Personally, I have had good results with basic Nitrile (Buna-N) O-rings.
I will use 90 durometer for any application where the O-ring can be extruded…like a gland seal (like piston seals). For face seals (valve to regulator interface, yoke or DIN, they are the same) or corner crush seals (tank neck or hose to 1st stage) either 70 or 90 durometer can be used, but I prefer 90 durometer for that rare chance that the fitting is not seated perfectly and there is a chance of extrusion.
A “properly mated” face seal or corner crushed seal should have very close to zero clearance for the O-ring to be extruded. Therefore, the 90 durometer is not as important and 70 durometer would have the advantage of better conforming to any irregularities (like surface scratches).
Just as a note: gland seals can be static (some of the barrels in second stages, etc.) or dynamic (any moving part, piston, swivels, etc.). Dynamic O-ring seals, other than gland seals, are very rare even in industrial applications.
The max O2 I dive is <40% and even with any high % of O2 (including 100%) a regulator would rarely be exposed to more a few hours a week to any high partial pressure of O2. So, I have not seen a need to change to Viton, but I am open to suggestions. O-rings in a tank valve will have long exposure to O2, but a regulator (for a non-commercial diver) would normally have very limited exposure (sadly, but true).
The way I feel about it, a regulator O-ring is working in a very benign environment as compared to an industrial O-ring. It is exposed to salt water, salt crystals, and some times sand, etc., but the temperature range is relatively small… chemical, oils, industrial dirt are also much harsher that any regulator will experience. Heck many industrial O-rings work 24 hours a day in environments to harsh for the human body.
There is also the issue with shelf live. All the Mil-spec O-ring packages I have seen have a date and they were supposed to be discarded after some time. Well, I have regulators with O-rings that are over 35 years old and they work fine. Again, I am very open to suggestions, but I haven’t yet seen a need to replace them.
I guess that I am a believer on the “if it isn’t broke don’t fix it”.
Let me add that IMO a small O-ring air leak is a major annoyance, but it is rarely of any concern. That being said I have very little tolerance for malfunctioning equipment…I don’t like leaks.
I am far from being an O-ring expert (nor did I stay on a Holiday Inn), but I did worked with a true O-ring expert close to 20 years ago (some people actually make a living from being an O-ring expert). I was going to look up information from the Parker O-ring book and other sources, but decided to just offer some opinions based on some experience instead…