Preparation and characterization of nanocomposites of poly-p-phenylene benzobisthiazole with graphene nanosheets

Abstract

Poly-p-phenylene benzobisthiazole (PBT)/graphene nanocomposite films were fabricated using a simple method with methanesulfonic acid (MSA) as the processing solvent. In this approach, graphene nanosheets were exfoliated in the MSA solution of PBT using ultrasonication and mechanical stirring, and then processed into thin films. The structure and composition of as-prepared graphene oxide (GO) and reduced graphene oxides (rGO) were characterized by Raman spectroscopy and X-ray photoelectron spectroscopy (XPS). The electrical conductivity of pristine PBT was enhanced by 10 orders of magnitude upon incorporation of 5 wt% rGO nanosheets. The enhanced conductivity of the nanocomposites was interpreted by the percolation model. The thermo-oxidative stability of PBT was improved with the incorporation of both GO and rGOs, while rGOs showed more pronounced effect. On addition of only 5 wt% GO, the tensile strength and Young's modulus of PBT increased by ∼4-fold and ∼7-fold, respectively, whereas relatively inferior tensile properties were observed for the PBT/rGO nanocomposites. The enhanced mechanical properties could be attributed to the molecular-level dispersion of the exfoliated GO nanosheets in PBT matrix, as evidenced by the results from morphological studies. A modified Halpin–Tsai model has been used to evaluate the reinforcement or orientation effects of GO/rGO on the Young's modulus of the nanocomposites.

---

1. Introduction

Polymer nanocomposites have attracted significant attention recently owing to their unique chemical and physical properties.¹ The fillers usually used for fabrication of the nanocomposites include exfoliated nanoclays,^2–4 carbon nanotubes (CNTs)^5–7 and graphene nanosheets.^1,8–10 Among the carbon-based nanofillers, CNTs have been explored to improve the mechanical, thermal and electrical properties of several polymers.^11,12 The improvement of tensile strength by ∼15% and elastic modulus by ∼30% were achieved with 1 wt% loading of multi-walled carbon nanotubes (MWCNTs) in epoxy resin.¹³ The reported electrical conductivities of polymer-based CNT composites are not significantly high, for instance, ∼10⁻⁶ S cm⁻¹ for CNT/poly(methyl methacrylate) composites¹⁴ and <1.31 S cm⁻¹ for a CNT/conducting polymer composite at room temperature.¹⁵ However, the major problem with the CNT-reinforced polymer composites is the inhomogeneous dispersion/distribution of CNTs in the matrix.

Graphene has attracted great interest in current research on material science and engineering owing to its exceptional thermal,¹⁶ mechanical,¹⁷ and electrical properties.¹⁸ The high specific surface area (2630–2965 m² g⁻¹)¹⁹ and large aspect ratio (>2000)²⁰ of graphite sheets lead to outstanding reinforcement for polymer matrices. To date, graphene-based filler materials have been applied in fabrication of polymer-based nanocomposites and achieved high reinforcement efficiency.^16,21 The most effective route to the preparation of polymer/graphene nanocomposites generally starts with the synthesis of GO. The exfoliation of graphene oxide in polar solvents atomically produces thin GO sheets, which can be considered as the analogue of the highly conducting crystalline graphene. Chemical or thermal reduction of GO can restore the graphene structure and electrical conductivity.¹⁰ The GO-derived graphene has superior electrical conductivity and stiffness compared to nanoclays.²² Moreover, the mechanical strength and electrical and thermal conductivities of pristine graphene are also comparable to those of single-walled carbon nanotubes (SWCNTs).^23,24 Therefore, the graphene-based nanofillers can be used as a promising reinforcing agent to fabricate polymer-based nanocomposites with superior properties. The electrical conductivity of polyethylene terephthalate (2 × 10⁻¹³ S cm⁻¹) was increased up to 7.4 × 10⁻² S cm⁻¹ with the addition of 2.4 vol.% graphene.²⁵ Zhou et al.²⁶ prepared poly(vinyl alcohol)/graphene nanocomposites by solution blending process, followed by sodium hydrosulfite reduction and found highest DC conductivity of 8.9 × 10⁻³ S cm⁻¹ at 3 wt% graphene loading. Song et al.²⁷ have recently fabricated exfoliated graphene/polypropylene nanocomposites with a 75% increase in yield strength and 74% enhancement in Young's modulus at only 0.42 vol% GO content. In our earlier work, we observed ∼12-fold and ∼18-fold increase of the tensile strength and tensile modulus of polyimide, respectively, upon incorporation of only 5 wt% GO.¹⁰ In our earlier work, the tensile strength and modulus of polyimide were enhanced by 12-fold and 18-fold, respectively, upon addition of 5 wt% GO.

PBT is an aromatic heterocyclic polymer being notable as a high-performance material.^28–30 The aromatic character and high stiffness of PBT chains endue its superior properties like thermal and environmental stability, mechanical strength, and chemical resistance.^31,32 As-extruded PBT films possess tensile modulus of 240 GPa and tensile strength of 1.5 GPa,³³ and exhibits high thermal stability up to 600 °C in nitrogen.³⁴ Although the PBT backbones consist of extended π-conjugation that could facilitate the movement of charge carriers, PBT polymer is a insulating material with conductivity of < 10⁻¹² S cm⁻¹ at room temperature.³⁵ However, there are very limited works on improvement of its electrical and electronic properties. Earlier, the electrical conductivity of PBT enhanced by several orders of magnitude through blending with conducting polymers such polyaniline and polypyrrole.^35,36 Structural modifications of PBT have also been explored to improve its electrical conductivity.^35,37 The aim of the present investigation is to improve the electrical conductivity of PBT without serious compromise with their excellent mechanical strength and high thermal stability. The π–π stacking interaction between conjugated PBT chains and graphene surfaces could improve the electrical conductivity of the prepared nanocomposites. To the best of our knowledge, the improvement of electrical conductivity of PBT by incorporation of graphene nanosheets has not been explored.

