ML19274F278
| ML19274F278 | |
| Person / Time | |
|---|---|
| Site: | Zion File:ZionSolutions icon.png |
| Issue date: | 05/30/1979 |
| From: | Resnikoff M ILLINOIS, STATE OF |
| To: | |
| References | |
| NUDOCS 7907130059 | |
| Download: ML19274F278 (26) | |
Text
.
1' b
1W UNITED STATES OF AMERICA
.iUCLEAR REGULATORY CO:OIISSION BEFORE THE ATOMIC SAFETY AND LICENSI:JG BOARD I;; THE MATTER OF:
)
)
COCKET : OS. 50-295 CO.'DIONWEALTH EDISON COMPANY
)
50-304 (ZION STATION CMITS 1& 2,)
)
- 9 C4,
N
- ccaer, U3Nac g,
g
- N 4 I373 )
SPENT FUEL POOL RERACKING D 5,7 h-.
n b
f o7:~ id' TESTIMONY OF, Marvin Resnikoff, on behalf of the State of Illinois -
Office of the Attorney General DATED:
May 30, 1979 O
sa.)
9907 ; o we G
e UNITED STATES OF AMERICA NUCLEAR REGULATORY COMMISSION BEFORE THE ATOMIC SAFETY AND LICENSING BOARD In the Matter of
)
)
COMMONWEALTH EDISON COMPANY
)
(Zion Station Units 1 and 2)
)
Docket Nos. 50-295
)
50-304 Amendments to Facility Operating
)
License Numbers DPR-39 and DPR-48
)
(Increased Spent Fuel Storage
)
Capacity)
)
DIRECT TESTIMONY OF MARVIN RESNIKOFF My name is Marvin Resnikoff. I am employed as a consultant for the Attorney General of the State of Illinois. A statement of my professional qualifications is attached to this affidavit.
This testimony addresses the possibility of water boiling in the spent fuel pool and the consequences of a partial loss of coolant accident. My conclusions are that it is possible for the spent fuel pool to boil with no heat exchanger operative and that, if no make-up water were added, an accident with major consequences could take place, even if the " greenest" spent fuel were cooled more than two years. This accident is possible under a major reactor accident scenario, or simply through neglect. Furthermore, boiling may take place under the conditions of a full pool load, with 1 1/3 cores of
" green" spent fuel and only one heat exchanger operative. In this case the boiling would be localized. If there were blockage thrcugh a baseplate hole under " green" spent fuel, the boiling could be quite
vigorous and lead to a metal-water reaction and possibly an accident with major consequences.
Under adiabatic conditions (no heat loss through the pool walls and no evaporative heat loss) which would be approximately the case in the Zion situation, the time to boiling from 150*F can vary from 6.3 to 12.9 hours1.041667e-4 days <br />0.0025 hours <br />1.488095e-5 weeks <br />3.4245e-6 months <br />, depending on the spent fuel pool heat load. The two cases considered are (i) a normal spent fuel load, with 1/3 core " green" spent fuel and a total of 2112 spent fuel assemblies, and (ii) the Worst case, with 1 1/3 cores " green" spent fuel and a total of 2112 spent fuel assemblies. Water would boil-off between the rates 46 and 94 gallons per minute. If no make-up water were added, the top of the spent fuel rack would be uncovered in a time period between 2.9 and 5.9 days from the initiation of boiling.
If the spent fuel assemblies were uncovered, the accident would then become very serious. The spent fuel assemblies would heat up and a metal-water reaction would take place between the steam and zircalloy cladding. Above 920*C this is an exothermic reaction, releasing more heat than the well-known interaction between sodium and water. The cladding would become brittle, crack apart, and fall into the boiling spent fuel pocl, causine the boiling to become more vigorous. The steam-zircalloy reaction would very rapidly produce an explosive mixture of hydrogen gas. A pressure spike of 28 psi at Three Mile Island, about 10 hours1.157407e-4 days <br />0.00278 hours <br />1.653439e-5 weeks <br />3.805e-6 months <br /> into that accident, was due to this steam zircalloy reaction or hydrogen explosion resulting from it. A similar spike at Zion spent s
. fuel pool would open up the entire spent fuel building and a major release of radioactivity would ensue. Because of the accumulated cesium and strontium, equivalent to 11 reactor cores in a fully loaded spent fuel pool, and the possibility that the spent fuel building would be open to the environment, the accident would be much more serious than a reactor melt-down accident, and could, depending on meteorological conditions, contaminate a major area of the country for a long period of time producing over a million immediate and long-term deaths, as well as genetic effects.
