f san

epence

Received 5 November 2014

Available online 11 March 2015

ech

Owe to their lubricious nature, these carbonous materials have attracted researchers to synthesize

ered

signicantly different chemical, physical and mechanical proper-

In MMCs, different metals or alloys of aluminum, magnesium,copper, or nickel are generally used as matrix materials. Amongthese matrix materials, aluminum alloys are the most widely usedmaterials, both in research and industrial viewpoints [3,4]. This is

in aluminum matrix composites. Al2O3, SiC, B4C is among the veryent in aluminum

article reinforcednto: (1) Solid stateVapor deposition.bricate aluminumreinforcements instate) and castingt of work has been

devoted to producing aluminum matrix composites by usingpowder metallurgy and to study their mechanical properties[6e10]. The basic route in the powder metallurgy (P/M) technique,in which all materials remain in the solid state, is mixing thepowders, compacting, and sintering of the compacted part toachieve the least possible porosities and the highest possible den-sity. In liquid state methods, the matrix is in the state of a liquid.There are different methods of liquid state processing to producemetal matrix composites. These methods include stir casting [11],

* Corresponding author.E-mail address: eomrani@uwm.edu (E. Omrani).

1

Contents lists availab

Composite

ev

Composites Part B 77 (2015) 402e420These authors contributed equally to this work.ties. The metal matrix composites (MMCs) reinforced by ceramicparticles or bers generally have superior properties, such as highspecic strength and modulus compared to unreinforced alloys. Ingeneral, composites have the combination of properties of con-stituents, such as they inherit ductility and toughness of the matrixand high modulus and strength of the reinforcements [1]. In thisregard, the applications of MMCs have been extended to use asstructural materials in aerospace, automotive, marine and defenseindustries [2].

common particles which are used as reinforcemmatrix composites.

There are different methods to synthesize pMMCs. Generally, these methods are classied iprocessing; (2) Liquid state processing; and (3)Two common methods which are applied to famatrix composites reinforced by particulatelarge scales are powdermetallurgymethod (solid(liquid state). In literature, a considerable amounmaterials which contain two or more distinct constituents with of composites, different reinforcements are used as reinforcementKeywords:A. Metal-matrix composites (MMCs)B. WearB. Mechanical propertiesE. Powder processingE. Self-lubricating

1. Introduction

Composite materials are enginehttp://dx.doi.org/10.1016/j.compositesb.2015.03.0141359-8368/ 2015 Elsevier Ltd. All rights reserved.lightweight self-lubricating metal matrix nanocomposites with superior mechanical and tribologicalproperties for several applications in automotive and aerospace industries. This review focuses on therecent development in mechanical and tribological behavior of self-lubricating metallic nanocompositesreinforced by carbonous nanomaterials such as CNT and graphene. The review includes development ofself-lubricating nanocomposites, related issues in their processing, their characterization, and investi-gation of their tribological behavior. The results reveal that adding CNT and graphene to metals decreasesboth coefcient of friction and wear rate as well as increases the tensile strength. The mechanismsinvolved for the improved mechanical and tribological behavior is discussed.

2015 Elsevier Ltd. All rights reserved.

or naturally occurring

due to their outstanding properties, such as lightweight, highstrength, high specic modulus, low thermal expansion coefcient,and goodwear resistance [5]. Depending on nal desired properties20 February 2015Accepted 3 March 2015Received in revised formnanocomposites for structural engineering and functional devices. Carbonous materials, such as graphite,carbon nanotubes (CNT's), and graphene possess unique electrical, mechanical, and thermal properties.Mechanical and tribological properties onanocomposites reinforced by carbon nand graphene e A review

Afsaneh Dorri Moghadam 1, Emad Omrani*, 1, PradeDepartment of Materials Science and Engineering, College of Engineering & Applied Sci

a r t i c l e i n f o

Article history:

a b s t r a c t

Rapid innovation in nanot

journal homepage: www.elself-lubricating metal matrixotubes (CNTs)

L. Menezes, Pradeep K. Rohatgi, University of Wisconsin, Milwaukee, WI 53211, USA

nology in recent years enabled development of advanced metal matrix

le at ScienceDirect

s Part B

ier .com/locate/compositesb

pospressure inltration [12], pressureless inltration [13] and squeezecasting [14]. nl et al. [15] investigated and compared the me-chanical properties of aluminum matrix composites reinforced byAl2O3 and SiC which are produced both by powder metallurgymethod and casting. The results show that the mechanical andtribological properties of composite which are produced by castingare higher than P/M method. In addition, casting is also a moreindustrial compatible technique to produce composites in largescales.

Among various reinforcements, recent emerging material, car-bonous materials, is found to have many favorable attributes suchas high thermal conductivity, low coefcient of thermal expansion,high damping capacity and good self-lubricant property [4]. Theconsiderable amount of research has been made to study the in-uences of embedding graphite particles into the metal matrix onthe tribological properties of aluminum alloys [16e18]. Metal ma-trix composites embedded by graphite or carbon bers have self-lubricating behavior since graphite act as a solid lubricant [19]. Inthis regard, solid lubricant as reinforcement tends to decrease thefriction coefcient of MMCs and improve tribological properties ofself-lubricating composite compared to composites reinforced byceramic particles. The graphite size, which is commonly used inMMCs fabrication and obtaining desired mechanical and self-lubricating properties are in the micron range [1,4,20e25].

Generally, size of reinforcement inuences the mechanicalproperties such as strength, ductility and fracture of self-lubricating MMCs. By increasing the reinforcement size, tensilestrength and ductility decrease simultaneously. MMCs reinforcedby larger particles are susceptible to formation of defects such ascracking during mechanical testing which results in a prematurefailure of the composites. Therefore, it is expected to have superiorproperties when the reinforcement size is in the nano range. It isexpected that by reducing the particle size in MMC's to the rangeof nanosize, some of the limitations such as poor ductility andelongation, poor machinability, and reduced fracture toughness ofMMCs can be solved. The reason for signicantly improved me-chanical properties is due to the combined effect of Orowanstrengthening and grain rening mechanisms and high tempera-ture creep resistance properties could make metal matrix nano-composites (MMNCs) very attractive, especially when lightweightmetals such as Al or Mg are used as the matrix material. Thestrengthening mechanisms which are involved in enhancing me-chanical and tribological properties of nano particle reinforcedmetal matrix composites have been discussed carefully elsewhere[4,26]. To briey discuss, there are a few common differentmechanisms that have been suggested to enable increasedstrength in metal composites; (1) Orowan strengthening fromdislocation bowing by reinforced particles, (2) HallePetchstrengthening from grain renement, (3) Forest strengtheningresulting from the Coefcient of Thermal Expansion (CTE)mismatch between matrix and particles, and (4) Taylor strength-ening by modulus mismatch between matrix and particles.MMNCs will benet from the Orowan mechanism only if adispersed second phase of nanometer size could be attained.HallePetch strengthening will in general one of the mostly activemechanisms which improve strength of MMNCs by incorporationof nanoparticles. Usually, the addition of nanoparticles renesgrains or connes grain growth. CTE and modulus mismatch areconsidered to be negligible when compared with Orowan andgrain renement in many recent observations [4,23,26,27].

Theoretically, it has been shown that the nanosized particlescan result in dramatic improvements in light of the very lowproportion to the quantities they are added. However, thestrength achieved in most solidication processed and powder

A. Dorri Moghadam et al. / Commetallurgy nanocomposites are still below the conventionalmonolithic alloys, which can be caused by possible nanoparticlesagglomeration, in capability of solidication front in capturingnanoparticles (pushing off by growing dendrites), and porosity.Therefore, before these materials can be introduced as anexcellent candidate for conventional metals and alloys, thesechallenges in synthesis and processing need to overcome. Gravitycasting techniques are prone to porosity, which could be onecause for low yield strength in nanocomposites. However, tech-niques such as Squeeze Casting can eliminate porosity. Anotherfactor is the degree to which the particles agglomerate and fail toachieve uniform dispersion. As far as ultrasonic dispersion isconcerned, de-agglomeration depends on how much power canbe transmitted to the melt. Therefore, processing technique isone of the factors which need to be considered in pursuing ad-vances in MMNC's [4].

Recently, research has been focused on nano-sized carbonousmaterials, such as carbon nanotubes (CNTs) and nano-graphite orgraphene [1] in order to attain enhancedmechanical, electrical, andtribological properties. For instance, carbon nanotubes have beendemonstrated to exhibit ultrahigh strength and modulus, and alsohave anisotropic electrical conductivity; when included in a matrix,they could pass on signicant property improvements to theresulting nanocomposites. Thus, the application of nanotechnologyto materials science and engineering opens up new opportunityand research direction for the development of novel smart metalmatrix nanocomposites. Carbon nanotubes and graphene possessexceptional mechanical strength as well as excellent electrical andthermal conductivities, and their incorporation in metallic matricesleads to composites with higher mechanical, electrical, and mag-netic properties. This has been led to an increasing interest inincorporating carbon nanotubes and graphene in MMCs to be themost effective reinforcing llers in synthesizing self-lubricatingcomposites for structural engineering and functional devices[28e30]. Table 1 summarizes a list of research on MMCs containingcarbon nanotubes and graphene. Carbon nanotubes and graphenein aluminum and copper alloys based composites enhanced theirstrength and tribological properties. The carbon nanotubes andgraphene were observed to reduce the grain size in aluminum al-loys, resulting in an additional higher strength. The incorporation ofcarbon nanotubes and graphene also increases the effectiveness ofpure aluminum.

Extensive researches have been done on polymer matrix com-posites reinforced by graphene in the last two decades. Almost all ofthese researches have shown that by adding graphene as rein-forcement to a polymer matrix, the properties of polymer matrixcomposites tremendously improved [37e42]. Tang et al. [37] haveshown that by adding 1 wt% graphene as reinforcement to poly-vinyl alcohol, the tensile strength and tensile modulus increased by178% and 139%, respectively. Additionally, only a very low volumefraction of graphene platelets (GPLs) increases the dielectric con-stant of polymermatrix composite. It has been also reported that bycombining GPLs and BaTiO3 and incorporating them into polymermatrix, the highest dielectric constant was reached without scari-fying the low dissipation factor [38]. Graphene based materialsprovide high thermal conductivity enhancement as well as theadvantage of improving barrier properties in comparison withpolymer matrix composites reinforced by CNTs [39,43]. Although,numerous polymer matrix nanocomposites reinforced by graphenenano-platelets and CNTs have been studied, still the challenge ofadding nanocarbonous materials into a metallic matrix and syn-thesize a fully uniform and dispersed structure is remained open toresearchers [41,44e47]. At the same time, another important pro-cessing issue in metal composite fabrication is the low interfacialstrength between the CNTs and the matrix. In CNT/polymer

ites Part B 77 (2015) 402e420 403nanocomposites, CNTs and polymer interact a molecular level

anot

ne

eanotionsingite

d

acti

osit

poswhile in the case of CNT/metal these types of bonding are notavailable.

