Low Temperature Reduction of Graphene Oxide Using Hot-plate for Nanocomposites Applications

Accepted Manuscript Title: Low Temperature Reduction of Graphene Oxide Using Hot-Plate for Nanocomposites Applications Author: Abdelrahman Hussein, Sourav Sarkar, Byungki Kim PII: DOI: Reference:

S1005-0302(16)00029-3 http://dx.doi.org/doi: 10.1016/j.jmst.2016.02.001 JMST 650

To appear in:

Journal of Materials Science & Technology

Received date: Revised date: Accepted date:

24-11-2015 14-1-2016 27-1-2016

Please cite this article as: Abdelrahman Hussein, Sourav Sarkar, Byungki Kim, Low Temperature Reduction of Graphene Oxide Using Hot-Plate for Nanocomposites Applications, Journal of Materials Science & Technology (2016), http://dx.doi.org/doi: 10.1016/j.jmst.2016.02.001. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Low Temperature Reduction of Graphene Oxide Using Hotplate for Nanocomposites Applications Abdelrahman Hussein, Sourav Sarkar and Byungki Kim* School of Mechatronics Engineering, Korea University of Technology and Education 1600 Chungjeol-ro, Byeongchun-myeon, Cheonan, Chungnam, 31253, Republic of Korea [Received 24 November 2015; Received in revised form 14 January 2016; Accepted 27 January 2016] *Correspondent author: Tel.: +82 1041093173; E-mail: [email protected] (Byungki Kim)

A green, easy to reproduce method to obtain thermally reduced graphene oxide (GO) is described. Only requirement is a heating source, like a hot plate, that can reach ~225 °C without any special setup requirements. Upon addition of graphene oxide, effective reduction could be achieved within 10 s. Starting flake size affects the yield of graphene, final structure and composition. A detailed characterization of the produced graphene using thermal analysis, spectroscopic methods, electron microscopy, X-ray diffraction and atomic force microscopy is presented. Application of the produced graphene as a filler to epoxy resin for mechanical reinforcement is also reported. Smaller flakes (D50 = 5.7 μm) showed improved ultimate tensile strength, fracture strain and plane strain fracture toughness compared to larger flakes (D50 = 47.9 μm) that showed negative effect. Both flake sizes showed a negligible effect on Young’s modulus. Keywords: Graphene; Nanocomposite; Mechanical properties; Fracture toughness

1.

Introduction 1 Page 1 of 21

Due to its unique properties, researchers are spending efforts to develop applications of graphene, thanks to its planar sp2 hybridized structure[1]. Applications of graphene includes, but not limited to, transistors[2], transparent electrodes[3], gas sensors[4] and electromagnetic shielding[5,6]. In addition to standalone applications, graphene was successfully applied as a filler for different polymers to enhance mechanical[7], thermal[8], barrier[9] and electrical properties[10]. Graphene could be successfully prepared by several methods such as mechanical cleavage and chemical vapor deposition[11]. For graphene based nanocomposites, chemically derived graphene is the preferred method due to its high yield[12]. Synthesis of chemically derived graphene starts by oxidizing graphite flakes into graphite oxide where oxygenated functional groups intercalate in between graphene layers of graphite and increase their spacing from ~0.34 nm to ~0.85 nm, weakening the van der Waals forces that bind the graphene layers together[13]. A next step is to exfoliate these intercalated layers into separate GO sheets, normally by ultrasonic treatment. Final step is to reduce the GO and restore the sp2 hybridization state. This step is done by using strong reducing agents like hydrazine[14] or a combination of solvents and high temperature treatment[15]. An alternative route, is subjecting GO to thermal shock[16] where exfoliation and reduction processes occur simultaneously. In this procedure, GO is subjected to high temperature (~1000 °C) thermal shock in inert atmosphere for short time of ~30 seconds[16]. The high temperature decomposes the attached oxygenated groups to water vapor and CO2 gas leading to a pressure build up that is sufficient to break the already weakened interlayer bonds[16,17]. For successful exfoliation process, as long as the decomposition rate is higher than the gas diffusion rate through the interlayers[17], sufficient pressure for exfoliation will build up which points to the importance of having sufficiently high heating rate, otherwise GO will reduce without exfoliation[18]. The aforementioned methods are either energy intensive, requires special setup and/or use dangerous chemicals. There have been many attempts to exfoliate GO at low temperature. Xian et al.[19] studied the rGO produced at 100 °C and 400 °C using a tube furnace in a nitrogen atmosphere. 2 Page 2 of 21

