Radiation crosslinked blends based on an ethylene octene copolymer (EOC) and polydimethyl siloxane (PDMS) rubber with special reference to the optimization of processing parameters

Abstract

This work focuses on the preparation of blends based on an ethylene octene copolymer (EOC) and polydimethyl siloxane rubber (PDMS), at a particular blend ratio through a melt mixing technique with special reference to the optimization of processing parameters. The optimized blend has been elastically reinforced through a dry curing process (radiation crosslinking) and the physico-mechanical properties were analyzed. It has been found that the rotor speed and the blending temperature of the mixer play very significant roles in controlling the strength properties of the blends, while the time of mixing has less effect on the ultimate properties as compared to the other parameters. It is also found that the viscosity ratio plays an important role in the development of the morphology of the blends. The droplet matrix morphology of the blends, particularly the size of the PDMS domains and the distribution of these domains in the EOC matrix have a remarkable influence on the overall mechanical strength properties of the blends. Through the optimization of the processing parameters, the size of the PDMS rubber domains is effectively decreased from 1.3 μm to 0.6 μm. This reduction in the particle size enhances the tensile strength by about 14.3%. It is found that radiation crosslinking at a radiation dose of 75 kGy improved the tensile strength by about 13% as compared to the optimized blend. All the irradiated blends have higher volume resistivity as compared to the optimized non irradiated blend.

---

1. Introduction

Polymer blends are physical mixtures of two or more different polymers that are not linked by any covalent bonds. The blending technique of polymers is a fascinating subject in the field of polymer processing. Blending of polymers is an easy and economical way of preparing high-performance polymer materials and polymer alloys with properties not typical to any one of the blend constituents. Quite often, the properties of the polymer blends depend upon the individual component properties, processing parameters, morphology developed and the volume proportion of each component in the blend.^1–5 In case of an immiscible blend, the performance property of the blend is predominantly dependent upon the blend morphology, i.e. size and shape of the deformable phases in the constituents.^6–9 Blend morphology is defined as the spatial arrangement of the blend components to acquire either droplet-matrix, staggered or co-continuous matrix structure. Morphology of the immiscible blends developed during melt mixing is linked with the viscosity ratio and the elasticity ratio of the blend constituents. While, the viscosity ratio controls the dispersion of the blend components, the elasticity ratio organizes the shaping of the phases. In developing the blend morphology, processing parameters play a significant role. Thus in order to study the phase morphology of polymer blends, a variety of empirical models have been earlier proposed by many scientists.^10,11 Jordhamo et al. developed an empirical model that explains the morphology development of polymer blends based on the viscosity ratio which may be given by,¹²where φ₁, φ₂ are volume fractions and η₁, η₂ are viscosity values of phase 1 and phase 2 constituents respectively. The eqn (1) results in the following morphology variants.¹³

• Y > 1: phase 1 is continuous or matrix and phase 2 is dispersed.

• Y < 1: phase 2 is continuous or matrix and phase 1 is dispersed.

• Y = 1: dual phase continuity.

The blend morphology that develops is primarily affected by the processing parameters, such as processing temperature, time, rotor speed and pressure. The inter relationship between the parameters mentioned above is complex, and the optimization using conventional statistical method is labor intensive as well as time-consuming. Hence, processing parameters were optimized using the design of experiments by applying Taguchi methodology for superior performance properties of the blend. In the recent past Suresh kumar et al. optimized the processing parameters for LLDPE–PDMS rubber blends using Taguchi methodology.¹⁴ Similarly, Jineesh and Nando used Taguchi methodology for the optimization of processing parameters for thermoplastic polyurethane–PDMS rubber blends, prepared by melt mixing technique.¹⁵

Polyolefin thermoplastic elastomers and their blends are widely used in the automotive and other sectors due to their wide range of functional properties. These materials have excellent mechanical strength, good chemical and environmental resistance, easy processability, low odor and moderate cost.^16,17 On the other hand, polydimethyl siloxane rubber (PDMS) has excellent electrical insulation properties and has been widely used for higher temperature applications. However, it has the disadvantages of high price, poor mechanical properties, poor resistance to wear and tear, oil and solvents, hence is used in specific applications in the form of blends and nanocomposites.^18,19 Santra et al.²⁰ commenced the work on PDMS rubber and polyolefin based thermoplastic elastomers by blending LDPE with PDMS rubber (1 : 1 proportion by weight) followed by compatibilizing it using EMA as a chemical compatibilizer.²¹ Subsequently Jana et al. studied the LDPE–PDMS blend system at all blend ratios and optimized the compatibilizer level throughout the blend ratios.^22–26 Jana et al.²⁷ utilized peroxide to crosslink the LDPE–PDMS system in situ during the molding operation. Recently, Giri et al. explored the LLDPE–PDMS blend system over a composition range varying from 70 : 30 LLDPE : PDMS to 30 : 70 LLDPE : PDMS.²⁸ They used EMA for compatibilizing the LLDPE–PDMS blend system at various blend ratios and crosslinked the LLDPE–PDMS system with electron beam irradiation and optimized the dosage of electron beam irradiation.^29–31

Recently, the authors have reported the chemical crosslinking of the EOC : PDMS rubber blends using peroxide and characterized them.³² However, peroxide crosslinking has numerous disadvantages such as slow curing, use of chemicals and lower physico-mechanical properties and bad odor. On the other hand irradiation crosslinking is a neat process involving no pollution. In addition, this process minimizes the deformation of the material and crosslinks the polymer in a faster, efficient and cleaner way. Various research groups studied the effect of electron beam irradiation on different EOCs and the corresponding blends.^33–35 Mishra et al.³⁶ studied the heat shrinkability and mechanical properties of ethylene octene copolymer after electron beam irradiation. Youm et al.³⁷ crosslinked the polyolefin by electron beam irradiation and studied the elastic properties. Naskar et al.³⁸ prepared thermoplastic vulcanizates of PP and EOC by electron induced reactive processing. Poongavalappil et al.³⁹ studied the effect of electron beam irradiation on the thermal, mechanical and rheological properties of EOC with high co-monomer content.

However no work has been reported on the radiation crosslinking of EOC–PDMS blends. Hence, in the present work emphasis has been laid on the preparation and characterization of EOC : PDMS blends on exposure to electron beam irradiation after optimization of processing parameters. Due to the high cost and low mechanical strength of PDMS rubber the blend ratio was chosen in such a way that the amount of PDMS should be low. Thus for the whole study a blend ratio (70 : 30 of the EOC and PDMS) was chosen viewing from the processing, expected properties and cost point of view. The primary experimental investigation may be divided into two parts. The first part deals with the optimization of the processing conditions and the second part deals with the optimization of radiation dose for the blend.

