In situ XRD studies of the process dynamics during annealing in cold rolled copper

S.Dey, N.Gayathri*, M.Bhattacharya and P.Mukherjee

Variable Energy Cyclotron Centre, 1/AF Bidhannagar, Kolkata 700064, India

Abstract

The dynamics of the release of stored energy during annealing along two different
crystallographic planes i.e.{111} and {220} in deformed copper have been investigated using
in-situ X-ray diffraction measurements at 458K and 473K respectively. The study has been
carried out on 50% and 80% rolled Cu sheets. The microstructures of the rolled samples have
been characterised using optical microscope and Electron Back Scattered Diffraction
measurements. The microstructural parameters were evaluated using Scherrer equation and
the modified Rietveld technique from X-ray diffractogram. The stored energy along different
planes were determined using the modified Stibitz formula from the X-ray peak broadening
and the bulk stored energy has been evaluated using differential scanning calorimetry. The
process dynamics of recovery and recrystallization as observed through the release of stored
energy has been modelled as a second order and first order process respectively.

Keywords: Deformation, Annealing, X-ray Diffraction, Electron Backscattered Diffraction,

Copper

*
Corresponding author:
Dr. Gayathri N. Banerjee,
Materials Science Studies Division,
Variable Energy Cyclotron Centre
1/AF, Bidhannagar,
Kolkata, 700064
Email: [email protected]
Phone: +91-33-23184460

1
1. Introduction:
Metallurgical research in the field of thermomechanical processing is mainly driven by
the need of industry. In order to control, improve and optimise the microstructure, grain
shape and texture of the final finished product, a detailed understanding of the physical
insight underlying the deformation and annealing phenomena is essential [1]. As a result, a
major need is to collect the quantitative information involving detailed microstructural
evolution during the deformation and annealing processes. The shape of the grains changes
with a creation of new grain boundary area by incorporation of the large number of
dislocations resulted during the deformation process [1]. The sum of the energy of these
dislocations and the new interfaces creating the domains within the grains represents the
stored energy of deformation [1]. The increase in the dislocation density is due to the
continued trapping of newly created mobile dislocations and their accommodation in the form
of the various microstructural features such as elongated grains, cells, domains, subgrains
etc. that are the characteristics of the deformed state [2]. The amount of stored energy in the
deformed material differs depending upon the configuration of dislocations and their density.
In a polycrystalline material, the stored energy of deformation is also highly orientation
dependent, since the deformation in the material depends on the slip activity which in turn
depends on the grain orientation [3-8]. The stored energy will thus vary from grain to grain
depending on its local crystallographic orientation [9,10]. Hence during recrystallization,
different crystallographic orientations will show different dynamics of the release of stored
energy compared to others because of a more favourable nucleation and/or growth [5,11].
Hence estimation of stored energy as a function of time along different crystallographic
planes during annealing of a deformed material will have a significant impact on selecting the
process parameters particularly annealing temperature and time, in order to achieve the
desired properties at the final finished stage.

The annealing process involves the decrease in stored energy and a corresponding
change in the microstructure. Hence the annealing behaviour of a material is not only
dependent on the overall stored energy but also on its spatial distribution [1]. Inhomogeneity
of stored energy on a local scale will affect the nucleation and the larger scale heterogeneity
will influence the growth of the recrystallized grains [1]. Hence, it is of high importance and
significance to study the annealing behaviour of a material with different deformations in the
light of accumulation of stored energy and distribution of defects. In this paper, we have
2
estimated the evolution of domains and the amount stored energy released at different
crystallographic planes during annealing of 50% and 80% cold rolled Cu. Several studies
have been done on the annealing of deformed Cu [12-15]. Ph. Gerber et al. [5] carried out
studies on evolution of global texture during recrystallization of cold rolled copper after
various rolling reductions. G. Mohamed et al. [16] investigated intergranular work hardening
state in moderately cold rolled copper and found that the stored energy varied substantially
with the initial orientation, leading to a strong variation of the recrystallization temperature.
The influence of cold deformation on the stored energy in other alloys have also been studied
with the help of neutron diffraction and Electron Back Scattered Diffraction (EBSD) [8,17].
D. Breuer et al. [18] estimated the density of dislocation and its arrangement in plastically
deformed copper using X-Ray. T. Rzychon et al. [19] determined the changes in
microstructure during compression and oscillatory torsion of deformed copper by XRD line
broadening. K. Piekos et al. [6,20] observed stored energy distribution with respect to crystal
orientation in polycrystalline copper experimentally as well as using stochastic vertex model
of recrystallization. They observed that the release of the stored energy suffered a
complicated process and was highly orientation dependent. L. Liu et al. [21] has measured
stored energy and recrystallization temperature in polycrystalline copper as a function of
rolling strain. C. Deng et al. [7] also estimated the bulk stored energy of cold rolled tantalum
by differential scanning calorimetry (DSC) along with the evolution of microstructure. R.D.
Doherty et al. [22] described the fundamentals of recrystallization with the understanding of
as-deformed state on different metals. D. Mandal et al. [15] studied the effect of grain
boundaries on the stored energy at different grain size of the cold worked copper. J. Schamp
et al. [23] concluded that the recrystallization is non-homogeneous across the cross-section of
the electrolytic tough pitch (ETP) copper wire by measuring the stored energy using DSC.
Other studies [2,24] have also been carried out to assess the static and dynamical stored
energy on deformed copper. However, the temporal release of stored energy at various
deformations and temperature with respect to different crystallographic planes and also the
process dynamics involved in releasing the stored energy have not been studied so far.

