 

 

-

 

 

 

 

>  ORNL-TM-2373
'
R REACTOR CHEMISTRY DIVISION
‘ i . ten Salts
Bubbles, Drops, and Entrainment in Molten Sal
2 ! , | -
i " H. W. Kohn

 

 

 DECEMBER 1968

 

 

 OAK RIDGE NATIONAL LABORATORY
p  Oak Ridge, Tennessee
| . : a "~ -operated by
o R . UNION CARBIDE CORPORATION

 

 

- . for-the: '
e o ~ U. S. ATOMIC ENERGY COMMISSION
. SR I LEGAL N_OTICE;,{',,?;  ‘ !

 

¢ 5 ; A. Makes any warranty or Yepresentation, expressed or impled, with respect to the acoy-
; ‘ ; ! racy, completeness, or usefulness of the information contained fn this Téport, or that the uge

of any information, apparatus, method, or Process diaclosed In this report may not infringe
; . L Privately owned rights; or .

 

  

. OCUM
BISTRIBUTION OF THIS O
Or employee of such contractor Prepares, i . .
! disseminates, op Provides accegs to, any information pursuant to his employment or contract )
+ with the Colnnuuion. or his employment with such contractor, oo :

 
 

 

 

 
 

 

 

.,u.l;,f

 
 

 

 

H

o

Bubbles, Drops, and Entrainment in Molten Salts

Harold W. Kohn
Reactor Chemistry Division

Oak Ridge National Laboratory
. Ogk Ridge, Tennessee 37830
To entreinl 1s defined as follows, "to carry along or over (especisal-

1yjmeohanica11y) es fine diops of liquid during distillation". One can

thus envision meny entrainment systems, some of which are pertinent to

‘ studiés*being:done here (viz. entrainment of process salt by 1iquid

bismuth) The iiscussion in this report however will be.confined to

studies of the entrainment of solid and of liquid particles by gases.,
Molten selt processes associated ‘with the MSRE usually use an

inert cover'gas. This gas is often moving along the surface and, in

some'processes, is even swept through the molten salt. Hence we can

‘expect some form of gasipartiole entrainment to play a part in most

molten salt experiments;' The situation is particulaerly aggravated in

- the MSRE pnmp bowl'where a considereble (L E/min. corresponding to a
: minimum L. S'V of 0.35 cm./min ) flow of helium is used to sparge xenon

- from the reactor fuel .

 

* . ; T R ‘ ' S ' o
Research sponsored by the U.S. Atomic Energy Commission under

- contract with the Union Carbide Corporation.

-lwebster’s Seventh New'Collegiate Dictionary, G & c. Merriam Co.,

Springfield, Mass., 1965

“ng C. Robertson, MSRE Design Operations Report Part I, Description

of Reactor Design, ORNL-TM-728. See pp. 139-155 for a complete

. description of\the-pumpibowl'operation, and Fig. 2.3, p. 13.

 
 

 

In order to have entrainment, some mechanism for the production of
.fine particles is also reqnired. Our literature search has shown thet :
bubbles and splashes are the principal sources of fine droplets end
spreys. -Again let us direct our attention to the MSRE pump bowi. This
boﬁl is a lentil shaped resefvoir containing about oﬁe_hundred'liters
ofumoiten salt. One inch gbove the surface of this salt pcol is &
sprey ring from vhich salt end helium_ere sprayed at a lively rate
(sixty-five gallons per minute). This lesds to entreirment of one to
two perceht of the helium bw'theifuel. It can give riee not oﬁly to
some directly formed asercsol, but alsorseconda:y droplets from spleshes.
jSince-the,fuel now also contains up to two percent by volume of helium
bubbles, on reaching the (pump bowl) surface, these may give rise to i
Jet droplets when they burst. d

These phenomena.may bear on the ultimate fate of the MSRE fission
products in the following way: there are reasons, derived from chem-

ical thermodynamics3 for believing that many of these fission products,

specifically Nb, Mo and Ru, are present in the fuel es metal. As metals;.