2. Experimental

2.1 Materials

Engage-8445 (EOC, an ethylene octene copolymer having an ethylene content of 84%) was supplied by DuPont Dow Elastomers (USA), having a specific gravity of 0.91. PDMS grade of silastic WC-50™ having specific gravity of 1.17 was procured from Dow Corning Inc. (Midland, MI, USA) as the other blend constituent. The physical characteristics of the individual components are given in Table 1.

Table 1 Physical characteristics of the materials

Polymer	Trade name	Crystallinity%	Octene content%	Density g cm^−3 ASTM D-792	MFI dg min^−1 ASTM D-1238 190 °C, 2.16 kg	DSC melting point (°C)	Glass transition temperature (Tg) (°C)
EOC	Engage 8445	37	16	0.91	3.5	103	−38
PDMS	Silastic WC-50™	—	—	1.15	—	—	−123

---

2.2 Design of experiments, orthogonal array and experimental design parameters

The Taguchi methodology offers a number of potential advantages over the normal statistical design.^40–42 Design of experiments (DOE) helps to reduce the number of experiments from very large to a limited number, thus saving time, energy and cost. Conventional statistical design parameter determines the optimum conditions on the basis of the characteristic properties. While Taguchi methodology finds the best balance in properties considering the least variability as the optimum conditions. The variability of a property is mainly affected by the noise factor, which is very difficult to control. Signal to noise ratio (S/N ratio) and orthogonal arrays are the primary tools used in the Taguchi design.⁴³ S/N ratios help to find out the robustness, which can further be used to determine the controlling factors that minimize the effect of noise on responses. The S/N ratio characteristics may be divided into mainly three categories. These are;

(1) Smaller is better. It is calculated from the following equation.

(2) Nominal is better. It is calculated from the following equation.

(3) Larger is better. It is calculated from the following equation.

where M is the average observed data, r is the number of observations and Y is the observed data. Taguchi methodology is a partial factorial design which means that it is a sample of the full experimental design. Thus, analysis by Taguchi methodology shall include investigations of confidence that can be induced in the results. To measure the confidence level analysis of variance (ANOVA) is used. It measures the variance of the data and the confidence from the variance. Mixing time, the temperature of blending and rotor speed was considered as the processing variable parameters. The effect of these three parameters at different levels was evaluated using L₉ orthogonal array based on Taguchi methodology. In a conventional full factorial experimental design, it would require 3³ = 27 experiments to study the influence of these three factors at three levels, whereas Taguchi's factorial design experiment reduces it to only 9 experiments, offering a great advantage in terms of number of experiments, time and cost. The analysis of the experimental data is carried out using the software MINITAB 15, specially meant for design of experiments (DOE). The experimental observations were transformed into plots of S/N ratios against different levels. The S/N ratios for tensile strength and elongation at break are expressed as larger the better. The factors and their levels for the melt mixing process are depicted in Table 2.

Table 2 Factors and their levels in the melt blending technique

Factors	Levels
	Level 1	Level 2	Level 3
Temperature (°C)	120	130	140
Rotor speed (rpm)	60	80	100
Time of blending (min)	4	6	8

---

2.3 Preparation of the blends

Blending of EOC and PDMS rubber in the proportion of 70 : 30 was carried out in a Brabender plasticorder by varying the processing parameters as per the L₉ table of Taguchi (Table 3). The addition of EOC into the Brabender plastograph is maintained at a lower rotor speed of 30 rpm. Subsequently, the mixing chamber was closed, and the rotor speed was increased to the value as given in Table 3. The melting of EOC was done for 2 minutes, and then the rotor speed was reduced to 30 rpm for the addition of PDMS rubber. Then, the mixing chamber was closed rapidly, and the rotor speed was increased as per Table 3. The total blending time was maintained as per Table 3. At the end of the mixing cycle, the molten mass was taken out of the chamber and sheeted out on a cold two roll mill. Then the molten mass was then compression molded at respective processing temperatures of melt mixing as per Table 3 in a compression molding hydraulic press at a pressure of 5 MPa for 3 minutes. For example, the blend EP1 was compression molded at 120 °C and EP5 at 130 °C at a pressure of 5 MPa for about 3 minutes.

Table 3 L₉ orthogonal array as per Taguchi methodology

Sample no.	Temperature (°C)	Rotor speed (rpm)	Time (minutes)
EP 1	120	60	4
EP 2	120	80	6
EP 3	120	100	8
EP 4	130	60	6
EP 5	130	80	8
EP 6	130	100	4
EP 7	140	60	8
EP 8	140	80	4
EP 9	140	100	6

---

2.4 Irradiation of samples

After the optimization of processing parameters, the blend which showed the best physico-mechanical properties was radiation crosslinked by varying the radiation dose from 25 to 150 kGy. The optimized blend was denoted as EP73 and EP73R100 denoted the radiation crosslinked blend exposed to a radiation dose of 100 kGy. The molded sheets were subjected to electron beam irradiation in air at room temperature using an electron beam accelerator (Model: Dynamitron, 369 supplied by RDI, USA) at Nicco Corporation Limited, Kolkata, India. An irradiation dose of 25, 50, 75, 100, 125 and 150 kGy was used for this study. The acceleration energy and beam current were 2.5 MeV and 17 mA respectively. The electron beam irradiated samples are designated as EP7R100 means 70 : 30 EOC : PDMS blend radiated at 100 kGy radiation dose.

2.5 Characterization of the blends

2.5.1 Gel content determination. The gel content of the radiation crosslinked samples was determined by extracting the sol component in boiling xylene for 24 hours using a soxhlet apparatus. Three samples were used in each case and the average values are reported.where W₁ and W₀ are the weight of the dried sample after extraction and the weight of the sample before extraction respectively. The extracted samples were dried in oven at 60 °C until they attained a constant weight.

2.5.2 Mechanical property evaluation. Dumbbell shaped specimens were punched out from the tensile sheets using a hollow dumbbell cutting die (ASTM die C) and the tensile properties were measured as per ASTM D 412 using a Hounsfield H10KS universal testing machine at a crosshead speed of 500 mm min⁻¹ under ambient conditions.

2.5.3 Melt rheology study. Rheological measurements of the individual components, i.e. EOC and PDMS at three different temperatures were found by an Anton Paar modular compact rheometer (MCR 102) at a shear rate varying from 10–100 s⁻¹. Since, the melt mixing is done at three different temperatures, i.e. 120 °C, 130 °C and 140 °C, the melt rheology study of the individual components were done at theses temperatures in the Anton Paar modular compact rheometer instrument.

