Chromatography Forum

Hi all,

I am working with a chiral molecule (R configuration). I have a limit for the S enantiomer of <0.5%. I perform a derivatization using a chiral reagent followed by analysis by normal phase HPLC. The chromatograhic separation is fine, but to my surprise, I found 1% of the S enantiomer in my samples (according to my API supplier, there should be none detected). On further investigation, I found that the optical purity of my derivatization agent is 98:2 S:R. I have found in a catalog the same reagent with an optical purity > 99.5:0.5 S:R, but it is 3 x the price ($400 per 5 mL).

So my question is, can the optical purity of the derivitization agent have an effect on the apparent purity of the chiral molecule I am trying to derivatize? Or am I wasting my money ordering the higer purity reagent?

Thanks for any ideas,

JohnM

You will get four compounds instead of two (assuming no side reactions). So your answer depends on the accuracy you want. (Of course, it might be possible that you know what all the compounds are and add the right ones together. I presume the accuracy would still be less, though, as compared to a ~100% derivatization reagent). Note that this argument holds for any derivatization reagent if the "dirt" in it also reacts with the analyte.

Well...

If you derivatize a mixture of enantiomers R and L with a reagent that contains the enantiomers R' and L', you will get 4 products: RR', RL', LR', and LL'. From this, you will get three peaks in standard chromatography, since the two in the middle can not be distinquished unless you use a chiral phase.

If your analyte is pure and your reagent is impure, you will get the peaks RR' and RL'. If your reagent is pure, and your analyte is impure, you will get RR' and LR'. You need to make sure by some other technique that the second peak is due to the impurity in the derivatization reagent, and not due to the impurity in your analyte. If you can do that you can live with impure reagents.

Thanks for the replies,

As it turns out, I do observe 3 peaks. I didn't notice at first - I was using a run time of 20 minutes since my peaks of interest are at 8 min and 11 min. The third peak elutes at about 53 min (very broad). This peak has an area about 2% of the main peak. I am expecting my new reagent next week and look forward to seeing what difference it might make.

JohnM

Considering how difficult it is to resolve such compounds, I do not believe that your peak at 53 minutes has anything to do with what you see at 8 and 11 minutes. I suspect that this might be your reagent peak, or some side product.

I have re-analyzed my samples simultaneously using the two different grades of reagent. When I use the >99.5:0.5 S:R reagent, I consistantly obtain 0.3% of my impurity. When I use the 98:2 S:R reagent, I still obtain about 1% of the impurity.

Looking at these results, I think the better solution is to avoid derivatization using a chiral column.