I am currently in the process of validating a GC method that I developed for analysis of residual organic solvents in a final pharmaceutical product. Part of the validation requires a Limit of Detection and Limit of Quantitation calculation. One way to determine the LOD/LOQ is to use a signal to noise ratio. A ratio of 3-2:1 being the LOD and a ratio of 10:1 being the LOQ. I am using the HP Chem Station software which allows me to enter any time range I choose for the s/n ratio calculation. I am unclear on what time range to use. I thought the range should be in the area of the peak(s) of interest. However when I set the s/n time range to encompass my peak of interest the software does not calculate a number for the s/n. If I set the time range to an area that does not encompass the peak of interest I do get a number for the s/n. I am interested in learning the correct way to calculate my LOD and LOQ. Thank you.

You should pick a "peak free" baseline region. The region you pick is for the purpose of calculating the noise only. The software will then use this value to give you a s/n relative to the peak of interest. It would be best to use the lowest standard to determine the s/n. You will want to report the LOD and LOQ (based on your ratios)in the units used for quantitation. The lowest standard will give you the best estimate when using the multiplier method versus using empirical determination of LOD and LOQ.

The EP clearly defines the procedure how to select a range for calculating the baseline noise for establishing the S/N ratio. The lachrome - HPLC software exactly follows the BP procedure