ML19339C914

Testimony Explaining Significant Events Leading to & Quantifying Amount of Hydrogen Produced During TMI-2 Accident.Accident Analysis & Prof Qualifications Encl

ML19339C914
Person / Time
Site:	Mcguire, McGuire
Issue date:	02/09/1981
From:	Miller A DUKE POWER CO.
To:
Shared Package
ML19339C912	List: ML19339C911 ML19339C914 ML19339C915 ML19339C916 ML19339C919 ML19339C920 ML19339C921 ML19339C922 ML19339C923 ML19339C925 ML19339C926
References
NUDOCS 8102120313
Download: ML19339C914 (22)

Text

9 UNITED STATES OF AMERICA NUCLEAR REGULATORY COMMISSION BEFORE THE ATOMIC SAFETY AND LICENSING BOARD In the Matter of )

)

DUKE POWER COMPANY ) Docket Nos. 50-369

) 50-370 (William B. McGuire Nuclear )

Station, Units 1 and 2) )

TESTIMONY OF A.D. MILLER REGARDING HYDROGEN PRODUCTION AT TMI

1. Q. What is the scope of this testimony?

A. This testimony will set forth the significant events leading to and quantify the amount of hydrogen production during the TMI accident.

2. Q. Explain the mechanism which resulted in the excessive production of hydrogen during the TMI accident?

A. The excessive hydrogen produced during the TMI-2 accident was a direct result of reactions between the zirconium in the fuel cladding and the coolant (water / steam). At high temperatures, zirconium reacts with water to form hydrogen and zirconium dioxide.

Zr + 2H 2O --9aZr0 + 2H 2

+ heat The rate of reaction is temperature dependent.

3. O. Describe the major events that resulted in the production of excessive quantities of hydrogen during the TMI accident?

After five minutes into the accident the events that l

would eventually result in productin of excessive l hydrogen had been set in motion,. i.e., a loss of coolant accident from the stuck open two inch power operated relief valve on the pressurizer coupled with

! operator termination of adequate ECCS operation.

Significantly, two to three hours elapsed before l

l B.L02120 $ b L

4 the core became uncovered and reached sufficiently high temperatures for significant hydrogen to be produced. The results of the hydrogen generation were felt approximately ten hours after the accident began in the form of a 28 psig pressure spike in containment which resulted from a partial burning of the generated hydrogen.

1 It is generally agreed that during the entire TMI accident, approximately 50% of all the circonium in the reactor was oxidized resulting in hydrogen production. Based on hydrogen accounting at the time of the pressure spike, two-thirds of the hydrogen produced was in the containment building atmosphere and one-third remained in the reactor coolant system.

4. Q. Have you prepared an analysis which supports the testimony?

A. Yes. I assisted in the preparation of NSAC-1,

" Analysis of Three Mile Island - Unit-2 Accident" (Revised March 1990) which is an analysis of the TMI accident. Specifically, my testimony is suppcrted by my analysis contained in the Appendix to NSAC-1 entitled " Hydrogen Phenomena" which is attached hereto and which I adopt as part of my testimon.

Y r

F l

\

\  :

l 9

- , , .y - - . _ .

,- *r-- *r' , - -e w e --+

l t

PROFESSIONAL CUALIFICATICNS OF ALAN D. MILLER I 3265 Cuesta Drive San Jose, California 95148 EDUCATICNAL BACKGROUND PhD Chemical Engineering, minor in Chemistry, University of Wisconsin, 1973.

B.S. Chemical Engineering, graduated "with distinction," Iowa State University, 1968.

EXPERIENCE June 1979 to Present - Project Manager - Nuclear Safety Analysis Center Responsible for organizing the Chemistry, Radiochemistry, Rad-waste Program Area for activities related to the Three Mile Island accident and coordinating these with EPRI, the Atomic '

Industrial Forum, and the Edison Electric Institute. This in-cludes both original work performed in-house and managing con-tractors' and consultants' work. Specific work includes hydro-gen flammability and detonability, radiation source terms, radio-iodine transport, reactor containment structural capabilities, and population radiation exposure assessment.