their vapor pressure would be vanishingly small, yet they appear to
favor the gas phase over the liquidh.' Since the MSRE fuel does not

wet the m.eta.l5 the spray from the sPray ring could be diSpersing a

metal fog directly, i.e., there is no reason to expect & droplet of salt

 

3b F. Baes, Thermodynamics, Vol. II, 257 (IAEA;.Viennad(1966))

hS. Kirslis and F. F. Blankenship, MSRP Semiannual Progress Report for
period ending Feb. 29, 1968, ORNL-4#52%, p. 9%. .

5p. J. Kreyger, S. S. Kirslis and F. F. Blankenship, MSRP Annual
‘Progress Report for Period Ending Jan. 31, 1964, ORNL-3591,
 

N

6

" plns'metal to adhere. :This non-wetting charscteristic also crestes a

possibility for flotation of the metal particles in the fuelﬁ. There

- is evidence, as discussed later, that such an interface scum would'be

preferentially removed (ejected into the gas phase) by Jjet droplets

| from bursting bubbles. .The'pump bowl liquid is quite agitated which

argues against this mechanism. Without a detailed study of flow
patterns, however, one cennot say for sure that areas amengble to the
existence of stable scum do not exist in the pump bowl. It is much
more iikely that a stable scum exists within the sampling ares since
this area is protected,from'the turbulence by & cylindricsl mist shield
wﬁich_extends_from the top of the'gas spece nearly to.the bottomrof the
1iquid. Therefore_consideration of the entrainment_process not only
points the way to an.explanatiOn of the peculiar disposition of-tne
fission products, it indicates that the samples, since they are taken
-within the mist shield;'na&'not be representative of the processes going ‘

S

on within the reactor pump bowl.
Summary of Previous Work

A literature search was completed using "drops",, entrainmen *,

end. "bubbles" as key words. Discussions with laboratory staff members

fsupplied additional references._‘

An actual measurement of entrainment of Csla? in a.boiling water
0

nx,reactor was made by Shor and co-workers . They pointed out the ‘many

 

H. W, Kohn, MSRP Semiannual'frogress'Report for Period Ending Feb. 29,
11968, omm-hash, P 127, also "Flotation", A. M. Gaudin, McGraw Hill

7@. Je Shor, H. T. Ward, D. Miller and W. A. Rodger, Nucl. Sci. and
Engr., 2, 126 (1957)

 
 

 

complexitieSIOf the problem including free or forced convection con-
‘ditions, contamina.tion of hest transfer surfa.ces 3 suspended and dissolved o 4
impurities e.t the interfece, power level, power density, and oPera.ting |
pressure. Even so they obtained good linear plots of the 10garitbxn‘of

 the decontemination factor and the power input, and of log D, vs. steam
velocity. | |

A grest deal of the definitive work on Jet droplets was done by

D. C. Blza.nc:l:ussrd8 and his assoclates in connection with oceanographic

vstudiesg 1h.

Most of the information is contained in Figs. 1 a.nd 2 of
this report and concerns bursting bubbles in sea water. One may convert
from mass of salt to .droplet size by remembering the sea water conta'ins |
from 3.15 to 3.5% salt. The other data has been added to the graph
using the 3.15% figure. The results of a bubble bursting at a surface

can be both spectacular end surprising. If we limit ourselves, as

'y

Blanchard and associates did, to bubbles two millimeters end less in

 

8Woods Hole Oceanographic Institute » DOV at the State College of New

York at Albany.

9D. C. Blanchard, "Progress in Ocesgnograsphy,” Vol. I. M. Sears, Ed.,
Pergemon Press, Inc., New York, 1963, p. T1-202.

lOD C. Bla.ncha.rd "From Raindrops to Volcanoes," Doubleda.y and Co.,

Garden City, N. Y., 1967.

