Current Opinion in Solid State and Materials Science 8 (2004) 31–37

Carbon nanotube polymer composites

R. Andrews *, M.C. Weisenberger 1

Center for Applied Energy Research, University of Kentucky, 2540 Research Park Drive, Lexington, KY 40511-8410 USA

Received 7 October 2003; accepted 29 October 2003

Abstract

The state of research into carbon nanotube/polymer–matrix composites for mechanical reinforcement is critically reviewed with emphasis on recent advances in CNT composite toughness. Particular interest is also given to interfacial bonding of carbon nano-tubes to polymer matrices as it applies to stress transfer from the matrix to the CNT. Potential topics of oncoming focus are highlighted.
2003 Elsevier Ltd. All rights reserved.

1. Introduction

Since the documented discovery of carbon nanotubes (CNTs) in 1991 by Iijima [1] and the realization of their unique physical properties, including mechanical, ther-mal, and electrical, many investigators have endeavored to fabricate advanced CNT composite materials that exhibit one or more of these properties [2,*3,*4]. For example, as conductive filler in polymers, CNTs are quite effective compared to traditional carbon black micro-particles, primarily due to their large aspect ratios [5]. The electrical percolation threshold was recently re-ported at 0.0025 wt.% CNTs and conductivity at 2 S/m at 1.0 wt.% CNTs in epoxy matrices [6]. Similarly, CNTs possess one of the highest thermal conductivities known [*7], which suggests their use in composites for thermal management [2]. The main focus of this paper, however, will be on the use of CNTs as discontinuous reinforce-ment for polymer matrices. The CNT can be thought of as the ultimate carbon fiber with break strengths re-ported as high as 200 GPa, and elastic moduli in the 1 TPa range [8,**9]. This, coupled with approximately 500 times more surface area per gram (based on equivalent volume fraction of typical carbon fiber) and aspect ratios of around 103, has spurred a great deal of interest in using CNTs as a reinforcing phase for polymer matrices.

* Corresponding author. Tel.: +1-606-257-0305/859-257-0267; fax: +1-606-257-0220/859-257-0220.

E-mail addresses: andrews@caer.uky.edu (R. Andrews), matt@ caer.uky.edu (M.C. Weisenberger).
1 Tel.: +1-859-257-0322; fax: +1-859-257-0220.

1359-0286/$ - see front matter 2003 Elsevier Ltd. All rights reserved. doi:10.1016/j.cossms.2003.10.006

However, after nearly a decade of research, their potential as reinforcement for polymers has not been fully realized; the mechanical properties of derived composites have fallen short of predicted values. Yet given the magnitude of the CNT’s mechanical proper-ties, significant improvement on current composites should be possible provided means to harness the CNT’s unique attributes exhibited at the nanoscale can be transferred to the macroscale. This essentially defines the fundamental challenge for applied CNT/polymer composites research. How does one effectively manipu-late nanoscale building blocks to assemble useful mac-roscale materials? Some have suggested a bottom-up approach possibly utilizing non-continuum effects at the CNT-polymer interface [*10,*11,*12], while macroscale composite synthesis typically utilizes more traditional continuum ideas. A better understanding of the rela-tionships between processing, interfacial optimization, and composite properties is a major goal of this area of research, which may lead to optimal reinforcement of polymer matrices with CNTs.

The microscale analogue to the current nanocom-posite material is a carbon fiber/epoxy laminate: essen-tially the same materials in dimensions 1000 times larger. A 20 nm thick polymer/aligned CNT nano-lamina’ may in fact be desirable, but currently extremely difficult to scale-up or make reproducibly. Lacking di-rect manipulation, when used as reinforcement in polymers, CNTs are typically first randomly dispersed in a solvent or polymer fluid/melt by sonication or shear mixing followed by further processing to create the composite. It should be noted that the energy input to

