Chromatography Forum

I'm developing a reverse phase gradient method for a compound using a Agilent SB-Aq 50mm x 1.8µm x 4.6mm column. The gradient is simply 80:20 -20:80 over 8 minutes using 100mmol ammonium formate pH adjusted to 3.2 using formic acid and ACN. Everything is acceptable at this condition. k' >2<20, Tailing=1.1 with resolution > 3 for all impurity peaks.

The only potential issue that I thought might need to be addressed was that the UV wavelength I used was not the absolute maximum. The compound has two maxima, 232nm and 252nm. I chose 252nm because at the 232 wavelength the baseline drift when running the gradient is much greater and made the chromatography problematic. I decided to try KH2PO4 at pH 3.2 since it has a lower UV cutoff and this would eliminate the drift at the 232nm wavelength. (I tried to keep this as short as possible, but also wanted to get all the details).

Long story short, I made a 10mmol PPM buffer at 3.2 and completely lost retention for this main compound. How can this happen when this pH works at 3.2 in the formate buffer system?

Just use 252nm. No need to make problems.

How can this happen when this pH works at 3.2 in the formate buffer system?

Without knowing anything about the chemistry of your analytes, I can "handwave": formate is somewhat hydrophobic (OK, not very hydrophobic, but more so that phosphate); some formate is sticking to the stationary phase and retaining your cationic analyte.

But your analyte is an acid, you say? No problem, we make up explanations to suit : some of the ammonium ion is sticking to the stationary phase and you're getting ion-pair retention of your anionic analyte.

Hi Mike H.,

Not to complicate things further...but doing so anyway...the pH of ammonium formate, pH adjusted to 3.2 when mixed with ACN and the pH of potassium dihydrogen phosphate, pH adjusted to 3.2 when mixed with ACN may actually NOT Be the Same.

A really weird idea...but likely quite true. Remember that mixtures of salts ("buffers") with water and organic solvent mixtures tend to move to the alkaline side...and that the pKas of formate and phosphate are different to begin with.

Lots of nice articles on this out there...LC-GC magazine comes to mind...I still wish that I had all of the issues I saved on hand or could find them electronically. [I moaned about this once to Mr. Jupille...the pH in organic solvent mixtures one(s) can likely be found on the Internet, though].

Lots of nice articles on this out there...LC-GC magazine comes to mind...I still wish that I had all of the issues I saved on hand or could find them electronically. [I moaned about this once to Mr. Jupille...the pH in organic solvent mixtures one(s) can likely be found on the Internet, though].

Tom just recently sent some email with a lot of useful links for chromatographer,

mabe the LC Troubleshooting bible could be what you're looking for:
http://www.lcresources.com/tsbible/

an archive of John W. Dolan's LC/GC troubleshooting articles

Hi Hollow, and everyone else,

Actually Tom Jupille put this into another currently running thread...

"If you are running existing validated methods, you run them as written, period. Column lifetime is a secondary issue.

If you are developing methods, the issue is complicated. Read these three articles by Bill Tindall that originally appeared in LC-GC:
http://www.lcresources.com/tsbible/20112002.pdf
http://www.lcresources.com/tsbible/20122002.pdf
http://www.lcresources.com/tsbible/21012003.pdf
_________________
-- Tom Jupille
LC Resources / Separation Science Associates
tjupille@lcresources.com
(800) 379-5221 ext 21"

These are the articles I was thinking of...my thanks to Mr. Jupille!