Dear colleagues,

I have posted on this topic before, so please excuse the repetition.

A colleague analyses dextrose (glucose) in intravenous fluids under standard hydrophilic interaction HPLC conditions:

Mobile phase: 75:25 acetonitrile:water
Flow rate: 1ml/min
Column: Zorbax 5um aminopropyl 250mm x 4.6mm
Temperature: Ambient
Detection: UV 190nm

Under these conditions dextrose elutes at about 7.7 minutes. Significant quantities of salts chiefly chlorides of sodium, potassium, calcium and magnesium are also present. These elute before dextrose and intefere with quantitation presumably because of the partial protonation of the amino group which now acts as an anion exchanger towards chloride.

I have suggested increasing the amount of acetonitrile in the mobile phase which will increase the dextrose retention factor and hopefully separate it sufficiently from the salts. However this has not worked. Neither has raising the pH of the mobile phase with ammonia to try to decrease the anion exchange ability of the amino group.

Treating the dextrose/salt sample with mixed bed ion exchange resin (1:1 sulfonic acid as H+: quaternary ammonium as OH-) removes all the salts effectively but unfortunately also binds dextrose. Even after washing with 10 or more bed volumes of water or mobile phase, considerable amounts of dextrose remain bound as evidenced by the decrease in peak area of a resin treated sample to one untreated with resin and diluted appropriately.

My next suggestion is to try just a strong anion exchange resin in the hydroxide ion form which should exchange with chloride OK. Other possibilities include a strong cation exchanger in the Ag+ form which should remove most of the chloride. Changing to an HPLC column containing the calcium form of a strong cation exchange resin is also a possibility since the salts should elute with the void but a column oven is needed and resources are limited.

Any other suggestions are very welcome. Thank you for your time,

Tony

What are the levels of glucose in the solution? I have had great luck with using a Biorad 87H column with sulfuric acid (dilute) as the eluent and a RI detector.

I definitely wouldn't use a strong based anion exchange are in the hydroxide form. A resin of this type will also retain dextrose (this is the basis for high pH anion exchange separation of carbohydrates).

Are you sure that the problem is due to chloride? Is the interference independent of which chloride salt is injected?

If you're sure that the problem is due to chloride, treatment of the sample with silver form resin would probably be the best course of action (assuming changing to a different stationary phase isn't an option).

Since I started using a YMC PBMN column with acetonitrile and water, I've not experienced any interference from salt.