120% recovery for LOQ (Matrix effect)

[Mike H.]

Has anyone observed a matrix effect occur only at the LOQ? Background: I made a GC method of aniline (actually aniline.HCl since our compound is an HCl salt). At the LOQ I get a 120% recovery of aniline, but at other points, I get 98% recovery. Besides potential error, what could cause this?

Note: Standard prep is in MeOH alone while the sample prep involves quantitative addition of NaOH to convert to plain aniline from HCl salt. I have also observed that addition of a small amount of NaCl to standard boosts the aniline response, but only at the LOQ point. What gives?

[James_Ball]

Has anyone observed a matrix effect occur only at the LOQ? Background: I made a GC method of aniline (actually aniline.HCl since our compound is an HCl salt). At the LOQ I get a 120% recovery of aniline, but at other points, I get 98% recovery. Besides potential error, what could cause this?

Note: Standard prep is in MeOH alone while the sample prep involves quantitative addition of NaOH to convert to plain aniline from HCl salt. I have also observed that addition of a small amount of NaCl to standard boosts the aniline response, but only at the LOQ point. What gives?

We normally think we are at the LOQ when we can quantitate within 50-150% so 120% should be decent accuracy at that level. When the calibration standard and the LOQ check are prepared differently then I would think 20% could easily come from the difference in prep and matrix.