MA TE RI A L S CH A R A CT ER IZ A TI O N 7 8 (2 0 1 3) 4 7–5 9

Available online at www.sciencedirect.com

www.elsevier.com/locate/matchar

Review

In-situ mechanical testing during X-ray diffraction

Helena Van Swygenhoven⁎, Steven Van Petegem1

Materials Science and Simulations, NUM/ASQ, CH5232 Villigen PSI, Switzerland

AR TIC LE D ATA ABSTR ACT

Article history: Deforming metals during recording X-ray diffraction patterns is a useful tool to get a deeper
Received 3 September 2012 understanding of the coupling between microstructure and mechanical behaviour. With
Received in revised form the advances in flux, detector speed and focussing techniques at synchrotron facilities,
4 December 2012 in-situ mechanical testing is now possible during powder diffraction and Laue diffraction.
Accepted 6 December 2012 The basic principle is explained together with illustrative examples.
© 2013 Elsevier Inc. All rights reserved.
Keywords:
X-ray diffraction
Mechanical testing
In-situ
Plasticity

Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
2. In-situ Tensile Testing During Powder Diffraction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
2.1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
2.2. Experimental Setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
2.3. Example I — Residual Strain in Nanocrystalline Metals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
2.4. Example II — Inter-granular and Inter-phase Strains in Bainitic Steel . . . . . . . . . . . . . . . . . . . . . . 51
2.5. Example III — Phase Transformation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
3. In-situ Compression During Laue Diffraction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
3.1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
3.2. Experimental Setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
3.3. What Can We Learn? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
3.4. Example — Slip in Directionally Solidified Mo-alloy bcc Micropillars . . . . . . . . . . . . . . . . . . . . . . . 55
4. Outlook . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57

⁎ Corresponding author. Tel.: + 41 56 310 29 31; fax: + 41 56 310 31 31.

E-mail address: [email protected] (H. Van Swygenhoven).
1
Tel.: + 41 56 310 29 31; fax: + 41 56 310 31 31.

1044-5803/$ – see front matter © 2013 Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/j.matchar.2012.12.010
48 MA TE RI A L S CH A R A CT ER IZ A TI O N 7 8 (2 0 1 3) 4 7–5 9

strain) is kept constant during accumulation of the neutron

1. Introduction diffraction pattern, which can be anything between a few
minutes up to several hours. For samples that are sensitive to
The macroscopic mechanical behaviour of a metal is to a relaxation/creep this may result in a changing microstructure
great extent determined by the crystal structure since the during data acquisition [5–7]. On the other hand, in some cases
latter defines the dislocation mechanisms responsible for neutron diffraction may be the more appropriate tool. A
plastic deformation [1]. Most metals used in daily applications discussion on this subject is beyond the scope of this tutorial.
are however not single oriented crystals but polycrystalline The reader is referred to Ref. [8] and references therein for more
alloys composed out of grains with different crystallographic information on this topic.
orientations and often also different chemical phases. Under- In-situ mechanical testing during X-ray diffraction has
standing the relation between such microstructures and their benefitted greatly from several recent developments at advanced
behaviour under an applied load or temperature requires synchrotron beam lines:
therefore characterization of the microstructure at different
length scales through determination of the crystal structure,
– Detectors: for in-situ experiments fast detectors with large
defect density, grain size distribution, texture etc [2]. Chal-
field-of-view are crucial. Due to the high brilliance at
lenging is that under the influence of external load and/or
synchrotron beam lines the time resolution of in-situ
temperature, the microstructure usually changes. Typically,
experiments is often limited by the detector read-out
plastic strain increases the stress required for further defor-
time, rather than by the actual acquisition time. With the
mation which is called strain hardening but the opposite
advent of new fast CCD detectors such as, for instance, the
i.e. strain softening is also possible. Among the mechanisms
FreLoN detector [9,10], single photon counters such as the
responsible for the behaviour under load or temperature are
Mythen detector [11] and other new technologies [12–14],
for instance dislocation accumulation and interaction mech-
time resolutions ranging from seconds down to millisec-
anisms, temperature and/or stress induced grain growth, phase
onds are now readily achievable.
transformations and texture formation. It is needless to say that
– Spot size: during the last decade remarkable progress has
the ability to study the evolution of the microstructure during
been made in the quality of both chromatic and achromatic
deformation, i.e. in-situ, can provide insights into the mecha-
X-ray focussing optics. State-of-the-art diffractive and re-
nisms responsible for the observed mechanical behaviour and
fractive lenses or reflective optics (such as for instance,
produce useful input for upgrading computational simulation
Kirkpatrick–Baez focussing mirrors) are nowadays standard
tools used by engineers to predict behaviours of metallic
tools to produce X-ray spot sizes well below 1 μm, without
components.
sacrificing too much on the flux. This allows investigating
X-ray diffraction techniques based on the very simple
very small samples and/or obtaining spatial resolved dif-
Bragg equation
fraction signals with very high spatial resolution.
– Precise mechanics: sub-micrometre spot sizes calls for high
nλ ¼ 2dhkl sinθ ð1Þ
precision mechanical stages with nanometre resolution and
long-term stability. This is particularly challenging when
which links the lattice spacing dhkl with the wavelength λ and working with large sample environments such as furnaces
the angular position of the diffraction peaks θ, have played or tensile machines. Ultra-high stability can be obtained
and are still playing a prominent role in the characterization with the help of active damping using laser systems [15].
of the microstructure. The techniques allow the determina-
tion of not only the crystal structure of all microstructural
There exist about 50 synchrotrons worldwide, many of
components but also the defect density, grain size and
them equipped with dedicated beam lines for in-situ me-
internal strain through analysis of peak position and peak
chanical testing. Most in-situ experiments are performed
shape [3,4]. Performed in-situ during application of load and/or
under typical ‘powder diffraction’ conditions (i.e. a mono-
temperature, it allows studying the character of the activated
chromatic beam interacts with a deforming polycrystalline
dislocation slip systems, the accumulation of defect densities,
sample). Recently, mechanical testing is also performed
the competition between inter and intra-granular stresses
during Laue diffraction, where diffraction of a polychromatic
resulting from the different elastic–plastic response of individ-
beam is followed during deformation. In what follows, both
ual phases and from elastic and plastic anisotropy within one
techniques are explained briefly and illustrated by means of a
phase.
few case studies.
For similar purposes, neutron diffraction is used as a
complementary technique. In view of in-situ studies one of
the major advantages of X-ray diffraction at synchrotron
sources over neutron diffraction is the temporal component; 2. In-situ Tensile Testing During Powder
the flux at typical synchrotrons is many orders of magnitude Diffraction
higher compared to what can be obtained at typical neutron
facilities. As such, typical in-situ measurements at synchro- 2.1. Introduction
trons can be performed on the fly, i.e. without holding times,
whereas many experiments at neutron diffractometers are What can we learn? Essentially the same as in standard
performed by alternating data acquisition. For instance, during ex-situ powder diffraction i.e. the evolution of grain size
an in-situ deformation test often the stress (or alternatively, the (distribution), crystallography, internal stress and defects
 MA TE RI A L S CH A R A CT ER IZ A TI O N 7 8 (2 0 1 3) 4 7–5 9 49

