I OFT
ORNL P
2922
to
N
.
v
.
EEEEEEEE
MICROCOPY RESOLUTION TEST CHART
ORINA-p-2022
Cort660607-6
MASTER
WAR 2 1 087
Nüte:
66-58
To be published in the Proceedings of the Third International wym-
posium on the Effecte of Radiation on Structural Metals, Atlantic
City, New Jersey, June 29 to July 1, 1966. Contents of this paper
should not be quoted or referred to without permission of the
authors.
0:57. Tas
H.C. $0.00 N.o
DOSE RATE, ANNEALING, AND STRESS-RELAXATION STUDIES
OF RADIATION HARDENING IN IRON
.:
N. E. Hinkle, s. M. Ohr, and M. S. Wechsler
:
Paper No. _50
::?
LEGAL NOTICE
This report was prepared as an account of Government sponsored work. Neither the United
· States, oor 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 proceso disclosed in this report may not Infringe
privatnly owned rights; or
B. Assumns 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.
As used in the above, "person acting on behalf of the Commission” includes way om-
ployee or contractor of the Commission, or employee of such contractor, to the extent that
such employne or contractor of the Commiosion, or employee of such contractor prepares,
disseminates, or provides access to, any information pursuant to his employment or :'ontrect
with the Commission, or his employment with such contractor.

SOLID STATE DIVISION
OAK RIDGE NATIONAL LABORATORY
operated by
UNION CARBIDE CORPORATION
for the
V. S. ATOMIC ENERGY COMMISSION
Oak Ridge, Tennessee
.
:
February 1967
..
.
.

.
RE
NO
.
ht

This paper wus mubmitted for publication
in the open Utoratur, at least months
prior to the isnuance date of this Micro-
card. Since the U.S.A.E.C. has no evi-
dence that it has been published, the pa-
per is being distributed in Microcard
form as & proprint.
DISTRIBUTION OF THIS DOCUMENT IS UNLIMITED
Dose Rate, Annealing, and Stress-Relaxation Studies
of Radiation Hardening in Iron
· N. E. Hinkle, S. M. Obr, and M. S. Wechsler
Solid State Division, Oak Ridge National Laboratory
Oak Ridge, Tennessee
.: ABSTRACT
A study of the specific effect of duse rate on the increase in
yield stress of a vacuum-melted iron was carried out by irradiating
tensile samples at dose retes from 2 x 1011 to 3 x 1023 neutrons/cm2
sec (E > 1 Mev) at 95°C in the poolside of the Oak Ridge Research Reactor.
For a dose of 4.6 x 1018 neutrons/cm2 (E > 1 Mev), a three-fold increase
in yield stress occurred, but there was no variation witb dose rate in
the range investigated. The possible significance of the absence of a
SE
dose rate effect is discussed in general terms.
The annealing of the radiation hardening was found to take place
at temperatures of 300-400°C with an activation energy of about 3 ev.
This range of temperatures coincides with that observed for the disso-
ciation of nitrogen from radiation-produced traps after electron irradia-
tion. Several models for the annealing process are considered.
Stress relaxation measurements on unirradiated and irradiated iron
and iron-silicon are described. Evidence for the existence of an inter-
nal or threshold stress, which must be exceeded for macroscopic
*Research sponsored by the U. S. Atomic Energy Commission under contract
with Union Carbide Corporation. :
dislocation motion, 18 cited. It 18 pointed out that the major cause of
radiation hardening 18 the increase in this stress. When account 18
taken of the increase in internal stress upon irradiation, the stress
relaxation experiments indicate no change in the activation volume for
plastic de formation.
Dnge Rate, Annealing, und Stress-Relaxation Studies
of Radiation Hardening In Iron
N. E. Hinkle, 3. M. Ohr, and M. S. Wechsler
The importance of determining the possiðle specific inflųence of the
dose rate in the hardening and embrittlement of iron and steel has long
teen recognized. From practical considerations, the question arises be-
cause of the different range of dose rates at which materials evaluation
experiments are carried out as compared to the dose rates at reactor pres..
..
."
..
.
..
.
kn3
.
www
....
sure vessels. In a typical. irraálation experiment, it is not unusual for
the dose rate (or flux) to be as large as 1013 neutrcns/cm2 sec (E > 1 Mev).
However, the dose rates calculated or measured at the wall of boiling-
water or pressurized-water reactor vessels (1) tend to be in the range of
109 - 2012 neutrons/cm2 sec (E >.1 Mev). In a series of experiments on
several mild steels, Barton, Harries, and Mogfora(2) found that there was
no effect of dose rate on the increase in yield stress upon irradiation
over the range 3 x 1021 to 3 x 1013 fiosion neutrons/cm2 sec for a dose.
of about 2.2 x 1027 fission neutrons/cm? and irradiation temperatures
from 100°C to 350 °C. Since the longest irradiation time in these experi-
ments was only about 200 hours, it was felt that longer term Irradiations
at higher doses may be needed to detect a dose rate effect. Therefore, a
series of irradiations was conducted at 95°C in the Oak Ridge Research
Reactor for times extending up to 5600 hours. Also, a vacuum-nelted
iron, Ferrovac E, was used instead of steel because as a single-phase
material its structure can be more easily controlled. Furthermore, it
was felt that the fundamental implications of the presence or absence of
.
a dose rate effect can be better assessed for the high-purity fron. As
18 shown below, no dose rate effect on the radiation hardening was
detected over a span of dose rater from 2 x 2071 to 3 x 1015 neutrons/
como sec (E > 1 Mev) at a dose of about 5 x 1928 neutrons/cm2 (E > 1 Mev).
