Big Returns From Small Fibers: A Review

of Polymer/Carbon Nanotube Composites

O. BREUER and UTTANDARAMAN SUNDARARAJ *

Department of Chemical and Materials Engineering

University of Alberta
Edmonton, Alberta T6G 2G6, Canada

This paper reviews recent studies conducted on carbon nanotube/polymer com-

posites. Carbon nanotubes are promising new materials for blending with polymers
with potential to obtain low-weight nanocomposites of extraordinary mechanical,
electrical, thermal and multifunctional properties. The size scale, aspect ratio and
properties of nanotubes provide advantages in a variety of applications, including
electrostatically dissipative materials; advanced materials with combined stiffness,
strength and impact for aerospace or sporting goods; composite mirrors; automotive
parts that require electrostatic painting and automotive components with enhanced
mechanical properties. The various processing methods for producing these nano-
composites are discussed, in particular melt mixing, solution processing and in-situ
polymerization. Some key results are summarized, relating to the mechanical, elec-
trical, thermal, optical and surface properties. Finally, the challenges for the future
are discussed in terms of processing, characterization, nanotube availability, nanotube
tailoring, and the mechanisms governing the behavior of these remarkable nanoscale
composites. Polym. Compos. 25:630–645, 2004. © 2004 Society of Plastics Engineers.

INTRODUCTION nanotubes at large scales is not yet at hand (15). How-

ever, it is only a matter of time before CNTs are mass-
S ince the discovery of carbon nanotubes (CNTs) in
1991 by Iijima (1), CNTs have been looked at ex-
tensively by researchers in various fields such as chem-
produced at low cost (16
22).
A carbon nanotube is a hexagonal network of carbon
atoms rolled up into a seamless, hollow cylinder, with
istry, physics, materials science, and electrical engi-
each end capped with half of a fullerene molecule (3,
neering. CNTs are unique nanostructured materials
18, 23). Although similar in chemical composition to
with remarkable physical and mechanical properties
graphite, CNTs are highly isotropic, and it is this topol-
(1
5). These properties have inspired interest in using
ogy that distinguishes nanotubes from other carbon
CNTs as a filler in polymer composite systems to obtain
structures and gives them their unique properties.
ultra-light structural materials with enhanced electrical,
There are two main kinds of nanotubes: single-walled
thermal and optical characteristics (6, 7). The prospect
nanotubes (SWNTs), individual cylinders of 1
2 nm in
of obtaining advanced nanocomposites with multifunc-
diameter, which are actually a single molecule; and
tional features, e.g., materials used for structures and
multi-walled nanotubes (MWNTs), which are a collec-
electrical conductors, has attracted the efforts of re-
tion of several concentric graphene cylinders, a “Russ-
searchers in both academia and industry (8
10). In-
ian doll” structure, where weak Waals forces bind the
dustry in particular recognizes many potential applica-
tubes together (24). A schematic of these species is
tions such as electrostatically dissipative materials and
shown in Fig. 1. The diameter, chirality and form of the
aerospace structural materials (8
13). Currently, one
nanotube determine its properties (25).
of the major obstacles to using nanotubes as a polymer
Since carbon-carbon covalent bonds are among the
filler is cost; however, advances in the synthesis of
strongest bonds in nature, a structure based on a per-
CNTs continue to rapidly improve both their quantity
fect arrangement of these bonds oriented along the axis
and quality (14), though growing structurally perfect
of the nanotubes produces a very strong material with
an extremely high strength-to-weight ratio. This quali-
fies CNTs (especially SWNTs) as the “ultimate” carbon
* To whom correspondence should be addressed. fiber (26, 27). The mechanical properties of various
© 2004 Society of Plastics Engineers
Published online in Wiley InterScience (www.interscience.wiley.com).
types of nanotubes have been extensively studied both
DOI: 10.1002/pc.20058 by experimental and computational means (28, 29).

630 POLYMER COMPOSITES, DECEMBER 2004, Vol. 25, No. 6

 Review of Polymer/Carbon Nanotube Composites

Fig. 1. Schematics of a single-walled carbon nanotube (a 5,5 armchair tube is shown) (a), a graphite sheet (b), a hemispherical end cap
(c) and a multiwalled carbon nanotube (d). The solid circles in each figure represent the carbon atoms (85). Published with permission
from Energeia.

Their tensile strength, approximately 100
600 GPa, is There are many possibilities for rolling a slice of graph-
about two orders of magnitude higher than that of cur- ene into a seamless cylinder, because when rolled into
rent high-strength carbon fibers (30
33), and their den- a nanotube, the hexagons may spiral around the cylin-
sity is around 1.3 g/cm3 (34), lower than the density of der, giving rise to “chirality,” a twist that determines
commercial carbon fibers (1.8
1.9 g/cm3). The en- whether the CNT behaves like a metal or a semicon-
hancement in strength implies that, for the same perfor- ductor (40). Metallic CNTs have ballistic transport,
mance, replacing commercial carbon fibers with CNTs which means there is zero resistance along the tube
will lead to significant reduction in the density and vol- (47), and they can have conductivities several times
ume of the structural composite parts used in aerospace that of copper. Because of their very low energy dissi-
and other high performance applications (35, 36). The pation, nanotubes carry tremendous current densities,
compressive strength of CNTs is approximately two or- higher than 100 MA/cm2 for MWNTs, which may be
ders of magnitude higher than the compressive strength compared to current densities of tens of kA/cm2 for su-
of any known fiber (37). CNTs are also one of the stiffest perconducting wires (48).
structures ever made; compared to carbon fibers, which The thermal conductivities of CNTs are highly aniso-
typically have Young’s moduli of up to 750 GPa, moduli tropic, diamond-like over the length of the tube and
of nanotubes range between 1 and 5 TPa (1, 30, 38). insulating in the transverse direction (7, 16). Theory
The most striking effect, however, is the combination predicts a room-temperature thermal conductivity of
of high flexibility and strength with high stiffness, a 6000 W/(m·K) for an isolated SWNT, while experimental
property that is absent in graphite fibers (39). In fact, if values are around 200 W/(m·K) (49).
we would apply nanotubes as mechanical springs, CNTs have important surface properties, which affect
these springs would be very stiff for small loads, but their dispersion within a polymer matrix. The strong
would turn soft for larger loads, accommodating large van der Waals interactions between nanotubes cause
deformations without breaking. They get can be flat- them to form bundles (50), 10
100 nm in diameter.
tened, twisted and buckled as they deform reversibly These bundles may break by shearing along the axis,
(38, 40). Their fracture strains range between 10% and causing failure between nanotubes at stresses much
30%, which is better than most carbon fibers, which below the intrinsic capabilities of a single nanotube.
have values of 0.1%
2% (35, 36). Their aspect ratios Nanotube surface structure can be tailored locally; for
can be as high as 10,000 (34), which can be an advan- example, we can functionalize or open the ends be-
tage for imparting conductivity and strength to a com- cause caps of a nanotube have a high concentration of
posite system. It is readily apparent that the combi- pentagonal defects increasing reactivity (1, 51).
nation of size, structure and topology provides the CNTs can be synthesized via many techniques in-
nanotubes with excellent mechanical properties (1). cluding arc methods, laser methods, chemical vapor
Carbon nanotubes have a unique electronic charac- deposition (CVD) and other methods like fluidized bed
ter, ranging from high-conductivity metallic behavior to reactors (52). The arc method was originally used for
semiconducting with a large band gap (1, 27, 41
46). fullerene production, but created CNTs in the process.

POLYMER COMPOSITES, DECEMBER 2004, Vol. 25, No. 6 631

 O. Breuer and Uttandaraman Sundararaj

The method continues to be used commercially but new and even larger species has been introduced re-
large-scale production has been difficult. In the arc cently: carbon tubes with a typical diameter of 5 m,
method, a high-current electric arc passes through for applications similar to those of CNTs; however, their
graphite electrodes in the presence of metal catalyst properties and advantages are yet to be assessed (63).
particles, and CNTs are one of the products formed. In Additional types of carbon nanofibers exist, in which
the laser method, a laser is shot at a carbon target in the graphite layers are stacked in various orientations
the presence of metal catalyst and carbon nanofibers with respect to the fiber axis, giving rise to assorted
are grown from metal catalyst particles created in the conformations rather than the concentric shell struc-
process. There are several different CVD processes that ture observed in CNTs or VGCFs (64
66). The internal
thermally decompose hydrocarbon gases and use structures vary as in the nanotube arrangement, form-
metal catalysts to grow nanotubes. CVD can produce ing truncated cones, segmented bamboo-like structures
well-aligned SWNTs or MWNTs but can also yield en- or stacked coins, to name a few. External morphology
tangled CNTs (52). Though there are no low-cost, large- variations include kinked and branched structures,
scale production methods, the traditional methods are carbon nanohorns, and bucky peapods within tubes
being developed further, and new methods like flu- (8). One of the most outstanding features of these
idized bed reactors are being investigated to create a structures is the presence of a large number of edges,
steady, reasonably priced CNT supply. The low CNT which contain sites readily available for chemical or
availability and their high prices have limited realization physical interactions. These structures have a chemi-
of polymer-CNT composites for many practical applica- cally active surface area in the range of 300
700 m2/g
tions. Part of the reason for the limited success may be (66). Their typical dimensions are 5
100 m in length
that CNTs need to be chemically modified so that they and 5
100 nm in diameter (66). These nanofibers may
can interact with other materials like polymers (53). be used in high-performance composite structures in
Vapor-grown carbon fibers (VGCFs), a species simi- place of conventional fibers. Another application is mo-
lar to MWNTs, comprise graphitic networks in concen- lecular sieves, where the small and precise distances be-
tric cylinders (54
62). An SEM micrograph of VGCFs is tween the nanometer-sized platelets (typically less than
shown in Fig. 2. They are larger than CNTs, with typi- 0.5 nm) and their chemical activity are exploited (64, 66).
cal diameters of 50
200 nm. The inner diameter is Yet another possible structure consists of carbon nan-
30
90 nm and the length is in the range of 50
100 otubes grown on conventional carbon fibers for multi-
m, so that the aspect ratios are in the 100
500 range scale, hybrid reinforcement (67). However, most of the
(30, 54
56). VGCFs have a tensile strength of 2.92 GPa recent work on polymer composites containing carbon
and a tensile modulus of 240 GPa (57), and their ther- nanostructures involves CNTs; therefore these will be
mal conductivity is the highest among all commercial the focus of our review. It should be noted that because
carbon fibers, with a value of 1950 W/(m·K) (58). of the availability of VCGFs, there are more studies re-
VGCFs have a much lower production cost than CNTs, porting mechanical property results and melt-mixing
and kilogram quantities are readily available (8, 58), characteristics of polymer/VCGF nanocomposites.
making VGCFs commercially attractive (59). Their dis- The explosion in interest in CNT nanocomposites can
advantage is in the larger amount of microstructural be seen through the increasing number of patents filed
defects in comparison to CNTs (30). Nonetheless, work in the past few years (68
75), relating to the fields of
has been done to evaluate their performance as a new electromagnetic shielding, reinforced textiles, and oth-
filler type, and to build a knowledge base that might be ers. This paper reviews the most recent developments
transferable to CNT composites (7, 19, 58, 60
62). A in CNT/polymer nanocomposites, focusing on the pro-
cessing, properties and applications of these materials.
We will discuss the problems and issues that must
be resolved to make these materials tailorable and
commercially viable.