32 R. Andrews, M.C. Weisenberger / Current Opinion in Solid State and Materials Science 8 (2004) 31–37

Abbreviations | | | CNT | carbon nanotube | TGA | thermal gravimetric analysis | MWNT | multiwall carbon nanotube | PE–B | polyethylene–butene | SWNT | singlewall carbon nanotube | PAN | polyacrylonitrile | CNF | carbon nanofiber | PS | polystyrene | ISS | interfacial shear strength | PMMA | polymethylmethacrylate | TEM | transmission electron microscope | PVA | polyvinylalcohol | SEM | scanning electron microscope | UHMWPE ultra-high molecular weight polyethylene | AFM | atomic force microscope | PMEMA methyl–ethyl methacrylate copolymer | XPS | X-ray photoelectron spectroscopy | | | | | | |

disperse the CNTs tends to break them into shorter segments [2,*13] decreasing their aspect ratio in the final composite while simultaneously increasing their disper-sibility. However, insufficient dispersion is often cited as a process limitation [14] and the key diminishing factor [*11,*13,15] on the composite’s mechanical properties. Efforts to improve CNT dispersion include: the use of surfactants [**16], and the oxidation or chemical func-tionalization of the CNT surface [17,18].

2. CNT functionalization

Recently, functionalization has been achieved via exposure of vapor grown Pyrografe III carbon nano-fibers (Applied Sciences Inc.) to a CO2/Ar plasma opti-mized with respect to time, pressure, power, and gas concentration resulting in 14.5 at.% oxygen in the first 10 atomic layers [*19]. The presence of functionalities such as hydroxyl, carbonyl, and carboxyl groups were de-tected by XPS. In this case, surface defects in the curved graphene planes of the nanofibers may have increased reactivity promoting the formation of the functionalities. Nitric acid treatment has also been reported to success-fully oxidize the surface of multiwall carbon nanotubes (MWNTs) as detected using diffuse reflection infrared Fourier-transform (DRIFT) spectroscopy [18]. Func-tionalization of the CNT surface cannot only lead to increased dispersibility of the CNTs in various organic solvents and polymers [*4,*20], but also to increase the strength of the interface between the CNT and the polymer matrix [*20]. However, chemical functionaliza-tion may disrupt the bonding of the graphene sheet, and thereby reduce the mechanical properties of the func-tionalized CNT in the final composite.

3. CNT/polymer interfaces

Increasing attention is being focused on the CNT surface, namely the interface between the CNT and surrounding polymer matrix. From micro-mechanics, it

is through shear stress build-up at this interface that stress is transferred from the matrix to the CNTs. Numerous researchers have attributed lower-than-pre-dicted CNT-polymer composite properties to a lack of interfacial bonding [*4,15,21]. If one considers the sur-face of a CNT, essentially an exposed graphene sheet, it is not surprising that interfacial traction is a concern. It is the weak inter-planar interaction of graphite that provides its solid lubricant quality, and resistance to matrix adhesion. This is exaggerated by the chemically inert nature of graphene structures. A June 2002 publi-cation on CNT composites noted In depth study on the stress transfer mechanism of the nanotube composites with different chemical and geometrical properties, ma-trix environments and loading conditions are essential

. . .’ [8]. Since then, some interesting published results have described progress on addressing this issue. The pull-out force necessary to remove a given length of an individual MWNT embedded in polyethylene–butene using an AFM was measured [**22]. The calculated average interfacial shear strength using a Kelly-Tyson approach of 47 MPa (typical values for carbon fiber/ epoxy are in the 30–80 MPa range depending on fiber modulus and surface treatment [23]) was sufficiently high to suggest that covalent bonding between defects in the outer shell of the MWNT and the polymer was occurring. It also suggested that the polymer chains close to the interface behaved differently than the bulk, a logical result when considering the CNT outer diameter is of similar magnitude to the radius of gyration for the polymer. In a report corroborating an interfacial region of non-bulk polymer, a sheathing layer’ of polycar-bonate on pulled out MWNTs was imaged, which gave further evidence of significant interfacial interaction between MWNTs and a polymer [**44]. It was also found that chemical functionalization of the MWNTs augmented the diameter of the polymer sheath sug-gesting chemical augmentation of interfacial bonding. Acid oxidation of MWNTs was again reported to attach carboxylic groups on the surface, which were then re-acted with epoxide-terminated molecules up to12 wt.% by TGA [*20]. These types of functionalized tubes could

R. Andrews, M.C. Weisenberger / Current Opinion in Solid State and Materials Science 8 (2004) 31–37 | 33 |

enhance reinforcement of epoxy resins. Carboxylated tubes have also been reported to augment the cure rate of epoxy resins at lower temperatures [18]. Unfortu-nately, direct and indirect measurement of polymer– CNT interfacial shear strength suggesting good bonding exists conflict with other reports which often cite clean pull-out of CNTs and poor interfacial bonding [*4,15,*24]. The magnitude of CNT strength (>10 times that of typical carbon fiber) may preclude embedded CNT tensile failure in large numbers resulting in the dominant failure mode to be CNT pull-out. Order of magnitude increases in interfacial shear strength may be required for the most efficient strengthening of polymers with CNTs. Optimizing the polymer–CNT interface for nanoscale mechanical reinforcement remains unclear, but the evidence available indicates that chemical means can be effective, and this is likely to be a major focus in the near term.

