Collision-induced activation: Towards industrially scalable approach to graphite nanoplatelets functionalization for superior polymer nanocomposites

Scale-up manufacturing of engineered graphene-like nanomaterials to deliver the industry needs for development of high-performance polymer nanocomposites still remains a challenge. Herein, we introduce a quick and cost-effective approach to scalable production of functionalized graphite nanoplatelets using “kitchen blender” approach and Diels-Alder chemistry. We have shown that, in a solvent-free process and through a cycloaddition mechanism, maleic anhydride can be grafted onto the edge-localized electron rich active sites of graphite nanoplatelets (GNP) resulting from high collision force, called “graphite collision-induced activation”. The mechanical impact was modelled by applying the point charge method using density functional theory (DFT). The functionalization of GNP with maleic anhydride (m-GNP) was characterized using various spectroscopy techniques. In the next step, we used a recyclable process to convert m-GNP to the highly-reactive GNP (f-GNP) which exhibits a strong affinity towards the epoxy polymer matrix. It was found that at a low content of f-GNP e.g., 0.5 wt%, significant enhancements of ~54% and ~65% in tensile and flexural strengths of epoxy nanocomposite can be achieved, respectively. It is believed that this new protocol for functionalization of graphene nanomaterials will pave the way for relatively simple industrial scale fabrication of high performance graphene based nanocomposites.


Grafting estimation
Page S5 Fig. S1. TGA thermograms of GNP, m-GNP and f-GNP at heating rate of 10 o C/min under nitrogen atmosphere. Page S5 Table S1. Mechanical properties of epoxy nanocomposites containing various contents of GNP and f-GNP.
Page S6  Table S2. Summary of the effect of different graphene chemical treatments on the tensile, the flexural properties, and glass transition temperature in several nanocomposites containing low loadings of graphene (e.g. 0.5 %wt) as reported in the literature. Page S12 Fig. S3. SEM image of fracture surface of epoxy/5% f-GNP nanocomposites containing high level of agglomerations (red arrow). Page S12 References Page S13

Materials and general experimentations
Graphite nanoplatelets (grade C, with a surface area of ~518 m 2 /g as measured by BET) was supplied as bulk dry powder by XG Sciences' xGnP®, Michigan, USA, which typically consist of aggregates of sub-micron platelets having a particle diameter of <10 μm and a thickness of less than a few nanometers. The received graphite nanoplatelets were refluxed in deionized water/ethanol solution for 72 h and then filtered and well-rinsed.
The washed graphite nanoplatelets was then heated for 3 min at 700 o C under argon atmosphere before storing in a vacuum oven at 100 o C to remove any impurities and moisture prior to use. Maleic anhydride (98%) was obtained from Sigma and used as received. An epoxy resin of DER 332 having equivalent weight of 175 g/eq under trademark of The Dow Chemical Company and hardener of tetraethylenepentamine was used as thermosetting epoxy polymer system (Sigma-Aldrich). All solvents used in this study were of analytical grade.