PROCESSING
Several processing methods are available for produc-
ing polymer/CNT composites based on either thermo-
plastics or thermosets. Although the techniques are in-
herently different, all of them try to address issues that
directly affect the composite properties, such as deag-
glomeration of bundles and ropes, CNT dispersion, align-
ment and interfacial bonding (10, 56, 76, 77). Figure 3
shows both good and poor dispersion in a VGCF/PS
composite. The effective utilization of carbon nanotubes
in composite applications depends strongly on the abil-
ity to disperse the CNTs individually and uniformly
Fig. 2. SEM micrograph of vapor-grown carbon fibers (VGCFs), throughout the matrix without destroying their in-
Pyrograph-III, from Applied Sciences Inc. tegrity or reducing their aspect ratio (78). Furthermore,

632 POLYMER COMPOSITES, DECEMBER 2004, Vol. 25, No. 6

 Review of Polymer/Carbon Nanotube Composites

thermal annealing in air or oxygen for selective

etching of amorphous carbons, and acid treatment
for eliminating catalyst residues (26, 86). Mechan-
ical techniques like centrifugal separation, size ex-
clusion chromatography and microfiltration may
be applied as well (87
90). Another approach is
using conjugated polymers, which serve as ex-
tracting agents for purification (91
95). Current
methods for nanotube purification tend to reduce
the amount of material by approximately 50% by
mass (98); therefore the development of more
rapid, high-yield techniques is required for scale-
up of the CNT purification process (88, 99).
2. Deagglomeration for dispersing individual nan-
otubes (49, 100): The most common method for
deagglomeration is ultrasonicating a CNT solution
(33, 49, 101); however, this process may also in-
troduce defects into the CNTs (24, 53). Electrosta-
tic plasma treatment is also used to separate
CNTs from the larger agglomerated particle mix-
ture of other CNTs and impurities (100). Electric
field manipulation and polymer wrapping have
also been proposed to overcome entangling of nan-
otubes, which results from their strong van der
Waals attractions (102). Ball-milling has been
used for breaking up the agglomerates in VGCFs
(57, 103) and in CNTs (104).
3. Chemical functionalization for improving nan-
Fig. 3. SEM micrograph of a fracture surface of VGCFs dis- otube/matrix interactions for processability and
persed in a polystyrene matrix: good dispersion (a) and poor property enhancement : The surface modification
dispersion (b). will allow the unique properties of CNTs to be cou-
pled with those of other materials in the final prod-
uct (7, 51, 56, 105
112). Several approaches to
functionalization have been developed, including
good interfacial bonding is required to achieve load
defect functionalization; covalent functionalization
transfer across the CNT-matrix interface (24, 27, 29,
of the sidewalls; and noncovalent exohedral func-
79). Finally, the 3D architecture will determine the
tionalization with polymers or surfactants as
properties — for example, skeleton-like for percolation,
supramolecular adducts (113
118). Ajayan and
or well-aligned for high strength, (80). Melt-mixing, in-
Schadler (6) used chemical vapor deposition (CVD)
situ polymerization and solution processing are the
to roughen the surface and promote adhesion
main processes discussed; however, new techniques
(119). Yet another possibility is to expose the nan-
such as deposition using layer-by-layer assembly have
otube composite to gamma radiation for altering the
also been applied (81).
chemistry at the interface for property enhance-
ment, as has been done in SWNT-PMMA (120).
Preprocessing
Bubert et al. (121) treated the surface of VGCFs by
Despite their intrinsic rigidity and high anisotropy, plasma, resulting in polar groups formed on the
most currently available forms of CNTs are isotropic surface and a modified surface morphology.
and fragile and contain several species. To prepare the
material for processing on a macroscopic scale, prepro- Melt-Mixing
cessing is required (82). The following steps are usually
Melt-mixing of CNTs into thermoplastic polymers
taken:
using conventional processing techniques, such as
1. Purification to eliminate non-nanotube material extrusion, internal mixing, injection molding and blow
(83
99): Most production processes generate a molding, are particularly desirable, because of the
range of carbonaceous particles such as amor- speed, simplicity and availability of these processes
phous carbon, fullerenes, and nanocrystalline in the plastics industry (78). These methods are also
graphites, and the final effluent includes transi- beneficial because they are free of solvents and conta-
tion metal catalysts (85). The most common meth- minants, which are present in solution processing
ods to remove unwanted by-products include methods and in-situ polymerization (10). Nanomaterials

POLYMER COMPOSITES, DECEMBER 2004, Vol. 25, No. 6 633

 O. Breuer and Uttandaraman Sundararaj

have a unique advantage in thermoplastic compound-

ing and molding, because in contrast to larger, micro-
scale carbon fibers, less fiber breakage occurs, and a
high L/D is maintained for CNTs (24, 30, 79, 122) and
VGCFs (56, 59, 123). Use of high-shear mixing and
longer processing times may enhance dispersion, and
when coupled with elongational flow, should yield
aligned nanotubes (8, 10, 33, 70). The increase in vis-
cosity with composition is much higher for nanofibers
than that reported for larger-diameter fibers or carbon
black (124), so high shear mixing is needed to overcome
high viscosities of the polymer/nanotube mixtures (7).
An additional advantage is the prospect of recycling
thermoplastic composites to improve cost-effectiveness
and to create environmentally benign, advanced nan-
otube materials (98). Nevertheless, there is still much Fig. 5. SEM micrographs of VGCFs after melt processing; the
to be learned about the ability of CNTs to withstand polystyrene matrix has been dissolved.
high shear and elongational flow processing (98) and
about optimization of processing parameters to provide
good CNT dispersion while minimizing the extent of geometry of the screw on the structure and properties
polymer degradation (125). The effect of mixing energy (59). Most of the fiber breakage occurs in the initial com-
on MWNT dispersion and fracture is shown in Fig. 4. pounding stage in the extruder, so no further changes
Fractured VGCFs as a result of melt mixing are de- were perceived after subsequent processing cycles such
picted in Fig. 5. as injection molding (59).
Ferguson et al. (126) and Schwartz and Bryant (127) Controlling the alignment is possible by using several
obtained good dispersion when melt-processing CNT melt-mixing methods. Spinning of extruded melt sam-
thermoplastics using a Buss Kneader for initial com- ples is one way of making well-aligned CNT/polypro-
pounding, followed by injection molding. The CNTs did pylene (PP) composites with good nanotube dispersion
not break or orient because of their size and geometry, (24, 55). Numerous fiber-aligning techniques have been
and therefore, conductivity of the nanocomposite was developed; for example, by increasing the residence
retained even after subsequent processing. The low time in the die channel, or altering the die design to
CNT loading of a few percent was also significant in pre- control the orientation of the fibers (61). Injection mold-
serving the original physical properties of the polymer ing was also found to induce significant alignment in
matrix (126). Nylon 12 and polycarbonate/acryloni- CNT/PP samples, as demonstrated by coefficient of
trile-butadiene styrene terpolymer (PC/ABS) blend thermal expansion (CTE) and electrical resistivity meas-
nanocomposites showed measurable conductivities at urements (54).
loadings less than 5 wt%, superior surface finish, low Polymer processing methods use large-scale quanti-
sensitivity to processing and the ability to be reground ties of raw materials, which are sometimes inappropri-
for recycling on conventional grinding equipment (127). ate when developing new composites based on costly
Extrusion followed by compression molding also fillers such as CNTs. A benchtop MiniMAX molder
yielded a random distribution of fibers, enhancing per- (Custom Scientific Instruments) is a common instru-
colation (124). Well-dispersed CNTs were also obtained ment used for processing gram-scale quantities of nan-
by extrusion and injection molding (128, 129). Studies ocomposites. The apparatus is equipped with a rotor,
on extrusion showed only a slight influence of the which can be submerged in a 12.7-mm-diameter
heated cup containing the material. Mixing is imparted
by the rotor, and some reorientation may be obtained
as well if the rotor is manually raised and lowered. The
material may subsequently be injection molded from
the cup into a mold, located below (57, 103). Results
obtained for a MWNT/PMMA composite showed good
dispersion and no apparent nanotube damage (10);
however, the low shear rate and absence of extensional
flow in this mixer may not always provide the desired
quality of mixing (130). Tibbetts and McHugh (57) ob-
served areas of insufficient wetting in VGCF blends
with either polypropylene or nylon mixed in a MiniMAX
molder. Lozano et al. (58, 60) used an intermediate-size
mixer (half of a standard Haake internal mixer—
Fig. 4. MWNT dispersion and tube breakage (78). Published Thermo-Electron Corp.) and obtained good distribution
with permission from Wiley-VCH. and dispersion (58).

634 POLYMER COMPOSITES, DECEMBER 2004, Vol. 25, No. 6

 Review of Polymer/Carbon Nanotube Composites

In-situ Polymerizatiom Aqueous solutions may also be used (155), depending

on the polymer (156). An aqueous poly(vinyl alcohol)
In-situ polymerization is a method used for improving
solution was mixed with an electrostatically stabilized
dispersion and integration between the phases (7, 131,
dispersion of CNTs in water, and was subsequently
132). The main advantage is that reinforcement is ob-
casted to form a composite film (157). Nanocomposites
tained on a molecular scale (100). This method has
may also be processed using latex as the matrix, utiliz-
been used with intrinsically conducting polymers: for
ing stable carbon nanotube aqueous suspensions as
example, solution electropolymerization of a conduct-
the filler system (158).
ing polymer film of CNT/polypyrrole (PPy), which is
Spin-coating is a common procedure in solution pro-
grown on an electrode while ionized CNTs serve as the
cessing. It has been used to make photovoltaic cells of
dopant (17); polyaniline (PANI) polymerized with CNTs
poly(p-phenylene vinylene) (PPV) and MWNTs. The PPV
(133
135); and poly(o-anisidine) (POAS) polymerized
was spin-coated on thin MWNT films (which were previ-
with CNTs by an oxidative technique (136); poly(pheny-
ously spin-coated on a glass substrate) and subse-
lacetylene) with CNTs (137). This method may also be
quently converted at high temperatures, stimulating PPV
used on aligned CNT arrays (138) to obtain polymer/
penetration of the MWNTs. A laminated structure was
nanotube coaxial wires, by electrochemically deposit-
formed, separating the polymer and nanotube layers,
ing a concentric layer of conductive polymer onto each
avoiding short-circuiting (159, 160). Spin-coating has
nanotube (139). In studies involving epoxy, the CNTs
also been used for making field emitters and photovoltaic
were dispersed in a low-viscosity matrix, and the ma-
cells with SWNTs dispersed in poly(3-octylthiophene)
trix was subsequently cured (9, 34, 122, 140
143).
(P3OT) (161, 162), and for MWNT-epoxy thin films (163).
Other studies using this technique employed SWNTs in
A solution processing technique has been exploited by
UV-curable polymers, with possible applications in mi-
Fournet et al. (164) for removing impurities as well as
croelectromechanical systems fabrication (144, 145),
dispersing the nanotubes—CNTs interact preferentially
and SWNTs in polyimide (146). Barraza et al. (147) pre-
with poly(m-phenylene-vinylene-co-2,5-dioctyloxy-
pared SWNT-filled polystyrene by mini-emulsion poly-
p-phenylene-vinylene) (PmPV), whereas the graphitic
merization, in which an adsorbed layer on the CNTs
impurities settle out of solution. The solution is rich in
enhanced dispersion.
CNTs, and these can be easily processed and spin-
coated (164).
Solution Processing
Coating CNTs in solution may also be a promising
Solution-based methods provide an advantage path to tailor the surface. An example of this has been
through low viscosities, which facilitate mixing and dis- shown by Chen et al. (17). A remarkably uniform PPy
persion of the CNTs (100). Many studies have used coating was obtained on individual CNTs, enabling
these methods for processing of both thermoset and controlled modification of the outer surface of CNTs.
thermoplastic polymers. Biercuk et al. (49) dissolved The nanotubes were also joined together into a network
epoxy in a well-dispersed, ultrasonicated CNT suspen- by conducting Ppy. These structures may have potential
sion. The solvent was evaporated, and the epoxy was applications in CNT-based nanoelectronic applications.
subsequently cured to form a nanocomposite in which Processing in solution also allows for control of CNT
the CNT dispersion was good, but not ideal. DeMora alignment (165, 166). Jin et al. (167) dissolved polyhy-
et al. (148) and Qian et al. (24) dissolved polystyrene in droxyaminoether into a sonicated MWNT/chloroform
either tetrahydrofuran or toluene, then added carbon suspension, and then solution-cast a film. The film was
nanotubes and finally evaporated the solvent. Safadi subsequently stretched to align the nanotubes. Aspect
et al. (149) and Watts et al. (150) added MWNT to a so- ratio and nanotube rigidity are key properties—align-
lution of PS dissolved in toluene, and made composites ment was reduced when longer and more flexible
both by film-casting and spin-casting. Jin et al. (151) SWNTs were used (167, 168). Ichida et al. (169) aligned
produced various types of polymer-coated and poly- an SWNT/polystyrene composite film by a similar me-
mer-grafted MWNT solutions, in some cases evapo- chanical stretching technique. Sometimes, simple
rating the solvent and subsequently melt-mixing with shearing is sufficient to induce alignment (137, 170).
another polymer (152). Yudasaka et al. (89) used a mix- Electrospinning is another process promoting nan-
ture of SWNTs and PMMA in monochlorobenzene otube alignment: a high-voltage electric field generated
(MCB) for dispersion, purification and subsequent spin- between the polymer solution and a metallic screen
casting of the material. Du et al. (153) demonstrated use forms nanocomposite fibrils, which contain aligned
of a stable nanotube-ethanol suspension to form a car- CNTs (35, 171). Electrophoresis may be applied as
bon nanotube composite sheet using electrophoretic well: applying a voltage between a carbon fiber and a
deposition. nanotube suspension produces a SWNT fiber, which
A different approach involves soaking nanotube sheets is subsequently spun through a polymer solution (172).
(Buckypaper) in polymer solutions, followed by drying. Insoluble polymers are sometimes used in solution-
Results showed intercalation of the polymers into the based methods. Barrera (7) processed purified SWNTs
free volume, and a significant increase in mechanical in toluene and then mixed in dry ABS powder, which is
properties relating to enhanced load transmission insoluble in toluene. The toluene evaporated, leaving
between nanotubes in the presence of a polymer (154). the ABS powder coated with CNTs as a precursor for

