Chromatography Forum

Again, if you prepare your buffer correctly you will end up with exactly the same solutiuion ([k+] as well as [phosphates]). Your precipitation problem is due to differences in concentration in your preps. Now if you report that the buffer you prepared is 25 mM you will mislead people, it´s an untrue statement.

Victor, do you know or do you assume that quinine HPLC is largely insensitive to ionic strength? It would be interesting to know.

I am preparing buffer this way:
5.7g of K2HPO4*3H2O in 1litre HPLC water (25mM??) then titrate to pH3.0 with phosphoric acid.
Then 80%buffer/20%MeCN isocratic.

I have got good peak shape on HP1050 yesterday with Discovery RP-Amide but get tailing when transferred to PE200 Series today (both isocratic) using same batch of mobile phase.

To HW Mueller
The original application from Supelco reads:
RP-Amide C16 15cm x 4.6mm id x 5um
methanol:25mM K2HPO4 pH3.0 (20:80)

(((( Another less clear detail for the same application by Supelco reads:
RP-Amide C16 15cm x 4.6mm id x 5um
80:20 25mM Potassium Phosphate (ph3.0) : MeOH ))))

Have you got a method of buffer preparation in mind when you see this - that is different to mine?

Going back to that wonderful string on buffer prep - perhaps the pH is critical and I'm out by 0.1-0.2 units?
(I have a busted pump motor on my HP1050 now - it was only a backup anyway)

The "less clear" statement is the more correct one.

How to do this has been mentioned several times before:
a. Use Beynon´s calculator (first method!), skip the ionic strength adjuster. It gives you the weights of H3PO4 and KH2PO4 to use.
b. Prepare a 25 mM H3PO4 and a 25 mM KH2PO4 (or here also K2HPO4) solution and mix the two until you get the desired pH (by taking, after mixing well, two portions out into two test tubes, inserting the electrode consecutively...., throwing the portions away, repeat if necessary). You can do this forever without changing the concentration. It is tedious, I now use almost always a., only if I need lotes of solutions with slightly different pH do I still consider b.
Mention which technique you use.

HWM- when I said the method was insensitive to phosphate or K+ concentration, I meant that the method should work over a fairly wide range of phosphate buffer concentrations and K+ concentrations. I mentioned this because WK maintained that it did not work starting with one buffer salt and not the other. However, I did not mean that the retention time is independent of the phosphate buffer concentration or K+ concentration.

If you prepare a pH 3 buffer by taking 25mM of say KH2PO4 and titrating with concentrated phosphoric acid of given SG you will get a solution with phosphate concentration more than 25mM (in fact it will be about 29mM phosphate). This method is reproducible and anyone can do it and produce the same buffer solution with the same phosphate concentration. The K+ concentration is not affected by the dilution caused by a few drops of acid so the {K+} is 25mM. You can titrate with 25mM phosphoric acid if you like and produce a solution with exactly 25mM phosphate but {K+}<25mM. Both methods will give reasonable analysis of quinine although the retention times may be slightly different.

HWM -I'm interested in your comments on the Beynon method. If you make up solutions according to these recipes and measure the pH, they are often apparently wrong by 0.1-0.2 pH unit or so. I think I can measure pH to within about +/- 0.04 of the true value. I have put this difference down to approximations in the programs used for these calculations. So until now I have believed that by measuring the pH I am more accurate than the calculations, but I'd be most interested if you do not agree. Certainly, the Beynon method should be more precise (if I am right it is less accurate). On the other hand, actually measuring the pH is a final check on any gross mistakes made in the preparation of the buffer. However, you could argue that gross mistakes could be made in the pH measurements....

OK HWM,
I have looked at Beynon method and to create 1L of phosphate ph3.0 at 20degC for use at 40degC:
0.003 mol acid (0.3458g adj for 85%)
0.0219 mol base (6.2731g adj for 3H2Os)
I have measured a pH of 7.8 for this????

Victor,
OK I misunderstood you on this RT effect, but there are indeed many substances which show little influence on variation of ionic strength. It is interesting to know which. Acid base dissociations, on the other hand, usually and obviously, have a stark dependancy on pH.

Again, if properly done the Phosphate and M+ will always have the same concentration for one pH, no matter where you start.

Check Bill Tindall´s statements on the accuracy of weighing vs pH meter. Obviously, if you have materials of prime standard quality (or near that) than weighing will be more accurate. Assuming that you are quoting a range +/-0.04, not a SD, then you still have to remember that your mean could be off. That 0.2 range for "Beynon" is a bit high if you used correct temp and ionic strength setting. If I have such a discrepancy the attention is first directed to the pH meter. Your +/-0.04 is ~ 0.1 range, thus a temp. effect, etc., could quickly add up to 0.2.
I also use the pH meter as a rough second (two is better than one, but if there is a discrepancy check your pH meter first).

WK,
without rechecking and recalculating.... even if you measured the solution with your pH meter at 20° instead of at 40°, you should not be at 7.8°. Maybe you used K2HPO4 as base instead of KH2PO4. (Sometimes it is advisable to doubt oneself first?)

SIELC,
What is the pH of the 0.25% Sulfuric acid? I assume %v/v?
I guess I have a limit on the LC - obviously so does the silica.
My manual says to avoid conc. sulfuric acid.

WK,

Any other relatively strong acid (TFA for example) can be used instead of sulfuric acid and we updated our website with the TFA method. 0.25% TFA will create pH around 2. You can use less acid if you want. Acid is used to control retention and peak shape.

SIELC,

What is the low pH stability limit of your columns? This is of interest because your phases have unusual groups bonded to them- i.e. I guess not the usual C18. While I'm here, what about the high pH stability limit- I guess that is the same as most silica based phases, assuming your phases are silica-based.

Primesep columns are silica based and our safe operational pH limit is 2-7. In 99% cases there is no need to run column outside this pH range. You can always find conditions to elute compounds using mobile phase with pH 2-7. Dual mechanism of interaction on the column (ion-exchange and reverse phase) allows using wide variety of additives to the mobile phase to control ion-strength. You can use organic and inorganic acids (TFA, formic, TFA, sulfuric, phosphoric, etc.) or various buffers (ammonium formate, ammonium acetate, phosphates, etc.). It is the matter of right buffer concentration, pH and amount of organic modifier. You might run column for the short period of time below pH 2 but need to flush it for overnight and long term storage.

I've given up and gone to 25mM K2HPO4 to pH7.5 / MeCN with Discovery RP-Amide.
What a carry-on!
This gives me peak shape of 1.2 at 5% peak height which is far better than the 2.5 at lower pHs.
Thanks for all your helpful hints.
WK

Well, this is what you are going to get if you use a packing with an embedded polar group. These things give you superior peak shapes, no matter what pH you are using.

Uwe,
Now you've opened another thread!
If your statement is correct then I should have got the same peak shape with low pH - in fact I got 2.5 instead of 1.2 (I achieved this 1.2 with another system).
I am wondering if there is a problem with the basic analytes and PEEK tubing with low pH? Since I have stainless tubing in the other system.
I have not got the tubing/fittings available to try stainless in my current system right now.
Regards
WK

There is not a good reason for the quinine peak to look bad on a packing with an embedded polar group - unless the peak was less retained and it is due to extra-column bandspreading or related things.