Examination into heat removal by the Unit-1 …...and outlet (after pressure drop due to heat removal and condensation of steam), i.e., this pressure difference provides the driving

Attachment 1-7-1

Attachment 1-7

Status of heat removal in the isolation condenser at Unit-1

1. Introduction

At Unit-1 the reactor was being cooled and its pressure was under control after the

earthquake by intermittent operation of the isolation condenser (IC; a schematic drawing is

shown in Figure 1). Immediately before the station blackout due to tsunami, the IC operation

was temporarily in a halted situation. After the station blackout, the shift operator noticed in

the Main Control Room the indicator lamps were “ON” indicating “CLOSURE” of the IC

(Channel A) isolation valves (MO-2A, MO-3A) outside the containment vessel. The operator

attempted to open the valves at 18:18 on March 11th and confirmed that steam was being

discharged from the reactor building based on witnessing it and hearing its sounds behind

the building. But the amount of steam produced was limited and steam generation ceased

after a while. The operator, having been concerned about depletion of the IC tank water

inventory, closed the return line isolation valve (MO-3A).

In general the IC cooling capacity is considered to deteriorate, when non-condensable

hydrogen gas is generated due to water-zirconium reactions, by the mixing of hydrogen gas

into the cooling lines. From the analysis done so far, the reactor water level was kept at the

level just below the top of active fuel (TAF) before 18:18 and a large amount of hydrogen

might not have been generated. But hydrogen gas might have been also generated by

radiolysis. It is necessary to clarify to what extent the IC heat removal capability deteriorated

at Unit-1.

According to the IC water level surveys conducted after the accident, the tank water level

of Channel A was 65% (vs. normal level at 80%) as of October 18th, 2011, showing there

was sufficient water inventory at the time of IC shutdown. If the return line isolation valve

(MO-3A) had NOT been closed at 18:25 on March 11th, the IC could have continued reactor

cooling. From these considerations, two issues are examined: Why did the amount of steam

generation decline and cease after a while after the IC Channel A isolation valves had been

opened (Unit-1/Issue-1)? What could be the possible influence on the accident progression

if the IC Channel A isolation valve outside the containment vessel had been kept open

(Unit-1/Issue-2)?

Attachment 1-7-2

Figure 1 Schematic diagram of isolation condenser system

2. Evaluation of heat removal capability

MAAP5.01 simulated the IC system and analyzed the accident progression taking into

account the IC operation history from the reactor scram to the station blackout. The results

are presented as Attachment 3 along with Progress Report 2.

After the station blackout, the IC might have been unable to keep its heat removal

capability due to the unexpected non-condensable gas trapped in the IC heat transfer tubes.

The earlier analysis (Attachment 3) therefore assumed that the IC had not been operable

regardless of the IC valve operations after 18:18 on March 11th.

In the current analysis, the IC was assumed to start up upon the valve opening operation

of IC Channel A at 18:18 and further assumed to continue working after the valve closing

operation at 18:25 (this is designated as the IC working case). All other conditions for

analysis were set the same as in the earlier analysis. Also concerning the gaseous phase

leaks from the containment vessel, the reactor building closed cooling water system (RCW)

was assumed to have been damaged, allowing the leaks, upon the RPV being damaged, as

was assumed in the earlier analysis (Attachment 3). Table 1 summarizes the IC operating

conditions in the earlier analysis and the current IC working case.

A B

From Make-Up water

system PCV

Attachment 1-7-3

Table 1 IC operating conditions

Time on

March 11th

Incident Earlier analysis IC working case

14:46 Earthquake

14:48 Reactor scrammed

14:52 IC(A)(B) started up automatically In operation

15:03 IC(A) stopped Not in operation

15:03 IC(B) stopped Not in operation

15:17 IC(A) restarted In operation






15:37 Station blackout

18:18 IC(A) valves 2A,3A opened Not in operation In operation

18:25 IC(A) valve 3A closed Not in operation In operation

21:30 IC(A) valve3A opened Not in operation In operation

3. Examination of the accident progression from the scram to the station blackout

From the scram to the station blackout, there are no differences between the earlier

analysis and the IC working case. Figure 2 compares the measured reactor pressures

recorded on the transient recorder and the MAAP results. The analysis results show bigger

changes compared with the measured data but both are more or less consistent. The IC

system is driven by the pressure difference between the values at its inlet (reactor pressure)

and outlet (after pressure drop due to heat removal and condensation of steam), i.e., this

pressure difference provides the driving force to statically let the steam flow in the IC tubes.

