Statistics.
reported results and errors are those
For those nuclides with
multiple sample determinations,
of the simple average.
the
Quality-Control Program
SAMPLE ALIQUOT ©
Results from dissolution of the
25-g aliquots are compared in Table 7
The usual MCL quality control
with those from the smaller aliquots.
program was expanded to examine the
The ratio of results from carrier
validity of wet-chemistry analysis
to carrier-free dissolutions is
of small sample (3 to 5 g) aliquots.
given for each of the atom ratios and
Specifically questioned was whether
isotopic concentrations.
the small aliquots were representative
reported with each entry result from
of the larger sample.
propagation of uncertainties in the
homogenized samples,
For wellthe small ali-
individual measurements.
quots were known to be representative.
——
Twenty-seven samples (5 vegetation
Errors
Soil samples
from both Bikini and Eneu show
excellent reproducibility in all major
and 22 soils) were selected for
isotopes.
carrier-free dissolution of 25-g
spectrometric measurement of minor
aliquots.
isotopes such as
The term carrier-free
Past experiences in mass
tou would lead
describes working solutions obtained
one to expect an even greater spread
by the dissolution of sample in the
than that observed.
absence of appropriate carriers and
results are reproducible to within
tracers.
10 percent is quite encouraging.
The major difference between
The fact that the
For
the carrier-free and standard carrier
the soils there is no indication of
dissolutions was the absence or
a statistically significant bias
presence of the carriers and tracers
resulting from the use of small
in the working solution.
Processing
24
of 70.4 r, Pu and
Tam samples from the
(5 g) sample aliquots.
carrier-free solution was accomplished
of samples of vegetation indicate
by adding Y carrier and
2
tracer to one aliquot and
Comparison of the measurements
Pu mass
the possibility of a slight bias
Am to
in the dterminations of 990, and
43
another, achieving isotopic exchange
239,240), ..
in the solution, and proceeding with
are not significant at the 95 percent
the standard methods for separation
confidence level.
and purification.
definite indication of significant
-13-
5010055
however,
these biases
There is a