In this work, chemically modified graphene reinforced PBT nanocomposites were fabricated through solution blending approach using MSA as a processing solvent for the liquid-phase exfoliation of graphene and to prepare homogenous solution of PBT. The mechanical, thermal and electrical properties of GO reinforced PBT nanocomposites were compared with those of the PBT/rGO nanocomposites. The temperature effects on the electrical conductivity of the prepared nanocomposites were evaluated.

2. Experimental

2.1. Materials

Graphite powder with an average particle size of 6 μm (purity = 99.95%) was obtained from Sigma Aldrich. The PBT polymer (M_w = 36 400) was obtained from SRI International, USA and used as received. The chemicals used for the oxidation of graphite, i.e., sulfuric acid (H₂SO₄), hydrochloric acid (HCl), hydrogen peroxide (H₂O₂), potassium permanganate (KMnO₄), anhydrous aluminum chloride (AlCl₃) and sodium nitrate (NaNO₃) were purchased from Sigma Aldrich, India. Sodium borohydride (NaBH₄) (purity = 98.5%) and MSA were used as received from Sigma Aldrich, India.

2.2. Preparation of graphene oxide and reduced graphene oxide

The modified Hummers method was used to synthesize GO from graphite powder.³⁸ Briefly, dried graphite powder (4 g), NaNO₃ (2 g) and concentrated H₂SO₄ (92 ml) were mixed together and kept stirring at <3 °C for 30 min, and then KMnO₄ (12 g) was added over 1 h period. The mixture was kept stirring for 24 h at room temperature, and then aqueous H₂SO₄ solution (300 ml, 5 wt%) was gradually added to allow the temperature to increase up to 98 °C. The reaction was continued for another 3 h. The color of the solution was turned to a yellowish brown. The mixture was then treated with H₂O₂ (50 ml, 30 wt%) to reduce excess unreacted KMnO₄. The obtained graphene oxide solution was repeatedly washed with dilute HCl solution and deionized (DI) water until the solution became acid free. The solution was then filtered through a membrane filter (50 mm diameter, 0.2 μm pore size) and dried under vacuum at 60 °C. Finally, the graphene oxide (GO) was obtained as a gray powder.

The as-prepared GO was reduced by combined chemical-thermal treatment. At first, 0.5 g GO was dispersed in 250 ml dimethylformamide (DMF) using ultrasonication. 250 ml of NaBH₄ (6 g) and 160 ml of anhydrous AlCl₃ (0.5 g) solutions were separately prepared in DMF. The prepared solutions were mixed together and refluxed at 150 °C for 4 h. The resultant suspension was filtered through a membrane filter and dried at 50 °C for 10 h. In the next step, the chemically reduced GO was treated at 400 °C in argon atmosphere for 1 h.

2.3. Preparation of PBT nanocomposites with GO and rGO

PBT/GO and PBT/rGO nanocomposites were prepared through following method: GO powder was dispersed in MSA (0.6 mg ml⁻¹) under ultrasonication for 2 h at room temperature to obtained GO-MSA suspensions with GO concentrations of 1, 3 and 5 wt%. Polymer solution was prepared by dissolving pre-determined amount of PBT in MSA. The GO-MSA suspension was slowly added to the PBT solution under vigorous stirring for 6 h. A black color viscous solution was obtained. The homogenous PBT/GO solution was poured onto a cleaned glass plate, and the nanocomposite film was cast by drawing a glass bar over the plate. The ∼20 μm thick film was peeled off from the glass substrate after immersing it into a tri-ethylamine/ethanol bath. The preparation of PBT/GO nanocomposites is presented in Scheme 1. The PBT/rGO nanocomposite films were prepared by same method as followed above.

Scheme 1 Illustration of the process for preparing PBT/GO nanocomposite films.

2.4. Characterization

Raman spectroscopic measurements were performed with a JobinYvon HORIBA Raman spectrometer, using a He–Ne laser operating at 632.8 nm as the excitation source. XPS spectra of the nanocomposites samples were taken on Perkin-Elmer PHI 5500 ESCA System, using a monochromatized Al Kα X-ray source (1486.6 eV). Wide-angle X-ray diffraction (WAXD) patterns of the nanocomposites were collected from a Rigaku Ultima IV X-ray diffractometer (Cu Kα radiation with λ = 1.54 Å) using an accelerated voltage of 40 kV and current of 40 mA. The morphology of the nanocomposites was studied by field emission scanning electron microscopy (FE-SEM, Zeiss-Leo Model 1530) with an acceleration voltage of 15 kV, and transmission electron microscopy (TEM, JEOL JEM 2100–2100F) operated at 200 kV. Cryogenically fracture surfaces of the nanocomposite samples were sputtered with gold prior to SEM analysis. For TEM analysis, thin sections were cut from the nanocomposite films using a Leica ultramicrotome with a diamond knife. The thermal stability of the nanocomposites was measured by thermal gravimetric analysis (TGA, TA, Q 400) in both nitrogen and air atmospheres with heating rate of 5 °C min⁻¹. The tensile measurements were carried out with an Instron 5848 Micro Tester at a crosshead speed of 1 mm min⁻¹. The electrical conductivity (σ) was determined using standard four-point probe technique using a Keithley 797A and calculated from the equation: σ = 1/tR_s (S cm⁻¹). The sample thickness (t) and film resistance (R_s) were measured using caliper and a four-point probe, respectively.