L), ¢. Blancherd, Nature, 175, 33h (1955).
12

C. F. Kei.ntzler, A. B. Arons, D. c. Blancha.rd and A. H. Woodcock
Tellus, 6, 1 (1954). _ _

AR

13p. ¢. Blanchard and A. H. Woodcock, Tellus, 9 15 (1957).

lhD c. Bla.ncha.rd, Nature, __’Q, 1048 (1951*) - S | *
 

 

 

 

  
   
    
    

 

 

 

1-06 . : ORNL—-DWG 68—6984
E T
: . :*——:-
3 x i 4600
5 - *
¥ - | | ok e
- I TOP DROP .
~ SECOND DROP | | )
E 4 4
110 -
1 ;
=k ;
< - -
U) pem
& | |
o103 L o TOP DROP _
2 - -« SECOND DROP 3
S - 4 THIRD DROP -
- s i FOURTH DROP 4
- « JAPANESE DATA -
- *« M.S. DATA A
x BRITISH DATA
102 |- * RUSSIAN DATA  —
ot e e 1
-0 - 000 2000

- DIAMETER OF BUBBLE ()
| Figure 1. Size of jet droplet as & function of bubble size. After
* Blanchard and Woodcock (Ref. 13). Other data has been added by
- - COnsidering sea water to be 3.15% salt. Molten salt data referred

to equivalent mass of a vater droplet.

 
 

 

 

 

   

 

 

 

ORNL—DWG 66-6983
| l. T T T I T T T 1 I 3 Y T T 1 . i
‘20F - 1
, | s ,
. . - -
- T n [ -
© * .
5 -
L /"\\ | | 7-
! N !
\
E | N
= T 2 N
T 10 | Y
L:|EJ i \ - 44600 o
* i STUHLMAN, DISTILLED WATER— | - e
o TOPDROP | | "
I « SECOND DROP - B
,. 1+ THIRD DROP , o ¥
- - « FOURTHDROP {1 |
s | -\ <. N " JAPANESE DATA
. L \* RUSSIAN DATA
TN \ , -
0 : — |
o 1000 2000 3000
DIAMETER OF BUBBLE () |
| -"Eigure 2. Rise height of jet cifoplets vs.» bubble diazﬁeter. Data 1

from Refs. 13, 29, 31, 36.
 

 

diameter, ‘we see that from five to seven Jet d.roplets can be ejected
per bubble 'burst, and that these drops mey be up to nearly 0.2 mm. in
.diameter, and all the drops might contain close to one milligram of
meterial. The topmost d.rop is flung nearly twenty centimeters into the
‘air; the others do not perfom.so spectacularly.

The high speed photographs shown in Refs. 12 and 15 delineate quite
'clearly and remarkably the history of the bursting bubble which is also
shown pictorially in Fig. 3. If the surface is clee.n,- the bubbles will
_burst almost immediately. .No agglomeration (growth of large bubbles at
the expense of small ones) was observed for sea water. .Liquid from the
film at the top of the bu'b'ble dra.ins until it ruptures ; & flow of
liq_uid down the side of the liquid ca.vity then ensues. This leads to
the formation of the Jet drop, (Fig. 3). A vortex ring is elso formed
and ejected downwa.rds, shown quite clea.rly by using India ink as a
tra.cerlG. However the foma.tion and beha.vior of this vortex ring is

- not at all clear. Recent studies” 20

have led to the conclusion that
it is formed by the Ra.yleigh_ jet and the drops which subsequently re-

enter the liquid. Jet d.roplets can also be formed by spla.shes in one

 

194, M. Worthington and R. S. Cole, Phil. Trans. Roy. Soc. (London)
&, 189, 137 (1897); A, 194, 175 (1900). |

l6-F'. MacIntyre, Je. Phys. Chem., ]_, 590 (1968).

. 3‘7 -. V Hobbs and A.. J. Kezweeny, Science, 55, 1112 (1967)

lBP. V. Hobbs and T. Osheroff, Science, 158, ll6l+ (1967)

lgw. Hall C. Me.xwell, Science, 160, 907 (1968).

