ORNL-TM-0202.txt

WMASTER

OAK RIDGE NATIONAL LABORATORY

7y

   

o
operated by
UNION CARBIDE CORPORATION
/™ NUCLEAR DIVISION
- for the

™ U.S. ATOMIC ENERGY COMMISSION
« ORNL- TM- 2021, Vol. 1
o0
o
>
T
-
i
=
O
>
a
Q
w
>
il
8

a‘* '

: EFFECT OF ALLOYING ADDITIONS ON CORROSION BEHAVIOR OF NICKEL -
MOLYBDENUM ALLOYS |IN FUSED FLUORIDE MIXTURES
Thesis)
Jackson Harvey DeVan
’
A This report is a portion of a thesis submitted to the Graduate Council of the University of Tennessee

in partial fulfillment of the requirements for the degree of Master of Science.

ERTRRUTION OF THIS DOCUMEN 15 UNLIMITED
 

LEGAL NOTICE

 

 

This report was prepored as an account of Government sponsorad work. Neither the United States,

nor the Commission, nor any person acting on behalf of the Commission:

A. Mokes any warranty or representotion, expressed or implied, with respect to the accuracy,
completeness, or usefulness of the information contained in this report, or that the use of
any information, apparatus, method, or process disclosed in this report may not infringe
privatsly owned rights; or

8. Assumes any liabilities with respect to the use of, or for damages resulting from the use of
any information, apparatus, method, or process diiclesed in this report.

As used in the above, '‘person acting on behalf of the Commission’’ includes any employse or

contractor of the Commission, or employse of such contractor, to the extent thot such employee

or contractor of the Commission, or employee of such contractor prepares, disseminates, or
provides access to, any information pursuant to his employment or contract with the Commission,

or his employment with such contractor.

 
LEGAL NOTICE

This report was prepared as an account of Government aponsored work. Neither the ¥United
States, nor the Commission, nor any person acting on behalf of the Commission:

A. Makes any warranty or rejresentation, expressed or implied, with respect to the accu-
racy, completeness, or usefulness of the information contained in this report, or that the use
of any information, apparatus, method, or process disclosed In this report may not infringe

privately owned rights; or ORNL - TM.._ 2 02 :|_
B. Assumes any liabilities with respect to the use of, or for damages resulting from the
use of any information, apparatus, method, or process disclosed in this report. VOl . I

Ag used in the above, “‘person acting on behalf of the Commission”’ includes any em-
ployee or contractor of the Comrnission, or employee of such contractor, to the extent that
such employee or contractor of the Commission, or employee of such contractor prepares,
disseminates, or provides access to, any information pursuant to his employment or contract
with the Commisgion, or his employment with such contractor.

Contract No. W-7405-eng-26

METALS AND CERAMICS DIVISION

EFFECT OF ALIOYING ADDITIONS ON CORROSION BEHAVIOR OF NICKEL-
MOLYBDENUM ALLOYS IN FUSED FLUORIDE MIXTURES

Jackson Harvey DeVan

MAY 1969

This report is a portion of a thesis submitted to the Graduate Cogncil
of the University of Tennessee in partial fulfillment of the requirements
for the degree of Master of Scilence.

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

WEIRIFULUON OF THIS DOCUMENE 15 UNHWM
iii

CONTENTS
Abstract . . . v v v v i e e e e e e e
Introduction . « .+« v ¢ ¢« ¢« ¢« + ¢ ¢ o o

Review of Related Work . « o« + o« o s o o
Corrosion by Fluoride Mixtures . . .
Corrosion Reactions . . « « . .

Reduction of UF, by Chromium . .

Corrosion of Nickel-Molybdenum Alloys.

Materials and Procedures . . « o o o o @

Tegt Materials . . . « . « « « . . .
Test Equipment . . . . . . . . . .
Salt Preparation . . . . . . « . . .

Operating Procedures « « « « ¢ « « o
Test Examination . .« ¢« ¢ « o o & &
Results and DisCussion . « v o « s o o &

Chromium « ¢« « &« &+ ¢ o o o o o o o o

Corrosion~Product Concentrations .

Metallographic Results . . . . .

Aluminum . v v ¢ 4 o« ¢ o o o« o o o

Corrosgsion-Product Concentrations .

Metallographic Results . « . . .
Titanium . o + & ¢ ¢ ¢ o o o + o o .
Corrosion-Product Concentrations
Metallographic Results . . . . .
Vanadium . . « ¢ ¢ ¢ v ¢ ¢ v 0 o 4
Corrosion-Product Concentrations
Metallographic Results . . . . .
Iron . o v ¢ ¢ 0 i e v e e e e e e
Corrosion-Product Concentrations

-Metallographic Results . . . . .

J
5

00 e oo DWW W

W W W wWww NN DN NN H P e
DO O OO0 0 W ®eEeuw KX P ;e oW
iv

Page
Nio-biu-zn . . . . . * ® - & . . » e . ” - * - - - L] - . . . » . - 32

Corrosion-Product Concentrations . + v o « o v o o o + o« + o 32

Metallographic Results . v v ¢ o o & o « o o o« o o+ o o« o o o 33
TUNESEEIL o v v v v o o v o o o o o o 4 s o e e e e e e e e e . 34
Corrosion-Product Concentrations . . + « o « o o o o o « o« o« 34

Metallographic Results . . « « « « ¢« v o o« o « o . . . ... 36
Relative Thermodynamic Stabilities of Alloying Constituents . . 36
General Discussion of Alloying Effects . « v v v v ¢ « v & » o . 38

Summary and ConcluSions .+ + « « « o o « & « o o o o o o o 0w oo o 40
ACKNOwledgZmENtS « v 4 s v 0 e e e e e e e e e e e e e e e e e e e . 4R
EFFECT OF ALLOYING ADDITIONS ON CORROSION BEHAVIOR OF NICKEL-
MOLYBDENUM ALLOYS IN FUSED FLUORIDE MIXTURES

ABSTRACT

Fused fluoride mixtures containing UF, have been developed
as fuel solutions for high-temperature nuclear reactors. To
develop container materials for such mixtures, we investigated
the corrosion properties of nickel-molybdenum alloys with vari-
ous solid-solution strengthening additions. These evaluations
utilized thermal convection loops which circulated salt mix-
tures between a hot-zone temperature of 815°C and a cold-zone
temperature of 650°C.

The alloys selected for study contained 17 to 20% Mo and
various percentages of Cr, Al, Ti, V, Fe, Nb, and W. Loops of
individual alloys were exposed to the salt mixture NaF-LiF-KF-UF,
(11.2-45,3-41.0-2.5 mole %) for periods of 500 and 1000 hr.
Measurements of the concentrations of corrosion products in
after-test salt samples indicated the corrosion susceptibility
of alloying additions to increase in this order: Fe, Nb, V,
Cr, W, Ti, and Al. However, metallographic examinations of
loop surfaces showed relatively light attack for all alloys
except those containing combined additions of aluminum and
titanium or aluminum and chromium.

A nickel-base alloy containing 17% Mo, 7% Cr, and 5% Fe,
designated Hastelloy N, was found to afford the best combina-
tion of strength and corrosion resistance among the alloy
compositions tested.

 

INTRODUCTION

Molten fluorides of uranium, thorium, or plutonium, in combination
with other fluoride compounds, have wide applicability as fuels for the
production of nuclear power.! Because of their high boiling points, these
mixtures can be contained at low pressures even at extremely high
operating temperatures. Their chemical and physical properties impart addi-

tional advantages such as excellent stability under irradiation and large

 

IR. C. Briant and A. M. Weinberg, "Molten Fluorides as Power Reactor
Fuels," Nucl. Sci. Eng. 2, 797-803 (1957).
solubility ranges for both uranium and thorium. These factors have
prompted design studies of molten fluoride fuel systems in conjunction
with thorium-uranium thermal breeders, uranium-plutonium converters, and
uranium burners.

The development of reactors which incorporate a circulating fluoride
salt is predicated on the availability of a construction material which
will contain the salt over long time periods and also afford useful
structural properties. The container material must also be resistant to
air oxidation, be easily formed and welded into relatively complicated
shapes, and be metallurgically stable over a wide temperature range.

In order to provide a material for initial reactor studies, several
commercially available high-temperature alloy systems were evaluated with
respect to the above requirements. As a result of these studies, Inconel,
a nickel-base alloy containing 15 wt % Cr and 7 wt % Fe, was found to
afford the best combination of desired properties and was utilized for
the construction of the Aircraft Reactor Experiment.2 Extensive corro-
sion tests,3’4 as well as posttest examinations of the ARE,” confirmed
the suitability of Inconel as a container material for relatively short-
term fluoride salt exposures. Corrosion rates encountered with this
alloy at temperatures above 700°C, however, were excessive for long-term
use with most fluoride fuel systems.

Utilizing experience gained in corrosion testing of commercial
alloys, we initiated an alloy development program to provide an advanced
container material for fluoride fuel reactor systems. The reference
alloy system was composed of nickel with a primary strengthening addition
of 15-20% Mo. This composition afforded exceptional resistance to fliuo-

ride attack but lacked sufficient mechanical strength and oxidation

 

W. D. Manly et al., Aircraft Reactor Experiment — Metallurgical

Aspects, ORNL-2349“(1§575.

°G. M. Adamson, R. S. Crouse, and W. D. Manly, Interim Report on
Corrosion by Alkali-Metal Fluorides: Work to May 1, 1953, ORNL-2337
(1959). .

“G. M. Adamson, R, S. Crouse, and W. D. Manly, Interim Report on -
Corrosion by Zirconium-Base Fluorides, ORNL-2338 (1961).

