ML20070C892

Aerosol Release from Sodium-Concrete Reactions. Prototype Hydrogen Filter Preliminary Test Results Encl

ML20070C892
Person / Time
Site:	Clinch River
Issue date:	10/31/1982
From:	Colburn R, Muhlestein L HANFORD ENGINEERING DEVELOPMENT LABORATORY
To:
Shared Package
ML20070C876	List: ML20070C873 ML20070C892 ML20070C917
References
NUDOCS 8212140380
Download: ML20070C892 (23)

Text

-.

s

,.. {

4 e

4 l

1

.e*

'9 L

1 i

'J J

B k

1 i

1 l

I l

AEROS0i. RELEASE FROM S0DILM-CONCRETE REACTIONS 1

)

j

.L.

D. MUHLESTEIN I

R. P. COLBURN 4

October.1982 i

i i

+

4 I

8 i

e i

1 f

1 i

e i

i I

j l

J i

8212140380 821207 PDR ADOCK 05000537 f., :. rf-A PDR

,2.-

s.

.,..~j.::

.m... ~ -....-

s..

~

s m 76 of Q.. n

...-- A'*

. &y. v El_.\\o.

...S:h

.c e 1

(

j'

~

f AEROSOL RELEASE FROM SODIUM-C0i4 CRETE REACTIONS L. D. Muhlestein L'

R. P. Colburn ABSTRACT i

The fomation and types of aerosols uhich my be generated as a result of sodium-concrete reactions are reviewed, and substanti-ated by experience gained through several years of experimental programs.

It is concluded that the only m terials present in any appreciable quantities in the cover gas space above the sod-ium pool are nitrogen and hydrogen gas, and sodium vapor.

4

f I

{

CONTENTS Abstract Figures D

1.0 SUPEARY AND CONCLUSIONS

2.0 INTRODUCTION

3.0 CHEMISTRY AND AEROSOL FORMATION 4.0 EXPERIMENTAL EVIDENCE GAINED FROM SODIUM-CONCRETE REACTIONS

5.0 REFERENCES

m G

e e

l l

i I

1 l

l l

=

. _ f...

,..

.-}&Q-

[

~;.(.

+

AEROSOL RELEASE FROM SODIUM-CONCRETE REACTIONS L. D. Muhlestein R. P. Colburn 1.0 SutmARY AND CONCLUSIONS The present study has reviewed general unpublished test results from numer-ous sodium-concrete reactions in an attempt to identify the materials which may exist above the sodium pool in the inert gas cover space. The review was aimed at addressing a concern that sodium-concrete reaction products could be carried through the sodium pool, and exist in the cover gas space as an aerosol.

Hydrogen gas, created from the sodium-water reaction, bubbles through the sodium pool to the cover gas space. Sodium vapor also exists above the sod-ium pool, and the amount depends on the sodium pool temperature. Other particulate matter resulting from sodium-concrete reactions generally remain in the dense reaction product layer. The reaction products of' concern have '

negligible vapor pressure at temperatures below the boiling point of sodium and thus are non-volatile. Thus, the only viable mechanism of transporting

~

particulates through the sodium pool and creating a suspended' aerosol in the cover gas space is by inefficient mechanical entrainment.

Quantitative infonnation regarding the chemical nature of materials which exist above the sodium pool is difficult to obtain because of their very reactive properties.

However, review of many pieces of information obtained, but not usually reported, from numerous sodium-concrete reaction tests led to the conclusion that the only materials present in any appreciable quanti-ties in the cover gas space are the inert gas, hydrogen, and sodium vapor.,

Pyrophoric " sodium fines" have been observed in ducts, and on test chamber floors which are thought to be a combination of condensed sodium vapor drop-lets and sodium hydride particles.

Exposure of these " sodium fines" to air quickly produces sodium oxide, sodium hydroxide, and, sodium carbonate, thus

~

making it difficult to provide a quantitative description of these materials.

1

.,$li.l.'

E-

.,;, -).

Y'

.e

.=., - -.

2.0 INTRODUCTION

In breeder reactor hypothetical core dfserfptive accident analysis, one q.

usually assumes that core debris and sodium have penetrated both the reac-tor vessel and guard vessel, and have come in contact with concrete in the lower cavity area. Sodium-concrete reactions can then occur and are sources for both energy and hydrogen release to containment. The heat from core debris, coupled with energy released from sodium-concrete reactions, will heat the sodium to near its boiling temperature. Thus, sodium vapor and hydrogen will leave the sodium pool.

