'
.
.'
'
"!
!'
"
""
.
.
.
"
',
"
."
..
.'.

ALF
D
EPW2.
"E
"
.
9
.
..
,
· Il
1
Y. 1.
.
-
.
. *
..
.
.....
.
..
.
..
-
.
..
Liit
"..'
".
L
i
!
4
..
F
.....
.
..
:
.
.
"
Y":
*
-
.
,
i
.
p!
'I
. 171
T.:,
.
k
#
4. ...
?
TI
'
.
J
.
1.:-
.
.
::
.
.:
i
!
!
2
..
.
.
27'!"
. . .
.
2
UNCLASSIFIED
ORNL
14+
.
19...;
:
ro
P.
w
P
:
::
.
***
.
..
...
-
N
..
.
..
.......
..
......
.
..
7
per
:
:
:
.

+
+
CEA
SA
.
"
.
.
.
-
:-*
::
L
.
'
*
:* ,
,
::::::
.
,
.
WAT
WINY
1
.
.
:...!
,"..-
:""
"
..
1152

:
:
:
:
..
A..
.';
...
TF:-
..
..
..
:.
*'
..
1
.1.1.
7
...
:
L
T
:.
WWW
#.
.. Yurt
.!
...
"..
ORN-0-1152
...
#"!
CONE-65041
OSNL NEC - OFFICIAL
APR 27 1980
,
TY)!110-IYIKO
Haper vo de presented at Meeting of Reiractory Composites Working Group
cu Georgia Institute of Technology, Atlanta, Georgia, April 12-14, 1965.
.
.
.
...
W
:1 R
1
the
MASTER
NE
TAR
SUDARY OF THERMOCHEMICAL DEPOSIITON STUDIES
AT OAK RIDGE NATIONAL LABORATORY*
AUGUST 1964- MARCH 1965
.

:
:
:
i
:
.
R. L. Heestand** J. I. Federer,
F. R. Patterson, and C. F. Leitten, Jr.
Metals and Ceramics Division
Oak Ridge National Laboratory
Oak Ridge, Tennessee
.:
:
1
..
.
.
.
.
cor-
erd worth Mother the Doctor
raport
, wman:
preveder pie
LEGAL NOTICE
de productos report may
m
my Hero w mpwed, e for deg meg for the
amatierten
t
to de
momento delom
,
prepared
my
warhorn
ante tormenta
A W
A
The
Wolontar contract
te Coll, or played monster peen,
nous in, wo dute per
t
tegu mund
moment,
e
wel
PATENT CLEARANCE OBTAINED. RELEAN TO
THE PUBLIC IS APPROVED. PROCEDURES
ARE ON EHLE HY THE REDNING SICHON.
W THE RESEO PROMEDICALE TO
V
.
.
OANI ÅEC - OFFICIAL
scurch sponsored by the U. S. Atomic Energy Commission under
Cvatrac with the Union Carbide Corporation.
ORNI-AIC - OFFICIAL
**speaker.
TRE

OINAC - Oruc
ABSTRACT
Progress in chemical vapor deposition studies at Oak Ridge
cutionem Laboratory for the period of August 1964 through March 1965
is reviewed. Particular emphasis is given to the studies carried out
on üeposition i tungster-rhenium alloys and deposition of theoretically
dense uranium dioxide. A brief resume of the status of other vapor
deposition and deposition of tungsten sheet for mechanical and physical
properties evaluation.
INTRODUCTION
The chemical vapor deposition program at the Oak Ridge National
Laboratory is primarily wirected toward developing fabrication tech-
colocy for high-temperature reactor fuel elements. The investigations,
Therezore, included deposition of both refractory metals and alloys
suitabie cor cladding fuel elements and deposition of refractory fuel
simpounds. Currently emphasis is being placed on deposition of tungsten
deposition process is being characterized over the temperature range of
450 to 750°C at a total syötem pressure of 10 torr and, in the case of
alloys, a hydrogen-to-WF6 + ReF6 mole ratio of 20:1. The deposits were
examined by chemical analysis, metallography, hardness and x-ray dif-
Tract.on. Although the goal of depositing homogeneous alloys has not
yet seen compietely attained the results of the study are encouraging
and provide a background for continuing efforts to optimize the process.
fi one-step conversion of uranium hexafluoride to uranium dioxide
nas been achieved by combining the uranium hexafluoride with hydrogen
dJ. I. Federer and C. F. Leitten, Jr., "Vapor Deposition and
Characterization of Tungsten-Rhenium Alloys," paper presented at the
America Suciety Symposium on Vapor Deposition Techniques for Fabricat:
of Fuels and Refractory Metal Tubing, San Francisco, California,
November 30-December 3, 1964. Submitted to Nuclear Applications.
ORNI - AIC - OFFICIAL