POLYMER COMPOSITES, DECEMBER 2004, Vol. 25, No. 6 635

 O. Breuer and Uttandaraman Sundararaj

additional processing. When this mixture was subse- inducing local stress concentrations, and the benefits
quently processed at high shear rates, a well-dispersed of the CNT properties are lost (179). Second, the nan-
nanocomposite was obtained (7). otubes should be well dispersed. If they are poorly dis-
Haggenmueller et al. (14) used a combination of sol- persed, the nanotubes will fail by separation of the
vent-casting and melt-mixing to disperse up to 8 wt% bundle rather than by failure of the nanotube itself,
SWNTs in poly(methyl methacrylate) (PMMA). Solvent- resulting in significantly reduced strength (120
122).
cast composite film was broken into pieces and hot- Using three-point bending tests, it was shown that
pressed into new films, which were again broken into functionalized CNTs improve strength in epoxy com-
smaller fragments and pressed again. This procedure posites (51). The tensile stress increases by 25% when
was repeated as many as 25 times. The dispersion was 1 wt% MWNTs are added to polystyrene (24). Safadi et al.
found to improve as the number of steps was in- (149) found that at less than 1 vol% of nanotubes, there
creased, with some of the films showing increased elec- is a 40% increase in strength of PS/MWNT film com-
trical conductivity. When drawing fibers from these posites. The modulus improves by a factor of 2 with re-
materials, good SWNT alignment along the fiber axis spect to the neat polymer when 2.49 vol% of CNTs are
was obtained, enhancing the mechanical properties added (149). Tibbetts and McHugh indicated that the ad-
(14, 172, 173). dition of VGCFs to a polypropylene (PP) matrix had no
effect on tensile strength, probably because of the in-
Other Methods creased brittleness of the polymer matrix, which resulted
Yet more exotic CNT composites can be obtained from PP’s inability to further crystallize on deformation,
using techniques such as chemical vapor deposition a behavior due to the molecular restrictions caused by
(CVD), by which arrays of CNTs can be grown onto the fiber dispersion (58). Alignment of CNTs in a PP ma-
poly(dimethylsiloxane) (PDMS) substrates. The thermal trix resulted in higher tensile strength but had no effect
shrinking property of PDMS at elevated temperatures on the modulus (61). At 5-wt% VGCF loading, modulus
allows fabrication of complex 3D networks of nan- and compressive strength of the spun PP increased by
otubes used for making flexible nanodevices such as 50% and 100%, respectively (55), as shown in Table 1.
highly sensitive chemical gas sensors (173, 174). An- Significant increase in stiffness was found at high tem-
other unique study utilizes CVD to grow CNTs directly peratures for MWNT/ PMMA composites, from 5-fold
onto carbon fibers that are later embedded in a polymer, to 27-fold, depending on the MWNT content (10). For
resulting in a hybrid multiscale composite in which the VGCFs, a modest amount of oxidation increases the
fiber/matrix interface is reinforced by the nanotubes tensile strength, whereas more etching may decrease it
(175). Wu et al. (176) used matrix-assisted pulsed laser (180). Impact strength was found to improve signifi-
evaporation to form composite films of SWNTs with cantly, even upon addition of extremely small amounts
polystyrene or polyethylene glycol. Hassanien et al. of nanotubes (30). Cooper et al. thought this was re-
(177) synthesized large-scale, well-aligned carbon nan- lated to the lack of interfacial adhesion, which in-
otube arrays onto which conducting polymers were creases the toughness and the impact strength, ac-
deposited electrochemically. cordingly (30). Biercuk et al. (49) found that the Vickers
hardness rose monotonically with SWNT loading up
NANOCOMPOSITE PROPERTIES to a factor of 3.5 at 2 wt%. Finally, using a finite ele-
It has already been established that CNTs possess ment method with a 3D nanoscale representative vol-
remarkable properties. The main challenge is to be able ume element based on continuum mechanics, Liu and
to implement these properties in composites on a Chen demonstrated the significant load-carrying ca-
macroscale, combining the choice of materials with the pacities of CNTs in a polymer matrix (181).
appropriate processing method (178). Recent research has shown that nanotubes aligned
perpendicularly to a crack are able to slow down its
Mechanical Properties propagation by bridging the crack faces. Thus, these
The strength of composites depends on two variables. materials may be used to improve the out-of-plane and
First, there should be a high degree of load transfer interlaminar properties of advanced composite struc-
between the matrix and the nanotubes. When the in- tures by increasing the matrix strength and linking up
terfacial adhesion between the phases is weak, the the individual laminar layers with these tiny, pin-like
nanotubes behave as holes or nanostructured flaws, structures (179).

Table 1. Properties of Fibers From Polypropylene (PP) and PP/CNT Composites (55)
(Adapted and Published With Permission From Elsevier).

Tensile Strength Tensile Modulus Elongation to Compressive

Sample (MPa) (GPa) Break (%) Strength (MPa)

PP-control 490  60 4.6  0.7 23  5 25  1

PP  5 wt% nano carbon fiber 570  70 7.1  0.9 16  2 48  10

636 POLYMER COMPOSITES, DECEMBER 2004, Vol. 25, No. 6

 Review of Polymer/Carbon Nanotube Composites

Fiber Surface Properties

To assess the performance of nanotubes as fillers,
the issue of load transfer through the interface must be
addressed. In traditional carbon fibers, surface treat-
ments are used to improve interfacial bonding by me-
chanical interlocking, chemical bonding and van der
Waals interactions. In the case of CNTs, especially
SWNTs, the question arises whether they can be func-
tionalized in a similar way without creating too many
defects that can compromise their properties (39). Dif-
ferent failure modes have been observed for various
CNT/polymer composites, giving clues about the inter-
Fig. 6. Surface electrical resistivity of MWNT/polymer com-
face quality. Wagner et al. (182) reported fragmentation posite films (78). Published with permission from Wiley-VCH.
of nanotubes as a result of tensile loading of an
MWNT/urethane/diacrylate nanocomposite film, prov-
ing good load transfer and interfacial shear strength as
high as 500 MPa (182). Another report showed that The CNTs make conjugated polymers mechanically
load transfer from the matrix to the nanotubes is in- stronger, more conductive and less susceptible to ther-
sufficient to fracture the nanotubes, resulting in pull- mal degradation (199, 200). Composite materials based
out of the nanotubes and mechanical fracture of the on CNTs in PmPV exhibited nearly an eightfold increase
polymer matrix (38). Lau and Shi (183, 184) showed the in electrical conductivity compared to the neat polymer,
reduction in flexural strength of a CNT-epoxy nano- without impairing the photo- and electro-luminescence
composite relative to neat epoxy, probably a result of a properties of the polymer (113).
weak interface. CNT/epoxy composites show increases in thermal
Burton et al. (54) studied the effect of surface treat- conductivity, K, of 110%
120% when 1% of SWNTs or
ments on VGCF/polypropylene (PP) composites. The 2% VGCFs are incorporated (see Fig. 7 ) (49, 140). This
VGCFs were oxidized to obtain optimal values of surface suggests easier formation of a percolation network for
area and surface energy to provide significant reinforce- SWNTs due to their larger aspect ratio as compared to
ment in PP composites. Sonication of polyphenylene- VGCFs, and the same is true for VGCFs versus tradi-
vinylene (PPV) in solution with CNTs yielded composites tional carbon fibers (123). The coefficient of thermal ex-
with excellent interfacial wetting, as seen by TEM (185). pansion in the longitudinal direction of flow (CTE) is
Another conjugated polymer, PmPV, has interesting significantly reduced when the nanocomposite is injec-
interactions with CNTs — good wetting, nucleation of tion molded (54). CTE reduction enables reduced cure
crystalline growth by defects in the nanotubes, and a shrinkage and hygrothermal instability in these sys-
structured wrapping of the polymer onto the lattice tube tems, resulting in better dimensional stability of the
due to polymer helicity (95, 186). Srivastava, Wei and material (8). The nanotubes delay the onset of thermal
Cho (32) used molecular dynamics tools to provide degradation of the polymer—for example, PMMA (10) or
insight on interfacial phenomena. They found that bond- PVOH (152, 154). Thermal stability may be affected by
ing between polyethylene and CNT is energetically favor- the presence of the nanotubes, which interfere with the
able, since chemical bonds can be formed at multiple mobility and crystallization of the polymer chains (56,
sites (32, 187, 188). 58, 202, 203), and enable the composite to be used for
high temperature applications (204). Wei et al. (187)
Electrical and Thermal Properties used classical molecular dynamics to show that addi-
tion of CNTs into a polymer matrix increases the glass
The excellent electrical and thermal properties of
transition temperature, and increases the thermal ex-
nanotubes are particularly important in nanocompos-
pansion and diffusion coefficients in the composite
ites. Conductivity can be obtained in a CNT/polymer
above Tg . These property enhancements have implica-
composite at extremely low CNT contents because of
tions in coating and painting applications, because
the large CNT aspect ratio. These composites exhibit
above the Tg the composite mobility is increased, al-
percolation-type behavior (78, 189
192) as shown in
lowing smoother flow of the material through nozzles.
Fig. 6. A wide range of values has been reported for per-
colation thresholds and conductivity of CNT compos-
Optical Properties
ites, depending on the processing method, polymer ma-
trix and nanotube type (193). Some results are shown It has been shown in CNT composites of -conjugated
in Table 2. Extremely low percolation thresholds are polymers that the polymer becomes photo-excited
attributed to the random orientation of the CNTs and under visible light and transfers electrons efficiently to
formation of network structures (14). nanotubes. This allows the preferential transfer of elec-
Significant enhancement of electrical properties is trons into the nanotube while leaving the holes to be
also obtained in nanocomposites based on intrinsically preferentially transported through the polymer. These
conducting polymers (94, 113, 133, 185, 191
201). systems alter the hole-transport mechanism and thus

POLYMER COMPOSITES, DECEMBER 2004, Vol. 25, No. 6 637

 O. Breuer and Uttandaraman Sundararaj

Table 2. Summary of Conductivity Results for Polymer/CNT Composites.

