ADDENDA
Terry,
I'm sure you noticed the slight contradiction between,
You do not need an external source of neutrons at all in order to burn refined yellowcake. The oxygen is not going to absorb many neutrons. In fact a "natural reactor" using only surface water as a moderator, continued for over 100,000 years at a site in Africa (Oklo) using only natural U.
"Then, how much enrichment is required for a given neutron source."
Zero, if you use heavy water or carbon as moderator
and the previous assertion,
"Normal fission gives about 2.4 new neutrons for every fission neutron. With Natural U, you immediately loose half of these to absorption by other 238U atoms, and if you are not using heavy water or graphite as a moderator you CANNOT go critical, no matter how much fuel you have, as there are too many losses to keep the Keff over 1."
OK, the resolution to this and your further question about enrichment is:
The natural reactor at Oklo occurred 2 million years ago when all the uranium on earth was of significantly higher enrichment then it is now. The natural U which we have now, in the form of yellowcake having undergone a fractional half-life since then, and presently at the enrichment of ~7 ppk would require a heavy water or carbon moderator to go critical, not light water. There will be no further "natural reactor" except possibly within the earth itself. With only light water as moderator, a significant external neutron source would be required now, based upon the amount of fuel and the geometry of the reactor, which was not the case at Oklo. Claro?
Jones
