changing retention time in LC-MS-MS method

[Fluterd]

Hi,

I would like some advice.

I have developped a LC-MS-MS-method with following parameters:
column: 2.5 µm, 3.0*50 mm (pH range: 1.5-10)
mobile phase: (a) water 10 mM ammoniumbicarbonate, pH 7.8 (b) acetontrile
Elution is carried out by using a gradient from 20 to 80%B.

A month ago the retention time of my polar component was 2.2 minutes (in several runs).
About two weeks ago it was about 2.3 minutes (again in several runs).
Now, a month later, retention time of this compound is 2.4-2.5.
The retention time of another component stays the same (3.3 minutes).

The pKa of the compound is not known, but could it be that it is in the range of the pH of my mobile phase and this explains the changes in retention times?
What worries me is that the retention time seems to becom later and later instead of random variation, or is this just coincidence...?

Thanks for your help!

[Uwe Neue]

you can't make a functioning buffer at a pH around 7 with ammonium bicarbonate. I think this is where the problem starts.

[Fluterd]

I read several articles using this mobile phase. They add 10 mM ammonium bicarbonate to water, this automatically sets the pH to 7.8. What do you suggest then when using a high pH mobile phase. Adjusting the pH to another pH value by using ammonium hydroxide?

[Uwe Neue]

Well, I also was first very excited about the creation of true buffers based on ammoniumbicarboante around pH 7. Then I found that the pH is changing over a few hours due to and during degassing.

Unfortuantely, there is no MS compatible buffer around pH 7. Either you avoid pH 7, or you live with the consequences of difficulties with the reproducibility.

[Fluterd]

To test whether it was really the pH 7 that was the problem, I just tested another mobile phase, same gradient:(a) water 10 mM ammonium bicarbonate, pH adjusted to 9.5 using ammonium hydroxide (b) acetonitrile.

Last week the retention time for the same compound was 2.37 and now, a week later, it is 2.47.
So even when avoiding pH 7.8 I have a shift in the retention times...

[Fluterd]

Another remark: I am using a column temperature of 40°C. Could this too high causing the degradation of ammonium bicarbonate...?

[yangz00g]

No matther at what pH. carbonate or bicarbonate is not good choice for buffer. The CO2 in the air will affect the buffer component balance and therefore pH over time. If you use this kind of buffer, you need prepare it freshly daily.

[HW Mueller]

A normal ammonium bicarb buffer should have enough potential CO2 to make this, practically, a one way street, namely CO2 escaping to the atmosphere. The backreaction would then be negligible.

[Fluterd]

Thanks for your replies! I am surprised to hear your negative advice on the use of ammonium bicarbonate as high pH mobile phase, since several articles used this successful in the LC-MS-MS analysis of basic compounds... The use of high pH mobile phase is not recommended then?

However, I still wonder if the problem I see is entirely caused by this buffer. I use daily prepared buffer. On one day the difference in retention time is minimal, but between different days (each with a freshly prepared buffer) there is this increase in retention time... Could this problem then fully be explained by the use of ammonium bicarbonate?

Thanks very much for your help!

[sam.pedraglio]

Just a general question: is the 0.1min shift in RT so critical for your analysis along different sessions (that include also new solvent preparation)?
In our experience we're so precise only on our UPLC's system on which 0.1min is a significant shift, on our HPLC's with short gradient we observe such a small shift quite often.

[Uwe Neue]

Well, this is actually a response to an older post that I had overlooked.

There is absolutely nothing wrong with using an ammonium bicarbonate buffer around pH 10. Works like charm, is a good buffer, is reproducible, and does not fall apart (no CO2 at this pH).