Behind the Scenes & Under the Hood: Virtuality's Antimatter Spacecraft Engine

By Amos Zeeberg (Discover Web Editor) | July 13, 2009 11:27 am

Phaeton VirtualityToday we present a very special installment of the Codex Futurius, Science Not Fiction’s look at the big scientific ideas in sci-fi: Kevin Grazier—JPL physicist and friend of SNF—gives an insider’s peek at the workings of and discussion around the Orion antimatter drive used to propel the Phaeton starship in Ron D. Moore’s recent TV movie, Virtuality. Grazier was a science adviser for the movie (which was intended to be the pilot for an ongoing show), so he was right in the middle of these discussions. The screenshot further down in this post shows the actual spreadsheet used in the production to see what stars would be reachable with the Orion drive. Without further ado, here’s some sci in your sci-fi:

DISCOVER: What kind of realistic technology could we use to get to nearby stars? Which stars would be feasibly reachable by such technologies?

Kevin Grazier: It’s a saying plastered on T-shirts and bumper stickers—the kind sold at both science-fiction conventions and physics departments nationwide:

186,000 miles per second:
It’s not just a good idea, it’s the law.

The speed of light, of all electromagnetic energy, in a vacuum is the ultimate speed limit in the universe. Nothing that has mass or carries information can travel faster.

This universal speed limit is a direct fallout from Albert Einstein’s special theory of relativity. Special relativity implies that the speed of light in a vacuum is a universal constant, but values that we tend to think of as constant in our daily experience—mass, length, and the rate of the passage of time—are not. Depending upon the relative velocity of two observers, these values will “adjust” so that both observers see the speed of light as a constant. Two observers travelling at high speeds relative to each other will find themselves in strong disagreement about measurements like the length of each other’s spacecraft and the rate of the passage of time.

Another consequence of special relativity is that, as an object travels increasingly faster, it behaves as if it has increasingly more mass. Therefore the amount of thrust it takes for an incremental change in velocity (known in the space program as a delta-V) is vastly greater at high speeds than at low. This effect is also highly nonlinear: It takes almost an order of magnitude more thrust to accelerate from .9c (nine-tenths of the speed of light) to .99c than it does to accelerate from .5c to .7c. An object travelling at the speed of light would act as if it had an infinite amount of mass and it would, therefore, require an infinite amount of energy (read: an infinite amount of thrust/fuel) to attain it.

This is, of course, a shame for civilizations (like ours) who want to explore planetary systems around other stars first hand. The distances involved are, well, astronomical. Just within the Solar System, it typically takes NASA probes 6 months to a year to reach Mars; it took Cassini 6 years, 9 months to reach Saturn. The (currently) fastest object created by humankind, the Voyager 1 spacecraft, will take 40,000 years, give or take a few thousand years, before it makes its closest encounter with its first star: AC+79 3888—currently located in the constellation Ursa Minor. At that speed few Time Lords, and even fewer humans, would survive the journey to even “nearby” star systems.

Current chemical rockets, and even the more efficient ion drives, cannot propel humanity to the stars at a reasonable speed, but there are concepts for interstellar spacecraft drives that are promising, that could be constructed in a practical sense, and you may be surprised how long the designs have been around. Stanisław Ulam, a Polish mathematician who participated in the Manhattan Project, proposed nuclear pulse propulsion back in 1947.

The idea is simple: explode a series of nuclear bombs behind a spacecraft. The explosions are directed against a thick steel “pusher plate”. The pusher plate is, in turn, connected to the spacecraft by a huge shock absorber to lessen the high G forces from the impulsive accelerations. In the straightforward terminology of Jimmy Johnson, the engineer on the Phaeton:

Basically, we gonna blow us up a bunch of big ass bombs off the ass-end of this here ship. Big ass bombs gonna vaporize some big ass alloy plates, and the translation of all that big ass energy’ll make us go real fast. Real fast. Yippe kai-ay, m…

The practical attempt to design and develop nuclear-pulse propulsion was performed by General Atomics in San Diego in the 1950s and 1960s. Ultimately the Nuclear Test Ban Treaty between the Unites States and Soviet Union made the testing for such a drive illegal, nevertheless over 50 years ago the design seemed practical and could be implemented within the bounds of existing technology. For more information, NASA and Star Trek designer Mike Okuda provided still more details on Project Orion, the U.S. government’s investigation into a nuclear-pulse spacecraft.

