Chromatography Forum

Hi, I am using 6890/1888 GC and headspace sampler with chemstation b.04.01.
My calibration standards are good and linear when looking at concentrations v area. But the response factors are pretty sporadic, which means none of my samples are getting calculated properly.

What is the source of this and is there a way to fix without redoing every standard? (I have 40 calibrations standards)

Nicoled,

The response factors for FID are, generally, pretty consistent. There are some general exceptions - methanol, chlorinated, etc.... but a wide range of compounds respond in a very similar manner.

So....., I would be a lot more suspicious of the headspace side of this equation than I would be of the FID side of the equation. Then again, if the headspace parameters are consistent, standards to samples should still work out.....

If you can provide it, more information gets a better response.

Best regards,

AICMM

AICMM wrote:
Nicoled,

The response factors for FID are, generally, pretty consistent. There are some general exceptions - methanol, chlorinated, etc.... but a wide range of compounds respond in a very similar manner.

So....., I would be a lot more suspicious of the headspace side of this equation than I would be of the FID side of the equation. Then again, if the headspace parameters are consistent, standards to samples should still work out.....

If you can provide it, more information gets a better response.

Best regards,

AICMM

Hi thank you for your reply!
I am using this method (https://www.restek.com/chromatogram/view/GC_FS0518)
My response factors across my calibration range are like this for most standards:
std 1 (5ug/mL) RF = 5.89495
std 2 (100ug/mL) RF = 9.44077

std 4 (2000ug/mL) RF = 15.13773.

I had been using a calibration range that had been decided by a different analyst, but the more I think about it, the more likely it seems that the response factors are so different because of the huge range of calibration points. I am narrowing them down now (2-250ug/mL now, with more points). Does this make sense?

" I had been using a calibration range that had been decided by a different analyst, but the more I think about it, the more likely it seems that the response factors are so different because of the huge range of calibration points. I am narrowing them down now (2-250ug/mL now, with more points). Does this make sense? "

Yes, I think you are correct. I graphed your numbers and got a Corr Coeff of around 0.9999, which sounds good. I think that is because you only have a 3 point calib curve and the range is large. A 2 point calib curve will give you a Corr Coeff of 1.0000. Adding another data point does not affect the Corr Coeff that much with a large calib. range. A smaller range with more points should definitely help. It goes to show that Response Factors can give you more information.

Dave