Chromatography Forum

Hi,

I have to analyze two opioid compounds by HPLC, compounds A and B, for assay. Their log P are 0.6(A) and 3.8(B), both have MW around 400 and their pKa for both is above 9. Inital testing on a C18 column 150x4.6 using a gradient composed of phosphate buffer pH3 and ACN and at 240nm show that compound A elutes with the solvent front and compound B has considerable retention. I need your advise to test other conditions, based on the properties described above, that will allow us to retain both compounds on the column and to make the run shorter (<10min). Thanks for the help.

So you have a quite polar (A) and a quite apolar (B) analyte.

Some things you might try:
- At how much % organic did you start your gradient? With a proper column (some sort of "AQ" or the like) you might start with 100% aqueous. Maybe that could retain your compound A?
- HILIC might be the answer to retain the polar compound A, but you might get problems with retaining B. Worth a try...
- You could use the old-fashioned way of ion-pairing for retaining compound A
- Raise the mobile phase pH above the pkA, i.e. in the range of >10. Don't try this with a "normal" C18 column, it will die faster than you're able to inject your samples. You need some sort of pH-stable column, i.e. the hybrid columns offered by Waters (XBridge, XTerra), Phenomenex (Gemini NX) or YMC (Triart).
- Go to a more polar stationary phase (CN, Diol?)

molecule A smells like hydrochlorothiazide, right?

i think isocratic elution will do? have the composition of 50/50(org/aqueous)and start it.is there is any reason why you chose pH 3.0.,because Ph 3.0 is no where related to pKa of your analyte.

i think isocratic elution will do? have the composition of 50/50(org/aqueous)and start it.is there is any reason why you chose pH 3.0.,because Ph 3.0 is no where related to pKa of your analyte.

50/50 for sure will NOT do as there is no retention for compound A, at least not with this stationary phase at this pH.

let the compound B gets retended.so any way if compound A is not getting retended the peak should come at void Rt right?

let the compound B gets retended.so any way if compound A is not getting retended the peak should come at void Rt right?

Right. And if I'm correct, he wants to analyze both. Quantifying a peak at the void is bad, bad, bad...

I can think of several possibilities (no guarantees on any of them, though!)

- try a more polar column (something listed as a "AQ", "embedded polar group" or "amide" column). Those often provide more retention for polar analytes and less retention for non-polars. If you can find a suitable column, this would be my first choice.

- get a base-stable column (a "hybrid" support or even PS-DVB) and try running in the pH 8 - 10 range in the hope that you can change the selectivity. That wouldn't be my favorite for a couple of reasons (column stability, and the tendency of alkaline buffers to absorb CO2 from the air and drift down in pH), but it's at least a possibility.

- start your gradient at a lower % organic (you didn't say what you used) to (hopefully) get retention for the polar compound then do a step gradient to a high enough %B to get a reasonably short elution time for the non-polar. Potential issues are method complexity, equilibration, and run time.

use of ion-pair agent may solve this problem?

Thank you all for your responses.

I am currently trying a Prontosil C18 column with an embeded amide group using 0.1% phosphoric and ACN as mobile phase and it is working well, I am working now on reducing the time in between peaks while maintaining peak purity. Thank you all for the help!

if your compounds are basic mixed-mode cation-exchange is the way to go. Look at these applications, may be you can find compounds with same functionalists:
http://www.sielc.com/Applications_By_Compound.html

Also you can try our step-by-stem method development guide:
http://www.sielc.com/MethodDevelopment_ ... dType.html

The guide will ask you series of questions and at the end recommend you column and conditions to try as initial step.

We also offer free method development for our customers. Contact me if you have a question.

Thank you Vlad for the information!