Functionalised graphene-multiwalled carbon nanotube hybrid poly(styrene-b-butadiene-b-styrene) nanocomposites

Functionalised graphene-multiwalled carbon nanotube hybrid poly(styrene-b-butadiene-b-styrene) nanocomposites

Accepted Manuscript Properties of functionalised graphene-multiwalled carbon nanotubes hybrid poly(styrene-b-butadiene-b-styrene) nanocomposites Qi Ch...

9MB Sizes 0 Downloads 31 Views

Accepted Manuscript Properties of functionalised graphene-multiwalled carbon nanotubes hybrid poly(styrene-b-butadiene-b-styrene) nanocomposites Qi Chau Tan, Robert A. Shanks, David Hui, Ing Kong, Associate Professor PII:

S1359-8368(15)00756-8

DOI:

10.1016/j.compositesb.2015.12.020

Reference:

JCOMB 3943

To appear in:

Composites Part B

Received Date: 11 September 2015 Accepted Date: 4 December 2015

Please cite this article as: Tan QC, Shanks RA, Hui D, Kong I, Properties of functionalised graphenemultiwalled carbon nanotubes hybrid poly(styrene-b-butadiene-b-styrene) nanocomposites, Composites Part B (2016), doi: 10.1016/j.compositesb.2015.12.020. 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.

ACCEPTED MANUSCRIPT

Properties of functionalised graphene-multiwalled carbon nanotubes hybrid poly(styreneb-butadiene-b-styrene) nanocomposites

1

RI PT

Qi Chau Tan1, Robert A. Shanks2, David Hui3 and Ing Kong1, 2*

Department of Mechanical, Materials and Manufacturing Engineering, The University of

2

SC

Nottingham Malaysia Campus, Jalan Broga, 43500 Semenyih, Selangor Darul Ehsan, Malaysia School of Applied Sciences, RMIT University, PO Box 2476, Melbourne VIC 3001

3

M AN U

Australia

Department of Mechanical Engineering, University of New Orleans, Lake Front, New Orleans,

LA 70138, United States.

TE D

*Corresponding author: Associate Professor Ing Kong

Department of Mechanical, Materials and Manufacturing Engineering

Jalan Broga

AC C

43500 Semenyih

EP

The University of Nottingham Malaysia Campus

Selangor, Malaysia

[email protected] Tel: +60389248362

Fax: +60389248017

1

ACCEPTED MANUSCRIPT

Abstract

A polymer nanocomposite suitable for industrial fabrication was prepared using the cost-

RI PT

effective solution casting method. A mix of functionalised graphene (f-G) and functionalised multi-walled carbon nanotubes (f-MWCNT) was dispersed into poly(styrene-b-butadiene-bstyrene) (SBS) matrix, where the ratios of individual fillers were varied in each samples. The

SC

synthesis of f-G is environmental friendly. It started from thermal expansion of expandable

graphite (EG), followed by modified Hummer’s method to produce graphite oxide (GO), and

M AN U

functionalisation of GO with 1-bromobutane (C4H9Br). The resultant f-G displayed stable and strong compatibility to non-polar organic solvents. The polymer nanocomposites were prepared and characterised for their morphology, thermal stability, mechanical properties and electrical conductivity. Polymer nanocomposites with mixed nanofillers showed improvement in thermal

TE D

and electrical properties than their single nanofiller counterparts. Both the properties have maximum improvement with same weight ratios of f-G and f-MWCNT, with each 1.5 wt% in SBS matrix. For mechanical properties, the improvement depends on the dominance of

EP

nanofiller. For the hybrid nanofillers, f-G improved the elastic behaviour of SBS, while f-

AC C

MWCNT gave the overall stiffness.

Keywords: A. Hybrid; A. Polymer-matrix composites (PMCs); B. Mechanical properties; B. Thermal properties; Graphene

2

ACCEPTED MANUSCRIPT

1. Introduction

In the vast field of nanotechnology, polymer nanocomposite has emerged as a prominent area of

RI PT

research and development. Polymer nanocomposite is defined as a polymer having fillers

dispersed in the polymer matrix in which at least one of them has its characteristic dimension measured within the nanometer range, 1–100 nm [1]. Many studies [2-4] of polymer

SC

nanocomposites showed superb enhancement of physical properties than the neat polymer. The increase in surface area-to-volume ratio when the particles get smaller, leads to an increasing

M AN U

dominance of behaviours of fillers over the overall properties in polymer nanocomposites. For that reason, majority of polymer nanocomposites just require very small ratio of the nanofiller.

The research field of carbon nanotubes (CNTs) has received a continuous growing interest since

TE D

these tubes were discovered in 1991 [5]. They have attracted attention as ideal reinforcement in composites with various matrices due to their unique structural, extraordinary mechanical, electrical, thermal and magnetic properties [6-9]. CNTs possess very high aspect ratio of >1000

EP

and is considered a 1D tube particles. The mixing of polymers and CNTs can open ways to develop engineering-flexible composites that have potential to be used in a number of

AC C

applications including electrostatic dissipation, electromagnetic interference shielding and bipolar plates of fuel cells [10, 11].

Graphene was first isolated in 2004 by Andre Geim and Konstantin Novoselov. They were awarded the 2010 Nobel prize in Physics for this [12]. Graphene is a hexagonal matrix of carbon that looks like chicken wire but is only one atom thick. Graphene is the stiffest and strongest

3

ACCEPTED MANUSCRIPT

material yet discovered (tensile modulus of 1 TPa and ultimate strength of 130 GPa). It has a greater surface area (2630 m2/g) and it is more electrically conductive (6000 S/cm) than any other material [13]. Graphene is impermeable to gases, resists high temperatures (estimated Tm =

RI PT

4900 K) and it is highly thermally conductive 5000 W/mK [14]. Very thin layer of graphene is the first material to the class of two-dimensional (2D) crystalline materials [15]. In order to

benefit from the outstanding intrinsic properties of both the CNTs and graphene, it is important

SC

to hybrid CNTs with graphene, where a combination of 1D and 2D intercalation of both

M AN U

materials results in a 3D nanostructure.

Poly(styrene-b-butadiene-b-styrene) (SBS) is a tri-block copolymer and a member of thermoplastic elastomers. As a synthetic rubber, it is widely used in many commercial applications due to its high elongation at break, abrasion resistance and durability [16]. SBS can

TE D

be used without vulcanization [17], which is an advantage as it does not degrade the mechanical and electrical properties obtained in composites. SBS can be manufactured by solvent casting using solvents such as toluene [18], p-xylene [19], ethanol and acetone [20], or by melt flow

EP

processes such as extrusion [21, 22].

AC C

The properties of polymer nanocomposites highly depend on the quality of mixing, stabilisation and homogeneity of the dispersed phase [1], particularly when nano-sized dispersion is desired. Good dispersion is important for mechanical reinforcement between fillers and polymer matrix. The challenge is prominent, as the preparation of effective dispersion of nanofillers is impeded by the nature of the fillers to agglomerate. Several reports had concluded that covalent modification of nanofiller is effective and essential to enhance filler dispersion [23, 24].

4

ACCEPTED MANUSCRIPT

Stability is another great challenge for polymer nanocomposites fabrication using solution casting. Differences in polarity between the matrix and filler can provide further complications by limiting the interaction between the composite constituents. Many non-polar organic solvents,

RI PT

such as alkanes, benzene with its homologs, and chloroform are extremely instable in dissolving nanofillers such as CNT and graphene sheets. This is due to the weak affinity between the

solvent molecules and the aromatic structures of both fillers. This effect is more intense for

SC

graphene due to the large surface area [12]. The processability is thus hindered, as most often non-polar organic solvents are used to solvate polymers. For this reason, various studies in

M AN U

preparing organophilic graphene sheets emerged, with most of them involving long and complex procedures.

