I had heard of filling tires with nitrogen, but pretty much blew off the whole subject as just a new way for people to be separated from money that can be better spent elsewhere. A massive 6 page article on a segment of the subject appears in the January 24, 2011 issue of Rubber and Plastics News, focusing mostly on the diffusion of oxygen and nitrogen into and out of a tire. I won't get into most of the details , but I am far more interested in the complexity of the subject that I never had given much thought to.
Surprisingly, the subject appears to have been ignored by serious academic researchers. It's mostly independent consulting engineers who have done the research, but in scrolling through their efforts, I really am not convinced any of it has been done correctly.
My interest would be almost entirely on whether internal oxidation  can be avoided by filling the tire with nitrogen, although I'm not even sure that this is that significant of a issue. I don't recall and haven't ever met anyone who lost a tire due to internal oxidation, although arguably some blowouts may have been the result of it. I would speculate that 99.999+ percentage of tires are replaced because of tread wear and external damage that cannot be repaired. So on that basis alone, is it even worth the effort to study it?
As I said, the complexity of the problem certainly would make it an interesting academic exercise, but that complexity is pretty much ignored by those who have studied it so far. The most common approach has been to take a tire, fill it with varying amounts of O2 and N2, heat it to 60 oC and see what happens. Nice try, but anyone who has worked with accelerated aging test knows that these tests, in the words of the Wicked Witch of the West, have to be done "Carefully...carefully". Jungle tests - exposing a part to extreme conditions - can produce totally erroneous results, something I've seen here at Aspen Research, including one client who we helped win millions in a lawsuit from their supplier.
Let me give you one example: olefins under a UV light are routinely exposed to a 18 minute dark period in every 2 hour cycle. Why? It produces more accurate results. Without the dark cycle, the oxygen levels at the surface can become reduced, thereby producing a decrease in the oxidation rate, despite the increase exposure to the UV light.
So if you are going to test a tire, do it right. Cycle the temperatures up and down to simulate real world conditions until you can prove that a consistent high temperature is just fine.
But to me, the crux of the whole issue is to ignore all of the variables and focus on just one question: how much oxygen is needed to do damage? The surface area of a tire is limited, and unless the whole tire is completely evacuated of oxygen and kept that way in the future by topping it off with 100% nitrogen, there will be more than enough oxygen present to form a monolayer on the inner surface of the tire. That should be enough to initiate the oxidation and start the damage, which can propagate quite fine on its own for a while. But regardless of the purity of the nitrogen inside the tire, oxygen will continually diffuse into the tire so that a fresh supply will always be available to keep things moving along.
I'm not too optimistic about the outcome of this experiment, but it certainly should be run, as science is based on data and experimentation, not logical thoughts.
 As an aside, I question the validity of the results, as one of the main equations used in the calculations, #3, is flawed (dimensionally inconsistent) as presented. It may be a typo, or not; for today, it really doesn't matter.
 Oxidation on the external surfaces is going to happen no matter what you fill the tire with.