:
*
To
to
1
.
I OFI
ORNLP
237 |

.
:
i
..
.
-
-
--
$
.
I
.
.
.:
.
:.
ce
_
---
=
-
--
..
--
..
.
I
.
.
.
!
.
CLEFEEEE
__
125 LA LA
MICROCOPY RESOLUTION TEST CHART
NATIONAL BUREAU OF STANDARDS -1963
Paper to be presented at the Third International Materials Symposium,
University of California, Berkeley, Calif. on June 13-17, 1966.
ORAL-B-237
GRAIN BOUNDARY REACTIONS DURING DEFORMATION* Conf.660619-3
ORNL - AEC - OFFICIAL
SEP 2 2 1966
ORNI - AEC - OFFICIAL
L. E. Poteat and c. S. Yust**

Metals and Ceramics Division
RELEASED FOR ANNOUNCEMENT
Oak Ridge National Laboratory
IN NUCLEAR SCIENCE ABSTRACTS
Oak Ridge, Tennessee
The intent of this paper is to report microstructural observations
H.C. $
-
-
-
-
made on specimens of thorium dioxide and uranium dioxide which have
.........
CHILI PRICES
experienced creep deformation, and to relate the observations to the
.04 ; MN
creep behavior of these materials. Fine grain polycrystalline specimens
5
in the form of right-circular cylinders were prepared for this study
from high-purity powders which were dry pressed and sintered. The test
specimens were 97.5% of theoretical density and had an average grain
diameter of 10 H, with fine residual porosity distributed uniformly
throughout the bodies on grain boundaries.
It is necessary, for the sake of clarity, to begin by summarizing
the creep data for these specimens. The creep deformation rate of
*
thoria and urania at several temperatures as a function of stress is
*.
**..
-
presented in Fig. 1; the significant point revealed is that the
-
-
...***
deformation mode changes as the creep stress increases. At 1770°C, the
thoria deformation behavior exhibits three distinct ranges, a completely
brittle range at high strain rates, a viscous creep range at low strain
*Research sponsored by the U. S. Atomic Energy Commission under
contract with the Union Carbide Corporation.
ORNL - AEC - OFFIN
**L. E. Poteat is currently associate professor, Department of
Ceramic and Metallurgical Engineering, C?'uson University. C. S. Yust
is metallurgist, Metals and Ceramics Division, Oak Ridge National Laboratory,
Oak Ridge, Tennessee.
LEGAL NOTICE
This report was prepared as an account of Goveroment sponsored work. Neither the United
States, aor the Commission, nor any person acting on behalf of the Commission:
A. Makes any warranty or representation, 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 dlaclosed in this report may not Infringe
privately owned rigbts; or
B. Assumes any liabilites with respect to the use of, or for dainages resulting from the
use of any information, apparatus, method, or process disclosed in this report.
As used in the above, "person acung on behalf of the Commission" includes any em-
ployee or contractor of the Commission, 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 accaO to, any information pursuant to his employment or contract
with the Commission, or his employment with such contractor.
ORNL - AES. - OFFICIAL
I
rates, and an intermediate range having a slope of approximately 5.
ORNL - AEC - OFFICIAL
t
',
-
.
Specimens deformed in the viscous and intermediate ranges are macro-
...-
-
.
.
-
scopically ductile, although microscopically it is observed that inter-
granular voids form concurrently with the deformation. The slope of
approximately 1 for the viscous creep range implies that the transport
of material is by a purely diffusional process, the slope of 5 for the
intermediate range is indicative of a dislocation climb mechanism of
material transport, while in the brittle fracture range only elastic
deformation occurs prior to fracture of the specimen. 'The data for
thoria at temperatures below 1770°C indicate the beginning of the inter-
mediate range at higher stress levels, while the uranium dioxide data
presented show the existence of both the viscous and the intermediate
ranges, Discussion of the vo, large grain polycrystal and vo, bicrystal
data will be presented after some additional remarks. .
The microstructures produced by the different deformation modes are
surnarized in Fig. 2. The compression axis is vertical with respect to
the figure. At the highest strain rate, 3,0 x 10hr, the microstruc-
Yo
p
'
*..
-
.-
ture is typical of brittle fracture, characterized by the absence of any
plastic deformation prior to failure, rupture along the plane of maximum
shear stress, and creation of sharp intergrunular cracks. As the strain
'
17.
-*-
i*
rate is lowered, a degree of plastic response is evident prior to failure,
*
*
*
and the intergranular voids formed take a more rounded appearance.
!
*723
menit
Finally, structures are obtained in the viscous creep range, typified
by that produced at a strain rate of 2.0 x 10-2 hr-7, in which the inter-
granular voids are seen to result from the formation and growth of chains
.1
--
interesindenszeniaus teisinda
en
ORNL - AEC - OFFICIAL
d
me
3
of pores on the grain boundaries.
In addition, the creep deformation
ORNL - AEC - OFFICIAL
ORNI - SEC. Assicure
