ORNL-TM-4286.txt

 

   
 
  
  

ORNL-TM-4286

 

 

 

ALLOY COMPATIBILITY WITH LiF-BeF, SALTS
CONTAINING ThF, AND UF,

J. W. Koger

 

 
 

 

 

This report was prepared as an account of work sponsored by the United
States Government. Neither the United States nor the United States Atomic
Energy Commission, nor any of their employees, nor any of their contractors,
subcontractors, or their employees, makes any 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.

 

 

 

»;

‘a
 

i

)}

ORNL-TM-4286

Contract No. W-7405-eng-26

METALS AND CERAMICS DIVISION

ALLOY COMPATIBILITY WITH LiF-BeF, SALTS CONTAINING
ThF; AND UF,

J. W. Koger

December 1972

 

- NOTICE Y
This report was prepared.as an account of work
sponsored by the United States Government. Neither
the United States nor the United States Atomic Energy |
Commission, nor any of their employees, nor any of
‘their contractors, subcontractors, or their employees,
makes any warranty, express or implied, or assumes any
legal liability or responsibility for the accuracy, com-
pleteness or usefulness of any information, apparatus,
product or process disclosed, or represents that its use | -
would not infringe privately owned rights,

 

 

 

 

OAK RIDGE NATIONAL LABORATORY .
Oak Ridge, Tennessee 37830
operated by
UNION CAREBIDE CORPORATION
for the
U.S. ATOMIC ENERGY COMMISSION

 

DISTRIBUTION OF THIS DOCUMENT 1S UKL

4

 
 

 

%)

 

 

®«>

'

 

 
 

iii

CONTENTS

A IaCt . o e e e e 1
IntrodUction . . ... ... e e 1
Molten Salt Chemistry . ... .. ...ttt et e e e e 3
Thermodynamic Data . . . .. e e e e e 6
Metal Corrosion ...................... ettt et et e P 7
Salt Preparation .. ..... ... ... e e 9
Materials Selection and Fabrication .......... ... ... .. it n
L00D OPerationS .. ...ttt ittt e e e 10
Results ........................ e 11
Hastelloy N ..o 11
L0 1S o e e e 11

00D IS A L et e e e 15

Loops 18 and 10 . .. . . i e 17

00D 2l L e e e e e e 23

L00D L6A ..o i e e e e e 27

Type 304L Stainless Steel ......... e e e e 28
e o 0 28
Maraging Steel . ... ... . e 33
Type 316 Stainless Steel . . ... ... . e e 36
L0 2 it e e e e e e 36

Other Ay S . . oot e e e e e e 40
Loopl18A....................... e e ettt e e 40

Conclusions .....oiviiiit it e, i it e et et ettt et 43

 
 

i

 

a

 

 
 

[L1]

 
 

™

J. W. Koger
ABSTRACT

Various compatibility tests between LiF-BeFj-based salts and Hastelloy N and certainsteels were
conducted in thermal convection loops. Temperature gradient mass transfer, as noted by weight losses
in the hot leg and weight gains in the cold leg, was evidenced in all tests. The weight changes of
corrosion specimens increased with increasing temperature and time.

From an operational standpoint, some difficulty was encountered in melting LiF-BeF,-ThF, salt
without mechanically failing the loop. The difficulty stemmed from the high melting point of the salt
and the high melting point of certain phases in the salt.

All the salts tested were quite compatible with Hastelloy N. Bismuth in contact with a fuel salt had
no effect on mass transfer in a Hastelloy N loop. We established that electrochemical methods to
determine the oxidation potential of molten fluoride salts could be used to predict the corrosion
behavior of a thermal convection loop. The values obtained by the electrochemical methods correlated
well with specimen weight change data. We showed that tellurium (as plated on Hastelloy N
specimens) does mass transfer in a molten fluoride "salt system and that an equilibrium between
tellurium in the salt and tellurium on the alloy can be established.

A type 304L stainless exposed to a fuel salt for 9.5 years in a type 304L stainless steel loop
showed a maximum uniform corrosion rate of 0.86 milfyear. Voids extended 10 mils into the matrix.
Chromium depletion was found.

The corrosion resistance of maraging steel (12% Ni—5% Cr—3% Mo~bal Fe) at 662°C was better
than type 304L stainless but worse than Hastelloy N under equivalent conditions. The uniform
corrosion rate was 0.35 mil/year. Subsurface voids were seen in the microstructure of the specimens
after 5700 hr, and microprobe analysis disclosed a definite depletion of chromium and iron.

Type 316 stainless steel exposed to a fuel salt in a type 316 stainless steel loop showed a maximum
uniform corrosion rate of 1 mil/year for 4298 hr, Mass transfer of chromium and iron did occur in the
system. :

For selected nickel- and iron-base alloys a direct correlation was found between corrosion
resistance in a molten fluoride salt and the chromium and iron content of an alloy. The more
chromium and iron in the alloy, the less the corrosion resistance.

INTRODUCTION

ALLOY COMPATIBILITY WITH LiF-BeF, SALTS CONTAINING ThF, AND UF,

A large portion of the support work of the Molten-Salt Breeder Reactor (MSBR) Program involved
studies of the compatibility and mass transfer of various alloys with LiF-BeF,-based molten salts. Table 1
gives the alloys involved, the salt compositions, and the details of the tests. Many of these tests were
conducted in thermal convection loops like that shown schematically in Fig. 1. The loops are essentially

Table 1. Thermai convection loop expenments“ involving LiF-BéFz-based molten salts

 

Salt composition + Maximum temperature AT

Loop No. O (mole®) o) o)

Time

(hr)

 

1258
15
15A
16%
16A
18
18A
- 19
19A
21
22

»

LiF-BeF5-Z1F 4-UF4-ThF4 (70-23-5-1-1) 688. . 100
LiF-BeF,-ThF4 (73-2-25) : 6T 55
LiF-BeF4-ThF4 (73-2-25) L 671 55
LiF-BeF,-UF4 (65.5-34-0.5) | - 704 170
LiF-BeF3-UF, (65.5-34-0.5) | 704 170
LiF-BeF-ThF4-UF4 (68-20-11.7-0.3) 704 170
LiF-BeF,-ThF4-UF, (68-20-11.7-0.3) 704 170
LiF-BeF5-ThF4-UF, (68-20-11.7-0.3) 704 170
-LiF-BeF3-ThF4-UF4 (68-20-11.740.3) 704 170
LiF-BeF;-ZrF4-UF4 (65.4-29.1-5.0) 610 110
LiF-BeF3-ThF4-UF4 (68-20-11.7-0.3) 650 110

83,516
2,003
39,476
37,942
5,878

15,930

2,278
4,745
24,515
13,798
4491

 

- > 9Loop 1258 constructed of type 304L stainless steel; loop 22 constructed of type 316 stainless steel; all other loops
u constructed of Hastelloy N.

bThis loop is discussed in another report, J. W. Koger, Effect of FeF, Addition on Mass Transfer in a Hastelloy

N-LiF-BeF 5-UF4 Thermal Convection Loop System, ORNL-TM-4188 (December 1972).

 
 

ORNL-DWG €8-~3987R3

g

CLAMSHELL
HEATERS

 

  
 

o
o
s

— r—— ——— t— — ——— — — — ———— S ‘r
h

INSULATION

CORROSION
SPECIMENS

 

 

 

 

’__‘]__—.-—.-_.._-__-—

 

FLUSH
TANK DUMP

 

 

 

 

 

 

Fig. 1. MSRP natural circulation loop and salt sampler.

o

w

 
By

 

test bed systems wherein metal spécimens suspended in the salt and salt samples can be removed and/or
replaéed at operating temperature without disturbing flow or introducing air contamination. The salt flow
in the thermal convection loops is generated by the difference in defisity of the salt in the hot and cold legs
of the loop and ranges from 1.5 to 7 ft/min depending on salt density and viscosity.

A thorough chemistry background and physical properties of the molten salts are covered elsewhere by
Grimes.!

MOLTEN SALT CHEMISTRY

Uranium tetrafluoride and uranium trifluoride are the only fluorides of uranium which appear useful as
constituents of molten fluoride fuels. Uranium tetrafluoride (UF,) is relatively stable, nonvolatile, and
nonhygroscopic. It melts at 1035°C (1895°F), but its freezing point is markedly depressed by useful
diluent fluorides. Uranium trifluoride disproportionates at temperatures above ~1000°C by the reaction

4UF, = 3UF, + U°, 3)

1t is chemically unstable?:3 at lower temperatures in most molten fluoride solutions and is tolerable in
reactor fuels only with a large excess of UF, so that the activity of U is so low as to avoid appreciable
reaction with moderator graphite or container metal.
Thorium tetrafluoride (ThF,) is the only known fluoride of thorium. It melts at 1111°C (2032°F), but
fortunately its freezing point is markedly depressed by fluoride diluents which are also useful with UF,.
Consideration of nuclear properties alone leads one to prefer as diluents the fluorides of Be, Bi, TLi, Mg,

" Pb, and Zr in that order. Simple consideration of the stability of these fluorides®:5 toward reduction by

structural metals, however, eliminates the bismuth fluorides from consideration. This leaves BeF, and TLiF
as the preferred diluent fluorides. Phase behavior of syktems based upon LiF and BeF, as the major
constitients has, accordingly, been examined in detail.® Fortunately the phase diagrams of LiF-BeF,-UF,
and LiF-BeF,-ThF4 are such as to make these materials useful as fuels.

The binary system LiF-BeF, has melting points below 500°C over the concentration range from 33 to
80 mole % BeF,.5:7 The phase diagram presented in Fig. 2 is characterized by a single eutectic (52 mole %
BeF,, melting at 360°C) between BeF, and 2LiF-BeF,. The compound 2LiF+BeF, melts incongruently to
LiF and liquid at 458°C. LiF+BeF, melts below 280°C.

The phase behavior of the BeF,-UF,%:7 and BeF;-ThF,® systems is very similar. Both systems show
simple single eutectics containing very small concentrations of the heavy metal fluoride. ThF, and UF; are
isostructural; they form a continuous series of solid solutions.