20p, v. Hobbs, Science, 160 907 (1968)

 
 

ORNL—DWG 6'8—'6982

 

Figufe 3, Time sequence diasgram of a bursting bubble, deduced from

high speed photographs 12 and theoretical calculationseh. After

MacIntyrels.
 

 

 

Cw

.y

"

and cited two observations

of two manners. A'drop hitting the surface invariably punches a clear

cylindrical hole which may or'may not have a sheer cylindrical wall

14,17,18

rising from it This wall either closes over, or forms a crown

which breaks up into.a,fihe,spray. If the hole does close, it soon

reopens, and a long Jet vhich breaks into jet droplets appears.
Blencherd end associatés considéred'alsb the formation of bubbles
from splash drops and the equilibrium between small bubbles and dis-
solved gas. The splash drdps producé very tiny (fifty micron) bubbles.
A father discouraging feature of this mechanism for material franspprt
is that ﬁnlgss thefe-is aﬁequate supersaturamion most of these bubbles
will go back inﬁo solution. However, in sd doing they will raise the
supérsaturation until a point is reached where such bubbles will grow.
Blanchard also found that the Jet droplets from bursting bubbles

9,10, 1%

carry a positive charge A greater charge may be induced by

applying a local field;'by using a positive field the charge sign on

17

the drops can be reversed; Hobbs and.Kezwéény on the other hand

fbund that the droplets fbrmed from splashes carried a negative charge
21,22 in support-of their measurements. This

differencezin'charge sign;seems7most,unusual since_thé mechanism of

- formation of the large drqplétsffrom,the Rayieigh_Jet seems identical
with its fbrmation from;ﬁubbies, 'Thé larger droplets,:those from the
Jet rgther thgn.frbm_the cr6wn,'carry the higher charge per unit mass,

 thus accéhtuating_thé difference. Hobbs has also observed a linear

 

i 21P Lenard, Ann Phys. Leipzig h6 58l (1892), ibid 47 h63 (1915)

2. T. Plerce and A. L. Wnitson, J. Atmos. Sei., 22, 31k (1965).

 
 

10

| relationship between the'nmnbei" of splash droplets and thé_ distance of
fall., He 'bé.s also observed a pesking in the maiMm rise height at a
particular liquid depth, ebout 8 mm. for water. |

An important consideration from our standpoint is that the jet
droplets cen also remove surface conteminants. This had been showm
experimentally by Blancha.rd25 (see also ref. 8, p. 107); the mechanism
of this removel and‘the- use of bubbles as e surfece micrqtome has also

| been discussed by MacIntyrelG.

A great deal of this work has also been discussed by Tobaah 28
these pu’qlice.tions were not a.vaila.ble at the time of this report.
(They are listed so that the bibliography will be coxnplete;) The Jet'
'drop experiment wes also performed in Bussia. using photographic tech-
niques by Gleim and associatess”’ 30; some of their data are included

in Figs. 1 and 2.

 

23b. . Blanchard, Science, 146, 396 (1964).

th. Hayami and Y. Toba, J. Oceanog. Soc. Japan, 1k, 1h5 (1958)

2%. Toba, J. Oceanog. Soc. Japan, yt, 151 (1958).

26Y. Toba, J. Oceanog. Soc. Japan, 15, 1 (1959).

27Y. Toba, J. Oceanog. Soc. Ja.pan, 15, 121 (1959).
28!. Toba, Meterological Soc. Japen, 1__&2, 63 (1962).
29 . G. Gleim, Trudy Novocherkassk Politekh. Inst., 25, 173 (1955)

3%. 6. Gleim, I. K. Shelomov, end B. R. Shidlovskii, Zhur. Prikled.
Khim., 32, 218 (1959). |
 

 

.

