Date of Award

Fall 2019

Document Type

Open Access Thesis


Nuclear Engineering

First Advisor

Travis W. Knight


Research reactors such as the Advanced Test Reactor (ATR) and the High Flux Isotope Reactor (HFIR) employ aluminum-clad fuel elements made up of many thin plates with uranium dispersed within. In most engineering applications, aluminum is considered to have favorable corrosion characteristics. It forms a thin oxide layer [Al2O3] under atmospheric conditions that is impenetrable to oxygen thus stopping any further corrosion. However, both aluminum metal and Al2O3 react with water to form hydrous oxides which are less protective against further corrosion and form significantly thicker layers than oxidation in dry air. As a result, aluminum-clad spent nuclear fuel (ASNF) hosts chemisorbed bound water on the fuel surface. In addition, adsorbed or physiosorbed water contributes to the total water within the oxide layer. This is a challenge for sealed dry storage of ASNF because the physiosorbed water and water in the hydroxides could be released as free water at high temperature or decomposed by radiolysis leading to further corrosion and a buildup of pressure within the cannister. The goal of this research is to study the formation of lab-grown oxides on aluminum samples as surrogates for those on ASNF, characterize those oxide layers, and quantify the conditions necessary to remove bulk, physiosorbed, and chemisorbed water. This knowledge will be used to set parameters for full-scale drying studies of ASNF later on. Testing of aluminum oxide powder samples by Thermogravimetric Analysis and Differential Scanning Calorimetry (TGA/DSC) has been performed on commercially available oxyhydroxide powders to determine the dehydroxylation temperatures to be expected in bulk tests. Gibbsite was found to decompose at about 300°C while dehydroxylation for fine and coarse boehmite averaged around 520°C, and 440°C respectively.

Aluminum coupons of Al-1100, Al-5052, and Al-6061 were immersed in distilled water at 20°C, 50°C, and 100°C to produce a hydrated oxide layer. Bulk drying tests conducted via Thermogravimetric Analysis (TGA) on these aluminum-cladding surrogate samples found dewatering for 20°C, 50°C, and 100°C samples to initiate at modest temperatures below 100°C. The amount of water removed depended on a combination of the heating period and maximum temperature. However, even in low temperature TGA runs, the total amount of water removed matched closely with higher temperature runs as long as the low temperature was maintained for a sufficiently long time. Imaging by Scanning Electron Microscope (SEM) and analysis by X-Ray Diffraction (XRD) took place throughout the research for a detailed understanding of the microstructure and crystal structure at each stage of the process. Based on the findings from this work it is believed that the current drying process of vacuuming the drying canister to 5Torr and heating to 220°C for 35 to 45 minutes in air cyclically is insufficient for removing the maximum chemically bound water. Instead, the drying process should involve heating the spent fuel elements continuously to 220°C or more staying below the suggested maximum of 250°C, for about 5 hours either by forced gas circulation or under vacuum with external heating. Even using these parameters, it is uncertain if not unlikely that the water trapped in crystalline structures on the outermost surface was fully liberated.

Rather, the evidence seems to suggest most of the mass loss seen in bulk drying tests is coming from the lower and/or the intermediate layers closer to the substrate’s surface.