Dioctyl phthalate (DOP, 98%), diethyl phosphite (99%), N-hexane (97%), toluene (98%), triphenyphosphine (95%), dichloromethane (99%), ethyl acetate (99%), calcium stearate (Zn 10%–12%) and sodium chloride (99.5%) were bought from Aladdin reagent (Shanghai) Co., Ltd. (China). S-1100 PVC resin (no additives) was provided by Jinling Chemical Industry Co. (Nanjing, China). Rubber seed oil (RSO) was provised by Kuming Jinning Chemical Industry Co., Ltd. (China).

RSO was dissolved in dichloromethane and put in the a round bottom flask, which was put in the ice bath. 3-Chloroperoxybenzoic acid was dropped in the mixture (mole ratio of double bonds to 3-Chloroperoxybenzoic acid is 1:0.5). The reaction was kept stirring for 5 h at room temperature to obtain the raw product. The obtained raw product was dissolved in dichloromethane and washed with brine and distilled water, then dried by anhydrous magnesium sulfate (MgSO₄) and purified using silica gel column chromatography to obtain ERSO.

ERSO (20 g) was dissolved in toluene (16 g), all of which were blended in a three-necked round-bottom flask. The mix of diethyl phosphate (6 g), triphenyphosphine (0.02 g) and toluene (16 g) were also put in the flask within 40 min at room temperature. Then, a 4 h-stir at 75°C was carried out. The reaction mixture was dropped to room temperature with sodium hydroxide (NaOH) solution and washed until pH = 7. FRP was obtained via removing water by rotary-evaporation, after a 3-time-wash through distilled water. The synthesis of FRP was showed in Fig. 1.

FRP-Plasticized PVC blends were prepared by melt processing and the formulations were illustrated in Table 1. Then, further extrusion into strips like dumbbell or rectangular was necessary for subsequent tests.

RSO, ERSO and FRP were respectively illustrated through Fourier transform infrared spectrometer (FT-IR) (Nicolet 6700), with the chosen wavelength between 4000 and 500 cm⁻¹.¹H nuclear magnetic resonance (¹H NMR) spectrums were presented with the solvent of CDCl₃ and internal standard of tetramethylsilane (TMS), respectively. Combustion tests were carried out according to GB/T 2406.1-2008. The FRP-modified PVC samples were designed into 80 mm × 10 mm × 4 mm. FTT200 cone calorimeter was used to analyze the combustion properties of samples that were cut into 100 mm × 100 mm× 2 mm, and ISO-5660 standard procedures would be referenced. The thermogravimetric analysis (TGA) was detected with surrounding N₂ in the flow rate of 50 mL/min. The heating rate of PVC blends was set to 10 °C/min from the beginning of 50°C to the final 600°C. The leaching tests related to plasticizers were examined in term of ASTMD1239-98 with N-hexane as the chosen solvent. The PVC blends were submerged into solvents at the temperature of 30°C for one week. Thereafter, the samples were removed and the surface solvent was wiped off. Then further drying was conducted at 50°C for whole day. The migration rate was calculated in term of the following Eq. (1):(1) Migrationrate=(W1−W2)/W1

Thereinto, W₁ and W₂ represent the front and rear weight, respectively.

In addition, the tensile rate of electronic tensile testing machine was set to 20 mm/min, and the dynamic mechanical analysis (DMA) was brought into effect to investigate the specific glass transition temperature (T_g), from −80°C to 100°C at a heating speed of 3 °C/min. Limiting oxygen index (LOI) values were measured on a JF-3 oxygen index measuring instrument.

FTIR spectra of RSO, ERSO and FRP were shown in Fig. 2. The characteristic peaks around 3007 cm⁻¹ were associated with -CH=CH- in RSO [24,25]. As Compared with the spectra of ERSO, the characteristic peak of CH=CH- vanished, opposite to the -C-O-C vibration which could be observed at approximately 972 and 1062 cm⁻¹. All mentioned indicated that the -C=C- double bonds were converted into epoxy groups [26,27]. The peak at 1032 cm⁻¹ was attributed to -P-O-CH- symmetric bending vibration, and -OH­­ was observed at 3470 cm⁻¹ in the FT-IR of FRP. The epoxy group at 972 and 1062 cm⁻¹ cannot be observed in the FT-IR of FRP [28,29], which means that the reaction between P-OH and ERSO was accomplished.

The ¹H NMR spectrums pertaining to RSO, ERSO and FRP are exhibited in Fig. 3. The strong peak at δ = 5.28 ppm is attributed to the protons of -CH=CH- in RSO [25,26]. When epoxidation was finished, the peak at δ = 5.28 ppm became weak in the ¹H NMR spectrum of ERSO, because the epoxidation reaction was not complete, and an emerged peak at 2.95–3.20 ppm was assigned to protons on the epoxy groups of ERSO [26,27], which indicated that some -CH=CH- bonds of RSO were converted into epoxy groups. In the ¹H NMR spectrum of FRP, a new proton signal of -P-O-CH₂- appeared at δ = 2.3 ppm [28,29], and the protons signals in the 5.28 ppm associated with -C-O-C- bonds showed weaker than ERSO, which indicated that chemical structure of FRP contains phosphate groups, which indicated that FRP was synthesized successfully.

The melting behavior was confirmed in the Haake chamber by gathering the processing data during the mix. Fig. 4 presents the torque values relevant to the proportion of plasticizer for the 5 min processing time. When more DOP was replaced with FRP, the torque of PVC blends increased from 11.4 to 18.4 N⋅m, respectively, corresponding to PVC-FRP-0 and PVC-FRP-20. In other words, 61% augmentation could be achieved, showing disadvantages in conduction of thermoplastic processing pertains to PVC blends.