In the present paper, issues above have been examined, where facile approaches of chemical

TE D

functionalisation and improved processes have been designed to ensure homogenous dispersion. The nanofillers used were CNT and graphene, which are ideally two materials of highest tensile strength and modulus discovered [25, 26] and proven superior electrical and thermal properties

EP

while the chosen polymer matrix was poly(styrene-b-butadiene-b-styrene) (SBS). Graphite oxide (GO) is a promising starting material for graphene functionalisation, traditionally by

AC C

Hummer’s method [27]. Recent years, modifications made to the original Hummer’s method often developed to yield quicker and more efficient oxidation of graphite [28, 29]. The significant advantages of this variation of modified Hummer’s method is the combined advantages from the papers, notably safe, high yield, eco-friendly and relatively inexpensive with the use of common chemicals. Large scale production and disposal facilities of reactants are also commercially accessible.

5

ACCEPTED MANUSCRIPT

The aim was to prepare and investigate the effects of hybrid nanofillers on the morphology, thermal stability, mechanical properties and electrical conductivity of the polymer nanocomposites under same total loading. Recent strategic approaches in polymer

RI PT

nanocomposites based on exploitation of π – π interactions among constituents at molecular levels to provide inherent ability for good dispersions [19, 30] was also performed in this paper with consideration of solvent and synergistic properties of both nanofillers and matrix. With

SC

aromatic rings contained in the styrene groups of SBS, functionalised graphene (f-G) and

functionalised multi-walled carbon nanotubes (f-MWCNT), essentially all the constituents,

M AN U

contain delocalised π electron systems where their π – π interactions were readily exploited.

2. Materials and methods

TE D

2.1 Materials

Expandable graphite (EG) was obtained from Graftech (GT) 220-50N (Ohio, USA). Pristine

EP

MWCNT with average diameter of 9.5 nm, purity of 90% was obtained from NanocylTM NC7000 (Belgium). Poly(styrene-b-butadiene-b-styrene) (SBS), CAS 9003-55-8,

AC C

tetrabutylammonium bromide (TBAB) powder, CAS 1643-19-2 were from Sigma Aldrich, Malaysia. Benzene, 98% sulphuric acid (H2SO4), 0.1M sodium hydroxide (NaOH), 30% hydrogen peroxide (H2O2), 1-bromobutane (C4H9Br) and hydrochloric acid (HCl) were obtained from R & M Chemicals, Malaysia. Potassium permanganate (KMnO4) and nitric acid (HNO3) were supplied by Bendosen Laboratory Chemicals (Malaysia). All reagents were of analytical grade and used directly without further purification. Ultrapure deionised water (Type I) prepared

6

ACCEPTED MANUSCRIPT

by Milli-Q® Integral 5 from Merck Millipore Water Purification System was used throughout the experiment.

RI PT

2.2 Synthesis of f-G

Expandable graphite (EG) was first heated with air in a preheated furnace at 1000 °C for 30 s to

SC

produce expanded graphite oxide (EGO). The size of EG were enlarged few hundred times and were loosely joined. Another set of filler, labelled as purified graphene (p-G) was obtained by

M AN U

further purifying the EGO at 1000 °C in tubular furnace under nitrogen atmosphere for 12 h. Further oxidation of both EGO and p-G were done by modified Hummer’s method as described below, which they were then labelled as graphite oxide (GO) and purified graphene oxide (p-

TE D

GO), respectively.

2.2.1 Modified Hummer’s method

EP

1.5 g EGO was first mixed with 100 ml 98% H2SO4 under stirring in an ice bath. 4 g KMnO4 was slowly added into the agitated mixture and the temperature of the reaction mixture was kept

AC C

below 25 °C until all KMnO4 was released fully. The reaction system was transferred to a 40 °C water bath and vigorously stirred for 45 min. Then, additional 500 ml water was prepared and the reaction mixture was added into it. 15 ml of H2O2 was added to react with the remaining KMnO4. Colour of the mixture turned from dark brown to yellowish, with violent effervescence. The mixture was then filtered and washed with excess dilute HCl to remove metal ions. During the filtration process, water was added a few times to wash the deposits and the pH of the filtrate was checked until it reached approximately pH 7. Finally, the deposit was dried at 60 °C in air 7

ACCEPTED MANUSCRIPT

until the paste became agglutinated. This method was repeated to yield 3 g GO. For comparison, p-GO was prepared with the exact method as mentioned above, using p-G as starting material.

RI PT

2.2.2 Functionalisation with 1-bromobutane

The functionalisation method was based on works done by [31] with slight modification. 0.75 g

SC

GO was loaded in a dried flask and 200 ml water was added. After vigorous stirring, the GO was dispersed using Hielscher ultrasonic processor UP400S (400 W, 24 kHz) set to 100% amplitude

M AN U

and 0.5 cycle/s for 15 min. Then, 50 ml of NaOH was added to the mixture. 80 ml of C4H9Br and 0.2 g of TBAB, a phase transfer catalyst were added into the flask under dry nitrogen atmosphere. The flask was then covered and heated in oil bath at 80 °C for 12 h. The yellowish brown unreduced dispersion changed into black, indicating the successful reaction process. After

TE D

that, the mixture was cooled to room temperature. Successive functionalisation can be determined by adding benzene or other immiscible non-polar solvents. Upon shaking, the functionalised graphene (f-G) would adhere to the non-polar solvent layer, leaving the water

EP

layer clean and transparent. To extract non-functionalised graphene, 100 ml benzene was added to the mixture. The f-G would be extracted into benzene (upper layer), leaving non-

AC C

functionalised graphene in water (bottom layer). Separation of both liquid layers was done by using a separatory funnel. Finally, the f-G were dried in oven at 60 °C for 72 h, and then pulverised into fine powder.

2.3 Synthesis of f-MWCNT

8

ACCEPTED MANUSCRIPT

Pristine MWCNT (p-MWCNT) were dispersed in a round-bottomed flask containing HNO3 solution. The suspension was dispersed with using Hielscher ultrasonic processor UP400S (400 W, 24 kHz) for 1 h. Then, it was refluxed at 80 °C with vigorous mixing for 24 h. The

RI PT

functionalised multi-walled carbon nanotubes (f-MWCNT) was collected by centrifugation, and then dried at 80 °C for 24 h. From the treatment in boiling nitric acid, the MWCNTs were broken into untangled and straighter forms with open ends. The TEM images of before and after

SC

functionalisation are shown in Fig. 1. Open ends were observed indicating the successful

M AN U

functionalisation of MWCNT [23].



TE D

2.4 Preparation of polymer nanocomposites

The f-G/f-MWCNT/SBS nanocomposite films were prepared by solution casting method, where benzene was used as the organic solvent for SBS. The composition of nanofillers was fixed at 3

EP

wt% with different ratios of f-G and f-MWCNT, while polymer matrix was set to be 97 wt%.

