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

To destroy an asteroid

  1. Jan 9, 2005 #1
    How many ways are there to destroy a hostile asteroid coming towards the earth? What are they?
  2. jcsd
  3. Jan 10, 2005 #2
    Excellent question.

    At present we have NO way of destroying or even deflecting an incoming asteriod.

    I know what you're thinking, "But we have TONS of nukes on missiles sitting in silos all over Montana." Well, you're right, but we do not have any nuclear equipped missiles - be it the Minuteman series or the Trident series - that are capable of even Earth orbit.

    The most likely and quickest chance would be to back-fit an old Titan-2 (which used to carry nukes when they were heavier... before we got better at making them smaller and lighter) which the Air Force still has sitting around.

    Another option would be to use a Delta-II or Delta-IV, but these were never designed to interface with a nuke.

    Even if ANY of these were to be possible, government bureauocracy would probably gum up the works as it would involve transport of a nuclear bomb to facilities unequipped to recieve and deal with them. Not to mention the fact that other countries would be going insane that we would be launching a nuke on a HUGE rocket. Plus, the US would be so scared that they wouldn't do it until it was too late for fear of ticking anyone off. So I wouldn't count on the US.

  4. Jan 10, 2005 #3
    I don't think people will care much for such systems until an asteroid causes a Tsunami-scale disaster. Statistically, I would expect many localized disasters to happen before any Armageddon-scale asteroid presents itself.
  5. Jan 10, 2005 #4


    User Avatar
    Science Advisor

    Just an ad-on to LB's post; even if we could get a nuke to the asteroid, hitting and asteroid with a nuke will not work, no matter how many movies you've seen it in. There are several methods being explored, including space-mirrors to reflect sunlight and heat up one side of the asteroid, in hopes of using out-gassing as a propelling to nudge the asteroid off course, or simply throwing a big dust cloud in the asteroid path to deflect it. These methods have the advantage that they would be absolutely useless as weapons, which makes them much safer technologies to have sitting in readiness for launch.
  6. Jan 10, 2005 #5
    Perhaps, we should just capture it in our orbit and study it???
  7. Jan 10, 2005 #6
    Use conservation of momentum and impulse to find the force required to put an asteroid in orbit, and you'll see that this isn't an option for significant asteroids. I know I should try calculating it for myself, but I'm tired and have to get up 6 AM for school tomorrow morning.
  8. Jan 11, 2005 #7
    It should be possible is to use nano-robotic devices to break the asteroid up into lots of tiny pieces. It's also possible to eject the pieces away from the asteroid as they are being chipped away, thus producing a small amount of thrust to nudge the asteroid's orbit.

    edit: actually, we may be able to use MEMS for this, and we can already build those now. We just need a suitable design...

    edit2: on second thought, power supply may be a problem..
    Last edited: Jan 11, 2005
  9. Jan 11, 2005 #8


    User Avatar
    Science Advisor

    Yes, a reaction driver is one of the strategies being investigated. However, I do not think it is a very likely one, since it requires a fully automated system to continue working for many years on its own.
  10. Jan 12, 2005 #9
    Depends on what you mean by destroy. I'm not sure what the energy needed would be to completely pulverize an asteroid. I'm not sure we have that much available stuff (technology) on hand to even deflect an asteroid at this point. Allot would depend on how far along it's path towards the Earth it is (i.e. when we detect it). I'm of course going on the assumption that we want to destroy this thing because it's threatening our existence? If detected early enough I suppose we cauld deploy something to slowly over time change it's orbit to miss the Earth. Maybe deflect it towards impact with another planet? Would a series of properly located nuclear blasts be able to do that? I've heard Allot also depends on what the make up of the asteroid is, also has something to do with what you would possibly use to deflect/destroy it.
  11. Jan 12, 2005 #10
    Well, at the basis, this could be as simple as a bunch of MEMS drilling/grinding devices. Although you'd want something more complex for a practical system, perhaps something more like MEMS versions of those huge machines they use for tunnel boring...

    As long as you have a very large number of machines working with reasonabe reliability, it doesn't matter if e.g. 10-20 % of them fail. But to be honest, I'm not entirely sure how well this will work.. I'll have to run some calculations later.
  12. Jan 12, 2005 #11
    The calculations, as promissed:

    Ok. I'll assume we're taking the approach of breaking the asteroid into lots of small pieces. The reasons for this are the following:
    1. It would take much more drilling work than is required to alter the asteroid's orbit (since more matter needs to be chipped away). Hence this will give a good upper limit to the time required for the operation.
    2. The required systems can be much simpler, since no matter needs to be accellerated for thrust.
    3. It is simpler to analize, since there are no orbital considerations.

    I will also assume nanotechnology is used, since I am more familiar with the area than with MEMS. However, the two are quite similar, and the calculation below should be easy to adapt for the MEMS case. Further, I am assumming an asteroid and not a comet, since comets are generally weaker and AFAIK also smaller than (the largest) asteroids, so they should be easier to deal with.

    First, some raw data and facts:

    - The typical operating frequency of a (high performance) nano-device, is on the order of 1 GHz or more [1,2].
    - The typical diameter of a complex nano-robotic device is (expected to be) around 100-1000 nm.
    - The expected structural materials for nano-robotic devices are carbon or silicon based, most likely diamondoid.