is accompanied by grain boundary sliding, demonstrated in these tests
in thoria by the observation of the displacement of a reference mark
across grain boundaries. Similar effects are noted in Vos although
the tendency to form intergranular voids is greatly diminished.
Armstrong and Irvine, reporting on the creep in bending of stoichio-
metric Uog, also state that grain boundary sliding occurs during creep.
The relationship between boundary void formation and translation of
grains along boundaries is therefore of great interest in understand-
ing the viscous creep mecanism in these materials.
Grain boundary sliding experiments have been performed on speci-
mens having a sufficiently large grain size to permit detailed study
of tlie boundary processes. Large grained polycrystalline rods of UO,
prepared at Oak Ridge National Laboratory by a floating-zone technique,
were the source of both large grain polycrystal specimens and bicrystal
specimens. The large grain polycrystals typically contained ó to 10
grains. The results of creep tests of several of the large grain
specimens are also shown on Fig. 1. At high stresses, where dislocation
notion within the grain lattice is indicated as the dominant deformation
process, the strain rate response is similar to that for fine grain
polycrystalline material. At relatively low stress, where viscous
creep' with associated boundary sliding is indicated, very low deforma-
tion rates are experienced in t. ' large grain polycrystal as compared
to fine grain specimens. Presumably, the small amount of grain boundary
area and the inability of the individual boundaries to slide independently
due to mutual restraint results in only deformation due to climo of dislo-
risvinju.Jj;- iivoo
cations. In fine grain polycrystalline naterial, greater deformation by
ORNI - AEC - OFFICIAL
4
.
ORNI - AEC - OFFICIAL
grain boundary sliding may be possible because of the much greater amount :
OD!!! ...rr.nr:......
of boundary are and consequent smaller sliding requirement of individual
grain facets to produce a given deformation. Deformation of a simple
bicrystal, however, gives the result indicated on Fig. 1, which approxi-
mates that of the fine grain polycrystalline bodies. Apparently, removing
the restraints or adjoining grains permits the simple boundary to slide
exter.sively by the viscous mechanism active in the polycrystalline speci-
mens, and detailed study of kicrystal sliding should yield information
concerning the sliding process in the fine grain polycrystal.
The boundary of a bicrystal vo, specimen has been examined by electron
transmission microscopy prior to sliding and is shown in Fig. 3. The
boundary is seen to be straight and free of gross impurities. The appli-
cation of compressive stress to a suitably oriented bicrystal results in
the generation of shear stress along the boundary, and the creation of
voids as a result of displacement along the boundary, Fig. 4. After an
amount of sliding small enough to avoid the formation of grain boundary
voids, the boundary configuration as seen by electron microscopy is shown
inen
salvestam
.
.w
in Fig. 5. The stress applied to produce the grain displacement in this
specimen vas 4,000 psi, at which stress only diffusional processes should
za to?-i farina Tatr
contribute to the sliding. Here large ledges are seen with heights of
approximately 1700 and 3500 Å, respectively; and the distance between
these jogs is about 5000 Å. At lower magnification, a dislocation
pattern is seen in the neighborhood of the jogged boundary which is
suggestive of the formation of a subgrain structure, Fig. 6.
Current concepts of the grain boundary suggest that the boundary can
be visualized as containing many atomic-sized ledges due to the fact that
ORNI - AEC - OFFICIAL
ORNL - AEC - OFFICIAL
-
-
--
-
-
-
-.
X
the grain lattices which join to form the boundary are not atomically
ORNI - AEC - OFFICIAL
sinooth.
In addition, we may
envision the atomic iit of the lattices
across a juncture as varying from place to place, being relatively good
at some points, poor at others, concepts which are not radical different
from the alternating islands or good fit and poor fit visualized by Mott,'
or the good-order and less-jell ordered regions pictured by Kft." The
application of shear stress to such a configuration creates a shear stress
gradient along the boundary and stresses the individual ledges.
In many instances, a tensile stress will be generated at the face uſ
a ledge which can be relieved by the creation of a vacancy at that point,
or more generally, a row of vacancies along the ledge face. Diffusion 0.1
the vacancies into the adjoining grain lattices may occur, the riet effect
of this vacancy diffusion being the motion of the ledge along the boundary.
Ultimately, coalescence of ledges is possible because of the variation in
to na
se
ledge velocity with ledge height or the interference of other imperfec-
meni
time on
tions; large ledges, such as have been observed in this work, may thus be
a
