You are on page 1of 15

UNCLASSIFIED

ORNL-2055

FUNDAMENTALS OF LIQUID-METAL CORROSION


W. D. Manly

.
LEGAL NOTICE
T h i s report

WOI

prspored a s o n account of Government sponsored work.

Neither the United Stotes,

nor the Commission, nor any person octing o n beholf of the Commission:

A.

Mokes

ony warronty

completeness,
any

01

or representation, express

OT

implied,

w i t h respect t o the accuracy,

usefulness of the information contained i n t h i s report,

informotion, apporatus,

p r i v a t e l y owned rights;

method,

or process disclosed

thot the use o f

i n t h i s report moy not infringe

8. Assumes any l i o b i l i t i e s w i t h respect t o the use of,

or for domoges r e s u l t i n g from the use

of

ony informotion, opparotus, method, or process disclosed i n t h i s report.

As used i n the obove, "person a c t i n g on behalf of the Commission" includes any employee or
contractor of the Commission t o the extent that such employee or contractor prepares, hondles
OT

distributes,

or provides o c c c s s to,

w i t h the Commission.

any i n f a m o t i o n pursuant t o h i s employment w contract

CLASSIFIED
ORNL-2055

Contract

No. W-7405-eng-26

METALLURGY DIVISION

FUNDAMENTALS OF LIQUID-METAL CORROSION

W. D. Manly

DATE ISSUED

OAK RIDGE N A T I O N A L LABORATORY


Operated by
UNION CARBIDE N U C L E A R COMPANY
A Division of Union Carbide and Carbon Corporation
Post Office Box P
Oak Ridge, Tennessee
MARTIN MARIE r T A ENERGY SISTCI(IS LIRRPRIES

CLASSIFIED

3 Y 4 5 b 0350095 7

UNGLASS IFIE0
ORNL-2055

Metallurgy and Ceramics

TID-4500 ( I Ith ed.)

INTERN A L DI STRl BU 7-10N

1. C. E. Center
2, Biology Library
3. Health Physics Library
-5, Central Research Library
& Reactor Experimental
Engineering Library
7-26. Laboratory Records Department
27. Laboratory Records, ORNL R.C.
28. A. M. Weinberg
29. L. B. Emlet (K-25)
30. J. P. Murray (Y-12)
31. J. A. Swartout
32. E, H. Taylor
33. E. D. Shipley
34. M. L. Nelson
35. W. H. Jordan
36. C. P. Keim
37. J. H. Frye, Jr.
38. R. S, L i v i n g s t o n
39. R. R. Dickison
40. S. C. L i n d
41. F. L. Culler
4 2 A. H. h e l l
43. A. Holl aender
44. M. T. Kelley

45.
46.
47.
48.
49.
50.
51.

52.

53-75.
76.

77.
78.
79.
80.
81.

82.
83.

84.
85.
86.
87.

88.
89.

K. 2. Morgan
J. A. Lane
T. A. Lincoln
A. 5. Householder

C. S. Harrill
C. E. Winters
D. S. Billington
D. W. Cardwell
W. D. Manly
E. M. King
A. J. Miller
D. D. Cowen
R. A. Charpie
M. L. Picklesimer
M. J. Skinner
G. E. Boyd
C. J. McHargue
P. U Reyling
G. C. Williams
N. J. Grunt (consultant)
E. Creutz (consultant)
T. S. Shevlin (consultant)
ORNL - -12 Technical Library,
Document Reference Section

EXTERNAL DISTRlBUTlQN

90. R. F. Bacher, California Institute of Technology


91. Division of Research and Development, AEC, O R 0
92-467. Given distribution as shown in TID-4500 under Metallurgy and Ceramics

category

DISTRIBUTION P A G E TO BE REMOVED IF R E P O R T I S GIVEN PUBLIC DISTRIBUTION

UNCLASSI FlED

iii

.........

FUNDAMENTALS OF LIQUID-METAL CORROSION


W.

