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

Questions about fusion (from a high-schooler)

  1. May 17, 2008 #1
    Hi! I have a major project coming up this next Thursday and am doing it on nuclear fusion used as power source. I've been reading the topic above (Fusion Summary?) and it is quite a read. I'll tell you, most of it is completely over my head, but it is one of the most interesting things I've read in a long time. (I'm glad the forum is civilized, as it seems to be a somewhat controversial issue.) :)

    I am in high school (in an Honor's Physics class) so try to keep it somewhat simple. I'm grossly ignorant compared to many of you, so bear in mind that there are a lot of things I don't know.

    Here goes!
    • There is not substance on Earth that can contain it, correct? It actually requires containment? I'm reading up on the different types right now.
    • It has no "true' waste products, meaning that everything that is released can be used?
    • The only thing more powerful is matter + anti-matter annihilation?
    • It isn't a chain-reaction, like fission. It needs enormous amounts of energy and pressure to start, but is that same amount needed to keep the atoms fusing?
    • How many moles/grams would be needed of the reactants to produce a substantial amount of energy? Can I figure this out by myself by adding up masses of D + T -> He(4) + N and finding the mass defect (right word?), then using E=mc^2 to find the energy? For some reason, I don't think I can use the standard, weighted-average masses in a periodic table. And where can I find the masses of isotopes? Is the mass of a He(4) made from fusion different from a regular He(4)?

    I'll be posting many more questions over the coming days, but these are the first few. Thanks!


    P.S. I hope it doesn't this doesn't come off as cheating; I won't ask you to research anything. Heck, if an answer is better explained in a paper, link it! And I will only ask about things that I don't understand or can't find an answer for. Thanks again!
    Last edited: May 17, 2008
  2. jcsd
  3. May 17, 2008 #2


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    Hello and welcome to PF !

    You can start reading about it on the level you ask for at wikipedia:

    and at the newly planned facility ITER http://www.iter.org/ (using magnetic confinement).

    I think it will already answer most of your questions.

    Given that one needs hundreds of millions of degrees, it is fairly obvious that no solid material container can be used (it would be vaporized immediately - or better, it would make the temperature of the fusion mix drop and the reaction would stop).

    The "waste" of the core reaction is He-4 gas, which you can use to inflate balloons for the kids :smile: But there will be radioactive waste products, but much less so than from a fission reactor (which, BTW, are much less of a true problem than is "generally known").

    I wouldn't say that the next most powerful thing is matter-anti-matter annihilation. It is just a rather energetic nuclear reaction. Each D + T delivers something like 17 MeV, which is pretty much: it means about 3.5 MeV per atom mass unit, that's more than 4 times better than a fission reactor can do (which produces about 170 MeV per uranium or plutonium atom, but which weight 40 times more than a D and a T). It is about 10 million times more than burning hydrogen.

    It means that 1 kg of D + T mixture will give you
    17 10^6 x 1.6 10^(-19) x 6.02 10^23/ 5 x 1000 = 324 T J (terra joules) = 3.24 10^14 J.

    Now, that means 10 MW during 1 year of thermal power. If 1/3 can be converted to electricity (thermal cycle) then it means that a 1 GW power plant (like your average nuclear power plant) would need about 300 kg of the mixture to do so.

    So with 300 kg of the mixture you would fuel a 1 GW electric power plant for a year while nuclear fission needs about 20 tons of enriched uranium, because current reactors don't use it well, and a coal fired plant would need about 3 MILLONS of tons of coal to do the same. So we find back our factor of about 10 million between "chemical" and "nuclear" energy. That said, we are still a long long way before this can practically be achieved...

    It is not a neutronic chain reaction like fission, but it is a chain reaction like normal fire: the idea is that the energy of the first reactions will heat up the mixture enough to allow for the next (well, it's more complicated...)

    BTW, helium-4 is helium-4, no matter where it comes from...
    Last edited: May 17, 2008
  4. May 18, 2008 #3


    User Avatar

    Staff: Mentor

    Just adding to what vanesch has mentioned, there are many types of fusion reactions, just like there are a number of different chemical reactions. However, perhaps the easiest fusion reaction to produce and generate energy is D +T -> He-4 + n. Unfortunately, the bulk of the energy produced (14.1 MeV) goes to the neutron and 3.5 MeV goes to the alpha particle which eventually combines with two electrons to form He-4.