3
and oxygen at elevated temperatures mů at reduced pressures. 2 The
reaciion product may be obtained as a submicron size powder, dendritic
crystallites, or a theoretically dense solid depending upon reaction
pressures and cas mixing techniques. The stoichiometry of the deposit
may be controlled by adjusting the ratios of the gaseous reactants.
It is anticipated that other refractory oxides may be produced
cither as coatings or Tree-standing theoretically dense bodies by the
use or similar techniques.
Preliminary studies of joining both wrought tungsten and pyrolytic
tungsten by tungsten vapor deposition have been completed. Joint designs
Save been optimized for flat stock and tube end closures. A more com-
piete program including mechanical properties and optimization of
deposition rates is anticipated.
Fabrication of pyrolytic tungsten plate for properties evaluation
is continuing. Difficulties were encountered during early runs in
octaining sufficient material for evaluation from a single experiment.
mese difficulties have been overcome and approximately 100 in.2 of
stock 0.060 to 0.080 in, thick is presently being produced in each run.
Experimental procedures and results for each of the above topics
are discussed separately in the following sections.
Vapor Deposition of Tungsten-Rhenium Alloys
Deposition studies for tungsten-rhenium alloys from their
respective hexafluorides were conducted in the equipment shown schemat-
ically in Fig. 1. The apparatus is similar to that previously described
ror tungsten depositions with the addition of a separate system for
2R. I. Heestand and C. F. Leitten, Jr., "Thermochemical Reduction on
Uranium Halides for the Direct Fabrication of Ceramic Fuels," paper pre-
sented at the American Society Symposium on Vapor Deposition Techniques
for Fabrication of Fuels and Reiractory Metal Tubing, San Francisco, Cali-
zornia, November 30-December 3, 1964. Submitted to Nuclear Applications.
38. L. Heestand, J. I. Federer, and C. F. Leitten, Jr., Preparation
and Evaluation of Vapor Deposited Tungsten, ORNL-3662 (Aug. 1964).
---ORNI - AEC - OFFICIAL
TY
!10-03 V-1822
1:30:9-JY-1050
VOLASSIFIC!)
()!!! .1? 03-6120
DEPOSITION FURNACE
SCAVENGER FURNACE
Ci.
....
...
mp3l3.1