AC C

10 g of SBS pellets were dissolved in 60 ml benzene and left stirring overnight at room temperature until they completely dissolved. Each f-G/f-MWCNT was dispersed separately before forming their respective polymer nanocomposites. Known amounts of f-G and f-MWCNT were added to a small beaker containing 80 ml benzene. Ultrasonication of f-G/fMWCNT/benzene was carried out with ultrasonic processor at 100% amplitude and 0.5 cycle/second for 20 min. Vaporisation of benzene was expected as heat was generated

9

ACCEPTED MANUSCRIPT

throughout the process. After that, f-G/f-MWCNT and SBS solution were combined and were further dispersed with ultrasonic bath for 6 h. The temperature of the water was controlled at ~25 °C by replacing the equipment water content every 2 h. Nanocomposites were then precipitated

RI PT

with excess methanol.

The precipitated f-G/f-MWCNT/SBS nanocomposites were left overnight to dry and then stored

SC

in a desiccator. After the nanocomposites were rock solid, they were cut into smaller pieces. Polymer nanocomposites films were fabricated with hot pressing using Labtech LP50 press

M AN U

machine. A 0.5 mm thick steel mold with rectangular opening of 12 mm × 8 mm2 was used as final shape of the films. The mold was ~20 wt% overfilled by pieces of nanocomposites. Polytetrafluoroethylene (PTFE) sheets were used to overlay both sides of the mold, and another two flat metal plates to cover the outer sides of the PTFE sheets. The temperature of the hot press

TE D

was set at 185 °C for both upper heater and lower heater. The period of pre-heating, venting, full pressing and cooling was set to 3, 2, 4 and 5 min, respectively. A set of neat SBS was prepared

EP

with similar procedures for comparison.

AC C

2.5 Characterisations

2.5.1 Solvent compatibility

For the preparation of filler dispersions in different solvents, the dried product was first ground with a mortar and pestle and then added to the solvent and sonicated in an ultrasound bath for 1 h. To allow direct comparisons between the dispersing behavior of the different solvents, a

10

ACCEPTED MANUSCRIPT

certain amount of powder (∼5 mg) was added to a given volume of solvent (∼10 ml) in such a way that the resulting nominal concentration was adjusted to 0.5 mg ml-1 for all of the solvents. The dispersions were tested by using the following organic solvents: nonpolar solvents include

RI PT

benzene, chloroform, toluene, hexane, diethyl ether, polar aprotic solvents such as dichloromethane, acetonitrile, diethyl sulfoxide (DMSO) and polar protic solvents include water, methanol. The compatibility of the filler with different solvents (nonpolar, polar aprotic and

SC

protic solvents) were compared.

M AN U

2.5.2 Scanning electron microscopy (SEM)

Morphology of nanocomposite fracture surface was observed using Field Emission Scanning Electron Microscopy (FESEM) equipment FEI Quanta400F under low vacuum mode. Fractured

TE D

samples are mounted vertically on conductive tape with the fractured surface pointing upwards.

EP

2.5.3 Transmission electron microscopy (TEM)

The morphology of both p-MWCNT and f-MWCNT were observed using a high-resolution

AC C

transmission electron microscope (HRTEM; JEOL-1010), with an accelerating voltage of 100 kV. The samples used for TEM observations were prepared by dispersing a small amount of MWCNTs in absolute ethanol followed by ultrasonic processoring for 10 min, then placing a drop of dispersion onto a 200-mesh carbon-coated copper grid, which was then allowed to dry in air.

11

ACCEPTED MANUSCRIPT

2.5.4 Fourier transform infrared spectroscopy (FTIR)

Fourier Transform infrared spectra (FTIR) were recorded on Perkin Elmer FT-IR Spectrum RX I

RI PT

and spectra analysis was done using Essential FTIR software package (eFTIR).

SC

2.5.5 Thermogravimetric analysis

Thermal stability of the polymer nanocomposites were interpreted by Thermogravimetric

M AN U

Analysis (TGA), by using Perkin-Elmer Simultaneous Thermal Analyzer (STA) 6000. A small mass of nanocomposite film, ~3 mg was analysed in an open alumina crucible (70 µl). The sample was heated from 30 to 700 °C under nitrogen at 20ml/min. At 700 °C, oxygen is switched

2.5.6 Electrical properties

TE D

on with flow rate of 25 ml/min. Heating was continued to 800 °C and held isothermal for 5 min.

EP

Electrical conductivity was characterised using FLUKE PM6304 Programmable Automatic RCL Meter. Both DC and AC currents were used, with AC testing frequencies at 1, 10 and 100 kHz.

AC C

Each specimen was cut from respective hot pressed nanocomposite into 20 mm x 5 mm2 films. For all measurements, RCL meter is set to connect in series and impedance were measured directly. All data showed dominance of both resistive and capacitive behaviour. Thus, simplified RC circuit model in series was used for calculations where inductance of nanocomposite was assumed to be zero.

12

ACCEPTED MANUSCRIPT

2.5.7 Mechanical properties

Mechanical properties of polymer nanocomposites samples were characterised by Lloyd LR5K

RI PT

Plus bench type tensile machine. The specimens were cut into dumbbell shape and procedures conform to ASTM D412 Type ‘D’, a standard test method for thermoplastic elastomers.

SC

2.5.8 Creep recovery

M AN U

Creep and recovery tests were performed by Dynamic Mechanical Analysis (DMA) conducted in tensile mode using TA Instrument Q800 Dynamic Mechanical Analyzer. The specimens with dimensions 12.8 x 4.0 x 0.6 mm3 were tested under ambient conditions (30 °C). The test was performed by subjecting films to an applied stress of 0.5 MPa for 20 min and followed by

TE D

recovery phase of 80 min with 0.01 MPa applied stress. The applied stress chosen was within the linear viscoelastic region of all the nanocomposites.

EP

2.6 Materials nomenclature

AC C

In the interest of clarity, the composites will be presented as “GX-TY”, where ‘G’ and ‘T’ refer to f-G and f-MWCNT and ‘X’ and ‘Y’ refers to the ratio of the weight percent of f-G (ranging from 0 to 3 wt%) to the weight percent of f-MWCNT (ranging from 0 to 3 wt%) as shown in Table 1.



13

ACCEPTED MANUSCRIPT

3. Results and discussion

RI PT

3.1 Surface modification

Functionalisation of GO to f-G is a dual substitution process. A schematic diagram of the

functionalisation is illustrated in Fig. 2. A strong base NaOH and phase transfer catalyst TBAB,

SC

both were introduced to the GO solution to start a reversible reaction. Na+ forms a strong bond that displaces the Br  of the TBAB, leaving the now positive tetrabutyl ammonium ion (TBA+)

M AN U

to adhere to the partially negative carboxyl groups (carboxylate ions in water). The Na+ forms strong ionic bond with Br  forming a stable salt compound, NaBr. When C4H9Br was introduced into the reaction, the weak bond between the terminal Br  and –C4H9 allows Br selectively

bond to strong TBA+, in which during the process leaving –C4H9 chains on the graphene sheets

TE D

[31] and TBAB revert to its original form. The end products are soluble salt, NaBr, TBAB and water which can be filtered easily and the dried product would be f-G.

EP



AC C

During the functionalisation process, C4H9Br and GO dispersed in water form two immiscible layers, with yellowish brown (colour of GO) layer floating on top, and a transparent layer of C4H9Br at bottom. It is important to set the temperature to be below the boiling point of C4H9Br, for example 80 °C as specified for this functionalisation process such that the rate of reaction is sufficient without excessive vaporisation of C4H9Br to the surrounding. The substitution reaction takes place at the interfacial layer and when the bubbles pass through the layer of water/GO. The

14

ACCEPTED MANUSCRIPT

unreduced dispersion changed into black, indicating the oxides were displaced by –C4H9 groups. Using phase transfer catalyst, TBAB also promotes the C–C formation under liquid-liquid phase and provides opportunity of environmental friendly organic catalytic process without the use of

3.2 Fourier transform infrared spectroscopy (FTIR)

SC

nanocomposites fabrication.