" The binary diagrams LiF-UF,° and LiF-ThF,'° are generally similar and much more complex than the
binary diagrams just discussed. ‘The LiF-UF, system has ihreé_ compounds (none of which melt

 

1. W. R. Grimes, “Molten-Salt Reactor Chemistry,” Nucl, Appl Technol 8 137 (1970).

2. W. R. Grimes, “Materials Problems in Molten Salt Reactors,” in Materials and Fuels for High Temperature Nuclear
Energy Applications, M. T, Simnad and L. R. Zumwalt, eds., the M.I.T. Press, Mass. (1964).

3. W.R. Grimes, MSR Program Semiannu. Progr. Rep. July 31, 1964, ORNL-3708, p. 214.

4. Alvin Glassner, The Thermochemical Properties of the Oxides, Fluorides, and Chlorides to 2500 °K, ANL-5750.

5. L. Brewer, L. A. Bromley, P. W. Gilles, and N, L. Lofgren, MDDC-1553 (1945) and L. Brewer in The Chemistry and
Metallurgy of Miscellaneous Materials, Thermodynamics, L. L. Quill, ed., McGraw-Hill, New York, pp. 76192 (1950).

6. R. E. Thoma, Phase Diagrams of Nuclear Reactor Materiais, 0RNL-2548 (Nov. 6, 1959). _

7. L. V. Jones, D. E. Etter, C. R. Hudgens, A. A. Huffman, T. B. Rhinehammer, N. E. Rogers, P. A. Tucker, and L.J.
Wittenberg, “Phase Equilibria in the Ternary Fused-Salt System LiF-BeF4-UFy,” J. Amer. Ceram. Soc. 45,79 (1962).

8. R. E. Thoma, H. Insley, H. A. Friedman, and C. F. Weaver, “Phase Equilibria in the Systems BeF;-ThF4 and
LiF-BeF,-ThF,,” J. Phys. Chem. 64, 865 (1960).

9. C. 1. Barton, H. A. Friedman, W. R. Grimes, H. Insley, and R. E. Thoma, “Phase Equilibria in the Alkali
Fluoride—Utanium Tetrafluonde Fused Salt System: 1. The Systems LiF-UF4 and NaF-UF4,” J. Amer. Ceram. Soc. 41, 63
(1958).

10. R. E. Thoma, H. Insley, B. S. Landau, H. A. Friedman, and W, R. Grimes, “Phase Equilibria in the Fused Salt
Systems LiF-ThF 4 and NaF-ThF4,” J. Phys. Chem. 63, 1266 (1959).

 
 

ORNL-DWG 79-5270R2

 

 

 

   
  

 

 

 

 

 

 

 

 

 

| [N { i 'fi o | | |
800 L s 500 T y ; ' -
848 X {EUTECTIC) = 0.3280 1 0.0004
/ Tmox = 4591 £0.2
800 - 450 - LIQUID -1 J
' X (EUTECTIC =
0.531£0.002
700 |- 400 LipBeF, + — -
5 LiF + _ LIQUID .~
& LIQUID
£ 00 ' 350 | 1 I 1 555 -
g 030 035 0.40 045 050 055
@
w
500 |- R L , —
= 458.9 £0.2°C t . - BeF, (B-QUARTZ TYPE)
- I | + LIQUID
} i '
400 i / —
. i , 363.5 +0.5°C
LiF + L - .
. . LiBeF3 + BeFp
LipBeF, o| LizBef +BeF, (B-QUARTZ TYPE) (
~QUARTZ TYPE), _|
300 |- %\' 280°C e )\
5| LigBeFg + u'| LiBeFy + BeF, (8-QUARTZ TYPE}
~ | UiBers & 3 : 227°C
200 | | l 3 1 L L I

 

 

 

 

 

o o 0.2 0.3 0.4 05 0.6 0.7 08 0.9 1.0
) S XB.thmole fraction) -

Fig. 2. The system LiF-BeF,.

congruently) and a single eutectic at 27 mole % UF,, melting at 490°C. The LiF-ThF, system contains four
binary compounds one (3LiF+ThF,) which melts congruently, with two eutectics at 570°C and 22 mole %
ThF, and at 560°C and 29 mole % ThF,.

The ternary system LiF-ThF,-UF,,'! shown in Fig. 3, has no ternary compounds and a single eutectic
freezing at 488°C ‘with 1.5 mole % ThF, and 26.5 mole % UF,. Most of the area on the diagram is

occupied by the primary phase fields of the solid solutions UF,-ThF,, LiF-4UF,-LiF-4ThF,, and

LiF-UF,-LiF-ThF, . Liquidus temperatures decrease generally to the LiF-UF, edge of the diagram.

The single-fluid molten-salt breeder fuel will need a concentration of ThF4 much higher than that of
UF,. Accordingly, the phase behavior of the fuel parallels that of the LiF-BeF,-ThF, system. Figure 4
gives the ternary system LiF-BeF,-ThF,; this system shows a single ternary eutectic at 47 mole % LiF and
1.5 mole % ThF,, melting at 360°C.5:7 The system is complicated to some extent by the fact that the
compound 3LiF+ThF, can incorporate Be?" ions in both interstitial and substitutional sites to form solid
solutions whose compositional ‘extremes are represented by the shaded triangular region near that
compound. Liquidus temperatures <550°C (1022°F) are available at ThF, concentrations as high at 22
mole %. The maximum ThF, concentration available at liquidus temperatures of 500°C (932°F) is above
14 mole %. Inspection of the diagram reveals that a considerable range of compositions with >10 mole %
ThF,4 will be completely molten at or below 500°C. |

As éxpe_cted from the general similarity of ThF, and UF, — and especially from the substitutional

~ behavior shown by the LiF-UF,-ThF, system (Fig. 3) — substitution of a small quantity of UF, for ThF,
- scarcely changes the phase behavior. Accordingly and to a very good approximation, Fig. 4 represents the

behavior of the LiF-BeF,-ThF,-UF, (MSBR fuel) system over concentrat:on regions such that the mole
fraction of ThF, is much greater than that of UF,.

 

-~ 11. C. F. Weaver, R. E. Thoma, H. Insley, and H. A. Friedman, “Phase Equlhbna in the System UF4-ThF, and
LlF-UF4-ThF4, J. Amer. Ceram. Soc. 43 213 (1960).

0y

 
 

 

    

 

3
ThF,
4
ORNL~LR-DWG 28215AR2
TEMPERATURE IN °C PRIMARY-PHASE AREAS
IN
COMPOSITION IN mole % 4 (0) UF,-ThF, (ss)
(b} LiF- 4UF,~LiF-4ThF, (ss)
{(c) LiF: 2Th(U)F (ss)
LiF-4The, (d) 7LiF-6UF, -7L|F 6ThF, (ss)
(e) 3LiF- Th(U)F (ss)
(f) LiF
LiF-2ThF,
PBIT R
£\
. _
TLiF- 6ThE,
P62 A
= \
P 597 A \
£ 565
3LiF-ThE [
£ 568
Q P 609
@\
AN ) .
\
-l % 51 62% 2500 _
| ". \ |
L \ \ £48 o,
845 4L|F UFy” P500" 490 P 610 PTTS. LiF-4UF, o
' 7L|F 6UF, ' .
Fig. 3. The system LiF-ThF4-UF4.
.

 
 

ORNL~LR-DWG 37420AR?

   

 

 

ThE, 41
00
W TEMPERATURE IN °C
COMPOSITION IN mole %
1050
LiF-4ThE,
LiF-The LiF-2ThE; |
3LiF-Th, ss 4 ! 4 1000~
LiF-2ThE,
. PB9? : : 950
2Lil-‘-B¢aF2 -
LiF-ThE, | 900
P 762 650
~8
P 597
£ 568
3LiF-ThE s
4 NN X 7
£ 565 NI %
700
650
AR 60
% DO\ \s \ ' ©
%, O w 550 526
LiF N/ N\ Ber
. P 2
848 2LiF-BeR; 500450 400 | 400 450 500 555
pPasg £ 360 '

Fig. 4. The system LiF-BeF,-ThF,.

THERMODYNAMIC DATA

The success of an MSBR is strongly dependent on the compatib‘ility of the container materials with the

molten salts used in the primary and secondary circuits of the reactor. Because the products of oxidation of .

metals by fluoride melts are quite soluble in the corroding media, passivation is precluded, and the

. corrosion rate depends on other factors, including ‘the thermodynamic driving force of the corrosion

reactions. Design of a practicable system utilizing molten fluoride salts, therefore, demands the selection of
salt constituents such as LiF, BeF,, UF,, and ThF, that are not appreciably reduced by available structural
metals and alloys whose components Fe, Ni, and Cr can be in near thermodynamic equilibrium with the
salt. ,
A continuing program of experimentation over many years has been devoted to definition of the
thermodynamic properties of many corrosion product species in molten LiF-BeF, solutions. Many of the
data have been obtained by direct measurement of equilibrium pressures for reactions such as

H,(g) + FeF,(d) = Fe®(c) + 2HF(g) @

and .
2HF(g) + BeO(c) = BeF, (1) + H,0(g), " (5)

»

w

m

 
 

 

Y

bA )

91

 

Table 2. Standard free energies of formation for
species in molten 2LiF-BeF,

733 to 1000°K

 

 

Material® —AG (kcal/mole) ~AGF(1000°K) (kcal/mole)
LiF(1) 141.8-16.6 X 10~ T°K 125.2
BeF,(1) 243.9-30.0 X 102 T°K 106.9
UF;(d) 338.0-40.3x 10> T°K 99.3
UF4(d) 4459-579 x 1073 T:K 97.0
ThF4(d) 491.2-624 X 103 T°K 107.2
ZtF4(d) 453.0-65.1 x 102 7T°K 97.0
NiF,(d) 146.9-36.3 X 1072 T°k 55.3
FeF,(d) 154.7-21.8 X 102 T°K 66.5
CrF,(d) 171.8-214 x 102 7°K 75.2
MoFg(g) 370.9-69.6 x 102 T°K 50.2

 

9The standard state for LiF and BeF, is the molten 2LiF+BeF, liquid. That
for MoFg(g) is the gas at 1 atm, That for all species with d is that hypothetical
solution with the solute at unit mole fraction and with the activity coefficient it
would have at infinite dilution.

where g, c, and d represent gas, crystalline solid, and solute, respectively, using the molten fluoride (I) as
reaction medium. Baes has reviewed all these studies' > and, by combining the data with the work of others,
has tabulated thermodynamic data for many species in molten 2LiF+BeF,. Table 2 records pertinent data
for the major components of MSRE and MSBR fuels and for corrosion products in moiten 2LiF-BeF,.