In this paper, an attempt has been made to characterise the microstructure of

electrolytic copper under optical microscope and EBSD in deformed and annealed condition.
The evolution of domains during annealing has been studied as a function of time using X-
ray Diffraction Line Profile Analysis (XRDLPA). The stored energy has also been estimated
qualitatively using modified Stibitz formula [9,25] which is only a first order approximation
3
and the error associated in estimation may be relatively high [26]. High Temperature X-ray
Diffraction technique has been used to study the annealing behaviour and the stored energy
variation with time in different crystallographic planes. The main aim of this work is to
understand the dynamics of the release of stored energy along different crystallographic
planes where the absolute value of the stored energy may not be necessarily important. The
overall stored energy of the samples has also been estimated by DSC at different
deformations. It is to be mentioned that the other important consequences of deformation
and/or annealing such as orientation of individual grains or texture have not been addressed
here.

2. Experimental details
The samples used in this study were electrolytic copper sheets (Table 1) which were
deformed at 50% and 80% by cold rolling at room temperature using a roller mill of diameter
16.5 cm at a speed of 20 rpm. Small samples were cut from the rolled sheet by jewellery
hacksaw and were mechanically polished using different grade silicon carbide (SiC) papers
and finally polished to mirror finish using fine grain diamond paste. Wide Angle X-ray
diffraction (XRD) profiles from as received and deformed samples have been recorded by
Bruker D8 Advance X-ray diffractometer using CuKα radiation. A range of 2θ starting from
40° to 100° and a scan step of 0.02° were used. The collection time given for each step was
0.5sec. Instrumental broadening correction was done using a standard defect free Si powder.

For the recrystallization study at the temperatures of 458K and 473K, small pieces of
the samples were cut from the rolled sheet and were thinned down to less than 0.2 mm to
satisfy the focussing condition of the Brag-Brentano geometry while using the high
temperature XRD stage. These samples were polished carefully and mounted on the high
temperature XRD stage consisting of a Platinum (Pt) strip heater using a small amount of
conducting silver paste. The whole stage was isolated from the surroundings by a vacuum
chamber which was maintained at a high vacuum better than 5×10-5 mbar during the
experiment. For carrying out the time dependent study of annealing at a particular
temperature, the sample was heated at a rate of 2 K/sec up to the desired temperature and
then the X-ray scan of a characteristic peak e.g. {111}, {220} was taken with different
soaking time. To understand the dynamics of the microstructural evolution, it is required to
collect the information of the evolution of microstructural features in-situ at a close interval

4
of time. In all cases, a freshly-prepared sample was taken to collect the data for a particular
{hkl} plane and also for different temperature. The time for collection of data of each {hkl}
plane was chosen after optimizing the peak to background ratio for obtaining an acceptable
peak. The 2θ range for the scans was also determined such that the desired amount of back
ground points are available on either side of the peak to carry out further analysis. Each peak
evolution was followed up to a time till no change could be observed in the integrated
intensity values. This indicated that the microstructural evolution (as observed by XRD) has
reached its saturation at that particular temperature.

The microstructure of the cold rolled and annealed samples were observed along the
plane containing the rolling and transverse direction using optical microscope (Carl Zeiss,
Inverted Optical Microscope) and FESEM (Supra 55, Carl Zeiss) equipped with a fully
automated EBSD system (Oxford instruments- HKL Channel 5). The Electron Back
Scattered Patterns (EBSP) were captured with a beam voltage of 20 kV, at a working
distance of 16 mm and a 70◦ tilt angle. The polished surface was electrochemically etched to
obtain a good microstructure. In each case, the sample was mounted with reference to the
rolling direction. The microstructure was also observed using EBSD for the samples after the
high temperature annealing studies.

The overall stored energy was measured on both the 50% and 80% rolled samples
using DSC/TG attachment of the Simultaneous Thermal Analyser (Netzsch STA 449F1).
Small pieces (less than 5 mm) of the rolled sheets were used for this study. The experiment
was carried out at a heating rate of 40K/min on both the samples up to 873K. After the 1st
run, the sample was cooled and reheated at the same rate to acquire the baseline for
calculating the quantitative stored energy. Argon gas was used to protect the sample against
oxidation during the entire experiment.

3. Results and Discussion

3.1 Deformation Study
Deformation causes the microstructural changes in various ways. The shape of the
grains changes along the direction of deformation and this leads to a large increase in the total
grain boundary area [1]. For cubic metals, the main mechanism of deformation is slipping
and twinning. The slipping mechanism is mostly predominant for the metals with high
stacking fault energy [27]. Though copper has stacking fault energy of 78mJm-2 [27], it is still
5
considered to be moderately high [28] and the mechanism of deformation of Cu is primarily
by slip followed by twinning. The essential difference between the deformed and the
annealed states lies in the dislocation content and their arrangement. It is well understood that
the evolution of the microstructure during recovery and recrystallization is determined by the
deformation microstructure based on the density, distribution and arrangement of the
dislocations [1]. Hence, it is of utmost importance to study the microstructure (grain size and
shapes), domains (dislocation cells within the cell blocks), stored energy along different
crystallographic planes and also the overall stored energy in the deformed material. In this
study we have characterised the deformed samples using various techniques to understand the
dislocation and its arrangement and its manifestation in the different length scales.

3.1.1 Deformation Microstructure

We have characterised the grain morphology of the as received and deformed samples
by optical microscopy and also by EBSD technique. Fig 1 shows the optical micrographs of
all the three samples (as received, 50% rolled and 80% rolled). Both the rolled samples
clearly reveal the effect of deformation on the grain morphology. The as-received sample
shows clear well defined grains with an average grain size of about 30 µm, which becomes
elongated with deformation. The deformation microstructure is more clearly evident in the
EBSD micrographs which are shown in Fig (2a, 2b and 2c) for all the three samples. The
50% rolled sample shows a smaller grain size with irregular grain boundaries whereas the
80% rolled sample clearly shows the elongated structure with highly irregular boundaries.
The small angle grain boundaries have been generated using the EBSP with a minimum and
maximum misorientation angle of 20 and 150 respectively. Using the software, the relative
frequency distribution of the misorientation angle between two consecutive EBSP has been
evaluated. The EBSD micrograph of the 80% rolled sample (Fig. 2c & 2c’) reveals a
significantly larger fraction of low misorientation angles as compared to 50% rolled sample
(Fig. 2b and 2b’). Another parameter that can be used to understand the effect of deformation
is the band contrast (BC) of the EBSP [29]. It is a quality factor derived from the Hough
transform that describes the average intensity of the Kikuchi bands with respect to the overall
intensity within the EBSP [30]. The values are scaled to a byte range from 0 to 255 (low to
high contrast). In general the grain boundaries and deformed regions tend to be darker than
the interior of the well recrystallized grains [30]. Similar parameters (such as pattern quality,
image quality etc.) defining the quality of the Kikuchi pattern generated in an EBSD have
been used recently to understand qualitatively the difference in the microstructure, in the

