Nanodiamond-polymer composites

Diamond & Related Materials 58 (2015) 161–171

Contents lists available at ScienceDirect

Diamond & Related Materials journal homepage: www.elsevier.com/locate/diamond

Nanodiamond–polymer composites Vadym N. Mochalin ⁎, Yury Gogotsi ⁎ Department of Materials Science and Engineering, Drexel University and A. J. Drexel Nanomaterials Institute, Philadelphia, PA 19104, United States

a r t i c l e

i n f o

Article history: Received 19 March 2015 Received in revised form 9 July 2015 Accepted 10 July 2015 Available online 23 July 2015 Keywords: Nanodiamond Polymer composites Nanocarbon Nanocomposites

Contents 1. Introduction . . . . . . . . . . . . . . . . . . 2. Advantages of nanodiamond in polymer composites 3. Nanodiamond — thermoplastic polymer composites 4. Nanodiamond—thermosetting polymer composites 5. Nanodiamond — elastomer composites . . . . . 6. Conclusions and outlook . . . . . . . . . . . . Prime novelty statement . . . . . . . . . . . . . . . Acknowledgment . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

1. Introduction Historically, polymer reinforcement has been one of the first and most obvious applications for nanoparticles, which emerged in the early 1990s. The first reports on polymer nanocomposites can be traced back to 1992 [1–5]. However, carbon-black-filled rubber, one of the premier materials engineering successes of the 20th century, can be considered as a polymer nanocomposite known from much earlier time. According to modern definition, polymer nanocomposite is a material consisting of a polymer matrix and filler nanoparticles, which have at least one dimension in the nanoscale range (less than 100 nm) and

⁎ Corresponding author. E-mail address: [email protected] (Y. Gogotsi).

http://dx.doi.org/10.1016/j.diamond.2015.07.003 0925-9635/© 2015 Elsevier B.V. All rights reserved.

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

161 163 164 166 169 169 169 169 170

can be “organic” or “inorganic” in the sense these terms are used in chemistry [6]. In a typical case, the polymer matrix constitutes the main part of the nanocomposite's weight or volume. However, sometimes the ratio of polymer to nanoparticles is reversed, in which case the composite material can be considered as a polymerimpregnated porous solid body [7] or as self-suspended nanoparticle fluid also called “nanoscale organic hybrid material” [8]. The nanofillers can be zero, one, or two-dimensional, having 0, 1, or 2 dimensions larger than 100 nm, correspondingly. Among different nanofillers, carbonaceous nanoparticles such as graphene (2D), carbon nanotubes (CNTs — 1D) and nanodiamond (0D) are very promising due to their unique and highly tailorable combination of properties, such as mechanical strength, electrical conductivity, aspect ratio, etc [9]. The main benefits of nanoparticles are related to their small size and a much larger surface-to-volume ratio as compared to micrometer-

162

V.N. Mochalin, Y. Gogotsi / Diamond & Related Materials 58 (2015) 161–171

sized particles. In polymer nanocomposites, the larger surface-tovolume ratio of the nanofillers results in a dramatic increase in the volume of the “interphase”, i.e., the volume of the polymer which is close enough to a nanoparticle so that its properties are influenced by the nanoparticle and are different from the bulk polymer. In fact, when the filler size is reduced to the nanoscale, the volume of the interphase may exceed the volume of the particle, whereas for micrometer-sized fillers the volume of the particles always exceeds that of the interphase [5]. This, in turn, allows nanofillers to exert significant impact on the polymer properties even at low concentrations. For example, for 1 vol.% dispersion of 4 nm diameter nanospheres in a polymer, assuming the interphase thickness 6 nm around each nanoparticle, the volume fraction of the interphase in the composite is ~ 63 vol.% — more than half of the total volume [5]. Therefore, instead of adding 15–50 vol.% of standard fillers, the same improvements in mechanical properties can be achieved with only 5 vol.% of a nanofiller, resulting in reduced cost while also reducing weight [10] and preserving valuable properties of polymer matrix, e.g., plasticity and toughness. At larger sizes, however, the nanofillers quickly become less effective: with 1 vol.% of 40 nm diameter nanospheres and same interphase thickness, the interphase fraction reduces to 12 vol.% [5]. Small size provides other, less obvious benefits. For example, when the nanofiller particles are smaller than the random coil size or radius of gyration of the polymer, good dispersion is achievable even in cases when the particles and the matrix might be otherwise considered incompatible [11]. Besides small size, nanofillers bring in useful properties of the material they are made of. For example, carbon nanotubes and graphene are conductive and can be used to improve electrical and thermal conductivity of the polymers; luminescent semiconductor quantum dots impart luminescence to the composite, novel 2D transition metal carbides/nitrides (MXenes) may provide higher mechanical strength and conductivity, while in contrast to graphene being hydrophilic and strongly interacting with water-soluble polymers [12]. Finally, reactive chemical groups exposed on the surface of some nanoparticles can be used to form covalent bonds (a strong “interface”) with the polymer matrix, which, in combination with the large surface-to-volume ratio of the nanoparticles (high fraction of the interphase) may result in super strong or highly thermally conductive polymer composites. Additionally, these surface functional groups can be used to improve the nanoparticle dispersion in the matrix through achieving favorable nanoparticle – polymer

interactions, as well as suppressing unfavorable nanoparticle – nanoparticle interactions. Composite properties depend on its composition and structural characteristics. Taking into consideration mechanical properties, for example, the Young's modulus enhancement depends on intrinsic properties of the matrix and the filler, as well as the interactions between them. Tensile strength and ultimate strain, on the other hand, are more sensitive to defects. Due to a complex interplay of these structural characteristics, as well as additional factors such as thickness and properties of the interface and interphase, nanofiller dispersion quality, intrinsic properties of the nanoparticles, etc., it is not surprising that creating a quantitative theory of nanocomposite properties is a very challenging task [13], necessitating the need to use simplified models (such as additive schemes) or adaptation of predictive models developed for traditional composites (for example composite micromechanics models), as well as extensive use of computational atomistic-level modeling [14,15]. However, special care should be exercised when micromechanics models developed for traditional composites with moderate concentrations of the microfiller are transferred to nanocomposites where the nanofiller concentration is typically much lower, because traditional theories of micromechanics do not account for a meaningful change in properties when so little material is replaced. Thus, the mechanics must be understood as arising from load transfer as much as from load bearing [1], emphasizing again the role of the interface between the components of the nanocomposite. In view of this, three-phase micromechanics models taking into consideration the matrix, the filler, and the interphase are being developed [16–19], producing an overall better agreement with experimental and atomistic computational modeling results. Current research in polymer nanocomposites has many facets, aiming at discovering novel nanofillers, improving dispersion and bonding (interface) between the nanoparticles and the matrix, developing nanocomposite theory, resolving manufacturing issues such as unacceptably high viscosity of some polymer-nanofiller melts, etc. Although our understanding of all these aspects has considerably improved over the past years, the promises of nanocomposites to a large extent still remain elusive. Similar to a majority of other engineering materials, the man-made nanocomposites are either stiff (high Young's modulus) but brittle (low strain at failure), like ceramics or extensible but weak, like rubber. At the same time, nature-made nanocomposites such as

Fig. 1. Blended high resolution TEM and atomistic model of a single ND particle, showing internal diamond structure with a typical twin defect and surface functional groups.