As a result, while the IC is working with its valves opened, the steam is cooled and

condenses in the IC, and then the condensed water returns to the reactor and re-evaporates

by absorbing decay heat. By repeating this cycle, heat energy is transferred to the IC shell

side and the reactor pressure decreases gradually. Along with the decreasing reactor

pressure, the steam flow to the IC tubes decreases and the amounts of steam supplied and

heat removed decrease, as seen in Figure 3 (amount of steam supplied to IC) and in Figure

4 (amount of heat removed), both by MAAP analysis.

The IC shell side receives heat from the reactor and its water temperatures increase.

Figure 5 shows the observed shell side water temperatures (chart readings) and MAAP

Attachment 1-7-4

results. The water temperature increase on the shell side is in agreement between

measurement and analysis. That means the IC heat removal capabilities are fairly well

simulated in the analysis.

0

1

2

3

4

5

6

7

8

9

10

3/11 14:47 3/11 14:57 3/11 15:07 3/11 15:17 3/11 15:27 3/11 15:37 3/11 15:47Date/time

0

1

IC弁

開信

号（0：閉

、1：開

)

Reactor pressure (MAAP5.01)Reactor pressure measured (1min interval data)Reactor pressure measured (0.01s interval data)IC valve 3A opening signal (1min interval data)IC valve 3A opening signal (0.01s interval data)IC valve 3B opening signal (1s interval data)IC valve 3B opening signal (0.01s interval data)

RP

V p

ress

ure

(M

Pa[

gage

])

IC v

alve

ope

nin

g si

gnal

(0:c

lose

, 1:o

pen)

Figure 2 Reactor pressures (observed data from the transient recorder and analysis results)

and the timings of IC valves opening

0

5

10

15

20

25

3/11 14:47 3/11 14:57 3/11 15:07 3/11 15:17 3/11 15:27 3/11 15:37 3/11 15:47

Date/time

Amount of steam supplied to IC Channel AAmount of steam supplied to IC Channel B

Am

ount

of

steam

supp

lied

to IC

(kg

/s)

Figure 3 Amount of steam supplied to IC (MAAP analysis)

Attachment 1-7-5

0

5

10

15

20

25

30

35

40

45

50

3/11 14:47 3/11 14:57 3/11 15:07 3/11 15:17 3/11 15:27 3/11 15:37 3/11 15:47

Date/time

Amount of heat removed by IC Channel A

Amount of heat removed by IC Channel B

Am

ount

of

heat

rem

ove

d by

IC

(M

W)

Figure 4 Amount of heat removed by IC (MAAP analysis)

0

20

40

60

80

100

120

140

160

3/11 14:47 3/11 14:57 3/11 15:07 3/11 15:17 3/11 15:27 3/11 15:37 3/11 15:47Date/time

Analysis (Water temperature of IC Channel A tank)Analysis (Water temperature of IC Channel B tank)Measured (Water temperature of IC Channel A tank)Measured (Water temperature of IC Channel B tank)

Wat

er

tem

pera

ture

of

IC t

ank

(℃)

Figure 5 IC shell side temperatures (measured (chart readings) and MAAP analysis)

Attachment 1-7-6

4. Evaluation of the accident progression after the station blackout

Concerning the IC operating conditions after the station blackout, the closing operation of

IC Channel A isolation valve at 18:25 was ignored in the IC working case, as explained in

Section 2 above, and the IC was assumed to have continued its operation. Figures 6

through 12 present the results of the IC working case and the earlier analysis after 18:18,

when the IC Channel A isolation valve was opened: Figure 6 Reactor water level changes;

Figure 7 Reactor pressure changes; Figure 8 Containment vessel pressure changes; Figure

9 Core temperature changes; Figure 10 Gas temperature changes in the reactor vessel;

Figure 11 Containment vessel temperature changes; and Figure 12 Amount of hydrogen

generated.

