Functionalized Graphenes and Thermoplastic Nanocomposites Based upon Expanded Graphite Oxide


  • Peter Steurer,

    1. Freiburger Materialforschungszentrum and Institut für Makromolekulare Chemie of the Albert-Ludwigs-University Freiburg, Stefan-Meier-Strasse 31, D-79104 Freiburg, Germany
    Search for more papers by this author
  • Rainer Wissert,

    1. Freiburger Materialforschungszentrum and Institut für Makromolekulare Chemie of the Albert-Ludwigs-University Freiburg, Stefan-Meier-Strasse 31, D-79104 Freiburg, Germany
    Search for more papers by this author
  • Ralf Thomann,

    1. Freiburger Materialforschungszentrum and Institut für Makromolekulare Chemie of the Albert-Ludwigs-University Freiburg, Stefan-Meier-Strasse 31, D-79104 Freiburg, Germany
    Search for more papers by this author
  • Rolf Mülhaupt

    Corresponding author
    1. Freiburger Materialforschungszentrum and Institut für Makromolekulare Chemie of the Albert-Ludwigs-University Freiburg, Stefan-Meier-Strasse 31, D-79104 Freiburg, Germany
    2. Freiburg Institute for Advanced Studies (FRIAS), Albertstrasse 19, D-79104 Freiburg, Germany
    • Freiburg Institute for Advanced Studies (FRIAS), Albertstrasse 19, D-79104 Freiburg, Germany. Fax: +49-761-203-6319.
    Search for more papers by this author


Exfoliation of expanded GO represents an attractive route to functionalized graphenes as versatile 2D carbon nanomaterials and components of a wide variety of polymer nanocomposites. Thermally reduced graphite oxides (TrGO) with specific surface areas of 600 to 950 m2 · g−1 were obtained by oxidation of graphite followed by thermal expansion at 600 °C. Thermal post treatment at 700 °C and 1 000 °C increased carbon content (81 to 97 wt.-%) and lowered resistivity (1 600 to 50 Ω · cm). During melt extrusion with PC, iPP, SAN and PA6, exfoliation afforded uniformly dispersed graphenes with aspect ratio > 200. In comparison to conventional 0D and 1D carbon nanoparticles, TrGO afforded nanocomposites with improved stiffness and lower percolation threshold. Recent progress and new strategies in development of functionalized graphenes and graphene-based nanocomposites are highlighted.

original image


Natural and synthetic graphites are versatile, environmentally friendly and chemically inert carbon materials exhibiting attractive combinations of high thermal stability with high thermal and electrical conductivity. In the graphite crystal lattice the hexagonally arranged sp2-bonded carbon atoms of interconnected benzene rings form single atom thick graphene sheets which are planar and stacked parallel to each other.1 The in-plane covalent bonds between the carbon atoms of graphene sheets are much stronger with respect to the weak bonds between the carbon atoms of two adjacent graphene sheets. This multilayer architecture accounts for some unusual properties typical for graphite materials. Since the graphene sheets are readily shifted against each other upon shearing, especially under pressure and at high temperatures, crystalline graphites are excellent lubricants. Moreover, graphite is readily intercalated and can host various atoms, molecules, metal complexes and salts between its expanded graphene sheets. Expandable graphite intercalates are of interest in fire protection. Upon very rapid heating of the graphite hydrogensulfate intercalate, formed by intercalating graphite with sulfuric acid, the pyrolysis vaporization of the hydrogensulfate forces the graphene sheets to come apart. Expanded graphite has very high porosity and an accordion-like shape with an apparent volume which is more than hundred times larger than that of the original graphite. The high specific surface area of expanded graphite exceeds 100 m2 · g−1. Applications of graphite materials range from pencils, break linings, electrodes, battery separators, and bipolar plates of fuel cells to crucibles, refractory and fire-proof products which are being used extensively in metal-producing industry. Crystalline graphites are well-known commercial polymer additives used in polymer melt compounding in order to improve polymer properties such as thermal and electrical conductivity, IR absorption, flame-proofing, barrier resistance, electromagnetic shielding, lubrication and abrasion resistance. For many applications it would be highly desirable to exfoliate graphite and to disperse the individual graphene sheets within the polymer matrix.