March 1978 to June 1979 - Project Manager - Electric Power Research Institute Managed projects in the radiation control and radioactive waste management area of the Engineering and Operations Department at the Nuclear Division. Wrote and evaluated requests for proposals, procurred funds, initiated contracts and managed the projects for R&D on icdine spiking, advanced radwaste treatment systems, de-

! contamination review, and Boiling Water Reactor offgas systems.

! December 1974 to March 1978 - Senior Engineer, General Electric

, Water chemistrv Unit t

I Supervised programs in the areas of high purity water chemistry and radioactive waste management. Considerable field experience in continous measurement of dissolved oxvgen, specific conductance, and pH in high purity water systems, automated sample systems, and ion exchange resin performance and degradation in water treatment l

i systems. Developed, designed, and implemented the software and I hardware for a computerized system for automated data acquisition /

l reduction and the interfacing of this system with the time shar-l ing system for general access to the data base. Successful com-l pletion of the General Electric Boiling Water Reactor Chemistry training course for the training of plant site chemists.

l i

5

_2_

April 1973 to December 1974 - Engineer, General Electric Chemical Engineering Unit Designed, constructed and operated a fluidized bed plus fixed bed catalytic nitrate destruction system for radioactive waste management. Developed analytical techniques for analyses of oxides of nitrogen. Prepared economic feasibility study and completed bench scale glove box investigation of a proprietary high temperature molten electrolyte electrowinning process.

BIBLIOGRAPHY

. Miller, A.D., "Short-Term Lessons Learned: Radiation Protection Recommendations - The NSAC Perspective,"

ANS, Washington, November 1980.

Miller, A.D., "Three Mile Island - Accident and Aftermath,"

ASCE, Iowa City, IA, September 1980.

Miller, A.D., " Radiation Source Terms and Shielding at TMI-2," ANS, Las Vegas, June 1980.

Miller, A.D., " Radiation Source Terms," Symposium on Im-proved Post-Accident Sampling Capability for Nuclear Power Stations, Oconomowoc, WI, February 1980.

Miller, A.D., "Three Mile Island - What Happened," American Society of Civil Engineers, Ames, IA, October 1979.

Remark, J.F. and A.D. Miller, " Review of Plant Decontamin-ation Methods," ANS, Sun Valley, September, 1979.

Shaw, R.A., M.D. Naughton, and A.D. Miller, " Radiation Exposure, Radiation Control and Decontamination," ANS, Sun Valley, September 1979.

Shaw, R.A., A.D. Miller, and M.D. Naugh ton , " Exposure and Radiation: U.S. Experience," IAEA/OECD-NEA, Interna-tional Symposium on Occupational Radiation Exposure in Nuclear Fuel Cycle Facilities, Los Angeles, June 1979.

Miller, A.D., " Water Chemistry Characterization of a Boiling Water Reactor, " Nucl . Tech. , 37, 111 (1978). -

Miller, A.D., "Electrogenerative Chloro- and Bromo-fluorina-tion of Olefins frcm Aqueous Media; the Electrogenera-tive Cell as a Chemical Reactor," J. Appl. Chem.

Biotechnol., 27, 176 (1977).

Indig, H.E. Weber, J.E., and Miller, A.D., " Monitoring Corrosion and Oxidation Potentials in a Boiling Water Reactor," Corrosion /77, San Francisco, CA, March 1977.

i Miller, A.D., Burley, E.L., Snyder, D.T., and Selby, K.A.,

" Water Chemistry Characterizations of a Boiling Water Reactor," ANS, New York, June 1977.

Miller, A.D. and Langer, S.L., "Electrogenerative Bromina-tion," J. Electrochem. Sec., 120 (12), 1965 (1973).

