I am trying to set up a GC/MS/MS method to quantitate metoprolol. I tried J&W column 15 m, 0.25mm ID, 0.25 um film with temp from 180 to 220 oC. Splitless inj of 3 uL with injector temp set at 250 oC. The SRM for MS/MS is 268 -->191 Positive CI mode with methane as reagent gas. No derivatizing was carried out. Two problems emerged:

1. The peak splits: A sharp peak followed by a broadened peak (they are barely resolved!). It could not be due to impurity, since another analog compound used as an internal standard has the same problem.

2. The sensitivity is extremely low. I can not see anything even at 10 ug/mL in methanol. What I got is about 100 ug/mL with poor chromatography as described above.

Anyone has suggestions? Thank you.

Mary Wanzit

Are you doing split or splitless injection? If you are doing splitless injection, then I would have to say that 3 uL injection volume is pretty large and you may be experiencing a backflash problem. What is your solvent? Try some injections with only 1 uL to see if the splitting goes away.

Could also be some type of focussing issue. Again, what's your solvent and also what phase are you using?

What type of liner do you have installed (straight tube, single or double taper)? Does it contain any glass wool?

Jason Ellis
Technical Support Engineer
Agilent Technologies

You are using a solvent with a large expansion volume, and with the short column you are using the head pressure is probably very low, making the solvent expansion even more of a issue. I would agree with Jason that backflash is a very real possibility. Even at 2 ml/min your head pressure would only be 14 psi.

The fact you are seeing poor response at high concentrations indicates that either the MS has no sensitivity or that analyte is not reaching the MS.

Methanol and water have extremely large evaporation volumes. Therefore, they should be avoided whenever possible to avoid backflash.

Thanks for your information.

I was on vacation for a few days, and I got back today and I am pleased to see so many responses to my post.

After I posted the message, I changed the solvent to toluene, and retuned mass spec. I could see 1 ug/mL. But I really want to have my lowest point in the calibration line be at 1 ng/mL. It seems that there is a long way to go.

What I used is splitless injection, the column pressure is 10 psi (the flow would be estimated at ~ 2 ml/min), the solvent used is toluene, column used J & W DB-17, 15 m, 0.25mm ID, 0.25 um film. The liner is straight tube with NO glass wool. I will try a larger liner. Currently, I used ThermoFinnigan Trace GC with 3-mm ID liner, and I will try the 5-mm ID liner. (However, I used the 3-mm ID liner and injected 2 ul of another compound in toluene with another project works extremely well). I can also try I ul injection as suggested by Jason.

I do not have a lot of experience with GC. Thanks for your help.

Mary Wanzit

Try running a split injection and see how the response compares with the splitless injection. I have done this in the past when the column length and id lead to a headpressure that is too low and allows to large a solvent expansion. If you get equal or higher response running a 20:1 split compared to splitless, there is a solvent expansion problem. I have done this at locations where inappropriate columns have been used to determine that the expansion volume is really the problem.