Diethylene glycol in glycerin

[JamHPLC]

Working with the USP 28 GC procedure I found the following problem: The signal obtained from an spiked solution of glycerin with diethylene glycol is about ten times the signal from a pure solution of diethylene glycol of the same concentration. The glycerin itself is free from diethylene glycol. The procedure uses a special liner ( inverted cup or speril liner) and I haven´t it. Could anyone help me with my problem?

[Consumer Products Guy]

My company runs the USP glycerin method, and uses the mini-Lam split liner (Restek #20990) or a spiral type cyclosplitter (Restek #20706). USP specifies these, doesn't offer up a choice. We have never spiked a sample with DEG, just followed as USP details, convoluted as it is.