From these data one can assess the extent to which UF;-bearing melt will disproportionate according to
the reaction

4UF4(d) = 3UF,(d) + U(d). | 6)

For the case where the total uranium content of the salt is 0.9 mole % (as in the Molten-Salt Reactor
Experiment), the activity of metallic uranium (referred to the pure metal) is near 10~'5 with 1% of the
UF, converted to UF, and is near 2 X 107'° with 20% of the UF,4 so converted.!® Operation of the
reactor with a small fraction (usually <2%) of the uranium present as UF; is advantageous insofar as
corrosion and the consequences of fission are concerned. Such operation with some UF; present should
result in formation of an extremely dilute (and experimentally undetectable) alloy of uranium with the
surface of the container metal. Operation with >50% of the uranium as UF; would lead to much more
concentrated (and highly deleterious) alloying and to formation of uranium carbides. All evidence to date
demonstrates that operation with relatively little UF; is completely satisfactory.

METAL CORROSION
The data of Table 2 reveal clearly that in r_eacti'ons with structural metals M)

2UF4(d) + M(0) = 2UF3(d) + MFs (@), - ™

chromium is much more readily attacked than iron, nickel, or molybdenum.!3.14

 

12. C. F. Baes, Jt., “The Chemistry and Thermodynamics of Molten Salt Reactor Fuels,” presented at AIME Nuclear
Fuel Reprocessing Symposium at Ames Laboratory, Ames, [owa, Aug. 25, 1969. Published in the 1969 Nuclear Metallurgy
Symposium, Vol. 15, by the USAEC Division of Technical Information Extension.

13. G. Long, Reactor Chem. Div. Annu. Progr. Rep. Jan. 31, 1965, ORNL-3789, p. 65.

14. J. W. Koger, MSR Program Semiannu. Progr. Rep. Aug. 31, 1970, ORNL4622, p. 170.

 
 

 

 

Nickel-base alloys, more specifically Hastelloy N and its modifications, are c_onéidere'd the most

‘promising for use in molten salts and have received most attention. Stainless steels, having more chromium

than Hastelloy N, are more susceptible to corrosion by fluoride melts but can be considered for some
applications. ' | '
Oxidation and selective attack may also result from impurities in the melt,

M+ NiF; > MF; +Ni, . | 8)

M + 2HF > MF, + H,, : - )
or oxide films on the metal,

NiO + BeF, - NiF, + BeO, o | (10)1

followed by reaction of NiF, with M.

Reactions (8), (9), and (10) will proceed essenhally to completlon at all temperatures within the MSBR
circuit. Accordingly, such reactions can lead (if the system is poorly cleaned) to a rapid initial corrosion
rate. However, these reactions do not give a sustained corrosive attack. '

Reaction (7) with UF,, on the other hand, may have an equilibrium constant which is strongly
temperature dependent; hence, when the salt is forced to circulate through a temperature gradient, a
possible mechanism exists for mass transfer and continued attack. Reaction (7) is of s:gmficance mainly in
the case of alloys containing relatively large amounts of chromium.

If nickel, iron, and molybdenum are assumed to form regular or ideal solid solutions with chromium (as
is approximately true), and if the circulation rate is very rapid, the corrosion process for alloys in fluoride
salts can be simply described. At high flow rates, uniform concentrations of UF, and CrF, are maintained
throughout the fluid circuit. Under these conditions, there exists some temperature (intermediate between
the maximum and minimum temperatures of the circuit) at which the initial chromium concentration of
the structural metal is at equilibrium with the fused salt. This temperature Tgp is called the balance point.
Since the equilibrium constant for the chemical reaction with chromium increases with increasing
temperature, the chromium concentration in the alloy surface tends to decrease at temperatures higher than
Tgp and tends to increase at temperatures lower than Tpp. At some point in time, the dissolution process
will be controlled by the solid state diffusion rate of chromium from the matrix to the surface of the alloy.
In some melts (NaF-LiF-KF-UF,, for example) the equilibrium constant for reaction (7) with chromium
changes sufficiently as a function of temperature to cause formation of dendritic chromium crystals in the
cold zone. For MSBR fuel and other LiF-BeF;-UF; mixtures, the température dependence of the mass
transfer reaction is small, and the equilibrium is satisfied at reactor temperature conditions without the
formation of crystalline chromium. Thus, the rate of chromium removal from the salt stream by deposition
at cold-fluid regions is controlled by the rate at which chromium diffuses into the cold-fluid wall; the
chromium concentration gradient tends to be small, and the resulting corrosion is well within tolerable
limits. A schematic of the temperature gradient mass transfer process is shown in Fig. 5.

i

5 ]

i

 
o

“t

 

ORNL-DWG 67-6800R

 

 

 

 

HOT SECTION

 

DIFFUSION TO SURFACE

 

SOLUTE ESCAPE THROUGH
NEAR-SURFACE LIQUID LAYER

—— ~ 7 DIFFUSION INTO BULK LIQUID

‘ —_——

TRANSPORT
TO COLD PORTION OF SYSTEM

 

 

\\\\\\\\\\\\\\\\\\\\\\\\\‘L\\

 

 

COLD SECTION ==
x SUPERSATURATION —
— { NUCLEATION
o GROWTH TO STABLE CRYSTAL SIZE

____OR

SUPERSATURATION AND DIFFUSION
THROUGH LIQUID

NUCLEATION AND GROWTH
ON METALLIC WALL — —— - —
OR DIFFUSION INTO WALL

 

 

  
  
   

 

 

 

 

Fig. 5. Temperature-gradient mass transfer.

SALT PREPARATION

The salt for the loops was prepared by the Fluoride Processing Group of the Reactor Chemistry
Division. Although starting materials of very high pu:i';y' were used in production of fused fluoride
mixtures, further purification was needed before the salts were used in loop systems. Two steps were
required: one for the removal of oxides and sulfides and one for the removal of metallic fluorides. For the
blanket and fuel salts, the oxides and sulfides were removed by gas sparging for several hours (usually 10 to
20) at 650°C with an anhydrous mixture of HF in hydrogen (1:4). The purification reactions are of the

~. type
02~ + 2HF = 2F~ +H,0. - (1)
The process is continued until the same amount of HF leaves th_e reaction vessel as enters. The reaction is

then considered complete

Gas sparging of the melt at 700 C with hydrogen alone for several hours (usually 20 to 40) was used as
the final phase of the punficatnon process to reduce FeF, and N1F2 concentrations to low values and to
remove any residual HF:

H, + MF, = M?° + 2HF. )

 
 

10

" The reduction of CrF, by hydrogen is too slow to be effective at process temperatures However, the
chromium concentration of the starting materials is very low. The by-product of the hydrogen spargmg is

HF, and the process is continued until the HF evolution is below a certain level.

MATERIALS_ SELECTION AND FABRICATION

The Hastelloy N material for loops 15, 16, 18, '19, and 21 was from heat No. 5097 obtained from

Stellite Division, Cabot Corp, and the composition is given in Table 3. The loops were fabricated from
0.750-in.-OD annealed tubing with a wall thickness of 0.072 in. Machined pieces were annealed at 1175°C
in H, before welding. The material was TIG welded in conformance with ORNL Specifications PS-23 and
PS-25 and inspected under ORNL Specification MET-WR-200. After weldrng, the loops were stress relieved
at 880°C for 8 hr.

Table 3. Composition of

 

 

 

Hastelloy N loop
material (heat 5097) -
Element Weight percent

Mo 16.5
Cr 6.9
Fe 4.5
Si 0.69
Mn 0.54
Ti 0.02
Ni Bal
LOOP OPERATIONS

The loops were heated by pairs of clamshell heaters placed end to end around two legs of the loop (Fig.
1). The input power was controlled by silicon controlled rectifier (SCR) units, and the temperature
controlled by a current proportioning controller. The loop temperatures were measured by Chromel-P vs
Alumel thermocouples that had been spot welded to the outside of the tubing, covered by a layer of quartz
tape, and then covered with stainless steel shim stock. Tubular electric heaters controlled by variable
autotransformers furnished the heat to the cold leg portions of the loops during startup of the loops.

Before filling with salt, the loops were degreased with ethyl alcohol, dried, and then heated to 150°C

under vacuum to remove any traces of moisture. A helium mass spectrometer leak detector was used to
check for leaks in the system.

The procedure for filling the loops consisted of heating the loop, the salt pot, and all connecting lines to
approximately 550°C and applying helium _pressure to the salt supply vessel to force the salt into the loop.

seal.
Normally the first charge of salt was held for 24 hr in the loops at the maximum operating temperature

“and then dumped. This flush salt charge was intended to remove surface oxides or other impurities left in

the loops. The loop was then refilled with fresh salt, and operation began. Once the loop was filled, the
heaters on the cold leg of the loop were turned off. As much insulation was removed as necessary to obtain
the proper temperature difference by exposing the cold leg to ambient air. Helium cover gas of 99.993%

- purity and under slight pressure (approx 5 psig) was maintained over the salt in the loops during operation.

Ajr was contmuously blown on freeze valves leading to the dump and flush tanks to provide a positive salt

o
 

 

 

 

11

With the exception of loop 1258 which only had specimens in the hot leg, the loops all contained 14 to
16 specimens 0.75 X 0.38 X 0.030 in., each with a surface area of 0.55 in.? (3.45 cm?). Generally two of
the specimens were titanjum-modified Hastelloy N. (Modifications to the standard Hastelloy N composition
were made to test for improved mechanical properties after neutron irradiation.) The composition of the
standard and modified Hastelloy N specimens is given in Table 4. Eight specimens were attached at
 different vertical positions on each of two Y-in. rods. One rod was inserted in the hot leg and another in
the cold leg. The rods were placed into or removed from the loop from standpipes atop each leg. The rods
were moved through a Teflon sliding seal compression fitting at the top of the standpipe and a ball valve at .
the bottom. Another ball valve on the loop above each leg assured removai or insertion without disturbing
loop operation or introducing air contamination.

