Compressive performance and crack propagation in Al alloy/ Ti2AlC composites

Full text

Authors’ Version
Compressive performance and crack propagation in Al alloy/
Ti2AlC composites
D.A.H. Hanaor1 *, L. Hu2, W.H. Kan1, G. Proust1, M. Foley3, I. Karaman2, M. Radovic2
1 School of Civil Engineering, University of Sydney, Sydney, NSW 2006, Australia
2 Department of Materials Science and Engineering, Texas A&M University, College Station, TX 77843, USA
3Australian Centre for Microscopy and Microanalysis, University of Sydney, Sydney, NSW 2006, Australia
* Corresponding author: dorian.hanaor@sydney.edu.au
Abstract:
Composite materials comprising a porous Ti2AlC matrix and Al 6061 alloy were fabricated by a
current-activated pressure assisted melt infiltration process. Coarse, medium and fine meso-structures
were prepared with Al alloy filled pores of differing sizes. Materials were subjected to uniaxial
compressive loading up to stresses of 668 MPa, leading to the failure of specimens through crack
propagation in both phases. As-fabricated and post-failure specimens were analysed by X-ray
microscopy and electron microscopy. Quasi-static mechanical testing results revealed that compressive
strength was the highest in the fine structured composite materials. While the coarse structured
specimens exhibited a compressive strength of 80% relative to this. Reconstructed micro-scale X-ray
tomography data revealed different crack propagation mechanisms. Large planar shear cracks
propagated throughout the fine structured materials while the coarser specimens exhibited networks of
branching cracks propagating preferentially along Al alloy-Ti2AlC phase interfaces and through
shrinkage pores in the Al alloy phase. Results suggest that control of porosity, compensation for Al
alloy shrinkage and enhancement of the Al alloy-Ti2AlC phase interfaces are key considerations in the
design of high performance metal/Ti2AlC phase composites.
Keywords: MAX phase; Ti2AlC; XRM; Tomography, Crack propagation

Authors’ Version
1. Introduction
MAX phases are a family of ternary carbides
and nitrides with the formula Mn+1AXn, where
M is a transition metal, A is a group III or IV
metal, X is either nitrogen or carbon and n =1,
2 or 3 [1]. This family of materials first came to
attention through the work of Nowotny and his
co-workers in the 1960s [2-5] and became the
subject of renewed interests after 1996 with a
study on synthesis and the unusual mechanical
properties of bulk Ti3SiC2 [6]. Following this,
the scope of this family of materials rapidly
expanded, with a growing range of systems and
compositions being studied. MAX phase
ceramics are of great interest owing to their
unusual combination of properties including
high thermal and electronic conductivity,
oxidation resistance, good machinability,
damage tolerance, and thermal shock resilience,
among others.
Among the numerous MAX phases, Ti2AlC has
drawn attention and has been the subject of
several investigations [7-11]. This material has
been shown to exhibit beneficial properties in
terms of machinability, electric conductivity
and fracture toughness [9, 12-14]. Of particular
interest is the crack healing ability observed in
Ti2AlC systems through the formation of a well
adhered alumina phase in heat treatment cycles
[15-17]. Following the typical structure of
materials in the M2AX subset of the MAX
phase materials (also termed the 211 group),
Ti2AlC comprises molecular layers of titanium
carbide with every third layer consisting of pure
aluminium, and belongs to the space group
P63/mmc [18]. These layers of ductile phase
facilitate nonlinear kinking behaviour and
plastic deformation, believed to occur through
basal plane slip.
An interesting combination of metallic- and
ceramic-like properties of MAX phases have
motivated the development of metal/MAX
phase composites [1, 19, 20]. A composite of
Cu and Ti3SiC2 was proposed as a new electro-
friction material [21], and MAX phases have
further been shown to increase the mechanical
strength of several metallic systems while
maintaining good thermal and electrical
conductivity [22-24]. In addition, the yield
strength of a Al/Ti3AlC2 composite was found
to be twice that of Al [25] while the
mechanical energy dissipation was found to be
significantly improved in Mg/Ti2AlC [26] and
NiTi/Ti3SiC2 composites [27] when compared
to its pure constituents.
Metal/MAX phase composites are frequently
processed by powder co-sintering methods.
One of the principal challenges in this approach
stems from the reaction between the metallic
and MAX phases that limits the use of high
temperature processes. In powder co-sintering,
temperatures are usually chosen just below the
melting point of the metal phase, which is
generally insufficient for the sintering of the
MAX phase material [25]. In order to obtain
metal/MAX phase composites with marginal
inter-phase reactions, molten metals were
infiltrated into MAX phase foams in a
pressureless infiltration technique to fabricate
Ti2AlC/Mg composites exhibiting higher
strength and mechanical energy dissipation
than other Mg composites [28, 29]. However,
such pressureless infiltration into foams is
encumbered by poor wettability of MAX phase
foams by some molten metals, which inhibits
adequate metal infiltration. The observed poor
wettability may yield weak bonding between
metal and ceramic phases, resulting in inferior
mechanical properties [30]. This problem can
be overcome by using pressure infiltration to
force molten metals into ceramic foams.
However, in many cases the reaction of the
molten metal with the MAX phase material is
sufficiently rapid that even in such pressure-