An Orion-style drive powered by thermonuclear explosions could theoretically reach speeds of .08c to .10c. That could get a spacecraft to the nearest stars within a human lifetime, but not within Phaeton’s 10-year mission. Virtuality is set in the mid-21st century, and it’s reasonable to assume some technological advances in the intervening time. Phaeton does not use thermonuclear explosions for propulsion, the charges dropped out the back are matter/antimatter charges (yes the thrust for Phaeton is, in essence, provided by photon torpedoes). The obvious assumption is that by the mid-21st Century, science has solved problems regarding the generation and containment of antimatter. One estimate has shown that Orion-style drive propelled by matter/antimatter explosions could attain speeds of .5c to .8c.

If Phaeton’s Orion Drive (named after the real-life nuclear concept) could propel it to 80 percent the speed of light, it could get to Sol’s nearest neighbor, Alpha Centauri (4.4 light-years away) in just 5 years, 6 months. That’s certainly a vast improvement, and shortens the round-trip mission time to several nearby stars to less than a human lifetime.

Only, it gets better.

Special relativity, which bit us in the asteroid when it comes to top-end velocity, does our crew a favor as our spacecraft attains speeds that are a high fraction of the speed of light. Recall that for objects travelling at relativistic speeds, values like mass, time, and length appear to “adjust” to keep the speed of light a constant. At high speeds, distances that we measure at “rest”, or at low speeds compared to c, appear to be shortened. This effect is called Lorentz contraction or Lorentz-Fitzgerald contraction.

Moving at a snappy .5c, the distance to Alpha Centauri is only 3.8 light-years (down from 4.4), and the apparent travel time is a bit over 7 years, 6 months. At 80 percent light speed, the distance is 2.6 light-years, and the travel time is 3 years, 3 months—less elapsed time for the crew than it would take for light to make the same journey.

Travelling at a speed of 0.7c is the “break even” point, where the combination of spacecraft velocity and Lorentz Contraction means you are travelling at “functional light speed” (the distance to Alpha Centauri in that frame would be 3.1 light-years and the travel time 4 years, 5 months). Of course time passes at different rates based upon their relative speeds as well, a phenomena called relativistic time dilation, so if Phaeton were travelling at a speed of .7c, for every year that passes for the crew, a year and five months would pass for The Edge of Never viewers back on Earth. Billie Kashmiri alludes to this in her confessional near the end.

With the phenomena of Lorentz Contraction as an aid, many more star systems become potential targets of a 10-year mission. There are sound scientific arguments why astronomers believe that any star that could potentially have a planet with life, in particular intelligent life, must be similar to our Sol: from mid-F range on the Herzsprung-Russell Diagram to mid-K. There are several stars in that size/temperature range in Sol’s neighborhood. Below is a screen capture of a spreadsheet that the producers of Virtuality used to select the target star for Phaeton’s mission (text color corresponds to the star’s color):

On the spreadsheet are the stars’ distances at rest, and at various fractions of light speed—with the corresponding travel time.

Virtuality planet spreadsheetClick image to embiggen.

Epsilon Eridani, the nearby star that the Phaeton is sent to explore, has one, perhaps two planets orbiting it, as well as at least three asteroid/planetesimal belts. If we assume that Phaeton’s Orion Drive can get her up to .8c, or 80 percent the speed of light, then because of Lorentz contraction the journey (normally 10.5 light-years) is only 6.3 light-years, and it takes just under 7 years, 11 months. So if the Orion Drive can reasonably get a spacecraft up to .8c, then Phaeton’s mission is actually closer to 16 years. If, however, the Orion Drive was capable of propelling Phaeton to .9c, or 90 percent the speed of light, then the distance to Epsilon Eridani is only 4.6 light-years, and the one-way flight time is 5.1 years.

So in order for Phaeton to get to Epsilon Eridani and back within the stated 10-year mission duration, we clearly see that the ship’s Orion Drive would have to propel her to over 90 percent the speed of light (.9c). For all the elements of Phaeton’s mission that might be practically attainable by the mid-21st Century, this is where a little science fiction enters the picture.

Thank you to Steve Cooperman, Doug Creel, and John Weiss for their helpful input and comments.


Comments (35)

  1. Jason

    Probably the best description of a pulse drive ever. It wasn’t a terrible movie either, if they kept their scientific foundation it could be a great promotional push toward stellar and interstellar exploration.

  2. Amos Zeeberg (Discover Web Editor)

    Thanks, Jason [on behalf of Kevin]. SNF takes its sci-in-sci-fi job srsly.

    Anybody have suggestions for topics we should cover in future Codex posts?

