Mechanistic insights into volatile ruthenium tetroxide formation during evaporation-to-dryness accidents of high-level liquid waste

高レベル液体廃棄物の蒸発乾固事故における揮発性四酸化ルテニウムの生成メカニズムに関する知見

吉田 尚生; 天野 祐希 ; 吉田 涼一朗 ; 大野 卓也 ; 田代 信介 ; 山根 祐一  

Yoshida, Naoki; Amano, Yuki; Yoshida, Ryoichiro; Ono, Takuya; Tashiro, Shinsuke; Yamane, Yuichi

四酸化ルテニウム(RuO)は、有毒で揮発性の化合物であり、原子力再処理における高レベル放射性廃棄物(HLLW)の蒸発乾固事故(EDA)の際に生成・放出される可能性がある。RuOに放射性同位元素(Ru、Ru)が含まれる場合、空気中の放射線による危険をもたらす。本研究では、現実的な組成の模擬HLLWを用いて、マトリックス効果、温度範囲、酸化剤など、環境に関連する条件下でのRuOの生成を検証した。その結果、従来のモデルの主要な経路と限界が明らかになり、事故シナリオの予測精度が向上した。これらの結果は、核燃料サイクル運用における揮発性放射性物質の放出管理のための、より信頼性の高い環境リスク評価を支援するものである。

Volatile ruthenium tetroxide (RuO) formed during evaporation-to-dryness accidents (EDA) of high-level liquid waste (HLLW) combines high chemical toxicity with radiotoxicity from isotopes such as Ru and Ru, making it a primary hazard in nuclear-fuel reprocessing. This study experimentally elucidates the mechanisms governing RuO formation while testing the validity of the pragmatic nitrosyl-ruthenium ([RuNO]) surrogate model, an experimental model that assumes Ru in HLLW exists as [RuNO], and evaluates gaseous Ru release using nitric acid solutions of [RuNO], commonly used in RuO release assessments by comparing it with the behavior of simulated HLLW (s-HLLW). We systematically investigated potential pathways, including oxidation by liquid-phase nitric acid (HNO(l)), gas-phase nitric acid (HNO(g)), and oxygen (O2); thermal decomposition of [RuNO]; and matrix effects of coexisting nitrates. The results identify oxidation by activated HNO(l), oxidation by HNO(g) and thermal decomposition of [RuNO] as the principal routes to RuO, with HNO(g) playing a far more significant role than previously recognized. Crucially, the RuO release profile from s-HLLW differed markedly from that predicted by the [RuNO] surrogate model, demonstrating that this model fails to capture the complex matrix effects and time-dependent chemical changes of Ru species inherent to real HLLW. These findings have significant implications for improving the accuracy of hazard assessments related to RuO release during HLLW EDAs, particularly by highlighting the limitations of commonly used [RuNO] surrogate models.