density. Additionally one can obtain information on load remain elastic take over the load (so-called load transfer),
transfer, inelastic behaviour, defect accumulation, etc. Fig. 1 resulting in an increased slope (Fig. 1C). Additional information
pictures schematically what may occur during an in-situ test. can be obtained from the peak profile (Fig. 1D), which provides a
Here Fig. 1A shows a stress-strain curve, Fig. 1B a particular footprint of type II and type III stresses in the material. For
hkl diffraction peak, 1c the lattice strain for two hkl reflections instance when different grains experience different stresses
as a function of applied stress and 1d the peak width in (type II) or when dislocation networks are developing during
arbitrary units. The peak width is determined by the plastic deformation (type III), the diffraction peak will broaden.
combined effect of many contributions, the most important This is reflected by the green curves in 1b and 1d.
ones are grain size, intra- and inter-granular stresses, stacking When following the lattice strain of all hkl-reflections in an
faults and twins. The average distance between the lattice in-situ powder diffraction pattern, information can be obtained
planes determines the peak position. Factors may slightly on, for instance, the anisotropic elastic and plastic response of
change the peak positions [3], but this is beyond the scope of the crystalline structure along various crystallographic direc-
this article. During the elastic deformation regime represented tions [16], the load transfer mechanisms among different
schematically by the blue curve in Fig. 1A, the diffraction peak, phases [17], the building up of residual stresses during
blue peak in Fig. 1B, will shift linearly with the applied stress thermo-mechanical loading cycles and the Bauschinger effect
(Fig. 1C) towards the position of the red peak (Fig. 1B). When when internal strain in unloading curves or tension–compression
measuring the diffraction peak position and comparing with the sequences is studied [18].
reference value before application of stress, the hkl-dependent
lattice strain εhkl can be obtained:
2.2. Experimental Setup

dhkl  d0;hkl
εhkl ¼ ð2Þ In the following section the in-situ setup as implemented at
d0;hkl
the Materials Science beam line of the Swiss Light Source is
described as an example [19]. Fig. 2 displays a schematic
In other words, the crystal lattice is used as an elastic strain drawing of the key elements: 1) a parallel monochromatic
gauge and this for all phases that contribute to the diffraction X-ray beam with high energy resolution (ΔE/E ~ 10−4), 2) the
pattern. As soon as the plastic yield is reached (green curve in Micro Tensile Machine (MTM), which can be mounted in
1a) the lattice strain will deflect from linearity (Fig. 1C). The multiple directions on a 4-axis goniometer, 3) a non-contact
slope of the hkl reflections related to those grains or phases that extensometer, 4) a large solid angle for the diffracted beam
are plastifying will decrease. The other grains or phases that and 5) a 1-dimensional solid-state microstrip detector with

Fig. 1 – Schematic presentation of peak shift and peak broadening in an in-situ diffraction study. The elastic regime is
represented by the blue lines, green indicates the plastic regime and red marks the yield point.
50 MA TE RI A L S CH A R A CT ER IZ A TI O N 7 8 (2 0 1 3) 4 7–5 9

high angular resolution (3.7mdeg in 2θ), large field-of-view specimens were cut by electro-discharge machining (EDM).
(120° in 2θ) and high frame rate (5 Hz). Because of the small After EDM, the samples were electrolytically polished to remove
size of the sample (few mm only) the macroscopic strain any damaged layer caused by the spark erosion process. All
should be measured using a non-contact method. In this diffraction peaks could be fit satisfactorily with a split Pearson
setup this is achieved with a CCD camera by tracking markers VII function yielding the peak position, peak broadening (here
on the sample or by using digital image correlation techniques described by the full-width at half-maximum — FWHM) and
[20]. Alternatively a laser extensometer could be used [21]. other parameters reflecting the peak shape.
Similar setups can be found at several other dedicated beam Thanks to the high flux at the synchrotron data acquisition
lines. times are of the order of a few seconds only. This allows
continuous tensile testing with moderate strain rates without
2.3. Example I — Residual Strain in Nanocrystalline Metals the need to stop the experiment for data acquisition. Fig. 3A
displays the stress-strain relationship for nanocrystalline Ni
Nanocrystalline metals with grain sizes below 100 nm are deformed with a strain rate of 10−4 s−1. Due to the applied
well known for their intriguing mechanical properties, such as stress the lattice spacing changes, which is reflected by a peak
high strength and hardness, increased strain rate sensitivity movement. For a given crystallographic direction hkl the
and low ductility. However, the underlying mechanisms relation between peak position 2θhkl and lattice strain εhkl (in
responsible for these properties are less known. When grain units of microstrain) is given by:
sizes approach the nanometre regime the grain boundary (GB)
area increases rapidly and it is therefore no surprise that GBs ε hkl ¼ − cotðθhkl ÞΔθhkl  106 ð3Þ
will play a dominant role on the activated deformation
mechanisms [22,23]. It is now generally accepted that, as
where Δθhkl represents the deviation of the peak position from
suggested by molecular dynamics, GBs act as source and sinks
the position before loading. Fig. 3B represents the evolution of
for dislocations in a single dislocation regime [24,25]. On the
the lattice strain for the {111}, {200}, {220} and {311} grain
other hand, inter-granular mechanisms such as grain bound-
families along the transverse direction. The dashed lines in
ary sliding and migration and stress driven grain coarsening
Fig. 3B represent linear fits for the data points recorded at
are expected to contribute to plastic strain at the smallest
grain sizes [26,27]. In-situ X-ray powder diffraction for
studying the relation between mechanical behaviour and the
nanocrystalline structure has been applied by several groups
internationally [28–32].
The material under investigation here is nanocrystalline Ni
(Goodfellows) synthesized by pulsed electro-deposition. It has
a number-averaged grain size of 26 nm as determined by dark
field transmission electron micrographs and a mild {100}
out-of-plane texture. Impurities mainly consist of sulphur
(500 ppm) and some Fe, Cu and Co (less than 200 ppm) [33].
Uni-axial tensile tests were performed on 3 mm dogbone-
shaped specimens with a cross-section of 200 ×200 μm2. The

