ORNL-TM-4188.txt

 

   
  
   
   
   
   
     

ORNL-TM-4188

0L
o

 

EFFECT OF FeF, ADDITION ON
'MASS TRANSFER IN A
HASTELLOY N—LiF-BeF, -UF,
THERMAL CONVECTION LOOP SYSTEM

J. W. Koger

THIS DOCUMENT CONFIRMED As

NCLASSIFIED
DI
BYVISION OF CLASSIF ICATION

DATE ZLle /75

 

 

 

 
 

 

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,

 

 

 

 

O

Ty
 

N

o

»

»

)

0

ORNL-TM-4188

Contract No. W-7405-eng-26

METALS AND CERAMICS DIVISION

EFFECT OF FeF, ADDITION ON MASS TRANSFER IN A
HASTELLOY N — LiF-BeF,-UF, THERMAL CONVECTION LOOP SYSTEM

J. W. Koger

 

NOTICE

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
theit contractors, subcontractors, or their employees,
“makes any warranty, express or implied, ot assumes any
legal liability or responsibility for the accuracy, com-
pleteness or uscfulness of any information, apparatus, | .
product or process disclosed, or represents that its use
would not infringe privately owned rights.

 

 

 

DECEMBER 1972

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

DISTRIBUTION OF THIS DOCUMENT IS UNLIMITE

 
 

 
 

 
 

L]

=~

[

CONTENTS
ADSIACT . ... oo it e e et i 1
Introduction . . .. ..o i e e e e et i e 1
Background . ... ... e e e e et 1
Experimental System ...........iiiiiiiiiii i e i e e 8
Results and Discussion .. ... ... ..ottt iiit it ieaiiet s insanaeeanaeaesnaaanns 11
003 173 113 Lo 3 T3P 19

1ii

 
 

 

 
 
 

 

£

i

”

EFFECT OF FeF, ADDITION ON MASS TRANSFER
IN A HASTELLOY N—LiF-BeF, -UF, THERMAL CONVECTION LOOP SYSTEM

J. W. Koger

ABSTRACT

The compatibility of Hastelloy N with high-purity LiF-BeF;-UF4 (65.5-34.0-0.5 mole %) in a
low-flow temperature-gradient system (maximum temperature 704°C, minimum temperature 538°C)
was shown to be quite good. (The maximum corrosion rate was 0.04 mil/year over 29,500 hr of
operation.) Subsequent experimental additions of FeF, increased the mass transfer of the system;
specifically, the maximum weight loss rate before FeF additions was 1 X 10™ mg cm™ hr™!, while
after addition the rate was 6 X 10”2 mgcem™ hr™?

Cracks which transformed into voids were found in the specimens after exposure to the salt
containing FeF,.

INTRODUCTION

The Molten Salt Reactor Program has been concerned with the development of nuclear reactors which
use fluid fuels that are solutions of fissile and fertile materials in suitable carrier salts. A major goal has been
to achieve a thermal breeder molten salt reactor (MSBR). One concept considered was a two-fluid MSBR.
The fuel would be 233UF, or 235UF, dissolved in a salt consisting of LiF and BeF, (66-34 mole %). The
blanket would be ThF, dissolved in a carrier of similar composition. Hastelloy N, a nickel-based alloy used
in the Molten Salt Reactor Experiment (MSRE) was favored as the material out of which the reactor would
be constructed. The design of the two-fluid MSBR showed the fuel salt entering the core at 538°C and
leaving at 704°C.1

As part of our materials program for molten salt reactor development, we studied the compatibility of
Hastelloy N with fuel salt. One such experiment was a thermal convection loop (NCL-16), which was
operated at a maximum temperature of 704°C and a minimum of 538°C. During the operation of NCL-16,
the MSRE was shut down and selected portions were examined. The Hastelloy N removed from the MSRE
apfieared sound, but all metal surfaces that had been exposed to fuel salt showed shallow intergranular
cracking when strained at 25°C.2 We subsequently used loop NCL-16 to investigate the possibility that the
attack in the MSRE was related to the localization of normal corrosion processes to grain boundaries. In
our study of cracking, we twice added 500 ppm FeF, to the loop and exa.mmed the corrosion spec1mens
for signs of cracking.

 BACKGROUND
In the beginning of the Molten Salt Reactor Program, several fluorides were considered as diluents for

the UF, fuel.3:4 After much investigation and consideratidn of nuclear properties and chemical
stability,$-6 BeF, and " LiF were selected as the diluent.

 

1. MSR Program Semiannu. Progr. Rep. Feb. 28, 1966, ORNL-3936, p. 172.

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

3. W.R. Grimes, MSR Program Semiannu. Progr. Rep. July 31, 1964, ORNL-3708, pp. 214-26.

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

5.- Alvin Glassner, The Thermochemical Properties of the Oxzdes Fluorides, and Chlorides to 2500°K, ANL-5750,
Argonne National Laboratory.

