Accuracy

[mbdixon]

Hello

I'm running a method which measures derivatised olive oil by using an internal standard ratioed to the C16 & C18 peaks. Now I'm about to inject samples that will have much larger results than normal (at least 2x) so I’m thinking of either injecting less sample or increasing the split at the injection to keep the peaks within the max voltage reading on the FID. I normally inject 1µl and have a split ratio of 50:1. Which would be more accurate. I would guess injecting less but is there a rule of thumb/calculation to follow with this sort dilemma?

[rb6banjo]

I would increase the split ratio. If you inject less, will be moving to a place where the syringe is less capable of accurately delivering the sample to the chromatograph. It's much easier to throw more of the sample away out the split vent. Good luck!

[Paulhurleyuk]

Is it possible to dilute your sample in some solvent that doesn't interfere ?

[chromatographer1]

Paulhurleyuk has the better idea.

Given the variables it is much better to dilute your preparation by half accurately than trying to change the split ratio which may change the discrimination or by varying the sample size.

best wishes,

Rod

[Peter Apps]

Whichever route you go, you need to calibrate the peak areas under the revised conditions. 2 x is not much of a dynamic range and if it gives you problems you are probably routinely working too close to the upper end of the calibration or the robustness of the method.

Peter

[mbdixon]

In the end I simply injected 0.5µl instead of 1µl and it worked fine. Cheers for the advice.