Chromatography Forum

Recently I have come across several papers that recommend ion exclusion columns for analysis of things like small sugars (which are classically difficult to retain).

This doesn't quite make sense to me. The Phenomenex Rezex line, for example, works by three mechanisms: size exclusion, charge exclusion, and reversed phase. But it seems to me the first two of these does nothing to retain a sugar molecule. So, ultimately, all we have is a reversed phase separation with some polar interactions.

Wouldn't we be just as well to use one of the reversed phase columns that has good ability to retain polar compounds.

Any thoughts are appreciated.

The advantages of these polymer-based columns is that they minimise sample preparation, and provide good separation of many sugars without much cleanup of aqueous samples ( dissolve in water or dilute acid ) . They are expensive, have limited pressure range, and need relatively-clean samples, but can be very durable and consistent, and are very good for monitoring fermentations and other processes.

The Aminex range from Bio-Rad ( which, I believe, the Rezex are clones of ) have a lot of technical information about methods and method development freely available at their WWW site.

From their site...

" Resin-based HPLC columns use the mechanisms of ion exclusion, ion exchange, ligand exchange, size exclusion, reversed phase, and normal phase partitioning. These multiple modes of interaction offer a unique ability to separate compounds.

The charge on the resin provides the capability for ion exclusion, while the polystyrene backbone allows hydrophobic interaction to take place. The extent of the interactions depends on the compounds being analyzed and the degree of selectivity required.

Reversed phase and ion pairing HPLC techniques require complex eluant conditions for effective separations. These methods work on the principle of modifying the compound to be analyzed until it is compatible with the column.

With resin-based HPLC columns, instead of modifying the compound to be analyzed, the column packing material is modified and chromatographic conditions are optimized to be compatible with the compound structure.

Therefore, resin-based columns often allow the use of an isocratic HPLC system, they simplify sample preparation methods, and
they require no sample derivatization. By cutting down sample preparation time, resin-based columns greatly reduce total analysis time. "

http://www.biorad.com/cmc_upload/Litera ... LIT42D.PDF

Bruce Hamilton

You will get better information from the 2009 Waters catalogue than what you have found at Phenomenex. On Page 139, the retention mechanisms of different sugar analysis tools is explained in common sense language ("large ones elute first" means that the basic principle of the separation mechanism is size exclusion), and on page 140 you get a nice view of the retention times that have been measured for a ton of sugars and similar compounds. These are primarily meant to help the user to select the right column for his application, but it can also help the curious to get a climpse of the possible retention mechanism(s).

Thank You. I will defenitely look into these leads.

Uwe: does size exclusion retain a small molecule any better than a regular porous reversed phase column.

...minimise sample preparation, and provide good separation of many sugars without much cleanup of aqueous samples ...very good for monitoring fermentations...

I agree about the sample prep, our lab centrifuged and filtered the samples, but I know many Fuel Ethanol Plants just used a paper filter to seperate the corn particles.

We used the Waters Fast Ethanol method (with an IC-Pak Ion Exchange 7.8 x 150 with a Shodex SH-G Guard Column), so we could get a lot of samples done quickly, but the trade off was resolution in the sugar area. This method doesn't give baseline resolution in the sugar area (Dextrin, Maltotrios, Maltose, Glucose).

Also I know that Sucrose will co-elute sompletely hidden in a peak (glucose?). We normally ran corn mash samples, but we sometimes got molasses samples too. Sucrose was defintly present, but not separated, no shoulders, nothing. I never got a chance to fiddle around with it enough to separate out the sucrose, it might require a different method all together... I have a suspicion other more exotic disaccharides were occasionally present, but the method wouldn't separate those either, like Trehelose for example.

Our Researcher was fond of the Biorad, but it required a longer run time (20 minutes). I never had to resort to switching back to his method, so I don't have hands-on experience with the Biorad, but he used that style happily for years. I only had two problems with our method. One was technician related: the column oven got turned off and the lactic acid peak slid right into the glycerol peak. The other was sample related; when the lactic acid got above 3% the glucose peak got ugly (shouldering and tailing). I'm sure that was related to the mobile being such a dilute acid (.001N H2SO4).

I agree with Uwe, the Waters catalog is awesome. I used mine like a reference book. Good Luck!

Another advantage would be that IC-Pak/Aminex/Rezex type columns uses 100% water as eluent. This is more environmentally friendly, plus it saves money.

Shodex has equivalent columns for the ethanol production monitoring, SH1011 column has the base-line resolution power for maltotriose, maltose, and glucose. http://www.shodex.com/english/dc030508.html

The application data with SH1821 column, http://www.shodex.de/index.php?seitenid=1&applic=1315

...uses 100% water as eluent...

Shodex has equivalent columns for the ethanol production monitoring, SH1011 column has the base-line resolution power for maltotriose, maltose, and glucose. http://www.shodex.com/english/dc030508.html

I agree, the water eluent is a huge plus. Thanks for the link, I might forward it onto my buddy still in the lab.

I did notice the succinic acid peak, this method must be for wine ethanol? The baseline resolution is nice, I think the Biorad method might have done it too, but the run time was twice as long. Since the 10 minute method gave "good enough" resolution for our normal purpose. But this might be better looking for those exotic sugars... hmmm