6
deformed, recovered and recrystallized stages of a sample [17,29-31]. The histogram of the
BC values obtained in the three samples are shown in Fig 2a’’, 2b’’ and 2c’’ respectively and
it can be clearly seen that the histogram becomes broader and the peak gradually shifts to a
lower grey value with increasing deformation.

3.1.2 X-ray diffraction studies

The X-ray diffraction pattern of the as received and the deformed samples are shown
in Fig 3. It is clearly observed that there is a systematic increase in the intensity of the {220}
peak and a decrease in the {111} peak of the 50% and 80% rolled samples respectively
compared to the as received sample. A significant broadening of the diffraction peaks of the
deformed samples is also observed indicating that the deformation has introduced changes in
the internal structure within the grains [1] causing the formation of substructure (domains).
During cold rolling, a large number of dislocations are introduced in the material but these
are not randomly distributed. These dislocations are accommodated into the material in a
variety of ways including creation of internal boundaries. It is seen that in case of copper
[1,28] rolled to low strains, the microstructure consists mainly of cells or domains (tangle of
dislocations) and also thin plate like microbands.

3.1.2(a) Domain size and microstrain evaluation

The X-ray diffraction line profile analysis (XRDLPA) of these samples was carried
out to evaluate the microstructural parameters such as size of the domains and microstrain
generated due to the deformation. The surface weighted domain size ( ) and the average

microstrain values <  > were obtained by the modified Rietveld technique using LS1
program [32]. The method of analysis has been described in the literature [33]. Fig 4 shows a
typical plot of modified Rietveld analysis of the highest deformed sample (80% rolled). The
microstructural parameters obtained by this technique for the 50% and 80% rolled samples
are shown in Table 2. The volume weighted domain size ( ) along different {hkl} planes
was also calculated using Scherrer equation [34] after the appropriate instrumental
broadening correction. The integral breadth (β) was calculated considering that the peak
profile follows the Lorentzian function. The peaks could not be fitted with a Gaussian or a
pseudo-Voigt function which implies that the strain contribution to the broadening is almost
negligible. The values of the corrected are also given in Table 2.

7
 It is clearly seen from the Table 2 that both and have decreased with increasing
deformation resulting in smaller coherent region or domains, but the value of the microstrain
is comparatively low and did not change with deformation. Copper being a material with
medium stacking fault energy [27,28], the mode of deformation in some grains may be
accommodated by two mechanisms, slipping and twinning. It is observed that the relative
shift in 2θ is not significant in case of both 50% and 80% rolled samples as compared to as
received sample. Hence, it may be considered that the stacking fault is almost absent even
after 80% deformation. With the increase in strain, the dislocation initially forms the cell
structure and subsequently the deformation is accommodated by twinning [1,30], the amount
of which increases with increasing deformation. However, we could not observe any
deformation twins in FESEM as these may be extremely fine. Humphrey et.al.[1] have
reported that the thickness of these twins are in the range of 0.2-0.3nm.

3.1.2(b) Calculation of Stored Energy along crystallographic planes

Rajmohan et.al. [9] have used a modified Stibitz [25] formula for calculating the
stored energy along different crystallographic orientations [9] using direction dependent
elastic modulus [Ehkl] & Poisson’s ratio [νhkl] as given below.
∆

= (1)
( )

where, is the Youngs modulus and is the Poisson’s ratio which are both
∆
direction dependent. The relative change in the lattice spacing can be obtained from the

broadening of the diffraction peaks from the following relation:

∆
= (2)

where, is the integral breadth after instrumental broadening correction at the Bragg angle
2 as explained in 3.1.2(a).

The values of the stored energy along the planes {111}, {200}, {220} and {311} have
been estimated using the direction dependent and and are presented in Table-3
as a function of deformation. It is evident from Table 3 that the stored energy is highly
orientation dependent and the value of stored energy is more in the highest density planes i.e.
{111} as compared to the other planes even in the as received samples. The stored energy has
increased in all the planes with increasing deformation with the change being maximum in

8
the {111} plane. The other planes show a relatively lower increase in the stored energy.
{111} being the slip plane can accommodate more dislocations and hence show a greater
increase in stored energy as compared to other planes. It should be noted that the error in full
width at half maximum value was high for the peaks with low signal to noise ratio ({311}
peak) and also low intensity values.

3.1.3 Stored energy by Differential Scanning Calorimetry

The stored energy provides the driving force for recovery and recrystallization [1,25].
Hence it is necessary to quantify the overall stored energy in both the samples which will
help to determine the annealing temperature. Fig 5shows the DSC curves of the 50% and
80% rolled samples heated at a rate of 40K/min after the baseline correction. It is clearly seen
that the peaks are obtained at temperature of 592K and 552K for 50% and 80% rolled
samples respectively. The area under the curve of the peak gives a measure of the stored
energy. It is found to be 38 J/mole and 46 J /mole for 50% and 80% rolled samples as seen in
Fig.5. These values corroborate with the studies done by L. Liu et. al. [21] as a function of
strain in pure copper.

It is evident that the stored energy of the 80%rolled sample is larger as compared to
the 50% rolled sample. Moreover, the release of the stored energy is found to occur at a lower
temperature for the 80% deformation. It is well known that the stored energy causes the
recovery and the recrystallization process to occur at a lower temperature [35].This may be
attributed to the fact that smaller activation energy is required to release the overall stored
energy for the 80% rolled sample, causing the process to start at a lower temperature.

