A method for the simultaneous determination of 21 elements in high purity silicon was developed by activation analysis. Extraction,
ion exchange, distillation and precipitation techniques were used combined with NaI(Tl) and Ge(Li) γ-spectrometry. Attention
was paid to the separation time, sensitivity and selectivity of the method.
32P was separated from the irradiated silicon matrix as ammonium phosphomolybdate and the activity was measured with a GM tube
or on a plastic scintillator. The second-order interference was taken into consideration.
The single comparator method has been extended to a triple comparator method, using60Co,114mIn and198Au. In this technique, thek-ratios of the elements to be analyzed, now determined against the three comparators, are corrected for each new ratio of
thermal to epithermal reactor neutron flux. These flux ratios are calculated from the absolute activities of the three comparators.
The thermal neutron activation cross-section and the resonance integral for the reaction113In(n,γ)114mIn have been determined.
A fast (10 min), non-destructive simultaneous determination of silicon and phosphorus in cast iron and steel by 14 MeV neutron
activation was developed. The 1.78 MeV28Al activity (T=2.24 min) induced by the reaction28Si(n, p)28Al is counted on a NaI(Tl) detector. Two measurements are made to correct for the 1.81 MeV56Mn activity (T=2.58 hr) from the iron matrix. However,28Al is also produced via31P(n, α)28Al. By (n, 2n) reaction, phosphorus yields also30P (T=2.6 min), the 0.511 MeV annihilation radiation of which is counted by two opposite NaI(Tl) detectors in coincidence.
Again, two successive coincidence measurements are carried out in order to take into account the53Fe activity (β+; T=8.9 min) from54Fe(n, 2n)53Fe. The28Al measurement is appropriately corrected via the computed phosphorus content. An oxygen flux monitor was used to normalize
to the same flux. Nuclear interferences have been examined. Special attention has been paid to the presence of copper. The
standard deviation for phosphorus being as high as ca. 0.09% P for a single determination, this technique can only be practical
as an independent phosphorus analysis for high phosphorus cast irons. The precision on the28Al measurement is 5% relative for 0.2% Si and 2.5% above 1% Si.
The triple comparator method is used for the analysis of impurities of high purity silicon by neutron activation. The ratios
of the specific photopeak activities of the isotopes investigated to the specific photopeak activities of the gold, indium
and cobalt comparators were determined. The triple comparator method avoids some tedious problems in the multi-element activation
analysis and it is very well suited for the determination of ‘non-expected’ elements.
An analysis has been elaborated to determine the long-living γ-emitting fission products of uranium. It consists of a sodium
bisulphate melt of the fission product solution or the U-fuel, followed by liquid-liquid extractions. Afterwards the isotopes
are absolutely counted with a standardized 3″×3″ NaI crystal. The total γ-spectrum of the original fission product solution,
taken with a NaI crystal or a Ge−Li detector, is also analyzed mathematically by mixed γ-spectrometry. From a short post-irradiation
of the fission product solution the concentrations of both235U and238U are determined. The absolute amount of fission products related to the U-concentration allows the calculation of the percent
atomic burn-up, the irradiation time, the cooling period, the flux of the reactor and the original degree of enrichment of
The neutron flux distribution in the vicinity of 30, 20 and 10 mm diameter targets is measured by irradiating concentric ring-type
iron monitors at different distances from the target and counting the induced56Mn activity. Considering the many uncertainties, satisfactory agreement was found between theory and experiment.
The effects of inaccurate sample sizes and sample positioning on 14 MeV neutron activation analysis results are estimated
for 30, 20 and 10 mm diameter targets. It appears that axial positioning is the most critical parameter and that using a larger
tritium target will yield an overall improvement of the reproducibility.