There has been concern expressed that other types of sodium-concrete reaction products could be carried through the sodium pool, and exist in the cover gas space above the sodium pool as an aerosol. The objective of the present study was to review'the types of reaction products which may form from sodium-concrete reactions, and examine which, if any, of these solid products may be carried through the sodium pool and, hence, exist in the cover gas space above the pool as aerosols.

Experience and data gained through several years of sodium-concrete reaction tests are examined to substantiate the chemical analysis.

l 3.0 CHEMISTRY AND AEROSOL FORMATION l

A pertinent summary of the basics involved with sodium-concrete reactions is presented in Reference 1.

Figure 1 shows a simplified schematic representing sodium oncrete reactions taken from Reference 1.

Basically, a deep sodium pool, with core debris, lies on top of concrete with the sodium pool and unreacted concrete separated by a reaction product layer. Above the sodium c.:fle poolisaninertdas= atmosphere. The possible chemical reactions which

~

can occur from sodium-limestone concrete reactions are listed in Table 1 (Table 2 of Reference 1).

From a review of the chemistry section of Refer-ence 1, the most probable reaction products are:

i

• Hydrogen, H2

~'

• Sodium Hydroxide, NaOH

' Sodium 0xide, Na 0 2

\\

,.-.~.[..

L

~.

J.

- %. i.

J TAB:.E 1 SODIUM-LIMESTONE CONCRETE CHEMICAL REACTIONS aH joo.g i

Possible Reactions (Kca))

1.

Na + H O + NaOH + 1/2 H

- 38 2

2

+ 5 2.

2NaOH + CACO 3 + Na 003 + Ca0 + H O 2

2

- II 3.

2Na + H O + CACO 3 + Na CO3 + Ca0 + H2 2

2 4.

2NaOH + MgC03 + Na2 03 + Mg0 + H O

- 12 2

^~ 00 5.

2Na + H O + MgCO + Na 3 + Mg0 + H2 2

3 2

6.

CACO 3 + Ca0 + CO2 7.

MgC0 + Mg0 + CO

+ 25 3

2

+

+ 75 8.

Na CO3 + Na2 2

2

- 37 9.

2NaOH + CO2 * "*2003+HO 2

-122 10.

4Na + 3 CACO 3

• 2"*2003 + 3Ca0 + C l'1.

4Na + CACO + 2Na 0 + Ca0 + C

- 60 3

2

-173 12.

4Na + 3MgC03 + 2Na2 3 + 3Mg0 + C

- 77 3 + 2N6 0 + Mg0 + C 13.

4Na + MgC0 2

14, 4Na + CO2 + 2Na20+C

-103 15.

Na 0 + CACO 3 + Na CO3 + Ca0

- 31 2

2 10

~ * ' '

16.

NaCO3 + SiO2 + Na2 3 + CO2 l

-122 17.

2NaOH + SiO2 + Na SiO3+HO 2

2

- 10 18.

2NaOH + A102 3 + 2NaA102+HO 2

- 30 19.

4Na + Na CO3 + 3Na 0 + C 2

2

'G e

4 I

l l

4 l

l

~~r n.,

a..

~-

.n

A-v

._m___%

SODIUM A

HYDROGEN GAS I

i

~Qw:xn: m_m=1 ODIUM REACTION I

\\

(TRANSPORT PRODUCTS

\\

\\

s g

l A

"%sangixi.

s g%

r+

i Y'

1. $UNREAQTED

')

?. CONCRETE r.

h h

b.

. )$.;

. b *

\\

1 n:.% %L *A*So h"S*O:*.

Aib*f?

VW w.

a HEDL E294-85.2 t.

FIGURE 1.

Sodium-Concrete Reaction Model Schematic.

Neg 8202343-1 i.:.

.., _v...

?', 'h

~

e

e O Sodium Carbonate, Na CO 2 3 O Calcium Oxide, Ca0 0 Magnesium Oxide Mg0

' Carbon Dioxide, CO

~

2 Carbon dioxide readily reacts with sodium to form sodium oxide and elemental carbon. Thus, the only gases which leave the sodium pool are hydrogen *and sodium.

Transport of other reaction products into the cover gas space above the sodium pool would be negligible compared to the mass of metallic sodium which is vaporized. The refractory reaction products Ca0, Mg0 and C have vapor pre-0 ssures less than 1 mmHg below 2000 C.

