EVALUATION OF THE PRECISION AND ACCURACY OF A URANIUM ISOTOPIC ANALYSIS USING GLOW-DISCHARGE OPTOGALVANIC SPECTROSCOPY

The measurement precision and accuracy were evaluated for an isotopic analysis of uranium using diode laser-excited optogalvanic spectroscop y. The ratio of U-235/(U-235 + U-238) was measured using the 776 nn ur anium isotope line for five samples ranging from depleted to enriched U-235 composition. The percent relative error (accuracy) with respect to thermal ionization measurements ranged from 2.38% for a nominally 2 0% enriched U-235 sample to slightly greater than 30% error for a depl eted one. Run-to-run, day-to-day, and sample-to-sample precision (repe atability) were measured using an enriched sample; the run-to-run prec ision ranged from +/-1.9% to +/-5.5% RSD, and the day-to-day precision was +/-2.6% RSD. The sample-to-sample precision, determined using thr ee different cathodes, was found to be +/-13.5% RSD. The ratios showed no trends or biases, varying about a mean of 0.21 and ranging from a high of similar to 0.23 to a low of 0.18. A search was conducted for a stronger atomic Line and a matching higher power diode laser in an at tempt to improve the signal-to-noise ratio for the depleted case. Usin g a 150 mW diode laser that produced nominally 42 mW chopped power at 831.84 mn, a depleted uranium oxide sample was found to contain 0.26% U-235, within 3.7% of the value found using thermal ionization mass an alysis. Using this same laser, the run-to-run reproducibility improved to +/-7.8% from +/-13.6%. This level of accuracy and precision is suf ficient for screening applications, where preliminary information abou t the isotopic composition of a sample provides the incentive for addi tional analysis using more precise techniques.