BURNOFF
INS
.w!
...
I de 17



COLD TRAP
.
.
.
.
VACUUM PUMP
SCRUBBER
tw
DRAIN
.
tom
Figere lo
Vapor-Deposition Apparatus
..
..
.
ORMI-A1C - OINICIAL
.
ORNI - AC-ORIICIAL
:::
FB :
A
:
. .
.:
v1
, 1 • cả tư *
*
*
*
* *
*
*
*
02H!-AIC - OP!!
.
metering rhenium hexafluorides. The metered gases are premixed prior
to entering the main deposition furnace where reduction of the fluorides
occurs. Products of the reaction are then deposited is the solid metal
or alloy on tre substrate with hydrogen fluoride as an effluent by-
product. Any unreacted fluorides are then reduced in the scavenger
urnuce which is maintained at 900°C. The remaining gas consisting of a
mixture oi lydrogen fluoride and excess hydrogen passes through the cold
tray to the vacuum pump which maintains system pressure. The cola trap
is maintained at approximately -70°C to prevent back diffusion of pump
cil into the deposition chamber, thus minimizing carbon contamination.
Ide gases finally pass from the cold trap through the water scribber
where hyárogen fluoride is removed. Copper tubing measuring 3/4-in. ID
was selected as the deposition substrate to facilitate removal from the
deposit by dissolution in acid.
Prior to starting codeposition experiments, exploratory studies
on rhenium deposition were made. The study was not sufficiently exten-
sive to determine optimum conditions for rhenium deposition, but the
effects of basic parameter changes were apparent. It was found that
rhenium deposited nonuniformly at temperatures of from 500 to 800°C,
rhenium hexafluoride-to-hydrogen ratios of 200 and 300:1, and a pressure
of 10 torr. The deposits consisted of rough nodular low-density growths
at the inlet to the reaction zone. Lowering the deposition temperature
to 400°C resulted in more uniform deposits. Approximately the same
results were obtained at 500 and 600°C by decreasing the hydrogen-to-
rezium hexafluoride ratios to 100:1 and 50:1, respectively. Previous
studies on deposition of tungsten by hydrogen reduction of tungsten
hexafluoride3 at 10 torr had indicated uniform deposits at 500°C; there-
fore, experiments in codeposition were conducted in the range of 450
to 700°C.
meters for rhenium hexafluoride due to interaction with glass and fluori-
nated hydrocarbons; therefore, mass flow meters of metal construction
ORNI - ALC-OSFICIAL
310-32.-1470
Chemical analysis of the deposits prepared using the above tech-
miques indicated that concentration gradients exist in the tubes. Since
reniuü exažluoride is more readily reduced than tungsten hexafluoride,
the deposits were high in rhenium at the inlet to the reaction tube.
The longitudinal variation in rhenium content is shown in Fig. 2 as a
üunction oi selected deposition conditions. Recovery values were
etermined by comparing the metal content of both metered fluorides with
the weicht of deposit obtained in the deposition furnace.
The curve labeled Wre-16 is typical of deposits prepared at tem-
peratures from 500 to 700°C - a high rhenium content near the inlet
zollowed sy a rapid decrease in rhenium content with distance from the
inici. Tre curves labelea WRe-18 and -19 represent deposits prepared
at 500 and 600°C, respectively, in which argon was included with the
reacting gases. By a mechanism that is not yet completely understood,
argon caused the rhenium to be distributed differently throughout the
deposit. Comparison with WRe-16 shows that the rhenium content was sub-
stantially increased in the "downstrean half" of the deposits. In fact,
the rhenium content in WRe-19 was essentially constant over the distance
ö to 12 in. from the inlet. The presence of argon, however, lowered the
metal recovery in the main deposition furnace as shown in Fig. 3.
Lowering the deposition temperature to 450°C also caused a significant
change in rhenium distribution as shown by the curve labeled WRe-20. The
high rhenium content ncar the inlet reflects the greater ease of reduction
of Relo compared to WF6 at 450°C. Analysis of the data from several
experiments reveals that approximately 50% of the rhenium, but only 55%
of the tungsten, was recovered in the main deposition furnace.
Deposition rates for tungsten-rhenium deposits were different from
those of unalloyed tungsten as shown in Fig. 3. The curve labeled W-600
shows deposition rates as a function of distance from the inlet for
unalloyed tungsten deposited at 600°C. In general, the deposition rates
for tungsten-rhenium deposits, as typified by curves WRe-16 and -19,
were greater near the iniet than for unalloyed tungsten. In addition,
comparison of curve Wre-19 with, -16 shows that deposition rates were
suppressed by the argon. The grain structure of these deposits was
SI
1.
ORNE-AIC - OFFICIAL
19131350 - 53V-INDO
MMO
-O3Y- THL
+
.
Fig. 2. Composition of Tungsten-Rhenium Alloys vs Distance from the Inlet,
of the Main Reaction durnace.
UNCLASSIFIED
ORN! --OWG 61-10145
80
menovanj - ** ----
WRC-16 WRC-18 WRC-19 Wie-20
TEMPERATURE (°C) 600 500 · 500 450
PRESSURE (milio) 10 10 10 10
Ha (cc/min)
1500 1500 1500
WRC-201
1500
WE (cc/min)
60 60
ReFc (cc/min)
Ar (cc/inin)
-
RECOVERY (%) 89