 

W. B. Cottrell et al., Disassembly and Postoperative Examination
of the Aircraft Reactor Experiment, ORNL-1868 (1958).
resistance at desired operating temperatures. To augment these latter
properties, additions of various solid-solution alloying agents were
evaluated, among them Cr, Al, Ti, Nb, Fe, V, and W. An optimum alloy
composition was selected on the basis of parallel investigations of the
mechanical and corrosion properties which were imparted by each of these
additions. The composition best suited to reactor use was determined to
be within the range 15-17 wt % Mo, 6-8 wt % Cr, 4—6 wt % Fe,
0.04-0.08 wt % C, balance Ni. Subsequent studies of the alloy, desig-
nated Hastelloy N, have shown it to be extremely well suited for appli-
cations demanding long-term compatibility with fluoride salts in the
temperature range 650-800°C.

The present research was concerned with the corrosion effects
resulting from additions of alloying elements to the nickel-molybdenum
system and an analysis of the thermodynamics of the corrosion process as

indicated by these alloying effects.
REVIEW OF RELATED WORK

Corrosion by Fluoride Mixtures

Corrosion Reactions

 

The corrosion of nickel-base alloys, containing iron and chromium,

by fluoride fuel mixtures has been found to result from a combination

of the following types of oxidation reactionss ©

7

1. Reactions’ involving impurities in the salt

2HF+QI_=CI‘F2 + Hp , (l)
NiF, + Cr = CrFp + Ni , (2)
FeF, + Cr = CrFp + Fe . (3)

 

*W. D. Manly et al., "Metallurgical Problems in Molten Fluoride
Systems," Progr. Nucl. Energy, Ser. IV 2, 164 (1960).

 

7"Solid-solution alloying elements are underlined.
2. Reactions involving impurities in or on the metal, for example
2Ni0 + ZrF, = ZrO, + 2NiFs (4)
followed by reaction (2).
3. Reactions involving components in the salt

2UF, + Cr = CrF, + 2UF; , (5)

1l

i

3UF, + Cr = CrF3 + 3UF; . (6)
The extent of the first four of these reactions, which proceed strongly
to the right and to completion rapidly, can be reduced by maintaining low
impurity concentrations iIn the salt and on the surface of the metal.
Reactions (5) and (6), on the other hand, are indigenous to fluoride
systems which derive their usefulness through the containment of UF,.
While the role of chromium has been investigated extensively in connec-

8 considerably less information is available

tion with these reactions,
regarding the thermodynamics of these reactions for the other alloying

elements which were of interest in the present study.

Reduction of UF, by Chromium

The reaction of UF,; with chromium is found to be strongly influenced
by the reaction medium employed.8 In melts composed principally of

NaF-Zr¥F, or NaF-BeFp, the reaction produces only divalent chromium, that is,
2UF, + Cr = 2UF3 + CrTy . (7)

However, in the case of NaF-KF-LiF-UF, mixtures used for this investiga-

tion, the reaction between chromium and UF, produces both divalent and

 

87. D. Redman, ANP Quart. Progr. Rept. Dec. 31, 1957, ORKL~2440,
pp. 78-82.

 
trivalent chromium. At equilibrium, approximately 80% of the total
chromium ions in the mixtures are observed to be trivalent, giving the

net reaction
2.8UF, + Cr = 0.2CrFp + 0.8CrF3 + 2.8UF3 . (8)
The equilibrium constant for this reaction is given by

(a )O.Z(Q )O.B(G )2.8
K = CrFs CrF, UF4 . (9)

CHPLICN

where & represents thermodynamic activity. Because of the relatively
small concentrations of CrF, and UFs which are attained in the salt solu-
tions at equilibrium, the activities of each of these components can be
closely approximated by their mole fractions, in accordance with Henry's
law. Thus, for a salt system of fixed UF, concentration, assuming the

reference states for salt components to be the infinitely dilute solution,

(NCng)O.Z(NCrF3)O.8(NUF3)2.8
K = . (10)
(g, )25 (o)

 

For a system initially containing no UFi, CrF,, or CrF;, it follows that
NCrF2 = 1/4NCrF3 = 1/14NUF3. In such systems where the change in UF,
concentration is small, Eq. (9) reduces to

_ 5/19
Ny, * Norp,) % (11)

where

4/19

o 14/19
iy - 5@;{) (TU% /
O~

The constant KN has been experimentally determined for the mixture
LiF-KF-NaF-UF; (11.2—41.0-45.3-2.5 mole %) by equilibrating it with pure
chromium (agr = 1) at 600 and 800°C (ref. 8). Under these conditions,
the constant is equivalent in value to the mole fraction of chromium ions
in the salt at equilibrium. The measured values of KN are listed in

Table 1.

Table 1. Eguilibrium Concentrationsa of Chromium Fluorides
for the Reaction of Pure Chromium with UF,

 

Concentration of Chromium

Equilibration Tons in NaF-LiF-KF-UF,
Temperature (11.2-41.0-45.3-2.5 mole %)
(°C)

 

ppm  mole fraction (KN)

 

600 1100 1.0 x 10~°
800 2600 2.4 % 1072

 

8J. D. Redman, ANP Quart. Progr. Rept. Dec. 31, 1957,
ORNL-244C, pp. 7882,

 

Because the chromium-UF, reaction is temperature dependent, chemi-
cal equilibrium beiween these two components is precluded in systems of
uniform alloy composition where the circulating salt continually experi-
ences a temperature change. In such systems chromium tends to be con-
tinually removed from hotter portions and deposited in cooler portions.
A theory relating the rate of this movement to the operating parameters

of the container system has been formulated by Evans.’

Corrosion of Nickel-Molybdenum Alloys

Because of the oxidation of chromium by fuel-bearing fluoride salts,

alloys employing large percentages of this element were not satisfactory

 

’R. B. Evans, ANP Quart. Progr. Rept. Dec. 31, 1957, ORNL-2440,
pp. 104—113.

 
as container materials except at temperatures where diffusion rates in
the alloys were relatively low. Accordingly, evaluations were made of
several commercial alloys in which chromium was not employed as a major
alloying addition. Based on these tests, alloys of nickel and molybdenum
appeared to offer the most promising container system for achieving rela-
tively high reactor operating temperatures. Unfortunately, commercially
available nickel-molybdenum alloys which exhibited excellent corrosion
properties were not well suited to contemplated reactor systems because
of three adverse characteristics: (l) poor fabricability; (2) a tendency
to age-harden and embrittle at service temperatures between 650 and
815°C (ref. 10); and (3) poor resistance to oxidation by air at elevated
temperatures. The scale formed on exposure of these alloys to high-
temperature air was of the type NiMoO,, which upon thermal cycling
between 760 and 350°C underwent a phase transformation and spalled as a
consequence of a resultant volume change.ll
By means of an alloy development program, however, it was considered
plausible to eliminate the undesirable features of the commercial materials
while retaining their inherent corrosion resistance. The initial objec-
tive of this program was to provide an alloy which did not embrittle under
the thermal treatments imposed by reactor operation. By experimenting
with various compositions of binary nickel-molybdenum alloys, 1t was
determined that lowering the molybdenum concentration to a level of 15-17%
served to avoid detrimental age-hardening effects in the alloy system.12
Although such an alloy system was satisfactory from the standpoint of
corrosion resistance, it was necessary to augment the oxidation and
strength characteristics of the system through additional solid-solution
alloying agents. The corrosion effects which resulted from these addi-

tions were the subjects of the present study.

 

1R, E. Clausing, P. Patriarca, and W. D. Manly, Aging Characteristics

of Hastelloy B, ORNL-2314 (1957).
iy, Inouye, private communication.

127, W. Stoffel and E. E. Stansbury, "A Metallographic and X-ray
Study of Ni Alloys of 20-30 Per Cent Mo," Report No. 1 under Subcontract
No. 582 under Contract No. W-7405-eng-26, Knoxville, Tenn., Dept. of
Chem., Eng. of the Univ. of Tenn. (1955).
MATERIALS AND PROCEDURES
Test Materials

The nickel-molybdenum alloy compositions selected for study were
supplied by the Metals and Ceramics Division at ORNL and, under subcon-
tract agreements, by Battelle Memorial Institute and Superior Tube
Company. Alloys furnished by ORNL were induction-melted under vacuum,
while those furnished by subcontractors were induction-melted in air
using a protective slag. Each alloy heat, which ranged in weight from
12 to 100 1b, was either forged or extruded into a 3-in.-diam tube blank
and was subsequently drawn into 1/2-in.-0D seamless tubing by the
Superior Tube Company. The cold-drawn tubing was annealed at 1120°C.
Table 2 lists the experimental alloy compositions used for this corrosion

study.
Test Eguipment

The method selected to evaluate the corrosion properties of these
alloys was based on the following considerations:

1. The method necessarily had to be tailored to the use of rela-
tively small quantities of material, since it was practical to produce
only small heats of the many alloys desired for study.

2. Tubing was considered to be a highly desirable form in which to
test the material, since production of the material in this form was
carried out as an adjunct to evaluating the fabricability of each alloy.

3. Previous demonstrations of the effects of temperature gradient
in the salt and the salt flow rate on the corrosion behavior of container
materials in fluorides made it mandatory that corrosion tests be con-
ducted under dynamic conditions, that is, conditions which provided for
the continuous flow of salt through a temperature gradient.

The thermal convection loop, which had been used extensively for
Inconel corrosion studies and had been developed into an extremely
straightforward and reliable test device, was judged to be the best form
of experimental device for meeting these requirements. This device con-

sists of a closed loop of tubing, bent to resemble the outline of a harp,
Table 2.