RI PT

organic solvents, as discussed by [30], another effective approach for industrial polymer

M AN U

FTIR spectrum of GO produced by this improved method yield sufficient oxides groups for further covalent modification of graphene sheets. The surfaces of GO are highly oxygenated, carrying hydroxyl, epoxide, and carboxyl groups as evidenced by FTIR in Fig. 3.

TE D



Carboxyl group is evidenced by the broad absorption peak at 3200 cm-1 to 3700 cm-1 which is

EP

attributed to the O–H stretch, and the 1720 cm-1 associated to the C=O stretch. The band at 1627 cm-1 is attributed to the C=C stretch that exist in the aromatic structures of graphene. 1160 cm-1

AC C

and 1035 cm-1 are recognised as the C–O stretch which resemble epoxides or hydroxyl groups that possibly formed on the deformed surface on the aromatic structure. FTIR spectrum of GO shows agreement to works of [32], [33] and [34].

The same modified Hummer’s method done to oxidise p-G, the p-GO was also shown in Fig 3. Comparing the spectra, GO with EGO as starting material gives more effective oxidation

15

ACCEPTED MANUSCRIPT

process, with more intense bands showing carboxyl, epoxy groups. The colour of p-GO appeared brownish black, much darker than GO, indicating lesser degree of oxidisation. In contrast, p-GO has preserved the overall structure of graphene. This is evidenced as the 3 subsequent peaks that

RI PT

exist around the 1650 cm-1 and 1550 cm-1 region which are associated with the C−C stretch of aromatic rings. From the spectra of GO and p-GO, it can be deduced that the oxidation process damages the graphitic structure. As further covalent modification requires high degree of

SC

oxidation, GO developed from EGO was preferred as starting material for functionalisation with

M AN U

C4H9Br.

The FTIR spectra clearly confirmed the –C4H9 groups were covalently bonded to graphene sheets. A basic indication of successful introduction of –C4H9 group is evidenced by introduction of a few intense bands compared to the FTIR spectra of GO. The shifting of the original broad

TE D

peak of GO to a narrower peak at 3130cm-1 is observed, indicating remaining carboxyl groups. 1600 cm-1 and 1401 cm-1 indicates the aromatic C=C stretch. This is further confirmed by 3010 cm-1 attributed to the sp2 aromatic C–H stretch. It can be explained that during the

EP

functionalisation process, partial structure of graphene is restored. The rightmost 722 cm-1 peak is considered the fingerprint of this functionalisation process as a rule that this characteristic peak

AC C

only appears in long chain alkanes, firmly indicating the presence of –C4H9 in f-G.

Both peaks 2927 cm-1, 2856 cm-1 in f-G resembles C–H stretch that is in common with carbonyl groups of aldehydes. Also, the original 1720 cm-1 band of GO shifted to 1747 cm-1, attributed to the general C=O stretch. 1164 cm-1 is a reminiscence of epoxy groups, similar to GO. These characteristic peaks show that the presence of H+ in oxygen containing groups is essential for the

16

ACCEPTED MANUSCRIPT

substitution process. Oxygen atoms are more electronegative than carbon atoms, but only with dissociation of hydroxyl in carboxyl groups, the catalytic process starts.

RI PT

3.3 Solvent Compatibility

Solution casting method provides enough freedom of space where nanofillers can disperse.

SC

However, the high aspect ratio of pristine CNT and graphene may agglomerate overtime due to magnified Van der Waals attraction. Functionalisation of MWCNT and graphene results in

M AN U

separation of individual tubes and exfoliation of sheets which have indirect influence on their overall compatibility in solvents. Further exploitation of such modification requires the right selection of solvent, as functionalisation process often result in different characteristics of

TE D

nanofillers [35].

The honeycomb arrangement of carbon atoms in f-MWCNT, f-G and styrene chains of SBS are similar to aromatic rings of benzene. Thus, the mutual π-π stacking between solvent molecules

EP

and three constituents are justified. The individual interactions between constituents are supported by: SBS matrix and CNT reported by [36, 37] and SBS with graphene reported by

AC C

[19, 38].

Fig. 4 shows the digital pictures of as-prepared filler dispersions in different solvents 1 week after ultrasonication. It was found that f-MWCNT shows moderate compatibility to benzene as shown in vial (a). Priority of functionalising MWCNT is to separate the individual tubes to reduce agglomeration. Dispersion of f-MWCNT in solvent is relatively easier due to their much

17

ACCEPTED MANUSCRIPT

smaller size compared to graphene. Acid treatment is considered the most conventional covalent



RI PT

modification of MWCNT [23].

In the case of GO (vial (c)), it appeared that GO was well dispersed in water but not in benzene,

SC

exhibiting hydrophilic property because of therein oxygen functional groups. Same amount of GO and f-G were dispersed into vial (d) and (e) each containing equal amount of water and

M AN U

benzene respectively. The stability of f-G in benzene was much comparable to GO dissolving in water. These results indicate that f-G exhibits very good hydrophobic property which can be well dispersed in non-polar polymers [39]. The enhanced interfacial interaction between f-G and polymer suggests the potential of processability of polymer nanocomposites in large scale

TE D

production.

Further investigating on the stability of f-G in different solvents suggest that polar protic solvents

EP

are the least favourable for f-G, polar aprotic solvents shows intermediate compatibility, while strong compatibility is observed to non-polar solvents. The finding aligns well with the results

AC C

reported by [31] as well. This arrangement of compatibility is opposite for GO. Heavily oxygenated GO is hydrophilic and readily suspends in polar solvents than in non-polar organic solvents.

Demonstration of this property was further proven by mixing f-G in vials containing different combination of immiscible solvents as shown in Fig.4. Each filler was dispersed into solvents

18

ACCEPTED MANUSCRIPT

with brief ultrasonic bath for 1 h. The ability of f-G to disperse in non-polar solvents outperforms polar protic solvents shown in vials labelled (f), (g), (h), (i) and (j). Polar protic solvents exhibit strong intermolecular hydrogen bonds and thus to achieve stability of fillers within the solvent,

RI PT

similar intermolecular forces have to be introduced. Meanwhile, when f-G was dispersed into non-polar and polar aprotic solvent, as shown in vials (l) and (m), partial f-G sunk to the bottom polar solvent layer. The f-G was observed to have suspended homogenously in both layers when

SC

sample bottles were shaken. After 1 h, large portion of f-G adhere to the top layer non-polar solvent. Some f-G have deposited to the bottom layer as f-G still has certain degree to adhere to

they sunk to the bottom due to weight.

TE D

3.4 Morphology

M AN U

polar aprotic olvents. Another reason may be of poorly functionalised graphene sheets such that

Morphology is an important study to investigate the dispersion of nanofillers. Dispersion is a well-known challenge since CNTs and graphene sheets have an inherent tendency to

EP

agglomerate due to large aspect ratio of both magnifying the Van der Waals forces [40]. The degree of dispersion is an important factor for mechanical reinforcement as well as to preserve

AC C

their electrons transport system.