2.5.4 Scanning electron microscopy studies. A JEOL-JSM 5800 scanning electron microscope (SEM) was used to study the morphology of the EOC–PDMS rubber blends before and after radiation crosslinking. The test samples were prepared by cryogenically fracturing the polymer strips under liquid nitrogen and then etching it in solvent xylene to remove the PDMS rubber phase. Etching in the solvent was done for 72 hours at room temperature. Before SEM analysis the samples were dried in a hot air oven at 70 °C for 24 hours and then sputter coated with a thin layer of gold in a vacuum chamber and subjected to SEM analysis at zero tilt angle.

2.5.5 FTIR spectroscopy. FTIR spectra on the thin films of the EOC : PDMS rubber blends before and after electron beam irradiation were recorded using a Perkin-Elmer Frontier spectrometer (UK) in ATR mode at room temperature over the frequency range of 4000–650 cm⁻¹.

2.5.6 X-ray diffraction (XRD) analysis. Changes in the crystallinity of EOC in EOC : PDMS blend before and after crosslinking were recorded by an X-ray diffractometer using monochromatic Cu Kα radiation of wavelength 1.5418 Å, in the 2θ range of 10–40 at a scanning rate of 5° min⁻¹ and at an operating voltage of 40 kV with a beam current of 30 mA.

The areas under the crystalline and amorphous regions were deduced in arbitrary units by X'pert highscore plus 2.1 and Origin Pro 8 software and the percentage crystallinity was calculated using eqn (6).

where, I_a and I_c correspond to the integrated intensities corresponds to the amorphous and crystalline phase respectively.

The crystallite size P (Scherrer equation) and the lattice size were calculated as follows⁴⁴

where β is the half height width of the crystalline peak, K is the shape factor for the average crystallite (∼0.9) and λ is the wavelength of the X-ray radiation.

3. Results and discussions

This section is divided into two parts. The first part discusses about the optimization of the processing parameters of EOC : PDMS blends by Taguchi methodology considering the tensile strength and elongation at break as the parameters. After optimization of the blend, the second part deals with the effect of electron beam irradiation on the physico-mechanical properties of the optimized blend (EP73).

3.1 Optimization of processing parameters of EOC : PDMS blend

3.1.1 Mechanical strength properties of blends. The modulus at 100%, 200%, and 300% elongation, elongation at break and tensile strength, of the blends are presented in Table 4. From the mechanical properties, it is observed that the blend EP5 and EP8 exhibit the best balance in physico-mechanical properties, hence may be chosen as the optimum blends for further study.

Table 4 Mechanical properties of blends

	Tensile strength (MPa)	Elongation (%)	Modulus, 100% (MPa)	Modulus, 200% (MPa)	Modulus, 300% (MPa)
EP 1	14.0 (0.2)	746 (22.1)	5.1 (0.1)	5.3 (0.1)	6.0 (0.1)
EP 2	14.6 (0.1)	746 (35.4)	5.2 (0.1)	5.4 (0.1)	6.1 (0.2)
EP 3	14.6 (0.2)	755 (23.9)	5.1 (0.2)	5.3 (0.1)	6.0 (0.1)
EP 4	14.0 (0.2)	717 (22.1)	5.2 (0.1)	5.5 (0.1)	6.2 (0.2)
EP 5	16.0 (0.2)	788 (28.7)	5.3 (0.1)	5.7 (0.1)	6.3 (0.1)
EP 6	14.6 (0.2)	714 (36.4)	4.9 (0.2)	5.2 (0.2)	5.9 (0.2)
EP 7	14.9 (0.2)	759 (40.3)	5.2 (0.1)	5.4 (0.2)	6.2 (0.2)
EP 8	15.8 (0.2)	780 (26.7)	5.3 (0.1)	5.5 (0.2)	6.3 (0.2)
EP 9	15.3 (0.2)	767 (39.3)	5.3 (0.1)	5.5 (0.1)	6.2 (0.2)

---

3.1.2 Scanning electron microscopy. SEM photomicrographs of xylene etched EOC–PDMS blends are shown in Fig. 1(a)–(i). Fig. 1(a)–(c) shows the photomicrographs of the blends processed at 120 °C. These photomicrographs illustrate that the blends those are processed at a lower temperature show a co-continuous type of phase morphology with larger domains, which is reflected in the lower mechanical properties of the blends (Table 4). The domain size of the dispersed PDMS rubber in Fig. 1(b) is relatively larger (1.2 μm) as compared to Fig. 1(e) and (h) (0.6 μm). The average domain size of the PDMS phase has been calculated by ImageJ software from the measurements of about 100 domains and taking an average of the sizes, which is presented in Table 5. The uneven distribution of the PDMS phase and the matrix co-continuity may be the primary factors for lower mechanical properties of these blends as depicted in Table 4.

Fig. 1 Scanning electron micrographs of xylene etched EOC–PDMS blends at a magnification of ×2000 and at a scale of 10 μm.

Table 5 Morphology and domain size distribution of different blends

	Morphology	Average domain size (μm)
EP 1	Co-continuous	—
EP 2	Elongated droplet matrix	1.24
EP 3	Co-continuous	—
EP 4	Elongated droplet matrix	1.30
EP 5	Droplet matrix	0.61
EP 6	Elongated droplet matrix	1.06
EP 7	Co-continuous	—
EP 8	Droplet matrix	0.68
EP 9	Elongated droplet matrix	0.81

---

Fig. 1(e) and (h) shows the photomicrographs of the blends processed at a higher temperature at 130 °C and 140 °C respectively. These micrographs shows a droplet matrix morphology, where the PDMS rubber domains are dispersed uniformly throughout the matrix. Also, these micrographs exhibit the lowest domain size of about 0.6 μm. This may be the reason for the higher mechanical strength properties of the blends processed at higher temperatures (Table 4).

3.1.3 Taguchi analysis of the EOC–PDMS blends. Signal to noise ratio (S/N) plays a significant role in optimizing the processing parameters. The phenomenon of larger is better and smaller is better are taken into consideration depending upon the nature of the parameters. For example, a higher value of S/N ratio implies that the signal is much higher than the random effect of the noise factors. The experimental design consistent with the highest S/N ratio always yields the best quality with minimum variance. In the Taguchi method, the term ‘signal’ represents the desirable output, and the term ‘noise’ represents the undesirable output. S/N ratios of processing parameters based on the analysis of tensile strength are reported in Table 6. The category of higher is better S/N ratios was used to select optimum processing parameters. Fig. 2 shows the main effect of S/N ratio on the tensile strength of the blends.

Table 6 Values of S/N ratios for tensile and impact strength properties of the blends

Performance property	Levels	Temperature	Rotor speed	Mixing time
Tensile strength	1	23.16	23.12	23.41
	2	23.42	23.76	23.29
	3	23.72	23.41	23.60
	Delta	0.56	0.64	0.31
	Rank	2	1	3
Elongation at break	1	57.49	57.39	57.46
	2	57.37	57.74	57.42
	3	57.71	57.44	57.70
	Delta	0.34	0.35	0.28
	Rank	2	1	3

---

Fig. 2 Main effect plot for signal-to-noise (S/N) ratios for the tensile strength (a) and elongation at break (b).