60
60
10
10
500
41
500
54
25
"
RHENIUM CONTENT (%)
IN
on the
in
WRe-18
t
entionalmente contatti
WRe-19
1
17h.
WRe-16
2
10
.
12
6
8
DISTANCE FROM INLET (in.)
.
A
ORNI - A8C - OSSICIAL
ORMI-AIC - OIFICIAL
A
**
er
.
..
..
.
..
.
.
.
.
.
.
.
-
1101110 - 33 V - 1180
W!!!!"..
.
***
..
.
.
.
.
14
.
UNCLASSIFIL...
3. Deposition Rates of Tungsten-Rhenium Alloys vs Distance from the
Inlet of the Main Reaction Hirnace.
1. ORNL-DWG 6410144

**
TEMPERATURE (°C)
PRESSURE ( HO)
Ha (cc/min)
WFG (cc/min)
Refo (cc/min)
Ar (cc/min)
RECOVERY (%)
W-600 WRC-16 WRe-19
600 600 600
10 10 10
1500 1500 1500
60 60 60
- 10 10
- -- 500
98 89 54
WRe-16
· DEPOSITION RATE (mils/hr)
.
.
__WRe-191
VW-600
10
12
4
6 8
DISTANCE FROM INLET (in.)
ORNI - ABC - OFICIAL
OPNL-71C - OTTICIAL
-AEC - OFFICIAL
se
gererally columnar as is round with pure tungsten deposition. With
increasing rhenium content more needle-like columnar grains are formed.
The interstitial content of two deposits as a function of rhenium
content and distance from the inlet of the reaction zone is shown in
Table 1. There is a general tendency for the impurity content to decrease
with decreasing rhenium content (increasing distance from the inlet).
Deposits containing greater than about 25% Re were attacked during
dissolution of the copper deposition tubes in nitric acid, reducing the
deposit to small pieces having considerable surface discoloration. Thus,
nigher cxygen values near the inlet may be the result of contamination
rom the acid dissolution.
In aüüition to chemical analyses and metallographic examinations,
e deposits were characterized by X-ray diffraction and hardness measure-
cienus. X-ray diffraction patterns were cötained using powder samples oi
various rzenium content. The results which have been more fully reported
Casewhere", are summarized in Table 2. Two phases were found in as-
aeosited samples: alpha-tungsten (body-centered cubic) which was the
Principai. phase in low rherium deposits and veta-tungsten (cubic) which
4iavcreu by high rhenium contents. The lattice parameters on the
alpha-tungsten phase approach the value ior pure tungsten, a = 3.1648 A,
with decreasing rhenium content, indicating that this phase is simply a
solid solution of rhenium in tungsten. The beta-tungsten structure
commoniy occurs as an intermediate phase in transition metal systems. 5
10 our knowledge, this is the first reported occurrence of the phase in
the tungster-rhenium system. This phase does not appear in previously
reporteü prase diagrams, although these diagrams have not been established
below abouü 1200°C. 6,? The possibility exists that the beta-turgsten
phase results from preferential substitution of rhenium on certain
crystallographic planes of tungsten to produce an ordered structure that
'
* I. Federer and R. M. Steele, Nature 205(4971), 587 (1965).
5. V. Nevitt, Trans. AIME 212, 350 (1958).
75. J. English, "Binary and Terrary Phase Diagrams ox Columbium,
olybdenum, Tantalum, and Tungsten," DAC Report 152, p. 92, Battelle
memorial Institute (April 1961).
10
Table 1.
Analysis of Impurities in Tungsten-Rhenium Deposits
Distance
from
Inlet
Deposition
Temperature
(°C)
Rhenium
Experiment
dude:
Impurities (ppm)
(in.) a
WRe-12
500
sco
46
120
on
<5
<5
<5
<5
<5
1179
.8 34
2 9
2 9
2 ó
2 15
3 2006
h 30
2 <5
WRe-22
600
<20
1. 46
3 29
5
10
9 6
di 4
1.5 37
3.5 15
5.5 5
7.5 20
9.5 1
2.5 , 32
405. 19
6.5 10
8.5 5
10.5 3
w
120
20
<20
<20
<20
<20
<20
1
20
30
<20
<20
<20
<20
<20
<20
<20
m in
û û û ûn
Ĥ A N
ca
ke-23
500
500
<20
<
6
3
i
1
4
<
34
55
14
8
41
<5
18
6
<5
<5
<5
motal length of deposition zone 12 in.
nagh values possibly associated with chemical separation of
acrosit from copper tube.
Tv11:30 - SY-
Hivi un
Table 2. Results of X-Ray Difiraction Studies on Typical
Tungsten-Rheniun Deposits
Bheniamond
(5)
C-Tungsten
Intensity a, A)
B-Tungsten
Intensity & TAJ
Sigma Intensity
As-Depositeå (600°C)
37
4.997
S
5.010
5.018
3.144
3.152
3.162
3.165
NI
3.165
3.165
NF
Lex
po = 9.633
= 4.985 A
wt .
S
Annealed (1500°C, 4 hr)
3.146 S
4.996
3.146
5.011
3.154 S
5.014
3.162 27 F
3.165
MF
3.164 NT
3.165
S
NOTE: S, strong; M, medium; W, weak; NF, not found.
- ORNI-AIC -OSTICIAS
is zot readily observed metallographically. Rhenium might be capable
0: nigh ocentration packing on the (111) of tungsten because of
