Journal of Nuclear Materials 447 (2014) 150–159

Contents lists available at ScienceDirect

Journal of Nuclear Materials

journal homepage: www.elsevier.com/locate/jnucmat

Fabrication of simulated plate fuel elements: Defining role of stress relief

annealing
D. Kohli a,b,⇑, R. Rakesh a, V.P. Sinha a, G.J. Prasad a, I. Samajdar b
a
Metallic Fuels Division, Bhabha Atomic Research Centre, Trombay, Mumbai 400085, India
b
Department of Metallurgical Engineering and Materials Science, IIT Bombay, Powai, Mumbai 400076, India

a r t i c l e i n f o a b s t r a c t

Article history: This study involved fabrication of simulated plate fuel elements. Uranium silicide of actual fuel elements
Received 27 July 2013 was replaced with yttria. The fabrication stages were otherwise identical. The final cold rolled and/or
Accepted 1 January 2014 straightened plates, without stress relief, showed an inverse relationship between bond strength and
Available online 10 January 2014
out of plane residual shear stress (s13). Stress relief of s13 was conducted over a range of tempera-
tures/times (200–500 °C and 15–240 min) and led to corresponding improvements in bond strength.
Fastest s13 relief was obtained through 300 °C annealing. Elimination of microscopic shear bands, through
recovery and partial recrystallization, was clearly the most effective mechanism of relieving s13.
Ó 2014 Elsevier B.V. All rights reserved.

1. Introduction An earlier study [13], as well as the present one, explored

Plate fuel elements are widely used in high flux research
reactors involved in the production of radioisotopes, irradiation
studies, etc. [1–7]. Roll bonding is the usual fabrication route
[1,3–8]. The final cold rolled plate is expected to have high stress
gradients. This has been shown through simulations [9–11], as well
as with experimental observations on plate fuels [12,13] and also
in accumulative roll bonding [14,15]. A preceding study [13] has
shown that such stress gradients are mainly in terms of out of
plane residual shear stress (s13). A direct correlation between s13
and bond strength between, aluminum cladding and simulated
fuel meat (of aluminum and yttria particles), was also indicated.
The stress relief process, typically given at the end of fabrication,
is a topic of considerable academic/applied interests [9,16–18]. Be-
cause of its apparent correlation with bond strength [12,13], stress
relief is naturally of concern to the fabrication of plate fuel ele-
ments. However, comprehensive studies relating stress relief with
property (such as bond strength) changes and microstructural
developments are virtually non-existent. It was hence decided to
initiate the present study: role of stress relief on bond strength
and microstructural developments.
Abbreviations: AA, Aluminum Association; ASTM, American Society for Testing
and Materials; EBSD, Electron Backscattered Diffraction; RD, Rolling Direction; ND,
Normal Direction; TD, Transverse Direction; GIXRD, Grazing Incidence X-Ray
Diffraction; XEC, X-ray Elastic Constant; FEG, Field Emission Gun; SEM, Scanning
Electron Microscope; OIM, Orientation Imaging Microscopy.
⇑ Corresponding author at: Metallic Fuels Division, Bhabha Atomic Research
Centre, Trombay, Mumbai 400085, India. Tel.: +91 22 25593929.
E-mail address: divyak@barc.gov.in (D. Kohli).
fabrication of simulated plate fuel elements. Yttria (Y2O3) replaced
uranium silicide as surrogate fuel in a typical (Fig. 1a) picture-
frame assembly. The fabrication stages (Fig. 1b) were otherwise
identical. The preceding study [13], by the same team, had docu-
mented aspects of microstructural developments at all (except
stress relief) fabrication stages. This study, on the other hand, tries
to rationalize careful observations on residual stresses with devel-
opments in microstructures and bond strength.
2. Experimental details
Fig. 1a shows the schematic of a picture-frame assembly. Com-
position of AA6061 cladding plates is given in Table 1. Surrogate
fuel meat consisted of a compact: aluminum and yttria (Y2O3) par-
ticles (60:40 by volume) of 20–100 lm size. This was cold com-
pacted at 900 MPa stress and a density of P95%. Y2O3 is selected
as surrogate fuel instead of actual uranium silicide (U3Si2). The
two materials have similar elastic modulus and coefficient of ther-
mal expansions. It may also be noted that a preceding study [13], in
the same journal, used this surrogate fuel and made observations
on plastic deformation. The present manuscript, on the other hand,
focuses on the stress relief annealing: the final stage of plate fuel
fabrication. Fig. 1b shows the process flow sheet used. Readers
may find further description in Ref. [13]. The final cold rolling
was often followed by industrial straightening operation [18].
The straightening consisted of 10 set of rolls placed on a rolling ta-
ble. This provided cyclic, albeit alternating, bending but did not im-
pose reductions in thickness. The final plates were subjected to
stress relief (followed by furnace cooling with similar cooling rates