Kaniyoor et al.[20,21] studied the effects of vacuum and argon/hydrogen atmosphere on the exfoliation of GO at 200 °C. Kottegoda et al.[22] produced few layer graphene after 10 min thermal treatment at 150—250 °C and subsequent annealing in argon atmosphere at 500 °C. Eswaraiah et al.[23] produced highly conductive graphene at 150—200 °C using focused solar radiation. In this article we report a facile, low temperature method to obtain thermally reduced graphene oxide (rGO) from different flake sized graphite oxide (GO). Using a hot-plate in ambient atmosphere and temperature ~225 °C, we successfully obtained rGO at a short time treatment of ~10 seconds. We believe that this method would make rGO more accessible especially for nanocomposites research and development. Our proposed method for obtaining rGO is ecologically green and easy to reproduce using a hotplate. We also studied the effect of starting graphite flake size on the final rGO quality. Our motivation for this is twofold: to determine the limiting graphite flake size to obtain meaningful amount of rGO. Secondly, small particles have less tendency for inter-particle interactions. Though it is widely accepted that large graphene flakes are desirable for enhanced reinforcement efficiency[24], large flakes tend to have larger inter-particle interaction. This leads to reduced reinforcing effect due to inefficient filler/matrix load transfer through strain shielding, in addition to particle agglomeration[25]. In that sense, using small sized flake might lead to obtain an enhanced overall reinforcing effect especially for large load fractions.

2.

Materials and Methods

2.1 Materials Graphite flakes (45 μm, grade 230U, ~5 μm grade Nano25) were kindly provided by Asbury Carbons (Fig. 1). Fuming HNO3 (>90%), KClO3 (98%) and Acetone (>99.8%) were purchased from Sigma Aldrich. H2SO4 (95%‒98%), and HCl (37%) were obtained from Alfa-Aesar. Epoxy resin, Epon 828 was DGEBA based epoxy resin from Momentive and hardener was Diethylenetriamine (DETA) from New Seoul Chemical Co. Ltd.

3 Page 3 of 21

2.2 Graphene preparation There are various methods to prepare GO[26] leading to different structures and oxidation levels[13]. The method of choice was the Staudenmaier’s method[27] because it is less aggressive than Hummer’s method so as not to destroy the aromatic rings of small flake graphite (private communications with the supplier), while at the same time it is not as mild as the Brodie method. The synthesis process (Fig. 2) is as follows: natural graphite (5 g) was added to acid mixture: 45 ml HNO3 to 87.5 ml of H2SO4 in a beaker while stirring on a magnetic plate to avoid agglomeration. The oxidizing agent KClO3 was added gradually during the four days course of reaction. The graphite slurry was then washed with 5% HCl solution several times and then in purified water. The slurry was vacuum filtered then dried in vacuum oven for 24 h at 60 °C. For the thermal reduction, a metallic container heated on a hot plate (J-HMS, JISICO) set to maximum temperature for 30 min to ensure thermal equilibrium. The temperature of the metallic container surface was (225±5) °C as measured by infrared thermometer gun (4470 Traceable®, Control Company). Dried GO was added and the metallic container was closed with a lid. Within ~5 seconds, the GO decomposed explosively with short flash of light and a massive volume expansion. The lid should be firmly closed as the pressure generated during reduction is enough to fling it and rGO would spill out the container. 2.3 Composite fabrication Calculated weight percentages of both large flake rGO (l-rGO) and small flake rGO (s-rGO) (0.05, 0.1 and 0.5 wt%) were dispersed in 10 ml DMF using ultrasonic for 30 min. The dispersion was mixed in epoxy resin using mechanical stirrer at 1000 rpm for 30 min and consequently with ultrasonic for another 30 min. The mixture was then degassed in vacuum oven for 30 min. DETA based hardener was then added (stoichiometric ratio 12%). The mixture was mixed in shear mixer (Kakuhunter SK-300S, Shashin Kagaku) at 2000 rpm for one min to avoid excessive heating and