Upon the IC startup at 18:18, the IC starts removing heat and the reactor pressure

decreases drastically. It should be noted that fuel temperatures are in an increasing trend,

because, according to the MAAP analysis, the reactor water level had cut the TAF level

before 18:18 when the IC started up and the fuel began to be uncovered. The IC can

remove the heat, but it does not inject external water to the reactor and it cannot restore the

level to flood the core. Current MAAP analysis shows that the reactor water and its

evaporation are not enough to prevent fuel temperatures from increasing. As the fuel

temperatures increase, the amount of hydrogen gas generated by water-zirconium reactions

sharply increases, this hydrogen gas trapped in the IC tubes blocks the steam flow, and thus

the IC heat removal capacity is quickly deteriorated and eventually lost. The reactor

pressure increases rapidly thereafter and shows the similar accident progression to the

earlier analysis.

No considerations were given to water radiolysis in the current MAAP analysis. In the IC

working case MAAP gives the results that the heat removal by IC startup at 18:18 decreases

very rapidly after around 19:00 when the water-zirconium reactions start generating

hydrogen gas and that it drops to almost zero at around 19:05 when about 20kg of hydrogen

is generated. The amount of non-condensable gas (hydrogen, oxygen) due to water

radiolysis is proportional to the decay heat. The amount of non-condensable gas from the

scram to 19:05 is about 1.5kg, which is less than 10% of the amount of hydrogen generated

by the water-zirconium reactions. It should also be noted that, upon the SRV activation, the

non-condensable gas is also discharged from the RPV together with steam, thus further

decreasing its amount remaining in the RPV. In conclusion, it is considered that the

influence of water radiolysis on the IC heat removal capability is very limited and the heat

removal capability deterioration is mainly due to water-zirconium reactions.

When comparing the results of the IC working case and the earlier analysis, it can be

noted that damage of core support plate and RPV is delayed in the IC working case. This

Attachment 1-7-7

means the heat removal by the IC after 18:18 could delay the accident progression but it

was not enough to stop the accident progression itself.

Damage of the main steam line flange gaskets takes place earlier in the IC working case

than in the earlier analysis. This is because the RPV atmosphere temperature was just short

of 450 deg C in the earlier analysis at which temperature the flange gaskets were assumed

to be damaged. This difference is considered to come from the difference in the conditions

in the reactor (for example, the reactor water level or fuel temperatures when the in-core

instrumentation dry tubes were damaged) or the difference in the amount of water-zirconium

reaction products in the analysis models. Certainly, in the two analyses there is a difference

in the timing of damage of main steam line flange gaskets and the ensuing reactor pressure

decrease, but the results in both are essentially the same.

Concerning the gas temperatures in the RPV, the IC working case gives generally higher

values. This is because, in the IC working case, the steam condensed in the IC can

evaporate again when returned to the reactor, and thus water-zirconium reactions are

considered to further continue due to this newly generated steam. But still, there is no big

difference in the timing for the reactor water level to cut the BAF level, when steam

generation ceased. Thus, only a small difference remains in the accident progression from

the earlier analysis.

Figure 13 shows the heat removal by the IC (Channel A) and the non-condensable gas

partial pressure, while Figure 14 gives the water inventory in the IC tank. The IC heat

removal slightly recovers at the timings of the in-core instrumentation dry tubes damage,

main steam line flange gaskets damage and reactor pressure decrease after the core

support plate damage. This comes from the analytical model in which part of the hydrogen

trapped in the IC returned to the RPV, but it is not certain whether such hydrogen return flow

to the RPV can physically take place. Regardless of whether hydrogen gas returned to the

RPV or not, the amount of water in the Channel A tank consumed for cooling is limited and

the amount of water consumed is only about 30 to 40%, according to the analysis.

Attachment 1-7-8

-10

-8

-6

-4

-2

0

2

4

6

8

10

3/1112:00

3/1114:00

3/1116:00

3/1118:00

3/1120:00

3/1122:00

3/120:00

3/122:00

3/124:00

3/126:00

3/128:00

3/1210:00

3/1212:00

3/1214:00

3/1216:00

3/1218:00

3/1220:00

3/1222:00

Date/time

Water level inside shroud (MAAP5.01IC working case)Downcomer water level (MAAP5.01IC working case)Water level inside shroud (MAAP5.01)Downcomer water level (MAAP5.01)Measured water level (Fuel range-A)Measured water level (Fuel range-B)Measured water level (Fuel range)Measured water leve (Wide range)

TAF (Top of Actie Fuel)

BAF (Bottom of Active Fuel)

*:The analysis results after the RPVrupture do not mean water levels arekept constant.