0022-3115/$ - see front matter Ó 2014 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.jnucmat.2014.01.002
 D. Kohli et al. / Journal of Nuclear Materials 447 (2014) 150–159 151

(a)

(b)
Fig. 1. (a) Schematic showing typical picture-frame assembly. This included aluminum (AA6061) cladding (cover plate and picture frame) and surrogate fuel meat. The latter
consisted of 60:40 (by volume) aluminum and yttria (Y2O3) cold compacted powder. (b) Process flow sheet for simulated plate fuel fabrication. In the actual plate fuel
fabrication, Y2O3 is replaced with uranium silicide (U3Si2). Process flow sheet is otherwise identical.

Table 1
Composition of AA6061 in wt.% alloying elements. quarter in all the dimensions. At least three samples were mea-
sured. Limited numbers of samples were also subjected to pull test
Mg Si Mn Fe Cu Cr Zn Ti Al
[20]. This established, also shown in Ref. [13], a similar trend be-
1 0.67 0.19 0.20 0.30 0.024 0.015 0.022 97.57 tween peel and pull test data.
Characterization involved residual stress and EBSD measure-
ments. Stress measurements were made with appropriate depths
of 0.7 °C/min) at various temperatures (200 °C, 300 °C, 400 °C and of X-ray penetration (to be discussed latter) of the rolling plane
500 °C) and times (0, 15, 30, 60 and 240 min). (plane containing RD and TD). On the other hand, EBSD scans were
Mid-width sections of the rolled plates, with and without stress taken on the long-transverse section (containing RD and ND). EBSD
relief, were taken for peel tests. This was carried out with ASTM samples were prepared by standard metallography followed by
D1876-01 standard [19], after proportionately scaling down one electropolishing. Electropolishing involved an electrolyte of
152 D. Kohli et al. / Journal of Nuclear Materials 447 (2014) 150–159

perchloric acid and methanol (20:80 by volume), 20 °C and at x = incident/grazing angle between the incident beam and
13 V dc. Controlled polishing with sub-micron colloidal silica was specimen surface.
used for residual stress sample preparation. It was confirmed that
electropolishing and colloidal silica polish returned similar values In GIXRD, reflections for different {h k l} or Bragg angles (h) are
of residual stresses. This eliminated possibilities of stress relaxa- obtained. This provides different w:
tion through controlled colloidal silica polishing.
Residual stress measurements were made on a Panalytical™ Wh k l ¼ hh k l  x ð1Þ
X’Pert PRO MRD system. A commercial software, X’Pert Stress
Residual stress measurements estimate appropriate interplanar
Plus™, was used. Fig. 2a and b shows the diffraction geometry hkl
spacings (d/w ):
for GIXRD. It is important to define:
/ = rotation around the specimen plane normal. hkl o
d/;w  d
w = inclination between specimen plane normal and the diffrac- eh/;wk l ¼ o
tion vector. d

Fig. 2. (a) Diffraction geometry for grazing incidence X-ray diffraction (GIXRD). (b) Definitions of w and /: details deliberated in the text. Also included are: angular
conventions of rotations between laboratory (L) and specimen (S) co-ordinate systems and conventions for tensorial components of the residual stress. (c) The GIXRD profiles
at different grazing angles (x). (d) Estimated s13 at different distances from the interface. Standard deviations of stress measurements are shown as appropriate error bars.
 D. Kohli et al. / Journal of Nuclear Materials 447 (2014) 150–159 153

Fig. 3. (a) Images from simulated plate fuel fabrication: with plates undergoing successful bonding or severe debonding. Multiple {h k l} GIXRD estimated (b) r11 and (c) s13 at
cladding-surrogate fuel interface versus bond strength. Data include plates from final fabrication (without stress relief): with and without post-fabrication straightening
operation. Standard deviations of stress measurements are shown as appropriate error bars.

