Texture Implicit in the macrotexture approach is that an

averaged texture for the whole sampled volume is

The ‘‘ crystallographic orientation ’’ of the component obtained. X-ray diffraction for macrotexture measure-
units in both natural and technological materials refers ment is a standard technique, established for many
to how the atomic planes in a volume of crystal are decades, and as such provides efficiently an overview
positioned relative to a fixed reference. This charac- of the texture. The texture is made up of many
teristic applies to all solids whose structure is crys- individual orientations, and a microtexture approach
talline, including minerals, ceramics, semiconductors, measures these individually by using electron dif-
superconductors, and metals. Almost all of these fraction in an electron microscope. With such an
materials are polycrystalline, where a ‘‘ crystal ’’ or instrument not only are individual orientations
‘‘ grain ’’ is a volume of material to which can be measured, but their spatial location can be viewed and
ascribed a single orientation. Grain orientations in directly related to the microstructure. Hence the
polycrystals are rarely randomly distributed. In most definition of microtexture encompasses not only
materials there is a propensity for the occurrence of single-orientation sampling, but also the spatial speci-
certain orientations, caused first during crystallization ficity and subsequent display of these data (Randle
from a melt or amorphous solid state, and sub- 1992).
sequently by further thermomechanical processes. The advent of microtexture has opened up new and
This tendency is known as ‘‘ preferred orientation ’’ or, exciting possibilities for texture research and practice,
more concisely, ‘‘ texture.’’ since in the study of fabricated and natural materials it
The importance and significance to materials of is frequently important to have a knowledge of both
texture lies in the fact that many material properties microstructure and texture. The traditional approach
are texture specific (Bunge 1987). Some examples of has been to make parallel, but separate, investigations:
properties that depend on the texture of a material optical microscopy, scanning electron microscopy
are Young’s modulus, Poisson’s ratio, strength, duc- (SEM)-based techniques, and image analysis to
tility, toughness, magnetic permeability, electrical analyze microstructure, and x-ray diffraction to obtain
conductivity, and, in noncubic materials, thermal the texture. A typical profile of a material might
expansion. The effect of texture on properties is therefore have included the grain size distribution plus
exploited in materials technology in order to produce the texture as determined by x rays. The principal
materials with specific characteristics or behavior. In disadvantage of this traditional approach was the lack
general, the exact mechanisms by which particular of a direct connection between the study of micro-
textures evolve are incompletely understood, although structure and texture. Hence a wealth of information
the empirical validation is sufficient for many processes concerning the spatial component of texture, including
to have become established commercial practice. the interface regions where grains join, was inac-
The traditional tool for texture evaluation is x-ray cessible. Furthermore, where phases having identical
diffraction. Since the mid-1980s this technique has or similar chemistry, but different crystallography,
been supplemented by electron diffraction, creating a coexist in the microstructure they could not be
new and expanding branch of texture analysis. This distinguished by any simple method. However, with
article provides a complete overview of state-of-the- microtexture techniques in both SEM and trans-
art texture analysis, with particular emphasis on mission electron microscopy (TEM), it is now possible
recently established practice. to obtain microtexture measurements in a routine
manner. Since microtexture is an increasingly popular
practice, and moreover involves more ramifications
than macrotexture, the former will be described in
1. Macrotexture and Microtexture Techniques
more detail in this article.
Almost all techniques for texture analysis exploit the The technique that has become synonymous with
diffraction of electron, neutron, or x radiation by a microtexture is electron backscatter diffraction
crystal lattice. A dichotomy in the classification of (EBSD), sometimes called backscatter Kikuchi dif-
texture analysis is then based on whether or not the fraction (BKD), an add-on package to SEM (Venables
radiation chosen can sample many orientations and Harland 1973, Dingley 1981, Randle and Engler
together or a single orientation at a time. This 2000). The most attractive feature of EBSD is its
characteristic in turn depends on the wavelength of the unique capability to perform concurrently rapid,
radiation employed. In general the wavelength of x (usually) automatic diffraction analysis to give
rays and neutrons (0n05–0n3 nm) means that the size of crystallographic data and imaging with a spatial
the probe samples large volumes containing many resolution of less than 0n5 µm, combined with the
grains, typically thousands, whereas the much smaller regular capabilities of SEM such as capacity for large
wavelength of electrons (0n001–0n01 nm) permits specimens, option of chemical analysis, and the ability
single-orientation sampling. These two approaches to image rough surfaces.
are known as ‘‘ macrotexture ’’ and ‘‘ microtexture,’’ Automated EBSD can be exploited by preselecting
respectively. a grid of sampling points whose spacing is finer than