Graphenes are 2D macromolecules with high specific surface area of 2 600 m2 · g−1 and attractive electrooptical properties.2, 3 Tiny amounts of graphene sheets are produced during graphite abrasion, for example when drawing a line with a pencil. In 2004 pure graphene was obtained successfully from graphite when peeling off single graphene sheets from the surface of graphite by means of an adhesive tape.4, 5 Today graphene (cf. Scheme 1), also referred to as exfoliated graphite, represents the new rising star at the horizon of materials science, condensed matter physics, and nanoelectronics.6 The discovery of the 2D graphene macromolecules has closed the gap in the family of carbon materials which has existed between 0 D carbon materials such as fullerenes, nano diamond, conducting carbon black nanoparticles, 1D carbon nanotubes, and 3D carbon materials such as diamond and graphite.7 As is evident from the graphene publication histogram displayed in Figure 1, the graphene community is expanding at an extraordinarily rapid pace. Since 2004 more than 4 300 papers on graphene have appeared. Graphenes are 2D crystals with very high stiffness, as expressed by their Young's modulus estimated to be around 1 000 GPa.8 As a consequence of graphene's high aspect ratios small amounts of graphenes of a few weight percent or less are sufficient to achieve percolation and network formation within a polymer matrix. The low percolation threshold of graphenes is of particular interest when aiming at improving the property profiles of polymer nanocomposites with respect to electrical conductivity,9 barrier resistance, electrooptics, polymer stiffness, abrasion resistance and fire retardency at very low carbon content. Since graphenes are optically transparent and electrically conducting, they can be applied to produce ultrathin films useful as transparent electrodes. Although the first successful route to graphene from graphite reported in 2004 has been very far from being industrially viable, significant progress has been made since then. Today graphenes and nanometer-scaled graphite platelets are considered to be attractive alternatives to the rather expensive carbon nanotubes.10 Since natural graphite is still abundant, it is highly cost effective with respect to the synthetic carbon nanotubes when functionalized graphenes are derived from graphite. Many approaches towards fabrication of functionalized graphenes and graphene nanocomposites are exploiting graphite oxide (GO) as versatile intermediate. Here we give a brief overview on different synthetic strategies leading to functionalized graphene nanoparticles and their thermoplastic nanocomposites based upon GO feed stocks. Special emphasis is placed upon the role of expanded GO, in particular thermally reduced graphite oxide (TrGO), in melt compounding of thermoplastics such as poly(propylene), polycarbonate, polyamide 6, and SAN. Morphology development, thermal, mechanical and electrical properties of melt extruded nanocomposites containing various amounts of oxidized graphenes, carbon black nanoparticles, and multiwall carbon nanotubes are compared.

Scheme 1.


Figure 1.

Histogram of the technical term “graphene” till 2007 (Science Finder data base accessed November 29, 2008).

Experimental Part


Polyamide 6 PA6 (Durethan B29, equation image = 20 000 g · mol−1) polycarbonate PC (Makrolon 3 108, equation image = 31 000 g · mol−1) were obtained as pellets from Lanxess AG and from Bayer AG. The poly(styrene-co-acrylonitrile) SAN (Luran VLP nature, equation image = 60 000 g · mol−1) is a commercial product obtained from BASF SE. Isotactic poly(propylene) iPP (HC101BF, equation image = 61 600 g · mol−1) was obtained from Borealis AG.

The graphite (Graphit KFL 99.5) was obtained from Kropfmühl AG, Passau, Germany. For comparison, multi wall carbon nanotube (MWCNT) from IoLiTec GmbH, Germany was employed to prepare benchmark nanocomposites. These MWCNTs were grown in vapor phase and their shell typically consisted of 8–15 graphene layers with a diameter of 10–30 nm and a length of 5–15 µm. They are obtained as agglomerates and exhibit curved intertwined entanglements. As second benchmark system nanocomposite were prepared using the conducting carbon black nanofiller (Carbon Black XE-2B) which was supplied by Evonik AG, Germany. All the other chemicals were purchased from VWR-Merck and used without further purification.

Preparation TrGO

TrGO was prepared in a two step oxidation/thermal reduction process using natural graphite (Kropfmühl AG, type KFL 99.5) as raw material. The graphite oxidation process of Hummers and Offeman was employed. For comparison GO was also prepared using the process reported by Brodie.

Oxidation with KMnO4/NaNO3 (Hummers and Offeman Process)26

The first step is an oxidation of graphite with KMnO4 and NaNO3 in concentrated sulfuric acid. This oxidation was carried out using 250 ml of concentrated sulfuric acid per 10 g of graphite as dispersion medium. To the stirred dispersion 5 g NaNO3 were added and after 1 hour of stirring it was cooled to 0 °C using an ice-water bath. Then 30 g of KMnO4 were added during 5 h. When the addition was completed, the resulting dispersion was stirred at room temperature for 2 h. The reaction was quenched by pouring the dispersion into 0.5 l of ice water and adding of some milliliters of H2O2 (5 wt.-%) until the excess KMnO4 was destroyed. The GO was filtered off and washed with aqueous HCl until no precipitation of BaSO4 occurred in the presence of aqueous BaCl2 solution. The GO was washed with water until the chloride test with AgNO3 was negative. The purified brown GO was dried by lyophilization to afford GO with the empirical formula C6O3H2. In the second step the dry GO (GO, 1.5 g) was thermally reduced to afford TrGO in a nitrogen atmosphere by rapidly heating GO up to approximately 600 °C using a metallic reactor heated with a gas burner. TrGO was obtained as a black powder of very low bulk density. The super heating is the prime requirement to achieve exfoliation of graphene sheets.

