Peak disappearing with ammonium formate buffer

[prk018]

I was using water and methanol as solvents to separate my compound of interest, found some papers for reference and learned that I should use ammonium formate buffer (10mM) for best results, so I switched to water and methanol with 10mM ammonium formate. I can't find my peak any more, and I'm confused because the LC in our lab is relatively new and we haven't really run any samples so it's unlikely that there's some sort of coating on filters or tubing that are interfering with my compound of interest, but the papers I found were all able to detect my compound of interest in much lower concentrations than my samples with the same solvents/methods. I'm wondering if there were any other things that I should have considered when using ammonium formate buffer, or really just in general?

[JI2002]

It's hard to speculate what could be the problem as no details were provided about the method. But the first thing coming to mind is pH. also injection of a high conc standard would allow easier peak identification.

[Andy Alpert]

What wavelength are you monitoring? Formate ion absorbs light below 240 nm. At 225 nm, a 10 mM solution of formate will produce a baseline that's 0.2 AU high. If you were running isocratically, then you could just subtract that high baseline. However, it sounds like you're running a gradient. At a wavelength below 240 nm, that's going to result in a significant change in refractive index of the mobile phase even if you have the same concentration of ammonium formate in both mobile phases. Result: Changing baseline. If you have less than 10 mM ammonium formate in mobile phase B, then in addition to the refractive index change you're going to have a significant decrease in the height of the baseline during the gradient. Do any of these factors sound like they might explain your problem?

Incidentally, what kind of column are you using?

[prk018]

Thanks for all the replies!

- I'm running a water/MeOH gradient (MeOH increasing from 5% to 90%), both have 10mM ammounim formate with pH~3.
- I have adjusted the injection volume so that about 10 times more sample got loaded onto the column, but I still didn't see my peak of interest.
- I'm monitoring at 430 nm so I'm probably not being affected by formate ion absorbing light.
- I'm using a c18 column, 1mm ID, 150mm length, 3um particle size.

[Zoraku]

I was using water and methanol as solvents to separate my compound of interest, found some papers for reference and learned that I should use ammonium formate buffer (10mM) for best results, so I switched to water and methanol with 10mM ammonium formate. I can't find my peak any more, and I'm confused because the LC in our lab is relatively new and we haven't really run any samples so it's unlikely that there's some sort of coating on filters or tubing that are interfering with my compound of interest, but the papers I found were all able to detect my compound of interest in much lower concentrations than my samples with the same solvents/methods. I'm wondering if there were any other things that I should have considered when using ammonium formate buffer, or really just in general?

I could be wrong but I'm a bit confused about how using a formate buffer allows for a detection at a lower concentration than not using a formate buffer?

The concentration at which you detect your molecule of interest should depend on the instrument, the detector, and the properties of your molecule pertaining to the detector you are using (how strong of a chromophore it is if using UV-VIS, how strong of a fluorophore it is if you are using Fluorescence detection, etc...)

I know ammonium formate buffer is pretty common in LC/MS. Were the publications you were looking at also using LC/MS and you are trying to translate that to LC-UV/Vis? Just a couple questions worth looking into to try and figure the problem out, as this could help justify why they could detect a lower concentration.

[prk018]

I wouldn't say that using a formate buffer necessarily allows for detection at a lower limit than not using it, but it really improves my baseline with my samples so the peak is easier to note. And you are right, the publications I found were describing LC-MS methods, I was not considering that before, so it definitely sounds worth it to check out other literature with LC-UV/vis methods that detect my compound of interest, thank you for the suggestion!

[Rndirk]

I agree with Zoraku.

The difference in acidity between pH 7 and pH 3 is huge. It is very possible that your analyte loses or shifts its absorbance by going to that pH.

Can you tell us what you are trying to improve to the original method?