Unretained Compounds

[brewer]

I'm looking for a couple of compounds that will not be retained under reversed phase conditions. They must ionize in ESI +ive mode.

Any suggestions?

[sassman]

It would help to know what the compounds are and what type of column you are using. Hydrophylic interaction chromatography might work for you. Another option is to use some type of volatile ion pairing reagent. Something like heptafluorobutyric acid. Although that will decrease response for positive ions, so you would have to optimize concentration of HFBA for retention and sensitivity.

[lmh]

The way you write your question, I get the impression you are actually looking for a non-retained dead-volume marker for use in reverse phase LC-MS. Am I right? If so, you could try something as simple as a little sodium formate (or just some salt in a formate-buffered run). I often find that very early peaks consist of a ladder of sodium formate or other salt clusters.

[brewer]

lmh, you are absolutely correct.

I am looking for something that I can do MRM on to identify dead volume and perform k' caculations with. I plan to incorporate this compound into my internal standard solution for method development test injections so I can tell when I'm pushing the methods too far in the search for faster separations.

Acetaminophen blasts right off the column(s), but I don't believe that it it a truly unretained compound. This is for discovery work, so I have some wiggle room, but I'd like to get as close as possible to a truly unretained compound. Sodium formate seems a bit small to do MRM on.

[brewer]

After re-reading your response, I see that you are speaking of an addition to the mobile phase that will be present in the chromatogram as an early elulting ladder of peaks. This is a possibility, but I would rather find something that I can do MRM on if possible.

[Kostas Petritis]

What is the composition of your initial mobile phase (i.e. any ion pairing reagents or some organic solvent?)

[brewer]

MPA = 0.1% formic acid in water
MPB = 0.1% formic acid in ACN

Initial gradient conditions = 20%B

[Kostas Petritis]

At these conditions several underivatized amino acids are eluted in the void volume. You can follow alanine at (90->44) or aspartic acid at (134->74). A lot more amino acids (up to the hydrophobicity of Val) will be eluted at the void volume. At 100% water formic acid Ala and Asp are eluted at the same retention time as uracil or NO3 (detection by ELSD).

I would expect something like Glucosamine to also elute in the void volume under these conditions...

[JA]

With respect to the comment regarding "pushing the methods too far in the search for faster separations", are you restrained by a retention factor specification in your method development protocol?

Expanding on the idea that eluting your compound too fast is bad with respect to the repeatability or reliability of quantitation, wouldn't the post-column sample infusion technique provide the best representation of when you've hit the limits.

[brewer]

Kostas, thank you for the helpful recommendations, I will try them.

JA, I am not strictly bound by a certain retention factor, but find some information is helpful. I will often convert a gradient scouting run to a very short isocratic separation. Since this is discovery analysis I am looking for a quick way to eyeball my retention and have some confidence that I will not be eluting in a region of inherently high suppression. Post column infusion for each method would indeed serve this purpose, but having the k number is nice too.

[lmh]

Are you sure you can have confidence that you're missing the suppression region merely on RT without looking at your samples? Just after the "no retention" area, depending on real sample composition, you may have a huge mass of assorted things, some of which will suppress. Having said that, if you're using good internal standards (i.e. coeluting and chemically similar, preferably isotopically-labelled), it may not matter.

Also, can you just measure the dead volume of your system once per column-size, and use this to calculate the corresponding injection retention time delay for any new flow-rates?

If you're using LC-MS with a diode-array detector, you probably also have a good estimate of dead volume already from the injection peak disturbance in the UV trace (assuming you haven't perfectly matched your injection and running solvents).

Good luck!

[brewer]

Are you sure you can have confidence that you're missing the suppression region merely on RT without looking at your samples? Just after the "no retention" area, depending on real sample composition, you may have a huge mass of assorted things, some of which will suppress.

You're right, the only way to be sure is via post column infusion.
Maybe I'm trying to take too much of a shortcut.