This major accident would take place if no heat exchanger were operative and no make-up water were added. This is possible under accident conditions or simply through neglect. To prevent this accident, it is recommended that the make-up water supply and cooling trains be fully automated and independent of reactor operation so that the spent fuel assemblies can continue to be cooled under any accident scenario. It is further recommended that the safety margin within the spent fuel pool be increased by not closely packing spent fuel immediate with the edge of the spent fuel pool, but allowing a region about the entire pool, and between the racks, for convective water to ficw.
As shown,with this safety margin the pool would contain 1746 fuel assemblies, and therefore would allow the applicant to store fuel through 1989. Finally, it is important to note that a reactor accident, and/or a spent fuel pool accident with a major release of radicactivity,
_ 4_
is greatly facilitated by the metal-water reaction, which adds heat, an explosive hydrogen mixture, and exposes the spent fuel pellets to the environment. Therefore, an added safety margin would take place at Zion if all circalloy clad fuel assemblies were replaced with stainless steel.
UNDER WHAT CONDITIONS MAY THE ZION SPENT FUEL POOL BOIL?
The Zion spent fuel storage pool receives spent fuel from two pressuri:ed water reactors.
The proposed licensing amendment would increase the capacity of the storage pool from 868 to 2112 spent fuel storage spaces, capable of holding one spent fuel assembly each.
A neutron poison material is contained within the stainless steel racks to prevent criticality.
These high density racks have a center-to-center spacing of 10.25 inches between spent fuel assemblies.
Each rack is closely spaced.
According to Fig.
3.
1-2, Licensing Report, each collection of spaces, racks, are only 2 13/16" apart.
The entire cluster of racks is only 9" from the east and west walls of the spent fuel storage pool, and actually is adjacent to the south edge of the storage pool.
According to Figure 3.2-2, each spent fuel assembly space has a bottom hole to enable convective water flow, though the hole size is not specified in the application.
The closer spacing of the fuel assemblies and the additional hea load introduce two important considerations in this application which were not previously reviewed in the FSAR.
Calculations must be
. performed to determine whether the bulk pool temperature will remain below boiling, assuming a certain heat load and perfect mi::ing.
Next one must determine that none of the individual fuel assemblies will boil.
Clearly the center fuel assembly of a rack of " green" spent fuel will be hotter because the coolant will not as easily reach the center, and because the center fuel assembly would be heated by the other fuel assemblies.
Our procedure will be the following.
After first determing the total heat load under two cases, (i) a normal, fully-loaded (2112 spent fuel assemblies) spent fuel pool, and (ii) the worst case situation, with 1 1/3 cores of " green" spent fuel (less than or equal to 10 days) in a fully-loaded fuel pool, we then calculate the bulk pool temperature and determine whether the pool will boil
.with one heat exchanger, or no heat exchangers in service.
The conclusion then is that the spent fuel pool can boil if no heat ex-changers were operative and no make-up water were added, even if the spent fuel had cooled for two years. The possibility that no heat exchangers are operative can occur in the case of a malfunction of the component cooling system, which exchanges heat with the spent fuel pool.
This requires a short digression into the operation of the component cooling system.
Or,
. lized boiling may take place under the conditions of a full pool load, with 1 1/5 cores of " green" spent fuel and only one heat exchanger operative.
. Total heat load.
To calculate the heat load, I use the standard ORIGEN computer code results rather than the Westinghouse reference used by the applicant (S.B. Gunst, et al, " measured and 235 233 239 Calculated Rates of Decay Heat in Irradiated U,
U, Pu and 232 Th",
Nucl. Sci & Engineering 56, pp.241 (1975).
The latter paper does not adequately account for cladding activation and actinide build-up.
In agreement with the NRC Staff, the maximum peak heat 6
loak during the 33rd refueling would be 22.2 x 10 Stu/h.
