Composites with carbon nanotubes and graphene: An outlook

See allHide authors and affiliations

Science  02 Nov 2018:
Vol. 362, Issue 6414, pp. 547-553
DOI: 10.1126/science.aat7439


Composite materials with carbon nanotube and graphene additives have long been considered as exciting prospects among nanotechnology applications. However, after nearly two decades of work in the area, questions remain about the practical impact of nanotube and graphene composites. This uncertainty stems from factors that include poor load transfer, interfacial engineering, dispersion, and viscosity-related issues that lead to processing challenges in such nanocomposites. Moreover, there has been little effort to identify selection rules for the use of nanotubes or graphene in composite matrices for specific applications. This review is a critical look at the status of composites for developing high-strength, low-density, high-conductivity materials with nanotubes or graphene. An outlook of the different approaches that can lead to practically useful nanotube and graphene composites is presented, pointing out the challenges and opportunities that exist in the field.

Carbon nanomaterials have had an unprecedented impact over the past three decades in defining the reach and applications of nanotechnology. Starting with the discovery of fullerenes and moving through the carbon nanotube (CNT) era to graphene and other two-dimensional (2D) materials, the academic world has been flush with new ideas, inventions, and innumerable attempts to find the killer applications for these remarkable nanostructures. Here we will discuss one such application, which when first introduced seemed close to embracing these materials but for various reasons has not met expectations: that of composite materials. In particular, the question of whether structures such as carbon nanotubes and graphene, touted as ideal reinforcements in composite matrices because of their mechanical properties, really are the right choices for mechanical reinforcement still remains largely unanswered. Moreover, the realm of functionalities that can be accessed by introducing nanotubes versus graphene in a matrix needs careful evaluation and thought. Although both are sp2 allotropes, the structure, morphology, and dimensionality of these two profoundly interesting carbon nanostructures are indeed quite different, and so is the nature of their interactions with the adjacent matrix (1). Hence, the overall composite mechanical behavior provided by these two reinforcement units could be distinct. It would indeed be useful to have selection rules that choose one over the other in composite applications, but no such rational approaches have emerged in the design of nanocomposites with CNT or graphene phases.

We will consider three important questions that may broadly define the fate of this field. Why should these unique structures be of any interest in composites, why have they not produced substantial progress after much effort, and what can be done to make them work as good reinforcements in composites? The discussion will go beyond just that of mechanical properties and also consider their excellent electrical and thermal properties.

Graphene represents the thinnest possible atomically flat layer, made of a planar hexagonal honeycomb lattice of strong C-C bonds that build up graphite when stacked via weak van der Waals interlayer forces. Graphene has a Young’s modulus near 1 TPa in the plane, reflecting the intrinsic carbon bond stiffness, but being very thin, it can be flexible in bending, twist, and other deformation modes (2). The in-plane electrical and thermal conductivities are the highest among known materials, but the through-thickness properties of stacked graphene are very poor. CNTs are rolled-up graphenes with the in-plane properties translated to axial properties, making them among the stiffest axial fibers ever created. Similar to graphene, they can also be easily bent, twisted, and buckled (3). As graphene stacks up to make multilayer structures, CNTs can also have a nested structure of tubes inside tubes [single-walled (SWNTs) to multiwalled nanotubes (MWNTs)], which has a notable effect in determining their mechanical properties. In general, although the local stiffness that can be measured is extremely high, owing to a near defect-free structure of graphene and nanotubes, a couple of major issues make reinforcement of these structures in matrices challenging. First, both graphene and nanotubes are particulate fillers, with their larger dimensions (lateral size of graphene or lengths of nanotubes) reaching several hundred micrometers at the most or, in exemplary cases, millimeters. Short fibers are generally poor load carriers in fiber composites, and this effect is clearly seen when composites are made with CNT or graphene dispersions. Second, the surfaces of both graphene and nanotube are atomically smooth, devoid of any dangling bonds or defects (except at the edges in graphene or tips in nanotubes), which means that strong matrix-filler bonds are hard to accomplish, leading to poor interfacial load transfer during mechanical deformation (1) and high electron and phonon scatter, compromising electrical and thermal properties. This interfacial problem was a major roadblock in carbon fiber composites before industry figured out the sizing of fibers via chemical modification, but for nanotubes and graphene, this problem is severe, and attempts to chemically functionalize CNT or graphene surfaces may substantially compromise their intrinsic properties (4). This compromise also relates to a third issue, which is the inhomogeneous dispersion of nanotubes and graphene in the matrix. Without proper surface treatments, CNTs or graphene tend to aggregate easily, owing to strong van der Waals interactions between them, to form poorly dispersed bundles or agglomerates in the matrix, which often leads to poor interfacial connectivity and formation of mechanical stress concentration or other functionally singular sites, resulting in severely affected composite performances. Noncovalent functionalization methods (5) could be used to partially overcome the dispersion challenge, but this approach is ineffective in solving the interface problem. Therefore, a systematic and careful engineering approach is needed to design CNT or graphene composites with optimal performances (Fig. 1). We can show the potential for using CNT or graphene composites for applications by plotting the hypothetical position of different CNT or graphene composites in a modified Ashby plot (Fig. 1B) and the position of future continuous CNT or graphene fiber composites. Finally, the ideal dispersion of material throughout the composite is very application dependent; typically for mechanical properties, one wants as high a loading of the filler as possible that is aligned in the direction of the load, whereas for electrical percolation, one aims for a random percolated network with as low a concentration as possible. For multifunctional applications, the required microstructures may therefore even be contradictory.