    - The density of diamond is 3.51 gm/cm^3 and that of graphite is 2.26 gm/cm^3 [3]. These correspond to 3510 and 2260 kg/m^3, respectively.
    - The density of silicon is 2330 kg/m^3 [4].

    - The largest known asteroid (or maybe 2nd largest [6]) is Ceres, and has a mass of about 4.35e-10 solar masses [5], which is about 8.66e20 kg (solar mass value taken direcly from google as 1.99e30 kg).
    - Ceres has a diameter of about 622 miles [7], or around 1000 km (conversion done with http://www.onlineconversion.com).
    - The largest known near-earth asteroid is 1036 Ganymed, about 41 km in diameter [8].

    - The mass of the NEAR Shoemaker probe was 818 kg [9].

    - The cell parameter of an iron crystal is 286.65 pm [10], which is about 2.87e-10 m.
    - The cell parameter of a nickel crystal is 352.4 pm [11], which is about 3.52e-10 m.

    Given the above data, I will assume the following values for the calculation:

    - Nanite operating frequency: Fn = 5e7 Hz = 50 MHz (a fairly modest value).
    - Nanite radius: Rn = 2.5e-7 m = 250 nm (500 nm diameter).
    - Nanite density: Dn = 3000 kg/m^3.

    - Asteroid radius: Ra = 1e5 m (approximately worst-case scenario, 200 km diameter).
    - Asteroid average cell parameter, assumming M-type (pure nickel/iron), which is more dangerous: La = 3e-10 m.

    - Average functioning nanite mass (allowing for medium failure rates): Mn = 500 kg.

    - Desired particle radius: Rp = 0.5e-3 m = 1 mm (small enough to be harmless, except perhaps for satellies).

    For simplicity, I'll assume the nanites, asteroid and so on are all perfect spheres. The equtions are then:

    Volume of a sphere:
    (1) V = 4/3 * pi * R^3

    Surface area of a sphere:
    (2) S = 4 * pi * R^2

    Area of a circle:
    (3) A = pi * R^2

    Number of particles that need to be produced (a - asteroid; p - particle):
    Np = Va / Vp = (4/3 * pi * Ra^3) / (4/3 * pi * Rp^3)
    (4) Np = (Ra/Rp)^3

    Volume of matter that needs to be grinded away to break up the asteroid into the desired particles - number of particles times particle surface area times nanite radius:
    Vm = Np * Sp * Rn = (Ra/Rp)^3 * 4 * pi * Rp^2 * Rn
    (5) Vm = 4 * pi * Rn * Ra^3 / Rp

    Average number of active nanites - total mass over nanite mass:
    Nn = Mn / (Vn * Dn) = Mn / (4/3 * pi * Rn^3 * Dn)
    (6)Nn = 3 * Mn / (4 * pi * Dn * Rn^3)

    Assuming each nanite removes one atomic layer per cycle, over an circular area with radius Rn, the volume-per-second of matter removed by all the nanites is:
    VPS = An * La * Fn * Nn = pi * Rn^2 * La * Fn * 3 * Mn / (4 * pi * Dn * Rn^3)
    (7) VPS = 3 * La * Fn * Mn / (4 * Dn * Rn)

    And finally, the time it would take to finish the job:
    t = Vm / VPS = {4 * pi * Rn * Ra^3 / Rp} / {3 * La * Fn * Mn / (4 * Dn * Rn)}
    t = 4 * pi * Rn * Ra^3 * 4 * Dn * Rn / (3 * La * Fn * Mn * Rp)
    (8) t = 16 * pi * Dn * Rn^2 * Ra^3 / (3 * La * Fn * Mn * Rp)

    Using (8) with the above values, we get:
    t = 8.38e8 s = ~9700 days = ~26.57 years

    Although this sound like much, this is actually not that bad considering the size of the asteroid and the comparitively small total mass of nanites. At this rate, it would take only about 80 days to decompose 1036 Ganymed if it was of similar composition (and roughly spherical, or of equivalent volume - what exactly does an asteroid "diameter" signify, anyway?). Changing its orbit would take less time, since only a fraction of the mass would need to be chipped away.

    Btw, considering the final form of the equation, it is easy to calculate the result for MEMS, we just need to know the MEMS' radius, density, and operational frequency... and also how thick a layer of matter they can remove per cycle (i.e. the equivalent value for La). Anyone has some sample values?


    [1] http://www.nanomedicine.com/NMI/2.3.2.htm
    [2] http://www.nanomedicine.com/NMI/2.4.1.htm
    [3] http://hyperphysics.phy-astr.gsu.edu/hbase/minerals/diamond.html
    [4] http://www.webelements.com/webelements/elements/text/Si/phys.html
    [5] http://aa.usno.navy.mil/ephemerides/asteroid/astr_alm/asteroid_ephemerides.html
    [6] http://www.planetary.org/html/news/articlearchive/headlines/2001/2001kx76.html
    [7] http://www.seasky.org/solarsystem/sky3k.html
    [8] http://www.nasm.si.edu/research/ceps/etp/asteroids/AST_near.html
    [9] http://space.skyrocket.de/index_frame.htm?http://space.skyrocket.de/doc_sdat/near.htm
    [10] http://www.webelements.com/webelements/elements/text/Fe/xtal.html
    [11] http://www.webelements.com/webelements/elements/text/Ni/xtal.html
    Last edited: Jan 12, 2005
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?

Similar Discussions: To destroy an asteroid