The S/N ratio decrease from 23.72 to 23.16 as the mixing temperature is increased from 120 °C to 140 °C. Similarly, in the case of rotor speed the S/N ratio increases from 23.12 to 23.76 with the temperature rise from 120 °C to 140 °C. So the rank for the rotor speed in the S/N ratio analysis is 1. And the rank for the temperature in the S/N ratio analysis is 2. Therefore, the optimum values for the temperature and rotor speed to have higher tensile strength are taken to be 140 °C and 80 rpm respectively. This clearly indicates that the temperature and the rotor speed are the primary parameters that play significant roles in the tensile strength of the blends of EOC and PDMS rubber. From this data, it is evident that significance of blending time on the tensile strength properties of EOC–PDMS rubber blends is lower as compared to the other processing parameters. Thus the blending temperature of 140 °C has been considered to be the optimum processing temperature, a rotor speed of 80 rpm is considered as the optimum rotor speed of the chamber and a blending time of 8 minutes has been selected as the optimum mixing time, because almost all properties reaches its highest value at this optimized processing parameters of blending.

Fig. 2(b) represents the effective plot for S/N ratios, considering the elongation at break as the performance property. From the Fig. 2(b), one can observe that both rotor speed and chamber temperature play significant roles in the elongation at break of blends. Here the S/N ratio increases from 57.39 to 57.74 as the rotor speed is increased from 60 rpm to 80 rpm. In the case of temperature, the S/N ratio is increased from 57.37 to 57.71 when the temperature is increased from 130 °C to 140 °C. From these plots of tensile strength and elongation at break, one can easily conclude that the rotor speed and temperature play significant roles in the elongation at break. Hence, the rank of the rotor speed and temperature are one and two, respectively which is same as in the case of tensile strength (Fig. 2(a)). The interaction parameter analysis and the variance study of EOC : PDMS blends can be found in the ESI† file.

3.1.4 Confirmation of test results. From the S/N ratio analysis, one can conclude that the optimum conditions for best tensile strength are 140 °C processing temperature; 80 rpm rotor speed and 8 minutes mixing time respectively. EOC–PDMS (70 : 30) blend has been prepared at the optimum processing conditions to confirm the statistical data and the tensile strength is measured. The experimental and predicted mean values for tensile strength exactly match with each other as given in Fig. 3. This confirms the experimental values of tensile strength very much coincides with the predicted value calculated from the S/N ratio analysis.

Fig. 3 Predicted and experimental values of tensile strength of the EOC–PDMS blend prepared at optimum processing conditions.

3.1.5 Rheological characterization of the blends. During melt blending in a Brabender plastograph, the actual shear rate at each position of the Brabender may be calculated according to the following equation,⁴⁵

γ = 2πRN/60h (9)

where γ is the shear rate developed inside the mixing chamber, R is the maximum radius of the rotor blade, N is the rotor speed during mixing, and h is the gap between the chamber wall and the rotor tip. Here, a photograph of the Brabender rotor was taken, and the parameters of the eqn (10) were calculated by ImageJ software. Since, the melt mixing is done at three different temperatures, i.e. 120 °C, 130 °C and 140 °C, the shear viscosity of the individual components, i.e. EOC and PDMS have been found at theses temperatures from the parallel plate rheometer at a shear rate varying from 10–100 s⁻¹. The average shear rate that is generated inside the Brabender chamber according to eqn (10) is calculated and presented in Table 7. The maximum shear rate was calculated with the parameters R = 18.6 mm, h = 1.25 mm and minimum shear rate, using R = 10.11 mm, h = 8.77 mm respectively. The shear viscosity of EOC and PDMS at this particular shear rate is found out from the shear viscosity vs. shear rate plot obtained from the parallel plate experiment depicted in Fig. 4(a) and (b).

Table 7 Maximum, minimum and average shear rate experienced inside the Brabender plastograph at various rotor speed

Rotor speed (rpm)	γmax	γmin	γave
60	93.5	13.1	53.4
80	124.1	17.8	71.3
100	155.9	22.2	81.9

---

Fig. 4 log viscosity vs. log shear rate plot of (a) EOC at different temperatures, (b) PDMS at different temperatures.

After finding out the individual viscosities of EOC and PDMS rubber, the viscosity ratio of various blends may be calculated using the following equation,

where ρ is the viscosity ratio of the blend.

From the viscosity ratio, the value of Y is calculated using eqn (1). From the value of Y, the theoretical phase morphology is calculated. These theoretical and experimental phase evolutions of the blends are summarized in Table 8. It is observed that the theoretical and experimental phase morphologies calculated from eqn (1) and obtained from SEM analysis respectively, are almost in tune with each other. In this blend system, the elastomeric phase is dispersed in the thermoplastic phase, which is evident from the scanning electron microscope (Fig. 1). Theoretically the morphology of the immiscible blends of an elastomer and thermoplastic after melt mixing can be expressed by plotting viscosities of the blends against the composition ratio.⁴⁶ Morphology of the blends change with the blend ratio as well as the viscosity ratio. The viscosity ratio calculated for EP1 and EP5 blends are presented in Table 8. At lower temperature (120 °C), the shear viscosity of EOC is higher than that of PDMS at a particular shear rate, i.e., 60 rpm which results in a lower viscosity ratio according to eqn (10). The higher and lower viscosities of EOC and PDMS rubber respectively affects the mixing process during blending, which results in lowering mechanical properties because of immiscibility. From Fig. 1(a) it is evident that, for a 70 : 30 (EOC : PDMS) blend, the viscosity ratio approaches the value of 0.33 and the morphology of the blend system tends to be co-continuous. This has been already confirmed by the scanning electron microscopy studies of the EP1 blend (Fig. 1(a)). As the mixing temperature increases, the viscosities of both EOC and PDMS decrease. But the viscosity decrease of the EOC is much faster as compared to that of the PDMS rubber; consequently the viscosity ratio rises to 0.97 for EP5 at 130 °C. This reduction in viscosity of EOC enables the mixing of EOC with PDMS easier enhancing the viscosity ratio. This is responsible for the higher mechanical properties of the blends. The blends (EP5 and EP8) which exhibit a viscosity ratio value close to unity develop a droplet matrix morphology that, on the other hand, imparts higher physico-mechanical properties.

