Seawater was pumped into the reactor cores and the SF storage pools for about two weeks, in an effort to cool the fuels. The total amount of seawater injected before March 25, when it was replaced by the injection of newly delivered pure water, was 2842, 9197, and 4495 kL for units 1, 2 and 3, respectively. It was estimated that as much as 32 tonnes of sea salt may have accumulated in the reactor units. Boric acid was added to the cores to func­tion as a neutron absorber to prevent re-criticality by the collapsed fuels [ 5] . Circulation of the water using the circulation lines was found to be impossible because there were appreciable leakages in reactor and/or con­tainment vessels, through which injected water continuously flowed out from the reactor building to the basement of the turbine building. The concentration of cesium-137 in the contaminated water exceeded the order of magnitude of 106Bq/cm3. By the middle of June 2011, a new water treat­ment system to decontaminate the highly contaminated water flowing from the cores had been established, and the circulation injection cooling using this facility was started in late June. This system (described in more detail in Section 24.5) cleanses the highly radioactive water, recovered from the basement of the building, and injects the decontaminated water back to the core. After removing radionuclides, water is desalinated before re-injection, reducing the salt content in the water to less than several ppm as of July 2012. The addition of seawater to the core, though it was unavoidable during the emergency, raised an issue of the adverse effects of the salt, in degrading of the vessels and other devices, as well as reacting with the fused fuels to form complicated chemical forms of debris.