6. L. Brewer, L. A. Bromley, P. W. Gfl!es, and N. L. Lofgren, MDDC-1553 (1945); L. Brewer in The Chemistry and
Metallurgy of Miscellaneous Materials; Thermodynamics, L. L. Quill, ed., McGraw-Hill, New York, 1950, pp. 75-192.

 
 

 

ORNL-DWG 71-5270R2

 

I I 'I I I I I I

 

 

  
   

  

 

 

 

 

 

 

 

 

 

 

 

 

 

 

900 - . 500_ - I T T . T =
/848 ' : X (EUTECTIC) = 0,3280+ 0.0004
% Tmax = 459.1 £0.2
800 . as0 |- LIQUID = -4
X (EUTECTIC=
0.531+0.002
700 400 LipBeF, + —
5 LiIF + ' . LIQUID
f; LIQUID .
& 600 }—- 350 ' . — 555
g . 030 035 040 045 0.50  0.55
< ,
W
& / -
m st~ 0 el : ‘ A
= 458.9+0.2°C ! o BeF, (B-QUARTZ TYPE)
i i + LIQUID
| !
400 |- I : -
' ) 363.5 £0.5°C
LiF + Lt - . :
. ' LiBeF3 + Befp
LizBeF, LipBefy + BeF, {8-QUARTZ TYPE) _
300 — ] E" ) 280°C . (a-QUARTZ TYPE)\_
5| LizBeFg + uP| LiBeFy + BeF, (B-QUARTZ TYPE) /
LiBeF3 @ gz7ec
200 l | | i - g | 1 1 ’
0 oA 0.2 03 . 04 05 0.6 0.7 0.8 0.9 1.0

XB.F’ {mole fraction)

Fig. 1. The system LiF-BeF,.

The phase behavior of systems based upon LiF and BeF, as the major constituents has, accordingly,
been examined in detail.”? Fortunately for the molten fluoride reactor concept, the phase diagram of
LiF-BeF, -UF, is such as to make it useful as a fuel. ] : :

The binary system LiF-BeF, has melting points below 500°C over the concentration range from 33 to
80 mole % BeF,.7:8 The phase diagram, presented in Fig. 1, 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, is formed by the reaction of solid BeF, and solid 2LiF-BeF, below
280°C.

The phase diagram of the BeF,-UF, system (Fig. 2) shows a single eutectic containing very little UF,.
That of the LiF-UF, system (Fig. 3) shows three compounds, none of which melts congruently and one of
which shows a low-temperature limit of stability. The eutectic mixture of 4LiF-UF, and 7LiF-6UF,

“occurs at 27 mole % UF, and melts at 490°C. The ternary system® LiF-BeF,-UF,, of primary importance

in reactor fuels, is shown as Fig. 4. The system shows two eutectics. These are at 1 mole % UF, and 52

mole % BeF, and at 8 mole % UF, and 26 mole % BeF,; they melt at 350 and 435 C respectlvely :

Moreover, the system shows a very wide range of compositions melting below 525 C.

The corrosion resistance of metals to fluoride fuels has been found to vary directly with the “noblhty”
of the metal — that is, inversely with the magnitude of the free energy of formatlon of fluorides involving
the metal. Accordingly, corrosion of multiéomponent alloys tends to be manifested by the selective
oxidation and removal of the least noble component. In the case of Hastelloy N, corrosion is selective with

fespect to chromium. The selective removal of chromium by fluoride mixtures depends on various chemical

 

7. R.E. Thoma, ed., Phase Diagram of Nuclear Reactor Materials, 0RNL—2548 (Nov. 6, 1959) ,
8. L. V. Jones, D. E. Etter, C. R. Hudgens, A. A. Huffman, T. B. Rhinehammer, N. E. Rogers, P. A. Tucker, and L. J.
Wlttenberg, *“Phase Equilibria in the Ternary Fused-Salt System LiF-BeF,-UFq,” J. Amer. Ceram. Soc. 45, 79 (1962).
 

A}

.x

o

1100

1000

900

@
o
o

TEMPERATURE (°C)
~
Q
o .

600

500

400

LiF

ORNL~LR-DWG 28598A

 

1100

1000

1
/.

 

 

i/.

o9

 

900

o

 

800

P o UF, + LIQUID

 

TEMPERATURE (°C)

700 £

 

 

’
/
J.
600 [ — ‘
é«o«-mumm-—uu—. —e—¢—so g ——— e

500 —QHIGHBQF2+L|QU D

 

 

 

 

 

 

 

 

 

 

 

 

QHIGHBeFZ +UFR;
i | |

 

10 20 30 40 50 60 70 80 90 UF,
UF4 (mole %)

Fig; 2. The system BeF,-UF,.