Filler Percolation Volume or Processing

Polymer Type Threshold Surface Resistivity Method Reference

PP VGCF 9
18 wt% 109 · cm Melt mixing in 60

internal mixer

PP VGCF 5 vol% 101
102 · cm Melt mixing, 62

injection molding

PP VGCF  1 vol% 100 · cm Melt mixing in 180

MiniMAX molder

PP Nylon VGCF  3 vol% 0.15 · cm Melt mixing in 103

MiniMAX molder

PP PS MWNT 0.05 vol% 105
1012 /square Melt mixing in 78

0.25 vol% 105 /square internal mixer

PS MWNT 2.49 vol% 103 /square Film casting 149

and spin casting

PC PC/15% MWNT 2 wt% 103 · cm Melt mixing in intermeshing 124

masterbatch co-rotating twin screw extruder

PMMA MWNT 0.5 wt% 103 · cm Spin coating 194, 195

SWNT 0.4 wt%

PA SWNT 3 vol% 10
1
100 Powder technology 196

· cm at 6 vol% methods

Field Emission
The power of nanotubes as field emitters has been ap-
parent from their inception (40, 48, 205
209), since
they show very low turn-on fields and high current den-
sities. Because they are so sharp, CNTs emit electrons
at lower voltages than electrodes made from other ma-
terials, and their strong carbon bonds allow nanotubes
to operate for longer periods without damage (43). The
full potential of CNTs as field emitters has not been fully
realized in practical devices because of the high surface
density required, which neutralizes their main advan-
tage, a high aspect ratio. Results show that the field-
emission threshold is lower and the emission current is
higher for composites in comparison to pure SWNT
mats. The explanation given for this is that the triple
junction between the SWNTs, polymer and vacuum
creates additional field emission (161).

Fig. 7. Enhancement in thermal conductivity relative to pristine ADVANTAGES AND APPLICATIONS

epoxy as a function of SWNT and VGCF loading (140). Pub- OF CNT NANOCOMPOSITES
lished with permission from The American Institute of Physics.
The mechanical properties of CNTs suggest that they
may be used as reinforcing fibers in high-toughness
the optical properties of the polymer. The interesting nanocomposites, where stiffness, strength and low
optical properties of nanotube-based composite system weight are important considerations (38). There are nu-
arise from the low dimensionality and unique elec- merous possible applications; some examples are aero-
tronic band structure of nanotubes. Such phenomena space structural panels, sporting goods, ultra-light-
do not occur in larger micron-scale carbon fibers (162, weight thin-walled space structures for use in space, and
163). Good quantum efficiency, about twice that of a high stiffness-to-weight space mirror substrates (51).
standard device, has been demonstrated in MWNT/ Applications relating to nonlinear optics include pro-
poly(p-phenylene vinylene) (PPV)-based systems. This tection of optical sensors from high-intensity laser
is due to the complex interpenetrating network of the beams. Here, we can utilize the broadband optical-
PPV chains with the MWNTs and the relatively high limiting properties of nanotubes and their composites,
work function of the MWNT film (158). as these materials are especially stable towards air and

638 POLYMER COMPOSITES, DECEMBER 2004, Vol. 25, No. 6

 Review of Polymer/Carbon Nanotube Composites

laser radiation (151). Short lifetime has been a signifi- offers more security during handling and transporta-
cant problem in the optical-limiting and light-emitting tion since it is less prone to dents, chips and scratches
devices consisting of polymers; incorporating nanotubes (56). Finally, recycling is possible without compromis-
may increase the lifespan, and thus the commercial ing properties, since the nanofiller is not expected to
value, of these devices (137). break down as a result of subsequent processing (123).
Inexpensive, renewable energy sources are the driv- A new application for these types of nanocomposites
ing force for research efforts on photovoltaic device is automobile bumpers. The bumper will have good me-
technologies (209, 210), so organic solar cells are an chanical properties and lower weight than the standard
important application. Additional applications involv- fiberglass bumpers, since only 1
5 wt% CNT will be
ing the optical and electronic properties are electron- needed compared to 30 wt% or more of fiberglass. The
emitting flat-panel displays, electromechanical actua- motivation for using CNT nanocomposites for automo-
tors, light-emitting diodes; supercapacitors, field-effect bile bumpers is their multifunctionality, enabling good
transistors, subpicosecond optical switches and optical mechanical properties, low weight and electrical con-
limiters (139, 163, 164, 197, 211
218). Nanotubes are ductivity (80, 124, 148, 225). The conductivity of the
also applied to modify the luminescent behavior of bumper will allow for direct application of electrostatic
optically active polymers (6). spray of base and clear coats, eliminating the need for
The nanoscale of the reinforcement enables thinner, an additional primer coat prior to painting. Additional
flatter structures to be produced— a composite mirror, advantages are savings in paint consumption and re-
for example. Conventional carbon fibers with a diame- ductions in volatile emissions from paint lines. Electro-
ter of approximately 8 m protrude from the surface of static painting may be extended even further to mirror
the composite mirror (referred to as “print-through”), shells, door handles and grills (56, 218). This has been
and reduce its optical quality. This is not the case for recently demonstrated by Hyperion Catalysis Interna-
CNTs, whose print-through is smaller than the wave- tional Inc., who manufactured plastic side view mirror
length of visible light (8). The small scale may also be housings for Ford Taurus™ in order to increase con-
utilized for spinning high performance fibers based on ductivity for electrostatic painting (8).
nanotubes combined with commodity polymers such The high thermal conductivity has a number of ap-
as polypropylene (24). plications, particularly in heat sinks for electronics,
CNTs are especially efficient as a conductive additive motors (49, 140) and tires (8). This property is espe-
because of their electrical properties, large aspect ratio cially important in thermal management applications
and their tendency to become entangled into a three- and in precision structures where thermally induced
dimensional, interconnecting network in the molten distortions reduce the ability to hold tolerances (8).
plastic (123, 219). Nanotube-filled polymers could Multifunctional applications include a housing for elec-
potentially be used for electrostatic dissipation (ESD), tronics in a satellite that needs to be mechanically sup-
electrostatic painting and EMI-shielding applications ported, to have excess heat transported away, and to be
(8, 54, 56, 220). They have advantages over metals: protected by EMI. Other examples include batteries,
greater flexibility, space and weight savings, mechanical and structures used to store hydrogen for fuel cells (8).
strength, better durability, and capability to tailor ther- Additional applications are: membranes for molecular
mal and electrical conductivity. These properties enable separations, which utilize the very small pores existing
production of advanced on-board shielding of electronic in the nanotubes (1, 226); abrasive resistance layers for
components and flexible electronics (221), and electro- surface engineering of composites (227); biomaterial
magnetic shields for mobile phone handsets (47). applications, such as drug delivery, which use the in-
Conducting polymer structures can be constructed ertness of SWNTs (176); nanosensors inside a polymer
at low loadings of nanotube fillers owing to lower per- for monitoring and quantifying deformation (143,
colation thresholds needed for high aspect ratio fillers 228
230) or for detecting polymer transitions (231);
(78, 125, 180, 190, 222, 223). Typical values are less light bulletproof vests; earthquake resistant buildings
than 5-wt% loading for nanotubes, as opposed to 8
20 exploiting the unique flexibility of the CNTs (40); smart
wt% for most carbon black containing compounds polymer coatings to protect components under extreme
(124). Low particle contents are also important for elec- physical conditions such as microsatellites (187, 188);
tronic cleanroom applications, where particle contami- and plastics requiring strict tolerances (dimensional
nation is a major problem (126). Conducting polymer stability) (121). Radio frequency (RF) permittivity spec-
systems can be designed to be multifunctional, com- tra show that CNT nanocomposites are also suitable for
bining electrical, thermal and structural properties use as microwave lenses, high-strength, low-weight elec-
(180, 224). tromagnetic interference shielding materials, antennas
CNT/thermoplastic ESD composites offer a number and waveguides (4).
of additional advantages over other materials. The nan-
otubes can be added in the compounding stage without TASKS FOR THE FUTURE
compromising the mechanical and processing properties Although nanotube-filled polymers show obvious
of the matrix. The finished product is lighter in weight, application for high-strength, lightweight and high-
appealing in surface finish, corrosion-resistant, easier performance composites, there have not been many
to handle, and less expensive to ship. The composite industrial successes showing their advantage over

POLYMER COMPOSITES, DECEMBER 2004, Vol. 25, No. 6 639

 O. Breuer and Uttandaraman Sundararaj

traditional carbon fibers. How can this be changed? applied to SWNT composites (7). Rapid upscaling of
Below, we discuss some of the challenges and ideas to CNT manufacturing on one hand, and development of
overcome them. new and efficient small-scale mixers on the other (236),
will resolve this issue.
Processing and Tailoring To date, several companies are working on com-
Because of their nanometer scale and high aspect mercially viable processes for mass production of
ratio, fabrication and investigation of nanotube/polymer CNTs, striving to lower costs. For example, Carbon
composites present considerable challenges (38). The Nanotechnologies Inc. (CNI) aspires to produce 1000 kg
main problem is in creating a good interface between on nanotubes weekly by 2004 (237). The goal is to
nanotubes and the polymer matrix to achieve the load achieve low-cost, high-yield, and large-scale produc-
transfer necessary for enhanced mechanical response in tion (19), resulting in high-purity material with control-
high-performance polymers (35, 145). CNTs are usually lable length, diameter and chirality (238). Several new
supplied in very entangled bundles, making handling methods have recently been suggested to reach this
and dispersion during composite processing a formida- goal (239
242). The most critical factor is production
ble hurdle (6, 8, 61, 78, 225). To optimize CNT compos- cost. Until cost is reduced to a competitive level with ex-
ites, we must maintain high CNT aspect ratios, orient isting fibers, large-scale use of nanotubes is unlikely
the CNTs, enhance interfacial interactions and elimi- (26). Additional issues are quality, which requires reli-
nate slippage between the CNTs in ropes or between the able characterization methods still under development,
layers in MWNTs (78). how to handle CNTs and health risks associated with
Common approaches to dispersion include both me- CNT exposure (8).
chanical (e.g., shear mixing) and chemical (e.g., surfac-
tants) means. Additional ways of overcoming these Characterization
problems are suggested in the literature (222): for ex-
Nanotube dispersion in polymers can affect the final
ample, breaking up the aggregate and suspending it in
nanocomposite properties. To optimize nanotube dis-
solution with the aid of ultrasonic mixers (232); or re-
persion, we need to investigate and quantify the dis-
inforcing the bundles by crosslinking the tubes within
persion. This can be done by visualizing the nanotubes
the bundles themselves to eliminate tube slippage and
themselves, the interface, and the effect of the CNTs on
to enhance rigidity. Another method for enhancing dis-
the surrounding matrix. It is not surprising that at the
persion may be to use CNT “fibers,” in which large num-
nanolength scale, characterization, testing and model-
bers of nanotubes are assembled in arrays or “yarns”
ing are much more difficult (8). Optical techniques,
such that the nanotube properties are translated from
typically used for characterizing dispersion, cannot dis-
the nano to the micro scale. Helicity can be introduced
cern between single CNTs. Electron microscopy tech-
to enhance load transfer between the neighboring nan-
niques are used but are not always easy to implement
otubes (233). Yet another approach is to uniformly coat
due to issues with sample preparation and a lack of
an aligned array of CNTs with polymer, rather than in-
contrast between the polymer and the CNT. Thus, new
corporating randomly entangled CNTs into a matrix
methods are required. Magnetic force microscopy (MFM)
(139, 234).
has been used, exploiting the interaction of the CNTs
Maintaining a strong nanotube/polymer interface is
with a magnetic field, allowing direct visualizaton of the
always important and crucial for good mechanical
nanotubes in the polymer (145). Scanning Kelvin mi-
properties (235). This can be done through selectively
croscopy measures the conductivity distribution in het-
functionalizing nanotube ends and mid-section, and
erogeneous materials with an oscillating probe and can
subsequently chemically binding them to the polymer
be used to characterize CNT distribution in an insulat-
(7, 115). Another approach is to overcome low shear
ing matrix (243).
properties of MWNTs and SWNT ropes by inducing
crosslinking between adjacent shells to increase the
Understanding the Mechanisms
strength and intra-tube load-transfer efficiency (39). In
addition, when MWNTs are used, efficient load bearing One important issue involving most nanocomposites
within the tubes themselves must be accomplished in is whether their unique properties can be attributed
such a way that shearing between the concentric to the small dimensions of the CNTs, or to their large
graphene layers is prevented (27, 40). aspect ratios. Simulations have been used to assess
whether the mobility of polymer segments changes in
Nanotube Availability the vicinity of CNTs because of their similar size scale.
Since very little SWNT material is available, most in- Simulations have also been used to determine whether
vestigators have been limited to working with milligram the significant surface area of CNTs promotes interac-
quantities. This may be problematic, since benchtop tions sufficient for load transfer between the phases
processing may not always be representative of larger- (15, 244, 245), and how chemical bonds between the
scale processing. This problem has been partially nanotube and matrix can enhance the nanocomposite
solved by working with vapor-grown carbon fibers shear strength (246).
(VGCFs), a cheaper and more readily available material The load-transfer mechanism is another issue to be
similar to CNTs, to build a knowledge base that can be addressed, using both experiments and simulation.