4 Page 4 of 21

premature curing. The mixture was poured in silicone mold and cured at 45 °C for 60 min and then post cured at 100 °C for 2 h. 2.4 Characterization We studied thermal decomposition behavior of GO using differential scanning calorimetry (DSC) (PerkinElmer, DSC4000) and Thermogravimetric Analysis (TGA) (TA Instruments, TGAQ500) to determine the exfoliation temperature and weight loss during reduction. Measurements were performed under nitrogen atmosphere with a heating rate of 2 °C/min. Chemical analysis to determine oxygen content and types of functional groups was performed using Fourier Transform Infra-Red (FT-IR) spectroscopy (PerkinElmer, Spectrum 100) and X-ray photo electron spectroscopy (XPS). Raman spectrum was recorded to investigate atomic scale defects (Horiba Jobin Yvon, LabRAM HR800, 532 nm laser). The interlayer distance between graphitic layers was measured using XRD (PANalytical, EMPYREAN). Morphology of the rGO flakes was observed using Scanning electron microscopy (SEM) (FEI, Helios600i) and transmission electron microscopy (TEM) (Jeol, JEM-2100F). Atomic force microscopy (AFM) studies and sheet thickness measurements were performed using Park XE-100 AFM. Particle size analysis of rGO was measured by using a Malvern Mastersizer 3000. Tensile testing was performed according to ASTM D638-02a using tensile testing machine (MTS 810 materials testing system) at crosshead speed of 1 mm/min. Mode I fracture toughness (KIc) was measured according to ASTM D5045-99 for compact-tension samples. KIc values were determined according to the relationship:

KQ 

f (x) 

PQ BW

1/2

(1)

f (x)

( 2  x )( 0 . 886  4 . 64 x  13 . 32 x (1  x )

3/2

2

 14 . 72 x  5 . 6 x ) 3

4

(2)

5 Page 5 of 21

x  a /W

where K Q is the conditional fracture toughness KIc (MPa m1/2), is specimen width (cm) and

a

(3)

B

is the specimen thickness (cm),

W

is the crack length (cm). For both tests, an average of at least four

samples is reported.

3.

Results and Discussion

3.1 Exfoliation of graphene oxide Upon adding graphene oxide particle on the hot plate, a popping sound generated as a sign for the explosive exfoliation reaction starting. The estimated heating rate is ~40 °C/min. During thermal exfoliation, there is competition between decomposition reaction of functional groups to CO2 gas and diffusion of these gas molecules through the galleries of GO. In order to have pressure build up for exfoliation, the decomposition rate should be larger than diffusion rate, requiring a heating rate >2000 °C/min (33 °C/min) based on kinetic studies[17]. Smaller flaks had a lower yield as a large portion converted to soot. We tried even smaller flakes (50—250 nm, grade Nano99, Asbury Carbons) but GO transformed completely into soot and we could not collect any amount of rGO for analysis. This makes ~5 μm the limiting size for graphite flakes to obtain rGO. 3.2 Thermal analysis Thermal analysis results are shown in Fig. 3. DSC scans has an exothermic peak at ~225 °C for sGO and ~207 °C for l-GO that are associated with thermal decomposition/exfoliation of GO. From the TGA scan ~40% weight losses for both samples accompany the exothermic DSC peaks roughly quantifying the amount of decomposed molecules. This thermal behavior is a characteristic of thermal reduction of GO[28] indicating that ~225 °C is a sufficient temperature for decomposition of GO. 3.3 Chemical analysis 6 Page 6 of 21

We performed chemical analysis to determine the extent of reduction and types of oxygenated groups present. Quantitative chemical analysis using XPS showed C/O atoms ratio of 2.42 and 2.70 for l-GO and s-GO, and, 6.46 and 6.80 for l-rGO and s-rGO, respectively, which are in good agreement with previously reported results[16,26]. As expected, s-GO has a larger oxygen content as small flakes have faster oxidation rate[16,18]. On the other hand, s-rGO has lower C/O ratio showing better reduction compared to l-rGO. This is due to a more uniform heat distribution within small flake allowing more functional groups to reach decomposition temperature. Fig. 4 gives more insight about the chemical state of graphite, GO and rGO. The starting graphite material does not show any functional groups. GO is highly bonded with different oxygenated functional groups[29,30]. The peak at ~1040 cm-1 corresponds to epoxy group (–O–). C–O vibrational mode appears at ~1360 cm-1. The peaks at ~1620 and 1720 cm-1 represents carbonyl and carboxyl groups respectively. Hydroxyl groups (–OH) are characterized by the broad peak at ~3350 cm-1. FT-IR spectra of rGO show residual functional groups having relatively lower intensity peaks with almost complete elimination of hydroxyl groups. Fig. 5 gives a better view of these residual groups. Components of C1s peak show a dominant peak of sp2 C–C aromatic ring (284.7 eV), C–O (286.4 eV), C=O (287.6 eV) and C(=O)–OH (289.2 eV) that was not detected from the FT-IR spectrum. 3.4 Raman analysis Fig. 6 shows two dominant broad peaks that are characteristic of thermally reduced GO[31]. The peak appearing at ~1580 cm-1 is called the graphitic (G) band. This band is associated with in-plane stretching vibration of sp2 carbon atoms[32]. The other band at ~1350 cm-1 is associated with 6-fold symmetric breathing vibration of aromatic rings. This band is Raman inactive until loss of symmetry due to disorder and defects formation, hence called the defects (D) band[31]. Thermal annealing partially restores the sp2 domains, however, they are still isolated by amorphous sp3 sites due to