SEM images of nanocomposites fracture surface are presented in Fig. 5. Fig. 5 (a) shows the morphology of neat SBS. It can be observed that the fracture surface of neat SBS shows slight wavy surfaces and voids that may be formed due to mechanical tearing. Fig. 5 (b) and (c) show the morphology of sample G0-T1 and G1-T0, each contain only single nanofiller. Comparing

19

ACCEPTED MANUSCRIPT

both, sample G0-T1 contains f-MWCNT which appear as bright dots, identifiable as the tips of the tubes, while f-G are depicted as long bright segments, resembling sheet-like structures



RI PT

embedded in the matrix.

SC

Spot magnification on the fracture site of sample G0-T1 reveals that the long f-MWCNT

entangle with each other and were held onto the surface of ripped polymer pieces (Fig. 5 (d)). On

M AN U

the other hand, similar viewing on sample G1-T0 shows the f-G being exfoliated to few-layered sheets in high magnification (Fig. 5 (e)). The crumpled morphology is totally different from the two-dimensional planar morphology of graphene, which is mainly attributed to high flexibility of the graphene sheets due to its large lateral size, thin thickness and large aspect ratio. The similar

TE D

morphology was observed by [41]. These crumpled surfaces are observed as a common feature of functionalised graphenes [42].

EP

From Fig. 5 (b)-(d), although implying the same fabrication technique, nanocomposites containing single filler, each f-MWCNT and f-G are difficult to be uniformly dispersed in the

AC C

matrix and are seen as large aggregates within the matrix. However, incorporating both fillers, under same magnification as shown in sample G1-T5 and G1-T1 (Fig. 5 (f) and (g)), the aggregates are smaller and has better quality of dispersion. The fracture surfaces also show wavier ridges, speculating a finer reinforcement. Therefore, the incorporation of both f-MWCNT and f-G has great influence on the dispersion. This was also claimed as the synergetic effect of the CNTs and graphene sheets [43, 44]. Herein,

20

ACCEPTED MANUSCRIPT

the SBS matrix have sufficient contact with f-MWCNT and f-G as observed in Fig. 5 (h), a spot magnification on the fracture site of sample G1-T1. Furthermore, the solvent, nanofillers and polymer chains, facilitate the dispersion through the inherent π–π interaction among aromatic

RI PT

rings and provide mutual reinforcement. The delocalised electrons serve as a binding factor among constituents and separation of nanofillers. As a result, it is expected that the mechanical

SC

properties of the composites could be significantly enhanced.

M AN U

3.5 Thermal stability

The thermal performances of final nanocomposites are necessary to analyse the combined effect of the modified fillers on the microstructure of the polymer. It is also important to ascertain the suitability of the nanocomposite product for different applications which would require exposure

TE D

to temperature during service life.

Fig. 6 shows the TGA curves of neat SBS and its nanocomposites. The degradation temperatures

EP

corresponding to different weight loss, 5 wt% (T5), 50 wt% (T50) and 95 wt% (T95) of neat SBS and its nanocomposites are shown in Table 2. Many studies have concluded that increasing filler

AC C

content results in higher thermal stability [43, 45]. However, isolated characterisation which focuses on the synergistic effect of both f-G and f-MWCNT in terms of thermal stability has never been done previously.



21

ACCEPTED MANUSCRIPT



It is observed that T5 for all samples are similar (~400 °C). From the curves, at temperature range

RI PT

up to 400 °C, all of the nanocomposites have equal level of stability with each of them slightly shifted the onset degradation temperature at an average of +5 °C, proving consistent level of enhancement for all sets of nanocomposites. Thermal stability for each samples started deviating

M AN U

give G1-T1> G1-T5> G0-T1> G5-T1> G1-T0> SBS.

SC

significantly at ~450 °C upon reaching T50. Arranging the levels of thermal stability by sample

Although total filler weight was fixed, ratios of both nanofillers affect the microstructure of nanocomposites and consequently have an impact on the thermal stability. It is noticed that fMWCNT have stronger influence in this property. Soon after f-G dominates the filler weight, the

TE D

thermal stability decreases (sample G5-T1 and G1-T0). This may be due to the large exposed surface of graphene which facilitate heat transfer.

EP

The degradation profiles of f-G dominated nanocomposites, samples G5-T1 and G1-T0 have a bulging profile between temperatures 500–560 °C. This is not obvious in f-MWCNT dominated

AC C

nanocomposites, which have rather uniform weight loss along that temperature range. This section in the profile is attributed to the removal of –C4H9 groups, as an obstruction before the degradation of graphene sheets. TGA using only f-G shows similar curve pattern at 400– 550°C [31]. Incorporating f-G into polymer matrix has translated the profile to a higher temperature region.

22

ACCEPTED MANUSCRIPT

All polymer nanocomposite have varying maximum degradation temperatures from 535 to 565 °C. In contrast, neat SBS completely degraded at ~500 °C. The gas was switched to oxygen at 700 °C and heating was continued to 800 °C to burn off the remaining carbon fillers as carbon

RI PT

dioxide or carbon monoxide, to ascertain that no char residue is left in the pan. This serves as a measure to offset the zero error in such that SBS, f-G and f-MWCNT do not react with nitrogen to form another compound, the lower limit of wt% at 800 °C serves as corresponding correction

SC

values for each samples.

M AN U

In general, the thermal stability of each nanocomposites was found to have consistent levels of improvement as compared to their neat SBS counterparts. This is due to the fixed total filler content throughout each sample.

TE D

3.6 Electrical conductivity

The large aspect ratio and delocalised π electrons of both graphene and CNT provide an

EP

opportunity for conductive polymer. Automatic RCL Meter determines the dominated electric behaviours of the material to be resistance, R and capacitance, C. Inductance is not dominating

AC C

even during testing with large frequencies (high oscillating frequencies induce voltages). Thus, the inductance term can be ignored, and general RC circuit model in series is used for calculation, shown as Eqn. 1. Z is impedance,  represents the imaginary unit and is the sinusoidal angular frequency in radians per second, given as 2 . Relating it to conductivity and considering scalar variables only, the conductivity, is given as Eqn. 2 where A and ℓ are the cross section area and length of specimens, respectively.

23

ACCEPTED MANUSCRIPT

=

1 (1)   ℓ

1  A  −  2 



(2)

RI PT

=R+

SC

The calculated results are shown in Table 3 and Fig. 7. The hybrid nanofillers demonstrated very

M AN U

significant synergistic effect on electrical conductivity. Using different test frequencies, the consistencies of results were compared to ensure capacitive behaviour of the samples have little contribution on overall polymer nanocomposites electrical properties. Similar to the arrangement of thermal stability, sample G1-T5 and G1-T1 again display better electrical conductivities, surpassing other samples. This may be due to the formation of 3D percolation network as could

TE D

be observed from SEM image (Fig. 5h), contributed by the dispersion state. Taking into account the electrical property of nanofillers, these two samples (G1-T5 and G1-T1) are dominated with f-MWCNT that are well-dispersed, undergo less chemical modification process, are speculated

AC C

EP

to have preserved the overall graphitic structure and hence the electrons transport network.





The electrical conductivity of f-G dominated samples G5-T1 and G1-T0 dropped dramatically. It was postulated that the electrons conjugation system of f-G was heavily damaged after the

24

ACCEPTED MANUSCRIPT

intense functionalisation process in order to introduce –C4H9 groups to graphene. This is considered as a drawback where processability was improved while sacrificing electrical properties. Comparing with neat SBS, all the samples containing nanofillers exhibit little

RI PT

improvements in electrical conductivity. Comparing with other reported journal [46], the electrical conductivity of the nanocomposites obtained in this study are a few orders of

magnitude lower. The possible reason for this is the total content of nanofiller is not substantial

SC

for good electrical conductivity and also the functionalisation of both carbon fillers. Chemical modification damages the graphitic structures hence the weakening of overall intrinsic electrical

M AN U

conductivity [23].