The maximum heat load for a full core off-load, from one Zion unit immediately after 1/3 core load from the other, such that i 1/3 6
cores is cooled less than 10 days is 45.5 x 10 Stu/h.
This latter number is 10% higher than the NRC Staff's estimate (p.2-4, Zion SER).
Loss or less effective use of spent fuel pool heat ex-changers.
The spent fuel pool cooling system exchanges heat with the component cooling system.
Under accident conditions, the component cooling system may be overloaded and may heat up.
As a result, the spent fuel pool heat exchangers may not remove enough heat to cool the spent fuel pool.
The component cooling system remcves heat from the following systems:
heat exchangers of the residual heat removal system, spent fuel pool, seal water, sample and let down systems, and the reactor coolant pumps, residual heat removal pumps, safety injection pumps, charging pumps, waste gas compressor and reactor support cooling.
As shown in Table 9.3-2, the largest load is due to the residual heat re-oval system and the spent fuel pool, with remaining systems requiring about 15% of the component cooling system flow.
The
. component cooling system consists of two loops serviced by two separate heat exchangers and a third heat exchanger common to the two icops.
The component cooling system exchanges heat, via three heat exchangers, with the service water system (Lake Michigan).
The design heat transfer of the 3 heat exchangers together is 3x 6
53 x 10 Btu /h (Table 9.3-3, Zion SAR).
In the event of an accident at Zion which requires the shutdown of both units, they would be cooled down rapidly using the steam and power conversion system (p.9.4-1, Zion SAR).
After 20 hours2.314815e-4 days <br />0.00556 hours <br />3.306878e-5 weeks <br />7.61e-6 months <br /> (p.(.4-3), cooling would be transferred to the residual heat removal system.
The cooling load would be 68.7 Stu/h at 20 hours2.314815e-4 days <br />0.00556 hours <br />3.306878e-5 weeks <br />7.61e-6 months <br /> after shutdown.
As shown in Figure 1 below, the total load would be 8
about 1.33 x 10 Stu/h, whereas the design heat transfer capability 8
.is only 1.59 x 10 Stu/h.
The effect is that the RHR systems would not cool Figure 1.
Component cooling load in shutdown mode.
Zion 1 Zion 2 shutdown shutdown RHR igad RER load 68.7 x 10 Btu /h 63.7 x 10 Btu /h 8
spent fuel total: 1.6 n 10 atu/h 22.2 x 10"pocl g
Stu/h
-15%
.23 x 10. Stu/h component j
cooling load = 1.83 x 10 Bru/h the reactor down as rapidly, and that the component coolant would be at a higher temperature.
This would less effectively cool the spent fuel pool therefore.
If a pump breakdgwn occurred in the component cooling syste, the situation would become more serious.
_a_
This component ecoling system problem is due to the fact that the additional heat load in the spent fuel pool is now above the design heat transfer capability for the 3 component cooling heat exchangers for this shutdown mode.
This was not the case under the original licensing.
Time to Boiling.
No Heat Exchangers.
To determine how rapidly the fuel pool will heat up (1) under adiabetic conditions, that is, with no heat loss, we need to know the weignt of water in the pool.
The volume of the pool is 4
3 8 x 10 ft (Table 9.5-2, Zion SAR) less the volume of the spent 3
3 fuel assemblies 5.32 x 10 ft gives us the volume of water plus storage racks, 7.42 x 10 ft Under the two heat loads mentioned above, the temperature rise is 9.82 F/h and 4.80 F/h, respectively,
'for the pool bulk temperature rise.
The time for the bulk temp-0 erature to reach boiling from 150 F is 6.3 to 12.9 hours1.041667e-4 days <br />0.0025 hours <br />1.488095e-5 weeks <br />3.4245e-6 months <br />.
The Zion SAR (p.9.5-2) indicates that the time recuired fcr the spent fuel pool to reach boiling is 30 hours3.472222e-4 days <br />0.00833 hours <br />4.960317e-5 weeks <br />1.1415e-5 months <br />; the Licensing Report mentions 3.2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> under the worst case situation.
Of course, certain of the hotter fuel assemblies, especially those located closer to the center, will begin to boil before the bulk pool temperature reaches boiling.
