what Earth resources they haven’t already destroyed in the propulsion process, so that by the time our battered world gets to M31, the few survivors must get off the planet and onto another one.

Robert Metzger, who for years wrote the science column in the Science Fiction Writers of America’s Bulletin, wrote a novel called Cusp (Ace, 2005) in which the Sun erupts with a massive, propulsive jet and begins a journey of its own, with the Earth suddenly encircled by enormous ring-like structures that help propel it along with the parent star. Here we’re in the company of quantum supercomputers (the ‘CUSP’ of the title) and technologies evolving into the Singularity so often speculated about in science fiction and elsewhere. Needless to say, the physical effects of moving the planet and star are as acute as they are in Schmidt’s novel.

In Bowl of Heaven (Tor, 2012), Gregory Benford and Larry Niven looked at ways to move an entire star to travel the galaxy — the sequel, ShipStar is just out (Tor, 2014), and nearing the top of my stack. Imagine half of a Dyson sphere curved around a star whose energies flow into a propulsive plasma jet that moves the entire structure on its journey. Here the notion of living space may remind you of Niven’s Ringworld, that vast structure completely encircling a star, though not enclosing it. The difference is that in the ShipStar scenario, most of the ‘bowl’ is made up of mirrors, with living space just on the rim.

I see the ShipStar model as a modified Shkadov thruster, a way of moving entire stars that the physicist Leonid Shkadov first described in 1987. In both cases, we’re talking about what can be called ‘stellar engines’ that use the resources of the star itself to create their propulsion. Would such a vast structure be detectible by another civilization? As with Dyson spheres, the size of the objects makes it feasible to consider picking them up in exoplanet transit data. Scottish physicist Duncan Forgan has considered the transit signature of a Shkadov thruster. As with the work of Richard Carrigan, the man whose searches for Dyson spheres have helped to define ‘interstellar archaeology,’ the Shkadov thruster could play a role in future SETI searches.

As is true of all such searches, we have to determine whether what we are seeing is fully explicable in terms of natural phenomena or whether there is a case to be made for technology, and I would rate the chances of our finding a Shkadov thruster quite low. But searching for artifacts in our existing astronomical databases is clearly a worthwhile idea. Certainly a civilization that had the power to move a star might find it a livable way to embark upon journeys lasting millennia. In such ways, a trip to another galaxy is not inconceivable even if tens of thousands of generations might live and die along the way. The key question: What compelling reasons might drive such a journey?

What I haven’t had the chance to get to today are astronomer Fritz Zwicky’s ideas on moving stars, an omission I’ll try to rectify next week. The Forgan paper mentioned above is “On the Possibility of Detecting Class A Stellar Engines Using Exoplanet Transit Curves,” accepted for publication in the Journal of the British Interplanetary Society (preprint). Leonid Shkadov’s paper on Shkadov thrusters is “Possibility of controlling solar system motion in the galaxy,” 38th Congress of IAF,” October 10-17, 1987, Brighton, UK, paper IAA-87-613.


Comments on this entry are closed.

  • Al Jackson June 27, 2014, 14:24

    I did not know until recently that Leonid Shkadov’s paper was presented at an IAC meeting in 1987.
    Because the first time I cam across it was in English Translation in
    Solar System Research, Vol. 22, No. 4, p. 210 – 214 in 1989…
    Tho the Russian version was published in:
    Astronomicheskii Vestnik (ISSN 0320-930X), vol. 22, Oct.-Dec. 1988, p. 333-339. In Russian.

    I thought I had sent a hard copy of the Solar System Research paper to Greg Benford in 1989? ;)

  • Alex Tolley June 27, 2014, 14:36

    The Richard Powers’ cover for “Sins of the Fathers”.

  • Gregory Benford June 27, 2014, 14:40

    The Forgan paper on “On the Possibility of Detecting Class A Stellar Engines Using Exoplanet Transit Curves” shows how hard finding one will be, indeed. A matter I noticed in all previous stellar engine models is their stability: hard to see how the simple Shkadov thruster can be.
    That’s why Larry Niven & I spent a lot of time in Bowl of Heaven, and more in Shipstar, describing how the jet which drives the Bowl-star system is stabilized dynamically. Plus we turned it into a major plot point, and along the way introduced some new kinds of aliens, all relevant to the longterm management of the Bowl, and its vast history.
    I do wonder if real aliens would ever build such a massive construct. There are structural demands we can’t meet now, but as with Dyson spheres, there are dynamic ways to manage those, as well.