Experiments were also carried out on the annealing behavior of the
irradiated iron. The yield stress 18 found to recover at temperatures
between 300°C and 100°C with an activation energy of about 3 ev.
Lastly, we wish to discuss strees-relaxation experiments in irradi-
ated and unirradiated iron from which certain information regarding the
dislocation dynamics may be deduced.
1. Dcee Rate and Annealing Experiments
The starting stock of vacuum-melted iron in the form of 1-1/4 inch
bars was swaged to 9/32-inch diameter and tensile specimens were machines
with a gage diameter of 0.146 inches and a gage length of one inch. The
specimens were given vacuum anneals at 885°C which resulted in a grain
size of 130m. Following the annealing, the samples were chemically ana-
lyzed and the results are given in Table 1. No major change in composi-,
tion upon annealing was found, with the possible exception of nitrogen
which increased from 1 ppm for the mill analysis to about 10 ppm after
annealing.
For the irradiations in the ORR, the samples were mounted in vertical
rows in helium-filled and welded aluminum boxes, such that they took
positions in six vertical planes at various distances from the reactor
face. The samples were beated by wire-wound heaters and placed inside
the cavity of ball-round cooling jackets (for examp].e see Fig. 1). In
this way the temperature of each sample was individually controlled at
..
95 + 5°C regardless of the distance from the reactor. The neutron fluxes
were measured at each location using nickel monitors, which were shielded
with cadmium in order to elininate the need for corrections due to 58co
burn-up upon thermal neutron absorption. A value of 420 mb was used for
the 2.9-Mev effective fission-spectrum threshold cross section, (3) which'
corresponds to a mean fission-spectrum cross section of about 100 mb."
The neutron spectrum was probably close to a fission spectrum for all
irradiations, since a void space always separated the experiment capsule
from the reactor face. The neutron doses and dose rates for neutrons
Above one Mev were calculated from the measured values using the nickel
dosimeters ase ming a flosion spectrum.
For a portion of the samples the irradiation times were adjusted at
-
-
-
-
--
the various dose rates so as to provide a constant dose of 4.6 x 10+
neutroas/cm2 (E > 1 Mev). This was achieved to within only about + 25%
and so the increases in yleid stress upon irradiation were adjusted to
a standard dose of 4.6 x 1018 neutrons/cm2 (E. > 1 Mev) on the basis of
limited measurements as a function of dose. These adjustments were
mostly within + 2000 psi. All measurements were made at room temperature
at a strain rate of 0.02 min-1.
The results of the dose rate experiments are shown in Fig. 2 for
Irradiations covering the range of dose rates from 2 x 1011 to 3 x 1013
neutrons/cm2 sec (E > 1 Mev) for an irradiation temperature of 95°C. For
the high-purity Iron, as for the mild steels, (2) no dose rate effect is
observed. Points are shown in Fig. 2 for unshielded and cadmium-shielded
samples. Since the increase in yield stress was not affected by the
cadmium shielding, we conclude that effects due to thermal neutrons, such
...
-
--
-
-
.
.
:
....ia
.
...
.
.
...
as transmitations, helium production due to the 10$ (n, a) L' reaction,
or (n, a) recoils, are unimportant as far as the radiation hardening is
concerned.
Post-irradiation annealing experiments were performed on samples
Irradiated at 94°C to 1.7 x 1018 neutrons/cm2 (E > 1 Mev) at a dose rute.
of 8 x 1022 neutrons/cm2 sec (E > 1 Mev). The results shown in Fig. 3 .
indicate that the radiation-induced increase in yield stress anneals
out at temperatures between 300 °C and 400°C. Each point in Fig. 3 repre-
sents a separata sample annealed at the indicated temperature for 2 hours.
Unirradiated samples were given the same thermal treatment and for anneal-
ing temperatures up to 590°C no effect on the yield stress was noted. The
curve shown in Fig. 3 11°8 in approximately the same temperature range
as the recovery curves for microhardness given by Bartlett and Eyre(5)
and Ibragimov and Lyashenko) for polycrystalline iron. However, the
annealing range indicated by Kunz and Holden(7) for recovery of the yield
stress of irradiated single crystals of iron appears to be shifted toward
lower temperatures. Figure 3 is also consistent with the results of
Chow et al. (0) who observed considerable recovery of tensile properties
upon annealing at 350 °C but no recovery at 250°C.
Figure 3 shows that the 2-hour annea.l at 360ºC achieved approximately
50% recovery of the yield stress. Anneals at 360°C were also performed:
for 15 min and 1000 min. As has been observed previously for iron (7) and
steel, (9) the resulting yield stress was linear on a log time scale and
the precise time for 50% recovery could thus be determined. Furthermore,
since the activation energy for annealing has been observed to be con-
stant, (7,9) it was possible to determine from the yield stresses after
2-hour anneals at 300 °C and 400°C the corresponding times for 50% recovery...
at these temperatures.
The times for 509 recovery are plotted in Fig. 4
versus reciprocal absolute annealing temperature and an activation energy
-
.
:
of 3.0 ev 18 obtained. This is in essential agreement with the 3.1 ev
determined by Kunz and Holden for iron single crystals, despite the
difference in the range of annealing temperatures mentioned earlier.
