Reproducibility

[Trishia]

I'm running a Berger SFC but it's a DAD Agilent detector so I figured this question would get more input this way...

I ran the same sample for time (sample containing two compounds);

Peak 1 Peak 2 Ratio
1 1237 3663.4 0.33766446
2 887.3 2674.1 0.33181257
3 847.5 2520.6 0.33622947
4 791.1 2348 0.33692504

RSD (%) = 0.78

So my RSD is fairly good (not surprising since I'm considering the ratios)... But are the numbers themselves too different from one run to another? I mean, we are using an autosampler (Alcott), with a fixed loop system (loop is 5 uL)... Is it normal that the areas vary so differently from run to run?

Thanks!

Trishia.

[Mark Tracy]

The areas do seem to have excess variability. A fixed-loop injection should give no worse than 2% rsd and more typically 0.5% rsd. A couple ideas: the loop needs to be overfilled 3x for best results; how much is actually dispensed? Air in the syringe leads to poor reproducibility, especially for viscous samples or rapid syringe draw speeds. I don't recall the design of the Alcott so this may not apply, but the volume between the needle and loop may be incorrectly calibrated. If the waste tubing is too far above or below the needle port, the sample can siphon out.