silk, bone, nacre, enamel, etc. exhibit unparalleled combinations of stiffness, strength, extensibility and toughness by exploiting hierarchical structures in which stiff nanometer-size crystallites are embedded and dispersed in soft (glyco)protein matrices. Currently, soft man-made nanocomposites (e.g., rubber) demonstrate a better commercial success, whereas hard man-made nanocomposites still have marginal performance improvements and would benefit from well dispersed, strongly interfaced hard nanofillers. Among different nanofillers, nanodiamond particles (NDs) usually produced by detonation (Fig. 1) hold a special place [20]. ND, also known as ultra-dispersed diamond or ultrananocrystalline diamond, is a member of a diverse family of nanocarbons that includes fullerenes, nanotubes, graphene, amorphous dense and porous networks. These nanoscale approximately spherical diamond particles with ~5 nm diameter and a narrow particle size distribution feature chemically inert cores with superior mechanical properties characteristic of bulk diamond, and fully accessible external reactive surfaces terminated by a large number of tailorable functional groups [21]. Unique characteristics combined with a moderate production cost and commercial availability favorably distinguish NDs from many other nanoparticles (nanotubes, graphene, polymers, metals, ceramics, etc.), which have been tried as fillers in polymer nanocomposites. ND was discovered and initially studied in the 1960s in the former USSR, where the first experiments on incorporation of this, then called “diamond containing” carbon into polymers, have been carried out by researchers at “Altai” center. These early studies dealt with impure and poorly characterized NDs, sometimes failed to provide details about the polymer used, and often produced inconsistent results. A review of a large body of literature on ND–polymer composites, published in Russian, can be found in [22–25]. The purpose of this thematic article is to summarize some of the most important, in authors' opinion, results and existing knowledge on ND–polymer composites. We will start with general advantages of ND for nanocomposites, followed by the analysis of representative results grouped by three main classes of polymers and focus on mechanical properties of the composites, occasionally referring to other properties.

18

i

influence on polymer properties. The role of the shape of micro- and nanofillers was analyzed in [5,26]. Assuming, as a first approximation, that the thickness of the interphase (t) is independent on the shape and size of the fillers, the ratio (k) of the interphase thickness to the smallest dimension of the particle (r or h) varies with particle size and shape as shown in Fig. 2b [5]. This results in an increase of δi when particle size reduces from micrometers to nanometers with maximum δi achieved for spherical particles at any size, especially pronounced in the nanometer range. In this size range, the calculated difference in δi between spheres, rods and plates is 10 to 100 times, respectively (Fig. 2b). Alishahi et al. have reported δND (spheres) 7 times that of δ GO (plates) and 2.5 times that of δ CNT or δ CNF (both rods) [26]. Nearly spherical shape of ND is also advantageous for processing of composite melts. Elongated nanofillers such as CNTs even at

b)

0.1 wt. % 0.5 wt. %

10

1.5

1015

ND 1.0

13

10

GO

1012

0.5

11

10

1

Spheres

1000 Rods 100 10

δι

CNT

16

Surface / Volume (1/m)

Number of particles

10

10

Main advantages of ND particles for nanocomposites stem from these unique properties: 1) diamond structure that provides superior Young's modulus, hardness, high thermal conductivity and electrical resistivity, low coefficient of friction, chemical stability, and biocompatibility; 2) small and uniform size, eliminating the need for costly fractionation procedures required for some other nanoparticles; 3) nearly spherical shape as compared to carbon nanotubes, graphene, etc. (Fig. 1 and below); 4) large and accessible external surface, maximizing interactions with the matrix, i.e. the interphase volume; 5) rich and tailorable surface chemistry, providing great flexibility for rational design of ND–matrix interface. Small size of a ND particle in all three dimensions (0D) translates into orders of magnitude higher number of the nanoparticles in the matrix at any given loading compared to 1D or 2D nanofillers, which have at least one much larger dimension [26] (Fig. 2a). As mentioned above, small size results in a much larger volume of the interphase in the composite. However, shape of the nanoparticles also makes a difference. Among nanoparticles of comparable size but different shapes, spherical (0D) nanofillers, such as NDs, maximize the interphase volume per unit of nanoparticle volume ðδi ¼ V int V Þ, i.e., 0D particles exert the strongest

2.0

CNF

17

14

2. Advantages of nanodiamond in polymer composites

1010

10 Plates k = 10 1 k=1

100

0.1 0.01

Particle size, nm

a) 10

163

k = 0.1 k = 0.01 1000

1E-3

109

0.01

Nanofillers

0.1

1

10

100

1000

Aspect ratio

c)

2r

2r

2r

h Sphere, r Aspect ratio = 1 k = t/r

Plate, 2r > h Aspect ratio = h/(2r) < 1 k = 2t/h

h

Rod, 2r < h Aspect ratio = h/(2r) > 1 k = t/r

Fig. 2. a) Surface-to-volume ratio and number of particles in the matrix at 0.1 and 0.5 wt.% loading for nanofillers of different dimensions and shapes (CNF — carbon nanofiber, GO — graphene oxide) [26]; b) the ratio of the interphase volume to the particle volume (δi) as a function of the particle aspect ratio and the ratio of the interphase thickness (t) to the particle size (r or h). As particles decrease in size to less than 100 nm, the interphase can dominate the volume and hence the physical properties of the composite, particularly for spheres [5]; c) definition of the aspect ratio and the ratio of the interphase thickness to the smallest dimension of the particle (k) (r is radius, h is height or length) for particles of different shapes.

164


Fig. 3. (a) Two major issues in polymer nanocomposites are related to poor dispersion (when nanoparticles gather together in large micron-sized agglomerates, creating defects in the polymer matrix rather than improving it) and weak interface between the nanoparticles (even if they are well dispersed) and the matrix, resulting in poor load transfer. Overcoming this issues, when nanoparticles become well dispersed and covalently bonded to the matrix, will result in superior composites with very high mechanical, thermal, and other properties; (b) schematic illustration of gas (red, left side of the panel) and wet (blue, right side of the panel) chemical modification techniques for nanodiamond [21]. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

low loadings increase the viscosity of polymer composite melts beyond processing limits, representing a serious manufacturing problem [27]. Furthermore, compared to spheres, cylindrical and plate-like shapes provide more favorable geometry for van der Waals interparticle interactions, resulting in undesirable stacking of graphene and bundling of CNTs. NDs, due to their spherical shapes, are less prone to increase polymer viscosity and do not form stacks or bundles, meaning that, at least in theory, they can be dispersed in the matrix better than 1D and 2D nanoparticles. The interfacial area of ND is also very large. Among typical nanofillers analyzed in [26], which include 5 nm diameter ND, 25 nm diameter × 20 μm length CNTs, 50 nm diameter × 30 μm length carbon nanofibers (CNFs), and 15 μm diameter × 8 nm thickness graphene oxide (GO) nanoplates, NDs have the largest surface area per unit of particle volume (Fig. 2a). In other words, the interface between the components of the composite can be maximized when ND is used as nanofiller. Thus, a larger number of particles, interphase volume, and the interfacial area per unit of the nanoparticle volume (Fig. 2) are the reasons for anticipated higher efficiency of ND (and spherical nanofillers in general) in enhancement of the mechanical properties of nanocomposites. Further enhancements can be achieved due to the rich surface chemistry of ND, a characteristic which advantageously distinguishes it from many other nanofillers, and which can be tailored to address the two main issues in polymer nanocomposites: i) nanofiller agglomeration and ii) weak interface between the nanofiller and the matrix (Fig. 3a) without compromising mechanical properties of the filler. In contrast to sp2 carbon, ceramic, or metal nanoparticles, ND exposes many functional groups on its surface (Fig. 1), which constitute an integral part of the structure, determining chemical properties [21,28,29], stability [30,31], and thermodynamics [32] of the ND particle. Purification of ND, normally achieved via oxidation [33,34], removes surface carbon layers along with attached heteroatoms and immediately creates new oxygen containing, mainly carboxyl (COOH) groups on fresh ND surface [35]. Some of the numerous ways of surface modification of purified COOH-terminated ND are summarized in Fig. 3b [21]. Importantly, ND surface modification can be performed to a large extent and variety, i.e., hundreds of functional groups of the same or different types can be formed on the surface of a ND particle [32,36–39] without any detrimental effect on its structure, in a big contrast to CNTs and graphene, where excessive attachment of functional groups disrupts the aromatic sp2 carbon structure, resulting in degradation of properties. Due to rich chemistry of carbon and sp3 hybridization, ND is the champion of surface chemistry among all nanofillers in terms of the variety of possible chemistries, as well as the number of surface functional groups per particle [28,29].