7 Page 7 of 21

extensive oxidation even after applying temperatures as high as 1100 °C[11,14]. A second reason for loss of symmetry is defects formation due to the abrasive nature of thermal exfoliation method[28]. A measure of disorder in graphitic materials is the D to G peak intensity ratio (ID/IG). l-rGo has ID/IG ~1.08 compared to ~1.02 for s-rGO, indicating higher defects density in l-rGO. Since s-rGO has lower oxygen content, and thus less sp3 domains, we attribute the higher ID/IG ratio to its higher defects intensity. Small flakes does not allow for efficient pressure build up, which results in less abrasive and inefficient exfoliation of s-rGO. These conclusions can be extracted from microstructural investigations as will be discussed in the next section. 3.5 Structural analysis Chemical characterization showed the reduction efficiency of GO and the type of residual functional groups in rGO. We assessed the extent of exfoliation as well as microstructural morphology by structural characterization. Fig. 7 shows the evolution of XRD patterns at different stages of graphene processing. Graphite has a sharp peak at 2θ = 26.5° corresponding to a layer spacing of 0.337 nm in the c-axis direction. It should be noted the FWHM (the full-width at halfmaximum) of s-graphite peak is broader (0.4582) than l-graphite (0.2726) with one order of magnitude less intensity, confirming smaller flake size[33]. Upon oxidation, inter-layer spacing increased to ~0.864 nm (2θ = 10.18°) as a result of the full intercalation of functional groups in the graphite basal planes. For l˗rGO, both peaks completely disappeared indicating complete exfoliation[16]. However, a faint, broad peak appears at 2θ ≈ 23.2° indicating partial restacking and consequently lower exfoliation efficiency. Fig. 8 shows the SEM images of s˗rGO and l˗rGO. Low magnification of l-rGO shows highly agglomerated irregular particles with particle size ~60 μm (Fig. 8(a)). s-rGO has particle size ~5 μm with a worm like structure indicating a uniform expansion in the c-axis direction (Fig. 8(c)). This supports our claim of uniform heat distribution for small flakes. Furthermore, worm like structure is a result of small flakes’ lower efficiency to build up pressure, preventing layers from fragmenting to 8 Page 8 of 21

smaller particles. Higher magnifications show highly porous 3D structure where the layers are exfoliated, however, there are connection points preventing full separation. Therefore, ultrasonic treatment is required for full separation of layers[17]. A 100 mg/ml dispersion of rGO in NMP was prepared by high amplitude ultrasonic bath while maintaining temperature below 10 °C[34]. A sample was withdrawn from the top of the dispersion for TEM analysis (Fig. 9). l-rGO has a more irregular topography (Fig. 9(a)) as compared to s-rGO (Fig. 9(b)). As mentioned previously, this is due to the more efficient pressure build up for larger flakes. SAED show the six-fold rotation symmetry of the (0001) plane. For l-rGO, the ring pattern indicates a polycrystalline structure, while spot pattern indicates single crystal structure of s-rGO. Contrast at bent edges show a multilayer structure of rGO[35]. This is further confirmed in the HRTEM (Fig. 9(c) and (d)). AFM sheet thickness measurements of l-rGO (Fig. 9(e)) show a sheet thickness in the range of 2 nm. The crumpled portion of the sheet (green line) shows a thickness of 4—8 nm. AFM of s-rGO (Fig. 9(f)) shows a relatively more flat sheet with thickness ~1.5 nm. A quantitative evaluation of flake size was performed through particle size analysis. The D10, D50 and D90 values are listed in Table 1. The average size D50 of the samples is 47.9 and 5.71 μm for l-rGO and s-rGO, respectively.