Fig. 2 – Schematic drawing of the in-situ deformation Fig. 3 – A) Stress-strain data for a continuous tensile test,
apparatus showing A) the incoming X-ray beam, B) the B) corresponding evolution of the transverse elastic lattice
tensile machine mounted on the goniometer, C) the CCD strain for the {111}, {200}, {220} and {311} diffraction peaks.
camera used to observe the sample during deformation, The dashed lines represent linear fits, C) deviation from
D) the diffracted beam, and E) the microstrip detector. linearity.
 MA TE RI A L S CH A R A CT ER IZ A TI O N 7 8 (2 0 1 3) 4 7–5 9 51

stresses below 0.8 GPa. The deviation between the data points because of the difficulty to absorb the dislocations slipping
and the linear fits is shown in Fig. 3C. through the grains [25,28–30].
Additional information can be obtained by performing load–
unload cycles, as shown in Fig. 4A. After each unload, the 2.4. Example II — Inter-granular and Inter-phase Strains
sample is kept free of applied stress for 10 min. Fig. 4B shows the in Bainitic Steel
evolution of the FWHM of the {111} and {311} diffraction peaks in
the unloaded state. Note that multiple diffraction patterns have Example I was dealing with an elementary metal. However,
been acquired in each unloaded state. During these periods the typical engineering materials contain often more than 1
FWHM slightly decreases, in particular at plastic strains above phase. This can generate significant internal stresses arising
1%. This is accompanied by a slight decrease in plastic strain from heterogeneous thermo-mechanical properties. The abil-
and may be attributed to relaxation effects that take place after ity to obtain phase selective information renders diffraction
unloading. In the early stages of plastic deformation, the FWHM techniques an additional advantage compared to other
clearly decreases. This trend is reversed for plastic strains above techniques to measure residual stresses. The following
0.5 %. example deals with the microstructural evolution during
Based on the observed evolution of the elastic lattice strain deformation of multiple phases in 1% CrMoV bainitic steel,
and peak broadening, the stress-strain curve of nanocrystal- which is a mid-carbon steel with ferrite, cementite and VC as
line Ni may be divided into three parts: I — elastic, II — main phases and which is known for its high temperature
microplastic and III — macroplastic, as indicated on Fig. 3A. In creep-resistance [37].
the elastic regime the different slopes in Fig. 3B reflect the Due to the low amount of carbon the volume fraction of
anisotropic nature of the elastic constants of Ni. After the cementite is relatively low (~3%). Therefore diffraction pat-
initial elastic regime, a microplastic regime is present, during terns with high statistics (>106 counts in the largest peak)
which the peak broadening decreases whereas the elastic have to be accumulated in order to investigate in detail the
lattice strain as a function of applied stress still remains cementite peaks. This can only be achieved within reasonable
approximately linear [34]. Here the decrease of the peak acquisition times at synchrotron facilities. Fig. 5 displays the
broadening can be ascribed to a relaxation of the internal corresponding diffraction pattern. The inset shows details of
stress. Note that in some nanocrystalline metals grain growth the diffraction pattern at the base of the (110) peak. All of the
is observed, which also leads to a decrease in peak broadening observed low intensity peaks coincide with cementite
[26,35,36]. After the microplastic regime, full plasticity sets in, reflections with an orthorhombic crystal structure. Electron
characterized by decreased strain hardening and changes in microscopy investigations reveal that the cementite particles
both elastic lattice strain and peak broadening. The deviation are located inside grains and at grain boundaries.
from linearity observed in Fig. 3B is similar to what is typically Fig. 6A displays the stress–strain curve during a tensile
observed for coarse-grained Ni, where it is related to the build deformation test. At each step the stress was held constant for
up of inter-granular stresses. The observation of a reduction 5 min, which allowed accumulating enough statistics to
in peak broadening during the microplastic regime is a matter investigate the behaviour of the small cementite peaks in
of debate. Together with insights obtained from molecular detail. Fig. 6B compares the behaviour of the (310) ferrite and
dynamics simulations [24,25] it is currently assumed that the (301) cementite reflections in the elasto-plastic regime.
during the microplastic regime the inter-granular stresses are The other diffraction peaks behave in a similar way. The
reducing due to grain boundary accommodation mechanism, lattice strain first increases linearly with the applied stress
whereas at larger strains these stresses build up again until the 0.02% yield, as indicated by the dashed line in Fig. 6B.

Fig. 5 – XRD Diffraction pattern with normalized intensity

obtained at the Materials Science Synchrotron Beamline for
1% CrMoV. The diffraction peaks all correspond to ferrite.
Fig. 4 – A) Stress–strain data of a series of load-unload cycles, The magnified region shows the details around the tails of
B) corresponding evolution of the FWHM of the {111} and the (110) peak. The additional, low intensity peaks can be
{311} diffraction peak after unloading. indexed as cementite (Fe3C).
52 MA TE RI A L S CH A R A CT ER IZ A TI O N 7 8 (2 0 1 3) 4 7–5 9

Fig. 6 – (A) Stress–strain curve during synchrotron XRD. (B) Lattice strain with applied true stress for the (310) ferrite (squares)
and the (301) cementite (circles) grain families parallel and perpendicular to the applied load, C) Ferrite (310) diffraction peak
with the scattering vector parallel to the loading direction: before loading; at 0.02% yield, before unloading and in the unloaded
state, D) similar as C) for cementite (301). The corresponding colour code identifying the position on the stress–strain curve is
given in (A). The intensity of the diffraction peaks is scaled to the most intense peak, i.e. ferrite (110).