Theoretical treatments of CNT pull-out were recently reported [*24,25]. Including one in which a ð10; 10Þ sin-glewall carbon nanotube (SWNT) pulled out from a polyethylene matrix was modeled via molecular dynam-ics simulations, and quite a low effective viscosity of 0.2 cP was found for interfacial sliding [*24]. A force of approximately 0.1 nN was required for pull-out to initi-ate for a single tube.

4. Embedded CNT strain characterization by micro-Raman spectroscopy

Earlier publications have associated Raman peak shifting up (to higher wavenumbers) or down (to lower wavenumbers) from the peak near 2700 cm_1 with compressive and tensile strain in the CNT respectively [26,27]. A recent report monitored tensile load transfer to MWNTs in a UHMWPE matrix by the Raman peak shifting at 2691 cm_1 [**28], and found four regions of strain behavior outlined in Table 1. Another report [*29] related the Raman peak shift at 1594 cm_1 to the axial strain in SWNTs embedded in epoxy by: Dxð1594 cm_1Þ | 1 | 1 | | | | | ¼ _cð _ msÞez | ð | Þ | | x0 | | | | |

where the Gruneisen parameter c ¼ 1:24 and the Pois-son ratio ms ¼ 0:28.

5. CNT reinforcement of polymers

Significant toughening of polymer matrices through the incorporation of CNTs has been reported [*4,**16, **28]. A loading of 1 wt.% MWNTs, randomly distrib-uted in an ultra-high molecular weight polyethylene film, was reported to increase the strain energy density by 150% and increase the ductility by 140%. Secondary crystallites, which nucleated from the MWNTs, were attributed a higher mobility and hence the increase in strain energy [**28]. A similar effect was found in aligned MWNT/polyacrylonitrile fibers containing 1.8 vol.% MWNTs with an approximately 80% increase in energy to yield and energy to break [*4]. A process of spinning 60 wt.% SWNT/polyvinylalcohol fibers with pre-drawn energy absorbing capacity nearly 3.5 times spider silk (165 J/g) was reported [**16]. Slippage between SWNT bundles was suggested as the mechanism responsible for the enhancement in the toughness. The addition of 1 wt.% MWNTs to isotactic polypropylene (iPP) was shown to affect crystal nucleation from differential scanning calorimetry and X-ray diffraction measure-ments [*30]. Compared with neat iPP, there was an in-crease in crystallization rate for the composite material with evidence of fibrillar crystal growth rather than spherulite growth (Fig. 1). These modifications in the morphology of a polymer matrix combined with the energy required for CNT debonding and pull-out suggest CNTs may augment the energy absorption or toughness characteristics of the composite.

A twofold increase in the tension–tension fatigue strength for an aligned SWNT/epoxy composite was found in comparison to typical carbon fiber/epoxy composites (Fig. 2) [*31]. Embedded CNTs may effec-tively prolong the formation of and/or bridge micro-cracking/crazing that can propagate and lead to fatigue failure. CNT reinforced polymer composites are seen as a potentially fruitful area for new, tougher or fatigue-resistant materials. Further investigations into the toughness and fatigue properties of these composites are

Table 1

Raman shift (2691 cm_1) vs. percent tensile strain in a MWNT/UHMWPE composite

Region | % e | Shift | Interpretation | 1 | 0–1 | Clear shift down | Tensile loading of MWNTs Elastic response | 2 | 1–10 | Much less apparent or intense shift down | Interfacial stick and slip’ yielding matrix | 3 | 10–15 | Somewhat more apparent shift down | Tensile loading of MWNTs; MWNT knots’ preventing further | | | | PE chain stretching | 4 | >15 | A shift up | Compressive loading of MWNTsElastic recovery from local | | | | matrix failure | | | | |

Four regions of behavior were found for D* Raman band shifting as a function of applied tensile strain to an ultra-high molecular weight polyethylene/MWNT composite film. Data and interpretation from [**28].