1
Texture

the grain size. If a color representation of the orien- the reflected intensities as a function of the diffraction
tation at each grid point is plotted, a crystallographic angle for gradually increasing angles θ and 2θ.
map of the microstructure is obtained. This is known Provided the crystal structure of the material is known,
variously as crystal orientation mapping (COMTM) or the individual peaks can be readily indexed with the
orientation imaging microscopy (OIMTM) (Wright et help of Bragg’s law.
al. 1993). The terms ‘‘ automated crystal orientation A complete x-ray texture goniometer system is
mapping,’’ ‘‘ orientation scanning microscopy,’’ and composed of an x-ray generator and x-ray tube, a
‘‘ crystal orientation imaging ’’ are other versions goniometer with a sample stage, and a detection
that have been less frequently adopted (Randle and system. Electronics and computer control of the
Engler 2000). Here it will be referred to succinctly as rotations and to read the recorded intensities are also
‘‘ orientation mapping.’’ required. Fully functional texture systems are readily
available commercially. To derive the texture of a
given volume, the orientation, with respect to an
external reference frame, of a set of lattice planes (hkl )
2. Experimental Methods for Texture Measurement has to be determined. The crystal is irradiated with
Two experimental techniques, x-ray and neutron monochromatic radiation at the proper Bragg angle
diffraction, are used for macrotexture measurements for reflection at the lattice planes (hkl ), and the
(Sect. 2.1) whereas the others are associated with detector is set at the angle 2θ with respect to the
microtexture. Although x-ray diffraction is used primary beam. Of course, a reflected intensity is only
mainly for macrotexture, it can in principle also be measured if the corresponding lattice planes are
used for microtexture measurements either if the grain arranged such that they lie parallel to the sample
size is of the order of millimeters (Laue technique) or surface, i.e., their normal is the bisector of the angle
if a sufficiently fine x-ray beam irradiates volumes less between incident and reflected beam. In order to
than the grain size. X rays generated in a synchrotron ensure reflection from other lattice planes, the sample
offer this possibility since they are characterized by an has to be rotated and\or tilted until the lattice planes
intensity several orders of magnitude higher than are in reflection condition, i.e., parallel to the sample
those generated by conventional x-ray tubes. The surface. The necessary rotation and tilt angles are a
application of synchrotron radiation in texture re- measure for the arrangement of the lattice planes
search is a new, promising field (Garbe et al. 1997). within the crystal, which means they are characteristic
Selected area channeling (SAC) in SEM is being of the orientation of the crystal with respect to the
largely superseded by EBSD for microtexture external sample frame.
measurements (Sect. 2.2) and so will not be discussed In order to determine an unknown crystal orien-
further here. TEM still has some applications for tation in practical applications, the sample is sys-
certain microstructures (Sect. 2.3), such as those tematically rotated in a texture goniometer about well-
associated with highly deformed structures. A few defined angles in such a way that all possible lattice
peripheral, nondiffraction microtexture techniques planes are successively brought into the reflection
such as acoustic, ultrasonic, magnetic, or optical condition and the reflected intensities are recorded as
methods also exist for special applications (Randle a function of these rotation angles. The rotation angles
and Engler 2000). are directly related to the pole figure angles α (radial)
and β (azimuthal), so that the reflected intensities can
be represented directly in a pole figure. The intensity
recorded at a certain sample orientation is directly
2.1 Macrotexture Methods: X-ray Diffraction and proportional to the volume fraction of grains which
Neutron Diffraction currently is in reflection condition. Thus, the pole
figure of a polycrystalline sample reflects the dis-
The principle of macrotexture pole figure measure- tribution of orientations in that sample, i.e., its texture.
ment is based on Bragg’s law for diffraction of Because of the much easier access to x-ray gonio-
radiation at the crystal lattice planes: meters than to neutron sources, the vast majority of
pole figure measurements is performed by means
nλ l 2d sin θ (1) of x-ray diffraction. Neutron diffraction offers some
advantages, however, which are mainly a result of the
where n is the order of the reflection, λ is the much lower absorption of neutrons by matter (Bacon
wavelength, d is the interplanar spacing and θ is the 1975). This allows investigation of centimeter-sized
‘‘ Bragg angle,’’ the angle of reflection. As each set of specimens, which yields the opportunity to perform
lattice planes has a different lattice spacing d, re- nondestructive texture measurements or in situ experi-
flections from various sets of lattice planes can be ments. Other advantages include the fact that that the
distinguished by setting the detector to the corre- grain statistics are better, which is of particular interest
sponding angle 2θ with respect to the incident radi- in coarse-grained samples or for detection of low
ation. A spectrum can then be obtained by recording volume fractions of a second phase.

