Ceramic matrix composites DJM

Ceramic matrix composites (CMCs) have been developed to overcome the
intrinsic brittleness and lack or mechanical reliability of monolithic ceramics, which are
otherwise attractive for their high stiffness and strength
1
. The issue is particularly acute
with glasses, as  the amorphous structure does not provide any obstacle to crack
propagation  and the fracture toughness is very low (<1MPa.m1/2
)
2
.  In addition to
mechanical effects, the reinforcing phase ,may benefit other properties such as electrical
conductivity, thermal expansion  coefficient, hardness and thermal shock resistance
1, 3
.
The combination of these characteristics with intrinsic advantages of ceramic materials
such as high-temperature stability, high corrosion resistance, light weight and electrical
insulation, makes CMCs very attractive functional and structural materials for a variety
of applications; they have  particular relevance  under harsh conditions  where other
materials (e.g. metallic alloys) cannot be used
4-6
.
A wide range of reinforcing fibres have been explored, including those based on
SiC, carbon, alumina and mullite
7-9
. However, carbon fibres are amongst  the highest
performance toughening elements investigated, since  the first reports  of their use in
ceramic matrices were published in the late 1960s
10
. The fracture toughness of carbon
and SiC fibre reinforced glass and ceramic matrix composites can be much better than
the native matrix, as demonstrated by a wealth of available data in the literature
4-6, 10, 11
 
(e.g.  17MPa.m1/2
  for SiC fibre reinforced glass-ceramic composites
5
).  Various
toughening mechanisms can be involved, including fibre debonding, fibre pull-out, and
crack bridging
11
.
Carbon nanotubes (CNTs) have received an enormous degree of attention in
recent years, and, in the context of composites, they  are often seen as the ‘next

generation’ of carbon fibre.  Although  their remarkable properties have suggested
applications as diverse as tissue scaffolds, field emission guns, and supercapacitor
electrodes
12
, the interest in composite materials is driven by both the mechanical and
functional properties that can be obtained at very low density (typically in the range 1.5-
2.0 gcm-3
). For individual perfect CNTs, the axial stiffness has been shown to match that
of the best carbon fibres (approaching around 1 TPa), whilst the strength is an order of
magnitude higher  (around 50 GPa)
13
. Their electronic properties depend subtly on the
exact structure but larger CNTs are essentially metallic conductors
14
; smaller CNTs can
offer unique optoelectronic properties, useful, for example, in non-linear optics
15
.
Ballistic  electron transport  effects  can be related to uniquely  high current  carrying
capacity  (up to 109
 Acm-2
) whilst the axial thermal conductivity is higher than  that of
diamond (>2000 Wm-1
K-1
)
16
. It is worth noting that surface areas of CNTs can be very
high  since, in the absence of agglomeration,  with every atom of a single walled
nanotube  lies on its surface; however, this factor can be a mixed blessing when
considering composite applications, as discussed further below.
One other significant characteristic of CNTs is their very high aspect (length to
diameter)  ratio which  is relevant to  load transfer with the matrix and, hence, effective
reinforcement.  Standard continuous-fibre composites have excellent anisotropic
structural properties combined with low density, but are expensive to process and are
limited to simple shapes
3
. Short-fibre composites, on the other hand, are easier to
produce in complex shapes but with conventional fibres, the aspect ratio is typically
limited to around 100, after processing
17
. In principle, the small absolute length of
CNTs, combined with their resilience in bending,  allows them  to be manipulated with
conventional processing equipment,  potentially  retaining their high aspect ratio;

however, in practice, length degradation is known to occur under high shear conditions.
The high aspect ratio of CNTs can also encourage the formation of percolating networks
that are relevant to functional properties, particularly electrical conductivity
18
; indeed
the lowest percolation threshold for any system has been observed in kinetically-formed
networks of CNTs in epoxy
19
.
Structurally, CNTs have diameters in the range of around 1 nm to a somewhat
arbitrary  upper limit of  50 nm,  and lengths of many microns (even centimetres in
special cases)
20
. They can consist of one or more concentric graphitic cylinders,
forming single or multi walled nanotubes (SWCNTs / MWCNTs). In contrast,
commercial (PAN and pitch) carbon fibres are typically in the 7 – 20 µm diameter range,
whilst  vapour-grown carbon fibres (VGCFs) have a broad range of  intermediate
diameters. Compared to carbon fibres, the best nanotubes can have almost atomistically
perfect structures; indeed, there is a general question as to whether the smallest CNTs
should be regarded as very small fibres or heavy molecules, especially as the diameters
of the smallest nanotubes are similar to those of common polymer molecules.
Consequently, it is not yet clear to what extent conventional fibre composite
understanding can be extended to CNT composites, or whether new mechanisms will
emerge.
Although the perfect CNT structure is very appealing, real materials are very
diverse and vary significantly in terms of dimensions, purity, surface chemistry,
crystallinity, graphitic orientation, degree of entanglement, and cost. These factors
directly affect the properties and processability of CNTs and they must be considered
when interpreting their performance in a given application.  In very broad terms, CNTs
can be divided into two classes depending on the synthetic route used to prepare them.