Miller, A.D., "Electrogenerative Halogenation," PhD Thesis, University of Wisconsin, 1973.

l l

1

)

i e

NSAC-1 HYDDOGEN PHENnMENA

1. INTRODUCTION 11 Purpose and Scope The issue causing the most concern and public apprehension during the incident at TMI involved hydrogen and the hydrogen bubble. The erroneous assumption that the accumu-lation of hydrogen within the primary system was or could become explosive led to speculations of a massive spread of contamination and consequent damages to the general population. As wa s later confirmed publicly, these speculations and fears about the " bubble" were totally unfounded. The presence of even small amounts of free hydrogen prevents accumulation of oxygen and thus any However, the possibility of hydrogen / oxygen explosion.  !!

amount of hydrogen produced was sufficient to cause legitimate concerns about core cooling and flammability in the reactor building atmosphere. An ignition, as measured by pressure and temperature spikes, did occur about 10 hours1.157407e-4 days <br />0.00278 hours <br />1.653439e-5 weeks <br />3.805e-6 months <br /> into the incident. Although equipment may have been damaged, the integrity of the reactor building was by no means challenged. The purpoEe of this appendix is to quantify the extent of the hydrogen generation and discuss possible scenarios. .

~

2. SOURCES OF HYDROGEN IN LIGHT WATER REACTORS in light Hydrogen may be produced by a number of mechanisms Among these are the radiolytic water reactor systems.

elevated decomposition of water, metal / water reaction at temperature, oxidation of materials of construction, and Significant amounts of decomposition of organic materials.

hydrogen may be produced only by radiolysis and the zirconium /wate'r reaction.

1

l l

2.1 Radiolysis of Water .

Absorption of energy from ionizing radiation will cause the  !

decomposition of water by a somewhat complicated mechanism I to form primarily hydrogen and oxygen.

2H2O -

2H2+O2 (Hydrogen peroxide is formed as an intermediate which is very rapidly decomposed to water and oxygen above about 200*

F.) The yield of this reaction is dependent upon the energy absorbed, the nature of the radiation, temperature, reaction products residence time, etc. i For example, boiling increases the radiolysis rate by 30 times over non-boiling. Gamma radiation from cobalt-60 in boiling water generater about 0.45 molecu).es of hydrogen per

' 100 ev of energy absorbed (l). Neutrons, however, in boiling water generate about 1.1 molecules per 100 ev absorbed. The presence of 0.04 ppm of dissolved hydrogen totally sup-presses radiolysis at 77'F in non-boiling systems while 0.7 ,

ppm stops radiolysis in boiling water at 525'F(2). Standard f practice in pressurized water reactors is to add about 1 ppm of hydrogen to the water in the reactor coolant system (RCS) to prevent radiolysis.

Throughout the incident at TMI-2, the dissolved hydrogen levels in the RCS were considerably above 1 ppm. Thus, radiolysis in the RCS was a source of neit'.ier hydrogen nor oxygen. (Radiolysis may have produced inconsequential .,g,.

l quantities of hydrogen in the reactor building sump where

~

dissolved hydrogen levels may have been considerably lower.)

. i 2

APPENDIX HYD

2.2 Zirconium / Water Reaction Above 1600*F zirconium alloys react with water to form hydrogen and zirconium oxide.

Zr + 2H 2O + ZrO2 + 2H2 The reaction rate increases with temperature and is very rapid above 2700*F. Stoichiometrically, about 8 standard cubic feet of hydrogen are produced per pound of zirconium oxidized.

3. EYDROGEN COMBUSTION 3.1 Limits of Combustion Near room temperature in air mixtures containing between 1 ppm and 10% moisture, the lower limit of upward flamma-bility is 4% hydrogen. The lower limit for downward flame propagation is 8% and the lower limit of detonability is 19%(3). Combustion of hydrogen mixtures between 4% and 9%

is often incomplete, leaving about 4% unburned (4).

