ORNL-TM-4870.txt

LOCKHEED MARTIN ENERGY RESEARCH LIBRARIES

| (2
llf I g Ef‘.

3 4456 O45017Y &

  

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

  

s

 
 

 

i

 
 

Printed in the United States of America. Available from
National Technical Information Service
U.S. Department of Commerce
5285 Port Royal Road, Soringfield, Virginia 22161
Price: Printed Copy $5.50; Microfiche $2.2

 

 

 

This report was prepared as an account of work sponsored by the United States
Government. Neither the United States nor the Energy Research and Development
Administration, nor any of their employses, nor any of their contractars,
subcontractors, or their employees, makes 2ny warranty, express or implied, or
assumes any legal liability or responsibility for the accuracy, completeness or
usefulness of any information, apparatus, product or process disclosed, or represents
that its use would not infringe privately owned rights.

 

 
ORNL-TM~4370
UC~-76 — Molten Salt Reactor Technology

Contract No. W-7405-eng-26

CHEMICAL TECHNOLOGY DIVISION

ENGINEERING DEVELOPMENT STUDIES FOR MOLTEN-SALT
BREEDER REACTOR PROCESSING NO. 20

Compiled by:

J. R. Hightower, Jr.

Other Contributors:

C. H. Brown, Jr.
E. M. Counce

R. B. Lindauer
H. €. Savage

JANUARY 1976

OAK RIDGE NATTONAL TARORATORY
Oak Ridge, Tennessee 37830
operated by
UNION CARBIDE CORPORATION
for the
ENERGY RESFARCH AND DEVELOPMENT ADMINLSTRATION

NERGY RESEARCH UBRARIES

A

3 uy5k Qus0L7H b
ii

Reports previously issued in this series are as follows:

ORNL-TM-3053
ORNL-TM-3137
ORNL-TM-3138
ORNL-TM~-3139
ORNL~TM~3140
ORNL-TM~-3141
ORNL-TM-3257
ORNL~TM~-3258
ORNL-TM~-3259
ORNL-TM-3352
ORNL~TM-4698
ORNL-TM-4863

Period
Period
Period
Period
Period
Period
Period
Period
Period
Period
Period

Period

ending
ending
ending
ending
ending
ending
ending
ending
ending

ending

December 1968
March 1969
June 1969
September 1969
1969
March 1970
June 1970
September 1970
1970
March 1971

December

Decembher

January through June 1974

July through September 1974
CONTENTS

- S’[I}B}{LARLES - . . . - - [% - - - - b . . . . » . - L] - . - . . . . . . . v

1. INTRODUCTION © & ¢ & v o v o 4 o 4 o o & o s o o o« o s o o & » 1

2. CONTINUOUS FLUGRINATOR DEVELOPMENT: AUTORESISTANCE HEATING TEST
A.—H:T"'B . " e - . T e s . . . e . . . . . . » . . . . . . . . . L 1

2.1 Experimental Equipment and Procedure, . . « « « « « « + o . 2
2.2 Experimental ResultsS. . + o « + 4 & o v o o o o o o o« o+ 4
2-2-1 I{un AHT—3“1_0 - . . . . . . . . - - . . . . - . . . » 4
2.2.2 Run AHT-3-11 . . & & v v v v e ¢ 4 o o o o + o & o s 4
2.2.3 Run AHT-3~12 . & ¢ v ¢ o v 4 & 4 4 o o o o s o o o 5
2 o 2 2 4 RUH AI'I'I‘"B“.]_B . Py . - - . » . . . - - a - » . . . . . 5
2 ,2-5 RUH AHT‘“B”IA » . . . . . . . . - . . » - . . . . . . 5
2.206 Rlll'l AHT"“B’lS » a . - » - . - . . . . . . » . « - . - 6
2 . 2 - 7 RUII AHT*3"16 » . . ° » . . . . . . . . . . . - . . . 6
2 - 2 » 8 RUII AHT_3M17 . . . . . . . . . . . . . . . . . . . . 6
2.2.9 Run AHT=3-18 . © ¢ ¢ v« 4 v o 4 4 o o o« o s o o o 7
2.2.10 Run AUHT=-3=19. . . o 4 & v v 4 v o o 4 o s o o s 4 7
2-2.11 R'U.II AHT‘"B”ZO. - . . - . . . . . . s . . . . . . . 7
2-2-12 RUI‘l AHT'—'B—Z].- . . . . . - . . » . . . . . . a . . 8
2.3 Discussion, + v v o 4 4« e 4 e e 4 e e e e e e e e e e 8
3. DEVELOPMENT OF THE METAL TRANSFER PROCESS. . ¢ « + ¢ o v o« o + o 9
3.1 CExamination of MTE-3 Equipment and Materials. . . « . . . . 9
3.2 Status of Metal Transfer Experiment MTE-3B. . . + « « + + o 12
L, SALT-METAL CONTACTOR DEVELOPMENT: EXPERTMENTS WITH A MECHANICALLY
AGITATED, NONDISPERSING CONTACTOR IN THE SALT~BISMUTH FLOWTHROUGH
FACILITY . . *» . . . . ® . = . L4 . . . - . . . . - . - . - » . 14
4.1 Preparation for Mass Transfer Experiment TSMC~7 . . . . . . 14
4.,1.1 Addition of beryliium to the system. « « « « « +» « .« 14
4,1.2 Prerun equilibration of salt and bismuth . . . . . . 14
4.2 Mass Transfer Experiment TSMC-4 . . ¢ « . « v v & « o « o« . 15
4.3 Experimental Results. ¢ « & ¢« ¢ & v ¢ 4 e o 4 « 4« « « « + « 15
- 5. SALT-METAL CONTACTOR DEVELOPMENT: EXPERIMENTS WITH A MECHANICALLY
AGITATED, NONDISPERSING CONTACTOR USING WATER AND MERCURY. . . . 17

9.1 THhEeOTY. v « o 4 o o o & o o « 2 s & o 5 o 4 o o o 2 « « « . .19

5.2 Experimental Results. « . . « . . « ¢ ¢ ¢ o v o o o 4 v « o« 23
Cn

iv

CONTENTS (Continued)

FUEL, RECONSTITUTION DEVELOPMENT: INSTALLATION OF EQUIPMENT
FOR A FUEL RECONSTITUTION ENGINEERING EXPERIMENT.

6.1 Equipment Documentation.

6.2 UY¥_ Generation

6
6.3 Instrumentation and Controls . . . . . . . .
REFERENCES.

Page
SUMMARTES

CONTINUOUS FLUORINATOR DEVELOPMENT: AUTORESISTANCE
HEATING TEST AHT-3

Twelve additional autoresistance heating runs were made with the AQT-3
equipment. Alr-watey cooling coils were installed on the test sections
of the test vessel to eliminate the need to remove insulation during each
run. Seven runs were made using both vertical and side~arm electrode
test sections, but steady temperature and resistance condifions were not
maintained for any appreciable time. Five runs using an electrode at the
top of the vertical test section and with the side-arm test section in a
frozen position were more successful; temperature control was better, and
it was found that steady conditions could be maintained at a much lower
salt resistance than previously believed possible. Apparently, the resis-
tivity of the salt being used was lower than literaturs values {foadicated

by a factor of 9 to 10, probably because of impurities.
: DEVELOPMENT OF THE METAL TRANSFER PROCESS

We have completed the installation of all equipment for the wmetal
transfer experiment MTE-3B in which we will continue to study the steps
in the metal transfer process for removing rare-earth fission products
from breeder reactor fuel salt. Necessary preoperational checkout of the

system is under way before the salts and bismuth will be charged.

During this report period, additional sawmples of the salt and bismuth
phases (at the three interfaces) from previously operated experiment MTE-3
were analyzed by X-ray diffraction to identify any interfacial impurities.
A high concentration of Th02 was found in the LiCl at the Bi~Li interface
in the stripper vessel. No oxides were detected at the LiCl--Bi-Th and

fluoride salt--Bi~Th interfaces in the contactor.
vi

SALT~-METAL CONTACTOR DEVELOPMENT: EXPERIMENTS WITH A
MECHANTCALLY AGITATED, NONDISPERSING CONTACTOR
IN THE SALT-BISMUTH FLOWTHROUGH FACTLTTY

The seventh tracer run, TSMC~7, has been completed in the mild steel
contactor installed in the salt-bismuth flowthrough facility in Building
3592. Prior to the run, approximately 1.5 g~equiv of beryllium
was added electrolytically to the salt phase to establish a uranium dis-
tribution coefficient of ~ 100. The salt aod bismuth phases were passed
through the contactor to ensure that chemical equilibrium was achieved

between the salt and bismuth.