3.7 Mechanical properties

TE D

The results obtained from uniaxial tensile testing of composites samples are shown in Table 4 and Fig. 8. For elastomers such as SBS, modulus is usually defined as the tensile stress at a particular strain value due to their small linear elastic region. Manufacturers of SBS typically

EP

record the tensile stress at 300% elongation as modulus as in Table 4. The composites samples exhibited remarkable improvements in mechanical properties relative to neat SBS. It was found

AC C

that even under same total loading amount of 3 wt% nanofillers, the moduli increment of different samples ranges at 25% – 74%, while the tensile strength (TS) improves for 1.4% – 24.2% relative to neat SBS. This strongly suggests the occurrence of synergistic effects in the mechanical properties of polymer nanocomposites.



25

ACCEPTED MANUSCRIPT



RI PT

The deformability of polymer nanocomposites is well enhanced from 24.8% – 45.1%, except for sample G0-T1 which breaks at lower strain. In cases of individual filler (samples G0-T1 and G1T0), it was found that f-MWCNT is very efficient in improving the modulus while f-G

SC

significantly improves the TS and elongation at break. Exploring the change in mechanical

properties of other samples also reveals this property with some synergistic improvements. For

M AN U

example, introduction of f-G into polymer nanocomposites while maintaining f-MWCNT at a higher loading, as in sample G1-T5, shows that the modulus, TS and elongation at break increase substantially than sample G0-T1. As dispersion state of polymer nanocomposites also influences the mechanical properties, revisiting the morphology of sample G1-T5 in Fig. 5 (f) shows this

TE D

agreement.

Further increased loading of f-G sacrifices the stiffness of polymer nanocomposites but result in

EP

tougher and more elastic behaviour. The elongation at break increase monotonically with the increase of f-G content as shown in Table 4. This may be contributed by the f-G nanosheets

AC C

which exhibit high stretching ability. Initially f-G is like crumbled sheets of paper held with strong mechanical interlocking with the SBS chains. When stretched they become flat while storing energy to revert to their original crumbled state. The load transfer efficiency in f-G is speculated to be very high. On the other hand, the tensile testing results concluded that the fMWCNT has restrictive effect, responsible to restrict deformation of SBS by inhibiting the movements of polymer chains, resulting the stiffer material [47].

26

ACCEPTED MANUSCRIPT

From the results, generally, the combination of both nanofillers gives tougher mechanical property. From Samples G1-T5, G1-T1 and G5-T1, each containing different ratios of f-G/fMWCNT, as f-G dominates in nanofillers, the modulus and TS decrease. Instead of synergistic

RI PT

effect, mutual restrictions may have occurred. The disproportionate sharing of stresses among fG and f-MWCNT postulates higher stress in f-G. f-G nanosheets are flexible, allowing

movements of polymer chains but the lower amount of f-MWCNT restricts the deformation

M AN U

within SBS matrix which in turn lowering the TS.

SC

when under high tension. These interactions may potentially introduce local stresses and defects

Apart from the uniaxial tensile testing, creep as a time and temperature dependent phenomenon, is important for material applications requiring long-term durability and reliability. Since polymers are viscoelastic materials that have combined mechanical behaviours of elastic solids

TE D

and viscous fluids, they respond to external force overtime in an intermediate manner between the two. Different time-dependant performance of nanocomposites were demonstrated by analysing common dependencies of the elastic, viscoelastic and viscoplastic characteristics, with

EP

typical profile as illustrated in Fig. 9. Viscoplastic effects are noticeable as irreversible strains

AC C

remaining after a certain time period after the stress removal.



The creep curves of all f-G/f-MWCNT/SBS nanocomposites and neat SBS were plotted on the same graph as in Fig. 10. This gives the order of samples G0-T1> neat SBS> G1-T0> G5-T1> G1-T5> G1-T1 by decreasing creep. It was found that for sample G0-T1, the loading of f-

27

ACCEPTED MANUSCRIPT

MWCNT alone has higher creep strain level than neat SBS, in other words it did not improve the creep performance of neat SBS. This is likely due to the low elasticity performance of fMWCNT within the SBS matrix. The creep characteristics of other nanocomposites are coarsely

RI PT

dependent to the loading ratio of each nanofillers, evidenced by the arrangement of creep

deformation. Nanocomposite of mixed nanofillers provide better resistance to creep deformation, with samples G1-T5, G1-T1 and G5-T1 having lower creep strain levels, whereas the 1:1 ratio

SC

nanofiller in sample G1-T1 surpassing other nanocomposite samples. Nanocomposites with individual nanofiller f-G and those with mixed nanofillers exhibited flatter creep curves

M AN U

compared to neat SBS sample. These samples also exhibit lower gradient, indicating a decrease of deformation of the viscous component. Besides, the recovery phases suggest that incorporating mixed nanofillers has most recoverable strain after the stress removal.

TE D



The Burger’s (or four parameters) model, a combination of Maxwell and Kelvin–Voigt elements,

EP

is one of the most suitable model to give the relationship between the morphology of the

AC C

composites and their creep behaviour [48].

Conclusions

In this study, an industrial approach of highly processable polymer nanocomposites was demonstrated. It started with the simple yet efficient functionalisation of graphene sheets with 1bromobutane to produce organic solvent compatible f-G. Polymer nanocomposites fabrication by solution casting was based on selections of solvents and polymers that promote compatibility by 28

ACCEPTED MANUSCRIPT π–π interaction. Both the methods of functionalisation and fabrication are highly scalable, fast, economic and environmental friendly and thus suitable for industrial manufacturing. Different ratios of f-G/f-MWCNT hybrid fillers with each constituted 3 wt% to respective SBS

RI PT

nanocomposites were also prepared and compared in details. Dispersion observed in SEM, thermal stability via TGA results and AC electrical conductivity from RC circuit have

determined to be at peak state with the 1:1 ratio of f-G and f-MWCNT. This synergy is a result

SC

of the enhanced dispersion by π–π interaction for better load transfer and the 3D nanostructure geometry of the nanofillers within the polymer matrix, observed in SEM. For mechanical

M AN U

properties, synergistic effects were observed with incorporation of both f-G and f-MWCNT, where the modulus and tensile strength increase after substituting a portion of the nanofiller with another. On the other hand, f-G proved to have superb enhancement in toughness, yielding highly stretchable SBS nanocomposites. Overall, under the same nanofiller weight, having

TE D

hybrid nanofillers potentially maximises the existing properties. In the case of f-MWCNT and fG, the optimum range is reasoned to be at 1:1 weight ratios or having equal amount of

EP

nanofillers.

AC C

Acknowledgements

The authors gratefully acknowledge the facilities, the scientific and technical assistance of Engineering Research Department, University of Nottingham Malaysia Campus and School of Applied Physics, National University of Malaysia.