34 R. Andrews, M.C. Weisenberger / Current Opinion in Solid State and Materials Science 8 (2004) 31–37

Fig. 1. SEM images crystalline morphology of isotactic polypropyl-ene. The difference in crystalline morphology for (a) neat isotactic polypropylene (iPP), and (b) iPP/1 wt.% MWNT composite is shown in these SEM images. Fibrillar morphology is observed for the com-posite while the neat iPP has a spherullite morphology [*30, p. 525, 526]. Nucleation Ability of Multiwall Carbon Nanotubes in Polypro-pylene Composites, Assouline E, Lustiger A, Barber AH, Cooper CA, Klein E, Wachtel E, Wagner HD, Copyright (2003) Wiley Periodicals Inc., Reprinted by permission of John Wiley and Sons Inc.

Fig. 2. SWNT fracture surface bridging. SWNT ropes are observed bridging a fatigue fracture surface in an epoxy matrix. Twofold in-creases in tension–tension fatigue strength over typical carbon fiber/ epoxy composites were observed [*31, p. 2178]. Reprinted form Car-bon 41, Ren Y, Feng L, Cheng HM, Liao K, Tension–Tension Fatigue Behavior of Unidirectional Single-Walled Carbon Nanotube Rein-forced Epoxy Composite, pg 2177–79, Copyright (2003) with permis-sion from Elsevier.

needed to understand the reinforcing mechanism at work.

Tensile strength and modulus enhancements are continually reported [32–34]; very little of the data achieve reinforcement predicted by a rule-of-mixtures approach especially at loadings beyond 10 vol.% [*4,17,*35] (For aligned, discontinuous fiber reinforce-ment, the Halpin–Tsai model [36] is often used, which approaches the rule of mixtures for large Ef =Em or l=d). For ENT=Em ¼ 100 and l=d ¼ 500, the Halpin–Tsai ap-proach is 95% that of the rule of mixtures for aligned composites. An effective l=d of the CNT, deduced from these sort of comparisons, can shed light on stress transfer to the CNTs [37]. Some have offered new explanations and factors contributing to understanding this discrepancy between experimental and predicted results. In a two paper series using finite element analysis and micro-mechanical methods, it was proposed that observed curvature of embedded CNTs or waviness’ significantly reduced their reinforcement capabilities (by factors from 50 to 200) compared to straight CNTs [*12,*35]. It was also noted that other indistinguishable factors contribute to the low values measured in ex-perimental data: including weak interfacial bonding, insufficient dispersion, and degradation of the CNTs due to processing [*35]. At very low strains, it was suggested that the effect of poor interfacial shear strength should not affect the composite’s modulus, implying a mea-surable elastic response below the strain detrimental to the CNT interface [*10]. It was also suggested that the diameter distribution of embedded MWNTs would have

Fig. 3. Effect of MWNT outer diameter on aligned MWNT composite tensile modulus. The tensile modulus of a MWNT composite is modeled as a function of embedded MWNT outer diameters for var-ious MWNT lengths and volume fractions. The strong dependency on MWNT outer diameter results from using an effective MWNT mod-ulus that accounts for only the outer shell carrying any of the tensile load [*10, p. 581]. Thostenson ET, Chow TW, On the Elastic Prop-erties of Carbon Nanotube-Based Composites: Modelling and Char-acterization, Copyright (2003) IOP publishing limited, Reprinted with permission.

| R. Andrews, M.C. Weisenberger / Current Opinion in Solid State and Materials Science 8 (2004) 31–37 | | 35 | | Table 2 | | | | | | | | | | | | Selected mechanical properties of CNT/polymer composites | | | | | | | | | | | | | | | | | | | | CNT | Matrix | Conc. (%) | Loading | Modulus | Yield | Strength | Tough- | Max. strain | ISS | Ref. | | | | | | | stress | | ness | | (Mpa) | | | | | | | | | | | | | | | MWNT | None | 100 | Tensilea | 910 GPa | | 150 GPa | | | | [**9] | | MWNT | None | 100 | Tensileb | 270–950 GPa | | 11–200 GPa | | 12% | | [39] | | MWNT | PAN | 1.8 vol.% | Tensile | +36% | +46% | +31% | +80% | _100% | | [*4] | | SWNT | PVA | 60 wt.% | Tensile | 80 GPa | | 1.8 GPa | 570 J/g | | | [**16] | | MWNT | PS | 5 wt.% | DMA | +49% | | | | | | [17] | | | | | (25 LC) | (aligned) + 10% | | | | | | | | | | | | (random) | | | | | | | | MWNT | PE–Bc | | AFMc | | | | | | 47 | [**22] | | | | | | | | | | | MPa | | | MWNT | UHMWPE | 1 wt.% | Tensile | +25% | +48% | +25% | +150% | +60 to +140% | | [**28] | | SWNT | PAN | 4 wt.% | AFMd | _ +100% | | e | | )37.5% | | [14] | | CNF | PMMA | 5 wt.% | Tensile | +50% | | +200% | | | | [34] | | MWNT | PMEMA | 1 wt.% | DMA | +200% | | | | | | [33] | |