  • Oz Monroe June 27, 2014, 21:05

    Awesome, I have that issue of Analog in my collection.

  • Eniac June 27, 2014, 22:02

    Stanley Schmidt:

    Perhaps the most immediately striking of these [effects] is to change the effective ‘up-down’ direction.

    With due respect to the author, I do not think that this particular issue is thought quite through. The Earth cannot be pushed this way. The Earth, because of its size, behaves like a glob of liquid, not a solid. That is why it is almost perfectly round. Its surface cannot be off the orthoganal of gravity more than the height of the highest mountains.

    If you could build an engine strong enough to generate noticable acceleration of the Earth, no solid material would be a match for its force. It would punch right through and leave the rest of the Earth sitting behind. Or, more likely, the engine would be crushed and incinerated while penetrating the mantle.

  • Eniac June 27, 2014, 22:06

    We havee been over Shkadov thrusters, and I believe we concluded that they would take billions of years to attain only a very modest velocity before the fuel burns out.

  • Chris W June 28, 2014, 11:44

    Eniac – “If you could build an engine strong enough to generate noticable acceleration of the Earth, no solid material would be a match for its force. It would punch right through and leave the rest of the Earth sitting behind.”

    Indeed, and even if you managed to spread the force over a large enough area, you would generate tidal forces that would rip the planet apart.

  • Adam June 28, 2014, 18:03

    One of the “Analog” covers for “Sins of the Fathers” shows Melbourne being flooded – my grandmother’s home town. Don’t often see Australian cities on SF covers.

  • Rob Henry June 28, 2014, 21:28

    Interesting comments by Eniac.

    To his first one I would like to add that the mantel flows freely for slow pushes, but resists sudden ones, so it will act very differently for a hundred year push than a longer one. This would give two very different maximum slope parameters, one for the short term and one for the long one. Also, the highest mountains are a misleading measure as they are partly supported by the buoyancy of the thick and light continental crust underpinning them. A much better measure would be how much higher an active volcano can build its shield above the plain. For the longer term slope, I suggest trying to work with the pace that Norway is rebounding after the last ice age.

    For Eniac’s second comment, I repeat that the Shkadov thrusters delta vee are just manoeuvring budgets.

  • Rob Henry June 28, 2014, 23:00

    Above I should have given ballpark figures. The initial rebound after the end of the ice age was rapid, but some parts are still rebounding at up to 1 mm/yr ten thousand years later, implying the pressure equivalent to >> 10m of granite but less than 20,000m of ice, can be sustained over huge, almost hemispheric, area for centuries. This gives figures of between 0.35 MPa and 200 MPa. The cross-sectional area of Earth is 125 million sq km, so our force is 45 – 25,000 EN. Now Earth weighs 6 x 10^24 kg, so we can accelerate at between .007 and 4 mm/s/s for at least a hundred years before we must ease off the pedal a bit.

  • torque_xtr June 29, 2014, 15:16

    Eniac, the limitations of a Shkadov thruster are formidable, but one could imagine it’s modified versions where they are much less limiting…

    What if some mass of the star is allowed to be used as the reactional?

    If thruster uses only light, than yes, the maximum delta-V is strongly defined by the energetics of the thermonuclear burning and the specific impulse of reflected radiation – is roughly the same for all stars (some percentage of “с”, which is the same as the rest energy liberation efficiency by stellar thermonuclear burning, multiplied by the efficiency of thruster reflectors) and the lifetime of the star is needed to achieve it. In case of hot blue stars it is possibly reasonable to use classic Shkadov thrusters as the means to transport Kardashev II civilization, and these stars simultaneously are the best for Dyson Swarms (on the basis of which the thruster is likely to be constructed) because of maximum power available.

