The source describing this event (Ordin, NASA) did not provide all details necessary to understand in- depth what happened. (1) At the start of this event, there is the action aiming at replacing a burst disk which had opened. Why this action was performed without inert gas protection is not explained. It is however important to notice, as explained by H2TOOLS, that liquid hydrogen storage systems are usually, or should be equipped with two independent pressure relief systems, with the possibility to switch between the two. Operators can then keep one in function and isolate the other to execute repair or maintenance works without the risk of hydrogen release. (2) Moreover, the source does not explain the decision to use liquid nitrogen to fight the fire. Was it the only fire suppression system on the site? Was it to extinguish the fire, or to keep cold neighbouring liquid hydrogen vessels? This last option is the most probable, because the use of liquid nitrogen to extinguish a hydrogen fire has demonstrated ineffective. Usually, under these conditions, the hydrogen is left releasing till emptying the vessel. On top of that, solid nitrogen could even have blocked the venting by freezing in contact with colder hydrogen, creating the condition for a BLEVE (boiling liquid evaporating explosion), as it happened in another case when water was used fight the over-pressurisation of a vessel (HIAD_609). (3) The temperature of the liquid nitrogen induced local thermal-mechanical stresses in the external shell of the neighbouring vessel, due to the contraction of the steel. If used intensively on the first vessel, it would have created on it the same condition for cracking.
When rupture disk was being replaced, the vessel was still loaded with LH2 , without purging gas. The H2-air mixture was probably ignited by static discharges. This suggest as ROOT CAUSE a failure in following the procedure. Rupture of the second vessel burst disk was caused by the low temperature exposure of the mild steel vacuum jacket. This time an emergency procedure failed, or the choice of the tank materials was wrong.