2
 Texture

and store the result at the rate of 5 sV" without any

2.2 Microtexture Methods: EBSD
operator intervention. Figure 1 shows the components
An EBSD system can be added as an attachment to of an EBSD system.
practically any SEM system, including a field emission The average spatial resolution and accuracy of
gun (FEG) SEM (Humphreys et al. 1999). EBSD EBSD are "200–500 nm and "1m, respectively. These
relies on positioning the specimen within the SEM parameters are influenced by the material, the
sample chamber such that a small angle, typically 20m, specimen\microscope geometry, accelerating voltage,
is made between the incident electron beam and the probe current, and pattern clarity. The fundamental
specimen surface. This simple expedient of tilting the factor that governs the absolute spatial resolution of
specimen reduces the path length of electrons which EBSD is the interaction volume of the electron beam
have been backscattered by lattice planes as they enter in the specimen. Because of the very thin, of the order
the specimen, thus allowing a far greater proportion of of nanometers, surface layer contributing to the EBSD
these electrons to undergo diffraction than if the pattern, the interaction volume approximates to the
specimen were flat (Venables and Harland 1973, area of the incident beam, i.e., the spot size. There is a
Dingley 1981). The resulting diffraction pattern can be linear relationship between accelerating voltage and
captured and interrogated in real time, and computer interaction volume for a specific element. Hence a low
algorithms allow the orientation of each diffraction accelerating voltage is chosen if good spatial resolution
pattern to be obtained and stored, from which raw is required. Although the spatial resolution is less
data a microtexture is constructed serially. A state-of- sensitive to probe current than to accelerating voltage,
the-art EBSD system can position a stationary probe if it is convenient to work at a particular accelerating
on the specimen, capture a diffraction pattern, index it, voltage then the probe current can be used to ma-

Figure 1
Components of an EBSD system (courtesy of O. Engler).

3
Texture

Figure 2
Example of an orientation map (originally in color) of grains in a microstructure, generated by EBSD. Colors\gray tone
correspond to orientations.

nipulate the resolution. The clarity of the diffraction establish the scale of the orientations and how it
pattern affects the precision with which Kikuchi lines relates to the microstructure, e.g., the grain size
can be located by the pattern-solving algorithm, which distribution, will clarify the requirements. For
in turn can degrade the accuracy of the measurement. example, in order to obtain nearest-neighbor grain
The three main SEM parameters that can be misorientations, the orientations must be sampled
altered in the mechanical set-up for EBSD are the with a step size that is less than the grain size. It may
specimen-to-screen distance, the specimen tilt, and the also be necessary to establish what phases are present
specimen height (‘‘ working distance ’’) in the micro- in the specimen. The microtexture of multi-phase
scope. Values for all these parameters are necessary to materials can be analyzed using EBSD (Engler et al.
calibrate the system and hence to index the diffraction 1995), and progress has been made in automated
pattern, which for most materials is performed auto- operation and mapping of them.
matically by the software. The specimen-to-screen Orientation mapping using EBSD, such as the map
distance is usually fixed by the microscope geometry. shown in Fig. 2 (which was originally in color), has
A tilt angle of 70m represents a good compromise with opened up many new possibilities for investigations of
regard to convenience and pattern contrast, and materials. The visualization aspect of orientation
therefore is usually used as a standard angle for EBSD. mapping is without doubt its most compelling and
For EBSD the major consideration when choosing the novel feature. A less conspicuous, but equally signi-
working distance is to locate the specimen so that ficant, merit of orientation mapping is that it embodies
electrons are backscattered from the specimen towards total quantification of the orientation aspects of
the camera, while avoiding risk of collision with microstructure, which complements the visual ap-
microscope hardware. The specimen position will praisal aspects. Hence texture or grain boundary
depend therefore on the geometry of the microscope, statistics can be accessed from a map and represented
and a typical working distance is in the range by standard methods. Orientation mapping offers the
15–25 mm. ability to compact into one dataset measurements
The orientation sampling schedule can be per- that had previously been obtained by disparate
formed either manually or automatically. In general methods (e.g., grain structure, texture distributions,
some preliminary exploration in manual mode to and grain boundary types), so that the need for

