MadSci Network: Astronomy

Re: What is required to initiate a fusion reaction.

Date: Wed Sep 15 18:58:21 1999
Posted By: Stephen Murray, Physicist
Area of science: Astronomy
ID: 936584076.As

Hi Ryan,

For the benefit of those who might not be as familiar with fusion as you, nuclear fusion is the combination of two lighter nuclei into a single, more massive one. The strong bonding between the protons in the more massive nucleus causes energy to be released during the fusion. This only continues to a point, however. In nuclei more massive than those near iron, the bonding per nucleon (neutron or proton) becomes weaker for more massive nuclei. That's why energy is also released when you break up a nucleus of an element such as uranium to form lighter nuclei.

We measure the strength with which the nucleons are held in the nucleus as the binding energy per nucleon (the energy that it would take to "pull" a nucleon out of the nucleus). The change in the binding energy is much larger when you compare very light elements to heavier ones than when you compare iron and uranium. That, and the fact that light elements tend to be more common has led people to view fusion energy as a very promising energy source.

There's a problem, though. In order to fuse two nuclei, they have to be brought very close together. That's because the strong nuclear force, unlike forces like gravity and electromagnetism, acts only on very short distances, comparable to the size of small nuclei. To get the nuclei that close, you have to overcome the fact that they are both positively charged, and so experience a strong electrical repulsion. To do this, they must collide at high speed, which translates into very high gas temperatures (millions of degrees). Higher gas densities also increase the rate of collisions, and so increase the energy output, which you would obviously want to maximize in a fusion power plant.

So, the goal has been to produce a long-lived plasma (a gas to hot that the electrons have been stripped away from the nuclei) that is extremely hot and as dense as possible. This has proven to be extremely difficult to attain. Some groups have used magnetic fields to try and contain the plasma. These efforts have struggled against the many instabilities which will tend to cause the plasma to become uncontained, in which case it will decrease in its density and cool by contact with the walls of the containment vessel. Others use lasers or particle beams to compress pellets of light elements to high density, causing fusion. Those efforts also are fighting instabilities in the compression, and technical issues of how to set up a system which can fire with enough frequency to actually generate useful energy.

With all of the difficulties, you want to find a fusion reaction that is as easy as possible to initiate. The most obvious one would be to combine normal hydrogen nuclei. The problem is that a nucleus containing two protons is highly unstable (electric repulsion overcomes the strong attraction). It only works in the extremely rare cases when, at the same time, one proton turns into a neutron. So, while this works fine in the Sun (because it has such a huge amount of hydrogen), it won't work in a fusion reactor on Earth.

The next possibility is to use two deuterium nuclei (Deuterium, also called "heavy hydrogen" is a so-called isotope of hydrogen. A nucleus of deuterium has one proton, like ordinary hydrogen, but also has a neutron). I'll use the usual shorthand, and refer to deuterium as D. The reaction is D+D-->T+p+energy. T stands for tritium, another isotope of hydrogen, which has one proton and two neutrons in its nucleus. The p stands for a proton, which is just a nucleus of ordinary hydrogen.

We actually use another reaction, between deuterium and tritium, which goes as D+T-->4He+n+energy. The n stands for a neutron, as you noted in your question, and the 4He stands for "helium-4" which is the ordinary form of helium. We prefer this reaction because, at a given temperature, deuterium and tritium nuclei colliding with each other are typically 100 times more likely to react than are two deuterium nuclei.

There are some problems here. One, as you noted, is that a neutron is released, and carries away much of the energy of the reaction. If some of that energy can be captured, then the efficiency of the reaction would be enhanced. But, there would remain the problem that neutrons are a very dangerous form of radiation. Because they are neutral, they can penetrate well into the body, doing extensive damage. Materials that are exposed to neutrons may also become "neutron activated." This means that some of the nuclei may absorb neutrons, and turn into radioactive nuclei. The containment vessel would therefore eventually have to be treated as radioactive waste.

There is also the problem that tritium itself is radioactive, with a very short half-life. Beyond making tritium somewhat dangerous to handle, it also means that it does not naturally occur on Earth, and has to be man- made. Presumably, this might be done using some of the energy generated by a fusion reactor.

The other reaction that you mention involves helium-3. This is a light isotope of helium, that has two protons and one neutron in its nucleus. It will react with deuterium as follows: D+3He-->4He+p+energy.

No tritium is required in the above reaction, and protons are much less dangerous than neutrons, and so you might think that it would be a much better reaction to use. There is the problem that 3He would have to be man-made, but that would be similar to the problem of tritium.

There's another problem with it, though. Helium has two protons in its nucleus, compared to one for hydrogen. The electrical repulsion is therefore twice as strong between hydrogen and helium nuclei than it is between two hydrogen nuclei. As a result, this reaction requires much higher temperatures if it is to occur at a significant rate. At a temperature of about 200 million degrees, a D-3He collision is about 100 times less likely to result in a reaction than a D-D collision, and therefore about 10,000 times less likely than a D-T collision! Only at temperatures above about 1 billion Kelvin do D-3He reactions become as likely as D-D reactions. They are still less likely than D-T reactions, but this is due to the unstable nature of tritium, which makes it very reactive.

Given the difficulties which we already face in getting fusion to work, we want to make use of the reaction which is the easiest to ignite. Perhaps, once we get the D+T reaction working, we could move on to more difficult reactions involving more common elements (D+D would be the most obvious choice).

How this would possibly drive interplanetary exploration/mining/commerce is, I think, anybody's guess right now (I'm certainly guessing, since I'm not well versed about the abundance of deuterium and 3He in the Solar System). At least initially, the elements for fusion are all present on Earth, or can be generated (tritium and 3He can be created in nuclear reactors, possibly not the most politically desirable option, but technically feasible). The possible use of sources from outside of Earth would depend heavily upon the costs involved in leaving Earth (which are enormous, at least initially) and the difficulty of obtaining the needed materials.

You can find more basic information about fusion at the Princeton Fusion Energy Education web site.

Current Queue | Current Queue for Astronomy | Astronomy archives

Try the links in the MadSci Library for more information on Astronomy.

MadSci Home | Information | Search | Random Knowledge Generator | MadSci Archives | Mad Library | MAD Labs | MAD FAQs | Ask a ? | Join Us! | Help Support MadSci

MadSci Network,
© 1995-1999. All rights reserved.