When the temperature is elevated and water content is high, Existing the oxygen content is often the limiting factor.

data indicate that between 300* F and 500* F with at least 50%

steam the lower limit of flammability is about 41 oxygen with detonability repressed below 9% oxygen (5), .

3.2 Pressure Increase from Hydrogen Combustion l

and 9%, combustion in large For H2 concentrations between 4%

containers results in pressure increases that depend'upon: I

1. the temperature of the pre-combustion mixture. l i

2. the ignition mechanism - multiplicity and energy.

3. the'H2 concentration.

3 3

APPENDIX HYD

4. the vessel wall material. ,

5 the vessel size and shape.

Theoretical calculations, based on the energy of combustion and assuming that all hydrogen is burned, provide an upper bound to the pressure increase. Between 8% and 10%

hydrogen, the transition region for downward flame propagation (4 & 7), the reaction may go to completion. This low concentration combustion exhibits slow flame propagation (1 to 10 m/sec.)(8). Large containers and ignition multi-plicity increase the flame velocity. This type of combus-tion yields a maximum pressure rise factor of about R.

When the concentration of hydrogen reaches about 19% in air with normal humidity, a change in combustion phenomena occurs with the possible transition to supersonic combustion (detonation)*. This transition requires either a detonative ignitor or the shock wave reflection from at least two hard surfaces (9). For example, detonation nearly always occurs for ignition of these compositions in pipes while in free balloons it almost never occurs.

Figure 1 illustrates the relationship between hydrogen concentration and pressure rise. Theoretical and experi-mental data are shown.

The pressure rise because of a detonation (hydrogen concen-

, trations between 19% to 70%) is about 17 times.the initial pressure (10). This does not include consideration of I

reflection amplification or the pre-detonation compression observed in piping systems.

• With bone dry air (< l ppm water), detonations have been

, observed as low as about 14% hydrogen.

APPENDIX HYD 4

4. EYDROGEN PHENOMENA AT TMI-2 Figures 2 and 3 show a pressure pulse of approximately (Figure 3 28 psi that occurred at 1350 on March 28, 1979.

shows this pulse as an apparent decrease in steam pressures.) The reactor building temperature measurements 40*F (Figure 4) indicated a nearly uniform increase of about found at any at the same time. No similar indications were other time during the event.

rise of 28 psi will result According to rigure 1, a pressure from ignition of about 8% hydrogen. This is also consistent with the rate of pressure increase (7).

Hydrogen generation in the containment by radiolysis may be estimated from Regulatory Guide 1.7. A conservative analysis can only account for less than about 0.1% contri- first bution to the hydrogen concentration during the the metal-water reaction is the only day of the incident. Thus, significant hydrogen source. No measurements of the reactor building atmosphere cc? position were made before the pressure transient.

The earliest reactor building atmosphere measurements were These were:

made on March 31 at 0600(6).

Sample 1 H2- 1.7%

I O2 - 15.7%

I N2 - 82.6%

1*7%

Sample 2 . H2-O2 - 16.5%

N2 - 81.8%

A series of samples taken April 1 and 2 averaged 2.1% H 2 (average of sixteen' measurements) and 18.6% O2 (eight measurements).' These are documented in the secondary 5

APPENDIX HYD

i j

chemistry log book. This disconnect in oxygen measurements po'ses a dilemma since the oxygen depletion in the reactor l

building is a direct measure of the magnitude of the hydrogen generation and subsequent burn. Later reactor building measurements taken on June 1(13) (0.6% H2, 14.5%

O)2 and August 2 (0.6 % H2 , 14.1% O2) tend to confirm the March 31 results.

The only additional data needed te closa material balances are the hydrogen inventory in the reactor coolant system, both dissolved and in the bubble. Using the known temper-ature and pressure of the system, the dissolved hydrogen can be calculated from Henry's law constants shown in Figure 5. The volume of the hydrogen bubble used was that reported by Metropolitian Edison (14). Figure 6 shows the bubble behavior with time both with and without accounting for effects of hydrogen solubility.