4.

Mechanical Analysis of the Nanocomposite Due to their high porosity, low temperature graphene is used for energy applications[36], namely,

supercapacitors. The application as reinforcements for polymers is one of the main applications of graphitic derivatives. Here we report their application as mechanical reinforcements for epoxy matrix. Fig. 10(a) shows that Young’s modulus was slightly affected by the addition of either s-rGO or l-rGO. Strength and fracture strain of nanocomposite at different loadings of l-rGO and s-rGO are shown in Fig. 10(b) and (c), respectively. l-rGO showed declined values of the strength and fracture strain compared to neat epoxy. On the other hand, s-rGO showed a simultaneous improvement in strength and fracture strain. For 0.05 wt% and 0.1 wt% s-rGO loading, the strength increased 23% and 24.3% respectively. However, for 0.5 wt% the strength falls back to 55 MPa which is almost the 9 Page 9 of 21

same value for neat epoxy. Based on the simultaneous enhancement of strength and fracture strain, the toughening effect of rGO was assessed using plane strain fracture toughness test (Fig. 10d). KIc value for neat epoxy is 1.08 MPa m1/2. Composites with l-rGO did not show a remarkable change in the property maintaining its value around 1.12 MPa m1/2 even at increasing weight percentage of lrGO. As expected for s-rGO, a significant increase in KIc to 2.12 and 1.74 MPa m1/2 at loadings 0.05 and 0.1 wt%, respectively. Increasing the amount of s-rGO to 0.5 wt%, KIc declined almost to the value of neat epoxy. Similar effect of improved toughening effect of reducing flake size was reported[37,38]. The toughening effect could be explained in the light of crack deflection theory[39,40]. When a propagating crack front encounters nanofiller, it deflects and twists changing from only mode-I crack opening to mixed mode. This leads to dissipation in energy, which is the main toughening mechanism. Smaller sizes of fillers act as more obstacles, creating longer tilting and twisting paths, thus consuming more energy. Larger particles act as stress concentration locations, facilitating micro cracks to generate and propagate.

5.

Conclusions (1) Thermally reduced graphene oxide could be successfully synthesized by facile, low energy consumption method. This method would make graphene more accessible and reduce time for graphene based nanocomposites research and development. (2) Reduction temperature of ~225 °C has exfoliation efficiency compared to high temperature ~900 °C. Although oxygen reduction efficiency is relatively lower, the produced graphene is readily chemically active for either functionalization or covalent bonding with polymer matrix. (3) Micron sized graphene flakes could be synthesized based on the starting graphite flake size. The limiting size for starting graphite flakes is ~5 μm. Submicron flakes totally converted to soot upon thermal reduction.

10 Page 10 of 21

(4) The obtained rGO was utilized as a filler for epoxy matrix. Large flakes had a negative effect on the mechanical properties of the nanocomposite. Small flakes had a simultaneous toughening and strengthening effect. This shows the potential application of low temperature s-rGO for toughening epoxy.

Acknowledgement The work was generously supported by the Space Core Technology Program through the National Research Foundation of Korea (NRF), funded by the Ministry of Science, ICT and Future Planning (No. 2013M1A3A3A02042257). Mr. Sangyun Kim and Mr. Hyeonkook Seo from the Electron microscopy laboratory at KAIST are acknowledged for their help in the TEM analysis.

References [1]

A.K. Geim, K.S. Novoselov, Nat. Mater. 6 (2007) 183–91.

[2]

F. Schwierz, Nat. Nanotechnol. 5 (2010) 487–496.

[3]

K.S. Kim, Y. Zhao, H. Jang, S.Y. Lee, J.M. Kim, K.S. Kim, J.H. Ahn, P. Kim, J.Y. Choi,

B.H. Hong, Nature 457 (2009) 706–710. [4]

F. Schedin, A.K. Geim, S.V. Morozov, E.W. Hill, P. Blake, M.I. Katsnelson, K.S. Novoselov,

Nat. Mater. 6 (2007) 652–655. [5]

M.S. Cao, X.X. Wang, W.Q. Cao, J. Yuan, J. Mater. Chem. C 3 (2015) 6589–6599.