237
Mass transfer experiment TSMC-7 was performed after 1 mg of 3 U308

and 11 mg of Mg0 were added to the salt in the salt feed tank. Salt and
bismuth flow rates were 152 and 170 cc/min, respectively, with an agitator

speed of 68 rpm.

Results from the flowing stream samples taken during the run indicate
that the salt-phase mass transfer coefficient was 0.0057 + 0.0012 cm/sec.

This corresponds to 657 of the wvalue predicted by the Lewis correlation.

SALT-METAL CONTACTOR DEVELOPMENT: EXPERTMENTS WITH A
MECHANTCALLY AGITATED, NONDLSPERSTING CONTACTOR
USING WATER AND MERCURY

Data from a series of five experiments performed in the water-mercury
contactor have been reanalyzed in an attempt to determine if the apparent
change in mass transfer coefficient during the execution of a run was due
to the controlling resistance to mass transfer changing from one phase to

the other.

A model was developed which assumed the reaction under consideration,

pp >t [1,0] + za[tg] > 7t [4,0] + PblHg] ,

to be instantaneous, irreversible, and occurring entirely at the water-—
mercury interface. The possibility that the control of mass transfer
switched from one phase to the other during a run was also considered in

developing the model.
vii

The model described above was applied to the data obtained from five
experiments. Several inconsistencies were found between the model and
the experimental data. We concluded that this model does not adequately

represent the system, and that further work is necessary in this area.

FUEL RECONSTITUTION DEVELOPMENT: INSTALLATION OF EQUIPMENT
FOR A FUEL RECONSTITUTION ENGINEERING EXPERIMENT

Equipment is described for absorbing UF6 gas into a flowing salt
stream containing 'UF4 and reducing the resultant UF5 to UF/ by hydrogena-
}

tion. Tnstallation of the equipment is under way.
1. INTRODUCTION

A molten-salt breeder reactor (MSBR) will be fueled with a molten
fluoride mixture that will circulate through the blanket and core regions
of the reactor and through the primary heat exchangers. We are developing
processing methods for use in a close-coupled facility for vemoving fis-
sion products, corrosion products, and fissile materials from the molten

fluoride mixture.

Several operations associated with MSBR processing are under study.
The remaining parts of this report discuss:
(1) experiments conducted in a simulated continuous fluorina-
tor for studying autoresistance heating in molten salt

and formation of frozen salt films on the fluorinator
walls;

(2) results of dnspection of equipment used in metal transfer
experiment MTE-3 for demonstrating the metal transfer
process for removal of rare earths from MSBR fuel carvier
salt;

. 2]
(3) measurements of mass transfer of 37U and 972r from MSBR

fuel carrier salt to molten bismuth in a mechanically
agitated contactor;

(4) measurements of mass transfer of lead and zinc between
aqueous solutions and mercury amalgams using a
mechanically agitated contactor; and

(5) design of experimental equipment to be used in engineering
studies of fuel reconstitution.’
This work was performed in the Chemical Technology Division during

the period September through December 1974.

2. CONTINUOUS FLUORINATOR DEVELOPMENT:
AUTORESISTANCE HEATING TEST AHT-3

R. B. Lindauer

A drain line was installed from the side arm to the wvertical section
of the test vessel in order to remove the salt heel from the side arm.
This was done to facilitate removal of the electrode section which had
melted off during autoresistance heating test run AHT-3-9; it was then

possible to resume autoresistance heating test No. 3 (AHT-3).
Air-water cooling coils were installed on the side arm and vertical
test sections of the test vessel to eliminate the need to remove (or
loosen) insulation during each run. The cooling coils were placed on
the test section in a coaxial position because of the arrangement of
existing heaters. Temperature control could be improved by using heating

and cooling zones which are transverse to the axis ol the test section.

Seven runs were attempted using the entire test vessel (AHT-3-10 to
~16). The best operation was attained in run AHT-3-15, but the resistance
decreased after about an hour of autoresistance heating. The maximum
power reached was 1130 W. The only temperature point above 350°C at this
time was at the junction of the electrode side arm and the vertical test

section.

An electrode was installed at the top of the vertical test section,
and five more ruas were made with the side arm frozen. Satisfactory
operation was achieved, but with an unusually low salt resistance (0.09
to 0.12 ). This was only about 10% of the resistance found during good
operation with LizBeF4. All attempts to transfer salt from the test ves-
sel, which had frozen salt on the wall, were unsuccessful. A different
autoresistance power supply with a higher current capacity was used in
the last two runs, since large currents (100 A) were being used at these
low resistances. It was found that steady conditions could be maintained
with a resistance of 0.095 @ using 110 A. It is bhelieved that both the

low resistance and the salt transfer difficulty are associated with crud

and impurities in the salt.

2.1 Experimental Equipment and Procedure

The equipment and flowsheet were essentially the same as described
previously.l A 1/2~in.-0D nickel drain line was installed from the low
point of the electrode side arm to the side of the 6-~in. vertical section
of the test vessel, 9 in. above the botltow. The line was installed to
permit draining the salt heel from the side arm so that the section of
electrode which melted off in the last experimentl could be recovered.
During normal operation this line will not be heated so that a salt plug

will form.
A new electrode was fabricated consisting of a 3/4-in. sched 40
nickel pipe with a 5-in.-long section which extended up the center of
the glanting portion of the electrode side arm. This electrode is there-
fore similar to the one which will be used in the next experiment, AHT-4,
in which the circulating salt will enter the test vessel through the

electrode.

Another change made to the equipment was the installation of cooling
coils on the electrode side arm and on the vertical test section. These
coils were strapped on the 6-in. pipe between the Calrod heating elements.
There were four separate cooling circuits. Two coils {top and bottom)
were installed on the slanting section of the side arm, connected in
parallel, and had the same air-water supply. Four coils (north, east,
south, and west) were installed on the vertical test section. These were
connected with three separate cooling supplies. A very low (less than 50
cmB/min) water flow was metered into the cooling air to increase the heat
removal capacity. Exit temperatures of each coil were monitored, and the
water flow rate was adjusted to the maximum which could be vaporized.
This was done by maintaining the exit cooling temperature slightly above
100°C. This cooling system eliminated the need to remove insulation from
the test section in order to form the frozen salt film. The cooling rate

was also increased.

The vent line in the center of the top flange of the 8-in.-diam
disengaging section was moved to allow this access nozzle to be used for
both an insulated thermocouple during runs AHT-3-10 to -16, and for the
autoresistance electrode during runs AHT-3-17 to -21. 1In the last five
runs, a 1/4-in.-diam nickel rod was used as the electrode with the rod
submerged 4 in. into the salt. The length of the test section was approxi-
mately 24 in., compared with 44 in. for the side-arm electrode. These
lengths assume a current path to the wall just below the gas inlet side

arme.

Operation of the system was similar to that previously described. The
entire test vessel was heated to between 500° and 550°C. Heat was then
turned off the test section, and cooling air and water were turned on. When

the wall temperature of the test section reached 350°C (about 2 hr), the
autoresistance power was inctreased stepwise until the electrical resis-
tance between the electrode and the molten salt at the bottom of the test
vessel reached a constant value. The autoresistance power was turned on
at a very low power (less than 10 W) at the start of cooling so that the
resistance could be calculated continuously from the voltage and current.
An instrument was installed to perform this division electrically, and
the quotient was recorded continuously on a 0 te 10 mV recorder. When
tne resistance had lecveled off, the cooling was adjusted by adjusting

the water and air flows. The voltage to the autoresistance circuit also

required adjusting to maintain a constant frozen film.

2.2 Experimental Results

2.2.1 Run AHT-3-10

test section. Insulation was loosened to cool the vertical test scctiom.
The electrical resistance through the salt was calculated periodically
from the voltage and current. Sufficient water was added to the cooling
aly to keep the exit air temperature near 100°C. When the test section
temperatures reached 350°C the autoresistance power was left on. The
resistance remained fairly constant at about 1.7 § for 30 minn as the power
was increased to 2.1 kW (60 V and 35 A). At this point, the current rose
sharply and the resistance dropped. The temperature at the top of the
electrode side arm vose 20°, indicaring an electrical short at this point.
After the hot spot was cooled for 15 min, autoresistance heating was
resumed. The resistance slowly decreased however, and sufficient power
could not be introduced to keep the vertical test section from freezing.
After run AHT-3-12, it was discovered that one of the temperature
recorders was reading 50°C too high, and the run had been started with

too low a temperature at the bottom of the test vessel.