It is expected that by adding graphene nano-platelets to a metalmatrix, the mechanical and tribological properties would beenhanced. In literature, only limited researches on metal matrixcomposites reinforced by graphene are available [10,31,32,48e53].To the authors' knowledge, so far, aluminum matrix compositereinforced by graphene has successfully produced only by powdermetallurgy method [31,48]. Recently, Wang et al. [31] have shown

Table 1Research reported in literature on metal matrix/nanocarbonous (including carbon n

Matrix Reinforcement Process

Aluminum Graphene Powder metallurgy technique

Aluminum Graphene Mixing AA2124 powder and grapheplatelets and then cold compactionat 525 MPa pressure

Copper nanographite Powder metallurgy technique whernanographite were dispersed in ethand then copper introduced to soluafter that drying the powder and u450 MPa pressure to make compos

Aluminum CNT Synthesizing by hot press and hotextrusion.

Aluminum CNT Sintering the mixture of aluminumand CNTs powders in a carbon molunder 50 MPa pressure

Aluminum MWCNT High energy ball milling, cold compand hot extrusion were employedto synthesizing composite.

Chromium MWCNT Cr/MWNT coatings were electrodepfrom electrolytes

A. Dorri Moghadam et al. / Com404that by adding 0.3 wt% Graphene nano-sheets to the aluminummatrix, the tensile strength of composite increased by about 62%.However, Bartolucci et al. [48] have shown that the tensile strengthand strain at failure of aluminum matrix composites reinforced by0.1 wt% graphene platelets are less than its pure aluminum matrix.In addition to these two researches, Chen et al. [49] have producedmagnesium matrix composite reinforced by graphene nano-platelets. They employed a novel method combining liquid stateultrasonic processing and solid state stirring to fabricate the com-posites. By using this novel method, they reported that the gra-phene nano-platelets (GNPs) could be dispersed uniformly intomagnesium matrix. The results showed that the micro hardness ofmagnesium matrix composite reinforced by GNPs has beenincreased by 78% compared to that of pure Mg prepared under thesame processing condition. They have also shown that the GNPsshow an excellent strengthening effect on the magnesium matrixcomposite [49,50,54e57].

Although the emerging research interest in smart materials suchas self-lubricating composites inspires both academia and industrythat the combination of these carbonous materials and metallicmatrices could potentially create composites that have high ther-mal and mechanical properties as well as exceptional wear resis-tance, there is still a need of understanding the nature, processing,and tailoring of these composites. This review initially covers a briefintroduction of various nano-reinforcements potentially are used inself-lubricating nanocomposites. Then, it addresses the currentprogress of research in self-lubricating nanocomposites followedby discussing the effect of these reinforcements on tribologicalproperties.2. Properties of carbonaceous nanomaterials

2.1. Carbon nanotubes (CNT)

If a sheet of carbon atoms is rolled, the carbon nanotubes willforms with a diameter of 1e2 nm which is called single-walledcarbon nanotubes (SWCNT). Other types of carbon nanotubes aredouble- and multi-walled nanotubes with diameters ranging from4 to 20 nmwhich are formed by rolling 2 or more carbon sheets as

ube and graphene) composites.

Properties Ref.

Increasing the tensile properties by embedding0.3 wt% graphene while graphene particles werepulled out from the fracture surface

[31]

Increasing the hardness and decreasingthe relative density. On the other hand,there is an optimum point that wearrate is minimized.

[32]

l,

Copper/nano-graphite exhibit better tribologicalproperties than copper/micro-graphite.Also increasing the volume fraction of reinforcementtends to improve tribological properties

[30]

No nanotubes damaged. No reaction products at theinterface between the matrix and carbon nanotubes.No signicant effect of annealing on the strength,while the strength of the pure aluminum matrixdecreases with annealing time.

[33]

No change in elongation while there is a signicantimprovement in tensile strength.

[34]

on Wear resistance and hardness of compositesignicantly increased while COF decreased.

[35]

ed Wear rate of composite decreased in comparedwith unreinforced chromium

[36]

ites Part B 77 (2015) 402e420they are schematically shown in Fig. 1. CNT has unique mechanicaland physical properties as well as lubricant nature. Owe to thesecharacteristics, it would be a promising candidate as reinforcementin a metallic matrix to enhance the properties of materials such asinherent stability at high temperature, high strength and stiffness,superior electrical and thermal conductivity and improved perfor-mance of metals in industrials components.

The results of the Brenner potential in predicting the modulusvalue of CNTs reveal that the modulus value of CNT is 1060 GPa thatit is very close to in-plane graphing [58]. A study was employedusing Molecular Dynamic (MD) simulation with a universal forceeld to measure the stiffness value of SWNTs and this turns into a1 TPa stiffness value for SWNTs [59]. Researchers have developedseveral methods to measure elastic modulus and strength of CNTsand one of the key techniques is measuring the amplitudes ofthermal vibration of nanotubes. This method measured an averagevalue of 1.8 TPa for elastic modulus with a large scatter in the re-sults ranging from 0.4 to 4 TPa. Using a similar technique, themodulus of 1.25 TPa for MWNTs grown by laser ablation was ach-ieved [60].

Although SWNTs have superior mechanical properties, but yetthey are not employed extensively as a reinforcement because theyare costly to be produced and puried. On the other hand, MWNTsare easier to produce. MWCNTs are composed of number ofcentered layers. One of the drawbacks of using MWNTs is thesusceptibility of the inner tubes to be pulled-out of the outer tubeby tensile stresses. This phenomenon is usually referred as tele-scoping effect or telescopic extension of multiwall carbon nano-tubes. In-situ TEM measurements on nanotubes demonstrated

(CVD) of graphene on metal carbides or metal surfaces [65,66], and

delamination. Presence of grooves on the worn surfaces of the

-, an

poswet chemical synthesis of graphene oxides followed by reduction[67,68].

The yield strength predicted for a single graphite layer using MDsimulation has reached an extreme value of 0.912 TPa [69]. Anotherstudy employed quantum mechanical approach revealed that theelastic modulus for armchair graphene and zigzag graphene are1.086 and 1.05 TPa, respectively [70]. The Young's modulus andthat this phenomenon occurs at ultra-low-friction state whichrevealed no wear or fatigue on the atomic scale. Telescoping failureand the higher probability of defects in MWNTs make them lessfavorable compared to SWNTs. However, still the strengthmeasured in MWNTS is much higher than that of high strengthmetals such as steel [61,62].

2.2. Graphene

Graphene is two-dimensional single atomic carbon sheet of sp2-bounded in which atoms densely packed in a honeycomb lattice.Graphite, the most common form of carbon, is stack of severalgraphene sheets along the c-axis with an interlayer spacing of0.34 nm. The bonding between carbon atoms is very strong whilethere are weak van der Waals interactions among the layers. Interms of thermodynamics, it was thought that exfoliation of layeredgraphite to freestanding atomic layer would not be possible [52].However, to date different approaches have been developed forsynthesizing graphene in large quantities, including thermalevaporation of silicon carbide [63,64], chemical vapor deposition

Fig. 1. Single-, double

A. Dorri Moghadam et al. / Comintrinsic tensile strength of graphene monolayer were experi-mentally tested by using nano-indentation of the atomic forcemicroscope (AFM). The Young's modulus and intrinsic tensilestrength obtained using these techniques are 1.1.02 TPa and130 GPa, respectively [71]. By using the same method, mechanicalproperties of graphene bilayer and trilayer have been determinedwhere Young's modulus is 1.04 and 0.98 TPa and intrinsic tensilestrength is 126 and 101 GPa, respectively [72]. These suprememechanical properties of graphene along with extreme thermalconductivity (5000 W m1 K1) [73], and super charge-carriermobility (200,000 cm2 V1 s1) [74] makes them an attractivematerial for researchers in the last decade to employ them asreinforcement into a metal matrix. The graphene has a plate shape;dispersion in any kind of matrices is easier in comparison withCNTs. Hence, the graphene too is a good substitution for CNTs asreinforcement for metal matrix composites [31]. Although gra-phene is dened as graphite single layers, graphene nanoplatelets(GNPs) or graphene nanosheets (GNSs) which are short stacks ofplatelet-shaped graphene sheets with an average thickness of thecomposites in the sliding direction at low normal loads shows thatthe abrasion wear mechanism becomes dominant.

Graphite is well-known reinforcement for metal matrix com-posites which acts as solid lubricant and makes the composite asself-lubricating composites [1,4,22,30,75,76]. When graphite isembedded into a metal matrix, the friction and wear behavior ofmetal/graphite composite signicantly improves compared to un-reinforced metal which lead to their increased industrial applica-tions where tribological properties are dominant. Damage5e100 nm are very common in the fabrication of metal matrixcomposites. Since, graphene in its single layer form cannot be easilybe stable in the free State, usually GNPs are used as reinforcementand then the sheets are exfoliated to achieve a single layerdispersed graphene in a matrix. This inexpensive material pos-sesses good thermal conductivity, electrical conductivity, me-chanical strength and more surface area than the expensive carbonnanotubes (CNTs).

3. Self-lubricating nanocomposites

During relative motion of two surfaces, different types of wearmechanisms, including adhesive wear, abrasive wear, delimitationwear, erosive wear, fretting wear, fatigue wear, and corrosive/oxidative wear may occur. The nature of wear mechanisms can beunderstood by studying the worn surfaces of materials. At lowloads and sliding speeds, abrasion is the dominant wear mecha-nism while at higher loads, the wear mechanism changes to

d multi-walled CNTs.

ites Part B 77 (2015) 402e420 405accumulation will be reduced in the presence of graphite particlesand hence decrease the wear rate of metal matrix insignicantextent. MMCs reinforced by graphite particles or bers are potentialstructural materials for aerospace and automotive owe to theirexcellent tribological properties.