640 POLYMER COMPOSITES, DECEMBER 2004, Vol. 25, No. 6

 Review of Polymer/Carbon Nanotube Composites

Experimental results show that the strength of CNT 3. R. Saito, G. Dresselhaus, and M. S. Dresselhaus, Phys-
composites is lower than that predicted by theory. This ical Properties of Carbon Nanotubes, Imperial College
Press, London (1999).
may point to some basic differences in stress-transfer 4. C. A. Grimes, C. Mungle, D. Kouzoudis, S. Fang, and P.
mechanisms between conventional composites and C. Eklund, Chem. Phys. Lett., 319, 460 (2000).
nanocomposites (8). Studies have examined interac- 5. B. S. Files and B. M. Mayeaux, Adv. Mater. Proc., 156,
tions between the polymer and tubes (247, 248), en- 47 (1999).
tanglement of the polymer around the tubes, polymer 6. P. M. Ajayan and L. S. Schadler, Polym. Prep., 42, 35
(2001).
infiltrating a nanotube bundle (244) and CNT pullout 7. E. Barrera, JOM, 52, 38 (2000).
from a polymer matrix (235, 249, 250). A constitutive 8. B. Maruyama and K. Alam, SAMPE J., 38(3), May/June,
modeling approach has been taken for predicting the 59 (2002).
bulk mechanical properties of SWNT/polymer compos- 9. A. Allaoui, S. Bai, H. M. Cheng, and J. B. Bai, Comp. Sci.
Technol., 62, 1993 (2002).
ites with various nanotube lengths, concentrations and 10. J. Sandler, M. S. P. Shaffer, Y.-M. Lam, C.-A. Keun,
orientations (251, 252). J. Nastalczyk, G. Broza, K. Schulte, and A. H. Windle,
http://www.msm.cam.ac.uk/polymer/members/js364
SUMMARY /js364Composites.pdf.
11. Z. Jin, K. P. Pramoda, G. Xu, and S. H. Goh, Chem.
Since carbon nanotubes were discovered in 1991, re- Phys. Lett., 337, 43 (2001).
searchers have envisioned them as important materials 12. C. Park, Z. Ounaies, K. A. Watson, K. Pawlowski, S. E.
Lowther, J. W. Connell, E. J. Siochi, J. S. Harrison, and
capable of dominating the 21st-century revolution in
T. L. St. Clair, Making Functional Materials with Nanotubes
nanotechnology. The excitement about CNTs stems Symposium, (Materials Research Society Symposium
from their unique properties; for example, an elastic Proceedings, Vol 706), pp. 91
96 (2002).
modulus higher than 1 TPa and strength near 200 GPa 13. R. Vajtai, B. Q. Wei, Z. J. Zhang, Y. Jung, G. Ramanath,
make SWNTs attractive candidates to act as fillers in and P. M. Ajayan, Smart Mater. Struct., 11, 691 (2002).
14. R. Haggenmueller, H. H. Gommans, A. G. Rinzler, J. E.
polymer-based composite materials. If incorporated Fischer, and K. I. Winey, Chem. Phys. Lett., 330, 219
into a polymer matrix, the resulting structures could be (2000).
substantially stronger and lighter than conventional 15. H. Dai, Surf. Sci., 500, 218 (2002).
aluminum, alloys and carbon fiber-reinforced polymer 16. F. T. Fisher and L. C. Brinson, http://www.tam.nwu.
edu/ftf234/nano/pubs/SEM2002_paper_web.pdf.
composites currently used in many applications. How-
17. G. Z. Chen, M. S. P. Shaffer, D. Coleby, G. Dixon, W. Zhou,
ever, the real value of CNTs lies in their range and D. J. Fray, and A. H. Windle, Adv. Mater., 12, 522 (2000).
breadth of properties, namely, their electrical, thermal, 18. M. S. Dresselhaus, G. Dresselhaus, P. Eklund, and R.
magnetic and optical properties. These properties pro- Saito, Physics World, January 1998, p. 33.
vide additional benefits, which enable development of 19. Y. Wang, F. Wei, G. H. Luo, H. Yu, and G. S. Gu, Chem.
Phys. Lett., 364, 568 (2002).
multifunctional, structural materials. Therefore, it is 20. F. H. Gojny, J. Nastalczyk, Z. Roslaniec, and K. Schulte,
easy to understand why CNT nanocomposites have pro- Chem. Phys. Lett., 370, 820 (2003).
duced so much interest among researchers, resulting in 21. J. B. Nagy, A. Fonesca, N. Pierard, L. Willems, G. Blister,
the large number of papers published recently (253). C. Pirlot, A. Demortier, J. Delhalle, Z. Mekhalif, K. Niesz,
C. Bossuot, J.-P. Pirard, L. P. Biro, Z. Konya, J.-F. Colomr,
In order to realize the potential of these materials,
G. Van Tendeloo, and L. Kiricsi, Electronic Properties of Mo-
some work must be done on developing mass produc- lecular Nanostructures: 15th International Winterschool/
tion techniques for CNTs and balancing the excellent Euroconference (AIP Conference Proceedings), 591, pp.
properties with reasonable costs. New technologies for 483
488, Kirchberg, Austria (March 2001).
manufacturing CNT composites must be developed, 22. D. E. Resasco, W. E. Alvarez, F. Pompeo, L. Balzano, J.
E. Herrera, B. Kitiyanan, and A. Borgna, J. Nanoparticle
and existing technologies should be improved to con- Research, 4, 131 (2002).
trol the dispersion and orientation of the CNTs within 23. M. S. Dresselhaus, Y. M. Lin, O. Rabin, A. Jorio, A. G.
the polymer. Methods to tailor the interface to aid dis- Souza Filho, M. A. Pimenta, R. Saito, G. G. Samsonidze,
persion and to provide good load transfer between the and G. Dresselhaus, Mater. Sci. Eng., C23, 129 (2003).
24. J. C. Kearns and R. L. Shambaugh, J. Appl. Polym. Sci.,
phases are also necessary.
86, 2079 (2002).
The field of materials science is currently undergoing 25. D. Qian, E. C. Dickey, R. Andrews, and T. Rantell, Appl.
a shift from developing traditional materials such as Phys. Lett., 76, 2868 (2000).
metals, ceramics, polymers and composites to a more 26. P. M. Ajayan and T. W. Ebbesen, Rep. Prog. Phys., 60,
revolutionary trend of developing nanostructures, 1025 (1997).
27. K.-T. Lau and D. Hui, Comp. Pt B Eng., 33, 263 (2002).
which are functionalized, self-assisting and occasionally 28. D. Qian, G. J. Wagner, W. K. Liu, M.-F. Yu, and R. S.
even self-healing materials. The question is whether Ruoff, Appl. Mech. Rev., 55, 495 (2002).
CNT polymer composites can meet these requirements: 29. E. T. Thostenson, Z. F. Ren, and T. W. Chou, Comp. Sci.
they appear attractive for many applications; however, Technol., 61, 1899 (2001).
30. C. A. Cooper, D. Ravich, D. Lips, J. Mayer, and H. D.
significant breakthroughs still need to be realized.
Wagner, Comp. Sci. Technol., 62, 1105 (2002).
31. M. Meyyappan and D. Srivastava, IEEE Potentials, Au-
REFERENCES gust/September, 16
19 (2000).
1. S. Iijima, Nature, 354, 56 (1991). 32. D. Srivastava, 2nd Annual Nanostructured Materials
2. P. M. Ajayan and O. Z. Zhou, in Carbon Nanotubes: Syn- 2001– Application and Commercialization, Chicago
thesis, Structure, Properties, and Applications, Chap. 13, (June 2001).
M. S. Dresselhaus, G. Dresselhaus, and Ph. Avouris, 33. M. Sennett, E. Welsh, J. B. Wright, W. Z. Li, J. G. Wen,
eds., Springer-Verlag, Berlin (2000). and Z. F. Ren, Appl. Phys. A., 76, 111 (2003).