Compositions of Experimental Alloys Used for Corrcsion Studies

 

Composition, wt %

Composition, at. %

 

 

Heat
i
Number Wi Mo Cr Fe Ti Al Wb W v Ni Mo Cr Fe Ti Al Nb W v
Series T
OR 30-1 80.12 16.93 2.83 85.50 11.10 3.41
-2 78.55 16.65 4.62 83.60 10.80 5.55
-4 73.65 16.37 9.21 78.30 10.60 11.04
-6  78.50 15.11 .40 g82.70  9.72  7.60
37A-1 77.0  20.39 2.62 83.30 13.50 3.20
43A-3 73.30 20.34 6.34 78.90 13.40 7.71
Series IT
OR 30-7 82.10 15.93 1.88 85.60 10.15 4.26
-8 80.30 17.80 1.89 85.90 11.60 2.47 _
-9 81.10 16.8 .09 88.10 11.20 0.72
-10 81.10 16.60 2.23 86.40 10.80 2.73
-11 79.80 16.53 3.68 85.10 10.80 4,12
-12 80.00 16.80 3.22 86.70 11.10 2.20
-19 79.00 16.90 4 .10 g87.10 11.40 1.44
-20 79.20 16.60 4,18 84.10 10.80 5.11
=21 78.90 16.40 471 85.50 10.90 3.62
S T23012 82.00 17.42 0.53 87.40 11.40 1.22
OR 1491 86.58 11.23 2.19 88.16 6.99 4. 85
Series ITL
OR 30-13 79.93 17.56 1.56 0.95 84.48 11.35 2.02 2.18
=14 79.53 16.50 1.52 2.45 82.11 10.42 1.92 5.51
=16 77.74 16.00 3.65 1.49 1,12 81.01 10.20 4.30 1.90 2.54
-22 77.65 15,90 5.69 1.16 0.60 80.27 10,05 6.64 2.6l 0.39
-33 74.07 15.15 5.C1L 5.07 0,70 77.26  9.66 5,90 5.56 1.59
B2897 76.13 20,50 1.25 1.32 83.61 13.77 1.68 0.92
B2898 76.30 20.50 2. 44 1.31 82.60 13.20 3.24 0.90
B3276 69,12 21,10 7.58 2.16 75,17 14,00 9,32 1.48
B3277 66.95 21.60 7.82 1.31 2.32 7L.72  14.20 9.4%7 3.06 1.57
S T23011 71.50 15.06 3.84 0,83 4.17 4.90 79.04 10.20  4.79 1.27 2.91 1.72
S T23013 74.42 15.20 0.58 4.57 5.23 g83.14 10.36 1.40 3.23 1.85
S T23014 80.86 16.70 2,19 0.57 85.22 10.76 2.71 1.30

 

®0R denotes heats furnished by the Metals and Ceramics Division; 8 T by Superior Tube Company; and B by Battelle

Memorial Institute.
10

two legs of which are heated and two of which are exposed to the cooling
effects of ambient air. Flow results from the difference in density of
the salt in the hot and cold portions of the loop.

The configuration and dimensions of the loop design are presented
in Fig. 1. All loops were fabricated of seamless tubing having an out-
side diameter of 0.500 in, and a wall thickness of 0.035 in. The tubing
was assembled by the Heliarc welding technique using an inert gas
backing.

During operation the loops were heated by a series of clamshell
resistance heating elements located as shown in Fig. 1. To fill the
loop required that we apply heat to both the cold- and hot-leg sections.
Auxiliary heating for this purpose was provided by passage of electric
current directly through the tube wall. When the loop was filled, as
determined by an electrical shorting probe near the top of the loop, the
auxiliary heat source was turned off and the heating elements were turned
on. Insulation was then removed from the cold leg to whatever amount
was required to establish a predetermined temperature gradient.

Loop temperatures were measured with Chromel-P—Alumel thermocouples
located as shown in Fig. 1. The thermocouple junctions, in the form of
small beads, were welded to the outside tube wall with a condenser dis-
charge welder and covered by a layer of quartz tape, which in turn was
covered with stainless steel shim stock. All tests were operated so as
to achieve a maximum mixed-mean salt temperature of 215°C and a minimum
salt temperature of 65C°C. The maximum salt-metal interface temperature,
which was attained near the top of the vertical heated section, exceeded
the maximum bulk salt temperature by approximately 95°C (ref. 13). The
salt-flow rate under these temperature conditions was established from

heat balance calculations to be in the range of 5 to 7 ft/min,

 

13Measurements of the maximum inside wall temperature could not
practically be made in each loop test; however, values of this tempera-
ture were recorded by means of heat balance calculations and imbedded
thermocouples using a specially instrumented test loop which exactly simu-
lated the geometry and temperature profile used in the corrosion

experiment.
11

T ORNL-LR~DWG 27228 R
I
!

| 4 T.C. NO. 4

 

 

T.C.NO.1-3

3-in.R

 

—twi

6-in. CLAM SHELL HEATERS —

 

CONTROL & NO. 2

 

 

2 -__"fij’ “

 

 

 

 

 

 

 

 

 

 

 

 

¢
(o
(]
|
| i
P | ;
IR c |
s
|1 |
Pl
! | } METALLOGRAPHIC SAMPLES |
|l : + THERMOCOUPLE LOCATIONS
||
i ! |
17 .E!
P 7 N
| | !
I 4 ‘
- |
I v i
| t |
] N
L '
L] !
‘. 7 ,
3-in. R\ 750 T.C. NO. 6 —
6-in. CLAM SHELL 7
e~ 1T HEATERS 37
/ e~ - .
/ .—"x__ w:"“-___-
/ T~ T B\

 

 

Fig. 1. Schematic Diagram of Thermal Convection Loop Used for
Evaluations of Experimental Nickel-Molybdenum Alloys. The locations of
thermccouples and test samples are shown.
12

Salt Preparation

The fuel mixture used in these studies was of the composition shown
in Table 3. We selected the LiF-KF-NaF-UF, composition (Salt 107) on
the basis that the oxidation of container constituents by a given con-
centration of UF; tended to be greater for this mixture than for other
mixtures of practical importance. Thus, achievement of satisfactory
compatibility with this mixture in effect provided a container material

of ultimate versatility with respect to all fuel mixtures.

Table 3. Composition of Fluoride Mixture
Used to Evaluate Experimental
Nickel-Molybdenum Alloys

Salt Number: 107
Liquidus Temperature: 490°C

 

 

Component Mole % Weight %
NaF 11.2 9.79
LiF 45,3 24 .4
KF 41.C 49, 4
Uy, 2.5 16.3

 

The fluoride mixtures were prepared from reagent-grade materials
and were purified by the Fluoride Processing Group of the Reactor Chemistry
Division. In general, the procedure for purification was as follows:
(1) the dry ingredients, except for UF,, were mixed, evacuated several
times for moisture removal, and then melted under an atmosphere of helium;
(2) the molten mixture was held at 815°C and treated with hydrogen for 4 hr
to purge hydrofluoric acid from the mixture; (3) the mixture was cooled to
205°C under a helium atmosphere and UF, was admitted. Upon the addition
of the UF,, the mixture was remelted, heated to 815°C, and then treated
again with hydrogen to purge the excess hydrofluoric acid.

All mixtures were prepared in 300-1b gquantities and apportioned into
50-1b containers, after which samples were submitted for analysis of

Ni, Fe, and Cr. It was required that each of these elements be present
13

in amounts less than 500 ppm as determined from individual batch analyses,
A second before-test analysis of each salt mixture was obtained from a

sample of the salt taken as it was being admitted to the test loop.
Operating Procedures

Fach loop was thoroughly degreased with acetone and checked for
leaks using a helium mass spectrometer. After thermocouples and heaters
were assembled and insulation was applied, the loop was placed in a test
stand, as shown in Fig. 2.

The salt charging pot was connected to the loop with nickel or
Inconel tubing, and both the loop and the charging pot were heated to
€50°C under a'dynamic vacuum of less than 50 p Hg. Helium pressure was
then applied to the charging pot in order to force the salt mixture from
the pot to the loop. After filling, scalt was allowed to stand in the
loop at 650°C for approximately 2 hr, so that oxides and other impuri-
ties would be dissolved from the container surface into the salt mixture.
This mixture was then removed, and a fresh salt mixture was admitted
from the fluoride charging pot. A helium cover gas under slight positive
pressure (approx 5 psig) was maintained over the salt mixture during all
periods of testing.

At the end of test, power to the loop was cut off and insulation
was stripped from the loop so as to freeze the salt mixture as rapidly

as possible.
Test Examination

Each loop was sectioned with a tubing cutter into approximately
6-in. lengths. Five Z2-in. sections were then removed from the loop posi-
tions indicated in Fig. 1 for metallographic examination. Two of the
remaining 6-in. sections, one from the hottest section of the loop
(specimen H) and one from the coldest section (specimen C), were obtained
for salt chemistry studies, and the remaining loop segments were held in
storage until all examinations of the loop were completed.

Salt removal was accomplished by heating each section in a small

tube furnace at 600°C in helium., The salt was collected in a graphite
 

   

 

 

 

-
»

 

 

 

 

" Fig. 2.
Stands.

(

 

  
15

crucible located below the furnace windings. The five 2-in, sections of
tubing were examined metallographically, and the salt samples were sub-
mitted individually for petrographic and chemical analyses. .If layers
of corrosion products were discovered on the tube wall, samples of the
tubing and salt contained in that section were also submitted for x-ray

diffraction examination.

