can PTV injection lead to false results?
Posted: Thu Apr 06, 2006 8:37 am
Some time ago, an external company performed a batch analysis with a GC method provided by us. They made two important deviations from that method:
* they used a PTV injection (split solvent vent; 60°C to 250°C hold 1min and from 250°C to 350°C hold 10min) instead of classic split injection at 260°C mentioned in our method
* no internal standard was used, although mentioned in our method
Our results from that batch give a purity of 97+%, whereas their result only give about 95%. Of course, now there is an argumentation going on between them and us. They claim the mistake is our analysis or another impurity which was not tested, and we claim to deviations are due to their change of method.
One other big issue is the fact that their analysis only yield a total concentration (purity + impurities) that is less than 98%.
The purity that had to be quantified is stable until 285°C and is known to stick in GC injection systems.
Because they used a PTV starting at 60°C instead of a split injector at 260°C, could it be that the sticking of the compound is more pronounced? They claim that test and calibration samples were injected using the same conditions and therefore could not be the reason for the lower purity. But what about the reproducibility of the sticking to the injector?
Since the purity is only stable until 285°C and they heated to 350°C, can it be that part of the purity is degraded before it entered the column?
If anyone could provide me some comments and/or suggestions regarding this, it would be highly appreciated.
thanks a lot in advance!
Regards,
Kurt
* they used a PTV injection (split solvent vent; 60°C to 250°C hold 1min and from 250°C to 350°C hold 10min) instead of classic split injection at 260°C mentioned in our method
* no internal standard was used, although mentioned in our method
Our results from that batch give a purity of 97+%, whereas their result only give about 95%. Of course, now there is an argumentation going on between them and us. They claim the mistake is our analysis or another impurity which was not tested, and we claim to deviations are due to their change of method.
One other big issue is the fact that their analysis only yield a total concentration (purity + impurities) that is less than 98%.
The purity that had to be quantified is stable until 285°C and is known to stick in GC injection systems.
Because they used a PTV starting at 60°C instead of a split injector at 260°C, could it be that the sticking of the compound is more pronounced? They claim that test and calibration samples were injected using the same conditions and therefore could not be the reason for the lower purity. But what about the reproducibility of the sticking to the injector?
Since the purity is only stable until 285°C and they heated to 350°C, can it be that part of the purity is degraded before it entered the column?
If anyone could provide me some comments and/or suggestions regarding this, it would be highly appreciated.
thanks a lot in advance!
Regards,
Kurt