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.
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.)
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.
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 :(