Oxidation with NaClO3 (Brodie Method)23

10 g of graphite were mixed with 85 g of fine NaClO3. In a round-bottom flask cooled with a mixture of ice and salt this mixture is cooled under vigorous stirring. Then HNO3 (100%) was added very slowly over 5 h. After additional stirring at room temperature for 30 min the flask was stored at room temperature over night. The next morning the green mass was heated for 10 h to 60 °C after reducing the agglomerates to small pieces. After cooling down the crude product was stirred in 2 l of water. After filtration and freeze drying the oxidation was repeated and bright, slightly yellow material was obtained. Thermal expansion was carried out as described above. The dispersion resulting from GO sonification in water is displayed in Figure 2.

Figure 2.

GO dispersion in water.

Melt Compounding

The polymers and their corresponding nanocomposites were processed using a DSM twin screw minicompounder (Xplore™) with a mixing chamber volume of 5 ml. All the samples were compounded at 100 rpm for three minutes. The melt mixing temperature was 210°C for SAN VLP and iPP HC101BF, 250 °C for PA6 B29 and 280 °C for PC Makrolon 3108, as measured by the thermocouple of the minicompounder. The MWCNT and conducting carbon black fillers were mixed and compounded together with the respective polymer. Tests using an ultrasonic lancet to deagglomerate the MWCNTs dispersed in organic solvents (acetone for SAN, CH2Cl2 for PC and xylene for iPP) containing the dissolved polymer did not reveal any positive effect. Because of the low bulk density of TrGO (see Figure 3), solution blends of TrGO with the polymer were prepared to obtain homogenous polymer-filler compound. These premixed compounds were first dried and then melt compounded in the minicompounder. The following solvents were used: acetone for SAN/TrGO, dichloromethane for PC/TrGO, both at room temperature and boiling xylene for iPP/TrGO. In the case of PA6, the filler was dispersed in acetone and then polyamide powder was added. Acetone was stripped off in vacuum leaving the polymer powder coated with TrGO. Heat treatment of TrGO was performed in a tube furnace type Heraeus RoF 4/50.

Figure 3.

GO (75 mg in 100 ml flask) before (left) and after (right) flash heating at 600 °C.

Measurement of Electrical Resistivity

The resistivity of fillers was measured by means of pressed discs with a diameter of 1.3 cm and a thickness of 0.5 to 1.0 mm. The pressure during the compression molding was 75 MPa for 5 min. The pressing was contacted in a distance of 1 cm with the measuring electrodes of a multimeter type VL 960 (Voltcraft, Hirschau). The specific resistivity was evaluated from the measured resistivity value using following equation where A is the cross section area and L the distance between the electrodes:

equation image((1))

The conductivity of the composites was measured on compression molded test specimens with DMA geometry meaning 5 cm × 0.2 cm × 0.6 cm. These test specimens were contacted with silver paste and measured by a Keithly electrometer 617 resulting in specific resistances up to 2 × 1010 Ω · cm. Some test specimens have been examined using a teraohmmeter Dr. Thiedig, Dipato 4 “Precision Pico-Ampere/Tera-Ohm-Meter” by BASF SE with a measuring range up to 1013 Ω · cm.

Transmission Electron Microscopy

The morphology of the nanocomposites was determined by transmission electron microscopy (TEM). For TEM measurements, ultra thin sections were prepared at room temperature (SAN and PC) using a Ultracut E ultramicrotome by Reichert and Jung and at −140°C (PA6 and PP) with a Leica Ultracut UCT using a diamond knife in each case. Measurements were carried out on a LEO 912 Omega (120 kV).

Environmental Scanning Electron Microscopy

Scanning electron microscopic imaging was performed with an environmental scanning electron microscope (ESEM 2020 from Electroscan Corp., Wilmington, USA). Imaging of the samples in dry state was conducted in a water vapor (5 Torr) with a voltage of 23 kV (LaB6-cathode). The samples were sputter-coated with gold-palladium (Pollaron Sputter Coater SC 7640, layer thickness ≈30 nm) before imaging. The detection of the secondary-electrons was occurred with a GSED (Gaseous secondary electron detector).

Results and Discussion

Routes to Functionalized Graphenes

For applications in polymer melt processing it is highly desirable to form and to disperse graphenes during processing of polymer nanocomposites in order to avoid potential safety and handling hazards associated with emissions of nanometer-scaled molecules with high modulus and high aspect ratio. Most of the numerous early attempts to obtain graphene dispersions directly from graphite, respectively, afforded only micro- and nanometer-scaled graphite platelets composed of fairly large stacks of graphene sheets. It should be noted that most properties typical for the 2D graphenes are lost when much more than 10 graphene sheets aggregate to produce graphite-like 3D architectures. As illustrated in Scheme 2, three strategies have been developed successfully to prepare graphene nanoparticles: (A) direct synthesis of graphenes from various hydrocarbon precursors, (B) formation of aqueous and non-aqueous functionalized graphene dispersions from GO using chemical reduction and optionally functionalization reactions, and (C) exfoliation of expanded and thermally reduced TrGO. In routes (B) and (C) GO is the raw material of choice because oxidation prevents graphene stacking and affords easy dispersion of functionalized graphenes in both aqueous and organic media. Functional groups of graphene are essential for promoting both graphene dispersion and interfacial adhesion of graphenes to the polymer matrix. Moreover, GOs are readily functionalized and grafted with polymers. Stable graphene dispersions are economically feasible graphene sources for the preparation of graphene nanofillers, functionalized graphene dispersions, and graphene based nanocomposites. Although the industrial graphene manufacturing is still at its infancy in the early bench scale and pilot plant stages, the recent progress is quite remarkable and is likely to render functionalized graphene highly cost and performance competitive especially with respect to carbon nanotubes.

