search


Top > Releases ・ Announcements > Press Releases > Status of TEPCO's Nuclear Power Stations after theTohoku-Chihou-Taiheiyou-Oki Earthquake > 2013 > Status of TEPCO's Nuclear Power Stations after the Tohoku-Chihou-Taiheiyou-Oki Earthquake (Daily Report as of 3:00 PM on September 12)

Status of TEPCO's Nuclear Power Stations after the Tohoku-Chihou-Taiheiyou-Oki Earthquake (Daily Report as of 3:00 PM on September 12)

Due to the Tohoku-Chihou-Taiheiyou-Oki Earthquake which occurred on March 11, 2011, TEPCO's facilities including our nuclear power stations have been severely damaged. We deeply apologize for the anxiety and inconvenience caused.
With regard to the accident at Fukushima Daiichi Nuclear Power Station, on April 17, 2011, we have compiled the roadmap towards restoration from the accident and on July 19 we accomplished the Step1 target "Radiation dose is in steady decline". Then on December 16 we confirmed the accomplishment of the Step 2 target "Release of radioactive materials is under control and radiation doses are being significantly held down".
In addition, on December 21, 2011, we have compiled the "Mid-to-long-Term Roadmap toward the Decommissioning of Fukushima Daiichi Nuclear Power Units 1-4, TEPCO".
In addition to the maintenance of the plant's stable condition, we will implement Mid-to-Long Term countermeasures towards the decommissioning of Fukushima Daiichi Nuclear Power Units 1-4 to enable evacuees to return to their homes as soon as possible and reduce the anxiety of the people in Fukushima and the whole nation as soon as possible.

Below is the status of TEPCO's Fukushima Daiichi Nuclear Power Station.

* The updates are underlined.

[Fukushima Daiichi Nuclear Power Station]
· Unit 1 to 4: Abolishment (April 19, 2012)
· Unit 5 to 6: Outage due to regular inspections before the earthquake

-At 10:00 AM on September 12, accumulated water transfer from the Unit 6 Turbine Building basement to a temporary tank was started. At 3:00 PM on the same day, the transfer was stopped.

-At 7:03 AM on September 12, cooling by the Unit 1 spent fuel pool alternative cooling system was suspended due to work to remove obstacles (such as debris) on the 1st floor of Reactor Building was suspended. The spent fuel pool water temperature was 27.5℃ when the suspension of cooling was stated. The suspension of cooling is scheduled to continue for about 130 hours until 5:00 PM on September 17, and, at the start of the suspension, we expected the pool water temperature to increase at a rate of 0.070℃/h. Therefore, we expect the pool water temperature to increase by approx. 9.1℃ during the suspension. With this increase, the pool water temperature will remain sufficiently below the operational limit value of 60℃, and there will be no problem in controlling the spent fuel pool water temperature.

-On September 12 dust sampling was performed at the ventilation facility in Unit 2 Reactor Building.

-Contaminated water transfer from the underground reservoirs was all completed as of July 1. However, we are continuing to take measures to prevent the expansion of contaminated water, and to conduct sampling activities.

<Measures to prevent the expansion of contaminated water>
· Since the decreases of all-β radioactivity densities in the leakage detection holes (at the northeast side of the underground reservoir No.1, the northeast side of the underground reservoir No.2, and the southwest side of the underground reservoir No.3) have been slow, operations to dilute the underground reservoirs No.1-No.3 by transferring filtered water or desalination-system (RO) treated water (the all-β radioactivity density: approx. 1×101Bq/cm3) into these reservoirs have been conducted as appropriate.

<Recent dilution operations>
Underground reservoir No.1 (since June 19): On August 3, approx. 60m3 of filtered water was injected.
Underground reservoir No.2 (since June 27): On August 1, approx. 60m3 of filtered water was injected.
Underground reservoir No.3 (since July 24): On August 12, approx. 107m3 of water in the drain hole (northeast) of this underground reservoir was injected.

