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

Antimatter mass production

  1. Oct 13, 2015 #1
    If you were to cover the equator of Mercury with solar panels, then used the poles for antimatter production, how much antimatter could you make in a day?

    http://www.centauri-dreams.org/?p=22962
    This site makes some interesting points about how the cost of producing antimatter could be lowered if we had dedicated facilities for making it (instead of as a by-product of high-energy particle physics experiments.)
     
  2. jcsd
  3. Oct 13, 2015 #2

    DEvens

    User Avatar
    Education Advisor
    Gold Member

    How wide a stripe along the equator? So what is the total effective area of panels? Don't forget Mercury is nearly spherical. What efficiency are the panels? So work out the electrical power you would produce.

    How are you producing anti-matter? What efficiency is the process? So that gives you the net power-to-anti-matter rate. And divide that power by c^2 and that gives you the rate of mass production.
     
  4. Oct 13, 2015 #3

    mfb

    User Avatar
    2016 Award

    Staff: Mentor

    CERN has whole accelerators dedicated to the production of trapped antiprotons.
    Their output? Thousands of atoms stored for less than an hour. And no current storage mechanism can be scaled up to milligrams (unless you want to cover the surface of Earth with traps).

    We are a really far away from producing and storing relevant quantities of antimatter. Going to Mercury doesn't help: sure, solar panels give more power per square meter there, but keeping them on Earth and producing more is cheaper by orders of magnitude. It's not like we would run out of space with all the deserts here on Earth.
     
  5. Oct 13, 2015 #4
    Ah. Thanks.
     
  6. Oct 13, 2015 #5

    ohwilleke

    User Avatar
    Gold Member

    One way to deal with the storage problem would be to shoot a steady beam of antiprotons in some direction in outer space (e.g. to a spacecraft that would use the energy created by annihilation on impact with its receiving grid).

    Does anyone have any idea what how much energy is expended per useable anti-matter proton? My intuition is that the process isn't very efficient.
     
  7. Oct 13, 2015 #6

    mfb

    User Avatar
    2016 Award

    Staff: Mentor

    Fast particle beams are actually much easier than trapped antihydrogen. Hitting a spacecraft with them is not.

    At 100 GeV proton collisions with a fixed hydrogen target, production rate is about 0.03 per collision (source), or ~3 TeV per antiproton. Other energies lead to worse ratios. Even if your accelerator is 100% efficient (for 100 GeV, 50% efficiency should be possible, maybe even more, so we can neglect the inefficiency here) and if you can capture all antiprotons (extremely unrealistic), that gives an efficiency of roughly 0.0007.
     
  8. Oct 13, 2015 #7
    I'm not too clear exactly what you have in mind, but assuming you could actually do this from an engineering point of view, that is to get all of your beamed antiprotons exactly on target and without them interacting with anything else during transit ...
    In that case I guess the plot is then to get these to annihilate with ordinary protons aboard the ship, (in a controlled manner), to create energy for propulsion (or something else).
    The annihilation process itself is 100% efficient and I think but am not sure that the result is gamma rays.
    How to make use of the radiation so produced? - I don't know if we have any current technology which could be anything close to helpful.
     
    Last edited: Oct 13, 2015
  9. Oct 13, 2015 #8

    ohwilleke

    User Avatar
    Gold Member

    Antimatter propulsion is pretty much synonymous with beyond current technology. I certainly don't know how one would actually go about using the gamma rays produced, but, given our near perfect understanding of QED, if it is possible, we know enough to make it happen or could with some focused research.
     
  10. Oct 14, 2015 #9

    mfb

    User Avatar
    2016 Award

    Staff: Mentor

    If you have a way to focus a particle beam onto a spacecraft, there is no need to use antimatter. Matter impacting at high speed releases energy (and momentum) as well, and it is much easier to get larger amounts of matter.

    Electron-positron annihilation leads to gamma rays of up to 511 keV, hadron annihilations lead to gamma rays of roughly 50,000 keV, muons and a few neutrinos.
     
  11. Oct 14, 2015 #10
    I ask because in the book I'm writing, most interstellar spacecraft have Alcubierre-White warp rings. The more power you put into them, the faster you can go. Most of the spacecraft with warp rings use fusion plants to get power, but I was planning to have couriers use antimatter so they could move much faster than anything else. It's not for conventional propulsion.

    Whilst I'm willing to use artistic license, I'd prefer to base it on real science if possible.

    Current solar cells have a max efficience of around 47%. Up that to 50% because it's the future.
    There are plans to put a 250 mile wide "Luna Ring" of solar panels on the Moon. This would generate 14,000 terrawatts of power. Mercury, meanwhile, has a circumference of 15,000KM (5,000 more than the Moon) and orbits at roughly 0.4AU, getting seven times more sunlight than we do out at Earth's orbit. I'm guessing 150,000 terrawatts output for the Mercury solar panel ring.

    I can't tell what the efficiency or the power requirements are from that. It tells me that the process could produce 87955200 antiprotons in a day, but nothing more. The best I can find on Google is that the RHIC collider when firing gold atoms uses about 9 watts.
     