Hanaor, D. A. H., L. Hu, W. H. Kan, G. Proust, M. Foley, I. Karaman, and M. Radovic. "Compressive performance and
crack propagation in Al alloy/Ti2AlC composites." Materials Science and Engineering: A 672 (2016): 247-256.
3
driven infiltration methods new phases are
likely to form. The new phases not only cage
the pores to prevent further infiltration, but also
degrade the constituents of the composites.
Minimising the extent of such reactions remains
a significant challenge towards the fabrication
of metal/MAX phase composites.
Aluminium alloys are attractive in ceramic-
metal composites for aerospace and
transportation applications, where weight
saving and thermal stability are important
considerations. Thus Al and its alloys have
been combined with ceramic phases of Al2O3
[31-33], B4C [30, 34, 35], and SiC [36, 37], in
composite materials. However, MAX phases
had not been used in Al-based composites until
recent studies on composites of aluminium
alloys with and Ti3AlC2, Ti2AlC and V2AlC [12,
38, 39] . The use of MAX phases in Al-based
composites has several additional advantages
relative to traditional ceramic components, e.g.
Al2O3, B4C, or SiC. Typical MAX phases
exhibit a higher fracture toughness (e.g. ~7
MPa·m1/2 for Ti3SiC2,) than Al2O3 (~4
MPa·m1/2), B4C (~3.7 MPa·m1/2), and SiC (~4.6
MPa·m1/2). Furthermore, unlike traditional
ceramics, MAX phases exhibit high thermal
and electronic conductivities originating from
atomic bonding with mixed covalent, ionic, and
metallic characteristics, allowing for more
versatile applications.
It has been shown that a current-activated,
pressure-assisted infiltration (CAPAI) is a
viable method for producing interpenetrating
Al alloy/MAX phase composites [40]. One of
the attributes of this method is that it facilitates
the fabrication of composite materials that
could not otherwise be fabricated using
conventional methods due to poor wettability
and interphase reactions [27]. Future
development of composites fabricated by such
means necessitates an improved understanding
of multi-scale structural-mechanical property
relationships in these materials. Here we
present the first report of structure, mechanical
performance and crack propagation in Al
alloy/Ti2AlC composite systems fabricated
with different meso-structures. We employ
micro-scale X-ray tomography (also known as
X-ray microscopy) to gain meaningful insights
into the deformation and failure of these
materials under compressive loads.
2. Materials and Methods
2.1. Ti2AlC foam
In order to fabricate Ti2AlC foams in three
distinct mesostructures, an appropriate
precursor MAX phase powder was used in
conjunction with sodium chloride pore formers
following reported protocols [38, 41]. In this
process Ti2AlC powder (Maxthal 211, Sandvik,
Sweden) with a particle size in the range 45–90
m, and three types of NaCl powders (Sigma-
Aldrich, USA), with particle size distributions
in the ranges 45–90 m, 180–250 m or 355–
500 m, were employed. The fabrication of
foams was conducted in three main steps: (i) a
mixture of the NaCl pore former (either coarse,
medium or fine) and the Ti2AlC powder was
blended in a 40/60 volume ratio by ball milling
and then pressed in a cylindrical die of 12.7mm
diameter at 800MPa; (ii) the NaCl pore former
was dissolved in distilled water by soaking
overnight, and (iii) the porous green Ti2AlC
body was sintered under flowing argon at
1400oC for 4 hours. Pore sizes in the foams
were determined by measuring the size of 50
pores in SEM images using the intercept
method, as specified in ASTM E112-13 [42],
from four SEM images in randomly selected
locations on each sample.
2.2. Composite fabrication
To prepare composite Al/Ti2AlC specimens, Al
alloy 6061 discs (McMaster-Carr, GA, USA)
with a diameter of 20 mm and a thickness of 4
mm were used in an infiltration process. In this
process the Ti2AlC foams with different pores
sizes were “sandwiched” in between two Al
alloy discs and placed in a graphite die, with

