Need help please! polysulfated disaccharide analysis

[Uwe Neue]

Don't know... Actually pyrrolidine would be the better choice. About the same pKa, boiling point 71 degrees, only a tad above the decomposition temperature of ammonium bicarbonate and above the temperature that you are currently using...
What is the highest temperature that the manufacturer recommends for these ELSD thingies?

[dkollmorgen]

The ELSD "thingies" can go up to 120C, but the analysis is considered "high temperature" at 80-100C. I wouldn't go that high anyhow, as my analyte is thermally sensitive - yet another challenge. Uwe, thanks for all of your responses. I appreciate you taking the time to help me out.

[Bryan Evans]

The Unison UK-Amino data uses a salt gradient (increasing ammonium hydrogencarbonate) for LC-UV.
I imagine these conditions are not conducive for LC-ELSD (difficult to evaporate).

The LC-ELSD method uses a gradient:
A: MeOH
B: TEAA

[Uwe Neue]

Bryan:
Ammonium hydrogen carbonate falls apart even in the solid state around 60 degrees into ammonia, carbondioxide and water. It is not a problem for the ELSD. It appears that one would not have a problem with pyrrolidine around 70 degrees either.

dkoll...:
Why do you worry about your analyte falling apart? You want an ELS signal, right? Who cares where it comes from, right?

[Bryan Evans]

Uwe,

I just meant that the mobile phase in the LC-UV data is 100% aq.

dkollmorgen,
If you try TEAA, it's best to use helium sparging and bypass the degasser.

[dkollmorgen]

Uwe, I have to worry about degradation of the analyte. I am attempting to measure the analyte as well as any degradation products present in the product I am analyzing. I certainly do not want to CREATE additional degradants.

Bryan, I am striving for high sensitivity and UV just doesn't cut it for the compound I am working with. I am having pretty good success with the ELSD as far as sensitivity goes, but now I need to work on the resolution of the parent compound and the degradant.

[Uwe Neue]

Hmmh: you get a detector response in the ELSD. It does not make a difference how the detector response is generated, if the molecule falls into pieces in the detector or not. For example, in MS you destroy the molecule, but you still measure the response quantitatively.

[dkollmorgen]

I understand that ELSD is a destructive detector, but I only want to analyze for degradants as they were present in the sample preparation. I absolutely cannot CREATE additional degradants by the detection mode. This would be an unacceptable method if this were the case.

Another question . . . . regarding buffer gradients . . . . what type of re-equilibration times would be typical when switching back to the original composition of buffer, especially on a polymer column? I don't seem to be seeing consistent RTs from injection to injection, but I'm not sure if that is due to the mobile phase volatility issue or due to the gradient re-equilibration. Any thoughts on this?
I am also still having trouble resolving my parent peak from the impurity. I have returned to the amm. bicarb. buffer at pH 9.5 for now. I'm thinking of trying a little acetonitrile to see if it will affect specificity. Methanol doesn't seem to do much.

[dkollmorgen]

My peaks are already eluting pretty early (~2-3 minutes), so increasing the temperature probably wouldn't help me much. They are pretty much co-eluting now. My peaks aren't really tailing with the current parameters, so there isn't anything to eliminate there. Thanks for the thoughts though. Much appreciated.

[Chromatographer2010]

In terms of re-equilibrating I have seen longer times required for polymer columns because they typically have a tendency to shrink and swell to some degree with gradients especially in the presence of organics. I would probably start with 10x column volumes + 1x system volume. Usually though if its re-equilibration related it becomes consistant after 3 or 4 runs even if its not re-equilibrated. Its like it reaches a constant state of non-equilibration.

In terms of the resolution and temperature you could go back to the conditions that gave you longer retention times but excessive tailing and try the higher column temperatures.

[dkollmorgen]

The longer RTs and excessive tailing were on a silica based column and using a "buffer" that is not ELSD compatible (non-volatile). Not an option to go back to those at this point. The RI sensitivity is terrible (that's where we were running those conditions)

[HW Mueller]

Since nobody is taking this up till now: The detector is supposed to give a response which can be calibrated. It is then totally irrellevant what substances give this response, it just need be reproducible. In other words, it does not matter whether the compound disintegrates or not, or to what it disintegrates. You will not know and not need to know, neither does the detector know, whether "degradants" form in the detector or not.

[dkollmorgen]

Well, I have been all over the board with pH (2.6 to 11.1), buffer strength, buffer type, but have been unable to separate the two impurities on either the polymer amino (aPhera amino) or the PRP-X100. We suspect that the amino functional group on the column is not strong enough to separate the two impurities (sucrose hepta- and sucrose octa- sulfates). I would like to solicit your opinions on column selection . . . do you think the Unison UK-amino would provide better selectivity for this application? What about a straight up strong anion exchange column . . .would that be any better? If so, recommendations? Is this the problem AT ALL? I see that Dionex has some unique anion exchange column chemistries, but of course they do not sell their columns to people who don't own their systems. We are an all Waters/Agilent lab. Please don't bother suggesting any of their stuff, as I can't buy it. To refresh, I am using ELSD detection so I need to use volatile buffers. What can you tell me about "ligand exchange" - is this an option? I don't really understand the concept, so I can't tell if this is a feasible alternative to what I am doing. Any other types of chromatography that might be useful? Any further assistance is greatly appreciated. I have been researching all over the place and coming up empty handed.

[Uwe Neue]

A completely different path is the ion-exclusion route. You would use one of the sulfonated polystyrene packings. Most of the separation methods for straight sugars use water, often at elevated temperature, as the mobile phase. For your particular problem, the mobile phases need to be manipulated differently. I will look a bit later to see if I can find an application example for your types of analytes, but I recommend that you look also.

[Bryan Evans]

Hi dkollmorgen -

Some thoughts:

a). The NH2 phase you used is 5um polymeric support.
Unison UK-Amino is 3um silica - efficiency will be > 2x.
So, this increase in efficiency may help with the separation.
(note: UK-Amino is more durable than other silica-based NH2 phases)

b). Mobile phase conditions. Did you also try the TEA acetate?
This is volatile & compatible with ELSD

c). SAX - I don't think this is a good option. The reason is because
you're limited as far as what mobile phase you can use (volatile).
I'm sure SAX will separate these compounds -
but you'll need a different detector.

d). ligand exchange - again, not a great option. The advantage to these
columns is 100% aqueous isocratic elution for LC-RI. But again,
not designed for ELS or MS detection.

Finally, it would be helpful if you could post some chromatograms.
If you can't - then can you email them to me?
Thank you.