Scheme 2.

Routes to functionalized graphenes: (A) Organic synthesis using precursors and nano templates, (B) chemical reduction and functionalization of GO exfoliated in aqueous dispersion, and (C) thermally reduced expanded GO (TrGO).

In his bottom-up approach (route A in Scheme 2) towards giant graphenes and molecular devices Müllen's group has developed multi-step organic syntheses, sophisticated graphene precursor chemistry, and nano templating methods for tailoring an impressive variety of 2D carbon macromolecules.2, 11, 12 In polymer processing application top down strategies towards functionalized graphenes (routes B and C in Scheme 2) are employed exploiting GO as preferred raw material. GO (cf. Scheme 3) is readily obtained by intercalation of graphite with sulfuric acid and subsequent oxidation using various oxidizing agents such as potassium permanganate, sodium chlorate, nitric acid, and others. GO contains epoxy and hydroxy groups within the graphene sheets and also carboxylic acid and ketone groups at the graphene edges (cf. Scheme 3). These sp3-bonded carbon atoms allocated in the graphene sheets disrupt very effectively the double bond conjugation and account for the waved sheet structures, resembling wrinkled silk, which is typical for GO.13 In contrast to the black graphite, graphite oxide (cf. Figure 2) has brownish (Hummers) or slightly yellow (Brodie) color. In 2006 Ruoff and coworkers have discovered that simple sonification of GO in water is sufficient to afford stable aqueous dispersions of functionalized graphenes.14, 15 Most likely ionic groups on the graphene surface account for electrostatic stabilization of these colloids. Characterization of colloid microstructures revealed the presence of nano platelets, among them also stacks of graphenes and larger aggregates.16 This dispersion approach is the key for many new applications of graphenes and manufacturing of easy to disperse functionalized graphenes. Upon chemical or thermal reduction graphite is partially recovered because carbon dioxide, carbon monoxide, and water are split off, thus removing some of the disruptive sites in the graphite sheets. As a function of the reduction processes it is possible to vary the degree of graphene functionalization. This GO based graphene dispersion technology has been the enabling technology for many modern graphene applications. For example, in a recent advance the graphene dispersion route to functionalized organosoluble graphenes has been applied successfully by Müllen's group in spin coating to produce ultrathin graphene films as transparent electrodes for solid-state dye-sensitized solar cells, aiming at substituting expensive ITO glass electrodes.17

Scheme 3.

Graphite oxide.

In the presence of dispersing agents such as polymers, surfactants or other nanoparticles, the uncontrolled aggregation of functionalized graphene during chemical reduction can be prevented, for example when adding polyelectrolytes.15 Functionalization of graphenes affords organophilic graphene modification required to disperse functionalized graphenes in non-polar medium. For example, the reaction with isocyanates converted the hydroxyl groups into urethanes and carboxylic acids into amides. This chemical graphene modification has enabled dispersion of such functionalized graphenes in non-aqueous media.18 Molecular dynamics simulations for functionalized graphenes containing 54 and 96 carbon atoms show that graphene sheets functionalized at their edges with short branched alkanes afford stable dispersion in oils.19 A variety of functionalized graphenes were produced following this dispersion strategy. Graphene dispersions are of interest in the development of coatings, ultrathin films as well as thermoset and polyurethane nanocomposites.

Since the direct injection of aqueous functionalized graphene dispersion is restricted to polymer solutions in polar solvents, water soluble polymers, and polymer melts of polar polymers such as polyamides, the development of modified graphene dispersion technology is in progress to render functionalized graphenes organophilic and to produce master batches, e.g., by blending together polymer latex, waxes, or water soluble polymers with graphene dispersions followed by chemical or thermal reduction, respectively. In a solvent-free process (route C in Scheme 2) TrGO containing partially exfoliated functionalized graphene assemblies is produced as micrometer-scaled filler. TrGO deagglomerates during melt compounding to produce functionalized graphenes dispersed in polymer melts. This process is solvent free. Key step is the partial reduction accompanied by thermal expansion. Only when the temperature increases extremely rapidly, the instant gas evolution will force the graphene sheets to exfoliate. The study of the thermal expansion mechanism of GO has revealed that exfoliation takes place when the decomposition rate of the epoxy and hydroxyl sites of GO exceeds the diffusion rate of the evolved gases, thus yielding pressures that exceed the van der Waals interactions holding the graphene sheets together.20 The TrGO surface functionalities and polarities vary with temperature. As a rule carbon content increases with increasing temperature. The groups of Aksay and Prud'homme at Princeton University employed super rapid heating of GO at 1 050 °C with heating rates of 2 000 °C · min−1 t in an argon atmosphere to achieve exfoliation and formation of single functionalized graphenes.21 The in situ surface modification is an attractive possibility to improve simultaneously graphene dispersion and graphene/polymer interfacial adhesion, thus eliminating the need for addition of special polymeric compatibilizers. Good interfacial adhesion is important to prevent pull-out of graphenes during straining which would impair energy dissipation at the crack tip and cause premature mechanical failure upon exposure of such nanocomposites to external stresses.