RESULTS AND DISCUSSION

Alloying effects were evaluated in terms of both the corrosion
products entering the salt mixtures and the metallographic appearance of
the alloy after test. Results have been grouped in this section

according to the alloying element studied.
Chromium

Corrosion-Product Concentrations

 

Effects of chromium additions were examined in six ternary alloys
with chromium contents of 3.2 to 11.0 at. %. One loop of each of these
compositions was operated with Salt 107 for 500 hr under the test condi-
tions described above, The compositicns investigated and the attendant
concentrations of chromium ions in salt samples taken at the conclusion
of these tests are shown in Table 4. The extent of reaction between
chromium and fluoride constituents, as indicaked by the chromium ion con-
centrations, increased with the amount of chromium in the alloy. This
increase is illustrated graphically in Fig. 3, where the data are com-

114 15 It may be noted

pared with data for Incone and for pure chromium.
that the chromium concentrations of the salts were less than those for
Inconel loops operated under identical temperature conditions. A hori-

zontal line, which represents the chromium ion concentration at

 

4G, M. Adamson, R. S. Crouse, and W. D. Manly, Interim Report on
Corrosion by Alkali-Metal Fluorides: Work to May 1, 1953, ORNL-2337
(1959),

17, D. Redman, ANP Quart, Progr. Rept. Dec. 31, 1957, ORNL-2440,
pp. 78-L82.
16

Table 4. Corrosion-Product Concentrations of Salts Tested with
Nickel-Molybdenum-Chromium Alloys

 

 

 

 

Alloy Composition Tegt Chromium Concentration in
Heat (at. %) : Salt Samples® (mole %)
Number Duration
(hr)
Cr Other Components Sample H Sample C Other
OR 37A-1  3.20 13.5 Mo, bal Ni 5C0 0.01%4 0.0180  ©.0213
OR 30-1 3.41 11.1 Mo, bal Ni 500 0.0222 0.0365  0.0291
OR 30-2 5.55 10.8 Mo, bal Ni 5C0 0.0375 0.0352  0.0376
OR 30-2 5.55 10.8 Mo, bal Ni 1000 0.0509 0.0509  0.0543
OR 30-6 7.60 9,72 Mo, bal Ni 500 0. 0606 0.0606  0.0566
OR 43A-3 7.71 13.4 Mo, bal Ni 500 0.0453 0.0476  0.0425
OR 30-4 11.04 10.6 Mo, bal Ni 5C0 0.0819 0.0814  0.0699

 

®Sample designations "H" and "C" are discussed on page 13; "other"
designates salt samples obtained from metailographic specimens.

equilibrium with pure chromium at 600°C, is seen in Fig. 3 to be above
the measured chromium ion concentrations for all alloys tested. Thus,
the observed concentrations were less than those required for the forma-
tion of pure chromium crystals in the coldest portion of the loops
(approx 650°C).

As discussed previously, the concentration of chromium ions under

the conditions of these tests should be governed by the relation
[Eq. (11)]

N . = _ 5/19

Cr ions (NCI'F2 * NCI"F3) - K2CYC£ 3

where NCr sons is the mole fraction of chromium ions in Salt 107 at
equilibrium with an alloy of given chromium activity, o, . If we assume

Cr

op? the resultant chromium ion

concentrations for these alloys should lie within a region which is

that NCr giveg an approximate measure of &

bounded above by the function determined at a temperature equivalent to
the maximum loop temperature and below by the function determined at the

minimum loop temperature. Bounds using experimental values of K
17

ORNL—LR—DWG 46946

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

3000
_—
///
- " — 200
2000
ms:fififii—f”"'
rATY
1600 = CONCENE e —{ 150
= \CTED
E PRED |
a e |
— !
§ 1200 =r,-,——=""—— PURE Cr-800°C
e — = T ] —_—t 100
< L 0
x 1om3:f”" 0
o
= o
Y ol ¢
800 2
o — £
8
= 600 T3 3”
2 — T —
o
z 3
—{ 40
400 2 i RANGE OF CHEMISTRY SAMPLES
H AND C - 500 hr TESTS
— , e AVERAGE CONCENTRATIONS iN ]
e - METALLOGRAPHIC SAMPLES —500 hr TESTS -
| ' E RANGE OF CHEMISTRY SAMPLES
H AND C —1000 hr TESTS
¢ AVERAGE CONCENTRATIONS IN —{ 20
200 |- METALLOGRAPHIC SAMPLES — 1000 hr TESTS _|
| L
003  0.04 0.06 0.08 0.0 0.2 0.4 0.6 0.8 10

ATOM FRACTION OF CHROMIUM IN ALLOY

Fig. 3., Concentration of Chromium Ions in Salt 107 as a Function
of Chromium Content in Nickel-Molybdenum-Chromium Alloys.

measured at 800 and 600°C (Table 1), which reasonably approximate the
maximum and minimum loop temperatures, are plotted in Fig. 3 and are
seen to include only two of the observed loop concentrations, both corre-~
sponding to chromium compositions greater than 7 at. %. However, except
for the alloy with least chromium content, all concentrations lie rel-
atively close to the lower bound and roughly approximate the slope of
this bound.

The deviation of measured corrosion-product concentrations in these
tests from the predicted concentration ranges may have resulted primarily
from inaccuracy in the assumption that N . equals & o However, assuming

C C

aCr to be accurately known, the corrosidE:product concentrations in these

tests would nevertheless have been lower than those predicted, since cor-

rosive attack would necessarily reduce the chromium content and hence the
13

chromium activity at the surface of the alloys relative to the original
chromium activity. We also examined the possibility that the test times
were not sufficiently long for the chromium concentration of the salt to
have completely attained the maximum or steady-state level. To evaluate
this latter point, sufficient material of one of the alloy compositions,
heat OR 30-2, was obtained to permit the operation of a 1000-hr test.
Results of this test, which are shown in Table 4 and Fig. 3, indicated
that only a small increase in chromium concentration occurred between the
500- and 1000-hr intervals. It was concluded, therefore, that the 500-hr
test provided a reasonably close estimate of the limiting corrosion-
product concentrations associated with the temperature conditions of these
experiments.

Additional data on the effects of chromium in nickel-molybdenum alloys
were afforded by tests of five alloys which contained chromium in combina-
tion with one or more other alloying elements, as shown in Table 5. Not-
withstanding the additional alloying agents, the concentrations of
chromium-containing corrosion products following 500-hr tests of these
alloys in Salt 107, as shown in Table 5, were comparable to the concen-
trations associated with the simple ternary chromium-containing alloys.
The corrosion-product concentrations for the alloys with multiple additions
are plotted in Fig. 4, and are seen to show the same general variation
with chromium content as the ternary alloys (see Fig. 3).

Also shown in Table 5 and Fig. 4 are the results of 1000-hr tests of
four of the alloys containing multiple additions. After all but one of
these tests, the chromium ion concentrations in the salts were slightly
higher than for 500-hr tests of the same alloys; in one test, the chromium
ion concentration was unaccountably lower. The chromium ion concentra-
tions of these multicomponent alloys again fell near those predicted on
the basis of equilibrium data for pure chromium at 600°C and were con-
siderably less than the concentration needed to deposit pure chromium at
600°C.

The chromium activities in all of the alloys tested would appear on
the basis of corresponding corrosion-product concentrations to be lower
than the activity of chromium in Inconel. However, it is lmportant to

note in Figs. 3 or 4 that any alloy in which the chromium activity
Table 5.

Containing Chromium in Combination with Other Alloying Elements

Corrosion-Product Concentrations of Salts Tested with Nickel-Molybdenum Alloys

 

 

Chromium Concentration in

 

 

11 iti . ,
Heat Alloy Composition, at. % Test Salt Samples,® mole 9%
- Duration
Number (hr)
Chromium Other Components Sample H Sample C Other

OR 30-16 4.30 2.5 Al, 1.90 Ti, 10.2 Mo, 500 0.0227 0.0236 0. 004
bal Ni

S T23011 4."79 1.27 Al, 2.91 Nb, 1.72 W, 500 0. 0490 0.0490 0. 0426
10.2 Mo, bal Ni

OR 30-33 5.90 1.59 Al, 5.56 Fe, 9.66 Mo, 1000 0.0689 0.0731 0.0620
bal Ni

OR 30-22 6. 64 0.43 Fe, 0.39 Nb, 2.61 Al, 500 0.0583 0.0555 0. 0490
10.05 Mo, bal Ni

OR 30-22 6. 64 0.43 Pe, 0.39 Nb, 2.61 Al, 1000 0.0416 0. 0402 0.0370
10.05 Mo, bal Ni

B3276 9.32 1.48 Nb, 14.0 Mo, bal Ni 500 0.0615 0.06el15 0.0578

B3277 9.47 1.57 Nb, 3.06 Al, 14.2 Mo, 500 0.0698 G. 0700 0. 0624
bal Ni

B3277 9.47 1.57 Nb, 3.06 Al, 0.073L1 0. 0657 0.0740

bal Ni

14.2 Mo, 1000

 

aSam_ple designations "H" and "C" are
obtained from metallographic specimens.

discussed on page 13; "other" designates samples

6T
20

UNCLASSIFIED
ORNL—LR—DWG 46945

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

3000
} | } o 1
| : —
! | ; ///
f /
| | } . - — 200
2000 N — e T e
. OoC L
- 80
b o I Lo u..,!__“_.__.___— ONS S WSO SO
—_ 800/ — e — 4,,, L_ CoNCEN/ — 150
o : PRE
& 1200 —t PURE Cr — 600°C i
= r—
g — - . —_— PRt B e B e S e (010
5 4000 [= A e
S T 3 - T
— — 80 ™
= |
S 800 : E~‘ PR INCONEL _156655 | g
— . S | ]
g NTRM\ON ; 52
= — CONCE = 460 o
2 L 3 : CcTed ’ °
S 800 ; | pREOVV_Z. - — f £
= l ' \
o —' * . ~ | | —
o I e |
(:5 T i
E, I RANGE OF CHEMISTRY SAMPLES | a0
400 4 H AND C - 500 hr TESTS
; AVERAGE CONCENTRATIONS IN
- ) , | : ® METALLOGRAPHIC SAMPLES -500 hr TESTS
| -
- i lqvl RANGE OF CHEMISTRY SAMPLES
| : | H AND € —1000 hr TESTS
| r ‘ g AVERAGE CONCENTRATIONS IN
; : METALLOGRAPHIC SAMPLES —1000 hr TESTS
L | : ‘ — 20
200 + . :
i i !
[ |
| |
; ; i i |
0.03 0.4 0.06 008 040 0.2 0.4 0.6 0.8 1.0

ATOM FRACTION OF CHROMIUM IN ALLOY

Fig. 4. Concentration of Chromium Ions in Salt 107 as a Function
of Chromium Contents in Nickel-Molybdenum Alloys Containing Multiple
Alloy Additions.

exceeded a value of 0,035 would support the formation of pure chromium
at 600°C in any case where the concentration of Cr¥F,; and CrF:; approached
equilibrium with the alloy at temperatures of 800°C or above. Thus,
unless the activity coefficients for chromium in these alloys are much
smaller than unity, the possibility for the deposition of pure chromium
in cold-leg regions cannot be completely excluded for any of the alloys
tested,

In the case of ZrF,- or ReF,-base salt mixtures, the temperature
dependence of the Cr-UF, reaction is much less than for Salt 107. It

16

has been shown-" that in these mixtures, alloys having a chromium

 

16y, R. Grimes, ANP Quart. Progr. Rept. March 10, 1956, ORNL-2106,
pp. 9699,

 
21

activity equivalent to or less than that for Inconel cannot support
chromium ion concentrations at 800°C which are sufficiently high to main-
tain pure chromium at 600°C., Consequently, mass transfer of chromium

by deposition of pure chromium crystals in the latter salt mixtures

should not be possible for any of the alloys listed in Tables 4 and 5.