4. Study of Recovery and Recrystallization

When a deformed material is annealed, recovery and recrystallization occur resulting
in the formation of defect free grains and thus restoring the property of the material to a
significant extent [2,27]. Both the recovery and recrystallization are competing processes and
both are driven by the stored energy in the deformed sample [1]. During the recovery process,
the microstructural changes are very subtle and occur on a small scale while new defect free
grains are nucleated and subsequently grow during the recrystallization process [1]. A
division between recovery and recrystallization is thus difficult to define; since recovery
mechanism plays a crucial role in nucleating recrystallization [22]. As recovery lowers the
driving force for recrystallization, a significant amount of prior recovery may in turn
9
influence the nature and kinetics of recrystallization [36-38]. The evolution of recovery and
recrystallization processes in a material depends on the extent of deformation (stored energy
in the sample) and also on the nature of the material ie, the stacking fault energy, impurities
etc [1]. As observed in the last section, the accumulation of stored energy is highly
orientation dependent and thus it is expected that the release of the stored energy along
different crystallographic orientations will show a different behaviour as a function of time.
A comprehensive study has been made in-situ to follow the release of stored energy with time
at a particular temperature to understand the dynamics of recovery and recrystallization.

The temperatures for the dynamic study were chosen based on the information from
the reported literature [5]. Ph. Gerber et.al. [5] have observed that for 90% and 70% rolled
pure Cu sample recrystallization of pure Cu starts at around 398 K and 448 K and ends at 473
K and 573 K respectively. Hence, in this study, the annealing was done at two different
temperatures (458 K and 473 K) to understand the dynamics.

The major slip system in copper is {111}<110> [27]. In this study, during the
deformation process (rolling) the {220} planes are found to be oriented along the rolling
plane which is evident from the increase in the peak intensity as seen in Fig 3.It is known that
during the rolling process, if the compressive stress is more than the longitudinal stress, then
the majority of {220} planes tend to orient towards the rolling plane [39].This is because, the
{111}<110> slip system becomes active tending to align the {111} planes perpendicular to
the rolling plane and since <110> is the slip direction, the corresponding {110} planes tend to
align parallel to the plane containing the rolling and the transverse directions. As a result, the
two major crystallographic planes which show predominant changes in the values of the
integrated intensity during the deformation process are the {111} and the {220} planes.
Hence the annealing experiment using high temperature XRD has been carried out on these
two crystallographic planes only.

4.1 In-situ dynamic studies of recovery and recrystallization using High Temperature XRD
4.1.1 Variation of domain size in different planes
The single peak analysis of the XRD peaks of {111} and {220} set of planes have
been performed to characterise the in-situ evolution of microstructure with time at different
elevated temperatures. The true integral breadth of {111} and {220} peaks as a function of
time were evaluated after correcting the Debye Waller factor and instrumental broadening at

10
458K and 473K. The volume weighted domain size for {111} and {220} was estimated using
the Scherrer’s equation [34] and plotted as a function of time. Fig. 6(a) and 6(b) reveal that at
two different temperatures, a distinct variation in the size of the domains in {111} and {220}
planes are observed as a function of time at the initial stage for both the deformed samples.
The domain size increased rapidly followed by a saturation for {111} and {220} peaks at
473K. But the rise in the domain size is observed to be gradual following a straight line at the
initial stage and then saturates with time at an annealing temperature of 458K. In the heavily
deformed sample, recovery occurs prior to recrystallization primarily with the changes of
dislocation structure in the material. It is to be noted that recovery and recrystallization are
competing processes and both are driven by the stored energy of the deformed state [1].
Initially there is a rearrangement of dislocations followed by annihilation which causes
increase in the domain size. This may be considered as the recovery stage and the onset of
saturation in the domain size as the start of recrystallization stage. The variation in the profile
of the evolution of domain size at two different temperature (Fig 6(a) and 6(b)) may have
occurred due to the difference in the amount of thermal energy supplied to the deformed
material. Hence, the increase in annealing temperature causes a significant amount of prior
recovery at a much lower time.

4.1.2 Variation of stored energy in different planes

Fig 7 shows the variation of the stored energy with time as calculated using equation
(1) for the 80% rolled sample from the diffraction peak broadening along the {111} planes in
a semi-log plot. The variation along {111} planes is monotonic (similar to an exponential
decay) at two different temperatures. It is worth noting that the saturation values of the stored
energy at the end of the experiments for both the temperatures are same (within the error bar)
which gives a good confidence in the experimental procedure. Fig 8(a) and 8(b) shows the
stored energy release with time along the {220} planes for the 50% and the 80%rolled
samples at the temperature458K and 473K respectively. It can be clearly seen that the nature
of the variation of the SE with time in the {220} planes is quite different compared to that of
{111}. The curve shows two distinct regions: a smaller decrease in the stored energy in the
early stage (stage I) and then an exponential decay (stage II), almost similar to the behaviour
of {111} set of planes (Fig. 7). The time corresponding to the change-over between the two
kinetics is significantly different in the two deformed samples (marked by arrows in Fig.8(a)
and 8(b)). The change in the process dynamics is found to occur at a lower time for 80%
deformation as compared to 50%, which is expected. In the initial stage, recovery occurs by
11
the rearrangement and annihilation of dislocations and the stored energy vs time curve shows
a different time dependence compared to that in the second stage which arises due to the
initiation of recrystallization processes seen Fig. 8. Since, the amount of stored energy is the
driving force of recrystallization, the onset of recrystallization occurs at a lower time for
higher deformation (80% deformation). This is also evident from the clear signature of the
initial increase in intensity for {220} peaks due to recovery followed by a sharp decrease due
to the formation of recrystallized texture as seen in Fig 9. This phenomenon could not be
observed for the {111} planes as the release of stored energy and the dynamic process of
annealing is so fast that it could not be followed within the laboratory time scale. The faster
dynamics in the {111} can be explained by considering the fact that this plane has the highest
the planar density of atoms and the initial stored energy is highest (Table 3) indicating that
this plane can accommodate more dislocations during the deformation process and
subsequently on annealing, the stored energy is released at a much faster rate.

