I've been thinking a lot about this in the last hour or so. At the time Grant was writing that material on pH of developers, his boss was working on HC110 and a few other developers, and my boss and I were working on the same type of problem with color developers. He may have been constrained in what he could publish due to ongoing work, but I do know that he and I discussed this topic.
It seems that a pH value of 10.0 or lower is most stable for borate or carbonate, with carbonate being better. The more alkali you have the more stable it is as well.
This is what goes on. HQ oxidizes to Qinone which reacts with sulfite to form HQ-monosulfonate and sodium hydroxide. The HQMS is lower in activity and the hydroxide raises pH and increases activity. It therefore depends on the amount of rise in pH and the HQMS lower activity which way the developer swings. At the same time, carbon dioxide in the air is reducing the amount of buffer and the pH slowly. I generally see an uptic in activity followed by a drop and leveling off under my mixing and storage conditions.
So, short term you see an increase, then a leveling off and then a slow decline in an unreplenished system. The shape and size of this curve depends on how full the bottle is in which it is kept, how often it is opened and a host of other things, but the initial reaction is inevitable due to the oxygen dissolved in all water. The only way to stop it would be to boil the water and then purge it with nitrogen during mixing. But, this initial reaction will take place between HQ and sulfite and then virtually stop once the oxygen in the water and the head of air / oxygen is used up. So there is an equillibrium that is reached.
So, the stability of B&W developers is based on a lot of factors and the results can go in a variety of directions depending on handling, storing, mixing, use and etc.... That is why Kodak and others suggest a shelf life for mixed developer in a closed container. It also explains in part, why there is a difference between a developer and a developer replenisher, as the replenisher must compensate for the normally observable changes in the developer due to air and also usage.
That is where my thoughts are right now. I'll keep on thinking and reading.
PE