Metallographic Results

Metallographic examinations of the ternary alloys shown in Table 4
showed little evidence of corrosive attack other than shallow surface
roughening. Void formation, which characteristically had been found in

7 was not detected

nickel-~chromium alloys under similar test conditions,?
in any of these alloys. The relative depths of attack which occurred
for the alloy containing 5.55 at. % Cr (heat OR 30-2) after 500- and
1000-hr tests, respectively, are compared in Fig. 5. Although the depth
of attack was similar at both time intervals, the intensity of surface
pitting was somewhat greater after the 1000-hr exposure. The intensity
of surface roughening was also found to increase with increasing chro-
mium concentration. This is seen in Fig. & which compares photomicro-
graphs of two alloys containing different chromium contents. Cold-leg
sections of these loops showed no evidence of corrosion and contained

no visible deposits of corrosion products.

The depths of attack observed metallographically for alloys con-
taining chromium in combination with other additions are shown graphi-
cally in Fig., 7. Attack in these alloys was manifested as surface
roughening after 500 hr and as a combination of surface pitting and
shallow void formation after 1000 hr. Depths of corrosion were in gen-
eral higher than for the ternary chromium-containing alloys, particularly

in alloys which contained aluminum additions.

 

171, s. Richardson, D. C. Vreeland, and W. D. Manly, Corrosion by
Molten Fluorides, ORNL-1491 (Sept. 1952).

 
T-11327

 

" F—
T T

[

 

o

0.014 INCHES
@ 250x

HO

B

(a) ifi!% , s ‘ SE N e |

o IN‘_‘I" -

|

"

 

. 0.014 INCHES
o 2s0x 1@

T 28

 

 

"r*j;j(b_ , Jwfiffj'
Flg. 5. Appearance of Hot Leg Surface of Ternary Nickel—Molybdenum
_ _A110y‘Contain1n 5.55 at. % Cr Following Exposure to Salt 107

‘Heat OR 30-2. f;) After 500 hr -and (b) after 1000 hr. - -

 
 

 

 

 

 

 

 

i
i
]
]

 

 

 

 

 

 

 

 

23

 

o

J

[

 

 

oy

:0.014 INCHES
I 250%

[

 

 

ke N -

&

 

 

0.014 INCHES

 

 

 

 

Fig. 6. Hot-Leg Sections of ‘Nickel-Moiybdenum—-clhrdzi_i_iti_m Thermal
Convection Loops Following 500-hr Exposure to Salt 107.

(a) 3 2 Cr=13.5 Mo—bal Ni "

 

(at. %), (v) 11.0 cr-10.6 Mo-bal Ni (at. %)
24

ORNL- LR—DWG 46944

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

S 500 hr - SALT 107 §
kol
< 4 §§§“
> NN 1000 hr - SALT 107 %
Q 3 D\
@ N \\\
o
@ /
& \ %
= N
Q // § /
I 4 - || | - |
& ) AV U sss
° o Lzza 2 Sk\
Mo | 10.36|12.89 | 11.35] 10.76 [ 1400 | 10.2 | 10.2 |13.77 | 14.20 | 10.42 | 9.66 | 10.05
Cr 9.32 | 4.30 | 4.79 9.47 590 |6.64
®
5 Fe 5.56 | 0.43
=2
Q
S Al [ 140|561 |2.48 | 1.30 2.54 | 1.27 3.06 | 5.51 | 1.59 | 2.61
2
& i 202|271 1.90 1.68 1.92
-
<I
Nb | 3.23 | 087 1.48 291 | 092 [ 1.57 0.39
w | 1.85 1.72

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Fig. 7. Depths of Corrosion Observed for Nickel-Molybdenum Alloys
with Multiple Alloy Additions Following Exposure to Salt 107. BRars
designating corrosion depths appear directly above the alloy compositions
which they represent. (Where bars have both positively sloped and
negatively sloped cross-hatching, the height of positively sloped cross-
hatching indicates depth of corrosion after 500 hr and combined height of
both types indicates depth after 1002 hr. )

Aluminum

Corrosion-Product Concentrations

 

Table & lists the concentrations of aluminum which were analytically
determined in fluoride mixtures operated for 500~ and 1000~hr periods
with nickel-molybdenum alloys containing single additions of aluminum.

In the case of alloys containing greater than 2 at. % Al, the corre-
sponding aluminum concentrations in the salt mixture were relatively
high, that is, in the range of 0.15-0.37 mole %. Only one alloy composi-
tion was subjected to tests at both 500 and 1000 hr; however, results

for this alloy suggest that effectively similar salt concentrations were
Table 6.

Corrosion-Product Concentrations of Salts Tested with

25

Nickel-Molybdenum-Aluminum Alloys

 

Alloy Composition

Heat

(at. %)

 

Number

Test
Duration

(hr)

Aluminum Concentration in
Salt Samples® (mole %)

 

 

Al Other Components Sample H Sample C Other
S T23012 1.22 11.4 Mo, bal Ni 500 0.000872 0.000872 b
OR 30-7 4,26 10,15 Mo, bal Ni 500 0.220 0.070 0.137
OR 30-7 4,26  10.15 Mo, bal Ni 10C0 0.214 0.105 0.178
OR 1491 4. 85 6.99 Mo, bal Ni 1000 0.374 0,374 0.374

 

%Sample designations "H" and "C" are discussed on page 13; "other"
designates samples obtained from metallographic specimens.

bNot determined.

realized for both time intervals. Table 7 lists the concentrations of
corrosion products which formed in salt mixtures circulated in loops
containing aluminum in combination with other alloying elements. These
concentrations were of the same general magnitudes as those observed for
the ternary Ni-Mo-Al alloys.

The corrosion-product concentrations showed no definable correspon-
dence with the aluminum content of the alloys. However, the propensity
for aluminum to react with interstitial contaminants, such as nitrogen,
in these alloys, makes the activity of aluminum in the alloys very depen-
dent on composition and metallurgical history. For this reason poor
correspondence between the corrosion-product concentration and total
aluminum concentration in the alloy would not be unexpected.

It is apparent from tests of both the ternary and multicomponent
alloys that aluminum additions less than 2 at. % give rise to very much
lower aluminum ion concentrations than additions above 2 at. %. This
observation may suggest that below a concentration of 2 at. % the bulk
of the aluminum addition has formed highly stable compounds with

interstitial contaminants in the alloy.

Metallographic Results

 

Hot-leg surfaces of ternary alloys containing aluminum in amounts

greater than 2 at. % underwent relatively severe attack by 8alt 107 in
Table 7.

Corrosion-Product Concentrations of Salts Tested with Nickel-Molybdenum Alloys
Containing Aluminum in Combination with Other Alloying Elements

 

Aluminum Concentration of Salt

 

 

 

Heat Alloy Composition, at. % Tes? Samples,? mole 4
Duration
Number (hI‘)
Aluminum Other Components Sample H Sample C Other

S TR23014 1.30 10.8 Mo, 2.7 Ti, bal Ni 500 <0. 0009 <0. 0009 <0. 0009

S T23013 1.40 10.4 Mo, 3.2 Nb, 1.8 W, 500 0.160 0. 0606 b
bal Ni

OR 30-33 1.59 9.7 Mo, 5.9 Cr, 5.6 Fe, 1000 0. 0660 0.0820 0.124
bal Ni

OR 30-16 2. 54 10.2 Mo, 4.3 Cr, 1.9 Ti, 500 0.358 0.326 0.334
bal Ni

OR 30-22 2.61 10.0 Mo, 6.6 Cr, 0.4 Fe, 500 0.382 0.570 0.244
0.4 Wb, 2.6 W, bal Ni

OR 30-22 2.61 10.0 Mo, 6.6 Cr, 0.4 Fe, 1000 0.249 0.259 b
0.4 Nb, 2.6 W, bal Ni

B3277 3.06 14.2 Mo, 9.5 Cr, 1.6 Nb, 500 0.502 0,525 0.481
3.1 W, bal Ni

B3277 3.06 14.2 Mo, 9.5 Cr, 1.6 Nb, 1000 0.473 0.394 0.374
3.1 W, bal Ni

OR 30-14 5.51 10.4 Mo, 1.9 Ti, bal Ni 500 C.394 0.439 0.499

OR 30-14 5.51 10.4 Mo, 1.9 Ti, bal Ni 10060 0.250 0.174 0.0960

 

aSample designations "H" and "C" are discussed on page 13; "other" designates samples
obtained from metallographic specimens.

b'Not determined.

9c
 

 

 

 

 

 

 

" 27 (

the form of surface pitting'and subsurface void formation. Void formation N
was evident to a depth of 0.002 in. in 500-hr tests of these alloys and
to 0.003 in. in 1000-hr teéts.. Figure 8 shows the appearapée of an alloy
containing. 4.27 at. % Al after a 1000-hr test exposure, However, an
alloy containing only 1.22 éf. % Al revealed negligible attack after a
500-hr exposure to Salt 107.