ORNL-LR-DWG 47457A

 

"

 

 

 

 

 

 

 

/

 

 

 

 

 

4LiF -UF,

 

\
LiF-UF,
LiF - UF,

 

 

 

 

 

 

 

 

 

0 20 30 40 50 60 70 80 90  UR,

UF4 (mole %)

Fig. 3. The system LiF-UF,.

 
 

 

ALL TEMPERATURES ARE IN °C

£ = EUTECTIC
P = PERITECTIC

[UR] = PRIMARY PHASE FIELD

  
  
 
 
 

LiF - UF,

 

 

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

ORNL-DWG 66—-7634

 
o

a

reactions, as follows:
1. Due to impurities in the melt, for example,
FeF, + Cr=CrF, + Fe (1)

2HF +Cr=CrF, +H, . - (2)

2. Dissolution of oxide films from the metal surface, for example,

2Fe’* (from film) + 3Cr = 2Fe + 3Cr**. L 3)

3. Due to constituents in the fuel, particularly,

Cr+ 2UF4 = 2UF3 + CIF2 . . _ (4)

If pure salt containing UF, (and no corrosion products) is added to a Hastelloy N loop operating
polythermally, all points of the loop initially experience a loss of chromium in accordance with the Cr-UF,
reaction, Eq. (4), and by reaction with impurities' in the salt (such'as HF, NiF,, or FeF,).

Impurity reactions go rapidly to completion at all temperature points and are important only in terms
of short-range corrosion effects. The UF, reaction, however, whose equilibrium is temperature-dependent,
provides a mechanism by which the alloy at high temperature is continuously depleted and the alloy at low
temperature is continuously 'enriched in chromium. All parts of the loop are attacked as the
corrosion-product (CrF,) concentration of the salt is increased by the impurity and UF, reactions.
Eventually the lowest temperature point of the loop achieves equilibrium with respect to the UF, reaction.
Howe}fer, in regions at higher temperature, because of the temperature dependence for this reaction, a
driving force still exists for chromium to react with UF,. Thus, the corrosion-product concentration will
continue to increase, and the temperature points at equilibrium will begin to move away from the coldest
temperature point. At this stage, chromium is returned to the walls at the coldest point in"the system. The
rise in corrosion-product concentration in the circulating salt continues until the amount of chromium
returning to the walls exactly balances the amount of chromium entering the system in the hot-leg regions.
Under these conditions, the two positions of the loop at equilibrium with the salt are termed the “balance

- points,” and they do not shift measurably with time. Thus, a quasi-steady-state situation is eventually

achieved in which there is a fixed chromium surface concentration at each point-in the loop and chromium
is transported at very low rates. This idea is supported by the fact that concentrations of CrF,, UF,, and

‘UF, achieve steady-state concentrations in the salt even though attack slowly increases with time. A

schematic of this mass transfer process is shown in Fig. 5.

Subsurface voids are often formed in alloys exposed to molten salts The formation of these voids is

 initiated by the ox1dat10n and removal of chromium from exposed surfaces. As the surface is depleted in

chromlum chromium from the interior diffuses down the concentration gradlent to the surface. Since
diffusion occurs by a vacancy process and in this particular situation, is essentially nondirectional, it is
possible to build up an excess number of vacancies in the metal. These precipitate in areas of disregistry,
principally at grain boundaries and impurities, to form voids. These voids tend to agglomerate and grow in
size with increasing time and/or temperature. Studies have demonstrated that such subsurface voids are not
interconnected with each other or with the surface. Voids of this same type have also been developed in
Inconel by high-temperature oxidation tests and high-temperature vacuum tests in which chromium is

 
 

 

" ORNL-DWG 67-6800R

 

 

 

 

 

HOT SECTION

DIFFUSION TO SURFACE

 

SOLUTE ESCAPE THROUGH
™ NEAR-SURFACE LIQUID LAYER

 

 

— ~ T DIFFUSION INTO BULK LIQUID

l _ .T__‘_.____,,.__‘__u__v

TRANSPORT
TO COLD PORTION OF SYSTEM

 

\\\\\\\\\\\\\\\\\\\\\\\\\‘l N

 

COLD SECTION —— —_— S
- Al {x SUPERSATURATION — —-i—

 

      

 

* NUCLEATION .
o GROWTH TO STABLE CRYSTAL SIZE

-_-__;OR L

SUPERSATURATION AND DIFFUSION
"THROUGH LIQUID -

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

 

 

 

Fig. 5. Temperature-gradient mass transfer.

selectively removed.? Voids similar to these have also been developed in copper-brass diffusion couples and
by the dezincification of brass.1® All of these phenomena arise from the so-called Kirkendall effect,
whereby solute atoms of a given type diffuse out at a faster rate than other atoms comprising the crystal
lattice can diffuse in to fill the vacancies which result from outward diffusion.