Similar activation energy plots for silicon-killed carbon steel(9) and
ASTM A-212B pressure-vessel steel(10) are also shown. For the steels,
the activatior energy appears to increase with decreasing irradiation
temperature. Also, there is a decided increase in the rate of the anneal.
ing with decreasing irradiation temperature. This may account partly for
the more rapid annealing observed by Kunz and Holden(7) for iron as com-
pared to the present results, since their irradiation was performed at
50 °C.
The activation energy of about 3 ev for 1ron agrees roughly with the
activation energy for self diffusion of about 2.5 ev measured by Borg
and Birchenall(11) and Lai and Borg(12) for alpha iron above the Curie
temperature. If this correspondence in activation energies were signif-
icant, the rate of annealing would be governed by the total rate of
jumping of lattice vacancies in thermal equilibrium. It has been pointed
out\9,19) that this supposition leads to the difficulty that the mumber
of vacancy jumps (or atom jumps) in the time necessary for the recovery
of yield stress would be incredibly small. From Fig. 3 the time for 50
percent recovery at 350 °C is about 100 min. The mumber of atom pumpo per
atom is given by(14) .... .
. .
.
.
J = It - Azv. exp (-2/xT) : t
.
where A = product of entropy factors - 10, 2 = coordination rumber = 8,
v. - 2013 sec -2, and k = Boltzmann's constant = 86.2 x 10-6 ev/deg. Then
for Q = 3 ev, T = 350°C, and t = 100 min, we find that only about 2 x 10-6
Jumps per atom are necessary for 50 percent recovery which is much too
small to be reasonable.
In a recent report F. A. Nichols(15) has suggested that radiation-
hardening is due to vacancy clusters or pores produced by the Irradiation
and the annealing is the result of the migration and coalescence of the
pores. In this case, the kinetics of the radiation annealing is governed.
not by volume difíusion, but by surface diffision at the inside wall of
the pore. The effective diffusivity for the pore is given by :15:10)
(1)
where D. is the surface self diffusion coefficient, r is the radius of
the pore, and an 18 the lattice parameter.
From scratch-smoothing
experimenta!!on alpha iron, the activation energy for surface diffusion
was found to be 2.5 ev, close to the 3 ev determined here for the anneal-
ing of the radiation hardening. However, for consistency with the 50
percent recovery times, the pore size, r, must be exceedingly small. **
Nichols 252ssociates the time, t, for 50 percent recovery with the time
for the pore to diffuse the average distance, R, between the pores. Then
by Eq. (1)
For our irradiation the dose, , was 1.7 x 1018 neutrons/cm2 (E > 1 Mev)
.
2
and for a primary collision cross section, one of 3 x 10-24 cm”, the
concentration of primary damage clusters would be no greater than one =
5.1 x 10-6. If the pore is assnciated with the damage cluster, the
average inter-pore spacing, R, is certainly greater than (D) 19 or
about 60 atomic spacings. Now, according to Blakely and Mykuraſ 17) Dg
at 350°C 18 about 6 x 20-26 cm2/sec and our time for 50 percent recovery
is 100 min from Fig. 3. When these values are substituted in Eq. (2),
the pore size, r, turns out to be sualler than 20, which casts doubt on
the validity of this approach' under these conditions. Furthermore, if
r were larger, say by a factor of 10, the predicted time to traverse
the inter-pore distance would, by Eq. (2), increase by a factor of 304,
and we would again be faced with the problem of understanding why the
observed time for 50% annealing is so short.
Recent transmission electron microscopy studies to, by on irradiated
iron have emphasized that the coarsening of spots ana loops seen in the
microscope takes place at the same temperatures (300 - 500°c) as for the
annealing of radiation hardening. The presumption is that the defects
which cause the hardening are submicroscopic (smaller than 204°). Bryner(18)
suggests that at high doses or upon raising the temperature following
irradiation the submicroscopic defect clusters migrate and coalesce into
larger clusters or loops which are less effective barriers to dislocations.
In some cases, the maximum loop area was found to depend on the concentra-
tion of interstitial impurities. On the other hand, Eyre and Bartlett(19)
take the view that the radiation hardening may be due to submicroscopic
clusters of iron interstitials, which are annihilated when vacancies break
free from vacancy-impurity interstitial complexes and migrate to the iron
:
me........
.
.
..
..
o no rendendo ... -
8
interstitial clusters. Another indication of the importance of inter-
stitial impurities in the radiation hardening in iron has been cited by
Barton, Harries, and Mogford, (2) who suggested that the susceptibility:
to radiation bardening in mild steels may depend upon the amount of
nitrogen that is originally free in solid solution and 18 therefore
available to associate with defects introduced during irradiation.
A clue to the role of nitrogen in the radiation hardening of iron
has been provided by the internal friction measurements of J. T. Stanley
et al. (20,211) who have observed evidence for the trapping of nitrogen
atoms in alpha iron by defects (presumably vacancies) created by elec-
tron irradiation at 170°K. Upon post-irradiation annealing the nitrogen
was observed to return to the solid solution in the same temperature range
(Fig. 5). as for the annealing of the radiation-induced hardness (Fig. 3).
we may estimate the energy, 1, for the dissociation of the nitrogen-
vacancy complex as follows. Let n be the number of trapped nitrogen
atoms. The rate of afssociation is given by
The -- • , exp (- 6)
whereby
where n, is the number of trapped nitrogen atoms at the outset, vo is
the vibration frequency, and the other symbols have their former mean-
ings. Figure 5 shows that hair of the nitrogen atoms are dissociated
after 15 min at 355°C. muling(22) v, - 1013 to 1015 sec-?, we find from
Eq. (3) that H = 2.0 to 2.3 ev or about 2 ev. This energy is still
".........