Table 8 Viscosity ratio value of various blends

	Temperature, °C	Rotor speed, rpm	Viscosity of PDMS, Pa s	Viscosity of EOC, Pa s	Viscosity ratio	Value of Y (Jourdharma et al.) (eqn (1))	Predicted morphology obtained from model	Actual morphology obtained from SEM
EP1	120	60	2111	6740	0.33	0.76	PDMS is continuous	Co-continuous
EP5	130	80	1710	1760	0.97	2.27	EOC is continuous	EOC is continuous
EP7	140	60	1418	3360	0.42	0.98	Co-continuous	Co-continuous
EP8	140	80	672	689	0.98	2.28	EOC is continuous	EOC is continuous

---

3.2 Radiation crosslinking of the optimized blend (EP73)

The optimized EOC : PDMS blend was further radiation crosslinked by exposing to varying doses of radiation from 25 to 150 kGy. Physico-mechanical properties and phase morphology of the radiation crosslinked blends were carried out and presented in the subsequent sections.

3.2.1 Gel content. The gel content of the neat EOC, neat PDMS and the optimized blend of EOC with PDMS (EP73) after irradiation were calculated by the method described in the Experimental section. Fig. 5 depicts the gel content (%) of irradiated blend and the individual components as a function of irradiation dosage. The gel content steadily increases for the EOC : PDMS blend, neat EOC and neat PDMS rubber with increase in the irradiation dosage. This is expected to be due to the formation of C–C crosslinks in the EOC and PDMS phases on exposure to electron beam irradiation. The highly reactive radicals and ions produced by the electron beam irradiation modify the molecular structure of the polymeric material and forms insoluble chemical crosslinks between the molecular chains. In addition to the chemical crosslinking of the polymeric chain, electron beam irradiation causes some degradation or chain scission of the polymeric chains. These processes occur simultaneously when the irradiation takes place and the one that dominates the other decides the final physico-mechanical properties of the blend. Domination of one particular process over the other is determined by the chemical structure of the polymer and the dosage of electron beam irradiation.

Fig. 5 Gel content of neat EOC, PDMS and 70 : 30 EOC : PDMS blend after irradiation.

From the Fig. 5, it is substantiated that the gel content of neat PDMS is higher at all radiation doses as compared to EOC. It is because of the fact that as compared to the polyolefin (EOC), PDMS rubber can easily form free radicals at vinylene and methyl groups at lower radiation doses and facilitates crosslink formation.³¹ As the radiation dose increases from 25 to 150 kGy, the gel content increases from 56.5 and 87.0 to 85.6 and 92.8 for EOC and PDMS respectively. At lower irradiation doses, there is a substantial difference in the gel content of EOC and PDMS rubber. This indicates that at lower radiation doses, the EOC is not getting crosslinked to a higher extent as compared to that of PDMS rubber. But as the amount of radiation dose increases, the difference in the gel content of EOC and PDMS becomes less. This is due to the fact that, as the extent of radiation increases more and more EOC gets crosslinked which results in a higher gel content. For the blends, the gel content sharply increases as the radiation dose increases from 25 kGy to 150 kGy. As the radiation dose increases from 25 to 150 kGy, the gel content of 70 : 30 EOC : PDMS blend increases from 79.9 to 89.9.

3.2.2 Physico-mechanical properties. To find out the effect of electron beam irradiation (EBR) on individual components, the neat PDMS rubber and neat EOC were irradiated by varying the radiation doses and the physico-mechanical properties were evaluated and presented in Table 9. From Table 9 it is clear that the effect of EBR on the physico-mechanical properties of PDMS rubber is more prominent as compared to EOC. Since virgin PDMS rubber has a very low modulus and lower green strength it was not possible to measure the strength properties. However, the tensile strength of PDMS rubber increases sharply up to 75 kGy of radiation dose, beyond which it decreases. Elongation at break decreases drastically, and it reaches 176% at a dosage of 150 kGy. This has been attributed to the higher degree of crosslinking resulting in tighter networks, which is responsible for the drastic reduction in the elongation at the break as radiation dose increases. On the other hand, neat EOC exhibits a tensile strength of 26.4 MPa as shown in Table 9. With the increase in radiation dose from 0 kGy to 25 kGy, the tensile strength marginally increases from 26.3 to 26.7 MPa, beyond which the tensile strength of EOC significantly decreases, and reaches a value of 13.3 MPa at 150 kGy (Fig. 6). This may be explained as due to the decrease in the overall crystallinity drastically which will be discussed in the subsequent section.

Table 9 Mechanical properties of irradiated EOC and PDMS rubber

Sample	Radiation dose (kGy)	Tensile strength (MPa)	Elongation at break (%)	Modulus (100%) (MPa)	Modulus (200%) (MPa)	Modulus (300%) (MPa)
PDMS	0	—	—	—	—	—
	25	7.9 (0.4)	1325 (35.7)	0.01 (0.04)	0.5 (0.1)	0.9 (0.1)
	50	9.1 (0.4)	750 (29.6)	0.02 (0.08)	1.0 (0.2)	2.0 (0.1)
	75	9.5 (0.4)	540 (33.3)	0.03 (0.09)	1.4 (0.2)	2.8 (0.1)
	100	7.9 (0.3)	347 (28.9)	0.03 (0.06)	2.1 (0.1)	4.4 (0.1)
	125	7.0 (0.4)	208 (26.7)	0.04 (0.06)	2.3 (0.1)	—
	150	6.0 (0.5)	176 (39.5)	1.6 (0.1)	6.4 (0.2)	—
EOC	0	26.4 (0.4)	1115 (36.1)	3.4 (0.2)	4.2 (0.1)	4.8 (0.1)
	25	26.7 (0.5)	988 (29.7)	4.1 (0.1)	4.4 (0.1)	5.4 (0.2)
	50	22.7 (0.4)	770 (36.7)	4.5 (0.1)	4.8 (0.2)	5.8 (0.2)
	75	21.1 (0.4)	865 (26.5)	5.3 (0.2)	5.3 (0.2)	6.5 (0.1)
	100	19.4 (0.4)	721 (33.7)	5.7 (0.1)	6.0 (0.1)	6.9 (0.2)
	125	17.8 (0.4)	701 (36.5)	6.3 (0.2)	6.6 (0.2)	7.8 (0.1)
	150	13.3 (0.5)	650 (22.9)	6.4 (0.1)	6.7 (0.1)	8.0 (0.1)

---

Fig. 6 Tensile strength of neat EOC, and 70 : 30 EOC : PDMS blend after irradiation.