Thus, there are five time frames for material balances to be performed:

1. Just prior to the hydrogen ignition, March 2A, 1979

2. March 31, 1979 (6)

3. April 1-2, 1979 (Log book data)

4. June 1, 1979 (13)

5. August 2, 1979 (15)

Note that the boundary for the material balances is the l reactor building. Thus any hydrogen which escaped the reactor building will also escape the material balance. It is estimated that between 2.5. and 10 million curies of noble gas (mostly Xe-133') were released during the first few days of the accident. A corresponding amount of hydrogen can be assumed to have also escaped since the hydrogen and noble gases were released from the core at about the same time.

Thus, since a maximum 10% of the noble gas released from the core was released to the environs, as much as 10% of tdue

.- t APPENDIX HYD 6 I m-____

hydrogep released by metal water reactor in the core was also released to the environs. This produces a systematic Similarly, error in the material balances of up to + 10%.

I depletion of the oxygen in the air of the reactor building by other than hydrogen oxidation (e.g. corrosion) will l produce errors in the negative director (i.e., less apparent zirconium / water reaction.

Basis for calculations:

Dry air contains 20.9% Oxygen by volume.

The reactor primary system volume is 1.18 X 104 cubic feet. (including the pressurizer)

The total zirconium inventory in the primary system is 53,000 lb.(ll)

The reactor building free volume is 2.05 X 106 cubic feet

• at 120*F, 14.7 psia, and 100% relative humidity, this corresponds to 1.5 X 106 standard cubic feet or 4200 lb moles of dry gas.

A summary of the five material balances is shown in Table 1 4.1 Material Balance Just Priar to Hydrogen Ignition From post-ignition analyses of, the containment atmosphere it is calculated that before the ignition the hydrogen concen-tration was 8%. This corresponds to about 340 pound moles of hydrogen or, stoichiometrically, about 16,000 pounds of zirconium reacting in the metal / water reaction (29% of the total zirconium inventory). The primary coolant with 825 psi of hydrogen overpressure (875 psia total pressure) of and at 280*F is calculated to contain about 34 lb moles dissolved hydrogen (3% metal-water reaction). Assuming a 1400 ft3 bubble (14) in the primary system at these con-ditions, (150 pound moles of hydrogen) represents another a

• Burns and Roe caledladed value 7 APPENDIX HYD

13 % of the total zirconium-water inventory. Thus, by mate-rial balance a total of 45% zirconium-water reaction occurred.

4.2 Material Balance Using March 31 Data As earlier noted, two samples of the reactor building atmo-sphere were taken at 0600 on March 31, 1979. The hydrogen concentration was 1.7% and the average oxygen concentration was 16.1 1 0.4%. The oxygen depletion was then 4.8 1 0.4%

representing 410 1 30 lb moles hydrogen or 35 i 3% metal-water reaction. As in Section 4.1 (using 823 ft3 for the bubble volume), the hydrogen inventories are 36 lb moles in the coolant and 86 lb moles in the bubble. These values correspond to 3% and 7% metal-water reaction, respec-tively. The residual hydrogen in the containment atmosphere is 72 lb moles representing 6% metal-water reaction. 'Thus the total indicated metal-water reaction based on these data is 51%.

4.3 Material Balance Using April 1-2 Data Measurements made April 1 and 2 indicate a hydrogen concen-tration of 2.1% and oxygen concentration of 18.6%. As

~

calculated above, the oxygen depletion corresponds to 200 lb moles hydrogen or 17% metal-water reaction. The 2.1%

residual hydrogen represents 84 lb moles or 8% metal-water reaction. The volume of the bubble at this time was very low (probably zero) while ths dissolved hydrogen in the coolant may have represented up to 3% metal-water reac-

..- tion. This totals 28% metal-water reaction.