(+/)) indicates an enhancement/diminishment from the neat matrix. Recently reported mechanical property enhancements for various CNT/polymer composites are presented with additional information on the method used to collect the data. Experimentally measured values of MWNT mechanical properties are also presented.

a In situ TEM tensile testing of individual MWNTs via a micro-fabricated device. b In situ SEM tensile testing of individual MWNTs via dual AFM tips.

c In situ AFM pull-out of individual MWNT from PE–B film. d Moduli measured using an AFM nanoindentation technique. e Compressive strength via loop test.

a very strong effect on the composite’s modulus (Fig. 3) [*10]. In the case of MWNTs, an effective modulus was deduced assuming that the outer shell carries essentially all the load [26] such that its modulus, ENT, was reduced in proportion to the ratio of the area of the annular outer shell thickness to the total cross-sectional area. A MWNT diameter distribution was therefore essential for accurate prediction of the composite’s modulus [*10].

In another study, increased inter-tube friction of SWNT bundles by the introduction of twist, which served to flatten’ the SWNTs, was predicted by molec-ular mechanics modeling to have increased the bundle’s load carrying capacity [38]. Some recent mechanical property results for CNT/polymer composites are sum-marized in Table 2.

6. A look forward

The ultimate goal remains macroscale CNT/polymer composites that are optimally reinforced. Increases in composite moduli are almost always observed even for very low loadings of CNTs. It should be feasible to in-crease the modulus of a polymer/CNT composite up to and beyond that achievable with high modulus graphite fiber. Alignment of CNTs has been achieved in poly-meric matrices most often by means of shear or elon-gational flow [*4,14,**16,17,**28], and recently by a magnetic field [40] resulting in film and fiber geometries. However, strengthening of polymer/CNT composites is less commonly reported. Due to the small number of defects per unit length, the most impressive mechanical property of the CNT is its tensile strength with recent

experimentally measured values up to 150 GPa for MWNTs [**9]. These high strengths imply break strains of approximately 10% with even higher break strains reported in the literature [8,41,42]. Recent reports indi-cate that tensile failure of MWNTs that carry all of the load within the outer shell is initiated by the formation of a Stone–Wales defect in which 2 C–C bonds are broken and 2 new C–C bonds are formed such that a pair of pentagons and a pair of heptagons results (Fig. 4) [41,42]. Increasing numbers of these dislocations leads to necking and ultimately, failure of the CNT [42]. Often, however, in a polymer/CNT composite a reduc-tion in strength is observed suggesting that the CNTs

Fig. 4. A Stone–Wales defect. A representation of a Stone–Wales defect in the graphitic structure of a CNT under tensile stress shows the rearrangement of the hexagonal graphitic structure into two pairs of heptagons and pentagons [42, p. 115401-1]. Wei C, Cho K, Srivistava D, Physical Review B 67, 115407, 2003, Copyright (2003) by the American Physical Society, Reprinted with permission.