formed along the boundary. The dimensions and spacing of the ledges
hechos
-
observed here are consistent with this picture of large jog for ation.
.
For example, an apparently straight boundary through the region depict d
Monte .
a
Winte insieme
izmitin
in Fig. 5 would have had to consist of an atom-sized ledge every three o:
four atom distances along the boundary to form the ledges observed,
assuming most of the coalescence has occurred. This 1 ; not an unreasın-
able initial distribution. It has been suggested that jogs may "orm by
intersection of slip with the boundary (G: "kins), or tey might be growi.
in (Chen and Machlin). Here we observe the formation of large jugs on
a boundary between crystals in which significant slip has not occurred,
si
O?N! - AEC - OFFICIAL
but on which they may have existed as small "as grown" jogs.
.
.
.
H
.
-
En
The generation of tensile stress at the face of a jog as grains
UO.. Inxn
ORNL - AEC - OFFICIAL
slide along a common boundary is one of the mechanisms by which the
formation of boundary cavities has been explained. McLean' has shown
that jobs as small as four atom distances high may form stable pore
nuclei. The lecigas observed in to, which are nany times greater in
height may therefore easily be the source of boundary pores. Adjoining
rains must slide on a common boundary for pores to be created, and we
have previously described the translation of grains along a boundary in
the following manner.° Referring again to the concept of the boundary
as alternating regions of relatively good and relatively poor fit, it is
not unreasonable to consider the result of shear stress along the boundary
to be the creation of a stress gradient, highest at good-fit regions where
the load is sustained. The stress at the good-lit regions is relieved by
a viiffusion of vacancies along the boundary which reduce the size of these
ri
areas. Ultimately, a stress directed local realignment of lattice p.lanes
occurs to re-establish a new good-fit area; and the summation of these
mu - Recent home, tou
local re, ignments over an entire boundary results in motion of one grain
lattice with rispect to the other.
Vacancies which may be emitted by
.
e
r
Siantista **; witamine tone in itine
the moving jogs would serve to facilitate the motion of the good-fit
areas by diffusing along the boundary.
Controversy exists on whether sliding is an elementary process or
sve manifestation of another well-established mechanism, such as slip.
The electron photomicrographs presented here reveal that some dislocation
motion 2.2 the volume adjacent to the grain boundary is associated with
vir..rie
the sliding process. It is not possible to assess the dislocation role
in the sliding mechanism in vo, at this time, but these results indicate
ORNL - AEC - OFFICIAL
...
- -
-
ORNI - AEC - OFFICIAL
that at least some cooperative deformation of the adjoining lattice
ORNL - AEC - OFFICIAL
accompanies grain boundary sliding.
ACKNOWLEDGMENTS
The authors gratefully acknowledge the contributions of L. L. Hall
for assistance in the performance of the experimental work, H. R. Gaddis
for the metallographic support, and C.K.H. DuBose for the electron
microscopy.
ت 1 - :( - الذرة
ORNL - AEC - OFFICIAL
REFERENCES
ORNI - AEC - OFFICIAL
ORNI - AEC - OFFICIAL
l. 1. M. Ar.nstrong and W. R. Irvine, J. Plucl. Mater. 7(2) 133–11 (1962).
2. 7. Garofalo, Fundamentals of Creep and Creep Rupture in Metals,
ac lillan Company, New York, 1955, p. 197.
3. E. F. Mott, Proc. Phys. Soc. (Ionkon) 60(4) 391-94 (1948).
4. T. S. Ke', J. Appl. Phys. 20(3) 274-80 (1949).
5. R. C. Gifkins, Acta vet. 4(1) 98 (1956).
6. C. W. Chen and E. S. Machlin, Acta Met. 4(11) 655 (1956).
7. D. McLean, J. Australian Inst. of Metals 8(1) 45-51 (1963).
8. L. E. Poteat and C. S. Yust, J. Amer. Ceram. Soc. (in press).
.
D. Mclear
J. Australian Inst. of Metals
. ...
in
t is-i
ait. Wh
Vitignorieren. Lt. VMHR
URINE -M6L-urricTAL
ORNL - AEC - OFFICIAL
LIST OF FIGURES
OPNL-AEC-OFFICIAL
Fig. 1. Summary of Deformation Data for Tho, and vo, Polycrystalline
and Bicrystal Specimens.
Temperature. Trorn left to right the macroscopic deformation
behavinr progresses fro:n brittle to very ductile.
Fig. 3. Electron Photomicrograph of a Typical vo, Bicrystal Boundary
Before Grain Boundary Sliding Occurs. 45,000 X
Fig. 4. Voids Produced at a Bicry: val Boundary in vo, by Grain Boundary
Sliding. Unetched, 250 X
Fig. 5. Jogs in a vo, Bicrystal Foundary After a Small Amount of Grain
Boundary Sliding. 69,000 X
rib. ó. Array or Dislocations in Vicinity of Jogged Bicrystal Boundary
in vog. 13,500 x
1.*
.
ORNI - AFC - OFFICIAL
ORNL - AEC - OFFICIAL
2
sie
Ak
.--
..
.
.
.
.*
-.Z
::-
-.
.
..
-
..
-
:
.
e
-
leiti is:
siz
Sit
**
41
je
s
*3
*
ä
brittle to very ductile.
macroscopic defor.nation behavior progresses from
at Constant Temperature. From left to right the
Fib. 2. Typical Microstruc-tures oỉ Thoria Bodies Deformed
11
DOMI - Arrrrimin
ORNI - AEC - OFFICIAL
7cmuel
ummach op dkgarmatron ota par Th ozanduo
line und BisCrystal Specimena

