Gas Chromatography-Peak areas

[mandytan]

Using a GC-FID (Gas Chromatography-Flame Ionization Detector) and a capillary column, an experiment was carried out. It was noticed that all the peak areas of the blank (ethanol), standards and sample were decreased significantly when compared with a previous analysis. (i.e. previous peak area~200 versus current peak area~ 50).Could it be that the decreased peak areas are due to the column not being installed correctly?
Although all the peak areas were decreased, by area normalisation, we could still obtain satisfactory results for our sample.
In your opinion, are these results considered acceptable?

[Peter Apps]

Poor column installation is certainly one possible explanation for the smaller peaks.

I would treat the results with suspicion unless the GC-FID in its "altered state" could be shown to be sufficiently accurate and repeatable. "Satisfactory" results for samples do not establish analytical performance.

A "satisfactory" result would only be correct if the sample itself is "satisfactory". How do you know that whatever changed in your GC-FID is not yielding a "satisfactory" result for an "unsatisfactory" sample ? And if you already know that the sample is "satisfactory", why bother to analyse it ?

Peter

[Don_Hilton]

With peak areas low, you could also have a worn septum. If the septum is leaking, I would expect the split generated by the leak to change injection to injection - and discrimination between compunds may change as well.