Page 1 of 1

split peak with mobile phase

Posted: Fri Jul 14, 2006 9:03 pm
by sluggo
Hello,

For the analysis of Paclitaxel we follow the USP method. Paclitaxel is diluted in methanol (acidified with acetic acid). Mobile phase is acetonitrile/water, and the column is a taxane specific PHP column.

Under some circumstances I need to use THF as a diluent for this product. THF is about the only solvent that disolves the drug and an excipient that forms the product matrix. I do not know of any way to separate the drug from the excipient once they are both dissolved together.

When I inject a sample of paclitaxel dissolved in THF, the chromatography has a split peak. Is there any explanation for this, or ways to avoid it? The split peak is not caused by excipients. I can make a standard solution of paclitaxel only dissolved in THF and get the same split peak. I will also try this on a C18 column, but have not done so yet.

thanks

Posted: Fri Jul 14, 2006 9:24 pm
by Mark Tracy
THF is the culprit. It is a very strong eluent for reversed-phase, and so the sample is partly self-eluting because the sample solvent is stronger than the mobile phase. There are a couple things to try: One, reduce the injection volume, but you might lose too much signal. Two, dissolve the drug formulation in the smallest possible volume of THF, then dilute it with a weaker solvent like MeOH:H2O; you will need to balance the chromatography versus precipitation in the sample.

Posted: Fri Jul 14, 2006 9:46 pm
by rhaefe
The THF causes the split peaks. THF is a much stronger solvent in RP-LC than acetonitrile or methanol. If the elution power of your solvent is significantly higher than your mobile phase it will lead to peak distortion or peak splitting. If you go back in the archives you will find more information about this.

best regards,

Posted: Fri Jul 14, 2006 9:48 pm
by rhaefe
Mark was quicker. Must have posted while I was typing. Well, sorry for the redundancy...

Posted: Fri Jul 14, 2006 10:16 pm
by james little
we often use strong solvents such as THF for injection.

Usually, 5 ul causes no problems. Even use methylene chloride/trifluoroacetic acid for some samples and use 5 ul. With this latter solvent mix, need to run immediately or the free OH groups form trfluoroacetyl esters. Also, must leave tops off bottle or will extract the cyclic siloxanes from cap and see in LC-MS analysis.