I am developing a new HPLC method using Prodigy ODS-3, 10cmx4.6mm, 5 um column, my mobile phase A is 0.1% TFA in water, mobile phase B is Acetonitrile, flow rate is 1ml/min, column temp is 35C. I run a linear gradient from 0 min to 20 min from 95%A: 5%B to 100%B. What I have noticed is that at wavelength of 228nm (which is the maximum absorbance of my compound of interest), the baseline starts to drift downward starting at 2min, from about 65mv to 10mv at 18min. This has really created a problem for my LOQ determination. My peak of interest comes out at about 10min, but since it is on the down slope, my S/N for LOQ is only 2. I tried to premix the solvents, or slow the gradient, it doesn't seem to be of much improvement. The only time I see any improvement is when I change the wavelength to 254nm or 264nm. But then, I have to inject a large amount of sample in order to get enough response for the main peak. Any suggestions will be highly appreciated!

Thank you!

You are probably seeing the decrease in TFA concentration in your mobile phase over the course of the gradient. TFA probably has a sufficient absorption at 228 nm and as the concentration is decreased, the background signal level also decreases. As you go to higher wavelengths, the absorption of TFA decreases so you no longer see the change in concentration. You could try a method using a non-absorbing acid such as phosphoric or try adding TFA to the acetonitrile to keep the TFA concentration consistent during the gradient.

Russ,
I substituted 0.1% phosphoric acid in place of TFA for mobile phase A at 228nm wavelength, it gives the same separation pattern, with only slight shift of retention times, and it dramatically improved the baseline. I was able to get a S/N of 10 at LOQ level. Thank you very much for the suggestions!

Great! Another example of improving the situation by getting off TFA.

Hans