Among many alloys, aluminum based composites are exten-sively used in various industries because of high strength to weightratio, superior tribological properties, and good corrosion re-sistivity. The explanation for the superior tribological properties ofaluminum/graphite composites can be explained by the wearmechanisms which occur in these systems. Aluminum alloys havelow yield stress and deforms extensively during sliding contactwhile graphite particles in aluminum/graphite composite improvethe deformation and fragmentation of the surface and sub-surfaceby providing a continuous lm of graphite on the contact surfacesafter short running-in period. The graphite lm hinders directmetal to metal contact and hence prevents seizure. Despite of goodtribological behavior of metal/graphite composite, poor mechanicalproperties is the disadvantage in the graphite reinforced MMCs.

Fig. 2. Comparison between aluminum matrix macro and nanocomposites reinforced by 15 vol% Al2O3 [80].

A. Dorri Moghadam et al. / Composites Part B 77 (2015) 402e420406These composites sometimes have lower mechanical propertiesthan unreinforced alloy [1,77]. In addition, graphite has a reverseeffect on electrical conductivity when copper alloys are reinforcedby micron sized graphite due to be hindering effect of particles inthe continuous copper matrix network, though it has a moderateelectrical conductivity. Another feature that causes to reduceelectrical conductivity of copperegraphite composite is the poorinterface bonding between copper and graphite particles whichleads tomore electron scattering [30]. Due to these shortcomings inusing graphite as reinforcement in metals, incorporation of nano-sized carbonous materials sought to be promising.

In general, it is desirable in terms of mechanical properties tohave matrix grain size in the range of nanometer to achieveenhanced hardness, yield strength, and tribological properties suchas wear resistance and friction coefcient [78]. Using nanosizeparticles as reinforcement also enhances both Young's modulus andtensile strength of composites as well as improving tribologicalperformance. Due to the fact that nanocarbonous materials havesuperior physical and mechanical properties, they have recentlyemployed as a novel reinforcement for self-lubricating metal ma-trix nanocomposite. Superior properties of MMNCs reinforced bycarbon nanomaterials is due to metallurgical factors, such asHallePetch effect by grain size renement, Orowan looping anddislocation generation resulting from thermal mismatch betweenthe matrix and reinforcements [52]. Previous studies revealed thatthe MMCs with smaller size reinforcements exhibit lower coef-cient of friction and wear rate, thus, it was concluded that theMMNCs have excellent tribological properties rather than metalFig. 3. a)Optical image after sessile drop of aluminum droplet on graphite substrate showsubstrate after the wetting angle reaches a nearly steady state [83].matrix micro-composites, as it is also experimentally conrmedand the results are presented in Fig. 2 [1,79,80]. More specically,the composite reinforced by nanoparticles (graphene) has lowerCOF than the composite reinforced by micro particles (graphite).Also, the hardness of composites reinforced by graphene is found tobe higher than the composite reinforced by CNTs [4]. Worn surfaceobservation suggested that the dominant wearmechanism for non-reinforced pure Al specimen has been delaminating wear accom-panied by some adhesive wear mechanism. However, worn sur-faces of the nano-particle reinforced composites were smootherand the total depth of deformations were smaller, grooves werener than the unreinforced aluminum alloy matrix specimens[81,82].

There is a great challenge in introducing carbonous materials tometal matrices. Generally, molten aluminum is not able to wetcarbonous materials, such as carbon bers (CFs), graphite particles,carbon nanotubes (CNTs) and graphene where the contact angle ofmolten aluminumwith graphite is between 140 and 160 [83]. Thereason for high contact angle between carbonous materials andmolten aluminum is due to the high surface tension of aluminum inliquid state. The surface tension of molten aluminum and carbonnanotubes are 955 mN/m and 45.3 mN/m, respectively [84]. Thevery high value of the surface tension of molten aluminumcompared to carbon nanotubes makes synthesis of aluminummatrix composite reinforced with carbonaceous materials a chal-lenging task. One of typical way to improve wetting behavior ofmolten aluminum on carbonous materials is by forming metalliccoatings, such as copper and nickel on reinforcements to reduce itsing non-wetting behavior b) Wetting of molten aluminum on copper-coated graphite

contact angle, as shown in Fig. 3 [85,86]. The formation of Al3Ni,Al3Ni2 and CuAl2 as an intermetallic compound plays a key role inachieving good wettability between aluminum with copper andnickel [28,87]. In the following sections, self-lubricating metal/CNTand Metal/Graphene (single layer or nanoplatelets) nano-composites have been introduced and their mechanical propertiesare discussed.

3.1. Metal-CNTs nanocomposites

The shear lag models, used in the case of conventional berreinforced composites, have also been applied to CNT composites.The stress is transferred to the ber (sf) through the interface and isrelated to the shear stress (tmf) between the ber and matrix givenby:

lf sf (1)

through using these techniques is usually weak, which causes theload transfer ineffectiveness.

The potential energies of interaction between two parallelinnitely long carbon nanotubes of the same diameter can besimplied greatly by assuming only van der Waals interactions ingraphitic systems. For a pair of parallel carbon nanotubes at a dis-tance of about 0.315 nm the cohesion energy has been calculated tobe about 37 kT nm1. So, an agglomerated CNT bundle needs about120 kT nm1 since three tubeetube contacts have to be broken toseparate a tube from a bundle. On the other hand, carbonecarbonbond energies lie about 190 kT. From this standpoint, it can beconcluded that the strong cohesion and very small difference be-tween carbonecarbon covalent bond energy, and CNTeCNT van derWaals energy, dispersing nanotubes is a difcult task which needscareful considerations. In polymer matrix composites, generally,ultrasonication aqueous media combined with an appropriatesurfactant is a technique to de-agglomerate CNTs. However, thetask becomes more sophisticated in metals processing. Although

A. Dorri Moghadam et al. / Composites Part B 77 (2015) 402e420 407Df 2tmf

where lf and Df are the length and diameter of the CNT, respectively.Carbon nanotubes with a larger aspect ratio assist larger loadtransfer and hence efcient utilization of reinforcement. For acritical length lc, the value of sf becomes equal to the fracturestrength of CNTs. For nanotube lengths l < lc, the fracture strengthof the composite is given as:

sFracC Vf sFracf

l2lc

VmsFracm (2)

Dislocation generation by thermal expansion mismatch be-tween matrix and CNTs and also Orowan looping mechanism playan important role in strengthening of aluminum/CNTs composites[88]. To achieve the theoretical value of strength prediction, uni-form dispersion of CNTs in the matrix during synthesizing is a greatchallenge [89]. Many attempts have been done in synthesizingCNTs reinforced metal matrix composites using traditional liquidcasting or powder metallurgy processes which have not shownpromising results [90e92]. In case of liquid mixing due to the poorwetting properties of CNTs and graphene with liquid aluminumand the difference in their densities, CNTs and graphene immedi-ately oat to the surface without being mixed. In P/M route and hotpressed sintered aluminum CNT composites, CNTs cause atremendous amount of residual stress in aluminum matrix whichcause a lower the tensile strength in the non-annealed compositesthan that of pure aluminum. Additionally, the bonding obtainedFig. 4. Variation of (a) relative density and hardness, and (b) tensile strength and elohigh intensity ultrasonication during liquid metal processing hasbeen employed in synthesizing nanocomposites, but still noexperimental or simulation data is available in CNT reinforcednanocomposites conrming the complete and successful disper-sion of these reinforcements in a metallic matrix [93e96].

In late 1990s, Kuzumaki et al. [33] fabricated aluminum/5 vol%CNT and aluminum/10 vol% CNT composites by mixing aluminumpowders and carbon nanotubes in ethanol followed by hot pressingand extrusion. However, no improvement in tensile strength ofaluminum/5 vol% CNT and aluminum/10 vol% CNT was reported incomparison to pure aluminum due to the poor dispersion ofnanotubes in aluminum matrix. Different factors, such as type ofnanotubes, functionalization of nanotubes, nanotube contents,matrix materials, and milling times can signicantly affect thedispersion of CNTs in a metallic matrix. Generally, to achieve auniform dispersion of CNTs in the nal composite, a homogenousdistribution of CNTs in the powder mixture at the starting stage isan important factor [52]. Uniform dispersion of CNTs in matrixtends to increase the hardness of composites compared to unre-inforced pure aluminum. If only a small amount of CNT isembedded into a metal matrices, the hardness of the compositeincreases due to the fact that themicro voids of metalmatrix will belled by CNTs. Beyond a specic CNT volume percentage, the excessCNTs which were not able to ll the micro voids will be agglom-erated with the aluminum particles. This agglomeration interruptsthe complete sintering and leads to the formation of defects whichultimately results in gradual reduction in hardness [97,98].ngation with carbon nanotube content for AA2024/MWNT nanocomposites [29].

electrochemical deposition, metal evaporation, and hydrogenreduction of metallic salts-graphite composite [103e105]. Howev-er, so far fully exfoliated graphite akes have not been obtained.Additionally, the difculty in large-scale synthesis of these com-posites becomes an obstacle in their production.

The powder metallurgy (P/M) route which is recently widelydeveloped for the fabrication of aluminum/CNT composites can beconsidered as an applicable pathway to fabricate aluminum/GNScomposites. Graphene oxide (GO) nanosheets have hydroxyl andepoxy groups on its surfacewhich helps to have better dispersion insolvents and formmore stable solutions than graphene. That makesthe GO nanosheets a more favorable reinforcement rather thanGNSs [106]. Fig. 5 shows themain difference between graphene andGO. Wang et al. [31] reported a route in synthesizing aluminum/GOcomposites using powder metallurgy in which four steps areinvolved:

(1) Exfoliating GO into several-layered or single-layered nano-sheets by sonicating GO aqueous dispersion in deionizedwater.

(2) Aluminum akes surface modication through ball millingfollowed by introducing a hydrophilic membrane on thesurface of the aluminum akes such as PVA.

(3) Adsorption and reduction of GO nanosheets by adding thepowder slurry of modied aluminum akes in deionized

posites Part B 77 (2015) 402e420Deng et al. [29] have investigated the physical and mechanicalproperties of aluminum AA2024/MWNT nanocomposites synthe-sized by cold isostatic press followed by hot extrusion. Duringsynthesis, AA2024 powders and MWNTs were rst mixed inethanol using ultrasonic mixing. Then, ethanol evaporated and thedried mixture ball milled and nally cold isostatic pressing and hotextrusion at 450 C were performed. Fig. 4a shows the changes inrelative density and Vickers hardness with CNTs content. Byincreasing theweight percentage of CNTs up to 1wt% of MWNT, therelative density and hardness of the nanocomposites increase. Itcan be clearly seen that the relative density of nanocompositecontaining 2 wt% MWNT sharply dropped which could be due tothe formation of nanotubes clusters. As shown in Fig. 4b, tensilestrength and elongation of aluminum/CNT composites depict thesame trend as the relative density. The AA2024/1 wt% MWNTexhibit maximum elongation and tensile strength and there is adrop in mechanical properties at high amount of reinforcements.