Fig. 4. (a) Bond strength with annealing time and temperature. Error bars indicate standard deviations from multiple peel-test measurements. Also included are
representative fractographs from samples with low and high bond strengths. (b) Rate of change of bond strength (0–30 min annealing) with annealing temperature.
154 D. Kohli et al. / Journal of Nuclear Materials 447 (2014) 150–159


1  2

2 By using a long incident beam path for small x, the above equation
¼ S1 ðr11 þ r22 þ r33 Þ þ S2 r11 cos2 / þ r22 sin / þ s12 sin2/ sin w
2 can be simplified as

1 1 sin x
þ S2 r33 cos2 w þ S2 ðs13 cos/ þ s23 sin/Þ sin2w ð2Þ C¼ ð5aÞ
2 2 l
rij and sij represent normal and shear components of a stress matrix In this l is termed as linear absorption coefficient and was taken as
(Fig. 2b). S1 and S2 are XEC, which can be related to the Poisson’s ra- 135.621 cm1 [29]. Controlled colloidal silica polish plus variable x,
tion (t) and Young’s modulus (E): allowed a calibration for C. This is shown in Fig. 2c. At x = 3.5°,
t Y2O3 peak was observed. Microscopic and estimated (Eq. (5)) values
S1 ¼  ð3Þ
E of C were approximately 4.5 lm cladding thickness. Such samples,
with 4.5 ± 0.5 lm cladding thickness from the cladding-surrogate
1þt meat interface, were used for subsequent stress relief annealing.
S2 ¼ ð4Þ
E For dislocation density estimation XRD system with line focus
E = 69.8 GPa was measured experimentally for the AA6061. For and high resolution (0.01° step size) h2h (where h is the Bragg an-
t = 0.34, the value from literature [21] was taken. gle) scans were used. The analysis was carried out in accordance
Standard d-sin2w measurements [22–25] involve a range of w with the 4th order variance method [30–33]. The estimated dislo-
values. In GIXRD [22,26–28], on the other hand, different {h k l} of- cation densities, at each point, were normalized with prior defor-
fer a range of d and w (Eq. (1)). A linear d-sin2w was then plotted mation values. EBSD measurements were done in FEI™ Quanta
with appropriate extrapolation [23,24] for w-splitting. 3D-FEG SEM using TSL-OIM™ system. Step size of 0.1 lm was used
This study used uniaxial stress analysis. Sample RD was aligned with identical beam and video conditions for all the scans. The
for u = 0°: allowing measurements of r11 and s13. 1, 2 and 3 EBSD scans were taken before and after stress relief annealing,
respectively represent the RD, TD and ND of the specimen using a suitably methodology described elsewhere [34].
(Fig. 2b). For peak identification Pearson VII profile fitting was
used, with appropriate user inputs for ‘minor’ shifts. In GIXRD, con- 3. Results
trolling x can offer different depths of penetration (C according to
Eq. (5) [26]: Fig. 2d plots s13 (out of plane residual shear stress) as a function
 1 of distance from cladding-surrogate meat interface. It is apparent
l l that a significant s13 gradient existed in the cold rolled plate. Near
C¼ þ ð5Þ
sin x sinð2hh k l  xÞ interface s13 was the highest. This study tried, with controlled