  3. CW

    Has Codex ever discussed ship shielding for interstellar travel? For example, Arthur C. Clarke’s novel “Songs of Distant Earth” theorized a mammoth ice shield to protect the vessel from micrometeors.

  4. Scott

    Do these calculations take in to account acceleration/deceleration rates? These drives don’t get up to speed immediately, and if it takes a light year or two to get up to .9c, it takes the same distance to slow down again at the end, and that distance is not as contracted, extending the time required.

    I remember a quick offhand mention of this effect in one of the Ringworld novels.

  5. grey

    Must be a NASA scientist because this technology more Star Trek than reality.

    Ablation plate using fusion explosions, bad idea.

    Photon torpedoes will only work for Capt. Kirk and crew.

    Seriously, maybe someday I will read about something that will work.

  6. Electron-positron annihalation is clean to photons – but it is only 0.027 wt-% of stuff. Hadron-antihadron annihalation loses 50% of its total mc^2 as neutrinos. Ya gotta build the engine TWICE as big. Maybe worry about blood chloride Cl-37 + nu to Ar-37, that electron capture-decaying with a 35 day half-life.

  7. Brian

    Uh, doesn’t the ion drive dismissed here have a lot more potential in it? The thrust values are low but if you’re going to be flying for several years anyway, you can invest some time in accelerating.

    I thought the fatal weakness of the Orion drive is that it simply doesn’t matter what you make the pusher plate out of, it’s vaporized by the nuclear (or matter/antimatter) fireball. Also it stretches credibility to believe that you can make a shock absorber good enough to transfer energy to the spaceship in a way that doesn’t flatten the occupants into a red floor coating.

    With the ion drive, the problem reverses. Instead of having acceleration rates that are too high, you are trying to beef up the drive in order to produce acceptable accelerations.

  8. Starcruiser

    it seems that when a science fiction spaceship needs that ultimate energy source for maximum specific impulse, it’s easy to fall back on that old standby, antimatter. Realistically, using matter/antimatter annihalation may not be as simple as throwing a BB sized pellet into a BB sized pellet of regular matter.

    There are some questions I have to how this would work to get the performance you describe.
    What is the actual efficiency of such a drive, in how much of the exploding matter/antimatter gets converted into forward thrust? What percentage of the total mass of the ship has to be ablation plate, and how much has to be anti-matter bombs? Of the antimatter bombs, how much of the mass is actual antimatter, compared to the other stuff needed to make the bomb work? Just how could such a bomb be designed to work at optimum efficiency of matter to energy conversion? What percentage of the antimatter is actually annihilated, vs, how much is wasted and just disperses into space?

  9. Klapaucius

    How would such a ship slow down? (haven’t seen the movie, my apologies if this was explained). Would it loop around the target star to transfer momentum?

  10. Lou

    How nice it is to fantasize about these star drives? Everyone knows their Bull unless one is talking about un-manned craft!




  11. scribbler

    Accelerate at the speed needed to give normal gravity. Decelerate at the same rate. Kills two birds with one stone.


    Carl Sagan had some destinations mapped out factoring in the time dilations and all. Once you set your target power needs to those to maintain only Earth gravity, ion drives become very attractive…

    As for matter/anti-matter explosions, I’d think it would be more like using TNT to propel a boat.

    My money’s on ion drives or nuclear reactors that ram jet gases either taken with us or those present in space or both.

    As for protection from both particles or radiation, water ice is handy. Still, to funnel space gas into a reactor would require magnetic fields that would double as both funnel and shield.

    Interesting mental puzzles…

  12. Joe C

    Whatever our future generations come up with (assuming their not trying to scratch out an existence after global warming or something causes civilization to collapse) they will probably think these ideas as quaint as victorian notions of taking a hot air balloon to the moon.

    Or, we’ll think the idea of sending actual people there is nutters when robots with downloaded human personalities can do the job so much better. We’ll be Post Human long before we build one of these drives…100 years to a star will seem trivial in not so many years, the only issue being the long wait for news back on earth. But the holo movies, unlike the space ship, will fly back to us at the speed of light and everyone on earth will be able to experience it as if we were there.

  13. G

    I agree with Scott, no acceleration or deceleration rates have been accounted for along with the communication back to earth. Once the craft reaches say 1/100th of a light year away from Earth, communication between the space craft and Earth is worthless. At this distance, communication takes nearly 4 days to reach its destination, same for a respose, meaning 99%+ of the voyage will have no communication.

