Decreasing Internal Standard recovery in method 8260

[pacerlaser]

Hello everyone, I have noticed that after calibration, the internal standards for method 8260 will decrease in response rather quickly. The response for fluorobenzene will be in the neighborhood of 450,000 for the midpoint of the calibration but will drop to 280,000 after about 10 analysis runs. I am using a 5973 MSD and a 6890 Gas Chromatograph. Has anyone else seen this in their analysis? I am curious as to what to try to remedy the situation. Thanks for the help!

[Bigbear]

How long has it been since you cleaned the source? I can only run 200-300 samples using EPA 524 between cleanings.

[pacerlaser]

The source was just cleaned recently (within 1-2 weeks)

[willnatalie]

Is there anything else occurring as well within the other peaks besides the IS?

Have you injected the calibration sample 10 times to see if the same thing occurs?

the next time a sequence is started all is ok again?

I cannot see how the MSD (with a clean source) would drop the signal as such especially if this is a reoccurring issue.

[Steve Reimer]

I have heard of this, especially with 5975s, where there is too much water getting to the source. Is your desorb time to 1 minute or less and your split ratio 50 to 1 or greater?

[LindseyPyron]

The problem you are seeing is due to water in the source. At the beginning of the day your response is high because your pump has had time to catch up overnight. As you run through the day, your response drops. You will need to look at increasing your split ratio and decreasing your desorb time to 0.5-1minute. If this does not help, let us know. There are a list of other things to try to optimize the system.

[pacerlaser]

In the past we have used a 4 minute desorb time and have since decreased it down to 2 minutes. Auditors in the past have cited method 524.2 as requiring a 4 minute desorb time however we were seeing the dropping of internal standard area counts and this was the reason for descreasing the desorb time down to 2 minutes. Most interesting is the first run the following morning, the internal standard area does not increase to the level of the calibration but still stays low. For example, if the IS area during the calibration was 400,000 and we run about 30 to 40 samples with the ending samples having an IS area of 250,000 and come in the next day to do a tune, the IS area will still be in the neighborhood of about 250,000-275,000. We are beginning to think the analyzer needs to be replaced. Thanks for all the great suggestions/help, it is much appreciated!

[queenann]

Interesting info. Wished I knew of it back when I did 8260. We just replaced the multiplier.

[pacerlaser]

Do you remember what kind of multiplier you replaced it with?

[LindseyPyron]

If you want to give our tech support guys a call, they can help you with a few ideas....no charge! 800-283-3210.