The corrosion specimens were withdrawn penodxcally along with salt samples to follow corrosion
processes as a function of time. During the removal periods all specimens were weighed and measured.

- Portions of the hottest and coldest specimens were removed and examined metallographically during tests,

and all specimens were examined metallographically after test.

Table 4. Composition of Hastelloy N speciméns

 

Weight percent
‘Mo Cr Fe Si Mn T Ni

 

 

Tirmodified 13.8 7.3 <0.1 0.05 0.13 047 Bal
Standard 172 74 45 0.64 055 002 Bal

 

RESULTS
Hastelloy N

Loop 15. Loop 15 was operated to determine the compatibility of a proposed blanket salt for a

~ two-fluid molten salt breeder reactor with Hastelloy N. Flgure 6 shows loop 15 on the left in place before

operation.

For loop 15 the fuel pot and associated lines had to be heated to 730° C and hlgher than normal inert
gas pressure applied to fill the loop with salt. During the initial salt transfer the fill line failed and resulted
in a spill. After repairs, we managed to fill the loop with the flush charge but found it impossible to initiate
circulation even after heating the loop’s hot leg to 760°C. The loop was drained and then refilled with clean
salt and put into operation. The loop was operated at a maximum temperature of 676 C with a
temperature difference of 54°C. A ‘helium cover gas under shght posmve pressure (~35 ps1g) was maintained
over the salt in the loops during operation.

On May 1, 1968, after 2003 hr of uneventful operation, the loop cooled due to a blown fuse, and all
salt was frozen. Heat was applied to the loop in order to thaw the salt, and a rupture occurred in the hot leg
loop piping 30 in. down from the hot leg surge tank (Figs. 7 and 8), thus ending the operation of loop 15.

The weight changes of the specimens in the hottest and coldest positions are plotted vs time of
operation in Fig. 9. The standard Hastelloy N specimen at a lower temperature lost more weight than the
titanium-modified specimen at the higher temperature. We attribute this increased corrosion resistance to
the lower iron concentration of the titanium-modified alloy (Table 4). Weight losses were measured for hot
leg specimens and weight gains for cold leg specimens, thus implying that temperature gradient mass
transfer had occurred. The weight changes in all cases were quite small.

 
 

12

|4
~N
-
ok
-
Q
-
o
I
o

LY

 

 

 

 

10N.

loops 15 and 16 before operati

L,

i

Lefttor

. 6.

F;
 

 

 

 

 

 

 

 

i
i

 

 

 

 

 

~
kS

 

 

13

Fig. 7. ¥Yiew of loop 15 and the ruptured area of NCL-15.

PHOTO 75636

 

 
 

 

14

10
0
©
B i
™~
o
-
o
X

w

 

 

 

 

 

Ruptured area of NCL-15.

ig. 8.

F
 

)

15

¥

ORNL—-DWG . 68—-44778

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

2T

O Ti MODIFIED HASTELLOY N
- 4 STANDARD HASTELLOY N
e
<
< |
£ 613°C COLDEST
a o #_ SPECIMEN
= 676 °C HOTTEST
3 +—a—T 0T ° SPECIMEN
5 Y
= 663 °C
I
Q —
w
= LOOP 15

0 500 1000 1500 2000 2500 3000
’ TIME (br) -

Fig. 9. Weight changes of Hastelloy N specimens in loop 15 vs time of operatibn in LiF-BeF,-ThF, (73-2-25 mole %).

Figure 10 shows microstructures of the specimens in the hottest and coldest positions, respectively.
There is little or no indication of corrosion or mass transfer effects. :

The salt was periodically analyzed for its major constituents, the major alloying elements of the -
container material, and oxygen and water. The chromium and iron increased with time, but the changes- 7
were very small (about 20 ppm mcrease) Nlckel showed a small overall increase, molybdenum remained at

<20 ppm during operation, oxygen ranged from <50 to 2 high of 244 to a value of 75 ppm just before the

rupture, and water was <500 ppm in each analysis. The latter four values are quite low and preclude any
possibility of air or water contamination. ‘

Table 5 gives pertinent details of loops that had operated in past years for various times with L1F-ThF4
(71-29 mole %) salt. The two Inconel loops that operated for a year (8760 hr) showed subsurface void
formation for distances up to 6% mils, but the INOR-8 (Hastelloy N) loop that oberated for a year showed
no voids. The improved behavior of INOR-8 was somewhat expected since there is less chromium present.
None of the loops showed any signs of plugging, and no large increases of any of the alloy constituents are
noted in the salt. Our experiment substantxated and quantified the previously observed good behavior of
Hastelloy N in LiF-ThF, salts.

From the operational viewpoint, this salt provided more dxfficultles than any other fluoride salt that has
been handled by the present Reactor Materials Group personnel. The first problem was the high

temperature (730°C) needed to transfer the fuel and the apparent presence of unmelted salt constituents. It

is thought that there was perhaps a separation of a high melting component which caused this problem. The
second problem was the rupture of the loop while trying to thaw the frozen salt. This thawing procedure
had been used with other salts quite successfully with no failures. | -

Loop 15A. The ruptured area of loop 15, which still contained frozen salt, was cut from the loop and ’
another piece of tubing was attached by means of compression-type fittings. The loop was then heated, the
salt was melted and dumped, and the test specimens were removed. The temporary fitting was removed,
and the joints to be welded were reamed to remove any remaining salt and corroded metal and filed to
provide the proper weld joint configuration. Welding was done in conformance with ORNL sPecificmions

 
 

 

I
|

 

 

 

 

'Fig. 10. Microstructures of Hastelloy N specimens
mole %). (¢) Hot leg, 676°C; (b) cold leg, 613°C.

 

 

for loop 15 after 2000 hr exposure to LiF-BeF,-ThF,

-
I

10.004 in.

1

 

10.003 in.

0.007 INCHES
S00X

10.005 in.

b

 

 

10,007 in.

&
| -

1
1

10.00¢ in,

I

 

10.003 in.

]

0.007 INCHES
500X

10.005 in.

 

 

10.007 in.

 

>

 
o)

 

 

 

 

17

Table 5. Thermal convection loops that operated with LiF-ThF4 (71—29 mole %) salt

 

 

Chemical
Maximum changes '
Loop Hours AT in salt .
ma _terial operated tem?eg;ture (o o) l(:pm) Hot leg condition
G Fe
Inconel® 1000 692 111 45 0  Light surface roughemng, few voids to a depth of <1 mil.
Inconel? 1000 676 75 70 0 Fewpitstol /2 mils, few voids to a depth of <1 mil.
Moderate surface roughening.
Inconel? 8760 733 90 95 20 - Intergranular voids to 6% mils. Moderate surface
roughening ‘
Inconel? 8760 676 79 300 0  Pits to 1 mil, voids to § mils. Moderate surface
, roughening
INOR-8? 8760 733 93 240 60  Heavy surface roughening.
INOR-8¢ 3114 690 72 65 50  Light surface roughening.
INOR-84 1000 - 682 61 40 120 Light surface roughening.

 

415 Cr, 7 Fe, bal Ni.

b17 Mo, 8 C1, 5 Fe, bal Ni.

€15 Mo, 6 Cr, 5 Fe, bal Ni.-

d15.8 Mo, 6.99 Cr, 4.85 Fe, bal Ni.

PS-23 and PS-25 and inspected under ORNL specification MET-WR-200. The first welding attempt was
unsuccessful because of poor penetration from excessive gas pressure in the loop. The second attempt was
successful and demonstrates that a successful weld can be made on material that has contained a molten
The loop was filled with salt from the original batch and designated 15A. After operation began,
specimens were found to have a “glaze™ or coating (probably a high-melting thorium compound) that was
impossible to remove without damaging the specimen. However, based on changes of chromium
concentration and metallographic studies, little if any mass transfer was indicated. Toward the end of the
run the coating on specimen surfaces disappeared, and the weight changes were measurable. For a total
specimen exposure time of 37,792 hr, the maximum weight loss was 4.7 mg/cm?, equivalent to a uniform
corrosion rate of 0.05 mil/year, and the maximum weight gain was 4.2 mg/cm?. Figure 11 shows the
microstructure of the hottest and coldest specimens. Some attack along the gram boundaries of the hottest
specimen is seen.

Loop 18 and 19. Loops 18 and 19 were constructed of standard Hastelloy N with removable specimens
in each leg and were operated with the salt mixture 68% LiF—20% BeF,~11.7% ThF,—0.3% UF, (mole
%). Loop 19 had a molybdenum hot finger containing bismuth at the bottom of the hot leg (Fig. 12) which
allowed contact between the flowing salt and the bismuth. The loops were operated to obtain data on the
compatibility of a fertile-fissile salt mixture with Hastelloy N and to determine if contactmg the salt with
liquid bismuth would affect the corrosion rate. ,

After a power outage at 4741 hr the bismuth finger on loop 19 was inadvertently heated to
temperatures over 1000°C. Subsequent examination disclosed localized melting of the Hastelloy N which
surrounded the molybdenum container. Portions of the loop and the _finger were removed for exanfination
and replacement parts fabricated for repair. :

- Prior to the 4741-hr shutdown the maximum weight loss for a standard Hastelloy N specimen at 704° C
was 0.8 mg/cm?. Assuming uniform dlssolutlon ‘this weight loss is equivalent to 0.07 mil/year, an
insignificant corrosion rate. The chromium content of the salt increased from 25 to 66 ppm while iron
decreased from 78 to 23. Figure 13 shows the metallographic appearance of the hottest and coldest

 
 

18

;
i
!
!
|
¥
i
1
{
i
i
i

 

 

 

   

 

|
|

 

Fig. 11. Microstructures of Hastelloy N specimens from loop 15A after 37,792 hr exposure to LiF-BeF,-ThF,
(73-2-25 mole %). (a) Hot leg, 676°C; (b) cold leg, 613°C. 500X.

 

 

 

 

 
i
!
|

 

 

 

 

 

 

 

 

 

19

PHOTO 75803A

&

~HASTELLOY N .
Mo-TZM

 

Fig. 12. Bismuth “hot finger” contained in molybdenum with Hastelloy N jacket on loop' 19.