POLYMER COMPOSITES, DECEMBER 2004, Vol. 25, No. 6 641

 O. Breuer and Uttandaraman Sundararaj

34. V. G. Hadijev, M. N. Iliev, S. Arepali, P. Nikolaev, and 68. P. Glatkowski, P. Mack, J. L. Conroy, J. W. Piche, and
B. S. Files, Appl. Phys. Lett., 78, 3193 (2001). P. Winsor, U.S. Patent 6,265,466 (2001).
35. F. K. Ko, W. B. Han, S. Khan, A. Rahman, and O. Zhou, 69. M. Dupire and J. Michel, U.S. Patent 6,331,265 (2001).
Proceedings of the American Society for Composites 16th 70. J. G. Gerard and H. V. Samuelson, U.S. Patent 6,426,134
Annual Technical Conference, Blacksburg, Va. (September (2002).
2001). 71. S. H. Foulger, U.S. Patent 6,417,235 (2002).
36. F. Li, H. M. Cheng, S. Bai, G. Su, and M. S. Dresselhaus, 72. M. Dupire and J. Michel, U.S. Patent 6,331,265 (2001).
Appl. Phys. Lett., 77, 3161 (2000). 73. A. Fisher, R. Hoch, D. Moy, C. Niu, N. Ogata, H. Tennent,
37. O. Lourie, D. M. Cox, and H. D. Wagner, Phys. Rev. and T. McCarthy, U.S. Patent 6,203,814 (2001).
Lett., 81, 1638 (1998). 74. Y. Zhao, Y. Chen, X.-C. Zhang, N. R. Raravikar, P. M.
38. C. Bower, R. Rosen, L. Jin, J. Han, and O. Zhou, Appl. Ajayan, T.-M. Lu, G.-C. Wang, and L. S. Schadler Feist,
Phys. Lett., 74, 3317 (1999). U.S. Patent Application US 2002/0176650 A1, 2002.
39. J. P. Salvetat, J. M. Bonard, N. H. Thomson, A. J. Kulik, 75. R. E. Smalley, D. T. Colbert, H. Dai, J. Liu, A. G. Rinzler,
L. Forro, W. Benoit, and L. Zupiroli, Appl. Phys. A., 69, J. H. Hafner, K. Smith, T. Guo, P. Nikolaev, and A. Thess,
255 (1999). U.S. Patent Application US 2002/0127169 A1, 2002.
40. C. Schonenberger and L. Forro, Phys. World, 13, 37 76. E. T. Thostenson and T.-W. Chou, J. Phys. D: Appl. Phys.,
(2000). 35, L77 (2002).
41. P. L. McEuan, Physics World, June 2000. 77. R. Andrews, D. Jacques, A. M. Rao, T. Rantell, F. Derby-
42. P. L. McEuan, M. Fuhrer, and H. Park, IEEE Transaction shire, Y. Chen, J. Chen, and R. C. Haddon, Appl. Phys.
on Technology, 1, 78 (2002). Lett., 75, 1329 (1999).
43. P. G. Collins and P. Avouris, Sci. Am., 283, 62 (2000). 78. R. Andrews, D. Jacques, M. Minot, and T. Rantell,
44. P. Avouris, Chem. Phys., 281, 429 (2002). Macromol. Mater. Eng., 287, 395 (2002).
45. J. Kong, C. Zhou, A. Morpurgo, H. T. Soh, C. F. Quate, 79. B. P. Grady, F. Pompeo, R. L. Shambaugh, and D. E. Re-
C. Marcus, and H. Dai, Appl. Phys. A., 69, 305 (1999). sasco, J. Phys. Chem. B., 106, 5852 (2002).
46. C. Dekker, Physics Today, 52, 22 (1999). 80. R. Mulhaupt and N. Schall, Kunststoffe, 91, 178 (2001).
47. K. Atkinson, A. Minett, and S. Roth, Extended Abstracts 81. A. A. Mamedov, N. A. Kotov, M. Porato, D. M. Guldi, J. O.
of the Fourth German-Vietnamese Seminar on Physics Wicksted, and A. Hirsch, Nature Mater., 1, 190 (2002).
and Engineering (GVS4), Dresden (June 2001). 82. B. Vigolo, A. Penicaud, C. Coulon, C. Sauder, R. Pailler,
48. E. J. Lerner, The Industrial Physicist, December 1999, C. Journet, P. Bernier, and P. Poulin, Mat. Res. Soc.
p. 22. Symp. Proc., 633, A12.1.1 (2001).
49. M. J. Biercuk, M. C. Llaguno, M. Radosavljevic, J. K. 83. M. Zhang, M. Yudasaka, A. Koshio, C. Jabs, T. Ichihashi,
Hyun, A. T. Johnson, and J. E. Fischer, Appl. Phys. and S. Iijima, Appl. Phys. A., 74, 7 (2002).
Lett., 80, 2767 (2002). 84. A. S. Lobach, N. G. Spitsina, S. V. Terehov, and E. D.
50. S. Lefrant, J.-P. Buisson, O. Chauvet, J.-M. Baibarac, Obraztsova, Phys. Sol. Stat., 44, 475 (2002).
I. Baltog, and P. Bernier, Making Functional Materials 85. M. Rao, R. Andrews, and F. Derbyshire, Energeia, 9, 1
with Nanotubes Symposium (Materials Research Society (1998).
Symposium Proceedings, Vol. 706 ), pp. 231
238 (2002). 86. J.-M. Moon, K. H. An, Y. H. Lee, Y. S. Park, D. J. Bae,
51. T. Tiano, M. Roylance, and J. Gassner, 32nd International and G.-S. Park, J. Phys. Chem. B., 105, 5677 (2001).
SAMPE Technical Conference, 192 (November 2000). 87. S. Bandow, S. Asaka, X. Zhao, and Y. Ando, Appl. Phys.
52. S. B. Sinnott and R. Andrews, Crit. Rev. Solid State, 26, A., 67, 23 (1997).
145 (2001). 88. G. S. Duesberg, J. Muster, H. J. Byrne, S. Roth, and M.
53. T. Saito, K. Matsushige, and K. Tanaka, Physica B., Burghard, Appl. Phys. A., 69, 305 (1999).
323, 280 (2002). 89. M. Yudasaka, M. Zhang, C. Jabs, and S. Iijima, Appl.
54. D. J. Burton, D. G. Glasgow, M. L. Lake, C. Kwag, and Phys. A., 71, 449 (2000).
J. C. Finegan, 46th International SAMPE Symposium 90. H. Ago, T. Kugler, F. Cacialli, W. R. Salaneck, M. S. P.
and Exhibition: Materials and Processes Odyssey, Long Shaffer, A. H. Windle, and R. H. Friend, J. Phys. Chem.
Beach, Calif. (May 2001). B., 103, 8116 (1999).
55. S. Kumar, H. Doshi, M. Srinivasaro, J. O. Park, and 91. J. N. Coleman, A. B. Dalton, S. Curran, A. Rubio, A. P.
D. A. Schiraldi, Polymer, 43, 1701 (2002). Davey, A. Drury, B. McCarthy, B. Lahr, P. M. Ajayan, S.
56. K. Lozano, JOM, 52, 34 (2000). Roth, R. C. Barklie, and W. J. Blau, Adv. Mater., 12,
57. G. G. Tibbetts and J. J. McHugh, J. Mater. Res., 14, 213 (2000).
2871 (1999). 92. R. Murphy, J. N. Coleman, M. Cadek, B. McCarthy, M.
58. K. Lozano and E. V. Barrera, J. Appl. Polym. Sci., 79, Bent, A. Drury, R. C. Barklie, and W. J. Blau, J. Phys.
125 (2001). Chem. B., 106, 3087 (2002).
59. O. S. Carneiro and J. M. Maia, Polym. Compos., 21, 960 93. B. McCarthy, J. N. Coleman, R. Czerw, A. B. Dalton,
(2000). H. J. Byrne, D. Tekleab, P. Iyer, P. M. Ajayan, W. J. Blau,
60. K. Lozano, J. Bonilla Rios, and E. V. Barrera, J. Appl. and D. L. Carroll, Nanotechnol., 12, 187 (2001).
Polym. Sci., 80, 1162 (2001). 94. B. McCarthy, A. B. Dalton, J. N. Coleman, H. J. Byrne,
61. R. J. Kuriger, M. K. Alam, D. P. Anderson, and R. L. Ja- P. Bernier, and W. J. Blau, Chem. Phys. Lett., 350, 27
cobsen, Comp. Pt. A., 33, 53 (2002). (2001).
62. S. A. Gordeyev, F. J. Macedo, J. A. Ferreira, F. W. J. van 95. B. McCarthy, J. N. Coleman, R. Czerw, A. B. Dalton, M.
Hattum, and C. A. Bernardo, Physica B., 279, 33 (2000). in het Panhuis, A. Maiti, A. Drury, P. Bernier, J. B. Nagy,
63. D. R. Mitchell, R. M. Brown Jr., T. L. Spires, D. K. Ro- B. Lahr, H. J. Byrne, D. L. Carroll, and W. J. Blau,
manovicz, and R. J. Lagow, Inorg. Chem., 40, 2751 (2001). J. Phys. Chem. B., 106, 2210 (2002).
64. P. E. Anderson, E. Engel, A. Crowe, C. Park, and N. M. 96. A. C. Dillon, T. Gennet, K. M. Jones, J. L. Alleman, P. A.
Rodriguez, WM ’00 Conference, Tucson, Ariz. (Feb. Parilla, and M. J. Heben, Adv. Mater., 11, 1354 (1999).
27
Mar. 2, 2000). 97. M. Holzinger, A. Hirsch, P. Bernoer, G. S. Duesberg, and
65. Y. Wang, S. Serrano, and J. J. Santiago-Aviles, Syn- M. Burghard, Appl. Phys. A., 70, 599 (2000).
thetic Met., 138, 423 (2003). 98. E. V. Barrera, L. P. Felipe Chibante, B. Collins, F.
66. R. T. K. Baker, http://itri.loyola.edu/nano/US.Review/ Rodriguez-Marcias, M. Shofner, J. D. Kim, and F. D. S.
09_03.htm. Marquis, Powder Materials: Current Research and In-
67. E. T. Thostenson, W. Z. Li, D. Z. Wang, Z. F. Ren, and dustrials Practices (conference), Indianapolis (November
T. W. Chou, J. Appl. Phys., 91, 6034 (2002). 2001).