Noguchi et al. [99] developed a two-step method in whichinitially CNTs were uniformly dispersed in an elastomer matrix;and in the second step, the elastomer matrix was displaced by Al.Low-energy ball milling in a Turbula mixer followed by hotcompaction, sintering and HIPing was another route which wasused to homogenize the mixture Al/CNTs and eliminate CNT clus-tering. Esawi et al. [100], showed the effect of mixing time andmixing speed on CNT cluster size using dry mixing in a Turbulamixer. He et al. [101] were also developed an in-situ route to syn-thesize CNTs inside the aluminum matrix by using a three-stepprocess including deposition-precipitation, reduction, and chemi-cally vapor deposition route. They have reported that the rst stepinvolves producing a Ni(OH)2eAl precursor through a deposi-tioneprecipitation route. In their method, Ni nanoparticles wereused as catalyst onwhich the CNTs will grow. Therefore, by uniformdistribution of Ni nanoparticles on the surface of Al powders, auniform distribution of CNTs could be guaranteed. Cha et al. [89]used another technique to mix CNTs and matrix material in a so-lution. In their fabrication process which basically involvesmolecular-level mixing of CNTs and the matrix material, a sus-pension of reinforcements, and dissolved metal ions in ethanol orwater is prepared. Then the suspension was dried and oxidizedmetal powders forms. Finally, the metal oxides are reduced undercontrol atmosphere to achieve a dispersed CNT reinforced metalmatrix composite.

3.2. Metal-graphene nanocomposites

Similar to aluminum/CNT composites, in order to achieve thefull potential of graphene nanosheets (GNSs) as reinforcement, ahomogenous distribution of GNSs in the aluminum matrix alongwith maintaining the structural integrity of the GNSs is essential.Aggregated graphene behaves no differently than particulategraphite platelets. The ultrahigh surface area that can be obtainedin a 2D graphene sheet is lost when these sheets are clustered[53,102]. In this regard, the main challenge in fabricating metal/graphene nanocomposites is to nd an approach to fully dispersethese sheets or exfoliate the single sheets of GNS to graphenemonolayer. Although, exfoliation of GNSs in polymeric matrices hasbeen studied extensively and successful results have been achieved,but yet in a metallic matrix it has been remain as the key challenge.The main obstacle in achieving a highly exfoliated structure inmetal-graphene nanocomposites is the high difference betweencarbon and metals surface energies. This high surface energy dif-ference does not let metal to easily wet the graphene sheets and fallthem apart. Several investigations have been carried out to nd thebest route to synthesize a fully dispersed and homogenous gra-

A. Dorri Moghadam et al. / Com408phene reinforced metal matrix composite includingwater to the GO aqueous dispersion. The mixed slurry colorchanges from brown to transparent during stirring. Heatingthe aluminum/GO composite powders decompose the hy-drophilic membrane and reduce the GO nanosheets to GNSs,until nally aluminum/GNS composite powders is obtained.

(4) Compacting and consolidation of aluminum/GNS compositepowders. Consolidation can be achieved by sintering in anargon atmosphere followed by hot extrusion.

Bastwros et al. [10] used graphite and exfoliated it to graphenein nitric acid and sodium chlorate solution. The intercalatedgraphite was achieved through sedimentation and nally theintercalated graphite was exfoliated to monolayer or few-layerFig. 5. Surface structure of the graphene and GO.

graphene oxide using ultrasonication. Then, they employed ballmill to mix Al6061 powder and graphene at different milling times.The composites were then synthesized by hot compaction in thesemi-solid regime of the Al6061. Ghazaly et al. [32] have synthe-sized the aluminum graphene at different weight percentage (0.5, 3and 5 wt%) by employing powder metallurgy technique. Aluminumpowder with the graphene nanosheets weremixed in a high energyball mill which resulted in the formation of nonuniform particles ofaluminum covered by graphene layer and disappearance of thegraphene nanosheets. A combination of cold compaction and hotextrusion at ~0.45Tm (305 C) were employed to synthesizealuminum/graphene self-lubricating nanocomposite. The resultsshow a decrease in density of nanocomposites by increasing theamount of graphene as shown in Fig. 6. Also, the variation ofhardness with increasing the amount of graphene is presented inFig. 6. As there was an expectation that addition of graphene up to3 wt%, the hardness of nanocomposites reinforced by grapheneincreasewhen compared to unreinforced alloy. The 3wt% graphenereinforced composite produced a 47.5% increase in hardness overthe base AA2124 alloy. By further increasing the amount of gra-phene (more than 3 wt%) nanocomposite, the hardness reduces butthe hardness values are more than the unreinforced alloy and lessthan the composite with reinforcements between 0.5 and 3 wt%. Bycomparing Figs. 4 and 6, it is obvious that aluminum/graphene hassuperior mechanical properties than aluminum/CNT.

To achieve individual graphene sheet, according to many avail-

extrusion [31]. The normalized tensile strength and ductility of Al/GNS composites in comparison with unreinforced pure matrix isshown in Fig. 7a. The tensile strength of aluminum/GNS nano-composite is 249 MPa. The nanocomposite showed a signicantimprovement (about 62%) in tensile strength compared to that ofunreinforced aluminum matrix (154 MPa). Higher tensile strengthof nanocomposite compared to unreinforced aluminum demon-strate that GNSs have a dominant role for increasing mechanicalproperties and reveal that the GNSs have a good potential to beused as reinforcement in aluminum matrix composites to improvethe mechanical properties. Generally, GNSs could contribute to thestrength improvement by grain size renement, dislocationstrengthening and stress transfer. GNSs hinder grain growth andhence result in having grain renement and also provide resistanceto the dislocation movement during thermal processing and plastic

A. Dorri Moghadam et al. / Composites Part B 77 (2015) 402e420 409able literatures, another method is attaching molecules or poly-mers to graphene sheets to decrease aggregation [51,107]. Xu et al.[53] suggested a combined technique in which a solvent, grapheneoxide, and an inorganic nano particle was incorporated to obtaingraphene as individual sheets inside a metal matrix. The inorganicnanoparticles sits in between graphene sheets and prevent re-agglomeration of them. . After the solvent is dried, dispersion ofisolated graphene sheets can be obtained having graphitic stackswith inorganic nanoparticles in the interlayer spacing.

Aluminum/0.3 wt% graphene nanocomposite was also synthe-sized by using a slurry based process followed by sintering and hotFig. 6. Relative density and hardness variations as a function of graphene addition toAA2124 nanocomposite [32].Fig. 7. a) Tensile properties of 0.3 wt% GNS/Al composite and the corresponding akyAl specimen b) Fracture surface of 0.3 wt% GNS/Al composite [31].

Fig. 8. Variation of (a) Vickers hardness, (b) Density, and (c) Compressive strength ofaluminum alloys with exfoliated graphite nanoplatelets particles contents for GNP/Alcomposites at different sintering temperature [108].

posites Part B 77 (2015) 402e420deformation, therefore, the tensile strength of aluminum matrixincrease in presence of GNSs. Another important reason of hightensile strength in presence of GNSs is the load bearing abilitywhere the graphene tolerate a substantial part of load duringplastic deformation. The fracture surface is shown in Fig. 7b. Inas-much as the fracture strength of a perfect single layer graphene is125 GPa, the theoretical strength of the aluminum/graphene is500 MPa by using the rule of mixtures for aluminum/0.3 wt% GNSs.The experimental result shows that strength of aluminum/0.3 wt%GNSs is 250MPa. The less strength in experimental than theoreticalstudy is due several parameters, such as to different alignment ofGNSs along the tensile direction, processing parameters, micro-structure, and weak interfacial bonding of the aluminum/GNSs[48].

Latief et al. [108] have synthesized aluminum/graphene com-posites using different percentages of exfoliated graphite nano-platelets particles by employing powder metallurgy method inorder to study the physical and mechanical properties of nano-composites. As observed in Fig. 8, the results revealed that theVickers hardness (Fig. 8a) and compression strength (Fig. 8c) in-creases with increasing graphene content up to 5 wt% in the purealuminum matrix while the density (Fig. 8b) decreases withincreasing graphene content. This can be expressed by followingequation [109]:

l 41 f r3f

(3)

where l, f and r are respectively the distance between the rein-forcement particles, particles volume fraction, and the particleradius (particle is assumed spherical). The shear stress for all testedalloys can also be calculated according to the following equation:

t0 Gbl

(4)

where, t0, G and b is the shear stress, the shear module, and theBurger's vector, respectively [110]. According to Equation (3), it canbe concluded that the distance between the graphene particlesdecreases by increasing their amount in the composite. Based onEquation (4), the shear stress that is required to move dislocationsbetween the graphene particles will increase when the distancebetween the reinforcement particles decreases which results in anincreased yield stress of materials [3,111]. In addition, the re-inforcements are an obstacle that causes to lock dislocationmovement extremely in metal matrix through dispersionstrengthening mechanism. This mechanism can increase the me-chanical properties of aluminum matrix composites reinforced bygraphene [109].

Bartolucci et al. [48] have compared the mechanical propertiesof aluminum/MWNT and aluminum/GNP composites. Fig. 9 depictsthe hardness of self-lubricating aluminum nanocomposites atdifferent samples after hot isostatic pressing and extrusion. It isobvious from the results of hardness that the aluminum reinforcedwith 1.0 wt% MWNT exhibits the highest hardness than aluminum/0.1 wt% graphene and pure aluminum. Fig. 10 shows a compressionbetween the tensile strengths of composite reinforced by MWCNTand graphene and also effect of fabrication method includingextruded and pressed composites. The metal matrix compositereinforced by nanotubes show the higher strength than reinforcedby graphene. For extruded composites, the tensile strength ofaluminum/nanotubes was about 12 percent greater than the purealuminumwhile the aluminum/graphene showed about 18 percentlower tensile strength as compared to the pure aluminum. In

A. Dorri Moghadam et al. / Com410addition, Fig. 10 shows the average strain-to-failure of the samples

Fig. 9. Vickers hardness data for the various materials and conditions [48].