2.2.2 Run AHT-3-11

P e S S

Conling coils were installed on the vertical test section before this
run. The system was heated to a higher temperature than normal to check

out the new coolers. The average cooling rate was increased from 55 to
85°/hr, but there was time for only 20 min of autoresistaoce heating, not
enough to reach steady conditions. The resistance dropped from 0.2 to
0.1 @ during this period. Again, because of the incorrect temperature
recorder, the bottom of the test vessel was allowed to drop below the

liquidus temperature (505°C).

2.2.3 Run AHT-3-12

Cooling was started with the test vessel aft a higher temperature
than normal--the test sections were about 600°C. A good cooling rate of
about 98°C/hr was achieved, and the autoresistance power was started
early in the day. Current was increased stepwise from 20 to 30 to 40 A,
The resistance started to drop at 40 A, although a maximum power of only
480 W was being used. The resistance was 0.25 §i. Since the resistivity
of the salt was believed Lo be 0.75 {-cm, this indicated a very thin
film. The power was reduced in an attempt to vaise the resistance, but

the test section froze over.

2.2.4 RBun AHT=-3-13

Before this run was started, it was discovered that one temperature
recorder was reading 50°C higher than the other. When this was corrected
it was found that the electrode side-arm test section was 50°C hotter
than the wvertical test section. There was insufficient time to balance
the temperatures; by the time the side arm was cooled to 350°C, the verti-
cal test section had cooled to 280°C. Autoresistance power was turned
on, although the resistance was still low, and shorting occurred in the

side arm.

2.2.5 Run AHT-3-14

Cooling of the electrode side arm was begun an hour before the
cooling was started on the vertical test section in an effort to form a
frozen film on the side arm before the vertical section became too cool;
however, the vertical section cooled at a much higher rate. The junction

0of the vertical and side-arm test sections cooled very slowly. By the
time this point reached 350°, the vertical test section had frozen com-
pletely. A recording resistance meter was installed before this run to

provide an immediate indication of resistance changes.

2.2.6 Run AHT-3-15

Autoresistance heating was started when the electrode side-arm test
section reached 350°C, although the vertical test section was still over
400°C. The resistance was unusually high, about 2 ¢, although the
junction of the vertical and side—arm sections was about 440°C. The
power was increased stepwlise with only slight changes in resistance
until 1130 W (38 V) was reached, at which point the resistance suddenly
dropped to 0.7 §. Autoresistance heating of greater than 100 W had been
applied for about 70 min, but temperatures on the test-section walls were

still decreasing rapidly.

2.2.7 Bun AH_I_’::‘_S:} é

An attempt was made to duplicate run AHT-3-15, but with the junction
of the vertical and side~arm test sections below 350°C. Autoresistance
heating was not applied until this temperature was reached. When voltages
approaching 20 V were applied, the resistance dropped from v 2 to ~ 1 Q.

It was still believed that the resistivity of the salt was about 0.75 O-
cm; the resistance was not allowed to get below 0.8 @, corresponding to a
3/4~in. film. TIf, at this time, it had been known that the resistivity

of the salt was much lower, satisfactory operation might have been achieved
with the electrode side arm by applying sufficient power to balance the
heat loss before temperatures became too low and the test section froze

over.,

2.2.8 Run AHT-3-17

The side arm was allowed to freeze completely, and autoresistance
heating was applied from an insulated 1/4-in.-diam nickel rod janserted
through the 8-in.~diam disengagement section above the vertical test scc~

tion. The rod extended 4 in. into the salt. The resistance again
increased to 800 W. This was similar to previous runs with the electrode
side arm. However, during this run, as the resistance dropped, heating
was maintained and relatively stable operation was achieved at a power
level of about 900 W. The resistance decreased slowly, reaching 0.11 @
at the end of the operating periocd. The current capacity of the auto-
resistance power supply (100 A) was nearly reached. Test-section wall

temperatures had leveled off at an average temperature of about 275°C.

2.2.9 Run AHT-3-18

In this run, we not only attempted to duplicate conditions of the
previous run, but also to reach steady conditions early enough in the day
to transfer the molten-salt core in the test section to the feed tank in
order to visually check on the salt~film thickness. Operation was very
similar to run AHT-3~17, with a resistance of 0.12 2 using 850 W of
autoresistance power. We were unable to tramnsfer the salt as planned,
although ten pressure-vacuum cycles were applied to the feed tank. The
test vessel bottom temperature and transfer line temperature indicated a
slight salt movement, and the applied pressure showed a slight indication
on the test vessel level instrument. The salt level did not fall below

the agutoresistance electrode.

2.2.10 Run AHT-3-19

The two previous runs were more or less duplicated in this run. The
test vessel bottom was heated to a higher temperature at the end of the
run to aid in transfer, but this was not successful. As in the two pre-
vious runs, the current capacity of the autoresistance heating suapply was
not sufficient, and wall temperatures continued to decrease at the end

of the run.

2,.2.11 Run AHT-3-20

The autoresistance power supply was increased from 100- to 200~A max
current; however, the new supply had only a 24-V max instead of 100 V.
During this run, the resistance was allowed to increase to 0.8  before

autoresistance heating was applied, and only 600 W of power was available.
This was not sufficient to prevent the molten core from freezing, as was
indicated by the test section liquid-level instrument going off the

scale and an increase in the resistance to > 3 .

2.2.12 Run AHT-3-21

The test-section wall was cooled as rapidly as possible and auto-
resistance heating was started when the salt resistance was 0.37 2. All
wall temperatures were below 350°C at this time. A maximum resistance
of 0.40 @ was reached when the maximum 24 V was applied to the electrode.
The vesistance then decreased and the power rose to a maximum of 2500 W.
The voltage was decreased, and stable operation was reached with a resis-—
tance of about 0,09 & and 1470 W. The wall rfemperatures also leveled out.
The resistance decreased very slowly and voltage and power were conse-
quently lowered to 1160 W. The temperature of the bottom of the test
vessel was being raised at this time to facilitate transfer of the salt
from the test vessel for inspection of the frozen filwm; this was probably
the cause of the decrease in resistance. After the power was decreased,
the resistance slowly increased to 0.095 Q. We were still unable to
transfer salt after repeated attempts, so heating of the entire test
vessel was begun to remove all of the salt to the feed tank. During the
heatup, inspection of the cell revealed a salt leak which was located at
the junction of the salt transfer line and the bottom of the test vessel.

Approximately 6 kg of salt was lost.

2.3 Discussion

Successful operation might have been achieved in most of these runs
if it had been known that the resistivity of the salt was not 0.75 Q-cm,
as stated in the literature,z but was considerably lower. In many cases,
the autoresistance power was reduced when the resistance was in the range
of 0.5 to 1.0 &; this was done to prevent melting the frozen film. Later
operation without the side arm showed that steady operation could be
majintained in the 0.09- to 0.15-0 range, and that the resistivity of the

salt was probably lower than 0.75 Q-cm by a factor of 5 to 10.
Operation with the electrode in the side arm was unsuccessiul for
two reasons. Cooling of both vertical and slanting test sections at the
same rate to avoid freezing one section before a complete film was formed
on the other section was quite difficult with the installed cooling zones.
In addition, the junction of the side arm and vertical section was diffi-
cult to cool, and shortiog occurred at this point until an air jet was
applied. Both of these difficulties should be lessened considerably by
circulating salt through the test vessel. This would be expected to pro~
mote a more uniform temperature over the entire test section. Conse-
quently, we have started designifig a system which will allow salt to be

circulated through the test vessel.

3. DEVELOPMENT OF THE METAL TRANSFER PROCESS

H. C. Savage

Engineering experiments to study the steps in the metal transfer
process for removing rare-earth fission products from molten-salt breeder
reactoy fuel salt will be continued in new process vessels which dupli-
cate those used in a previous experiment, MTE~3.3 Installation of the
equipment for the new experiment, degignated MTE~2B, was completed during
this report period. Experiment MTE-3B utilizes mechanically agitated
contactorsé to achieve effective mass~transfer rates of the rare-earth
fission products between the salt and metal phases in the metal transfer

process {(as was done in ezperiment MTE-3).

3.] Examination of MTE-3 Equipment and Materials

We have previously reported the results of analyses of the salt and
bismuth phases from experiment MTE*-ZLb These analyses were made in an
attempt to determine whether or not the Jlower-than~expected mass transfer
coefficients observed in experiment MTEwBB were due Lo the presence of
films (composed of solids) at the salt-bismuth interfaces formed by entry
of impurities, such as oxides, into the system. These analyses, which
included chemical analyses, metallographic examivation, electron beam

scanning, and X-ray fluorescence avalyses, indicated the presence of
& 3
10

significant concentrations of iron (up to 3500 ppm) and thorium at the

interfaces.