36 R. Andrews, M.C. Weisenberger / Current Opinion in Solid State and Materials Science 8 (2004) 31–37

may promote crystalline defects in the matrix or, con-sidering their size, act as defects themselves. This sug-gests a critical volume fraction of CNTs is required for hindering matrix strain allowing strengthening to occur as typically observed in composites. Classically this re-sults from the composite’s failure criterion being fiber rupture at the fiber break strain, which is typically smaller than the matrix break strain. It remains to be seen whether or not a 10% elastic tensile strain can be transferred to CNTs embedded in a polymer, fully exploiting their load carrying capacity, and at this strain, most polymeric matrices will have begun to flow plastically. Furthermore, many authors report higher loadings of CNTs in a composite do not perform as well as lower loadings [*4,34] suggesting an increase in voids or other defects as the loading of CNTs increases, likely due to the difficulty of homogeneously dispersing con-centrated CNT/polymer melts or solutions. All this begs the question as to what is the failure criterion applicable to CNT/polymer composites, a question complicated by CNT misalignment, debonding and pull-out, end-effects and dispersion, to name a few. Strengthening a polymer matrix with CNTs may best be realized in an aligned concentrated CNT nano-lamina’, where all polymer is near a CNT interface [*43], provided the interphase polymer does indeed have higher mechanical properties than the bulk. In situ TEM tensile testing of these nano-lamina could provide insight into this issue. Lay-up of nano-lamina, or perhaps self-assembly could lead to optimally reinforced macroscale CNT/polymer com-posites.

Acknowledgement

The authors wish to acknowledge financial support from the NSF MRSEC Grant DMR-9809686.

References

The papers of particular interest have been high-lighted as:
* of special interest; * of outstanding interest.

[1] Iijima S. Helical microtubules of graphitic carbon. Nature 1999;354:56–8.

[2] Biercuk MJ, Llaguno MC, Radosvljevic M, Hyun JK, Johnson AT. Carbon nanotube composites for thermal management. Appl Phys Lett 2002;80:15.

[*3] Ounaies Z, Park C, Wise KE, Siochi EJ, Harrison JS. Electrical properties of single wall carbon nanotube reinforced polyimide composites. Compos Sci Technol 2003;63:1637–46, This work describes SWNT polyimide composites with low percolation thresholds and conductivities exceeding that needed for anti-static applications.

[*4] Weisenberger MC, Grulke EA, Jacques D, Rantell T, Andrews R. Enhanced mechanical properties of polyacrylonitrile/multi-wall carbon nanotube composite fibers. J Nanosci Nanotech 2003;3(6), A process for making MWNT/PAN fibers is presented with data pointing to an 80% increase in composite fiber toughness.

[5] Colbert DT. Single-wall nanotubes: a new option for conduc-tive plastics and engineering polymers. Plastics Additives Compd 2003:18–25.

[6] Sandler JKW, Kirk JE, Kinloch IA, Shaffer MSP, Windle AH. Ultra-low electrical percolation threshold in carbon-nanotube-

epoxy composites. Polymer 2003;44(19):5893–9.

[*7] Kim P, Shi L, Majumdar A, McEuen PL. Thermal transport measurements of individual multiwalled nanotubes. Phys Rev Lett 2001;87(21):2155021–4, Thermal transport measurements of individual carbon nanotubes using a microfabricated sus-pended device yield a room temperature thermal conductivity of 3000 W/m K.

[8] Lau K-T, Hui D. The revolutionary creation of new advanced materials––carbon nanotube composites. Composites Part B

2002;33:263–77.

[**9] Demczyk BG, Wang YM, Cumings J, Hetman M, Han W, Zettl A, et al. Direct mechanical measurement of the tensile strength and elastic modulus of multiwalled carbon nanotubes. Mater Sci Eng 2002;A334:173–8, In situ TEM tensile testing of individual MWNTs is achieved using a microfabricated device that applied pure tension to the tubes. A break strength of 150 GPa was found.

[*10] Thostenson ET, Chou T-W. On the elastic properties of carbon nanotube-based composites: modelling and characterization. J Phys D: Appl Phys 2003;36:573–82, This work suggests the effect of diameter on MWNT modulus is very significant, and uses a diameter distribution to partition the moduli of MWNTs in a composite deducing an effective modulus.

[*11] Odegard GM, Gates TX, Wise KE, Park C, Siochi EJ. Constitutive modeling of nanotube-reinforced polymer com-posites. Compos Sci Technol 2003;63:1671–87, A constitutive model for a SWNT-polymer system is presented that takes into account atomic interactions between the SWNT and the polymer using a molecular dynamics and an ’equivalent-continuum model’.

[*12] Bradshaw RD, Fisher FT, Brinson LC. Fiber waviness in nanotube-reinforced polymer composites––II: Modeling via numerical approximation of the dilute strain concentration tensor. Compos Sci Technol 2003;63:1705–22, The degree to which curvature of embedded CNTs affects the composite modulus is considered quantitatively resulting in significant reductions in modulus for increasing curvature. Somewhat of a refinement of part I.