I} a

11

- The oceanogrephers were interested primarily in bubbles less than
two millimeters in dismeter and this led them into a mild controversy

with e rival group of engineers (Imperial College, London) who were more

interested in the bursting of larger (2-5 mm.) bubbles. The 2-3 mm.
_size_range is eritical'sinee it is generally acceptedBl that in water,

-bﬁbbles greater than 3 mm. in diameter are inherently unstebleBl. The

Imperial College group has shown at least for large bubbles that in

“eddition to the Jet droplets that accompany bubble breaknp, there is =a

fine spray resulting from the breakup of the bubble dome33’3h. For

bubbles greater than 5 mm,'diameter,-all the droplets came from rupture

‘of the bubble dome. They also discovered that a temperature increase

caused & marked decrease in the number of jet droplets. The question

was pretty much resolved by another group of engineers (Birmingham)35

who showed that most of the_mess of the‘drqplets from bubbles 2 mm.
and less in size came from the jet droplets, whereass for large bubbles
most of the mess comes from-breakup of the bubble dome. They also

demonstrated thet entrainment_decreased repidly with bubble size.

 

310 -Stuhlman, Physics, ‘ h55 (1932)

) '320. Miysgi, Philm Mag., 2“, 112 (1925)

5 3D M. Meritt, N.-Dom‘browski, and F. H. Knelm&n, ’I‘rans. Inst Chem.

Engr., 32, 24k (195k).

Bhf. H. Knelman, N. Dambrowski, and D. M. Meritt, Nature, 17§, 261
195k4). |

35§, H. Garner, S. K. M. Ellis and J. A._Lacey, Trens. Inst. Chen.

Engr., 32, 222 (1954).

 
 

 

12

Two groups of Japenese iﬁiestigatorslhawe also studied entréinment
due to jet.drops. One group studied-Jet drop formation in wster at
20°c, 2.1% n-butylic acid (a pronounciational misprint‘no doubt) and
in 50% glycerin éolﬁtion36. Contrary to the results of Garmer et al.
 31ycéro1 and n-butyric acid solutions gave curveé'similar.inform to
| those from water, but the maximum rise height of the droplet was lower
tban it was for water.- The second group investigsted tﬁe.felationshiﬁs
bet#een bubble size, drqplet‘sizes,_and physical properties of:the liquid,
particularly viscosity end surface tension37. These date are presented
as graphs and equations in their publication. As they point out, how- .
ever, it is nearly impossible to'vary only & éingle’parametér (e.g.
surface tension) without simultaneously changing enother (e.g. viscosity).
All these investigations havé followed thérformulation‘of Davie538
in equating the vertical force P.(27#r) with the force on thé'maSS of
‘liquid set in motion (mg). This_lea#es no roﬁm for the effect of vis-
cosity even though such an effect has been observed'repeatedlyjs’is.
(A viscosity effect might find its way'into the constant in the empiricel

16 that the impulse

equation in ref. 36.) It was recently pointed out
force arising from the rupture of the bubble-can be dividedinto-two
parts involving inertiel and viscous forces and for water those are
.approximétely equal. The equations involved, however, are not partic-

ularly smenable to direct solution, especially for s molten salt system.

 

36N; Mitsuishi, Y. Matsuda, Y. Yamamoto, Y. Oyama, Kegaku Kogaku,
22, 680 (1958). | | |

3Ts. Atba end T. Yemada, AICKE Journel, 5, 506 (1959).

3BF. Davies, Proc. Inst. Mech. Engrs., 1kh, 198 (19Lk0).

oty
 

 

[r— ’\

wi

‘!I)

| recently been'demonstrated'experimentally

 

13
' One'othef possibility for entrainment, "volumetric evaporation",
mﬁst be mentioned39’hO;This phenoﬁeoon, invol#ing the entrainmeﬁt of
very-fine (suhmicroscopic)jparticles by repidly evaporating liquids has
7 24 by measuring the lose of a
non-volaetile material, pdfasSium dichramaie,’during evaporation of a

