Sorry, Heather, but science is going against you with this one....and also against what I observe regularly (running a teaching darkroom, we have many bottles of developers, stop, fixers, etc.). This always happens to partially filled tightly sealed plastic bottles of developers -- and as anon notes -- never to bottles contain non-oxidizing chemicals. And if the bottle is glass, one hears the intake of air as it gets sucked into the bottle as one unscews the cap.
It is not 21% of the oxygen, but almost 100% of it -- which comprises of 21% of the normal air around us.I do not believe that 21% of oxygen manages to get into the chemicals.
If a chemical reacts with oxygen, it will react with it until there is no more oxygen it can react to (or until the chemical is completely oxidized). This is the reason the shelf life of a developer in a half-filled bottle is only a third of that of a full bottle (both tightly sealed). And why a tray of Dektol, even if not used, will die within 24 hours...all that surface area and an unlimited amount of oxygen to react with.
The problem is made worse every time one opens up the bottle to use a little of the developer -- the partial vacuum created by the removal of the oxygen from the air above the liquid, sucks fresh air into the bottle, providing even more oxygen for the developer to react with and killing the developer faster and faster.
ic-racer. What you say is partly true. But there is not an equilibrium in a freshly sealed half-filled bottle of developer. Chemical reactions are taking place...oxygen gas is being "consumed" (changed from O2 to new chemicals with oxygen atoms freshly attached to them) by the chemicals of the developer. Once all the available oxygen gas is "consumed' and removed from the air space above the liquid, an sort of equilibrium is reached, if that would is the proper word for an unstable chemical (developer). From that point, changes in temperature will affect the vapor pressure inside the bottle -- but it is insignificant compared to the reduction of the pressure inside the bottle due to the removal of oxygen gas.
And there is some reduction in pressure inside of a sealed bottle of a hot liquid once it cools down -- enough to suck in the sides of a plastic bottle if there was some air space in the bottle. But that is not what is happening to a partially filled bottle of developer. Chances are that the liquid was room temp when the cap was put on, after all, it is partially empty because one has been using it over an amount of time...time enough to cool down.
I remember my 7th grade science teacher having a gallon gasoline can on a bunston burner on his desk -- after the little bit of water inside of it boiled, he took it off the flame and screwed on the cap, and went on with the class. As the can cooled, the sides started to cave in dramatically and with much noise. An impressive experiment -- got our attention!