[6]

B. Wen, M.S. Cao, M.M. Lu, W.Q. Cao, H.L. Shi, J. Liu, X.X. Wang, H.B. Jin, X.Y. Fang,

W.Z. Wang, J. Yuan, Adv. Mater. 26 (2014) 3484–3489. [7]

M.A. Rafiee, J. Rafiee, Z. Wang, H. Song, Z.Z. Yu, N. Koratkar, ACS Nano. 3 (2009) 3884–

3890.

11 Page 11 of 21

[8]

A. Yu, P. Ramesh, M.E. Itkis, E. Bekyarova, R.C. Haddon, J. Phys. Chem. C 111 (2007)

7565–7569. [9]

H. Kim, Y. Miura, C.W. Macosko, Chem. Mater. 22 (2010) 3441–3450.

[10]

D. Zheng, G. Tang, H.B. Zhang, Z.Z. Yu, F. Yavari, N. Koratkar, S.H. Lim, M.W. Lee,

Compos. Sci. Technol. 72 (2012) 284–289. [11]

V. Singh, D. Joung, L. Zhai, S. Das, S.I. Khondaker, S. Seal, Prog. Mater. Sci. 56 (2011)

1178–1271. [12]

J.R. Potts, D.R. Dreyer, C.W. Bielawski, R.S. Ruoff, Polymer. 52 (2011) 5–25.

[13]

D.R. Dreyer, S. Park, C.W. Bielawski, R.S. Ruoff, Chem. Soc. Rev. 39 (2010) 228–240.

[14]

S. Stankovich, D.A. Dikin, R.D. Piner, K.A. Kohlhaas, A. Kleinhammes, Y. Jia, Y. Wu, S.T.

Nguyen, R.S. Ruoff, Carbon 45 (2007) 1558–1565. [15]

S. Dubin, S. Gilje, K. Wang, V.C. Tung, K. Cha, A.S. Hall, J. Farrar, R. Varshneya, Y. Yang,

R.B. Kaner, ACS Nano. 4 (2010) 3845–3852. [16]

H.C. Schniepp, J.L. Li, M.J. McAllister, H. Sai, M. Herrera-Alonso, D.H. Adamson, R.K.

Prud'homme, R. Car, D.A. Saville, I.A. Aksay, J. Phys. Chem. B 110 (2006) 8535–8539. [17]

M.J. McAllister, J. Li, D.H. Adamson, H.C. Schniepp, A.A. Abdala, J. Liu, M. Herrera-

Alonso, D.L. Milius, R. Car, R.K. Prud'homme, I.A. Aksay, Chem. Mater. 19 (2007) 4396–4404. [18]

E. Matuyama, J. Phys. Chem. 58 (1954) 215–219.

[19]

H. Xian, T. Peng, H. Sun, J. Wang, J. Mater. Sci. 50 (2015) 4025–4033.

[20]

A. Kaniyoor, T.T. Baby, S. Ramaprabhu, J. Mater. Chem. 20 (2010) 8467.

[21]

A. Kaniyoor, T.T. Baby, T. Arockiadoss, N. Rajalakshmi, S. Ramaprabhu, J. Phys. Chem. C

115 (2011) 17660–17669.

12 Page 12 of 21

[22]

I.R.M. Kottegoda, X.W. Gao, L.D.C. Nayanajith, C.H. Manorathne, J. Wang, J.Z. Wang, H.K.

Liu, Y. Gofer, J. Mater. Sci. Technol. 31 (2015) 907–912. [23]

V. Eswaraiah, S.S. Jyothirmayee Aravind, S. Ramaprabhu, J. Mater. Chem. 21 (2011) 6800–

6803. [24]

R.J. Young, I.A. Kinloch, L. Gong, K.S. Novoselov, Compos. Sci. Technol. 72 (2012) 1459–

1476. [25]

N. Sheng, M.C. Boyce, D.M. Parks, G.C. Rutledge, J.I. Abes, R.E. Cohen, Polymer. 45

(2004) 487–506. [26]

H.L. Poh, F. Šaněk, A. Ambrosi, G. Zhao, Z. Sofer, M. Pumera, Nanoscale. 4 (2012) 3515–

3522. [27]

L. Staudenmaier, Berichte Der Dtsch. Chem. Gesellschaft. 31 (1898) 1481–1487.

[28]

M.J. McAllister, J.L. Li, D.H. Adamson, H.C. Schniepp, A.A. Abdala, J. Liu, M. Herrera-

Alonso, D.L. Milius, R. Car, R.K. Prud'homme, I.A. Aksay, Chem. Mater. 19 (2007) 4396–4404. [29]

D.W. Lee, L. De Los Santos V., J.W. Seo, L.L. Felix, A. Bustamante D., J.M. Cole, C.H.W.

