The effect of implantation and heat treatment on the structural evolution and migration behaviour of Selenium in glassy carbon

Abstract

The storage and management of nuclear waste is perhaps the most controversial aspect of nuclear power production, as there is no permanent solution to this problem that has existed for centuries. In the last decade, our group at the University of Pretoria, South Africa, has been studying the suitability of glassy carbon (GC) as a nuclear waste storage material. Some of these studies focused on investigating the migration of strontium, caesium, silver, cadmium, indium, europium, and xenon in GC. No research information on the migration of selenium (Se) in GC in the literature, hence this study. A radioisotope of Se, 79Se is a fission product found in spent fuel and reprocessed nuclear fuel. Its release into the environment has associated health hazards. In this study, Se ions of 150 keV were implanted into GC substrates to a fluence of 1 × 1016 ions/cm2 at room temperature (RT), 100, 150 and 200 °C. The samples implanted at RT and 200 °C were characterised by transmission electron microscopy (TEM) to compare the radiation damage level with respect to the implantation temperatures. Some as-implanted samples were sequentially annealed at the low-temperature regime (300 – 700 °C) and high-temperature regime (1000 – 1200 °C) for 5 h in steps of 100 °C. A new set of as-implanted samples were isochronally annealed at 1000, 1100 and 1200 °C for 5 h cycles. The migration of Se was monitored by Rutherford backscattering spectrometry (RBS) and secondary ion mass spectroscopy (SIMS). Raman spectroscopy was used to monitor the microstructural changes in the GC substrates. Se implantation induces radiation damage at relatively comparable depths in the GC samples implanted at RT and 200 °C. The damaged layer in the RT sample corresponds to about 8.5 dpa, greater than 0.2 critical dpa, which will totally distort the GC microstructure. The microstructures of the as-implanted GC samples were damaged by Se ion implantation, which increases with increasing implantation temperatures. The sample implanted at RT has a more graphitic disorder and the 200 °C sample has the least damaged microstructure and is less graphitic. Annealing the as-implanted samples at 300 – 700 °C resulted in a limited recovery of the GC substrate and appreciable recovery was observed at the high-temperature regime, 1000 – 1200 °C. No measurable diffusion of Se atoms occurred in all the as-implanted samples after annealing at the low-temperature regime (300 – 700 °C). At 1000 °C, the RT Se profile broadens, indicating the diffusion of Se atoms. Further annealing at 1100 and 1200 °C resulted in the asymmetrical broadening of Se profiles towards the surface of the RT sample, accompanied by about 5 and 32 % losses of Se atoms, respectively. The Se profiles of the RT sample at 1100 and 1200 °C exhibited tailing towards the bulk, indicating the migration of Se in the bulk of the GC substrate. The diffusion coefficients of Se were calculated to be 4.79 × 10-20 and 5.90 × 10-20 m2s-1 after annealing at 1000 and 1100 °C, respectively. No measurable diffusion of Se occurred in the sample implanted at 100 °C at the high-temperature regime, 1000 – 1200 °C. Segregation of Se at the surfaces of the samples implanted at 150 and 200 °C and sequentially annealed at 1000 – 1200 °C was observed, accompanied by the loss of Se and migration in the bulk of these substrates. Overall, the SIMS profiles of the new sets of as-implanted samples isochronally annealed at the high-temperature regime were somewhat similar to those obtained by RBS, with minor differences. The differences in the RBS and SIMS profiles were attributed to the resulting different microstructures of these two samples (i.e., the sequentially and isochronally annealed sample types). Generally, the migration behaviour exhibited by Se atoms in the bulk region of the as-implanted samples (after annealing) can be explained in terms of trapping and de-trapping of the Se atoms by defects induced during implantation. The high-temperature annealing caused the annealing of defects in the less radiation damage region, creating pathways for Se atoms to migrate deeper into the bulk of the as-implanted and annealed GC samples. One of the initial concerns in this research was the migration of Se atoms in the bulk of the GC substrates, as this will limit the use of GC as a potential nuclear waste storage container. With a minute concentration of Se atoms estimated in the bulk region of the GC, the integrity of GC (as a nuclear storage container) cannot be limited.