The removal of the least noble constituent is often preferential along grain boundaries. In time, given a
continuing electrochermcal process, this will lead to crevices in the grain boundaries. Diffusional processes
within a crevice may lead to its broadening and ultimately to the formation of pits. However, if the root of
the crack is anodically polarized relative to the walls, knife-line attack will continue. Such a condition may
arise if the walls of the crevice become covered with a very noble material (mckel or molybdenum) This
covering by a noble constituent can occur either by the noble material remaining on the wall when the least
noble constituent is removed or by dlssolutxon of all the alloy constituents with subsequent prempltatlon of

‘the more noble constituents.

 

9. A. DeS. Brasunas, “Sub-Surface Porosuy Developed in Sound Metals during High-Temperature Corrosion,” Metals
Progr. 62(6), 88 (1952).

10. R. W. Balluffi and B. H. Alexander “Development of Porosnty by Unequal Diffusion in Substltutlonal Solutnons
SEP 83, Sylvania Electric Products (February 1952). .

-
 

Table 1. Thermal convection loops that have operated with LiF-BeF,-UF, salts

 

 

 

» Maximum Hours
Salt Alloy temperature Remarks
CO) operated
% LiF-BeF,;-UF, Inconel 6002 _ 677 1000 General subsurface voids to
(53-46-1 mole %) | 2% mils
Inconel 600 677 8664 Heavy intergranular voids to 7 mils
Inconel 600 - 677 8760 Moderate to heavy intergranular
voids to 7 mils
Inconel 600 732 8760 Heavy intergranular voids to
- 15 mils
Inconel 600 732 8760 Heavy intergranular voids to
- 15 mils
Hastelloy N? : 677 © 1000 No attack
Hastelloy N - 677 8760 No attack
Hastelloy N . 732 8760 Light surface pitting
LiF-BeF, Inconel 600 677 1000 Few voids <1 mil
(71-29 mole %) * Inconel 600 677 1000 Few voids <1 mil
Inconel 600 - 677 8760 Light to moderate intergranular
voids to 5 mils
Inconel 600 732 8760 Moderate to heavy intergranular
. ' ' voids to 6% mils
Hastelloy N 677 1000 Light surface roughening
Hastelloy N - 677 3114 Light surface roughening
: Hastelloy N 732 8760 Heavy surface roughening
LiF-BeF5-UFq4 -Inconel 600 677 1000 General intergranular attack
(62-37-1 mole %) <1 mil-
Inconel 600 732 1000 Intergranular voids to 3 mils
Inconel 600 - 732 1000 Intergranular voids < 2 mils
Inconel 600 132 1000 Intergranular voids to 4 mils
Inconel 600 677 8760 Heavy intergranular and general
' voids to 5 mils
) Inconel 600 732 8760 Heavy intergranular voids to 14 mils
* Hastelloy N 677 1000 No attack
Hastelloy N 677 8760 Light surface roughening
Hastelloy N 732 8760 Light surface roughening
LiF-BeF,UF,4 Inconel 600 677 1000 Intergranular voids <1 mil
(60-36-4 mole %) Hastelloy N 617 1000 Light surface roughening
Hastelloy N 677 8760 Moderate surface roughening
LiF-BeF,-UF,4 Hastelloy N 677 1000 No attack
(70-10-20 mole %) Hastelloy N 732 1000 Moderate surface roughening
415% Cr—7% Fe—bal Ni. ,
57% C1—5% Fe—16% Mo—bal Ni..
. Table 1 lists the results of previous Hastelloy N and Inconel 600 thermal convection loop tests using

salts made up of LiF, BeF,, and UF,.11~18 There were no corrosion specimens in the loops, so no weight
change data are available. Yet, it is interesting to compare the behavior of the various salts, the various
alloys, and different times and temperatures. In all cases the Hastelloy N showed better corrosion

 

resistance, and, in general, the higher peak temperature and longer times resulted in greater corrosion. The

11. MSR Quart. Progr. Rep. Sept. 1, 1957 0RNL~2378 p- 3.

s 12.
13.
14.
15.
16.
17.

E/ .

[ 3]

Ibid., Oct. 31, 1957, ORNL-2431, pp. 23-29.

Ibid., Jan. 31, 1958, ORNL-2474, pp. 51 -54.
Ibid., Qct. 31, 1958, ORNL-2626, pp. 53—-53.
Ibid., Jan. 31, 1959, ORNL-2684, pp. 75-76.

Ibid., Apr. 30, 1959, ORNL-2723, pp. 51-54.
Ibid., July 31, 1959, ORNL-2799, pp. 47-55.

Ibid., Jan. 31 and Apr. 30, 1960, ORNL-2973, pp. 33-36.