4
 Texture

elaborate correlation techniques between various 3. Data Representation

separately determined parameters is obviated. The
Because of their individual and spatially related
morphology of grains is depicted using orientation
nature, there are more data representation options for
mapping, which allows a direct measure of the true
microtexture than for macrotexture. In this section the
grain size and distribution on the particular section
basic theoretical aspects pertinent to both will be
through the microstructure. The drawbacks of orien-
addressed first, and then representation methods for
tation mapping for texture work are that it requires
the two branches will be separately considered.
much SEM system time and large amounts of
redundant data are often generated.

3.1 Descriptors of Orientation

2.3 Microtexture Measurements: TEM
An orientation is defined as ‘‘ the position of the
The main advantage of TEM-based techniques for crystal coordinate system with respect to the specimen
texture measurement is their excellent spatial res- coordinate system ’’. Specification of an orientation
olution, which is at least one order of magnitude better requires three independent variables (because it is
than for SEM. Evaluation of diffraction patterns three-dimensional) and, practically, these can be
obtained by means of convergent beam electron expressed in several different ways, which will be
diffraction (CBED) is advantageous compared to the discussed below. The fundamental mathematical
more traditional selected area diffraction (SAD) tech- orientation descriptor is a matrix which embodies the
nique, as the former yields higher spatial resolution as rotation of the specimen coordinates onto the crystal
well as much higher angular accuracy. However, SAD coordinates (Bunge 1987).
is still of some use since it allows direct measurement The orientation matrix is a square matrix of nine
of pole figures of small volumes in TEM. In general, numbers and is obtained as follows. The first row of
TEM is relatively little used for mainstream texture the matrix is given by the cosines of the angles α , β ,
determination and is underdeveloped compared to γ between the first crystal axis, [100], and each of" the
"
EBSD. "
three specimen axes X, Y, Z, in turn, where X, Y, Z

Figure 3
Illustration of the relationship between crystal axes and specimen (reference) axes, which in turn specify the orientation
matrix (courtesy of K. Dicks).

5
Texture

and last column of the orientation matrix by nearest,

or ‘‘ idealized ’’, Miller indices. For the above example
this would be written as o123qf634g. This shorthand
notation is very popular in metallurgy.
The projected poles descriptor leads to the most
popular method for texture representation, the pole
figure or inverse pole figure. The pole figure is a
stereographic projection (an equal-area projection can
also be used) of poles of plane families from each
orientation, where the reference axes of the stereogram
correspond to the macroscopic specimen geometry
(Fig. 4). The position of the pole on the projection is
given by two angles: α, which describes the azimuth of
the pole, and β, which characterizes the rotation of the
pole around the polar axis, as shown in Fig. 4. Rather
than representing the orientation of the crystal co-
ordinate system in the specimen coordinate system,
i.e., in a pole figure, the orientation of the specimen
coordinate system can be projected onto the crystal
coordinate system. This representation is called the
inverse pole figure. Inverse pole figures are often used
for axial symmetric specimens, where only one of the
Figure 4 axes is prescribed. According to the crystal symmetry
Definition of the pole figure angles α and β (courtesy of it is not necessary to show the entire pole figure, but
H. Davies). one stereographic unit triangle suffices.
The Euler angles refer to three rotations which,
relate to well-defined directions in the specimen such when performed in the correct sequence, transform the
as those associated with the processing geometry specimen coordinate system onto the crystal coor-
(Fig. 3). The second row of the orientation matrix is dinate system—in other words specify the orientation.
given by the cosines of the angles α , β , γ , which are The rotations are: } about the normal direction ND,
# # # the third
between [010] and X, Y, Z in turn. Similarly, "
transforming the transverse direction TD into TDh and
row of the matrix comprises the cosines of the angles the rolling direction RD into RDh; Φ about the axis
α , β , γ between [001] and X, Y, Z. Hence the RDh (in its new orientation); and } about NDd (in its
$ $ $matrix g, including an example, is # the Euler angles
complete new orientation), where } , Φ, } are
" #
(Bunge 1987). The effect of the sequence of these three
gl rotations can be followed in Fig. 5. An orientation
cosα cos β cos γ g g g expressed in terms of its Euler angles can be displayed
" " " "" "# "$
cosα cos β cos γ lg g g as a point in a three-dimensional coordinate system
# # # #" ## #$ whose axes are given by the three Euler angles. This is
cosα cos β cos γ g g g
$ $ $ $" $# $$ called Euler space.
0n768 k0n582 0.267 An orientation can also be the rotation of the crystal
coordinate system through a single angle about a
l 0.384 0.753 0.535 specific axis, sometimes known as the angle\axis pair.
k0n512 k0n308 0n802 The main use of this descriptor of orientation is that if
(2) the reference orientation is another orientation, rather
than the specimen coordinate system, the ‘‘ mis-
The orientation matrix allows a crystal direction to be orientation ’’ between the two grains is obtained. This
expressed in terms of the specimen direction to which is then used in grain boundary texture work. The
it is parallel, and vice versa. Since an orientation needs angle\axis pair is also the starting point for for-
only three independent variables to specify it, the mulation of the Rodrigues vector, another descriptor
matrix, having nine numbers, clearly contains non- of orientation. The Rodrigues vector R combines the
independent elements. The pertinence of the orienta- angle θ and axis of rotation r having components r r r
tion matrix is that it is the mathematical tool for into one mathematical entity: "#$
calculation of other descriptors of orientation. These
descriptors are the ideal Miller or Miller Bravais R l tan(θ \2)r (3)
indices, projected poles, Euler angles, angle\axis pair, then
and Rodrigues vector. All these descriptors, apart
from the first, lead to graphical depiction. R l tan(θ \2)r R l tan(θ \2)r R l tan(θ \2)r
The ideal Miller indices notation expresses the first " " # # $ (4)$