Material Balance for June 1 Data 4.4 By June 1, the hydrogen concentration had been reduced to 0.6% by operat,on i of the hydrogen recombiners. The oxygen APPENDIX HYD 8

By oxygen depletion, about 540 lb concentration was 14.5%. Residual moles of hydrogen were formed (47% metal water).

hydrogen is another 26 lb. moles (2% metal water).

is minimal (< 1 lb Dissolved hydrogen in the primary coolant is mole). A total amount of metal-water reaction of 49%

indicated.

l 4.5 Material Balance for Au=ust 2 in the On August 2, the hydrogen and oxygen concentrations These respectively.

reactor building were 0.6 and 14.1%,

correspond to 2% and 50% zirconium-water reaction.

5. CONTAINMENT CAPABILITY The hydrogen ecmbustion event produced a peak pressure thus,of design pressure is 60 psig; 28 psig. The containment structural integrity due there was no threat to containment to the combustion of the hydrogen resulting from the oxida-tion of an estimated 29% of the zirconium.

The possibility of the hydrogen concentration increasing substantially above the lower limit of flammability without abundance of ignition ignition is reduced by the great located throughout the reactor building. These sources for valves and sources include approximately 100 motors various rotating equipment and contacts associated with .

position indication and plant parameters.

even if 100% of the total Recent work has shown that zirconium in the core had reacted'the reactor containment Ultimate failure of the would have remained intact. 150 to 190 psia via tendon containment would occur at failure. .-

APPENDIX EYD 9

m 1

TABLE 1 4

MATERIAL BA. LANCE

SUMMARY

Pound Moles Hydrogen Equivalent REACTOR BUILDING RCS NUMBER TIME 02 BUBBLE DISSOLVED TOTAL H7 '

1 Preburn 340 0 150 34 520 2 March 31 72 410 86 86 600 3 April 1-2 89 200 < 10 36 330 4 June 1 26 540 0 < 1 570 5 August 2 26 580 0 < 1 610

% Zirconium-Water Reaction REACTOR BUILDING RCS NUMBER TIME H2 O2 BUBBLE DISSOLVED TOTAL 1

1 Preburn 29% 0% 13% 3% 45%

2 March 31 6% 35% 7% 3% 51%

3 April 1-2 8% 17% < 1% 3% 28%

4 June 1 2% 47% 0% <0.1% 49%

5 August 2 2% 50% 0% <0.1% 52%

O e

o e

, /

APPENDIX HYD 10

m REFERENCES

" Boiling Water Reactor Technology - Status of the Art 1

Report", C. R. Breden, ANL-6572, 1963.

J. A. Ghormley and T.

W.

2 A. O. Allen, C. J. Hochanadel, Davis, " Decomposition of Water and Aqueous Solutions Under Mixed Fast Neutron and Gamma Radiation", J. Phys. Chem., 5 6,,

575 (1952),

I. L.

3 " Survey of Hydrogen Combustion Properties" by Drell, and Belles, F. E. NACA Rept. 61383 A. L: et 4 "Some Observations on Near Limit Flames" by Ferno, al, Proc. Int. Sym. on Combustion, #13 1970 5

SAR Advanced Design Memorandum 56, Preliminary Explosion 7, 1967 R. L. Mathews, D. A. McCune, March Test Results.

6 Letter Lavallee to Zebroski, Nuclear Safety Analysis Center, June 5, 1979 7 NEDO-LOB 12, Hydogen Flammability and Burning Characteristics Peterson, April G.

in BWR Containments, B. C. Slifer, T.

1973 8 "Some Effects of Inert Diluents on Flame Speeds and R., Fourth Intl. Symp.

! Temperature" Morgan, G. H.; Kane, W.

on Combustion, 1953.

Contract AF 18 (600)-1687, Arthur D. Little Inc.

9 l NEDE 11146, 10 Pressure Design Bases for,off Gas Systems,

' General Electric Co.

L., April 17, 1979.