29

ACCEPTED MANUSCRIPT

References

[7]

[8]

[9]

[10]

[11]

[12] [13] [14] [15]

[16]

[17]

RI PT

SC

[5] [6]

M AN U

[4]

TE D

[3]

EP

[2]

Ajayan PM, Schadler LS, Braun PV. Nanocomposite science and technology. Weinheim, Germany: John Wiley & Sons; 2006. Kuilla T, Bhadra S, Yao D, Kim NH, Bose S, Lee JH. Recent advances in graphene based polymer composites. Progress in Polymer Science. 2010;35(11):1350-75. Potts JR, Dreyer DR, Bielawski CW, Ruoff RS. Graphene-based polymer nanocomposites. Polymer. 2011;52(1):5-25. Spitalsky Z, Tasis D, Papagelis K, Galiotis C. Carbon nanotube–polymer composites: Chemistry, processing, mechanical and electrical properties. Progress in Polymer Science. 2010;35(3):357-401. Iijima S. Helical microtubules of graphitic carbon. Nature. 1991;354:56-8. Gardea F, Lagoudas DC. Characterization of electrical and thermal properties of carbon nanotube/epoxy composites. Composites Part B: Engineering. 2014;56:611-20. Jyoti J, Basu S, Singh BP, Dhakate SR. Superior mechanical and electrical properties of multiwall carbon nanotube reinforced acrylonitrile butadiene styrene high performance composites. Composites Part B: Engineering. 2015;83:58-65. Tijing LD, Park C-H, Choi WL, Ruelo MTG, Amarjargal A, Pant HR, et al. Characterization and mechanical performance comparison of multiwalled carbon nanotube/polyurethane composites fabricated by electrospinning and solution casting. Composites Part B: Engineering. 2013;44(1):613-9. Jiang Q, Wang X, Zhu Y, Hui D, Qiu Y. Mechanical, electrical and thermal properties of aligned carbon nanotube/polyimide composites. Composites Part B: Engineering. 2014;56:408-12. Teng C, Ma C, Huang Y, Yuen S, Weng C, Chen C, et al. Effect of MWCNT content on rheological and dynamic mechanical properties of multiwalled carbon nanotube/polypropylene composites. Composites Part A: Applied Science and Manufacturing. 2008;39(12):1869-75. Pan Y, Li L, Chan SH, Zhao J. Correlation between dispersion state and electrical conductivity of MWCNTs/PP composites prepared by melt blending. Composites Part A: Applied Science and Manufacturing. 2010;41(3):419-26. Nobel Foundation. Nobel Prize in Physics 2010. 2010. Beech R. Nanoscale Graphene Platelets taking its place as an emerging class of nanomaterials. Nanotechnology Law and Business. 2011;Spring. Kim H, Miura Y, Macosko CW. Graphene/Polyurethane Nanocomposites for Improved Gas Barrier and Electrical Conductivity. Chemistry of Materials. 2010;22(11):3441-50. Rao CNR, Sood AK, Subrahmanyam KS, Govindaraj A. Graphene: The New TwoDimensional Nanomaterial. Angewandte Chemie International Edition. 2009;48(42):775277. Costa P, Silvia C, Viana JC, Lanceros Mendez S. Extruded thermoplastic elastomers styrene–butadiene–styrene/carbon nanotubes composites for strain sensor applications. Composites Part B: Engineering. 2014;57:242-9. Munteanu SB, Brebu M, Vasile C. Thermal and thermo-oxidative behaviour of butadiene– styrene copolymers with different architectures. Polymer Degradation and Stability. 2005;89(3):501-12.

AC C

[1]

30

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

[18] Costa P, Silva J, Sencadas V, Simoes R, Viana JC, Lanceros-Méndez S. Mechanical, electrical and electro-mechanical properties of thermoplastic elastomer styrene–butadiene– styrene/multiwall carbon nanotubes composites. J Mater Sci. 2013;48(3):1172-9. [19] Czajka M, Shanks RA, Kong I. Preparation of graphene and inclusion in composites with poly(styrene-b-butadiene-b-styrene). Science and Engineering of Composite Materials. 2015;22(1):7-16. [20] Lorenz H, Fritzsche J, Das A, Stöckelhuber KW, Jurk R, Heinrich G, et al. Advanced elastomer nano-composites based on CNT-hybrid filler systems. Composites Science and Technology. 2009;69(13):2135-43. [21] Siengchin S, Karger-Kocsis J. Binary and ternary composites of polystyrene, styrene– butadiene rubber and boehmite produced by water-mediated melt compounding: Morphology and mechanical properties. Composites Part B: Engineering. 2013;45(1):1458-63. [22] Wu J-H, Li C-H, Wu Y-T, Leu M-T, Tsai Y. Thermal resistance and dynamic damping properties of poly (styrene–butadiene–styrene)/thermoplastic polyurethane composites elastomer material. Composites Science and Technology. 2010;70(8):1258-64. [23] Bose S, Khare RA, Moldenaers P. Assessing the strengths and weaknesses of various types of pre-treatments of carbon nanotubes on the properties of polymer/carbon nanotubes composites: A critical review. Polymer. 2010;51(5):975-93. [24] Georgakilas V, Otyepka M, Bourlinos AB, Chandra V, Kim N, Kemp KC, et al. Functionalization of Graphene: Covalent and Non-Covalent Approaches, Derivatives and Applications. Chemical Reviews. 2012;112(11):6156-214. [25] Lee C, Wei X, Kysar JW, Hone J. Measurement of the Elastic Properties and Intrinsic Strength of Monolayer Graphene. Science. 2008;321(5887):385-8. [26] Yu M-F, Lourie O, Dyer MJ, Moloni K, Kelly TF, Ruoff RS. Strength and Breaking Mechanism of Multiwalled Carbon Nanotubes Under Tensile Load. Science. 2000;287(5453):637-40. [27] Dreyer DR, Park S, Bielawski CW, Ruoff RS. The chemistry of graphene oxide. Chemical Society reviews. 2010;39(1):228-40. [28] Chen J, Yao B, Li C, Shi G. An improved Hummers method for eco-friendly synthesis of graphene oxide. Carbon. 2013;64(0):225-9. [29] Chen T, Zeng B, Liu JL, Dong JH, Liu XQ, Wu Z, et al. High throughput exfoliation of graphene oxide from expanded graphite with assistance of strong oxidant in modified Hummers method. Journal of Physics: Conference Series. 2009;188(1):012051. [30] Makosza M. Phase-transfer catalysis. A general green methodology in organic synthesis. Pure and Applied Chemistry. 2000;72(7):1399-403. [31] Shen J, Li N, Shi M, Hu Y, Ye M. Covalent synthesis of organophilic chemically functionalized graphene sheets. Journal of Colloid and Interface Science. 2010;348(2):37783. [32] Shen J, Shi M, Ma H, Yan B, Li N, Hu Y, et al. Synthesis of hydrophilic and organophilic chemically modified graphene oxide sheets. Journal of Colloid and Interface Science. 2010;352(2):366-70. [33] Pokharel P, Pant B, Pokhrel K, Pant HR, Lim J-g, Lee DS, et al. Effects of functional groups on the graphene sheet for improving the thermomechanical properties of polyurethane nanocomposites. Composites Part B: Engineering. 2015;78:192-201.