D. Manly

ABSTRACT
The corrosion of structural metals or alloys i n liquid metal i s for the most part due
t o the dissolving of the various constituents of the metals or alloys by the liquid metal.
The manner i n which t h i s dissolution proceeds gives r i s e to many types of attack,ranging
from a simple solution-type attack t o a deep intergranular attack with the preferential
leaching of one constituent of an ~ l l o y , Examples of the many types of corrosion experienced with solid-metal-liquid-metal systems are illustrated. The role of impurities
on the corrosion i s discussed.
In n static one-metal system the solution step stops when the solubility l i m i t i s
reached; however, in more complex systems the corrosion of metals can be continued
because of removal of the dissolved materials from the liquid metal by a temperature
gradient and/or by dissimilar-metal transfer.
Thus the transfer of metal froni the hot
region of a plumhing system and i t s subsequent deposition in the cold zone, and the
transfer of material from one part of the system and i t s alloying with a metal of different
Composition in another part of the system w i l l greatly increase the corrosion as compared t o the results obtained in static systems. Examples of temperature-gradient mass
tronsfer and dissimilar-metal transfer are reviewed, and factors governing the occurrence
of these phenomenu are outlined.

Corrosion has been defined by U. R. Evans as


follows: The word corraTion denotes destruction
o f metal by chemical or electrochemical action; a
familiar example i s the rusting o f iron. Thus, corrosion by t h i s definition denotes the transfer o f
electrons, But t h i s i s not usually the case in liquidmetal attack, and, therefore, to define the destruction o f a solid metal by a l i q u i d metal as corrosion,
this definition must be broadened to allow for the
solution or solubility o f a solid metal in a l i q u i d
metal wherein no transfer of electrons i s involved.
Liquid-metal corrosion for the most part simply
depends on the solution rate and the extent of
solubility of the solid metol in the l i q u i d metal.
However, many coniplicating factors can influence
the solution rate or the attainment of the solubility
I imit. The formation of surface intermetallic compounds 3nd of oxide or nitride films are good examples of such factors.
Other factors are: impurities in the liquid metals which can increase
the solution rate, and temperature gradients and
multimetallic systems which can cause an increase

i n the amount of attack over that expected t o saturate the liquid metal because of the mass transfer
of material under the driving force of the temperature gradient or the concentration gradient.
In the short time that interest hos existed i n
the corrosion characteristics of liquid metals, many
testing methods have been developed t o determine
the interaction between liquid metals and solid
metals. These methods range from the very simple
static capsule test t o the expensive and camplicated forced-circulation loop tests which carefully
reproduce the operating conditions. The tests and
the manner in which data are derived from them
have been previously described.2
Several types of corrosive attack of solid metals
by liquid metals have been ~ b s e r v e d . ~ In t h i s
discussion the types of liquid-metal attack w i l l
be arranged into the following cotegories:
1. simple solution,
2. alfoying between liquid metal and solid metal,
3. intergranular penetration,
4. impurity reactions,

U. R. E v a n s , An Outline of Corrosion Mechunisms,


Including t h e E l e c t r o c h e m i c a l Theory. p 3 of Cotroszon
Handbook, H. J. Uhlig, ed, Wiley, N e w York, N. Y., 1948.

2D. C. Vreeland, E. E. Hoffman, and W. D. Manly,


Nuclponics 11, No. 11, 36--39 (1953).
3A. des. Brasunas, O R N L - 1 6 4 7 (Secret).

5.
6.

temperature-gradient mass transfer,


concentration-gradient mass transfer, or dissimilar-metal mass transfer.
Examples of a l l s i x kinds of liquid-metal corrosion
w i l l be presented and described i n detail i n subsequent paragraphs.
Several variables affecting liquid-metal corrosion
must be considered, and they are: (1) temperature;
-

(2) temperature gradient; (3) c y c l i c temperature


fluctuation; (4) surface area t o volume ratio; (5)
purity of l i q u i d metal; (6) f l o w velocity, or Reynolds
number; (7) surface condition of container material;
(8) number of materials i n contact w i t h the same
l i q u i d metal; (9) condition of the container material,
such as the presence of a grain-boundary precipitate, the presence of a second phase, the state
of stress of the metal, and the grain size.
Temperature i s one of the most important variables, because the higher the temperature the
higher the s o l u b i l i t y of the s o l i d metal i n the
l i q u i d metal. Also, as the temperature increases,
diffusion rates increase, which i s q u i t e important
i n certain types of liquid-metal corrosion.
When a temperature gradient e x i s t s i n a liquidmetal system, the difference i n s o l u b i l i t y of the
dissolved metal i n the liquid metal i n the hot and
c o l d parts of the system gives r i s e t o temperaturegradient mass transfer.
The c y c l i c temperature fluctuation i s helpful i n
explaining erroneous static-corrosion results since
i n a supposedly isothermal condition i n a poorly
controlled furnace the liquid-metal-solid-metal
interface temperature can fluctuate quite appreciably around a mean temperature.
Thus, at the
high temperature, material goes i n t o solution and
subsequently at the lower temperature comes out
of solution and precipitates i n the b u l k liquid, or
forms dendrites, or forms a uniform layer on the
container wall. The C u - B i system i s an example
of this; at 500 t 5OC t h e corrosion rate i s several
times greater than that a t 500 t 0.5OC.
The r a t i o of the exposed metallic surface area
t o the volume of l i q u i d metal i s a controlling factor
i n the amount of corrosion experienced i n a solidmetal container i n a static system since the metal
w i l l corrode s u f f i c i e n t l y t o saturate the l i q u i d
metal a t the operating temperature; therefore, as
the r a t i o of surface area t o volume decreases, the
amount of corrosion increases.
P u r i t y of the l i q u i d metal can have quite an
effect on the rate at which the s o l u b i l i t y l i m i t i s
reached and can markedly affect the wetting tend-