Physico-mechanical properties such as tensile strength, elongations at break, modulus at different percentage of strain of 70 : 30 EOC : PDMS blend at varying radiation doses are presented in Table 10. The tensile strength is increased from 16.0 MPa to 18.1 MPa upon exposure to a radiation dose of 75 kGy. That means there is a 13.2% increase in the tensile strength on exposure of 75 kGy radiation. This increase in the tensile strength is explained as due to the formation of effective chemical crosslinks of intramolecular, intermolecular and interfacial type in the PDMS rubber phase as well as in the amorphous regions of the EOC phase. Whereas beyond 75 kGy, there is a significant drop in the tensile strength; from 18.1 MPa to 11.1 MPa with an increase in radiation dose from 75 kGy to 150 kGy. From these results it may be concluded that, the optimum dose of radiation for 70 : 30 EOC : PDMS blends shall be 75 kGy in order to get higher physico-mechanical properties.

Table 10 Mechanical properties of irradiated 70 : 30 EOC : PDMS blend

Sample	Radiation dose (kGy)	Tensile strength (MPa)	Elongation at break (%)	Modulus (100%) (MPa)	Modulus (200%) (MPa)	Modulus (300%) (MPa)
EP73	0	16.0 (0.3)	1093	3.5	3.8	4.5
	25	16.5 (0.4)	968	3.6	3.8	4.9
	50	17.4 (0.4)	928	3.8	4.5	5.4
	75	18.1 (0.4)	845	4.4	5.1	5.8
	100	16.0 (0.3)	795	4.5	5.3	6.1
	125	14.1 (0.4)	692	4.6	5.7	7.1
	150	11.1 (0.5)	533	4.6	5.7	7.3

---

When a polymer is treated with an ionizing radiation, two processes come into action simultaneously. One is the formation of chemical crosslinks between the molecular chains, and the other one is the chain scission that destroys the molecular structure. Although both processes occur simultaneously upon irradiation, at lower radiation doses, the crosslinking predominantly occurs and at higher radiation doses chain scission become more pronounced.³¹ Hence for the EOC : PDMS blends beyond 75 kGy radiation doses, chain scission is initiated resulting in lower physico-mechanical properties. It has been reported earlier by Giri et al. that, at higher radiation doses the main backbone chain of the polymer (here it s polyolefin) may break into many small fragments,³¹ resulting in the formation of small clusters of crosslinked network. Such fragments consist of a larger number of polymer molecules crosslinked tightly; however, they may act as separate physical entities. The intercluster force may not be sufficient to hold them under externally applied separating forces during the tensile test.³¹ This result in lower physico-mechanical properties (Fig. 6).

From the Table 10, it is observed that the elongation at break continuously decreases upon irradiation, and reaches a value of 795% at 100 kGy, i.e. a decrease in elongation at break by 27%. After 100 kGy, the irradiation causes a rapid reduction in the elongation at break and it reaches 533% at 150 kGy (51% decrease in elongation at break). Such a reduction in EB can be attributed to the formation of three-dimensional network structure upon irradiation, which prevents the structural reorganization during elongation. The increase in crosslink density restricts the internal chain mobility of the polymer when the external force is applied, and hence the irradiated blends show lower EB.

Also, it is noted that upon irradiation the modulus at different percentage of strain gradually increases. As the radiation dose changes from 0 to 150 kGy the 100% and 200% modulus increases from 3.6 MPa to 4.6 MPa and 3.8 MPa to 5.7 MPa respectively. As the radiation dose increases from 0–150 kGy, the 300% modulus of 70 : 30 EOC : PDMS blends increases from 4.5 MPa to 7.3 MPa (Table 10).

3.2.3 Scanning electron microscopy. To understand the phase morphology of the blends developed, SEM photomicrographs of xylene etched EOC–PDMS blends before and after irradiation at various doses of radiation are carried out and presented in Fig. 7(a)–(d) and 8(a) and (b). Fig. 7(a) represents the SEM photomicrographs of 70 : 30 EOC : PDMS blend showing droplet matrix morphology, the domains being spherical and elongated in shape of PDMS rubber dispersed uniformly throughout the EOC matrix. The average size of the PDMS rubber domains is about 0.6 μm as per Fig. 7(a). Fig. 7(b)–(d) represents the 70 : 30 EOC : PDMS blends irradiated at doses of 50 kGy, 100 kGy and 150 kGy electron beam irradiation. As one can perceive clearly from these SEM photomicrographs that, it is hard to remove the PDMS rubber phase from the blend after irradiation due to the crosslinking of the PDMS rubber phase; both intermolecular and intramolecular crosslinks. In addition to that there are interfacial crosslinks between the PDMS and EOC which does not allow the PDMS rubber to get removed from the matrix during etching. Although a few holes are visible in the magnified image of the photomicrograph Fig. 8(a) and (b) at lower doses of radiation (black and white arrows indicates the holes after etching in the solvent), after 150 kGy of irradiation it is not possible to remove the PDMS rubber domains from the blend as it has been fully crosslinked (Fig. 8(b)).

Fig. 7 Scanning electron micrographs of (a) 70 : 30 EOC : PDMS, (b) irradiated 70 : 30 EOC : PDMS (50 kGy), (c) irradiated 70 : 30 EOC : PDMS (100 kGy), (d) irradiated 70 : 30 EOC : PDMS (150 kGy), solvent (xylene) etched.

Fig. 8 Scanning electron micrographs of irradiated 70 : 30 EOC : PDMS (50 kGy) at two different magnifications, solvent (xylene) etched.

3.2.4 FTIR analysis. FTIR study of the EOC : PDMS blends before and after electron beam irradiation has been carried out to understand the interaction between EOC and PDMS rubber and the spectra are presented in Fig. 9. Fig. 9(a) and (b) corresponds to the FTIR spectra of 70 : 30 EOC : PDMS blends before and after electron beam irradiation in the range from 4000 to 650 cm⁻¹. It is observed that in the non-irradiated 70 : 30 EOC : PDMS blend there is a broad, intense doublet peak at 1095 cm⁻¹ and 1024 cm⁻¹ corresponding to Si–O–Si and Si–O–C stretching vibrations respectively.²¹ The peak at around 1300 cm⁻¹ corresponds to the CH₃–Si stretching vibration of PDMS rubber. Further, the multiple bands between 860 cm⁻¹ and 790 cm⁻¹ are due to the methyl rocking vibrations and Si–C stretching vibrations. The characteristic peak observed at around 1465 cm⁻¹ is due to the C–H stretching vibrations of the methylene units of the ethylene octene copolymer (EOC).⁴⁷ Further, there are two strong peaks at 2910 and 2847 cm⁻¹ attributed to symmetric and asymmetric C–H stretching vibrations of –CH₂ groups in EOC and PDMS rubber. It is noticed from Fig. 10(b) that, the intensity of the Si–O–Si and Si–O–C stretching vibrations a doublet at around 1095 cm⁻¹ and 1024 cm⁻¹ and the intensity of CH₃–Si stretching vibration peak (at 1300 cm⁻¹) gradually reduces with increasing radiation dose. This indicates the presence of intramolecular and intermolecular crosslinks in the PDMS rubber.⁴⁸ Further, the intensity of the peaks at 2910 and 2847 cm⁻¹ found to decrease as the radiation dose increases from 0–150 kGy. This may be due to hydrogen abstraction and resulting crosslink formation at higher radiation doses.