Thermally Reduced Graphite Oxide (TrGO)

The process for producing thermally reduced expanded GO (TrGO) has two steps and exploits technology established in commercial scale for expanded graphite: (i) graphite intercalation with sulfuric acid and oxidization to produce GO; (ii) chemical or preferably thermal reduction partially converting GO in functionalized graphenes. The oxygen content of TrGO was lowered and the carbon content increased by thermal post treatment at 700 °C and 1 000°C, respectively, under nitrogen atmosphere. Table 1 summarizes the properties of the graphite materials prepared in this study. Bench scale oxidation of graphite and formation graphite oxide (GO) is well known and dates back to pioneering advances of Brodie in 1859 22–25. Typically either potassium permanganate together with sodium nitrate (Hummers) or sodium chlorate (Brodie) was employed as oxidizing agent (for details see Experimental Part). The resulting brownish GO (Hummers method) contained around 40 wt.-% oxygen and did not have any measurable conductivity because the conjugated π-system of the polycyclic aromatic system was severely disrupted when incorporating the sp3 carbon atoms of epoxy and tertiary alcohols. According to the literature the specific resistance of GO is above 1010 Ω · cm.26 The presence of epoxy and hydroxy groups was confirmed by means of 13C NMR spectroscopic analysis and was in accord with the data reported in the literature.27, 28

Table 1. Composition and properties of graphite, GO and TrGO (H: Hummers methods, B: Brodie methods for preparing GO).
Materiala)Content of CContent of HResistivity δBETMB-Absorption
wt.-%wt.-%Ω · cmm2 · g−1m2 · g−1
  • a)

    H: Hummer's GO route, B: Brodie's GO route, TrGO-700: thermal post treatment at 700 °C, TrGO-1000: thermal post treatment at 1 000 °C.

Graphite 99.5%99.760.011≪100≪100
GO (H)57.921.9>1010≪100≪100
TrGO (H)81.000.51 6006001 000
TrGO-700 (H)90.880.71006001 000
TrGO-1000 (H)95.230.3506001 000
GO (B)61.101.0>1010≪100≪100
TrGO (B)84.200.68309501 200
TrGO-700 (B)91.900.75009501 200
TrGO-1000 (B)97.100.4659501 200

In the second TrGO process step, the reduction of GO was achieved either chemically or thermally. Thermal reduction affords TrGO which is also referred to as functionalized graphene sheets (FGS). Several groups have pioneered this approach.29, 30, 20, 21 The groups of Aksay and Prud'homme have investigated the formation of functionalized graphenes and nanocomposites based upon expanded graphite oxide.20, 21 Chemical reduction employs reducing agents such as hydrazine, Vitamin C and others. The thermal reduction by means of very rapid heating to temperatures of 600 and 1 000 °C afforded TrGO. Depending on the pyrolysis temperature, TrGO can contain a significant amount of oxygen (up to 20%, cf. Table 1) and has completely different appearance and properties with respect to GO. With respect to their specific surface area, color and structure both TrGO and chemically reduced GOs resemble CNTs. The TrGO surface area was measured to be 800 to 1 000 m2 · g−1 by means of BET (Brunauer, Emmett, Teller) test31 and 1 000 to 1 500 m2 · g−1 according to the methylene blue absorption (MB) method. The MB measurement is a solution method in which the graphene sheets are solvated. This in accord with earlier observations by Boehm et al. who reported that TrGO had a specific surface area up to 1 500 m2 · g−1, which is very similar to that of DWCNTs and activated charcoal. Taking into account the specific graphene surface area of around 2 600 m2 · g−1, the degree of exfoliation can be estimated to be around 50%.30 In this thermal reduction process the decomposition of functional groups such as the epoxy-group can account for structural defects impairing planarity of the functional graphene sheets.20 However, this reduction process is thermodynamically favored because of partial recovery of the polycyclic aromatic ring system. In an alternative process, TrGO with high C-content was obtained by thermal post treatment at 700 °C and 1 000 °C for the duration of 8 h. It was possible to increase the carbon content from 84.2 wt.-% to 97.1 wt.-% (cf. Table 1).

As is apparent from the ESEM images displayed in Figure 4, both the unmodified natural graphite and GO have very similar layered structures. Dramatic volume expansion and high porosity with accordion-like morphology is typical only for TrGO. This accounts for the very low TrGO bulk densities which cause feed problems in melt compounding and require formation of master batches. This problem was solved by solution blending TrGO with the polymers. Alternatively, also coating of polymer powder such as polyamide proved to be a feasible mode of achieving effective mixing. After expansion, the specific resistivity dropped to 1 600 Ω · cm. Upon thermal post treatment at 1 000 °C in a tube furnace under nitrogen for 8 h it was possible to reduce the specific resistivity to 50 Ω · cm. As is apparent from Figure 5 the specific resistivity decreased with increasing carbon content. As proven by extensive ESEM and BET analyses, this thermal post treatment did not impair the very high specific surface area and the morphology of TrGO (cf. Table 3).