Na CO decomposes rather than volatil-2 3 0

izing, and it has a dissociation pressure of only 19 mHg. at 1000 C.

Na 0 2

0 0

is reported to sublime above 1275 C.

NaOH has a boiling point of 1390 C.

Thus, they are non-volatile at temperatures of concern. The only significant mode of transport would be the mechanism of mechanical entrainment of these species in sodium droplets ejected from the sodium pool as gas bubbles burst on the sodium pool surface or in hydrogen gas bubbles themselves.

While this mode of mechanical entrainment may have radiological significance for species such as fuel particles, the amount of particle mass transported into the gas space by this mechanism would be small.

Experience gained in numerous sodium-concrete reaction experiments demonstrate that sodium and sodium-concrete reaction products, except hydrogen, form two l

separate and distinct layers. These di,stinct layers are evident when the system has been cooled, and may have resulted as soluble materials percipi-tated during the cooling process.

However, evidence indicates that the bulk of the reaction products form a dense imiscible layer on top of unreacted concrete, and would not be mixed in the sodium pool. Thus, these products, in all probability, could not be entrained in the sodium vapor.

Some reaction products (notably-NaOH and Na 0) are soluble in sodium, however, they are not 2

volatile at the temperatures of the sodium pool.

Hence, the only means of entraining these materials in either the hydrogen gas or sodium vapor would be by extremely inefficient mechanical entrainment processes.

I f((l I..'.;. ' '

y.,

9 Thus, the only materials present in any appreciable quantities in the gas space imediately above the sodium pool are the inert gas, hydrogen, and sodium vapor. The inert gas is typically nitrogen, but may have trace amounts (1 to 2%) of oxygen. This oxygen would rapidly react with sodium vapor to form sodium oxide. Thus, initially trace amounts of sodium oxide may also be present above the sodium pool as an aerosol.

Large amounts of hydrogen can be generated by the sodium-water reaction.

Thus, recombination of sodium and hydrogen to fom solid sodium hydride (NaH) would be possible in regions where the temperature is below the sodium 0

hydride decomposition temperature (T 400 C).

The Sieverts constant for hydrogen in liquid sodium is 12 ppm H2per(P)1/2,

and it is essentially independent of temperature. This gives a solubility of 120 ppm hydrogen in equilibrium with 1 atmosphere of H, and less for lower 2

partial pressures. Typically the H Pressure would be less than 1 atmosphere.

2 The themal dissolution of NaH(S) follows th'e relationship:

Log 10 P = 13.86 6 60 (P in N/m2, T in O ).

K 5

2 For P = 1 atmosphere H2(1.03x10 N/m ); T = 422 C.

0 Thus solid hydride is not stable above 422 C, and the maximum hydrogen content 0

of the sodium above 422 C is set by the Sieverts constant. As sodium cools below a temperature where solid hah is stable with the available partial pressure of hydrogen, a substantial amount of hydrogen may combine with sodium.'

^

I However, the kinetics for this reaction is a steep function of temperature as.

given by the reaction:

O ),

Log 10 K = 6.16 -

(K in m3 H /sec N. T in K

NTP 2

~

where K = rate of hydrogen consumption per square meter of sodium surface.

Thus the rate of NaH formation in a cooling aerosol of sodium droplets will be a complex function of the droplet size, hydrogen partial pressure, and rate of aerosol cooling.

l 4;.,. Jl 6

.v...

~

,..)

.... ~_

~

.,,. l. jgy. e

.:.r._

Sodium vapor could condense on colder surfaces and form a solid deposit.

A solid deposit of sodium hydride could also form but at much higher temp-0 0

eraturesthansolidsodium(400Cvs100C). Other solid particulate would be negligible.

4.0 EXPERIMENTAL EVIDENCE GAINED FROM SODIUM-CONCRETE REACTIONS Experience will be cited from numerous sodium-concrete reactions which sub-stantiates the conclusion that sodium vapor is the dominant material above the sodium pool.

Figure 2 shows schematically the main types of test con-

~

figurations used for various size tests at HEDL.

It should be realized that the amount of sodium vapor produced dramatically depends on the temperature of the sodium in the sodium reservoir.

In,the small-scale tests, hydrogen and nitrogen gas and sodium vapor would exit through a 5.08-cm (2-in.) exhaust vent into an air filled cell.

Sodium vapor condensed on the'unheeted exhaust vent and grew in mass until the gas built up enough pressure to blow essentially spericil, fluffy metallic sodium balls out of the vent. These nearly spherical sodium balls landed on top of the surrounding casing.

