Mobile phase buffering capacity

[Edde]

I have to analyze nitrite by derivatization with N-acetylcysteine to S-nitroso-N-acetylcysteine, then measuring the latter derivatized compound by C18 RP-HPLC and UV detection. According to literature, the derivatization reaction requires the addition of 4M HCl. After the completion of reaction, 200 uL of the reaction mixture is run isocratically with a 25% ACN (in water, buffered by K2HPO4 to pH 2.0) mobile phase.

My question is whether the mobile phase can buffer the highly acidic reaction mixture so that it will not harm the C18 column.

What are the signs and symptoms of a C18 column deteriorated by a few injections of the above-mentioned acidic sample?

Please help.

[tom jupille]

Your mobile phase is already at pH 2. You didn't specify the buffer concentration, nor the final HCl concentration in the sample. If you have sufficient buffer, and if there is sufficient mixing volume between injector and column, then you could be OK.

Symptoms would be deterioration in peak shape and/or changes in retention. If the chromatography gets worse with each sample, then you have your answer. If the chromatography gets worse with time (independent of the number of injections), then you have a problem with the low pH of the mobile phase.

[Edde]

Thanks a lot.

The calculated HCl conc. of the sample is 0.16M and hence the pH should be 0.8 (10 uL 4M HCl added to a final mixture volume of 250 uL).

The mobile phase contains 10 mM of NaH2PO4 in acetonitrile–water, 15:85, v/v.

The isocratic run is at 1mL/min and injection volume 200 uL. How do I know whether there is sufficient mixing volume between injector and column?

I think a C18 RP HPLC column should work at pH 2-8. So is it ok to use pH 2 mobile phase? Would it be better to adjust to pH 2.5?

I am worryng about the column after a few injections. What can I do to maintain it now?

Many thanks.

[tom jupille]

You originally mentioned "according to the literature". That implies that you are following a published, validated method. If it has been validated and you are following it *exactly* then you should not have any problems. By "exactly", I mean just that, including using exactly the same kind of column and with your instrument set up exactly the same way.

If you are changing things, you can anticipate a number of *possible" problems:

- 10 mM is a *very* low buffer concentration for a 200 microliter injection of pH 1.8 sample. If I were setting up a method, I would use 100 mM. That said, if it's a validated method and it works, then it works!

- Given the big disparity between what you are injecting and your mobile phase, I don't think *any* amount of mixing volume will help, so don't worry about that aspect. Again, if it works, it works.

- At low pH, the bonded phase typically comes off; that reduces the amount of C18 and increases the amount of silanol activity. The most common symptom is peak shape problems. Retention usually decreases (bot sometimes increases -- depends on whether the analytes interact more strongly with the C18 or the silanols!). Generally speaking, column lifetime problems start to show up with mobile phase pH values below about 2.5, but there are columns that are designed for lower pH operation (that's why it's important to follow a successfully validated method *exactly*).

You will get the longest lifetime if you flush the column with buffer-free mobile phase for overnight (or longer). The price you pay is that you will have to re-equilibrate the column with the buffered mobile phase before you can run again.

My guess is that if the method works, you *will* have a short column lifetime, but remember that the column *is* a consumable item. Even if you only get 3-4 weeks and/or a couple of hundred injections from the column, the bulk of your analysis cost will still be operator time.

"Columns are cheap; your time is valuable."

[Edde]

Thanks for your detailed reply.

I don't understand why the author of the article used such a low pH for derivatization reaction.

Therefore, I tried pH 2.9 and ran on the same mobile phase. The UV peak height seemed a little bit (less than 2 fold) higher, so I used pH 2.9 instead.

Next, I tried higher buffer conc. and lower organic composition (from 10mM KH2PO4 in 15% ACN to 50mM KH2PO4 in 10% ACN because the latter composition is the routine MP we use for other analysis). The peak shape is a bit broadened with the latter MP.

Why does peak broadening happen? The sample is essentially aqueous and acidic. The 2 mobile phases have same pH.