Hanaor, D. A. H., L. Hu, W. H. Kan, G. Proust, M. Foley, I. Karaman, and M. Radovic. "Compressive performance and
crack propagation in Al alloy/Ti2AlC composites." Materials Science and Engineering: A 672 (2016): 247-256.
4
graphite foils separating the discs from the die.
This “sandwich” set-up facilitates a more
uniform infiltration of molten metal. Infiltration
was carried out using a spark plasma sintering
system (SPS 25-10, GT Advanced
Technologies, CA, USA). In this system, the
chamber was evacuated and held at 10-6 torr for
10 minutes before heating. A direct current was
pulsed at 10 ms intervals from 0 to 1250 A over
4 min to give a heating rate of 200 °C/minute
before stabilizing at a current of 860 A to give
a 1 min soak at 750 °C. The complete
infiltration process including heating/melting,
soaking, and cooling/solidification was carried
out over 10 minutes. The temperature was
calibrated and measured using procedures
described elsewhere [27].
By employing pore formers with three different
size distributions in the synthesis of Ti2AlC
foams, we examined three distinct meso-
structures of Al alloy/Ti2AlC composites,
referred to as fine, medium and coarse,
exhibiting Al alloy phase segments of different
sizes. The term meso-structure is used as it
exists at a level between the microstructure of
the individual material constituents and the bulk
composite macrostructure. Fine, medium and
coarse composite samples were denoted by
suffixes F-,M- and C- respectively, in this
paper.
2.3. Characterization
Density and porosity (both open and closed) of
all samples were determined by an alcohol
immersion method, as specified in ASTM C20-
00 [43]. Theoretical density values of 4.11
g/cm3 and 2.70 g/cm3 [44, 45] for Ti2AlC and
Al alloy, respectively, were used to calculate a
theoretical density of 3.55 g/cm3 for composites
containing 40 vol% Al, using the rule of
mixture. Actual material densities are shown in
Table 1.
To assess the influence of structure on the
mechanical performance and failure of the
composite materials, the deformation behaviour
of materials in compression was tested. The
compressive stress-strain curves of the
specimens under cyclic loading were obtained
using an MTS 810 (MTS Systems, MN, USA)
servo hydraulic test frame at a strain rate of 7

10-4 s-1. All specimens for compressive testing
were cut by electrical discharge machining to
dimensions of 3.5 mm × 3.5 mm × 7 mm. All
specimens were machined to have flat and
parallel ends within ± 25 μm. As-processed and
post-compression samples are denoted by the
suffixes AP and PC, respectively.
Preliminary microstructural analysis of
specimens was carried out using a Hitachi TM
3030 scanning electron microscope (SEM),
with an accelerating voltage of 15 kV and a
working distance of approximately 7 mm. The
distribution of the different phases and the
morphology of cracks in PC specimens were
examined using this method. Electron
backscatter diffraction (EBSD) analysis was
further conducted to determine the phases
present. EBSD scans were carried out using a
Zeiss Ultra field emission gun SEM equipped
with Oxford Instruments AZtec integrated EDS
and EBSD system, with X-Max 20 mm2 silicon
drift EDS detector and Nordlys-nano EBSD
detector. The scans were done with a step size
varying from 0.2 to 0.5 µm depending on the
specimen (0.2 µm for the M-AP specimen, 0.3
µm for C-AP and 0.5 µm for F-AP). All the
scans were done with an accelerating voltage of
20 keV. Data acquisition and analysis were
done using Oxford Instrument's AZTec HKL
and HKL Tango software.
To gain reliable insights into failure
mechanisms in Al alloy/Ti2AlC composites, it
is necessary to capture 3D microstructural
information from as-processed and post-
compression specimens. As cutting cross
sections from specimens may result in further
cracks, deformation and/or loss of material, it is
preferable to conduct non-destructive 3D
structural reconstruction. To achieve this,
micro-scale X-ray computed tomography was
carried out using a Zeiss MicroXCT-400.
Micro-scale X-ray computed tomography, also

Hanaor, D. A. H., L. Hu, W. H. Kan, G. Proust, M. Foley, I. Karaman, and M. Radovic. "Compressive performance and
crack propagation in Al alloy/Ti2AlC composites." Materials Science and Engineering: A 672 (2016): 247-256.
5
known as μ-CT, as carried out here constitutes
a type of X-ray microscopy (XRM) as it uses
optical elements in conjunction with a
scintillator to acquire X-ray data at higher
magnifications and spatial resolution [46, 47].
Several XRM scans with micrometre scale
resolution were facilitated using a 150 kV, 10W
X-ray beam. Specimens were rotated 360° in
the sample chamber with 2D X-ray projections
captured at 0.2° intervals with an exposure of 5
seconds per frame. Projections were captured
using both a lens magnification of 4x and 20x
and geometric positioning to result in pixels of
linear size 4.95 and 1.01 μm, respectively. 2D
X-ray projections were reconstructed using
XMReconstructor software v7.0.2817 to yield
isotropic voxels in 3D with the same linear
dimensions. The reconstructed XRM data
contain greyscale data representing material
attenuation of X-rays. These data were
interpreted using Avizo Fire 8 software (FEI
Visualization Sciences Group). In this work we
applied an entropic thresholding algorithm to
quantitatively approximate the relative
proportions of Ti2AlC, Al alloy and air based on
image greyscale histograms [48, 49]. An
illustration of the thresholded phase distribution
is shown in Figure 1. On the basis of 2D
orthogonal slices and 3D reconstructions, the
interaction of cracks with different phases was
evaluated in post compression material.
Volumetric segmentation analysis further
facilitated the assessment of crack behaviour on
the basis of the surface area to volume ratio.
Figure 1. Orthogonal slice from XRM based micro-CT (a) greyscale reconstruction results (b)
three-phase thresholded data.
3. Results
3.1. Specimen properties
Ti2AlC foams with various pore sizes, i.e. 42–
83 m, 77–276 m and 167–545 m, were used
for Al alloy infiltration to prepare Al
alloy/Ti2AlC composites with fine, medium,
and coarse structures, respectively. These
structures were achieved using NaCl particles
as pore formers and serve as precursors to fine,
medium and coarse composite materials. All
three foams exhibited connectivity of Ti2AlC
grains and formation of sintering necks. These
precursors were fabricated using the same
volume percent (40 vol.%) of NaCl particles
and thus have comparable overall porosities of
40.8, 41.6 and 39.9 vol.%. Thus the infiltration
of these three foams with Al alloy resulted in
roughly 40 vol% Al alloy/Ti2AlC composites
with various interpenetrating phase sizes
(Figures 2(a)–(c)).