· On September 11, leaked water in the leakage detection holes at the underground reservoirs No.1-No.3 was transferred to the temporary aboveground tank, and leaked water in the drain holes at the underground reservoirs No.1 and No.2 was transferred into these underground reservoirs.

<Sampling>
On September 11, sampling was performed in the drain holes of the underground reservoirs No.1-No.7 (14 locations), the leakage detection holes of the underground reservoirs No.1-No.4 and No.6 (sample could not be collected at 2 out of 10 locations), and the observation holes of the underground reservoirs (22 locations). No significant change was found in the analysis results compared to the analysis results on the previous samples (taken on September 10).
Analysis was conducted on the groundwater bypass pump wells No.7-No.12, and resulted in no remarkable change from the analysis on the previous samples (taken on September 7).
Further, analysis for tritium was conducted on water sampled on September 4 in the drain holes of the underground reservoirs No.1-No.7 (14 locations), the leakage detection holes of the underground reservoirs No.1-No.4 and No.6 (sample could not be collected at 2 out of 10 locations). The analysis results showed no significant change compared to the previous results (on the water sampled on August 28).

-On August 19, puddles were found inside a dike around the H4 area tanks in the power station and outside of a drain valve of the dike.
We found water spread at the bottom level of tanks near Tank No.5 in the Group I in the H4 area. Therefore, we checked the water level of this tank, and found out that the water level has fallen by approx. 3m than the normal level (the amount of water: approx. 300m3). We started collecting the water remaining inside the dike and already collected some of the water. However, since it seemed that the water has flowed out of the dike through the drain valve, we are collecting soil in the surrounding area and continuing to conduct an investigation to find out the range reached by the water. Later, we found streaky traces of flows on the wall surface of a drainage channel located east of the H4 area tanks. The maximum surface dose equivalent rate measured at this location was 6.0mSv/h (γ and β rays (70μm dose equivalent rate)). As this information indicates the possibility that contaminated earth and sand, etc. may have flowed into the drainage channel, we are planning to conduct a detailed investigation and evaluation concerning these traces.

On August 22, transfer of water stored in Tank No.5 in Group I in the H4 area and water collected in a temporary tank (water accumulated inside the dike) into Tank No.10 in Group B in the H4 area was completed.

On August 22, we conducted full inspections (appearance inspections and dose measurement) on the flanged tanks in the other areas, which are of the same type as the tank from which water has leaked. Neither leak nor puddle was found by the appearance inspections on the tanks and the drain valves. However, 2 locations locally showing high dose rates were found around the H3 area tanks. The surfaces of these locations were dry, and we confirmed that there has been no water having flowed into the inside of the dike or the outside of the dike. We also confirmed that the water levels of these tanks remain unchanged after they received water. Additionally, we conducted soundness inspections (visual appearance inspections and water level confirmation) on the flanged tanks that are in use for storage of accumulated water from Units 5 and 6. The inspections were completed on August 26 and showed no abnormality.

During our inspection of Tank No.5 in Group I in the H4 area from which water has leaked, we found out the following:
· 3 tanks including this tank (Tank No.5 in Group I, Tank No.10 in Group I, and Tank No.3 in Group II in the H4 area) were initially installed in the H1 area.
· Ground subsidence occurred in the H1 area at the foundations on which these tanks had been placed, and these tanks were planned to be installed in the H2 area, but actually, have been placed in the H4 area despite our plan.
Although it is still unclear whether there is a causal relationship between the water leaking of Tank No.5 and the tank's having experienced ground subsidence occurred in the H1 area at the foundation on which it had been placed, we conducted water transfer from inside Tank No.3 in Group II in the H4 area to Tank No.10 in Group B in the H4 area to reduce the risk of leakage.
<The latest transfer operations>
· At 3:57 PM on August 25, transfer from Tank No. 10 in Group I in the H4 area to Tank No. 10 in Group B in the H4 area was started. At 2:07 AM on August 27, the transfer was completed.
· At 10:30 AM on August 29, transfer from Tank No. 3 in Group II in the H4 area to Tank No. 10 in Group B in the H4 area was started. At 2:07. At 11:03 AM on September 2 water transfer was suspended due to the implementation of rainfall countermeasures.