..
vila
Jedan
ORNL-DWG 66-5795
2000
3000
STRESS (psi)
4000 6000 7500
91,000 15,000
...
NIIN
Thoz270-
4E.
SLOPE = 5.0
2-1790°C
4.1
4.2
_
_
_
X
SLOPE = 4.0.
2-1666°C
DEFORMATION RATE (hr).
No no no no ô no ģ no ô

-1600°C
-1535°C
-1430°C
ThO2-
10.-1535°C.
VO, BICRYSTAL 1430°C
UO2-1430°C
TO
TFUO 2, LARGE GRAIN
POLYCRYSTAL, 1535°C
3.1
3.2
3.3
3.4
3.5
3.9
4.0
4.1
4.2
4.3
3.6 3.7 3.8
LOG STRESS
12
PHOTO 83760
ORNL - AEC - OFFICIAL
...!!! , copy',

1
:
..
miri,
Biļ
-
-
-
-
All
-0.007 INCHES
500x Tom
0.007 IMCHES
W 500X
l
é = 3 x 10' hr--
é = 3 x 100 hr-

yu
NA
1
..
21
.
J.;
9
WY.
C
....me amorowerowe
-
V
*
-
CA
.
2
i
*
-
W.
A
LC
2
MI:
.
4.
Y
.....
.
.
.!
.
...'
-
-
of
11
wo
*
'
Vui
;, ';:
-
ORNI - AEC - OFFICIAL
li
2.007 INCHES
500X
Un
9.COM
SOOX
é = 3 x 10- hr-
é = 2 x 10-2 hr-1
T = 1770°C
#
13
ORNI - AEC - OFFICIAL
ORNI - AEC - OFFICIAL

' 41.
,
X
1:
X
.
"
S
1946
.
.
..o me imunitet
..
.
.
RE
.
.
4
.
1
LO
AH)
...
ta
.
.
1
.
TAY
43
. 70
-
S
.
2
,
7
.
AE
.
.
Sh.
*",
o
T
1
ma
NI
.
10
A
NY
47
Fig. 5. D.lcctzor. Protomicrograph of a Typical.
to, Bicrystal sounüary Before Grain Boundary
Sliding Occurs. 45,000 X
ou - zleconen
ORNI - AEC - OFFICIAL
ORNI - AEC - OFFICIAL
14
ONI - AEC-osirin!

Mily
"
.
doi
V
9
.
..
*
--
-
--
•
.
.
-
.
..
---
-0.016, 110HF.S
TRE TOT 1250x
TA TT T
Fig. 4. Voids Produced at a Bicrystal Eoundary
in vo, by Grain Boundary Sliding. Unetched,
250 X
ORNL - AEC - OFFICIAL
ORNI - AEC - OFFICIAL
24 15
ORNI – AEC - OFFICIAL
ORNI - AEC - OFFICIAL

AILY
.
.
NA
Itt
A
.
-
..
.
-
Fig. 5. Jogs in a vo, Bicrystal Boundary After
a Small Amount of Grain Boundary Sliding.
69,000 X
O?:4! - ".FC-;?FFIC!A!
ORNI - AEC - OFFICIAL
R
1.6
ORNI - AEC - OFFICIAL
ORNI - AEC - OFFICIAL

Wow
hy
.
IP 11
.
.
to
...
"T
?
.
..
V
he
2
x
?
1
.
24
72
2011
.
4
WE
XV.
hogy
W
!
.
V
..
.
Www
-
-
-
Fig. 6. Array of Dislocations in Vicinity of Joggeå
Bicrystal Boundary in uog. 13,500 X
-
-
-
-
.
.
*
ORNI - AEC - OFFICIAL
ORNL - AEC - OFFICIAL

ÉR
IT.
*
.
JK
es
2
AY
.
.
-
.
:-
-
LE
SK
.
2
-
-
-
-
-
....nihe
men
m
-:
.
O
END
DATE FILMED
29/ € / 1