We will now consider some examples of ND–polymer composites where ND addition provided benefits in terms of improved mechanical and, occasionally, other properties. 3. Nanodiamond — thermoplastic polymer composites In first attempts to reinforce thermoplastic polymers, non-modified as-produced NDs were simply mixed with the polymers. It has been shown that even in low concentrations and without surface functionalization, ND sometimes can reinforce some thermoplastics. Although this “trial and error” approach adopted in early studies is not recommended because the chances to achieve the improvement are slim, it is still in use due to its simplicity. For example, improved elastic strain at uniaxial extension was observed due to the addition of glass beads in combination with only 0.1 wt.% of as-received ND to polydimethylsiloxane. According to viscosity measurements of the melts, ND changes the conformation of polydimethylsiloxane macromolecules, resulting in better mechanical performance of the composite [40]. This is an example of interphase-mediated property change. In another study, increased Young's modulus and glass transition temperature (Tg) were achieved due to the addition of 0.25 wt.% of as-received ND to polyurethane-2-hydroxyethylmethacrylate (PU-PHEMA). The improvements in this case, explained by a reaction between ND COOH groups and the isocyanate groups formed during the polymerization of PU-PHEMA [41], provide an example of interface-mediated property change. In most situations, however, modification of ND is required to achieve improved performance in composites. To increase affinity of ND to polyethylene (a polymer consisting of a –CH2– backbone with no side chains) alkyl groups of variable length were grafted to ND. DSC measurements performed on these polyethylene–ND composites indicate an increased crystallinity, as well as higher crystallization and melting temperatures of the nanocomposites with increasing ND content and alkyl chain length. In parallel, 2.5 times higher Young's modulus and 4.5 times higher hardness were also measured for these composites by AFM, with a larger increase corresponding to longer grafted alkyl chains. These improvements were explained by a significant increase in crystallinity of the polymer — an example of how a proper ND surface functionalization can be used to design an interphase [42] (Fig. 4a, b). Electrospun PAN and PA11 nanofibers with high contents of ND show dramatically improved mechanical properties [43]. ND used in this study was purified by air oxidation [33], followed by HCl treatment to reduce the content of metals and hydrolyze anhydrides and lactones formed on ND surface during oxidation in air in order to maximize the number of carboxylic groups on the surface, which was confirmed by


165

Fig. 4. (a) Young's modulus and (b) hardness of ND–low density polyethylene composites prepared using NDs with chemically attached different alkyl chains [42]; (c) AFM image of NDpoly(iso-butyl methacrylate) brush prepared via atom transfer radical polymerization [54] (left side of the panel) and possible mechanisms of ND interactions with PVA chains [50] (right side of the panel); (d) hardness of PC and PMMA with same contents of different functionalized NDs [53].

FTIR. The electrospun fiber mats were then melted, yielding uniform ND–polymer films on the substrate. A 400% increase in Young's modulus and a 200% increase in hardness were measured by nanoindentation for these films (a few comprehensive reviews on nanoindentation in polymer nanocomposites were published recently [44,45]). Importantly, even at ND concentrations up to 20 wt.% the films remain optically transparent while strongly absorb UV radiation. This study has demonstrated that ND–polymer nanofibers with very high ND content (up to 80 wt.%) can be produced by electrospinning. At high ND concentrations the composite fibers behave as ceramic fibers, in particular, showing brittle failure typical for ceramic materials, with no necking or crazing, that are typical for pure PAN or CNT-containing PAN nanofibers. Also, pronounced agglomeration of ND observed at higher concentrations indicates that properties of these composites can be even higher when dispersion of ND is improved. These ND–polymer films are promising candidates for optically transparent UV absorbing scratch- and wearresistant coatings and paints, and show that ND can be used to suppress photoageing of polymers [43]. For coatings specifically, a strong adhesion between the substrate and the coating composite film is required. ND can improve the adhesion through creating nano-roughness at the coating-substrate interface, as well as through other mechanisms. A significantly improved adhesion of the Nylon-11 coating to steel due to the addition of 7 wt.% purified ND has been reported [46], further emphasizing the potential of ND in polymer-based coatings and paints. In many studies ND has been shown to be non-toxic and biocompatible [47,48], which renders it an excellent material for biomedical applications. In the context of this paper, it is important to mention that ND can be used to reinforce polymers for bone surgery and tissue engineering. Most polymers developed for such applications, in particular biodegradable polymers, suffer from poor mechanical properties, severely

limiting their potential for the replacement of metals in applications such as bone fracture fixation. Several studies have investigated the use of ND to reinforce polyvinyl alcohol (PVA), which is used in soft tissue replacements, artificial cartilage, skin, pancreas, as well as catheters and hemodialysis membranes, and a number of other biomedical applications. A 40% improved tensile modulus and a 70% larger fracture energy were measured due to the addition of 1 vol.% of ND-containing detonation soot purified by treatment with o-xylene to remove adsorbed organics. The observed improvements were explained by favorable interactions between oxygen containing groups (including COOH, C_O and alike), exposed on the surface of ND-containing soot and OH groups of PVA [49]. It is likely that removal of adsorbed hydrocarbons and other weakly bonded species from the surface of NDcontaining soot by boiling it in o-xylene facilitates weak interactions (mainly hydrogen bonding) between these groups and the functional groups of polymer matrix (Fig. 4c), which collectively result in a quite strong nanofiller-matrix interface. In a similar study, using “Gohsenol NH-18” PVA, significantly larger improvements in Young's modulus (up to 3 times), tensile strength (up to 30%), yield strength, and fracture energy were measured. The improvements were explained by an 8% increased crystallinity of PVA and hydrogen bonding between PVA hydroxyl groups and ND's oxygen containing surface functional groups and adsorbed water [50] (Fig. 4c). While the measurements above were performed on bulk PVA–ND samples, the Young's modulus of PVA–ND composites has also been studied locally by nanoindentation. The compression modulus and hardness improved by more than 90% and 78% respectively, due to the addition of only 0.6 wt.% acid-purified ND, which has an increased number of carboxylic groups. In parallel, DSC data showed a 14% increase in polymer crystallinity [51].