[*13] Bai JB, Allaoui A. Effect of the length and the aggregate size of MWNTs on the improvement efficiency of the mechanical and electrical properties of nanocomposites––experimental investi-gation. Composites Part A: Appl Sci Manufact 2003;34(8):689– 94, MWNT epoxy composites are considered for mechanical and electrical properties. MWNTs are suggested to block molecular motions of the polymer matrix. Longer MWNTs were found to be more effective for lowering the percolation threshold.

[14] Ko F, Gogotsi Y, Ali A, Naguib N, Ye H, Yang G, et al. Electrospinning of continuous carbon nanotube-filled nanofiber yarns. Adv Mater 2003;15(14):1161–5.

[15] Bai J. Evidence of the reinforcement role of chemical vapour deposition multi-walled carbon nanotubes in a polymer matrix.

Carbon 2003;41:1309–28.

[**16] Dalton AB, Collins S, Munoz E, Razal JM, Ebron VH, Ferraris JP, et al. Super-tough carbon-nanotube fibres. Nature 2003;423:703, This work reports the synthesis of very concen-

R. Andrews, M.C. Weisenberger / Current Opinion in Solid State and Materials Science 8 (2004) 31–37 | 37 |

trated (60 wt.%) SWNT/PVA fibers that display toughness values that are incredibly large in magnitude.

[17] Thostenson ET, Chou T-W. Aligned multi-walled carbon nanotube-reinforced composites: processing and mechanical characterization. J Phys D: Appl Phys 2002;35:L77–80.

[18] Bae J, Jang J, Yoon S-H. Cure behavior of the liquid-crystalline epoxy/carbon nanotube system and the effect of surface

treatment of carbon fillers on cure reaction. Macromol Chem Phys 2002;203(15):2196–204.

[*19] Bubert H, Haiber S, Brandl W, Marginean G, Heintze M, Bruser V. Characterization of the uppermost layer of plasma-treated carbon nanotubes. Diam Relat Mater 2003;12:811–5, Oxidation of MWNT surfaces with and CO2/Ar plasma is described along with characterization of these surfaces via XPS and depth profiling. 14.5 at.% surface O.

[*20] Eitan A, Jiang K, Dukes D, Andrews R, Schadler LS. Surface modification of multiwalled carbon nanotubes: toward the tailoring of the interface in polymer composites. Chem Mater 2003;15(16):3198–201, Oxidation of MWNTs with acid is followed by chemistry to attach epoxide groups to the MWNT surface for incorporation into an epoxy matrix with enhanced load transfer.

[21] Nan C-W, Shi Z, Lin Y. A simple model for thermal conductivity of carbon nanotube-based composites. Chem Phys

Lett 2003;375:666–9.

[**22] Barber AH, Cohen SR, Wagner HD. Measurement of carbon nanotube-polymer interfacial strength. Appl Phys Lett 2003;82(23):4140–2, Repeatable measurements of the force required to remove a known length of an individual MWNT from a polymer were done in an AFM resulting in an interfacial shear strength of 47 MPa.

[23] Donnet J-B, Wang TK, Peng JCM, Rebouillat S. Carbon fibers. 3rd ed. (revised and expanded ed.). New York: Marcel Dekker Inc.; 1998. p. 185–7.

[*24] Frankland SJV, Harik VM. Analysis of carbon nanotube pull-out from a polymer matrix. Surf Sci 2003;525:L103–8, The pull-out of a SWNT from PE was simulated using molecular dynamics and an interfacial viscosity for pull-out to occur was found to be 0.2 cP.

[25] Lau K-T. Interfacial bonding characteristics of nanotube/ polymer composites. Chem Phys Lett 2003;370:399–405.

[26] Schadler LS, Giannaris SC, Ajayan PM. Load transfer in carbon nanotube epoxy composites. Appl Phys Lett 1998; 73(26):3842–4.

[27] Ajayan PM, Schadler LS, Giannaris C, Rubio A. Single-walled carbon nanotube-polymer composites: strength and weakness.

Adv Mater 2000;12(10):750–3.

[**28] Ruan SL, Gao P, Yang XG, Yu TX. Toughening high performance ultrahigh molecular weight polyethylene using multiwalled carbon nanotubes. Polymer 2003;44(19):5643–54, This work reports 150% increases in strain energy density for 1 wt.% MWNTs in UHMWPE. Secondary crystals nucleating from the MWNTs are attributed higher mobility under stress. Raman spectroscopy is used to characterize both the embedded MWNTs and the matrix indicating 4 regions of strain that occur.