6
 Texture

which is the volume used for data display (Randle and

z Day 1993).

3.2 Representation of Macrotextures

The direct output from macrotexture measurements is
the reflected intensity in the form of the two pole figure
angles α and β, i.e., the distribution of poles in the
traditional pole figure. Hence the pole figure is the
x obvious choice for a critical assessment of experi-
mental macrotexture results. Since the raw data arise
as intensities of reflections from particular planes, in
order to compare pole figures from different samples it
is necessary to normalize these orientation densities
to standard units that are not dependent on the
experimental set-up.
y
The commonly used convention is to express the
data in multiples of a random or uniform distribution
Figure 5 of orientations, which means that the pole figure of a
Definition of the Euler angles } , Φ, } (courtesy of randomly oriented standard sample would be ‘‘i1 ’’
K. Dicks). " # at all points. Pole figure regions with higher intensities
indicate that more lattice planes are aligned in those
directions than in a sample with random texture, and
vice versa. The expression ‘‘ random texture ’’ therefore
means ‘‘ no texture,’’ i.e., no preferred orientation.
The most commonly used way of contouring the
orientation density in a pole figure is to draw lines of
equal intensity, isodensity lines. Inverse pole figures
are similarly represented.
A macrotexture pole figure comprises in fact only
two-dimensional information, whereas a complete
orientation requires three independent variables. If
required, full macrotexture data need to be calculated
from a set of several pole figures representing different
planes from the same specimen. This is known as
‘‘ pole figure inversion,’’ and can be achieved by several
complex mathematical techniques (Randle and Engler
1998). The result is an ‘‘ orientation distribution
function,’’ ODF, which is displayed almost always in
Euler space. Similarly to pole figures, the data are
expressed directly as density distributions. For con-
venience on the printed page, the three-dimensional
Euler space is shown as a series of sections (Fig. 7).
Figure 6
The fundamental zone of Rodrigues–Frank space for cubic
materials. 3.3 Representation of Microtextures
Both the discrete nature of microtexture data and their
An R-vector, which can represent either an orientation direct link to the microstructure introduce a much
or a misorientation, is displayed in a three-dimensional broader and more complex scenario for data evalu-
space known as Rodrigues–Frank (RF) space or ation and representation than that for macrotextures.
Rodrigues space (Frank 1988). The axes of The nature of microtexture gives rise to two fun-
Rodrigues–Frank space can be chosen to be either the damental lines of inquiry: statistical distribution of
specimen axes (for representation of orientations) orientation and misorientation data, and spatial dis-
or the crystals axes (for representation of mis- tribution of these data, related to the microstructure.
orientations) (see Grain Boundary Geometry: For most investigations both these analysis steps will
Measurement). Figure 6 shows the ‘‘ fundamental be required, and the data will be selected from
zone ’’ of Rodrigues–Frank space for cubic crystals, relatively small regions of interest. Analysis of mis-