3. Results and discussions

The structure and composition of as-prepared GO and rGO were analyzed by XPS. Fig. 1a shows the XPS spectra of GO and rGO. The O 1s peak intensity for GO is significantly higher than that of graphite, indicating that many oxygen containing groups are introduced onto the graphite surfaces during oxidation process, and these oxygen-containing groups disrupt the sp²-carbon network of the graphite. The significantly lower carbon-to-oxygen (C/O) atomic ratio for GO indicating a fairly high oxidation degree on the GO sheets. As seen in the Fig. 1a, the chemical–thermal reduction treatment largely suppressed the intensity of the O 1s peak (532 eV), while the peak corresponding to C 1s (284.2 eV) dominated. The C/O atomic ratio of GO is significantly increased from 2.16 to 9.54 after the two-step reduction treatment. The above results clearly indicates that the majority of oxygen functionalities are removed during the chemical–thermal reduction of GO. Fig. 1b shows the Raman spectra of GO and rGO and the changes in relative intensity of two main peaks, i.e., D and G peaks, after reduction of GO. The D band of GO located at 1354 cm⁻¹ and that of rGO at 1348 cm⁻¹ arises from a defect-induced breathing mode of sp² rings.³⁹ The G band at around 1592 cm⁻¹ for GO and at 1587 cm⁻¹ for rGO is due to the first order scattering of the E_2g mode.³⁹ The intensity of the D peak is related to the size of the in-plane sp² domains.⁴⁰ The intensity ratio of D and G peaks (I_D/I_G) is inversely related to the average size of the sp² clusters.^39,40 After GO was reduced to rGO, the I_D/I_G value of GO significantly increased from 0.92 to 1.51. The higher intensity ratio (I_D/I_G) of rGO than GO is attributed to the elimination of oxygen-functionalities and formation of highly defected carbon lattice upon reduction of GO.⁴¹ Hence, the chemical-thermal reduction treatment removed oxygen-functionalities from the GO nanosheets with the greatest efficiency.

Fig. 1 XPS (a) and Raman (b) spectra of pristine graphite, GO and rGO.

Fig. 2 shows the XPS spectra of pristine PBT and its nanocomposites. The atomic concentrations (%) of different elements obtained from XPS spectra of the nanocomposite samples are presented in Table 1. From the results, it is obvious that the oxygen concentration on the PBT/GO nanocomposite surface is higher than that of pristine PBT that is related to the surface oxygen functional groups in the GO. However, the lower surface oxygen concentration in the PBT/rGO nanocomposites compared to PBT/GO confirmed the reduction of oxygen functionalities in the rGO nanosheets. The chemical environment of the N-atom in PBT backbone was changed upon incorporation of graphene nanosheets into PBT matrix, as indicated by the shifting of the N 1s peak toward higher binding energies for nanocomposite samples (inset Fig. 2). The observed binding energy shift for the PBT/GO nanocomposites is clearly indicating the presence of an effective interfacial interaction between the N-heterocylic moieties in PBT backbone and the oxygen containing functional groups of GO (Table 2).

Fig. 2 Full range XPS spectra of pristine PBT and its nanocomposites containing different concentrations of GO and rGO. The inset shows high resolution N 1s XPS spectra of pristine PBT and its nanocomposites.

Table 1 The atomic concentrations (%) of different elements obtained from XPS spectra of the nanocomposites

Sample	Atomic (%)
	C 1s	O 1s	N 1s	S 2s + S 2p
Pure PBT	80.1	1.1	9.4	9.4
PBT/GO (1%)	76.6	10.0	7.2	6.2
PBT/GO (5%)	65.2	19.6	6.9	8.3
PBT/rGO (1%)	82.1	3.7	7.3	6.9
PBT/rGO (5%)	81.0	6.6	6.1	6.3

---

Table 2 WAXD results for pristine PBT, PBT/GO and PBT/rGO nanocomposites

Sample	Side-to-side interchain distance (nm)	Face-to-face interchain distance (nm)
Pure PBT	0.2834 (2θ = 15.78)	0.1705 (2θ = 26.87)
PBT/GO (1%)	0.2953 (2θ = 15.13)	0.1731 (2θ = 26.44)
PBT/GO (5%)	0.3108 (2θ = 14.36)	0.1775 (2θ = 25.74)
PBT/rGO (1%)	0.3245 (2θ = 13.74)	0.1757 (2θ = 26.02)
PBT/rGO (5%)	0.3421 (2θ = 13.02)	0.1802 (2θ = 25.32)

---

Raman spectra of pristine PBT and its nanocomposites are presented in Fig. 3. The Raman spectrum of the pristine PBT film exhibits four well-defined Raman-active bands at 1598, 1526, 1482 and 1285 cm⁻¹. The band at 1598 cm⁻¹ is assigned to stretching vibration of the phenylene ring in the heterocycle (benzobisthiazole). The peak at 1526 cm⁻¹ related to stretching vibration of the phenylene ring coupled with benzene ring deformation in the benzobisthiazole group. The peak at 1285 cm⁻¹ corresponds to stretching vibration of the C–C bond connecting benzobisthiazole and phenylene. In the Raman spectra of PBT/GO and PBT/rGO nanocomposites, the characteristic Raman bands of the PBT matrix along with two typical graphitic bands (D band at ∼1352 and G band at ∼1572 cm⁻¹) are detected. As noticed in the Raman spectra of nanocomposites (Fig. 3), the characteristic vibrational bands of the PBT matrix are shifted towards higher frequencies with respect to those of the pristine PBT, which is a consequence of physical constraints introduced to the polymer chains by the presence of GO/rGO nanosheets. The blue shift could be attributed to the formation of non-covalent π–π interaction between conjugated PBT backbones and basal planes of graphene. The Raman spectra of the PBT/rGO nanocomposites exhibit higher extent of blue shift compared to PBT/GO nanocomposites, indicating that the rGO plane with more graphitic sp² network might strengthen the π–π interactions compared to GO nanosheets. As shown in Fig. 1b and 3, the Raman peak intensities of the D and G bands are relatively weaker for the nanocomposite systems compared to those observed for the neat GO/rGO, probably due to the lower content of GO/rGO in the nanocomposites. The blue shift of D band from 1354 cm⁻¹ for neat GO to 1365 cm⁻¹ for PBT/GO (5 wt%) nanocomposites is indicating better exfoliation of the GO nanosheets in PBT matrix.⁴² The similar blue shift phenomenon is also observed for the PBT/rGO nanocomposites.

Fig. 3 Raman spectra of (a) pristine PBT and its nanocomposites containing (b) 1 wt% GO, (c) 5 wt% GO, (d) 1 wt% rGO and (e) 5 wt% rGO.