reasonabnü sinilarity to the (1000) of rheniun. The small variation
oi tie veia-tungsten lattice parameters over a wide range of rhenium
concentrations may be due to the ability of tungsten to accomodate
rhenium on preferential sites. This substitutional behavior would
explain the hardness increase corresponding to increasing rhenium con-
tent of the deposits.
Signa phase, which is reported to occur in alloys containing from
26 to 630 Re,? has not been round in any of the as-deposited alloys
within this range oi compositions. As shown in Table 2, the sigma
phase was identified in the deposit containing 37% Re following a
vacuum anneal at 1500°C. Otherwise, this heat treatment resulted only
in small changes in the lattice parameters of the other deposits. No
evidenca os elemental rhenium was found in as-deposited or heat-treated
Sampies.
liardness values of two deposits are shown in Table 3. The sharp
increase in hardness with increasing rhenium content coupled with the
Cenerally excellent purity of these deposits support the x-ray evidence
oi alloying at the deposition temperatures of 500 and 600°C. As pre-
viously mentioned, the high hardness of these alloys could be associated
with a preferential substitution of rhenium on certain crystallographic
planes of tungsten.
Further studies to improve the homogeneity of deposits are under
way. 22:ects row being investigated include zone heating of substrates
and modification of the gas injection system.
Deocition of Uranium Dioxice From: Uraniuin Hexafluoride,
Oxygen and Hydrogen
To date, our work has centered on the thermochemical deposition of
UO2 by the reaction of hydrogen and oxygen or steam with UF, as follows:
(1)
SU- JY-TRO
Table 3. Room Temperature Hardness of
Tungsten-fheniun Deposits
Rhenium
(%)
Haraness
(DPH)
Re-12
1775
1960
1730
570
370
350
325
o un año tu o t ñ ☆
WRe-16
1680
2000
425
350
380
Ünalloyed Tungsten
450
ORNE-AEC - OSSICIAL
14
ORNIHrac
(2)
(3)
(4)
(5)
The reaction of these materials has previously been proven thermo-
dynamically feasible when accomplished in a series of steps as shown
in rcüc vion paths 2 and 3 or 4 and 5; however, the process was directed
toward deposition on fluidized goed particles rather than as a free
standing, derse deposit, and difficulties were encountered in obtaining
a high-density low-fluorine deposit.
UF 6 + 7.2 → UF4 + 2H7 ,
UF4 + 2H26 - V02 + 487 .
UF 6 + 2H20 → UO2F2 + 4HF,
UO 2F2 + H2 → UO2 + 2HF .
me overall reaction as shown in Eq. (1) is believed to proceed simul-
Cuneous-y oy the paths shown in Eqs. (2; and (3) or (4) and (5); however,
ja matery be cotained as a dense, solid deposit; a crystalline needle-like
üüposiü; or a powder with neither UF4 ncr UO2F2 detectable in the system.
van analysis was made of the thermodynamic and kinetic factors that
wtect the rate and efficiency of the one-step reaction, and it was
found that i'accors favoring direct reduction were: (1) low system
pressure, (2) temperatures sufficiently high to limit formation or con-
densation ci possible intermediate compounds, and (3) excess amounts of
both hydrogen and oxygen to drive the reaction to completion. Based on
Ciese assumptions, experiments were ruri to confirm the erfects of temper-
ature, ressure, and gas composition on the reaction product.
A schematic (Fig. 4) shows the apparatus consisting of a resistance
heated, 22-40% Rh tube furnace, a nigh-density alumina reaction tube,
borosilicüie glass flowmeters, a cold trap cooled to approximately -50°C
ANA
cür hyà: cen fluoride removal. It was necessary to inject the UF6
Cirectiy unto tne hot zone of the furnace to prevent preliminary formation
ci' the intermediate products UF4 or UO2F2 in the cold zones, thus losing
materia: by condensation. Figure 5 shows the arrangement of the water-
cooled injector used to introduce the UF6 into the hot zone of the system
ORNI - ALC - OFFICIAL
BI. E. Knudsen, H. E. Hootman, and N. M. Levity, A Fluia Bed Process
zor the Direct Conversion of Uranium Hexafluoride to Uranium Dioxide,
AL-6066, Argonne National Laboratory (1963).
M
W AYANG
#
....
"
.
...!" .
..
.. ..
R
TID1310 - OV-1N8O
TV1013!() 937 - 110
E AK *
UNCLASSIFIED
ORNL-DWG 64-10778
".
.
.
.
.
Fig. 4. Schematic of the Fucl Deposition Apparatus.

