A. Materials

PE/PET composite fibers were purchased from SINOPEC Shanghai Petrochemical Company. The dimethylformamide (DMF), AN, CPDB and other chemical agents, all of analytical grade, were purchased from Sinopharm Chemical Reagent Co., Ltd.

B. RAFT graft polymerization

The 5%–20% monomer solutions were prepared by dissolving acrylonitrile in DMF. Different amount of chain transfer agent CPDB were added to the solutions. The mixtures of acrylonitrile (AN) and CPDB were contained in glass vials. The PE/PET fibers were added to the vials, and deoxygenated by bubbling with nitrogen gas for 15 min. The samples were irradiated at room temperature in a ⁶⁰Co source to different doses at dose rate of 0.18 kGy/h or to 12 kGy at different dose rates of 0.12, 0.16, 0.19, 0.23 and 0.3 kGy/h. Different initial monomer concentrations and initial molar ratios between monomer and RAFT agent ([AN]₀/[CPDB] ₀) were used. For comparison, a group of samples without CPDB were irradiated to complete grafting polymerization. The grafted samples were extracted by DMF to remove homopolymer, and vacuum-dried for 24 h.

C. Degree of grafting (DG)

The degree of grafting is calculated by

where W_g is the dry weight of grafted polymer, W₀ is the initial weight of PE/PET fibers.

D. Characterizations

A RAFT mediated grafting polymer was measured on a gel permeation chromatography (GPC) system to analyze the molecular weight and polydispersity index (PDI). The grafted polymers were characterized by FT-IR spectra on a Nicolet Avatar FT-IR spectrometer in the range of 3500–500 cm^-1. Thermogravimetric (TG) analysis was performed on a TG 209 F3 Tarsus (NETZSCH, Germany) instrument at a heating rate of 10 ℃/min under nitrogen atmosphere. Morphology and diameter of the fibers were analyzed with an scanning electron microscope (SEM) instrument (FESEM, JEOL, JSM6700F). Mechanical properties were tested on an Electronic Single Silk Strength Tester (LaiZhou Electron Instrument Co., Ltd.).

A. Kinetics of graft polymerization

The relationships between DG and absorbed dose, dose rate, monomer concentration and CTA concentration were studied. The dose-DG effect of PE/PET-g-PAN is shown in Fig. 1(a), from the samples with an initial AN concentration of 15% (v/v) and [AN]₀/[CPDB]₀=800:1, irradiated to different doses at 0.18 kGy/h. The DG increased linearly with the absorbed dose. Free radicals which initiate the graft polymerization were generated by the γ-rays on surface of PE/PET fibers and in the monomer solution. The propagating chains combined to the thiocarbonyl group of RAFT agents, which in turn released the living group radicals and re-initiated polymerization. All the RAFT combining and releasing steps were reversible, when an equilibrium was established [26], the rate of free radical generation was proportional to the absorbed dose under a fixed dose rate. The DG increase was proportional to the number of free radicals, so the DG increased linearly with dose when the graft polymerization was performed in presence of the RAFT agent (CPDB). However, without CPDB in the grafting system under the same conditions as the samples in Fig. 1(a), the DG increase was not linear (Fig. 1(b)), because the equilibrium mentioned above did not exist.

Fig. 1.Full-screen View

(Color online) DG of PE/PET-g-PAN versus absorbed dose at [AN]₀/[CPDB]₀=800:1 (a) and without CPDB (b), at dose rate of 0.18 kGy/h and initial AN concentration of 15%(v/v).

Fig. 2(a) shows the effect of dose rate on grafting rate from the 12 kGy-irradiated samples with an initial AN concentration of 15% (v/v) and [AN]₀/[CPDB]₀=800:1. As the rate of free radical generation is proportional to the dose, the grafting rate, i.e., the amount of increased weight of PE/PET-g-PAN per hour, increased in a linear relationship with the dose rate. Fig. 2(b) shows also a linear relationship between the initial monomer concentration and DG, from the samples with [AN]₀/[CPDB]₀=800:1, irradiated to 12 kGy at 0.19 kGy/h.

Fig. 2.Full-screen View

(Color online) Grafting rate versus dose rate (a) and initial monomer concentration (b), at absorbed dose=12 kGy and [AN]₀/[CPDB]₀=800:1.

The effect of RAFT agent concentration on DG was also studied, with different initial molar ratios of [AN]₀/[CPDB]₀ from 400:1 to 1500:1. As shown in Fig. 3, the DG changed slightly. The radiation chemical yields (G-value) of PE [27] and acrylonitrile [5] are 4 and 2.4–5.6 radicals per 100 eV, respectively. With AN concentration of 15%(v/v), assuming that the weight of substrate polymer is 1g, it can be estimated that the number of free radicals generated by AN monomer and the substrate polymer is about 1.8×10¹⁶. The minimum initial molar ratio between RAFT agent and monomer is 1:1500, so the number of CPDB molecules is about 2.8×10¹⁹, being far more than the number of free radicals. Therefore, when the free radicals are generated, they will combine with the RAFT agents immediately to establish the equilibrium, and the grafting rates are about the same in this molar ratio range.

