Big GCMS problem Help needed!!

[FIDER]

Hi everybody!

I am new in GCMS and analysed standards of compounds from different chemical groups (sterols, n-alkanes) and I obtain the same problem in precision. In a series of injections (n=6) RSD% is often 5% to 10% but sometimes I experience drift of one sample of a value 15-20%. Is it normal for GCMS (Agilent 7890A/5975)?
A have checked autosamler, injector (split/splitless and cold on column is the same), column, MS calibration, try to analyse in scan and sim and still the same problem. When I analysed the same on GC/FID everything was OK. I have two GCMS's (new) (Agilent) and both behave similar.
Do you experience similar problems? Maybe it is a matter of proper software configuration. What shoud I look for in GCMS chemstation. Agilent users please help .

Greetings!!!

Fider

[Amirav]

Fider
What is the size of your hydrocarbons, is the problem bigger with the heavier and lower volatility samples?
One possibility is that you have ion source induced peak tailing (unlike in FID) which behaves as an iceberg so that a small tail results in a large TIC signal loss and high RSD. At low ion source temperatures each injection leaves some minor deposit from the sample and column bleed at the ion source which affects the peak tailing and TIC area of the next sample.
The best parameter to explore is the ion source temperature (what is it). Raise it by 50C and your RSD and problem should be lower (unless for ion source degradation for the sterols).
How important is <10% RSD for you
Aviv

[FIDER]

Aviv, thanks for reply!!

Indeed heavier compounds have higher RSD and little tail. My present Ion Source and analyzer temperatures were 260 and 170. I analyze cholesterol in isotherm 270C so I will raise to 290 (max 350) and 190 (max 200) (what shoud be proper analyzer temp?) and watch the diffference. My target is to obtain parameters that beetween two injections have always 5% or lower difference (need low method uncertainty).

Fider

[Amirav]

Fider
The mass analyzer temperature should be irrelevant to the RSD thus can be as before 170C. The ion source temperature can be raised and it should help with the hydrocarbons TIC and RSD but you will lose molecular ion abundance. Cholesterol has another problem in that it degrade in the ion source to cholestane as observed by having too high ratio of m/z=368/386 (over the library) and also bigger tail of the 368 than of the 386. This could prevent you from having low RSD. (I worked a lot on Cholesterol)
If low RSD are essential for you I could offer you some further advice if you write your Email for off-line communication
Aviv

[FIDER]

Aviv,

I was hoping to avoid derivatisation step so cholesterol was not degraded in inlet and column. Could I do something with the Ion Source to minimize decomposition?

e-mail: fider@onet.eu

Fider

[sepscientologist]

If your FID is working well why don't you just use it? I typically do dual injection onto FID and MS. I use the MS for qualitative
and FID for ALL quantitative analysis. The FID is far superior in terms of precision and dynamic range.

[PaulDDodson]

Fider,
Without knowing your parameters I would have to assume that you have a high enough injector temp otherwise you'll get mass discrimination because the lower molecular weights boil at lower temps and will leave the inj. before all get on the column. You also mentioned that the FID gave a good %rsd but remember that FID's have a response factor (RF) because different compounds resond differently. -Paul

[FIDER]

sepscientologist

Unfortunatelly, FID and MS are installed in different GC systems so I could not do dual injections. I assume that if you quantitate by FID, your precision in MS in rather high?

Fider