At the boiling point, with continuing.: eat input to the pool, the water will begin to boil.
The amount of water boiled off per minute is 93.6 gpm to 45.7 gpm.
_9_
Time to Evaporate Water to Top of Fuel Rack.
- !o Heat Exchangers.
The top of the fuel rack is 170.3" off the pool floor.
4 The depth of the pool is 40.4'.
For 5.19 x 10 ft" of water, the time required to boil off this volume will vary between 69.1 and 141 hours0.00163 days <br />0.0392 hours <br />2.331349e-4 weeks <br />5.36505e-5 months <br />.
At this point, with the water at the top of the fuel assemblies, an accident with major consequences can ensue.
I return to this point shortly.1 One Heat Exchanger.
Worst Case Heat Load.
In this case, we consider 1 1/3 cores of fuel which has cooled for 10 days of less, and one heat exchanger operative.
Assuming complete mining of water and the worst case loading, the bulk pool temperature can be calculated.
The model for this case is to take the diffepence between the heat entering the pool (due to the decay heat within the fuel rods) and the heat leaving the pool entering the one heat exchanger (this will be proportional to the difference in temperature between water entering and exiting the heat exchanger).
This difference between incoming and outgoing heat is equal to the heat remaining within the pool.
As the pool temperature rises, the heat leaving the pool through the heat exchanger will rise, until the heat leaving the pool will equal that entering the pool.
The temperature will level off at that point.
The change of temperature within the heat ex-changer will then be 35.5 F.
If the inlet temperature is 107 1,,
e the bulk pool temperature would be 142.5 F.
However, in the high-density arrangement projected by the applicant, it is possible for the center-located fuel assemblies to heat up more than the outer
fuel assemblies (those near the wall).
The calculations by.iSC (NSC-COM-0220-LO21) show that for a blocked tube, the temperature within a space can rise 94 F within the tube, which raises the temperature above the boiling point.
Though this is the case for a blocked tube, (not our case), there are several aspects of the problem which have not been considered by NSC which leads one to believe that the center temperatures will be greater than the o
32 F rise, for the case of non-blocked bottom holes predicted by USC.
The applicant's models, PCOHLT and CIRCUS, are too simple.
They only assume convective flow through a tube, with a downcomer at the pool edge.
As far as can be determined, they do not consider:
1) the size of the drain hole at the bottom of the assembly space.
This drain hole has the effect of deterring the flow of coolant water.
The size of the drain hole has an important effect on heat up of fuel assemblies in air ~ and will have a similar effect in water.
2) the fact that downcomer size has an important effect on cooling. As the spent fuel arrangement is projected, the assembly racks would be adjacent to the sough wall of the pool, and the space would be a mere 9" on the east and west sides.
The spacing between the racks is less than 3".
This will considerably hinder the convective flow.
The Sandia report (Table VI)" shcws the considerable difference between a one foot width downcomer, and wall-to-wall racks.
3) the radiative heat flows, only convective.
The effect of this oversimplification is that the heating effect cf fuel assemblies adjacent to center assemblies is not taken into account.
The radiative effect will increase the temperatures of the center fuel assemblies.
The correct model is indicated in Figures 7 and 2 of the Sandia report.4 Of course, the Sandia report calculates heat-up due to a loss-of-coolant accident, that is, heat up in air, but the model is similar for water.
4) the effect of imperfect ventilation.
The Sandia report clearly shows that the air will heat up and will affect the temperature of the cladding (Figure 21, Sandia report).
In sum, a correct calculation has not been performed by the applicant, but based on the results for a blockel tube con-vective flow, in which the temperature will go above boiling, it is my conclusion that localized boiling will occur for this high-density configuration.
The important question is, what will the temperature of the cladding be under those conditions?
This calculation has not been performed.
.THE CONSEQUENCES OF UNCOVERED FUEL If the tops of the fuel assemblies were to become uncovered, a serious accident would ensue.
This situation is not evaluated in detail in the Sandia report where only heat up in air is considered.
Illustrative calculations are performed here.
The possible mechanisms for heat remeval from the decay heat are convection, radiation and conduction.
Convective flow is the dominant contribution.