SOIution;of this material from a porous Celite sphere. Previous con-

Jeetures shout this mechagism‘of entrainment were based on changes of

- the heat transfer coefficient. Although the physical arrangement in a

liquid-gas system is quite'different-from that described by Gauvin, the
constant loss of helium inithe MSRE supplies & volatile - non-volatile

relationship which could lead to "volumetric evaporstion“.
Discussion

Table 1 lists pbysicsl constants for the systems investigated in

these publications and also the best values for the molten salt systems

we are interested in, The'last two columns and three lines show the

éffect of changing the teﬁperature in water systems. This is seen to

have a large effect on the Jet drops, the effect on £ilm drops (at dis-
- tances greater than l Cl. from the surface of the liquid) is much less
:marked. Raising the temperature by 20 C lovers the viscosity by 33%,
o'which one would expect to enhance Jet dr0p fbrmation, yet it does not
. seem to have as big an effect as simultaneously lowering the surface

l tension.by only h~l/2%.r'Nsvertheless; since this lowers.the‘number of

 

| :39A. V. Lykov, Int. .Cﬁan_.Eng., 3,.195 (1963).

 
Table I

Physical Properties of Several Liquids
(Molten salt data is from S. Cantor et. al, ORNL-TM-2316 (1968))

| Substance (am./m1.) y dynes/em. v centipoise N .65% NLB.lla Ref. No.
Water 25° 0.997 71.97 0.89 - .8.25 43.0 35
35° 0.99k 70.38 0.7 3.5 275 35
- u5° - 0.990  68.7h | 0.60 2,04 18.3 35
50% glycerin 1.1263 69.9 6.05 | | f | 36
2.1% Bu. Acid = ~L.0 o 092 | . | 36
Benzene 0879 289 0.65 | =
LiMOsKMs .93 120 | |
LiF-BeF» 1000%K 2.15 | 180-~195 5

~ %Refers to the number of Jet drops intercepted by a sensitized microscope slide h.hh dm. above
the bubbling solution. The subscripts are the size in mm. of the bursting bubbles. See references
for more ccmplete data. ' ' - | ' ‘ ‘

Measured here crudely' Extrapolation from individual melt data gives ~12, 3 CePs which 1s
‘obviously incorrect. , ,

T

 
 

oy

15

jet drops, it is not at all epparent, due to the high viscosity, thet

bubbles_in a.moltenhﬁait will eject Jet drops. - We perfbrned a few crude

experiments, catching jet;drops on & glass microscope slide at varying

heights from & molten salt (KNO3-LiNO3 eutectic a£'170-180°b)_solution;

b1

and using 8 vibrating capillary to produce the bubbles ~. Bubble sizes

were measured by combining photographic and.microscopic techniques.
This solution behawes-much 1ike glycerin (see Fig. 4) and the relation
between bubble size, jet drop size, and rise. height, is not sufficientl&
different from the relation in water to make us suspicious of the
results.

It has been observed previously k2 that cover glasses over nitrate
melt become clouded with salt. It was_postulated that hhis was due to

the ejection of microscoPic droplets of salt ejected from bubbles of

oxygen generated hy,the:reaction:’

N0z~ = No2~ + 3 Os.

We undertook to repesat this work uwsing a physical arrangement which

would avoid the condensation of volatile impurities as a mechanism of

deposition. Five 2 ml beakers containing molten LiNO5-KNO5 eutectic

‘were'placed on a glass ﬁlaiform in a'hOO ml. besker sustained at 185QC.
The nitrate melts contained, respectively nothing else, NaNOE, HEO, ;

‘Hnoa, &nd NaHCOa. All slides showed e fog, the slides covering the

320 and HN03 impurity melts showed a weight gain of l.mg, the one

,covering the NaROa impurity result gained only 0.2 mg.' The other two

- rwere chipped in handling and hence showed weight 1osses.-

 

thh MacIntyre, Rev; s¢1; Inst.; 38, 969 (1967).
L2 | |

J. Braunstein, priuate communication.

 
 