 
 

20

 

S

X006 «|

IHONI LOO

 

C

 

o X00% i
SINONI L00

 

 

 

Q

 

 

 

 

 

®
3
E
0
Q@
=~
-
v
&
~
o
v
S’
<
5
3
L
E
o™
ke
O
A
]
i
o
b
=]
2
o
)
o
-
y—
e
3
E .
finw.
3
2 o
=5
w3
L ]
mo
—
RS
o/ 0
£
QS
mn...
@ o
QO
8%
mo
=
L e
oS
o b
L =
oy e
~
~
L
 

»

 

 

 

»

N

 

21

specimens. Little attack or deposition is seen. The bismuth content of the salt was <20 ppm for the last
1500 hr. After the temperature excursion the salt and the bismuth at the top of the finger were both
analyzed for certain metal impurities, and the results are giveri in Table 6.

Table 6 shows that bismuth extracted impurities from the salt in quantities proportional to the
solubility of these structural metals in bismuth, Analysis of the dumped salt from loop 19 disclosed 225
ppm bismuth. - ' -

At the time of this analysis, in loop 18 the maximum weight loss of a standard Hastelloy N specimen
was 0.8 mg/cm? at 704°C in 3400 hr, a very low rate, equivalent to 0.08 mil/year assuming uniform attack.
The chromium concentration in the salt in this loop had increased from 21 to 90 ppm. We showed no
difference in mass transfer in the two loops at this time. ‘

The specimen from the hottest position of loop 19 was bent around a Y;-in.-diam tube and examined
metallographically (Fig. 14). There was no indication of cracking or weakening of grain boundaries, and
there were no signs of salt or bismuth attack.

Table 6. Analysis of salt and bismuth
from hot finger portion of loop 19

 

 

Major Impurities Concentration Solubigity at
constituent (ppm) 600 C¢
Bismuth Ni 3000 6.0%
Cr 80 140 ppm
Fe 10 50 ppm
Mo 0.2 <1 ppm
Salt - Bi 225

 

aC. J. Klamut ef al, “Material and Fuel Technology for an
LMFR,” pp. 433-72 in Progress in Nuclear Energy, Senes IV, vol. 2,
Technology, Engineering and Safety, Pergamon, London, 1960.

I-

T~y

 

len

IS 500X

n

=

 

 

;

Fig. 14. Microstructure of stressed Hastelloy N specimen from loop 19 exposed to LiF-BeF,-ThF4-UF,
(68-20-11.7-0.3 mole %) for 4741 hr at 700°C.

 

 
 

 

22
= Repairs were made to loop:19 and the molybdenum finger, and the system was restarted with fresh salt |
and bismuth with theloop ‘designated as 19A.. Loop 19A operated 24,515 hr; and the weight changes are

shown in Fig. 15. Little if any attack' was seen.-.on . the hottest ‘specimen. Assuming uniform: loss the e
maximum::weight -loss:iwas equivalent to ‘only: 0.02' mil/year.- A modified: Hastelloy N .alloy (Ni-13.0%
Mo0—8.5% C1—0.1% Fe—0.8% Ti-+1:6% Nb) lost less weight than a standard alloy at the same position. We .

attribute this difference to the low iron in the modified alloy, compared with about 5% in the standard
alloy. The chromiur content of the salt increased 194.ppm in' 18,000 hr, while iron decreased 70 ppm;.and
there was no detectable:bismuth in‘the-salt.:During the last exposure period, the specimens in‘the hot leg
w’erje-:inadver“t'é’ntly_{_plaéed 4:in.ilower than usual: This put the bottom ‘specimen into the molten bismuth
during operation. On removing the stringer, this bottom specimen .was missing. The end ‘of the stringer
appeared fo:have been'in-contact with the bismuth but was not dissolved. Thus it is likely that the Very
small specimen ‘attachment pins-dissolved during'the run, and the-specimen fell off the stringer. As we had
surmised earlier in loop 19, the bismuth in contact with the salt appearéd to have had no effect on our inass
transfer results as loops 18 and 19 showed about the same amount of mass transfer (Fig. 16).

Loop 18 operated a total of 15 930 “hr and the maximum welght loss was 1.5 mg/cm? at 704°C,
equivalent to a corrosion. rate-of 0.05. mll/year Chrormum in the salt mcreased from 20 to 190 ppm while
iron decreased from 80 to: 20 ppm o -

Pomer ey

ORNL~DWG 71-6990CR

/cma)

D HASTELLOY N

o

© WEIGHT CHANGE (mgq
o

 

0 2000 4000 6000 " 8000 7 10,000 12,000 " 14,000 * 16,000 ' 18,000
OPERATING TIME (hr) ' ‘
Flg 15. Weight changes of Hastelloy N from loop 19 exposed to LlF-Ber-ThF4-UF4 (68-20-11.7-0.3 mole %) for
vanous times. .

ORNL-DWG T0-4934R

N w

1
-

- WEIGHT CHANGE (mg/cm2)
o o '

 

|
o

0 1000 2000 3000 4000 5000 -
S OPERATING TIME (hr)

.. Fig. 16, Comparative corrosion rates of Hastelloy N in LiF-BeF;-UF4-ThF salt with and without bismuth.
 

 

1)

*

"

w

 

23

Loop 21. Loop 21, a Hastelloy N thermal convection loop (Fig. 17) with removable specimens in each
leg and containing the Molten-Salt Reactor Experiment reference fuel salt (Table 1), operated for 13,798

h: The loop was equipped with electrochemical probes which measured the U**/U* ratio, and hence

provided an instantaneous measure of the oxidation potential of the salt. The probes were designed and
operated by J. M. Dale and A. S. Meyer of the Analytical Chemistry Division. Previous use of these probes
had been limited to small static systefné, and this experiment was intended to evaluate their possible use for
on-stream analysis in a large system. This loop differed from our other thermal convection loops in that the
salt flow up the hot leg went directly into_é large surge tank. In our other loops the salt was circulated only
through tubing which was in a harp-shaped:‘ configuration. From the large surge tank the salt then flowed
out the bottom of the tank into a crossofier. leg and into the cold leg. The electrochemical probes passed
through mechanical fittings (separated from the metal by Teflon), which were on risers 10 in. from the
surge tank (to protect the Teflon from the heat of the salt) and then entered the salt in the large hot leg
surge tank, as shown in Fig.'18. There were five access ports for the probes.

The loop was filled on July 20, 1971, and held under isothermal conditions at about 650°C until July
26, at which time the cold leg temperature was established and Hastelloy N specimens were inserted into
both the hot and cold leg of the loop for corrosion studies. Analyses of the U** concentration of the melt
were started on July 22 and were continued on a routine basis until the specimens were removed from the
loop on August 23. More details about these analyses can be found in refs. 15, 16, and 17.

The original measurements in the melt indxcated that about 0.02% of the total uranium was present as
U3*. The U concentration showed a gradual increase until a value of 0.05% was reached at 475 hr of loop
operation. At this point the rate of formation of U increased, and the concentration reached a value of
about 0.15% when the metal specimens were removed from the loop. The maximum weight loss of a hot leg
specimen was 0.2 mg/cm?, and the maximum weight gain of a cold leg specimen was 0. 05 mg/cm?

To evaluate whether material floating on the salt was affecting the potential measurements, the
electrode assembly was modified to allow the salt to be flushed from around the electrode with helium.
This assembly was installed in the loop after the metal specimens were removed. The electrode was encased
in a '-in. nickel tube, open at the bottom, whlch protruded below the surface of the melt. During
computer-controlled runs the salt entered the electrode compartment from below the melt surface and
remained there for the time of measurement The salt was then flushed from the compartment, and the
electrode remained in a helium gas envelope when not in use. This appeared to keep floating material away
from the electrode and also should have increased the llfetlme of the electrodes or electrode area-defining
insulators which may be added at a later tlme :

After the system was started the analyses were performed unattended through computer control. At
initiation of the analyses a signal from a logic circuit that we added to the computer started a timer which
caused the pressure in the electrode assembly to be released, and salt entered the electrode compartment.
The computer then operated the voltammeter to perform five analyses for U:"*, and the results plus some
diagnostic information were printed out on a Teletype after each analysis. At the end of the run the average
U3 concentration and standard deviation were calculated and printed out. When the run was completed,

‘the timer shut off causing the salt to be flushed from the electrode assembly. As an example of the

precision of the analysis system, a standa_rd deviation of 2.1% was obtained for the results of the hourly

‘analyses made over a one-weekend period when the U** concentration appeared to be relatively stable.

 

15. J. M. Dale and T. R. Mueller, MSR Program Semiannu. Progr. Rep. Feb. 28, 1971, ORNL4676, p. 138.
16. J. M. Dale and A.S. Meyer, MSR Program Semiannu. Progr. Rep. Aug. 31, 1971, ORNL4728, pp. 69-70.
17. J. M. Dale and A. S. Meyer, MSR Program Semiannu. Progr. Rep. Feb. 29, 1972, ORNL-4782, pp. 77-79.

 
 

 

24

ORNL-DWG 68-3987R3

     
      
   
   
 

STANDPIPE

 

—TWIN
BALL VALVES

 

CLAMSHELL

 
 
 

 

 

 

 

HEATERS
30in. .
L/INSULATION
CORROSION
SPECIMENS
i |
| i
‘ SAMPLER
\
0
L b1
INCHES

 

 

 

- FLUSH
, TAUNK : buwmp

TANK

 

 

 

 

 

 

Fig. 17. Schematic of loop 21.
 

 

 

»

»

*»

*

_hr and are plotted as the squares in Figs. 19 and 20.

 

Fig. 18. Upper portion of loop 21.

The percent of the total uranium whlch was present as U for the first 3600 hr of loop operation is
shown in Figs. 19—21. The irregular appearance of the data at 600 hr was due to large temperature
fluctuations in the electrode tank. The first measurements with the shielded electrode were made at 1200

At about 1250 hr it was first noted that analyses with the shlelded and unshlelded electrodes gave

'dlfferent values of the U* concentration. This was later shown to be an effect of the nickel tube around

the shielded electrode and not to a difference in the electrode response. At 2350 hr the nickel tube shield
was raised out of the melt, and the measurements w1th the two electrodes were in agreement as shown in
Fig. 20.