These dynamic processes of annealing responsible for release of the stored energy can
be mathematically modelled in the following way. Recovery and recrystallization are the
primary process involved in the change in the stored energy of the deformed material.
Recovery involves the movement of the dislocations and subsequently their annihilation
causing a decrease in the stored energy, and recrystallization involves the process of
nucleation and growth of new defect free grains which is also again driven by the stored
energy in the material. The kinetics of these two processes will hence determine the overall
variation of the stored energy as a function of time. In a medium stacking fault energy
material, the mobility of the dislocation is reduced [40] and hence the movement of the
dislocations can thus be considered to be restricted to quasi two dimension and can be
modelled as a second order process. The recrystallization on the other hand is restricted to the
boundaries of the grains and the interfaces, and hence can be modelled as a first order process
[1,41]. As we clearly see from the experimental results above, the kinetics of the release of
stored energy is different along the {111} and {220} crystallographic planes in the
experimental duration. Hence, it can be conjectured that the time scale of the two processes
involved in the release of the stored energy is different along the two planes. This is expected
because, {111} set of planes have a higher stored energy and would thus have a lower
activation energy compared to the {220} set of planes. The stored energy variation is thus
fitted to the following equation.

12
 ( )= + +  (3)
where, τ1 and τ2 are the time constants of the second order (recovery) and the first order
(recrystallization) process respectively. A time ‘ 0’ has been included in the two terms to
account for the experimental delay time (required to be introduced to account for the time
during which both the processes have already began before the start of the experimental
observation). Coefficients A and B give the relative contribution of recovery and
recrystallization process to the dynamics. The stored energy variation has been fitted using
this equation and the values obtained are tabulated in Table 4.The fitting curves are
superposed on the experimental data in Fig 7, 8(a) and 8(b).It can be seen from the Table 4
that the time constants of the two processes are varying systematically with temperature as
well as the deformation. In the 80% rolled sample, the time constants of the recovery and
recrystallization processes are much less along the {111} plane compared to the {220} plane.
In fact, at T=473K, the curve could be fitted with only the first order process with a small
time constant of 423 secs. Along the {220} plane, the 80% rolled sample has lower time
constants compared to the 50% rolled sample. This is also expected since the 80% rolled
sample has a higher stored energy thus enabling the recovery and also the recrystallization
process to start at a earlier time compared to that of the 50% rolled sample. From table 4 it
can be seen that for the 50% rolled sample, annealed at 458K, the time constants of the two
processes are quite large and is comparable to the maximum experimental time. It can also be
seen from Fig.8(b) that the release of the stored energy is not complete in the sample
subjected to 458K within the experimental time. This can also be corroborated with the EBSP
analysis and the misorientation angle distribution of the same sample as discussed below. The
large differences in the time constants along the {111} and {220} planes clearly indicate that
there is a marked difference in the kinetics of the two processes ie., recovery and
recrystallization along the different crystallographic planes and the activation energy of the
processes is also different in the different planes.

4.1.3 Microstructure of the annealed samples

The microstructure of the annealed samples using EBSP shows clear indication of well
recrystallized grains as shown in Fig.10(a) and Fig.10(b). It is evident from the observations
that the irregular shaped boundaries as seen in Fig.2(b) and 2(c) have changed to sharper
boundaries with larger misorientation among the grains. This is especially more significant in
the 80% rolled sample. This also corroborates with the XRD peaks [Fig.11] which show clear

13
changes in the peak intensity of {200} and {220} planes indicating the formation of
recrystallized texture. It can be seen that for the 80% rolled sample, both the misorientation
angle distribution (Fig 10(b´), which shows a significant number of high misorientation angle
boundaries) and the band contrast distribution (Fig 10(b´´) which shows a peak shift towards
the high grey values and a much sharper distribution function) clearly shows that new defect
free grains have formed. In the case of the 50% rolled sample, the misorientation angle
distribution (Fig 10(a´)) and the band contrast (Fig 10(a´´)) clearly indicates that significant
recrystallization has not taken place; there are formation of new grains with lower
misorientation angle and only a slight shift of the band contrast distribution function to higher
grey values.

5. Conclusion
Detailed characterisation of deformed materials has been carried out with respect to
microstructure (grain size, domain size, microstrain), stored energy (overall and along
specific planes), temporal evolution of microstructure and release of stored energy at
different temperatures to study the annealing phenomenon. The deformed Cu samples have
been used to understand the kinetics of the release of stored energy along different
crystallographic planes during the annealing process. The microstructure was characterised
using EBSP which revealed irregular grain boundaries with increasing deformation. The
microstructural parameters as evaluated using XRDLPA technique showed a decrease in
domain size whereas the microstrain remained almost constant. The overall stored energy
was found to increase with deformation. The stored energy along the different
crystallographic planes was evaluated using the modified Stibitz formula and it was seen that
the accumulation of stored energy is much higher in {111} set of planes compared to the
other planes. The kinetics of the release of the stored energy along the {111} and {220} set of
planes were followed using in-situ XRD measurements carried out at 458K and 473K on the
deformed samples. The process dynamics of the recovery and the recrystallization could be
explained clearly from the observed time dependent variation of the stored energy in the
{220} set of planes. The two processes were found to follow second order and first order
dynamics respectively.

14
 REFERENCES

[1] F.J.Humphreys, M.Hatherly, Recrystallization and Related Annealing Phenomena.