 
 

salts containing no more than 1 mole % UF, at 677°C only produced light surface roughening on the
Hfiste]loy N. A little more attack was produced at 732°C and by the salts with the larger amounts of UF,.
Based on these results for our specific salt and temperature conditions, Hastelloy N should be quite
resistant to attack. '

EXPERIMENTAL SYSTEM

The thermal convection loop is an excellent corrosion test system.that is intermediate in complexity
and cost between isothermal capsules and pumped loops. The loop is particularly suited for small-scale tests
that involve flow and terriperature gradient mass transfer. The flow in the systerfi results from the difference
in density of the liquid in the hot and the ‘cold leg. A schematic of a thermal convection loop is shown in
Fig. 6, and an actual photograph of loop NCL-16 is seen in Fig. 7.

Thermal convection loop NCL-16 contained 14 specimens, 7 in each leg. Twelve specimens were
titanium-modified Hastelloy N, and two specimens were standard Hastelloy N. Their compositions are given
in Table 2. The loop itself was constructed of standard Hastelloy N. The test specimens were 1.9 X 0.95 X
0.076 cm and weighed approximately 1 g, with a surface area of 3.5 cm?. They were measured to within
0.0025 cm to obtain surface area and were triply weighed to within 0.01 mg. The specimens were attached
by wires to the speéiméh fixture, which consisted of 0.32-cm-diam rod welded to 0.63-cm-OD Hastelloy N
tubing. Salt for analysis was dip-sampled from the harp portion of the loop into a hydrogen-fired copper
container attached to 0.63-<m-OD copper tu'bing. Both the specimen fixture and the copper salt sampler
were lowered into the loop through standpipes. The standpipes consisted of 1.0-in. (2.54-cm) sched 40 type
304L stainless steel pipe with a 1.9-cm ball valve on one end and a sliding Teflon seal at the other through
which the 0.63-cm-OD tubing extended. Before they were opened to the loop environment, the standpipes
were evacuated and backfilled with helium. On removal the épecimen fixture or salt sample was pulled into
the standpipe, isolated from the loop, and allowed to cool to room temperature.

The initial preparation of the fuel salt, LiF-BeF,-UF, {65.5-34.0-0.5 mole %), first involved weighing
and mixing the pure constituents in a nickel-lined container. Two steps were required for purification of the.
fuel salt: one for removal of oxides and sulfides and one for the removal of metallic fluorides. The oxides
and sulfides were removed by gas sparging for several hours at 650°C with an anhydrous mixture of
hydrogen fluoride in hydrogen (1:4). The impurities reacted directly with hydrogen fluoride, and the
process was continued until the same amount of hydrogen fluoride left the reaction vessel as entered. The
reaction was then considered complete.

To remove metallic fluorides, particularly FeF, and NiF,, hydrogen gas sparging of the melt at 700°C
for 24 hr was used. The reduction of CrF, by hydrogen is too slow to be effective at process temperatures,
but analysis of the melt for chromium after sparging indicated a very low concentration. The by-product of
hydrogen sparging is hydrogen fluoride, and the process was continued until the hydrogen fluoride
evolution was below a certain level. ' |

Table 2. Composition of Hastelloy N specimens in NCL-16

 

Weight percent
Mo Cr Fe Si - Mn Ti - Ni .

 

 

Timodified =~ 138 73 <01 005 013 047  Bal
Standard 172 74 45 064 055 002  Bal

 

O
 

ORNL-DWG 68-3987

[ )]

 
  
 

STANDPIPE

 

BALL VALVES

  
    

CLAMSHELL
HEATERS

  
   
 
  

INSULATION

  

CORROSION
SPECIMENS

 
  
     

 

 

 

 

 

SAMPLER

 
 

FREEZE
VALVES

 

 

FLUSH
TANK . . DUMP

TANK

 

 

 

 

 

 

a . Fig. 6. Schematic of thérmal convection loop.

 
 

 

 

10

 

Fig. 7. Natural circulation loops NCL-15 and NCL-16 prior to operation.

»t
-

11

Before filling with salt, the loop was degreased with ethyl;alcohol, dried, and then heated to 150°C
under vacuum to remove any traces of moisture. A helium mass -‘spec'trometer leak detector was used to
check for leaks in the system.

The procedure for filling the loop consisted in 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.

Air was continuously blown on freeze valves leading to the dump and flush tanks to provide a positive salt
seal. All fill lines exposed to the salt were Hastelloy N, and all temporary connections from fill line to loop
were made with stainless steel compression fittings.

The first charge of salt was held for 24 hr in the loop at the maximum operation temperature and then
dumped into the flush tank. This flush salt charge was intended to remove surface oxides or other
impurities left in the loop. The loop was then refilled with fresh salt, and operation was begun. 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. This
temperature difference caused the salt to flow in the loop. A salt sample was then taken, and the specimens
were inserted into the loop. Helium cover gas of 99.998% purity and under slight pressure (approx 5 psig)
was maintained over the salt in the loop during operation.