Fig. 5. (a) s13 and bond strength as function of annealing time for (a) 200°, (b) 300°, (c) 400° and (d) 500 °C annealing. Appropriate error bars, indicating standard deviations,
are included. s13, residual shear stress at cladding-meat interface, were obtained by GIXRD at 4.5 ± 0.5 lm cladding thickness from the cladding-surrogate meat interface.
 D. Kohli et al. / Journal of Nuclear Materials 447 (2014) 150–159
colloidal silica etch-polish (described earlier), to estimate near
interface s13 for a cladding depth of 4.5 ± 0.5 lm. Straightening at
times led to complete debonding and almost always to a drop in
bond strength. Extreme example of the former is shown in
Fig. 3a. Though estimated average bond strength did not relate to
r11 (Fig. 3b), Fig. 3c demonstrates an inverse relationship between
bond strength and s13. This validates a critical observation of the
earlier study [13]: the loss of bond strength through developments
of near interface out of plane residual shear stress (s13).
Fig. 4a collates the evolution of bond strength with stress relief.
Stress relief clearly improved the bond strength: nature of fracture
surface changed from brittle-ductile to fully ductile. Though bond
strength values were not remarkably different beyond 60 min
stress relief, initial 30 min clearly had different rates of bond
strength improvements – see Fig. 4b. The fastest improvement
was at 300° stress relief. This appears counter-intuitive.
Fig. 5 shows effects of stress relief at different annealing tem-
peratures. It is clear that drops in s13 matched corresponding
improvements in the bond strength. Fig. 6a confirms, for the initial
30 min annealing, that the rate of improvement in bond strength
has a direct correlation with rate of relaxation of residual shear
stress (s13). This was further analyzed with Zener–Wert–Avrami
function [35–37]:
   m 
st;T
13 Q tural developments and stress relief, it is important to reiterate
¼ exp  C  t  exp  ð6Þ
s13
t¼0 kT
t;T
where s13 is the residual stresses at time t (in s) and temperature T
(in K), s13 is the initial residual stresses in MPa, k is the Boltzmann
t¼0
155
constant, Q is the activation energy in eV, C is material constant and
t;T
m is ‘stress relaxation exponent’. A plot of log (ln (st¼0
13 =s13 ) versus
log time (Fig. 6b) gives the value of stress relaxation exponent
(m). Fig. 6c shows that m and rate of increase in bond strength
had a similar scaling with annealing temperatures. Both had a ‘high-
est’ at 300 °C.
To understand the rationale behind best stress relief kinetics at
300 °C, direct EBSD measurements were made, see Fig. 7. These re-
vealed temperature dependent (for 30 min annealing) develop-
ments in annealed microstructures. Annealing at 200 °C did not
significantly affect the deformed grain structure, while partial
recrystallization was noted after 300 °C stress relief. On the other
hand, 400 and 500 °C annealing led to an ‘apparent’ full recrystal-
lization. These observations raised an interesting point. They had
shown that partial recrystallization was more effective than an
‘apparent’ full recrystallization. A rationale for this counter-intui-
tive observation is sought in the subsequent section on discussion.
4. Discussion
The present study brought clear observations on temperature
dependent stress relief. Before initiating discussions on microstruc-
observations on improved bond strength. Rate of change in bond
strength, for initial 30 min annealing, was best at 300 °C (Fig. 4b).
This was clear and statistically reproducible. The only plausible
explanation is non-monotonic (with annealing temperature)
156 D. Kohli et al. / Journal of Nuclear Materials 447 (2014) 150–159

Before Stress Relief After Stress Relief

(30 min.)

200ºC

RD

TD 300ºC

400ºC

500ºC

Fig. 7. Direct observations on the microstructure developments during stress relief (30 min annealing). IPF (inverse pole figure) images (with data points below 0.1 CI marked
black) of identical regions before and after the stress relief treatments. 300 °C stress relief showed partial recrystallization, while microstructures above 300 °C appeared fully
recrystallized.

changes in residual stresses and/or effective diffusivities. The effec-
tive diffusivity [38,39] can be expressed as:
d
Deff ¼ Dgb þ gDpipe þ Dl
d
 
Q
D ¼ D0 exp 
RT
where Dgb = grain boundary diffusivity, Dpipe = pipe (dislocation) dif-
fusivity, Dl = lattice diffusivity, d = grain size, d = grain boundary
width and g = cross-sectional area of dislocations per unit area of
matrix. Eq. (8) represents temperature (T) dependence of diffusivi-
ties, Q being the activation energy for diffusion.
Annealing temperature/time brought in changes in grain size
d (Fig. 8a) and in dislocation densities (Fig. 8b). These data were
combined with other constants (see Table 2) and suitable mea-
sures of d and g. The value of d was taken [38] as 0.5 nm while
g was calculated considering 10 atoms in a dislocation cross-
section with matrix containing 1013 atoms/mm2 [38]. Estimated
effective diffusivities, as summarized in Fig. 8c, shows clear
temperature dependence. Annealing at 500 or 400 °C had 3
and 2 orders of magnitude higher diffusivities than 300 °C.
Possible differences in diffusivities and diffusion bonding thus
cannot explain highest rate of increase in bond strength
(Fig. 4b) after 300 °C annealing. Hence the experimental observa-
tion on best bond strength after 300 °C annealing can only be
explained from non-monotonic temperature dependence of
ð7Þ residual stresses and stress relief.
This study established, see Fig. 3, that s13 developments were
primarily through cold working. Kinetics of subsequent stress relief
ð8Þ (Fig. 6a) was clearly temperature dependent. Intermediate anneal-
ing temperature of 300 °C offered fastest s13 relief and bond
strength improvement. Arguably, high temperature annealing
may offer better diffusivity and corresponding improvements in
metallurgical bonding. However, this also have the potential of cre-
ating residual stresses through thermal mismatch. This study used
very slow furnace cooling (approximately 0.7 °C per minute),
which is expected to reduce (but not eliminate) thermal stresses.
The temperature dependence of the stress relief annealing
(Fig. 6a) was, however, apparent irrespective of the cooling
profiles.
An effort has been made to explain these observations
through a hypothesis and complimentary phenomenological evi-
dences, see Fig. 9. As shown in Fig. 9a(i), during unconstraint
deformation of cladding and surrogate fuel meat: higher strain
in cladding is expected. It is recorded in [13] that initial hot
working offered significant difference in the strain partitioning.
Strain in the cladding was more. Subsequent deformations
 D. Kohli et al. / Journal of Nuclear Materials 447 (2014) 150–159 157