    The energy from the neutron must be converted to thermal energy (heat) in a blanket outside of the fusion reaction, and some of that energy will then be used to produce electricity. The energy of the alpha particle is imparted directly into the plasma by electromagnetic processes (coulombic and magnetic interactions) and perhaps some useful electrical energy may be obtain if and when the alpha particle leaves the plasma and is collected. One consequence of the neutrons from fusion is the 'activation' of the surrounding structural materials, and in that sense, there is some 'waste' that people will eventually have to deal with.

    Antimatter-matter is certainly a much more powerful reaction, but there are no significant sources of anti-matter on the earth that we can use as an energy source. It takes a lot of energy to artificially produce anti-matter in the form of anti-protons, but it requires much more energy than we can recover.
  5. May 18, 2008 #4
    Hi! Thank you for the welcome, but the pleasure is mine. This place is cooler than boiling water in low pressure, lol.

    Great answers, those were just what I was looking for.

    I was looking at the ITER site and it seems to have three steps instead of two for the D+T reaction. It goes D+T -> He-5 (whoa!) -> He-4 + N. I've done some research and apparently the He-5 atom is extremely unstable and almost immediately decays to He-4.

    Deuterium = 2.01355321270 u
    Tritium = 3.0160492 u
    He-5 = 5.01222 u
    D + T = 5.0296024127 u

    So the energy, if I'm thinking this correctly, is the lost mass (converted to energy) when you fuse it into He-5 because D+T>He-5. So where is this energy, exactly? And where does the location of that energy change when it decays into He-4 + N?

    And when they say energy is released, it is more like the energy is in the Helium and neutron, right? More so in the neutron.

    The nuclear "waste" isn't actually the products, but what the products hit, right? I think I read somewhere that after a small fusion reaction (can't remember which test site?) that the site was radioactive for a year and tests could only be done remotely?

    That calculation you made, to find out the energy, what do those numbers represent?


    When you say "blanket", what does this mean? I just want to clarify one thing: confinement is the idea of us containing the energy release (neutrons and alpha particles) into something that we can harness and collect, right? It isn't what makes D+T plasma, right? Like we get some tritium and some deuterium, make them do their stuff (lol), and then use either magnetic containment and inertial containment to keep them safe? Or do we use magnetic containment and inertial containment to make them into plasma? And then use something else to keep them safe?

    Sorry for all the questions, I'm kind of asking while I'm learning, which probably isn't a good idea, but I just want to know before I go further in.


  6. May 18, 2008 #5
    Kind of an odd question: plasma is a gas, right? How do we make plasma? For example, from the deuterium gotten from electrolysis? Like, what needs to be done to make it a plasma? And why do you need plasma for fusion? Or is plasma only needed for magnetic-containment fusion reactors? I asked the same question in my previous post, but mention in one more time: how do we heat plasma to the temperatures needed for fusion to occur?

    Last edited: May 18, 2008
  7. May 18, 2008 #6


    User Avatar
    Science Advisor

    Some say plasma is a gas, others say it is a seperate state of matter (solid, liquid, gas, plasma). Basically, what makes it a plasma is when the electrons stop orbitting nuclei and go flying off on their own. There are probably several ways to do this, but the one I always hear about is that one simply raises the energy state (the temperature) of the element untill the electrons dissociate.

    Because a plasma is basically just a swarm of protons with no electrons attached, all plasmas are strongly magnetic (positive). This is extremely handy for magnetic containment, but it is not the only advantage for fusion. In order to bring two atomic nuclei into direct contact with one another (close enough for the strong nuclear forced to take over), one must find some way to get past their electron-shells. Unless those shells can be stripped off , in which case the Bear nuclei can be forced together directly.

    As for how the deuterium and tritium are heated to high enough temperatures to achieve a plasma state, it is my understanding that this is a multistep process using different techniques, including friction, compression, and bombardment with radio waves.
  8. May 18, 2008 #7
    Thanks for the reply!

    OK, so first it has to be heated to remove the electrons thus turning it into plasma. Then, it again must be heated to fuse. I found a good explanation of heating the plasma at splung-dot-com.

    I think you were spot on. :) I think the same things are used to heat them into plasmas and then to heat them to fuse. Or is it?