FRP contains many flame-resistant phosphate groups, which could be used as a alternative to DOP as the flame-resistant plasticizer. The flame-resistance of plasticized PVC blends was conducted by the LOI tests and the variation with regard to FRP content was described in Fig. 5. When DOP was replaced with FRP in plasticized PVC blends, the LOI value increased from 24.3% for PVC-FRP-0% to 33.1% for PVC-FRP-20. Only the latter one exhibited greater value than 26.0%, to a certain extent, displaying self-blackout capacity. Generally, FRP is not suitable to be considered as a preferable flame retardant plasticizer in the PVC blends [30,31].

PVC chemical structure contains a large amount of chlorine, which makes it hard to burn. However, it will combust and release much heat when PVC products plasticized with DOP, being blended into PVC. Fig. 6 showed the heat release rate (HRR) curves of plasticized PVC blends. There exists a notable difference in HRR curves between PVC-FRC-0 and PVC-FRC-20. Almost all HRR curve of PVC-FRC-0 is higher than PVC-FRC-20, which indicats that PVC-FRC-0 released heat faster than PVC-FRC-20. Once ignited, PVC-FRC-0 produced flammable volatiles and released a lot of heat. However, the PPVC-FRC-20 was harder to burn than PVC-FRC-20. Thermal degradation of phosphate ester groups containing FRP produces a large amount of phosphoric acid and polyphosphate [31–33], which can promote carbon formation by removing water from FRP. The char residue will block heat transfer and the pyrolysis process. In addition, the total heat release (THR) curves of plasticized PVC blends were shown in Fig. 7. It can be found that the whole THR curve of PVC-FRC-0 was higher than PVC-FRC-20, which indicated that PVC-FRC-0 released more heat than PVC-FRC-20. The THR value decreased from 39 MJ/m² (pertaining to PVC-FRC-0) to 22 MJ/m² (pertaining to PVC-FRC-20) when 20 phr DOP was replaced with FRP in the plasticized PVC blends. All mentioned above revealed that FRP had capacity of improving the flame retardancy.

The thermogravimetric loss data of plasticized PVC blends with different mass of plasticizers were shown in the Figs. 8 and 9, and Table 2. As is demonstrated in the Fig. 8, the degradation temperature (T_d: Temperature at 5% weight loss) of PVC blends increased with more DOP replaced by FRP, from the initial content of 5% FRP. When PVC blends plasticized with 40 phr DOP, T_d is 252°C, while the T_d reached 272°C when PVC blends plasticized with 20 phr DOP and 20 phr FRP. In addition, the temperature at 10% weight loss (T₁₀) and the 50% one (T₅₀) showed the same increasing trend when DOP was replaced with FRP. This indicated that FRP can improve thermal stability of PVC blends. All plasticized PVC blends presented 3 stages in N₂ with regard to thermal degradation curves. None of the samples illustrated conspicuous weight loss until approaching 150°C. At stage one (from 150°C to 320°C), the weight loss process could be ascribed to the FRP migration together with the PVC dechlorination [34,35]. The temperature range belonging to the second stage was 320°C–450°C, which is mainly understood as the conversion process of conjugated olefins to aromatics. Finally curves between 450°C and 600°C corresponding to the fact that polyenes were cracked into olefin, as the third stage came out [34,35]. FRP had a considerable effect on preventing the weight loss of plasticized PVC blends. There exist numerous polar groups in plasticizer molecules that form hydrogen bonds with α-H of the PVC chain, deferring the PVC dechlorination [36,37].

T_g partly mirrors the plasticizing property of the plasticizers. Fig. 10 shows the DMA curves of plasticized PVC blends containing different masses of FRP. Each curve has only a single peak, corresponding to a homogeneous system. It has been reported that the T_g value of the original PVC is approximately 85.8°C, indicating that adding DOP or FRP to PVC can prominently reduce the T_g value. When DOP was replaced with FRP gradually, T_g increased from 23°C to 47°C, which indicated that plasticized ability of FRP was lower than DOP.

Extraction resistance is an important quality in evaluating the migration property of plasticizers. The excellent extraction resistance can maintain the long-term stability of mechanical, thermal, electrical properties and flame-retardancy of plasticized polymer products, and prevent the potential threat to human health caused by the migration of plasticizers in the use of food and medical devices. Fig. 11 manifests the weight loss of plasticized PVC with DOP and FRP in N-hexane, the advantage of using N-hexane as solvent is that organic solvent can accelerate the migration of plasticizer, and then its migration performance can be evaluated in a short time [5]. With the gradual decrease of DOP and increase of FRP in PVC blends, the weight loss of plasticized PVC blends significantly reduced from 22.6% to 2.8%. FRP showed superior solvent extraction resistance than that of DOP. FRP with a larger relative molecular weight and higher branching degree has better migration resistance than DOP.

The tensile test was conducted at the temperature of 25°C to appraise FRP’s plasticizing effect, with the detailed results being exhibited in Table 3. The affixion of plasticizer usually increases elongation at break and decreases the tensile strength of plasticized PVC materials. Furthermore, when DOP was replaced gradually with FRP, the elongation at break of plasticized PVC blends decreased from 450% to 19%, seeing an opposite trend to the tensile strength, from 12.3 to 28.9 MPa. The results indicated though the ester groups and phosphate groups in the structure of FRP have the lubrication effect on PVC chains, making plasticized PVC blends more flexible [38,39], but FRP showed a lower plasticizing effect than DOP.