Choice of column: C8, C18, C30, TSK-GEL Amide80, etc...

[Amesy]

I'm trying to purify a relatively polar, uknown natural product (elutes with ~60% MeOH, 40% H2O with a C18 RP column). Unfortunately, 2-3 other compounds co-elute from the C18 column.--This has been the case for several different MeOH:H2O and ACN:H2O gradient conditions, so I don't think the C18 column is gonna do the separation trick.

I tried NP HPLC (Agilent RX-SIL column), but it seems that my compound became stuck to the column; the compound has a very strong UV absorbance, and this wasn't seen when I ran the mixture through NP HPLC, even after flushing with EtOAc for a long time...Also, the mixture lost bioactivity following NP HPLC, which seems to indicate the compound was lost somewhere in route.

So, now I really want to stick with RP HPLC to purify this monster, but I'm trying to decide what column to try. At my disposal are:
-Several C8 columns: Alltima C8, Agilent SB-C8, Agilent RX-C8
-A C30 column: Develosil Phenomenex C30 CH0-5690
-Agilent Zorbax SB-C3 column
-A TSK-GEL Amide80 column

Does anyone have any insight/suggestion as to which one of these might work best? What is the advantage/disadvantage of a C3 vs C8 vs C18 vs C30 column? Am I possibly better off trying an Amide80 column instead of sticking with the C-based columns? Any insight (or instinct) is appreciated! Thanks much!

[ym3142]

try,
isocratic;
temperature;
THF;
Ph column;
CN column;
different pH;
ion pair;


find some structure info: alkloid or acid?

good luck

[paul duque]

hello,friend

when you compare c8 vs c18 you probably can see that your compounds elute with more or less retention,but not selectivity

I can tell you ,try with chemical changes in your hplc system,i.e pH,phenyl columns,or organic modifier in your mobile phase....this changes probably give diferent elution profiles...

best regards
Paul

[XL]

I agree with Paul's comments. You will have better chance with columns with dramatically different chemistries, such as CN, Phenyl, Diol, perfluorinated phase, etc.

You didn't disclose the types of charge the compound of interest and interferences have. If they bear different type of charges, you can consider using one of the mixed-mode columns in the market. Or if the compound is a base (not a quat), a high-pH stable RP column running at pH10 or 11 might do the trick.

If you can share with us some details, such as type of compounds and/or the chromatogram, I might have some specific suggestions.

[shaun78]

A polar ... natural product says to me that it is big.

Can you infuse the sample on an MS to get an idea of the weight, then select a size exclusion column that exhibits a fractionation range that includes the major molecular fragment?

Also, your use of stuck to the column and did not come off also makes me believe that this molecule is big; perhaps too big to fit through the packing of a normal analytical column (80 or 120 angstrom. I would start with 300 and go up from there if I were you).

Just a thought...

Shaun

[Uwe Neue]

Another trick to consider is the use of a standard C18 and a C18 with an embedded polar group. In general, I have seen similar retention, but a larger difference compared to the difference between a C18 and a phenyl packing. This would always be my first choice in stationary phase selection, and together with solvent and pH, this often does the trick in changing the peak spacing to you desire.

[promochrom]

Hi Amesy,

You may also think of a column with mixed mode (reverse phase +bonded ion pair group or ion-exchange group). You can see more information from the website www.sielc.com or www.gc-lc.com.

promochrom

[Uwe Neue]

The C3 or the C30 have some chance to give you a bit of a difference. Hower, both types of columns have their own problems, either low retentivity or or high silanol activity.

Mixed-mode phases can do a lot for you, but they are more complicated to handle, if you want to take advantage of both functions. I still think that EPG phases or phenyl phases are your best choices together with solvent chage and pH change. The last one, pH change, is clearly the best, if your compounds are ionizable, no matter what else you might consider.

[Alex Buske]

Hi,

Elutig with 60% MeOH isn't really polar, so RP is a good choice. You have already checked solvent influence. Maybe you could try sone THF in the mobile phase.
Are the coeluters similar compounds (similar UV spectrum)? In that case I would give the C8 certainly a try. And then phenyl or perfluorrophenyl or some embedded polar.
With NP you have a much wider range of solvents. Do you know the "prisma" model? I have just a citation in german.
If the compounds were from different classes pH variation might help.
What are your column dimensions? Separation of closely related compound might just be impossible on a 50mm column.

Alex