Fig. 1

(A) A schematic illustration of a futuristic CNT or graphene polymer composite that consists of continuous CNT fiber preform (fabric) in a polymer matrix and chemically modified CNT or graphene as matrix modifiers. The continuous CNT or graphene spun fibers can exceed the mechanical properties of carbon fibers that are used for state-of-the-art structural reinforcements in composites. The best load-carrying capability ever reached yet could be achieved by engineering the continuous CNT or graphene fiber preforms in composite design. It can be envisioned that the multifunctional CNT or graphene composites are realized by marrying the best features of the continuous CNT or graphene fibers and the modifiable polymer matrix with chemically functionalized CNT or graphene particulates. (B) Ashby plot of Young’s modulus plotted against tensile strength comparing the mechanical properties of conventional polymer composites, including glass fiber–reinforced plastic (GFRP) and carbon fiber–reinforced plastic (CFRP), with CNT or graphene–based polymer composites. Thick outlines represent families of each material. In this projected Ashby plot, both mechanical properties of CNT or graphene polymer composites, which are considered to be particulate composites, are scattered around in between those of GFRPs and polymers. Those properties are shown to be below the ones of CFRPs that are continuously reinforced composites conventionally used for load-carrying structures. However, the mechanical properties of CNT spun fibers can indicate that CNT or graphene fibers can replace carbon fiber and, if properly designed and manufactured, the continuously reinforced CNT or graphene spun fiber composites might be used in making ultralight yet superstrong structures in the near future. The Ashby plot was made by taking many reported values from the literature and also our estimated scenario (e.g., for the CNT fiber composites).


Carbon nanotube and graphene fillers, interfaces, and load transfer

A great deal of effort has been made to develop lightweight, strong composite materials with CNTs and graphene as reinforcement, and although these are considered to be discontinuous short fillers, they possess outstanding mechanical properties. The extremely high Young’s modulus of CNTs and graphene, their nanoscale dimensions, along with particular geometries that offer high specific surface area, present unprecedented opportunities to efficiently tailor the interface properties between the reinforcements and composite matrices. CNT or graphene nanocomposites may not be as strong or stiff as a continuously reinforced composite such as typical carbon fiber laminates that are currently used in primary load-carrying structural applications. However, if the extraordinary potential and multifunctionality of CNTs and graphene are fully realized and their nanocomposites properly designed, they might still become game-changing composite materials. In addition to the challenges of dispersion, viscosity control, and sizing of the CNTs and graphene without compromising intrinsic properties, challenges still exist in accurately characterizing the reinforcement’s mechanical properties. These properties include Young’s modulus and strength, and also the interfacial shear strength at the filler-matrix interface, which will determine the critical length of the fillers required for the most efficient load-transfer capability. Although these factors are critically important in designing short-fiber composite systems, unfortunately most attention seems to be focused on improving practical issues such as dispersion in matrix materials and increasing the loading fractions of these fillers without suffering viscosity-related processing issues.