Hanaor, D. A. H., L. Hu, W. H. Kan, G. Proust, M. Foley, I. Karaman, and M. Radovic. "Compressive performance and
crack propagation in Al alloy/Ti2AlC composites." Materials Science and Engineering: A 672 (2016): 247-256.
6
Table 1. Properties of composite materials with differing meso-structures.
Sample
Interpenetratin
g phase size1,
m
Volume percent of constituents, vol.%
Compressive
strength4,
MPa
Failure
strain4, %
Ti2AlC2
vol%
Al alloy3
vol%
Open pores2
vol%
Closed pores2
vol%
Fine
42–83
60.1 ± 0.9
37.0
0.3 ± 0.2
2.6 ± 0.4
668 ± 28
1.27 ± 0.11
Medium
77–276
58.4 ± 1.2
35.9
0.5 ± 0.4
5.2 ± 0.3
610 ± 30
0.97 ± 0.06
Coarse
167–545
59.2 ± 1.4
34.6
0.7 ± 0.6
5.5 ± 0.4
563 ± 68
0.93 ± 0.04
1Determined from SEM images following ASTM E112-13 [42].
2Measured by alcohol immersion following ASTM C20-00 [43].
3Determined from the balance of Ti2AlC and pores.
4Averaged from 3 test specimens
Table 1 outlines the size of Al phase,
volumetric contents of the constituents (Ti2AlC,
Al alloy), and open and closed porosity in fine,
medium and coarse Al/Ti2AlC composites.
Despite the short processing time, these data
show that more than 97% of the open porosity
in the foams was infiltrated with molten metal.
The volume percent of closed pores in the
composite materials is significantly higher than
open pores. Closed pores most likely originate
from shrinkage voids formed in the alloy during
rapid cooling and pre-existing closed pores in
the Ti2AlC foams prior to infiltration.
EBSD phase maps of the three different
specimen types are given in Figure 2,
illustrating the variation in mesostructure
between the materials. From the presence of
Ti3AlC2 and Al3Ti phases at interface regions it
is evident that some reaction occurs between
the MAX phase and the Al alloy during the
infiltration process. These interface regions
between the two phases are mainly composed
of Al3Ti (shown in yellow) which most likely
arises through Al diffusion from the liquid
phase.
Figure 2: EBSD phase maps of (a) F-AP, (b) M-AP and (C) C-AP. Each colour represents a specific phase:
red – Al; blue – Ti2AlC; Orange – Ti3AlC2; yellow – Al3Ti. All the maps are at the same magnification.
Stress-strain curves of the three Al/Ti2AlC
composite materials tested to failure in
compression are shown in Figure 3. Both
strength and failure strain of the composites
decrease with increasing interpenetrating phase
size. The compressive strength and failure
strain of the fine structured composites are
approximately 20% and 35%, respectively,
higher than those of the coarse/medium
structured composites
.