......
.
...
....
...
.
.is -
,.,..,?..
"'.'....
.
significantly less than the observed 3.0 ev for the annealing of the
yield stress, but this may be a consequence of the higher nitrogen
concentration (about 5.2 x 10-4) for the internal friction measurements(21)
a3 compared to about 4 x 10-5 for the tensile samples.
. Even if nitrogen plays as central a role as is suggested here in
the radiation hardening of iron, there remains the question of whether
the nitrogen-vacancy complex itself is the hardening agent or whether,
as suggested by Eyre and Bartlett,(19) the interstitial clusters are
responsible. In the former case, the recovery is attendant upon the
simple dissociation of the complex, whereas for the latter mechanism
the vacancies freed from the complexes must migrate to the interstitial
clusters and annihilate them. It is difficult at this time to choose
between the two possibilities but the time for the dissociation of the
nitrogen-vacancy complex (15 min at 355°C, Fig. 5) 18 not greatly
different than that for the yield stress annealing (100 min at 350°C,
Fig. 3), which would lend some confidence in the simpler mechanism.
However, a number of factors indicate that further work is needed
to clarify the effect of interstitial solutes trapped by radiation pro-
duced defects. Some indication has recently been obtained from magnetic
measureigents (25) that the temperature for the release of nitrogen from
Jefect traps is lower after neutron irradiation than after electron
irradiation. There is also the question of the influence of other
interstitial solutes such as carbon or oxygen, which were present in the
material investigated at higher concentrations than nitrogen (Table 1).
The work of Damask et al. (24-26) on neutron irradiated Fe-0.01 wt.% C
indicated that carbon-defect complexes dissociate below 230°C, well
10
10
.
.
.
.
below the annealing range shown in Fig. 3. On the other hand, magnetic
measurements on Fe-C alloys with lower carbon contents appear to place
the release of carbon from defect traps at 300°C after neutron irradia-
tion and 580°C after electron irradiation, and this bas led to the
postulation of different types of radiation - produced trapping sites
for the two types of radiations 23). Not much is known about the
interaction between oxygen interstitials and radiation - produced defects,
and in the absence of further information, it is probably safe to assume
that oxygen is tightly bound either to grain boundaries or to iron atoms
or substitutional impurities as oxides within the matrix.
The absence of any dependence on dose rate over a span of three
orders of magnitude (Fig. 2) must be provided for in the picture. of the
.
mechanism of radiation hardening. The resistivity measurements in iron
after low temperature bombardment (cf. Figs. 8 anu 10 in Ref. 27) indi-
cate considerable defect migration below our irradiation temperature of
95°C. At this temperature, then, the defect structure results from a
dynamic process in which vacancies and interstitials are constantly
being created and annihilated. If the annihilation process were complete-
ly random, as is assumed in theories of radiation-enhanced diffusion, 147
the dynamic equilibrium number of defects reached would depend upon the
::
dose rate. However, if the annihilation process depended upon the density
and arrangement of defects within a damage cluster or displacement spike,
the resulting defect structure would be largely independent of the dose
rate provided that the damage clusters were spaced far enough apart so
that there would be no interaction between them. At a dose of 5 x 1040
neutrons/cm? and for a primary collision cross section of 3 x 10-24 cm?, .
5
W
Teri
.. ..il
the damage clusters would be spaced an average of about 40 atomic spacings:
apezt, which is apparently sufficient to avoid interaction.
However, this
LA
«
r
suggests that a dose rate effect may take place at higher doses. Another
question suggested by these experiments is the possibility of a dose rate
effect upon irradiation with electrons or gamma rays, where the damage
consists of a more random distribution of point defects.
i
A
inanz
.
- ..
-
- -
-- - -
-
2. Stress Relaxation Experiments
Since the motion of dislocations in metals provides the basis for
plastic deformation, it is important to understand better how dislocation
dynamics is affected by radiation. Dislocation dynamics may be studied
by a stress relaxation technique, 0) in which a tensile sample is ex-
-
.
J
.
..
.
tended to a given prestrain at a constant strain rate, the tensile machine
18 stopped, and the relaxation of stress is observed as a function of
time until a steady value, debi.gnated as the "internal stress", 04, is
reached. Since the total strain rate 1s held to zero during the stress
relaxation,
te +
p
= 0
.
The elastic strain rate may be written
te -
and for the plastic strain rate, dislocation theory gives
o =
dopr
where b is the Burgers vector, p is the mobile dislocation density, v is
.
.
the average dislocation velocity, and a is a numberical constant of value
12 :
near 0.5 to convert from shear strain to tensile strain. When v 18 ex-
pressed in terms of stress, Eq. (4) results in an equation for the stress
as a function of time. From etch-pitting experiments, (29,30) an empirical
fit to the stress dependence of v 18 given by
. Mom
(7)
or by
V. V. exp
NO
(8)
These expressions require that the stress be reduced to zero in order that
v and, by Eq. (6), the plastic strain rate go to zero. However, the
stress does not relax to zero in a stress relaxation experiment and hence
it is necessary to insert a threshold or internal stress, 0%, below
which no macroscopic motion of dislocations takes place. Equations (7)
and (8) then become
and
The insertion of o, in Eq. (8) was also shown (5+) to be necessary in order
that the dislocation velocity measurements on Fe-S1650) extrapolate to
av value close to the shear wave velocity.