The second time the specimens were removed was November 29 1971. The specimens had been
exposed a total of 2513 hr, and the maximum weight loss was 0.2 mg/cm? while the maximum weight gain
was 0.05 mg/cm?. The last time the specimens were removed was February 8, 1972. The total exposure
time was 4169 hr, and the maximum weight loss was 0.23 mg/cm?, which would correspond to a corrosion

 
 

 

26

 

 

ORNL- DWG. 71-14377R
¥ ' T

 

 

 

 

 

 

LOGP OPERATION, hours

Fig. 20. Uranium(II) in loop 21 fuel salt, 9/8/71 to 10/28/71.

T T T T T T T 1 ]
030 g
o2sf 3

£ [ .

~ 0.20+ .
& i ]

. D |, . i
5 >°F < :
o - -
= & F ]
a o040 ¥ -

[ e ]
L :? * .
0.05F ‘,,w“-"f" .
0 [ 1 1 1 | i I L 1 1 ! 1 ]
0 200 400 600 800 ' 1000 1200
LOOP OPERATION, hours :
Fig. 19. Uranium(@H) in loop 21 fuel salt, 7/20/71 to 9/8/71.
- . ORNL-DWG. 71-14378R
0 T T L— T | ' | ' | '
030 -
0.25 ]
= 2 i
-g}a" ]

- [ ]
2020 -
") - -

2 : :
; i +d 3]

8 Q.15 ~ . ~
i : ‘ W, o W * "
[T} - + + * -
e 010 | s vt W ¢ —

C a ]

C o ° ]

A o :

005 _— : b _:

- ol ! : | P . L . 1 ; .
{1200 1400 1600 1800 2000 2200 2400
 

 

 

3

o

v

"

 

27

- ORNL-DWG, 71-14379R
Y 1 1 T T T T T T T T

 

 

 

 

030 " -

'-' + * ?.-::{::’ + :

- {!{’.‘?, * r

025 |- o, T 3

B . ha i

L + *t -

3 020 At ..:

< - ’ ]

+ N .
) +*

) r .

= 045 - —

& f ]

o i .

@ [ 1 ]

ld_" 0.10 [ * -':

b 4

L W‘A“,. b .

0.05 - -

0 [ ) ] 1 1 ] ] 1 1 ) ] 1 ]

2400 3600

2600 2800 3000 3200 3400
: LOOP OPERATION , hours

Fig. 2k. Uranium(IIl) ifl loop 21 fuel salt, 10/28/71 to 12/17/71.

rate of 0.02 milfyear, assuming uniform material removal. The chromium content of the salt increased 40
ppm. Except for the first time, when the U3 concentration was already at a low value, the insertion of the
metal specimens appeared to introduce an oxidant which caused a decrease in the U* concentration (1316
hr on Fig. 20 and 3200 hr on Fig. 21). Special precautions were taken for the next specimen insertion to
determine if this operation could be performed without oxidizing the U%. At 1725 hr the shielded
electrode, which had previously been removed for examination, was put back into the loop without
immersing it into the salt. As shown in Fig. 20, the U3* concentration decreased and then rapidly recovered,
probably due to dilution effects caused by the salt flow.

- In summary, the results of the ratio measurements generally showed excellent reproducibility and were
consistent with the known factors of loop operation. Despite certain operational problems such as the
offset in the potential of shielded electrodes, smaller offset voltages observed between the reference and
working electrodes, and the establishmént of accurate calibrations for the determination of corrosion
products, experience with this test system gives confidence that this relatively sxmple transducer system can

‘be widely applied to other test systems and ultimately to in-line reactor streams.

Loop 16A. Loop 16A was operated to determine the kl_netlcs of tellurium transp'ort in a temperature
gradient loop system where the tellurium is introduced as a metallic coating on'Hastcll.oy N. The reason for
the experiment was the discovery of cracks in stressed Hastelloy N that had been exposed to fuel salt in the .
MSRE,'® and subsequent evidence that tellurium entering salt as a fission product could cause cracks of
this nature. The test bed for this experiment was a Hastelloy N thermal convection loop that had previously

‘operated ~38,000 hr as loop 16A with the fuel mixture 65.5% LiF—34% BeF,—0.5% UF, (mole %).

Operatmg conditions for loop 16A are noted in Table 1.
“For the initial experiment (which lasted 3400 hr), a Hastelloy N specxmen coated with 1.6 mg/cm?

tellurium was added to the normal complement of standard Hastelloy N specimens. In the first 1300 hr of

operation the weight loss of the tellurium-coated specimen equaled the amount of tellurlum coated on the
specimen. In the last 2100 hr the weight loss of the tellurium-coated specimen was the same as the hottest

 

18. H.E. McCoy and B. McNabb, Intergranular Cracking of INOR-8 in the MSRE, ORNL-4829 (November 1972).

 
 

 

28

Hastelloy N specimen located in a neighboring position at the same temperature. Thus, all the tellurium was
apparently removed in the first 1300 hr. Half the tellurium removed was found in the salt and half on a
specimen in the cold leg. These results indicate that tellurium has some solubility in the salt and is capable
of being redistributed by temperature-dependent mass transfer, The specimens were removed from the loop
and were stressed and examined metallographically for cracks. No cracks were found. _

We then placed radioactive tellurium-coated Hastelloy N specimens in the hot and cold leg of the loop
and we used weight change measurements and radioactive analysis to monitor the movement of the
tellurium. The isotope used was ' 27™Te, and it was electroplated on the specimens. Specimens were placed
in the vapor space, in stagnant salt, and in the circulating salt of the loop. The specnnens were removed
after 42, 159, 396, 701, 1318, and 1792 hr exposures. We expected no corrosion of the Hastelloy N by the
fuel salt in these time periods, so all weight changes were attributed to movement of the tellurium.

After the first 42 hr in the loop, all specimens lost at least 66% of their initial activity. Our standard
specimen which had remained at room temperature in air lost 4% of its activity. After 159 hr in the loop,
all specimens had lost 75% of their initial activity. The standard specimen had lost 8%. In the next two
weighing and measuring periods, we saw an increase of activity on the cold leg specimens and a continued
decrease of activity on the hot leg specimens. In the last two examination periods, there were almost no
weight or activity changes. Weight changes correlated quite well with the activity measurements; specimens
in the hot leg lost weight, and specimens in the cold leg gamed weight.

“Therefore, after large initial activity losses, we observed mass transfer of tellurium from the hot leg to
the cold leg analogous to our normal mo'vementv of chromium in a Hastelloy N—salt systeni. In a Hastelloy
N system, until an equilibrium between the salt and the alloy is established at one point in the system, all
points of the loop lose material. After this equilibrium is established at the balance point, losses are seen in
the hot leg accompanied by gains in the cold leg. The tellurium in this system behaved in somewhat the
same manner. |

Type 304L Stainless Steel

Loop 1258. Loop 1258, constructed of type 304L stainless steel (11.07% Ni, 18.31 % Cr, 0.028%C, bal
Fe) and containing removable corrosion specimens in the hot leg, operated nine years (80,000 hr) and thus
provided long-term stainless steel—fuel salt compatibility data. An LiF-BeF, -ZrF4-UF4-ThF4 (70-23-5-1-1
mole %) salt circulated in the loop at a maximum temperature of 688°C and a minimum temperature of

, 588°C. The loop w1th its one surge tank is seen in Fig. 22. Corrosion specimens were present in the hot leg

only.

The loop was started July 29, 1963. Four of ten specimens were removed October 31, 1963 after 2244
hr exposure and are plctu;ed in Fig. 23. The weight losses were 11.6, 11.3, 8.1, and 3.3 mg/cm?. In terms
of uniform material removal, the maximum corrosion rate was over 2 mils/year. Figure 24 shows a
micrograph of a speéimen removed after 15,000 hr at 680°C. We see a maximum of 2 mils grain boundary
penetration. In January 1967 after approx 30 000 hr, the remainder of the original specimen set was
removed, and a second set, consisting of ten new specimens, was placed in the hot leg. The hottest specimen
was replaced after 3700 hr, and the rest of the specimens remained in the loop (periodic removals for
examination) until its shutdown. The specimen exposed at the highest temperature, 688°C, was removed
for detailed metallurgicél arialysis after 5700 hr (between May 1967 and February 1968). Figure 25 shows
all the spec1mens at this time (9400 hr exposure) No gross corrosive effects are evident. The microstructure
of the hottest specimen is seen in Fig. 26. Voids extend into the matrix for a maximum distance of 25 mils
along grain boundaries. The microprobe analysis of this specimen for chromium and iron is shown in Fig.
27. Note that the chromium concentration of the .alloy is decreased for a distance of 28 um. Thus
chromium depletion does appear to be occurring in this system. \
 

 

Fig. 22.

 

Loop 1258.

 
 

 

 

 

30

PHOTO 11 820

-

 

0.035 INCHES
INT100X

Je

 

 

 

_J

Fig. 24. Microstructure of type 304L stainless steel specimen exposed to LiF-BeF 5-ZrF 4-ThF4-UF, (70-23-5-1-1 mole
%) for 15,000 hr at 680°C. '

 

 
31

 

»

 

|
i

 

Fig. 25. Type 304L stainless steel specimens from loop 1258 exposed to LiF-BeF,-ZrF,-ThF 4-UF 4 (70-23-5-1-1 mole
%) for 9400 hr at temperature ranges from 688°C (top specimen) to 668°C (bottom specimen).

 

 
 

 

 

32

After 62,600 hr operation of the loop (September 1970) we replaced the heaters and thermocouples of
the loop. Operation of the loop had become increasingly difficult as heater and therinocouple failures
became more frequent. When the heaters were removed, we found many green chromium oxide crystals on
the loop exterior directly under the heaters. These crystals were the result of a fairly common phenomenon
in Nichrome-wound heaters where chromium, because of its high vapor pressure, is selectively vaporized
from the heater wire over a long period of time. This vaporization changes the heating characteristics of the

 

 

 

Fig. 26. Microstructure of type 304L stainless steel specimen for loop 1258 exposed to LiF-BeF,-ZrF4-ThF4-UF,
(70-23-5-1-1 mole %) for 5700 hr at 688°C.