A substantial challenge has been the proper characterization of fillers and filler-matrix interfaces during loading and assessing the efficiency of load transfer in composites. Raman spectroscopy has been the most powerful technique for both the characterization of carbon-based nanomaterials and the assessment of their mechanical properties (6). Raman spectroscopy can also be used to evaluate their mechanical properties by observing the shift of the Raman bands when the materials are subjected to deformation (see Box 1 and Fig. 2).

Box 1

Raman spectroscopy and CNT or graphene composites.

sp2 allotropes of carbon have electronic structures that lead them to undergo very strong resonance Raman scattering so that well-defined-spectra can be obtained even from individual CNTs or single layers of graphene (16, 38). These structures show characteristic G and 2D bands in their Raman spectra, and where there are internal or edge defects, a D band may also be seen. Subtle differences between the spectra of SWNTs, DWNTs (double-walled carbon nanotubes), and MWNTs and between those of mono-, bi- and multilayer graphene enable the spectra to be used as fingerprints of the materials. The Raman bands shift during deformation for both CNTs and graphene (6). Such downshifts during tensile deformation are large and relatively easy to measure and enable one to monitor the deformation of CNT or graphene in nanocomposites inside a polymer matrix (6). The shift rates (in cm−1/% strain) are found to scale with the Young’s modulus of the nanocarbons, which enables such band shifts to be used as a universal stress sensor for composite mechanics (39).

Fig. 2 Dependence of the 2D Raman band position upon strain during the deformation of CNTs in an epoxy resin matrix composite (0.1 wt %).

The higher slope for the SWNTs reflects their higher Young’s modulus. (Inset) The shift of the 2D band for the SWNTs. [Reprinted from (7) with permission from Springer Nature]

In polymer composites, we are normally concerned with the matrix-reinforcement interface. However, in layered reinforcements, it is also necessary to consider the van der Waals bonding between the walls of carbon nanotubes and the individual layers in multilayer graphene. This bonding is relatively weak compared with the strong covalent bonding within the graphene layers. When MWNTs and multilayer graphene are used in composites, their ability to reinforce is therefore limited by easy shear between the walls or layers, respectively. It is possible to track internal stress transfer between the walls of carbon nanotubes and between the layers of graphene from stress-induced Raman band shifts. Imperfect stress transfer is manifested as Raman band broadening during deformation and a lower Raman band shift rate compared with the single-walled or monolayer material. Comparing Raman band shifts under stress for SWNTs and MWNTs in epoxy nanocomposites shows that interwall stress transfer efficiency is only around 70% for MWNTs (7). Lower Raman band shift rate in multilayer graphene obtained with increasing number of graphene layers shows a similar trend to that in MWNTs. Raman spectroscopy has been effective in evaluating the efficiency of stress transfer at the interface before and after chemical modification of CNT or graphene surfaces. Several chemical modification schemes have been proposed (1, 8, 9) (Fig. 3), which improves the interfacial strength between CNT or graphene fillers and the polymer matrices.

Fig. 3 A schematic representation of four different chemical modification schemes for CNT or graphene interacting with polymer matrix.

Shown are π-π interactions (i.e., noncovalent interactions), chemical bonding, van der Waals force, and electrostatic interactions. (Inset) A scanning electron micrograph (SEM) image of the fracture surface of SWNT epoxy composites showing good dispersion and pullout of the nanotube bundles from the matrix (9). CNT or graphene surfaces with high specific surface area are favorable for chemical interactions with other molecules, which can allow for several chemical modification schemes. The ends of CNTs and edges of graphene facilitate additional chemical interactions with the polymer matrix. The chemically modified CNT or graphene surfaces can promote more uniform dispersion, control loading fractions of CNT or graphene without suffering high viscosity, and improve interfacial strength between the fillers and polymer matrix in the composites. Although the nanoscale dimensions present a formidable change, it is vital to understand the interactions between the CNT or graphene fillers and matrix toward optimum performance in composite design. In particular, the characterization and quantification of the interfacial shear strength will be essential.