This is shown quite clearly in Figure 13, Sandia Report.
Within each storage space, the steam heats up as it rises.
This convection will not icwer pond heat loads and leads to cladding temperatures well in excess of 920 C, the significance of which will be seen shortly.
In this case of forced convection by steam, the temperature of steam leaving the top of the fuel assembly depends on the fraction of the assembly exposed and not on the pond heat load.
Taking the average specific heat of steam as 1/2 cal (1 bar pressure, T from 100 C to 300 C) g K,
and the latent heat of boiling water (1 bar pressure) as 540 cal /g, we have (540 e
) t 100 T= (1/2 l-e
)
where T(o ) is the temperature of the steam leaving the upper part C
of the fuel element, and e is the fraction of active length of fuel exposed.
The results are shown in Table 1:
Table 1. Temperature of Steam Leaving Fuel Element Exposed Fraction of Steam Temperature
_ Active Length, e T (OC) 0.3 561 0.4 817 0.5 1175 0.6 1713 0.7 2609 T
. The fuel pellet will greatly exceed the above temperatures for fuel which has coo' led for less than two years.5 For cladding temperatures above 920 C, a metal-water 0
reaction takes place Zr + 2H O
) ZrO2 + 2H (1) 9 2
This reaction is exothermic and releases 215 kcal/ mole Zr.
The Sandia report discusses the reaction Zr + 0 --> Zr0 which is also 2,
exothermic and releases 262 kcal/ mole Zr.
- Iowever, in the steam' environment, Ea. (1) is preferred If an adequate supply of steam 7
is available, the reaction follows a parabolic rate law:
5 dg = - 3.97 x 10 exp ( - 22889 )
(2) dt (r.-r)
T where r= radius of reacting interface (m) r= initial radias (m) = 0.53cm o
T= interface temperature ( K) t=
time (sec)
If we consider an early stage of the reaction, when r/rg
= 0.99 and assume a cladding temperature of 1175 C for 1/2 the fuel assembly exposed (see Table 1), we find (r
= 0.53 cm), from Eq. (2) g
- dr = 1.02 x 10~
m/sec ct 3
The mass burned per m length of fuel rod (density 6. 55 :4T/m ggr 3r)
~
2.23 x 10 kg/sec 2
The heat output /m length of fuel rod (.215 kcal/ mole) is 0.22 sw.
The heat output due to this exothermic metal-water reaction (2112 assemblies, 204 rods per assembly, 1/2 of active length exposed) is 17 3.'M for the entire storage pool.
The 'r consumption for the fuel storage pool is theri 17.6 kg/sec.
The M ev luti n for the storage 2
pool.is 0.44 kg/sec.
The H O consumption for the storage. pool is 2
6.95 kg/sec.
The above metal water reaction is steam-limited.
A pond heat load of 13.3 MW, with 1/2 the fuel element within the water, generates steam at the rate of 2.96 kg/sec (1/2 c f 93. 6 gpm).
This steam generation rate is less than the H O consumption rate of 2
6.95 kg/sec in the metal-water reaction.
Of course, some of the 173 MU generated will radiate the pcol water and generate larger quantities of steam.
The hot Z
.aetal may shatter and fall into the pcol, as well.
It is important to emphasize that the fuel pool generates decay heat of 13.3 MW, but the metal-water reaction produces heat at the rate. of 173MW.
The temperature of the building will rise tery rapidly, and a large amount of hydrogen will be produced.
A hydrogen explosion could take place which would open the spent fuel storage pool to the outside environment.
If all the cladding in a fuel rod were to react, mass
= 0.542 kg, the total energy released would be 1.49 kwh.
,r/ rod u
If all this energy were retransferred to the 2.21 kg fuel pellets in a rod (assuming the specific heat of the fuel pellets to be
.3 kwh/kg K) the fuel pellets would rise to 7.81 :: 10 C.
According 0
to Lewis, fuel melting occurs at 2800 C, and vaporization occurs at C
3300 C.
If the reaction goes sufficiently fast, then some fuel melting will occur.
-lS-As pointed out in the beginning of this section, this calculation is meant to be illustrative and recuires more detailed study.
It is clear that the steam-circalloy reaction produces a substantial energy release and it mut be supposed that radionuclides will be released from the spent fuel.
