Engineering topics

Zhao, Xiaodong

Publications and source records attributed to Zhao, Xiaodong.

3 records

DOE OSTI2026

Trisodium Phosphate Phases and Solubility in Alkaline Solutions Relevant to Radioactive Waste Processing

The U.S. Department of Energy’s Hanford Site faces significant challenges in managing millions of gallons of legacy radioactive waste, where phosphate precipitation can obstruct pipelines during retrieval and processing. To resolve long-standing inconsistencies in reported solubility and clarify factors governing trisodium phosphate hydrates, we examined the Na3PO4:NaOH:H2O system. Powder and single-crystal X-ray diffraction revealed that commercial precursors undergo transformations that produce multiple hydrates, including three previously unreported phases comprising an ordered polymorph of Na3(PO4)·12H2O·1/6NaOH, Na3PO4·5H2O, and Na3PO4·9H2O. Computational modeling indicated that the ordered dodecahydrate is more stable than its disordered counterpart, suggesting kinetic persistence of structural disorder. Solubility measurements were conducted with solutions prepared either from anhydrous Na3PO4 or from Na3(PO4)·12H2O·1/6NaOH and revealed that release of interstitial NaOH from the hydrate precursor elevated solution alkalinity, thereby significantly reducing phosphate solubility relative to solutions prepared with anhydrous Na3PO4 across 20–44 °C. These results begin to reconcile inconsistencies in prior solubility data and clarify how phase composition dictates phosphate precipitation under alkaline conditions.

Graham, Trenton R. (ORCID:0000000189078004)↗

DOE OSTI2024

Understanding Trace Iron and Chromium Incorporation During Gibbsite Crystallization and Effects on Mineral Dissolution

Incorporation of pollutants, e.g., heavy metals, or critical elements, e.g., lithium, as impurities in mineral phases can significantly affect their mobility or sequestration in the environment. Even when present at low concentrations, impurities can alter the solubility and reactivity of the host mineral. Here, in this study, we investigate the incorporation of trace amounts of iron (Fe 3+ ) and chromium (Cr 3+ ) during the crystal growth of the aluminum (Al 3+ ) hydroxide, gibbsite, a major component of bauxite ores, an important soil mineral, and a dominant mineral phase in stored radioactive wastes. Using a comprehensive suite of analytical techniques, we show that both Cr 3+ and Fe 3+ can be incorporated into the gibbsite lattice during coprecipitation by replacing Al 3+ in octahedral sites. These small amounts are consistent with limited to no structural isomorphism shared between Al 3+ and Cr 3+ /Fe 3+ hydroxide precipitates, nor room temperature miscibility of their isostructural M 2 O 3 oxide forms, in contrast with oxyhydroxide forms where Al 3+ and Fe 3+ share similar structural topologies. Despite the limited uptake of Cr 3+ /Fe 3+ , we show that these impurities have significant implications for gibbsite dissolution behavior. The limited uptake of Cr 3+ /Fe 3+ (e.g. 0.43% Cr 3+ and 0.4% Fe 3+ ), we show that these impurities have significant implications for gibbsite dissolution behavior and subsequent reactivity in complex environments.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

DOE OSTI2024

Accurate and Data‐Efficient Micro X‐ray Diffraction Phase Identification Using Multitask Learning: Application to Hydrothermal Fluids

Traditional analysis of highly distorted micro X‐ray diffraction (μ‐XRD) patterns from hydrothermal fluid environments is a time‐consuming process, often requiring substantial data preprocessing and labeled experimental data. Herein, the potential of deep learning with a multitask learning (MTL) architecture to overcome these limitations is demonstrated. MTL models are trained to identify phase information in μ‐XRD patterns, minimizing the need for labeled experimental data and masking preprocessing steps. Notably, MTL models show superior accuracy compared to binary classification convolutional neural networks. Additionally, introducing a tailored cross‐entropy loss function improves MTL model performance. Most significantly, MTL models tuned to analyze raw and unmasked XRD patterns achieve close performance to models analyzing preprocessed data, with minimal accuracy differences. This work indicates that advanced deep learning architectures like MTL can automate arduous data handling tasks, streamline the analysis of distorted XRD patterns, and reduce the reliance on labor‐intensive experimental datasets.

97 MATHEMATICS AND COMPUTING↗