It is the purpose of this section to describe briefly further exper-
iments which support the concept of an internal stress'anni permit the
activation volume for plastic de formation to be evaluated for irradiated
... 13
and unirradiated samples. During the stress relaxation process, the
stress decreases rapidly at first and then approaches asymptotical.l.y
the steady value, 04 Thus, q, should correspond to the flow stress
in the limit as the applied plastic strain rate approaches zero. This
idea was checked by conducting tensile tests as a function of strain
rate on iron samples of several grain sizes. 54) Stress relaxation
tests were also carried out on the same samples following the tensile
tests.
The results in Fig. 6 show that the internal stress for a
given grain size does not depend on the prior strain rate, and the
lower yield stress does indeed approach the internal stress asymptot-
ically upon decreasing the strain rate.
In order to test Eq. (9)directly, we have conducted stress relaxation
experiments on Fe-si (3.2 wt. % Si) for which the dislocation velocity
measurements of Stein and Low(30) are available as a function of
stress. Since our work was on polycrystalline samples whereas Stein
and Low used single crystals, we made measurements as a function of
grain size in order to extrapolate to the value appropriate for the
single crystal. The results (Fig. 7) indicate a linear variation of
the internal stress, 04, and the yield stress, Oqy, versus d-1/2 where
d 18 the grain diameter. From the intercept at -1/2 = 0, the inter-
nal stress of the unstrained single crystal of Fe-Si 18 deduced to
be 22 kg/rom. In Fig. 8, the dislocation velocity data of Stein and ..
Lowl 30) are plotted versus 0-0, according to Eq. (9). Except for a
few points for very low velocities, the fit is quite good. Furthermore,
the value of m= 8 18 quite close to the value of m, = 7 from stress
14
relaxation measurements on Fe-51.351 In addition, it 18 possible to
analyze the results in terms of two components, my and me, and it was
suggested?) that the more mobile component, men = 7, corresponds to
the motion of edge dislocations. This is in accord with the result in.
:
Fig. 8, since Stein and Low's measurements were on edge dislocations. .
.
It 18 interesting that approximately the same values of my and my are
obtained for the Ferrovac-E iron. 15479) concerning the effect of
irradiation on the parameters 04 and m, it has been shown (39) that on.
18 increased upon irradiation, but m is unchanged. The fit. of the data
in Fig. 8 to Eq. (10) was also attempted, using the same value of 0.4
22.0 kg/m2, as deduced from Fig. 7. However, a good fit was not
obtained.
The activation volume v* is a measure of the average spacing and
size of the obstacles that are impeding dislocation motion. It has
been shown (36) that r* may be experimentally determined using the
relation
v*
= KT
I do IT
(11)
where k i8 Boltzmann's constant, T. is the absolute temperature, &, is
the plastic-strain rate, and o is the applied tensile stress. During
the stress relaxation test, én is given by - ao from Eqs. (4), (5),
and (6) and it decreases from the prior applied strain rate to a value
approaching zero. Thus, from the shape of the stress relaxation curve,
v* can be evaluated using Eq. (11). This was done for an unirradiated
Ferrovac-E iron sample and for one that was irradiated to a dose of 3.3
x 2018 neutrons/cm2 (E > 1 Mev). Figure 9 shows that when the activation ::
volume 18 plotted against 0-04, the magnitude and the stress dependence
of v* are unaffected by the irradiation. Using change-of-strain-rate
tensile tests on irradiated Ferrovac-E iron, Smidt(37) reported a decrease" .
in v* upon irradiation. However, in that case the comparison was not
made at constant effective stress, 0-0,. We have found154) that the
effective stress for yielding in the irradiated iron 18 higher (by about
4 kg/mm2) than that for the unirradiated material, probably because of
a decrease in the mobile dislocation density. Since v* decreases rapid :
ly with increasing effective stress, the measured v*. for irradiated
samples will appear to be lower than for unirradiated samples. However, .. .,
it is emphasized that when the comparison 18 made at constant effective
stress, no effect of the irradiation on v* is found. Thus the radiation
induced defects do not appear to be effective as obstacles in impeding
short-range dislocation motion. Nevertheless, their presence does cause
the internal stress to increase throughout the crystal and this long-
weats
www.
range effect appears to be the major source of 'radiation hardening.
Conclusions
No effect of dose rate over the span from 2 x 1071 to 3 x 1013
neutrons/cm sec (E > 1 Mev) on the increase in yield stress of Ferrovac-E
iron is observed. The samples were irradiated at 95°C to a dose of 4.6 x
1010 neutrons/cm2 sec (E > 1 Mev) and experienced a three-fold increase
in yield stress.
The result implies that no interaction takes place
between damage clusters produced by primary collisions.
:
16
The annealing of radiation hardening in iron occurs at 300-400°C
with an activation energy of about 3 ev. Since this energy is close to
the self-diffusion activation energy, the motion of single vacancies in
thermal equilibrium is implied. However, the annealing proceeds too
quickly for the annealing to be due to the motion of single vacancies.
Also, a model based on the motion of vacancy clusters or pores leads
to too small, a pore size. Finally, the annealing times and temperatures:
appear to correlate with the release of nitrogen from radiation-produced
traps, which suggests the importance of interstitial nitrogen in the
annealing of 'radiation hardening in iron.