Properties of carbon nanotube and graphene composites

The simplest way to look at the effect of nanofiller addition to a matrix is from the resulting mechanical properties of the nanocomposites, most commonly assessed through stress-strain curves (Fig. 4A). A telling example is natural rubber, which is a relatively strong elastomer with an extension to failure of ~1000%, where the addition of both the MWNTs and graphene leads to a major change in the stress-strain curves and a large increase in the stiffness even at relatively low loading of the fillers (10, 11). In both cases, however, the strength of the rubber is not increased, and the strain to failure is even reduced. The effectiveness of the reinforcements can be best assessed by plotting Young’s modulus of the nanocomposite, Ec (as the stress at 100% strain for rubbers with nonlinear stress-strain curves), against the filler loading, and an approximately linear increase in Ec with filler volume fraction is seen in both cases. The effect of adding carbon black (commonly used nanofiller) shows a much smaller increase in the modulus of rubber (about a factor of 3 lower than for graphene).

Fig. 4 Schematic illustrations of representative mechanical and electrical properties of CNT or graphene nanocomposites.

(A) A schematic representation of tensile stress-strain behaviors of three different polymers as CNT or graphene loading fraction increases: thermoset polymer (TS, dark blue region), thermoplastic polymer (TP, blue region), and elastomer (ES, light blue region). Schematic of representative tensile stress-strain curves are compared for pristine thermoset, thermoplastic polymers and elastomer, and its nanocomposite with CNT or graphene reinforcements, respectively. For the pristine polymers, the stress-strain curves are depicted with yellow lines, while for the corresponding CNT or graphene nanocomposites, the curves are illustrated with black lines. As the CNT or graphene loading increase (depicted as horizontal white arrows), the stress-strain curve of each pristine polymer shifts from the yellow line to the black line. This shift indicates that with the increase of CNT or graphene loading fraction, Young’s modulus of the polymer can increase proportionally; by contrast, the elongation at break can be substantially decreased. A slight decrease in tensile strength is also shown. Both elastomer and thermoplastic polymer exhibit a similar behavior of ductile-to-brittle transition at high loading fractions of CNTs or graphene, whereas thermoset nanocomposites become increasingly brittle over the pristine polymer. The shift of the stress-strain curves with CNT or graphene reinforcements clearly shows that there is a compromise between Young’s modulus and toughness, which can be quantified by characterizing the area of the stress-strain curve. (B) Loss modulus and electrical conductivity with respect to interfacial volume fraction in CNT or graphene nanocomposites. With increasing interfacial volume and strength between CNT or graphene fillers and polymer matrix, the loss modulus, which represents viscoelastic damping properties, can linearly increase as a result of filler-matrix slippage until the dispersion quality can be maintained. However, at higher loading fraction of CNTs or graphene, the mobility of polymer chains gradually decreases, eventually leading to no further increase in loss modulus. As for electrical conductivity, as the interfacial volume fraction of CNTs or graphene increases, the conductivity of the nanocomposites changes from insulating to percolative conductive regimes, displaying an S-shape curve. Once the random conductive network is formed, the conductivity of the composites rises sharply, which is characterized as the percolation threshold. Beyond that, the composites become conductive. Typical low percolation thresholds of CNTs or graphene in polymer composites are highly advantageous for many electrostatic and electromagnetic applications.

The mechanics of deformation of fiber-reinforced composites are now well established, but this is not the case for nanocomposites. An obvious way forward is to adapt the continuum mechanics theories developed for macroscopic composites, such as those reinforced by carbon or glass fibers, to composites at the nanoscale. Some researchers have taken an opposing viewpoint and have suggested that polymer nanocomposites are actually quasi-homogeneous molecular blends that should be regarded as molecular composites or self-reinforced composites (12, 13) and that the properties are controlled by interactions on the molecular scale between the nanoparticles and the polymer matrix.