During a patrol on August 31, we found 4 locations showing high dose equivalent rates (γ and β rays (70μm dose equivalent rate)). We consider that there has been no leakage to the outside of the dike because the water levels of all of the relevant tanks have not decreased and also because the drainage valves have been closed.
One of the 4 locations, the connecting pipe section between Tank No.5 and Tank No.6 in Group IV in the H5 area has the heat insulation material placed in the upper part of the pipe, and one drop of water fell to the floor surface when this heat insulation material was pressed. The dose equivalent rate at a location on the floor surface to which the water fell was measured and confirmed to be approx. 230mSv/h. Although water then stopped dropping from the connecting pipe, a discolored part (in a dry condition) of approx. 20cm×20cm was found on the floor surface under the pipe section. Then, we checked the status of water dropping from the connecting pipe between Tank No.5 and Tank No.6 in Group IV in the H5 area with the heat insulation material having been pulled out. Then, we found out that a flange part connecting an isolation valve on the Tank No.5 side (there are 2 isolation valves connecting these tanks and connecting pipe) and the connecting pipe was dripping one drop per approx. 90 seconds. In response, on the same day, we wrapped the adsorption mat around this flange part and covered it with plastic-sheet protection, while placing a drain receiving pan under the flange part on the floor. Both of the 2 isolation valves on the respective Tanks No.5 and No.6 sides of this connecting pipe were found to have been closed. On September 1, we tightened up 12 flange bolts at this flange part, and we confirmed that no water leakage was occurring. For the confirmation purpose, the water levels of Tank No.5 and Tank No.6, in Group IV in the H5 area, were measured and found unchanged.

<Results of the latest patrol>
During a patrol on September 11, no location showed a high dose equivalent rate (β and γ rays (70μm dose equivalent rate)). Locations near the dike floor continued to show low dose equivalent rates since rainwater accumulated inside the dike (approx. 4 to 5 cm deep) worked as a shield. Additionally, we conducted a visual inspection, and all of the tanks were found without any abnormality such as leakage (expect for the leaks that resulted in accumulated water inside the dike).

As a result of the full inspections (appearance inspections and dose measurement) conducted on August 22 on the tanks in the areas other than the H4 area, we found tanks (Tank No.4 in Group B and Tank No.10 in Group A, both in the H3 area) each having a part locally showing a high dose rate. Although no water dropping was found on the outside of these tanks, we are planning to transfer water inside the tanks to an RO waste liquid supply tank to reduce the risk of leakage. The transfer is scheduled to be conducted between August 29 and September 17.

Following the leakage from a tank this time, we sampled water near the exit of the drainage channel of the south water outlet, in the drainage channel B near Fureai Intersection, in the drainage channel C near the main gate, and at the drainage channel C OP. 30m exit (on September 11), and conducted nuclide analysis on the water. The analysis results showed no remarkable change from the previous measurement results (on the samples taken on September 9). One remarkable result of the analysis is that the all-β density at the drainage channel C OP. 30m exit (C-2) was 220Bq/L, being about 12 times higher than the previous value (19Bq/L). However, this value is within the range of past fluctuations. Additionally, the densities of cesium-134 and cesium-137 detected there were 24Bq/L (the previous value: below the detection limit value, which was 20Bq/L) and 80Bq/L (the previous value: below the detection limit value, which was 26Bq/L), respectively. Given that these values are at about the same levels as those in the samples taken at this location before cleaning of the drainage channel B, and that we conducted cleaning in an area near the junction of the drainage channels B and C, it is possible that the cleaning has affected these values. We will continue the monitoring. No remarkable change was found in the other analysis results compared to the previous results.

Note that, due to cleaning of the drainage channel B that has been conducted since September 7, we did not conduct sampling at the junction of the drainage channels B and C near the H4 area and at the 3 points (B-1 to B-3) inside the drainage channel B from September 7 to 11.