16

ORNL-DWG 68-6985

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

22 ~ - i [
20 ® LiNO, ~KNO, EUTECTIC |
18 * /\ Hz0 —— H,0: BLANCHARD AND —|
1 WOODCOCK
] C,H,OH: MITSUISHI ef o,
6 — - +=  CgHg: STUHLMAN ~ —
- CxHx(OH),: MITSUISHI
= 14 LiNO5KNO THIS WORK —
= : ‘ \
o
g A,
7
5 /N
\
o / \ \
S / N —2.1% n—C4H,COOH
- ‘
S 8 !
o | .
T l/ % C4H(OH)y
6 / '
¥ | |
4 / , ' ‘\
/4 \ \ \ |
WL NN\
2 /// | \‘\\
§ ‘ . \\
O — ‘ = »
0 1 2 3 a4 5. 6

BUBBLE DIAMETER (mm)
~ Figure k. Droplet height vs. bubble diameter. Data from references

13, 31, 36 and this laboratory.

Y
 

,"“.

L

,"bubbles could remove surface meterial as previously discussed

 

17

This phenomenon appeared to be analogous to the volumetric evapor-

39, 40

ation mentioned earlier « Ve undertook to repeat the experiment,

, this time using five beakers containing, respectively. no additive,

gold colloid, silwer, a surface coating of talc, and a surface coating

'of graphite._ The same sort of results were Obtained. The particles

were visible in.the'microscope, and were barely resolvable, thus

‘measuring about O.ky in diemeter. None of the added.materials seemed

-to interfere with the deposit vhich was easily visible after a day

The.deposit over the carbon,was not gray, chemical analysis showed no

entrainment of thersurface or colloidal materials. USing the Japanese

data as a guide, such droplets should arise from bubbles less than 51

in dismeter. Such bubbles,,however, should be subjected to an extremely

high pressure (7/r) 6 x 108 dynes/ 2) and should thus all go back

into the solution until_the,supersaturation becomes indeed very high.

The question was resolved by an analysis of themainmatrixmaterial

vhich proved to be almost exclusively NEgNOg. The very slow volatiliza-

,tion of this minor impurity should not be confused with the rapid

| volatilize.tion required for volumetric evaporation. ‘, o

“, One experiment was tried to see if the Jet drops from bursting
,16

A ManIntyre bubbler (bubbles 2 mm. in diameter) was set 1 cm. below the
-surface of a LiNOaéKNDS eutectic melt, the surface was sprinkled with
N fine (< 0. §u) carbon and several jet drops were collected. After ex-

'posure to the air, the eutectic had absorbed enough weter to turn to

& liquid. Meny, but not all, of the drops uere then seen to contain

carbon (Fig. 5). A small area of the melt surface vhere the bubbles

 
 

 

 

 

Figure 5.

18

Jet droplet artifacts

PHOTO 92241

 

entrained cerbon.

 

 

 

Ay
 

 

 

 

-,

Y

 

had been bursting wes observed to be free of carbon.

& . "V = L -
e - EEATNL
S tET R R ViR
AooEg s !

- ‘Conclusions

The'production and behavior-of drops froﬁ splashes and bubbles has
beeﬁ reviewed_and deséribed.' Possible applicatidn of these phenomena
to the MSRE pump bowl has been considered. - |

There are thrée sorts 6f drops which can contribute to entrain-
ment.in fhe MSRE, Jjet drqps-fram splashes and from small récirculated
bubbles, film drops frbm.the;breakup of the film cap of larger bubbles
end crown-drops from the breaskup of the crown.from.splash drops. In
additioh, theré might be aéonsiderable diréct asrosol formation due
to spray from the spray ring, but we have, at present, no way of

evaluéting this. The smount of salt available for entraimment from

. these sources is much lerger than that actually lost from the reactor,

hence most of the droplets must be returned somehow to the liquid streem.