RESULTS AND DISCUSSION

Prior to its use in the cracking studies, loop NCL-16 operated for 29,500 hr with the fuel salt circulating
in the system. The maximum weight loss for this period was 2.9 mg/cm?, and the largest weight gain was
1.7 mg/cm?. Assuming uniform loss, the maximum corrosion rate was 0.04 mil/year. The chromium
content of the salt had increased 500 ppm, and the iron had decreased about 100 ppm. The changes in
chromium and iron concentration in the salt during the first 12,000 hr are shown in Fig. 8. These changes
suggest that besides the UF, reaction, Eq. (4), the FeF, reaction, Eq. (1), also played a large part in the
mass transfer of chromium in the system. Because the mass transfer involved mainly chromium transfer and
the mass transfer rates were low, it appears that solid-state diffusion of chromium in the alloy controlled
the overall process. Titanium-modified Hastelloy N specimens (Ni—12% Mo—7% Cr—0.5% Ti) had smaller
weight losses than standard Hastelloy N specimens (Ni—-16% Mo—7% Cr—5% Fe) under equivalent

ORNL-DWG €9-4762R

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

300 : ; — "
_ e
Cr 4|

250 .”
= : _NCL-16 :
g //}/ |
£ 200 =
g pre
2150 A
s |/ \ A
[ o
& 100
S —
8 \
- 50 - \‘]\ Fe N -

0 [

 

0 2000 4000 6000 8000 10,000 12,000
: 'TIME (hr) ‘

Fig. 8. Concentration of iron and chromium in the fuel salt in loop NCL-16.

 
 

 

12

4]

ORNL-DWG 68-11T79RA

LCOP 16

N

i

£
76°C 704 °C
: HO;I'TEST SPECIMENS

o Ti-MODIFIED HASTELLOY N
& STANDARD HASTELLOY N

WEIGHT CHANGE (mg/cm?)
' 1
- o

f
N

 

|
o

o 1000 2000 3000 4000 5000 6000 7000 8000 - 9000
TIME (hr) '

Fig. 9. Weight change vs time for standard and titanium-modified specimens in loop NCL-16 exposed to fuel salt
(LiF-B ng-UF4, 65.5-34. 0-0.5 mole %) at various temperatures.

conditions. These differences are seen in Fig. 9 and are generally attributed to the absence of iron in the
modified alloys. Even though iron is more noble than chromium, some iron is removed during the corrosion
process; thus, an iron-free alloy is more resistant to molten salts.

In our investigation of the relationship of salt impurities to the cracking that had been observed in
Hastelloy N for the MSRE, we initially added 500 ppm FeF, to the salt. Specimens were removed and
weighed and portions of specimens were examined metallographically 450 and 1100 hr after the first
addition. Then an additional 500 ppm FeF, was added. Specimens were then examined three times after
this second addition.
~ After each specimen removal, we found weight changes typical of all our temperature-gradient mass
transfer systems, with weight losses in the hot section and weight gains in the cold segtion.'Figure 10 shows
the weight changes of selected specimens as a function of time. (Not shown are the results at the conclusion
of operation after about 36,400 hr.) Figure 11 shows the changes completely around the loop. Note that
the balance point (the point at which there is no weight change) did not shift. Note, also, that the weight
changes after the FeF, additions were relatively large. The changes during the 450 hr after the first addition
equaled those during the previous 10,000 hr. Weight changes during the next 650 hr were two or three
times those for the first 450 hr and were larger than those obtained during 29,500 hr of operation before
any additions.

Optical micrographs of various specimens under different conditions are shown in Fig. 12. Very little
attack or deposit was seen before the FeF, additions. Metallographic examination after the initial FeF,
additions disclosed grain boundary attack, which altered the polishing characteristics of the specimen, but
no cracks were visible. Examination of the hottest specimen 800 hr after the second addition revealed more
grain boundary attack but still no cracks. The surface of the specimen was “lacy” due to severe chromium
- removal from the alloy. This specimen was bent, and some cracking was induced in the depleted area, but
no cracks penetrated the matrix. Specimen examination after 2900 hr exposure to FeF, disclosed that the
weight losses were six times as great as in the previous 29,500 hr operation of the loop, and cracks were
now visible to a depth of 0.5 mil. The cracks seemed to be similar to those seen in the MSRE samples but
were much shallower. Over the next few thousand hours the “cracks” became voids in the hotter specimens
in which no cracks as such were seen. Figure 13 shows an optical micrograph of all the specimens at the end
of the test. No voids were seen in the hottest specimens, as the surface layer was completely removed by
 

o

13

ORNL~-DWG T72-2021R

WEIGHT CHANGE (mg/cm?)

ADDED 500 ppm FeF;,

® STANDARD ALLOY, ALL OTHERS MODIFIED WITH 0.5% Ti

 

o) 8000 16,000 24,000 32,000
OPERATING TIME (hr}

Fig. 10. Weight changes of Hastelloy N specimens exposed to LiF-BeF;-UF, (65.5-34.0-0.5 mole %), with FeF,
added, in NCL-16 as a function of time and temperature.