The microstructure of the cryogenic fracture surface of pristine PBT and its nanocomposites containing 5 wt% GO and rGO was characterized by SEM, shown in Fig. 4. As shown in Fig. 4a, the pristine PBT is characterized with a smooth and featureless fracture surfaces. In contrast, the SEM images of the nanocomposites reveal rough surfaces with several protruded GO or rGO nanosheets (Fig. 4b and c). The rough fracture surface might be a result of crack distortion and thus, absorb more energy during fracture. As shown in the SEM image of the PBT/GO nanocomposite, the wrinkled GO sheets are uniformly distributed in PBT matrix and align parallel to the film surface. The boundary between the PBT matrix and the GO is obscured as the dispersed GO nanosheets are covered with a thick polymer layer, implying the interfacial adhesion induced by the surface functional groups of GO and favorable π–π interactions between PBT matrix and graphene.⁴³ The interfacial interaction favors stress transfer between polymer matrix and graphene sheets, leading to enhance the mechanical properties of the nanocomposites (Table 4). The rGO nanosheets are also well dispersed in the PBT matrix (Fig. 4c). The favorable dispersion of GO/rGO in PBT matrix is also verified by the TEM analysis, shown in Fig. 5. Fig. 5a and b display a homogeneous dispersion of GO and rGO nanosheets in the PBT matrix, respectively, which is similar to those observed in the SEM images. The GO nanosheets appear highly oriented with almost no large agglomerates. The preferential orientation of GO sheets could be ascribed to the better interfacial interactions between the PBT matrix and GO. The TEM image of rGO-filled nanocomposite exhibit some multilayer sheets (lower aspect ratio) (Fig. 5b), suggesting that the reduction process leads to partial restacking of the graphene sheets. In addition, the removal of oxygen functionalities weakens the polar interaction between rGO nanosheets and MSA molecules during liquid-phase exfoliation process. As a result, there are some non-exfoliated rGO sheets in the nanocomposite.

Fig. 4 SEM images of cryogenic fractured surfaces of (a) pristine PBT, (b) PBT/GO (5 wt%) and PBT/rGO (5 wt%) nanocomposites. The inset shows low magnification images.

Table 3 Selected results of TGA measurements

Sample	Nitrogen atmosphere			Air atmosphere
	Tonset (°C)	Tmax (°C)	Char yield at 900 °C (wt%)	Tonset (°C)	Tmax (°C)	Char yield at 900 °C (wt%)
Pure PBT	612	664	63	497	513	0.02
PBT/GO (1%)	671	751	58	548	582	0.14
PBT/GO (3%)	633	719	57	530	566	0.32
PBT/GO (5%)	620	703	51	524	552	0.54
PBT/rGO (1%)	674	758	63	556	593	8.72
PBT/rGO (3%)	706	774	61	576	617	3.09
PBT/rGO (5%)	719	798	63	614	645	6.00

---

Table 4 Mechanical properties pristine PBT and its nanocomposites with GO and rGO

Sample	Tensile strength (MPa)	% Increase of tensile strength	Young's modulus (GPa)	% Increase of Young's modulus	% Elongation
Pure PBT	700 ± 12	—	2.24 ± 0.04	—	8.96 ± 0.12
PBT/GO (1%)	1404 ± 17	100	4.31 ± 0.04	92	4.17 ± 0.16
PBT/GO (3%)	2029 ± 11	189	10.61 ± 0.02	473	2.98 ± 0.26
PBT/GO (5%)	2667 ± 21	281	14.85 ± 0.08	562	1.81 ± 0.22
PBT/rGO (1%)	1006 ± 15	44	3.74 ± 0.06	67	4.72 ± 0.21
PBT/rGO (3%)	1168 ± 27	67	5.77 ± 0.09	158	3.49 ± 0.36
PBT/rGO (5%)	1529 ± 24	118	8.96 ± 0.12	300	3.56 ± 0.32

---

Fig. 5 (a) TEM image of PBT/GO (5 wt%) nanocomposite, (b) TEM image of PBT/rGO (5 wt%) nanocomposite, and (c) typical tapping-mode AFM image of graphene sheets deposited on mica substrate from an MSA dispersion.

The crystalline structures of pristine PBT and its nanocomposite films were characterized by WAXD, the results of which are presented in Fig. 6. The WAXD pattern of pristine PBT showed two typical diffraction peaks at 2θ = ∼15.7° (labeled as peak A) and ∼26.8° (labeled as peak B), corresponding to the (200) and (010) crystalline plane, respectively. The periodicity for peak A and peak B stand for the side-to-side and face-to-face distances between two adjacent polymer chains, respectively. For pristine PBT, the side-to-side and face-to-face distances are 0.283 and 0.170 nm, respectively, calculated by using Bragg's equation (nλ = 2d sin θ). The WAXD spectra of PBT/GO or rGO nanocomposites display two crystalline peaks for PBT matrix without any new diffraction peaks. This observation suggests that the GO and rGO nanosheets are homogeneously dispersed in the PBT matrix and the crystalline structure of PBT largely remains in the presence of graphene. Compared with pristine PBT, the diffraction peaks of the nanocomposites shifted toward lower 2θ values indicating an increase in the interchain/interplanar distances between two neighboring PBT chains. The π–π stacking interactions between the graphene plane and the conjugated PBT chain weaken the interactions between PBT backbones and enlarge the interchain distances. The larger interchain distances for PBT/rGO nanocomposites suggest that the π–π interactions between the PBT macromolecule and the rGO basal plane is stronger than those between PBT and GO. This stronger π–π interaction could ensure higher electrical conductivity of the PBT/rGO nanocomposites.

Fig. 6 WAXD patterns of pristine PBT and its nanocomposites containing 1 and 5 wt% GO and rGO.

