Immense effort has been exerted in the materials informatics community towards enhancing the accuracy of machine learning (ML) models; however, the uncertainty quantification (UQ) of state-of-the-art algorithms also demands further development. Most prominent UQ methods are model-specific or are related to the ensembles of models; therefore, there is a need to develop a universal technique that can be readily applied to a single model from a diverse set of ML algorithms. In this study, we suggest a new UQ measure known as the Δ-metric to address this issue. The presented quantitative criterion was inspired by the k-nearest neighbor approach adopted for applicability domain estimation in chemoinformatics. It surpasses several UQ methods in accurately ranking the predictive errors and could be considered a low-cost option for a more advanced deep ensemble strategy. We also evaluated the performance of the presented UQ measure on various classes of materials, ML algorithms, and types of input features, thus demonstrating its universality.
This work presents details of the synthesis, properties and structure of a novel neptunium carbonate (NH4)[NpO2CO3], a member of the M[AnO2CO3] (M = K, (NH4), Rb, Cs) class of compounds. Carbonates play an important role in the migration of actinides in the environment, and thus are relevant for handling and disposal of radioactive wastes, including spent nuclear fuel and vitrified raffinates. Knowledge of the crystallographic structure of these compounds is important for models of the environmental migration behavior based on thermodynamic descriptions of such chemical processes. (NH4)[NpO2CO3] crystals were obtained during long-term hydrothermal treatment of Np(VI) in aqueous ammonia at 250 °C. X-ray diffraction (XRD) and X-ray photoelectron spectroscopy (XPS) show that a single-phase sample containing only Np(V) was obtained. Structural features of (NH4)[NpO2CO3] were elucidated from single crystal X-ray diffraction and confirmed by vibrational spectroscopy. The results obtained are of interest both for fundamental radiochemistry and for applied problems of the nuclear fuel cycle.
99Tc is one of the predominant fission products of 235U and an important component of nuclear industry wastes. The long half-life and specific activity of 99Tc (212,000 y, 0.63 GBq g−1) makes Tc a hazardous material. Two principal ways were proposed for its disposal, namely, long-term storage and transmutation. Conversion to metal-like technetium matrices is highly desirable for both cases and for the second one the reasonably high Tc purity was important too. Tetramethylammonium pertechnetate (TMAP) was proposed here as a prospective precursor for matrix manufacture. It provided with very high decontamination factors from actinides (that is imperative for transmutation) by means of recrystallisation and it was based on the precise data on TMAP solubility and thermodynamics accomplished in the temperature range of 3–68 °C. The structure of solid pertechnetates were re-estimated with precise X-ray structure solution and compared to its Re and Cl analogues and tetrabutylammonium analogue as well. Differential thermal and evolved gas analysis in a flow of Ar–5% H2 gas mixture showed that the major products of thermolysis were pure metallic technetium in solid matrix, trimethylammonium, carbon dioxide, and water in gas phase. High decontamination factors have been achieved when TMAP was used as an intermediate precursor for Tc.
Investigation of shocked materials provides unique information about behavior of substances in extreme thermodynamic conditions. Near surface nuclear tests have induced multiple transformations of affected soils. Examination of nuclear glasses and relics of entrapped minerals provides a unique database on their behavior under an intense temperature flash. In this work, several types of nuclear fallout particles from historic tests at the Semipalatinsk test site are investigated using complementary analytical methods. Distribution of radionuclides in all types of samples is highly heterogeneous; domains with high content of radionuclides are often intermixed with non-active materials. There is no general correlation between chemical composition of the glassy matrix and content of radionuclides. In aerodynamic fallout, the main fraction of radionuclides is trapped in the outer glassy shell. Relics of quartz grains are always devoid of radionuclides, while glass regions of high activity have different composition. In contrast to underground tests, iron-rich minerals are not necessarily radioactive. In most cases, the glassy matrix in anhydrous and is strongly polymerized, and the Q3 silicate groups dominate. Temperature-induced transformations of entrapped minerals are discussed. Investigation of zircon grains shows absence of a direct correlation between degree of decomposition into constituting oxides, morphology of resulting baddeleyite, and maximum experienced temperature. For the first time, temperature history of a nuclear ground glass is estimated from Zr diffusion profiles from decomposing zircon grain.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.