  12. Oct 15, 2015 #11

    mfb

    User Avatar
    2016 Award

    Staff: Mentor

    1018=1000000000000000000 antiprotons per second, not 1018.
    The efficiency of ~0.0007 quoted above is a hard limit for accelerator-based production.

    Why do you want to put the solar panels on a planet? That lowers their efficiency compared to a structure in space.

    The collision power might be 9 watts, the accelerator is probably using megawatts of power.
     
  13. Oct 15, 2015 #12
    Putting them on a planet means you can use in-situ resources to build and repair the panels and production facilities. A lot of resources would be needed.

    I think the 1018 number came from me copy and pasting it. I didn't notice.
     
  14. Oct 15, 2015 #13

    mfb

    User Avatar
    2016 Award

    Staff: Mentor

    I don't think Mercury provides an advantage compared to asteroids. The asteroid composition could be even better.
     
  15. Oct 15, 2015 #14
    If there were any humans on-site (and the Luna Ring plans say there would be) then they would have constant gravity. Presumably by this point there would be at least some colonies on Mercury anyway, probably at the poles where there's water ice. To do the same for a space installation, you would need a spin ring. Though that wouldn't give them gravity when they were actually working on the panels or the robots.
     
  16. Oct 18, 2015 #15
    Current production costs for antimatter requires an energy budget of about 1,000,000,000 to 1. That's 1 billion times more energy to create antimatter than what is released by antimatter.

    Realistic production efficiencies would improve that by an order of magnitude, but that's about it until we have some breakthrough in science that we don't know about.

    Hopefully, you won't be using solar panels from Solyndra, but best current panel efficiencies are 24%. However, panels rapidly degrade over time and with that much solar flux (bare in mind that there will be a much, much broader spectrum of solar radiation than the wavelengths used for PV cells attacking the substrate) the efficiency will rapidly diminish and probably require frequent replacement.

    Mercury is tidally locked, so you only need to work on the day side, but it's hot at 420°C. That's hotter than my soldering iron and there are no electronic components that can operate in that environment to date. So, a whole new materials technology needs to be created there, too.

    I'd say that you just have to accept a miracle happens here with regard to antimatter production because there is no known mechanism to produce it in bulk at this time. Getting into details like that is just waddling into shark infested waters.

    It's bad enough that producing antimatter is a stretch of the imagination, but if you start embellishing on how you produce it you will just open your story to more and more bad science, in my opinion.
     
  17. Oct 18, 2015 #16

    mfb

    User Avatar
    2016 Award

    Staff: Mentor

    Not in a 1:1 relation. It still has days and nights.

    The temperature rapidly drops with depth in the upper layers (towards some equilibrium over the day/night cycle), using the surface as heat sink might work.
     
  18. Oct 18, 2015 #17
    I stand corrected!

    It rotates three times for every two of its solar years. So, you end up with a lot of down time for the panels, halving their production efficiency.
     
  19. Oct 19, 2015 #18
    Fair points. Mercury is too close. As for the inefficiency of making antimatter, that's not really an issue. Antimatter is a battery, rather than a fuel source. If you have the power to spare for making it, and you need something with very high energy density, why not make it?

    Since I brought up the Luna Ring concept, I've been thinking about it more (although I was thinking about it because you'd said "why put the panels on a planet".) What if, having built a much larger solar array elsewhere to power Earth (even at current power requirements, the Luna Ring would be a terrawatt short, and power requirements are only going to increase) the Luna Ring was left redundant. So we have a load of solar panels just lying around on the Moon, not being used, and someone in power decides that their pet antimatter project could make use of them.

    Using the figure of 14 terrawatts, how much antimatter could we make? Well, if we use the billion to one figure, we get 14,000 watts worth of antimatter. An order of magnitude more efficient and you get 140,000 watts.

    H'mm. I suppose I should also try and find out how much power a warp drive would take to run in the first place.

    (EDIT: Actually, according to
    http://www.utne.com/science-and-technology/solar-power-from-the-moon-luna-ring.aspx
    Luna Ring would produce 220 terrawatts per year.)
     
    Last edited: Oct 19, 2015
  20. Oct 19, 2015 #19

    Drakkith

    User Avatar

    Staff: Mentor

    Your units don't match. The unit Watt is a measure of power, not of energy. One watt is equal to 1 joule of work performed/energy consumed in one second. So a 100 watt light bulb uses 100 joules of energy per second, 6,000 joules of energy per minute, and 360,000 joules per hour.

    1 joule's worth of energy could create about 6.65 x 1015 antiprotons, for a total mass of 1.11x10-11 kilograms.

    So a terrawatt of power, over 1 years time, could produce about 3x1020 joules, which could create around 6.65 x 1035 antiprotons, or 3,330,000,000 kilograms worth of antimatter.
     
  21. Oct 19, 2015 #20

    Ryan_m_b

    User Avatar

    Staff: Mentor

    One reason to not make it is that it would be extremely dangerous in large quantities. A few grams of antimatter would release as much energy as the Hiroshima bomb. If you want to store kilos that's a huge bomb (~20kg would be equivalent to the most powerful nuclear bomb ever built). Mass storage of antimatter would have to be really good and really far away from anything/one important.
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook




Similar Discussions: Antimatter mass production
  1. Anyone play Mass Effect? (Replies: 26)

Loading...