[*29] Puglia D, Valentini L, Armentano I, Kenny JM. Effects of single-walled carbon nanotube incorporation on the cure reaction of epoxy resin and its detection by raman spectros-copy. Diam Relat Mater 2003;12:827–32, SWNTs are shown to accelerate the cure reaction of and epoxy resin. A relationship

between Raman peak shifting and actual tensile strain in embedded SWNTs is presented.

[*30] Assouline E, Lustiger A, Barber AH, Cooper CA, Klein E, Wachtel E, et al. Nucleation ability of multiwall carbon nanotubes in polypropylene composites. J Polym Sci Part B 2003;41:520–7, MWNTs are shown to nucleate faster crystal growth in iPP by Ozawa kinetic theory of nonisothermal crystallization using DSC data. Fibrillar morphology is ob-served in the presence of MWNTs.

[*31] Ren Y, Li F, Cheng H-M, Liao K. Tension–tension fatigue behavior of unidirectional single-walled carbon nanotube rein-forced epoxy composite. Carbon 2003;41:2159–79, The fatigue strength of a SWNT epoxy composite is found to be approx-imately two times larger than that of carbon fiber epoxy.

[32] Coleman JN, Blau WJ, Dalton AB, Munoz E, Collins S, Kim BG, et al. Improving the mechanical properties of single-walled carbon nanotube sheets by intercalation of polymeric adhesives. Appl Phys Lett 2003;82(11):1682–4.

[33] Velasco-Santos C, Martinez-Hernandex AL, Fisher FT, Ruoff RS, Castano VM. Dynamical-mechanical and thermal analysis of carbon nanotube-methyl–ethyl methacrylate nanocompos-ites. J Phys D: Appl Phys 2003;36:1423–8.

[34] Zeng J, Saltysiak B, Johnson WS, Schiraldi DA, Kumar S. Processing and properties of poly(methyl methacrylate)/carbon

nano fiber composites. Composites Part B: Eng, in press. [*35] Fisher FT, Bradshaw RD, Brinson LC. Fiber waviness in

nanotube-reinforced polymer composites-I: Modulus predic-tions using effective nanotube properties. Compos Sci Technol 2003;63:1689–703, The degree to which curvature of embedded CNTs affects the composite modulus is considered quantita-tively resulting in significant reductions in modulus for increasing curvature.

[36] Mallick PK. Fiber reinforced composites: materials, manufac-turing, and design. 2nd ed (revised and expanded ed.). New York: Marcel Dekker, Inc.; 1993. p. 129–30.

[37] Cadek M, Coleman JN, Barron V, Hedicke K, Blau WJ. Morphological and mechanical properties of carbon-nanotube-reinforced semicrystalline and amorphous polymer composites. Appl Phys Lett 2002;81(27):5123–5.

[38] Qian D, Liu WK, Ruoff R. Load transfer mechanism in carbon nanotube ropes. Compos Sci Technol 2003;63:1561–9.

[39] Wagner HD, Lourie O, Feldman Y, Tenne R. Stress-induced fragmentation of multiwall carbon nanotubes in a polymer matrix. Appl Phys Lett 1998;72(2):188–90.

[40] Kimura T, Ago H, Tobita M, Ohshima S, Kyotani M, Yumura M. Polymer composites of carbon nanotubes aligned by a magnetic field. Adv Mater 2002;14(19):1380–3.

[41] Wei C, Cho K, Srivastava D. Tensile yielding of multiwall carbon nanotubes. Appl Phys Lett 2003;82(15):2512–4.

[42] Wei C, Cho K, Srivastava D. Tensile strength of carbon

nanotubes under realistic temperature and strain rate. Phys Rev B 2003;67:1154071–6.

[*43] Koratkar NA, Wei B, Ajayan PM. Multifunctional structural reinforcement featuring carbon nanotube films. Compos Sci Technol 2003;63:1525–31, Thin films of tightly packed aligned MWNTs were used as inter-layers in composites to enhance inter-ply bonding, laminate stiffness, and structural damping.

[**44] Ding W, Eitan A, Fisher FT, Chen XH, Dikin DA, Andrews R, et al. Direct observation of polymer sheathing in carbon nanotube–polycarbonate composites. Nano Lett 2003;3(11): 1593–7.