DC electrical conductivities of the PBT/GO or rGO nanocomposites before and after thermal annealing were measured with a four-probe technique. The high thermal stability of the PBT polymer was allowed to optimize the effects of thermal annealing on electrical conductivity over a wide range of temperatures. The DC conductivity of the nanocomposite films plotted as a function of annealing temperature is shown in Fig. 7. The PBT/GO nanocomposites exhibit remarkable enhancement of DC conductivity upon thermal treatment, but the extent of improvement is dependent on the GO content. As annealing temperature raise from 25 to 400 °C, the conductivity of the PBT/GO nanocomposites containing 1 wt% GO increased by three orders of magnitude, whereas the conductivity of the nanocomposite with 5 wt% GO content enhanced by six orders of magnitude. The thermal annealing process has significantly restored the π-conjugated network on the GO's basal plan, i.e. conversion of sp³-carbon to sp²-carbon, which in turn increased the electrical conductivity of the GO nanosheets and likely promotes the π–π interactions with PBT macromolecules to facilitate electron transport in the nanocomposites with improved conductivity. In contrast to GO-filled nanocomposites, the PBT/rGO nanocomposites reveal only a small improvement in conductivity (i.e., two orders of magnitude) upon thermal annealing, shown in Fig. 7b. This could be attributed that the major fraction of oxygen functionalities in the rGO nanosheets were eliminated during the chemical–thermal reduction process, and small residual fraction of unreduced groups were thermally reduced during annealing process. The low order improvement of conductivity after being thermal annealing suggests that the rGO nanosheets could be used as efficient conductive nanofiller to fabricate nanocomposites materials with stable electrical conductivity over a wide temperature range. The PBT/rGO nanocomposite with higher rGO loading exhibits greater extent of increment in electrical conductivity with increasing annealing temperature, which could be attributed to the presence of higher concentration of unreduced carbon moieties (sp³-carbon) on rGO surfaces and the rearrangement of graphene sheets.⁴⁴

Fig. 7 Effects of thermal annealing on the electrical conductivities of formulated (a) PBT/GO and (b) PBT/rGO nanocomposites. Insets show the electrical conductivity as a function of graphene content for PBT/GO and PBT/rGO nanocomposites.

The electrical conductivity of the nanocomposite films plotted as function of GO/rGO content is shown in the inset of Fig. 7. The pristine PBT (σ ≈ 10⁻¹² S cm⁻¹) is an insulating material.³⁵ As shown in the inset of Fig. 7, the incorporation GO into PBT matrix increased the electrical conductivity by four orders of magnitude up to 6.78 × 10⁻⁸ S cm⁻¹, which could be attributed to higher conductivity of the GO (i.e., ∼10⁻⁵)^45,46 compared to PBT. The PBT/rGO nanocomposites with 5 wt% rGO exhibits 10¹⁰ times enhancement in electrical conductivity when compared to the pristine PBT. This could be ascribed to the favorable electron transport between rGO nanosheets via tunneling through the PBT layers. The π–π stacking interaction between the π-orbitals of conjugated PBT chains and the sp²-orbitals in the rGO plane probably reduced the electron transport barrier energy between the rGO nanosheets. The conductivity of the PBT/rGO nanocomposite sharply increased with increasing rGO content, whereas the enhancement was relatively less for the GO loaded nanocomposites. The most likely reasons for this observation are: (i) the rGO nanosheets have high electrical conductivity compared to GO and (ii) the higher fraction of sp³-carbon in the GO plane restricted π–π stacking interaction with the PBT chains and thus hampered the formation of conducting network in the nanocomposite. The highest conductivity of 7.42 S cm⁻¹ was achieved for the thermally annealed (at 400 °C) PBT/rGO nanocomposite with 5 wt% rGO content, and this conductivity value is higher than that of the many other polymer/graphene nanocomposites.^9,47–51

The percolation theory has been used to describe the dependence of the electrical conductivity on filler (GO/rGO) volume fraction in the ϕ > ϕ_c region. Above the conductivity percolation threshold, the electrical conductivity (σ) of the nanocomposite based on the power law model is generally expressed as:

σ = σ_f(ϕ − ϕ_c)^ν (1)

where σ_f, ϕ, and ν are the electrical conductivity of filler, the volume fraction of GO/rGO, and the critical exponent describing the rapid variation of σ near percolation threshold (ϕ_c), respectively. For conversion from weight fraction to volume fraction, the GO density of 2.13 g cc⁻¹ as measured by a pycnometry method and the rGO density of 2.28 g cc⁻¹ as taken from the literature^52,53 were used. The best fit of the electrical conductivity values using the power law equation [eqn (1)] is depicted in Fig. 8. In Fig. 8, both PBT/GO and PBT/rGO nanocomposites exhibited better co-relation between the DC conductivity and the GO/rGO volume fraction, as could be deduced from the linear correlation coefficients (R²) of 0.9868 and 0.9916, respectively. The fitting results showed that the electrical percolation threshold (ϕ_c) values of the PBT/GO and PBT/rGO nanocomposites are 0.0031 (0.66 wt%) and 0.0024 (0.54 wt%), respectively. These ϕ_c values are on the same order with other polymer/graphene nanocomposite systems.^54–57 However, the present threshold values are substantially lower than the recently reported value for poly(p-phenylene benzobisoxazole) (PBO)/graphene nanocomposite (i.e., 3.7 wt%).⁵⁸ The ν value of 1.919 for PBT/GO nanocomposite is close to the universal value of 2.0 for 3D conductive network and in agreement with the prediction of the percolation theory. The fitting result for PBT/rGO nanocomposite yielded a relatively high ν value, i.e., 2.878, indicating that the electron conduction arises from tunneling among individual graphene sheets separated by the non-conductive polymer matrix, i.e., the sheets form a 3D conductive network.^59,60 The reduced tunneling resistance is likely to be a major factor for the lower percolation threshold observed for the PBT/rGO nanocomposite.^52,54

Fig. 8 The best fitting curves for PBT/GO and PBT/rGO nanocomposites using eqn (1).