ency of the l i q u i d metal on the s o l i d metal.


T h e f l o w velocity, or Reynolds number, i s only
important i n a forced-circulation pump loop, i n that
a decrease i n the thickness of the lamellar layer
i n the hot and cold legs occurs as the f l o w velocity
increases. Thus, when metal atoms pass from the
hot leg t o the c o l d leg, the d i f f u s i o n paths are
shorter i n these t w o regions.
T h e surface condition of the container material
i s not a too-important variable as long as i t i s
free of surface films. After equilibrium i s reached
between the l i q u i d metal and the s o l i d metal, t h e
surface condition should have no effect, and i t s
primary effect i s t o change the rate a t which the
l i q u i d metal i s saturated. The greater the surface
area that i s exposed, the faster the l i q u i d becomes
saturated.
When t w o or more s o l i d metals are i n contact
w i t h a l i q u i d metal, the phenomenon of dissimilarmetal mass transfer, or concentration mass transfer,
should be considered even when the temperature
i s constant.
Equilibrium i s attained i n a multicomponent, multiphase system only when the chemic a l potential of each component i n any given phase
i s equal to i t s chemical potential i n every other
phase. A redistribution of the components between
t h e phases of the system w i l l therefore tend t o
occur u n t i l the condition of the equality of the
chemical potentials i s satisfied. T h e driving force
for t h i s process, thus, i s associated w i t h the
equalization of the chemical potentials, which
r e s u l t s i n a decrease of the free energy of the
system.
The condition of the container material, such as
grain s i z e and other variables, must be considered
because the w e t t a b i l i t y of individual grains and
t h e wettability of grain boundaries are different.
A continuous carbide or oxide network a t t h e grain
boundary can greatly increase the corrosion that
w i l l be experienced.
I n a l l liquid-metal corrosion tests, thesevariables
must be considered and controlled t o obtain information that w i l l be useful i n the over-all understanding of the s u i t a b i l i t y of s o l i d metals as containers for various l i q u i d metals.
T h e f i r s t example of liquid-metal corrosion t o be
discussed i s simple-solution~type attack.
I n the
case of a pure metal, an example of the even removal of metal from the surface t o saturate the
l i q u i d metal i s illustrated i n Fig. 1, which shows
t h e surface of a titanium specimen after being exI n the case
posed t o lead a t 1000C for 4 0 hr.

of a complex alloy the attack can a l s o be a simplesolution type,


An example of t h i s i s shown i n

Fig. 2, which illustrates the corrosion of type 304


low-carbon stainless steel in sodium after 40 hr
Another attack that might be termed
a t 1000C.
simple solution i s the decarburizing action of
lithium and sodium. Figure 3 illustrates the decarburization of type 430 stainless steel after 40
hr a t 1000C in lithium.

If all the phase diagrams of Iiquid-rnetal-solidmetal systems were available, an ascertainment


could be made of the depth of attack that would

occur in o static system as a result of simple solut i o n by examining the solubility l i m i t of the solid
metal in the liquid metal at the operating temperature.
However, there would be no conception of
the rate a t which the solubility l i m i t was achieved.
Thus, in looking at the F e - L i phase diagram in
Fig, 4, it can be seen that the amount of attack
of iron by lithium should be quite small i n a static,
isothermal system; and corrosion tests have proved
this.
Therefore, in simple-solution type attack
the amount of damage the solid metal w i l l receive
depends on the ratio of surface area t o volume of
the system, but the rate at which the attack occurs

9
1

Fig. 1.
Even-Surface Removal from Titanium
Specimen After 40 hr in L e a d at 1000C.
1OOX.