Visual examination of the 2-in.-diam plugs removed from the three
interfaces (fluoride salt—--Bi~Th, LiCl--Bi-Th, and LiCl--Bi-Li) indicated
the presence of a layer of material ~ 1/32-in, thick at the interface
between the LiCl and Bi-Li in the stripper vessel. No foreign material
was seen at the interfacial areas between the fluoride salt—--Bi-Th and

LiCt1-~Bi-Th in the contactor.

During this report period, samples of material were removed from the
vicinity (within 1/32 in.) of each of the three salt-metal interfaces in
the MTE-3 equipment and analyzed by X-ray diffraction. The results of
the six analyses are given in Table 1. The only oxides in the system
were at the interface between LiCl and the Li-Bi stripper alloy in the

stripper vessel. A high concentration of ThO. was found in the TiCl at

2
this interface. Although no thorium should have been present in the
stripper during most of the operating time, thorium could be expected to
combine with any oxides present when the fluoride salt (containing

thorium) was entrained into the chloride salt.

No oxides were detected at the LiCl--Bi-Th interface in the contactor.
The analysis of the mass transfer rates observed during this experiment
suggested that some hindrance of mass transfer was occurring at this
interface, possibly caused by oxide films. Analyses of the material near
this interface, however, do not support the theory ihat oxide films at

the LiCl--Bi-Th interface slowed down the mass transfer.

The samples taken for X-ray diffraction analysis were extremely small
(about 2 to 3 mm3 of material was removed from the interface for each
sample), and effects of segregation during freezing will make interpre-
tation of the results of these analyses very difficult. For instance,
it is interesting to note that no LiCl was detected in the sample of LiCl

taken from the interface in the stripper vessel.

Bismuth was detected in each of the salt samples analyzed by X-ray
diffraction. 1t is likely that during freezing, bismuth was forced into
the already frozen salt and was not present in the salt when both phases

were molten.
11

Table 1. Results of X-ray diffraction analyses of material removed
from the vicinity of salt-metal interfaces in experiment MTE-3

 

 

A st = e e

Identified Estimated composition
Phase material (mole %)%
1iCl in stripper Bi 25-50
L13ihf7 10-30
Th02 40-80
unidentified solid -
solution
Li-Bi alloy in stripper Bi 75-95
BeF2 2-10
Bi—-Th in contact with Bi 90
LiCl in contactor
LiCl in contactor Bi 30-70
LiCl 30-70
LlBth7 10-30
Fluoride salt in con- Bi 20-40
tactor Li.ThF 40-80
3 7
L17Th6}31 5-15
Bi~-Th in contact with Bi. 60-90
fluoride salt in con- Th (unidentified L
tactor
compound)
Fe (no X~ray data trace

available)

 

a . . -
X-ray diffraction can only detect > %~ 5 mole Z.

It is to be noted that iron and/or iron oxides were not found by X-

ray diffraction, since the amounts present (based on chemical analyses of

similar samples) are below the limit of detectability by X-ray diffraction

(v 5 mole %).

An additional chemical analysis was made of the 1/32-in.-thick layer

of material found between the LiCl and the Bi~Li in the stripper vessel.
This material had a different stiucture and a dark gray or black appear-
ance. The followiag resulis were veported (wt 7%):

Bi - 51.24 ' — unable to analyze

Th ~ 3.7 Fe - 14.0

Li - 5.0 O, - unable to analyze due to high B8i

Bi - 0.23 Cl ~ 394 ppm

The very high ircn concentration (14 wt %) is unexpected. Previous anal-
yses of other samples of similar appearing material indicated irou concen-

trations of about 0.25 wt 7.

It is not possible at this time to draw firm conclusions about the
relationship of these observations to the low mass trvansfer rates seen
in experiment MTE~3. The transfer of fluovide salt into the chloride salt
just prior to shutdown, and the length of time between shutdown and
inspection (from February 1973 tu February 1974), introduce much uncer-
i~
L

tainty in an accurate interpretation of the analyses. We plan to closely

J

monitor the salt and metal phases duvrinyg operaiion of exveriment MTE~
L

L

o

B

<

for buildup of impuritics (iron and oxides). Sampling phases at or uear

cach interface will be atrempted.

3.2 Starus of Metal Transier bxperiment MTE—-3B

Installation of the new process vessels and squipment for metal trans-
fer experiment MTE-3B was completed during this report period. Figure 1
is a photograph of the process vessels afier electric heating elements and
thermocouples were installed. We have completed the process piping
installation, thermocouple and heater hookup, and the calibration of the
temperature recorders and controllers, pressure gages, selecited thermo-
couples, and all gas and cooling-water rotameters, The experiment is now
ready for preoperational checkout, leak testing of the vessels and piping,
pressure tests at the operating temperatuve of 650°C, and hydrogen treat-
ment of the interlor of the carbon stee! process vessels to remove oxide

impurities before charging salts and bisauth.
 
     
  
   

s
A S A S

e P

—

Pig. L.
ment MTE-3B.

L

 

— a8 PHOTO NO. 2706-74A |
|

.

Ko ,SALT—METALi
/ i CcONTACTOR °

Photograph of processing vessels for metal-transfer experi-
14

4. SALT-METAL CONTACTOR DEVELOPMENT: EXPERIMENTS WITH A
MECHANICALLY AGITATED, NONDISPERSING CONTACTOR
IN THE SALT-BISMUTH FLOWTHROUGH FACILITY

C. H. Brown, Jr.

We have continued operation of a facility in which mass transfer
rates between molten LiF—BeFZ—ThF4 (72-16-12 mole %) and molten bismuth
can be measured in a mechanically agitated, nondispersing contactor of
the "Lewis" type.6 A total of seven experimental runs have been completed
to date. Results from the first six runs have been reported previously.

Preparation for and results obtained from the seventh run, TSMC-7, are

discussed in the following sections.

4.1 Preparation for Mass Transfer Experiment TSMC-7

Prior to the run, it was necessary to: (1) add beryllium to the salt
to adjust the uranium distribution coefficient, and (2) contact the salt
and bismuth by passing both phases through the mild steel contactor to
ensure that chemical equilibrium was achieved between the salt and bis-

muth.

4.1.1 Addition of beryllium to the system

 

As discussed previously,7 it is desirable to maintain the uranium
distribution coefficient at a relatively high level (> 30), so that the
actual value of the distribution coefficient will not affect the calcu-
lation of the overall mass transfer coefficient. Approximately 1.37
g-equiv of beryllium was added electrolytically to the salt phase
in the treatment vessel (T5), which raised the uranium distribution

ratio to > 97.

4.1.2 Prerun equilibration of salt and bismuth

237 .
A U308 tracer was utilized to measure mass transfer rates across

the salt—bismuth interface in the stirred interface contactor while the
system was otherwise at chemical equilibrium. In order to ensure chemical

equilibrium between the salt and bismuth, both phases were passed through
15

the contactor prior to run TSMC-7. Three attempis at phase equilibration

runs were necessary before a satisfactory flowthrough was achieved.

During the first two attempts, a leak developed in the transfer line
from the bismuth feed tank (T1l) toe the contactor. This transfer line was
completely replaced, along with the associated Calrods and thermal insu-
lation. The third attempt at phase equilibration was successful with
salt and bismuth flow rates of 152 cc/min and 170 ce/min, respectively.

The agitator was operated at ~ 100 rpm.

4.2 Mass Transfer Experiment TSMC-4

After the phase equilibration described above was accomplished, the
salt and bismuth were transferred to their respective feed tanks.

237

An 11-mg quantity of tracer, consisting of 1 mg of U,0, and 10 mg

378

of MgO which had been irradiated in the ORR for 72 hr, was placed in a
steel addition vessel. It was then inserted in the salt-feed tank where
it was sparged with ~ 0.5 scfm of argon for 1 hr to facilitate dissolu~

tion of the tracer in the salt phase.

The volumetric flow rates of salt and bismuth to the contactor were
set at 152 cc/min and 170 cc/min, respectively, by controlled pressuri-
zation of the feed vessels (T3 and Tl). The stirrer rate was set at 68
rpm for the run. Seven sets of samples were taken of the salt and bis~

muth effluent streams from the contactor.

4.3 Experimental Results

The samples taken during the run were analyzed by first counting the
2 -
37U (207.95 keV B ). The material

in the sample capsules was then dissolved, and the activity of 237U was

sample capsules for the activity of

counted again. The counting data obtained in this manner are shown in

Table 2.