One issue that arises is the difficulty of realizing the promised high stiffness of the nanotubes or graphene in polymer-based nanocomposites. The effective stiffness of the fillers can be assessed by using the rule of mixtures, and the experimentally derived stiffness of CNT or graphene fillers in composites is almost always observed to be far below the anticipated ∼1 TPa modulus. In the case of CNTs, realization of the high stiffness only occurs when the CNTs are extended and aligned in polymer fibers or tapes (14). Values approaching ∼500 GPa have been found, and this can be rationalized by using a modified rule of mixtures that takes into account filler size (length of CNTs and diameter for graphene flakes) and orientation (15):Embedded Image(1)where ηo is the Krenchel orientation factor and ηl is the length factor. The parameter Eeff is the effective Young’s modulus of the filler that depends only on its structure. For example, it will be ~ 1 TPa for monolayer graphene but is lower for few-layer graphene and graphene oxide (16). A similar reduction in Eeff occurs when going from SWNTs to MWNTs as the result of interwall sliding. The orientation factor ηo is 1 for aligned CNTs and flakes but much smaller for randomly oriented CNTs. Hence, the mechanical properties of CNTs are more sensitive to tube orientation, and randomly oriented graphene flakes offer better reinforcement than randomly oriented CNTs. The length factor ηl (0 ≤ηl ≤ 1) reflects the efficiency of stress transfer from the matrix to the filler that is controlled by both the geometry (e.g., aspect ratio) of the filler and the strength of the filler-matrix interface. It approaches unity for long CNTs or large-diameter flakes with strong filler-matrix interfaces. The high values of Eeff found for aligned CNTs in fibers and tapes are therefore the result of long, extended CNTs with a high degree of alignment.

The very high strengths reported for CNTs and graphene are also not realized in nanocomposites. The celebrated experiment (17) that measured a strength of 130 GPa for graphene (“200 times stronger than steel”!) has been confirmed theoretically, but the experiment involved the nano-indentation of a monolayer graphene membrane and the deformation of less than a few thousand carbon atoms. Single-nanotube deformation experiments involve similar numbers of atoms. The strength of all materials falls as their size increases owing to the presence of defects, which is one reason why high strengths are not found in macroscopic nanocomposite specimens. Backing out the effective strength of the filler from the fracture strength of the nanocomposite is possible, using a rule of mixtures equation for fracture analogous to that for stiffness. Modest values of strength are generally found in randomly aligned nanocomposites, but values as large as 88 GPa have been reported for SWNTs in highly aligned polymer tapes (18). Strength values of ~10 GPa extracted from fracture data for graphene nanocomposites (19), although being well below the quoted 130 GPa graphene strength value, compare favorably with that of carbon fibers. Dispersion of the nanofillers also plays a major role in their effective strengths in nanocomposites, with high values only found for uniform dispersions.

In addition to being processed into extended structures for potential high-strength and high-stiffness composites, CNTs and graphene may best be applied in their particulate state as dispersed matrix modifiers or fiber modifiers in composites. The CNTs or graphene dispersed polymer composites exhibit time-dependent viscoelastic behavior under mechanical loadings, owing to the nature of polymers experiencing molecular rearrangements to release induced local stress or strain as a time-dependent behavior. The presence of CNTs or graphene can affect the polymer chains’ mobility particularly in the vicinity of the reinforcements in composites. The viscoelastic behavior of nanotube and graphene composites has been reported, leading to strong matrix-modifying effects of these fillers. With only 0.1 weight % (wt %) of graphene platelets (GPLs), epoxy-GPL composites creep substantially less than SWNT-epoxy and MWNT-epoxy composites at the same loading fraction, mainly because of stronger interaction between graphene and epoxy. Functionalization of graphene (e.g., oxidation) can further promote better interaction with polymer matrix (20), leading to a higher glass transition temperature. As a consequence of the stronger interaction at the interface, relaxation process and storage moduli of the functionalized graphene epoxy composites decay more slowly over the pristine graphene composites. It has been observed that without compromising modulus and strength, an order-of-magnitude enhancement in damping properties can be achieved in CNT-polycarbonate nanocomposites, mainly resulting from the interfacial slippage between CNTs and the polymer matrix (20). In general for CNT or graphene polymer composites, the intertube, interlayer interaction and sliding, and the sliding between polymer chains, are all contributing factors to the strong viscoelastic behavior observed in composites (Fig. 4B).