hu:
H2--
DEPOSITION FURNACE
ja minera
COLD TRAP
7
4-2015
*
.
wees.
.-..-......
....
....
........
. *****.
BURNOFF
.
.
H2O
A-
VACUUM PUMP
SCRUBBER
ORHIMIC - OFFICIAL
OENI-NIC - OPTICIAL
.
..LIS
.
T.E
1
1
1
-23V-IKO
1191310 - 93 V - INZO
UNCLASSIFIED
ONN- OG 64.9982
. za..
"
!
PAVY
Fig. 5. Details of Uranium Hexafluoride Injection System.
1
.rtriting A
.
1
1
.

!
.
.
1
LOW DENSITY ALUMINA
SUBSTRATE TUBE)
- HIGH DENSITY ALUMINA ,
;; CONTAINMENT TUBE !
ARGOIV
--HO IN
-UFO
...
.
.
.
...
.....
..
.....
H20 OUT
-02, H2

VO2 DEPOSIT
.
.
.
- FURNACE
V
'
S
.
ORHI-MIC-OIFICIAL
ORNI ATC - OIIICIAL
xi..!
.
.
..
.

17
Eco Con Low-density alumina deposition substrate that was used to
cucinevate removal of samples. In addition, argon is used as a blanket
CÔ separate the reactants at the injector tip and to prevent buildup of
102 on the injector.
io tio deposition temperature. Following adjustment of the argon, hydro-
Cete, üric oxygen flows, the UTé is introduced and adjusted to operating
3-ster prescure. No further system changes other than minor i low adjust-
we w ure required throughout the rün. On shutdown, the gas flows are
..vorne at temperature and pressure is reduced to < i torr. The furnace
iw chci allowed to cool to ambient temperature prior to removal of the
wpciv
wizont limits of investigation in the experiments consist of
zvücuion temperatures from 900 to 1500°C, pressures from 2 to 20 torr,
nad w ceci composicions ranging fro. 30 to 80 vol % H2, 10 to
ven 02 or steering and I to 10 vol % UF. System parameters have
.vü Ben oprinized in regard to all variables; however, several trends
9 Eparenc. In the temperature range of 1200 to 1500°C and with
Dessutom wciow 10 torr, UO2 is formed when gas composition limits are
held w in 60 to vol 112, 10 to 40 voi 02 or steam, and I to
no voljo UF:.
IT UF', is injected directly into the hot zone sc. that mixing occurs
asi che desired temperature range, the deposit may be obtained either as a
mint aendritic, crystalline growth or as a solid, dense, uniform deposit.
in the solids, it has been found that stoichiometry may be con-
occia By adjusting the feed-gus composiu.on at a given temperature and
:0 2. Reproducible deposits have been made for oxygen-to-uranium
music ringing from U02.001 to 102.166• Fiuorine contents are generally
biela b ow 50 ppm. Bulk densities for solid deposits have been found to
weivoc 02 theoreticaily dense oxide (10.96 $/cc).
thin the gas conposition limits established for depositing bulk
CO2, rü ibuna that < 1% of the feed material passes through the system
ww dovodioce or unreacted fuel compounds. However, when UFG concern
weininni en ühe i'eed material was raised above 10%, in addition to the
OF HI - ABC - OFFICIAL