ORNL-DWG 72-2020
15

 

 

10
HEATED AND INSULATED
& 5
&
g
= 0
o 0 9086 hr
z o 19656 hr
x e 29509 hr
° -5 FeF, ADDED 1300
I . wh—
% o 29957 hr u
< 30606 hr <
® -0 FeF, ADDED 1200 &
* 31445 hr '2
4 32396 hr fi
- 1"
i5 7 00 %.
)
-
-20 _ , —1 1000
o 10 20 30 40 50 60 70 . 80 90 100
DISTANCE {in.)
b -+ et - ‘ -
UPPER COLD LEG BEND LOWER BEND HOT LEG
CROSSOVER (VERTICAL) CROSSOVER (VERTICAL)

Fig. 11. Weight changes of Hastelloy N specimens exposed to LiF-BeF,-UF, (65.5-34.0-0.5 mole %), with FeF,
added, in NCL-16 as a function of position and time.

the corrosion process. Note the deposits on the specimens from the cold leg. The maximum weight loss rate

before impurity additions was 1 X 10™ mg cm™ hr™', while after addition the rate was 6 X 107> mg
-2 o1

cm “ hr.

 
 

 

704°C

676°C

560°C

   

10,686 hr

 

+0.6 mg/em?
500X

29,509 hr

 

14

29,509 hr 29,509 hr
448 hr-with 500 ppm FeF, 1097 hr with 500 ppm FeF,

 

 

2

—2.5 mg/cm
500X

1]
 

15

" ORNL DWG. 73-4386
29,509 hr 29,609 hr 29,509 hr 29,509 hr
1097 hr with 500 ppm FeF» 1097 hr with 500 ppm FeF5

1097 hr with 500 ppm FeF, 1097 hr with 500 ppm FeF,
839 hr with 500 ppm FeF5 1800 hr with 500 ppm FeF 5 5150 hr with 500 ppm FeF, 5800 hr with 500 ppm FeF,

s additional additional additional additional

  

—~19.0 mg/cm?2 —30.2 mg/cm2
500X 500X

~15.5 mg/cm?
200X

[

   

—25.0 mg/em?
500X

Fig. 12. Optical micrographs of specimens exposed to LiF-BeF3-UF, (65.5-34.0-0.5 mole %) at various temperatures

g and times.

 
     

     

91

 

RN T

Fig. 13. Optical micrographs of specimens from various portions of the loop at the end of the test. Approximatety 36,400 hr exposure.

 

 
 

o>

.y

17

- Table 3. Activity coefficients estimated for various fluorides
- in molten LiF-BeF; solutions at 1000°K

 

 

. AF°f P LT g Approximate a
Species ~ {kcal per fiuorme) Equlllbnum stgdled " mole fraction Y
UF, -95.3 UF4 + 2H,0 = UO; + 4HF 0.02 0.55
UF,3 -1014 UF4 + '4H, = UF; + HF 0.0001 50
- FeFy ~665  + TFeFy+Hy=Fe+2HF 0.005 .16
CtFy, -74 . CrFy3+Hy;=Cr+2HF - 0.001 0.5

 

@Based on the crystalline solid as reference state.

‘/3 -in. Hastelloy N specimen rod from the hot leg was inadvertently broken dunng one of the
specimen removal penods The rod had been exposed to the unaltered fuel salt for 10,301 hr, to the salt
with 500 ppm FeF, for 1100 hr, and to the salt with an additional 500 ppm FeF, for 3254 hr. The total
exposure time was 14,655 hr. The rod was sectioned and examined at different positions which represented
different exposure temperatures (Fig. 14). Voids were quite evident at the hottest positions, with the
amount and depth of the voids decreasing with decreasing temperature.

~ Observations of the specimen rods correlate quite well with the speclmen behavior. The balance point
(no gains or losses) as determined from the specimen welght change is about 621°C, so all but one of the
rod specimens should have been attacked, and this is what we found.

It is interesting to compare values for actual mass transfer in a loop system such as NCL-16 with
calculated values based on quite basic chemical equilibrium experiments. Data obtained from studies of
hydrogen reduction equilibria by Long!® and by Blood2? can be used to compute reasonable (and
consistent) estlmates of the activity coefficients of the several species in fluoride salt melts. T hese activity
coefficients based on the crystalline solid as reference state are shown in Table 3.