Second ed, Elsevier (2004)
[2] E. Nes, Modelling of work hardening and stress saturation in FCC metals, Prog.
Mat. Sci. 41 (1998) 129-193.
[3] W.G. Burgers, Recrystallization Power and Shear Hardening in Aluminium Single
Crystals, Nature 131 (1933) 326-327
[4] C.S. Barrett, L.H.Levenson, The structure of aluminium after compression, Trans,
Metall. Soc. A.I.M.E. 137 (1940) 112-127. C.S. Barrett, T.B. Masalaki, The structure
of metals, third (revised) ed., Oxford Permagon, 1988
[5] Ph. Gerber, J. Tarasiuk, Th. Chauveau, B. Bacroix, Ph. Gerber, J. Tarasiuk, Th.
Chauveau, B. Bacroix, Acta Materialia, Acta Materialia 51 (2003) 6359–6371
[6] K. Piękoś, J. Tarasiuk, K. Wierzbanowski, B. Bacroix, Use of Stored Energy
Distribution in Stochastic Vertex Model, Materials Science Forum 571-572
(2008)231-236
[7] C. Deng, S.F. Liu, X.B. Hao, J.L. Ji, Z.Q. Zhang, Q. Liu, Orientation dependence
of stored energy release and microstructure evolution in cold rolled tantalum, Int.
Journal of Refractory Metals and Hard Materials 46 (2014) 24–29.
[8] A.L. Etter, M.H. Mathon, T. Baudin, V. Branger, R. Penelle, Influence of the cold
rolled reduction on the stored energy and the recrystallization texture in a Fe–53%Ni
alloy, Scripta Materialia 46 (2002) 311–317.
[9] N. Rajmohan, Y. Hayakawa, J. A. Szpunar, J.H. Root,Neutron diffraction method
for stored energy measurement in interstitial free steel, Acta Mater. 45 (1997) 2385-
2494.
[10] A. Borbely, J.H. Driver, Proc. Int. Conf. on recrystallization and grain growth,
Aachen. Eds. G. Gottstein and D. Molodov, Springer-Verlag, Berlin, 635.
[11] B. Bacroix, O. Castelnau, A. Miroux, H. Re´gle´, Proceedings of the 21st Risø
International Symposium on Materials Science, Roskilde, Danemark, 2000.
[12] M. Richert, J. Richert, A. Hotloś, W. Pachla and J. Skiba, Structure and
properties of copper deformed by severe plastic deformation methods, Journal of
Achievements in Materials and Manufacturing Engineering, 44 (2011) 50-56.
[13] H.W.F. Heller, C.A. Verbraaks and B. H. Kolster, Recrystallization at grain
boundaries in deformed copper bicrystals, Acta Metal. 32 (1984) 1391-1406.
15
[14] Bevis Hutchinson, Stefan Jonsson and Lena Ryde, On the kinetics of
recrystallization in cold worked metals, Scripta Metallurgica 23 (1989) 671-676.
[15] D. Mandal and I. Baker, The Influence of Grain Boundaries on the Stored Energy
of Cold-Work and Recrystallization Kinetics, Materials Science Forum 207-209
(1996) 521-524.
[16] G. Mohamed and B. Bacroix, Role of stored energy in static recrystallization of
cold rolled copper single and multicrystals, Acta Mater. 48 (2000) 3295-3302.
[17] Yanick Ateba Betanda, Anne-Laure Helbert, FrançoisBrisset, Marie-
HélèneMathon, Thierry Waeckerlé, ThierryBaudin, Materials
Science&EngineeringA614(2014)193–198
[18] D.Breuer, P. Klimanek and W. Pantleon, X-ray determination of dislocation
density and arrangement in plastically deformed copper, J. Appl. Cryst. 33 (2000)
1284-1294.
[19] T. Rzychoñ and K. Rodak, Microstructure characterization of deformed copper
by XRD line broadening, Archives of Materials Science and Eng.28 (2007) 605-608.
[20] K. Piękoś, J. Tarasiuk, K. Wierzbanowski,Generalized vertex model of
recrystallization application to polycrystalline copper, Computational Materials
Science 42 (2008) 584–594.
[21] L. Liu and I. Baker,The dependence of recrystallization temperature and stored
energy on rolling starin in polycrystalline copper,Scripta Metallurgica et Materialia 28
(1993) 197–200.
[22] R.D. Doherty, D.A. Hughes, F.J. Humphreys, J.J. Jonas, D. Juul Jensen,M.E.
Kassner, W.E. King, T.R. McNelley, H.J. McQueen, A.D. Rollett, Current issues in
recrystallization: A review,Materials Science and Engineering A238 (1997) 219–274.
[23] J. Schamp, B. Verlinden and J. Van Humbeeck, (Recrystallization at ambient
temperature of heavily deformed ETP Copper wire, Scripta Materialia 34 (1996)
1667-1672.
[24] D. Rittel, A.A. Kidane, M. Alkhader, A. Venkert, P. Landau, G.
Ravichandran,On the dynamically stored energy of cold work in pure single
crystaland polycrystalline copper, Acta Materialia 60 (2012) 3719–3728.
[25] G.R. Stibitz, Energy of lattice distortion, Phys Rev 49 (1936) 859

