Introduction

Currently, the study of nanofiller-reinforced polymeric materials has attracted more attention. A wide variety of fillers such as carbon nanotubes (CNTs), graphene and graphene oxide (GO), fullerenes, layered silicates and inorganic nanoparticles were added to polymer matrix which contributed to the enhancement of mechanical, thermal stability and gas barrier properties [1–4]. In contrast to the traditional inorganic nanoparticles (such as boron nitride, aluminum and silica, etc.) [5, 6], the functionalized graphene with layered structure can effectively reinforce the properties of the composite materials [7].

The GO was provided with remarkable mechanical and thermal properties attributed to their extremely high specific surface area which make them ideal candidates for the reinforcing phase in polymer nanocomposites [8, 9]. The GO is a chemically modified graphene which is decorated with hydrophilic oxygen functional groups, such as hydroxyl, epoxy, and carboxylic groups. Those hydrophilic functional groups make the GO to spontaneous stable dispersion in aqueous solution and organic solution of polarity [10–12]. Also, the surface modification of GO has turned into an important method which has not only improved the dispersion, but also improved a strong interface between GO and the matrix. And also many different functional groups can be introduced onto the graphene sheets [13]. Incorporation of these modified GO into polymer matrices can easily prepare a variety of excellent functional composites. Ge et al. [14] reported that triethylenetetramine-modified graphene oxide/chitosan composite was synthesized by microwave irradiation method which could significantly improve adsorption of Cr in effluents, and highest adsorption capacity of Cr reaches up to 219.5 mg g−1. Ribeiro et al. [2] indicated that the chemical modification of the GO by TEPA had improved the thermal and mechanical properties of composites. The type and extent of chemical modification of GO are the critical factors. Fan et al. [15] showed that GO is converted into graphene by the chemical modification using alkyl imidazolium ionic liquids (ILs) as lubricant additives to greatly enhance friction and anti-wear behaviors.

Liquid crystals polymer (LCP) as the functional of soft materials has attracted considerable attention in recent decades. With regard to future developments, the main attention is not only focused on academia but also on industry application due to its excellent thermal stability, excellent mechanical and photoactive [16–20]. In particular, thermotropic main-chain LCP copolyester is significantly important to the polymer’s transverse strength and compressive strength. It represents a class of high-tech polymers which combine high mechanical performances with low density [21, 22]. Noteworthily, polymer composites with LCP improved toughness, thermal conductivity, and other thermal properties without sacrificing other performance. These outstanding properties provide the opportunity to be used widely in applications such as aerospace, energy storage and electronic packaging [23–26].

Epoxy resin as one of the most important typical polymer matrices in composite has drawn attention due to its excellent mechanical properties, good thermal stability and low shrinkage [27, 28]. Hence, it is used widely in industry applications including structural materials, adhesives, coatings and additive-type flame retardant [29]. However, cross-linked resins generally have weak resistance to crack propagation and low fatigue resistance to heat and impact due to the inherent brittleness, which limited the effective usage of epoxy composites [30, 31]. The functionalized GO can effectively improve the properties of epoxy resins, such as toughness, glass transition temperature and impact resistance without sacrificing their intrinsic properties [32–34].

The aim of this study is to develop a new functionalized GO which was modified by biphenyl liquid crystalline polyester (BLCP-GO) as a reinforcing filler to improve the mechanical and thermal properties of epoxy. The BLCP-GO dispersed in the epoxy resins to form the crosslinking networks in the curing reaction. Meanwhile, BLCP-GO are chosen as the fillers for preparing the epoxy composites with 0.5, 1.0, 1.5, and 2.0 wt% BLCP-GO loading. The epoxy composites have been characterized in terms of their morphological, thermal, and mechanical properties to study the effect of the BLCP-GO on their dispersion in the epoxy matrix and the interaction with the epoxy.