Fig. 3.Full-screen View

(Color online) DG of PE/PET-g-PAN versus the molar ratio between monomer and RAFT agent.

The kinetics results are in accordance with typical living polymerization and the degree of grafting can be controlled by changing experimental conditions.

B. Characterization of the PE/PET-g-PAN fibers

1. GPC results

Since the molecular weight of grafting polymer in the composite fiber is difficult to analysis, the homopolymer in solution of the sample of DG=4.2% (Fig. 1(a)) was measured by GPC. As shown in Fig. 4, the polydispersity index of the homopolymer is 1.24, which is in agreement with the results of conventional RAFT polymerization.

Fig. 4.Full-screen View

GPC image of the homopolymer formed in solution of the PE/PET-g-PAN fibers, at DG=4.2%, dose rate=0.18 kGy/h, initial AN concentration =15%(v/v), and [AN]₀/[CPDB]₀=800:1.

2. FT-IR spectra

Figure 5 shows FT-IR spectra of the pristine PE/PET composite fiber and PE/PET-g-PAN fibers of DG = 4.2% and 17%. The absorption bands at 2919 cm^-1 and 2851 cm^-1 are attributed to the asymmetric and symmetric stretching of -CH₂- groups, respectively [28]. The bending deformation of -CH₂- groups is demonstrated by the absorption band at 1463 cm^-1 [28]. The absorption bands at 1720, 1100 and 720 cm^-1 representing the stretching vibrations of ester groups and the para-substitution of benzene ring represented by the band at 870 cm^-1 are attributed to the PET fibers contained in the original composite fibers [29]. The PAN graft chains is confirmed by the nitrile groups in acrylonitrile,and is represented by the new peak at 2243 cm^-1 in the spectra of grafted fibers [9]. The spectra also show that the PE/PET-g-PAN of DG=17% has a larger peak than that of DG=4.2%. The FT-IR results indicate that acrylonitrile is grafted onto PE/PET composite fibers successfully.

Fig. 5.Full-screen View

(Color online) FT-IR spectra of pristine PE/PET composite fibers, PE/PET-g-PAN fibers with DG = 4.2% and 17.0%.

3. TGA analysis

Thermal stability of pristine PE/PET fiber and grafted fibers was studied by TGA measurements. Fig. 6 shows that the PE/PET fiber started degrading at 336 ℃, which is in accordance with the degradation temperature of PE. The residue should be attributed to PET core of the fibers because PE gives no residue at 700 ℃ [29, 30]. From TG curves of the grafted fibers, it can be seen that there are two degradation stages of the grafted fibers. The degradation stage at lower temperature is attributed to the weak bonds formed on the pristine polymer by graft polymerization. The second degradation stage is due to the grafted PAN chains [31]. Its starting temperature decreases with increasing DG. It can be observed that the residue increased from 10.5% at DG=4.2% to 17.1% at DG=17.0%, which is due to carbonization of PAN grafted chains.

Fig. 6.Full-screen View

(Color online) TGA spectra of pristine PE/PET composite fibers, PE/PET-g-PAN at different DGs.

4. Scanning electron microscopy

Surface morphology of the RAFT mediated grafted PE/PET-g-PAN fibers (DG=10.0%) was compared with a PE/PET-g-PAN sample prepared via conventional grafted polymerization (without CPDB) with a similar DG (10.3%). SEM images of the samples are shown in Fig. 7. Surface of the RAFT mediated PE/PET-g-PAN fibers was smoother than the other sample. This is because the RAFT mediated polymerization is controllable through the formation of intermediate radicals and the polymerization rate is steady, hence the simultaneous growth of the grafted chains and the same length at the same instant of time.

Fig. 7.Full-screen View

SEM images of RAFT mediated PE/PET-g-PAN fibers with a DG of 10.0% at low (a) and high (b) magnification, and conventional grafted PE/PET-g-PAN fibers with a DG of 10.3% at low (c) and high (d) magnification.

Figure 8 shows SEM images of the pristine PE/PET fibers and the RAFT mediated grafted fibers of DG=17.0%. The diameters of the pristine and grafted fiber are 17.27 μm and 19.72 μm, respectively. It can be seen that the fibers with DG of 17% still has a smooth surface. As mentioned above, the RAFT mediated graft polymerization was controlled by RAFT agent and the grafting rate was slow, so the grafted chains propagated moderately without accumulating at one place.

Fig. 8.Full-screen View

SEM images of RAFT mediated PE/PET-g-PAN fibers with DG of 17.0% at low (a) and high (b) magnification, and pristine PE/PET composite fibers at low (c) and high (d) magnification.

5. Mechanical Properties

Mechanical properties of PE/PET-g-PAN fibers, which are expected to be maintained to maximal extent, are shown in Fig. 9. It can be seen that the breaking strength and breaking elongation of single silk change slightly compared to the pristine PE/PET fibers. This is because that mechanical properties of the composite fibers are mostly attributed to the PET inner core and PET is very stable under radiation [32].

Fig. 9.Full-screen View

(Color online) Breaking strength and breaking elongation of single fiber with different DGs.