Chromatography Forum

Dear forum members,

I am trying to develop a LC-MS/MS method for ethyl sulfate. However, budget is tight, so I would prefer to use the columns available in the lab.

Tested conditions are:

(1) reversed phase (C18 and pentafluorophenyl column) with several mobile phases (acetonitrile vs methanol and aqueous buffers at pH 4 and 9). There is NO retention (as expected, considering the polarity of the molecule)

(2) Acquity UPLC BEH Amide column (used as a HILIC column), tested conditions:
95% acetonitril 5% aqueous buffers at both pH 4 and 9
95% acetonitril 2% aqueous buffer 3% methanol
95% acetonitril 5% methanol
There is NO retention, while I would expect retention!?!

The columns, the LC-MS/MS system and the mobile phases are fine, since other analyses have no problems... How can I get retention on the amide column?

Thank you very much for your appreciated advice!!

ACQUITY UPLC Amide column: Go back to 95% MeCN, 5% water, without buffers. Buffers are not likely to be needed.
Mixture of MeCN with MeOH are not likely to work.

Unfortunately, the use of pure water instead of an aqueous buffer does not increase retention...

Other suggestions maybe?

if you're running Hilic, what's your sample in? You probably want as much ACN as you can get away with, as little water, and no methanol. I can't speak for your column and LC system, but my attempts at HILIC, admittedly using amino columns, have usually failed miserably at the first scent of methanol. It's easy to forget that the sample solvent can have serious consequences for retention, let alone peak-shape.

Injection volume is indeed acetonitrile, so I guess this is not the problem...

Could a lower flow rate have an effect? I now use a 1.7 µm, 2.1 * 50 mm ACQUITY UPLC Amide column with a flow rate of 0.5 ml/min. Perhaps that at a lower flow rate there is more interaction between analyte and column and thus more retention or is this not correct?

Thanks!

Perhaps that at a lower flow rate there is more interaction between analyte and column and thus more retention or is this not correct?

Alas, it's not correct. Chemical interactions are generally *much* faster (as in, orders of magnitude) compared to the residence time in the system. Even the rare cases of slow interaction usually don't affect retention; they typically show up as peak shape issues.

if you're running Hilic, what's your sample in? You probably want as much ACN as you can get away with, as little water, and no methanol. I can't speak for your column and LC system, but my attempts at HILIC, admittedly using amino columns, have usually failed miserably at the first scent of methanol. It's easy to forget that the sample solvent can have serious consequences for retention, let alone peak-shape.

I'm using HILIC for the first time now... While I need to look more closely at the data next week, it looks like I can get away with a 3ul injection of sample in 100% aqueous media.

I know 5 uL of 50% ACN and water works fine on the column I am using ... 5um 4.6x150mm amine (Carbohydrate) column.

If one needs a large volume injection for sensitivity then i would guess that an 100% ACN sample solvent solvent would be best (and/or running a SLOW gradient starting at very low percentage of water... if you are not using an RI of course!)

- Karen

Thanks for the help, unfortunately the ethyl sulfate still is unretained...

Some things I will try the next weeks (suggested by Waters representative):
* use of isopropanol instead of water
* increase buffer capacity
Anybody any experience?

I still have a hard time to understand why a molecule with no retention on C18 is also unretained on a Amide column...