Fig. 8. (a) Grain size, (b) normalized dislocation density (estimated dislocation density normalized by the value in fully recrystallized state), (c) estimated effective diffusivity
(Eq. (7)) are plotted as a function of annealing time and temperature. (a) Contains standard deviations in grain size from multiple measurements.

Table 2
Values of constants (Eqs. (7) and (8)) used for estimating effective diffusivities.
D0 (cm2/sec)
Grain boundary diffusion [40] 0.1
Dislocation pipe diffusion [41] 0.18
Bulk/lattice diffusion [40] 0.035
brought down such differences. During the final cold rolling
cladding and surrogate meat had similar strains. In other words,
initial unconstrained deformation of the picture-frame assembly
latter became constrained. A constrained deformation, with dif-
ferences in elastic–plastic properties, is expected to develop
residual stresses [42]. It was stipulated that constrained defor-
mation of cladding and surrogate meat resulted in ‘lateral’ stres-
ses (Fig. 9a(ii)). In extreme situation, generation of near interface
microscopic shear bandings was also hypothesized (Fig. 9a(iii)).
The presence of near interface s13 component (Fig. 2d) and
microscopic shear banding (Fig. 9b) supports the simplified sche-
matics/hypothesis of Fig. 9a. It is apparent that developments in
strong s13 also accompanied formation of the shear bands.
Annealing at 300 °C selectively eliminated such shear bands. This
is shown through direct EBSD observations (Fig. 9b). This was
clearly the most effective s13 relief.
Initial stages (30 min) of 400 and 500 °C annealing led to the
Q (kcal/mol)
formation of large grains, see Fig. 7. Visually they appeared fully
14.4 recrystallized. Such grains can originate through classical recrys-
20.34
28.8
tallization [42,43] plus strain induced boundary migration (SIBM)
[43–46]. SIBM was shown to leave orientation gradients [44]. This
appears to be the only rationale for clear orientation gradients in
grains after ‘apparent’ fully recrystallization (Fig. 9c). s13 stress re-
lief remained incomplete, see Fig. 5. Prolonged annealing, and vir-
tual elimination of orientation gradients, brought the stress relief
to completion.
This study provides an important correlation between near
interface out of plane residual stress (s13) and bond strength. This
is of practical relevance. More importantly, however, the present
study brings in important phenomenological correlations between
microstructure and stress relief. If stress is considered as a ‘distor-
tion’ of the unit cell (changes in inter planar spacings and angles)
[42], then such a distortion needs to be supported by appropriate
features of deformed microstructures. Similarly, stress relief is a
‘practical’ topic: which also should have microstructural origin.
The present study supports these arguments and perhaps initiates
future discourse on the ‘microstructural origin of stress relief
annealing’.
158 D. Kohli et al. / Journal of Nuclear Materials 447 (2014) 150–159

(a)

(b)

(c)
Fig. 9. (a) Hypothesizing developments in lateral stresses and shear banding during fabrication of simulated plate fuel. Details included in the text. (b) Direct EBSD
observations and corresponding image quality maps showing elimination of shear bands at 300 °C (30 min) stress relief. (c) ‘Apparent’ full recrystallized (500 °C and 30 min
annealing) grains had orientation gradients. Such gradients are shown with respect to average grain orientations in appropriate color codes’’.

