Page 29 - Inorganic Mass Spectrometry - Fundamentals and Applications

lMass Spectrometry
a
~
~ ~ e Ioni~ati~n 19
estimate of the amount of the target element in the sample, information not always
available. !
Another way of addressing variations in bias is throughthe use of an internal
calibration ratio. In this technique, there are two isotopes of the analyte element
whose ratio is well known that can be used to calculate the specific bias applicable
to the analysis in question. It has long been used in geological applications; in
strontium analysis, for example, the 86Sr/88Sr ratio is invariant in nature and can be
used to calculate the bias necessary to apply to conect the 87Sr/@Sr ratio [2]. Dietz
et al. were the first to suggest use of internal calibration through addition of a
two-isotope spike to the sample [61]; they tested it with uranium. If the ratio of the
two reference isotopes is well known, it can serve as the comparison necessary to
effect internal calibration. This ratio is measured for each run and its value
compared to the known. The bias factor required to bring the measured ratio into
agreement with the known is calculated and applied to all other ratios. The theory
underlying internal calibration has been described by Dodson [62,63]. It has been
applied to uranium [64,65] and to molybdenum [66], plutonium [67], and lead
[68]. In principle it should be applicable to concentration measurements of any
element with three isotopes, two for the spike and one for the sample; it can also
be used to refine the value of a ratio of two isotopes, both of which are either ab-
sent or present only in low abundance in the spike. One of the most attractive
features of the double spike is that it should produce results independe~t of
individual laboratory calibration methods. Instrumental bias is one of the major
causes of disagreement between laboratories, but use of the same double spike
requires that all laboratories make corrections based on the same isotopic ratio
independently of differences in their methods. This matter was addressed in a
recent study [69].

A long-standing and still-current challenge in thermal ionization mass spectrome-
try is to improve ionization efficiency. This is usually defined experimentally as
the ratio of ions collected to the number of atoms loaded; it thus includes all
aspects of the* ionization, extraction, transmission, and collection processes.
One obvious way to improve ionization efficiency is to make sure the
sample is as clean as possible. A heated filament provides a constant amount of
energy, and any devoted to evaporating or forming ions of contaminant species is
lost to the desired process. Sodium, potassium, calcium, and other readily ionized
elements are bad actors; the fact they are also ubiquitous makes the problem just
that much more difficult. Every element presents its own challenges, and much
effort has been invested in purifying target elements of interest. Loading a chem-
ically pure sample on the filament is one way to improve ion emission.