The high heat will vaporize certain of the radionuclides.
If more than about 18% of the cladding is oxidized, then the cladding may fragment due to thermal shock.
The metal-stezm reaction will easily produce sufficient heat to reduce the strength of the concrete and sufficient hydrogen to reach an explosive mixture.
Concrete itself will lose strength with increasing temperature.
U Complete loss of strength occurs at 1050 K.
As shown previously, this temperature will easily be exceeded.
The production of hydrogen proceeds very rapidly.
An explosive mixture is reached in a matter of seconds. Assuming that 5
the air space within the building is 1.6 x 10 (double the pool volume) the lower flamnability level (4% H by volume) is reached following g
the evolution of 14 kg H2 An explosive mixture is reached in twice the tim'.
The, evolution rate for H is 0.77 Kg/sec when 1/2 e
2 the length of the fuel assemblies are exposed.
At this rate, it takes 18.2 seconds to reach a flammable level, and twice that to reach an explosive mixture.
Clearly the combination of weakened concrete and an H 2
explosion will produce a substantial breach 'in the spent fuel building providing a pathway to the outside environment.
In the next section,
-lG-Before proceeding to that calculation, we note in passing that there is sufficient water in the poll to allow the metal-water
~
reaction to oxidice all the circonium.
I estimate 234 MT of Zr in the cladding of 2112 fuel assemblies.
The metal water reaction, if it proceeds to completion, will generate 10.2 MT of H fr
- 2 92.4 MT H 0.
However, the pool contains 2.27 x 10 MT of H 0, 3
clearly enough to oxidize all the Zr.
It is important to also add that this entire accident scenario assumes only that the spent fuel pool cooling is cut off, through the loss of heat exchangers.
Another initiating event for a major accident can also be the direct loss of cooling through severe cracking of the pool walls with resultant leakage of water.
According to the Sandia Report, this will, within hours, give rise to a metal-air reaction, with the release of more heat (262 kcal/
mole v. 215 kcal/ mole) than the metal-water reaction.
It is clear that for spent fuel which has cooled more than two (2) years, the metal-water reaction would not immediately occur.
The Cr. water reaction would occur in the " greener" spent fuel, and the resultant heat would cause the older spent fuel to react.
THE RELEASE OF RADI0 ACTIVITY A'iD MAGNITUDE OF RESULTANT HEALTH EFFECTS In this section we estimate the radioactive inventory of certain radionuclides in the Zicn pool, (the percentage release of e
1 radioactivity, and the resultant health effects) assuming the worst case scerario.
For selected radionuclides, the inventory of radioactivity, and the amounts released, are listed in Table 2:
Table 2.
Inventory of Radioactivity and Amount Released inventory Release Total Ci nuclide (Ci)
Fraction Released Sr-89 6.00 E07 5%
6.02 E06 Sr-90 6.04 E07 5%
Ru-106 7.77 E07 90%
6.99 E07 Cs-134 5.61 E07 90%
1.24 E08 Cs-137 8.17 E07 90%
Ce-144 1.30 E07 5%
6.90 E06 Pm-147 3.00 E00 5%
1.50 E06 Pu-238 9.02 E04 1%
1.01 E03 Pu-239 1.06 E04 1%
It is assumed that iodine, krypton and xenon will also be released quantitatively, but this discussion will concentrate on the cesium and plutonium releases only.
Detailed dispersion calculations have not been performed for this testimony.
The initial plume rise due to che heat released is uncertain, as is the duration of the release.
However, to under-stand the impact of such large releases of radioactivity, we consider an illustrative example for the radionuclide cesium.
Cesium has a t
-la-low vaporization temperature, on the order of 1200 C, and 90% release fraction is assumed.
Depending on the meteorological conditions, the material would carry from the site in a general cigar shaped pattern, depositing itself on the ground with a certain deposition velocity, from 0.01 to 0.003 m/sec.
The surface contamination patterns would depend on these meteorological conditions.
In our case, I assume uniform distribution of Cs over an area that would produce a whole body dose of 100 rads in 30 years.
Cs-137 has a half-life of 30 years.