16
[26] A. Borbely, J. H. Driver and T. Ungar, An X-ray method for the determination of
stored energies in texture components of deformed metals; Application to cold
worked ultra high purity iron, Acta. Materialia 48 (2000) 2005-2016.
[27] George E. Dieter, Mechanical Metallurgy. Third edition, McGraw-Hill
International Edition. 1988
[28] Chao Voon S. Lim Department of Material Science & Engineering, Carnegie
Mellon University Phd thesis, Length Scale Effect on the Microstructural Evolution
of Cu layers in a Roll-bonded CuNb Composite.
[29] Tim Maitland, Scott Sitzman, Scanning microscopy and Nanotechnology, Ed.
Weilie Zhou, Zhong Lin Wang, Springer, 2006
[30] Adam J. Schwartz, Mukul Kumar, Brent L. Adams and David P. Field
(Eds),Electron Backscatter Diffraction in Materials Science, Second ed., Springer,
USA, 2009.
[31] S.Wronski, J.Tarasiuk, B.Bacroix, K.Wierzbanowski, H.Paul, Microstructure
heterogeneity after the ECAP process and its influence on recrystallization in
aluminium, Materials Characterization 78 (2013) 60–68.
[32] L.Lutterotti, P.Scardi, Simultaneous structure and size-strain refinement by the
Rietveld method, J.Appl. Crystallogr. 23 (1990) 246-252.
[33] P.S. Chowdhury, N. Gayathri, P. Mukherjee, M. Bhattacharya, A. Chatterjee, A.
Dutta, P. Barat, In situ Studies of Evolution of Microstructure with Temperature in
Heavily Deformed Ti-modified Austenitic Stainless Steel by X-ray Diffraction
Technique, Mat. Sci. and Eng. A 528 (2011) 967-972
[34] P.Scherrer, Bestimmung der Grösse und der inneren Struktur von Kolloidteilchen
mittels Röntgenstrahlen, Nachr. Gott. 2 (1918) 98-100.
[35] M.B.Bever, D.L.Holt, A.L.Titchener, The stored energy of cold work, Progress
in Materials Science 17 (1973) 5–177.
[36] P.A. Beck, Annealing of cold worked metals, Adv. Phys. 3 (1954) 245-324.
[37] M.B.Bever, Creep and Recovery ASM, Cleveland (1957) 14
[38] A.L. Titchener and M.B Bever, The stored energy of cold work, Prog. Metal
Phys., 7 (1958)247-338.
[39] Donald O. Thompson, Dale E. Chimenti, Review of Progress in Quantitative
Nondestructive Evaluation, Springer Science & Business Media, 9 2013 -Technology
& Engineering - 2144 pages

17
[40] F. Monthelillet and Jean-Philips Thomas, in Metallic Materials with High
Structural Efficiency, Ed by Oleg N. Senkov, Daniel B. Miracle, Sergey A. Firstov,
Kluwer Academic Publishers, Netherlands, 2004 pp. 357-386.
[41] T. LL. Richards, S.F. Pugh and H.J.Stokes, The kinetics of recovery and
recrystallization of copper from hardness and thermoelectric-power measurements,
Acta Metallurgica 4 (1956) 75-83.

18
 Table 1. Composition of the Copper sample

Element O Ag S Fe Ni Sn As Pb Sb Cu
Amount in 270 25 14 10 8 5 4 3 3 Bal.
ppm

Table2. Volume weighted domain size (calculated using Scherrer formula) of different {hkl}
planes and surface weighted domain size and microstrain (calculated using modified Rietveld
technique) of the as received and deformed samples

Sample Domain Domain Domain Domain Domain Microstr

size of size of size of size of size Ds ain
{111} at {200} at {220} at {311} at (nm)
RT (nm) RT (nm) RT (nm) RT (nm)
As 48.73 27.40 33.86 25.34 ---- ----
received
50%rolle 35.53 27.96 26.96 19.63 29.4 8.6x10-4
d
80%rolle 32.76 21.39 26.15 16.32 24.5 8x10-4
d

Table 3.Stored Energy of different {hkl} planes of the As-received, 50% and 80% rolled
samples

Sample SE of {111} at SE of {200} at SE of {220} at SE of

RT (J/Mole) RT (J/Mole) RT (J/Mole) {311} at
RT
(J/Mole)
As received 10.48 7.48 5.22 4.79
50% rolled 19.92 7.15 8.09 7.73
80% rolled 23.17 12.24 8.63 11.35

19
 Table 4. Parameters obtained by fitting eq. 3 to the observed stored energy variation as a
function of time.

Sample Temperature Peak SEt=∞ t0 τ1 τ2 A B

(K) (J/mole) (secs) (secs) (secs) (J/mole) (J/mol
e)
80% 458 {111} 2.3 1200 704 1420 409.3 23.7
rolled
473 {111} 2.0 577 - 423 0 23.5
458 {220} 0.6 1200 7565 9784 3.8 4.4
473 {220} 0.7 577 2461 2657 3.5 6.6
50% 458 {220} 0.9 1200 12160 17500 3.3 4.6
rolled
473 {220} 1.6 577 1729 3615 1.3 10.6

Figure Captions
Fig. 1. Optical Micrographs of (a) As received (b) 50% rolled and (c) 80% rolled copper
Fig. 2. EBSD Micrograph of (a) As-Received, (b) 50% rolled and (c) 80% rolled; Frequency
distribution of misoriented angles of (a´) As-Received, (b´) 50% rolled and (c´) 80% rolled ;
Frequency distribution of band contrast (a´´) As-Received, (b´´) 50% rolled and (c´´) 80%
rolled
Fig. 3. X-ray diffraction patterns of the samples.
Fig. 4. Typical plot of modified rietveld analysis of the 80% rolled sample
Fig. 5.Typical DSC curves of 50% rolled and 80% rolled samples
Fig. 6. Domain Size variation on (a) {111} planes and (b) {220} planes at two temperatures
458 K and 473 K.
Fig. 7. Stored Energy variation in {111} planes at two temperatures 458 K and 473 K on 80%
rolled Cu. The red line shows the fit curves according to eqn 3.

20
Fig. 8. Stored Energy variation for {220} plane at two temperatures 458 K and 473 K on (a)
50% rolled and (b) 80% rolled. The red line shows the fit of the data according to eqn 3.
Fig. 9.Stored Energy and peak Intensity variation for {220} plane at temperature 458 K on
80% rolled Cu.
Fig.10. EBSD Micrograph of annealed samples (a) 50% rolled and (b) 80% rolled; Frequency
distribution of misorientation angle of the annealed samples (a´) 50% rolled and (b´) 80%
rolled; Frequency distribution of band contrast of the annealed samples (a´´) 50% rolled and
(b´´) 80% rolled.
Fig. 11 X-ray diffraction patterns of the 80% rolled sample before and after annealing.
(Arrows indicate the decrease in the (220) intensity and the increase in the (200) intensity).

