HP GC-MSD in SIM mode: Baseline noise reduction?

[jdaknis]

I am attempting to quantitate trace levels of chloroform via GC-MSD. I'm using a 6890 GC w/ HP5973 MSD in SIM mode monitoring m/z = 83 (+/- 0.3 amu). Dwell time = 500 ms.

The chromatogram below is from a standard injection, chloroform concentration in DMF = 3.7 ng/mL. Chloroform r.t. = 2.24 min.



My first question is one concerning the baseline noise. Is this pattern typical when performing trace analysis w/ SIM? Can anyone suggest ways to minimize the effect of such noise on my integration? (In case you haven't guessed, I'm fairly new to SIM!)

Background: The GC method I'm running was validated for GC-FID w/ a LOQ of 95 ppm. W/ MSD, I'm getting %RSD < 3% for 6 replicate injections of 7.5 ng/mL standard and linearity is good (R^2 = 0.998 over 3.7 - 11.2 ng/mL range).



The problem I face is that my sample solutions appear to contain less than my lowest standard concentration. (Sample peak areas range from 120 to 218). S/N (pk-pk) = 11.5 for my 3.7 ng/mL standard, but only 1.7 for my least-concentrated sample.

Am I simply pushing the LOD for this system, or is there a way to further increase the sensitivity?

Any thoughts would be greatly appreciated. Thanks in advance!

FYI:

2 uL injection
20:1 split (Helium, split flow = 40 mL/min)
Column: Restek RTX-5 30m x 0.32 x 1 um film

[JI2002]

Could try m/z 85 and see if that give you a better baseline. Not sure why you are using split injection when you have sensitivity issue. Might try splitless injection.

[chhubert]

I guess you use a fairly large dwell time. Dwell time determines the no. of data to construct a chromatographic peak. Normally a data point of 10-15 is required to construct a peak for accurate quantitation. To know whether sufficient data point for the peak, determine the scan number at the start and end of the peak. The difference, i.e. the number of scan is the data point of the peak. In normal instance, I use dwell time of less than 100ms.
For best sensitivity, in the GC side, splitless injection is definitely a better choice than split mode.

[Okkie]

If already mentioned tips are not sufficient.
You can try a standard addition method.
Okkie