Alloys containing aluminum combined with other alloying components
also exhibited both surface pitting and subsurface void formation after
exposure to Salt 107. The;depths of attack for these alloys are shown
graphically in Fig. 7.  Additions of up to 2 at. % Al resulted in only
light attack except in alloys_where chromium additions were also present.
Alloys which contained over 2.5 at. % Al in combination with chromium or
titanium revealed pfqnounéed'attack,ih the form of subsurface voids to

depths ranging from 0.002 to 0.0045 in. Cold-leg sections of all loops

 were unattacked and contained no insoluble corrosion products.

]
]

o 7-13132

 

s Jon o =

|

 

=

0.014 INTHES
£ 250X

 

 

        

. Fig. 8. Hot-Leg Surface of Ternary Nickel-Molybdenunm Alloy
Containing 4.27 at. % Al After 1000-hr Exposure to Salt 107. Heat OR 30-7.

 

 
28

Titanium

Corrosion-Product Concentrations

 

Titanium-containing alloys which were investigated are iisted in
Table &, together with attendant titanium corrosion-product concentra-
tions. Except for the first composition shown, which contained
2.47 at. % Ti, alloying agents in addition to titanium were present in
all of the alloys evaluated. The corrosion properties of the single
ternary alloy were studied at both 500- and 1000-hr time intervals.

Analyses of salts operated in loops of this alloy revealed titanium

Table 8. Corrosion-Product Concentrations of Salts Tested with
Nickel-Molybdenum Alloys Containing Titanium

 

 

 

 

Alloy Composition Test Titanium Concentration in
Heat (at. %) Duration Salt Samples® (mole %)
Number (hr)
Ti  Other Components Sample H Sample C  Other
OR 3C-8 2.47 11.6 Mo, bal Ni 500 0.039%94 0.0386 0.0380
OR 30-8 2.47 11.6 Mo, bal Ni 1000 C. 0404 0. 0495 G, 0505
B2E97 1.68 13.8 Mo, C.9 Nb, 10C0 0.0373 C.0383 0.0283
bal Ni
OR 30-16¢ 1.90 10.2 Mo, 4.3 Cr, 5C0 C.0378 0.0348 0.0500
2.5 Al, bal Wi
OR 30-14 1,92 10.4 Mo, 5.5 Al, 500 C. 0449 0. 0489 0.0378
bal Ni
OR 30-14 1.92 10.4 Mo, 5.5 Al, 1C00 0.0373 0.053C 0.0434
bal Ni
OR 30-13 2.02 11.3 Mo, 2.2 Al, 500 C.0389 0. 0404 0. 0484
bal Ni
S T23014 2.71  10.8 Mo, 1.3 Al, 500 C.0187 0.0242 0.0187
bal Ni
B2898 3.24 13.2 Mo, 0.9 Nb, 500 0.0353 0.0252 b
bal Ni

 

aSample designation "H" and "C" are discussed on page 13; "other”
designates samples obtained from metallographic specimens.

bNot determined.
 

 

 

29

concentrations of 0.038-0.040 mole % after 500 hr and 0.040-0.055 mole %
after 1000 hr. vThese*analyses-cbrrespohd closely to the analyses of
salts circulated in loops constructed of the other titanium-bearing
alloys, in which titanium contents ranged from 1.68-3.24 at. %. Only
the test of the 2.71 at. % alloy effected a titanium salt concentration
significantly different from the ternary alloy, the concentration in

the former test being unaccountably lower than for the other tests.

Metallographic Results

' The ternary titanium-bearing alloy revealed only light attack in
‘the form of surface pitting after both the 500- and 1000-hr test inter-
vals. The photomicrograph in Fig. 9 shows the appearance of a typical
hot-leg section of this alloy after the longer test interval. Depths of
attéck'which WEre‘dbsérvea in7the‘remainder‘of the titaniumdbearing
alloys are graphically depicted in Fig. 7. Except where aluminum addi-
tions were present in these alloys in amounts greater than 2 at. %s
depths of attack were in all cases less than 0.003 in. and‘géherally

 

 

fn

 

T

C.014 INCHES
Jx 250x

 

 

 

 

 

 

 

 

 

 

 

 

 

Fig. 9. Hot-Leg”Surface of Ternary Nickel-Mplybdenum.Alloy Con-
taining 2.47 at. % Ti After 1000-hr Exposure to Salt 107. Heat OR 30-8,

 

 
30

were less than 0,002 in., Attack in all cases was manifested as general
surface pitting and shallow vcid formation. No attack was seen in the
cold-leg sections of any of the titanium-containing lcops nor were cold-

leg deposits detected.
Vanadium

Corrosion-Product Concentrations

 

Loops were operated with two ternary alloys containing vanadium
additions of 2.73 and 5.11 at. %, respectively. As shown in Table 9,
the vanadium concentration detected in a salt mixture tested with the
former alloy after 500-hr exposure was less than 0.005 mole %. However,
salts operated with the alloy of higher vanadium content contained
0.027 mole % V in a 500-hr test and 0.019-0.020 mole % V in a 1000-hr
test.

Metallographic Results

 

Metallographic examination of the alloy with 2.73 at. % V revealed
very light attack in the form of surface roughening., Hot~leg surfaces
of the alloy with 5.11 at, % V exhibited attack in the form of void
formation to a depth of 0.002 in. in the 500-hr loop and 0.004 in. in
the 10C00-hr loop. A photomicrograph illustrating attack incurred by the
latter loop is shown in Fig. 10,

Iron

Corrosion-Product Concentrations

 

Only two loop tests were completed with alloys containing iron as a
major addition. Results of both tests, one of which operated for 500 hr
and the other for 1000 hr, are summarized in Table 9. In the 500-hr
loop, which contained 4.12 at., % Fe, after-test salt samples were ana-
lyzed to contain 0.013-0.015 mole % Fe. In the 1000-hr loop, which con-
tained 5,56 at. % Fe together with aluminum and chromium additions, the
after-test salt samples contained an even lower iron concentration. It

would appear that corrosion products formed by reactions involving
Table 9. Corrosion-Product Concentrations of Salts Tested with Nickel-Molybdenum
Alloys Containing Vanadium and Iron

 

Corrosion-Product Concentration

 

 

 

Heat Alloy Composition (at. %) Tes? in Salt Samples® (mole %)
Duration
Number (hr)
v Fe Mo Cr Al Ni Sample H Sample C Other
Vanadium Concentration
OR 30-10 2.73 10.8 bal 500 <0.005 <0.005 <0, 005
OR 30-20 5.11 10.8 bal 500 0.0274 0. 0274 0.0261
OR 30-20 5.11 10.8 bal 1000 0.0198 0.0194 b
Iron Concentration
OR 30-11 4,12 10.8 bal 500 0.0146 0.0134 0.0152

OR 30-33 5.56 9.7 5.9 1.6 bal 1000 0.00517 0. 00387 0.00517

 

fSample designation "H" and "C" are discussed on page 13; "other" designates samples obtained
from metallographic specimens.

bNot determined.

¢
 

32

 

0
tal
X
o
£
4
<
o

 

 

 

 

 

Fig. 10, Attack at Hot-Leg Surface of Ternary'Nickel—Mblybdenum

 Alloy Containing 5.11 at. % V. Alloy was exposed to Salt 107 for 1000 hr.

Heat OR 30-20.

chromlum and aluminum in this latter test served to inhibit the reactlon

with iron.

rMetallbgraphic Results

Metallographic ekaminations'df specimens from the 500-hr loop showed

no ev1dence of attack other than 1ight surface roughening. . Examinations
of the 1000-hr test, as 1ndicated in Flg. 7, showed corr031ve attack to
a depth of O 003 in. in the form of small subsurface v01ds.

g

CorrosionéProduct ConcentratibnSffjj,3+f>"

Loop tests of 500~ and 1000-hr duratlons were carrled out w1th ter-'

',nary alloys containlng 2.20 and 3. 62 at % Nb respectlvely..-As shown
_1n Tdble 10, the nldblum.concentratlons in ‘salts exposed. to these alloys
'_1ncreased from a value near O. 015 mole % in the 500-hr tests to values
_of_from'0.022 to 0.025 mole %_;n the 1000-hr tests. - Salts tested with

Ten

 

 
33

Table 10. Corrosion-Product Concentrations of Salts Tested
with Nickel-Molybdenum-Niobium Alloys

 

 

 

 

Alloy Composition Niobium Concentration in
Heat (at. %) Test Sal 3
. Duration alt Samples® (mole %)

Number (hr)

Nb Mo Ni Sample H cample C Other
OR 30-12 2.20 11.1 bal 500 0.00363 b 0.0135
OR 30-12 2.20 11.1 bal 1000 0.0225 0.0207 C.0233
OR 30-21 3.62 10.9 bal 500 0.0155 0, 0174 0.0L53
OR 3C-21 3.62 1C.9 bal 1000 G.C0244 0.0228 0.0259

 

a, . . 1 .
Sample designations "H' and "C" are discussed on page 13; "other”
designates samples obtained from metallographic specimens.

bNot determined.

alloys which contained niobium combined with other additions showed very
much lower niobium concentrations than did the simple ternary alloys.

As seen in Table 11, concentrations in the multiple-addition tests were
less than 0.002 mole % for niobium contents as high as 3.62 at. %. Thus,
it appears that corrosion products formed by reactions with other com-
ponents in these alloys, namely titanium, aluminum, and chromium, were

effective in inhibiting reaction of the salt with niobium,.