The large interfacial volume available can be used to design interface-directed dynamic composites. A few recent studies have shown highly intriguing morphology evolution of rubbery polymer of polydimethylsiloxane (PDMS) to become more ordered around CNTs under repeated loading, resulting in improved bulk composite stiffness for locally crystallized polymer composites (21). Discernable stiffening was observed for the composite during cyclic compressive stressing, a phenomenon not observed for the neat polymer. Such self-stiffening behavior was also observed in graphene-based PDMS composites. Because of its relatively higher chain mobility compared to chemical cross-links, the physically cross-linked network allows polymer chains to undergo realignment and reorientation along graphene surfaces under cyclic compression loading, resulting in increased interface stiffness (22). The engineering of unique interface-related properties should allow us to build materials that are stimuli-responsive, dynamically reconfigurable, adaptable, and self-strengthening under loading. CNT or graphene particles seem to be ideal fillers for creating such dynamic interfaces in polymers, which can be further tuned through chemical modification of the filler surfaces. In this regard, quantitative interface evaluation between CNT or graphene and matrices enabled by nanomechanical measurement (see Box 2 and Fig. 5) is very important for more effective interface engineering for improving overall composite performances.

Fig. 5 Nanomechanical characterization of CNT-epoxy interfaces.

Shear-lag fits give a value for interfacial shear stress (IFSS) of ~30 ± 7 MPa for pristine carbon nanotubes and epoxy matrix (25). (Inset) SEM observation of a single nanotube pullout from epoxy using an AFM probe (43). It was observed that average IFSS increases as the embedded length decreases, which is consistent with the shear-lag theory and demonstrates the development of interfacial shear stresses at the ends of these fibers. [Reprinted from (43) with permission from The Royal Society]

Box 2

Interfacial strength and nanomechanics in CNT or graphene composites.

Single-fiber pull-out tests have been used since the early development of composite materials technology to measure the shear strength of ductile interfaces capable of developing a constant shear stress prior to fiber pull-out, and the fracture energy of brittle interfaces that experience crack initiation and propagation prior to fiber pull-out. Accordingly, attempts have been made to perform similar experiments on CNT or graphene composites. A successful attempt was made (40) to attach a CNT to an atomic force microscopy (AFM) probe that was then dipped into a liquid epoxy, and a pull-out test was carried out after the epoxy was cured. More recently, microfabricated devices have been developed to perform the pull-out experiments with better control (41). The average interfacial shear strength reported by these tests corresponds to the range of ~6 to 200 MPa for CNT-epoxy interfaces (42), although such quantitative nanomechanical measurements remain a big challenge for graphene/polymer interfaces.

In addition to their mechanical properties, nanotubes and graphene have several other functionalities that make them attractive for use in composites. Important examples are their thermal stability and high electrical and thermal conductivities. Nanotubes are used in dispersions in thermoplastic polymers to improve their electrostatic discharge or electromagnetic interference shielding properties (23). Graphene exhibits an electrical conductivity comparable to that of CNTs (105 to 106 S cm−1). However, the percolation threshold for electrical conductivity of graphene in a polymer composite is much higher than for CNTs in composites. The percolation threshold of CNTs in high-density polyethylene matrix is 0.15 wt %, compared to 1 wt % for graphene in the same polymer (24). This is because 1D CNTs form a better conductive network than 2D graphene in composite materials (Fig. 4B). In contrast to the electrical conductivity of CNT or graphene composites, the improvement in thermal conductivity appears to be marginal for both. Even though individual CNTs or graphene have thermal conductivities as high as ~3000 W m−1 K−1, such particulate composites have a more efficient network, including intertube or interlayer scattering and heat transfer resistance at the filler-matrix interface (25). Attempts can still be made to enhance the thermal conductivity of the composites by substantially increasing the weight fraction of graphene or CNTs in a matrix. For the same loading fraction, graphene composites exhibit greater interaction at the interfaces between filler and matrix, which requires a higher activation energy for the glass transition, resulting in better thermal stability of polymer composites.