Figure 4.

View of A) natural graphite (scale bar: 10 µm), B) GO (scale bar: 10 µm) and C) TrGO (scale bar: 15 µm) imaged by means of ESEM.

Figure 5.

Specific resistivity of TrGO as a function of the TrGO carbon content. The carbon content was increased by thermal post treatment of TrGO at 700 and 1000 °C, respectively (cf. Table 2).

Melt Compounding of Nanocomposites with Various Carbon Nanofillers

Among the large families of fillers, layered materials such as organoclay and graphite are attracting considerable attention because micrometer-scaled layers of such fillers can be exfoliated upon shearing in an extruder to disperse nanometer-scaled platelets with very high aspect ratio. Such nanoplatelets can improve stiffness/toughness balance, heat distortion temperature, flame proofing, barrier resistance, electrical conductivity, abrasion resistance.32 Progress in graphite exfoliation33–35 and formation of graphene-based nanocomposites35–38 has been reviewed. The group of Drzal has developed processing technology to convert expanded graphite into graphite nanoplatelets (10 × 1 000 nm) containing stacks of around ten graphenes. They discovered that aspect ratio, concentration of the graphite nanoplatelets and poly(propylene) crystallization conditions can be tuned to control poly(propylene) crystallization and electrical conductivity.39 Several groups have employed expanded GO to prepare thermoplastic nanocomposites based upon polyethylene,40 maleic-anhydride grafted poly(propylene),41 polystyrene,42 and ethylene/methyl acrylate/acrylic acid copolymers.43 In comparison to conventional GO, the expanded GO with its much higher degree of exfoliation and high specific surface area gave improved stiffness and electrical properties at lower GO content. Macosko and coworkers employed functional graphenes from expanded GO in melt processing of poly(ethylene-2,6-naphthalate).44 Although the functionalized graphenes were dispersed effectively within the polyester matrix, the additional matrix reinforcement was lower with respect to that of commercial graphite filler. This was attributed to the presence of wrinkled graphene structures resulting from defects in the functionalized graphene sheets.

In this study isotactic poly(propylene) (iPP), poly(styrene-co-acrylonitrile) (SAN), polyamide 6 (PA6) and polycarbonate (PC) were melt compounded in a twin-screw miniextruder together with TrGO. As bench marks also conventional nanofillers such as conducting carbon black nanoparticles (CB) and multi wall carbon nanotubes (MWCNT) were included. The nanofiller content was varied between 0 and 12 wt.-%. Since the TrGO had very low bulk density (Figure 3), solution blending (SAN, PC, iPP) and TrGO coating of PA6 powder were employed to premix TrGO with the polymers and to eliminate potential hazards associated with TrGO fine dust emission. The nanocomposite compositions, their mechanical properties and resistivities are summarized in Table 2. As is apparent from Figure 6 and 7, displaying TEM images of the PC, SAN, PA6, iPP nanocomposites containing 7.5 wt.-% TrGO, all TrGO fillers are very effectively exfoliated and uniformly dispersed in the polymer matrix. The addition of dispersing aids and polymer compatibilizers was not required. This was quite surprising because PC, SAN, iPP and PA6 had different polarities, different melt viscosities, and very different processing temperatures. Although the TEM thin sections do not allow reconstruction of the 3D architectures it is clearly visible that single graphene sheets together with ultrathin stracks of graphene sheets are obtained. Large graphite platelets were not detected. The in-situ exfoliated functionalized graphenes have very high aspect ratio > 200 and appear to form network structures at 7.5 wt.-% content. In contrast, the commercial MWCNT was much more difficult to disperse and failed to give uniform dispersions when using the same processing conditions.