Experimental

Materials

The epoxy resin used in this study is diglycidylether of bisphenol A (DGEBA, E-51, epoxy value = 0.51 and viscosity value = 12 Pa s at 25 °C) supplied by Yueyang Chemical Plant (Yueyang, China). Nature graphite flake (325 mesh, 99 %) was purchased from Hengrui Graphite Co., Ltd. (Qingdao, China). H2SO4 (98 %), HCl (36 %), H2O2 (30 %), KMnO4, KI, NaOH, P2O5 and K2S2O8 were provided by Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China). N, N′-Dimethylformamide (DMF), 4,4′-dihydroxydiphenyl sulfone (DDS) and pyridine were obtained from Xiya Chemical Reagent Co., Ltd. (Chengdu, China). 4-Dimethylaminopyridine (DMAP), p-phthaloyl chloride, 6-chloro-1-hexanol, and N,N′-diisopropylcarbodiimide (DIC) were purchased from Aladdin Chemistry Co., Ltd. (Shanghai, China). Deionized (DI) water was used in all the process of aqueous solution preparations and washings. All of the reagents were at analytical grade without further purification.

Synthesis of the biphenyl liquid crystalline polyester (BLCP)

The synthetic route of the 4,4′-di(6-hydroxy hexyloxy) biphenyl (BP6) is described as Scheme 1. The BP6 was synthesized by the nucleophilic substitution reaction of 4,4′-dihydroxydiphenyl and 6-chloro-1-hexanol in a molar ratio of 1:4. Briefly, NaOH (1.60 g, 0.04 mol), KI (1.66 g, 0.01 mol) as catalyst and 90 ml anhydrous ethanol solvent were added into a 250-ml three-necked, round-bottomed flask equipped with magnetic stirrer and thermometer [35]. After NaOH was completely dissolved under electromagnetic stirring, 4,4′-dihydroxydiphenyl (3.72 g, 0.02 mol) was added into the flask. Then, 6-chloro-1-hexanol was slowly dropped into the above-mentioned mixture and heated to 75 °C for 24 h. After the reaction, the solution was cooled to room temperature. The precipitate was collected by filtration and then washed with distilled water (DI) several times. The crude product was recrystallized with DMF and was separated by filtration and vacuum-dried at 60 °C to obtain a white, needle-like crystal (BP6).

Scheme 1
scheme 1

The synthetic routes of the BLCP

Full size image

In a three-necked flask, 4.25 g (0.011 mol) BP6 was dissolved with 90 ml DMF solvent and 1 ml pyridine as catalyst was dropped into the mixture under stirring. p-phthaloyl chloride (2.03 g, 0.01 mol) was dissolved in 10 ml DMF and slowly dropped into the reaction solution to keep stirring at ambient temperature for 2 h, and then further reacted at 80 °C for another 12 h. The cold reaction mixture was precipitated into distilled water. The solid was washed with distilled water and then dried at 60 °C in a vacuum drying oven to provide the yellow powder.

Synthesis of BLCP-grafted graphene oxide (BLCP-GO)

The GO were prepared according to the modified Hummers method, in which the natural graphite flakes were oxidized in concentrated sulfuric acid, phosphorus pentoxide, potassium permanganate and hydrogen peroxide [19, 36, 37]. The GO suspensions were prepared via ultrasonication to form colloidal aqueous solutions and freeze-dried as shown in Scheme 2. Then, 0.1 g GO was dispersed in 100 ml DMF under ultrasonication for 30 min and was placed in a 250-ml three-necked flask. After that, 5 g BLCP, 2 g DIC and 0.05 g DMAP were placed in the above-mentioned mixture at ambient temperature and kept stirring for 2 h, and then heated to 40 °C for 48 h. After filtration, the crude product of the BLCP grafted-GO (BLCP-GO) was washed several times with DMF to ensure the complete removal of the free BLCP, then with ethyl alcohol to remove residues of DMF solvent, and the filtered material was dried in a vacuum oven at 60 °C.

Scheme 2
scheme 2

The synthetic routes of biphenyl liquid crystalline polyester grafted GO

Full size image

Preparation of BLCP-GO/epoxy composites

The BLCP-GO/epoxy composites were prepared as follows: a specific amount of BLCP-GO as filler was dispersed in acetone and treated with ultrasound for 1 h. Then, the BLCP-GO dispersion was added to a stoichiometric amount of preheated epoxy. The mixture was heated up to 60 °C with stirring under the condition of vacuum to ensure the removal of the acetone solvent entirely and uniform dispersion. After that, curing agent DDS (30 g/100 g of epoxy resin) was added under vigorous mechanical stirring. Then, the mixture was poured into mold and placed in a vacuum oven to remove air bubbles for 30 min. Finally, the mold was placed in a convection oven to cure at 140 °C for 2 h, 160 °C for 2 h and 180 °C for 2 h. The epoxy composites were prepared with different weight fractions (0, 0.5, 1.0, 1.5, and 2.0 wt%) of BLCP-GO by above methods.

