By Anonymous on Thursday, February 5, 2004 - 05:02 am:
I'm trying to report an LOD for an LC/MS system and was wondering what the typical protocol is. My peak elutes at about 5 minutes and is 15 seconds wide. in order to determine the SD of the blank (and then the LOD), I've looked at a 15 seconds @ 5minutes. I then plot out signal at m/z 506. My problem is that the blank yields virtually no signal and so I divide zero by the slope of the calibration curve. Thus my "LOD" is zeptomolar and I know thats not right. Is there something Im doing wrong? Should I just report the lowest standard I can see (50 nanomolar)?
Thanks
By Anonymous on Friday, February 6, 2004 - 06:36 am:
Hello,
standard deviation of residuals in your calibration curve is a good estimate of SD of the blank and intercept value of calibration curve is an estimate of blank value.
With repeated injections of your lowest concentration you can obtain a standard deviation for it. If relative SD (C.V.%) is <10% 50 nM is your LOQ (limit of quantification)