Table 2. Thermoplastic nanocomposites of SAN, PC, PA-6 and iPP containing TrGO, MWCT and conducting carbon black (CB).
SampleYoung's modulusElongation at breakElectrical resistivity
MPa%Ω · cm
SAN Luran VLP2 350 ± 305.2 ± 0.61.0 × 1016
SAN/4% TrGO2.7 × 109
SAN/5% TrGO3 160 ± 402.2 ± 0.23.7 × 105
SAN/6.25% TrGO1.0 × 105
SAN/7.5% TrGO1.3 × 104
SAN/9% TrGO7.5 × 103
SAN/10% TrGO4 120 ± 401.1 ± 0.1not determined
SAN/12% TrGO8.2 × 102
SAN/4% CB6.5 × 1012
SAN/5% CB2 510 ± 202.4 ± 0.33.8 × 1011
SAN/7.5% CB2.0 × 104
SAN/8.25% CB3.4 × 102
SAN/9% CB2.0 × 101
SAN/10% CB2 670 ± 102.2 ± 0.1not determined
SAN/12% CB2 670 ± 102.2 ± 0.11.1 × 101
SAN/5% MWCNT2 500 ± 102.7 ± 0.2out of range
SAN/10% MWCNT2 600 ± 302.2 ± 0.3out of range
SAN/12% MWCNT1.4 × 105
PC Makrolon 31081 480 ± 10104.4 ± 6.91.0 × 1018
PC/2.5% TrGO1.3 × 107
PC/5% TrGO2 250 ± 702.4 ± 0.22.1 × 105
PC/7.5% TrGO2.2 × 104
PC/10% TrGO2 670 ± 1101.5 ± 0.13.2 × 103
PC/12% TrGO1.0 × 103
PC/2.5% CB2.8 × 1010
PC/5% CB1 640 ± 2024.5 ± 6.08.4 × 101
PC/10% CB1 740 ± 309.3 ± 3.9not determined
PC/12% CB1.1 × 101
PC/2.5% MWCNT2.5 × 108
PC/5% MWCNT1 590 ± 2014.7 ± 4.41.1 × 105
PC/7.5% MWCNT8.0 × 102
PC/10% MWCNT1 670 ± 4010.3 ± 0.1not determined
PC/12% MWCNT1.8 × 102
PP HC101BF980 ± 50572.9 ± 48.01.0 × 1018
PP/5% TrGOH1 400 ± 907.3 ± 2.14.0 × 103
PP/10% TrGOH1 500 ± 807.8 ± 1.85.3 × 103
PP/5% CB1 340 ± 4010.8 ± 3.43.8 × 1011
PP/10% CB1 410 ± 307.1 ± 1.13.0 × 101
PP/5% MWCNT1 180 ± 4020.3 ± 6.48.9 × 102
PP/10% MWCNT1 320 ± 806.8 ± 2.33.2 × 101
PA6 Durethan B291 650 ± 60128.7 ± 83.41.0 × 1016
PA6/5% TrGOH2 180 ± 107.5 ± 0.6out of range
PA6/7.5% TrGOH4.5 × 109
PA6/10% TrGOH2 430 ± 104.3 ± 0.54.6 × 104
PA6/12% TrGOH1.4 × 104
PA6/5% CB1 890 ± 10047.0 ± 25.4out of range
PA6/7.5% CB4.5 × 109
PA6/10% CB1 990 ± 104.6 ± 0.44.6 × 104
PA6/12% CB1.5 × 104
Figure 6.

TEM images of SAN (above) and PC (below) nanocomposites containing 7.5 wt.-% TrGO with two different magnifications; 1 µm (left) and 200 nm (right).

Figure 7.

TEM images of iPP (above) and PA6 (below) nanocomposites containing 7.5 wt.-% TrGO with two different magnifications of 1 µm (left) and 200 nm (right).

The resistivity of the nanocomposites containing TrGO was measured and compared to those of nanocomposites containing MWCNT and CB (cf. Table 3). The resistivity measurement range in our experimental set-up was limited to values below 2 × 1010 Ω · cm. Figure 8 shows the percolation thresholds, as measured by resistivity changes. The percolation thresholds are summarized in Table 3. The thresholds measured by rheological measurements were somewhat lower and will be reported in more detail elsewhere. As pointed out by Macosko,44 the threshold value depends upon sample geometry and sample preparation. All nanocomposite test specimen were prepared using identical conditions (see Experimental Part). In PC (1018 Ω · cm), the resistivity of 1.3 × 107 Ω · cm was measured at 2.5 wt.-% TrGO content. The thresholds for PC nanocomposites containing TrGO, CB and MWCNT were very similar. However, CB and MWCNT afforded higher specific resistivity. The additional thermal treatment of TrGO at 1 000 °C for 8 h reduced specific resistivity to 50 Ω · cm which is comparable to that of compressed CB and natural graphite. Incorporation of 12 wt.-% TrGO into SAN afforded the SAN/TrGO nanocomposite with a specific resistivity of 50 Ω cm, being the same as that of the pure filler. In SAN (1016 Ω · cm), the electrical percolation of graphene nanosheets was observed at a TrGO content of 4 wt.-% with resistivity of 2.7 × 109 Ω · cm. At the same content of CB the resistivity was much higher (1013 Ω · cm). Addition of MWCNT to SAN did not lead to a detectable decay of resistivity until 12 wt-% filler content when using the same processing conditions. In the case of iPP (1018 Ω · cm) the threshold iPP/TrGO nanocomposites was < 5 wt.-% TrGO. Since the TrGO based graphenes are not aligned during melt extrusion and still contain considerable amounts of defects, the conductivities of polymeric TrGO nanocomposites is orders of magnitude lower with respect to that achieved when planar graphenes are stacked in highly ordered ultrathin coatings.17 However, the obtained resisitivities are sufficient to achieve electromagnetic shielding for all four polymers.

Table 3. Percolation threshold of nanocomposites.
PolymerPercolation threshold from electrical resistivity [wt.-% TrGO]
Figure 8.

Specific resistivity of SAN, PC, iPP and PA6 nanocomposites as a function of the TrGO content.