31

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

[34] Ponnamma D, Sadasivuni KK, Strankowski M, Guo Q, Thomas S. Synergistic effect of multi walled carbon nanotubes and reduced graphene oxides in natural rubber for sensing application. Soft Matter. 2013;9(43):10343-53. [35] Kim DH, Yun YS, Jin H-J. Difference of dispersion behavior between graphene oxide and oxidized carbon nanotubes in polar organic solvents. Current Applied Physics. 2012;12(3):637-42. [36] Lu L, Zhou Z, Zhang Y, Wang S, Zhang Y. Reinforcement of styrene–butadiene–styrene tri-block copolymer by multi-walled carbon nanotubes via melt mixing. Carbon. 2007;45(13):2621-7. [37] Liao K, Li S. Interfacial characteristics of a carbon nanotube–polystyrene composite system. Applied Physics Letters. 2001;79(25):4225-7. [38] Liu Y-T, Xie X-M, Ye X-Y. High-concentration organic solutions of poly(styrene-cobutadiene-co-styrene)-modified graphene sheets exfoliated from graphite. Carbon. 2011;49(11):3529-37. [39] Qiu F, Hao Y, Li X, Wang B, Wang M. Functionalized graphene sheets filled isotactic polypropylene nanocomposites. Composites Part B: Engineering. 2015;71:175-83. [40] Uddin ME, Layek RK, Kim NH, Hui D, Lee JH. Preparation and properties of reduced graphene oxide/polyacrylonitrile nanocomposites using polyvinyl phenol. Composites Part B: Engineering. 2015;80:238-45. [41] Wang H, Xie G, Fang M, Ying Z, Tong Y, Zeng Y. Electrical and mechanical properties of antistatic PVC films containing multi-layer graphene. Composites Part B: Engineering. 2015;79:444-50. [42] Schniepp HC, Li J-L, McAllister MJ, Sai H, Herrera-Alonso M, Adamson DH, et al. Functionalized Single Graphene Sheets Derived from Splitting Graphite Oxide. The Journal of Physical Chemistry B. 2006;110(17):8535-9. [43] Yang S-Y, Lin W-N, Huang Y-L, Tien H-W, Wang J-Y, Ma C-CM, et al. Synergetic effects of graphene platelets and carbon nanotubes on the mechanical and thermal properties of epoxy composites. Carbon. 2011;49(3):793-803. [44] Chatterjee S, Nafezarefi F, Tai NH, Schlagenhauf L, Nüesch FA, Chu BTT. Size and synergy effects of nanofiller hybrids including graphene nanoplatelets and carbon nanotubes in mechanical properties of epoxy composites. Carbon. 2012;50(15):5380-6. [45] Li Y-L, Kuan C-F, Chen C-H, Kuan H-C, Yip M-C, Chiu S-L, et al. Preparation, thermal stability and electrical properties of PMMA/functionalized graphene oxide nanosheets composites. Materials Chemistry and Physics. 2012;134(2–3):677-85. [46] Du J, Zhao L, Zeng Y, Zhang L, Li F, Liu P, et al. Comparison of electrical properties between multi-walled carbon nanotube and graphene nanosheet/high density polyethylene composites with a segregated network structure. Carbon. 2011;49(4):1094-100. [47] Martin-Gallego M, Bernal MM, Hernandez M, Verdejo R, Lopez-Manchado MA. Comparison of filler percolation and mechanical properties in graphene and carbon nanotubes filled epoxy nanocomposites. European Polymer Journal. 2013;49(6):1347-53. [48] Dey A, Basudha PK. Applicability of Burger Model in Predicting the Response of Viscoelastic Soil Beds. GeoFlorida 2010: Advances in Analysis, Modeling & Design. 2010:2611-20.

32

ACCEPTED MANUSCRIPT

Figure Captions

Figure 1

Figure 3 FTIR spectra of p -GO, GO and f-G.

Figure 4

M AN U

Schematic diagram of functionalisation process of f-G.

SC

Figure 2

RI PT

TEM images of (a) pristine MWCNT and (b) acid-treated MWCNT (f-MWCNT).

TE D

Digital photos of as-prepared dispersion 1 week after sonication. Photograph showing different solvent compatibility of nanofillers.

EP

Figure 5

SEM images of fractured surfaces of nanocomposites (a) neat SBS (2000×); (b) G0-T1

AC C

(2000×); (c) G1-T0 (2000×); (d) G0-T1 40000×; (e) G1-T0 40000×; (f) G1-T5 (2000×); (g) G1-T1 (2000×); (h) G1-T1 (40000×).

Figure 6 TGA curves of neat SBS and its nanocomposites.

33

ACCEPTED MANUSCRIPT

Figure 7 AC conductivity of neat SBS and its nanocomposites. Right axis corresponds to the values of

RI PT

100 kHz test frequencies.

Figure 8

Mechanical properties of neat SBS and its nanocomposites. Right axis corresponds to the values

SC

of elongation at break.

M AN U

Figure 9

Typical strain decomposition from creep and recovery testing by DMA.

Figure 10

AC C

EP

TE D

Creep recovery curves of neat SBS and its nanocomposites.

34

ACCEPTED MANUSCRIPT

Table Captions

Table 2 Thermal stability of neat SBS and its nanocomposites.

M AN U

Table 3

SC

Experimental parameters investigated in this work.

RI PT

Table 1

AC Conductivity of neat SBS and its nanocomposites at different test frequencies.

Table 4

AC C

EP

TE D

Mechanical properties of neat SBS and its nanocomposites.

35

ACCEPTED MANUSCRIPT

Table 1

SBS G0-T1 G1-T5 G1-T1 G5-T1 G1-T0

0 0 0.5 1.5 2.5 3

f-MWCNT (wt%)

RI PT

f-G (wt%)

0 3 2.5 1.5 0.5 0

AC C

EP

TE D

M AN U

SC

Sample Designation

ACCEPTED MANUSCRIPT

T5 (°C)

T50 (°C)

T95 (°C)

SBS G0-T1 G1-T5 G1-T1 G5-T1 G1-T0

395 397 396 401 398 403

460 467 469 470 467 466

AC C

EP

TE D

M AN U

SC

Sample

RI PT

Table 2

492 535 542 565 534 541

ACCEPTED MANUSCRIPT

Table 3

AC Frequency (S/cm)

9.236E-07 1.882E-06 2.153E-06 1.181E-05 2.555E-06 1.262E-06

1.390E-05 2.130E-05 3.456E-05 3.959E-05 2.014E-05 2.081E-05

SC

10 kHz

AC C

EP

TE D

M AN U

SBS G0-T1 G1-T5 G1-T1 G5-T1 G1-T0

1 kHz

RI PT

Sample 100 kHz

9.237E-05 2.039E-04 3.742E-04 5.025E-04 2.607E-04 1.320E-04

ACCEPTED MANUSCRIPT

Table 4

300% Modulus (MPa)

Tensile Strength (MPa)

SBS G0-T1 G1-T5 G1-T1 G5-T1 G1-T0

3.20 5.10 5.57 4.54 3.97 3.77

21.0 23.0 23.5 23.1 21.3 26.1

SC M AN U

TE D EP AC C

Elongation at break (%)

RI PT

Samples

448 407 559 574 635 650

ACCEPTED MANUSCRIPT

RI PT

Figure 1

Closed end

AC C

EP

TE D

M AN U

100 nm

SC

Open end

100 nm

ACCEPTED MANUSCRIPT

Figure 2

M AN U

SC

RI PT

GO from modified Hummer’s Method:

AC C

EP

TE D

Functionalisation:

ACCEPTED MANUSCRIPT

Figure 3

GO

AC C

EP

TE D

M AN U

SC

f-G

RI PT

p-GO

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

Figure 4

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

Figure 5

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

Figure 6

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

Figure 7

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

Figure 8

ACCEPTED MANUSCRIPT

Recovery

SC

Viscoelastic +

M AN U

Elastic

Elastic

Viscoelastic +

Plastic

EP

TE D

Irreversible creep

AC C

Strain (%)

Creep

RI PT

Figure 9

Time

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

Figure 10