642 POLYMER COMPOSITES, DECEMBER 2004, Vol. 25, No. 6

 Review of Polymer/Carbon Nanotube Composites

99. A. G. Rinzler, J. Liu, H. Dai, P. Nikolaev, C. B. Hoffman, 125. R. Andrews, http://www.carbon2001.org/agenda/

E. J. Rodriguez-Macias, P. J. Boul, A. H. Lu, D. Hey- papers/session%2037/37.3.pdf.
mann, D. T. Colbert, R. S. Lee, J. E. Fischer, A. M. Rao, 126. D. W. Ferguson, E. W. S. Bryant, and H. C. Fowler, SPE
P. C. Eklund, and R. E. Smalley, Appl. Phys. A., 67, 29 ANTEC ’98, 1219 (1998).
(1998). 127. K. K. Schwartz and E. W. S. Bryant, SPE ANTEC ’95,
100. Y. K. Kim, A. F. Lewis, P. K. Patra, S. B. Warner, and P. 1358 (1995).
Calvert, National Textile Center Research briefs-Materials 128. J. A. King, K. W. Tucker, J. D. Meyers, E. H. Weber, M.
Competency, p. 24 (June 2001). L. Clingerman, and K. R. Ambrosius, Polym. Compos.,
101. C. Pirlot, I. Willems, A. Fonesca, J. B. Nagy, and J. Del- 22, 142 (2001).
halle, Adv. Eng. Mater., 4, 109 (2002). 129. M. Sennett, E. Welsh, J. B. Wright, W. Z. Li, J. G. Wen,
102. R. Tucknott and S. N. Yaliraki, Chem. Phys., 281, 455 and Z. F. Ren, Making Functional Materials with Nan-
(2002). otubes Symposium (Materials Research Society Sympo-
103. I. C. Finegan and G. G. Tibbetts, J. Mater. Sci. Res., 16, sium Proceedings Vol. 706 ), pp. 907
1002, Watersdale,
1668 (2001). Pa. (2002).
104. Y. A. Kim, T. Hayashi, Y. Fukai, M. Endo, T. Yanagisawa, 130. U. Sundararaj, C. W. Macosko, A. Nakayama, and
and M. S. Dresselhaus, Chem. Phys. Lett., 355, 279 T. Inoue, Polym. Eng. Sci., 35, 100 (1995).
(2002). 131. Z. Jia, Z. Wang, C. Xu, J. Liang, B. Wei, D. Wu, and
105. A. Hirsch, Angew. Chem. Ind. Ed., 41, 1853 (2002). S. Zhu, Mater. Sci. Eng., A271, 395 (1999).
106. H. Goering, H.-E. Maneck, U. Knoll, S. Gajewski, K.-W. 132. C. Park, Z. Ounaies, K. A. Watson, R. E. Crooks, J. Smith
Brzezinka, R. Mach, H.-D. Klotz, A. Schonhals, and J. F. Jr., S. E. Lowther, J. W. Connell, E. J. Siochi, J. S. Har-
Friedrich, http://www.nanotube.org/abs/GoerinH.html. rison, and T. L. St. Clair, Chem. Phys. Lett., 364, 303
107. Y. Chen, R. C. Haddon, S. Fang, A. M. Rao, P. C. Eklund, (2002).
W. H. Lee, E. C. Dickey, E. A. Grulke, J. C. Pendergrass, 133. M. Cochet, W. K. Mase, A. M. Benito, M. A. Callejas, M.
A. Chavan, B. E. Haley, and R. E. Smalley, J. Mater. T. Martinez, J.-M. Benoit, J. Schriber, S. Lefrant, and O.
Res., 13, 2423 (1998). Chauvet, http://www.cmp-cientifica.com/cientifica/
108. C. Velasco-Santos, A. L. Martinez, M. Lozada-Cassou, frameworks/generic/public_users/tnt2k1/Abstracts-
A. Alvarez-Castillo, and V. M. Castano, Nanotechnology, poster/Martinez.pdf.
13, 495 (2002). 134. W. K. Maser, A. M. Benito, M. A. Callejas, T. Seeger,
109. S. B. Sinnott and J. Nanosci, Nanotechnology, 2, 113 M. T. Martinez, J. Schreiber, J. Muszynski, O. Chauvet,
(2002). Z. Osvath, A. A. Koos, and L. P. Biro, Mater. Sci. Eng.,
110. U. Dettlaff-Weglikowska, J.-M. Benoit, P.-W. Chiu, C23, 87 (2003).
R. Graupner, S. Lebedkin, and S. Roth, Curr. Appl. 135. J. Deng, X. Ding, W. Zhang, Y. Peng, J. Wang, X, Long,
Phys., 2, 497 (2002). P. Li, and A. S. C. Chan, Eur. Polym. J., 38, 2497 (2002).
111. M. Ichida, S. Mizuno, H. Kataura, Y. Achiba, and K. 136. B. Valter, M. K. Ram, and C. Nicolini, Langmuir, 18,
Nakamura, Making Functional Materials with Nanotubes 1535 (2002).
Symposium (AIP Conference Proceedings), 590, pp. 137. B. Z. Tang and H. Xu, Macromolecules, 32, 2569 (1999).
121
24, Kamakura, Japan (January 2001). 138. Z. F. Ren, Z. P. Huang, D. Z. Wang, W. Z. Li, J. G. Wen, S.
112. D. E. Hill, Y. Lin, L. F. Allard, and Y.-P. Sun, Int. J. Nano., X. Yang, Y. Tu, and J. H. Chen, SAMPE J., 37, 64 (2001).
1, 213 (2002). 139. M. Gao, S. Huang, L. Dai, G. Wallace, R. Gao, and Z.
113. A. Star, J. F. Stoddart, D. Steuerman, M. Diehl, A. Boukai, Wang, Angew. Chem., 112, 3810 (2000).
E. W. Wong, X. Yang, S.-W. Chung, H. Choi, and J. R. 140. J. Hone, M. C. Llaguno, M. J. Bierck, A. T. Johnson,
Heath, Angew. Chem. Int. Ed., 40, 1721 (2001). B. Batlogg, Z. Benes, and J. E. Fischer, Appl. Phys. A.,
114. X. Gong, J. Liu, S. Baskaran, R. D. Voise, and J. S. Young, 74, 339 (2002).
Chem. Mater., 12, 1049 (2000). 141. P. M. Ajayan, O. Stephan, C. Colliex, and D. Trauth,
115. M. Bratcher, B. Gerstein, H. Ji, and J. Mays, Making Science, 265, 1212 (1994).
Functional Materials with Nanotubes Symposium (Materi- 142. L. Vaccarini, G. Desarmot, R. Almairac, S. Tahir, C.
als Research Society Symposium Proceedings, Vol. 706), Goze, and P. Bernier, in Electronic Properties of Novel
pp. 323
328, Watersdale, Pa. (2002). Materials–Molecular Nanostructures: XIV International
116. D. H. Yu, Y. Z. Yue, Jesper de C. Christiansen, S. Ka- Winterschool/Euroconference (no. 544), 521, H. Kuz-
zoui, and N. Minami, Proceedings of the Nordic Meeting many, J. Fink, M. Mehring, and S. Roth, eds., American
on Materials and Mechanics, pp. 147
151, Denmark Institute of Physics, Melville, N.Y. (2000).
(May 2000). 143. J. Sandler, M. S. P. Shaffer, T. Prasse, W. Brauhofer,
117. J. L. Bahr and J. M. Tour, Chem. Mater., 13, 3823 K. Schulte, and A. H. Windle, Polymer, 40, 5967 (1999).
(2001). 144. Q. Zhao, M. D. Frogley, and H. D. Wagner, Comp. Sci.
118. V. Georgakilas, K. Kordatos, M. Prato, D. M. Guldi, M. Technol., 62, 147 (2002).
Holzinger, and A. Hirsch, J. Am. Chem. Soc., 124, 760 145. J. Xie, N. Zhang, M. Guers, and V. K. Varadan, Smart
(2002). Mater. Struct., 11, 575 (2002).
119. Y. Lin, A. M. Rao, B. Sadanadan, E. A. Kenik, and Ya- 146. K. Lillehei, C. Park, J. H. Rouse, and E. J. Siochi, Nano
Ping Sun, J. Phys. Chem. B., 106, 1294 (2002). Lett., 2, 827 (2002).
120. J. P. Harmon, P. A. O. Muisener, L. Clayton, J. D’An- 147. H. J. Barraza, F. Pompeo, E. A. O’Rear, and D. E. Resasco,
gelo, A. K. Sikder, A. Kumar, M. Meyyaooan, and A. M. Nano Lett., 2, 797 (2002).
Cassell, Surface Engineering 2001–Fundamentals and 148. M. DeMora, C. Foley, M. Hamilton, B. Regner, B. Sakad-
Applications (Mater. Res. Society Symposium Proceed- jian, and B. Villarreal, http://kcgl1.eng.ohio-state.
ings, Vol. 697 ), pp. 425
435, Boston (November 2001). edu/~CE_REGNER/Proposal.htm.
121. H. Bubert, S. Haiber, W. Brandl, G. Marginean, M. 149. B. Safadi, R. Andrews, and E. A. Grulke, J. Appl. Polym.
Heintze, and V. Bruse, Diamond and Related Materials, Sci., 84, 2660 (2002).
12, 811 (2003). 150. P. C. P. Watts, W. K. Hsu, G. Z. Chen, D. J. Fray, H. W.
122. L. S. Schadler, S. C. Giannaris, and P. M. Ajayan, Appl. Kroto, and D. R. M. Walton, J. Mater. Chem., 11, 2482
Phys. Lett., 73, 3842 (1998). (2001).
123. J. Sandler, P. Werner, M. S. P. Shaffer, V. Demchuk, V. 151. Z. Jin, X. Sun, G. Xu, S. H. Goh, and W. Ji, Chem. Phys.
Altstadt, and A. H. Windle, Comp. A., 33, 1033 (2002). Lett., 318, 505 (2000).
124. P. Potschke, T. D. Fornes, and D. R. Paul, Polymer, 43, 152. Z. Jin, K. P. Pramoda, S. H. Goh, and G. Xu, Mater. Res.
3247 (2002). Bull., 37, 271 (2002).

POLYMER COMPOSITES, DECEMBER 2004, Vol. 25, No. 6 643

 O. Breuer and Uttandaraman Sundararaj

153. C. Du, D. Heldebrant, and N. Pan, J. Mater. Sci. Lett., 182. H. D. Wagner, O. Lourie, Y. Feldman, and R. Tenne,
21, 565 (2002). Appl. Phys. Lett., 72, 188 (1998).
154. J. N. Coleman, W. J. Blau, A. B. Dalton, E. Munoz, S. 183. K.-T. Lau and S.-Q. Shi, Carbon, 40, 2961 (2002).
Collins, B. G. Kim, J. Raazal, M. Selvidge, G. Vieiro, and 184. K.-T. Lau, S.-Q. Shi, and H.-M. Cheng, Comp. Sci. Tech-
R. H. Baughman, Appl. Phys. Lett., 82, 1682 (2003). nol., 63, 1161 (2003).
155. M. S. P. Shaffer and A. H. Windle, Carbon, 36, 1603 185. S. A. Curran, P. M. Ajayan, W. J. Blau, D. L. Carroll,
(1998). J. N. Coleman, A. B. Dalton, A. P. Davey, A. Drury, B.
156. M. S. P. Shaffer and A. H. Windle, Macromolecules, 32, McCarthy, S. Maier, and A. Strevens, Adv. Mater., 10,
6864 (1999). 1091 (1998).
157. M. S. P. Shaffer and A. H. Windle, Adv. Mater., 11, 937 186. D. W. Steuerman, A. Star, R. Narizzano, H. Choi, R. S.
(1999). Ries, C. Nicolini, J. F. Stoddart, and J. R. Heath, J.
158. M. Dufresne, M. Paillet, J. L. Putaux, R. Canet, F. Car- Phys. Chem. B., 106, 3124 (2002).
mona, P. Delhaes, and S. Cui, J. Mater. Sci., 37, 3915 187. C. Wei, D. Srivastava, and K. Cho, Nanoletters, 2, 647
(2002). (2002).
159. H. Ago, K. Petritsch, M. S. P. Shaffer, A. H. Windle, and 188. C. Wei, K. Cho, and D. Srivastava, MRS Spring Meeting
R. H. Friend, Adv. Mater., 11, 1281 (1999). 2001, p. 1., www-mc.stanford.edu/chogroup/publica-
160. H. Ago, M. S. Shaffer, D. S. Ginger, A. H. Windle, and tion/pub54.pdf.
R. H. Friend, Phys. Rev. B., 61, 2286 (2000). 189. K. Yoshino, H. Kajii, H. Araki, T. Sonoda, H. Take, and
161. I. Alexandrou, E. Kymakis, and G. A. J. Amartunga, S. Lee, Fullerene Sci. Technol., 7, 695 (1999).
Appl. Phys. Lett., 80, 1435 (2002). 190. J-M. Benoit, B. Corraze, S. Lefrant, W. J. Blau, P.
162. E. Kymakis and G. A. J. Amaratunga, Appl. Phys. Lett., Bernier, and O. Chauvet, Synthetic Metals, 121, 1215
80, 112 (2002). (2001).
163. X. Xu, M. M. Thwe, C. Shearwood, and K. Liao, Appl. 191. P. C. P. Watts, W. K. Hsu, D. P. Randall, H. W. Kroto,
Phys. Lett., 81, 2833 (2002). and D. R. M. Walton, Phys. Chem. Chem. Phys., 4, 5655
164. P. Fournet, J. N. Coleman, B. Lahr, A. Drury, W. J. (2002).
Blau, D. F. O’Brien, and H.-H. Horhold, J. Appl. Phys., 192. J. Benoit, B. Corraze, and O. Chauvet, Phys. Rev. B.,
90, 969 (2001). 65, 241405R (2002).
165. B. Vigolo, P. Launois, M. Lucas, M. Badaire, P. Bernier, 193. J.-M. Benoit, B. Corraze, S. Lefrant, P. Bernier, and O.
and P. Poulin, Making Functional Materials with Nan- Chauvet, Making Functional Materials with Nanotubes.
otubes Symposium (Materials Research Society Sympo- Symposium (Materials Research Society Symposium Pro-
sium Proceedings, Vol. 706), pp. 3
8 (2002). ceedings), 706, 85 (2002).
166. B. Vigolo, A. Penicaud, C. Coulon, C. Sauder, R. Pailler, 194. C. Stephan, T. P. Nguyen, S. Lefrant, L. Vaccarini, and
C. Journet, P. Bernier, and P. Poulin, Electronic Proper- P. Bernier, AIP Conference Proceedings, 544, 363
ties of Molecular Nanostructures: 15th International Win- (2000).
terschool/Euroconference (AIP Conference Proceedings), 195. C. Stephan, T. P. Nguyen, B. Lahr, W. Blau, S. Lefrant,
591, pp. 562
567, Kirchberg, Austria (March 2001). and O. Chauvet, J. Mater. Res., 17, 396 (2002).
167. L. Jin, C. Bower, and O. Zhou, Appl. Phys. Lett., 73, 196. I. A. Tchmutin, A. T. Ponomarenko, E. P. Kinichnaya, L.
1197 (1998). I. Tkachenko, A. P. Lisitskaya, G. I. Kozub, and O. N.
168. F. Tsui, L. Jin, and O. Zhou, Appl. Phys. Lett., 76, 1452 Efimov, http://194.85.238.10/conf/iwfac2001/page_
(2000). 219.pdf.
169. M. Ichida, S. Mizuno, H. Kataura, Y. Achiba, and K. 197. L. Dai and A. B. W. Mau, Adv. Mater., 13, 899 (2001).
Nakamura, Making Functional Materials with Nanotubes 198. E. Kymakis, I. Alexandou, and G. A. J. Amaratunga,
Symposium (AIP Conference Proceedings), 590, pp. Synthetic Metals, 127, 59 (2002).
121
124, Kamakura, Japan (January 2001). 199. G. B. Blanchet, C. R. Fincher, and F. Gao, Appl. Phys.
170. E. K. Hobbie, H. Wang, H. Kim, S. Lin-Gibson, and E. A. Lett., 82, 1290 (2003).
Grulke, Phys. Fluids, 15, 1196 (2003). 200. B. E. Kilbride, J. N. Coleman, J. Fraysse, P. Fournet,
171. H. L. Schreuder-Gibson, K. Senecal, M. Sennett, L. M. Cadek, A. Drury, S. Hutzler, S. Roth, and W. J. Blau,
Samuelson, Z. Huang, J. G. Wen, W. Li, Y. Ti, D. Z. J. Appl. Phys., 92, 4024 (2002).
Wang, S. X. Yang, Z. F. Ren, and C. Sung, “Character- 201. B. McCarthy, J. N. Coleman, S. A. Curran, A. B. Dalton,
istics of Electrospun Fibers Containing Carbon Nan- A. P. Davey, Z. Konya, A. Fonesca, J. B. Nagy, and W. J.
otubes,” in Fullerenes 2000 —Volume 10, Chemistry and Blau, J. Mater. Sci. Lett., 19, 2239 (2000).
Physics of Fullerenes and Carbon Nanomaterials, The 202. L. Vallentini, J. Biagotti, J. M. Kenny, and S. Santucci,
Electrochemical Society, pp. 210
221 (2000). J. Appl. Polym. Sci., 87, 708 (2003).
172. P. Poulin, B. Vigolo, and P. Launois, Carbon, 40, 1741 203. M. Cadek, J. N. Coleman, V. Barron, K. Hedicke, and
(2002). W. J. Blau, Appl. Phys. Lett., 81, 5123 (2002).
173. B. Vigolo, P. Poulin, M. Lucas, P. Launois, and P. Bernier, 204. J. R. Wood, Q. Zhao, M. D. Frogley, E. R. Meurs, A. D.
Appl. Phys. Lett., 81, 1210 (2002). Prins, T. Peijs, D. J. Dunstan, and H. D. Wagner, Phys.
174. H. T. Ng, M. L. Foo, A. Fang, J. Li, G. Xu, S. Jaenicke, Rev. B., 62, 7571 (2000).
L. Chan, and S. F. Y. Li, Langmuir, 18, 1 (2002). 205. J.-M. Bonard, J.-P. Salvetat, T. Stockli, L. Forro, and
175. E. T. Thostenson, W. Z. Wang, Z. F. Ren, and T. W. Chou, A. Chatelain, Appl. Phys. A., 69, 245 (1999).
J. Appl. Phys., 91, 6034 (2002). 206. C. H. Poa, S. R. P. Silva, P. C. P. Watts, W. K. Hsu,
176. P. K. Wu, J. FitzGerald, A. Pique, D. B. Chrisey, and H. W. Kroto, and D. R. M. Walton, Appl. Phys. Lett., 80,
R. A. McGill, Mat. Res. Soc. Symp., 617, J2.3.1 (2000). 3189 (2002).
177. A. Hassanien, M. Gao, M. Tokumoto, and L. Dai, Electronic 207. Y. W. Jin, J. E. Jmg, Y. J. Park, J. H. Choi, D. S. Jung,
Properties of Molecular Nanostructures: 15th Interna- H. W. Lee, S. H. Park, N. S. Lee, J. M. Kim, T. Y. Ko, S.
tional Winterschool/Euroconference (AIP Conference Pro- J. Lee, S. Y. Hwang, J. H. You, J.-B. Yoo, and C.-Y. Park,
ceedings), 591, pp. 501
506, Kirchberg, Austria (March J. Appl. Phys., 92, 1065 (2002).
2001). 208. S. Hoffman, C. Ducati, B. Kleinsorge, and J. Robertson,
178. V. Lordi and N. Yao, J. Mater. Res., 15, 2770 (2000). Appl. Phys. Lett., 83, 4661 (2003).
179. K.-T. Lau and D. Hui, Carbon, 40, 1605 (2002). 209. E. E. Donaldson, M. F. Durstock, B. E. Taylor, D. W.
180. G. G. Tibbetts, J. C. Finegan, and C. Kwag, Mat. Res. Tomlin, L. C. Richardson, and J. W. Baur, Proceeding of
Soc. Symp. Proc., 733E, T2.3.1 (2002). the SPIE–The International Society for Optical Engineer-
181. Y. J. Liu and X. L. Chen, Mech. Mater., 35, 69 (2003). ing, Vol. 4465, pp. 85
93, San Diego (August 2001).

