Hi Gerapas,
This is pretty weird. Reminds me of making buffers with anhydrous sodium phosphate salts...the way we did this was to prepare a solution of sodium hydroxide and water and add the sodium phosphate to the sodium hydroxide solution. In this way, the pH meters we used were able to "equilibrate" or find a pH value of the newly-prepared buffer more quickly so that we didn't have to wait for a longer period of time for the sodium phosphate to dissolve. I understand...tetrabutylammonium hydrogen sulfate has some buffering capacity (probably the acidic hydrogen has a pKa of around 2-3), but maybe adding the TBAHS to a dilute NaOH solution instead of water and then titrating with NaOH could make a difference in the preparation. There may be some value in working in kind of a "backwards" way...make small quantity of dilute (say 10 mL) NaOH solution up and add an amount of TBAHS in the same proportion as for the EP method...dissolve it up and measure pH...that way you will get a feel for how to make a recipe for the 10g/L pH 8.0 solution.
Reminds me also of trying to obtain pH measurements on water purified in the field by the Army...we had to think about what we were doing as our results weren't consistent with other labs in a round-robin. We were using a pretty nice electrode and meter--a fast response-style electrode, can't recall the vendor at the moment--but we were pretty dumb at first. We had to remember that there is Carbon Dioxide in the air...we were using a stirring rate that was WAY TOO FAST (not yelling...emphasizing)...the effect was that we were stirring air into these water samples...and as a result of the CO2, the pH was falling and falling...getting more acidic, and the even the so-called fast response electrode just would not "find" a pH value to settle on, sometimes for 45 minutes!! Eventually all of the labs in the round-robin decided to work with their pH meters under normalized stirring rates...nitrogen-filled bags were tried for a time by all of the labs and we decided eventually that that degree of care wasn't needed for the study..and then everyone was within 0.05 pH unit...hundreds of water samples, 27 different labs, can't recall how many analysts. We all decided that the measurement to record would be taken if there was no drift in the observed pH value for a period of 30 seconds.
These are just some ideas...another idea is that there is a protocol suggested by J.A. Illingworth (1981) in Biochem. J., Vol. 195, pp. 259-262, "A Common Source of Error in pH Measurements." I mean not to say that I believe that your pH meter is not well...this is a way Independent of Standardizing with pH Buffer Standards to check that the junction of your electrode is really well. This is another thing that kind of cropped up during my days at the Army...we found that electrodes that "calibrated perfectly well" were not Actually Working So Well, after all. Junction problems can crop up and remain undetected...and everything seems to be just fine.
Good Luck--I hope that this is some help to you.
