RI versus ELSD for Sugar Analysis

[lmh]

I suppose it would be downright inflammatory to mention capillary electrophoresis for sugars and oligosaccharides, combined with the tagging kit that I think Beckman Coulter made - it was some sort of fluorescent thing bristling with sulphate groups, conferring charge (for CE) and fluorescence (for detection, using a laser for excitation energy), and as I remember provided very high sensitivity, the narrowest peak-widths imaginable, and very beautiful separation. But no one in the Western world wants to admit the existence of capillary electrophoresis (and that includes me).
You'd have thought that something as common as sugars would be easy, but they're uncomfortable by all methods.

[cigtoxdoc]

I don't see the problem here with RI detectors going back to the days of the Waters Associates R-401. I just looked at a method that I approved in 1982 for the determination of glucose, fructose, and sucrose in tobacco (a difficult matrix). "Method was validated for 2.50 to 100 rag of each sugar per gram of
tobacco. The limit of detection is 0,5 mg for each sugar. Column was 4.6 mm l.D. x 25 cm long column packed with spherical 5 pm Spheri-5 NH2 packing.
Catalog Number AS-5A from Brownlee Labs., Santa Clara, CA. The mobile phase for the liquid chromatography is 80% acetonitrile and 20% water." RI was a Waters R-401.

John

[Consumer Products Guy]

Remember: drift and unstable baselines with RI detector will be dependent on how much signal your assay provides compared to the baseline. So much easier if higher levels/signals of analyte are present, tougher at low levels.

[chemist23]

I don't see the problem here with RI detectors going back to the days of the Waters Associates R-401. I just looked at a method that I approved in 1982 for the determination of glucose, fructose, and sucrose in tobacco (a difficult matrix). "Method was validated for 2.50 to 100 rag of each sugar per gram of
tobacco. The limit of detection is 0,5 mg for each sugar. Column was 4.6 mm l.D. x 25 cm long column packed with spherical 5 pm Spheri-5 NH2 packing.
Catalog Number AS-5A from Brownlee Labs., Santa Clara, CA. The mobile phase for the liquid chromatography is 80% acetonitrile and 20% water." RI was a Waters R-401.

John

"2.50 to 100 rag" what does the rag stand for?

I am curious to know how you prepared the tobacco samples (concentration, flitering, etc.).

I am attempting to develop a method for tobacco matrix as well, including galactose, maltose and sucrose.

issues I encounter:
-Galactose peak tailing
-Low sensitivity (0.1 mg/ml to 0.5 mg/mL is the lowest I can get)
-Lactose solubility. Lactose doesn't dissolve very good in water therefore need to prepare lower concentration with respect to other sugars.
-RT of galactose and maltose shift slightly in sample matrix.

[Vlad Orlovsky]

My choice for detector for sugars goes in the following order CAD-ELSD-RI. Considering the art affects in RI and no gradient, it is obvious that CAD and ELSD are better choices. If you are analyzing sugars in HILIC mode, you definitely need a gradient to elute mono-, di- tri-saccharides in a much shorter method. You can always do derivatization and analyze them by UV. We have several columns for analysis of sugars, which can operate at lower ACN content than regular HILIC columns.