21
 ( )
(a) (b) ( )
(c)

Fig. 1. Optical Micrographs of (a) As received (b) 50% rolled and (c) 80% rolled copper
 0.7 0.05

0.6

Relative Frequency
0.04

Relative Frequency
0.5
0.03
0.4

0.3 0.02

0.2 0.01
0.1
0.00
0.0
0 10 20 30 40 50 60 0 50 100 150 200 250
Misorientation angle (degs) Band Contrast gray values

Fig. 2 (a) Fig. 2 (a´) Fig. 2 (a´´)

0.7 0.05

0.6
0.04

Relative Frequency
Relative Frequency

0.5
0.03
0.4

0.3 0.02

0.2
0.01
0.1
0.00
0.0
0 10 20 30 40 50 60 0 50 100 150 200 250
Misorientation angle (degs) Band Contrast gray values

Fig. 2 (b) Fig. 2 (b´) Fig. 2 (b´´)

0.7 0.05

0.6 0.04
Relative Frequency
Relative Frequency

0.5
0.03
0.4
0.02
0.3

0.2 0.01

0.1 0.00
0.0 0 50 100 150 200 250
0 10 20 30 40 50 60
Misorientation angle (degs) Band Contrast gray values

Fig. 2 (c) Fig. 2 (c´) Fig. 2 (c´´)

Fig. 2. EBSD Micrograph of (a) As-Received, (b) 50% rolled and (c) 80% rolled ; Frequency
distribution of misoriented angles of (a´) As-Received, (b´) 50% rolled and (c´) 80% rolled ;
Frequency distribution of band contrast (a´´) As-Received, (b´´) 50% rolled and (c´´) 80%
rolled
 (220)
5
1.0x10

s)
8.0x10
Intensity (count
4
6.0x10
(200)

4 (111) (311)
4.0x10

4
2.0x10
80% Rolled
0.0 50% Rolled
40
50
60
   d 70
80 As Received
e g re 90
es ) 100

Fig. 3. X-ray diffraction patterns of the samples.

Fig. 4 Typical plot of modified rietveld analysis of the 80% rolled sample
 0 .0 4 1 s t ru n T peak = 5 9 5 K
2 n d ru n

DSC/(mW/mg)
D iffe re n c e
0 .0 2

0 .0 0
-0 .2 5

-0 .3 0
5 0 % ro lle d
-0 .3 5
500 550 600 650 700
0 .0 2 T peak = 5 5 2 K 1 s t ru n
2 n d ru n
0 .0 1 D iffe re n c e
DSC/(mW/mg)

0 .0 0

-0 .2 5
-0 .3 0
-0 .3 5
8 0 % ro lle d
-0 .4 0
450 500 550 600 650 700
T e m p e ra tu re (K )

Fig. 5. Typical DSC curves of 50% rolled and 80% rolled samples
 120 120
110 110
100 473 K 100 473 K
90 90

Domain Size(nm)
Domain Size (nm)

80 80
70 70
458 K
60 60
458 K
50 50
40 40
{111} 30
30 {220}
20 20
(a) 80% Rolled (b) 80% rolled
10 10

0 10000 20000 30000 40000 50000 0 10000 20000 30000 40000 50000
Time (Secs) Time (secs)

Fig. 6. Domain Size variation on (a) {111} planes and (b) {220} planes at two temperatures 458 K and 473 K
 25
{111}
Stored Energy (J/mole)

458 K 80% rolled

20

15

10

473 K
5

0
100 1000 10000 100000
Time (secs)

Fig. 7. Stored Energy variation in {111} planes at two temperatures 458 K and 473 K on 80%
rolled Cu. The red line shows the fit curves according to eqn 3.
 14
{220} 10 {220}
Stored Energy (J/Mole)

12

Stored energy (J/mole)

50% Rolled 80% Rolled
8
10
458 K
8 6
458 K 473 K
6 4

4 473 K 2
2
(a) (b)
0
100 1000 10000 100000 100 1000 10000 100000
Time (secs) Time (secs)

Fig. 8. Stored Energy variation for {220} plane at two temperatures 458 K and 473 K on (a) 50% rolled and (b) 80% rolled. The red line shows the fit of
the data according to eqn 3
 9
Stored Energy (J/Mole)

8
{220}
7
6 80% rolled
5
4
3
2
1
0
22500
Intensity

20000

17500

15000

100 1000 10000 100000

Time (Secs)

Fig. 9. Stored Energy and Peak Intensity variation for {220} plane at temperature 458 K on
80% rolled Cu
 0.7 0.05

0.6
0.04

Relative Frequency
Relative Frequency
0.5
0.03
0.4

0.3 0.02

0.2 0.01

0.1
0.00
0.0
0 5 10 15 20 25 30 35 40 45 50 55 60 0 50 100 150 200 250
Misorientation angle (degs) Band Contrast gray values

Fig. 10(a) Fig. 10(a´) Fig. 10(a´´)

0.7 0.05

0.6

Relative Frequency
0.04
Relative Frequency

0.5
0.03
0.4

0.3 0.02

0.2 0.01

0.1
0.00
0.0
0 10 20 30 40 50 60 0 50 100 150 200 250
Misorientation angle (degs) Band Contrast gray values

Fig. 10 (b) Fig. 10 (b´) Fig. 10 (b´´)

Fig. 10. EBSD Micrograph of annealed samples (a) 50% rolled and (b) 80% rolled;
Frequency distribution of misoriention angle of the annealed samples (a´) 50% rolled and (b´)
80% rolled ; Frequency distribution of band contrast of the annealed samples (a´´) 50% rolled
and (b´´) 80% rolled
 140000 (200)
80% Rolled
120000

Intensity (counts)
100000
(111)
80000
60000 (220)

40000 (311)

20000
0
40 After annealing
 60
d
eg
ree 80 Rolled
s) 100

Fig. 11 X-ray diffraction patterns of the 80% rolled sample before and after annealing. (Arrows indicate the decrease in the (220) intensity and
the increase in the (200) intensity).