The overall salt-phase mass transfer coefficient was calculated using
three different equations derived from an overall mass balance around the

7
contactor. The average measured mass transfer coefficient, with the
Table 2. Counting data obtained from run TSMC-7

 

Solid analysis Solution analysis Solid analiysis Solution analysis
Sample for 2370 for 237y Samplie for 237y For 237g
code? {counts/qg} {counts/qg) coded (counts/qg) {counts/qg)

 

Samples taken prior to run

 

 

 

 

 

313-B-5 2.68 x 107 6.39 x lo* 311-5-5 < 4.2 x 107 < 1.8 x 10Y
314-B--_ 2.11 x 107 6.77 x 1ot 312-5-5 < 5.2 x 107 < 1.5 x 10"
317-B-1 2.25 x 10% 6.79 x 10" 315-5-3 < 6.6 x 10° < i.8 w10
318-B-1 2.57 x 10% 6.47 x 10" 316-3-3 < 7.6 x 107 < 9.5 x 103
Samplies taken prior to run but after addition of tracers
319-5-3 3.95 x 10° 5.46 x 107
320-5-3 3.95 x 10° 5.61 x 105
Samples taken during run
321-B-FS 1.20 x 10° 3.91 x 10° 329-S-F§ 2.21 x 10% 2.89 x 10°
322-3-F8 1.36 x 107 4.33 x 10° 330-5-58 2.85 x 10% 3.32 x 10°
323-B-¥s 1,47 x 10° 4.81 x i0° 331-8-FS 2.75 x 10° 3.26 x 10°
324-B-FS 1.81 x 107 5.03 x 10° 332-5-75 2.89 x 10° 3.97 x 10°
325-B-FS 1.62 x 10° 5.06 x 10° 333-5-7g 2.82 x 108 2.96 x 10°
326-B-FS 1.79 x 10° 5.05 x 10° 334-5-FS 2.93 x 10° 3.90 x 106
327-B-FS 1.78 x 10° 5.17 x 10°
sampies taken after run

335-B-1 8.65 x 10* 2.22 x 10° 33¢-5-3 -- -~
336-B-1 6.66 x L0Ob 2.18 x 10° 340-5-23 -- -
337-B-2 1.24 x 10° 3.78 x 10° 341-5-4 -- -—
338-B-2 1.34 x 107 3.86 x 10° 342-5-4 1.64 x 108 2.19 x 10°
343-B-5 i.99 x 10° 5.05 x 10° 345-5-5 < 9.8 x 107 < 8.7 x 103
344-B-5 1.84 x 107 5.58 x 10° 346-5-5 < 5.3 x 107 < 1.7 x 1%
Each sample 1is cesignated by a code corresponding to A-B-C, where A = sample number; B = material in

sample (B = bismuth, S = salt): and C = sample origin: 1 =71, 2 =72, 3 ="T3, 4 = T4, 5 = T, PS8 =
flowing stream sample.

o1
17

corresponding standard deviation, is 0.0057 + 0.0012 cm/sec. This value
. —_— . 6

corresponds to 657 of the value predicted by the Lewis correlation. A

Lewis plot of the results for this run, along with the results from cuns

TSMC~3 through -6, are shown in Fig. 2. The nomenclature used in Fig. 2

is:
k = individual phase mass transfer coefficient, cm/sec,
. . . . 2
v = kinematic viscosity, ¢m /sec,
2 . .
Re = Reynolds number (ND /v), dimensionless,
D = stirrer diameter, cm,

N = stirrer rate, 1/sec,

subscripts 1, 2 = phase being considered.

In Fig. 2 it can be seen that the mass fransfer group based on uranium
is 45 + 13% of the Lewis correlation for runs 3, 5, and 7, while that same
group is 103 + 47 of the Lewis correlation for runs 4 and 6. The mass
transfer results based on zirconium are consistently slightly less than
the values based on uranium. 7This discrepancy is probably related to the
inability to correct for the self absorption of the 743.37 keV B~ from the

97Zr in the s0lid bismuth samples.

We believe that at a stirrer speed bétween 160 and 180 rpm, the ini-
tiation of entrainment of salt into the bismuth begins to cccur; the
apparent increase in mass transfer coefficient is a manifestation of an
increase in the surface area for mass transfer due to surface motion.
Experiments with water—-mercury and water-methylene bromide systems support

this belief.

5. SALT-METAL CONTACTOR DEVELOPMENT: EXPERIMENTS WITH A
MECHANICALLY AGITATED, NONDISPERSING CONTACTOR
USING WATER AND MERCURY

C. H. Brown, Jr.

The reference flowsheet for the proposed MSBR processing plant calls
for the extraction of rare earths from the fluoride fuel carrier salt to
an intermediate bismuth stream. One device being considered for performing

this extraction is a mechanically agitated, nondispersing contactor in
18

ORNL-DWG 75-15156

 

100

- [ T T T T T TTT7

- /g‘rsmcw -

- TSMC-6/ & rcric-a .

i g TSMC-6 -
LEWIS / ®TSMCS

CORRELATION—

L \7’ @TSMC-3 =
/ &TSMC-5
/
N 10 - / -
- / ®TSMC-7 ~

60K,

/ ®2°7y

 

 

/i | i | L1 L 1tif

|
OO0 10000 10000
Re|+72/% Rep

 

Fig. 2. Comparison of mass transfer coefficients measured in the
salt-bismutlhh contactor using 2379 and ?/7r tracers with values predicted
by the lLewis correlation.
19

which bismuth and fluoride salt phases are agitated to enhance the mass
transfer rate of rare earths across the salt-bismuth interface. Previous
reports ~10 have shown that the following reaction in the water-mercury
system is suitable for simulating and studying mass transfer rates in
systems with high density differences:

i

24
P77 [H,0] + zaltg] > zn” [H,0] + Pb[ig] . (1)

i 8 .
A large amount of data have been reported £for the water-mercury system
in which it was assumed that the limiting resistance to mass transfer
existed entirely in the mercury phase, as suggested by literature corrve-

lations.

An experiment was previously describedll that was designed to iden-
tify the phase in which the controlling resistance to mass transfer
resided. In these experiments, it was noticed that the mass transfer
coefficient appeared to vary during the course of a run if the initial
concentrations of Zn in the mercury phase and Pb?’+ in the water phase
were not equal. The mass transfer coefficient would abruptly change values
after the experiment had been in progress for about 12 to 18 min. During
this report period, the data from the experiments that had been performed
in the water-mercury contactor were reanalyzed in an atfempt to determine
if the apparent change in mass transfer coefficient was due to the con-
trolling resistance changing from one phase to the other during the course

of an experiment.

5.1 Theory

. 11 .
As was previously reported, the reaction being studied, Eg. (1), is
considered to be a fast, irreversible, ionic reaction which occurs entirely

at the mercury-water interface.

The rate of transport of reactants from the bulk phase to the inter-

face where they react is given by:

N =

1 e
7n° K (Cne,p — €

Zn®,i

) s (2)
20

2+ =
N Ky

) A(CPb2+ — C,.2+ ), (3)

0 , B Pb™ ,i

where

= individual phase mass transfer coefficient, cm/sec,
= rate of mass Cransport to the interface, g/sec,

2

interfacial area, cm ,

= concentracion,

= O B 2w
|

= denotation of bulk phase concentration,
i = decotation of iaterfacial conceatration, g/em”,
Hg,H20 = refers to the phase being considered.

. , . 2+ .
Since one mole of 7n° is equivalent to one mole of Pb according to Eq.

(1), then NPb2+ = NZn°° Substituting Eqs. (2) and (3) into this expres-
sion yields the following equality,

) = k 2+BWC 2+ )

4)
ppt i) (4)

1.0 Cpp

k -0
Hg(CZn°,B Zn°,1i ”

L .
As earlier reported, it the controlling resistance to mass transfer
is assumed to be in the ore phase, the interfacial reactant concentration
in that phase is very swall with respect to the bulk reactant concentra-—

tioa.

In the transient experiment which we bave performed,ll it may be
possible for the limiting resistance to mass transfer to switch from one
phase to the otner during the course of an experiment. Since both inter-
facial reactant concentrations are very small with respect to their bulk
phase concentrations, the following equality exists at this transition

point:

k o o, = K 2+ .
Hg CZH , B kHZO CPb , B (5)
If the controlling resistance to mass transfer is in the aqueous

, 2t
phase, two equations, one for the concentration of Pb and one Ior the

. 2+ .
concentration of Zn , can be derived:
 

'CPb2+(t) kazo A
1n | ===} = — —— t , and
Cpp2+(0) 1.0
- . k.. A
g LPb2+(O) C2n2+(t> _ MV“EEO -
3 C - -
I pp2+(0) | Yi.0
2
where
V = volume of phase, cm’,
t = time, sec, and

parentheses iudicate quantities which are functions of time.