ORNL-DWG 72-1121

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

100 T
I |
80 i
..
‘ o I .
O—O-U-O—U—O—Q-O-Q—O-U-Q—O-O-O-O-O Fe
¥ 60 i '
z | |
2 402 |
-
=
© 30 |
=
8 | - Cr
s O=O~0=0=0r-O~OO—0—0—O OO~ OO O-O—py- OO0 575
o 20 |
]
10 I
o |
0O 28um 100 200 300

DISTANCE (zzm)

Fig. 27. Chromium and iron concentration gradient in a type-304L stainless steel specimen from loop 1258 exposed to
LiF-BeF »-Z1F 4-ThF 4-UF 4 (70-23-5-1-1 mole %) for 5700 hr at 688°C:

o
 

 

 

.

 

33

wire and certainly contributed to some of our operating problems. The salt was successfully thawed and
loop operation continued. |

The weight changes of the second specimen set after 32,000 hr are seen in Fig. 28. Note the dependence
of weight loss with time and temperature. The microstructures of three specimens after 45,724 hr salt
exposure are shown in Fig. 29. Voids exténd for at least 10 mils into each specimen. At loop shutdown the
second specimen set had been exposed to the salt for 49 ,057 hr. The maximum weight loss was 95.9
mg/cm?, equivalent to a uniform corrosion rate of 0.86 mil/year.

For both specimen sets, the weight losses and depth of void formation increased approxnnately in
proportion to the one-half power of the operating time. This suggests that the corrosion process was
selective toward chromium and was controlled by the solid-state diffusion rate of chromium.

Maraging Steel

Because of interest in iron-base alloys with lower chromium contents for possible containment of
molten salts, we tested a maraging steel (12% Ni—5% Cr—3% Mo—bal Fe) specimen in loop 1258. The
specimen was annealed 1 hr at 800°C and air cooled. Table 7 compares the weight loss and corrosion rate of
this specimen with type 304 stainless steel and Hastelloy N under similar conditions of exposure. As
expected, the maraging steel showed better corrosion resistance than the stainless steel because of its lower
chromium content. However, the corrosion behavior of the Hastelloy N was still superior to both.

Figure 30 shows the metallogréphic appearance of the maraging steel specimen after exposure for 5700
hr at 662°C. Most of the attack is manifested as surface pitting, but there are also subsurface voids such as
were seen in the stainless steel specimens of this loop. The void formation is attributed to removal of
chromium atoms from the surface. This results in a concentration gradient and causes like atoms from the
underlying region to diffuse toward the surface, thus leaving behind a zone enriched with vacancies. In time
these vacancies accumulate and form visible voids.

The specimen of maraging steel was examined using the electron-beam microprobe.!® Using an
accelerating voltage of 20 kV and specimen current of 0.01 #A, x-ray intensities were recorded (Fe Ka, Ni
Ka, Cr Ka, and Mo La) in both as-received and corroded specimens. Since x-ray intensities of a given
element are directly related to the weight percent of the element, a comparison of intensities should reflect
the change in composition in the attacked zone in the corrosion specimen. Table 8 shows the comparison
for each of the four elements. Figure 31 shows the cathode ray tube display and corroborates the
quantltatwe data. One can see the depletlon of iron and chromium and the increase in nickel. Molybdenum

‘ seerns to remain about the same.

 

 19. Microprobe examination made by R.S. Crouse, H. Mateer, and T. J. Henson, Metals and Ceramics Division,

Table 7. Companson of weight losses of alloys at
approx 663 °C in similar flow fuel saltsin a

 

 

temperature gradient system
f Al ‘ Weight loss (mg/cm?) Average corrosion’
d 2490t~ 3730hr rate (mils/year)
' Maraging steel 30 48 ' 0.55
Type 304 stainless steel 6.5 10.0 1.1
Hastelloy N 0.4 0.6 0.06

 

 
10

WEIGHT L 0SS (mg/cm2 )
o
o B

S
Q

80

 

 

LOOP 1258

® EQUIVALENT TO 4mii/year
o EQUIVALENT TO 1.5 mil/year

IMEN

{
REPLACED : WEIGHT
CHANGES CONTINUED

2 4 6 8 10 12 14 1€ 18 20 22 24 26 28
SPECIMEN TIME [N SYSTEM (1000 hr)

~ Fig. 28. Weight changes of. type 304L stainless steel specimens from loop 1258 exposed to LiF-BeF;-ZrF 4-ThF4-UF,4
(70-23-5-1-1 mole %) for various times and temperatures.

ORNL-DWG 68-6087BR2

 

 
 

 

 

 

 

 

i
i
!
1
|

i
|
|
!

 

[}

-n

35

0.038 INCHES
100X 10040 in.

10.030 in.

5o m

 

100X

 

0.033 INCHES

I

oo

 

2 v-118200 .

 

19010 in.

 

0.035 INCHES
100X

 

oo

Fig. 29. Microstructures of type 394Lf'stainle.§s steel specimens from loop 1258 exposed to LiF-BeF;-ZiF 4-ThF4-UF,
(70-23-5-1-1 mole %) for 45,724 hr at (a) 685°C, (b) 679°C, and (c) 674°C. Reduced 28%.

 
 

 

36

'-

Iro

 

[

0.007 INCHES
500X

o

1=

 

 

Fig. 30. Microstructure of a maraging steet (12% Ni—5% Cr—3% Mo—bal Fe) exposed to LiF-BeF,-Z1F4-ThF4-UF,
(70-23-5-1-1 mole %) for 5700 hr at 662°C. As polished.

Table 8. X-ray intensities

 

 

 

of constituents of
maraging steel
Intensity ratios,
Element attack/as received
Fe 0.77
Ni 3.25
Cr 0.32
Mo 1.00
Type 316 Stainless Steel

~ Loop 22. Loop 22 was a type 316 stainless steel thermal convection loop which operated for 4491 hr
with a LiF-BeF,-ThF,-UF, salt mixture identified in Table 1. The harp portion of the loop was 1-in.-OD
tubing with 0.109-in. wall thickness. The material was from heat'X-22662, U.S. National Tube, and the
ORNL Inspection Engineering Number was IR-9141, Y-12 No. 8194. The nominal composition of the alloy
was 16.6% Cr, 13.6% Ni, 2.25% Mo, bal Fe. The loop was bperated to determine the compatibility. of
stainless steel with a well-purified fuel salt. The weight changes of the specimens as a function of position
and temperature are given in Fig. 32. Figure 33 shows the microstructure of the hottest specimen after
1410 hr. Subsurface voids are seen for a distance of 1 mil. After 2842 hr exposure the specimen at 650°C
had lost a total of 6.2 mg/crrll2 , equivalent to a uniform corrosion rate of 0.96 nfil/year.

 

 
 

37

 

 

 

 

 
 

- ., -\fl.flmh\\cmwa l“..a
Dt 9 o1V .
M
M
™~
o
> o
J
., g
>
o
O
0 SF
Tt 2 -
...m«.\...,- ; >
L i b a g
< ‘ , < v
e L 2
a X =
o =
o
X
B
Q

 

 

 

 

 

   

 

 

araging steel. Reduced 10%.

tion of attack zone in m

ig. 31. Charact
 

 

 

38

ORNL-~DWG 73-9846R

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

C4 C5 C6 C7 C8 HB H? HE H5 Ha
10, T T T T | I
T 5 : 1200
Q e — E:
o o = D emEE = < <
E o == Sl X 100 &
£ o= N
¢ — / - Nl N g
r . A . \\\ ‘E_’
5 -5 0 599 hr 1000 &
= / © 1410 hr 2 w
& a 2842hr -
© 4298 hr :
-5
0 10 20 30 40 50 80 70 80 90 100

DISTANCE (in.)

'Fig. 32. Weight changes of type 316 stainless steel specimens in loop 22 exposed to LnF-Bng-ThF4-UF4
(68-20-11.7-0.3 mole %) as a functlon of position and temperature.

 

 

Fig. 33. Microstructure of type 316 stainless steel specimen in loop 22 exposed to LiF-BeF,-ThF,4-UF,
(68-20~ll .7-0.3 mole %) after 1410 hr at 650°C. 500X.

!

The main corrosion reaction in this system is 2UF4.+ Cr = CrF; + 2UF;; thus, in order to determine if
we could decrease the mass transfer of the stainless steel, we added UF; to the salt after the 2842-hr

~exposure measurements to make the salt less oxidizing. In the next 645-hr time period the mass transfer -

was somewhat less, and the overall corrosion rate was 0.83 mil/year. In the final 811-hr time period, the
weight loss of the hottest specimen increased such that the uniform corrosion rate was >1 mil/year.
However, the overail mass transfer, as evidenced by weight changes of other corrosion specimens, remained
quite low. | |

The average uniform corrosion rate for the hottest specimen was 1 mil/year for the entire 4298-hr
period of loop operation. Figure 34 shows the microstructures of hottest and coldest specimens at the end
of the run. Figure 342 shows a 100X cross section of the specimen with voids extending about 1 mil into

&
 

39

 

Y-7 5_690

 

 

 

-
-

i
J
|
]
!

 

Fig. 34. Microstructures of type 316 stamless steel specimens in loop 22 exposed to LiF-BeF,-ThF4-UF,
(68-20-11.7-0.3 mole %) after 4298 hr. Hot leg, 650 °c: (@) cross section of specimen, 100)( (b) as polished, 500X; (c)
| etched, 500X. Cold leg, 560°: (d) as polished, 500X. Reduced 24.5%.

 
 

 

 

40

-the surface. Figures 34b and 34c¢ are 500X views as polished and etched respectively. In F1g 34c¢ note that
the etching reveals a 3-mil-wide affected area. This complete area is probably depleted of chromium. Figure

344 shows the specimen in the coldest position (560°C). Some deposits are seen.

