Simultaneous isotopic analysis of fission product Sr, Mo, and Ru in spent nuclear fuel particles by resonance ionization mass spectrometry.

Fission product Sr, Mo, and Ru isotopes in six 10-µm particles of spent fuel from a pressurized water reactor were analyzed by resonance ionization mass spectrometry (RIMS) and evaluated for utility in nuclear material characterization. Previous measurements on these same samples showed widely varying U, Pu, and Am isotopic compositions owing to the samples' differing irradiation environments within the reactor. This is also seen in Mo and Ru isotopes, which have the added complication of exsolution from the UO2 fuel matrix. This variability is a hindrance to interpreting data from a collection of particles with incomplete provenance since it is not always possible to assign particles to the same batch of fuel based on isotopic analyses alone. In contrast, the measured 90Sr/88Sr ratios were indistinguishable across all samples. Strontium isotopic analysis can therefore be used to connect samples with otherwise disparate isotopic compositions, allowing them to be grouped appropriately for interpretation. Strontium isotopic analysis also provides a robust chronometer for determining the time since fuel irradiation. Because of the very high sensitivity of RIMS, only a small fraction of material in each of the 10 µm samples was consumed, leaving the vast majority still available for other analyses.

Simultaneous Isotopic Analysis of U, Pu, and Am in Spent Nuclear Fuel by Resonance Ionization Mass Spectrometry.

Solid samples of spent nuclear fuel were analyzed for actinide isotopic composition by resonance ionization mass spectrometry. Isotopes of U, Pu, and Am were simultaneously quantified using a new method that removes and/or resolves the isobaric interferences at 238U/238Pu and 241Pu/241Am without sample preparation other than cutting and mounting small (∼10 µm) samples. Trends in burnup and neutron capture product distributions were correlated with the sampling positions inside the reactor. The results show the skin effect, in which the core and near-edge regions of a fuel pellet exhibit strong differences in actinide concentrations and isotope distributions due to differences in the neutron energy spectra between the pellet rim and the core. While no elemental concentration measurements were made, the ability to measure the 238Pu/239Pu ratio in the presence of a 7400× excess of 238U enabled an estimate of the enhancement in Pu concentration due to the skin effect at the rim of the pellet.

New Resonance Ionization Mass Spectrometry Scheme for Improved Uranium Analysis.

Resonance ionization mass spectrometry (RIMS) combines tunable laser spectroscopy with mass spectrometry to provide a high-efficiency means of analyzing solid materials. We previously showed a very high useful yield of 24% for analysis of uranium using three lasers to excite and ionize atoms sputtered from metallic uranium and uranium dioxide. A new resonance ionization scheme using only two lasers achieves a higher useful yield of 38% by accessing both the ground electronic state and a low-lying electronic state of atomic uranium that is significantly populated by sputtering. The major loss channel in analyzing uranium dioxide is the formation of UOx molecules during sputtering. Prebombardment of the surface with 3 keV noble gas ions prior to analysis reduces the surface and results in a sputtered flux with a greatly enhanced proportion of atomic U. This method of surface reduction results in uranium useful yields as high as 6.6% for uranium dioxide analysis, compared to 2% from previous work.

High Useful Yield and Isotopic Analysis of Uranium by Resonance Ionization Mass Spectrometry.

Useful yields from resonance ionization mass spectrometry can be extremely high compared to other mass spectrometry techniques, but uranium analysis shows strong matrix effects arising from the tendency of uranium to form strongly bound oxide molecules that do not dissociate appreciably on energetic ion bombardment. We demonstrate a useful yield of 24% for metallic uranium. Modeling the laser ionization and ion transmission processes shows that the high useful yield is attributable to a high ion fraction achieved by resonance ionization. We quantify the reduction of uranium oxide surface layers by Ar+ and Ga+ sputtering. The useful yield for uranium atoms from a uranium dioxide matrix is 0.4% and rises to 2% when the surface is in sputter equilibrium with the ion beam. The lower useful yield from the oxide is almost entirely due to uranium oxide molecules reducing the neutral atom content of the sputtered flux. We demonstrate rapid isotopic analysis of solid uranium oxide at a precision of <0.5% relative standard deviation using relatively broadband lasers to mitigate spectroscopic fractionation.

Ion microscopy with resonant ionization mass spectrometry: time-of-flight depth profiling with improved isotopic precision.

There are four generally mutually exclusive requirements that plague many mass spectrometric measurements of trace constituents: (1) the small size (limited by the depth probed) of many interesting materials requires high useful yields to simply detect some trace elements, (2) the low concentrations of interesting elements require efficient discrimination from isobaric interferences, (3) it is often necessary to measure the depth distribution of elements with high surface and low bulk contributions, and (4) many applications require precise isotopic analysis. Resonant ionization mass spectrometry has made dramatic progress in addressing these difficulties over the past five years.

Extinct technetium in silicon carbide stardust grains: implications for stellar nucleosynthesis.

The isotopic composition of ruthenium (Ru) in individual presolar silicon carbide (SiC) stardust grains bears the signature of s-process nucleosynthesis in asymptotic giant branch stars, plus an anomaly in 99Ru that is explained by the in situ decay of technetium isotope 99Tc in the grains. This finding, coupled with the observation of Tc spectral lines in certain stars, shows that the majority of presolar SiC grains come from low-mass asymptotic giant branch stars, and that the amount of 99Tc produced in such stars is insufficient to have left a detectable 99Ru anomaly in early solar system materials.