166


Acid-purified ND was shown to improve mechanical properties of another common biodegradable polymer, poly-(methyl methacrylate) (PMMA) which is used for example in dental implants. The introduction of the acid-purified ND in concentrations as small as 0.1 wt.% resulted in increased impact strength and fracture toughness of PMMA. At ND content of 0.8 wt.%, Young's modulus was 80% higher and the Tg increased by 20 °C. Similar to PVA, the improved properties are explained by hydrogen bonding between the polymer carbonyl groups and ND OH groups [52]. TEM micrographs of the ND–PMMA nanocomposites showed significant agglomeration of NDs, suggesting that further improvements are possible when well dispersed single ND particles become available. Polycarbonate (PC) and PMMA were reinforced with as-received and various functionalized NDs. Hardness and Young's modulus of these nanocomposites were increased by up to 100%, when the appropriate surface functionality was selected (Fig. 4d). The largest improvements in mechanical properties were obtained when using amide functionalized ND for PMMA and amino functionalized ND for PC [53] illustrating the importance of ND surface chemistry in designing the interface to different matrices. It is worth mentioning that in many cases it is difficult to disentangle the contributions of ND surface chemistry into improved ND–matrix interface and better ND dispersion in the matrix, which normally translates into a larger volume of the interphase. Judiciously chosen surface chemistry is often beneficial for both, resulting in a stronger interface, while at the same time maximizing the interphase. Poly-Lactic Acid (PLA) is a well-known biodegradable polymer used in tissue engineering. Storage modulus, tensile modulus, and the tensile strength of PLA were improved by the addition of 5 wt.% ND. The reasons for the improvements might be due to the increased crystallinity of the matrix (measured by DSC) as well as the attractive molecular interactions between ND and PLA molecules as suggested by the increased thermal stability of the composite measured in TGA [55]. An enantiomer of PLA, poly-L-(Lactic-Acid) (PLLA), which is used in manufacturing fixation devices for bone fracture surgery, was reinforced by octadecylamine (ODA)-modified ND. Long hydrophobic chains of ODA were grafted to ND surface [56] in order to increase the affinity of ND towards the polymer matrix and improve the dispersion of the nanofiller, while at the same time reducing undesirable nanoparticle–nanoparticle attractive interactions by replacing ND polar functional groups prone to hydrogen bonding and other types of stronger interactions with alkyl chains, which can only interact via weak van der Waals forces [57,58]. Successful covalent binding of ODA to ND was confirmed by detection of newly formed amide bonds in FTIR [56] and NMR [58] and indirectly by AFM [59]. In contrast to pristine ND, the resulting ND–ODA material is hydrophobic [56], resulting in good miscibility with hydrophobic polymers and solvents [58]. TEM images of PLLA–ND–ODA films show single ND– ODA particles and loosely bonded ND–ODA agglomerates dispersed in the matrix [58,60]. Good dispersion translates into high mechanical properties: PLLA–ND–ODA composites containing up to 10 wt.% ND–ODA showed an up to 2 times higher Young's modulus and up to 8 times higher hardness when compared to neat PLLA, measured using nanoindentation [58]. As expected, unmodified ND reinforced PLLA to a lesser extent. The bulk compression modulus of a PLLA– ND–ODA composite containing 10 wt.% ND–ODA was increased by 22%, and a 316% increase in fracture energy was observed at the same time. It is speculated that ND–ODA induced crazing, as evidenced by light and TEM microscopy, which is responsible for the large increase in strain to failure and fracture energy [60]. It is important to mention that ND can act as multifunctional nanofiller. Besides mechanical reinforcement, ND–ODA is blue fluorescent when illuminated with UV radiation [56,58], and promotes biomineralization [60], offering additional benefits in bone surgery and tissue engineering. In a similar biodegradable system, poly-L-(lactide-co-ε-caprolactone) or poly(LLA-co-CL), a copolymer, in which ε-caprolactone reduces glass

transition temperature and increases the elongation at break of L-lactic acid, the reinforcing effects of three different NDs, including milled acid-purified oxygen terminated ND, ND with grafted polylactide (ND– PLA), and benzoquinone functionalized ND (ND–BQ) were investigated [61]. As expected [58,60] acid purified hydrophilic oxygen-terminated ND did not disperse well in hydrophobic poly(LLA-co-CL) and resulted in degradation of mechanical properties at concentrations of 5 wt.% or higher. ND–BQ had essentially no influence on Young's modulus, stress, and ultimate strain. In contrast, ND–PLA showed a clear trend towards enhanced Young's modulus with increasing amount of ND–PLA. At 10 wt.% ND–PLA in poly(LLA-co-CL) the Young's modulus was ~6 times higher, while retaining enhanced elasticity at break provided by poly(LLA-co-CL) in comparison to PLLA. ND–BQ and ND–PLA demonstrated good dispersion in THF and in poly(LLA-co-CL) matrix, which in case of ND–PLA translated into improved mechanical properties, explained by favorable interactions and entanglement between ND grafted PLA chains and the matrix, an effect that is absent in case of ND–BQ. The change in mechanical properties upon addition of ND–PLA paralleled the increase of the composite glass transition temperature from 7 to 18 °C [61], indicative of the increased crystallinity of the matrix, i.e. an interphase effect of ND–PLA. Grafting polymer chains to nanofillers has been used to improve their miscibility with the matrix and maximize the nanoparticle–matrix interactions (stronger interface) due to entanglement and attractive interactions between the matrix macromolecules and the polymer chains attached to the nanoparticle. ND with covalently linked polymer chains (ND–polymer brushes) were synthesized using the atom transfer radical polymerization of poly (iso-butyl methacrylate) at the ND surface. The brushes were characterized by TGA, FTIR, NMR and AFM. Single ND–polymer brushes visualized using AFM (Fig. 4c) can have diameters of up to 300 nm, i.e. ~100 times larger than the average diameter of single ND particles [54]. Using a similar approach, polyimides were grown on ND particles. It was found that the XRD diffraction peak of polyimide at 4.9° disappeared in ND composites, providing evidence that long range interactions between polyimide chains are disrupted due to the introduction of ND. The produced nanocomposite had a 25% higher Young's modulus and a 15% higher hardness at a ND content of 5 wt.% [62]. Long chains on ND surface anchor the particle in the matrix, increase affinity between the nanoparticle and the matrix and, by mixing with its molecules, influence the structure of the host near the ND, resulting in changes of the interphase and the interface. 4. Nanodiamond—thermosetting polymer composites A common thermosetting polymer is epoxy, widely used as matrix material for carbon-fiber reinforced composites [63] in aerospace, ship building, and sports industries. Many nanofillers have been investigated to reinforce epoxy systems. Comparison between ND–epoxy and CNT– epoxy composites at similar nanofiller loadings in the low concentration range demonstrated a significant increase in glass transition temperature, 37 °C and 17 °C for NDs and CNTs, respectively. The fracture surfaces of both composites showed a better resistance to crack propagation compared to neat epoxy. Tensile properties of ND- and CNT–epoxy composites showed enhancement of 6.4% and 2.9%, respectively. The nanocomposites also showed an increase in microhardness by 41% for NDs and 12% for CNTs. The authors conclude that at similar nanofiller concentrations in the range 0.1–0.5 wt.% NDs showed superior enhancement compared to the CNTs in epoxy matrix composites [64]. To further improve the mechanical properties of the epoxy, ND– epoxy composites with ND content of up to 35 vol.% were produced. Hardness and Young's moduli of these composites measured by nanoindentation were higher by 300% and 700% respectively, reaching modulus values of up to 20 GPa and resulting in an increased scratch


a)

b)

c)

d)

167

Fig. 5. (a) TEM photograph of a ND–epoxy sample with 50 wt.% ND and a schematic illustrating TEM image formation for this composite in the inset, showing that the real 3D structure of the composite may be different from what can be inferred from the 2D TEM image; (b) interparticle distance (τ) as a function of the nanofiller volume fraction showing that τ in ND containing composites is shorter and direct contacts between the 5 nm NDs occur at lower concentrations compared to larger 10 nm diameter silica nanoparticles. Dashed vertical line marks the point at which ND–ND separation is 3 nm, i.e., less than the ND particle size. It is reasonable to assume that the majority of NDs form direct contacts starting at this concentration; (c) schematic of covalent incorporation of ND–NH2 into a molecular structure of epoxy polymer; (d) Young's modulus of neat epoxy (green squares), ND–NH2–epoxy composites in which the content of the molecular curing agent (PACM) was reduced to compensate for the additional amount of NH2 groups introduced with ND–NH2 (blue circles), and composites cured solely with ND–NH2 (orange triangles) plotted as a function of stoichiometry (r). The stoichiometric point (i.e. all amino groups react with all epoxide groups in the system) is defined as r = 1. The uncompensated effect of the NH2 groups introduced by 6 vol.% ND–NH2 on Young's modulus at different stoichiometries is shown by dotted gray line [68]. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

resistance [7]. The traditional view, where a polymer is considered as the matrix and nanoparticles are considered as the filler, need to be reversed in situations where the nanofiller is present in high concentrations. At such high ND loadings as indicated above, the composite material should be rather considered as a ND network infiltrated by a polymer acting as a binder (Fig. 5a). Direct contacts between the ND particles in such composites result in improved thermal conductivity [7]. It is important to emphasize that due to their small sizes, shorter interparticle distances and direct contacts between NDs are achieved at lower ND contents compared to larger nanofillers, e.g., 10 nm diameter silica (Fig. 5b). Tribological studies showed that an alumina counterbody was damaged by the ND–epoxy agglomerates contained within these composites, suggesting very high hardness of the agglomerates, which can eventually replace micron-sized diamond particles in drilling and cutting tools. Average macroscale friction coefficients of epoxy composites containing 7.5 vol.% ND were reduced 4 times approaching 0.1 [65]. While high loadings of ND result in remarkably high hardness and Young's moduli of the epoxy–ND composites, lower ND concentrations can be used as well to improve the mechanical properties. The bulk Young's modulus measured in tensile tests was 25% higher upon addition of 0.5 wt.% as-received ND, which also increased the decomposition temperature. However, due to poor ND dispersion, the storage modulus of the epoxy composite was significantly reduced [66], emphasizing the importance of a good dispersion to optimize the mechanical properties