Reduced

33.5%.

Fig. 3. Decarburization of Type 430 Stainless


750X.
Steel After 40 hr in Lithium at 1000C.

Reduced 33%.

UNCLASSIFIED
ORNL-LR-DWG

14457

4 200
V

e io00
W

W
OI

5k

800
600

400

200

2.

40

304 LowCarbon Stainless Steel by Sodium at 1000C. 75X.


Reduced 34%.
Fig.

Attack After

hr of Type

Li

(0

Fig.

1
20

LITHIUM (SOLIDI

30

40

50

+ IRON (SOLID)]
60

70

80

90

Fe

IRON ( w t %)

4. Fe-Li Phase Diagram.

can be influenced greatly by other variables, such


as impurities in the system,
The next type of corrosion t o be discussed i s
the alloying that occurs between liquid metals and
solid metals.
For t h i s t o result, there must be
some solubility of the liquid metal i n the solid
metal.
The Ag-Pb phase diagram (Fig. 5) i s an
example of a system in which the liquid metal i s
soluble i n the solid metal.
I n some experiments
the liquid metal dissolves considerably in the solid
metal with the formation of an intermetallic compound. When vanadium i s tested for 400 hr i n lead
a t 1000C, an intermetal I ic compound i s formed
between the vanadium and lead, as can be seen

in Fig. 6. When type 446 stainless steel i s tested


i n lead at 1000C, it i s found that after 400 hr
lead has diffused into the alloy predominantly a t
the grain boundaries and has formed a compound
(Fig. 7).
Sodium w i l l penetrate solid copper a t
the grain boundaries and w i l l form an intermetallic
compound, which i s considerably harder than the
base metal, copper, as illustrated i n Fig, 8.
One of the more serious types of corrosion that
can occur i s the deep intergranular penetration
brought about by the removal of one constituent
from an alloy.
The best example of t h i s i s the
selective removal of nickel from austenitic stainless steels, and Fig. 9 shows a type 347 stainless
steel after 400 hr of testing in lead a t 80OOC.
Other examples of t h i s type of carrosion are given
in Figs. 10 and 11, which illustrate the attack on
type 304 low-carbon stainless steel after 400 hr

in lead and i n lithium, respectively, a t 1000C.


considerable portion of the attack i s attributed
t o the removal of nickel caused by the alloying

Fig. 6. Vanadium (at Left) Tested for 400 hr in


Intermetallic compound formed
L e a d at 1000C.
between the vanadium and lead. 200X. Reduced

14%.

UNCLASSI FlED
ORNL-LR-DWG (4458

SILVER ( a t % )
50 60
70

IO 20 30 40

1000

80

90

900

I400

-2
W

1200 Ej
1000

?I

800

600

Pb

IO

20

30

400
40

50

SILVER ( w t

60

70

80

90

. 1

Ag

%I

Fig. 5. Ag-Pb Phase Diagram.

Fig. 7. Type 446 Stainless Steel Tested for 400


hr in L e a d a t 1000C. B O X . Reduced 32%.

uNCLASS1FI ED
Y-7067

-.

"61

Y- 144?7

'

c
ii

01

A
b

Fig. 8. Copper Capsule-Sodium T e s t a t 18OOOF for 100 hr. Sodium penetration into s o l i d copper at
t h e grain boundaries with the resultant formation of an intermetallic compound.

Fig. 9. Selective Removal o f Nickel from Type


347 Stainless Steel After 400 hr i n L e a d at 800C.
250X. Reduced 35%.

Fig. 10. Type 304 Law-Carbon Stainless Steel


After 400 hr i n Lead at 1000C. Removal of nickel
from the alloy shown, 1OOX. Reduced 33%.

of the nickel with the iron container wall, and it

32.5%.