    So, I think I've got the basic idea down:

    Obtain tritium from Li + N and deuterium from heavy water + electrolysis. Heat them to transform them into plasma via ohmic, neutral beam, or RF. Since they are positively-charged, put them in an magnetic confinement torus (a tokamak, right?). This keeps the plasma from touching the walls and keeps it circling the magnetic field lines. So, it is basically floating, right? Heat the floating plasma with ohmic, neutral beam, or RF. Now that is heated sufficiently, fusion can occur. You have a release of helium (or is it alpha particles?) which needs time to cool before it is released. You also have a release of highly-energetic neutrons which hit the walls (the blanket!). The walls are internally cooled and you can now transfer the heat. The neutrons (after transferring the energy) can now combine with lithium to make more tritium. And now there is a happy cycle. :)

    Correct me if I have said anything incorrect?

    A few more questions: where does pressure come in? Or do you not need pressure if there is enough heat? Why isn't fusion happening right now? That is badly worded, but why can't this be happening today? It seems like everything should be able to work. Why do people say fusion can't be contained? It kind of seems like it has been worked out? Do we just need time to set this all up?


    P.S. Nice sig, lol.
    Last edited: May 18, 2008
  9. May 18, 2008 #8


    User Avatar

    Staff: Mentor

    Plasma refers to a highly ionized gas, in which many or most of the electrons are free from the nuclei, and in that sense, it is a mixture of an electon gas and an ion/nuclei (protons, deuterons, tritons, alpha particles, . . .) gas. Ions and electrons can be confined magnetically, but neutral atoms will 'leak' through the magnetic field.

    To ionize the gas, it can be heated by a variety of methods, such as microwave (radiowave), neutral beam, magnetic compression and ohmic heating.
    See - http://en.wikipedia.org/wiki/Tokamak_design#Plasma_heating

    In a reactor, the blanket surrounds the plasma, usually to absorb neutrons and thermal energy from the fusion reaction.

    Pressure arises from the kinetic energy of the electrons and ions/nuclei - basically, P = nkT where n is the ion/electron density, k is Boltzmann's constant, and T is temperature.
    Last edited: May 18, 2008
  10. May 18, 2008 #9
    Excellent, thank you. And this is the same heating that is done to raise the plasma to 10 million Kelvin, right? That is what I've read, just want to confirm.

    Now, from all the articles I've read, few seem to mention pressure. Doesn't fusion require extremely high pressure? Or is that already achieved, somehow? At ITER, they talked about how there would be difference in pressure between the minor and major diameters, but it still didn't address how they would get such high pressure.


  11. May 18, 2008 #10


    User Avatar

    Staff: Mentor

    The high pressures of the plasma are a conseqence of the high temperatures of the ions and electrons, and their density. One wants to get as high as possible density, because the reaction rates are proportional to the density squared, BUT pressure increases proportionally to temperature and particle density. The magnetic field must be strong enough to contain the plasma ( a limit on the magnets ) and the structure must be strong enough ( a material strength limit ) to support the forces imposed by the high pressures.
  12. May 18, 2008 #11
    Ahhhhh....So increasing one increases the other, so to speak? I think I get it.

    Well, I'm about finished with the section on magnetic confinement (my main area), so I'm going to do a few slides on inertial-laser confinement and then maybe one on inertial electrical. I have more time than I thought, so I might do a quickie slide on cold fusion.

    Thank you so much for all the help you've given me so far!

  13. May 19, 2008 #12


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    That's in fact the same! Alpha particles are helium-4 nucleae. But in the beginning of radioactivity, people hadn't figured that out yet, they had simply noted that there were 3 kinds of "radioactive radiations", which they called "alpha", "beta", and "gamma" (the 3 first letters of the greek alphabet). It turned out that "alpha" were helium-4 nucleae, "beta" were electrons, and "gamma" were photons, but the names stuck.
  14. May 19, 2008 #13


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    You shouldn't bother too much about this "matter into energy". In fact, when you burn a piece of wood, you also turn "matter into energy", only, the difference it too tiny to be seen. Just think of it more classically: it takes more energy to break a helium-4 and a neutron, than it takes break a tritium and a deuterium into their basic constituents (2 protons and 3 neutrons). As energy is conserved, that means that if you turn a tritium and a deuterium into a helium-4 and a neutron, you have some "redundant energy", and that goes into the energy of motion (kinetic energy) of the product particles. This is exactly the same as what happens when you burn wood: the molecules in the end (mainly CO2 and H2O) are more difficult to separate in the atoms than the molecules in the wood and oxygen. Hence the CO2 and the H2O have extra energy (but this time it is not in their motion, but rather in their vibrations) when they are the product of burning.