The thermal behavior of PBT and its nanocomposites containing different concentrations of GO and rGO was studied in both non-oxidative and oxidative conditions by using TGA. Fig. 9a and b display the TG curves of the pristine PBT and its nanocomposite samples in nitrogen and air atmosphere, respectively. Table 3 presents the results obtained from TGA measurements. The TGA of the pristine PBT showed a typical one-step decomposition with an onset temperature of 612 °C at 1 wt% loss. The sharp weight loss at above 600 °C is associated with the catastrophic decomposition of PBT polymer. The presence of GO/rGO nanosheets noticeably improved the thermal stability of PBT, as both the onset degradation temperature (T_onset) and maximum degradation temperature (T_max) are shifted toward higher temperatures and the decomposition rate becomes substantially slower. This could be attributed to the formation of an inflammable GO/rGO network in the polymer matrix, which act as an effective barrier to inhibit the emission of decomposition products during combustion.⁶¹ In addition, the high thermal conductivity of graphene sheets might facilitate heat dissipation within nanocomposites and consequently improved the thermal stability of the nanocomposites. The thermal degradation of PBT/GO nanocomposites in N₂ atmosphere involves multiple stages of weight loss: the initial weight loss at the temperature region of 140–230 °C is ascribed to the decomposition of labile oxygen functional groups on the GO surface, while the slow weight loss in the second region (240–350 °C) is attributed to the removal of more stable oxygen functionalities. It is noteworthy that the weight loss occurring in the PBT/rGO nanocomposites is significantly smaller than that of the PBT/GO nanocomposites in the range of 140–380 °C. This indicates that the majority of oxygen containing groups in the rGO nanosheets were already removed during two step reduction process, and remaining small fraction of unreduced groups were subjected to thermal decomposition during TG analysis. At the temperature above 620 °C, the decomposition of the nanocomposites is associated with the catastrophic degradation of the PBT matrix. The PBT/rGO nanocomposites have ∼50 °C higher T_max and 6–10% higher char yield compared to PBT/GO nanocomposites. The results suggest that the PBT/rGO nanocomposites have significant potential as a high performance material. The removal of oxygen functionalities during reduction process leads to decrease the defect density on the rGO surface and thus improves the thermal stability of the individual rGO nanosheets, which is consistent with the thermal stability of PBT/rGO nanocomposites.

Fig. 9 TG curves of pristine PBT and its nanocomposites in (a) nitrogen and (b) air atmosphere.

The degradation patterns of the PBT/GO or rGO nanocomposites in air atmosphere are nearly identical to those observed in N₂. The nanocomposites appeared to be thermally more stable than pristine PBT in air atmosphere. The PBT/rGO nanocomposites exhibit higher degradation temperatures (T_onset & T_max) with more effective improvement in the char yield, interfering that the rGO-filled PBT nanocomposites has better thermo-oxidative stability compared to GO-reinforced nanocomposites.

Fig. 10 illustrates the stress–strain curves of the pristine PBT and its nanocomposite films. The slope of the stress–strain curves increases with increasing the graphene (GO/rGO) concentration. The tensile properties of the pristine PBT and formulated nanocomposites are tabulated in Table 4. The tensile properties of PBT were significantly enhanced by incorporation graphene. For instance, the incorporation of 5 wt% GO increased the tensile strength and Young's modulus to 2667 MPa and 14.85 GPa, respectively, corresponding to increases of ∼4 and ∼7 times higher than that of the pristine PBT. The superior tensile properties of the PBT/GO nanocomposites could be attributed to the effective stress transfer from the polymer matrix to the graphene nanosheets. Moreover, the interfacial adhesion between the PBT chains and the GO nanosheets might promote the polymer reinforcing efficiency of GO to produce nanocomposites with superior mechanical performance. The good dispersion of GO within PBT matrix and strong interfacial interaction restrained the segmental movement of the polymer chains upon application of the tensile stress, which lead to decreased the elongation at break of the nanocomposite films but enhanced the modulus. The tensile strength of PBT/GO nanocomposites increased from 1404 to 2667 MPa (∼90% increase) as the GO content increased from 1 to 5 wt%. Similarly, the Young's modulus of the nanocomposites enhanced from 4.31 to 14.85 GPa with increasing GO loading from 1 to 5 wt%. The incorporation of rGO nanosheets also increased the tensile strength and modulus of the PBT polymer, whereas the extent of enhancement of the tensile properties was found to be less upon reinforcement of the rGO compared to that of GO (Fig. 10 and Table 4). The limited enhancement in tensile strength and modulus of the PBT/rGO nanocomposites might be a consequence of the incomplete exfoliation of rGO nanosheets and/or the presence of few-layer stacking of rGO nanosheets in the matrix, as shown in the TEM image (Fig. 5b). These results suggest that the reinforcing efficiency of the GO nanosheets is higher than that of the rGO into PBT.

Fig. 10 Tensile stress–strain curves of pristine PBT and its nanocomposites.

The Halpin–Tsai model, which is most widely used model to estimate the modulus of the nanocomposites with unidirectional or randomly distributed graphene sheets, was used to simulate the modulus of the PBT/GO or rGO nanocomposites. For the random distribution of graphene sheets, the modified form of the Halpin and Tsai equation can be written as:^62,63

wherewhere the E_r, E_g and E_p are Young's modulus of the nanocomposite, graphene and polymer, respectively. Young's modulus of GO and rGO are taken as 207.6 GPa⁶⁴ and 1060 GPa,^52,65 respectively. l_g, t_g and V_g are the length, thickness and volume fraction of the graphene. The thickness and average length of graphene nanosheets are about 1.14 nm and 1.25 μm as determined from the atomic force microscope (AFM) analysis (Fig. 5c). For unidirectional (parallel) orientation of graphene nanosheets, the modified Halpin–Tsai equation is expressed as:

A comparison between the experimental data and predicted Young's modulus values is demonstrated in Fig. 11. For PBT/GO nanocomposites, the experimental E values are close to the theatrical simulation for the hypothesis that graphene sheets are aligned parallel to the surface of nanocomposite (Fig. 11a). This agreement is clearly evidenced by the SEM and TEM images of the nanocomposites. The unidirectional alignment of GO nanosheets can provide efficient stress transfer from the polymer matrix to graphene nanosheets through effective interfacial phase. In the case of PBT/rGO nanocomposites, the experimental Young's modulus is consistent with that predicted by unidirectional Halpin–Tsai model at relatively low graphene content (≤1 wt%) (Fig. 11b). However, the experimental data was found below the calculated one at higher rGO content, which could be attributed to the aggregation or stacking of the rGO nanosheets in the nanocomposites. The π–π stacking interactions between adjacent rGO nanosheets compete with that between the rGO plane and PBT macromolecules, which results few-layer stacking of rGO nanosheets as observed in the TEM images of the nanocomposite (Fig. 5b).

