# Concentrating Sunlight to initiate fusion reaction

The idea is to collect sunlight over a large area, and concentrate it down to the nano scale.
For the sake of discussion lets say you concentrate all the light you collect down to 1000 nanometers (1µm).

Questions is... how much surface area sunlight would you need to collect and focus to 1µm, in order to deliver enough energy density to initiate a fusion reaction.

• There are well known thermodynamics limits on this kind of thing. You can't heat the target any hotter than the visible surface of the sun (5700ish K; AKA not hot enough). In any case, the problem in hot fusion is not getting things hot enough (several way of doing that are known) it is keeping the plasma confined. Jun 30, 2013 at 22:29
• So in the following thought experiment.... Jul 1, 2013 at 6:17
• (forgive the absurd hypothetical) let's say an advanced civilization created astronomically scaled parabolic mirror that was able to capture a large percentage of the suns energy, and then was able to focus and deliver all of that energy to a tiny bit of surface area of a planet/moon etc.. You're suggesting that the yottawatts of energy that would be delivered to a few square meters of a solid body in space, would result in only 5700ish K at that surface? Seems like there would be an awful lot of left over energy that wasn't converted to heat, how would it be expressed upon impact. Jul 1, 2013 at 6:28
• Any optical path is a two way thing, and if the target were hotter than the sun more energy would radiate backward along that path than forward. There are also optical limits on how good the focusing system can be that prevent you from getting the focus you are assuming: diffraction effects from the edges of the optical system spread the beam. Jul 1, 2013 at 14:00
• @user26473 Your wording is "a tiny bit of surface area", and we're saying that no, you can't focus to an arbitrarily small area. A parabolic mirror can theoretically focus parallel beams of light to a single point, so you can focus as small as the wavelength of light. The problem is that the sun's rays aren't parallel. As you concentrate the light in space, the angles become more sporadic, and no optics can beat this. That's because the spatial/angular spread is a manifestation of entropy. Jul 1, 2013 at 14:40

Just to summarize the correct comments by dmckee and Alan.SE:

The second law prevents you from using the Sun (or anything) to heat an object to greater than the surface temperature of the Sun. Otherwise you could take a box of gas at equilibrium split it into two halves, use lenses and mirrors to focus the radiation from the left half on the right half, and raise the temperature of the right half. Then you could use that temperature difference to run an engine, thereby extracting work from an equilibriated gas in blatant violation of the 2nd Law.

If you do not see how optics prevents you from doing this, you should ask that as a separate question. (Although dmckee and AlanSE have already said the main points in the comments.)

In any case delivering large quantities of energy to objects is not the constraint on fusion. (Militaries are quite good at that sort of thing). The issue is confining the resulting hot mess, and other subsidiary engineering challenges.

Finally, I do appreciate the elegance of using the Sun to power a fusion reaction. But wouldn't it be simpler to use regular solar power to power a fusion reaction? No giant lenses necessary! (Solar cells: cheap - Giant lenses: expensive.)

• fusion needs temperature and pressure, and the temperature limit we've talked about also limits pressure in accordance with the Stefan–Boltzmann law. Surface power is limited by $\sigma T^{4}$, and for mass-less photons, the constant $h$ converts between power and pressure, just like it does between energy and momentum. In terms of energy quality (as suggested by the question), sunlight is useless for fusion. Gross quantity of energy is sufficient, of course. That is saying that you can run a tokomak from a grid powered by solar panels. That is true. Jul 1, 2013 at 19:14
• Solar cells really $\neq$ cheap, but point made Jul 18, 2017 at 22:40

There is another way to do what you are suggesting. And that is to use an already concentrated form of energy -- lasers.

There are a variety of forms of high energy laser fusion projects underway. Inertial Confinement Fusion is the fancy term for the leading method. See ICF details here I would post some illustrations, but I'm not allowed to post images yet :(

• Yes the idea here is to setup a stable, continuous beam of energy that could provide continuous fusion initiation, so that the fusion reaction doesn't need to be self sustaining, and confined, which of course is a problem with fusion attempts. Jul 1, 2013 at 6:37

Visible part of spectrum does not get involved to nucleus. Maybe electrons. Which can only make plasma or increase pressure if it is constrained in a cube. If pressure can reach the critical level, then deuterium can be used for fusion(like pressurizing through lazer bombardment).

So, you need sudden increase of pressure right(to increase probability of tunneling and imploding of D-T pellet)? Which also needs high temperatures.

http://en.wikipedia.org/wiki/Radiation_pressure says on earth, 9.15 N per kilometer-square

Deuterium pellet initiating pressure is http://link.springer.com/article/10.1023%2FA%3A1020858108669#page-1

http://www.wolframalpha.com/input/?i=150+MPa 150MN per meter square so you need :

16,393,442,622,950 meter-square of lightened area. 16 Tera Meter Square

This number is for 1-meter-squared area target. Deuterium pellet is micro-meter sized so there are 1,000,000,000,000 pellets on 1m² area, which means only 16meter-square is enough to just to pressurize one of pellets. Then, if you want to also increase the temperature by a great deal(maybe to heat 1micro-meter pellet to 100M Kelvin), you would need %3 of earth surface.

For 100M Kelvin, you may need to multiply the result maybe by a million.

Focused and concentrated LASER array is just more applicable for now.

If what you think was applicable, scientists would have used that already.

• Are your calculation assuming a Square Meter target? I am assuming a micrometer target. If so, based on your calculations you would only have to collect about 16 Square kilometers. Which is much more reasonable. Especially if space based. Jul 1, 2013 at 6:48
• You may be right about that, thinking right now. Jul 1, 2013 at 11:46
• It is 16 meter square now. Jul 1, 2013 at 11:57
• Also it is not needed to be continuous beam, that if you can accumulate somehow, only a few cm² could do. Lasers are better. Jul 1, 2013 at 12:01
• These are just to for pressure requirements. To heat to 100M Kelvins, you need at least several million times of this. Jul 1, 2013 at 12:19