GC-MS Where are the missing ions?

Chromatography Forum: LC-MS & GC-MS Archives: GC-MS Where are the missing ions?
Top of pagePrevious messageNext messageBottom of pageLink to this message  By wilson on Monday, June 16, 2003 - 10:19 am:


I am trying to run EPA method 524.2 (drinking water) that uses 4-BFB to check the tune. GC is 5890 with EPC, with a 5972 MSD. In the literature we found instrument settings that would speed up the run time. This included turning up the column pressure by a factor of about 3 to 37 cm/sec. After doing this, the 4-BFB QC checks will not pass. As you scan across the peak, the ion ratios change dramatically (okay), but at the apex, some even disappear. The 4-BFB is introduced with manual injection to column. We've played with the sampling rates, etc. With no luck. When we return to the previous conditions, the problem goes away. Anyone have any suggestions? The literature method was using a 5973 MSD, and some here believe we are in a 'you just can't do that with our mass spec' situation. I don't know enough about the differences between the two detectors to know...but I hope it's not true.

Top of pagePrevious messageNext messageBottom of pageLink to this message  By Anonymous on Monday, June 16, 2003 - 05:30 pm:

Did you do the BFB autotune for your HP5972MSD under the same column pressure as you run the samples? The tune performance is greatly affected by the vacuum, or in turn the column flow rate. You may find a significant increase of EMV when high column flow rate is used (especially when you are using diffusion pump instead of turbo pump).

Top of pagePrevious messageNext messageBottom of pageLink to this message  By Anonymous on Wednesday, June 18, 2003 - 11:51 am:

sounds like the sensitivity increased. Lowering the concentration may help in both cases.

Other than lowering sampling rate, I can't think of anything else for the spectral tilting problem.

as for the missing ions...

data dumping (I don't think this is the official term for it), usually happens when the ion detected is very abundant and it doesn't collect data for the next ion (one lower in m/z). I saw this alot with the 5972, and since it could be explained, wasn't usually a problem for identification. When it was a problem, lowering sample concentration always helped.

5973 detectors I've seen are quite different than 5972 in two ways. One, the 5973 is usually a network connection to the computer which can handle much faster data transfer. could that reduce missing ions???

Other main difference is 5973 has a different tune setting which is more sensitive for higher ions. The 219 in 4-BFB is about 130% abundance to the 69 m/z. Though the 5973 has a way of doing the old autotune way (with 219 being about 50% of 69m/z).

Perhaps you can speed up run time with a shorter column than with faster flow rate.

Most column manufacturers have a column "hotline" that can give expert suggestions on a better column for your needs.

hope some of this helps.

Top of pagePrevious messageNext messageBottom of pageLink to this message  By Anonymous on Thursday, June 19, 2003 - 07:26 am:

You don't really give enough information to give a complete answer, but the problem is most likely flow related. The 5973 with a turbo pump, especially if it is a high end turbo, will handle high column flows better than a 5972 with a diffusion pump. The 5972 can only handle 2 mL a minute of column flow, and if that high a flow is used the sensitivity is decreased due to source design and lack of pumping capacity. The mass spectra will also be distorted, and I am guessing the ions you see disappearing are the isotopic peaks at low abundance. The loss of sensitivity from overpressuring the system is probably responsible for this.

If you want to speed up the run the best way to do this is to use a column with a smaller id. This will speed up the analysis while still keeping the column flow rate at 1 mL or minute or less, which is the flow the 5972 works best at. I would suggest either a 0.2 or 0.18 mm id column, both of which should work quite well in your instrument.

The 5972 is a better instrument for volatiles analysis than the 5973 due to the tuning issues mentioned above. The 5972 tunes much closer to literature ion ratios for PFTBA, so it is easier to meet and hold tune critera with a 5972.

A final question, how much do you need to increase the speed of the analysis? Right now your purge and trap cycle time is probably about 20 to 25 minutes per sample. If you cut your GC cycle time down to 15 minutes per sample the GC will be sitting there waiting for the purge and trap to finish processing the next sample. The ideal situation is to get the GC method and purge and trap method set up so the cycle times match up and the GC comes ready as the purge and trap comes ready to desorb.

Good luck, and if these posts do not help please give additional information such as column used, dimensions, flow rates, temperature profiles, etc.

Add a Message

This is a private posting area. A valid username and password combination is required to post messages to this discussion.