I n l i q u i d metals, impurities, such as oxygen,


nitrogen, and carbon, can have an appreciable
effect upon the rate of attack, and in some cases
the whole mode of attack t a n be changed because
of the effect of the impurity on the surface tension
or because of the reactivity of the impurity. An
example i s the attack of stainless sfeels by lithium
when nitrogen i s the principal contaminant. Figure
13 illustrates the shallow attack produced by pure
lithium a t 8WoC i n 100 hr on type 316 stainless
steel. However, if the lithium i s contaminated w i t h
a small quantity of nitrogen, the complete tube
wall, comprising 0.035 in., w i l l be penetrated by
the lithium during the same type of test. T h i s i s
due t o the nitrogen-contaniinated lithiurn reacting
with the carbides that form the grain boundary network, since a test i n type 316 stainless steel with
lithium that was Contaminated with nitrogen resulted in shallow attack when the testing temperature was above the solution temperature of the
carbides. The true effect of nitrogen on cqrrosion
by I ithium. i s not understood.
High-temperature
alloys can be severely carburized by I iquid metals,
especially sodium and lithium, if the liquid metols
have been stored under kerosene or have acquired
carbonaceous material from some other source.
The carburization, after a long service period, of
type 347 stainless steel by sodium Contaminated
with carbon i s shown in Fig. 14. I n corrosion by
sodium, oxygen impurities can have an appreciable
effect on the rate a t which the solubility l i m i t i s
attained. In lead the oxygen contamination, if anything, decreases the rate of corrosion since most

Fig. 12. Hot L e g of a 75% Ni-25% Mo ThermalConvection L o o p Operated for 200 hr with L e a d at
800C and with a Temperature Gradient of 300OC.
150X. Reduced 35%.

Fig. 13. Very Shallow Attack Produced by Pure


L i t h i u m at 800C in 100 hr on Type 316 Stainless
Steel. 250X. Reduced 33%.

i s assumed that the attack would have been less


if the specimens and containers had been the same
material.
The selective removal of nickel from
a 75% Ni-25% Mo alloy also occurred i n a sample
from the hot leg of a thermal-convection loop which
operated for 200 hr w i t h lead a t 800C and with
In t h i s
a 300OC temperature gradient (Fig. 12).
case the nickel was preferentially removed from
the hot zone and deposited i n the cold zone of the

Ioop.

..

Fig. 11. Type 304 Low-Carbon Stainless Steel


Removal o f
After 400 hr in L i t h i u m at 1000C.
nickel from the alloy illustrated. 25OX. Reduced

yet, the rate-controll ing step in temperature-gradient


mass transfer has not been found,
To obtain more data on mass transfer i n liquid
lead, a series of quartz thermal-convection loops
were operated with various alloys and the elements

UNCLASSIFIEI
y-na6

1
'

alloy^.^

comprising the

Results of these tests

4J. V. Cathcart and W. D . Manly, A Technique / o r


Corrosion Testing in Liquid Lend, ORNL-1737 (Aug. 5 ,

1954).

U NCLASSI FlED
ORNL-LR-DWG
14459

40

LEAD ( a t %1

20

40 50 6070 90

30

2000
1800

1600

LL

1400

3
n

1200
1000

w
n

800

al

600

Fig. 16.
P l u g from an Inconel-Lead ThermolConvection Loop That Operated for 125 hr at 800OC
with a Temperature Gradient o f 200'C.
3.5X. Reduced 11%.

CU

10

Fig.

20

18.

30

40

50

60

70

80

90

Pb

400

LEAD ( w t %I

Cu-Pb Phase Diagram.


UNcLASSlFlEb

ORNL-LR-OW6

5W3

HOT LEG

MELLAR LAYER

COLD LEG

SUPERSATURATION
HUCLEATION

0 GROWTH TO STABLE CRYSTAL SIZE


OR
AN0 DIFFUSION T

Fig. 17. Iron Crystols Formed i n a Type 410


Stainless Steel Thermal -Convection Loop After 40
hr Operation with Lithium at

1000C.

Fig. 19. Various Steps Through Which Metallic


Atoms Must Go i n Temperature-Gradient Mass
Transfer.

200X.

have indicated that the rate-controlling step for


mass transfer by lead a t the v e l o c i t i e s used in
the thermal-convection loop takes place in the hot
leg and i s probably a solution step. For example,
it was found that the formation of an intermetallic
compound, or other type of diffusion barrier, i n the
hot zone greatly increases t h e time that elapses
before a plug occurs i n the I O O P . ~
The l a s t type of liquid-metal corrosion t o be discussed i s dissimilar-metal mass transfer or concentration-gradient mass transfer, The best example
that has been obtained a t the Oak Ridge National
Laboratory i s the interal loying between molybdenum
and n i c k e l (Fig. 20). I n t h i s case the molybdenum
sample was being tested in sodium contained i n
a n i c k e l crucible, and after 100 hr at 1000C a
sufficient quantity of n i c k e l had transferred through
the sodium and deposited and alloyed on the molybdenum surface to produce the Ni-Mo intermetallic
compound and solid solution. Another example of
t h i s phenomenon resulted i n the course of testing
s i l i c o n with lithium in an iron capsule. Figure 21
shows that a l l the s i l i c o n has gone into solution
i n the lithium and has been carried t o the w a l l of
the iron container where alloying of the iron with
the s i l i c o n occurred.
Dissimilar-metal transfer
can cause many misleading results as can be seen
i n Fig. 22, which illustrates the difference in the

