Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Light powered spaceship

  1. Feb 10, 2010 #1
    This thread is a spin-off from another thread that got me interested in an idea. Previously I didn't know this, but apparently you can use a flashlight to power a spaceship (since light has momentum). I understand the process would be horribly inefficient, but nevertheless it got interested in how feasible it would be...whether now or in the future.

    So I went and ran some calculations, and I ended up with some interesting numbers. So interesting, that it makes me think I must be wrong. After all, I used my limited knowledge of simple kinematic formulas to arrive at my results. So I'd like to ask the forum to help me point out my mistakes...maybe I missed a decimal place somewhere, or applied formulas completely wrongly.

    I found on Wikipedia that it would take about 300MW worth of light to generate a Newton of force. I then used E=MC^2 to figure out how much energy we could get if we were to annihilate matter/ antimatter...and basically my results told me that if we were able to annihilate 6500kg of mass, we could accelerate a 10000kg spaceship to about half the speed of light. With a flashlight.

    To figure out the acceleration of the spaceship, I figured out the starting acceleration (at 16500kg) and final acceleration (at 10000kg), and took the average. Is this correct? Seems overly simplistic...

    Another question, I have found many separate pages referencing the following:


    Basically, some company was able to get something like 3000x more force from a "flashlight". This would reduce the amount of matter we need to annihilate to less than a kilogram. Now that just sounds completely out of this world. Is it even physically possible to increase the net force from photons? Does it not violate any physical laws?

    The original thread that got me thinking about this:

    Last edited: Feb 10, 2010
  2. jcsd
  3. Feb 10, 2010 #2
    299,792,458 m/s is the speed of light.
    149,896,229 m/s is half that.

    (1 newton) / (10 000 kilograms) = 0.0001 m / s2
    ((149896229 m/s) / 2) / (0.0001 (m / (s^2))) = 47 500.2632 years

    Your going to need a brighter bulb :).
    at 3000x the force its about 16 years.

    I don't see where your going with the mass required. I can't open the link, but if it has 3000x the output i doubt it has the same input... look at a v8 vs a v4. the v8 has more power but it uses more gasoline. Even if it had the same input, it would still take 2.25 kg, not less than a kg.

    Also, I havn't got to relativity yet, but I beleive that as you go faster your relative mass increases, making it more difficult for you to accelerate (hence, you can't reach the speed of light).
  4. Feb 10, 2010 #3


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    Using the relativistic rocket equation, it works to about 0.46c. This is assuming that all of the energy can be directed as light propelling the craft.
    No, it wouldn't reduce the amount of matter needed. What determines that is the "exhaust velocity", which in this case is the speed of light and doesn't change with how much force is output. What the force determines is how quickly your craft will reach its final speed, not the final speed itself for a given amount of fuel.
    Last edited by a moderator: May 4, 2017
  5. Feb 10, 2010 #4
    I fixed the link. Google "photon thruster" or something like that and you will see various articles on it.

    And of course I realize we will need a more powerful light bulb :) I simply wanted to show the relationship between power and force.

    As for the 3000x the output, I didn't mean that it takes 3000x the power, I mean it puts out 3000x the force with the same power input! And it doesn't even look like they put that much time into it...
  6. Feb 10, 2010 #5


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    I finally found an article that describes the system. Basically, it is a modified light-sail concept. The gain comes from bouncing the light back and forth between the spacecraft and launch platform, rather than just once as in the case of the light sail. It isn't "take your engine with you" rocket propulsion, it is a "leave your engine at home" system.

    This allows you to get the most out of your light in terms of propulsion. In other words, it lets you get near to the efficiency needed to propel that 10,000 kg ship up to .5c by converting 6500 kg of matter to energy, whereas with conventional system waste the vast majority of the energy they use.
  7. Feb 10, 2010 #6
    This concept only really works if you use a laser - as in most of the energy goes where you want it. With most other light sources light goes in all directions and you lose a lot of energy.
  8. Feb 11, 2010 #7
    Yeah I just realized that, too. You would basically need two spaceships travelling in opposite directions. And then, to what distance really can the lasers bouncing between them remain effective.