Compared to the vast number of studies on polymer matrix composites reinforced by CNTs and graphene, ceramic or metal-based composites have received much less attention, possibly because of the fabrication challenges. In addition, mechanical reinforcing effects are more obvious in polymer-based composites. The primary motivation for the development of CNTs or graphene-based ceramic composites is enhancing toughness or resistance to crack growth because ceramics are already stiff and strong. For graphene-reinforced ceramic composites, the primary toughening mechanism is the increased energy dissipation due to pullout of graphene nanosheets (26). Other toughening mechanisms observed include crack deflection at the matrix-reinforcement interface and crack bridging. Similar toughening mechanisms were also observed in CNT-reinforced ceramic composites. Additionally, improved thermal and electrical conductance are often attractive properties of ceramic composites. On the other hand, light metals such as Al, Mg, and Cu are commonly used as matrices for CNTs or graphene-reinforced metal composites for mechanical property enhancement (27). It is worth noting that the interfacial strength between CNT or graphene fillers and metal matrices is substantially affected by the nature of interactions that include chemisorption (e.g., Ni or Ru) and physisorption (e.g., Pt or Ir) (28), and even reactions leading to carbide formation (e.g., Al) (27). One important lesson in both ceramic and metal composites is that the processing-induced changes in the matrix, such as grain refinement, are as important in determining the final properties of the composites as the amount of CNTs and graphene reinforcements present. Other functional properties, such as anticorrosion, antioxidation, piezoelectric properties, and biocompatibility, are being explored in both ceramic and metal composites with CNTs or graphene. The large reduction in weight, in addition to property enhancements, is a key reason for adding CNT or graphene dispersants in metal and ceramic matrices.

As discussed above, the Achilles’ heel for structural nanocomposites is the difficulty in processing them at meaningful filler loadings. This issue arises from the high interfacial volume between the matrix and the filler, leading to a low rheological percolation threshold. Consequently, the matrix viscosity is found to increase substantially at very low concentrations of CNTs or graphene content, albeit with thixotropic behavior, meaning that the viscosity will drop upon shear (29). Furthermore, the dimensionality of the particle has an important role. Indeed, the 1D nature of nanotubes, which makes them ideal for electrical percolation, means that nanotubes become nearly unprocessable in a conventional polymeric dispersion at loadings of a few volume %, whereas the 2D nature of graphene means that the platelets can slide over each other in shear. This allows systems with ~20 wt % GNPs to be processed, giving graphene a competitive advantage over CNTs for higher filler load applications. One successful route for addressing this viscosity increase at moderate concentrations has been to produce liquid crystalline phases, typically nematic, that require near-monodispersed particle size distribution and an excellent degree of dispersion (30).

A key approach to achieving high loading fractions is to preform the nanoparticles into an organized architecture (e.g., fibers) before introducing the matrix so that traditional composite production techniques can be subsequently used. In a direct analogy to polyaramid coagulation spinning, graphene, graphene oxide, and nanotubes have been spun from the liquid crystalline phases, by using either superacid, organic solvent, or aqueous/surfactant dispersions (31). Alternatively, polymer composite fibers can be produced where the polymer [e.g., poly(vinyl alcohol) or polyacrylonitrile] is used as a binder to improve the interparticle stress transfer, with the binder optionally being pyrolyzed (32). Finally, specifically for CNTs, fibers can be drawn from their growth substrate or zone (33), which bypasses issues with the liquid-phase rheology, allowing extremely long and continuous CNTs to be used. The 1D nature of CNTs means that they are more appropriate for fibers than graphene; in particular, very–high–aspect ratio CNTs can be used in the fibers, leading to high intertube stress transfer. To date, the mechanical properties of CNT fibers have been considerably better than for graphene fibers. Furthermore, graphene and nanotube fibers could have exceptional multifunctionality, including in applications such as energy harvesting, energy storage, and sensing, that will be essential for future demands of smart fabrics and structures.

Another route to using fiber layup production techniques is hierarchical composites in which CNTs or graphene are grafted onto macroscale carbon or glass fiber by either direct growth or deposition (34). These “hairy fibers” show increased out-of-plane improvement in interlaminar shear stress and electrical and thermal conductivity. Alternatively, hybrid systems can be produced where two or more from graphene, nanotubes, and fibers are used within a system through mixing, interacting, and avoiding some of the viscosity and aggregation issues discussed above. For example, nanotubes could combine with graphene to give the low electrical percolation of the former with the high electrical conductivity of the latter. Similarly, additions of graphene or nanotubes to fiber-polymer composites can change the glass transition temperature, and give fire retardancy properties, barrier properties, structural damage sensing, and increased thermal and electrical conductivity. A recent study (35) demonstrated that the addition of 2 wt % graphene to the matrix of a carbon fiber epoxy composite can lead to an increase in axial stiffness of ~10 GPa (∼9%) as the result of the alignment and constraint of the graphene flakes between the fibers. This technology has been reported (36) as being used in the case of the ultralightweight “graphene watch” produced by Richard Mille, the Swiss luxury watchmaker, in conjunction with McLaren Formula 1 motor racing.