    But if the light is somehow focused on a (very) small spot on the star’s surface instead of simple reflection, so that the plasma there is hyperheated and expands with the multiple times the star’s escape velocity, it creates the modified star thruster akin to the described in the “how to move the Planet Earth”, with the lower specific impulse, but much higher mass flow, which allows to accelerate the star faster at the cost of mass reduction. Of course the thruster should be more advanced than a simple spot heater, to avoid the omnidirectional expansion of plasma and it’s collision with the Dyson swarm elements – ideally it should somehow eject the stellar mass in some form of a collimated jet)

    But assuming that for type II civilization it is possible, in the extreme case of O-type star reduced to M-dwarf, a very high “fuel-to-payload ratios” could be achieved, equal to the mass ratio of the heaviest stars to the lightest, or up to 1000. That could possibly increase the maximum delta-V of a Shkadov thruster up to fivefold and place it in the range of a substantial fraction of the speed of light!

    (there could be issues of the needed decceleration, of the drag force by the interstellar/intergalactic medium on the Dyson Swarm elements, the benefits of combining the Shkadov-modified thruster and Bussard ramjet by somehow redirecting the medium towards the central star, and many more, and all that sounds quite crazy, but where the boundary of infeasibility for the Kardashev II civilization lies? :-) )

    PS quick calculations show that the mass of intergalactic medium in the column swept by a big Dyson swarm directly in the intergalactic voyage (1000000 ly long and 100 AU wide, 1e48 m^3 by volume) is on the order of the mass of the Moon, which is low enough to become the issue and to enable Shkadov-Bussard thrusters except possibly only for the largest “inter-hypercluster” voyages by the largest swarms-thrusters…

  • Rob Henry June 30, 2014, 16:28

    Adding once more to what I wrote… The last ice sheet was approximately a couple of kilometres thick over the continental portions, and its removal resulted in a biphasic rebound. One of these was geologically indistinguishable from being instant, the other gradual, the first corresponding to Eniac’s Earth flow, the second to my rigid Earth..

  • stephen July 2, 2014, 10:54

    I remember in elementary school an issue of My Weekly Reader had an article about a scientist suggesting we move the Earth out of the solar system with giant rockets. I wondered how we’d keep warm if we did that. We’d be dependent on geothermal energy.

    Speaking of which, some scientists think the Earth’s core might be 10 million degrees. Is that right?

    Has anybody worked on the possibilities of doing this with a multiple star system? One of my favorites is Beta Capricorni, which may consist of seven stars.

  • James M. Essig July 2, 2014, 12:11

    Now if only a non-Fermi-Dirac matter to energy conversion process could be found and applied on a scale commensurate with Stellar Engines! That would be great.

    As someone who used to read the latest fringe literature, there was often a theme that nuclear reactions of at least some forms release an undiscovered potentially harmful form of radiation. Capture and use of any related stellar radiation from stars would be awesome!

    I always keep an open mind even regarding the fringe. We as a species should be open to call cultural scientific memes just incase there is something profound that our theoreticians and experimentalists have missed.

    I keep hoping for some new news regarding nuclear energy and perhaps new QCD physics when the LHC resumes operation next year after the latest upgrade process is complete.

    Regardless, thanks for doing these threads on Stellar Engines, Paul, and also to all those who have commented on Paul’s posts.

    Going cosmic by stellar engine travel can be an agenda that employs untold numbers of our deep future descendants. As someone who enjoys the time-less theme of creation on the cosmic scale, stellar engines are of special attraction to me as a subject. This attraction is not only intellectual, and deeply emotional and affective for me as well.

  • Michael July 2, 2014, 13:45

    @stephen July 2, 2014 at 10:54

    ‘Speaking of which, some scientists think the Earth’s core might be 10 million degrees. Is that right?’

    Nope, I think it is around 6000 to 7000 degrees Celsius. Although the decay reactions are in the billions of degrees they are few and far between.

  • Rob Henry July 2, 2014, 17:59

    Michael, when I last looked at this matter it was a huge unknown. Yes, the conventional view was about that 6,000 K, but our core could well be 10 thousand, or even slightly more.