Hanaor, D. A. H., L. Hu, W. H. Kan, G. Proust, M. Foley, I. Karaman, and M. Radovic. "Compressive performance and
crack propagation in Al alloy/Ti2AlC composites." Materials Science and Engineering: A 672 (2016): 247-256.
7
Figure 3. Compressive stress/strain performance
of Al alloy/Ti2AlC composites with different
meso-structures up to yield point
3.2. SEM analysis
Figure 4(a) shows that the fine-structured
material exhibited a single dominant crack
subsequent to compressive yielding (F-PC),
traversing throughout the entire material bulk.
The diagonal crack plane exhibited the typical
morphology of a shear type failure at roughly
45 ° relative to the loading direction, and
exhibited step-like kinks over its length in
which fracture debris appeared to be present,
shown in Figure 4(b). This debris appeared to
consist of a mixture of both Al alloy and Ti2AlC
phases as evident from the EDS results shown
in Figure 5(a), with MAX phase grains
exhibiting the layered microstructure typical of
this material [12] , shown in Figure 5(b).
Figure 4. SEM micrograph of F-PC material (a) low magnification (b) magnification of crack debris
Figure 5. SEM/EDS analysis of F-PC material (a) EDS spectrum of crack region (b) magnification of
layered Ti2AlC material
In contrast to fine structured material, the
medium and coarse structured materials did not exhibit a single dominant crack. Rather these
materials exhibited a network of smaller cracks

Hanaor, D. A. H., L. Hu, W. H. Kan, G. Proust, M. Foley, I. Karaman, and M. Radovic. "Compressive performance and
crack propagation in Al alloy/Ti2AlC composites." Materials Science and Engineering: A 672 (2016): 247-256.
8
with some, as seen in Figure 6(a), exhibiting
tearing propagating preferentially through the
Al alloy/Ti2AlC interfaces. The tendency of
cracks to propagate around Al regions was
observed in both coarse and medium structured
material as well, shown in Figures 6(b) and
6(c), respectively.
Figure 6. SEM micrographs of post
compression materials (a) Jagged crack in
M-PC (b) Crack interaction with Al in M-PC
(c) Crack interaction with Al in C-PC
3.3. Tomography data
As SEM data only represents material surfaces,
micro-scale X-ray tomography data is analysed
in order to gain a meaningful insight into
material structure and failure behaviour. Axial
2D slices were extracted from reconstructed X-
ray data of all specimens. These ortho-slices,
taken parallel and perpendicular to the loading
direction, are consistent with propagation found
from SEM analysis, showing crack propagation
throughout the volume, rather than just at the
surface, and shed further light on the structure
and compressive yielding of the materials
studied in the present work. The three phases
that can be seen in the X-ray tomography data
are, in order of increasing greyscale (and thus
material attenuation), MAX phase, Al alloy and
air. The latter, which appears as black or near
black in the orthogonal slices shown in Figure
7, is apparent in cracks, shrinkage voids within
Al regions and empty closed MAX phase pores
that were not filled with Al. It was further found
that cracks existed in the coarse precursor
Ti2AlC foams and were filled with Al during
CAPAI. This resulted in elongated Al regions
in as processed coarse structured (C-AP)
materials as seen in Figure 7(d).

Hanaor, D. A. H., L. Hu, W. H. Kan, G. Proust, M. Foley, I. Karaman, and M. Radovic. "Compressive performance and
crack propagation in Al alloy/Ti2AlC composites." Materials Science and Engineering: A 672 (2016): 247-256.
9
Figure 7. XRM slices acquired at using a 4x objective lens, showing view perpendicular to loading
direction (a) F-PC (b) M-CP (c) C-PC (d) C-AP
On the basis of a review of parallel and
transverse slices reviewed from each post-
compression specimen, the interaction of
propagating cracks with the MAX phase and Al
alloy components was assessed. For medium
and coarse structured specimens it was clearly
evident that cracks propagate preferentially
through the more brittle MAX phase rather than
through the more ductile interpenetrating alloy,
as illustrated by figures 6 and 7. Upon
encountering Al filled pores, cracks generally
tend to be deflected and propagate through
MAX phase-Al alloy interfaces.
3.4. Volumetric X-ray analysis
Projections acquired from XRM were used to
reconstruct volumetric representations of the
porous MAX phase matrix constituent in fine-,
medium- and coarse-structured Al/Ti2ALC
composites. These reconstructions shown in
Figure 8 reveal the failed Ti2AlC phase
excluding the interpenetrating metal.

Hanaor, D. A. H., L. Hu, W. H. Kan, G. Proust, M. Foley, I. Karaman, and M. Radovic. "Compressive performance and
crack propagation in Al alloy/Ti2AlC composites." Materials Science and Engineering: A 672 (2016): 247-256.
10
Figure 8. Volume rendering of MAX phase
On the basis of entropic thresholding, the
approximated phase proportions of the three
constituents (Al alloy, Ti2AlC and air) are
shown in Table 2 for the different materials
analysed. The post compression material shows
a higher air fraction owing to the presence of
cracks. The presence of unfilled pores and Al
shrinkage voids further explains for the

