Reflective reviews on Japanese high-level waste (HLW) vitrification – Exploring the obstacles encountered in active tests at Rokkasho

Vitrification is currently the most widely used technology for the treatment of high-level waste (HLW) generated from spent nuclear fuel (SNF) reprocessing. Glasses for the solidification of HLW have been developed over half a century at different times in each country. Likewise in Japan, repeated system evaluations of borosilicate waste glass matrices and the liquid-fed joule-heated ceramic-lined melter (LFCM) process have been found to be applicable to a wide range of HLW compositions. The Rokkasho vitrification facility located in northern Japan encounters significant operational issues owing to the presence of sparsely soluble elements including noble metals (NMs) and separated phases, such as the yellow phases (YPs) constituents, in glass melts. These elements cause serious disruptions, jeopardizing the stable functioning of the LFCM. The active vitrification test was suspended because of the formation of YPs in the glass melts and subsequent sedimentation of NMs at the bottom of the melter. First, we analyzed the testing data to identify the fundamental causes behind the unstable LFCM operation. Our investigation revealed that the presence of significant amounts of Mo and S, along with NMs, specifically Ru, Rh, and Pd, in the Rokkasho HLW stream triggers the formation of molybdate and sulfate salt phases (YPs), which subsequently enabled the sedimentation of NMs at the bottom of the melter. Second, we selected eight major tasks related to difficulties of LFCM operation in the active tests and studied the findings in each task considering the preconditions of task success. Moreover, we strengthened our analysis by subjecting each task to 22 hypotheses tests in a systematic manner to minimize the risk of issues associated with generalizability. Third, we combined these findings and classified them into two categories based on the action system theory on management: variable system (Y) and actor’s system (X), assuming that the indirect or mechanism/environment factors of X should influence on the directly observed factors of Y. Our approach involved implementing a reflective actor’s system (X’) to improve the variable system for the controllable system (Y’), thereby mitigating unexpected events. We then summarized the main implications of this method to ensure predictable and reliable outcomes for the associated tasks. Finally, we concluded the potential effectiveness of strategic actions that could leverage organizational capabilities and processes to improve task performances in the context of a controllable system given the implications we identified.