* Cleaning techniques used: inductive heating, solvent cleaning, and heating in a reducing Bunsen burner flame (refs. 100, 101).
Electron energy in particular proved to be a highly critical parameter. Changing the electron energy settings from 14.0 to 14.1 eV caused more than a 10% difference in average relative peak intensities. Since actual electron impact energies are not readily defined to within a few tenths of an electron volt and changes of that magnitude might easily be caused by contamination of the ion source or by electronic instabilities during pyrolysis, fluctuation in electron energy may be a major factor limiting long-term reproducibi1ity. Other possible troublesome factors are adsorption phenomena on the walls of reaction tube and expansion chamber, as well as changes in ion energies and mass resolution by contamination of ion source and quadrupole rods. Controlled heating of the ion source could reduce these problems. In view of all these possible or actual sources of variation, the degree of inter-laboratory reproducibility achieved with instruments showing considerable differences in ion-source and quadrupole design is remarkable, as will be discussed in the following paragraphs.
Was this article helpful?