trouble with standard recovery

[Annemarie]

Hi,
I'm having an issue with my GC. It's an Agilent 6890 FID. I am analyzing for ethanol and methanol in water samples. I normally run a set of standards at the beginning and at the end of each run. Usually, the end standards are just a little low from sitting on the sampler after they have been pierced. No problem. Recently, I have been getting low recoveries (90%) on the beginning standards and close to 100% on the end standards. Why are my initial standards so low? I change the liner, septum and o-ring before every run. After the issue began, I clipped the guard column (no improvement) then I changed the gold seal-still no improvement.
Does anyone have any ideas?
Thanks for your help.
Annemarie

[chromatographer1]

Your initial std areas are low. THUS

Less material is getting to the detector: THUS

1. absorption sites are removing some of the analytes before they get to detector
or
2. the amount being injected is incorrect initially.
or
3. your first std prep is made or sampled incorrectly.
or
4. your split ratio changes after the initial injection of standards.
or
5. the age of Aquarius ended with Jupiter crashing into Mars.

Which is the correct answer?

That is the question you should ask, and how to fix each situation if it is true.

1. replace injection liner with clean one. replace part or all of column. check reproducibility of std if several injections of 1/4 the amount is performed (0.25uL instead of 1.0uL)
2. check syringe for leaks and for the amount which is taken up if autosampler is used, and the amount left in syringe after injection
3. run same vial before samples and after last std for check of duplication of std vial results, also run several stds or std injections initially to see if immediate changes are seen
4. check column flow rate and splitter flow rate before and after each injection. clean and or replace pneumatic lines/traps, check septum and septum purge flow rate
5. drink tea of bat eyes and lizard tails and dance the dance of death to scientific error demon

Good luck, Don't tell me if #5 is the correct answer.

Rod

[Don_Hilton]

...in water samples.

Water has a high expansion volume in the GC inlet - and is not my favorite GC solvent for that and other reasons.

What volume are you injecting? What pressure and temperature are the inlet? What is the volume of the liner you are using? Have you changed inlet temprature, or made a change that would affect the inlet pressure? Have you changed from one type of liner to another? --- these questions are to check for overfilling the inlet liner upon injection and changes that might lead to changes in discrimination.

How have you standardized the system to measure recovery and how are you calculaing recovery?

What kind of detector are you using?