Noisy baseline with ammonium acetate gradient HPLC method

[Rob Burgess]

From a historical perspective we've recently started using a gradient HPLC method for related impurities that contains the following mobile phase composition:

MP A: 25 mM ammonium acetate + 0.05% TFA in water
MP B: 0.05% TFA in ACN

Gradient: 20 to 95% B in 40 mins

I have no history of the method and need to ask the "developer" for his choices.

However, I am concerned at the short term very noisy baseline obsereved routinely when this method is run at 275 nm. Is there any underlying rerason for this? A colleague has advised me that acetate methods are notoriously more noisy than say, phosphate methods. Has anyone else observed this generalisation.

[ConstantinS]

There are 2 possible reasons for ypur problems.

Firstly, of caurse carbonyl groups of acetate and TFA adsorb light near 280 nm. Using acetate+TFA buffers is not good idea for gradient elution.

At least, you may use TFA/TFA solt buffer without acetate. But phosphate buffer is surely better for gradients.

The second, do you fill bottle 1 with 20% eluent and bottle 2 with 80% eluent, or you fill b. 1 with pure buffer and b. 2 with pure modifier? The second way is wrong..

[Uwe Neue]

Based on the spectra that I have seen, I have my doubts about a strong absorbance of TFA and acetate at 275 nm. While I do not have a good explanation for your observations, I doubt that the problem is your buffer.

[Mark Tracy]

I agree with Uwe. If ammonium acetate were absorbing light, you would see a terrible slope to your baseline since it is only in one mobile phase component. My experience with acetate buffers is that they are OK above 240 nm, and tolerable below if you don't use too much. This buffer is around pH 5.5 where phosphate is hopeless. The TFA is probably in there as a weak ion pairing agent to tune up the selectivity. I would investigate the usual suspects: dirty flow cell, tired lamp, inadeqate gradient mixing, flakey check valve, bubbles.

You know the idea of using a pH gradient like this is kind of interesting, and not often used in RP-HPLC.