[Babcock_Hall]

I recently prepared several high-ionic strength phosphate buffers in different ways.  Unlike biological buffers (with which I am more familiar), the purpose of this buffer is to wash an organic layer.  We are trying to leave 4-nitrophenol in the organic layer and pull our product, a carboxylic acid, into the aqueous layer.  Then we acidify and pull the carboxylic acid back out.  Getting to approximately the correct pH is important (we are aiming for about 6.25), and having a high ionic strength may be helpful.  However, I am using this solution in place of saturated bicarbonate in a literature protocol, and so the conditions I have chosen may or may not be optimal.

One way was to take almost saturated NaH₂PO₄ (0.8 moles in ~0.15 L final volume) and add 10 M NaOH (roughly 45 mL).  This generated a good deal of heat, and so I iced the solution to bring it back to room temperature.  My rough calculations suggest that the apparent pK_a of phosphate is about 6.15, which is roughly one pH unit below the thermodynamic pK_a IIRC.  I have also tried making this buffer up taking saturated solutions of the monosodium and disodium salts and mixing until the correct pH is obtained.  However the disodium salt is much less soluble in my hands than the monosodium is (the CRC Handbook solubilities are not entirely helpful IMO).
(1) Does this sound reasonable for a dissociation constant at very high phosphate concentration (about 4 M in phosphate)?
(2) Is there a better way to make this buffer up?  I would like to keep the ionic strength as high as possible, although I am not certain that it is critical to do so.
(3) Is it possible to make a reasonable prediction about the apparent pK_a of 4-nitrophenol (the side product we are trying to remove) at very high ionic strength?  If it is significantly depressed, we may need to rethink what our optimal pH.

[AWK]

Potassium phosphates are much better soluble in cold water

[Babcock_Hall]

Good point; thank you.