Page 299 - Inorganic Mass Spectrometry - Fundamentals and Applications
P. 299
Analysis of ~oncond~ct~~e Sa~ple Types 285
the R value from dc powering was 98% vs. 75% for the rf. Operation of the rf dis-
charge at higher powers was projected to produce more comparable values. Analy-
sis of a mixture of rare earth elements (REEs) indicated that the degree of atomic
to
ion production was again very sensitive sampling conditions (i.e., position, pres-
surelpower) in the case of rf, and not so much so for the dc case. In fact, R values
between the REEs varied with rf conditions, but not for dc, seemingly depending
on the M-0 bond strengths. Relative sensitivity factors for the rf and dc plasmas
were found not to be appreciably different under controlled conditions, as the rf
source showed very good stability (-5% RSD) and reproducibility (~15% RSD).
As described in Sec. 7.3, which dealt with the use of the secondary cathode
methodology, one of the unexpected areas in which GD-MS may be applied (so
long as the sample can be analyzed intact) is the direct analysis of polymeric ma-
in-
terials. A simple method providing elemental, end-group, or molecular weight
formation could have impact in a number of industrial situations. Coburn and
coworkers actually reported the production of molecular fragment mass spectra
of
fluoro- and hydrocarbon polymers by rf GD-MS [69]. Twenty years later, Marcus
and coworkers noted a number of unexpected spectral features in the analysis of
small glass samples mounted to the DIP via double-sided adhesive tape. On fur-
ther study, it was shown that these species were in fact representative of the com-
position of the tape. This observation has opened up an entirely new line of study
in the use of rf GD-MS as a tool that provides high-sensitivity determinations of
polymer composition.
of
The first studies by this group involved an evaluation discharge parame-
ters and their respective roles the structure and intensity mass spectra derived
in
of
from PTFE-based polymers [70]. The spectra were found to be nearly identical to
those obtained by SUMS, as shown in Fig. 7.9, but with ion beam currents that were
six orders of magnitude more intense vs. A). The polymer fragment
by rf GD-MS.
ion currents are in fact similar to those obtained in metals analysis
Plasma stabilization times of less than 5 min were easily achieved, mdcing the
analysis very competitive with SIMS. In contrast to those in the work of Schelles
and Van Grieken [24], the spectra were rich with molecular fragments as opposed
the
to mainly atomic ions, with the CF3+ fragment in rf spectrum -1000 times more
intense than the C? in the dc case. Discharge power was not seen dramatically to
change the spectral character, only the ion signal levels. On the other hand, changes
in discharge gas pressure, and more correctly the subsequent changes in dc bias,
cause substantial variations in the spectral character. Specifically, high Ar pres-
sures (i.e., lower dc-bias values) are less effective in liberating polyatomic species,
and so the spectra tend to be more atomic in nature. A depth profile of a metallic
layer on a polymer substrate was presented
to demonstrate applications such as the
analysis of printed circuit boards.
Two additional studies have focused on the area of polymer analysis by rf
GD-MS. The problem of sample melting was addressed through the design of a
cryogenically cooled sample holder [7 l]. Just as in the case of Schelles and Van