Double chromatographic peaks

[xmiao]

I analyze CYP inhibition samples using LC-MS/MS with an internal standard. The problem that I encountered is that there are always two peaks observed for the substrate metabolite with the same mass, however, there is only one peak for the internal standard within the same run. I changed some conditions, which could resolve the problem and so substrate metabolite was eluted as one peak, such as, (a) diluting the samples with water; (b) starting from high aqueous composition for the HPLC gradient; (c) reducing injection volume from 15uL to 10uL. They all can help, but I cannot explain why two chromatographic are observed without the above treatment. Any feedback would be greatly appreciated.
XM

[Uwe Neue]

When you inject the analyte in a "strong" solvent, it will migrate with the solvent peak through some part of the column, in the extreme even to the column outlet.

This problem diminishes or goes away, if you make the sample solvent a weaker eluent by diluting it with water.

It will also become less, if there is enough water in the starting conditions of your gradient to dilute the sample efficiently.

The same is true, if you inject less, since there is always some amount of pre-column bandspreading in any instrument that effectively dilutes the sample, before it reaches the column. A smaller volume is diluted easier.

The problem also depends on the sample solvent. DMSO is worse than methanol, for example.

[xmiao]

Hi, Uwe,

Thank you very much for your reply and very kind explanation. Sorry, I have a following question on the "strong" solvent. How can you define a solvent is "strong" or not? As for the CYP samples that I analyzed, it seems normal and it is ACN: water (30:70) containing some phosphate buffer, which is from the CYP inhibition process and cannot removed by the protein precipitation process. Can phosphate buffer make the solvent "strong"? Again, I greatly appreciate your help.
XM

[mbulentcakar]

Hi Xmiao,

"Strong" solvent is relative. Clearly relative to the mobile phase. If you dilute your sample in your mobile phase, it will be the best injection solvent. When this is not possible the solvent you used should be not too strong relative to your mobile phase. Also when you use a weaker injection solvent you may inject more sample without any peak shape distortions.
For reversed phase, 50% MeOH (rest is water or any buffer) roughly bears same strength with 40%ACN and 30%THF.

Good Luck

[Peter Apps]

Hi XM

Is it possible that your analyte has two optically active centres ? If it has it may form diastereomers that will separate in a non-chiral separation, usually to give two peaks that are similar shapes.

The effects of injection volume, injection solvent etc usually give distorted and asymmetrical peaks.

Peter

[xmiao]

Thanks to mbulentcakar, Peter and Uwe.

Actually, the analyte is hydroxytolbutamide, which is a metabolite of tolbutamide used for monitoring P450 enzyme inhibition. There are no chiral centers for this compound, so diastereomers cannot be formed. Also, the injection solvent is ACN: water (30:70), and I had to start from 2% of organic (MeOH) to solve the above double-peak problem. The strange thing is that I never observed any distorted peaks for the internal standard compound no matter what HPLC conditions I used. The internal standard compound was eluted after hydroxytolbutamide. Is this problem compound-dependant, e.g. hydrophobic compounds are less affected by the strength of injection solvents? XM

[HW Mueller]

Well, think about it: The compound with lower retention will partially be washed into the column further than a higher retained species by a diluent that has more retentive power than the mobile phase.....