Fig. 9 FTIR spectra for 70 : 30 EOC : PDMS blends before and after irradiation in the range of (a) 4000 to 2400 cm⁻¹, (b) 2400 to 650 cm⁻¹.

Fig. 10 X-ray diffraction patterns of (a): virgin EOC, (b): 70 : 30 EOC : PDMS blend before and after radiation crosslinking showing the angular range (2θ) between 15° and 30°.

3.2.5 X-ray diffraction study (XRD). In order to find out the effect of radiation crosslinking on the crystalline structure and overall crystallinity of virgin EOC and EOC : PDMS blend, X-ray diffraction analysis was carried out for the irradiated samples and the various parameters are listed in Table 11. The sharp peaks observed for neat EOC at 21.4 2θ and 23.7 2θ correspond to the (1,1,0) and (2,0,0) planes of orthorhombic crystals (JCPDS number: 00-060-0984). A sharp drop in crystallite sizes (Fig. 10(a)) was observed for the blend (EP73) as compared to the virgin EOC which may be attributed to the higher lattice strain developed because of the introduction of 30 parts of heterogeneity (PDMS) to the virgin EOC (Table 11). Thus the addition of 30 weight% of PDMS to virgin EOC causes the reduction in the crystallite size from 5.1 to 3.8 and degree of crystallinity from 35.1 to 19.8.

Table 11 Structural characteristics of EOC, 70 : 30 EOC : PDMS blends before and after radiation crosslinking

Sample	Crystallite size P1 (1,1,0) Å	Crystallite size P2 (2,0,0) Å	Crystallite size anisotropy (P1/P2)	Lattice strain at (1,1,0) (%)	Degree of crystallinity (%)
EOC	5.14	5.94	0.86	0.03	35.1
EP73	4.15	3.29	1.26	0.45	19.8
EOCR50	3.56	3.01	1.18	0.52	19.2
EOCR100	3.11	2.50	1.24	0.60	15.5
EOCR150	1.25	0.76	1.64	1.26	11.3
EP73R50	3.78	3.24	1.16	0.44	17.3
EP73R100	2.08	1.35	1.54	0.89	15.6
EP73R150	1.78	0.86	2.06	1.44	13.9

---

Similarly, reduction in the peak intensity and overall crystallinity on irradiation of the EOC : PDMS blends occur. It is also observed that, for the virgin EOC, increase in the radiation dose reduces the intensity of the peak reducing the overall crystallinity. As the magnitude of radiation dose increases from 0 to 100 kGy, the overall crystallinity is reduced from 35.1 to 15.5. Further, the overall crystallinity reduces to 11.3 as the EOC is irradiated at a radiation dose of 150 kGy.

For the optimized 70 : 30 EOC : PDMS blend (EP73), crystallite size and degree of crystallinity further reduces as the blend is irradiated with higher radiation doses. This may be due to higher crosslinking of EOC phase upon irradiation. In this case, the degree of crystallinity and crystallite size are reduced from 19.8 to 15.6 and from 4.15 to 2.08 respectively on irradiation of 100 kGy dosages to the 70 : 30 EOC : PDMS blend. After 100 kGy of radiation, the crystallinity and crystallite size are reduced from 15.6 to 13.9 and from 2.08 to 1.78 respectively as the blend is irradiated at 150 kGy doses of radiation.

3.2.6 Volume resistivity analysis. Fig. 11 illustrates the effect of radiation dose on the volume resistivity of the 70 : 30 EOC : PDMS blend. It has been found that for the blend volume resistivity increases with electron beam dosage up to 100 kGy. Beyond this the volume resistivity decreases sharply. However, all the irradiated blends have higher electrical resistance as compared to the virgin blend (EP73). When the radiation dose increases from 0 to 100 kGy the volume resistivity increases from 2.1 × 10¹⁵ ohm cm to 15.2 × 10¹⁵ ohm cm for EP73 blend. This increase in the volume resistivity may be associated with the formation of crosslinked structure on exposure to electron beam radiation. The electron beam irradiation caused numerous crosslinking points in the EP73 blend that may to act as barriers to prevent the electrical charge movement between the polymer chains. The increase in number of traps in a material due to pronounced crosslinking may cause severe restriction in the charge mobility and consequently improving the volume resistivity of the irradiated samples.^48–51 However, very high radiation dose not only leads to crosslinking but also causes chain scission to a certain extent. Such molecular chain fragmentation forming polar and ionic segments on irradiation may result in the reduction of electrical resistivity beyond 100 kGy.^52,53

Fig. 11 Volume resistivity of the 70 : 30 EOC : PDMS blend after radiation crosslinking.

4. Conclusions

Blends of EOC and PDMS rubber are prepared by melt mixing technique and the processing parameters have been optimized by applying Taguchi methodology. This optimized blend is elastically reinforced through radiation crosslinking by varying the dose from 25 to 150 kGy, and the physico-mechanical properties were analyzed. It is found that the rotor speed and the blending temperature of the mixer play significant roles in controlling the strength properties of the blends. While, the time of mixing has less effect on the ultimate properties of the 70 : 30 EOC : PDMS blends as compared to the other parameters. The optimum processing conditions such as temperature, rotor speed and blending time in the mixer were found to be 140 °C, 80 rpm and 8 minutes respectively. It is also found that the viscosity ratio plays a significant role in the morphology development of the blends. The blends (EP5 and EP8) which have a viscosity ratio close to unity develop a droplet matrix morphology that results in higher physico-mechanical properties. Through the optimization of the processing parameters, the size of the PDMS rubber domains are effectively decreased from 1.3 μm to 0.6 μm thus enhancing the tensile strength by about 14.3%. It is found out that radiation crosslinking improves the tensile strength by about 13% as compared to the optimized blend at a radiation dose of 75 kGy. All the irradiated blends have higher volume resistivity as compared to the optimized blend.