A. Dorri Moghadam et al. / Composwhere the results revealed samples reinforced by nanotubes andgraphene displayed the lowest ductility.

4. Tribological behavior of self-lubricating nanocomposites

Tribology is an investigation on wear and friction performanceof materials. It is the science of interacting surfaces in relativemotion [112]. When under an external load, two materials are incontact with each other, the asperities of two surfaces come intoclose contact and during movement, deterioration of the surfacesoccurs which is known as wear. During the sliding process of softermaterials against harder materials, atoms will be taken off from thesofter one and these atoms tend to locate themselves in the as-perities of harder surface. As results, a cold welding occurs incontact surface and interatomic junctions across the interfaceforms. By continuing the sliding process, fracture can take place atthe junctions and causes the detachment of the fragments fromadhering asperities. Friction force causes shear at interatomicjunctions during movement of two surfaces under an applied force.Fig. 10. Ultimate tensile strengths and strain-to-failure of pure Al, Ale1 wt% MWNT,and Ale0.1 wt% graphene [48].Archad [113] expressed a formula for wear of materials that de-scribes the volume of wear loss (V) of materials due to adhesivewear:

V c PLH

(5)

where c, P, L and H are wear coefcient, applied load, sliding dis-tance, and hardness of the softest contacting surfaces, respectively.

Usually to avoid friction and consequently deterioration ofmaterial under wear, liquid or solid lubricants are employed.However, in cases such as high vacuum environment, high-speedconditions, high applied loads, and very low or high tempera-tures, liquid and grease type lubricants are undesirable. In such atribological systems the common liquid and grease type lubricantsdo not show desired performance or durability [52]. Anotherapproach is replacing the liquid and grease type lubricants withsolid lubricant coatings that they are used to decrease coefcient offriction and wear rate. The coatings are applied on the surface ofmaterials by depositing via chemical or physical vapor depositiontechniques to form a coating layer [114,115]. The disadvantages ofsolid lubricant coating are limited lifetime, difculty in replenish-ment, oxidation and aging-related degradation, and poor adhesion.Therefore, to avoid the drawbacks of both the liquid and grease typelubricants and the solid lubricant coatings, embedding carbonousmaterials in the metal matrix seems promising.

Generally, metal matrix composites have lower coefcient offriction (COF) compared to unreinforced matrix [1,4,35,116e118].Furthermore, adding ceramic particles to the metal matrices lead toan increase in wear resistance of the matrices [1,4,118e121]. Themain reason for increasing of wear resistance of metal matrixcomposite is attributed to low friction coefcient of metal matrixcomposite compared to the unreinforced metals. For conventionalmetal matrix composites, the reinforcements act as load bearingcomponents at contact surface which tend to protect the surfacefrom ploughing during sliding. Generally, the hardness of rein-forcement greatly affects the wear loss and hence, the wear volumeof MMCs. The wear loss of MMCs depend on several intrinsicproperties such as the reinforcements dispersion state, distributionof reinforcement, size of reinforcing particles, and interfacial bondbetween matrix and particles [52]. When bonding between matrixand reinforcement is poor, the hard ceramic particles are easilypulled out from MMCs and then they will be trapped between thesliding surfaces and act as third body abrasives and help to increaseworn surface damage and wear rate. Among the composites,composites reinforced by carbonous materials show better tribo-logical properties compared to composites reinforced by ceramicreinforcements, such as SiC and Al2O3 due to the lubricative natureof carbonous materials that make them a potential reinforcementfor self-lubricating composite. The conventional self-lubricatingcomposites are embedded by graphite particles or carbon bers [1].

The main reason for signicant decrease in COF and wear rate isdue to formation of a lubricant lm between the contact surfacesbecause of presence of carbon-based solid lubricant in the MMCs.Thus, the lubricant lm prevents direct contact between slidingsurfaces and reducing wear [77]. In addition, due to the presence oflubricant lm which prevents direct contact, the transfer of atomsfrom the asperities of softer surface to the asperities of hardersurface will be reduced that hence, it leads to decrease in coldwelding of atoms of softer materials with atoms of harder materialsduring sliding and then subsequent fracture of atomic junctions[30]. As noted before, although the graphite particles in the metalmatrix improve the tribological performance, it tends to reduce themechanical properties of the composites. Hence, recently, the nano

ites Part B 77 (2015) 402e420 411solid lubricants are used as the dominant reinforcement for the

metal matrices in self-lubricating composites. This is because themetal matrix composite reinforced by nano solid lubricant haveexcellent self-lubricating behavior with low coefcient of frictionand wear rate as well as high mechanical properties [122].

4.1. Effect of CNT

Carbon nanotubes (CNTs) are one of the important solid lubri-cants that is increasingly employed in novel self-lubricatingnanocomposite materials because of the superior properties ofCNT including electrical, optical, and mechanical properties andhave a key role in enhancement of wear resistance and reduction inCOF of metal matrix composite reinforced by CNTs. Metal matrixnanocomposites reinforced by CNT is an excellent candidate forindustrial applications because of its excellent mechanical proper-ties, lightweight and superior tribological properties. It is expectedthat the utilization of Multi-Wall Carbon Nano Tubes (MWCNTs) inthe composites will increase the industrial applications due to their

reasonable cost. Thus, there have been many investigations todevelopMMNCs reinforcedwithMWCNTs using various fabricationroutes [1]. Previous investigations exhibit superior tribologicalproperties of metal/CNTs composites as a result of the reduction inwear rate and the coefcient of friction (COF) due to the lubricatingnature of CNTs. CNTs form a lubricant lm between contact surfacesduring sliding.

It has been reported that when there is a strong bonding be-tween functionalized MWNTs and an epoxy matrix the outer shellsof the tube remain embedded in the matrix following pull-out[123,124]. However, in case of metal matrix, the MWCNTs areattached by a very week van der Waals forces, there is no directobservation that support the telescopic extension of multiwallcarbon nanotubes occurs. The week bonding between CNT-metalled to easily slide or roll between the contact surfaces and mini-mize a direct contact between the surfaces, thus results in decreasein friction coefcient of the composite. The improvement in wearresistance is attributed to the role of CNTs as spacers that prevent

A. Dorri Moghadam et al. / Composites Part B 77 (2015) 402e420412Fig. 11. Variations of a) Brinell hardness (HB) b) wear rate and coefcient of friction

with MWNTs content for melt-inltrated MWNT/AleMg composites under an appliedload of 30 N and a sliding velocity of 1.57 m/s [13].Fig. 12. Effect of volume fraction of CNT on friction coefcient and wear loss of Al/CNTnano composite at normal load 30 N and sliding speed 0.12 m/s [122].

direct contact between rough surfaces [28]. Generally, severalmaterial parameters, such as amount of reinforcements, size of

as shown in Fig. 11a. The effect of volume fraction of CNTs on thefriction coefcient and wear rate of the composite is shown in

Fig. 13. SEM images of wear tracks for (a) pure aluminumwith a grain size of ~150 nm and composites containing MWCNTs of (b) 1.5 vol%, (c) 3.0 vol%, (d) 4.5 vol%, and (e) 6.0 vol%,respectively [122].

A. Dorri Moghadam et al. / Composites Part B 77 (2015) 402e420 413reinforcement, and spatial distribution have direct effect on tribo-logical properties of self-lubricatingmetal/CNTs composites [1,125].

Zhou et al. [13] fabricated AleMg/MWCNT composite by using apreform with MWCNT and then using pressureless inltrationmethod to inltrate molten metal into the MWCNT preform.Embedding the MWCNT into AleMg alloy increases the hardness ofthe composites compared to unreinforced aluminum alloy.Furthermore, by increasing the MWCNTs volume percentage, thehardness of composite initially increases. Then, further increase inMWCNTs content has an adverse effect and decreases the hardnessFig. 14. Variations of COF of 4.5 vol% MWNT/Al composite with (a) applied load at sliding spe% MWNT/Al composite with (c) applied load at sliding speed 0.12 m/s and (d) sliding speeFig. 11b. It can be seen that although hardness has an optimumvalue with increasing volume fraction of CNTs in the composite, yetthe coefcient of friction decreases even at high CNT content. Athigh amount of CNTs the direct contact between the metal surfacesis hindered which ultimately results in better tribological proper-ties. The favorable effects of CNTs on tribological behavior ofcomposites depend on their excellent mechanical properties, welldispersity in the composite and the efciency of CNTs as rein-forcement. X-ray diffraction (XRD) analysis of contact surface re-veals that the wear particles are mainly aluminum oxide. Duringed 0.12 m/s and (b) sliding speed at applied load 30 N. Variations of wear loss of 4.5 vold at applied load 30 N [122].

delamination. As shown in Fig. 13(a) and (b), grooves and materialdelamination were observed on the worn surfaces for purealuminum and aluminum reinforced by 1.5 vol% MWCNT. Thiswould conrm that micro-ploughing and delamination are the twomain dominant wear mechanisms of pure aluminum andaluminum/1.5 vol% MWCNTs. By increasing the volume percentageof MWCNTs to 3 and 4.5 vol%, less deep grooves can be observedand the surface is much smoother compared to the one it wasobserved in Fig. 13(c) and (d). The surface of aluminum reinforcedby MWCNTs of 6.0 vol% demonstrates rougher worn surface thanaluminum/4.5 vol% MWCNTs. At this amount of CNT the debris areable to easily separate out from the surface which can be justiedby presence of pores as shown in Fig. 13(e). This can be the mainreason for increasing of wear loss and COF at high volume per-centage of MWCNTs. Furthermore, They conrmed that the coef-cient friction of aluminum composite signicantly decreased byadding CNTs where COF is 0.35 and 0.06 for pure aluminum andaluminum/4.5 vol% MWNT, respectively, under an applied load of30 N and a sliding speed of 0.12 m/s. In addition, there is a variationbetween COF and volume content of CNT where COF decrease byincreasing the amount of CNTs. The effect of applied load andsliding velocity on coefcient of friction and wear loss was also

Fig. 15. Wear rate and weight loss variation as a function of graphene content inAA2124 matrices [32].

A. Dorri Moghadam et al. / Composites Part B 77 (2015) 402e420414wear process, laminated oxide lms formed at contact surfaces thatthey subsequently broke up and aked off due to low adhesionbetween the oxide lms and aluminum matrix. While the oxideparticles which form at contact surface are harder than thealuminum matrix and are able to increase abrasive wear. As thealuminum matrix gradually wear out during sliding process, theCNTs which was initially embedded in the matrix now are pulledout and exposed on the contact surface and form a lubricant lm onworn surface. Those solid lubricating lms signicantly reduce theadhesive wear cause by oxide particles compared to unreinforcedaluminum.