The gamma dose rate due to surface contamination declines each year due to the movement of Cs-137 below the soil surface, providing increased shielding.
The calculational methods are presented in NASH-1400, Reactor Safety Study, Appendix VI, p.ll-20.
The dose rate integrated to 30 years approximately equals 0.64 rad per q Ci/m'.
Given a release of 1.24 E08 Ci of Cs (Table 2), we find a dose rate of 100 rads in 30 years if the Cs is spread over 3.1 E04 square miles, an area the size of Indiana, and easily encompassing the Chicago, Gary metropolitan area.
If we assume on the order of 10 million persons within the 31,000 sq.mi. area, and each person receiving 100 rad in 30 years, then I would expect approximately 600,000 long term cancer deaths 4
in this population.
This assumes 6 cancers per 10 person-rems whole body dose.
There would be additional genetic effects and
~
increased susceptibility to disease.
We would expect the actual distribution of Cs to be different from this illustrative example which assumes Cs uniformly spread over 31,000 sq.mi.
The actual distribution should be greater near the Zion facility, and less further away.
It is my expectation that if a slow wind, blowing
. towards Chicago and Pasquill Category F, obtained, the health effects would be greater than discussed here since more immediace health effects would occur.
The risk estimate for an immediate death is 4
about 1000 rads for an immediate death, not 6 per 10 rads.
The above health effects are entirely due to Cs.
Other radionuclides will also be released.
The Pu released can cause lung cancers.
On the order of mg amounts within the lung can cause 3
9 lung cancer. If 10 gm are released, the potential exiscs for 10 lung cancers.
Of course, only a small amount of the Pu released will reach human lungs.
Ruthenium, which would be vaporized in an accident, can also cause lung cancers.
According to Table 2, a large amount of Sr will also be released which will cause bone cancers.
These cal-culations have also not been performed here.
The net result would be a large number of cancers and health ef fects, and a large area of the country contaminated for many years.
SUGGESTED METHODS OF PREVENTING A LOSS OF COOLING ACCIDENT The above worst case accident can take place if cooling capacity of the spent fuel pool is lost.
Since the spent fuel assemblies are so densely packed as the racking is designed, this loss of cooling can result in ; serious accident.
Recommendations are made belcw which would prevent such an accident under any circumstances.
1)
The pumps for the spent fuel pool ecoling train "are operated manually from a local station"9 Cqder a major reactor accident in which the site must be vacated, the pool may overheat and cause a larcer accident than a reactor meltdown.
The cool could contain eleven reactor cores for this reason it is recommended that the make-up water supply and the heat exchangers be fully automated and independent of reactor operation.
The spent fuel pool must be cooled under any accident scenario.
2)
Because the spent fuel assemblies are so closely packed, there is not sufficient space for convective flow.
It is rectamended that opacing be provided between each storage rack (collection of storage spaces), and that the racks not be placed' immediate with the edge of the spent fuel pool, but that a region be provided around the entire collection of storage racks to allow for a downcomer region.
The suggestion is that a row or column, the space of one spent fuel assembly storage space, be provided between storage racks and around the entire collection of storage
' racks.
With this added safety margin, the pool would contain 1746 spent fuel assemblies and would allow the applicant to store fuel through 1989.
3)
The accident becomes very serious because of the oxi-dation of the ziracloy cladding.
As shown, while the spent fuel in the pool itself, under the worst case accident, gives rise to 13.3 MU in decay heat, the zircalloy-steam reaction gives rise 173 MN and a large cuantity of hydrogen.
The heat released would degrade the concrete and give rise to a hydrogen explosion which would release a large amount of radioactive material directly to the environment.
Clearly, this :iracioy-steam reaction can be avoided by using another material for cladding, such as stainless steel.
NOTES 1)
It is interesting to point out that the applicant believes that the spent fuel pool cooling system "can be rer.ioved from service intermittently" (Note (2), Table 9.3-2, Zion SAR).
2)
See Figure 17 in NUREG/CR-0649, " Spent Fuel Heatup Following Loss of Water During Storage.",
A.S.
Benjamin, et al.,
March, 1979.
3)
Ibid.
4)
Ibid.
5)
Ibid.
See figure 17 and discussion in 55.