Beyond the 0.02% yield, the cementite lattice strain increases stress state with opposing sign (Fig. 6B). The origin of the large
considerably, resulting in a large residual tensile lattice strain broadening of the cementite peaks cannot be deduced from
for the axial direction and a large residual compressive lattice this in-situ experiment solely. Supplementary powder diffrac-
strain for the transverse direction in the unloaded state. tion measurements on cementite powder chemically separat-
More information can be obtained by investigating the shape ed from the ferrite matrix before and after deformation
of the diffraction peaks in detail. Fig. 6C and d display, provide no evidence for plasticity in the cementite. Instead,
respectively, the (310) ferrite and the (301) cementite diffraction the large irreversible broadening of cementite peaks is
peaks in the axial direction before straining, in the elastic and presumably related to the existence of a broad distribution
plastic regimes and finally in the unloaded state. The corre- of stress states within the same cementite grain family. In
sponding positions along the stress/strain curve are indicated in literature other examples can be found where changes in peak
Fig. 6A. The intensities of the cementite peaks were normalized broadening are, in part, related to different stress states in
to the intensity of the most intense ferrite peak. At the onset of different grains [39,40]. By using 3DXRD it could even be
macroscopic yield, all peaks starts to broaden. At maximum explicitly shown that the different stress states in various
load the broadening of the cementite peaks is significantly sub-grains and in dislocation cell walls of deformed Cu can
larger compared to the ferrite peaks. Furthermore a strong peak account for the observed peak shape [41].
asymmetry has been developed. Upon unloading the peak
positions shift back to lower d-values but especially the 2.5. Example III — Phase Transformation
cementite peaks remain very broad.
The above results can be interpreted as follows: in the Because of its high sensitivity to changes in the crystal structure
elasto-plastic transition the ferrite matrix is shedding part of in-situ X-ray diffraction is an ideal tool to investigate phase
its load to the cementite, which hence might be regarded as a transformations during deformation. In the following example
reinforcing phase [38]. The broadening of the ferrite peaks is the evolution the microstructure during a mechanically in-
mainly due to an increase in dislocation density, although duced martensitic transformation is studied [42].
strain gradients because of incompatibilities between ferrite The material under investigation is a nanocrystalline austen-
and cementite may also contribute. After removal of the itic steel with nominal composition (in wt%) of Fe-18Cr-8Ni-2W,
applied load matrix and cementite are both left in a residual with and without 0.5% Y2O3 particles, further denoted as base
 MA TE RI A L S CH A R A CT ER IZ A TI O N 7 8 (2 0 1 3) 4 7–5 9 53

alloy and ODS (oxygen-dispersed alloy). Fig. 7 shows the stress-

strain curves for continuous uni-axial testing. The inset in 3. In-situ Compression During Laue Diffraction
Fig. 7B shows a section of the recorded diffraction pattern,
highlighting the (420) austenite and the (321) martensite peaks 3.1. Introduction
before deformation and just before failure. Newly emerging
martensite peaks grow at the expense of austenite peaks. All Laue diffraction is one of the oldest diffraction techniques,
the martensite peaks show a large broadening, suggesting the which however lost the interest from material scientists
presence of small coherent scattering volumes. Fig. 7B also because it was only applicable for single crystals. However,
shows the evolution of the volume fraction of the martensite with the enhanced performance and focusing conditions at
phase in the base and the ODS alloys with applied stress. The synchrotron light sources, Laue diffraction has gained back in
onset of the martensitic transformation coincides with the interest, as is witnessed by the papers in special dedicated
onset of macroscopic yield. The transformation saturates at a volumes such as Ref. [44]. White beam X-ray Laue diffraction
stress level of 1050 MPa when about 80% of the austenite is is sensitive to the underlying microstructure. Ex-situ it is
transformed into martensite in both alloys. The transformation used, for instance, in combination with a scanning technique
rate is, however, different for both alloys: the corresponding and newly developed differential aperture techniques, to
plastic deformation range at which 80% martensite is formed spatially resolve crystalline structures in two or three di-
was estimated to be about 1.35% in the base alloy and 0.85% in mensions with submicron resolution [45,46]. Properties that
the ODS alloy. These results indicate that in this material the can be measured include the local crystalline phase, local
kinetics of martensitic transformation is mainly determined by texture (orientation), and the local strains. When performed
the applied stress, rather than plastic strain, as is often under load, Laue diffraction can reflect the dynamics of the
assumed. Also in this case one can obtain information on the self-organization process of dislocations leading to the
build-up of inter-phase and inter-granular stresses, as is done formation of sub-grain structures [47]. Continuous streaking
in, for instance, Ref. [43]. of Laue diffraction peaks can be related to internal strain
gradients, while discontinuous streaking evidences the pres-
ence of dislocation walls, forming geometrically necessary
boundaries [48]. The path along which a Laue peak moves
during deformation can be linked to a specific active lattice
plane within the rotational Taylor model [47]. From the
collective motion of all Laue spots together, the rotation of
the local crystallographic orientation can be quantified.
Currently only a few beam lines are equipped with
mechanical testing devices to perform in-situ micro-
diffraction on micron-sized objects. In 2005 the first in-situ
micro-compression device was installed at the MicroXAS
beamline of the Swiss Light Source [49]. Since recent similar
experiments can be performed at BM32 at the ESRF [50]. At
ID1 at ESRF there exist an in-situ AFM device with which 3D
reciprocal space maps during deformation of sub-micron-
sized objects can be recorded [51].
Here we explain the setup at the MicroXAS beamline of the
Swiss Light Source. This device was originally developed to
study size effects in plasticity of micron sized single crystal
volumes, it is however also a very useful experiment to
characterize the sequence of activated slip systems in
complex microstructures.

3.2. Experimental Setup

Fig. 8 demonstrates the principle of the micropillar compression

testing technique: micro-pillars prepared using focussed ion
beam (FIB) milling, are compressed with a conventional
indenter tip with a flat end. Fig. 8D shows SEM images of Ni
micropillars with a diameter of 8 μm before and after deforma-
tion. By slip trace analysis the nature of the slip lines can be
Fig. 7 – A) Stress–strain data for nanocrystalline austenitic characterized. Information on the evolution of the microstruc-
alloys with and without yttria particles. B) Stress dependence ture inside the micro-pillars can only be obtained by X-ray
of volume fraction of the martensite phase. The inset shows micro-diffraction.
the intensity of the (420) austenite and the (321) martensite Fig. 8A and B shows schematic drawings of the in-situ
peak before deformation and before failure at 2% strain (base micro-compression device (MCD) installed at the MicroXAS
alloy). beam line of the Swiss Light Source. It shows the key elements
54 MA TE RI A L S CH A R A CT ER IZ A TI O N 7 8 (2 0 1 3) 4 7–5 9

Fig. 8 – A) Schematic drawing of the MCD setup at the MicroXAs beam line, B) MCD device in detail, C) schematic drawing of the
microcompression testing principle showing a row of free-standing micro-pillars, D) SEM images of Ni micropillars before and
after deformation revealing presence of localized slip.