11 Letter Duncan, R. H. to Zebroski, E.

Spray Systems 12 " Design Considerations of Reactor Containment Part 1", T. H.

Row, et al, ORNL-TM-2412, April,, 1969.

Letter Lavallee to Zebroski, June 4, 1979 13 March 31, 1979-14 Preliminary Annotated Sequence of Events, through April 30, 1979, Metropolitian Edison Company, July 20, 1979 D. Miller, August 15 Personal Communication, K. Frederick to A.

27, 1979

?

[ *

/.

i APPENDIX HYD 11

i.

70 39 Calculated Ap\

60 -

17 50 - -

15 C

3 Ic- X

$ 40 e

Calculated T, 13 (

t E.

2 $

8 5 3 30 - - 11 g E .

E

/ e 9

- Experimental Ap

/ Upward l '

Upward and downward 10 - propagation propagation -

7 only

! I I 5 O

0 4 8 12 16 20 Hydrogen (%)

Figure HYD 1. Pressure Rises Resulting From ignition of Hydrogen-Air in 12 Foot Diameter Sphere at'18'C

/

Appendix HYD

l 32 28 -

Hydrogen bum y 24 -

e 5

20 -

5 E

c 16 -

5

= 12 -

di 5 e _

8 2

4 -

0 I I I I l l

_4 10 12 14 2 4 6 8

O Time AfterTurbineTrip(hours)

Figure HYO 2. Reacter Building Pressure Versus Time' O

.- Y Accendix HYD

i l

270 -5 o- o OTSG "A" 265 -

• OTSG "B" -

o 250 -

c

~

C C

i o 255 -

e 10 O

^

l

~C o _O m C

• S o O

'm 250 -  ! o 15 'g b I o / b o 4 s I $

o I o o 245 - o - 20 l

o o 240 - -

25 l

235 - - 30 l

l I I I I i 13:50:20 13:50:30 13:50:40 13:50:50 13:51:00 13:51:10 13:51:20 I

Time Figure HYD 3. Effect of H2Ignition on Measured Steam Pressurts l

( i l Appendix HYD

l l

190 5 Supply air 180 -

6 Supply air 7 Ambient steam and feedwater 170 - penetration e sump area 13 160 -

9 Ambient let down cooler area 150 - 12 13 14 14.11 7 130 --

E 7' ' 8 120 -

9% A 9 -

8 5 N 110 -

11 100 - 6 10 Ambient drain tank area 3 11 Ambient elevation 3531 12 Ambient elevation 353 2 13 Ambient elevation 330-1 80 -

14 Ambient elevation 330-2 I  !

70 1400 1500 I 1300 Time .

Figure HYD 4. Reactor Building Temperatures During Hydrogen Ignition

-o-e Y

Accendix HYD

i

(_ l l

i

)

l i

l X 104 g 14

=

2

_c C

g 12 -

t

.c 8

$ 10 -

2 o

E N

E e 8 -

e E

i

.c k

'o e 6 -

B E

- e E

3a 4 -

J . ,Tf: , .' 3 ' ,

A sJ4ydrogen h D &n ;j t, vit c C Er h O NNegen 3ylQ;x 2 -

3

• 0 e

- I i O

100 200 300 400 Temperature CF)

Figure HYD 5. Henry's Law Constants Versus Temperature e'

Appendix HYD

r 1000 Gas bubble volume estimate y {.. O without solubility correction

&g _ a with solubility ccrrection g

cc

~

e m "

E 600 -

t e

~

O e 400 -

3 O(V 1 ..

g ..

= a Sm - m 1 ..

-- O 1

\ .s "

0 -

1 -

l i i i l i e e I , I i i i

! i i e

- 200 ,

" April 3 Apdf4 April 1 April 2 March 31 -

Time Figure HYD 6. Calculated Volume of Hydrogen " Bubble"in Pressure Vessel Versus Time O

e* $

Appendix HYD

, _