Characterization

Fourier-transformed infrared spectroscopy (FTIR) was recorded between 4000 and 450 cm−1 on a Nicolet 470 spectrophotometer (spectra resolution: 0.125 cm−1, 32 scans) using KBr pellets at room temperature. The morphology of composites was examined using field emission scanning electron microscopy (FE-SEM, JSM-6701F, Japan) at an accelerating voltage of 20 kV, and the cryogenic fractured surfaces of composites were sputter-coated with gold–palladium alloy before viewing. The liquid crystalline transitions and optical textures were observed using a Leica DMxRP polarizing microscope with an INSTEC STC-200 hot stage. Small amounts of samples (2–3 mg) were pre-melted on a microscope slide, then covered with a piece of cover glass to form a uniform thin film. The formation and development of the LC phase were examined under polarized light. Thermal gravimetric analysis (TGA) was performed with a NETZSCH STA 449C at a heating rate of 20 °C min−1 from 50 to 700 °C under nitrogen atmosphere. Differential scanning calorimetry (DSC-204, NETZSCH, Germany) was performed at temperature from 30 to 250 °C at a heating rate of 10 °C min−1. All tests were performed in a nitrogen atmosphere with a sample weight of about 8 mg. Dynamic mechanical analysis (DMA) was performed on DMA Q800 (TA Instruments, USA), operating in a single cantilever bending mode at an oscillation frequency of 1 Hz and displacement amplitude of 10 µm. Testing temperature was set as from room temperature to 250 °C at a heating rate of 3 °C min−1. The impact strength was measured on a tester of type XJJ-5, which is with no notch in the specimen according to National Standard of China (GB1043-79). The specimen was with a thickness of 4 mm and width of 10 mm and length of 80 mm. The tensile strength was examined on an electron omnipotence tester of type RGT-5. The tensile experiments was used a three-point bending mode of the universal testing machine with a crosshead speed of 2 mm/min according to the National Standard of China (GB1040-92). All the presented results are an average of five specimens.

Results and discussion

Characterization of the fillers

FTIR spectra of the GO, BLCP and BLCP-GO are shown in Fig. 1. It can be seen in the three infrared curves, the wide and strong absorption peak at 3405 cm−1 attributed to hydroxyl group (–OH) stretching [8]. For the GO, the peak at 1722 and 1621 cm−1 was associated with the C=O carbonyl bending and the peak adsorption of water [38]. At 1384 cm−1was presented a strong and sharp peak corresponding to hydroxyl vibration. A peak appearing at 1150 cm−1 is associated with the C–O stretching vibration of the epoxy group. For the spectrum of BCLP, the peaks at 2865 and 2938 cm−1 could be attributed to –CH2 symmetrical and asymmetrical stretching vibrations. The band of 1690 cm−1 is due to the C=O stretching. The bands at 1606, 1500 and 824 cm−1 were due to aromatic structure [35, 39]. The bands at 1261 and 1247 cm−1 were associated with the –CH2 out-of-plane bending vibration. Comparing the spectra of GO and BLCP with that of BLCP-GO, it was found that the characteristic absorption peaks of the BLCP appear on the spectrum of product BLCP-GO. The original peak of GO curve at 1722 cm−1 disappeared. However, a new peak was observed at 1709 cm−1. This is due to carboxyl group of GO by modification into the ester group. The peak at 1621 cm−1 disappeared in the GO curve due to some oxygen-containing functional groups included in the reaction process and the liquid crystal molecules that inhibit water.