needed to perform in-situ Laue micro-diffraction: 1) the equipped with a 20 μm flat diamond tip. This allows
intense synchrotron source providing an X-ray beam with deforming micron-sized objects with displacement resolu-
large bandwidth, 2) non-dispersive focussing system, 3) a tions of the order of nanometres and force resolutions below
deformation rig suitable for handling micron-sized objects, 1 μN.
4) a two-dimensional X-ray detector. In the following some 4) The diffracted X-ray beam is detected by a 16 bit charged
more details are given: couple device (CCD) detector that is positioned in a Laue
transmission geometry at typical sample-to-detector dis-
1) When using a broad wavelength spectrum (5–23 keV at tances of ~50 mm, resulting in a field-of-view of ±40degrees.
MicroXAS) Bragg's law is fulfilled simultaneously for a Note that the diffraction signal represents an integration
number of crystallographic planes, resulting in a set of of the information in the total illuminated volume,
diffraction spots on the detector, without the need to which for this geometry is defined by beam size × sample
rotate either the sample or the detector. thickness.
2) In order to optimize the signal to background ratio and to
avoid unwanted signals it is imperative to focus the white A critical step is the sample preparation technique. The
beam at least to the size of the object under study (ideally even micropillars should be free-standing, i.e. the X-ray beam
smaller, which allows performing spatial resolved measure- should be able to hit the micropillar under an angle of 90°
ments). This is usually achieved by using Kirkpatrick-Baez (relative to the axis of the pillar) in order not to interfere with
(KB) mirrors, which allow focussing a polychromatic beam the much larger compression head. Further the surrounding
down to ~0.5 μm–1 μm. material should not shadow that part of the diffracted beam
3) The MCD device consists of a set of 11 linear stages, 2 tilting that is recorded by the detector. A typical sample geometry is
stages and 1 rotary positioner, which are used for precise shown in Fig. 8C. As such, when using FIB to prepare the
alignment of the micro-pillars with the compression head pillars significant amounts of material have to be removed.
and with the X-ray microbeam. The actual compression test Here care should be taken not to introduce too much damage
is performed with a standard 1D Triboscope three plate in the micro-pillars, which is particularly important for a pillar
capacitive transducer from Hysitron Inc. The transducer is with a diameter below a few micrometres [52].
 MA TE RI A L S CH A R A CT ER IZ A TI O N 7 8 (2 0 1 3) 4 7–5 9 55

3.3. What Can We Learn? fluorescence mapping; the sample is moved in front of the
X-ray beam and the fluorescence signal of Mo is mapped in 2
First the recorded diffraction peaks should be assigned to dimensions as a function of its position. This allows aligning
particular lattice planes. There exist several methods to do the pillar with a precision of ~ 200 nm. The samples are
this so-called indexation procedure [53]. The position of the deformed in-situ with a loading rate of ~ 1 μN/s.
Laue diffraction spots is directly related to the average These particular pillars are chosen because of their initial
orientation and shape of the crystallographic unit cells in dislocation-poor microstructure. This is confirmed by the
the illuminated volume. In a fitting procedure these param- Laue diffraction patterns before deformation: the shape of
eters are refined leading to information on both crystal diffraction spots is similar to that of a standard defect-free
orientation and the elastic strain tensor. Because in white thin Si wafer. Fig. 9 displays the stress-strain data from a
beam diffraction information on the hydrostatic component series of load-unload cycles. Note that for microcompression
of the strain tensor is lost, one can only retrieve the deviatoric experiments the strain is usually determined from the
strain tensor [54]. Note that this can only be done in a reliable displacement of the indenter head. As micropillars produced
and reproducible manner provided that the position of the by FIB often exhibit a taper it is difficult to calculate a true
diffraction spots can be precisely determined with subpixel strain and is therefore usually avoided. A more detailed
resolution [55], which is often not possible in case of discussion on this topic can be found in Ref. [59]. The
significant strain gradients within the illuminated volume. evolution of the (101) diffraction spot is visualized in Fig. 10;
During an in-situ study the change in the orientation of the its behaviour is representative for the other diffraction spots.
crystal can be followed with high precision (~0.01degree). This The load in the first cycle reached 220 MPa and after
may render information on the nature of the activated slip unloading there was no measurable residual strain in the
systems. For instance, Taylor's theory stipulates that, when pillar. Furthermore the Laue peaks before and after the first
slip occurs in compression on a given slip plane, the normal of load–unload cycle had similar widths, indicating an elastic
this plane will rotate towards the compression axis [56]. Such regime. However, during loading some minor peak broaden-
a rotation can easily be tracked by plotting the crystallograph- ing is observed, indicating contributions from minor mis-
ic direction parallel to the compression axis in the crystal's alignments between the pillar axis and the indenter.
stereographic projection [56,47]. In the second loading cycle, upon additional straining, the
Also the shape of the diffraction spots can be related to the diffraction peaks move further along a rotation direction
underlying microstructure, however the interpretation is not corresponding to slip on the (112) plane with streaking along
trivial. The mostly used theory is currently provided by the the [111] direction. The activation of this slip system is also
work of R. Barabash and G. Ice. Here a differentiation between confirmed by post-mortem slip trace analysis using a scan-
several types of dislocation ensembles is made: i) statistically ning electron microscope. At 350 MPa, the peak starts
stored dislocations (GND) leading to isotropic peak broaden- spreading in other directions. This is a sign of slip activity on
ing, ii) geometrically necessary dislocations, resulting in other slip systems. Further loading introduces extensive
long-range rotations between sections of the crystal lattice, splitting of the peaks, resulting in a large orientation spread
which leads to peak streaking, iii) geometrically necessary as evidenced in pattern 45, and making it difficult to calculate
boundaries, i.e. GNDs organized into boundaries, resulting in further rotation of the compression axis. The streaking
peak splitting, iv) etc. The nature of the peak streaking and/or directions are no longer unique, suggesting activity on slip
splitting can be related to a GND density in case the curvature systems with a different 〈111〉 Burgers vector. Interestingly,
can be described by GNDs residing on one single slip system. A these slip systems have the same Burgers vector as
detailed discussion on this topic is beyond the scope of this the dislocations that were first activated, stressing the
paper. Details can be found in Ref. [48]. importance of cross-slip. Similar in-situ experiments have
been carried out on Mo-alloy pillars that were 1) deformed
3.4. Example — Slip in Directionally Solidified Mo-alloy bcc
Micropillars

It is well established that in body-centred cubic (bcc) metals at

temperatures well below a critical temperature Tc the flow
stress and plastic deformation mechanisms are primarily
controlled by the glide of ½〈111〉 screw dislocations on {110}
planes. Above Tc bcc metals behave fcc-like, i.e. the selection
of the slip system is primarily determined by the Schmid
factor [57]. However, at intermediate temperatures or in the
presence of precipitation the situation is less clear. In-situ
micro-compression can shed light on this matter because it
allows detecting sequentially activated slip systems in a small
single crystal volume.
The micro-pillars are produced from a directionally solid-
ified NiAl\Mo eutectic alloy [58], utilizing a combination of
focussed ion beam milling, etching and electropolishing [52]. Fig. 9 – Stress–strain curve of a load–unload experiment on
The micro-pillars are located in the X-ray beam using an as-prepared pillar.
56 MA TE RI A L S CH A R A CT ER IZ A TI O N 7 8 (2 0 1 3) 4 7–5 9