644 POLYMER COMPOSITES, DECEMBER 2004, Vol. 25, No. 6

 Review of Polymer/Carbon Nanotube Composites

210. S. H. Jhang, S. Y. Kim, J. H. Park, T. Ahn, D. S. Kim, 232. R. Bandyopadhyaya, E. Nativ-Roth, O. Regev, and
and Y. W. Park, Electronic Properties of Molecular Nano- R. Yerushalmi-Rozen, Nano Lett., 2, 25 (2002).
structures: 15th International Winterschool/Euroconfer- 233. R. B. Pipes and P. Hubert, Comp. Sci. Technol., 62, 419
ence (AIP Conference Proceedings), 591, pp. 489
492, (2002).
Kirchberg, Austria (March 2001). 234. J. H. Chen, Z. P. Huang, D. Z. Wang, S. X. Yang, J. G.
211. H. S. Woo, R. Czerw, S. Webster, D. L. Carroll, J. Bal- Wen, and Z. F. Ren, Appl. Phys. A., 73, 129 (2001).
lato, A. E. Strevens, D. S. O’Brien, and D. J. Blau, Appl. 235. C. A. Cooper, S. R. Cohen, A. H. Barber, and H. D. Wag-
Phys. Lett., 77, 1393 (2000). ner, Appl. Phys. Lett., 81, 3873 (2002).
212. K. H. An, W. S. Kim, Y. S. Park, J.-M. Moon, D. J. Bae, 236. O. Breuer, U. Sundararaj and R. Toogood, Polym. Eng.
S. S. C. Lim, Y. S. Lee, and Y. H. Lee, Adv. Funct. Mater., Sci., 44, 868 (2004).
11, 387 (2001). 237. P. Ball, Nature, 414, 142 (2001).
213. K. H. An, K. K. Jeon, J. K. Heo, S. C. Lim, D. J. Bae, and 238. S. Iijima, Physica B., 323, 1 (2002).
Y. H. Lee, J. Electrochem. Soc., 149, A1058 (2002). 239. C. J. Lee, S. C. Lyu, H.-W. Kim, C.-Y. Park, and C.-W.
214. Y.-C. Chen, N. R. Raravikar, L. S. Schadler, P. M. Yang, Chem. Phys. Lett., 361, 469 (2002).
Ajayan, Y.-P. Zhao, T.-M. Lu, G.-C. Wang, and X.-C. 240. X. Zeng, X. Sun, G. Cheng, X. Yan, and X. Xu, Physica
Zhang, Appl. Phys. Lett., 81, 975 (2002). B., 323, 330 (2002).
215. J.-Y. Kim, M. Kim, H. Kim, J. Joo, and J.-H. Choi, Opt. 241. Y. Wang, F. Wei, G. Gu, and H. Yu, Physica B., 323, 327
Mater., 21, 147 (2002). (2002).
216. Y.-C. Chen. T.-M. Liu, G.-C. Wang, X.-C. Zhang, N. R. 242. L. Dong, J. Jiao, C. Mosher, and S. Foxley, Mat. Res.
Raravikar, Y.-P. Zhao, L. S. Schadler, and P. M. Ajayan, Soc. Symp. Proc., 728, S8.21.1 (2002).
Technical Digest. Summaries of papers presented at the 243. T. Prasse, A. Ivankov, J. Sandler, K. Schulte, and
Conference on Lasers and Electro-Optics. Conference W. Bauhofer, J. Appl. Polym. Sci., 82, 3381 (2001).
Edition (IEEE Cat. No.02CH37337), 1, 660 (2002). 244. S. J. V. Frankland and D. W. Brenner, Mat. Res. Soc.
217. P. C. P. Watts, W. K. Hsu, V. Kotzeva, and G. Z. Chen, Symp. Proc., 593, 199 (2000).
Chem. Phys. Lett., 366, 42 (2002). 245. S. J. V. Frankland and D. W. Brenner, Polym. Preprints,
218. S. M. O’Flaherty, R. Murphy, S. V. Hold, M. Cadek, J. N. 40, 721 (1999).
Coleman, and W. J. Blau, J. Phys. Chem. B., 107, 958 246. S. J. V. Frankland, A. Caglar, D. W. Brenner, and
(2003). M. Griebel, J. Phys. Chem. B., 106, 3046 (2002).
219. B. Miller, http://www.fibrils.com/conplas.htm. 247. M. I. Panhuis, A. Maiti, A. B. Dalton, A. van den Noort,
220. J. Amaraskera, A. Burmnell, C. Lietzau, and K. Balfour, J. N. Coleman, B. McCarthy, and W. J. Blau, J. Phys.
Polym. Prep., 42, 36 (2001). Chem. B., 107, 478 (2003).
221. C. Chu and R. P. Boehmer, http://www.winbmdo.com. 248. M. In het Panhuis, A. Maiti, J. N. Coleman, A. B. Dalton,
222. P. M. Ajayan, L. S. Schadler, C. Giannaris, and A. Rubio, B. McCarthy, and W. J. Blau, Electronic Properties of Mo-
Adv. Mater., 12, 750 (2000). lecular Nanostructures: 15th International Winter-
223. S. H. Foulger and R. V. Gregory, National Textile Center school/Euroconference (AIP Conference Proceedings),
Research Briefs–Materials Competency (June 2001). 591, pp. 497
500, Kirchberg, Austria (March 2001).
224. E. Barrera and K. Lozano, JOM, 52, 32 (2000). 249. S. J. V. Frankland and V. M. Harik, Mat. Res. Symp.
225. P. Calvert, Nature, 399, 210 (1999). Proc., 733E, T6.2.1 (2002).
226. I. Vesselenyi, A. Siska, D. Mehn, K. Niesz, Z. Konya, J. 250. S. J. V. Frankland, A. Cagler, D. W. Brenner, and
B. Nagy, and I. Kiricsi, J. Phys. IV, 12, Pr4 (2002). M. Griebel, Mat. Res. Soc. Symp. Proc., 633, A14.27.1
227. B. Wang, http://www.eng.fsu.edu/departments/in- (2001).
dustrial/wsfiles/nanotube.pdf. 251. G. M. Odegard, T. S. Gates, K. E. Wise, C. Park, and E.
228. O. Lourie and H. D. Wagner, J. Mater. Res., 13, 2418 J. Siochi, “Constitutive Modeling of Nanotube-Reinforced
(1998). Polymer Composites,” ICASE Report No. 2002-27,
229. O. Lourie and H. D. Wagner, Appl. Phys. Lett., 73, 3527 NASA/CR-2002-211760.
(1998). 252. G. M. Odegard, T. S. Gates, L. M. Nicholson, and K. E.
230. M. D. Frogley, Q. Zhao, and H. D. Wagner, Phys. Rev. B., Wise, Comp. Sci. Technol., 62, 1869 (2002).
Condens, Matter Mater. Phys., 65, 113413/1 (2002). 253. J. Ouellette, The Industrial Physicist, December 2002/
231. Q. Zhao, J. R. Wood, and H. D. Wagner, J. Polym. Sci. B: January 2003, p. 18.
Polym. Phys., 39, 1492 (2001).

POLYMER COMPOSITES, DECEMBER 2004, Vol. 25, No. 6 645