Why U235 over U238? There are 3 isotopes of uranium that can be found in nature[1]: U234, U235 and U238. For a chain reaction to last there must be a high amount of neutrons contents and by comparison, U238 contains 92 protons and  146 neutrons[2] while U235 contains 92 protons and 143 neutrons[3]. My question is why not use U238? Why bother to enrich U235 using gaseous diffusion technique[4]?
 A: The requirement for a material to be fissile (to be able to sustain a chain fission reaction) is not simply that it have neutrons, but that each fission releases enough neutrons of the right energy to trigger further reactions.
As it so happens (and I know of no simple reason for this, it's just the way the equations work), U-238 isn't fissile. You can break it with a fast neutron, but it won't release enough fast neutrons for this to continue.
U-235, on the other hand, is very sensitive to being broken up by slow neutrons (almost 600 times more sensitive according to Wikipedia). Moreover, it often releases multiple neutrons when breaking up. And it is easier to slow down too-fast neutrons (this is what moderators do) than to speed up too-slow ones.
It is just an unfortunate coincidence that U-238 has 3 more neutrons than U-235, and the textbook uranium chain reaction involves releasing 3 free neutrons, so people think U-238 is the fissile material. In reality, U-235 releases the 3 neutrons (it has plenty to spare anyway), and so we need to work hard to acquire the rather rare U-235 in order to harvest the power of chain reactions.
A: Neither of the answers addresses what fissile really means, why to use fissile materials, and why not to non-fissile materials. 


*

*Fissioning any fissionable isotope inherently releases a probabilistic number of neutrons, the average of which ranges from 2-4 per fission (bottom table (v bar)), and this number is not a major criterion for choosing an isotope. 

*Contrary to some comments here, U-238 can fission and nuclear stability is not solely related to high neutron counts.
The key part of the definition of fissile is that a fissile material can sustain fission with neutrons of any energy. This means that any neutron in the local neutron population can cause fission, though, in the case of most isotopes (U-235 included), lower "thermal" energy neutrons are more likely to cause fission. This point is understood with a basic understanding of each isotopes' energy dependent neutron cross-sections and critical energy. 
Critical energy is the excitation energy level of the nucleus above which fission can occur.


*

*In fissile isotopes, like U-235, the critical energy is at or below the excitation of the nucleus when it has absorbed a neutron. So any absorbed neutron can cause fission in these nuclei, though they still have a capture probability (U-235 can absorb a neutron and become U-236 instead of fissioning).

*In fissionable and non-fissile isotopes, like U-238, critical energy is greater than the excitation of the nucleus when it has absorbed a neutron, so the neutron must bring additional energy to cause fission.

*Low critical energy and being fissile is generally linked to odd neutron counts.
Energy dependent neutron cross-sections 


*

*The plot in the link shows that only very high energy neutrons can cause fission in U-238 (far right brown line)(note the logarithmic axes), which correlates the critical energy concept; only neutrons above a certain energy are bringing enough to the U-238 nucleus to get it above its critical energy. 

*It also shows that any neutron energy can cause fission in U-235. 


Answer 


*

*Since too few neutrons are born from fission at the energy required to fission U-238 (and other non-fissile isotopes), a reaction with only U-238 is not sustainable. 

*All neutrons can cause fission in U-235, so its reaction is sustainable.

A: It depends not on the number of neutrons it has but the number of neutrons that it releases in each step of the chain reaction. The isotope is chosen based on this.