Hanaor, D. A. H., L. Hu, W. H. Kan, G. Proust, M. Foley, I. Karaman, and M. Radovic. "Compressive performance and
crack propagation in Al alloy/Ti2AlC composites." Materials Science and Engineering: A 672 (2016): 247-256.
11
presence of air in both AP and PC materials. As
entropic thresholding is an indirect method
relative to the alcohol immersion method, the
data given in Table 1 is likely to be a more
accurate representation of as-processed
materials. The largest void fraction is found for
coarse material post-compression. This is likely
the result of the wider cracks formed in this
material during failure.
Table 2. Phase proportions approximated by volumetric X-ray tomographic analysis
Material
Volumetric composition As-processed / Post Compression (%)
Ti2AlC
Al 6061
Air
Fine
64.14/61.84
32.14/32.5
3.72/5.66
Medium
59.54/62.05
37.57/33.82
2.88/4.13
Coarse
59.56/63.81
38.05/30.25
2.38/5.94
3.5. Pore segmentation
The distribution of the aluminium alloy
material was assessed by threshold-separating
this phase and applying a watershed
segmentation algorithm to define individual Al
alloy-filled pores. The results of this
segmentation are shown in Figure 9. The
aluminium alloy phase is interconnected, as it is
formed through an infiltration process in the
predominantly open pore structure of the MAX
phase foams, and thus segmentation is
somewhat ambiguous. Nevertheless, the
structure and the Al phase distribution are
readily observable in the segmentation results.
The volume and interface area of individual
segments are naturally smaller for finer
structures. The specific interfacial area of the
alloy phase is represented by the parameter α’
given in terms of μm-1 and corresponds to the
ratio of interface area to pore volume.
1
1
in
i
i
in
i
i
A
V








……………….(1)
where Ai and Vi are the interfacial area and
volume of individual segments. From the
analysis of this parameter it is found
unsurprisingly that the specific interface area of
the Al alloy phase increases monotonically with
finer mesostructures.

Hanaor, D. A. H., L. Hu, W. H. Kan, G. Proust, M. Foley, I. Karaman, and M. Radovic. "Compressive performance and
crack propagation in Al alloy/Ti2AlC composites." Materials Science and Engineering: A 672 (2016): 247-256.
12
Figure 9. Segmented volume rendering of Al alloy phase for fine, medium and coarse materials.
Arbitrary segment colours.

Hanaor, D. A. H., L. Hu, W. H. Kan, G. Proust, M. Foley, I. Karaman, and M. Radovic. "Compressive performance and
crack propagation in Al alloy/Ti2AlC composites." Materials Science and Engineering: A 672 (2016): 247-256.
13
Table 3. Aluminium phase segmentation
Material
Total
interface
(μm2)
Mean
segment
volume (μm3)
Mean segment
interface
(μm2)
α’ AP/PC (μm-1)
Fine
11.5

108
0.379

106
3.65

104
9.63

10-2 / 10.78

10-2
Medium
6.15

108
1.216

106
4.39

104
3.62

10-2 / 4.48

10-2
Coarse
3.92

108
3.241

106
8.10

104
2.50

10-2 / 3.40

10-2
4. Discussion
From examination of SEM micrographs, XRM
slices, and volumetric analysis, it is evident that
while the fine structure yields through a
relatively straight shear crack in Al/Ti2AlC
processed by Al melt infiltration of the porous
Ti2AlC foams, the medium and coarse materials
exhibit fine branching cracks. This is further
evident from the quantitative analysis of phase
segmentation. From analysis of the
segmentation data we find that for fine, medium
and coarse structured materials the specific
interface area of the Al alloy phase increases
subsequent to failure by approximately 12%,
24% and 36% respectively. This trend is
indicative of two possible tendencies: (i) a
greater extent of cracks; (ii) a preferential
tendency for cracks to propagate through or
around Al alloy inclusions. From the
micrographic interpretation of ortho-slice
sections there appears to be little tendency for
preferential propagation within Al alloy
regions. However an exception exists for the
case of crack propagation through elongated Al
alloy inclusions likely formed through the
filling of pre-existing cracks in the MAX phase
matrix, which are more common in coarse
structured material. Thus it is likely that both
mechanisms (i) and (ii) play a certain role in the
observed trend.
While absent in the fine material, Al filled pores
exhibiting large shrinkage voids are common in
coarse and medium structures and these are
likely to have acted as crack nucleation sites as
evident by the correlation of lower compressive
strength to higher porosity and the cracks seen
in the XRM reconstruction of post compression
materials. Such voids can be seen in XRM
ortho-slices shown in Figure 7 (c) and (d).
Minimising void formation in CAPAI synthesis
is likely to improve toughness and may be
achieved by employing a finer meso-structure
or by increased infiltration pressure.
The tendency for cracks to propagate through
Al alloy/Ti2AlC interfaces suggests that
fracture toughness can be improved by
enhancing the bonding across these interfaces,
thus increasing the tensile strength and energy
dissipation. The unintentional formation of
Al3Ti and Ti3AlC2 near interfaces during
infiltration, as revealed by EBSD, may weaken
these regions and minimising their occurrence
is likely to enhance fracture toughness. One
method that may be appropriate towards this
end would be the gas phase deposition of Al
through PVD methods on MAX phase
precursor to create a well-bonded intermediate
layer with minimal diffusion impurities prior to
CAPAI.
5. Conclusions
The compressive performance of a novel
metal/MAX phase composite system was
examined and interpreted with reference to
crack propagation observations in materials of
varied meso-structure. By employing recently
developed techniques of X-ray microscopy we
were able to gain three dimensional insights
into the structure and failure behaviour of these
materials. Understanding the role of meso-