We conducted analysis for γ nuclides, all-β and tritium on water in observation holes (E-1: on the north of the dike that surrounds tanks including Tank No.5 in Group I in the H4 area from which water has leaked, E-2: on the south of the dike that surrounds tanks including Tank No.5 in Group I in the H4 area from which water has leaked) newly installed near the H4 area tanks.

With regards the groundwater observation hole (E-1) around the H4 area, tritium in the samples taken on September 9 and 10 was detected at 29,000Bq/L and 64,000Bq/L, respectively, indicating that the tritium density there went up from 4,200Bq/L in the sample taken on September 8. As it is highly possible that the leakage from the H4 area tank has affected the density, we will continue analysis on groundwater around the area and investigation to find out the range reached by the leaked water. With regards the groundwater observation hole (E-2), the analysis resulted in values that are about the same as that in the sample taken on September 8.

-We installed observation holes east of the Unit 1-4 Turbine Buildings, and have been conducting sampling and analysis of groundwater from the observation holes. On June 19, we announced that tritium and strontium were detected at high densities in the observation hole located between Units 1 and 2. Therefore, we have been conducting intensified monitoring and have been pumping up groundwater at the well point and the water collection point (south) on the east of Unit 1 and Unit 2 Turbine Buildings.

<The latest groundwater transfer operation>
At 3:50 PM on August 31, the water transfer to the Unit 2 vertical shaft C from the well points and the water collection pit (south) was suspended, and the water transfer to the Unit 2 Turbine Building was started at 3:55 PM on August 31.
So that water in the Unit 2 vertical shaft B (water collected due to the closure of the trench) can be transferred to Unit 2 Turbine Building during the daytime from September 3 to 13, groundwater transfer from the well points and the water collection pit (south) is being conducted while the transfer destination is sequentially changed.
<The latest operation>
· At 1:33 PM on September 7, the transfer destination of groundwater in the well points and the water collection pit (south) was changed back to Unit 2 Turbine Building from the Unit 2 vertical shaft C at.

-We conducted a purification test of wells (sub-drain pits) located next to the Units 1-4 buildings at Fukushima Daiichi NPS, and detected radioactive materials in water accumulated in the pits. One possible cause of the entrance of radioactive materials there is fallout. In order to find out the influence of fallout, we are planning to install new observation wells around the Units 1-4 buildings.

Yesterday (on September 11), we sampled water in the newly installed sub-drain observation well (1T-4) at the sea-facing side of Unit 1 Turbine Building, and the newly installed sub-drain observation wells (2T-1 and 2T-2) at the sea-facing side of Unit 2 Turbine Building. We would like to announce measurement results for radioactive materials such as cesium and all β as we obtained the results today (on September 12).

<Sub-drain observation well 1T-4 (at the sea-facing side of Unit 1 Turbine Building)>
- Sampled on September 11
Cesium-134: ND (0.46Bq/L)
Cesium-137: 0.88Bq/L
All-β: 9500Bq/L

<Sub-drain observation well 2T-1 (at the sea-facing side of Unit 2 Turbine Building)>
- Sampled on September 11
Cesium-134: ND (0.36Bq/L)
Cesium-137: 0.66Bq/L
All-β: ND (24Bq/L)

<Sub-drain observation well 2T-2 (at the sea-facing side of Unit 2 Turbine Building)>
- Sampled on September 11
Cesium-134: ND (0.47Bq/L)
Cesium-137: ND (0.60Bq/L)
All-β: 830Bq/L

Meanwhile, the tritium density in water in the sub-drain observation well (4T-1) at the sea-facing side of Unit 4 Turbine Building, sampled on September 9, was 1800Bq/L.

<Sub-drain observation well 4T-1 (at the sea-facing and south side of Unit 4 Turbine Building)>
- Sampled on September 9
Tritium: 1800Bq/L

* Revised past progress

The aforementioned attachments are only available in Japanese.
We apologize for any inconvenience this may cause.

In Order to view the PDF documents, you will need a software product called Adobe® Acrobat® Reader installed on your computer. You can download this software product for free from Adobe's Web site by clicking the left button:


to TOP