Because the sodium balls were formed by condensing small sodium vapor droplets, their density was much less than that of pure sodium.

Hence, the balls had a fluffy, light appearance.

Before the balls could be chemically analyzed they were exposed to an air atmosphere and quickly oxidized.

Figure 3 shows a picture of some of these sodium balls.

For the intermediate-scale tests, the test article was placed in an inerted test cell which was either maintained as a closed volume or purged to control the pressure. The inerted test cell provided a large chamber in which sodium,

vapor could condense and either plate out en the cell surface or fall to the chamber floor.

For tests where the sodium pool temperature was maintained 0

0 near the sodium boiling point (850-870 C compared to 883 C), large quantities of " sodium fines" were observed on the test floor. These " sodium fines" were i

pyrophoric in nature and would react very rapidly when exposed to oxygen. This rapid reaction was a source of aggrevation impeding entry into the test celt. '

Numerous cycles of nitrogen and air had to be introduced into the test cell

.to sufficiently react all of these " sodium fines" without significantly, increasing the test cell temperature.

~

n ~*.... J.~;. s o

• f.

d i' "*

h s *

.e e

s s - --

x -

m. -

_=

ugt:

.{'

Y

?.

4 1

h m-f L x n-l I

=

m.

l NWl r-w I

z n_ _ _,

_h k

?

,,. w= =

A,

' - 3S l g

RuLa T.yyll

~

.c.

1. , !L. Ww=-

=% f '

.4 y

.i ~.

.,x_.u

'. ".., -? ::-

q -

; ~ -

x

. - f...%.s,-

y. ar-

.-c-

'o l

. I.

- -r - - ~.. e. _ ~,,,3/ ". t',. ;..'

w.-.

f2L

~*

i:

AIR INERT CELL NITROGEN OR AIR INERT G4S I

s

[

INERT SAS NITROGEN w u sssss w s/sssss w sssssssssssssssss 9

SODIUM SODIUM h-SODILM CONCRETE CONCRETE CONCRETE

~

w i

SMALL-SCALE INTERNEDIATEdSCALE LARGE-SCALE J"

FIGURE 2:-

SCHEMATIC OF TEST ARRANGEMENTS ~USED c' ~

FOR SODIUM-CONCRETE REACT: IONS 9

+~

f n

-3

.'5/

,)

. g\\

i

'.,I ( f(

There are two possibilities regarding the nature of these " sodium fines".

A First, it was assumed that they consisted principally of small condensed

,j tvium droplets with a large total surface area such that they would quickly J.,

W^l react with oxygen. Second, the cell atmosphere temperature and test cell Y,

perimeter temperature were always below the sodium hydride decomposition temperature. Thus,f depending on the sodium concentration, hydrogen concen-i

,(

tration, and temperature, sodium hydride could have formed; and would have i

bn part of the deposited material on the test cell floor. Sodium hydride

~

is'also very pyrophoric when exposed to oxygen. Thus, chemical analysis of these aerosol deposits has been difficult to do, and consequently very little quantitative information is available regarding their chemical nature prior to exposure to oxygen. However, it it now felt that the aerosol deposits are a combination of metallic sodium droplets and sodium hydride I

particles. This fact seems to be substantiated by hydrogen generation data measured during sodium-concrete reaction tests which suggests that hydrogen in the closed test cell was being depleted by sodium hydride formation. '"

~'*r In some of the tests, the test cell was purged through a 1.91 cm (3/4-in.)

~

line, or a 5.08 cm (2-in.) line. The gas flow rate through the small vent 3

l lines was usually very low (1.1 to 2.2 standard m /hr). -In all cases the large test cell volume acted as a settling chamber with most of the " sodium fines" settling on the test floor.

The 1.91 cm (3/4-in.) line by itself would occasionally plug with the aerosol material.

Disassembly of the line and examination of the plug material clearly demonstrated that the material was very fine particles similar to those found on the test cell floor. The particles were very pyrophoric in nature, and appeared to be a combination of condensed sodium vapor or' sodium'

. hydride particles which had accumulated in the cold vent line.

~

~

i.

For the large-scale tests nitrogen and hydrogen gas and sodium vapor would exit from the inert gas space above the sodium pool through a'5.08 cm (2-in.)

vent pipe into an air filled or inerted test cell. There were numerous " '

0.95 cm (3/8 in.) to 1.27 cm (1/2-in.) sample lines confiected'to the gas space of the sodium reservoir or to the vent pipe.