Choi et al. [122] have shown that the wear loss and coefcient offriction decrease with increasing the CNTs content. However,beyond a critical amount, 4.5 vol% CNTs in their investigation, thewear rate and friction coefcient increase as shown in Fig. 12. Thedeteriorated wear properties in the composites at high volume ofMWCNTs may be associated with the presence of voids and cracks

due to the very high amount of CNTs which could act as a source of

Fig. 16. SEM micrographs of worn surfaces of AA2124 a) unreinfoinvestigated and it is shown in Fig. 14. The investigations haveshown that the COF and wear loss increases with increasing normalload for aluminum/5 vol% MWNT composite at sliding speed of0.12m/s. However, the coefcient of frictionwas still lower than 0.1.At higher load, severe wear is the dominant mechanism whichresults in increasing friction coefcient and wear loss and subse-quently severe surface damage. On the other hand, the coefcientof friction and wear loss slightly decreases with increasing slidingspeed at an applied load of 30 N as shown in Fig. 14.

4.2. Effect of graphene

High strength, lightweight and lubricating nature of graphenemade it suitable as reinforcement for self-lubricating ultrahighstrength metal matrix nanocomposites. As this is fairly a novelmaterial and it is difcult to uniformly disperse in metals as well asits complex microstructure, there are only a few studies whichinvestigated the tribological properties of graphene in a metallicmatrix. Ghazaly et al. [32] who have investigated the effect ofrced, b) 0.5, c) 3 and d) 5 wt% graphene nanocomposite [32].

tribological properties under dry wear test compared to unrein-forced and other amount of graphene reinforcements as shown inFig. 15. SEM micrographs of worn surfaces of unreinforcedaluminum alloy and Al/graphene nanocomposites are shown inFig. 16, which clearly demonstrate the presence of longitudinalgrooves in all samples. In addition, by comparing theworn surfaces,it is obvious that the scratches, craters, and delamination ofAA2124/3wt% graphene composite is less than that of unreinforcedalloy. Thus, unreinforced alloy and AA2124/3wt%graphe-necompositeare in the severe and mild wear regime, respectively.Shallow parallel grooves and ridges formed on the worn surface ofAA2124/0.5 and 5 wt% graphene nanocomposite due to micro-ploughing. Thus, the dominant wear mechanism is severe plasticdeformation of the matrix that results in high wear rate. Entrapped

A. Dorri Moghadam et al. / Composites Part B 77 (2015) 402e420 415weight percentage of graphene on mechanical properties, alsostudied its effect onwear rate of self-lubricating AA2124 aluminumalloy matrix nanocomposites. The results showed that self-lubricating composite reinforced by 3 wt% graphene has better

debris between delaminated surfaces was observed at highmagnication onworn surfaces of unreinforced AA2124 alloy whilethere is no wear debris on the worn surfaces of nanocomposite asillustrated in Fig. 17. Alumina fragmented lms or strain hardened

Fig. 17. SEM micrographs for the worn surfaces of AA2124-a) 0, b) 3 and c) 5 wt%graphene nanocomposite [32].

Table 2Comparison of physical and mechanical properties betweenmicro graphite particlesand nanographite particle [30].

Composite Relative density(%)

Hardness(HV)

Electrical conductivity(%IACS)

Cu-15% Gr 92.3 0.13 72 1.6 65 1.5Cu-5% NG 95.82 0.14 94 1.9 79.8 1.9Cu-10% NG 96.41 0.13 90 1.0 72.4 1.0Cu-15% NG 96.92 0.12 81.5 1.6 70 1.3Cu-20% NG 88.42 0.15 56 2 38.7 2.5Fig. 18. Variation of a) Coefcient of friction with normal load at sliding speed 0.77 m/sb) Variation of coefcient of friction with sliding speed at 36 N [30].

particles are the two main sources of debris. This debris is from theheavily milled consolidated powders which were detached underthe load during the wear test. By comparing the worn surfaces athighmagnications, it is obvious that the surface of nanocompositecontaining 3% graphene is smoother than that of unreinforced alloy

Inasmuch as copper has good electrical and thermal conduc-tivities and graphite has lubricious nature, copper/graphite com-posites have variety of application in industries. Conversely, themechanical properties of copper composites decrease in the pres-ence of graphite reinforcement. To solve the impact of micro sizedgraphite particles, Rajkumar et al. [30] employed powder mixing,compaction and microwave sintering methods to synthesize cop-per nanocomposite reinforced by nano-graphite (NG) particleswith an average particle size of 35 nm to form copper/5e20 vol%NG nanocomposites. The graphite particles were coated with cop-per using electrodeless plating method. The nanographite particleshave not been exfoliated in this investigation and cannot beconsidered as single or few layer graphene sheets. Table 2 com-pares the physical properties, such as relative density and electricalconductivity and also mechanical properties, such as hardness ofsintered copper/graphite composite and copper/nano-graphitecomposites. As comparison shows, the nanocomposites had bet-ter hardness and electrical conductivity compared to micro-composites. As stated earlier, the volume percentage of nano-particles has an effect on physical and mechanical properties ofself-lubricating composites. The amount of nano-particles is alsoinuence the relative density as shown in Table 2. The relativedensity increases with increasing the volume percentage up to15 vol% of nano-graphite due to ability of nanoparticles to ll up theporosity cavities. When the nano-graphite amount is increasedover 15% volume fraction, the relative density and hardnessreduced due to the reduction in the distance between particles,which consequently facilitate nanoparticles agglomeration.

Fig. 19. Variation of wear rate of composites with normal load at sliding speed 0.77 m/s [30].

A. Dorri Moghadam et al. / Composites Part B 77 (2015) 402e420416and the composite reinforced with 5wt%graphene. Furthermore,the surface of AA2124/3 wt% graphene composite was covered bylubricant lms that results in reducing friction and wear due to thesoft nature of the lubricant lm. Conversely, deep grooves and se-vere damage exist on the worn surfaces of AA2124/5 wt% graphenecomposites which delaminated in the direction of sliding that ex-plains the signicant increase in wear rates and weight loss.Fig. 20. SEM image of worn surface of coppere15% nanographite composites at dFig. 18 illustrates the variation of normal load and sliding speedwith coefcient of friction at different volume percentage ofgraphite particles. Fig. 19 shows the variation of wear rate withnormal load for copper based composites reinforced by micro andnanographite particles. Results revealed that, at constant volumefraction, embedding nano-particles decreases the coefcient offriction and wear rate compared to the composite reinforced byifferent sliding speeds at 36 N a) 0.77 m/s, b) 1.77 m/s and c) 2.77 m/s [30].

micro sized graphite particles. Higher hardness, lower porosity andner microstructure are the reason for the improved wear resis-tance of nano-graphite reinforced composites. Further, the nano-graphite particles reinforced composites are more effective on thedegree of self-lubrication compared to micron-size graphite parti-cles reinforced composites. The amount of nano-graphite particlesalso inuences the tribological properties of self-lubricating coppercomposites. The increase in volume percentage of nano-graphite upto 15 vol% tends to decrease the wear rate and COF because of theformation of a uniform and continuous layer of solid lubricant lm.This lubricant lm reduces the rate of deformation of the matrixand improves the tribological behavior. When the amount ofreinforcement increases, the decrease in the COF is associated withincrease in the availability and uniformity of lubricant layer. Thelubricant layer causes to minimize the metal to metal contactsbetween the copper matrix composite and steel counter surface. Incontrast, when the volume fraction of nano-graphite is more than15 vol%, a large amount of agglomeration was observed that tends

to incomplete spreading of graphite at the contact zone, and hence,increases thewear rate. Increasing of COF at high volume fraction ofnano-graphite is a result of increasing the deformation and fractureat the contact surface of copper matrix and increasing the amountof copper debris at contact surfaces.

It can be seen from Figs. 18 and 19 that the wear rate and co-efcient of friction both increasewith increasing applied load whilethe coefcient of friction decrease with increasing the slidingspeed. Increasing normal load also increases the amount of copperwear debris at the contact zone and hence inuences the rate ofincrease in the coefcient of friction with normal load. In thesegures, it can be seen that the coefcient of friction of self-lubricating composite signicantly decreases with increasing thesliding speed up to 1.77 m/s because of formation of uniformlubricant lm. By increasing the sliding speeds beyond 1.77m/s, thecoefcient of friction slightly increase or become constant for 5 and10e15 vol%, respectively. This is due to peel off of the self-lubricating lm on the contact surface at high sliding speed.

c) co

A. Dorri Moghadam et al. / Composites Part B 77 (2015) 402e420 417Fig. 21. a) Distribution of nanographite in matrix, b) distribution of graphite in matrix,

f) conceptual wear generation model for nanographite and graphite reinforced composite reand copperegraphite respectively [30].ntact prole nanographite composite, d) contact prole of graphite composite, e) and

spectively, g) and h) typical wear debris at 48 N and 0.77 m/s for copperenanographite

[12] Rohatgi PK, Gupta N, Alaraj S. Thermal expansion of aluminumey ash

posFurthermore, sliding speed does not affect the coefcient of frictionof copper/nano-graphite with high amount of nano-graphite con-tent due to the contact surface that is uniformly covered with thehighly adherent graphite layer. As shown in SEM micrograph ofworn surface at different sliding speed, at constant normal load, inFig. 20, the lubricant lm on copper/15 vol% nano-graphite is notcontinuous at lower sliding speed. While increasing the slidingspeed, a lubricant layer uniformly form on the surface of compositethat decreases COF as a direct result of a decrease in direct surfaceto surface contact. However, a gradual increase in COF wasobserved for 20 vol% of nano-graphite composite by increasing thesliding speed that it leads to lower mechanical properties such ashardness due to increase in temperature at the interface. Further, ittends to more grain fracture during sliding. This phenomenon ismore intensive at higher sliding speeds.

The mechanism of wear under normal loads suggested byHuang et al. [126]. Fine graphite particles form an adherent layer atthe contact zone and under high normal loads; these nano-graphitefrom composite is squeezed out to the contact zone. Owe to theirsmaller size, nanographite particles are able to penetrate deep in-side the asperities of composite and counter surface during thesliding process. During sliding, the nano-graphite particles couldhave lled most of the asperities of the composite surface. So agraphite layer forms at the pin (composite) edisc (steel) interface.The layer formation process continues up to the formation of thickadherent graphite layer. In case of micron size graphite particles,they can also undergo similar process; however due to their largersize they are not able to penetrate into the very narrow grooveswhich formed during the wear process or gaps between the as-perities of sliding contact easily.