Fig. 1
figure 1

FT-IR spectra of GO, BLCP and BLCP-GO

Full size image

The thermal behaviors and liquid crystal properties of synthesized BP6, BCLP were investigated by differential scanning calorimetry (DSC) and polarizing optical microscope (POM) in Fig. 2. The DSC heating curve of BP6 shows two distinct endothermic sharp peaks at 99.1 and 176.5 °C. They correspond to the BP6 melting temperature (T m) and isotropization temperature (T i), respectively. It revealed the change of a crystalline solid into the liquid crystalline state at 99.1 °C (T m), and then the temperature reached to 176.5 °C (T i), which is the liquid crystalline structure to isotropic phase. The image of POM showed that the sample of BP6 began to melt and gradually turned to fan of liquid crystal texture when it was heated to 110 °C which is in inset of Fig. 2a. The texture disappears and the isotropic phase was observed above 185 °C. Furthermore, the phase transition temperatures observed by POM are conformed to the DSC determination results. It can be confirmed that the presence of obvious liquid crystalline mesophases for these BP6 and it exhibited relatively wide temperature range of the liquid crystal phase. This is attributed to the long flexible chain on both sides of the biphenyl, which is easy to orientation [35, 40]. For the DSC curve of BLCP which is shown in Fig. 2b, three endothermic peaks appear in the field of vision, which correspond to melting point temperature (at 99.9 °C), texture transformation temperature (at 134.7 °C) and clearing point of temperature (at 151.4 °C). The POM observation of BLCP indicated that the sample shows bright fine-grain woven structures after melting (inset I). With increasing temperature, woven texture structures suddenly turn into fan texture structure (inset II) and subsequently, the liquid crystalline phases of sample disappeared and turned into isotropic phase [41, 42]. The similar woven texture structures can be observed when it was cooled (inset III). Also, POM and DSC results are consistent.

Fig. 2
figure 2

a DSC curve of BP6, inset POM figure of BP6; b DSC curve of BLCP, inset POM figures of BLCP

Full size image

The TGA curves of GO, BLCP, and BLCP-GO under nitrogen and air atmosphere are shown in Fig. 3. Under nitrogen atmosphere, as shown in Fig. 3a, it was also found that the GO weight loss about 12 wt% at 100 °C, which is due to evaporation of the remaining water. However, BLCP-GO has no loss of weight at this stage. It means that introduction of BLCP will reduce the ability of absorbed water. After that, GO was loss of weight sharply about 28 wt% between 160 and 220 °C, which was attributed to the decomposition of labile oxygen groups [12] to produce H2O, CO, CO2. Compared with the GO, the weightlessness of BLCP-GO has weakened. This is mainly due to the removal of the oxygen groups residual of the GO [38]. At higher temperatures, the decomposition temperatures (T d) of BLCP and BLCP-GO are 275 and 365 °C, respectively. The T d of the BLCP-GO is 90 °C higher than that of BLCP. The degradation mechanism of BLCP and BLCP-GO is the fracture of C–C bond and C–O bonds, to produce massive quantities of molecules such as oligomers, low-molecular polyester, low alkane, hexylidene, CO, CO2 etc. These results suggest that the thermal stability of the GO has been improved after modification with BLCP. This is because the oxygenic groups of GO have been replaced by BLCP chain to form a stable chemical bond, and they strongly interacted with GO sheets leading to the decrease of the molecular motion and the increase of the rupture temperature of the polymer backbone. However, under air atmosphere, as shown in Fig. 3b, the thermal stability of BLCP-GO has deteriorated. The thermal decomposition temperature drops and residual carbon amount is only 6 wt%.

Fig. 3
figure 3

TGA curves of GO, BLCP and BLCP-GO under a N2 and b air atmosphere

Full size image

Figure 4 shows the SEM images of GO and BLCP-GO nanosheets. It can be seen from Fig. 4a that the GO are showing transparent and overlapping in some instances. Corrugation and scrolling are parts of the intrinsic nature of GO, which result from the fact of the two-dimensional membrane structure becoming thermodynamically stable via bending. The SEM image of BLCP-GO are shown in Fig. 4b. It can be found that some obvious BLCP were grafted to the GO nanosheets leading to loss of transparency.