  14. Brian


    I could not disagree more. Human beings crave communication and will do almost anything to get it.

    Ever heard of snail mail? Letters routinely take 4 days or longer by ground delivery, and yet were the standard for long distance communication for over 100 years.

    I’ve worked in remote locations and news from “home” is a big deal. People want letters, newspapers, magazines, books, whatever they can get. Work in a camp often comes to a complete stop when the mail comes in. It’s an important morale booster; easily as important for morale as good food.

    Maybe a spoken conversation will not be viable at a distance, sure. That breaks down pretty quickly, even at Mars type distances. Interactive communication is possible but becomes cumbersome. However unless the spacecraft is so big that people are taking their entire social circle and all interests with them, spacefaring people will want news from Earth. Even if it takes several years to get there.

    Don’t forget that electromagnetic communications can (and would) be sent as gathered. The recipients can receive a nearly continuous stream of communications in a “pipeline” sequence. There don’t have to be any gaps.

    At a distance of several Light Years from Earth, all communications received would be that many years old. Despite this there would be an uninterrupted stream of data, flowing in both directions. This would maintain a connection between the Earth and the voyagers, and interest in each other. The turnaround becomes a huge issue but otherwise the communication link can be fully functional.

    No, the only thing that would likely kill a communications link would be if Earth stopped caring about it’s long distance astronauts, or vice versa. As long as the interest is there they will find a way.

  15. chris

    Hmm, how about we create a “ship ballast” out of something that has negative mass?

    So as the ship accelerates the mass is not increased due to the decreasing mass of the ballast?

    I think the concept of utilizing “a giant shock absorber” to be silly.

    Instead control the amount of force applied, this is the same thing as the effect of the shock absorber except less moving parts and mass.

    Also, early interstellar craft will likely use multiple forms of thrust, where efficiency and mass of the thrust are more important.

    Such as using both a solar sail and an ion drive?

    But I suppose the really notable thought is that these endeavors must have a purposeful reason. Generally given our nature, I would say that we’d need to use most of the resources of our solar system before we’d truly have much need/want for interstellar travel.

    That is unless we find ways to break special relativity.

  16. Hi All

    Accelerating at 1 gee to 0.9c takes 1.43 years ship-time and you travel 1.254 light-years (non-Lorentz contracted light-years that is), so that’d need to be factored in.

    As for nukes vaporising the pusher plate – nope. They set off BIG nukes next to metal spheres coated lightly with graphite – the spheres survived, tho a thin layer of graphite was ablated. That’s what gave the ‘Orion’ designers confidence it would work back in the late 1950s and all the subsequent tests pretty much confirmed it.

    A few problems arise. To get decent efficiency the pulse-units need to produce high quality directed plasma. With fission bombs a range of ‘filler’ materials gave the blast the desired directionality. But m-am reactions produce LOTS of gamma-rays and converting their energy into directed plasma isn’t straightforward. I guess the F in the SF is some ‘unobtainium’ to reflect the gamma-rays.

    Protection against the interstellar medium is a must. Would’ve been nice to see some details on what the writers assumed for that – more ‘unobtainium’? Or magnetic deflection and a forward-facing ‘ioniser beam’ to charge up the impinging ISM? The ‘unobtainium’ can then reflect away the resulting x-ray bremstrahlung.

    Finally, travelling at 0.9 c means that you’ll only receive signals from Earth from part of the journey time. Travelling 10.5 light-years at 0.9 c means you take only 1.167 years longer to get there than light does – thus you’ll only get 1.167 years worth of transmissions during the journey and have to wait for the rest to arrive. You only ‘catch up’ to the apparent future on Earth by flying back home.

    These exact figures assume instant acceleration but the result is similar even when we factor acceleration in. In the case of a 1 gee acceleration over the whole trip then something curious happens. Imagine the Earth is sending us a clock-signal. On arrival at Epsilon Eridani how much time has elapsed as measured by the clock-signal from Earth? Just 1.79 years. Yet our ship-board clocks will say 4.94 years…

    Ponder that.

  17. scribbler

    OK, pondered a bit…

    So, how much time will have passed on Earth, being that in this example, it’s “relatively” stationary?

  18. Panagiotis Karatasios

    I don’t understand why they chose to use this propulsion with antimatter explosions and not a matter- antimatter reactor which would send them to 0.93c. Of course a matter -a antimatter reactor has its problems (how to produce and store the amount of antimatter needed) but the same is true for this nuclear propulsion system.