depth of corrosion when type 304 stainless steel


was tested w i t h lithium i n two types of containers.
In one case the type 304 stainless steel specimen
was placed i n a type 304 stainless steel container,
In the second
and the attack was rather shallow.
case the type 304 stainless steel was tested i n an
iron container, and the attack was much more severe, which was caused by the selective removal
of n i c k e l that was transported t o the iron w a l l
where it alloyed.
A schematic v i e w of the manner i n which dissimilar-metal mass transfer takes place is seen
i n Fig. 23. In t h i s case atoms of metal A go into
solution and move t o the surface of metal B either

by diffusion and/or by movement of the liquid.


When they reach the surface of metal 8, they come
out of solution, a l l o y w i t h metal 8, and diffuse
inward.
The driving force for dissimilar-metal
transfer i s the decrease i n the free energy that
i s achieved through the alloying of the two metals.

5J. V. Cathcart

and W.

D.

Manly, The Mass-Transfer

Various Metals and A l l o y s in Liquid


Lead, ORNL-2008 (Jan. 10, 1956).

P ~ o p e r t i e s of

The greater the difference in the chemical pot e n t i a l s of A and of 6 i n the two solid phases,
the greater w i l l be the driving force for the occurrence of mass transfer. The rate of mass transfer i s dependent on the temperature since an i n crease i n temperature will, of course, increase
the diffusion rate i n both the l i q u i d and the solid
phases. A high s o l u b i l i t y of metal A in 6, or 6
i n A, or both, w i l l also f a c i l i t a t e mass transfer.

Fig. 20. lnteralloying Between Molybdenum and


N i c k e l by Dissimilar Mass Transfer After 100 hr
a t lDOOC Vickers hardness impressions (SO-g
load) indicate relative hardness of the Ni-Mo alloy.
S O X . Reduced 325%.
CLASSIFIED

Y-1216

ONE

IRON C A P S U L E WITH L U M P S i IN L i .
T R I P L I C A T E TEST AT 4 0 0 0 C FOR 4 hr.
APPROX. 2 5 % Si, 75% L i ; 1 4 - I L

OF

Fig. 21. Dissimilar Mass Transfer That Occurred


Between Silicon and Iron i n an Iron Capsule Containing Lump Si i n L i Tested a t 1000C for 4 hr.

UNCLASSI
.. - . .- Fl
. ED

7
.c :,

..

. .

-...

e
0

.
Fig.

j
010

22. The Effect o f Type of Container on the Corrosion of Type 304 Stainless Steel i n L i t h i u m at

1DOOOC for 400 hr.

UNCLASSIFIED
OWL-LR-DWG
5642

METAL
A

ION

METAL

MOVE

OR

DIFFUSION INTO
METAL B AND
ALLOY FORMATION

Fig. 23. Schematic View of the Manner i n Which


Dissimilar-Metal or Concentration-Gradient Mass
Transfer Occurs.
CONCLUSIONS
The various types of liquid-metal corrosion have
been explained, and examples have been given;
the driving forces for the various types of liquidmetal corrosion have been discussed. Liquid-metal
corrosion has been described as mainly the solut i o n of the s o l i d metal in the l i q u i d metal. The

10

rate-controlling step for the various types of liquidmetal corrosion has not been determined. For the
successful containment of the l i q u i d metals, the
amounts of impurities, such as oxygen, nitrogen,
and carbon, should be carefully controlled. Future
work in the f i e l d of liquid-metal corrosion must
be directed toward increasing the understanding
of the rate-controlling steps of dissimilar-metal
transfer and temperature-gradient mass transfer.
When this has been accomplished, corrective measures can then be taken t o keep these types of
corrosion to a minimum.

ACKNOWLEDGMENTS
The author would l i k e t o acknowledge the assistance of J. V. Cathcart, E. E. Hoffman, and other
members of the Oak Ridge National Laboratory
working in the f i e l d of liquid-metal corrosion. The
metallographical work used i n t h i s paper was performed by a group under the direction of R. J. Gray.

You might also like