For each test the sodium pool 0

temperature was maintained from 500 to 850 C so that substantial sodium was A

10

.'. T

.c' L..

~

vaporized. The inert gas space and vent line were always above 400'C so that little or no sodium-hydride could form in the vent line.

Invariably the 0.95 cm (3/8-in.) lines would plug with metallic sodium which condensed in the cold lines. This was partially resolved by slightly inclin-1, ing the lines and heating them above 100'C. This tended to reflux the condensed sodium vapor back into the sodium pool.

The inert gas space was also equipped with a pressure relief vent which consisted of a weighted plug resting on top of a larger vent line. For each test, sodium vapor condensed in the closed vent line welding the weighted plug to the walls of the vent line, and forming a solid sodium plug.

Aerosol samples were taken by flowing gas through a 0.95 cm (3/8-in.) sample line attached to the 5.08 cm (2-in.) vent through a sma11' filter. When the filter was removed in such a manner that oxygen was not allowed to enter the filter chamber, the material collected on the filter paper was clearly metallic sodium which had been deposited as sodium vapor.

In the large-scale test arrangement temperatures in tha. sodium. pool, inert gas space, and vent line, after test initiation, were larger than the sodium hydride decomposition temperature. Consequently, no sodium hydride appeared to be present. All evidence supports the conclusion that the materials exit-ing through the vent line were nitrogen gas, hydrogen gas, and sodium vapor.

5.0 REFERENCES

1.

L. D. Muhlestein and A. K. Postma,." Sodium-Concrete Reaction Executive Sumary Report: Application to Limestone Concrete," HEDL-TME 82-15 Hanford Engineering Development Laboratory Richland, WA, June 1982.

2.

A. C. Whittingham, "An Equilibrium and Kinetic Study of the Liquid Sodium Hydrogen Reaction and Its Relevance to Sodium Water Leak Detection in LMFBR Systems," Jor, of Nucl. Mat. 60, 1976 pg 119-131.

11

??

~

e d-

i Prototvne Hvdraaan Filter Tests introduction The Project has provided a letter report on the 1208 Instrumentation The purpose of this report is to provide preliminary development progran.*

information on the results of the prototype hydrogen filter test perf ormed in September 1982 as a part of the TMDB instrianentation development progren.

The objective of the prototype test was to demonstrate the perf ormance of the sodium aerosol filters for the hydrogen monitoring system during prototypic TEDB conditions. This test was designed to simulate the total filter exposure anticipated for TEDB conditions by 100 hours0.00116 days <br />0.0278 hours <br />1.653439e-4 weeks <br />3.805e-5 months <br /> of exposure to sodium aerosol and 600 hours0.00694 days <br />0.167 hours <br />9.920635e-4 weeks <br />2.283e-4 months <br /> of exposure to high temperature. The test involved 3 filter configurations which are described in Table 1.

Test Results For the fIrst 40 hours4.62963e-4 days <br />0.0111 hours <br />6.613757e-5 weeks <br />1.522e-5 months <br /> of the test, the three filter assembiles perf armed adequately. The filter assemblies were repeatedly blown back successfully whenever the pressure drop reached 100 Inches of water, thus demonstrating the adequacy of the filter blow back technique.

At 40 hours4.62963e-4 days <br />0.0111 hours <br />6.613757e-5 weeks <br />1.522e-5 months <br />'into the test, water vapor injection into the test chenber was stopped and carbon dioxide Injection began.

At this time, the filter blow back became increasingly dif fIcult and, at 45 hours5.208333e-4 days <br />0.0125 hours <br />7.440476e-5 weeks <br />1.71225e-5 months <br />, the blou back technique could no Ionger malntaln the pressure drops less than 100 Inches of water. The test was terminated at this time. Table 2 provides the primary paraneters of the test and Table 3 summarizes the results of the test. Figures 1 and 2 provide a comparison of the prototype test chenber temperature and aerosol concentration versus the TEDB Base Case conditions.

Following cool-down, filter assemblies 1 and 2 were disassembled, examined, photographed, and flow tested.

(Filter assembly 3 was preserved for future examination and testing.) The filter disassembly showed that:

e Filters were mechanically intact, Areas of corrosion and small holes were present in the filter o

media, Filter surf aces appear darkened and glossy.

o Settling chenbers and filter surf aces appear relatively free of e

caked aerosol but I d-up.

