Selectivty of Neutral Compounds in Mixed Mode Separations

[SIELC_Tech]

Methanol might give you different selectivity, but with a strong acid in the mobile phase (TFA) and high organic (over 50% of MeOH) you might observe ester formation between TFA in the mobile phase or/and MeOH and acidic function of the stationary phase. This will effect reproducibility in a long run (ester formation is reversible but still you have to take extra precautions). I would rather try THF or other alcohol which has slower rate of ester formation.

[HW Mueller]

SIELC_tech,
"variations" refered to your different columns.
You may not say directly that a single column can "do everything" but you act it by your suggestions.

Let me state differently what I am trying to say: For many of the suggestions made a better solution could well be to use different columns, optimized for one type of chromatography (one type of interaction), in series. In the case of the initial question I would probably do two separate chroms, or use a slightly polar, top notch C-18, find separation conditions for the neutrals with H2O/org, initially almost ignoring the amine, then substituting a buffer for the water to shift the amine to where I want it.

Oh yes, there are a few million unsolved cases out there in clinical chemisty, alone. For instance the question asked the other day: Prolactin in serum/plasma, or ouabain, or cortisol at barely detectable levels (UV) in serum/plasma.

[SIELC_Tech]

What if amine is not "shifting"? Are you going to use two methods? What about other polar compounds (or non polar), do you want to do 5 methods for 10 compounds or you want to use one universal method for polar and ionic compounds (not to mention inorganic cations and anions in mixtures with organic molecules)
I am just wondering if you have any experience outside of academy. People in industry might have different goals. It is all about money and delivering results on time. This board is a great thing: scientists with analytical problems can decide what to do-spend few weeks/month trying to optimize existing method or do something different and save some money for the company and time for other things (education, family or whatever).

Regarding the applications you mentioned: low level of compounds has nothing to do with the column but rather with the detection technique and may be appropriate mobile phase. We are doing free screening for all our customers if they can provide us with material. So if somebody is willing to send us these samples we are ready to challenge this task.

[Yury Zelechonok]

To HW Mueller

On a subject of optimum column... You offered “twoâ€

[Uwe Neue]

MG,
Finally somebody said something intelligent and useful. The switch from acetonitril to methanol is excactly the type of solution that is useful. On a standard C18, it results in a fairly large selectivity change for such compounds.

An ester formation in the mobile phase is of little concern with such a high water content.

[SIELC_Tech]

Uwe,

My comment about ester formation is related to Primesep 100 column which Rob is using for his separation. If you have a mixture (pre-mixed) of MeOH/water/TFA=50/50/0.1 you will observe ester formation which will effect amount of TFA in the mobile phase and retention time on Primesep 100 column, which is a mixed mode column, plus Primesep 100 column has a strong acid on the surface which will get esterified at high MeOH level-not a lot, but enough to effect reproducibility of runs.

[Uwe Neue]

OK, even if you are correct, so what is the problem?

[SIELC_Tech]

There is no problem. Rob is analizing three compounds-two neutral and one amine-on Primesep 100 column. If he will switch to MeOH he might observe (over time) slight change in retention of amine due to ester formation on the phase (at high concentartion of MeOH) and deminish of TFA in the mobile phase (only if they are premix), but with MeOH he might have a chance to separate neutral compounds. I would try to use different buffer first and then switch to THF or IPA.

[Uwe Neue]

Ahh, ester formation with the phase ... now you are making sense ...

[HW Mueller]

Thanks Yury for restating, quite well, what I have said all along.
SIELC-Tech, please read Yury´s statement, and see whether you can get an accurate cortisol level of your blood with one of your columns and UV detction. But to be fair, please don´t do it after you read my contribution.... the cortisol would be way up, even a RIA could determine it ~correctly.

On methanol vs. ACN, that´s so obvious that I assumed it, and other obvious experiments, had been done without success.

[Rob Burgess]

All,
Sorry I haven't been able to contribute to this debate further. I was off for a couple of days last week.
It seems that this is turning out to be a continued interesting topic of discussion.
May I reiterate that the problem we had was with separating the neutral solutes: a sugar ester and a fluorinated steroid. With three different strength AcN mobile phases there was no selectivity at all between these species. However with conventional RP on a C18 column the selectivity was in excess of 3. How can such a dramatic difference be described / attributed. I would hope to expect that the PrimeSep column with a C13 hydrophobic component stationary phase would offer at least some selectivity for these neutral species.
I could try using MeOH for alternative selectivity as we usually mix -on line via the pumps. Would this be viable on the PrimeSep columns?
Is there any other mixed mode alternatives out there? Zirconia phases with different buffer systems perhaps? The reason I'm so hung up on this approach is that we'd like to move away from using ion-pair reagents and try increasing our column loading amounts through higher injection volumes. Also with the absence of ion-pair reagents it may be possible to use either MS or ELSD as alternative detection techniques.

[Yury Zelechonok]

Use MeOH or THF alone or in combination with MeCN. Additional effect on selectivity can be obtained, and often very significant, with replacing buffer. For MS application switch from TFA to ammonium formate or ammonium acetate; for UV from TFA to phosphoric acid or sulfuric acid or as it was said before screen several different cations in phosphate. If nothing helps then there is different column chemistry available such as Primesep P which is phenyl column with the same ion-exchange chemistry as Primesep 100. This column should give you alternative selectivity toward neutral compounds and retain basic compounds as Primesep100 does.

[Victor]

Rob,

I'm missing something here. If you can separate the two netural species successfully on a conventional colmn, presumably the problem is with the basic drug. What pH adjustments have you tried to move this away from the other two on the conventional column? Is your question about the lack of resolution of neutrals on a Primesep column an academic point (if an interesting one?)

[Rob Burgess]

There isn't a problem with our "separation" on the conventional column when we use SDS as an ion pair reagent.
However, for sensitivity purposes we'd like to 1) be able to use either a MSD or ELSD as the sugar ester has a very weak UV chromaphore and 2) be able to increase our column injection volume (before we encounter peak shape problems of the basic solute) starting from a "fixed" 50/50 AcN/water dissolving solvent.
As you are probably aware, ion pair reagents are not compatible with MSD / ELSD.

[SIELC_Tech]

Rob,

We are going to order a few esters of sugars (whatever is available from Aldrich), couple of steroids and show you how the retention time chages with the nature of buffer and the nature of organic portion of the mobile phase. This approach might show you that you can use the same mobile phases for your separation on Primesep 100 column.

If you provide us with email and your phone number we an do it in more efficient way, otherwise we will update you through this thread.