(&)

(7)

[f the mervcury phase contains the controlling resistance to mass

- . - . L, 2t
transfer, two other equations, one for the concentration of Pb

i ., 2 .
for the concentration of Zn~ , ¢an be derived:

 

[ C G) —C +{(0) 4+ C +(t k A
(Jzno(") Pb2+‘(J) P1£32+( ) ~ He —
1o = w‘“”fi"“'t, and
C D(O) V‘,
3 Zn He

 

 

v CZn°(O) *‘Czn2+(t) } kH A |
In c. . (0) Ty t
i 7n° Hy

Equations (6)-~(9) are all of the form,

y = -mt ,
where
y = the natural logarithm of a concentration ratio,
= a mass transfer coefficient times the interfacial area
divided by the phase volume, and
£t = time,

and one

(8)

(9)

(10)

This indicates that a logarithmic plot of the concentration ratio vs time

should yield a straight line, with the slope belng proportional to the

mass transfer coefficient. If the control of mass transfer switches fFrom
22

one phase to the other during the execution of an experiment, then the
slope of the semilog plot mentioned above would indeed change, as was

observed.

Equation (5) can be used with experimental values of k and

He? kH 0’

o

concentrations of an+ and Pb2+, which are evaluated at the time of
apparent change of mass transfer coefficient and at the beginning of the
experiment, to determine the phase in which the limiting mass fransfer
resistance lay at the beginning of the run. The first step in the detor-
mination of the mass transfer limiting phase at the beginning of a run

is to find the pair of individual phase mass transfer coefficients, one
before and one after the transition point, which satisfy Eq. (5), with
the bulk reactant concentrations evaluated at the transition point. Two
combinations of mass transfer coefficients are possible, giving Eq. (5)

the following two forms:

(1) . (2)
ng C oo = kH 0 CPBZT , and (11)
tr 2 tr
(2) (L
k C. o = k C. . 24 . (12)
Hg Z1i e H20 Pb r

where superscripts 1 and 2 refer to the slope in Eq. (10) before and after
the transition, respectively, and the subscript tr refers to the transi-

2
tion point. Equation (6) or (7) is used to determine k (1) and k(”),
1.0 H,0

(1) and k(2) 2 2

and Eq. (8) or (9) is used to determine ng’ He

Once it has been determined which of the two equations, FEq. (11) or
(12), is satisfied for a particular run, the mass transfer limiting phase
at the beginning of the run may be determined by further application of an
inequality similar to Eq. (5). The individual phase mass transfer coef-
ficient in the limiting phase times the bulk reactant concentration in
that same phase must be less than the same product in the nonlimiting
phase. Therefore, if Eq. (11) is satisfied, then the direction of the

following inequality determines the initial mass transfer limiting phase,
23

> . (2) .
CZn°(O) - ]‘1{20 CPb2+(O) . (13)

(1)
kHr
Similarly, if Eq. (12) is satisfied, then the limiting phase is determined

by the direction of the inequality:

(2) > (1) \
ng €, ne(0) < kHZO Cop2+(0) (14)

5.2 Experimental Results

Five experiments were run in the water-mercury contactor to determine
if the apparent change in mass transfer coefficient was due to controlling
resistance changing from one phase to the other during the course of an
experiment., The initial mercury~phase zinc concentration was held
constant at 0.1 M. Phase volumes and agitator speed were also held

constant at 1.8 liters and v 150 rpm, respectively.

The experimental data obtained from these five runs, the conditions
under which the runs were made, and the transient response of the system
in terms of aqueous lead and zinc concentrations are shown in Table 3.

The data were analyzed by use of the equations in the previous section.

Results of this analysis for determination of the phase which con-
tained the limiting resistance to mass transfer at the beginning of each
run are shown in Table 4, along with the measured value of the mass trans-
fer coefficient in the limiting phase. These results indicate that the
limiting resistance to mass transfer was initially in the mercury phase in
the runs made without initial aqueous-phase lead concentration eaqual to or
less than the initial mercury-phase zinc concentration. For the runs made
with an initial lead ion concentration greater than the initial zinc
amalgam concentration, the aqueous phase controls mass transfer. The
results from run 148 (in which the initial reactant concentrations are
nearly equal) are difficult to analyze, because only a slight change of

slope in a plot of Eq. (10) was detected.

Tn runs 145 and 149, the initial lead ion concentration was low with

respect to the zinc amalgam concentration. This would indicate that the
24

Table 3. Concentration vs time data from the water-mercury contactor

Initial Hg-Zn analgam concentration = 0.1 M
Stirrer velocity = 150 rpm
Phase volumes = 1,8 liters each
Paddle diameter = 7.62 cm
Paddle height = 1.91 cm

 

Run Elapsed time Lead ion Zinc ion
numbe {min} concentration (M) concenitration (M)
145 0 0.0197 0.0003
3 0.0171 0.0038
6 0.0141 0.0070
9 ¢.0112 0.0095
12 0.0092 0.0115
15 0.0071 0.0134
18 G.0053 0.0158
21 0.0042 0.0170
24 0.0033 0.0181
27 0.0025 0.0186
30 G. 0019 0.0196
146 0 0.141 0.00021
3 0.131 0.0187
6 0.117 0.034¢6
5 0.101 0.0478
12 0.0883 0.0604
15 0.0738 0.0750
i8 0.0589 0.0826
21 0.0589 0.0887
24 0.0550 0.0210
27 0.0526 0.0948
30 ¢.0512 0.0979
147 0 0.184 0.00104
3.33 0.171 0.0338
6 0.144 0.0520
9 0.132 0.0678
12 0.121 0.0811
15 0.109 0.0872
18 0.103 0.0%02
21 0.101 0.0941
24 0.0999 0.0979
27 0.0975 0.0971
30 0.0956 0.100
148 o 0.0960 0.0007
5 0.0749 0.0199
10 0.05h68 0.0375
15 0.0419 0.0516
20 0.0331 0.0631
25 0.0249 0.0704
30 0.0176 0.0784
35 0.0148 0.0834
40.33 0.0102 0.0899
45 0.0670 0.0906
50 0.0046 0.0922
55 0.0029 0.0922
60 0.0017 0.0948
1499 0 0.0192 0.00045
3.17 0.0156 0.00512
6.07 0.0122 0.00795
9.25 0.0970 0.0106
12 0.0741 0.0130
15 0.00536 0.0148
18 0.00384 0.0162
21.08 0.00249 0.0177
24 0.00211 0.0188
27 0.00127 0.0190
30 0.00069 0.0199

 
25

Table 4. Initial lead ion concentration and mass transfer
coefficients in the phase initially containing the
limiting resistance to mass transfer for runs 145 to 149

 

Mass transfer coefficients in the

 

 

 

Initial lead phase controlling mass transfer

Run ion concentration initially (cm/sec)
number (_M;) ng kH? 0

145 0.0197 0.0012
149 0.0192 0.0013
148 0.0960 0.0056
146 0.141 0.0061
147 0.184 0.0046

 

aqueous phase should initially contain the limiting resistance, since the
initial lead dion concentration times the mass transfer coefficient should
be less than the initial zinc amalgam concentration times the mass trans-—
fer coefficient in that phase. Similarly, in runs 146 and 147, the ini-
tial lead ion concentration was large with respect to the zinc amalgam
concentration; this would indicate that the mercury phase should initially

contain the limiting resistance to mass transfer.

According to the model we have developed, if the aqueous phase ini-
tially contains the limiting resistance to mass transfer, the following

inequality is satisfied:

kyy Cgno(0) > kHZO Cpy24(0) (15)

During the course of an experimental run, the inequality shown in Eq. (15)
should always be satisfied, since the lead ion concentration decreases
with an equal decrease in the zinc amalgam concentration. This indicates
that the aqueous phase should control mass transfer throughout the run;
no change of slope, corresponding to a change in the controlling phase,
should be denoted in a plot of Eq. (6). Figure 3 shows the parameters of

Eq. (6) plotted for the data obtained in run 146, A definite change of
26

ORNL DWG 75-4970Ri

 

(.2 —

14—

1.0

08 -

 

08—

0.7 —

)

(o)

; 0.6 |—

/ot i)
Cpp

_gnK

0.5

0.4 —

0.3 —

0.2 —

Ol i—

 

 

0.0 | 1 1 { | | ] ] 1 | |
O 3 6 3 {2 15 18 24 24 27 30

 

TIME (min)

Fig. 3. Logarithm of the concentration ratio vs time for ruan 146
in the water—-mercury contactor.
27

slope is recognizable at a value of the abscissa (time) between 15 and 13
min, which would indicate a change in the phase containing the limiting

resistance to mass transfer.

