the gist of harnessing nuclear energy is converting radiation energy into electricity.

isnt beta radiation a stream of electrons? it happens when a neutron in a nucleus decays to form a proton and give out an electron in the process. the new proton remained in the nucleus. there's no fission or fusion in the process.

the nature of the radiation (alpha, beta or gamma) depends on the radioactive element inside the lead box. your description suggests that the electrons released from the radioactive element hits the semiconductor and create a current. however, the idea of having a metal box to contain ions is dodgy. ions are unstable at ambient condition. Moreover, ions normally occur as an ion-pair where two oppositely charged ions group together in gasesous state to balance each other charge. there's no source of anion available to balance the formation of positive charge, if the electrons released are being directed out of the box through an electric circuit.

i dont think it's beta radiation i could be wrong. after-all, i have limited knowledge on nuclear science.

So pacemaker wearer-ers already use this idea. Cool, it works. Now where would I find out how to scale that up in size? The interim goal here is the "forever LED". Well not forever, just for the halflife of whatever I use. I obviously can't get plutonium. Is there a special section of webelements that I should look at to see which radioactive element would suit my experiment the best? My mom would love some everlasting garden lights for mother's day, and I want a nuclear battery. Seems like this goes hand in hand  

EDIT:
I was just browsing webelements again, and I ran across Actinum and Radium. Both are very bad for you. Webelements told me little other than they are "radioactive". Sometimes I got as far as some elements are good at putting out alpha (heavy, positive), or beta (lightweight, negative), or gamma (will shoot through a yak). For battery purposes, I want beta since I equate negative with electrons. I'm going to go out on a limb and say that I most likely can't buy any of those fun elements. So for experimental purposes, smoke detector supplied americium will have to do. Webelements said that stuff is a good gamma emitter too though, which scares me. How much gamma is too much?

Gamma radiation is one of those things where if you can reduce your exposure to it, it's good.  I guess it all depends on the energy of that gamma radiation.  Look at Rhenium-187.  It emits gamma rays, as do most radioactive isotopes, but they are PATHETICALLY weak.  As a result, it poses no danger.  Gamma rays are the result of the nucleus of the decomposing atom 'readjusting itself', so to speak.  The particles inside move around and form a less energetic state, and that excess energy is shot off as a gamma ray.  It is really difficult to find a radioactive isotope that doesn't emit gamma rays.  Also, the only real way to block gamma rays is with lead metal, and putting lead metal into the ground isn't a good thing to do.  (As the acid rain will slowly corrode the lead, and you don't want any lead being taken up by any plants or vegetables in the area).  Another thing you have to remember is that when dealing with radioactive isotopes, you have to be wary of criticality.  I'm fairly certain you wouldn't be going to that extreme, but if you were to put too much of one substance in too close of an area, you could have it go critical and that would be bad.  

ANY isotope which decays by emitting a neutron (I.E can be fissioned) can go critical.  Neptunium can go critical, Uranium-233 can go critical, thorium can go critical, americium can go critical, I believe, and there are many more.

Exactly.  To go critical you generally need a large amount of the substance.  You need around 10 kg of Pu to go critical, though amounts as low as 6 kg can go critical if the bomb-maker is insanely good.  Neptunium takes about 34 kg to go critical, and basically, if you have enough of an isotope for it to go critical, just being near it would irradiate you to death.

i wouldnt worry too much about americium 241 (im assuming that's what you'll be using becuase of its availibility) because if you got enough of it to make a critical mass, putting you hand within the range of the alpha radiation would instantly give you a radiation burn. If you some how do acuire enough radioactive material ("hello, chemical supply store? i need about 50 kg of plutonium, preferably of reagent grade") to pose a neutron hazard, i recommend alternating peices of you material with peices of cadmium.

Below (or perhaps above, i dont know where this attachment is ending up) is my plan for a nuclear battery. First, the Americium emits alpha radiation, which then hits the copper-activated Zinc sulphide, emitting a glow, just like in radium watches. Because the Americium salt and the Zinc salt are in a mixture, almost all of the alpha radiation emitted by the Am hits the ZnS, making visible light. The visible light then hits a sillycone (i like spelling it that way) solar panel, making energy. The quarts is there so that visible light from the Am and ZnS hits the sillycone, but not alpha. This is a design (but they used a more sane alpha emitter) that is used by some for small electronics that barely use energy, such as a watch, or a pacemaker, or something like that. And also, they didnt use silicone solar panels, they simply used a sheet of silicone with a transparant conductor sandwiched between it and the quartz.

By the way, on the subject of solar panels, a company is now developing a new solar panel that not only absorbs infrared, red, yellow, orange, and green, as silicone panels do (did you notice that they are blue? its becuase they reflect blue, these ones are black), but also some ultraviolet, making lots and lots of electricity. It is also cheap and easy to make, and can even be painted on to a surface.

EDIT: Turns out it's below.