    17 10^6 x 1.6 10^(-19) x 6.02 10^23/ 5 x 1000 = 324 T J (terra joules) = 3.24 10^14 J.

    17 10^6 is 17 MEGA electron volt

    1.6 10^(-19) is the energy in Joule of 1 electron volt (unit conversion)

    6.02 10^23 is avogadro's number: the number of atoms in A gram of the substance, where A is the atomic number (1 for hydrogen, 2 for deuterium, 3 for tritium, 4 for helium-4,...)

    5 is the total atomic number for a tritium and a deuterium couple

    1000 is the number of gram in a kilogram.

    So 6.02 10^23/5 is the number of D-T couples in 1 gram of the mixture.

    this x 1000 is the number of D-T couples in 1 kilogram of the mixture

    17 10^6 x 1.6 10^(-19) is the energy, in Joule, liberated by the fusion of one single D-T pair.

    everything together is the energy, in Joule, liberated by 1 kilogram of the mixture.

    BTW, 1 Joule delivered in 1 second is 1 Watt.

    In a year, there are 365 x 24 x 3600 seconds (number of days, times number of hours per day, times number of seconds in one hour).
  15. May 19, 2008 #14


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    Yup, that's essentially it.

    As Astro said, the pressure is pretty high, but also the amount of material in the reactor is very very small!

    Well, there are several problems. First of all, electric heating (sending a strong electric current through the gas) becomes less and less efficient as temperatures get higher and higher. Hence, RF and neutral beam heating, because with the electric current we cannot get it hot enough. The reason is actually that the electrical resistance of a plasma drops strongly with temperature, so at high temperatures the plasma almost looks as a superconductor, and doesn't heat anymore.

    And now the difficulties come in: all these extra heating techniques render the plasma instable in the magnetic field. It starts to wobble and vibrate, and doesn't stay well in place. People have found methods of trying to stabilize it again, but usually they can only contain it for a short time (seconds to minutes). During this time, SOME fusion can occur.

    And this is the fundamental problem: we have a small amount of plasma in the reactor, which can stay there only for a short time before it becomes unstable, and which doesn't undergo as much fusion as one would like during that time. And it turns out that the energy we can extract, up to now, is less than all the energy we put in (with the current, and the RF heating and so on). So, yes, fusion is happening, but no, it is not yet a working power source. ITER should be the first machine in which one DOES gain net energy in the process in a reliable way. http://www.iter.org/a/index_faq.htm
  16. May 19, 2008 #15
    Wow, thank you so much. That information is great, I can't imagine getting that somewhere else. Those are the questions that articles seem to gloss over. That equation is definitely going into the PowerPoint, lol. :)

    I have one quick question about inertial laser confinement. How exactly are they going to contain the fusion? Or how will they actually absorb the energy? In magnetic, they have the blanket, but how do they propose doing it laser inertial? From the video, it seems like the entire test site seems to blow up, lol (from NIF). The same goes for Z-Pinch.

    I mean, the site precludes it becomes a star (wow), but I'm fairly certain stars don't explode from the fusion reactions because of their enormous gravity and that doesn't sound possible in this instance.

  17. May 19, 2008 #16


    User Avatar

    Staff: Mentor

    Well, the fusion pellet is very small compared to the volume of the chamber in which the fusion pellet ignited. One concept would put a layer of liquid lithium, like a water fall, cascading down the inner surface of the ignition chamber. The hot liquid would be collected and pumped to a heater exchanger where the heat (thermal energy) would be transferred to another working fluid which would in turn be used to drive a turbine/generator set.

    Similar a blank could be used on the outside of the ignition chamber, but then the first wall gets a significant amount of radiation and high energy particles, and it would eventually become brittle.

    I have not seen a magnetic-based energy extraction system for an ICF system however.
  18. May 19, 2008 #17
    Interesting, interesting.

    Ah, have the lithium on the walls and go down a drain near the bottom. Wouldn't the lithium absorb the energy from the neutrons and actually transform into tritium? Or does that not happen with liquid lithium? It is a concept, so... I'm almost finished with the presentation, only a quick slide on cold fusion and inertial electrostatic confinement is left!

