UC Berkeley

Efficient extraction and separation of actinides, lanthanides, and fission products from acidic or high ionic strength media is desirable for applications in nuclear waste reprocessing, environmental remediation, and laboratory scale processing of accelerator targets or forensics samples. The adaptation of traditional solvent extraction agents for use in solid phases would potentially reduce or eliminate the generation of mixed hazardous, radioactive waste streams. This dissertation presents the synthesis and evaluation of organically functionalized silica for the extraction of U(VI) from aqueous solutions. Diethylphosphonate and acetamide diethylphosphonate were grafted to an ordered mesoporous silica substrate, Santa Barbara Amorphous (SBA)-15. The phosphonate functional groups were chosen based on their demonstrated utility in solid and liquid phase extraction studies. SBA-15 has shown great potential as a framework for organic modification due to its high surface area, tunable pore size and particle morphology, resilience in acidic media, and hydrothermal robustness.

The work presented here is motivated by the need to obtain a better fundamental understanding of 1) the functionalization process, 2) the structure of the organometallic complexes formed at the functionalized silica surface, and 3) the limitations of metal extraction. Most previous studies on heavy metal extraction using organically modified silica have relied on measurement of macroscopic properties. In contrast, this work uses a combination of batch contact and nuclear magnetic resonance spectroscopic techniques to more thoroughly characterize the functional layer, and to probe the interaction between uranyl and the surface. In doing so, a greater understanding of the limitations of these types of materials has been achieved.

The fundamental interaction of U(VI) with diethylphosphatoethyl triethoxysilane (DPTS) functionalized SBA-15 mesoporous silica is studied by macroscopic batch experiments and solid-state nuclear magnetic resonance (NMR) spectroscopy. DPTS-functionalized silica has been shown to extract U(VI) from nitric acid solutions at or above pH 3. Extraction is dependent on pH and ionic strength. Single-pulse 31P NMR on U(VI) contacted samples revealed that U(VI) only interacts with a fraction of the ligands present on the surface. At pH 4 the U(VI) extraction capacity of the material is limited to 27-37% of the theoretical capacity, based on ligand loading. Single pulse 31P NMR spectroscopy on U(VI)-contacted samples was combined with batch studies to measure a ligand-to-metal ratio of approximately 2:1 at pH 3 and 4. Batch studies and cross-polarization NMR measurements reveal that U(VI) binds to deprotonated phosphonate and/or silanol sites. 31P-31P DQ-DRENAR (Double-Quantum-based Dipolar Re-coupling effects Nuclear Alignment Reduction) NMR studies were used to compare the average dipolar coupling between phosphorus spins for both U(VI)-complexed and non-complexed ligand environments. These measurements reveal that U(VI) extraction is not limited by inadequate surface distribution of ligands, but rather by low stability of the surface phosphonate complex.

Acetamide diethylphosphonate (AcPhos)-functionalized silica has been shown to have a high affinity for U(VI) in pH 2-3 nitric acid solutions. Previous work has focused on actinide and lanthanide extraction under various conditions, but has shown poor reproducibility in the functionalization process. For this work, four AcPhos-functionalized SBA-15 materials were synthesized and evaluated based on their U(VI) sorption capacity and their resilience to contact with pH 3 nitric acid solutions. Materials synthesized with a pyridine base catalyst were shown to form a greater fraction of polymeric structures at the silica surface, which correlated with higher structural integrity upon contact with acidic solutions. The single-pulse 31P and 1H NMR spectra of these materials show evidence of the formation of phosphonic acid groups, as well as hydrogen-bonding interactions between ligands or with the silica surface. Additionally, these materials were found to have a significantly higher U(VI) sorption capacity and Keq than the materials synthesized without pyridine, most likely due to the ion-exchange properties of the phosphonic acid groups. The 31P-31P DQ-DRENAR NMR technique was again employed to compare the average strength of dipolar coupling interactions between phosphorus atoms for the four different materials. Because the strength of dipolar coupling interactions depends on the number and proximity of neighboring spins, this technique provides information about the average density of ligands on the surface. The conventional functionalization procedure yielded materials with the lowest average surface ligand density, while those using extended reaction times and the pyridine base catalyst yielded materials with higher surface ligand densities.