Freshwater injection started

Reached TAF around 18:00

on March 11th

Seawater injection started

RPV ruptured

Freshwater injection resumed

Core Support Plate damaged


RPV ruptured

Reached BAF around 20:20

on March 11th

IC started up at 18:18

on March 11th andcontinued its operation(assumption)

Reac

tor

wat

er

leve

l (m

)

Figure 6 Reactor water level changes (IC working case (red) vs. earlier analysis (blue))

0

2

4

6

8

10

3/1112:00

3/1114:00

3/1116:00

3/1118:00

3/1120:00

3/1122:00

3/120:00

3/122:00

3/124:00

3/126:00

3/128:00

3/1210:00

3/1212:00

3/1214:00

3/1216:00

3/1218:00

3/1220:00

3/1222:00

Date/time

RPV pressure (MAAP5.01 IC working case)

RPV pressure (MAAP5.01)

Measured pressure (Reactor-A)

Measured pressure (Reactor-B)

Pressure decreasedupon IC start-up

Gaseous leak from in-core instrumentationtube (assumption)

Gas phase leak fromgaskets of flanges onMain Steam lines(assumption)


RPV ruptured*1 Gaseous leak fromin-core instrumentationtube (assumption)

*2 Gaseous leak from gasketsof flanges on Main Steam lines(assumption) Core Support Plate damaged

*1*2

RPVruptured

IC started up at 18:18 on

March 11th and continuedits operation (assumption)

RP

V p

ress

ure

(M

pa[a

bs])

Figure 7 Reactor pressure changes (IC working case (red) vs. earlier analysis (blue))

Attachment 1-7-9

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

3/1112:00

3/1114:00

3/1116:00

3/1118:00

3/1120:00

3/1122:00

3/120:00

3/122:00

3/124:00

3/126:00

3/128:00

3/1210:00

3/1212:00

3/1214:00

3/1216:00

3/1218:00

3/1220:00

3/1222:00

Date/time

D/W pressure (MAAP5.01 IC working case)

S/C pressure (MAAP5.01 IC working case)

D/W pressure (MAAP5.01)

S/C pressure (MAAP5.01)

D/W pressure measured

S/C pressure measured

Gaseous leak from in-core instrumentation tube (assumption)

Gaseous leak from gaskets of flanges on Main Steamlines (assumption)


RPV damagedPCV leakage (assumption)

S/C vent valve opened

S/C vent valve closed

Seawater injection started

Seawater injectionhalted

Gaseous leak from in-core instrumentation tube (assumption)

Gaseous leak from gaskets offlanges on Main Steam lines(assumption)

Freshwater injection

Freshwater injection resumed

RPV rupturedPCV leakage（assumption）



March 11th and continuedits operation (assumption)

PC

V p

ress

ure

(M

pa[a

bs])

Figure 8 PCV pressure changes (IC working case (red) vs. earlier analysis (blue))

0

500

1000

1500

2000

2500

3000

3500

3/1112:00

3/1114:00

3/1116:00

3/1118:00

3/1120:00

3/1122:00

3/120:00

3/122:00

3/124:00

3/126:00

3/128:00

3/1210:00

3/1212:00

3/1214:00

3/1216:00

3/1218:00

3/1220:00

3/1222:00

Date/time

Core max. temperature (MAAP5.01 IC working case)

Core max. temperature (MAAP5.01)

Core damage started

around 18:40 on March 11th

Core damage started

around 19:00 on March 11th

IC started up at 18:18 on March

11th and continued its operation(assumption)

Core

max

imum

tem

pera

ture

(deg.

C)

Figure 9 Core temperature changes (IC working case (red) vs. earlier analysis (blue))

Attachment 1-7-10

0

100

200

300

400

500

600

700

800

3/1112:00

3/1114:00

3/1116:00

3/1118:00

3/1120:00

3/1122:00

3/120:00

3/122:00

3/124:00

3/126:00

3/128:00

3/1210:00

3/1212:00

3/1214:00

3/1216:00

3/1218:00

3/1220:00

3/1222:00

Date/time

In-RPV temperature (gas phase) (MAAP5.01 IC working case)

In-RPV temperature (gas phase) (MAAP5.01)

Gaseous leak from gaskets of flanges onMain Steam lines (assumption)


Flange gaskets damage temperature on MainSteam lines（450℃・assumption）



In-R

PV

tem

pera

ture

(gas

phas

e)

(deg.