6)
E.E. Lewis, Nuclear Power Reactor Safety, John Wiley & Sons (1977),
p.
306.
7)
Ibid.
8)
(2)
J.P.
Callahan, et al, " uni..xial Compressive Strengths of Concrete for Temperatures Reaching 1033 K",
Nuclear Engineering and Design 45, pp. 439 (1978).
9)
Zion SAR, p.
9.5-4 p.
1
ATTACHMENT A PROFESSIONAL QUALIFICATIONS OF MARVIN 5ESNIKOFF I graduated from the University of Michigan at Ann Arbor, Michigan, with a PhD in high energy theoretical physics, May, 1965.
My undergraduate schooling in math and physics was taken there as well. I have taken and taught courses in many areas of physics and math including nuclear physics, neutron diffusion, thermodynamics and fluid mechanics.
Following my graduation from the University of Michigan, I worked as a research associate for 2 years at the University of Maryland. From the years 1967 to 1973, I was employed as an Assistant Professor in the Department of Physics, State University of New York at Buffalo, and following that, for one year on a Fulbright Fellowship at the University of Chile, Santiago, Chile. During this period, from 1965 through 1973, I did research in elementary particle physics, publishing several papers and supervising one doctoral candidate.
I am principally employed as the co project director and.
staff scientist for an educational campaign on radioactive waste in New York State being run by the Sierra Club, an environmental and conservation organization. My business address is 3164 Main Street, Buffalo, New York 14214. I am also presently employed by Rachel Carson College of the State University ~of~New York at Buffalo, as
~ ~ _ -
a lecturer and principal researcher on a grant from the U.sT'Environt.entil Protection Agency to study the economics of plutenium recycle. Finally, t
. I am also serving as a consultan't totheStateor}LcwerSaxony, West Germany, on a panel of international experts called the Gorleben International Review, to study the plans of the West German nuclear industry to locate a fuel reprocessing and fabrication plant and waste disposal center near the town of Gorleben, West Germany.
I have been working on matters concerning reprocessing and spent fuel pools for the past five years. I was technical coordinator for the Sierra Club in their intervention in the NRC proceedings concerning Nuclear Fuel Services reprocessing plant located in West Valley, New York, 35 miles south of Buffalo. This work involved coordination with scientists examining separate actions of the NFS' Safety Analysis Report, and an examination by myself of the ventilation, radiation shielding and chemical engineering of that facility.
I was science consultant to the New England Coalition on Nuclear Pollution in the licensing amendment concerning the Vermont Yankee scent fuel pool. I served as a science consultant to Environment-alists, Inc. in the NRC proceedings concerning the proposed Barnwell reprocessing plant. I reviewed the safety report for the spent fuel storage pool and prepared interrogatories for Natural Resor"ces Defense Council concerning that facility. On behalf of the Attorne; General of the State of Illinois, I made a site visit to the GE Morris plant to examine the spent fuel pool. In addition I was consultant to the
. Attorney General of New York concerning the air shipments of plutonium into Kennedy Airport, calculating the effects of an accident including the release of plutonium, the spread of the plutonium cloud, and the possible lung cancers. I gave an invited paper at the 1976 Toronto meeting of the American Nuclear Society.
I have testified approximately 20 times before Congress, New York and California Legislatures, New York and California Energy Commissions on matters concerning reprocessing. I've testified before the NRC concerning Table S-3, or the Uranium Fuel Cycle proceedings.
I have written extensive testimony on Nuclear Fuel Services reprocessing plant for the GESMO or plutonium recycle proceedings before the,NRC.
Specific papers, other than testimony that I have written concern plutonium recycle, decommissioning and uranium recycle and air transport of plutonium. I wrote the first paper, to my knowledge, that shows that the economics of plutonium recycle are marginal. The paper on the decommissioning of nuclear power reactors showed that time period for a reactor vessel and internals to decay to safe levels is hu.' reds of thousands of years rather than the 180 years that the nuclear industry had previously maintained. I am presently working en a paper on technetium-99 and uranium recycle.
h
I have prepared the foregoing affidavit and swear that it is true and correct to the best of =7 knowledge.
L
- /
u.'s p) l'6G7Di RESNFCF 7,-
O e
e