Probably a stupid car question...

[Lissa]

On a (slightly) unrelated note, just how dense does the shielding have to be to impart Bremsstrahlung from Nitrogen-16 decay? I mean, just how energetic is that beta particle event?

From Wolfram|Alpha... Nitrogen 16. During its decay back to O16, it gives off 5-7 MeV high energy gamma radiation. From Wiki, "Condensate from the condenser is typically retained for 10 minutes to allow for decay of the 16N. This eliminates the need to shield and restrict access to any of the feed water piping or pumps."

Errr...there's no gamma with the decay, just a beta. Given, you have an electron moving at relativistic speeds, but it's just an electron which will be stopped by your skin if the decay happens external to your body. If the N16 decays in your lungs however, there is definite damage potential.

For gamma / neutron radiation, here is source for computing necessary shielding.

And you'll note that there are given thicknesses required to stop any gamma or neutrons. It's well known that you need 7+ thicknesses to cut the radiation to near 0 levels and it's going to depend on how powerful a gamma or neutron source is (case in point, when I was at the University of Arizona, we had a 5 Cu Co60 source. We had a couple feet of lead to act as shielding, but even still, when the source was raised to irradiate a sample, you could still measure a slight rise in radiation outside the room where the source was stored (when placing samples in the irradiation chamber while the source was lowered, we were required to wear finger dosimeters and we had to make sure that when we reached in, we had to do so that the lead shielding blocked our body and head). Given the amount of lead we needed for just a 5 Cu source, the amount of lead/uranium/other dense material for a reactor, even a small one, would be considerable.

I still believe it is possible with better engineering. For example, Toshiba has developed a 10MW compact reactor measuring 20 feet x 6 feet. I like the part where they use U238 as a part of the shielding, but over time it becomes enriched and contributes to the total fuel - an ingenious type of breeder reactor. A little too big, and far more power than is needed for a car sized vehicle (150 Kw). So, not an easy proposition to scale this down to 1/66 the energy output, and size by 1/10, but hardly the stuff of science fiction.

The problem isn't better engineering, we're designing reactors now that are passively safe, it takes an act of sabotage to make them meltdown now. The problem still remains, depending on your source strength, you're going to need a certain amount of shielding and that shielding can weigh a lot.

Another aspect is this, the Russians have created suitcase and hand gernade size nukes, we know minaturization is possible, the problem is the shielding and it always will be. You can't use coulombic forces to stop a neutron or a gamma like you can with a beta or an alpha, so the only way to really stop them is material that will increase the chance of interaction and potentially back scatter to the source till the harm said radiation is nil.