Stress relaxation experiments indicate that a threshold or internal
stress must be exceeded before macroscopic dislocation motion will takė
place. The activation volume for plastic deformation is found to be
unaffected by neutron irradiation, when account is taken of the increase
in effective stress. The major cause of radiation hardening seems to
be the increase in the long range internal stress produced by the
radiation-induced defects.
-
·
Acknowledgments:
We wish to thank J. T. Stanley for helpful discussions and permission
to discuss results on iron-nitrogen alloys in advance of publication.
The
assistance of N. K. Smith in the experimental work is also gratefully
acknowledged.
..
17
TABLE I
CHEMICAL ANALYSIS OF FERROVAC-E IRON TENSILE SPECIMENS
Element
.. Original
Mill Analysis
Concentration after anneal at 885°C
for 2 hrs
Datch 1 Batch 2 Batch 3 Batch 4
(ppm)
Fe
.
base metal
---
:. 40
74
---
40
77
9
--- ---
40 . 30
66 72
12
..........
.
..
SI
..
....
*
-
10
20
400
200
10
400
700
10
400
100
10
100
.
10
18
.
References
**
1. J. J. Di Munno and A. B. Holt, "Radiation Embrittlement of Reactor,
· Vessels," Nuclear Safety, vol. 4, p. 34, (1962).
2. P. J. Barton, D. R. Harries, and I. L. Mogford, "Effects of Neutron
Dose Rate and Irradiation Temperature on Radiation Hardening in Mild
Steels," J. Iron and Steel Inst. , Vol. 203, p. 507, (1965). .
3. "Tentative Method for Measuring Neutron Flux by Radioactivation Tech-
niques," E-261, 1966 Book of ASTM Standards, Part 31, American
Society for Testing and Materials, Philadelphia.
4. T. 0. Passell, 'The Use of Ni58 and Fe 54 as Integrators of Fast
Neutron Flux," p. 50.2 in Neutron Dosimetry, Proceedings of the Sym-
posium on Neutron Detection, Dosimetry, and Standardization Vol. 1,
International Atomic Energy Agency, Vienna, 1963.
5. B. L. Eyre and A. F. Bartlett, "An Electron Microscope Study of
Neutron Irradiation Damage in Alpha Iron," Phil. Mag., Vol. 12,.
p. 261, (1955).
6. S. S. Ibragimov and V. S. Iyashenko, 'The Effect of Fast Neutrons on
the Properties of Metals," Physics of Metals and Metallography,
Vol. 10, 23. 27, (1960).
7. F. W. Kunz and A. N. Holden, "The Effect of Short-Time Moderate Flux
· Neutron Irradiations on the Mechanical Properties of Some Metals,"
Acta Met., Vol. 2, p. 816, (1954).
8. J. G. Y. Chow, S. B. McRickard, and D. H. Gurinsky, 'Effect of Fast-
Neutron Irradiation on the Mechanical Properties of Pure Iron,"
p. 277 in Radiation Damage in Solids, International Atomic Energy
Agency, Vienna, 1962. .
.
.
.
ieder
w
19: : .... ..
9. R. W. Nichols and D. R. Harries, "Brittle Fracture and Irradiation
Effects in Ferritic Pressure Vessel Steels," p. 162 in Symposium on
Radiation Effects on Metals and Neutron Dosimetry, ASTM-STP-341,
American Society for Testing and Materials, Philadelphia, 1963.
20. R. G. Berggren, W. J. Stelzman, and T. N. Jones, "Radiation Effects
in Pressure Vessel Steels," p. 3 in Radiation Metallurgy Section
Solid State Division Progress Report for Period Ending February 1966,
ORNL 3949, April 1966.
11. R. J. Borg and C. E. Birchenall, "Self Diffusion in Alpha Iron,"
'Trans. Met. Soc. AIME, Vol. 218, p. 980, (1960).
12. D. Y. F. Lai and R. J. Borg, "Diffusion in Body-Centered-Cubic Iron,"
Trans. Met. Soc. AIME, Vol. 233, p. 1973, (1965).
13. M. S. Wechsler, "Radiation-Embrittlement of Metals and Alloys,"
p. 296 in The Interaction of Radiation with Solids, North-Holland
Publishing Company, Amsterdam, 1964.
14. M. S. Wechsler, 'Fundamental Aspects of Radiation Effects on Diffusion-
Controlled Reactions in Alloys," p. 86 in Symposium on Radiation Effects
on Metals and Neutron Dosimetry, ASTM-STP-341, American Society for
Testing and Materials, Philadelphia, 1963.
15. F. A. Nichols, "Theory of Radiation Embrittlement and Recovery of
Radiation Damage in Ferritic Stecls," Phil. Mag., Vol. 14, p. 335, (1966).
16. R. S. Barnes, 'Mechanisms of Radiation-Induced Mechanical Property
Changes," p. 40 in Flow and Fracture of Metais in Nuclear Environments,
ASTM-STP-380, American Society for Testing and Materials, Philadelphia,
1965.
20
:
-
1
-
17. J. M. Blakely and H. Mykura, "Studies of Vacuum Annealed Iron Sur-
faces," Acta Met., Vol. 11, p. 399, (1963).
J. S. Bryner, "A Study of Neutron Irradiation Damage in Iron by
Electron-Transmission Microscopy," Acta Met., Vol. 14, p. 323, (1966).