Other Alloys

Loop 18A. After a problem resulting from a heater short to flie loop tubing, loop 18 was drained and

the hot leg replaced. New salt was added, and the loop was designated 18A. The operating conditions for
loop 18A are listed in Table 1. We placed several alloys with improved resistance to cracking by tellurium in
fuel salt in the loop (NCL-18A) to determine their relative compatibility. Tables 9 and 10 show the alloys
tested, their compositions, and the total weight change. Figure 35 shows the time dependence of the weight
changes. The weight change results show a correlation between corrbsion resistance and alloy constituents
that form stable fluorides. There is, of course, a rather good correlation between weight changes and
chromium or iron content. The nickel-base alloys are far more resistant to corrosion than the iron-base

alloy. It is not possible to obtain absolute corrosion information from a test with so many materials

present, but the detrimental effects of increasing chromium and iron are obvious. If the loop had been
constructed entirely of one of the test alloys, the corrosion rate of the alloy might not have been as large as
we noted in our experiment. The less-corrosion-resistant material always suffers in comparison with the
corrosion-resistant material when they are in the same system. |

‘The microstructures of the alloys are shown in Fig. 36. No attack is seen on the Hastelloy N, and there
is little attack on the Inconel 600. You will note a layer of deposit on the Monel with voids underneath.

Table 9. Compositions of test alloys

 

Ni C Fe Mo Cu Al Nb Mn

 

Monel 65 35

Hastelloy N ~70 7 5§ 16

Inconel 600 ~75 1§ 7

Inconel 606 ~70 20 2 2 3
Inconel 601 ~60 23 14 1

Inconel 690 ~50 30 15 _
Incoloy 811E ~30 21 47 “

 

Table 10. Weight changes of alloys exposed to
LiF-BeF,-ThF4-UF4 (68-20-11.7-0.3 mole %)

 

 

at 690°C for 2776 hr
. . Total active
Chromium Iron alloying Weight change
Alloy content - content constituents? (mg/cm?)

Monel 0 1 2 +0.78
Hastelloy N 7 5 14 -0.15
Inconel 600 15 7 25 -0.49
Inconel 606 20 2 27 —6.6
Inconel 601 23 14 38 : -16.0
Inconel 690 30 ‘15 45 —-55.5
Incoloy 811E 21 47 70 -116.7

 

| ‘@Constituents that tend to t‘om(l’ rather stable fluorides. All corrosion
specimgns were annealed 1 hr at 1121 C except Monel, which was annealed 1 hr
at 800 C. '

O
 

-/

 

41

Since the composition of the Monel alloy is most unlike that of any of the other alloys, it is reasonable to
assume that after some initial attack, some material deposited on the Monel by an activity gradient transfer

~ mechanism, thus causing the weight gain that we measured. Thus the Monel was not quite as corrosion

resistant as our weight changes showed. The cross-section view of the Inconel 606 is unusual since at one
end of the specimen we see heavy attack while at the other end almost no attack. This corrosion specimen
was not placed any differently in the loop, so its behavior is not completely understood. We would suspect
that there might be a homogeneity problem in the specimen. The attack of the Inconel 601, Inconel 690,
and the Incoloy 811E was more straightforward and resulted in large areas of void formation: 5§ mils for
Inconel 601, 15 mils for Inconel 690, and 18 mils for Incoloy 811E. |

ORNL-DOWG 73-9847R

 

 

 

 

 

 

 

 

 

 

 

 

 

20 {
- MONEL o
INCONEL 606 |
o | HAST N \— o
__'—'—u—._.__ »
\__—__-_—
_20 \ :INCONEL 600 _| _,
| INCONEL 601
& INCONEL 690
[*]
~
g -40 [— —
W \INCOLOY 8HE
2 \ o
<
L
O
- 60 ‘
I
9 .
w i
© =80 7 — \ ‘
-100 | \
-120 " _
0 1000 2000 3000

OPERATION TIME {hr).
8mg/cm? AT THIS TIME =impy

Fig. 35. Weight changes of various alloys exposed to LiF-BeF,-ThF4-UF, (68-20-11.7-0.3 mole %) at 690°C in loop
18.

 
 

42

"@fi” 1 v-75701

i

 

 

 

 

 

 

 

L e d
e
Pk 8- 4

2%

Y-75699

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Y-75700
-~
Fig. 36. Microstructures of various alloys exposed to LiF-BeF;-ThF,-UF, (68-20-11.7-0.3 mole %) at 690°C for
2776 hr. (2) Hastelloy N, 500X;.(b) Monel, 500X; (¢} Inconel 600, 500X; (d) Inconel 606, 100X; (e) Inconel 601, 100X;
() Inconel 690, 100X; () Incoloy 811E, 100X. Reduced 24.5%. 4
 

 

 

 

43

CONCLUSIONS

1. Temperature gradient mass transfer, as noted by weight losses in the hot leg and weight gains in the
cold leg, was evidenced in all systems tested. :

2. The weight changes of corrosion specimens increased with increasing temperature and time.

3. Some difficulty was encountered in melting LiF-BeF,-ThF, salt which was frozen in a loop. The
difficulty stemmed from its high melting point and the high melting point of certain phases in the salt.

4. Repairs and welds were made on tubing which had been exposed to molten fluoride salts.

5. All the salts tested were compatible with Hastelloy N. ' '

6. Bismuth in contact with a fuel salt had no effect on mass transfer in a Hastelloy N loop.

7. We established that electrochemical methods to determine the oxidation potential of molten
fluoride salts could be used on a thermal convection loop. The values obtained by the electrochemical
methods correlated well with specimen weight change data.

8. We showed that tellurium (as plated on Hastelloy N specimens) does mass transfer in a molten
fluoride salt system and that an equilibrium between tellurium in the salt and tellurium on the alloy can be
established. ‘

9. A type 304L stainless steel exposed to a fuel salt for 9.5 years in a type 304L stainless steel loop
showed a2 maximum uniform corrosion rate of 0.86 mil/year. Voids extended into the matrix for 10 mils.
Chromium depletion was found.

10. The corrosion resistance of a maraging steel (12% Ni—5% Cr—3% Mo—bal Fe) at 662°C was better
than type 304L stainless but worse than Hastelloy N under equivalent conditions, The uniform corrosion
rate was 0.55 mil/year. Voids were seen in the microstructure of the specimens after 5700 hr, and
microprobe analysis disclosed a definite depletion of chromium and iron.

11. Type 316 stainless steel exposed to a fuel salt in a type 316 stainless steel loop showed a maximum
uniform corrosion rate of 1 mil/year for 4298 hr. Mass transfer did occur in the system.

12. For selected nickel- and iron-base alloys a direct correlation was found between corrosion
resistance in a molten fluoride salt and chromium and iron content of an alloy. The more chromium and
iron in the alloy, the less corrosion resistance. | |

 
 

 

 

 
 

45

INTERNAL DISTRIBUTION
(79 copies)

(3) Central Research Library (5)

ORNL — Y-12 Technical Library
Document Reference Section
(10) Laboratory Records Department
Laboratory Records, ORNL RC
ORNL Patent Office ‘
G. M. Adamson, Jr.
C.F. Baes
C. E. Bamberger
S. E. Beali
G. Bohlmann
B. Briggs
Cantor
L. Compere
H. Cook
L. Culler
E. Cunningham
M. Dale
H. DeVan
R. DiStefano
R. Engel
E. Ferguson
H. Frye, Jr.
O. Gilpatrick
R. Grimes
G. Grindell
O. Harms
N. Haubenreich
(3) M. R. Hill
W. R. Huntley
H. Inouye
P. R. Kasten

E.
R.
S.
E.
w.
F.
J.
J.
L.
J.
J.
D.
J.
L.
Ww.
A.
w.
P.

EXTERNAL DISTRIBUTION
(24 copies)

BABCOCK&WILCOX COMPANY, P 0. Box 1260, Lynchburg, VA 24505
B. Mong

 BLACK AND VEATCH, P. O. Box 8405, Kansas City, MO 641 14

C. B. Deering

BRYON JACKSON PUMP, P. O. Box 2017, Los Angeles, CA 90054
G. C. Clasby

ORNL-TM-4286

W. Koger

J. Lawrence
L. Lotts

S. Lundy

N. Lyon

G. MacPherson
E. MacPherson
R. Martin
W. McClung
E. McCoy
J. McHargue
A. McLain
McNabb

E. McNeese
S. Meyer

B. Parker
Patriarca
M. Perry

W. Rosenthal
C. Savage

L. Scott

H. Shaffer
M. Slaughter
P. Smith

A. Strehlow
E. Thoma

B. Trauger
M. Weinberg
R. Weir

C. White

V.

J.

E.
A.
T.
R.
H.
R.
W.
R.
H.
C.
H.
B.
L.
A.
R.
P.
A.
M.
H.
I

I

G.
G.
R.
R.
D.
A.
J.
1.
L Wl]SOl’l

CABOT CORPORATION, STELLITE DIVISION, 1020 Park Ave Kokomo, IN 46901

T. K. Roche

 
 

 

 

46

C

CONTINENTAL QIL COMPANY, Ponca City, OK 74601
J. A. Acciarri '

EBASCO SERVICES, INC., 2 Rector Street, New York, NY 10006
- D. R. deBoisblanc ‘
-T. A. Flynn | -

THE INTERNATIONAL NICKEL COMPANY, Huntington, WV 25720 |
J. M. Martin

UNION CARBIDE CORPORATION CARBON PRODUCTS DIVISION 12900 Snow Road, Parma, OH 44130
R. M. Bushong

USAEC, DIVISION OF REACTOR DEVELOPMENT AND TECHNOLOGY, Washmgton DC 20545
David Elias ‘
J.E.Fox -
Norton Haberman
C. E. Johnson
T. C. Reuther
S. Rosen
Milton Shaw
J. M. Simmons

USAEC, DIVISION OF REGULATIONS, Washington, DC 20545
A. Giambusso

USAEC, RDT SITE REPRESENTATIVES, Oak Ridge National Laboratory, P. 0. Box X, Oak Ridge, TN 37830
D. F. Cope
Kermit Laughon ‘ ~
C. L. Matthews ‘

USAEC OAK RIDGE OPERATIONS, P. O. Box E, Oak Ridge, TN 37830
Research and Technical Support Division

USAEC, TECHNICAL INFORMATION CENTER, P. O. Box 62, Oak Ridge, TN 37830
(2)

i

-

It

. *u.s. GOVERNMENT PRINTING OFFICE: 1973-747-302/70