Filters lost approximately 2% of their weight.

e

• Letter HQ:S 8210%, J. R. Longenecker to P. S. Check, "TEDB instrumentation Development," dated September 29, 1982.

Test Conclusions The prototype test demonstrated the feasibility of desiping filters to accommodate the high temperature aerosol Iceding anticipated for 19608 condiflons for significent periods of time (See Table 2). The Project is currently beginning a series of laboratory-scale bench tests to obtain a more complete understanding of the phenomena involved in limiting the time of effectiveness of the individual units in the test.

In addition, the Project is evaluating whether it is realistic to egoct free carbon dioxide to exist in the vicinity of the hydrogen filter assemblies during TEDB conditions.

Following completion of these tests and studies in January 1985, the Project will re-evaluate the design of the filter assemblies and perform additional testing of candidate filter assemblies.

In the unlikely event that a single filter assembly cannot be desipod to accommodate all the TleDB environmental conditions for the time required, alternate design configurations, such as a series of parallel filters, could be utilized to meet the 11608 conditions for the required length of time.

TABLE 1 FILTER TYPES AND PARAMETERS HEE SETTLING CHAMBER DESIGN FILTER 101 SINTERED WIRE MESH OPEN BOTTOM FILTER 102 SAME AS 101 WITH A CLOSED BOTTOM NICKEL POWDER LAYER BONDED TO UPSTREAM SURFACE OF SINTERED WIRE MESH FILTER 103 SINTERED NICKEL POWDER OPEN BOTTOM i

O i

1

TABLE 2 PROTOTYPE TEST CONDITION

SUMMARY

Maximum Temperature 1140'F 3

Maximum Aerosol Concentration (1) 55 g/m 3

Average Aerosol Concentration (1)

% 40 g/m Steam Flow s 15 vol %

Start at 11.2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> Stop at 41 hours4.74537e-4 days <br />0.0114 hours <br />6.779101e-5 weeks <br />1.56005e-5 months <br /> CO Flow 1.9 vol %

2 Start at 39.8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> -

Stop at 47.6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br /> Aerosol Composition NaOH, Na 0 at 4.8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> 22 Na 00 at 40.8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> 2 3 Na CO at 45 hours5.208333e-4 days <br />0.0125 hours <br />7.440476e-5 weeks <br />1.71225e-5 months <br /> 2 3 t-(1) At chamber tenperature l

1

TABLE 3 SU W RY OF PROTOTYPE FILTER RESULTS 2

Filter Area, m 0.053 0.054 0.058 Flow Rate, cc/ min (1) 3000(2) 3000 700 Flow Duration, hrs 42 42 44.7 Calculated Mass Filtered, g 528 904 220 Time at 1st Blow Back, hrs 10.0 2.1 11.4 Total Number of Blow Back (3) 33 35 14 2

Loading at 1st Blow Back, kg/m 3.0 0.24 0.07 2

Final Loading, kg/m 10.0 16.8 3.8 Filter Efficiency, %

95.0 92.1 98.8 (1) At 70*F; the TMBDB filter flow is estimated to be 700 cc/ min at 70'F (2) Flow reduced to 1500 cc/ min at 21 hrs (3) When filter AP increased to 100 inches H O 2

1 Temperature (*F) l 1600 1400 j

i Test CW Te 1200 U

-Test Filter Flow Stopped 1000 800

~

~

TMBOB Containment Atmosphere Temperature (With Margins included) 400 200 L

I I

I I

I I

I I

0 O

20 40 60 80 100 120 140 160 Tkne (Hours)

Figure 1. Comparison Of TM808 Containment Atmosphere Temperature To Test Chamber Temperature 194331661

i Aerosol Mass Concentratinn 3

(gm/m At Chamber Temperature) i 100 l

l 80 -

/ jTM8D8 Base Case j

V

/ \\

/

\\

60 -

Filter Testh

/s,s'

\\

1

\\

/

\\

/'N' r

\\

I

's f

i

/

/n

\\/

/

Ns/

4

\\

/

I i

i

/

\\

/

1 l

l 20 y

,/

/,a**'

Staat Steam Start CO2 Stop Steam f

p njection iniection 3 pInjection I

/

I Y I

I I YY I

I 0

I o

10 20 30 40 50 60 70 Time (Hours) l I

Figure 2. Aerosol Mass Concentration At Chamber Temperature

e 0

6 1

i