5. Conclusions Acknowledgements

An inverse relation between measured bond strength and esti-
mated near interface out of plane residual shear stress (s13) was
established during the final stages of simulated plate fuel element
fabrication. The relationship was observed after the final fabrica-
tion (cold rolling and straightening), as well during stress relief
annealing. The latter enabled drop in s13 with corresponding in-
crease in bond strength.
Developments in ‘lateral’ stresses (s13) were supported by near
interface microscopic shear bands. Annealing at 300 °C eliminated
such bands through partial recrystallization and recovery. This was
more effective than stress relief at higher (400 and 500 °C) temper-
atures. The latter formed large grains, possibly through combina-
tions of classical recrystallization and SIBM. These grains,
however, had orientation gradients: and incomplete relief in s13.
Prolonged annealing plus elimination of orientation gradients were
needed to accomplish full stress relief.
The authors would like to acknowledge the EBSD and stress
measurements from the National Facility (DST-IRPHA) at IIT Bom-
bay. Support from BRNS (board of research in nuclear science) is
greatly appreciated. Support provided by the team members at
Plate Fuel Development and Fabrication group at MFD (metallic
fuel division, BARC) was critical: special acknowledgments to Mr.
M.R. Shaikh, S.S. Gotad, B.K. Mondal, B.N. Pisal and R.M. Malagi.
References
[1] F. Thummler, H.E. Lilienthal, S. Nazare, Powder Metall. 12 (1969) 1–22.
[2] James A. Lake, Donald K. Parsons, John L. Liebenthal, John M. Ryskamp, Gary N.
Fillmore, Deslonde R. Deboisblanc, Nucl. Instrum. Methods Phys. Res. A249
(1986) 41–52.
[3] M.K. Meyer, T.C. Wiencek, S.L. Hayes, G.L. Hofman, J. Nucl. Mater. 278 (2000)
358–363.
[4] Dennis D. Keiser Jr., Adam B. Robinson, Jan-Fong Jue, Pavel Medvedev, Daniel
M. Wachs, M. Ross Finlay, J. Nucl. Mater. 393 (2009) 311–320.
 D. Kohli et al. / Journal of Nuclear Materials 447 (2014) 150–159
[5] M.K. Meyer, G.L. Hofman, S.L. Hayes, C.R. Clark, T.C. Wiencek, J.L. Snelgrove,
R.V. Strain, K.H. Kim, J. Nucl. Mater. 304 (2002) 221–236.
[6] Dennis D. Keiser Jr., J. Gana, J.F. Juea, B.D. Miller, C.R. Clark, Mater.
Characterization 61 (2010) 1157–1166.
[7] G.L. Hofman, L.A. Neimark, F.L. Olquin, 1986 International Meeting on Reduced
Enrichment for Research and Test Reactor, U.S.A., November 3–6, 1986.
[8] C.R. Clark, S.L. Hayes, M.K. Meyer, G.L. Hofman, J.L. Snelgrove, Transactions of
session 2, in: 8th International Topical Meeting on Research reactor and fuel
management, March 21–24, 2004.
[9] Hakan Ozaltun, J. Nucl. Mater. 419 (2011) 76–84.
[10] Gregory K. Miller, Douglas E. Burkes, Daniel M. Wachs, Mater. Des. 31 (2010)
3234–3243.
[11] Hakan Ozaltun, Herman Shen and Pavel Medvedev, Proceedings of the ASME
2011 International Mechanical Engineering Congress & Exposition, November
11–17, 2011, Denver, Colorado, USA.
[12] C. Liu, M.L. Lovato, D.J. Alexander, K.D. Clarke, N.A. Mara, W.M. Mook, M.B.
Pime, D.W. Brown, in: 34th International Meeting on Reduced Enrichment for
Research and Test Reactors, October 14–17, 2012.
[13] R. Rakesh, D. Kohli, I. Samajdar, V.P. Sinha, G.J. Prasad, Fabrication of simulated
plate fuel elements:defining role of out-of-plane residual shear stress, J. Nucl.
Mater 445 (2014) 200–208.