Hanaor, D. A. H., L. Hu, W. H. Kan, G. Proust, M. Foley, I. Karaman, and M. Radovic. "Compressive performance and
crack propagation in Al alloy/Ti2AlC composites." Materials Science and Engineering: A 672 (2016): 247-256.
14
structure in the strength of metal/MAX phase
composites provides new pathways towards the
tailoring of mechanical properties in this type of
system.
In the present work a finer interpenetrating
metallic phase and lower porosity were found
to strengthen the composite materials. This
suggests that adjusting the size distribution of
the metallic phase and minimising the void ratio
in CAPAI processed materials may facilitate
improved mechanical performance in these
materials.
The major findings are summarized as follows.
1. While fine structured composite
materials yields through a relatively
straight shear crack, the medium and
coarse materials exhibit yielding
through large numbers of fine
branching cracks.
2. Cracks propagate predominantly
through the MAX phase. Upon
encountering Al filled pores, cracks
tend to be deflected and propagate
through MAX phase-Al alloy
interfaces.
3. The tendency for cracks to deflect
through Al alloy/Ti2AlC interfaces
suggests that fracture toughness can be
improved by enhancing the bonding
across these interfaces, thus increasing
the tensile strength and energy
dissipation that occurs in crack
propagation.
Acknowledgement
We acknowledge access to XRM facilities of
the Australian Microscopy & Microanalysis
Research Facility at the Australian Centre for
Microscopy & Microanalysis at the University
of Sydney. This work was further supported by
the U.S. Air Force Office of Scientific
Research, MURI Program (FA9550-09-1-
0686) and US National Science Foundation
(NSF–1233792) to Texas A&M University.
The authors would like to thank the program
manager Dr. David Stargel for his support. In
addition, the authors are also grateful for the
support of the International Program
Development Fund and DVC
Research/International Research Collaboration
Award, at the University of Sydney.

Hanaor, D. A. H., L. Hu, W. H. Kan, G. Proust, M. Foley, I. Karaman, and M. Radovic. "Compressive performance and
crack propagation in Al alloy/Ti2AlC composites." Materials Science and Engineering: A 672 (2016): 247-256.
15
References
[1] M.W. Barsoum, MAX phases: properties of machinable ternary carbides and nitrides, John Wiley &
Sons, 2013.
[2] V.H. Nowotny, Progress in Solid State Chemistry, 5 (1971) 27-70.
[3] W. Jeitschko, H. Nowotny, Monatshefte für Chemie und verwandte Teile anderer Wissenschaften,
98 (1967) 329-337.
[4] E. Reiffenstein, H. Nowotny, F. Benesovsky, Monatshefte für Chemie und verwandte Teile anderer
Wissenschaften, 97 (1966) 1428-1436.
[5] H. Wolfsgruber, H. Nowotny, F. Benesovsky, Monatshefte für Chemie/Chemical Monthly, 98 (1967)
2403-2405.
[6] M.W. Barsoum, T. El‐Raghy, Journal of the American Ceramic Society, 79 (1996) 1953-1956.
[7] J. Frodelius, M. Sonestedt, S. Björklund, J.-P. Palmquist, K. Stiller, H. Högberg, L. Hultman, Surface
and Coatings Technology, 202 (2008) 5976-5981.
[8] A. Zhou, M. Barsoum, S. Basu, S. Kalidindi, T. El-Raghy, Acta materialia, 54 (2006) 1631-1639.
[9] R. Benitez, W.H. Kan, H. Gao, M. O'Neal, G. Proust, M. Radovic, Acta Materialia, 105 (2016) 294-
305.
[10] M.W. Barsoum, M. Radovic, Annual review of materials research, 41 (2011) 195-227.
[11] M. Radovic, M.W. Barsoum, American Ceramics Society Bulletin, 92 (2013) 20-27.
[12] X. Wang, Y. Zhou, Journal of Materials Science & Technology, 26 (2010) 385-416.
[13] R. Bhattacharya, R. Benitez, M. Radovic, N.C. Goulbourne, Materials Science and Engineering: A,
598 (2014) 319-326.
[14] P.N. Parrikar, R. Benitez, H. Gao, M. Radovic, A. Shukla, Materials Science and Engineering: A, 658
(2016) 176-184.
[15] S. Li, G. Song, K. Kwakernaak, S. van der Zwaag, W.G. Sloof, Journal of the European Ceramic
Society, 32 (2012) 1813-1820.
[16] H. Yang, Y. Pei, G. Song, J.T.M. De Hosson, Journal of the European Ceramic Society, 33 (2013)
383-391.
[17] J. Byeon, J. Liu, M. Hopkins, W. Fischer, N. Garimella, K. Park, M. Brady, M. Radovic, T. El-Raghy,
Y. Sohn, Oxidation of metals, 68 (2007) 97-111.
[18] M. Barsoum, T. El-Raghy, M. Ali, Metallurgical and Materials Transactions A, 31 (2000) 1857-1865.
[19] M. Barsoum, T. Zhen, S. Kalidindi, M. Radovic, A. Murugaiah, Nature Materials, 2 (2003) 107-111.
[20] L. Hu, Texas A&M College Station, Texas, USA, 2015.
[21] Y. Zhang, Z. Sun, Y. Zhou, Material Research Innovations, 3 (1999) 80-84.
[22] L. Peng, Scripta materialia, 56 (2007) 729-732.
[23] J. Zhang, J. Wang, Y. Zhou, Journal of Materials Research, 26 (2011) 372-383.
[24] Z.L. Zhang, H.X. Zhai, Y. Zhou, Z.Y. Huang, M.X. Ai, Key Engineering Materials, Trans Tech Publ,
2008, pp. 989-991.
[25] W. Wang, V. Gauthier-Brunet, G. Bei, G. Laplanche, J. Bonneville, A. Joulain, S. Dubois, Materials
Science and Engineering: A, 530 (2011) 168-173.
[26] S. Amini, C. Ni, M.W. Barsoum, Composites Science and Technology, 69 (2009) 414-420.
[27] L. Hu, A. Kothalkar, G. Proust, I. Karaman, M. Radovic, Journal of Alloys and Compounds, 610
(2014) 635-644.
[28] S. Amini, M.W. Barsoum, Materials Science and Engineering: A, 527 (2010) 3707-3718.
[29] S. Amini, C. Ni, M.W. Barsoum, Composites Science and Technology, 69 (2009) 414-420.
[30] J. Liu, J. Binner, R. Higginson, Z. Zhou, Composites Science and Technology, 72 (2012) 886-893.
[31] M. Kouzeli, D. Dunand, Acta materialia, 51 (2003) 6105-6121.
[32] B. McWilliams, J. Yu, E. Klier, C.-F. Yen, Materials Science and Engineering: A, 590 (2014) 21-29.
[33] T. Srivatsan, J. Mattingly, Journal of materials science, 28 (1993) 611-620.