Barnes, J.M. Cole, J. Phys. Chem. B. 114 (2010) 5723–5728. [30]

M. Acik, G. Lee, C. Mattevi, A. Pirkle, R.M. Wallace, M. Chhowalla, K. Cho, Y. Chabal, J.

Phys. Chem. C 115 (2011) 19761–19781. [31]

K.N. Kudin, B. Ozbas, H.C. Schniepp, R.K. Prud’Homme, I.A. Aksay, R. Car, Nano Lett. 8

(2008) 36–41. [32]

A.C. Ferrari, J. Robertson, MRS Proc. 593 (1999) 299–304.

[33]

G. Sun, X. Li, Y. Qu, X. Wang, H. Yan, Y. Zhang, Mater. Lett. 62 (2008) 703–706.

[34]

I. Zaman, T.T. Phan, H.C. Kuan, Q.S. Meng, L.T.B. La, L. Luong, O. Youssf, J. Ma, Polymer

52 (2011) 1603–1611. 13 Page 13 of 21

[35]

B. Wen, X.X. Wang, W.Q. Cao, H.L. Shi, M.M. Lu, G. Wang, H.B. Jin, W.Z. Wang, J.

Yuan, M.S. Cao, Nanoscale 6 (2014) 5754–5761. [36]

C. Zhang, W. Lv, X. Xie, D. Tang, C. Liu, Q.H. Yang, Carbon 62 (2013) 11–24.

[37]

X. Wang, J. Jin, M. Song, Carbon 65 (2013) 324–333.

[38]

Q. Zhao, S.V. Hoa, J. Compos. Mater. 41 (2006) 201–219.

[39]

K.T. Faber, A.G. Evans, Acta Metall. 31 (1983) 565–576.

[40]

M.A. Rafiee, J. Rafiee, I. Srivastava, Z. Wang, H. Song, Z.Z. Yu, N. Koratkar, Small 6 (2010)

179–183.

Figures Captions Fig. 1. SEM images of precursor graphite flakes: (a) large flakes and (b) small flakes. A close look at the highly agglomerated flakes shows that they are composed of smaller size flakes. Fig. 2. Schematic of synthesis process for hot-plate thermally reduces graphene oxide. Fig. 3. Thermal decomposition of GO: (a) large flakes and (b) small flakes. The TGA signal is super imposed over the DSC scan. Fig. 4. FT-IR of graphite, GO and rGO: (a) large flakes and (b) small flakes. Fig. 5. High resolution XPS of the C1s peak for rGO: (a) large flakes and (b) small flakes. Fig. 6. Raman spectrum and vibration modes of rGO. Fig. 7. Evolution of XRD spectrum from graphite to rGO: (a) large flakes and (b) small flakes. The (002) plane peak shifts form 2 ~26° for graphite to 2 ~10° for GO indicating inter-planar distance increase from 0.336 nm to 0.864 nm. The peak vanishes for rGO.

14 Page 14 of 21

Fig. 8. SEM images of l˗rGO (a, b) and s˗rGO (c, d) at different magnifications. Fig. 9. (a) and (b) TEM with SAED patterns of rGO sheets, (c) and (d) HRTEM, (e) and (f) AFM with sheet thickness measurement. Fig. 10. Mechanical properties of the nanocomposite: (a) modulus, (b) strength, (c) fraction strain and (d) KIc. Table 1 Particle size distribution of rGO Sample

D10 (μm)

D50 (μm)

D90 (μm)

l-rGO

15.2

47.9

122

s-rGO

1.36

5.71

12.6

D10, D50 and D90 are particle sizes at which 10%, 50% and 90% of the sample is below this given size

15 Page 15 of 21

Fig. 1

Fig. 2

16 Page 16 of 21

Fig. 3

Fig. 4

17 Page 17 of 21

Fig. 5

Fig. 6

18 Page 18 of 21

Fig. 7

Fig. 8

19 Page 19 of 21

Fig. 9

20 Page 20 of 21

Fig. 10

21 Page 21 of 21