Fig. 4
figure 4

SEM image of a GO and b BLCP-GO

Full size image

Morphology of composites

Figure 5 shows the SEM images of the impact fracture surfaces of neat epoxy and the BLCP-GO/epoxy composites. It can be seen that Fig. 5 provides three groups of representative images with different magnification. The fracture surface of the neat epoxy exhibits river patterns observed from Fig. 5a, b. Also, fracture surface was very smooth and single crack direction, which indicates that neat epoxy almost presents no plastic deformation and weak resistance to crack initiation. This is a typical brittle fracture and consistent with poor tensile strength [6, 22, 43]. The fracture surfaces of 1.0 wt% BLCP-GO/epoxy composites are shown in Fig. 5c, d. The fracture surface is very rough, which presents uneven or fuzzy interface and spreads the crack extension. The reason is that BLCP-GO hinders crack extension to make materials ductile fracture. This indicates that the addition of modified GO changes the way fracture of epoxy resin. GO flake linked the substrate together like a bridge, leading to the flexible interface and discouraging the formation of brittle fracture [31, 44, 45]. At the same time, the rigid segment structure of liquid crystal acting as a steel reinforcement exhibited ductile behavior when the material ruptured. However, with the filler loading increasing, BLCP-GO nanosheets exist in the form of agglomerates and affect uniformity of material as shown in Fig. 5e, f.

Fig. 5
figure 5

SEM images of fractured surface of composites: a, b neat epoxy, c, d 1.0 wt%; and e, f 2.0 wt%

Full size image

Thermal properties of epoxy composites

Figure 6 shows the TGA and DTG curves of neat epoxy and its composites. As can be seen from the curves, all of the samples present a single thermal decomposition platform and exhibit similar thermal behavior, which means that the addition of BLCP-GO does not change the degradation mechanism of epoxy matrix. It can be seen that initial decomposition temperature of neat epoxy at 351 °C in Fig. 6a. Compared with neat epoxy, the decomposition temperature (T d) of the composite materials has improved when the modified GO was added into the epoxy matrix. The T d of composites increases by 28 °C when BLCP-GO content is 0.5 wt%. As DTG curves shown in Fig. 6b, the composite with 0.5 wt% content of BLCP-GO has the maximum degradation temperature at 421 °C [46, 47]. The decomposition temperatures (T d) and the maximal decomposition temperature (T max) are listed in Table 1. Thermal properties data indicated that the addition of BLCP-GO improves the thermal property of epoxy composites. It can be attributed to two factors. First, the BLCP-GO sheets act as physical interlock points and physical tangles with the epoxy matrix, thus hindereing the thermal motion of epoxy matrix molecule [28]. Second, the crosslinking between nanosheets and epoxy groups in the curing reaction restrained the mobility of chain, which leads to enhanced thermal properties of the composites [48, 49].

Fig. 6
figure 6

a TGA and b DTG curves of the neat epoxy and its composites, inset in a: drawing of partial enlargement

Full size image
Table 1 Thermal stabilities of epoxy composites
Full size table

Figure 7 shows the DSC curves for neat epoxy and its composites, which exhibit the change in T g for composites. It can be seen that the T g of pristine epoxy takes place at 163.6 °C. The addition of BLCP-GO nanosheets can enhance the T g of the epoxy composites, which shifts to a higher temperature compared to neat epoxy. The T g values of the BLCP-GO/epoxy composites with 0.5, 1, 1.5, and 2 wt% BLCP-GO fillers are 175.1, 176.9,180 and 174.8 °C, respectively. The maximum value of T g is 16.4 °C higher than that of the neat epoxy resin when the filler content is 1.5 wt%. It’s attributed to good interaction between the nanosheets and epoxy matrix. When the filler content is higher than 1.5 wt%, the T g of the BLCP-GO composites was decreased with increasing the content of BLCP-GO filler. The main reason was BLCP-GO agglomerate in the epoxy matrix. Once nanosheets agglomerate, the interactions are more between BLCP-GO nanosheets rather than BLCP-GO nanosheets and epoxy matrix. The agglomerated BLCP-GO nanosheets could not impose any restrictions on the mobility of epoxy, and thereby higher loading BLCP-GO nanosheets resulting in lower T g values. But the T g values of BLCP-GO composites with 2.0 wt% are still higher than that of neat epoxy at higher filler content.

Fig. 7
figure 7

DSC curves of the neat epoxy and epoxy composites

Full size image

The storage modulus and tan δ of neat epoxy and its composites were measured by DMA to study the effects of BLCP-GO filler in Fig. 8. The incorporation of the BLCP-GO filler did not deteriorate the thermomechanical behavior. Within the experimental temperature range, the storage modulus of epoxy composites is higher than that of neat epoxy. It reflected the interface interaction between BLCP-GO filler and epoxy resin matrix flexibility and effectively. As shown in Fig. 8a, the storage modulus of the epoxy was increased by the addition of the BLCP-GO fillers, especially in the glass region. In the test temperature range, all the storage modulus of the composites was much higher than that of neat epoxy but exhibited irregular change with increasing of the loading fillers. It is attributed to the inherent hardness and strong storage modulus of filler. Well-dispersed BLCP-GO fillers filled the spaces in the epoxy matrix and acted as the crosslinking points. On one hand, the high surface area of hybrid fillers has a huge interface with epoxy resin matrix and provides an effective way to transfer the actual stress from the matrix. On the other hand, the LC domains of filler participate in crosslinking networks with the matrix to form strong covalent bonding leading to hinder the movements of the polymer chains [2, 47, 48]. Figure 8b shows the tan δ curve of neat epoxy and its composites. The T g of neat epoxy and its’ composites could be obtained accurately from the tan δ peaks. The T g of the BLCP-GO/epoxy composites with 0, 0.5, 1.0, 1.5, and 2.0 wt% are 176.8, 194.0, 191.4, 193.4 and 189.9 °C, respectively. The variation trend of T g measured by DMA shows a slight distinction with the results of DSC. The main reason for the discrepancy is based on two different principles measurement of T g. There are different T g results between DSC and DMA. For DMA measurement, the highest T g value is up to 194.0 °C at 0.5 wt%, which increased by 17.2 °C compared to the neat epoxy. The filler effect and the crosslink effect can improve the thermal stability of the composite materials [50]. On the other hand, the free volume of the composites was significantly reduced when the BLCP-GO filler was added.

Fig. 8
figure 8

a Storage modulus and b tan δ of the neat epoxy and epoxy composites

Full size image

Mechanical properties of epoxy composites

Figure 9 shows the impact strength and tensile strength of neat epoxy and its composites with different content. As can be seen from it, the impact strength and tensile strength of composite materials were obviously improved with the addition of BLCP-GO. Within a certain content of BLCP-GO, the impact strength and tensile strength were improved steeply with the increasing of BLCP-GO and reached the maximum, then decreased again at higher loading. The peaks of impact strength and tensile strength reached the maximum with content of 1.0 and 1.5 wt% BLCP-GO, which improved by 103 and 60 % compared to neat epoxy, respectively [43]. Two reasons may account for this phenomenon. First, this is attributed to provide an effective way that the actual stress is expected to be easily transferred from the matrix to GO nanosheets. Second, the residual epoxy groups of GO form the chemical crosslinking with epoxy matrix, which hinders the stress concentration effect to improve impact strength and tensile strength [51, 52].

Fig. 9
figure 9

Impact strength and tensile strength of neat epoxy and epoxy composites

Full size image

Figure 10 shows the flexural strength and modulus of the neat epoxy and its composites. The results show an improvement in strength and modulus of the composites compared to neat epoxy with increasing of BLCP-GO but the change trend is different. For instance, the maximum flexural strength (131.9 MPa) and modulus (2442 Mpa) were observed in 1.0 and 1.5 wt% BLCP-GO content, respectively, which increased by 66.1 and 57.8 % compared to neat epoxy [19, 51]. However, the flexural properties of the epoxy composites decreased when the filler loading exceeds the critical level. The mechanical properties of the epoxy composites were enhanced due to the good compatibility between BLCP-GO and epoxy matrix, which has good dispersion and is beneficial for the efficient stress transfer in epoxy matrix [13]. Table 2 summarizes the mechanical properties data.

Fig. 10
figure 10

Flexural strength and modulus of neat epoxy and its composites

Full size image
Table 2 The mechanical properties of neat epoxy and its composites
Full size table

Conclusions

Epoxy composites with BLCP-GO as inclusions were prepared successfully. The epoxy composites have excellent thermal and mechanical properties due to the strong interfacial interaction. The epoxy composite with only 0.5 wt% BLCP-GO increased the initial decomposition temperature (T d) by 28 °C and glass transition temperature (T g) by 17.2 °C comparing with neat epoxy. Meanwhile, the composites with the 1.0 wt% BLCP-GO exhibit an increase in impact strength, tensile strength, flexural strength and flexural modulus by 103, 52, 66 and 56 %, respectively. This work clearly indicates that the novel BLCP-GO fillers play a critical role in the improvement of epoxy composites.