Hanaor, D. A. H., L. Hu, W. H. Kan, G. Proust, M. Foley, I. Karaman, and M. Radovic. "Compressive performance and
crack propagation in Al alloy/Ti2AlC composites." Materials Science and Engineering: A 672 (2016): 247-256.
16
[34] D.M. Hulbert, D. Jiang, U. Anselmi-Tamburini, C. Unuvar, A.K. Mukherjee, Materials Science and
Engineering: A, 488 (2008) 333-338.
[35] J. Jung, S. Kang, Journal of the American Ceramic Society, 87 (2004) 47-54.
[36] D.L. McDanels, Metallurgical transactions A, 16 (1985) 1105-1115.
[37] T. Wilkes, M. Young, R. Sepulveda, D. Dunand, K. Faber, Scripta materialia, 55 (2006) 1083-1086.
[38] L. Hu, R. Benitez, S. Basu, I. Karaman, M. Radovic, Acta Materialia, 60 (2012) 6266-6277.
[39] M.T. Agne, M. Radovic, G.W. Bentzel, M.W. Barsoum, Journal of Alloys and Compounds, 666
(2016) 279-286.
[40] L. Hu, A. Kothalkar, M. O'Neil, I. Karaman, M. Radovic, Materials Research Letters, 2 (2014) 124-
130.
[41] L. Hu, M. O’Neil, V. Erturun, R. Benitez, G. Proust, I. Karaman, M. Radovic, Submitted Manuscript,
(2016).
[42] ASTM, ASTM International, E112-13 (2013).
[43] ASTM, ASTM International, C20-00 (2010).
[44] M.W. Barsoum, Progress in Solid State Chemistry, 28 (2000) 201-281.
[45] R.B. Ross, Metallic materials specification handbook, Springer Science & Business Media, 2013.
[46] Y. Sylvester, L. Hunter, B. Johnson, R. Estrada, 3D Systems Integration Conference (3DIC), 2013
IEEE International, IEEE, 2013, pp. 1-7.
[47] L.L. Lavery, J. Gelb, A.P. Merkle, A. Steinbach, Microscopy Today, 22 (2014) 16-21.
[48] J.N. Kapur, P.K. Sahoo, A.K. Wong, Computer vision, graphics, and image processing, 29 (1985)
273-285.
[49] T. Pun, Computer Graphics and Image Processing, 16 (1981) 210-239.