19. B. L. Eyre and A. F. Bartiett, "An Electron Microscope Study of
Neutron Irradiation Damage in Alpha-Iron," Phil. Mag., Vol. 12, p.
261, (1965).
20. J. T. Stanley, J. M. Williams, and M. S. Wechsler, "The Effect of
.
.
.
-
Radiation on Atomic Kearrangements in Fe-N, Cu-Ni, and Cu-Al Alloys,".
.
-
.
.
.
in Radiation Effects in Materials, edited by W. F. Steely, Gordon
and Breach, New York, 1966.
J. T. Stanley and W. E. Brundage, "Effect of Irradiation on the Pre-
.---.
.--
cipitation of Nitrides in Alpha Iron," p. 2 in Radiation Metallurgy
Section Solid State Division Progress Report for Period Ending August
1965, ORNL-3878, January 1966.
22. J. D. Fast and B. M. Verrijp, "Diffusion of Nitrogen in Iron," J. Iron
Steel Inst., Vol. 176, p. 24, (1954).
M. Wuttig, J. T. Stanley, and H. K. Birnbaum, "Interstitial Solute
Trapping in Irradiated and Quenched Iron," submitted for publication.
24. F. R. Fujita and A. C. Damask, "Kinetics of Carbon Precipitation in
Irradiated Iron II - Electrical Resistivity Measurements," Acta Met., .
Vol. 12, P. 331, (1964).
25. R. A. Arndt and A. C. Damask, "Kinetics of Carbon Precipitation in
Irradiated Iron III - Calorimetry," Acta Met., Vol. 12, p. 341,
(1964).
21
26. H. Wagenblast, F. E. Fujita, and A. C. Damask, "Kinetics of Pre-
cipitation in Irradiated Iron IV - Electron Microscope Observations,"
Acta Meto, Vol. 12, p. 347, (1964).
27. M. S. Wechsler, "Radiation Effects, Diffusion and Body-Centered Cubic
Metals," p. 291 in Diffusion in Body-Centered Cubic Metals, American
Society for Metals, Metals Park, Ohio, 1965.
28. F. W. Noble and D. Hull, "Stress Dependence of Dislocation Velocity
from Stress Relaxation Experiments," Acta Met., Vol. 12, p. 1089,
(1964).
29. W. G. Johnston and J. J. Gilman, "Dislocation Velocity, Dislocation
Densities, and Plastic Flow in LiF Crystals," J. Appl. Phys., Vol. 30,
p. 129, (1959).
30. D. F. Stein and J. R. Low, Jr., "Mobility of Edge Dislocations in
Silicon-Iron Crystals," J. Appl. Phys., Vol. 31, p. 362,-.(1960).
31. J. J. Gilman, "Dislocation Mobility in Crystals,". J. Appl. Phys.,
Vol. 36, p. 3195, (1965).
32. N. E. Hinkle and N. K. Smith, "Tensile Tests on Irradiated Iron," .
- -
-
.
-
p. 27 in Radiation Metallurgy Section Solid State Division Progress
-
-
-
Report for Period Ending August 1965, ORNL-3878, January 1966.
-
2
.
"Dislocation Dynamics and Internal Stress'in Iron and
...Silicon Iron by the Stress Relaxation Method," submitted for publi-
cation.
34. S. M. Obr, "A Study of Radiation Hardening in Iron by Stress Relaxa-
tion Techniques," p. 16 Radiation Metallurgy Section Solid State
Division Progress Report for Period Ending August 1965, ORNL-3878,
January 1966.
. ns-....
r
.
.....
...
22
35.
S. M. Ohr, N. E. Hinkle, J. M. Williams, and M. S. Wechsler, "Dis-
location Dynamics in Irradiated Iron," in Radiation Effects in
Materials, edited by W. F. Sheely, Gordon and Breach, New York,
1967.
36. 1. Conrad, "On the Mechanism of Yielding and Flow in Iron," J. Iron
Steel Inst., Vol. 198, p. 364, (1961).
37. F. A. Smidt, "Effects of Irradiation on Thermally Activated Flow
in Iron," J. Appl. Phys., Vol. 36, p. 2317, (1965).
23
Figure Captions
1.
Completed Assembly of ORR Poolside Experiment Prior to Enclosure.
3.
Lower Yield Stress versus Dose Rate for Ferrovac-E Iron of 130m
Grain Size. Irradiation temperature, 95°C. Neutron dose, 4.6 x
1078 neutrons/cm? (E > 1' Mev). Test temperature, 30°C. Strain
Rate, 0.02 min-1.
Effect of Two-Hour Anneals at Indicated Annealing Temperatures on
the Yield Stress of Irradiated Ferrovac-E Iron Tensile Samples of
130u -Grain Size. Irradiation temperature, 94°C. Irradiation dose,
· 1.7 x 1018 neutrons/cm? (E.> 1 Mev). Irradiation dose rate, 8 x
10+2 neutrons/cm? (E > 1 Mev). Test temperature, 30°C. Strain
rate, 0.02 min-1
Annealing Time for 50 Percent Recovery of Yield Stress versus
Reciprocal Absolute Annealing Temperature for Irradiated Iron and
Steel.
Re-solution of Nitrogen in Alpha Iron upon Isochronal Annealing
(t = 15 min) Following 2-Mev. Electron Irradiation to 1.5 x 1019 .
„electrons/cm? at 170°K, Based on Internal Friction Measurements at
65°C. (After Stanley and Brundage(21)).
.
.
.
.