Metallographic Results

 

Neither ternary alloy containing niobium showed significant attack
in metallographic examinations of 500-hr tests; however, the presence of
very small subsurface voids to depths of approximately 0.001 in. was
detected in examinations of the 1000-hr tests, as illustrated in Fig. 11.
Corrosion results determined for alloys containing niobium together with
other additions are summarized in Fig. 7. Except where aluminum and
chromium were both present, attack in these alloys was less than

0.002 in. in depth.
34

Table 11, Corrosion-Product Concentrations of Salts Tested with
Nickel-Molybdenum Alloys Containing Niobium in Combination
with Other Alloying Elements

 

Alloy Composition Niobium Concentration in

 

 

 

Test
Heat (at. %) Duration Salt Samples? (mole %)
Number (hr)
Nb Other Components oample H Sample C Other

OR 30-22 0.39 10.0 Mo, 6.6 Cr, 500 0.00129 0.00181 0. 00207
2.6 Al, bal Ni

OR 30-22 0.39 10.0 Mo, 6.6 Cr, 1000 0.00052 0.00104 0.00052
2.6 A1, bal Ni

B2898 0.90 13.2 Mo, 3.24 Ti, 500 <0. 0005 <0. 0005 b
bal Ni

B2897 C.92 13.8 Mo, 1.7 Ti, 1000 0. 00104 0. 00052 C.001z¢2
bal Ni

B3276 1.48 14.0 Mo, 9.3 Cr, 500 0. 00078 0.00052 0.00078
bal Ni

B3277 1.27 14.2 Mo, 9.5 Cr, 500 <0. 0005 <0.0005 <0, 0005
3.1 A1, bal Ni

B3277 1.27 14.2 Mo, 2.5 Cr, 100C 0.00259 C.00181 0. 00104
3.1 Al, bal Ni

S T23013 3.23 1C.4 Mo, 1.4 Al, 500 <0. 0005 <0. 0005 <0.0005
1.8 W, bal Ni

 

a
Sample designations "H" and "C" are discussed on page 13; "other"
designates samples obtained from metallographic specimens.

bNot determined.

Tungsten

Corrosion-Product Concentrations

Single alloying additions of tungsten to the nickel-molybdenum sys-
tem were evaluated at levels of 0,72 and 1l.44 at. %, respectively. Loop
tests of both alloys were conducted for 500 hr, at which time tungsten
concentrations in the salt mixtures had reached levels of 0.029 to
0.032 mole %. In 500-hr tests of two alloys which contained tungsten in
addition to chromium, niobium, or aluminum, the concentration of tungsten
detected in salt samples was below 0,010 mole %. Compositions of these
alloys and their attendant salt corrosion-product concentrations are

shown in Table 12,
s

o

35

 

 

q‘

I T i~ =

[on

 

=

im O, 014 INCHES
% 250%

i
o

 

 

 

Fig. 11. Appearance'of;Voi&s in Nickel-Molybdenum Alloy Containing
3.62 at. % Nb after 1000-hr Exposure to Salt 107. Heat OR 30-21.

Table 12. Corrosion;Préduct'Concentrations of Salts Tested with
Nickel-Molybdenum Alloys Containing Tungsten

 

Alloy Comp051t10n

 

Tungsten Concentration in

 

 

 

designates samples obtalned from metallographlc specimens. -

bNot determined

Test a
Heat (at. %) Duration Salt Samples (mole %)
Number — (hr)
W Other Components _ Sample H Sample C  Other
OR 30-9  0.72 11.2 Mo, bal Ni 500  0.0318 . .0.0287  0.0325
OR 30-19 1.44 1L.4 Mo, bal Nt 500 0.0384  0.0311  0.0310
523011 1.72  10.2 Mo, 1.3 Al, 500 . 0.00500  0.00983 b
- 2,9 Nb, 4.8 Cr, | |
; o : ~ bal Ni ,_w_""_'w : | - .
S T23013 1.85 10.4 Mo, 1.4 41, 500 0.00223  0.00105 b
Samples designatlons "H" and "C" are discussed on page 13, "other"

 
36

Metallographic Results

 

Metallographlc exemanation of loops constructed of nlckel-molybdenum-
tungsten alloys revealed only”sllght,surface pitting, as illustrated in
Fig. 12. The addition of aluminum?together with tungsten led to heavy
surface pitting;”while the,addition_of;elpminum, chromium, and niobium

produced subsurface voids to a depth'of 0.0025 in., as shown in Fig. 7.

 

T-11961

s e N

|

 

o

0.014 INCHES
1@ 250x%

 

 

 

Fig. 12. Hot-Leg Surface'OffNickel-Mblybdefium.Alloy'Contdining_
1.44 at. % W after 500-hr Exposure to Salt 107. Heat OR 30-19.

-Relative'Thermodynamic SteoilitieeeofAlloying Constituents
Slnce the salt mlxtures supplied for th1s 1nvest1gatlon were very
7un1form 1n compositlon, one can presume that the same relative act1v1t1es
"of'UFl and UF3-eXJsted*at the start of each test. It follows, therefore,
that the relative extent of reaction which occurred between the salt
mlxtures and an alloylng element

xM + = :
M+ yUF, = MF + yUF;

X

 
37

for which
o
O”MXF “ur,
- xy
K, = = (12)
M UF'3

should be directly related to the activity of the fluoride compounds,
MxFy’ involving that element.

An indication of the relative activities of these compounds is provided
by a consideration of their standard free energies of formation, as

derived from the reacticn

I =
M + £ Fa(g) MF (13)
where
_ %
RT
M T
Xy

In Table 13 are listed the standard free energies of formation, per gram-
atom of fluorine, of fluoride compounds at 800 and 600°C associated with
each of the alloying elements investigated. Values are given for the
most stable compounds (that is, those with most negative free energies)
reported18 for these elements, and are listed in order of decreasing
stabilities. The resultant order suggesfs that corrosion-product
concentrations associated with each element at a given activity should
have increased in the following order: W, Nb, Fe, Cr, V, Ti, and Al.

In Fig. 13, the general ranges of corrosion-product concentrations
actually observed for these components, when present as single alloying

additions, are plotted as a function of alloy content. It is seen that,

 

18y, Glassner, The Thermodynamic Properties of the Oxides, Fluorides,
and Chlorides to 2500°K, ANL-5750.

 
38

Table 13. Relative Thermodynamic Stabilities of Fluoride
Compounds Formed by Elements Employed as Alloying
Additions (Data Compiled by Glassner)®

 

 

 

Most Suable AG8OOOC AGSOO o

Element Fluoride N\

Compound (-~E§E£h-——;\ (___EQEEE____‘

p g atom of F/ g atom of F/
Al AlF4 —&7 —92
Ti TiF3 ~85 —90
Vv VF> —80 84
Cr CrFs —72 =77
Fe Fels —66 —66
Nb NbF 5 —58 —60
W WF 5 —46 ~48

a

A. Glassner, The Thermodynamic Properties of the
Oxides, Fluorides, and Chlorides to 2500°K, ANL-5750.

with the exception of niobium and tungsten, the concentrations per atomic
per cent of addition do increase in the exact order predicted. Only
tungsten noticeably deviates from this predicted behavior; the causes
for this deviation have not been established, although the number of

tests completed on alloys with tungsten additions were quite limited.

General Discussion of Alloying Effects

The corrosion-product concentrations associated with either iron,
niobium, or tungsten alloying additions were much lower when these ele-
ments existed in multicomponent alloys than in simple ternary alloys.
The reason for this behavior is undoubtedly associated with the presence
of the more reactive alloying additions in the multicomponent alloys.
Consider, for example, a system containing comparable additions of

chromium and iron, for which the corrosion reactions are

Cr + 2UF, = CrFp + 2UF;3: AGI (]_4)

FeF, + 2UF3: AG.. - (15)

Fo + Uk IT
500

400

300

200

60

50

40

CONCENTRATION DETECTED IN SALT {mole % x 103)

30

20

10

39

ORNL ~ LR— DWG 46943

 

 

 

Al/

 

 

 

 

 

 

/

—— \—"

 

S 7
/

 

y

 

,

2
/7

 

 

7/
Nb/
0 T
/ /

/
/1N / 3/}/

 

 

 

 

 

 

 

 

 

 

 

Fig. 13.

2 3 4 S 6 7 8 9 | 2
ALLOY CONTENT (at. %)

Comparison of Corrosion-Product Concentrations Formed in

salt 107 by Various Alloying Additions as a Function of Alloy Content.
40

Since AGI is considerably more negative than AG the equilibrium UF;

)
concentration produced by the first reaction isliigher than that which
would be produced by the second reaction. According to the law of mass
action, therefore, the FeF, concentration at eguilibrium should be

reduced in the presence of chromium compared to a system containing iron
only.

While this means of corrosion inhibition was effective for those
alloying components less reactive than chromium, its effect was not
apparent among elements of relatively high reactivities. Thus, chromium
corrosion-product concentrations were the same in the case of alloys con-
taining both aluminum and chromium as in the case of alloys containing
chromium alone. Similarly, the presence of aluminum did not noticeably
reduce the titanium-corrosion products associated with titanium-containing
alloys. For most of the multicomponent alloys, the amount of chromium and
titanium, on an atomic percent basis, exceeded that of aluminum; conse-
quently their activities could have approached the activity of the alumi-
num component. Nevertheless, the high aluminum concentrations of fluo-
ride mixtures after these tests suggest a level of UF:z higher than would
have been produced by either chromium or titanium alone, so that some
inhibitive action would be predicted. The fact that none occurred may
suggest that the measured aluminum concentrations were somewhat higher

than actually existed,!”®

SUMMARY AND CONCLUSIONS

The corrosion properties of solid-solution alloying elements in the
Ni—17% Mo system were investigated in molten mixtures of NaF-LiF-KF-UF,
(11.2—41.0-45.3-2.5 mole %). The corrosion susceptibility of alloying
additions was found to increase in the following order: Fe, Nb, V, Cr,

W, Ti, and Al, With the exception of tungsten, the susceptibility of

 

19Considerable difficulty was in fact encountered in analyses of
this element by wet chemical methods. Analyses of a limited number of
samples were consequently made using semiquantitative spectirographic
techniques. Concentrations obtained by the latter technique were lower
by a factor of approximately 2/3 than the values that were reported in
Tables € and 7.
41

these elements to corrosion increased in approximately the same order as
the stabilities of fluoride compounds of the elements.