Fig. 11 Comparison between the theoretical values predicted by the Halpin–Tsai model and the experimental Young's modulus data obtained for (a) PBT/GO and (b) PBT/rGO nanocomposites.

4. Conclusions

In this work, PBT based nanocomposites with exfoliated GO and reduced GO were fabricated through solution blending approach, using MSA as a solvent for the liquid-phase exfoliation of graphene nanosheets in the PBT solution. The PBT/rGO nanocomposite films were found to possess high electrical conductivity with a low percolation threshold of 0.54 wt%. The electrical conductivity PBT/rGO (5 wt%) nanocomposite was recorded ∼10¹⁰ and ∼10⁶ times higher than that of the PBT and PBT/GO nanocomposite, respectively. The conductivity of the PBT/GO (5 wt%) nanocomposite increased by more than six orders of magnitude after being annealed at 400 °C, while that of PBT/rGO (5 wt%) by only two orders of magnitude. The thermal stability of PBT, under both non-oxidative and oxidative atmospheres, is remarkably improved with the addition of graphene. The tensile strength and Young's modulus of PBT/GO (5 wt%) was ∼4 and ∼7 times higher than those of the pristine PBT. Therefore, the prepared PBT/GO or rGO nanocomposites could be potentially used in various electronic/electrical applications.