Additionally, when the size of graphite particles comes down inthe range of nanosize, at a same volume fraction, the mean freepath between the graphite particles also decreases (Fig. 21a)compared to same volume fraction of micron sized graphite par-ticles (Fig. 21b). This will cause smaller size asperities and also lessspace between the asperities compared to micro graphite rein-forced composite (Fig. 21d) during the wearing process which canbe lled nano-graphite particles as shown in Fig. 21c. Thecompletely lled nanographite particles produce more uniformgraphite layer that reduces the direct contact between the twowearing body and will cause reduction in the frictional coefcient.As conrmed by SEM (Fig. 21h and f), nanocomposite reduces thewear debris size, as shown in Fig. 21e when compared to micro-composites (Fig. 21f) [30].

5. Conclusion

Carbon nanotubes (CNTs) and graphene have superior proper-ties, including large aspect ratio, exceptional high Young's modulusand strength, and excellent electrical and thermal conductivity.These unique properties attract researcher to use them as rein-forcement for metal matrix composites to enhance properties ofcomposites and make them high strength, lightweight and self-lubricating. These favorable properties can be achieved only if thereinforcements are dispersed uniformly and not agglomerated inthe matrix. Based on the literatures available, so far, a few studieshave been conducted on metal matrix composite reinforced byCNTs and graphene because of difculty in synthesizing anddispersing. Most researches were focused on fabrication and someof them have been investigated the mechanical properties and self-lubricating properties. Mechanical properties of composites wereremarkably increased by adding CNTs and graphene. Importantphenomena for increasing the mechanical properties of metalmatrix composites reinforced by CNTs and graphene are dislocation

A. Dorri Moghadam et al. / Com418generation by thermal expansion mismatch between the matrixcenosphere composites synthesized by pressure inltration technique.J Compos Mater 2006;40:1163e74.

[13] Zhou S-M, Zhang X-B, Ding Z-P, Min CY, Xu G-L, Zhu W-M. Fabrication andtribological properties of carbon nanotubes reinforced Al composites pre-pared by pressureless inltration technique. Compos Part A 2007;38(2):301e6.

[14] Sukumaran K, Ravikumar KK, Pillai SGK, Rajan TPD, Ravi M, Pillai RM, et al.Studies on squeeze casting of Al 2124 alloy and 2124-10% SiCp metal matrixcomposite. Mater Sci Eng A 2008;490:235e41.

[15] Unlu BS. Investigation of tribological and mechanical properties Al2O3eSiCreinforced Al composites manufactured by casting or P/Mmethod. Mater Des2008;29:2002e8.

[16] Baradeswaran A, Perumal E. Wear and mechanical characteristics of Al 7075/graphite composites. Compos Part B 2014;56:472e6.

[17] Baradeswaran A, Perumal AE. Study on mechanical and wear properties of Al7075/Al2O3/Graphite hybrid composites. Compos Part B 2014;56:464e71.

[18] Iacob G, Ghica VG, Buzatu M, Buzatu T, Petrescu MI. Studies on wear rate andmicro-hardness of the Al/Al2O3/Gr hybrid composites produced via powdermetallurgy. Compos Part B Eng 2015;69:603e11.

[19] Liu X-B, Liu H-Q, Liu Y-F, He X-M, Sun C-F, Wang M-D, et al. Effects oftemperature and normal load on tribological behavior of nickel-based hightemperature self-lubricating wear-resistant composite coating. Compos PartB Eng 2013;53:347e54.and reinforcements, and also Orowan looping mechanism. Further,a few research papers are available on tribological behavior of self-lubricating nanocomposites reinforced by graphene and CNTs.A signicant reduction inwear rate and coefcient of friction can beachieved in presence of CNTs and graphene up to a critical volumefraction. The reinforcements prevent a direct contact between twosurfaces by forming a lubricant lm between the contact surfaces.Beyond a critical volume fraction of CNTs and graphene, the wearrate and COF increases. At higher amount of reinforcement, themechanical properties also decrease due to agglomeration thatleads to the formation of some defects which ultimately decreasethe tribological performance of the nanocomposites. Several ma-terial parameters, such as amount of reinforcements, size of rein-forcement, and spatial distribution have an effect on mechanicaland tribological properties. There is an optimum amount of rein-forcement in the composites to have excellent mechanical andtribological properties.

References

[1] Rohatgi PK, Tabandeh-Khorshid M, Omrani E, Lovell MR, Menezes PL.Tribology of metal matrix composites. Tribology for scientists and engineers.Springer; 2013. p. 233e68.

[2] Miracle DB. Metal matrix composites e from science to technological sig-nicance. Compos Sci Technol 2005;65:2526e40.

[3] Torralba JM, Costa CEd, Velasco F. P/M aluminum matrix composites: anoverview. J Mater Process Technol 2003;133(1e2):203e6.

[4] Dorri-Moghadam A, Schultz BF, Ferguson J, Omrani E, Rohatgi PK, Gupta N.Functional metal matrix composites: self-lubricating, self-healing, andnanocomposites-an outlook. JOM 2014;66(6):872e81.

[5] Sandoz-Rosado EJ, Tertuliano A, Terrell EJ. An atomistic study of the abrasivewear and failure of graphene sheets when used as a solid lubricant and acomparison to diamond-like-carbon coatings. Carbon 2012;50(11):4078e84.

[6] Tabandeh Khorshid M, Jenabali Jahromi SA, Moshksar MM. Mechanicalproperties of tri-modal Al matrix composites reinforced by nano- andsubmicron-sized Al2O3 particulates developed by wet attrition milling andhot extrusion. Mater Des 2010;31:3880e4.

[7] Yao B, Hofmeister C, Patterson T, Sohn Y-H, Van den Bergh M, Delahanty T,et al. Microstructural features inuencing the strength of trimodal aluminummetal-matrix-composites. Compos Part A Appl Sci Manuf 2010;41:933e41.

[8] Esawi AMK, Morsi K, Sayed A, Taher M, Lanka S. Effect of carbon nanotube(CNT) content on the mechanical properties of CNT-reinforced aluminiumcomposites. Compos Sci Technol 2010;70:2237e41.

[9] Kondoh K, Umeda J, Watanabe R. Cavitation resistance of powder metallurgyaluminum matrix composite with AlN dispersoids. Mater Sci Eng A2009;499:440e4.

[10] Bastwros M, Kim G-Y, Zhu C, Zhang K, Wang S, Tang X, et al. Effect of ballmilling on graphene reinforced Al6061 composite fabricated by semi-solidsintering. Compos Part B Eng 2014;60:111e8.

[11] Schultz BF, Ferguson JB, Rohatgi PK. Microstructure and hardness of Al2O3nanoparticle reinforced AleMg composites fabricated by reactive wettingand stir mixing. Mater Sci Eng A 2011;530:87e97.

ites Part B 77 (2015) 402e420[20] Kestursatya M, Kim J, Rohatgi P. Wear performance of copperegraphitecomposite and a leaded copper alloy. Mater Sci Eng A 2003;339(1):150e8.

pos[21] Liu Y, Rohatgi P, Ray S. Tribological characteristics of aluminum-50 vol pctgraphite composite. Metall Trans A 1993;24(1):151e9.

[22] Rohatgi P, Ray S, Liu Y. Tribological properties of metal matrix-graphiteparticle composites. Int Mater Rev 1992;37(1):129e52.

[23] Liu Y, Lim S, Ray S, Rohatgi P. Friction and wear of aluminium-graphitecomposites: the smearing process of graphite during sliding. Wear1992;159(2):201e5.

[24] Krishnan B, Raman N, Narayanaswamy K, Rohatgi P. Performance of anAleSi-graphite particle composite piston in a diesel engine. Wear1980;60(1):205e15.

[25] Badia F, Rohatgi P. Dispersion of graphite particles in aluminum castingsthrough injection of the melt. 1969.

[26] Ferguson J, Sheykh-Jaberi F, Kim C-S, Rohatgi PK, Cho K. On the strength andstrain to failure in particle-reinforced magnesium metal-matrix nano-composites (Mg MMNCs). Mater Sci Eng A 2012;558:193e204.

[27] Vogt R, Zhang Z, Li Y, Bonds M, Browning N, Lavernia E, et al. The absence ofthermal expansion mismatch strengthening in nanostructured metal-ematrix composites. Scr Mater 2009;61(11):1052e5.

[28] Bakshi S, Lahiri D, Agarwal A. Carbon nanotube reinforced metal matrixcomposites e a review. Int Mater Rev 2010;55(1):41e64.

[29] Deng CF, Wang DZ, Zhang XX, Li AB. Processing and properties of carbonnanotubes reinforced aluminum composites. Mater Sci Eng A2007;444(1e2):138.

[30] Rajkumar K, Aravindan S. Tribological behavior of microwave processedcopperenanographite composites. Tribol Int 2013;57:282.

[31] Wang J, Li Z, Fan G, Pan H, Chen Z, Zhang D. Reinforcement with graphenenanosheets in aluminum matrix composites. Scr Mater 2012;66(8):594e7.

[32] Ghazaly A, Seif B, SalemHG.Mechanical and tribological properties of AA2124-graphene self lubricating nanocomposite. Light Met 2013;2013:411e5.

[33] Kuzumaki T, Miyazawa K, Ichinose H, Ito K. Processing of carbon nanotubereinforced aluminum composite. J Mater Res 1998;13(9):2445.

[34] Kwon H, Park DH, Silvain JF, Kawasaki A. Investigation of carbon nanotubereinforced aluminum matrix composite materials. Compos Sci Technol2010;70(3):546e50.

[35] Bastwros MMH, Esawi AMK, Wi A. Friction and wear behavior of AleCNTcomposites. Wear 2013;307(1e2):164e73.

[36] liu B, Zeng Z, Lin Y. Mechanical properties of hard CreMWNT compositecoatings. Surf Coatings Technol 2009;203:3610e3.

[37] Tang Z, Lei Y, Guo B, Zhang L, Jia D. The use of rhodamine B-decoratedgraphene as a reinforcement in polyvinyl alcohol composites. Polymer2012;53:673e80.

[38] Wang Z, Nelson K, Miao J, Linhardt RJ, Schadler LS. Effect of high aspect ratioller on dielectric properties of polymer composites: a study on bariumtitanate bers and graphene platelets. IEEE Trans Dielectr Electr Insulation2012;19:960e7.