    So, a major con of the neutronic fusion reactions is the release of these radioactive neutrons. Did you hear about the Z Machine? 3.7 billion Kelvin! Holy Cheez-Its! I hear that is enough for aneutronic fusion. I'm hoping that this is in my lifetime, because I'm actually pretty excited. About physics!

    The "Fusion Summary" thread is starting to make more and more sense, freakishly. :)

    Thank you, again.

  19. Jun 12, 2008 #18
    Ok first i just want to point out that you must be an exceptional student and very smart to be understanding this as a high school student... i would suggest looking at engineering schools for college, maybe MIT :-)

    I believe these guys might know this stuff a little more in depth than i do and they might have already answered some of these questions but i did not see...

    basically thermodynamically temperature and pressure are directly related so by increasing the temperature you do increase the pressure and visa versa... but they also increase the pressure with the electro magnets forceing the plasma to be more and more dense... so think of it this way... you are compressing all the plasma making it more and more dense and all the atoms closer together... as the pressure increases so does the temperature... so you got all these protons freaking out and being forced towards each other... when they collide they have so much energy that they fuse and create helium... now this shows einsteins genius... he knew this all before we could even begin to try this... helium weighs less than the two hydrogen atoms... so some of the mass dissapeared... or did it? energy and mass are forms of the same thing so the missing mass is converted into energy.... E=mc^2 so even a little bit of missing mass equals a huge amount of energy.

    i have a theory on this subject... we require a huge amount of energy to pressurize this plasma and keep it from touching the outer walls with the electromagnets... the sun solves this problem of pressurizing and containing it with one simple (really complex haha) force... GRAVITY... if we can somehow manipulate gravity it could reduce the need for that amount of electrical energy to contain and pressurize it and then the energy obtained could be more than the energy put in and we would have self sustaining fusion just like the sun! im not sure we will be able to contain self sustaining fusion without gravitational manipulation.

    another thing about antimatter which is the super future of energy... is one like they said it is extremely hard to create and usually occurs random in supercolliders... the next problem is that once we create it we need to contain it... if you have a positively charged electron and a negatively charged proton you can contain them because they are charged but once you put them together to make actual anti MATTER instead of just anti-particles, you can no longer contain it because it doesnt have a charge anymore and once it touches matter it converts to %100 energy.
  20. Jun 12, 2008 #19
    Hehe, thank you. Oh, I'm sure it is much more in-depth than this, though. I feel like I've only touched the surface. Still very excited and a bit let down that many people quickly ignore nuclear power, I mean, fission is still really nice! :)

    I think I get it: higher temperature, which means faster moving particles, means higher pressure. This is because P = F/A and when there is more F (F=ma, more acceleration due to heat), there is a higher P. A shame he died so young, I wonder if he knew how much he helped?

    By the way, the PowerPoint was a success, 200/200! :D I really couldn't have done it without you guys, thank you so much! I wish there was a way to repay all this help. I could upload the PowerPoint, if you guys wanted?

    I can see what you mean: I mean, most, if not all, the energy is going just to initiate and contain fusion; if there was a "cheaper" way (gravity!), that would be fantastic. But, in my mind, however, seems a bit off in the distance.

    Now, antimatter: that would be a wicked Physics project. If I had to do another, that would totally be it.

  21. Jun 12, 2008 #20


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    Just a remark about "anti matter power production", that leads nowhere, even if we could produce copious amounts of anti matter and confine it (one can now, but in tiny quantities). Indeed, if one has to *produce* the anti-matter, then by conservation of total energy, this means that one has *already spend* at least the energy contained in its mass (E = mc^2), and also in the necessary mass of normal matter that one must create at the same time (conservation of baryon and lepton number). So at most, if everything were 100% efficient, you could by matter anti-matter annihilation, just get out of it what you put into its production. No gain.

    This would be similar as if we were to take helium for starters, split it in two deuterium nucleae and then fusion them again. We would spend at least the amount of energy gained in the fusion, to split the helium (with an accelerator or I don't know what). It is only because deuterium is *freely available in nature* that we can "just fuse it" and gain energy. And it is because uranium is *freely available in nature* that we can fission it and gain energy. But there's no anti-matter freely available in nature. So we cannot gain energy from it. If we have to make it first, then the laws of nature forbid us to gain any energy from it. With deuterium or uranium, we can because the stuff is available.
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?

Similar Discussions: Questions about fusion (from a high-schooler)
  1. Question about fusion (Replies: 11)

  2. Questions on Fusion (Replies: 28)