References

1. T. Ramanathan, A. A. Abdala, S. Stankovich, D. A. Dikin, M. Herrera-Alonso, R. D. Piner, D. H. Adamson, H. C. Schniepp, X. Chen, R. S. Ruoff, S. T. Nguyen, I. A. Aksay, R. K. Prud'Homme and L. C. Brinson, Nat. Nanotechnol., 2008, 3, 327–333.
2. M. Baniassadi, A. Laachachi, F. Hassouna, F. Addiego, R. Muller, H. Garmestani, S. V. Ahzi, V. Toniazzo and D. Ruch, Compos. Sci. Technol., 2011, 71, 1930–1935.
3. J. Faucheu, C. Gauthier, L. Chazeau, J. Y. Cavaillé, V. Mellon and E. B. Lami, Polymer, 2010, 51, 6–17.
4. J. F. Feller, S. Bruzaud and Y. Grohens, Mater. Lett., 2004, 58, 739–745.
5. N. Grossiord, M. E. L. Wouters, H. E. Miltner, K. Lu, J. Loos, B. V. Mele and C. E. Koning, Eur. Polym. J., 2010, 46, 1833–1843.
6. S. H. Park and P. R. Bandaru, Polymer, 2010, 51, 5071–5077.
7. N. G. Sahoo, S. Rana, J. W. Cho, L. Li and S. H. Chan, Prog. Polym. Sci., 2010, 35, 837–867.
8. T. Kuilla, S. Bhadra, D. Yao, N. H. Kim, S. Bose and J. H. Lee, Prog. Polym. Sci., 2010, 35, 1350–1375.
9. J.-H. Park, A. Choudhury, B. L. Farmer, T. D. Dang and S.-Y. Park, Polymer, 2012, 53, 3937–3945.
10. H. W. Ha, A. Choudhury, T. Kamal, D. H. Kim and S. Y. Park, ACS Appl. Mater. Interfaces, 2012, 4, 4623–4630.
11. M. Moniruzzaman and K. I. Winey, Macromolecules, 2006, 39, 5194–5205.
12. B. Fiedler, F. H. Gojny, M. H. G. Wichmann, M. C. M. Nolte and K. Schulte, Compos. Sci. Technol., 2006, 66, 3115–3125.
13. J. Zhu, J. D. Kim, H. Peng, J. L. Margrave, V. N. Khabashesku and E. V. Barrera, Nano Lett., 2003, 3, 1107–1113.
14. N. R. Raravikar, A. S. Vijayaraghavan, P. Keblinski, L. S. Schadler and P. M. Ajayan, Small, 2005, 1, 317–320.
15. W. Feng, X. D. Bai, Y. Q. Lian, J. Liang, X. G. Wang and K. Yoshino, Carbon, 2003, 41, 1551–1557.
16. Y. Zhu, S. Murali, W. Cai, X. Li, J. W. Suk, J. R. Potts and R. S. Ruoff, Adv. Mater., 2010, 22, 3906–3924.
17. A. K. Geim and K. S. Novoselov, Nat. Mater., 2007, 6, 183–191.
18. O. C. Compton and S. B. T. Nguyen, Small, 2010, 6, 711–723.
19. H. K. Chae, D. Y. Siberio-Perez, J. Kim, Y. Go, M. Eddaoudi, A. J. Matzger, M. O'Keeffe and O. M. Yaghi, Nature, 2004, 427, 523–527.
20. S. Stankovich, D. A. Dikin, R. D. Piner, K. A. Kohlhaas, A. Kleinhammes, Y. Y. Jia, Y. Wu, S. T. Nguyen and R. S. Ruoff, Carbon, 2007, 45, 1558–1565.
21. Z. Xu and C. Gao, Macromolecules, 2010, 43, 6716–6723.
22. M. Alexandre and P. Dubois, Mater. Sci. Eng., R, 2000, 28, 1–63.
23. D. Li and R. B. Kaner, Science, 2008, 320, 1170–1171.
24. E. T. Thostenson, C. Y. Li and T. W. Chou, Compos. Sci. Technol., 2005, 65, 491–516.
25. H. B. Zhang, W. G. Zheng, Q. Yan, Y. Yang, J. W. Wang, Z. H. Lu, G. Y. Ji and Z. Z. Yu, Polymer, 2010, 51, 1191–1196.
26. T. Zhou, F. Chen, C. Tang, H. Bai, Q. Zhang, H. Deng and Q. Fu, Compos. Sci. Technol., 2011, 71, 1266–1270.
27. P. Song, Z. Cao, Y. Cai, L. Zhao, Z. Fang and S. Fu, Polymer, 2011, 52, 4001–4010.
28. A. L. Rusanov and L. G. Komarova, Polym. Sci.: A Comp. Ref., 2012, 5, 537–596.
29. T. Zhang, J. Jin, S. Yang, G. Li and J. Jiang, Carbohydr. Polym., 2009, 78, 364–366.
30. B. Song, L. H. Meng and Y. D. Huang, Appl. Surf. Sci., 2012, 258, 5505–5510.
31. S. H. Hsiao and W. T. Chen, J. Polym. Sci., Part A: Polym. Chem., 2003, 41, 914–921.
32. X. D. Hu, S. E. Jenkins, B. G. Min, M. B. Polk and S. Kumar, Macromol. Mater. Eng., 2003, 288, 823–843.
33. L. Feldman, R. S. Farris and E. L. Thomas, J. Mater. Sci., 1985, 20, 2719–2726.
34. K. E. Newman, P. Zhang, L. J. Cuddy and D. L. Allara, J. Mater. Res., 1991, 6, 1580–1594.
35. L. S. Tan, K. R. Srinivasan, S. J. Bai and R. J. Spry, J. Polym. Sci., Part A: Polym. Chem., 1998, 36, 713–724.
36. S. Kim, D. A. Cameron, Y. Lee, J. R. Reynolds and C. R. Savage, J. Polym. Sci., Part A: Polym. Chem., 1996, 34, 481–492.
37. S. J. Bai, K. R. Srinivasan, L. S. Tan, R. J. Spry and G. E. Price, J. Appl. Phys., 1999, 85, 280–286.
38. W. S. Hummers and R. E. Offeman, J. Am. Chem. Soc., 1958, 80, 1339.
39. A. C. Ferrari and J. Robertson, Phys. Rev. B: Condens. Matter Mater. Phys., 2000, 61, 14095–14107.
40. Y. Guo, X. Sun, Y. Liu, W. Wang, H. Qiu and J. Gao, Carbon, 2012, 50, 2513–2523.
41. J. I. Paredes, R. S. Villar, F. P. Solis, A. A. Martinez and J. Tascon, Langmuir, 2009, 25, 5957–5968.
42. S. Qin, D. Qin, W. T. Ford, J. E. Herrera and D. E. Resasco, Macromolecules, 2004, 37, 9963–9967.
43. Z. H. Tang, Y. D. Lei, B. C. Guo, L. Q. Zhang and D. M. Jia, Polymer, 2012, 53, 673–680.
44. B. H. Cipriano, A. K. Kota, A. L. Gershon, C. J. Laskowski, T. Kashiwagi, H. A. Bruck and S. R. Raghavan, Polymer, 2008, 49, 4846–4851.
45. C. J. Kim, W. Khan and S. Y. Park, Chem. Phys. Lett., 2011, 511, 110–115.
46. S. Park, J. An, R. D. Piner, I. Jung, D. Yang, A. Velamakanni, S. T. Nguyen and R. S. Ruoff, Chem. Mater., 2008, 20, 6592–6594.
47. J. Liang, Y. Wang, Y. Huang, Y. Ma, Z. Liu, J. Cai, C. Zhang, H. Gao and Y. Chen, Carbon, 2009, 47, 922–925.
48. E. Tkalya, M. Ghislandi, A. Alekseev, C. Koning and J. Loos, J. Mater. Chem., 2010, 20, 3035–3039.
49. H. J. Salavagione, G. Martinez and M. A. Gomez, J. Mater. Chem., 2009, 19, 5027–5032.
50. H. Pang, T. Chen, G. Zhang, B. Zeng and Z. M. Li, Mater. Lett., 2010, 64, 2226–2229.
51. Y. A. Balogun and R. C. Buchanan, Compos. Sci. Technol., 2010, 70, 892–900.
52. H. Kim, A. A. Abdala and C. W. Macosko, Macromolecules, 2010, 43, 6515–6530.
53. M. Qu, F. Deng, S. M. Kalkhoran, A. Gouldstone, A. Robisson and K. J. Van Vilet, Soft Matter, 2011, 7, 1066–1077.
54. J. R. Potts, D. R. Dreyer, C. W. Bielawski and R. S. Ruoff, Polymer, 2011, 52, 5–25.
55. N. Du, C. Y. Zhao, Q. Chen, G. Wu and R. Lu, Mater. Chem. Phys., 2010, 120, 167–171.
56. H. Kim, Y. Miura and C. W. Macosko, Chem. Mater., 2010, 22, 3441–3450.
57. H. J. Salavagione, G. Martinez and M. A. Gomez, J. Mater. Chem., 2009, 19, 5027–5032.
58. Y. Chen, Q. Zhuang, X. Liu, J. Liu, S. Lin and Z. Han, Nanotechnology, 2013, 24, 245702–245712.
59. Y. K. Yang, C. E. He, R. G. Peng, A. Baji, X. S. Du, Y. L. Huang, X. L. Xie and Y. W. Mai, J. Mater. Chem., 2012, 22, 5666–5675.
60. G. A. Gelves, M. H. Al-Saleh and U. Sundararaj, J. Mater. Chem., 2011, 21, 829–836.
61. T. Kuila, S. Bose, C. E. Hong, M. E. Uddin, P. Khanra, N. Kim and J. H. Lee, Carbon, 2011, 49, 1033–1037.
62. X. Zhang, T. Liu, T. V. Sreekumar, S. Kumar, V. C. Merre, R. H. Hause and R. E. Smalley, Nano Lett., 2003, 3, 1285–1288.
63. D. Qian, E. C. Dickey, R. Andrews and T. Rantell, Appl. Phys. Lett., 2002, 81, 5123–5125.
64. J. W. Suk, R. D. Piner, J. An and R. S. Ruoff, ACS Nano, 2010, 4, 6557–6564.
65. C. Lee, X. Wei, J. W. Kysar and J. Hone, Science, 2008, 321, 385–388.

---

Footnote

† Present address: Department of Chemistry & The Alan G. MacDiarmid NanoTech Institute, The University of Texas at Dallas, Richerdson, TX 75080, USA.

---