Post-crystallization alteration of natural uraninites: Implications for dating, tracing, and nuclear forensics

Abstract

Abstract This study presents systematic chemical (U, Pb, Ca, Si, Fe) mapping coupled with in situ analyses of major, minor and trace elements, U/Pb, 207Pb/206Pb, and O isotopic compositions of natural uraninites (UO2) from two samples of the high-grade uranium ore from the Cigar Lake unconformity-related uranium deposit (Athabasca Basin, Saskatchewan, Canada). The studied uraninites are characterized by major chemical and isotopic heterogeneities expressed at small scale (µm to tens of µm), from almost pristine zones to strongly altered material. The 206Pb/238U and 207Pb/235U ratios of the different areas are widely spread and depict two similar and well-defined Discordia, providing an upper intercept age of crystallization at ca. 1300\u202fMa (1299\u202f±\u202f4 and 1308\u202f±\u202f14\u202fMa, respectively) and lower intercepts at 38\u202f±\u202f13 and 72\u202f±\u202f22\u202fMa, respectively. The freshest areas are characterized by sub-concordant 206Pb/238U and 207Pb/235U ratios, identical chemical compositions and similar very low δ18O values (−39.3 to −31.4‰). These data indicate that the two uraninites both crystallized at ca. 1300\u202fMa, from the same fluid and under identical physico-chemical conditions. Alteration is characterized by (i) the progressive incorporation of Ca, Si, and Fe, reaching several wt.%, which substitute to the radiogenic Pb and cause a progressive decrease in the Pb/U isotopic ratios. The radiogenic Pb is also substituted by water during the alteration, (ii) concomitant variations in trace element contents (As, Mn, V, LREEs, Sr, Th, B, Ba, Nb, for example) and (iii) heavier δ18O signatures (−22.5 to −8.91‰), typical of meteoric waters, in the altered zones. This combined approach demonstrates that fluid-driven post-crystallization exchanges affected each uraninite during recent fluid flow events (ca. 40\u202fMa and 70\u202fMa respectively). The relatively high dispersion of the Pb/U ratios in relation to the Discordia for both samples is considered as linked to a local (nm to µm-scale) differential mobility between lead and uranium within the uranium oxides. The chemical changes affecting elements previously considered as immobile in uraninite, such as REEs, indicate that these elements are not preserved during the post-crystallization alteration process studied here. Alteration processes may therefore have a major impact on the classical geochemical tracers, such as REE patterns or LREE/HREE ratios, currently used in nuclear forensic studies. The isotopic and chemical tracers currently used to track back the origin, age and history of natural uraninites should therefore be considered with a high degree of caution to avoid misleading and erroneous conclusions. Moreover, the comparison of calculated U-Th-Pb chemical, 207Pb/206Pb, and Pb/U isotopic ages shows that the use of age clustering for determining U-Th-Pb chemical ages and 206Pb/207Pb ages is not appropriate for constraining crystallization stage(s) of altered uraninites and for deciphering the different fluid events that potentially altered or recrystallized the uraninites over time. This study also indicates that estimation of the crystallization age of uraninite from substitution trends of Pb to Ca is not applicable to unconformity-related U deposits and results in overestimated ages because of an initial integration of calcium in the uraninite lattice at the time of crystallization.