.
.
5.
Lower Yield Stress and Internal Stress versus Strain Rate for
7.
Ferrovac-E Iron of Three Grain Sizes. Test temperature, 24°C.
Internal Stress, 04, and Lower Yield Stress, Orys versus Grain
Diameter in Iron Silicon. Test temperature, 25°C. Strain rate,
0.016 min-1. .
8.
24
Dislocation Velocity versus Effective Strese in Iron Silicon at
Room Temperature, Based on Etch-Plt Data of Stein and Low (Ref. 30).
Activation Volume from Stress Relaxation Measurements versus Effec-
tive Stress in Unirradiated and Irradiated Ferrovac-E Iron of 130
Grain Diameter. Dose, 3.3 x 1028 neutrons/cm? (E > 1 Mov). Irradia-
tion temperature, 95°C. Test temperature, 30°C.

.ר
; ידיד.
..-
,...-.-
-
-
-
י
--
--
-.- .. ............... --- ---
- .. .....
--
-
..
דם.
:א.:..
: ..
:.
.
4
4
.
ייד
. 1
...
.
.
24
..
ה
בית
- ל--
י
י..
.
.
Figure 1
...
uni
.
ill
.
i
.
il
.
...
או או או 'רזי יזwזוטי
,
...4....
-
-
-
-
. .
. .
. . .
....
.. - .
- -
- . - -
ש
מו
או.איגוא
.
--
-
----
-
-
-
-
-
-
---
..
-
--
.
.
--
ז
-
דר
...
..
-- -,.
- ..
.
... - ..
ORNL- DWG 66-3577
(103)
TMT TTTTI
FILLED SYMBOLS REFER TO
CADMIUM - SHIELDED SAMPLES
LOWER YIELD STRESS (psi)
26
- UNIRRADIATED
101 2
e
5 1012 2
1013 2
NEUTRON DOSE RATE (neutrons /cm2 .sec, E > 1 MeV)
5
1014

Figure 2 :
..."
.
.
ORNL-DWG 66-4694
.:
(x 100)
-
IRRADIATED
LOWER YIELD STRESS (psi)
PERCENT RECOVERY
27

CONTROLS
-torot
(A = 590°C)
0
100
200
300
T, ANNEALING TEMPERATURE (°C)
400
.
.
:
Figure 3
.
.
ORNL-DWG 66-4699
TA, ANNEALING TEMPERATURE (°C)
550 500 450 400 350 300
2 x 104

ORNL A212-B PRESSURE
VESSEL STEEL
ACTIVATION 24 v 2.7 | 3.3
ENERGY
ANNEALING TIME FOR 50 PERCENT RECOVERY (min)
290°C
190
150 194
IRRADIATION
TEMPERATURE.
7 ORNL VACUUM-.
EMELTED IRON
.
UKAEA-HARWELL SILICON-
KILLED CARBON STEEL
-
10-1
L
.
4.4 .2 1.3 1.4 1.5 1.6 1.7 1.8.
TA, RECIPROCAL ABSOLUTE ANNEALING TEMPERATURE (x103)
:
Figure 4
+
300
(x10-3) (
T
325
T
ORNL-DWG 65-8519R 2
TEMPERATURE (°C).
350 375 400
TT

NITROGEN IN SOLUTION (wt %)
..
-
-
-
-
-
570
620
670
ANNEALING TEMPERATURE (K)
Figure 5
30
ORNL-DWG 65-8313
· 5x10-5 1.
STRAIN RATE (min-")
5x10-4 2x10-3 2x10-2
TTM
2x102
2 x100
4.6

[(x103
log STRESS (psi)
STRESS (psi)
- FERROVAC-E IRON-
0,5 GRAIN DIAMETER, 174
4,4 GRAIN DIAMETER, 604
10. GRAIN DIAMETER, 130u
OPEN SYMBOLS-LOWER
YIELD STRESS
CLOSED SYMBOLS -
THRESHOLD STRESS
SL
- 3
- 2
log STRAIN RATE (min)
:
Figure 6
.
..
.
'
7
.
'
-
-.
.. .
.
.
.
.
.
'
-
-
-
-
-
-
..
.
-
-
.
-
-
-
-
-
-
.
-
----
*--....."
-
-
-
-
-
-
-
-
--.-.--....
....
.
STRESS (kg/mm2)
.
.
Figure 7
012 (cm2).
1020
31
:
30
ORNL-DWG 65-10816R2

ORNL-DWG 65-10815R3

0; = 22.0 kg/mm2
log H = - 10.6
9
log V, DISLOCATION VELOCITY (cm/sec)
m= 8.01
20
:
:
5 . 10
0-0, , EFFECTIVE STRESS (kg/mm?)
Figure 8
-
.
T
.
.
.
.
.
.
.
.
.
.
..
-
-
:
-
33.
-
ORNL-OWG 66-1313 R2

...
1.
o UNIRRADIATED (0= 7.0 kg/mm2)
• IRRADIATED (0; = 21.3 kg/mm2;
. v*, ACTIVATION VOLUME (03)
Logomolo0000-4000
.
-
-
-
-
-
2
4
5
.
--
-
-
- -
0-9; ,EFFECTIVE STRESS (kg/mm2)
----
-
-
Figure 9
-
- -
-
-
- -
-
!:
2
inim
HAS
m
-...
.....
-
8 / 16 /67
DATE FILMED
END








-
be
-
-