ORHI-AIS-Orrical
18
VO2 iconed in the hot zone, excess intermediate fuel compounds passed
Carousia re system and were observed in the cold trap. For steam or
oxygen concentrations in excess of 40%, trace amounts of intermediate
urarium conpounds were again observed in the cold trap. Unreacted
material under the above conditions may be due to the relatively small
system size. Raising the system pressure above 20 torr yields UO2F2
powder under conditions that produced only 102 at lower pressure. Within
the current limits of investidution, temperatures above 1300°C give no
significant deposition advantage.
The above results indicate that for a high-temperature low-pressure
hydroiysis reduction of UF, the steam or oxygen concentration promotes
the formation of UO 2F2, which is reduced by hydrogen to UO2. Conversely,
nydrogen reduction of UF6 to UF4 proceeds at the reaction temperature
but is suppressed by the presence of steam or oxygen. Thus, by selecting
the proper combination of parameters, the desired product, UO 2; may be
formed. A theoretically dense, uniform, low-fluorine deposit has been
achieved using the one-step conversion process with control of stoichio-
metry. In addition, it has also been demonstrated that 102 may be
produced as an extremely fine powder or a dendritic, crystalline material.
Both of these are potential feeå material for subsequent fuel fabrication.
Iiivestigations of deposition parameters for VO2 in all three forms are
Joining of Tungsten by Vapor Deposition
The method for depositing tungsten metal by thermochemical reduction
o tungsten hexarluoride with hydrogen has been investigated for joining.
both wrought and pyrolytic tungsten. The process for tube deposition was
used' and only minor charges in fixturing and reactant flow distribution
were necessary to produce joints suitable for demonstrating the process.
Several joint ãesigns, including flat plate and tube end closures, were
Tabricated and samples were prepared for Materials Advisory Board (MAB)
type ductile-to-brittle transition testing.
-AEC - OFFICIAL
Is joestand, j. I. Federer, ana C. F. Leitten, Jr., Preparation
wa Everton of Vapor Depositeä Tungsten, ORNL-3662 (Aug. 1964).
-------OG