Fig. 10 – Evolution of the (101) diffraction peak with applied stress during the first two loading and unloading cycles presented
in Fig. 9. The lines represent the expected Taylor rotation for the 10 slip systems with the highest Schmid factor. Note that the
diffraction spot in panel A is located at the cross-point of the dashed lines and is only a few pixels wide.

prior to dealloying or 2) treated by FIB. Both sets of pillars further ongoing miniaturization of electronic and mechanical
exhibit pronounced different behaviour as compared to the components (MEMS and NEMS) [63] it can be anticipated that
defect-free pillars. In fact, it could be shown that the types of more dedicated setups will be constructed in the near future.
activated slip systems depend strongly on the nature of the Bragg coherent X-ray diffraction imaging (Bragg CDI) is a
defects introduced prior to the compression experiment. A promising tool to further push the limits in spatial resolution
detailed discussion on this topic is beyond the scope of this down to the nanometres scale [64]. Due to its steep rise in
tutorial. The interested reader is referred to Refs. [60] and [61] for popularity the intensive and complex reconstruction tech-
more information on this particular topic. niques that come along with this technique become more and
more efficient. Currently CDI is a relatively slow technique
and not particularly suited yet for most in-situ studies.
However, the advent of the X-ray free electron lasers with its
4. Outlook
mind-blowing brilliance may open new pathways for in-situ
CDI, or at least the push the time resolution of ‘conventional’
In-situ mechanical testing during conventional powder X-ray
in-situ diffraction methods further down.
diffraction is a well-established technique. When studying
Despite the recent advancements in experimental tech-
relatively large samples or thin films appropriate setups are
niques only little progress has been made in data analysis
readily available at the many synchrotrons around the world.
methods. Indeed, in-situ mechanical testing on advanced
More complex testing devices are currently under development
materials calls for the development of new models. In what
in order to study the behaviour of materials under conditions
follows two exemplary cases are described:
that go beyond the simple uni-axial test. For instance, recently a
in-situ bi-axial testing device has been developed, which allows Example I. Quantized crystal plasticity
studying the evolution of microstructure during strain path
changes or the influence of a bi-axial stress state [62]. Due to plastic anisotropy the stress/strain fields rapidly
In-situ testing on micron-sized specimen is currently avail- become very complex upon plastic deformation. Often, the help
able only at a few specialized beam lines. However, given the of extensive modelling is needed to interpret the results from
 MA TE RI A L S CH A R A CT ER IZ A TI O N 7 8 (2 0 1 3) 4 7–5 9 57

the diffraction experiments. For instance, the well-known Example II. Interpretation of Laue diffraction
elasto-plastic self-consistent model is often employed to patterns
understand the evolution of intergranular strains within certain
grain families along axial and transverse direction [65]. While It is far from trivial to link the shape of Laue diffraction spots
this works well for large grained materials, it fails to capture the with the underlying dislocation ensemble. Because of the
trends observed in small-grained materials. complex nature of this approach even the most advanced
As an alternative a quantized crystal plasticity (QCP) was methods rely on relatively simple dislocation organization
recently developed. Detail about the model can be found in Ref. models [48]. In case only one slip system is present, Laue spot
[66]. The model uses material parameters and the stress-strain streaking may be interpreted in terms of a GND (geometrically
curve as input. When a good fit between experimental and necessary dislocation) density. However for more complex
simulated properties are found, the lattice strain along certain strain fields no unique solutions can be derived and more
crystallographic directions can be calculated and directly complex measurement and analysis techniques are needed
compared with the lattice strain observed in the experiments. [68]. Currently under development in several groups are true
This is demonstrated in Fig. 11A and B for the case of bottom-up approaches: the calculation of Laue diffractions
nanocrystalline Ni (see Section 2.3) and coarse-grained Ni. It patterns from complex strain fields obtained from computer
shows the evolution of residual lattice strain (i.e. after simulations. For instance, Song et al. calculated Laue diffraction
unloading) as a function of plastic strain. The QCP model patterns from configurations obtained by a dislocation based
captures the behaviour and magnitudes of the residual lattice crystal plasticity formulation [69]. Devincre and co-workers use
strain in various grain families fairly well. The QCP model a slightly different approach. Here the dislocation ensembles
suggests that the role of the Schmid factor is reduced at the are obtained from 3-dimensional discrete dislocation dynamics
nanoscale. In the early stages of plastic deformation there exists (3DDD) computer simulations [70].
stress redistribution within the grain families, in addition to the
conventional load transfer between the grain families [67]. This
phenomenon can be ascribed to the heterogeneous nature of Acknowledgements
the microstructure of nanocrystalline metals.
The authors gratefully acknowledge the help of the present and
past PhD students of the Materials Science and Simulation
group at the Paul Scherrer Institut: Z. Budrovic, S. Brandstetter,
R. Maaß, J. Zimmermann, M. Weisser and S. Pierret. This work
would not be possible without the beam line scientists at the
Swiss Light Source: B. Schmitt and A. Cervellino from the MS
beam line, and D. Grolimund and C. Borca from MicroXAS.

REFERENCES

[1] Weertman J, Weertman JR. Elementary dislocation theory.

New York: Oxford University Press; 1964.
[2] Argon A. Strengthening mechanisms in crystal plasticity.
New York: Oxford Series on Materials Modelling; 2008.
[3] Cullity BD. Elements of X-ray diffraction. 3rd ed.New Jersey:
Prentice Hall; 2001.
[4] Warren BE. X-ray diffraction, Dover books on physics. New
York: Dover; 1990.
[5] Mulay RP, Clausen B, Agnew SR. In-situ neutron diffraction
study of the Bauschinger effect. Mater Trans A 2011;42A:60–70.
[6] Withers PJ, Holden TM. Diagnosing engineering problems
with neutrons. MRS Bull 1999;24:17–22.
[7] Fan GJ, Li L, Yang B, Choo H, Liaw PK, Saleh TA, et al. In situ
neutron-diffraction study of tensile deformation of a bulk
nanocrystalline alloy. Mater Sci Eng A 2009;506:187–90.
[8] Kisi EH, Howard CJ. Applications of neutron powder diffraction.
New York: Oxford University Press; 2008.
[9] Labiche JC, Mathon O, Pascarelli S, Newton MA, Ferre GG,
Curfs C, et al. Invited article: the fast readout low noise
camera as a versatile X-ray detector for time resolved
dispersive extended X-ray absorption fine structure and
diffraction studies of dynamic problems in materials science,
chemistry, and catalysis. Rev Sci Instrum 2007;78:091301.
Fig. 11 – Evolution of residual transverse lattice strain [10] Labiche JC, Mathon O, Pascarelli S, Newton MA, Ferre GG,
obtained A) during load/unload experiments on Curfs C, et al. Evaluation of imaging performance of a taper
nanocrystalline and coarse-grained Ni and B) from optics CCD ‘FReLoN’ camera designed for medical imaging.
configurations simulated by QCP modelling. J Synchrotron Radiat 2006;13:260–70.
58 MA TE RI A L S CH A R A CT ER IZ A TI O N 7 8 (2 0 1 3) 4 7–5 9