The corrosion-product concentrations produced by either iron, nio-
bium or tungsten alloying additions in the above salt mixture were much
lower when these elements existed in combination with chromium, titanium,
or aluminum than when they existed as single alloying additions. In con=-
trast, corrosion-product concentrations associated with chromium, tita-
nium, or aluminum were unchanged by the presence of other alloying
constituents.

The metallographic examinations of all alloys investigated under
this program showed considerably less corrosion than Inconel under equiv-
alent conditions. Surface roughening or shallow pitting was manifested
in hot-leg sections of all the loops tested. ©Shallow subsurface voids
were also seen in aluminum- and niobium-containing systems. Alloys con-
taining more than one alloying addition invariably were attacked to
depths greater than alloys containing each of the additions individually.
The greatest depths of attack, which ranged from 0.003 to 00,0045 in. in
1000-hr exposures, were incurred in alloys which contained combined addi-
-tions of aluminum and chromium or aluminum and titanium. Alloys without
either aluminum or titanium in no case exhibited attack to depths of
more than 0,002 in, and generally showed attack in the range from
0.001 to 0,002 in.

In the case of the majority of alloys tested, the depth of attack
increased at a greater rate between O and 500 hr than between 500 and
1000 hr. This finding is in asgreement with the concentrations of corro-
slon products, which in general increased only slightly between the peri-
ods of 500 and 1000 hr. Both results suggest that nearly steady-state
conditions were established within the first 500 hr of test operation.

In view of the generally favorable results of these tests, one has
considerable freedom in the choice of potential alloying components for
nickel-molybdenum alloys for molten salt service., Only titanium and
aluminum appear to afford potential corrosion problems, particularly if
used as combined additions or in combination with chromium. Thus, the
choice of an optimum alloy composition for any given application can be

judged mainly on the basis of strength and fabrication requirement.
42

ACKNOWLEDGMENTS

The many loop experiments reported here were carried out under the
skilled hand of M. A. Redden. The author is also indebted to W. 0. Harms

for his editorial assistance and guidance as thesis advisor.
1-3,
4=5,

6—15.
lé.
17.
18.
19.
20,
21.
22.
23.
24,
25,
26,
7.
28.
29.
30.
31.
32.
33,
34,
35,
36.
37,
38.
39,
40,
41,
42,
43,

45,
46,
47.
48,
49,
50.
51.
52.
53,
5.
55.
56.
57,

43

INTERNAL DISTRIBUTION

Central Research Library
ORNL Y-12 Technical Library

Document Reference Section
Laboratory Records
Laboratory Records, ORNL-RC
ORNL, Patent Office

HEFNPLORERDREIRNOANIOERDHEASSINEAANHAD TR X

Adams
Adamson
Affel
Anderson
Apple
Baes
Baker
Ball
Bamberger
. Barton
Bauman

. Beall

. Beatty
Bell
ender

. Bettis
Bettis
Billington
. Blanco

. Blankenship
. Blomeke
Blumberg

G. Bohlmann
J. Borkowski
E. Boyd
Braunstein
A. Bredig

B. Briggs

R. Bronstein
D. Brunton
A. Canonico
Cantor

L. Carter

I. Cathers
B. Cavin
Cepolina

M. Chandler
H. Clark

R. Cobb

D. Cochran

EEEE e E R

OsEHmnEOOE NS GQEY

58.
59.
€0,
el.
62.
63.
ST
65,
66,
67,
68.
69.
70.
71-80.
81.
82.
83.
84,
g5,
86.
87.
88.
89.
20.
91.
92.
93.
%.
95.
926.
97.
98.
99,
100.
101.
102,
103,
104,
105.
106.
107.
108,
109,
110.
111.
112,

oW aoasgsunduodgaxdyrEiigdn Qg mryudHGEOH PN OO oG E REO

QR m< =

0
»

SD_tIJZmOQCE}b"UD:‘ZQDU.O';E:EUW’;E:':N';C:brdigtljw:'d?drd?f—:}mgififi.fii_fi.z.fljw.b‘}_'

ORNL-TM-2021
Vol. I

Collins
Compere
Cook
Cook
Corbin

Crowley
Culler
Cuneo
Cunningham
Dale
Davis
DeBakker
DeVan
Ditto
Dworkin
Dudley
Dyslin
Eatherly
Engel
Epler
Evans I11
Ferguson
Ferris
Fraas

. Friedman
frye, Jr.
Furlong
Gabbard
Gallaher
Gehlbach
Gibbons
Gilpatrick
Grimes
Grindell
Gunkel
Guymon

. Hammond
Hannaford
Harley
Harman
Harms
Harrill
Haubenreich
Helms
Herndon
113,
114,
115-117.
118.
119.
120,
121.
122,
123,
124,
125,
126.
127,
128.
129,
130.
131.
132,
133.
134,
135,
136,
137.
138.
139,
140.
141.
142.
143,
144,
145.
146.
147.
148,
149.
150,
151.
152,
153,
154,
155,
156,
157,
158,
159,
160,
161.
162.
163.
164,
165,
166.

Fj?:?ijrfl?:?jg)Fj?4?j?1?j?j;:?=Fj?)?=?ig ?¢f)?4C)t4H]$>11C4U)C1C+m HORE9WHH s 0o @m=249Y

Hess
Hightower
Hill
Hoffman
Holmes
Holz

. Horton
Houtzeel

L. Hudson

R. Huntley
Inouye
Jordan
Kasten
Kedl
Kelley
Kelly
Kennedy
Kerlin
Kerr
Keyes
Kiplinger
Kirslis
Koger
Korsmeyer
Krakoviak
Kress
Krewson
Lamb

Lane
Larson
Lawrence
Lin
Lindauer
. Litman
Llewellyn
Long, Jr.
Lotts
Lundin
Lyon
Macklin
MacPherson
MacPherson
Mailen
Manning
Martin
Martin
Mateer

. Mauney
McClain

W. McClung
E. McCoy

L. McElroy

=0 E W

-

tflffi_’;dpb‘f)tij@.t“':'?;fil‘?‘l:‘*gfl"dtd'mutdbtrjilmk-%tfli:m-f.c—ckflz'JUCAF-EIde'.Ii

44

167.
168.
169,
170.
171.
172.
173.
174,
175.
176.
177.
178,
179.
180.
181.
182,
183,
184,
185,
186,
187.
188,
189.
190.
191,
192.
193.
194,
195,
196,
197.
198,
199,
200,
201.
202.
203.
204,
205,
206,
207,
208,
209,
210.
211.
212.
213.
214,
215,
216.
217,
218.

.(}:é‘..m.’;rj??:?d?ligf—cgl.hjgflfibhdfimfimmlflbbdb:flCqb‘QO

o

-

U

HEEEGUDEEDHYOQHE O & aE S

rIj.O.OEDOPDUE*I;‘FQ@;E.‘&EWO@@&!DSZW@QSUMQW

oo E e

a,

CDEDL—**“UCAZZDZ‘.

D

';Eitdb;meO

McGlothlan
McHargue
MclNeese
McWherter
Metz
Meyer
Moore
Moulton
Mueller
Nelms
Nichol
Nichols
Nicholson
Qakes

atriarca

Perry
Pickel
Piper
Prince
Ragan
Redford

ichardson

Robbins

. Robertson

Romberger
Ross
Savage
Schaffer
Schilling
Scott
Scott
Seagren
Sessions
Shaffer
Sides
Slaughter
Smith
Smith
Smith
Smith
Smith

iewak

Steffy
Stoddart
Stone
Strehlow
Sundberg
Tallackson

. ¥, Thomason
219,
220.
221.
222.
223,
224,
225,
226,
227,
228,
229,

240,
241,
242,
243,
244,
45,
246,
247,
248,
249,
250-251.
252,
253,
254,
255,
256,
257,
258,
259,
260,
261,
262,
263.
264,
265,
266280,

FEEAESNIAYIQREEa0

M.
L.
R.
E.
F,

TrOBOS ORINUQYEREEE "Iy 90

Fc

L. M.
D. B.
W. E.
G. M.
J. S.
H. L.
C. F.
B. H.
A, M.
J. R.
W. J.

45

Toth 230, K. W. West
Trauger 231. M. E. Whatley
Unger 232, J. C. White
Watson 233. R. P. Wichner
Watson 234, L. V. Wilson
Watts 235, Gale Young
Weaver 236, H. C. Young
Webster 237. J. P. Young
Weinberg 238. E. L. Youngblood
Weir 239. F. C. Zapp
Werner

EXTERNAL DISTRIBUTION

Allaria, Atomics International

Asquith, Atomics International

Cope, RDT, SSR, AEC, Oak Ridge National Laboratory
Deering, AEC, OSR, Oak Ridge National Laboratory
Dieckamp, Atomics International

ziambusso, AEC, Washington

Haines, AEC, Washington

Johnson, AEC, Washington

Kitterman, AEC, Washington

Larkin, AEC, Oak Ridge Operatiouns
McIntosh, AEC, Washington

Martin, Atomics International

Mason, Atomics International

Meyers, Atomics International
Reardon, AEC, Canoga Park Area Office
Roth, AEC, Oak Ridge Operations

Shaw, AEC, Washington

Simmons, AEC, Washington

Smalley, AEC, Washington

Stamp, AEC, Canoga Park Area Office
Stansbury, the University of Tennessee
Sweek, AEC, Washington

Taboada, AEC, Washington

Wilson, Atomics International

Laboratory and University Division, AEC, Oak Ridge Operations
Division of Technical Information Extension