While the addition of MWCNT and CB afforded only marginally improved Young's modulus, the addition of TrGO increased significantly the Young's modulus of SAN and PC (cf. Figure 9 and Table 2). This enhanced stiffness was achieved at the expense of elongation at break, which decreased with increasing TrGO content. More research is needed to examine the role of the interfacial coupling and micromechanics.

Figure 9.

Young's modulus as a function of nanofiller content for PC and SAN nanocomposites.


The development of 2D carbon materials derived from natural graphite offers attractive opportunities for the development of polymer nanocomposites with improved property profiles such as high thermal and electrical conductivity, improved heat distortion temperatures, IR absorption, gas barrier performance, electromagnetic shielding, fire protection, antistatics, and matrix reinforcement. Most of the 0D and 1D carbon materials such as conducting carbon black nanoparticles, carbon nanotubes, and nano diamonds, require special precautions due to potential hazards associated with handling and inhalation of nanoparticles. In contrast, many micrometer-sized layered materials such as organoclays and modified graphites can be exfoliated during processing, producing nanometer-scaled platelets when exposed to shear. Significant progress has been made in graphite chemistry and nanocomposite development. Today GO is widely recognized as attractive intermediate because oxidation disrupts the aromatic graphene structure, thus promoting exfoliation, dispersion, and functionalization of graphenes. Exfoliation can be achieved in aqueous medium and in polymer melts. Since the pioneering advances in 2004, the aqueous and non-aqueous GO- based graphene dispersion chemistry is progressing rapidly. Such dispersions can be used to produce and functionalize graphenes, including grafting them with polymers. Most approaches employ chemical reduction to restore part of the polycyclic aromatic rings. Thermal reduction is achieved in the case of expanded GO, produced by partial pyrolysis of GO in a process similar to that of expanded graphite. The presence of functional groups promotes dispersion and interfacial adhesion in many polymer matrices. Single oxidized graphene sheets are obtained in polymers of greatly different polarity when TrGO is melt compounded. The carbon content of TrGO can be varied from 80 to 97% by thermal post treatment without sacrificing high specific surface area and easy exfoliation. The electrical conductivity of both TrGO and the corresponding TrGO nanocomposites increases with increasing carbon content. More insight is needed to better understand the graphite oxidation process and the influence of graphite functionalization on polymer properties. More effective TrGO processes are needed in view of cost competitiveness with other nanofillers. The in situ functionalization during oxidation is the key to control self-assembly and graphene superstructure formation in polymer melts. The assembly of functionalized graphenes at polymer interfaces will be an important challenge in the development of new polymer blends and polymer composite materials.


The authors would like to thank BASF SE for technical assistance and many helpful advices and discussions concerning the resistivity measurements, and Kropfmühl AG for contribution of various graphite materials.

Biographical Information

Rolf Mülhaupt, born in 1954, studied chemistry at the University of Freiburg in Germany (1973–1978) and got his PhD under the supervision of Prof. P. Pino in 1981 at the industrial and engineering laboratory of ETH Zürich in Switzerland. After industrial polymer research at Du Pont Central Research in Wilmington, Delaware, USA (1981–1985) and Ciba Plastics and Additive Division in Marly, Switzerland (1985–1989), he was appointed full professor of macromolecular chemistry and director of the Institute of Macromolecular Chemistry at the University of Freiburg in 1989. Since 1992 he is the managing director of the Freiburg Materials Research Center (FMF) and since 2000 member of the Heidelberg Academy of Sciences. His research is concerned with polymerization catalysis and polymer synthesis, engineering plastics, nanoparticles, nanocomposites, functional processing, and specialty polymers. He is member of the Executive Advisory Board of Wiley's macromolecular journals.

original image

Biographical Information

Rainer Wissert, born in 1978, studied chemistry at the University of Freiburg in Germany (1999–2005). During his study he puts emphasis on macromolecular chemistry. Since 2005 he works as PhD under the supervision of Prof. Rolf Mülhaupt at the Institute of Macromolecular Chemistry at the University of Freiburg. His focus of research is the preparation of polymer nanocomposites based on matrix reinforcing fillers and conductive carbon fillers. At the Freiburg Materials Research Center (FMF), he is responsible for surface characterization of inorganic and polymeric materials by environmental scanning electron microscopy (ESEM).

original image

Biographical Information

Peter Steurer, born in 1978, studied chemistry at the University of Freiburg in Germany (2000–2003 and 2004–2006) and at the ENSC of Montpellier in France (2003–2004). Since 2006 he works as PhD student under the supervision of Prof. Rolf Mülhaupt at the Institute of Macromolecular Chemistry at the University of Freiburg. His focus of research is the preparation of graphene based carbon fillers, their characterization and incorporation into polymers.

original image

Biographical Information

Ralf Thomann, born in 1963, studied chemistry at the University of Freiburg in Germany and got his PhD under the supervision of Prof. R. Mülhaupt at the Institute of Macromolecular Chemistry at the University of Freiburg. Since 1997 he is working as a research microscopist specialized on electron- and atomic force microscopy.

original image