It bas also been shown that Jet drops can preferentiaslly remove a
surface film.ffdm a_molten_éaliﬁsﬁrface.. This surface ﬁicrotoﬁe efféct
may well be-cogtributingtd the apparent. loss of fission products to

the gas stream.

| _'”Aékﬁoﬁiedgement'

- The author_wishes td:ﬁhankubfs. S. S..Kirslis,and F.fF;'Blankenship

 _for>heipfu1dis¢ussidn,a#a:ﬁis:wife fbr_assistancé witbthé 1itérature

search.

 
 

| -

 

 
 

 

1.
2-1".

6-T.

8»58
59.
60.

61-65.
66.
67.

- 68.
69.
T0.
T1l.
T2.
T3
Th

76,
T1-87.
88

89.
90.
91-
92.
95
ok.

95.
9.

97.
98.

99.

100.

. 102,

103,

10k,
- 105.
106,

- 107.

108,

- 109.

110.
111.
112,
113.

21

INTERNAL DISTRIBUTION

Biology Library B
Central Research Library
Laboratory Shift Supervisor
ORNL Y-12 Technical Library
(Document Reference Section)
Laboratory Records Department

Laboratory Records, ORNL R.C.

H. G. MacPherson
G. E. Boyd

¥. R. Bruce

F. L. Culler

A. F. Rupp

'R. B, Briggs

M. W. Rosenthal
K. Z. Morgen

E. H. Taylor

E. S. Bettis

M. A. Bredig

P. N. Haubenreich
P. R. Kasten

W. R. Grimes

. B. G. Bohlmann

H. ¥. McDuffie
G. M. Watson

' F. F. Blankenship
‘R. D. Ackley

R. E. Adams
S. I. Auerbach

C. F. Baes

C. E. Bamberger

'C. J. Barton
- R. L. Bennett

W. D. Bond
J. Braunstein

" R. E. Brooksbank

W. E. Browning, Jr. - -
G. D. Brunton

T. H. J. Burnett

H. M. Butler

S. Cantor

R. M. Carroll

G. I. Cathers
J. M. Chandler

C. F. Coleman -
‘B. L. Compere

K. E. Cowser '
W. M. Culkowski (AEI:)

114,
115.
116.
1170
118.

119.

120.
121,
122.
123,

124,

125 .
126.
127.
128-168.
169,
170.
171.
172,
173.
17k,
175 .
176.
177.
178.
179 .
i80.
181.
182.
183,
18L.,
185.
186.
187,
188,
189.
190.
191.
. 192 »
193.

194,
- 195.

196.
197.
1198,
199.

200.

201“215 .
216.

D.

R.

V.
R.
L.

 

ORNL-‘IM-23T3

R. Cuneo -

J. Davis

de Lagunsa

B. Evans, III
L. Fairchild

Birney Fish

H.
F.
P.
R.
A.
G.
G.
S.
H.
R.

‘A.

A.
W.

- J.

L.

J.

G.

L.

G..
T.

K.

A.

A. Friedman
Gifford (AEC)
A. Hasas

G. Hsire

R. Irvine

H. Jenks

W. Keilholtz
S. Kirslis

W. Kohn

E. Leuze

L. Lotts

P. Malinauskas
L. Marshall, Jr.
P. McBride

E. Morse

G. Morgan

W. Parker

M. Parsley

D. Robbins

H. Row

A. Romberger
D. Ryon

Dunlap Scott

J...
! A.

M.
J.
.B,.
R.
E.

F.

T.
R.

L.

Je
cC.

Ce
Po

E.
R.

H. Shaffer
J. Shor

D. Silverman

W. Snider

A, Soldano
A. Strehlow
G. Struxness
H. Sweeton
Tamura

E. Thoma, Jr.
M. Toth

Truitt

D. Watson

F. Weaver

H. Emmett (consultant)
A. Mason (consultant)
F. Newton (consultant)

Howard Reiss (consultant)
DTIE, OR
Leb. and Univ. Div., ORO