MI - AEC - OFFICIAL
19
In studies of joint design effects it was noted that deposits in
joint angles of 90° or less gave cracking at the intersection of the
angles or radii equal to 0.5 in. tanceni to the root of the wela, samples
without cracking were ootained. On metallographic examination of the
Laterzace between pyrolytic and wrought tungsten at high magnifications
(500X) no visual evidence of voids, iracture, or impurity buildup could
be detected.
Joints of end caps to tubes were also demonstrated by fabricating
a closca-end tungsten tube by pyrolytic deposition, cutting off the end,
and rejoining the end again by pyrolytic deposition. End closures of
this type were leak-tested by a helium mass spectrometer leak detector
and round to be leak-tight. Several flat plate specimens or pyrolytic
Chagsies. wrought tungsten and wrought turigsten to wrougai tungsten
were in oricated for bend testing. Testing of the joints is presently
under way.
Deposition oz Pyrolytic Tungsten Plate Material for
Mechanical Properties Evaluation
Work has continued to develop techniques for depositing tungsten
diate .ving high chemical purity and or reproducible physical charac-
Cerisüics. Problems investigated include the elimination of nodular
vzowths within deposits, matching of ceposition substrates to the
aeposi optimization of the deposition process to yield uniformly
whick verial for subsequent rechanical and physical property evaluation.
02 prime importance is the eliminat:un of rough nodular growths within
the deposited, tungsten. These nodules lead to abrupt changes in such
physical characteristics as density purity and grain orientation.
It is assumed that growth of the nodules is associated with initial
impurities in the system or the formation and precipitation of inter-
mediate compounds.
zormation of these growths have been reduced by lowering deposition
temperatures to less than 600°C and maintaining a leak-free system.
OFNI – AC - OFFICIAL
08:11-AEC - OSSICIAL
150 - 318-IMNO
Previously tungsten has been deposited on copper substrates for
convenient removal by acid dissolution. It has been found that copper
is ncü suitable as a substrate for deposition of flat plate. Cooling
from the deposition temperature results in warping of the deposited
cungoten due to the wide difference in thermal expansion of the materials.
Preliminary deposition experiments using molybdenum foil inserts
in rectangular copper tubes showed that the warpage of the deposit could
be eliminated. However, difficulties were encountered in fabricating
che liners from molybdenum foil. Consequently, molybdenum deposition
r.andreis encased in copper are now being made using powder-metallurgy
cecaniques.
previously reported, difficulties are encountered in depositing
ions canovas or tungsten of uniform thickness.20 This problem stems
Proin temperature variations in the deposition furnace and from fuel gas
apletion as the deposit progresses down the mandrel. In the nonstatic
Cype og system usea to deposit tungster, the longitudinal uniformity of
wie deposit was shown to be a function of the reaction efficiency. These
wysteris utilize a stationary gas ejector and fixed-temperature profile
in the deposition furnace. Thus, at high deposition efficiencies (which
are readily attainable) it has not been possible to remove the generated
ayárogen fluoride gas from the substrate and supply an equal amount of
fresh gas among the entire marürel length. Since the deposition effi-
ciency is a 30 a function of H2-W concentration, pressure, temperature,
mandrel condition and mandrei geometry, it is possible to select con-
citions that optimize the efficiency of the reaction to yield uniform
üepciäis. With the aforementioned parameters near optimum in our non-
svatic syocom, uniiorm deposit thickness was approached only when
approxima Culy 50% of the WF'G passed through the system unreacted. The
reactici utriciency was primarily controlled by reducing the system
Pressure and introducing a heavy inert gas to facilitate removal of the
hydrogen müoride from the substrate wurface.
17. C. Thurber, Metals and Ceramics Div. Ann. Prog. Rept.
June 30, 1964, ORNL-3450, pp. 132-139.
Further improvements in deposition uniformity were achieved by
preheating the fecü gases and moving the high-temperature deposition
zone along the length of the mandrel while holding the remainder of the
mandrel at a lower temperature. Previously, this moving hot-zone tech-
nique was proven satisfactory for depositing long, tubular lengths of
tungsten.' Variable temperature was afforded by several taps along the
Sensh of a series-wound tube furnace. Recent deposit-thickness-
uniformity studies were made in 1 1/2-in.-diam copper tubes 16 in. long.
mo date, the schedule presented in Table 4 appears to be near optimum
L'or deposit uniformity with thicknesses being (0.050 $ 0.003) in, within
a 12-. length. Continuing work is being directed toward higher
purification or the source gases and analyses of any foreign compounas
which appear in the system.
1
Table 4. Deposition Schedule for Moving-Temperature Zone
Cús ilowa (cc/min)
WF' H2 Ar
37
312 to-
WF6 :
(ratio)
Pressure
(torr)
Temperature (°C)
Moving Static
200
3000
1000
5
600
450
"Gases preheated to approximately 200°C.

-
.
T-
..
.
'
L
A
..
.
:
UKU
...
.
14
+1.
-
-
I.
1
..
PE
w
01d
13
14
27
...
i
RY
M
1.
12
..
.
.
?
....
1
1
.
".'
..
.
DATE FILMED
5 / 26 /65
1.5
Y
LA
..
.
12
2
FE
-LEGAL NOTICE
This report was prepared as an account of Government sponsored work. Neither the United
States, nor the Commission, nor any person acting on behalf of the Commission:
A. Makes any warranty or representation, expressed or impliod, with respect to the accu-
racy, completeness, or usefulnose of the information containod in this report, or that the use
of any information, apparatus, method, or process disclosed in this roport may not infringe
privately owned righto; or
B. Assumos any liabilities with respect to the use of, or for damager resulting from the
use of any information, apparatus, method, or process disclosed in this report.
As used in the above, "person acting on behalf of the Commission" Includes any om-
ployee or contractor of the Commission, or omployee of such contractor, to the extent that
such employee or contructor of the Commission, or employee of such contractor propares,
disseminates, or provides access to, any information pursuant to his employment or contract
with the Commission, or his employment with such contractor.
i
.
By
V
West:
US
si;
i
1..
END
.
C