C)

Figure 10 Gaseous temperature changes in RPV

(IC working case (red) vs. earlier analysis (blue))

0

100

200

300

400

500

600

700

800

3/1112:00

3/1114:00

3/1116:00

3/1118:00

3/1120:00

3/1122:00

3/120:00

3/122:00

3/124:00

3/126:00

3/128:00

3/1210:00

3/1212:00

3/1214:00

3/1216:00

3/1218:00

3/1220:00

3/1222:00

Date/time

D/W temperature (MAAP5.01 IC working case)

S/C temperature (liquid phase) (MAAP5.01 IC working case)

D/W temperature (MAAP5.01)

S/C temperature (liquid phase) (MAAP5.01)

IC started up at 18:18 on March 11th andcontinued its operation (assumption)

PC

V t

em

pera

ture

(de

g. C

)

Figure 11 PCV temperature changes

(IC working case (red) versus earlier analysis (blue))

Attachment 1-7-11

0

200

400

600

800

1000

1200

1400

1600

1800

2000

3/1112:00

3/1114:00

3/1116:00

3/1118:00

3/1120:00

3/1122:00

3/120:00

3/122:00

3/124:00

3/126:00

3/128:00

3/1210:00

3/1212:00

3/1214:00

3/1216:00

3/1218:00

3/1220:00

3/1222:00

Date/time

Total hydrogen (MAAP5.01 IC working case)

Total hydrogen (MAAP5.01)

Core damage startedaround 18:40 on

March 11th

Core damage startedaround 19:00 on

March 11th



Hyd

roge

n p

rodu

ction (

kg)

Figure 12 Hydrogen gas generation (IC working case (red) vs. earlier analysis (blue))

0

5

10

15

20

25

30

35

40

45

50

3/1114:00

3/1116:00

3/1118:00

3/1120:00

3/1122:00

3/120:00

3/122:00

3/124:00

3/126:00

3/128:00

3/1210:00

3/1212:00

3/1214:00

日時

0

0.05

0.1

0.15

0.2

0.25

0.3

0.35

Amount of heat removed by IC Channel A

Non-condensable gas partial pressure in RPVIC started up at 18:18on March 11th andcontinued its operation(assumption)

RPV rupturedPCV leakage（assumption）


*1 Gaseous leak from in-coreinstrumentation tube(assumption)

*2 Gaseous leak from gasketsof flanges on Main Steam lines(assumption)

*1*2

Am

ount

of

heat

rem

ove

d by

IC

Chan

nel A

(M

W)

Par

tial

pre

ssure

in R

PV

(-)

Figure 13 Amount of heat removed by IC Channel A and non-condensable gas partial

pressure in RPV (IC working case)

Attachment 1-7-12

0%

20%

40%

60%

80%

100%

3/1114:00

3/1116:00

3/1118:00

3/1120:00

3/1122:00

3/120:00

3/122:00

3/124:00

3/126:00

3/128:00

3/1210:00

3/1212:00

3/1214:00

3/1216:00

3/1218:00

3/1220:00

3/1222:00

Date/time

Water inventory in IC Channel A tank

Water inventory in IC Channel B tank

Gaseous leak from in-coreinstrumentation tube (assumption)


RPV damagedPCV leakage（assumption）



March 11th and continuedits operation (assumption)W

ater

inve

nto

ry in IC

tan

k (-

)

Figure 14 Amount of water inventory in IC tanks

Attachment 1-7-13

5. Conclusion

The accident progression was examined for the case in which the IC was assumed to

have remained operable beyond 18:25 when the IC had been actually deactivated after

being activated at 18:18. The analysis results showed that, even if the IC had continued its

operation, the hydrogen gas generated by water-zirconium reactions would have trapped in

the IC tubes and deteriorated their heat removal capability eventually to null. In the IC

working case assuming continued IC operation, the timing of RPV damage could be delayed,

but no big difference in the accident progression could have taken place from the actual

situation of Unit-1 to date.

Documents

Examination into heat removal by the Unit-1 …...and outlet (after pressure drop due to heat removal and condensation of steam), i.e., this pressure difference provides the driving