Future of CNT or graphene composites

After more than two decades of efforts to create nanotube and graphene composites in various concentrations and with various matrix combinations, the question still stands: How can the spectacular properties of CNT or graphene structures be accessed in their composite structures? The answers have come mainly from trial-and-error–based attempts to control dispersion, filler fraction, and, to a limited extent, interface control. The results thus far do not point to competitive advantages of CNT or graphene particulate additives to composites as real structural reinforcements leading to high strength or increased modulus when compared to carbon fibers. Such an outcome would require continuous fibers and better interfacial engineering, which could still be possible with spun fibers from nanotubes and chemically modified extended graphene structures. Light weight of CNT or graphene fibers could be the best advantage provided by these structures in composites. In addition, in large markets such as the automotive industry, where there is a need for cheap reinforcements that bridge the properties of glass and carbon fibers and to lower production costs, these nanocarbon structures may provide a solution. For all composites, including high-performance ones, the matrix- or fiber-modifying capabilities of these fillers might prove an important advantage, through increasing thermal stability, electrical and thermal conductivities, viscoelastic properties, toughness, and creation of dynamic interfaces toward building multifunctional composites. Indeed, graphene or nanotubes are already used commercially on a large scale as a substitute for traditional conductive additives (e.g., carbon black) for applications such as static dissipative polymers. Figure 6 shows a summary of the important parameters that might determine CNT or graphene additions to composite materials and possible properties that could be enhanced for a range of more advanced applications. The progress in fiber spinning with CNTs and continuous filaments, sheets, and 3D scaffolds with CNT or graphene elements could see a resurgence of CNT or graphene composites for structural applications. If this possibility is realized, then CNT or graphene fiber structures could not only eventually replace today’s carbon fibers but also provide new functionalities such as energy storage and conversion elements (e.g., CNT or graphene fibers acting as reinforcements and electrodes) (37), real-time structure monitoring, sensing, and vibrational and acoustic damping in composites. These attributes would also be ideally suited for thin composite structures (for which micrometer-size carbon fibers may be too thick) or for providing flexibility and resilience in superflexible structures. Unexplored properties of CNT or graphene composites, such as radiation resistance, could also prove vital in determining whether these unique carbon nanostructures prove serious contenders in composites. Finally, beyond the intrinsic scientific and technological issues, the widespread availability of these nanomaterials in large volumes and at relatively low cost will dictate the ultimate success of nanotube and graphene composites.

Fig. 6 A schematic summarizing various CNT and graphene additions to composites with anticipated enhanced properties for the prospective applications along with six key parameters in CNT and graphene composites.

There has been substantial progress in synthesizing continuous fiber spinning with CNT and graphene, and also various yet controllable CNT and graphene structures. The implementation of these unique structures of CNT and graphene as reinforcements in composite materials could provide the opportunity to translate the exceptional properties of the individual CNT and graphene into a variety of engineering applications. It can be envisioned that, if carefully designed and optimized, CNT andgraphene composites could not only replace continuously reinforced carbon fiber composites as state-of-the-art structural composites but also provide new functionalities such as stimuli-responsiveness, energy storage and conversion elements, real-time structural health monitoring, sensing, and vibrational damping in composites. A range of perspective applications including structural applications, multifunctional composites, electrical and thermal management, energy-absorbing composites, and self-stiffening composites with the required properties are summarized along with six key parameters including strength/stiffness, lightweight, interface, energy transfer, conductivity, and dispersion in the composite design and manufacturing.


References and Notes

Acknowledgments: We thank S. K. Lee for her help in creating some of the figures. Funding: J.L. acknowledges support from the U.S. Department of Energy, Office of Basic Energy Sciences, under grant DE-SC0018193. P.M.A. and J.L. acknowledge the NSF Nanosystems Engineering Research Center for Nanotechnology-Enabled Water Treatment (ERC-1449500) for support. J.S. acknowledges support from the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIP) (2018R1A2B2001565). I.A.K. and R.J.Y. acknowledge funding from the European Union’s Horizon 2020 research and innovation program under grant agreement no. 785219. Competing interests: The authors declare no competing interests.

Stay Connected to Science

Navigate This Article