[39] Potts JR, Dreyer DR, Bielawski CW, Ruoff RS. Graphene-based polymernanocomposites. Polymer 2011;52:5e25. Jeffrey R. Potts.

[40] Shiu S-C, Tsai J-L. Characterizing thermal and mechanical properties of gra-phene/epoxy nanocomposites. Compos Part B Eng 2014;56:691e7.

[41] Lee JH, Marroquin J, Rhee KY, Park SJ, Hui D. Cryomilling application ofgraphene to improve material properties of graphene/chitosan nano-composites. Compos Part B Eng 2013;45(1):682e7.

[42] Ionita M, Pandele AM, Crica L, Pilan L. Improving the thermal and mechanicalproperties of polysulfone by incorporation of graphene oxide. Compos Part BEng 2014;59:133e9.

[43] Kim H, Miura Y, Macosko CW. Graphene/polyurethane nanocomposites forimproved gas barrier and electrical conductivity. Chem Mater 2010;22(11):3441e50.

[44] Moriche R, Prolongo S, Sanchez M, Jimenez-Suarez A, Sayagues M, Ure~na A.Morphological changes on graphene nanoplatelets induced during disper-sion into an epoxy resin by different methods. Compos Part B Eng 2014;72:199e205.

[45] Qin W, Vautard F, Drzal LT, Yu J. Mechanical and electrical properties ofcarbon ber composites with incorporation of graphene nanoplatelets at theberematrix interphase. Compos Part B Eng 2015;69:335e41.

[46] Park O-K, Kim S-G, You N-H, Ku B-C, Hui D, Lee JH. Synthesis and propertiesof iodo functionalized graphene oxide/polyimide nanocomposites. ComposPart B Eng 2014;56:365e71.

[47] Qiu F, Hao Y, Li X, Wang B, Wang M. Functionalized graphene sheets lledisotactic polypropylenenanocomposites. ComposPartBEng2015;71:175e83.

[48] Bartoluccia SF, Paras J, Raee MA, Raee J, Lee S, Kapoor D, et al. Graphe-neealuminum nanocomposites. Mater Sci Eng A 2011;528:7933e7.

[49] Chen L-Y, Konishi H, Fehrenbacher A, Ma C, Xu J-Q, Choi H, et al. Novelnanoprocessing route for bulk graphene nanoplatelets reinforced metalmatrix nanocomposites. Scr Mater 2012;67:29e32.

[50] Raee MA, Raee J, Wang Z, Song H, Yu Z-Z, Koratkar N. Enhanced me-chanical properties of nanocomposites at low graphene content. ACS Nano2009;3(12):3884e90.

[51] Stankovich S, Dikin DA, Dommett GH, Kohlhaas KM, Zimney EJ, Stach EA,et al. Graphene-based composite materials. Nature 2006;442(7100):282e6.

[52] Tjong SC. Recent progress in the development and properties of novel metalmatrix nanocomposites reinforced with carbon nanotubes and graphenenanosheets. Mater Sci Eng R 2013;74(10):281e350.

A. Dorri Moghadam et al. / Com[53] Xu C, Wang X, Zhu J. Grapheneemetal particle nanocomposites. J Phys ChemC 2008;112(50):19841e5.[54] Sengupta R, Bhattacharya M, Bandyopadhyay S, Bhowmick AK. A review onthe mechanical and electrical properties of graphite and modied graphitereinforced polymer composites. Prog Polym Sci 2011;36(5):638e70.

[55] Zhao X, Zhang Q, Chen D, Lu P. Enhanced mechanical properties of graphene-based poly (vinyl alcohol) composites. Macromolecules 2010;43(5):2357e63.

[56] Wang X, Hu Y, Song L, Yang H, Xing W, Lu H. In situ polymerization ofgraphene nanosheets and polyurethane with enhanced mechanical andthermal properties. J Mater Chem 2011;21(12):4222e7.

[57] Song P, Cao Z, Cai Y, Zhao L, Fang Z, Fu S. Fabrication of exfoliated graphene-based polypropylene nanocomposites with enhanced mechanical and ther-mal properties. Polymer 2011;52(18):4001e10.

[58] Robertson DH, Brenner DW, White CT. On the way to fullerenes: moleculardynamics study of the curling and closure of graphitic ribbons. J Phys Chem1992;96(15):6133.

[59] Yao N, Lordi V. Young's modulus of single-walled carbon nanotubes. J ApplPhys 1998;84(4):1939.

[60] Krishnan A, Dujardin E, Ebbesen TW, Yianilos PN, Treacy MMJ. Young'smodulus of single-walled nanotubes. Phys Rev B 1998;58(20):14013.

[61] Cumings J, Zettl A. Low-friction nanoscale linear bearing realized frommultiwall carbon nanotubes. Science 2000;289(5479):602e4.

[62] Ode Y, Nakajima M, Yang Z, Saito Y, Fukuda T. In-situ TEM investigation ofstepping motion of telescoping carbon nanotube. In: System integration (SII),2010 IEEE/SICE international symposium on: IEEE; 2010. p. 366e71.

[63] Berger C, Song ZM, Li T, li X, Ogbazghi A, Feng R, et al. J Phys Chem B2004;108.

[64] Berger C, Song ZM, Li XB, Wu XS, Brown N, Naud C, et al. Electronicconnement and coherence in patterned epitaxial graphene. Sience2006;312:1191e6.

[65] Reina A, Jia X, Ho J, Nezich D, Son H, Bulovic V, et al. Layer area, few-layergraphene lms on arbitrary substrates by chemical vapor deposition. NanoLett 2009;9(8):3087.

[66] Li X, Cai W, An J, Kim S, Nah J, Yang D, et al. Large-area synthesis of high-quality and uniform graphene lms on copper foils. Science2009;324(5932):1312.

[67] Li X, Cai W, Colombo L, Ruoff RS. Evolution of graphene growth on Ni and Cuby carbon isotope labeling. Nano Lett 2009;9(12):4268.

[68] Zhao L, Rim KT, Zhou H, He R, Heinz TF, Pinczuk A, et al. Inuence of coppercrystal surface on the CVD growth of large area monolayer graphene. SolidState Commun 2011;151(7):509.

[69] Tsai J-L, Tu J-F. Characterizing mechanical properties of graphite using mo-lecular dynamics simulation. Mater Des 2010;31(1):194.

[70] Zheng Q, Geng Y, Wang S, Li Z, Kim J-K. Effects of functional groups on themechanical and wrinkling properties of graphene sheets. Carbon2010;48(15):4315.

[71] Lee C, Wei X, Kysar JW, Hone J. Measurement of the elastic properties andintrinsic strength of monolayer graphene. Science 2008;321(5887):385.

[72] Lee C, Wei X, Li Q, Carpick R, Kysar JW, Hone J. Elastic and frictional prop-erties of graphene. Phys Status Solidi (B) 2009;246(11e12):2562.

[73] Balandin AA, Ghosh S, Bao W, Calizo I, Teweldebrhan D, Miao F, et al. Su-perior Thermal conductivity of single-layer graphene. Nano Lett 2008;8(3):902e7.

[74] Bolotin KI, Sikes KJ, Jiang Z, Klima M, Fudenberg G, Hone J, et al. Ultrahighelectron mobility in suspended graphene. Solid State Commun2008;146(9e10):351e5.

[75] Menezes PL, Rohatgi PK, Lovell MR. Self-lubricating behavior of graphitereinforced metal matrix composites. Green tribology. Springer; 2012.p. 445e80.

[76] Menezes PL, Reeves CJ, Rohatgi PK, Lovell MR. Self-lubricating behavior ofgraphite-reinforced composites. Tribology for scientists and engineers.Springer; 2013. p. 341e89.

[77] Akhlaghi F, Zare-Bidaki A. Inuence of graphite content on the dry slidingand oil impregnated sliding wear behavior of Al 2024egraphite compositesproduced by in situ powder metallurgy method. Wear 2009;266(1e2):37e45.

[78] Perez-Bustamante R, Estrada-Guel I, Antnez-Flores W, Miki-Yoshida M,Ferreira PJ, Martnez-Sanchez R. Novel Al-matrix nanocomposites reinforcedwith multi-walled carbon nanotubes. J Alloys Compd 2008;450(1e2):323e6.

[79] Cizaire L, Vacher B, Mogne TL, Martin J, Rapoport L, Margolin A, et al.Mechanisms of ultra-low friction by hollow inorganic fullerene-like MoS2nanoparticles. Surf Coatings Technol 2002;160:282e7.

[80] Jun Q, Linan A, Blau PJ. Sliding friction and wear characterristics of Al2O3 Alnanocomposites. In: STLE/ASME 2006 international joint tribology confer-ence. San Antonio, Texas, USA; 2006. p. 59e60.

[81] Nemati N, Khosroshahi R, Emamy M, Zolriasatein A. Investigation of micro-structure, hardness and wear properties of Ale4.5 wt% CueTiC nano-composites produced by mechanical milling. Mater Des 2011;32(7):3718e29.

[82] Sameezadeh M, Emamy M, Farhangi H. Effects of particulate reinforcementand heat treatment on the hardness and wear properties of AA 2024-MoSi2nanocomposites. Mater Des 2011;32(4):2157e64.

[83] Oh S-I, Lim J-Y, Kim Y-C, Yoon J, Kim G-H, Lee J, et al. Fabrication of carbonnanober reinforced aluminum alloy nanocomposites by a liquid process.J Alloys Compd 2012;542(25):111.

ites Part B 77 (2015) 402e420 419[84] Nuriel S, Liu L, Barber AH, Wagner HD. Direct measurement of multiwallnanotube surface tension. Chem Phys Lett 2005;404(4e6):263.

[85] So KP, Jeong JC, Park JG, Park HK, Choi YH, Noh DH, et al. SiC formation oncarbon nanotube surface for improving wettability with aluminum. ComposSci Technol 2013;74:6.

[86] Lim J-Y, Oh S-I, Kim Y-C, Jee K-K, Sung Y-M, Han JH. Effects of CNF dispersionon mechanical properties of CNF reinforced A7xxx nanocomposites. MaterSci Eng A 2012;556:337.

[87] Neubauer E, Kitzmantel M, Hulman M, Angerer P. Potential and challenges ofmetal-matrix-composites reinforced with carbon nanobers and carbonnanotubes. Compos Sci Technol 2010;70(16):2228.

[8