References

1. A. Garton, Polym. Eng. Sci., 1982, 22, 124–129.
2. S. Newman and D. R. Paul, Rubber modification of plastics in polymer blends, Academic press, New York, 1978, vol. 2, ch. 7, pp. 330–336.
3. M. Kawasumi, N. Hasegawa, M. Kato, A. Usuki and A. Okada, Macromolecules, 1997, 30, 6333–6338.
4. Y. Wang, Q. Zhang and Q. Fu, Macromol. Rapid Commun., 2003, 24, 231–235.
5. L. A. Utracki and Z. H. Shi, Polym. Eng. Sci., 1992, 32, 1824–1833.
6. S. Wu, Polymer, 1985, 26, 1855–1863.
7. J. R. Stell, D. R. Paul and J. W. Barlow, Polym. Eng. Sci., 1976, 16, 496–506.
8. P. M. Subramanian, Polym. Eng. Sci., 1987, 27, 1574–1581.
9. A. J. Lovinger and M. L. Williams, J. Appl. Polym. Sci., 1980, 25, 1703–1713.
10. G. N. Avgeropoulos, F. C. Weissert, P. H. Biddison and G. G. A. Boehm, Rubber Chem. Technol., 1976, 49, 93–104.
11. T. S. Omonov, C. Harrats, P. Moldenaers and G. Groeninckx, Polymer, 2007, 48, 5917–5927.
12. G. M. Jordhamo, J. A. Manson and L. H. Sperling, Polym. Eng. Sci., 1986, 26, 517–524.
13. R. Rajesh, N. K. Singha and K. Naskar, Polym. Bull., 2011, 66, 95–118.
14. M. S. Sureshkumar, K. Naskar, G. B. Nando, Y. K. Bhardwaj and S. Sabharwal, Polym.-Plast. Technol. Eng., 2008, 47, 341–345.
15. A. G. Jineesh and G. B. Nando, J. Elastomers Plast., 2012, 44, 189–204.
16. J. K. Mishra, K. J. Hwang and S. H. Chang, Polymer, 2005, 46, 1995–2002.
17. J. Graces, D. Moll, J. Bicerano and R. Fibiger, Adv. Mater., 2000, 12, 1835–1839.
18. E. A. Cherney, IEEE Trans. Dielectr. Electr. Insul., 2005, 12, 1108–1115.
19. L. H. Meyer, E. A. Cherney and S. H. Jayaram, IEEE Electr. Insul. Mag., 2004, 20, 13–21.
20. R. N. Santra, B. K. Samantaray, A. K Bhowmick and G. B. Nando, J. Appl. Polym. Sci., 1993, 49, 1145–1158.
21. R. N. Santra, S. Roy, A. K. Bhowmick and G. B. Nando, Polym. Eng. Sci., 1993, 33, 1352–1359.
22. R. N. Jana, H. P. Bhunia and G. B. Nando, Thermochim. Acta, 1997, 302, 1–9.
23. R. N. Jana, A. K. Bhattacharya, G. B. Nando and B. R. Gupta, Kautsch. Gummi Kunstst., 2002, 55, 660–664.
24. R. N. Jana and G. B. Nando, J. Appl. Polym. Sci., 2003, 90, 635–642.
25. R. N. Jana, P. Mukunda and G. B. Nando, Polym. Degrad. Stab., 2003, 80, 75–82.
26. R. N. Jana and G. B. Nando, J. Elastomers Plast., 2005, 37, 149–163.
27. R. N. Jana and G. B. Nando, J. Thermoplast. Compos. Mater., 2008, 90, 225–241.
28. R. Giri, K. Naskar and G. B. Nando, Plast., Rubber Compos., 2012, 41, 341–349.
29. R. Giri, M. Sureshkumar, K. Naskar, Y. K. Bharadwaj, K. S. S. Sharma, S. Sabharwal and G. B. Nando, Adv. Polym. Technol., 2008, 27, 98–107.
30. R. Giri, K. Naskar and G. B. Nando, Radiat. Phys. Chem., 2012, 81, 1930–1942.
31. R. Giri, K. Naskar and G. B. Nando, Mater. Express, 2012, 2, 37–50.
32. R. Padmanabhan, K. Naskar and G. B. Nando, Mater. Res. Express, 2015, 2, 105301,  DOI:10.1088/2053-1591/2/10/105301.
33. W. Głuszewski, Z. P. Zagórski and M. Rajkiewicz, Radiat. Phys. Chem., 2014, 105, 53–56.
34. S. Perraud, M. F. Vallat and J. Kuczynski, Macromol. Mater. Eng., 2003, 288, 117–123.
35. S. Zhang, B. Sun, X. Jiang, L. Li, Z. Meng and M. Zhu, Radiat. Phys. Chem., 2015, 106, 242–246.
36. J. K. Mishra, Y. W. Chang, B. C. Lee and S. H. Ryu, Radiat. Phys. Chem., 2008, 77, 675–679.
37. J. S. Youm, J. C. Kim and K. S. Yang, Fibers Polym., 2012, 13, 1165–1169.
38. K. Naskar, U. Gohs, U. Wagenknecht and G. Heinrich, eXPRESS Polym. Lett., 2009, 3, 677–683.
39. S. Poongavalappil, P. Svoboda, R. Theravalappil, D. Svobodova, M. Danek and M. J. Zatloukal, J. Appl. Polym. Sci., 2013, 128, 3026–3033.
40. R. D. Yang, R. R. Mather and A. F. Fotheringham, J. Mater. Sci., 2001, 36, 3097–3101.
41. S. K. Park, K. D. Kim and H. T. Kim, Colloids Surf., A, 2002, 197, 7–17.
42. J. M. Liu, P. Y. Lu and W. K. Wong, Mater. Sci. Eng., B, 2001, 85, 209–211.
43. R. K. Roy, A Primer on the Taguchi Method, Van Nostrand Reinhold, New York, 1990, ch. 2, pp. 18–20.
44. S. Bhandari, N. K. Singha and D. Khastgir, J. Appl. Polym. Sci., 2013, 129, 1264–1273.
45. J. K. Kim and H. Lee, Polymer, 1996, 37, 305–311.
46. L. A. Utracki, J. Rheol., 1991, 35, 1615–1637.
47. K. Sirisinha and M. Darinya, J. Appl. Polym. Sci., 2004, 93, 1179–1185.
48. J. Dutta, T. Chatterjee, G. Dhara and K. Naskar, RSC Adv., 2015, 5, 41563–41575.
49. F. Ziaie and M. A. Bolorizadeh, Nukleonika, 2007, 52, 77–81.
50. M. Sabet, A. Hassan, M. U. Wahit and C. T. Ratnam, Polym.-Plast. Technol. Eng., 2010, 49, 589–594.
51. L. Boudou and J. Guastavino, J. Phys. D: Appl. Phys., 2002, 35, 1555–1561.
52. C. Oproiu, M. Toma, S. Marghitu and A. Jianu, Nucl. Instrum. Methods Phys. Res., Sect. B, 2000, 166, 669–675.
53. D. Kostoski, J. Dojcilovic, L. Novakovic and E. Suljovrujic, Polym. Degrad. Stab., 2006, 91, 2229–2232.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra19049e

---