Measurements
Brunauer-Emmett-Teller (BET) surface area was measured by a Micromeritics TriStar 3000 using adsorption isotherm of nitrogen at 77 K and Malvern Mastersizer 2000 (UK) was used for determining particle sizes. Deionized water was used as a dispersion media for particle size measurements. The reflective index of 2.42 was used in the calculation of particle size distribution of graphene nanopatlates. Each experiment was carried out three times and error bars were not determined as the differences were insignificant. Fourier transform infrared (FTIR) analysis was performed by a Bruker Vertex 70 FTIR spectrometer in ATR mode with a resolution of 4 cm -1 . Raman measurements were conducted using a Renishaw InVia Raman Microspectrometer (Renishaw, Gloucestershire, UK) with diode laser at 514 nm at room temperature. All crbon-13 nuclear magnetic resonance ( 13 C-NMR) measurements were done on a Bruker AVANCE III NMR spectrometer operating at 75.4 MHz. Samples were packed into 4mm zirconia MAS rotors and magic-angle-spinning (MAS) at 10 kHz was used. 13 C measurements with direct carbon excitation and proton decoupling during acquisition were done to get comparable carbon spectra. For these a 13 C excitation 90 o pulse of 3 microseconds was used, applying a recycle delay of 24 sec for 6k sampled scans, totaling in 51 h measurement time per sample. X-ray photoelectron spectroscopy (XPS) analysis was performed using an AXIS Nova spectrometer (Kratos Analytical Inc., Manchester, UK) with a monochromated Al Kα source at a power of 180 W (15 kV  12 mA) and a hemispherical analyser operating in the fixed analyser transmission mode. The total pressure in the main vacuum chamber during analysis was typically 10 -8 mbar. Survey spectra were acquired at a pass energy of 160 eV. To obtain more detailed information about chemical structure, oxidation states etc., high resolution spectra were recorded from individual peaks at 20 eV pass energy (yielding a typical peak width for polymers of 0.8 -1.0 eV). All elements present were identified from survey spectra. The atomic concentrations of the detected elements were calculated using integral peak intensities and the sensitivity factors supplied by the manufacturer. Binding energies were referenced to the C 1s peak at 284.5 eV for graphitic carbon. Precision (i.e. reproducibility) depends on the signal/noise ratio but is usually much better than 5%. The latter is relevant when comparing similar samples. Water contact angles were measured using a KSV Model According to this model, the composite tensile modulus (EC) can be predicted by the following equations 1-4 .
Where the T and L parameters can be calculated by following equations: For all the above formulations, , Vf,, Em, and Ef are shape factor, volume fraction of filler, tensile modulus of matrix, and tensile modulus of filler, respectively. For the Halpin-Tsai model, the tensile modulus of GNPs, Ef, was equal to the modulus of exfoliation in the graphite c-axis (throughthe-plane) in the order of 36.5 GPa and the filler shape factor, , is equal to 0.667 (L/d) for platelets [3][4][5][6] .
In order to study flexural strength and modulus, 3-point bending test also were conducted according to ASTM D790-02 using an Instron universal testing machine. The tests were performed with a 10 kN load cell at a cross-head speed of 2 mm/min and the span-to-depth ratio was maintained at 16:1. The maximum flexural stress at failure on the tension side of a flexural sample, calculated from Eq. 4, was considered as the flexural strength (S) of the material. Moreover, flexural modulus (E) was determined from the slope (m) of the initial straight-line portion of the loaddeflection curve according to Eq. 5.

Grafting estimation
Surface functionalization degree could be approximately quantified using the following equation 7,8 : Where ΔW is difference of %weight loss for pure GNP with m-GNP during degradation process at an inert atmosphere, up to 600 o C, which is the temperature, assuming that graphitic structure of graphite nanoplatelets remains untouched and existing weight loss only results from the molecules on the surface. The M is the molecular weight of maleic anhydride. For our graphene nanoplatelets systems, ΔW and M are ~8.5% and 98.06 g/mol, respectively. Accordingly, ~0.94 mmol maleic anhydride equaled to 92.1 mg was obtained to be attached on per gram of pure GNP.     31 Reduced GO (0.1 wt%) d +31 +40 ----------Frictional pull-out/ Improvement of mechanical interlocking-adhesion at the nanofiller-matrix interface/ Crack deflection 32 Thermally reduced graphite oxide Integrated 3D structure of GF can totally eliminate the issues of uniform dispersion/ Interlocking mechanism/ Enhancing the interfacial adhesion/ Inducing local crack tip blunting/ Formation of microcracking bifurcation dilatation fracture process zone 35 a) Percentage increase "+" or percentage decrease "-" of mechanical properties compared to neat epoxy. b) Percentage increase "+" or percentage decrease "-" of glass transition temperature compared to untreated nanoparticle/epoxy nanocomposite. c) The value states the percentage differences of Tg compared with neat epoxy glass transition temperature. d) The value in brackets indicates that the content of the graphene-based nanofiller which is either lower or higher than 0.5 wt%. e) An average amount for mechanical properties data are reported for any length of polyether amine coupling agent which is grafted to GO. f) The prepared epoxy/GO mixtures or nanocomposites are denoted as EPx-GOy, where x is the TGPAP weight percentage in the epoxy oligomer, and y is the GO content in parts per hundred parts of epoxy resin without a curing agent (phr).