[14] S.H. Lee, Y. Saito, T. Sakai, H. Utsunomiya, Mater. Sci. Eng. A325 (2002) 228–
235.
[15] S.H. Lee, Y. Saito, N. Tsuji, H. Utsunomiya, T. Sakai, Scripta Mater. 46 (2002)
281–285.
[16] K.H. Eckelmeyer, Advances in Surface Treatment, vol. 4, Pergamon Press, 1985.
[17] M. Movahedi, H.R. Madaah-Hosseini, A.H. Kokabi, Mater. Sci. Eng. A 487 (2008)
417–423.
[18] T. Ericsson, The effect of final shaping prior to heat treatment, in: G. Totten, M.
Howes, T. Inoue (Eds.), Handbook of Residual Stress and Deformation of Steel,
ASM International, 2002, pp. 150–158.
[19] ASTM: D 1876-01, Standard Test Method for Peel Resistance of Adhesives (T-
Peel Test).
[20] ASTM: C 633-01, Standard Test Method for Adhesion or Cohesion Strength of
Thermal Spray Coatings.
[21] J.S. Cheon, Y.S. Kim, Material Properties of Aluminum Alloys and Pure
Zirconium for Use in High-density Fuel Development for Research Reactors,
ANL/RERTR/TM-12-6, 2012.
[22] U. Welzel, J. Ligot, P. Lamparter, A.C. Vermeulen, E.J. Mittemeijer, J. Appl. Cryst.
38 (2005) 1–29.
159
[23] P. Van Houtte, L. De Buyser, Acta Metall. Mater. 41 (2) (1993) 323–336.
[24] I.C. Noyan, J.B. Cohen, Residual Stress-Measurement by Diffraction and
Interpretation, Springer-Verlag, Berlin, 1987.
[25] M.E. Fitzpatrick, A.T. Fry, P. Holdway, F.A. Kandil, J. Shackleton, L. Suominen
Determination of Residual Stresses by X-ray Diffraction-Issue 2, DTI
Measurement Good Practice Guide No. 52, 2005.
[26] A. Baczmanski, C. Braham, W. Seiler, N. Shiraki, Surf. Coat. Technol. 182 (2004)
43–64.
[27] P. Gergaud, S. Labat, O. Thomas, Thin Solid Films 319 (1998) 9–15.
[28] J. Peng, V. Ji, W. Seiler, A. Tomescu, A. Levesque, A. Bouteville, Surf. Coat.
Technol. 200 (2006) 2738–2743.
[29] B.D. Cullity, S.R. Stock, Elements of X-ray Diffraction, third ed., Prentice Hall,
2001.
[30] T. Unger, A. Borbely, Appl. Phys. Lett. 69 (1996) 3173–3175.
[31] I. Groma, Phys. Rev. B 57 (1998) 7535–7542.
[32] G. Guiglionda, A. Borbely, J.H. Driver, Acta Mater. 52 (2004) 3413–3423.
[33] R. Khatirkar et al., ISIJ Int. 51 (2011) 849–856.
[34] A. Albou, S. Raveendra, P. Karajagikar, I. Samajdar, C. Maurice, J.H. Driver,
Scripta Mater. 62 (2010) 469–472.
[35] Jérémy Epp, Holger Surm, Thomas Hirsch, Franz Hoffmann, J. Mater. Process.
Technol. 211 (2011) 637–643.
[36] O. Vohringer, Advances in surface treatments, Intr. Guidebook on Residual
Stress, vol. 4, Pergamon Press, Oxford, 1987. pp. 367–396.
[37] Weizhi Luan, Chuanhai Jiang, Vincent Ji, Mater. Trans. 50 (2009) 1499–1501.
[38] D.A. Porter, K.E. Easterling, Phase Transformations in Metals and alloys, second
ed., Chapman & Hall, UK, 1992.
[39] P.G. Shewmon, Diffusion in Solids, second ed., McGraw-Hill, New York, 1963.
[40] M.F. Ashby, Acta Metall. 20 (1972) 887–897.
[41] Jurgen Mimkes, Thin Solid Films 25 (1975) 221–230.
[42] B. Verlinden, J. Driver, I. Samajdar, R.D. Doherty, in: R.W. Cahn (Ed.), Thermo-
Mechanical Processing of Metallic Materials, ISBN-978-0-08-044497-0,
Pergamon Materials Series – Series, Elsevier, Amsterdam, 2007.
[43] F.J. Humphreys, M. Hatherly, Recrystallization and Related Annealing
Phenomena, second ed., Pergamon Press, Oxford, 2004.
[44] S. Raveendra, H. Paranjape, S. Mishra, H. Weiland, R.D. Doherty, I. Samajdar,
Met. Trans. A 40A (2009) 2220–2230.
[45] Paul A. Beck, Philip R. Sperry, J. Appl. Phys. 21 (1950) 150–152.
[46] Peter Bate, Bevis Hutchinson, Scripta Mater. 36 (1997) 195–198.