From these values and the values of AF °f for the several compounds (where AF° =RT 1n K), we may
assess the extent of the reaction

wn+m#&m+wn,,'g*;“7{.g'_' L g @)

where the UF4, CrF,, and UF; are in solutlon in the fuel salt- and the chromitm is at the relatlvely low
activity it has in unaltered Hastelloy N. For this reaction,

auF,’ aCng (N7)UF3 (N‘r)cu-*,

 

K, 24x10fl_ (5)

aum dce. (NY)UF4 acy

If, mmally, the salt were comp!etely pure and the metal contained no oxide (so that all CiF, was
generated by th1s reaction), then

 

19. MSR Program Semiannu. Progr. Rep. Jan. 31, 1964, ORNL-3626, pp. 119-29.

20. C. M. Blood, Solubility and Stability of Structural Metal Difluorides in Molten Fluoride Mixtures, ORNL-CF-
61-54 (Sept. 21, 1961); C. M. Blood et al., “Activities of Some Transition Metal Fluorides in Molten Fluoride Mixtures,”
in Proceedings of the International Conference on Coordination Chemistry, 7th, Stockholm and Uppsata, June 25-29,
1962, Butterworths, London, 1963. :

 
 

18

 

73-4385

ORNL DWG.

 

|
=

u __O0.0_

687°C

1

-

Wl €00'0]  X00S |

S3HINI L00°0

u S000)

Ui

£00°0)
]

 

 

 

Fig. 14. Hastelloy N specimen rod -exposed to LiF-BeF,-UF4 (65.5-34.0-0.5 mole %) for 14,655 hr (4354 hr with

1000 ppm FeF, in the salt).
 

 

s

19

For reaction of the fuel from NCL-16 (Nyf, = ayfr, = 0.005) with Hastelloy N (ac, = 0.083), the
equilibrium indicated in Eq. (6) is satisfied at

Neep, =07X 107 57
and
Nyp, =14X 107,

Accordingly, less than 3% of the UF4 would be reduced to UF3, and the chromium fluoride concentration

- of the melt would be 70 ppm (as Cr). In our system we produced 400 ppm Cr as CrF, in 29,500 hr; thus
‘other sources such as FeF, were available for the oxidation of chromium. For the reaction

FeF, + Cr==Fe + C1F, , ’ (1)
according to ref. 20,

_ NCngNFe

=——2 " -6000, e,
N " NciNrer, 0

where the chromium is in solid solution in the Hastelloy N (¢ = 0.083) and the iron is crystalline iron at
unit activity. Thus, the reaction should proceed until the ratio '

=500, ' (8)

 

NFng

which means that large amounts of CrF; can be produced by very little FeF,. Depletion of chromium at
the surface of the alloy will lower this ratio. If we attribute 70 ppm Cr as CrF, to oxidation by UF4 and
330 ppm Cr as C1F, to oxidation by FeF,, which decreased by about 100 ppm in 29,500 hr, the calculated
equilibrium constant for the FeF, reaction in the first 29,500 hr is 40. During the periods of FeF,
additions the calculated equilibrium constant ranged from 12.5 to 36. Thus the depleted alloy caused the
equilibrium constant to be lowered an order of magnitude.

CONCLUSIONS

1. The compatibility of Hastelloy N with LiF-BeF,-UF, (65.5-34.0-0.5 mole %) which contains few
impurities in a low-flow temperature gradient system (maximum temperature 704°C, minimum
temperature 538°C) is quite good. The maximum corrosion rate was 0.04 mil/year. | _ '

2. Additions of impurities such as FeF; increase the mass transfer of the system; specifically, the
maximum weight loss rate before impurity additions was 1 X 10™* mgem™ hr ™!, while after addition the
rate was 6 X 10”2 mgem ™ hr7!, |

3. Cracks which transformed into voids were found in the specimens exposed to the impurity-laden
salt. However, these cracks were not equivalent in appearance to those noted in Hastelloy N in the MSRE;

thus attack by FeF, was not the primary cause of the crack formation.

 
 

 
 

)
¥

 

 

(3) Central Research Library

INTERNAL DISTRIBUTION
(79 copies)

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
E. Beall

. G. Bohlmann
. B. Briggs

Cook

- B. Mong

EXTERNAL DISTRIBUTION
(24 copies)

 BABCOCK & WILCOX COMPANY P. O. Box 1260, Lynchburg, VA 24505

BLACK AND VEATCH P.O. Box 8405, Kansas City, MO 64114

C. B. Deering

BRYON JACKSON PUMP, P 0. Box 2017 Los Angeles CA 90054

G. C. Clasby

ra&>uwwobhfimg>ww>rw;omwswmwa>mu

ORNL-TM-4188

 

Lawrence

l"[-
c
3 S
g<$

MacPherson

.W.
. J.
. L.
. S.
.N. Lyon
. G.
. E. MacPherson
. R.
.W.

E

g

A. McLain
M

. S. Meyer

. B. Parker
. Patriarca

.M. Perry

. W. Rosenthal
. C. Savage

. Shaffer
. Slaughter
. Smith

. Trauger
. Weinberg

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

T. K. Roche

21

 
 

22

CONTINENTAL OIL 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, Huntmgton 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 20543

A. Giambusso
USAEC, RDT SITE REPRESENTATIVES, Oak Ridge National Laboratory, P. O. 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

O

"y

O,