Mass transfer theory predicts that the value of the individual phase
mass transfer coefficient is constant. The experimental results given in
Table 4 indicate that the measured mass transfer coefficient for the
mercury phase in runs 145, 148, and 149 increased in direct proportion to
the initial lead ion concentration. Similarly, the values reported for
the aqueous phase mass transfer coefficient in runs 146 and 147 vary
inversely to the initial lead ion concentration. It is evident that for
this system the model we have developed is inadequate, and further studies

need to be made to completely understand the system.

6. FUEL RECONSTITUTION DEVELOPMENT: INSTALLATION OF EQUIPMENT
FOR A TUEL RECONSTITUTION ENGINEERING EXPERIMENT

R. M. Counce

The reference flowsheet for processing the fuel salt from a molten-
salt breeder reactor (MSBR) is based upon removal of uranium by fluorina-

: . . 12 . . .
tion to UFG as the first processing step. The uranium removed in this

step must subsequently be returned to the fuel salt stream before it
returns to the reactor. The method for recombining the uranium with the
fuel carrier salt (reconstituting the fuel salt) is to absorb gaseous

UF6 into a recycled fuel salt stream containing dissolved UF, by utilizing

4
the reaction:

UFg oy + UF,qqy = 20F5qy - (16)

The resultant UF5 would be reduced to UF4 with hydrogen in a separate

vessel according to the reaction:

+ 1/2 H + HF ) (17)
(2)

Vs cay 2¢g) = Faqa)
28

We are beginning engineering studies of the fuel reconstitution step
in order to provide the technology necessary for the design of larger
equipment for recombining UF6 generated in fluorinators in the processing
plant with the processed fuel salt returning to the reactor. During this
report period, equipment that was described previously13 was fabricated
and is being installed. This report contains documentation of equipment

and adds detail to the previous report.

The major components of the fuel reconstitution engineering experi-
ment (FREE) are shown schematically in Fig. 4, and in move detail in Figs.
5 and 6. The equipment for this experiment consists of a 36-liter feed
tank, a UF6 absorption vessel, a H2 reduction column, an effluent stream
sampler, a 36~liter receiver, NaF traps for collecting excess U'F6 and

disposing of HF, gas supplies for argon, hydrogen, and UF and means

63
for analyzing the gas streams from the reaction vessels.

6.1 Equipment Documentation

The feed and receiver tanks are almost identical in construction,
with the exception that the feed tank has an additional nozzle. The
similarity of these vessels is seen in Figs. 7 and 8. The feed tank has
five nozzles, as seen in Fig. 9, as follows:

(1) a salt inlet nozzle (1/2-in. sched 40 pipe with a fitting

for a sleeved 3/8-in. tube that does not extend into the
tank) ;

(2) a-salt exit nozzle (1/2-in. sched 40 pipe with a fitting
for a sleeved 3/8-in. tube that extends to 1/2 in. of
the bottom of the tank);

(3) a sparge nozzle (3/8-in. sched 40 pipe with a fitting for
a sleeved 1/4-in. tube extending to within 1/2 in. of
the bottom of the tank);

(4) an off-gas and pressurization nozzle (3/8-in. sched 40
pipe with a fitting for a sleeved 1/4-in. tube that
does not extend into the tank); and

(5) a sampler port (3/4-~in. pipe with a 3/4~-in. ball valve
and cap).
The return tank has nozzles 1 through 4 of the above, as seen in Fig. 10;

however, the only pressurization capacity is through the sparge line.
29

ORNL DWG 74-1i666R!

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

-
- E ]
E—=
UFS HF UFG
ANALYSIS ! |ANALYSIS HF TRAP
' TRAP o
.
e
‘ BUILDING
OFF~(GAS
Ufg . ;————’ SYSTEM
b g } EFFLUENT
STREAM
* P * SAMPLER
FREEZE ; i
VALVES B
Ar-——-—] !
t : UFG 1
: ABSORPTION 1
i VESSEL :
: :
| !
[ '
FEED TANK RECEIVER
Hy

 

 

 

L

H, REDUCTION
COLUMN

Fig. 4. Schematic flow diagram for the fuel reconstitution engi-

neering experiment (FREE).
ORNL -DWG 7%-37R2

30

 

 

 

-+ N,
- UFg
- Hp
—=~ VACUUM
OFF-GAS
- Ar
Ha OFF-GAS

B
—

 

 

 

 

 

 

 

HF TRAP

RECEIVER
TANK

          

 

 

 

 

 

 

 

 

 

 

 

ABSORPTION

 

 

 

 

VESSEL

Ha ;
REDUCTION |
COLUMM

TAMNK

FEED

EE.

for FRE

i gram

5. Flow disa

Fig.
31

ORNL-DWG 75-35R2

Fi 101 1S A HASTINGS
MASS FLOWMETER
MODIFIED TO GIVE
A 10-50 mA

OUTPUT SIGNAL

-— ARGON

 

SYSTEM

     

CALIBRATION
TRAP

UFg TRAP

 

UFg TANK

Fig. 6. UF6 generation and metering system' flow diagram for FREE.
SR

YRS S

GAK RIDGE NATIONAL LABORATORY
3 4 5 6 7 8 2

Eie .

ia

Feed tank for FREE.

PHOTO 3144-74

 

k
 

  

PHOTO 3141-74

 

 

 

OAK RIDGE NATIONAL L ABORATORY
i A # & 7 4 ¥ . " ‘

Receiver tank for FREE.

ee
PHOTO 3151-74

    

 
 
 

™

i
<

[ \ %
(5]
wn

PHOTO 3148-74

      

e OAK RIDGE ! ATIONAL LABORATORY
) f 2 3 4 4 6 & 8 | 9 \O |

  

 

 

Fig. 10. Top view of receiver tank for FREE.
36

A side view of the UF6 absorption vessel is shown in Fig. 11, and a
top view is shown in Fig. 12. The nozzles in Fig. 12 are identical with
nozzles 1 through 4 of the return tank with one exception. The off-gas
nozzle is a 3/8-in. sched 40 pipe with a 3/8-in. Swagelock fitting which

serves as the off-gas line from that vessel.

The H2 reduction column is shown in Figs. 13-15. Two nozzles are

shown in a top view of the column (Fig. 14):
(1) a salt inlet nozzle (as previously described for the
feed tank); and

(2) an off-gas nozzle (as described for the UF
vessel).

6 absorption
Figure 15 shows the 3/8-in.-pipe side arm (the gas inlet and distributor).
The hydrogen supply line, a 1/4-in. tube, is welded to this side arm. Also
shown is the 3/8-in.-pipe salt exit port. The salt exit line, a 3/8-in.

tube, is welded to the exit port.

The flowing stream sampler is shown in Fig. 16. Access to the liquid
phase is obtained through the 3/4-in. ball valve. The sampler is equipped
with a 3/8-in. tube fitting for an off-gas line. A 3/8-in. tube fitting

for a "cover gas'" line is not shown.

The UF, and HF traps are identical in construction except for the

6
length. The UF. trap is 55-5/8 in. long, while the HF trap is 31-5/8 <in.
long. The UF6 trap is shown in Figs. 17 and 18. As seen in these figures,

the trap has a 3/8-in. fitting on the top and bottom flanges for gas inlet
and outlet, respectively. The HF trap is not shown because of its simi-

larity to the UF, trap.

6

6.2 UF6 Generation

The uranium hexafluoride for the fuel reconstitution engineering
experiment will be acquired in 5-in.-diam by 36-in.-high cylinders. These

cylinders contain up to 55 1b of UF The triple point of UF, occurs at

6 6
22 psia and 147.3°F. UF6(g) will be supplied to this experiment by main-

taining the temperature of the UF, cylinder at 220°F, and the vapor pres-

6
sure will be 70 psia. Heating will be accomplished by use of 17.2-psia

steam.
Pig. Ll.

 
 

37

PHOTO

NATIONAL LABORATORY

UF6 absorption vessel for FREE.

3149-74

 
PHOTO 31

50-74

 

 

= OAK RIDGE NATIONAL LABORATORY
0 ! 2 5 4 5 6w T 8 9 10 (R

~ ’ ' ' ' i I ! 1 t i t I ! | i |

Fig. 12. Top view of UF, absorption vessel for FREE.
O
Fig.

