hep on seperating interference peaks from main peak

[kwfx]

Hi, HPLC pros:

I have two peaks from placebo that interfere with my main peak. They are not from diluent and the system is clean. The overlay of placebo and API shows these peaks are right underneath the API. My assay method is isocretic, ACN/MeOH/H2O=800:190:10, diluent is Ethanol:H2O=90:10, flow rate :1.5mg/ml, column temp: 22degree. 3.5x250mm C18 column. My current retention time is 11min.

I can't increase H2O ratios in mobile phase because my diluent would be stronger and I assume that would mess up my peak shape. (I don't have room to play with diluent due to the nature of the sample). I prefer not to change column and hope to seperate these interference peaks from main peak without too many trail and errors. So:

1. Which parameters shall I try first?
2. If change ACN and MeOH ratio what ratio do you suggest to start with? I don't know huch much 1% organic change will affect retention time? Does anyone have an idea?
3. If change temperature, what suggested temp shall I go for?

Thanks a lot.

[tom jupille]

You may be asking the impossible.

First of all, your mobile phase is only 1% water. That's rather extreme, and suggests that a shorter-chain or more polar column might be more appropriate. As a back of the envelope estimate, a 250 x 3.5mm column should have an internal volume of about 1.5 mL, which implies a dead time of 1 minute at 1.5 mL/min (by the way, are you sure about the diameter? 3.2mm is more common). That puts the k' of your peak(s) out around 10, which is about as high as you usually want to go.

In general, a 10% change in aqueous/organic ratio will change retention by a factor of about 2.5-3, so a 1% change in aqueous/organic ratio will change retention be a factor of about 1.1 (that looks funny, but the relationship is logarithmic). That generalization applies when the aqueous/organic ratio is "reasonable" -- say from 90/10 to 10/90; at the extremes, it's no more than a wild guess.

As to the methanol/ACN ratio, there is no way to guess what the effect of a change is without knowing how that ratio was arrived at. If the method has been developed systematically, the selectivity of binary mixtures (ACN/water; MeOH/water) would have been determined, and the final ratio chosen to interpolate between those extremes. In the absence of that selectivity data, all you can do is to make a big change in the ratio and see what happens.

Retention in reversed-phase typically decreases as you increase temperature, but the effect on selectivity (peak spacing, which is what you're interested in) has to be empirically determined. You are most likely to see selectivity changes when your sample has a mixture of acids and bases and your mobile phase pH is reasonably close to at least some of the pKa values. If you want to try it, make a *big* change in temperature (say, to 50 degrees) and see what happens. If the selectivity changes, you can interpolate from there.

I know this isn't what you hoped to hear, but the best advice I can give you is to start over with a more appropriate column and do the development systematically.

[SIELC_Tech]

Is your compound that hydrophobic??? I think that even lipids will elute at this conditions. DO you have any basic groups on the molecule?

[kwfx]

Thanks a lot for your reply. The API is vitamin D5.

I just got another purity method from our client, which uses ACN/H2O=85:15 and requires a different column. This method shows a few labeled impurities but I don't have the column required.

So my question is :

If I want to try this purity method with a different column and maybe have to twick mobile phase, will the change of column or change of mobile phase affect RRT if impurities? I am worried if I can still known which impurity is which when using a different column with the same HPLC parameters. (Assuming all peaks will be completely seperated).

Thanks.

[tom jupille]

No, there is no guarantee that your peaks will elute in the same order with a different column and/or solvent.