of ND–polymer composites. A study on the mode I and II fracture toughness of ND–epoxy composites has shown that, besides an improved Young's modulus and hardness, the mode II fracture toughness of epoxy–ND composites with 0.1 wt.% ND is increased. This is because ND is thought to hinder shear deformation, improving fracture toughness [67]. For covalent binding to epoxy, aimed to form the most strong ND– polymer interface, ND terminated with reactive amino groups was synthesized by linking ethylenediamine to ND–COOH surface via amide bond (yielding ND–CONH(CH2)2NH2, hereafter named ND–NH2). The rationale behind using amino terminated ND is that similar to molecular curing agents, reaction of ND–NH2 with epoxy resin is expected to result in a covalently bonded network of ND and epoxy molecules (Fig. 5c) [36]. However, to get full advantage of covalent ND–polymer interface, it is critically important to have covalent bonds all the way from ND surface to the macromolecules of the matrix. Therefore, first covalent bonding between the diamine molecules and ND particles was confirmed by FTIR, TG and DSC. When ND–NH2 reacted with the epoxy resin, a strong covalent ND–epoxy interface was formed as evidenced by DSC, which was used to monitor the reaction. As a result, Young's modulus of a composite containing 3.5 vol.% ND–NH2 was improved by 60% [68]. Also, it was found that in order to manufacture ND–NH2–epoxy composites with uniformly dispersed NDs it is important to keep ND–NH2 dispersed in a compatible and inert solvent without drying [36,68].

168


Tetrahydrofuran (THF) was chosen for this purpose as it provides a good dispersing medium for ND–NH2, dissolves epoxy resin and, according to a previous report, does not react with components of the epoxy system [69]. Moreover, it can be evaporated from the system by moderate heating over few days. This study [68] also discusses the interference of amino groups of ND–NH2 with the amino groups of the curing agent (PACM) used for this epoxy system and emphasizes that it is necessary to adjust the amount of the curing agent to compensate for the amino groups introduced by ND–NH2 in order to maintain the right stoichiometry in the system and to maximize the Young's modulus of the composite. Young's moduli of neat epoxy samples containing an excess of amino groups, equivalent to the number of amino groups introduced by 6 vol.% ND–NH2 have been calculated and plotted in Fig. 5d (gray dotted line). As can be seen from these results, without the compensation the stoichiometry of the system (r) is changed, resulting in deviations (mainly reduction) of Young's modulus in the entire range of stoichiometry (an increase at low r is due to addition

of nanodiamond which itself has high Young's modulus offsetting the negative effect of skewed stoichiometry). To avoid the undesired reduction of Young's modulus in the epoxy–ND–NH2 composites it is critical, therefore, to compensate for the additional NH2 groups. With proper compensation, the Young's moduli of the composites are consistently higher compared to neat epoxy as well as to non-compensated epoxy–ND–NH2 composites (Fig. 5d, blue circles) [68]. A remarkable result of this work is the first experimental demonstration of curing of epoxy resin solely by ND–NH2 (no molecular curing agent added), which is due to the large number of reactive amino groups introduced on ND surface [36]. These ND–NH2 cured composites showed extremely high Young's modulus of up to 20 GPa (Fig. 5d, orange triangles). The tribological properties of these covalently bonded epoxy–ND– NH2 composites were studied as well. A composite with a high content of ND–NH2 (33 vol.%) has a friction coefficient of 0.06 ± 0.02, a 54% reduction when compared to composites produced with up to 50 vol.% of

Table 1 Representative examples of properties improvements in nanodiamond–polymer composites. Polymer

ND content

Property

Quantitative increase

Ref.

Polydimethylsiloxane Polyurethane-2-hydroxyethylmethacrylate

0.1 wt.% 0.25 wt.% 11 wt.%

Polyamide-11

20 wt.%

Polyamide-11 (Nylon-11) Polyvinyl alcohol

7 wt.% 1 vol.% (detonation soot)

Polyvinyl alcohol (Gohsenol NH-18)

5 wt.%

Polyvinyl alcohol

0.6 wt.%

Poly-(methyl methacrylate)

0.1 wt.%

N/A N/A N/A 2.5 times 4.5 times 400% 200% N/A N/A 40% 70% 2.9 times 1.3 times 0.2 times 0.2 times 1.3 times 8% 90% 78% 14% N/A N/A 80% 20 °C 1.2 times 2.1 times 1.3 times 1.6 times 19% 75% (@130 °C) 10.1 °C higher 2 times 8 times 316% 6 times 9 °C higher 15 times 25% 15% 6.4% 41% 37 °C higher 700% 300% N/A 0.25 times N/A N/A 25% N/A 700% 60% 28 °C lower

[40] [41]

Polyethylene

Elastic strain Young's modulus Glass transition temperature Young's modulus Hardness Young's modulus Hardness UV absorbance Adhesion Tensile modulus Fracture energy Young's modulus Tensile strength Strain at break Toughness Thermal conductivity Crystallinity Young's modulus Hardness Crystallinity Impact strength Fracture toughness Young's modulus Glass transition temperature Hardness Young's modulus Hardness Young's modulus Tensile modulus Storage modulus Onset of thermal decomposition Young's modulus Hardness Fracture energy Young's modulus Glass transition temperature Young's modulus Young's modulus Hardness Tensile strength Hardness Glass transition temperature Young's modulus Hardness Scratch resistance Friction coefficient Wear resistance Abrasion Young's modulus Fracture toughness Young's modulus Young's modulus Glass transition temperature

0.8 wt.% Poly-(methyl methacrylate)

5 wt.%

Polycarbonate

5 wt.%

Poly-(lactic acid)

Poly-L-(lactic acid)

5 wt.% 3 wt.% 1 wt.% 10 wt.%

Poly-L-(lactide-co-ε-caprolactone)

10 wt.%

Polyimide

50 wt.% 5 wt.%

Epoxy (DER 332 + TETA)

0.1 wt.%

Epoxy (Epon828 + PACM20)

35 vol.%

Epoxy (Epon828 + PACM)

7.5 vol.%

Epoxy (DGEBA + Jeffamine D2000)

0.5 wt.%

Epoxy (Epon828 + PACM)

30 vol.% 3.5 vol.% 20 vol.%

[42] [43]

[46] [49] [50]

[51]

[52]

[53] [53] [55]

[60]

[61]

[62] [64]

[7]

[65]