    Still, the 6500kg and .5c speed I figured out using the "bring your engine with you" concept. I mean, even a fraction of this is beyond most anything we can imagine, and we're talking about a rocket a fraction the weight of conventional rockets (which can get up to 3,000,000kg)

    Am I the only one that finds this conecept amazing? I mean, have never heard anybody even mention anything about it, and yet it seeems like a feasible way to travel between stars. Sure we would need petawatt lasers and thousands of kilograms of antimatter...but I don't see this as impossible, or even more than a few generations away. It definitely seems more likely than filling a rocket with millions or billions or whatever tons of propellant....

    I went ahead and checked out how fast we could get going with the more available fission technology. A 10,000kg spaceship on a rocket the size of the one that took us to the moon, if loaded with Uranium, would be able to get up to something like 65 million kilometers an hour. A similar rocket powered by fusion: 500 million kilometers per hour. Even if we could only get 10% of this, that's still significant fraction of the speed of light, and imo, quite amazing....
    Last edited: Feb 11, 2010
  9. Feb 11, 2010 #8


    User Avatar
    Science Advisor
    Gold Member

    3250kg of anti-matter is no small amount! At current energy prices (~$50 per megawatt-hour), even with perfect theoretical efficiency (which is 1/2 given that to produce 1 gram of anti-matter you also have to produce 1 gram of matter), anti-matter would cost 2.5 million dollars per gram to produce. At that cost, 3250kg of anti-matter would cost 8.125 TRILLION dollars (or ~57% the GDP of the USA).

    Current efficiency is more like 4E-8, which would yield a cost of 31.25 Trillion dollars per GRAM (~45% WORLD GDP). A total cost of 102 QUINTILLION dollars for 3250kg of the stuff. That is far more than the World's GDP.

    Remember, we don't just see anti-matter lying around in space where we could mine it! We have to produce it artificially anti-proton by anti-proton.
  10. Feb 11, 2010 #9
    You can't really use that argument. How much did a gram of enriched Uranium cost in 1940?

    Right now, a fraction of a gram of antimatter is no small amount. But I'm obviously not talking about building this thing now. I'm just trying to see if it's even feasible to have a propellantless rocket. Once something is feasible then it's just a matter of time before it becomes reality...
  11. Feb 11, 2010 #10


    User Avatar
    Science Advisor
    Gold Member

    Right, I'm not saying it's impossible to use matter-antimatter as a source of fuel (ever), I'm just saying that 3250kg of the stuff is no small amount.

    Enriched uranium and anti-matter are completely different. It costs a minimum amount of energy to make the anti-matter because of E=mc^2. This energy must be produced by a method of production that is less efficient than matter-anti-matter annihilation.

    Carrying around 3250kg of anti-matter around with you is also highly dangerous. If just 100 grams of the anti-matter annihilated, you'd be getting an energy output of 2 megatons of TNT (or about 150 Hiroshima bombs).

    Anyways, my point is not to say that this is impossible, but only that the feasibility of a light propelled rocket may be actually higher than getting your hands on 3250kg of anti-matter. So, don't assume that 3250kg of anti-matter is a small detail that you can gloss over.
  12. Feb 11, 2010 #11


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    You have to keep in mind that this is a theoretical maximum. It doesn't factor in how much of the energy is wasted, or used to maintain the containment system needed for the antimatter or even how massive a containment system for that much antimatter would be. The vast majority of that 10000 kg kg could be just the engine and containment system.
    Plenty of people have mentioned it, it is just that, at this point we have no idea how to create and store that much antimatter or make an efficient photonic rocket. So, for now, it remains purely in the realm of science fiction.
    Your numbers are way too high. A fusion rocket could get an exhaust velocity of 100,000 meter/sec. With that, and the mass ratio of the Saturn V to 10,000 kg (and this is assuming that all of the mass of the Saturn is fuel.) you get a final velocity of 2 million kph. This is less than 1/5 of 1% of c.
  13. Feb 11, 2010 #12


    User Avatar
    Science Advisor

    Isn't there a problem here concerning conservation of momentum? If the wavelength of the photons stays constant, then the momentum stays constant as well, so how can there be any net transfer of momentum? If one were using massive particles for this, then if there were any transfer of momentum, there would be a corresponding decrease in the KE of the reflected particles.

    I can see how such a device can work if the photons are absorbed by the target, or even if they were absorbed and re-emitted with lower wavelength, since either of those cases results in a net transfer of momentum.