7
Texture

Figure 7
Example of an ODF in Euler space for a cubic material (copper). Data are shown as (a) discrete points (obtained by
EBSD) and (b) contours (obtained by x rays) (courtesy of O. Engler).

orientation data is a very important subset of micro- of 10 mmi10 mm and an average penetration depth
texture (Randle 1993) (see Grain Boundary Geometry: of about 0n05 mm, an x-ray macrotexture of a sample
Measurement). with an average grain size of 25 µm represents the
For statistical distribution of orientation and mis- orientation information of about 3i10& grains. In the
orientation data the chief aim is only to represent and case of neutron diffraction this number is even higher
quantify the orientation distribution in a region of by two orders of magnitude, which is due to the much
interest in the microstructure, either by sampling every higher penetration depth of the order of 10 mm. In
orientation present or by statistical sampling through- contrast, the number of grains sampled by EBSD is
out a large region of the specimen. If sampling points usually an order of magnitude smaller than that for x
do not represent every grain, then there is no access rays. The appropriateness of EBSD for overall texture
to grain boundary misorientation or microstructure estimation depends on the ability to assess how
information—in fact the data are equivalent to an many orientation measurements are required and the
‘‘ average ’’ texture, i.e., the microtexture counterpart formulation of a representative sampling schedule.
to a macrotexture. Sometimes, since automated EBSD With regard to general evaluation of statistical
is very rapid and is able to measure directly the ‘‘ true ’’ microtexture data in the form of either orientations or
grain orientation, the statistical analysis mode is used misorientations, it is usual to display first the popu-
advantageously in place of a traditional macrotexture lation using one of the methods described in Sect. 3.1.
technique to generate a measure of the overall texture. Secondary processing steps can then be applied to
However, it is of prime importance to consider both extract quantified information appropriate to the
the size and distribution of the sample population in investigation, e.g., what proportion of specified orien-
order to obtain a statistically reliable result since the tations within a certain angular tolerance are present,
number of grains encountered is usually smaller than either by number or by projected grain area. Many of
the number of grains contributing to x-ray or neutron these processing steps are available as part of com-
pole figure measurements. Assuming a sampling area mercial software packages or alternatively the user can

8
 Texture

Figure 8
Example of a f111g microtexture pole figure from a cubic material (courtesy of R. Davies).

design customized options. Some of the representation soon become unacceptably crowded and lucidity is
options are those traditionally used for macrotexture, lost. Instead the data can be shown as density
e.g., the pole figure\inverse pole figure and Euler space distribution contours.
(Fig. 7), whereas other options such as Rodrigues The most valuable aspect of microtexture is the
space and misorientation evaluation derive directly direct link that it provides between orientation and
from microtexture data. The fundamental differences microstructure. Hence consideration of the statistical
between data collection for microtexture and orientation distribution of the sampled region(s) (or
macrotexture feed through to necessitate different subsets of them) is usually only part of the total
approaches for data handling in both cases. microtexture investigation. Typically, orientation data
A ‘‘ raw ’’ microtexture population comprises dis- collection is accompanied by a record of sampling
crete individual orientations, which are usually dis- coordinates in the microstructure, and the location of
played as such (Fig. 8). When the number of data these individual data points is marked on the rep-
points is very large (which is often the case for resentation, which is usually a pole figure. A similar
mapping) and\or the texture is very sharp, it becomes strategy has traditionally been common practice in
inconvenient to display the measurements—in any TEM. Clearly, this method is only viable for relatively
space—individually because the representation will small data subsets since the correlation between