[67] [68]


as-received ND [65]. This friction coefficient value is close to carbidederived or diamond-like carbon films, demonstrating the potential of the composite for tribological applications. Small additions of ND (less than 1 wt.%) to cross-linked high density polyethylene (PEX) resulted in composites with notably enhanced strength, toughness, elastic modulus, as well as thermal conductivity, specific heat capacity, and thermal stability. The improvements were explained by the “enhanced adhesion with the matrix” (an interface effect), resulting in a better load transfer between the filler and the matrix. This enhanced adhesion is mainly mediated by dispersive interactions (not polar interactions), as follows from the analysis of components of ND–polymer free interaction energy obtained from contact angle measurements. Well dispersed NDs substantially increased degree of polymer crystallinity while at the same time promoting formation of larger polymer crystals — both positively influence the mechanical properties of composites. For higher ND concentrations, when ND agglomeration becomes pronounced, a degradation of these properties was observed. Thermal conductivity of the ND–PEX composites was found to be significantly increased due to the ND related heat capacity increase. The authors note that the mechanism of ND induced thermal conductivity enhancement in this case is substantially different from the CNTs induced enhancement, where the main contribution into thermal conductivity is due to enhanced thermal diffusivity and not increased heat capacity. Additionally, the ND–PEX composite with the lowest filler content showed a better thermal stability, which was explained due to hindered polymer chain mobility under confinement by the well dispersed ND particles, resulting in lower chemical reactivity of the polymer and therefore, enhanced thermal stability of ND–PEX [70]. 5. Nanodiamond — elastomer composites Elastomers are cross-linked, amorphous polymers above their Tg, possessing high elastic deformation and resuming their original shape after deforming force is removed. They are heavily used in modern life, especially in the automobile industry (tires, braking systems, chassis, interior parts, etc.). In many cases, improvements in mechanical properties of elastomers are required to extend their lifetime and further broaden their applications. Improved mechanical properties, such as cohesive strength, rupture, and wear resistance were reported for a variety of ND filled elastomers including fluorinated elastomers and rubbers [71]. The effect of surface functionalization of ND on the mechanical properties of polysiloxane films was studied as well, where mechanical properties such as engineering stress and tensile strength were increased due to the addition of silanized ND. The improvement in mechanical properties was attributed to a reduction in ND agglomerate size during a silylation reaction, which in addition removes adsorbed water from the ND surface, rendering the material hydrophobic [23]. A recent review on ND- and ND-containing soot reinforced rubbers provides more information on the subject [72]. Further studies on the reinforcing mechanisms of ND in elastomeric polymer matrices are required. The covalent incorporation of ND into elastomers is a promising yet underexplored approach that could yield materials with increased wear and rupture resistance, and at the same time showing larger strain-to-failure values. 6. Conclusions and outlook Nanodiamond is advantageous and in many respects a unique material for polymer nanocomposites. Main advantages of nanodiamond, compared to other nanofillers, are: 1) superior mechanical, electric, optical, and thermal properties of the diamond core; 2) small and uniform size and nearly spherical shape of its particles; 3) large and accessible external surface, maximizing interactions with the matrix, i.e. the interphase formation; 4) rich and tailorable surface chemistry, providing

169

great flexibility for rational design of the ND–matrix interface; and 5) additional properties, such as biocompatibility, lack of toxicity, potential to create brightly fluorescent nanoparticles, etc., that impart multifunctionality. All these properties are important for polymer nanocomposites. For example, the spherical shape of ND maximizes the interphase volume per nanoparticle volume, contributing to improved mechanical properties of the composite; small particle size minimizes the interparticle distance, resulting in close contacts between the nanoparticles at lower concentrations, thus yielding a higher thermal conductivity and better mechanical strength, etc. Simply listing these properties already shows the tremendous potential of ND in composites. Table 1 summarizes some representative examples of properties improvement in nanodiamond– polymer composites available in literature. The reader is referred to the original papers for many additional details as to the type of ND, surface modification, ND concentration range, ND dispersion and composite fabrication technique, etc., which in each particular case are different. However, as with any nanomaterial, harnessing its full potential takes a good deal of understanding and control. With regard to ND, this means the need of purification of commercial powders and a careful control over their surface chemistry in order to achieve good dispersion in polymers and design an optimal interface between the components of the composite. One recent development in ND dispersion is mixing NDs with polymer nanoparticles [73]. The importance of both good dispersion and interfacing ND to the matrix has only recently been appreciated as a critical factor in developing ND–polymer composites and should be further emphasized. Many novel multifunctional ND–polymer composites are still to be developed. A unique combination of properties offered by ND should be more widely explored in composites for electronics, packaging, membranes, biomedical applications, etc. Those may include, for example, tissue engineering scaffolds and biomedical polymer devices incorporating NDs to impart mechanical strength, bioimaging modality (due to luminescent NDs), drug delivery modality (by adsorption/desorption or chemical linking/release of the drugs to/from NDs), enhanced biomineralization, etc. Finally, synergism between NDs and other fillers should be explored. There is a handful of results in this area [74–77] and it should be studied further. For example, incorporation of NDs in traditional epoxy–carbon fiber composites may reinforce the matrix between the fibers and thus increase the mechanical properties and failure tolerance of the composite [75,76]. At the same time, the use of ND for this purpose is clearly more advantageous than the use of CNTs or graphene nanoplatelets, because ND can be introduced in higher concentrations without increasing the viscosity of the resin beyond processability limits, and small spherical ND particles won't be filtered by carbon fibers, allowing uniform infiltration and dispersion. Prime novelty statement Among different nanoparticles, nanodiamond particles are especially attractive for polymer nanocomposites. The interest to this application of nanodiamond is growing, as evidenced by a large number of recent publications on the subject. This thematic article is not a comprehensive review on the subject. It is rather an attempt to highlight the main advantages of nanodiamond in polymer nanocomposites for a newcomer to this area and summarize some of the most important, in authors' opinion, results and existing knowledge on nanodiamond– polymer composites. Acknowledgment This material is based upon work supported by the British Council and the UK Department for Business, Innovation and Skills through the Global Innovation Initiative.