    Is there something wrong with my analysis here? What am I missing?
  14. Feb 12, 2010 #13
    Absolutely. But we should also keep in mind that the theoretical spaceship I'm talking about is hundreds of times less massive than something that flew 40 years ago. That gives us plenty of room for real-world factors.

    I was still on the same topic with fusion and fission power, in that I had in mind a fission or fusion powered "flashlight" that would propel the spaceship. The effective exhaust velocity for that would be the speed of light. Since we all know that antimatter is still way out of grasp, I wanted to see how the concept would work with more current technology. With fission, all we would really need is this "flashlight" of sufficient power. With fusion, obviously I'm thinking a little bit further ahead.

    Anyway, the whole point of this thought exercise was trying to see if it's possible and feasible to have, sometime in the future, a self-powered, propellant-less spaceship (since in the other thread I was immediately shot down for even having a glimmer of a thought of this concept). My conclusion so far, despite necessary technological advancements and a more thorough analysis, is that it definitely IS.
  15. Feb 12, 2010 #14
    This is all a little above my head because I'm only through physics 2, but light has momentum, therefore there must be some force on the photons to give them momentum (right?), so theres also a force in the opposite direction on the flashlight. Why can't we just put a flashlight on the back of the ship and point it opposite where we want to go?
  16. Feb 12, 2010 #15
    Also, I remember that, in an interview with someone in CERN, it was said that the process they use to make the antimatter is specifically for making small quantities, not designed to be an industrial type process for quantity (because they're just doing research - they don't need mass quantities of it to power something). It is likely, in my opinion (I base this on nothing), that eventually we will have a much better (and cheaper) way to mass produce antimatter.

    Any thoughts?
  17. Feb 12, 2010 #16
    Yeah that's exactly the point of this thread. It will take ALOT of light, like absolutely ridiculous amounts of light, probably on the orders of magnitude greater than all the power plants on earth put together could pump out...but compared to what's possible and what humans have achieved, it's definitely doable.

    Large-scale antimatter production is also doable, although it will require either a source of power orders of magnitude greater and/or cheaper than we currently have (such as fusion), or a way to harness it from space somehow.

    I was wondering another thing...doesn't antimatter annihilation release mainly photons? If so, I'm wondering if there might not even be a need for a stupidly powerful laser/ flashlight. Simple annihilation of mass inside some sort of mirror-lined "combustion chamber" might do the trick...
  18. Feb 12, 2010 #17


    User Avatar
    Staff Emeritus
    Science Advisor
    Homework Helper

    Since the photon changes direction each time it is reflected, it's momentum changes.
    Also, the wavelength is not constant, but gets longer each time it is reflected from the receding spacecraft.
  19. Feb 13, 2010 #18


    User Avatar
    Science Advisor

    Ok, I was a bit imprecise ... my contention was that the *magnitude* of the momentum was constant ... clearly the direction changes upon reflection.

    I guess I understand what you are saying about the wavelength change too ... you are saying there *is* some momentum/energy transfer to the ship, which causes it to increase its outward velocity a little bit, and so there is a small Doppler (red) shift of the reflected photon, correct? I neglected the Doppler shift in my initial analysis ... so I think I understand how it could work now.
  20. Feb 13, 2010 #19
    Yes, the electromagnetic radiation exchange momentum with the spaceship. If this one has a speed v and total reflecting back, then the frequency f2 of light (or in general EM radiation) after reflection becoms:

    f2 = f1*(1 - v/c)/(1 + v/c)

    where f1 is the frequency before reflection.

    Note, however, that not only frequency, but intensity too is decreased because of a moving frame of reference (= after reflection), so the previous formula is not enough to to evaluate the energy/momentum change of the radiation. The light intensity changes with a similar formula (don't remember if the same or with a Sqrt).
  21. Feb 14, 2010 #20
    Solar sails, as previously mentioned, are already a form of powering a craft by light. It uses radiation pressure to propel itself across space. For more info see:

    Bonsor, Kevin. "How Solar Sails Work." 12 October 2000. HowStuffWorks.com. http://science.howstuffworks.com/solar-sail 14 February 2010.
    Last edited: Feb 14, 2010
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook