Simultaneous regulation of morphology, crystallization, thermal stability and adsorbability of electrospun polyamide 6 nanofibers via graphene oxide and chemically reduced graphene oxide

Abstract

Modulating the processability, structure and properties simultaneously remains a challenge for the preparation of high-performance electrospun nanofibers. The influence of incorporating graphene oxide (GO) and chemically reduced graphene oxide (RGO) nanosheets on the rheological behaviors of polyamide 6 (PA 6) solutions was systematically investigated. A small amount (0.5 wt%) of GO could dramatically increase the steady viscosity of PA 6 solution at 0.001 s⁻¹ (η_0.001) by a factor of four, while the same content of RGO resulted in a threefold decrease. Based on the modulation of viscosity of the PA 6 spinning solutions by GO and RGO nanosheets, GO/PA 6 and RGO/PA 6 nanofibers with various structures and properties were successfully prepared via electrospinning. The addition of 1 wt% GO and RGO nanosheets improved PA 6 fibrous uniformity and fineness, and the spinnable concentration (SC) range of PA 6 was effectively broadened. The difference in interfacial interaction between nanosheets and the PA 6 matrix led to different crystallization behaviors of the electrospun PA 6 nanofibers. The formation of γ-form crystals was promoted by GO but inhibited by RGO, which offers a convenient strategy to modulate the mechanical properties of PA 6 nanofibers. In addition, the thermal stability and adsorption capacity of PA 6 nanofibers was enhanced by adding GO or RGO.

---

1. Introduction

Polymer nanofibers, namely ultrafine fibers with diameters in the range of 10–1000 nm, can be facilely and effectively prepared by electrospinning, possessing extensive applications in filtration systems, water treatment, tissue engineering, wound dressings, sensors, electronic and optical devices, etc.^1–7 Recently, two chief directions in the development of polymer nanofibers have been attracting more attention. On the one hand, spinning technology is being optimized in order to accomplish efficient control of the nanofibrous morphology. On the other hand, regulation of the fibrous microstructure provides guidance for the implementation of high-performance or multi-functionality. Among the spinning process parameters, the spinning solution viscosity, mediated mainly by polymer concentration, is a determinative factor for the fibrous morphology.⁸ A solution of too low viscosity tends to form droplets or beaded fibers due to the low entanglement number (n_e),^9,10 while a solution of too high viscosity cannot form nanofibers due to nozzle blockage induced by condensation,¹¹ implying that there is a spinnable concentration (SC) range for polymer solutions to obtain electrospun fibers with well-controlled morphology. Extending the SC range is helpful to meet the multipurpose demands for polymer nanofibers. Moreover, the microstructure of the nanofibers formed during electrospinning, such as crystal modification and orientation, is vital in determining the properties relevant to application, which is the focus of much research.¹²

Polyamide 6 (PA 6) has been employed as a kind of conventional raw material for electrospinning nanofibers with applications in many industrial fields due to its superior fiber forming ability, low cost, biocompatibility and good mechanical strength.^13–15 The electrospinning conditions, morphology and microstructure of electrospun PA 6 nanofibers have been widely studied.^9,14,16 In particular, PA 6 is a typical semicrystalline polymer with polymorphic phases,^17,18 and changes of its crystal form and crystallinity can dramatically affect the properties of the electrospun nanofibers,¹⁹ which has attracted much attention.^20–23 However, how to modulate the processability, structure and properties of electrospun PA 6 nanofibers simultaneously remains a challenge and has not attracted enough attention.

Incorporating functional nanoparticles into nanofibers has been one of the most exciting research issues in the field of electrospinning for the last ten years, and is considered as an effective strategy to form high-performance multi-functional nanofibers.²⁴ Several nanoparticles, such as nanoclays, passivated Au nanoparticles and carbon nanotubes, have been incorporated into PA 6 to prepare composite nanofibers.^25–28 Recently, graphene as a 2D nanomaterial has attracted tremendous attention due to its excellent electronic, thermal, and mechanical properties, making it a suitable candidate for applications in various areas such as solar cells, supercapacitors, sensors, hydrogen storage, electrodes and nanofillers for nanocomposites.^29–33 Pant et al. obtained spider-wave-like bimodal fibers by blending a suitable amount of graphene oxide (GO) with PA 6 solution.³⁴ They further reduced the GO sheets incorporated in the nanofibers to reduced graphene oxide (RGO) using a hydrothermal treatment and obtained RGO/PA 6 composite nanofibers with better electrical conductivity.³⁵ Li et al. used aligned electrospun graphene/PA 6 nanofibers as reinforcement to prepare PMMA-based nanocomposites with good transparency and improved mechanical properties.³⁶ Besides these, the excellent thermal performance and large specific surface area of graphene are expected to offer superior thermal stability to the PA 6 nanofibers and make the composite nanofibers a promising adsorbent. However, the influences of GO and RGO nanosheets on the morphology, crystalline structure and performance of PA 6 nanofibers have still not been systematically investigated, particularly for the associated study between these three.

In a previous work, we developed a strategy to broaden the SC window of water-soluble polar polymers and modulate the morphology and diameter of nanofibers via adding a small amount of GO or RGO, based on a novel viscoelastic modulation method.³⁷ Inspired by this, the previous strategy is employed in the semicrystalline and non-water soluble PA 6 system to beneficially extend its scope of application. Herein, an obvious rheological difference of GO and RGO suspensions is reported, and accordingly, the influence of GO and RGO on the SC window and morphology of PA 6 nanofibers is explored. Moreover, alteration of the crystalline structure and enhancement of the thermal and adsorption performance of PA 6 nanofibers induced by GO or RGO nanosheets are investigated. Therefore, the simultaneous regulation of morphology, crystalline structure and performance of PA 6 nanofibers is effectively realized through this method.

2. Experimental

2.1 Materials

Graphite (99.8%) with an average particle size of 0.45 μm was obtained from Alfa Aesar Co. Ltd., UK. Concentrated sulfuric acid (H₂SO₄), concentrated hydrochloric acid (HCl), potassium persulfate (K₂S₂O₈), phosphorus pentoxide (P₂O₅), potassium permanganate (KMnO₄), hydrogen peroxide (H₂O₂, 30%), formic acid (88%) and ascorbic acid were purchased from Shanghai Chem. Reagent Co. Ltd., China. PA 6, under a trademark of CM1017 (weight averaged molecular weight M_w = 627 kg mol⁻¹), was purchased from Toray Co. Ltd., Japan. All the materials were used without further purification.

2.2 Preparation of GO and RGO

GO was prepared from natural graphite using a modified Hummers’ method.^36,38 In brief, K₂S₂O₈ (16.8 g), P₂O₅ (16.8 g) and graphite powder (20 g) were added to concentrated H₂SO₄ (80 mL) at 60 °C and the oxidation reaction was performed at 80 °C for 5 h. The suspension was diluted with deionized water, vacuum-filtered and dried in air at room temperature. In the second oxidation step, the pretreated graphite powders were added to concentrated H₂SO₄ (460 mL) at 2–3 °C in an ice bath and, under continuous stirring, KMnO₄ (60 g) was added gradually. The mixture was then heated to 35 °C and stirred for 2 h, followed by diluting with deionized water (1.6 L) and further stirring for 2.5 h at 85 °C. The reaction was terminated by adding deionized water (1.5 L) and 30% H₂O₂ (500 mL), followed by washing with 1 M HCl solution. The products were water-washed and separated via centrifugation until pH of the decantate reached 6. The sample of GO was obtained after freeze-dehydration for 2 days. RGO was synthesized from the as-prepared GO sheets.³⁹ The GO (50 mg) and ascorbic acid (400 mg) were suspended in deionized water (100 mL) under stirring for 24 h at 90 °C. RGO was obtained after the product was vacuum-filtered and washed. The obtained GO and RGO were dispersed in formic acid (88%) with the aid of ultrasound to yield stable suspensions. The GO/PA 6 and RGO/PA 6 suspensions were prepared by simply adding PA 6 into the GO or RGO suspensions obtained, respectively.

2.3 Electrospinning

Nanofibers were prepared by electrospinning at a voltage of 18 kV, with a flow rate of 0.1 mL h⁻¹ and a distance of 16 cm between the collector and tip of a 20 mL syringe [bearing a 0.7 mm inner diameter metal needle which was connected with a high voltage power supply (GDW-a, Tianjin Dongwen High-voltage Power Supply Plan, China)]. The flow rate of solution was controlled by a pump (WZ-50C2, Zhejiang University Medical Instrument Co. Ltd., China) and the electrospinning was performed at 25 °C and at 50% relative humidity. The fibers, collected on a tin foil, were vacuum-dried for 24 h at 40 °C.

2.4 Batch adsorption experiments

Batch adsorption experiments were conducted by adding a certain amount (∼10 mg) of nanofibers into 10 mL Cu²⁺ solutions at 30 °C, and equilibrating the solutions under agitation for 24 h. After equilibration, the supernatant was analyzed for Cu²⁺ content according to the UV-Vis spectra after coordination with 25 mM ethylene diamine tetraacetic acid.

2.5 Characterization

Fourier-transform infrared (FT-IR) spectra were collected using a spectrometer (VECTOR 22, Bruker Optics, USA). The UV-Vis absorption spectra were measured on a UV-Vis spectrophotometer (Lambda 35, Perkin-Elmer, USA). Raman spectra were collected on a Raman spectrometer (ALMEGA-Dispersive, Thermo Nicolet, USA) with 514.5 nm excitation. X-ray diffraction (XRD) was carried out on a diffractometer equipped with Cu Kα radiation (λ = 0.154 nm) (Rigaku D/max 2550, Shimadzu, Japan). The thermal stability was evaluated by a thermogravimetric analyzer (TGA, Q1000, TA, USA) at a heating rate of 5 °C min⁻¹ for nanosheets and 10 °C min⁻¹ for nanofibers under nitrogen. Rheological measurements were carried out on an advanced rheometer (AR-G2, TA Instruments, USA) equipped with parallel plate grippers of 40 mm in diameter. The gap distance was set as 0.5 mm for all tests. The morphology of the nanosheets and fibers was observed under a scanning electron microscope (SEM, S4800, Hitachi, Japan) and an atomic force microscope (AFM, Bruker-Dimension Edge, USA). Transmission electron microscopy (TEM, JEM-1200EX, JEOL, Japan) was employed to investigate the dispersity and ordering of GO or RGO in the fibers by depositing electrospun nanofibers onto TEM copper slot grids. Thermograms were recorded using a differential scanning calorimeter (DSC, Q100, TA, USA), which was calibrated using indium.

3. Results and discussion

3.1 Rheological behaviors of spinning solutions

Electrospinning of GO/PA 6 and RGO/PA 6 nanofibers requires a uniform dispersion of nanosheets in the spinning solutions. GO can be easily exfoliated and dispersed in formic acid due to the strong interaction between the abundant polar groups (e.g. hydroxyl, carboxylic acid and epoxy groups) of GO and the carboxyl groups of formic acid (Fig. S1†). However, the strong tendency for π–π stacking and low solubility leads to the agglomeration of RGO nanosheets in the absence of capping reagents. Here, L-ascorbic acid (L-AA), a mildly nontoxic reagent, is employed as the reductant of GO and the oxidized products of L-AA can play a role as a capping reagent to stabilize the RGO nanosheets simultaneously.^39,40 Fig. 1 shows that the GO nanosheets are exfoliated to a monolayer with an average height of 1.0 nm and a lateral size of 5–10 μm (Fig. 1A and B). After reduction, the majority of RGO nanosheets are still monolayers with an average height of 0.4 nm and the lateral size is reduced to smaller than 5 μm (Fig. 1C and D). The reduction from GO to RGO is further confirmed by extensive analyses of FTIR, UV-vis, Raman spectra, XRD, and TGA (see Fig. S2–S6† for details).

Fig. 1 SEM images of GO (A) and RGO (C) deposited on aluminum foil, AFM ichnographies and the corresponding height sensor graphs of GO (B) and RGO (D) deposited on a mica sheet.

There is a large amount of H-bonding between macromolecular chains so that the steady viscosity (η) of the PA 6 solution is sensitive to the shear rate ([small gamma, Greek, dot above]). Fig. 2A shows a plot η against [small gamma, Greek, dot above] for the PA 6 solution (C = 20 wt%) and its suspensions with various nanosheet contents with respect to the PA 6 concentration (C_F). The PA 6 solution exhibits a strong shear thinning region at [small gamma, Greek, dot above] < 1 s⁻¹ due to the breakage of H-bonds,^37,41 which is followed by a Newtonian flow region at [small gamma, Greek, dot above] > 10 s⁻¹. The addition of GO and RGO nanosheets dramatically changes the rheological behaviors. A weak shear thinning region appears instead of the original Newtonian plateau of the PA 6 solution at [small gamma, Greek, dot above] >10 s⁻¹. A scaling relation of η ∼ [small gamma, Greek, dot above]^−x is used to describe shear thinning in the strong and weak regions. The values of x are listed in Table 1. In the strong thinning region, the flow curves of the suspensions are nearly parallel to that of the PA 6 solution, revealing a similar predominant thinning mechanism of H-bonding breakage. At [small gamma, Greek, dot above] > 10 s⁻¹, the nanosheets bring a weak shear thinning that becomes more remarkable upon increasing C_F, revealing a mechanism associated with flow-induced anisotropy of the nanosheets of a high aspect ratio.^37,42

Fig. 2 (A) Steady viscosity (η) against shear rate ([small gamma, Greek, dot above]) for PA 6 solution (C = 20 wt%) and GO/PA 6 and RGO/PA 6 suspensions, (B) η at [small gamma, Greek, dot above] = 10⁻³ s⁻¹ and 10³ s⁻¹ as a function of C_F.

Table 1 Rheological exponents for PA 6 solution and nanosheet suspensions

Suspensions	x^a	x^b
a Strong shear thinning region.b Weak shear thinning region.
PA 6	0.53 ± 0.05	0.008 ± 0.001
0.5 wt% GO	0.58 ± 0.03	0.051 ± 0.003
1.0 wt% GO	0.61 ± 0.03	0.083 ± 0.004
1.5 wt% GO	0.65 ± 0.01	0.110 ± 0.002
0.5 wt% RGO	0.53 ± 0.04	0.020 ± 0.004
1.0 wt% RGO	0.49 ± 0.04	0.049 ± 0.004
1.5 wt% RGO	0.56 ± 0.03	0.071 ± 0.003

---

The influence of the nanosheets on η is examined at two different rates [small gamma, Greek, dot above] = 10⁻³ s⁻¹ and 10³ s⁻¹, shown in Fig. 2B. A small amount (0.5 wt%) of GO yields a fourfold increase in η_0.001, while RGO at the same loading leads to a threefold decrease in η_0.001. Increasing C_F causes a further increase in η_0.001, and the effect of GO is more significant than RGO. Similar viscosity variation with a low amplitude occurs at [small gamma, Greek, dot above] = 10³ s⁻¹. The results show that the viscosity of PA 6 solution can be effectively modulated in a wide range via simply controlling GO and RGO loadings, which is a beneficial extension of our previous work about polyvinyl alcohol (PVA) solution in a non-water soluble polymer system.³⁷

The amide group of PA 6 can interact with various nanofillers via hydrogen bond or donor–acceptor complexes.^34,43,44 GO, with a mass of polar groups and a large specific surface, brings strong interfacial H-bonding to replace the original inter-chain H-bonding among the PA 6 chains. The interfacial H-bonding could immobilize the PA 6 chains and create chain bridging between individual nanosheets, resulting in a marked increase of η with respect to C_F, together with the hydrodynamic effect. On the other hand, the steric segregation of RGO nanosheets with a large lateral size can heavily destroy the inter-chain H-bonding of PA 6, while little new interfacial H-bonding between the nanosheets and polymer chains form due to the few oxygen-containing groups of RGO. As a result, the total H-bonding density of the RGO/PA 6 suspension is lower than the PA 6 solution, causing a significant decrease of η at C_F = 0.5 wt%. At high C_F, the hydrodynamic effect dominates over the contribution of the reduced density H-bonding, causing a gradual viscosity increment with C_F. In brief, the total amount of inter-chain H-bonding of PA 6 and the interfacial H-bonding between PA 6 chains and nanosheets plays a crucial role in determining the suspension rheology.

3.2 SC window and morphology of nanofibers

The concentration and viscosity of the electrospinning solution determines the SC window of PA 6 to obtain nanofibers with good morphology. For polar polymers, however, the SC window is too narrow to meet the different needs.^37,45 For instance, considering the environmental impact and cost, electrospinning at a high concentration is desired for reducing solvent use.⁴⁶ To fabricate superfine fibers or composite nanofibers with a uniformly dispersed nanofiller, electrospinning at a low concentration is favored.⁴⁷ Therefore, broadening the SC window is vital and practical for the preparation of nanofibers with the desired morphology and performance. Solutions of linear polymers in good solvents can be usually divided into four concentration regimes, including the dilute, semidilute unentangled, semidilute entangled, and concentrated regimes.^48,49 The critical entanglement concentration, C_e, is regarded as the minimum concentration required for the electrospinning of beaded nanofibers, while 2–2.5 times the value of C_e is the minimum concentration required for the electrospinning of uniform, bead-free fibers.⁵⁰ For solutions free of specific polymer–solvent interaction, C_e can be estimated by the entanglement number (n_e), defined as:¹⁰where (M_e)_soln and M_e = 1.980 kg mol⁻¹ (ref. 51) are entanglement molecular weights in solution and in the melt state, respectively. M_w and ϕ_p are the average molecular weight and polymer volume fraction, respectively. Then, C_e of PA 6 solution is estimated as 6.0 wt% according to the equation:where ρ_p (1.15 g mL⁻¹) and ρ_soln (1.22 g mL⁻¹) are the density of PA 6 and the solution, respectively.

Fig. 3 shows SEM images and diameter distribution histograms of PA 6 nanofibers electrospun at C = 15 wt% (∼2.5C_e) and 30 wt% (5C_e), respectively, corresponding to the minimum concentration to obtain bead-free nanofibers, which is in agreement with previous reports^10,50 and the upper limit of spinnable concentration. All nanofibers show a unimodal distribution of diameter and are round, which is different from some other reports which obtained “fishnet-like” nanowebs or ribbon-like fibers.^34,52–54 The formation of nanowebs in PA 6 is ascribed to the electrically induced double layer in combination with the polyelectrolytic nature of the solution,⁵³ while the ribbon-like fibers result from the collapse of hollow tubes due to the rapid solvent removal from the surface of the jet.⁵⁴ Here, the existence of water (12 wt%) lowers the solvent evaporation rate and consequently restrains the surface of the jet from collapse, yielding the round-shape. Moreover, the PA 6 solution becomes neutral due to the addition of water.^55,56 Meanwhile, the relatively low voltage used here is also unfavorable for the formation of nanowebs. The average diameters of nanofibers spun at C = 15 wt% and 30 wt% are 68.78 and 177.09 nm, respectively, which are smaller than other reports.^9,16 To investigate the effect of the rheology of the spinning solution on the morphology of the nanofibers, η at [small gamma, Greek, dot above] = 1 s⁻¹ (approximating the flow rate of electrospinning at steady flow) was measured by the mode of peak holding test, as shown in Fig. 4 and Table 2. The SC window (15–30%) corresponds to η = 0.40–6.61 Pa s.

Fig. 3 SEM images (A1, B1) and diameter distribution histograms (A2, B2) of PA 6 nanofibers electrospun at C = 15 wt% (A1, A2) and C = 30 wt% (B1, B2).

Fig. 4 Peak hold curves of various spinning suspensions at [small gamma, Greek, dot above] = 1 s⁻¹.

Table 2 Viscosity values of spinning solutions and structural parameters of corresponding fibers

Sample	η at 1 s^−1 (Pa s)	Diameter (nm)	Half-peak width (nm)
15% PA 6	0.40 ± 0.01	68.78	15.89
30% PA 6	6.61 ± 0.01	177.09	79.84
10% PA 6	0.07 ± 0.01	—	—
GO/10% PA 6	0.37 ± 0.04	54.00	30.67
35% PA 6	24.38 ± 0.04	—	—
RGO/35% PA 6	9.68 ± 0.03	364.28	168.54
20% PA 6	0.93 ± 0.03	78.15	36.64
GO/20% PA 6	1.76 ± 0.01	73.03	20.47
RGO/20% PA 6	0.74 ± 0.02	70.20	25.45

---

At C = 10% (η = 0.07 Pa s), only beads and a few beaded fibers could be collected (Fig. 5A) due to the deficiency of molecular entanglement (n_e < 4); at C = 35% (η = 24.38 Pa s), only a few thick and interconnected nanofibers could be collected (Fig. 5C) and the electrospinning could not last for long, caused by nozzle blockage. Taking advantage of the modulating effect of GO and RGO on the suspension rheology, the viscosity of the spinning solution can be easily controlled. As shown in Fig. 4 and Table 2, a small amount (1 wt%) of GO nanosheets gives rise to a fourfold increase of η from 0.07 Pa s to 0.37 Pa s, and electrospinning yields bead-free nanofibers with diameters of 54.00 nm and a half-peak width of Gaussian distribution of 30.67 nm (Fig. 5B). Thereby, GO can be regarded as a kind of thickener to dramatically expand the lower limit of SC, which favors the preparation of superfine fibers. On the other hand, at C = 35%, marked condensation and nozzle blockage effects during electrospinning could be avoided by adding RGO due to the decrease of η from 24.38 Pa s to 9.68 Pa s. Consequently, uniform and continuous nanofibers with diameters of 364.28 nm and a half-peak width of 168.54 nm could be produced (Fig. 5D). RGO can thus be employed as a viscosity reducer and antigelation additive to significantly expand the upper limit of SC windows, which is especially vital for large-scale electrospinning with reduced solvent waste.

Fig. 5 SEM images of PA 6 (A, C), GO/PA 6 (B) and RGO/PA 6 (D) nanofibers electrospun from solutions or suspensions at C = 10 wt% (A and B) or C = 35 wt% (C and D). The insets in (B) and (D) show diameter distribution histograms of the nanocomposite fibers with C_F = 1 wt%.

Besides the broadening of the SC range, GO and RGO influence the fibrous morphology of the nanocomposite fibers electrospun at C = 20 wt%. Fig. 6 shows SEM and TEM images and diameter distribution histograms of the nanofibers electrospun without or with 1 wt% GO and RGO nanosheets. The addition of nanosheets leads to reductions of the nanofiber diameter and its distribution, which is reflected by the fitting parameters of the Gaussian distribution (Table 2). The TEM images show that nanosheets with a lateral size larger than the nanofiber diameter can successfully inset in the nanofibers and orient along the nanofiber axis, which might be ascribed to the flow-induced strong anisotropy due to the nanosheet orientation. However, the surface of the fiber becomes rough due to the existence of nanosheets. The favorable stretching of PA 6 chains along the orientation direction of the nanosheets and the improvement of conductivity induced by GO and RGO can perfect the uniformity, along with the decrease of diameter.⁵⁷ Being different from traditional methods for improving the uniformity of fibrous diameter distribution, such as increasing the molecular weight and concentration of the polymer, which inevitably increases the fibrous size, adding GO or RGO offers an effective strategy to improve the uniformity and decrease the fibrous diameter simultaneously.

Fig. 6 SEM images (A1, B1 and C1), TEM images (A2, B2 and C2) and diameter distribution histograms (A3, B3 and C3) of PA 6 (A) and its GO (B) and RGO (C) nanocomposite nanofibers electrospun at C = 20 wt% and C_F = 1 wt%.

3.3 Crystalline structure

The influences of GO and RGO on the crystalline property of nanofibers are explored. PA 6 is known to exist in two crystal forms, named α and γ.^17,18 The thermodynamically stable α form is a fully extended planar zigzag conformation.⁵⁸ In contrast, in the metastable γ form, the molecular conformation is helix,^12,59 and all amide bonds lie in the same direction. The two phases exhibit different mechanical properties and temperature dependences.¹⁹ According to Ito et al., the α phase exhibits a higher modulus below T_g but a more rapid decrease above T_g in comparison with the γ phase, and the γ phase is therefore able to improve the heat distortion temperature and toughness of PA 6.^19,60 Thus, altering the relative fraction of these two crystalline phases is of vital importance for mediating the mechanical properties of PA 6. The ratio of α and γ phases depends on the method of preparing samples and post treatment. In general, rapid crystallization (e.g. quenching and high-speed spinning) favors the γ form, while slow crystallization (e.g. slow cooling and solution crystallization) favors the α form.²⁰ Annealing above 150 °C, treating with an aqueous phenol solution and post-drawing can cause the γ to α transformation,^18,20,26 while the reverse can be achieved by iodine treatment.⁶¹ These strategies all involve post treatments, making the process much more tedious. On the other hand, the crystalline structure of PA 6 nanofibers can also be altered by adding nanofillers. It is reported that the presence of multi-wall carbon nanotubes (MWCNTs) favors the formation of the α phase, while nanoclays improve the fraction of γ phase, while the improving effects of these nanofillers on the morphology and performance of PA 6 nanofibers are not very obvious.^26,62

Fig. 7 demonstrates DSC heating and cooling scans of PA 6, GO/PA 6 and RGO/PA 6 nanofibers, and Table 3 shows the melting and crystallization temperatures (T_m and T_c). Also shown for comparison are casting films. The melting curves of the nanofibers show a shoulder at about 215 °C (T_m1) and a major peak at about 223 °C (T_m2), which are ascribed to the melting points of the γ- and α-form crystals, respectively. The absence of the T_m1 shoulder during the melting of the casting films suggests that the γ phase is formed during electrospinning where the solvent volatilizes rapidly. T_m2 and T_c of the PA 6 film are 218.0 °C and 185.9 °C, respectively. The addition of the nanosheets elevates T_m2 by 2–4 °C, which is consistent with earlier findings in PA 6,6.⁶³ The PA 6 nanofiber exhibits a T_m2 that is about 5.5 °C higher than the film, which could be ascribed to the molecular orientation produced during electrospinning. The presence of nanosheets in the nanofibers does not influence T_m2, suggesting a minor effect on the molecular packing order in the α-form crystals in comparison with the marked effect of high-speed electrospinning. The well dispersed nanosheets have a nucleation effect, causing T_c to increase by 4–5 °C in the films and by 2.2 °C in the nanofibers.

Fig. 7 DSC heating (A) and cooling (B) curves of casting films and nanofibers of PA 6, GO/PA 6 and RGO/PA 6, electrospun at C = 20 wt% and C_F = 1 wt%.

Table 3 T_m1, T_m2 and T_c of PA 6 and its nanocomposites based on DSC curves

Sample	Tm1 (°C)	Tm2 (°C)	Tc (°C)
20% PA 6-film	—	218.0	185.9
GO/20% PA 6-film	—	221.8	190.9
RGO/20% PA 6-film	—	220.1	191.0
20% PA 6-fiber	213.9	223.5	185.9
GO/20% PA 6-fiber	216.3	223.1	188.1
RGO/20% PA 6-fiber	215.9	223.9	188.2

---

Fig. 8 shows the XRD patterns of the films and nanofibers, confirming the different crystalline structures by DSC. All the casting films show two diffraction peaks at 2θ = 19.9° and 24.0° (Fig. 8A) that are associated with the (200) and mixed (002/202) planes of the α phase formed during solution crystallization. On the other hand, the nanofibers display an obvious diffraction peak at 2θ = 21.2° (Fig. 8B), associated with the (001) plane of the γ phase. It is surprising that GO and RGO have completely different effects on the content ratio (γ/α) of the γ to α forms, as shown in Fig. 9. As the content of GO is increased from 0 to 1.5 wt%, the γ/α value increases by 36.4% from 1.10 to 1.50. On the contrast, this value drops by 39.1% with 1.5 wt% RGO. In another word, GO favors the formation of γ-form crystals, while RGO inhibits it.

Fig. 8 XRD patterns of casting films (A) and nanofibers (B) of PA 6 with different amounts of GO and RGO.

Fig. 9 γ/α values of nanofibers with different amounts of GO and RGO.

The crystallization in casting films is very slow and there is enough time for PA 6 chains to adopt their conformation to form the stable α phase. On the contrary, the rapid solvent volatilization during electrospinning leads to the appearance of the minor γ phase along with the major α phase. The opposite effects of GO and RGO on the crystalline structures of the PA 6 nanofibers could be ascribed to the different surface properties and molecular mobility induced by various interfacial interactions. GO, with a mass of polar groups, can form strong GO-PA 6 H-bonding and cause conformational changes of the PA 6 chains. As a result, GO promotes a greater amount of the γ phase with twisted H-bonding to form, similar to nanoclays,⁶² leading to a higher γ/α value. On the other hand, RGO, with less polar groups and a large lateral area, could restrain the intermolecular H-bonding of PA 6, releasing unbonded chains that tend to rearrange to form the stable α phase. Moreover, RGO improves the conductivity of the solution and the jet suffers a stronger stretching force during electrospinning, facilitating the γ to α phase transformation,^26,64 as reported for the FeCl₂-induced decrease of the γ phase.⁶⁵ Thus, the introduction of RGO in the electrospinning solution promotes the γ to α transformation and lowers the γ/α value in comparison with PA 6 nanofibers. As aforementioned, the relative fraction of these two crystalline phases is involved in the mechanical properties of PA 6.^19,60 The opposite effects of GO and RGO thus provide a new strategy to mediate the mechanical properties of PA 6 nanofibers.

3.4 Thermal stability

The effects of GO and RGO on the thermal stability of the nanofibers were investigated by TGA under a nitrogen environment, as shown in Fig. 10. The PA 6 nanofibers demonstrate a single-stage thermal decomposition, which is coincident with other reports.^66,67 The degradation curves of the composite nanofibers are clearly shifted towards high temperatures in comparison with the neat PA 6 nanofibers. The temperatures at 5% (T_5%) and 10% (T_10%) weight loss and the maximum decomposition temperature (T_max) corresponding to the peak temperature (T_p) on the DTG curve (Fig. 10B) for PA 6 and its composite nanofibers are summarized in Table 4. As is shown, the incorporation of a small amount (1 wt%) of nanosheets can markedly improve the thermal stability of the nanofibers, which is much more efficient than nanoclays and MWCNTs.^66–68 This result is probably caused by the excellent thermal resistance of the nanosheets with a large lateral area and strong secondary interactions between the PA 6 chains and nanosheets.^69,70

Fig. 10 (A) TGA and (B) DTG curves of PA 6, GO/PA 6 and RGO/PA 6 nanofibers electrospun at C = 20 wt% and C_F = 1 wt%.

Table 4 Results of TGA analysis of PA 6, GO/PA 6 and RGO/PA 6 nanofibers

Sample	T5% (°C)	T10% (°C)	Tmax (°C)
PA 6	324.7	341.6	404.9
GO/PA 6	359.7	377.0	424.5
RGO/PA 6	367.6	382.7	423.0

---

3.5 Adsorbability to copper ions

Electrospun nanofibers have been proved to be an efficient adsorbent for heavy metal ions.⁷¹ The addition of nanosheets might improve the adsorption capacity of PA 6 nanofibers, due to the large specific surface area and metal coordination ability of GO and RGO.^72,73 Fig. 11 shows the adsorption isotherms of nanofibers for copper ions (Cu²⁺), and fitted curves with Langmuir⁷⁴ and Freundlich⁷⁵ isothermal adsorption models. The parameters of the isotherm models calculated are presented in Table S1.† As shown in Fig. 11, the equilibrium data of all samples are better depicted by the Langmuir model (R² > 0.993) compared to the Freundlich model (R² > 0.845), indicating that the sorption of Cu²⁺ by nanofibers is a monolayer coverage. The maximum adsorption capacities (Q_m) of PA 6, GO/PA 6 and RGO/PA 6 nanofibers are 4.77, 5.33, 4.96 mg g⁻¹, respectively, suggesting that a small amount (1 wt%) of GO and RGO nanosheets can efficiently increase the adsorption capacity of PA 6 nanofibers towards Cu²⁺ by 11.7% and 4.0%, respectively. The GO/PA 6 nanofibers have a higher Q_m value because GO, with more functional groups, can coordinate with metal ions. Introducing a functional nanofiller such as GO opens up a new convenient way to prepare nanofibers with high adsorption capacity, which is usually achieved by the tedious post functionalization of as-spun fiber mats.⁷⁶

Fig. 11 Adsorption isotherms of Cu²⁺ onto PA 6, GO/PA 6 and RGO/PA 6 nanofibers electrospun at C = 20 wt% and C_F = 1 wt%.

The morphology, structure and performance of PA 6 nanofibers are correlative to each other and the simultaneous regulation of these three is important to guide applications. Traditional methods, including parameter controlling and post treatment, which often involve special conditions and tedious operations, tend to affect only one or two aspects of them. For example, altering the solution and process parameters, such as the viscosity of solution, voltage, and distance can optimize the morphology of the nanofibers;¹⁶ controlling the solvent evaporation during electrospinning can obtain various crystal phases;²¹ and after sputtering coatings of tin-doped indium oxide, the conductivity and light transmittance of PA 6 nanofibers are improved.⁷⁷ However, by adding GO and RGO nanosheets, the morphology, crystalline structure and performance of PA 6 nanofibers can be regulated simultaneously, so the complicated controlling of multiparameters and post treatments are avoided. This method can be expected to provide a simple and convenient way to guide the simultaneous regulation of the morphology and performance of nanofibers electrospun from other polymers.

4. Conclusions

The rheological behaviors of PA 6 spinning solutions could be effectively modulated in a wide range via simply controlling the loading of GO or RGO. A small amount (0.5 wt%) of GO can dramatically increase the η_0.001 by fourfold, while the same content of RGO results in a threefold decrease, which is related to the formation of interface H-bonding between the PA 6 chains and nanosheets in the first case, and the reduction in the intermolecular H-bonding in both cases. The SC range of the PA 6 electrospinning solution is broadened from (15–30%) to (10–35%) by adding 1 wt% GO as a thickener or 1 wt% RGO as a viscosity reducer, offering an effective strategy to improve the uniformity and decrease the fiber size at the same time. A metastable γ-form crystal appears during the crystallization of PA 6 nanofibers, which is promoted by GO, but restrained by RGO. Meanwhile, the thermal stability of the nanofibers is enhanced greatly by nearly 20 °C after adding GO or RGO. And the incorporation of 1 wt% GO and RGO can increase the adsorption capacity of PA 6 nanofibers towards Cu²⁺ by 11.7% and 4.0%, respectively. This work thus affords a new convenient way to modulate the morphology, crystalline structure, mechanical and thermal properties, as well as the adsorbability of electrospun PA 6 nanofibers simultaneously, and can be further employed as a common strategy to guide the preparation of functional nanofibers electrospun from other polymers.

Acknowledgements

This work was supported by the National Natural Science Foundation of Zhejiang Province (Grant No. R14E030003), the National Natural Science Foundation of China (Grant No. 51573157, 51333004, 51373149 and 51403113), and the Major Projects of Science and Technology Plan of Guizhou Province (Grant No. (2013) 6016).

References

1. D. H. Reneker and I. Chun, Nanotechnology, 1996, 7, 216–223.
2. D. Li and Y. N. Xia, Adv. Mater., 2004, 16, 1151–1170.
3. J. S. Im, S. C. Kang, S.-H. Lee and Y.-S. Lee, Carbon, 2010, 48, 2573–2581.
4. J. Q. Wang, K. Pan, E. P. Giannelis and B. Cao, RSC Adv., 2013, 3, 8978–8987.
5. M. Li, H. Wang, S. Wu, F. Li and P. Zhi, RSC Adv., 2012, 2, 900–907.
6. G. Wang, Q. Dong, T. Wu, F. Zhan, M. Zhou and J. Qiu, Carbon, 2016, 103, 311–317.
7. G. Wang, Q. Dong, Z. Ling, C. Pan, C. Yu and J. Qiu, J. Mater. Chem., 2012, 22, 21819–21823.
8. J. M. Deitzel, J. Kleinmeyer, D. Harris and N. C. B. Tan, Polymer, 2001, 42, 261–272.
9. C. Mit-uppatham, M. Nithitanakul and P. Supaphol, Macromol. Chem. Phys., 2004, 205, 2327–2338.
10. S. L. Shenoy, W. D. Bates, H. L. Frisch and G. E. Wnek, Polymer, 2005, 46, 3372–3384.
11. E. R. Kenawy, J. M. Layman, J. R. Watkins, G. L. Bowlin, J. A. Matthews, D. G. Simpson and G. E. Wnek, Biomaterials, 2003, 24, 907–913.
12. R. Dersch, T. Q. Liu, A. K. Schaper, A. Greiner and J. H. Wendorff, J. Polym. Sci., Part A: Polym. Chem., 2003, 41, 545–553.
13. H. Wang, Y. Li, Y. Zuo, J. Li, S. Ma and L. Cheng, Biomaterials, 2007, 28, 3338–3348.
14. J. S. Stephens, D. B. Chase and J. F. Rabolt, Macromolecules, 2004, 37, 877–881.
15. I. Steyaert, G. Vancoillie, R. Hoogenboom and K. De Clerck, Polym. Chem., 2015, 6, 2685–2694.
16. P. Heikkila and A. Harlin, Eur. Polym. J., 2008, 44, 3067–3079.
17. D. R. Holmes, C. W. Bunn and D. J. Smith, J. Polym. Sci., 1955, 17, 159–177.
18. Y. Kinoshita, Makromol. Chem., 1959, 33, 1–20.
19. M. Ito, K. Mizuochi and T. Kanamoto, Polymer, 1998, 39, 4593–4598.
20. Y. Liu, L. Cui, F. X. Guan, Y. Gao, N. E. Hedin, L. Zhu and H. Fong, Macromolecules, 2007, 40, 6283–6290.
21. C. B. Giller, D. B. Chase, J. F. Rabolt and C. M. Snively, Polymer, 2010, 51, 4225–4230.
22. K. H. Lee, K. W. Kim, A. Pesapane, H. Y. Kim and J. F. Rabolt, Macromolecules, 2008, 41, 1494–1498.
23. C. Wang, S.-Y. Tsou and H.-S. Lin, Colloid Polym. Sci., 2012, 290, 1799–1809.
24. C.-L. Zhang and S.-H. Yu, Chem. Soc. Rev., 2014, 43, 4423–4448.
25. L. Li, L. M. Bellan, H. G. Craighead and M. W. Frey, Polymer, 2006, 47, 6208–6217.
26. M. V. Jose, B. W. Steinert, V. Thomas, D. R. Dean, M. A. Abdalla, G. Price and G. M. Janowski, Polymer, 2007, 48, 1096–1104.
27. L. Francis, F. Giunco, A. Balakrishnan and E. Marsano, Curr. Appl. Phys., 2010, 10, 1005–1008.
28. J. Li, L. Tian, N. Pan and Z.-j. Pan, Polym. Eng. Sci., 2014, 54, 1618–1624.
29. F. Schedin, A. K. Geim, S. V. Morozov, E. W. Hill, P. Blake, M. I. Katsnelson and K. S. Novoselov, Nat. Mater., 2007, 6, 652–655.
30. G. K. Dimitrakakis, E. Tylianakis and G. E. Froudakis, Nano Lett., 2008, 8, 3166–3170.
31. S. Bai and X. Shen, RSC Adv., 2012, 2, 64–98.
32. W. Wan, Z. Zhao, T. C. Hughes, B. Qian, S. Peng, X. Hao and J. Qiu, Carbon, 2015, 85, 16–23.
33. Q. Dong, G. Wang, B. Qian, C. Hu, Y. Wang and J. Qiu, Electrochim. Acta, 2014, 137, 388–394.
34. H. R. Pant, C. H. Park, L. D. Tijing, A. Amarjargal, D. H. Lee and C. S. Kim, Colloids Surf., A, 2012, 407, 121–125.
35. H. R. Pant, B. Pant, C. H. Park, H. J. Kim, D. S. Lee, L. D. Tijing, B. S. Hwang, H. Y. Kim and C. S. Kim, Fiber Polym., 2013, 14, 970–975.
36. B. Y. Li, H. H. Yuan and Y. Z. Zhang, Compos. Sci. Technol., 2013, 89, 134–141.
37. Y. Q. Tan, Y. H. Song and Q. Zheng, Nanoscale, 2012, 4, 6997–7005.
38. W. S. Hummers and R. E. Offeman, J. Am. Chem. Soc., 1958, 80, 1339.
39. M. J. Fernandez-Merino, L. Guardia, J. I. Paredes, S. Villar-Rodil, P. Solis-Fernandez, A. Martinez-Alonso and J. M. D. Tascon, J. Phys. Chem. C, 2010, 114, 6426–6432.
40. P. Karrer, Chem. Rev., 1934, 14, 17–30.
41. H. W. Gao, R. J. Yang, J. Y. He and L. Yang, J. Appl. Polym. Sci., 2010, 116, 1459–1466.
42. A. Guimont, E. Beyou, G. Martin, P. Sonntag and P. Cassagnau, Macromolecules, 2011, 44, 3893–3900.
43. H. R. Pant, M. P. Bajgai, K. T. Nam, Y. A. Seo, D. R. Pandeya, S. T. Hong and H. Y. Kim, J. Hazard. Mater., 2011, 185, 124–130.
44. H. R. Pant, M. P. Bajgai, K. T. Nam, K. H. Chu, S.-J. Park and H. Y. Kim, Mater. Lett., 2010, 64, 2087–2090.
45. C. S. Ki, D. H. Baek, K. D. Gang, K. H. Lee, I. C. Um and Y. H. Park, Polymer, 2005, 46, 5094–5102.
46. W. K. Son, J. H. Youk, T. S. Lee and W. H. Park, Mater. Lett., 2005, 59, 1571–1575.
47. J. C. Sy, A. S. Klemm and V. P. Shastri, Adv. Mater., 2009, 21, 1814–1819.
48. R. H. Colby, L. J. Fetters, W. G. Funk and W. W. Graessley, Macromolecules, 1991, 24, 3873–3882.
49. W. E. Krause, E. G. Bellomo and R. H. Colby, Biomacromolecules, 2001, 2, 65–69.
50. M. G. McKee, G. L. Wilkes, R. H. Colby and T. E. Long, Macromolecules, 2004, 37, 1760–1767.
51. L. J. Fetters, D. J. Lohse, D. Richter, T. A. Witten and A. Zirkel, Macromolecules, 1994, 27, 4639–4647.
52. B. Ding, C. R. Li, Y. Miyauchi, O. Kuwaki and S. Shiratori, Nanotechnology, 2006, 17, 3685–3691.
53. R. Nirmala, K. T. Nam, S.-J. Park, Y.-S. Shin, R. Navamathavan and H. Y. Kim, Appl. Surf. Sci., 2010, 256, 6318–6323.
54. H. Fong, W. D. Liu, C. S. Wang and R. A. Vaia, Polymer, 2002, 43, 775–780.
55. S.-Y. Tsou, H.-S. Lin and C. Wang, Polymer, 2011, 52, 3127–3136.
56. S.-Y. Tsou, H.-S. Lin, P.-J. Cheng, C.-L. Huang, J.-Y. Wu and C. Wang, Eur. Polym. J., 2013, 49, 3619–3629.
57. G. M. Chen, S. H. Liu, S. F. Zhang and Z. N. Qi, Macromol. Rapid Commun., 2000, 21, 746–749.
58. K. Song and J. F. Rabolt, Macromolecules, 2001, 34, 1650–1654.
59. A. Okada, M. Kawasumi, I. Tajima, T. Kurauchi and O. Kamigaito, J. Appl. Polym. Sci., 1989, 37, 1363–1371.
60. D. M. Lincoln, R. A. Vaia, Z. G. Wang and B. S. Hsiao, Polymer, 2001, 42, 1621–1631.
61. H. Arimoto, J. Polym. Sci., Part A: Gen. Pap., 1964, 2, 2283–2295.
62. Y. B. Cai, Q. Li, Q. F. Wei, Y. B. Wu, L. Song and Y. Hu, J. Mater. Sci., 2008, 43, 6132–6138.
63. F. Navarro-Pardo, G. Martinez-Barrera, A. L. Martinez-Hernandez, V. M. Castano, J. L. Rivera-Armenta, F. Medellin-Rodriguez and C. Velasco-Santos, Materials, 2013, 6, 3494–3513.
64. S. Murase, M. Kashima, K. Kudo and M. Hirami, Macromol. Chem. Phys., 1997, 198, 561–572.
65. N. Kimura, B.-S. Kim and I.-S. Kim, Polym. Int., 2014, 63, 266–272.
66. K. P. Pramoda, T. X. Liu, Z. H. Liu, C. B. He and H. J. Sue, Polym. Degrad. Stab., 2003, 81, 47–56.
67. J. Li, L. Tong, Z. Fang, A. Gu and Z. Xu, Polym. Degrad. Stab., 2006, 91, 2046–2052.
68. Q. Li, D. W. Gao, Q. F. Wei, M. Q. Ge, W. M. Liu, L. J. Wang and K. Q. Hu, J. Appl. Polym. Sci., 2010, 117, 1572–1577.
69. F. Z. Rafique and N. Vasanthan, J. Phys. Chem. B, 2014, 118, 9486–9495.
70. H.-H. Liu, W.-W. Peng, L.-C. Hou, X.-C. Wang and X.-X. Zhang, Compos. Sci. Technol., 2013, 81, 61–68.
71. H. Y. Ma, B. S. Hsiao and B. Chu, Curr. Org. Chem., 2013, 17, 1361–1370.
72. R. Sitko, E. Turek, B. Zawisza, E. Malicka, E. Talik, J. Heimann, A. Gagor, B. Feist and R. Wrzalik, Dalton Trans., 2013, 42, 5682–5689.
73. M. Machida, T. Mochimaru and H. Tatsumoto, Carbon, 2006, 44, 2681–2688.
74. I. Langmuir, J. Am. Chem. Soc., 1918, 40, 1361–1403.
75. H. Freundlich, J. Am. Chem. Soc., 1906, 62, 121–125.
76. P. Kampalanonwat and P. Supaphol, ACS Appl. Mater. Interfaces, 2010, 2, 3619–3627.
77. Q. Wei, H. Wang, Y. Xu and B. Deng, J. Coat. Technol. Res., 2010, 7, 511–514.

---

Footnote

† Electronic supplementary information (ESI) available: Photographs of GO and RGO suspensions, FTIR, UV-Vis, and Raman spectra, XRD and TGA of GO and RGO, Langmuir and Freundlich isotherm parameters. See DOI: 10.1039/c6ra05255j

---