170


References [1] T. Twardowski, Introduction to Nanocomposite Materials: Properties, Processing, Characterization, DESTech Publications, 2007. [2] E.P. Giannelis, Polymer layered silicate nanocomposites, Adv. Mater. 8 (1) (1996) 29–35. [3] Y. Fukushima, S. Inagaki, Synthesis of an intercalated compound of montmorillonite and 6-polyamide, J. Incl. Phenom. 5 (4) (1987) 473–482. [4] P.C. LeBaron, Z. Wang, T.J. Pinnavaia, Polymer-layered silicate nanocomposites: an overview, Appl. Clay Sci. 15 (1–2) (1999) 11–29. [5] K.I. Winey, R.A. Vaia, Polymer nanocomposites, MRS Bull. 32 (4) (2007) 314–319. [6] W.J. Work, K. Horie, M. Hess, R.F.T. Stepto, Definition of terms related to polymer blends, composites, and multiphase polymeric materials (IUPAC recommendations 2004), Pure Appl. Chem. 76 (11) (2004) 1985–2007. [7] I. Neitzel, V. Mochalin, I. Knoke, G.R. Palmese, Y. Gogotsi, Mechanical properties of epoxy composites with high contents of nanodiamond, Compos. Sci. Technol. 71 (5) (2011) 710–716. [8] S. Srivastava, J.L. Schaefer, Z. Yang, Z. Tu, L.A. Archer, 25th anniversary article: polymer–particle composites: phase stability and applications in electrochemical energy storage, Adv. Mater. 26 (2) (2014) 201–234. [9] Y. Gogotsi, V. Presser, Carbon Nanomaterials, 2nd ed. CRC Press, 2013. 514. [10] J. Markarian, Automotive and packaging offer growth opportunities for nanocomposites, Plast. Addit. Compd. 7 (6) (2005) 18–21. [11] M.E. Mackay, A. Tuteja, P.M. Duxbury, C.J. Hawker, B. Van Horn, Z.B. Guan, G.H. Chen, R.S. Krishnan, General strategies for nanoparticle dispersion, Science 311 (5768) (2006) 1740–1743. [12] M. Naguib, V.N. Mochalin, M.W. Barsoum, Y. Gogotsi, 25th anniversary article: MXenes: a new family of two-dimensional materials, Adv. Mater. 26 (7) (2014) 992–1005. [13] V. Mittal, Modeling and Prediction of Polymer Nanocomposite Properties, WileyVCH, 2013. 320. [14] M. Esmaeili, R. Ansari, V. Mottaghitalab, B. Xi, G.M. Spinks, G.G. Wallace, A.I. Minett, CNT/polymer nanocomposites, in: L.-S. Liu, A. Ballada (Eds.),Engineering of Polymers and Chemical Complexity, vol. I, Apple Academic Press 2014, pp. 149–255. [15] B. Chen, J.R. Evans, H.C. Greenwell, P. Boulet, P.V. Coveney, A.A. Bowden, A. Whiting, A critical appraisal of polymer–clay nanocomposites, Chem. Soc. Rev. 37 (3) (2008) 568–594. [16] X.L. Ji, J.K. Jing, W. Jiang, B.Z. Jiang, Tensile modulus of polymer nanocomposites, Polym. Eng. Sci. 42 (5) (2002) 983–993. [17] L. Mishnaevsky Jr., G. Dai, Hybrid and hierarchical nanoreinforced polymer composites: computational modelling of structure–properties relationships, Compos. Struct. 117 (2014) 156–168. [18] S. Saber-Samandari, A. Afaghi-Khatibi, Evaluation of elastic modulus of polymer matrix nanocomposites, Polym. Compos. 28 (3) (2007) 405–411. [19] E. Sharifzadeh, I. Ghasemi, M. Karrabi, H. Azizi, A new approach in modeling of mechanical properties of nanocomposites: effect of interface region and random orientation, Iran. Polym. J. 23 (11) (2014) 835–845. [20] I. Neitzel, V. Mochalin, Y. Gogotsi, Advances in surface chemistry of nanodiamond and nanodiamond–polymer composites, in: O.A. Shenderova, D.M. Gruen (Eds.), Ultrananocrystalline Diamond: Synthesis, Properties, and Applications, 2nd ed.Elsevier, Amsterdam 2012, pp. 421–456. [21] V.N. Mochalin, O. Shenderova, D. Ho, Y. Gogotsi, The properties and applications of nanodiamonds, Nat. Nanotechnol. 7 (1) (2012) 11–23. [22] V.Y. Dolmatov, Polymer–diamond composites based on detonation nanodiamonds. Part 1, J. Superhard Mater. 29 (1) (2007) 1–11. [23] V.Y. Dolmatov, Polymer–diamond composites based on detonation nanodiamonds. Part 2, J. Superhard Mater. 29 (2) (2007) 65–75. [24] V.Y. Dolmatov, Polymer–diamond composites based on detonation nanodiamonds. Part 3, J. Superhard Mater. 29 (4) (2007) 199–205. [25] V.A. Gerasin, E.M. Antipov, V.V. Karbushev, V.G. Kulichikhin, G.P. Karpacheva, R.V. Talroze, Y.V. Kudryavtsev, New approaches to the development of hybrid nanocomposites: from structural materials to high-tech applications, Russ. Chem. Rev. 82 (4) (2013) 303. [26] E. Alishahi, S. Shadlou, S. Doagou-R, M.R. Ayatollahi, Effects of carbon nanoreinforcements of different shapes on the mechanical properties of epoxybased nanocomposites, Macromol. Mater. Eng. 298 (6) (2013) 670–678. [27] A. Allaoui, N.E. Bounia, Rheological and electrical transitions in carbon nanotube/ epoxy suspensions, Curr. Nanosci. 6 (2) (2010) 153–162. [28] A. Krueger, D. Lang, Functionality is key: recent progress in the surface modification of nanodiamond, Adv. Funct. Mater. 22 (5) (2012) 890–906. [29] A. Krueger, The chemistry of nanodiamond, in: O.A. Williams (Ed.), Nanodiamond, Royal Society of Chemistry 2014, pp. 49–88. [30] L. Lai, A.S. Barnard, Stability of nanodiamond surfaces exposed to N, NH, and NH2, J. Phys. Chem. C 115 (14) (2011) 6218–6228. [31] L. Lai, A.S. Barnard, Modeling the thermostability of surface functionalisation by oxygen, hydroxyl, and water on nanodiamonds, Nanoscale 3 (6) (2011) 2566–2575. [32] G.C.C. Costa, O. Shenderova, V. Mochalin, Y. Gogotsi, A. Navrotsky, Thermochemistry of nanodiamond terminated by oxygen containing functional groups, Carbon 80 (2014) 544–550. [33] S. Osswald, G. Yushin, V. Mochalin, S.O. Kucheyev, Y. Gogotsi, Control of sp2/sp3 carbon ratio and surface chemistry of nanodiamond powders by selective oxidation in air, J. Am. Chem. Soc. 128 (35) (2006) 11635–11642. [34] I. Petrov, O. Shenderova, V. Grishko, V. Grichko, T. Tyler, G. Cunningham, G. McGuire, Detonation nanodiamonds simultaneously purified and modified by gas treatment, Diamond Relat. Mater. 16 (12) (2007) 2098–2103.