13.

H

2

reduction column for FREE.

PHOTO 3143-74

 

6¢€
40

    

PHOTO 3142-74

" —

 

 

Fig. 14. Top view of H2 reduction column for FREE.
 

column for FREE.

PHOTO 3145-74

_L\
Fig,

16.

Flowing stream sampler for FREE.

PHOTO 3146-74

 

&%
Fig, 17.

UF

6

trap for FREE.

PHOTO

3140-74

 

e
b

 

PHOTO 3147-74

OAK RIDGE NATIONAL [ ABORATOR

 

Fig. 18. Top or bottom view of UF, trap for FREE.

6
45

The steam heats the cylinders directly through 3/8-in. copper tubing
which is coiled around the entire cylinder, as illustrated in Fig. 6.
The lower 18 in. of the cylinder and tubing is covered with a conductive
cement to minimize response time. The cylinder and associated valves
are then enclosed in a cabinet insulated with 2 in. of Fiberglas insula-

tion.

6.3 Instrumentation and Controls

The principal objectives of the instrumentation and control system
are to provide closely regulated flows of molten salt, UF6, HZ’ and

diluent gases to the process, and to analyze the resultant off-gas.

Flow rate of salt to the UF6 absorption tank is controlled by regu-
lating the change of liquid level in the feed tank. This liquid level is
inferred from the pressure of argon that is supplied to a dip-leg bubbler

in the tank.

Figure 19 is a schematic diagram of the control system that regulates
the flow of salt to the UF6 absorption vessel. An adjustable ramp genera-
tor and an electric—-to-pneumatic converter are used to linearly decrease
the set point of a controller that senses liquid level in the feed tank
and controls the level by controlling the flow rate of pressurizing argon
to the gas space of the feed tank. The result should be a uniformly
decreasing salt level and, hence, a uniform discharge rate of salt from
the tank. This control system should be unaffected by small increases in
back pressure caused by plugging of transfer lines, decreasing feed tank
level, etc., or leakage of pressurizing argon (a small argon bleed is

provided to improve pressure control).

The UF6 control system is a simple feedback system involving a

Hastings Mass Flow Meter as the flow rate sensor (see Fig. 6). This sig-
nal from the mass flow meter is altered to a 10 to 50 mA signal to feed an
electronic Foxboro recorder—controller. The recorder-controller actuates
a 1/4-in. Research Control Valve. All control instrumentation and equip-

ment having contact with the UF, stream (with the exception of the feed

6
lines) are enclosed in an insulated cabinet maintained at 212°F.
46

ORNL-DWG 75-36

 

 

 

 

 

 

 

 

 

 

 

 

_____ J N N REMOTE SET-POINT
D BELLOWS
RAMP E/P FOXBORO
GENERATOR CONVERTER CONTROLLER
20 min —» 30 min

 

 

 

L4
ey

A o,

77
g
e

 

 

 

 

 

 

 

 

 

3-15 PSI AIR
D/P CELL AN RNy X J
N\
Ll I
¥
OFF-GAS J

 

 

 

<% ARGON

 

 

 

 

 

 

 

 

 

Fig. 19. Schematic diagram of control system for metering salt
from the feed tank for FREE.
47

The hydrogen and diluent gas streams are metered through Matheson flow

meters accompanied by differential pressure indicators. The off~gas anal-

ysis is made by Gow-Mac gas density detectors. The gas concentrations can

be inferred from the gas density analysis.

The liquid-phase system temperatures are maintained at the desired

temperatures by automatic controllers. These controllers regulate appro~

priate voltage transformers that supply power to Calrod tubular heaters,

10.

11.

12.

7. REFERENCES

R. M. Counce and J. R. Hightower, Jr., Engineering Development Studies

 

for Molten-Salt Breeder Reactor Processing No. 18, ORNL-TM-4698
(March 1975), p. 37.

 

MSR Program Semiann. Progr. Rept. Aug. 31, 1970, ORNL-4622, p. 101.

 

Chem. Technol. Div. Annu. Progr. Rept. March 31, 1973, ORNL-4883,
Pp. 23-25.

 

H, 0. Weeren et al., Engineering Development Studies for Molten-Salt
Breeder Reactor Processing No. 9, ORNL-TM-3259 (December 1972), pp.
205-15.

 

 

H. C. Savage, Engineering Development Studies for Molten—Salt Breeder
Reactor Processiang No. 19, ORNL-TM-4863 (July 1975), p. 1.

 

 

J. B. Lewis, Chem. Eng. Sci. 3, 248~59 (1954).

J. A. Klein et al., Engineering Development Studies for Molten—Salt
Breeder Reactor Processing No. 19, ORNL-TM-4863 (July 1975), p. 21.

 

J. A. Klein, Engineering Development Studies for Molten-Salt Breeder
Reactor Processing No. 18, ORNL-TM-4698 (March 1975), p. 1.

 

J. A. Klein et al., MSR Program Semiannu. Progr. Rep. Aug. 31, 1972,
ORNL“A‘BBZ, pe 1_71-

 

J. A. Klein, Engineering Development Studies for Molten-Salt Breeder
Reactor Processing No. 15, ORNL-TM-4019 (in preparation).

 

 

C. H. Brown, Jr. et al, Engineering Development Studies for Molten-

 

Salt Breeder Reactor Processing No. 19, ORNL-TM~4863 (July 1975).

R. M. Counce, MSR Program Semiannu. Progr. Rep. Aug. 31, 1974,
ORNL-5011, p. 127.
48

13. R. M. Counce, Engineering Development Studies for Molten-Salt
Breeder Reactor Processing No. 19, ORNL-TM-4863 (July 1975), p.
38.

 

 
INTERNAL DISTRIBUTION

MSRP Director's Office

C. F. Baes, Jr.
C. E. Bamberger

J. Beaus

M. Bender

M. R. Bennett

E. S. Bettis

R. E. Blanco

J. 0. Blomeke

E. G. Bohlmann
J. Braumnstein
M. A. Bredig

R. B. Briggs

H. R. Bronstein
R. E. Brooksbank
C. H. Brown, Jr.
K. B. Brown

J. Brynestad

S. Cantor

D. W. Cardwell
W. L. Carter

W. H. Cook

R. M, Counce

J. L. Crowley

F. L., Culler

J. M. Dale

F. L. Daley

J. H. DeVan

J. R. DiStefano
W. P. Eatherly
R. L. Egli, ERDA-OSR
J. R. Engel

G. G. Fee

D. E. Ferguson
L. M. Ferris

L. 0. Gilpatrick
J. C. Griess

W. R. Grimes

R. H. Guymon

49

41-56.
57.
58.
59.
60.
61.
62.
63.
64 .
65.
66.
67.
68.
69.
/0.
/1.
72.
73.
74,
75.
76.
77.
78.
79.
80.
81.
82.
83.
84.
85.
86.
87.
38.

89-91.
92.

93-102.
103.

-

*

.'ZL-oSUr'*OEHDWZGEZEGQWSPO';DEE.‘WPUZIPGEWU:J.C-i

*

R.

ORNL-TM~-4870
UC-76 — Molten Salt Reactor Technology

N momOmeO s s o

:—l

P.
D.

Lo o=

Hightower, Jr.
Hitch

Hortoun
Huntley

Kee

Kelmers

Laing
Lindauer
MacPherson
McClain

. McCoy

Malinauskas

. Matthews, ERDA-OSR

Meyer

. Moore

Nichols
Owen

. Postma
. Rosenthal
. Savage

Scott
Skinner
Smith
Smi.th

Spiewak

. G.
. K.
. M.
. B.
. E.
. R.

K.

Stewart
Tallent
Toth
Trauger
Unger
Weir
Wilkinson
Wymer

Central Research Library
Document Reference Section
Laboratory Records
Laboratory Records (ORNL-RC)
50

CONSULTANTS AND SUBCONTRACTORS

104, J.
105. C,
106. J.
107, W.
108. R.
EXTERNAL

Davis
Richards

DISTRIBUTION

109. Research and Technical Support Division, ERDA, Oak Ridge Opera-
tions Office, P. 0. Box E, Oak Ridge, Tenn. 37830
110. Director, Reactor Division, ERDA, Oak Ridge Operations Office,

P. 0. Box E, 0Osk Ridge,
111-112. Diyector, ERDA Division
Washington, D. C. 20545

Tenn. 37830
of Reactor Research and Development,

113-214. For distribution as shown in TID-4500 under UC-76, Molten Salt

Reactor Technology

v U5 GOVERNMENT PRINTING OFFICE: 1976-748-189/208