[11] Bergamaschi A, Cervellino A, Dinapoli R, Gozzo F, Henrich B, [32] Cheng S, Stoica AD, Wang XL, Ren Y, Almer J, Horton JA, et al.
Johnson I, et al. The MYTHEN detector for X-ray powder Deformation crossover: from nano- to mesoscale. Phys Rev
diffraction experiments at the Swiss Light Source. J Synchrotron Lett 2009;103:035502.
Radiat 2010;17:653–68. [33] Brandstetter S. Deformation mechanisms of nanocrystalline
[12] Ponchut C, Rigal JM, Clement J, Papillon E, Homs A, nickel studied by in-situ X-ray diffraction. PhD Thesis EPFL no
Petitdemange S. MAXIPIX, a fast readout photon-counting 4158; 2008.
X-ray area detector for synchrotron applications. J Instrum [34] Brandstetter S, Van Swygenhoven H, Van Petegem S, Schmitt
2011;6:C01069. B, Maass R, Derlet PM. From micro- to macroplasticity. Adv
[13] Le Bourlot C, Landois P, Djaziri S, Renault PO, Le Bourhis E, Mater 2006;18:1545–8.
Goudeau P, et al. Synchrotron X-ray diffraction experiments [35] Gianola DS, Van Petegem S, Legros M, Brandstetter S, Van
with a prototype hybrid pixel detector. J Appl Crystallogr Swygenhoven H, Hemker KJ. Stress-assisted discontinuous
2012;45:38–47. grain growth and its effect on the deformation behavior of
[14] Basolo S, Berar JF, Boudet N, Breugnon P, Caillot B, Clemens nanocrystalline aluminum thin films. Acta Mater 2006;54:
JC, et al. Application of a hybrid pixel detector to powder 2253–63.
diffraction. J Synchrotron Radiat 2007;14:151–7. [36] Fan GJ, Fu LF, Qiao DC, Choo H, Liaw PK, Browning ND. Grain
[15] Holler M, Raabe J, Diaz A, Guizar-sicairos M, Quitmann C, growth in a bulk nanocrystalline Co alloy during tensile
Menzel A, et al. An instrument for 3D X-ray nano-imaging. plastic deformation. Scr Mater 2006;54:2137–41.
Rev Sci Instrum 2012;83:073703. [37] Mayer K-H. The development of creep-resistant steels. In:
[16] Daymond MR, Priesmeyer HG. Elastoplastic deformation of Abe F, Kern T-U, Viswanathan R, editors. Creep-resistant
ferritic steel and cementite studied by neutron diffraction steels. Cambridge: Woodhead Publishing; 2008. p. 15.
and self-consistent modelling. Acta Mater 2002;50:1613–26. [38] Weisser MA, Evans AD, Van Petegem S, Holdsworth SR, Van
[17] Young ML, Almer JD, Daymond MR, Haeffner DR, Dunand DC. Swygenhoven H. In-situ room temperature tensile deformation
Load partitioning between ferrite and cementite during of a creep resistant bainitic 1%CrMoV steel using synchrotron
elastic–plastic deformation of an ultrahigh-carbon steel. Acta and neutron diffraction. Acta Mater 2011;59:4448–57.
Mater 2007;55:1999–2011. [39] Dubois JB, Thilly L, Renault PO, Lecouturier F, Di Michiel M.
[18] Thilly L, Renault PO, Van Petegem S, Brandstetter S, Schmitt Acta Mater 2012;58:6504–12.
B, Van Swygenhoven H, et al. Evidence of internal [40] Gil Sevillano J, Alkorta J, Gonzalez D, Van Petegem S, Stuhr U,
Bauschinger test in nanocomposite wires during in situ Van Swygenhoven H. Adv Eng Mater 2008;10:951–4.
macroscopic tensile cycling under synchrotron beam. Appl [41] Pantleon W, Wejdemann C, Jakobsen B, Lienert U, Poulsen
Phys Lett 2007;90:241907. HF. Mater Sci Eng 2009;523:55–63.
[19] Van Swygenhoven H, Schmitt B, Derlet PM, Van Petegem S, [42] Schloth P, Weisser MA, Van Swygenhoven H, Van Petegem S,
Cervellino A, Budrovic Z, et al. Following peak profiles during Susila P, Subramanya Sarma P, et al. Two strain-hardening
elastic and plastic deformation: a synchrotron-based mechanisms in nanocrystalline austenitic steel: an in situ
technique. Rev Sci Instrum 2006;77:013902. synchrotron X-ray diffraction study. Scr Mater 2012;66:690–3.
[20] Gianola DS, Eberl C. The micro- and nanoscale tensile testing [43] Muransky O, Sittner P, Zrnik j, Oliver EC. In situ neutron
of materials. JOM 2009;61:24–35. diffraction investigation of the collaborative
[21] Sharpe Jr WN. Applications of the interferometric deformation–transformation mechanism in TRIP-assisted
strain-displacement gage. Opt Eng 1982;21:483–8. steels at room and elevated temperatures. Acta Mater
[22] Kumar KS, Van Swygenhoven H, Suresh S. Mechanical 2008;56:3367–79.
behavior of nanocrystalline metals and alloys. Acta Mater [44] Barabash RI, Ice GE. Small-scale materials behavior from
2003;51:5743–74. X-ray microdiffraction and imaging. JOM 2010;62:21-1.
[23] Meyers MA, Mishra A, Benson DJ. Mechanical properties [45] Larson BC, Yang W, Ice GE, Budai JD, Tischler JZ.
of nanocrystalline materials. Prog Mater Sci 2006;51: Three-dimensional X-ray structural microscopy with
427–556. submicrometre resolution. Nature 2002;415:887–90.
[24] Van Swygenhoven H, Derlet PM, Frøseth AG. Nucleation and [46] Levine L, Larson BC, Yang W, Kassner M, Tischler JZ,
propagation of dislocations in nanocrystalline fcc metals. Delos-Reyes MA, et al. X-ray microbeam measurements of
Acta Mater 2006;54:1975–83. individual dislocation cell elastic strains in deformed
[25] Van Swygenhoven H, Weertman JR. Deformation in single-crystal copper. Nat Mater 2006;5:619–22.
nanocrystalline metals. Mater Today 2006;9:24–31. [47] Van Swygenhoven H, Van Petegem S. The use of Laue
[26] Zhang K, Weertman JR, Eastman JA. Rapid stress-driven grain microdiffraction to study small-scale plasticity. JOM 2010;62:
coarsening in nanocrystalline Cu at ambient and cryogenic 36–43.
temperatures. Appl Phys Lett 2005;87:061921. [48] Ice GE, Barabash RI. White beam microdiffraction and
[27] Brandstetter S, Zhang K, Escuadro A, Weertman JR, Van dislocations gradients. In: Nabarro FRN, JP Hirth JP, editors.
Swygenhoven H. Grain coarsening during compression of bulk Dislocations in solids, vol. 13. New York: Elsevier; 2007.
nanocrystalline nickel and copper. Scr Mater 2008;58:61–4. p. 500–601.
[28] Thilly L, Van Petegem S, Renault PO, Lecouturier F, Vidal V, [49] Maaß R, Van Petegem S, Van Swygenhoven H, Derlet PM,
Schmitt B, et al. A new criterion for elasto-plastic transition in Volkert CA, Grolimund D. Time-resolved Laue diffraction of
nanomaterials: Application to size and composite effects on deforming micropillars. Phys Rev Lett 2007;99:145505.
Cu\Nb nanocomposite wires. Acta Mater 2009;57:3157–69. [50] Kirchlechner C, Kiener D, Motz C, Labat S, Vaxelaire N,
[29] Budrovic Z, Van Swygenhoven H, Derlet PM, Van Petegem S, Perroud OJ, et al. Dislocation storage in single slip-oriented
Schmitt B. Plastic deformation with reversible peak Cu micro-tensile samples: new insights via X-ray
broadening in nanocrystalline nickel. Science 2004;304:273–6. microdiffraction. Philos Mag 2011;91:1256–64.
[30] Brandstetter S, Derlet PM, Van Petegem S, Van Swygenhoven [51] Scheler T, Rodrigues M, Cornelius TW, Mocuta C, Malachias
H. Williamson–Hall anisotropy in nanocrystalline metals: A, Magalhães-Paniago R, et al. Probing the elastic properties
X-ray diffraction experiments and atomistic simulations. of individual nanostructures by combining in situ atomic
Acta Mater 2008;56:165–76. force microscopy and micro-X-ray diffraction. Appl Phys Lett
[31] Li HQ, Choo H, Ren Y, Saleh TA, Lienert U, Liaw PK, et al. 2009;94:023109.
Strain-dependent deformation behavior in nanocrystalline [52] Zimmermann J, Van Petegem S, Bei H, Grolimund D, George
metals. Phys Rev Lett 2008;101:015502. EP, Van Swygenhoven H. Effects of focused ion beam milling
 MA TE RI A L S CH A R A CT ER IZ A TI O N 7 8 (2 0 1 3) 4 7–5 9 59