[35] B.J.M. Etzold, I. Neitzel, M. Kett, F. Strobl, V.N. Mochalin, Y. Gogotsi, Layer-by-layer oxidation for decreasing the size of detonation nanodiamond, Chem. Mater. 26 (11) (2014) 3479–3484. [36] V.N. Mochalin, I. Neitzel, B.J.M. Etzold, A. Peterson, G. Palmese, Y. Gogotsi, Covalent incorporation of aminated nanodiamond into an epoxy polymer network, ACS Nano 5 (9) (2011) 7494–7502. [37] A.N. Zhukov, F.R. Gareeva, A.E. Aleksenskii, A.Y. Vul, Surface charge of detonation nanodiamond particles in aqueous solutions of simple 1:1 electrolytes, Colloid J. 72 (5) (2010) 640–646. [38] L. Schmidlin, V. Pichot, M. Comet, S. Josset, P. Rabu, D. Spitzer, Identification, quantification and modification of detonation nanodiamond functional groups, Diamond Relat. Mater. 22 (2012) 113–117. [39] W.S. Yeap, K.P. Loh, Functionalization of nanodiamond for specific biorecognition, in: D. Ho (Ed.), Nanodiamonds: Applications in Biology and Nanoscale Medicine, Springer Science + Business Media, LLC 2010, pp. 117–125. [40] A.S. Gavrilov, A.P. Voznyakovskii, Rheological characteristics and relaxation properties of polymer–nanodiamond composites, Russ. J. Appl. Chem. 82 (6) (2009) 1041–1045. [41] V. Bershtein, L. Karabanova, T. Sukhanova, P. Yakushev, L. Egorova, E. Lutsyk, A. Svyatyna, M. Vylegzhanina, Peculiar dynamics and elastic properties of hybrid semi-interpenetrating polymer network-3-D diamond nanocomposites, Polymer 49 (4) (2008) 836–842. [42] A.-Y. Jee, M. Lee, Thermal and mechanical properties of alkyl-functionalized nanodiamond composites, Curr. Appl. Phys. 11 (5) (2011) 1183–1187. [43] K.D. Behler, A. Stravato, V. Mochalin, G. Korneva, G. Yushin, Y. Gogotsi, Nanodiamond–polymer composite fibers and coatings, ACS Nano 3 (2) (2009) 363–369. [44] R.F. Gibson, A review of recent research on nanoindentation of polymer composites and their constituents, Compos. Sci. Technol. 105 (2014) 51–65. [45] A.M. Díez-Pascual, M.A. Gómez-Fatou, F. Ania, A. Flores, Nanoindentation in polymer nanocomposites, Prog. Mater. Sci. 67 (2015) 1–94. [46] A. Stravato, R. Knight, V. Mochalin, S.C. Picardi, HVOF-sprayed Nylon-11 + nanodiamond composite coatings: production & characterization, J. Therm. Spray Technol. 17 (5) (2008) 812–817. [47] V. Paget, J.A. Sergent, R. Grall, S. Altmeyer-Morel, H.A. Girard, T. Petit, C. Gesset, M. Mermoux, P. Bergonzo, J.C. Arnault, S. Chevillard, Carboxylated nanodiamonds are neither cytotoxic nor genotoxic on liver, kidney, intestine and lung human cell lines, Nanotoxicology 8 (Suppl. 1) (2014) 46–56. [48] A.M. Schrand, J. Johnson, L. Dai, S.M. Hussain, J.J. Schlager, L. Zhu, Y. Hong, E. Osawa, Cytotoxicity and genotoxicity of carbon nanomaterials, in: T.J. Webster (Ed.), Safety of Nanoparticles. From Manufacturing to Medical Applications, Springer 2009, pp. 159–187. [49] T.S. Kurkin, A.N. Ozerin, A.S. Kechek'yan, O.T. Gritsenko, L.A. Ozerina, G.G. Alkhanishvili, V.G. Sushchev, V.Y. Dolmatov, The structure and properties of polymer composite fibers based on poly(vinyl alcohol) and nanodiamond of detonation synthesis, Nanotechnol. Russ. 5 (5–6) (2010) 340–351. [50] S. Morimune, M. Kotera, T. Nishino, K. Goto, K. Hata, Poly(vinyl alcohol) nanocomposites with nanodiamond, Macromolecules 44 (11) (2011) 4415–4421. [51] U. Maitra, K.E. Prasad, U. Ramamurty, C.N.R. Rao, Mechanical properties of nanodiamond-reinforced polymer–matrix composites, Solid State Commun. 149 (39–40) (2009) 1693–1697. [52] P. Protopapa, E. Kontonasaki, D. Bikiaris, K.M. Paraskevopoulos, P. Koidis, Reinforcement of a PMMA resin for fixed interim prostheses with nanodiamonds, Dent. Mater. J. 30 (2) (2011) 222–231. [53] A.Y. Jee, M. Lee, Mechanical properties of polycarbonate and poly(methyl methacrylate) films reinforced with surface-functionalized nanodiamonds, J. Nanosci. Nanotechnol. 11 (1) (2011) 533–536. [54] L. Li, J.L. Davidson, C.M. Lukehart, Surface functionalization of nanodiamond particles via atom transfer radical polymerization, Carbon 44 (11) (2006) 2308–2315. [55] Y.-Q. Zhao, K.-T. Lau, J.-k. Kim, C.-L. Xu, D.-D. Zhao, H.-L. Li, Nanodiamond/poly (lactic acid) nanocomposites: effect of nanodiamond on structure and properties of poly (lactic acid), Compos. Part B Eng. 41 (8) (2010) 646–653. [56] V.N. Mochalin, Y. Gogotsi, Wet chemistry route to hydrophobic blue fluorescent nanodiamond, J. Am. Chem. Soc. 131 (13) (2009) 4594–4595. [57] J.B. Hooper, K.S. Schweizer, Theory of phase separation in polymer nanocomposites, Macromolecules 39 (15) (2006) 5133–5142. [58] Q. Zhang, V.N. Mochalin, I. Neitzel, I.Y. Knoke, J. Han, C.A. Klug, J.G. Zhou, P.I. Lelkes, Y. Gogotsi, Fluorescent PLLA–nanodiamond composites for bone tissue engineering, Biomaterials 32 (1) (2011) 87–94. [59] S. Guo, S.D. Solares, V. Mochalin, I. Neitzel, Y. Gogotsi, S.V. Kalinin, S. Jesse, Multifrequency imaging in the intermittent contact mode of atomic force microscopy: beyond phase imaging, Small 8 (8) (2012) 1264–1269. [60] Q. Zhang, V.N. Mochalin, I. Neitzel, K. Hazeli, J. Niu, A. Kontsos, J.G. Zhou, P.I. Lelkes, Y. Gogotsi, Mechanical properties and biomineralization of multifunctional nanodiamond–PLLA composites for bone tissue engineering, Biomaterials 33 (20) (2012) 5067–5075. [61] Y. Sun, A. Finne-Wistrand, T. Waag, Z. Xing, M. Yassin, A. Yamamoto, K. Mustafa, D. Steinmüller-Nethl, A. Krueger, A.-C. Albertsson, Reinforced degradable biocomposite by homogenously distributed functionalized nanodiamond particles, Macromol. Mater. Eng. 300 (4) (2015) 436–447. [62] Q. Zhang, K. Naito, Y. Tanaka, Y. Kagawa, Grafting polyimides from nanodiamonds, Macromolecules 41 (3) (2008) 536–538. [63] G. Lubineau, A. Rahaman, A review of strategies for improving the degradation properties of laminated continuous-fiber/epoxy composites with carbon-based nanoreinforcements, Carbon 50 (7) (2012) 2377–2395.

V.N. Mochalin, Y. Gogotsi / Diamond & Related Materials 58 (2015) 161–171 [64] S. Roy, K. Mitra, C. Desai, R. Petrova, S. Mitra, Detonation nanodiamonds and carbon nanotubes as reinforcements in epoxy composites — a comparative study, J. Nanotechnol. Eng. Med. 4 (1) (2013) 011008. [65] I. Neitzel, V. Mochalin, J.A. Bares, R.W. Carpick, A. Erdemir, Y. Gogotsi, Tribological properties of nanodiamond–epoxy composites, Tribol. Lett. 47 (2) (2012) 195–202. [66] Z. Spitalsky, A. Kromka, L. Matejka, P. Cernoch, J. Kovarova, J. Kotek, M. Slouf, Effect of nanodiamond perticles on properties of epoxy composites, Adv. Compos. Lett. 17 (1) (2008) 29–34. [67] M. Ayatollahi, E. Alishahi, S. Shadlou, Mechanical behavior of nanodiamond/epoxy nanocomposites, Int. J. Fract. 170 (1) (2011) 95–100. [68] I. Neitzel, V.N. Mochalin, J. Niu, J. Cuadra, A. Kontsos, G.R. Palmese, Y. Gogotsi, Maximizing Young's modulus of aminated nanodiamond–epoxy composites measured in compression, Polymer 53 (25) (2012) 5965–5971. [69] V.I. Raman, G.R. Palmese, Influence of tetrahydrofuran on epoxy–amine polymerization, Macromolecules 38 (16) (2005) 6923–6930. [70] E. Roumeli, E. Pavlidou, A. Avgeropoulos, G. Vourlias, D.N. Bikiaris, K. Chrissafis, Factors controlling the enhanced mechanical and thermal properties of nanodiamondreinforced cross-linked high density polyethylene, J. Phys. Chem. B 118 (38) (2014) 11341–11352.

171

[71] V.Y. Dolmatov, Composite materials based on elastomer and polymer matrices filled with nanodiamonds of detonation synthesis, Nanotechnol. Russ. 4 (9–10) (2009) 556–575. [72] A. Shakun, J. Vuorinen, M. Hoikkanen, M. Poikelispää, A. Das, Hard nanodiamonds in soft rubbers: past, present and future — a review, Compos. A: Appl. Sci. Manuf. 64 (2014) 49–69. [73] N.F. Attia, J.P. Rao, K.E. Geckeler, Nanodiamond–polymer nanoparticle composites and their thin films, J. Nanopart. Res. 16 (4) (2014) 1–12. [74] K.E. Prasad, B. Das, U. Maitra, U. Ramamurty, C.N.R. Rao, Extraordinary synergy in the mechanical properties of polymer matrix composites reinforced with 2 nanocarbons, Proc. Natl. Acad. Sci. 106 (32) (2009) 13186–13189. [75] A. Kausar, Design and study of epoxy composites based on polycaprolactone and nanodiamond functionalized carbon fibers, Am. J. Polym. Sci. Eng. 3 (1) (2015) 50–60. [76] F. Zhao, R. Liu, X. Yu, H. Ding, X. Qu, Q. Zhang, Carbon fiber grafted with nanodiamond: preparation and characterization, J. Nanosci. Nanotechnol. 15 (8) (2015) 5807–5815. [77] E. Roumeli, A. Avgeropoulos, E. Pavlidou, G. Vourlias, T. Kyratsi, D. Bikiaris, K. Chrissafis, Understanding the mechanical and thermal property reinforcement of crosslinked polyethylene by nanodiamonds and carbon nanotubes, RSC Adv. 4 (85) (2014) 45522–45534.