  19. I have a question, achieving .9c speed now sounds “easy” everybody talks about acceleration but nobody is talking about how does the spacecraft slow down from such a permanent speed?


  20. Brian


    The speed isn’t permanent at all. The simple rule is that you accelerate continuously until, approximately halfway to the destination, you turn the ship around and continuously decelerate.

    If you do not do this you’ll overshoot your target (and that’s assuming you miss the destination. If your aim is true you slam into the destination!). Overshooting is not like missing a turnoff on a highway; at these speeds you’ll have NO CHANCE WHATSOEVER of slowing down in a reasonable timeframe.

    However the “accelerate halfway, decelerate halfway” rule will probably need some modification in reality. If the propulsion system is a reaction mass system you may need to conserve mass. In that case you accelerate until you’ve used half your reaction mass.

    If the propulsion system has a top end speed and you reach that early, you might as well shut it down. I’m talking about Newtonian mechanics here.

    If you start hitting relativistic speeds (yeah, yeah, they’re all relative…), then your ship gains apparent mass until it hits infinity at the speed of light. Therefore somewhere above .9c, even if your drive has more apparent power, your ship resists accelerating more and more. Eventually it’s not worth trying any further.

    Finally, your drive system may have power limitations, or you may wish to devote it’s power to non-propulsion activities. Power conservation or allocation may cause you to take your foot off the accelerator.

    The effect of all these flight plan modifications is that you spend an interval in the middle of the trip coasting. You still accelerate for a set period at the start of the voyage and decelerate for an identical period of time at the end. This ignores so-called black swan events, like inventing a better propulsion drive during the trip itself, or being boarded by aliens (!).

  21. Awsome!

    I was thinking about some like that but it’s better to read it from somebody else.


  22. hmm, I don’t remember antimatter being mentioned in Virtuality? I understood it to be the usual orion drive, using nuclear fission bombs.

    I actually don’t like the idea the way they present it on the show, because they show it being done by a private company, I certainly hope the United States Armed Forces would not trust a private company to take proper care of thousands of nuclear weapons. If they were using the Orion Drive, the ship should have been USS Phaeton (military designation meaning “United States Ship”, a commissioned Navy warship) not just Phaeton.

    The Mini Mag Orion concept would be better, where you have fissile material, eject it out the back of the craft, and use a magnetic field to confine it (achieving critical mass) until fission occurs. That way there is no nuclear weapon aboard the ship.

    Personally my favorite is VASIMR, it is a plasma drive that can use hydrogen. Then you can use your friendly neighborhood gas giant planet to refuel, and again, no nuclear weapons involved. :)

    Also I thought it was a little weird on the show that they were acting like the VR would be their only source of entertainment underway. What about movies, books, music, etc.? We get by that way on submarines with no problem.

    –Brian (former us navy submariner)

  23. We should ask: why do fast moving objects gain mass? Or time pass more slowly in their volume? That may be answered by adding the relativistic Lorenz-Einstein transform functions together and data mapping that across the quantized wavefunctions for frequency and wavelength. This RQT (Relative Quantum Topological) analysis develops the picoyoctometric 3D animated video images of the force and energy fields of electromagnetic photons by inclusion of quantum symmetry numbers in their topological wavefunctions.
    This model displays spacons, chronons, and magnetic field elements as they react to conserve momentum and hence fulfill relativistic quantum physics. The flowing mechanics of Lorenz-Einstein relativity are visible there in picoyoctometric detail. Images of the h-bar magnetic energy waveparticle of ~175 picoyoctometers may be seen at

  24. StephenBoston

    @Dale (#28): OK, I’m sure it’s a joke and all, plus the link was dead, and I surely didn’t understand the details, but…

    It sounds like you’ve managed to unify Relativity with Quantum Mechanics, then? (to “fulfill relativistic quantum physics” in “picoyoctometric detail” no less?)

    In that case, congratulations, Mr. Future Nobel Laureate! :)

  25. RevStarlust

    All the posts here are almost a year old but here goes a potential waste of energy! Lou was worried about radiation for manned exploration of interstellar space. If we used these fission fusion bomb drives Why couldn’t we improve the exhaust speed with a magnetic bottle? We should have a huge excess of electrical power because if I designed such a craft it would be built in LEO and be the size of a battle ship or even better an aircraft carrier with many redundant reactors each with enough electrical generating power for a small city for life support and magnetic shields to supplement mass shields for cosmic radiation and domestic radiation created by the ship. I don’t understand why we dont build small POC robotic ships today.


Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!


See More

Collapse bottom bar