and pre-straining on the microstructure of directionally [62] Geandier G, Thiaudiere D, Randriamazaoro RN, Chiron R,
solidified molybdenum pillars: a Laue diffraction analysis. Scr Djaziri S, Lamongie B, et al. Development of a synchrotron
Mater 2010;62:746–9. biaxial tensile device for in situ characterization of thin films
[53] Various indexation software packages can be found at the mechanical response. Rev Sci Instrum 2012;81:103903.
website of the Collaborative Computational Project No. 14. [63] Bhusban B, editor. Springer handbook of nanotechnology. 3rd
http://www.ccp14.ac.uk. ed. Berlin: Springer; 2010.
[54] Ice GE, Pang JWL. Tutorial on X-ray microLaue diffraction. [64] Pfeifer MA, Williams GJ, Vartanyants IA, Harder R, Robinson
Mater Charact 2009;60:1191–201. IK. Three-dimensional mapping of a deformation field inside
[55] Petit J, Bornert M, Hofmann F, Robach O, Micha JS, Ulrich O, et al. a nanocrystal. Nature 2006;442:63–6.
Combining Laue microdiffraction and digital image correlation [65] Clausen B, Lorentzen T, Leffers I. Self-consistent modelling of
for improved measurements of the elastic strain field with the plastic deformation of FCC polycrystals and its implications
micrometer spatial resolution. Procedia IUTAM 2012;4:133–43. for diffraction measurements of internal stresses. Acta Mater
[56] Hosford WF. The mechanics of crystals and textured 1998;46:3087–98.
polycrystals. Oxford: Oxford University Press; 1993. [66] Li L, Anderson PM, Lee MG, Bitzek E, Derlet PM, Van
[57] Hirsch PB. Fifth international conference on crystallography. Swygenhoven H. The stress–strain response of nanocrystalline
Cambridge: University Press; 1960. p. 139. metals: a quantized crystal plasticity approach. Acta Mater
[58] Bei H, Shim S, Pharr GM, George EP. Effects of pre-strain on 2009;57:812–22.
the compressive stress–strain response of Mo-alloy [67] Li L, Van Petegem S, Van Swygenhoven H, Anderson PM.
single-crystal micropillars. Acta Mater 2008;56:4762–70. Slip-induced intergranular stress redistribution in
[59] Volkert CA, Lilleodden ET. Size effect in the deformation of nanocrystalline Ni. Acta Mater 2012;60:7001–10.
sub-micron Au columns. Philos Mag 2006;86:5567–79. [68] Larson BC, El-Azab A, Yang W, Tischler JZ, Liu W, Ice GE.
[60] Zimmermann J, Van Swygenhoven H, Marichal C, Van Experimental characterization of the mesoscale dislocation
Petegem S, Borca C, Bartova B, et al. Slip in directionally density tensor. Philos Mag 2007;87:1327–47.
solidified Mo-alloy micropillars — part I: nominally [69] Song X, Hoffman F, Korsunsky M-A. Dislocation-based
dislocation-free pillars. Acta Mater 2012;60:4604–13. plasticity model and micro-beam Laue diffraction analysis of
[61] Zimmermann J, Van Swygenhoven H, Van Petegem S, Borca polycrystalline Ni foil: a forward prediction. Philos Mag
C. Slip in directionally solidified Mo-alloy micropillars—part 2010;90:3999–4011.
II: pillars containing defects. Acta Mater 2012;60:4614–22. [70] Devincre et al. Private Communication. 2012.