9
Texture

measuring site and orientation is lost if the distribution 5. Concluding Remarks

has to be expressed as density contours. The ultimate
The choice of both the experimental technique used to
refinement of texture\location correlation is the orien-
obtain texture information and the method used to
tation map where each pixel in the map has both
represent it will depend ultimately on factors such as
orientation and location associated with it. However,
what equipment is available and the nature of the
a map as such is a visual entity; secondary processing
information sought. X-ray goniometry and EBSD are
is necessary to extract quantitative information.
by far the most popular choices for macrotexture and
microtexture analysis, respectively. If it is feasible,
benefits can often derive from using more than one
4. Applications of Microtexture technique. There is much active development in the
field of microtexture, especially in refinements to the
Microtexture allows access to a whole new stratum of
speed and resolution of EBSD.
knowledge, since the texture and morphological
aspects of microstructure and properties become
fused. The following phenomena can be explored: See also: Thin Films : Texture Effects on Mechanical
Local property effects. Microtexture allows the Properties
study of properties in specific regions of the micro-
structure, e.g., in clusters of grains or adjacent to a
physical feature such as a surface or a crack. Bibliography
Morphological\geometrical grain parameters. Corre-
Bacon G E 1975 Neutron Diffraction. Clarendon Press, Oxford
lations can be made between the size, shape, distri- Bunge H-J 1987 Three-dimensional texture analysis. Int. Mater.
bution, and location of grains with respect to their ReŠ. 32, 265–91
orientations. A simple example is that large grains Dingley D J 1981 A comparison of diffraction techniques for
may have a different texture to small grains. Orien- scanning electron microscopy. Scanning Electron Microsc. 4,
tation mapping can also be used to give true grain 273–86
size and distribution measurements. Engler O, Hirsch J, Lu$ cke K 1995 Texture development in
Direct ODF measurement. The traditional route for dual—phase alpha\beta brass-part II. Recrystallization
ODF measurement relies on calculation from several textures. Z. Metallkd. 86, 475–84
pole figures which have been determined by macro- Frank F C 1988 Orientation mapping. Metall. Trans. 19A,
403–8
texture methods. These procedures admit inherent Garbe S, Poulsen H F, Jensen D J 1997 Local texture analysis
inaccuracies which are obviated when a microtexture using high energy synchrotron radiation. In: Liang Z et al.
technique is used to obtain the orientation distri- (eds.) Proc. 11th Int. Conf. Textures of Materials. Inter-
bution. For this case full grain orientations are national Academic Publishers, China pp. 100–5
obtained directly rather than calculated from crystal Humphreys F J, Huang Y, Brough I, Harris C 1999 Quantitative
plane distributions as in the x-ray case. Hence ODFs metallography by electron back-scattered diffraction.
obtained by microtexture measurements give the true J. Microsc. 195, 1–10
orientation distribution of the sampled grains. Randle V 1992 Microtexture Determination and its Applications.
Orientation Šariations within indiŠidual grains. Institute of Materials, London
Randle V 1993 Grain Boundary Geometry in Polycrystals.
Although by classical definition a grain has a single, Institute of Physics Publishing, Bristol, UK
distinct orientation, in practice the orientation within Randle V, Day A 1993 Use of Rodrigues–Frank space for
a grain may vary in a continuous or discontinuous representation of microtexture and grain boundary para-
manner, particularly close to interfaces. A micro- meters. Mater. Sci. Tech. 9, 1069–78
texture technique is able to characterize these fine- Randle V, Engler O 2000 Macrotexture, Microtexture and
scale textures on a local level. Orientation Mapping. Gordon and Breach, UK
Phase relationships. Multiphase materials can be Venables J A, Harland C J 1973 Electron back-scattering
studied to produce concurrently the texture of each patterns—a new technique for obtaining crystallographic
phase, the distribution of its orientations, and orien- information in the scanning electron microscope. Philos. Mag.
27, 1193–200
tation relationships between individual crystals of each
Wilkinson A J, Hirsch P B 1997 Electron diffraction-based
phase. techniques in the scanning electron microscopy of bulk
Interfacial parameters and properties. Microtexture materials. Micron 28, 279–308
gives rise to a ‘‘ texture between grains,’’ which is also Wright S I, Adams B L, Kunze K 1993 Application of a new
sometimes referred to as ‘‘ mesotexture.’’ This in turn automatic lattice orientation measurement technique to poly-
gives access to studies of links between inter- crystalline aluminium. Mater. Sci. Eng. A160, 229–40
facial structure and properties (see Grain Boundary
Geometry: Measurement). V. Randle

10
 Texture

Copyright ' 2001 Elsevier Science Ltd.

All rights reserved. No part of this publication may be reproduced, stored in any retrieval system or transmitted
in any form or by any means : electronic, electrostatic, magnetic tape, mechanical, photocopying, recording or
otherwise, without permission in writing from the publishers.
Encyclopedia of Materials : Science and Technology
ISBN: 0-08-0431526
pp. 9119–9129

11