Room temperature and low temperature toughness improvement in PBA-g-SAN/α-MSAN by melt blending with TPU

Abstract

Poly(butyl-acrylate)-g-poly(styrene-co-acrylonitrile)/α-methyl-styrene-acrylonitrile/thermoplastic polyurethane elastomer (PBA-g-SAN/α-MSAN/TPU) ternary blends were prepared with different composition ratios via melt blending. This work is mainly focused on improving the toughness of PBA-g-SAN/α-MSAN blends at room temperature (25 °C) and low temperature (0 °C and −30 °C). The results of notched Izod impact strength tests revealed that TPU had good toughening efficiency at 25 °C and 0 °C. With addition of 30 phr TPU, the impact strength of the blends increased from 5.3 kJ m⁻² to 26.5 kJ m⁻² at 25 °C, and increased from 2.9 kJ m⁻² to 18.7 kJ m⁻² at 0 °C. Obvious brittle-tough transition of the ternary blends was observed with increasing TPU content at 25 °C and 0 °C. The toughening efficiency of TPU was not notable when the impact temperature dropped to −30 °C, but the impact strength of the blends still increased with increasing TPU content. The glass transition temperature (T_g) of the blends was mainly responsible for the temperature dependence of toughening efficiency, which was proved by dynamic mechanical thermal analysis. It was also found the addition of TPU could lead to the slightly decreased T_g of the blends in the low temperature region. The probable reason for this is the unique structure formed in ternary blends, which was proved by attenuated total reflection infrared analysis and contact angle tests. The impact-fracture surfaces of blends were observed by scanning electron microscopy. The introduction of TPU brought with it a slight decrease of tensile strength, but the elongation at break of the blends increased greatly. Flexural strength, flexural modulus and heat resistance showed a decreasing tendency with increasing TPU content.

---

1 Introduction

Polymer blending, an effective and economical way of developing novel polymer materials, is effective for obtaining materials with various properties through simple process technology. It has attracted much attention from the scientific and industrial communities. The properties of polymer blends are influenced by many factors, such as the composition of the blends, the properties of each component, the compatibility between the constituents and processing conditions.^1–4

The poly(butyl-acrylate)-g-poly(styrene-co-acrylonitrile) (PBA-g-SAN) copolymer is one of the most widely used impact modifiers with a core–shell structure. PBA particles constitute elastomer core, while styrene and acrylonitrile monomers are grafted onto the core by copolymerization to form the rigid shell.⁵ The unique core–shell structure endows PBA-g-SAN with toughness, strength, and high efficiency to absorb impact energy. PBA-g-SAN has similar structure to acrylonitrile-butadiene-styrene (ABS), so it also has the analogical properties of ABS, such as desirable mechanical properties, chemical resistance, dimensional stability, good surface smoothness, and easy processability.^6,7 However, butadiene rubber, one of components in ABS is highly susceptible to aging problems for the appearance of double bonds in its main chain.⁸ This weakness strongly limits its wider application. PBA-g-SAN is synthesized with the saturated acrylate rubber to provide superior weatherability and high impact strength simultaneously.⁹ Thus PBA-g-SAN is more attractive in numerous outdoor applications. In order to prepare the appropriate composite materials for desired properties, many polymer blends containing PBA-g-SAN were investigated, such as PBA-g-SAN/polyamide-6 (PA-6),¹⁰ PBA-g-SAN/polyvinyl chloride (PVC),¹¹ PBA-g-SAN/polyethylene terephthalate (PET),¹² PBA-g-SAN/SAN¹³ and so on.

Usually, commercially used PBA-g-SAN resin was prepared by combining PBA-g-SAN with SAN. However, as the shell phase, SAN is completely miscible with PBA-g-SAN, which limited the raise of heat distortion temperature (HDT).¹⁴ Previous study has shown that with the addition of 50 wt% SAN, the HDT of the blends only increased 5 °C (measured under maximum pressure of 1.80 MPa).¹³ Recently, α-methyl-styrene-acrylonitrile (α-MSAN) was chosen to replace SAN to blend with PBA-g-SAN. Compared with SAN, α-MSAN has a much stronger rigid structure, which is attributed to the strong steric effect of the benzene ring and additional α-methyl. Zhang's work¹⁵ showed α-MSAN was effective in enhancing HDT of rigid PVC. Zhu¹⁴ found the partial miscibility of PBA-g-SAN/α-MSAN blends maintains the high glass transition temperature (T_g) of α-MSAN. Compared with PBA-g-SAN, the HDT of PBA-g-SAN/α-MSAN increased 20 °C when the addition of α-MSAN was larger than 70 wt%. Although PBA-g-SAN/α-MSAN has good heat resistance and stiffness, the toughness is damaged by large scale addition of α-MSAN. For expanding the applied range of PBA-g-SAN/α-MSAN, it is significant to explore an efficient way to improve the impact resistance of PBA-g-SAN/α-MSAN, including room temperature toughness and low temperature toughness.

Since 1960 Goodrich firstly realized the industrial production of thermoplastic polyurethane elastomer (TPU), it has been widely used in areas of biomedical applications, automobile, and electronic for its excellent mechanical properties and biocompatibility.^16–18 Being one of the mostly used thermoplastic elastomer, TPU was usually applied to toughen polylactide (PLA).^17,19–21 TPU is a kind of linear segmented block copolymer consisting of hard segments (HS, adduct of diisocyanate and chain extender) and soft segments (SS, long flexible polyester or polyether chains). Thermodynamical incompatibility between HS and SS leads to microphase separation of TPU. The amorphous SS have a low temperature T_g which helps the TPU present as rubbery state in a wide temperature range. SS are responsible for high elasticity of TPU. The HS are held together by interchain hydrogen bonds which make the TPU form a physical crosslinking structure. The HS also act as reinforcing fillers and give strength to the TPU.^21,22 The unique microstructure of TPU combines the mechanical properties of vulcanized rubber with the processability of thermoplastic polymers.²³

In this work, polyester based TPU was chosen to toughen the PBA-g-SAN/α-MSAN (25/75, W/W) binary blends. Notched Izod impact test was performed over a wide range of temperatures (25 °C, 0 °C and −30 °C) to verify the enhancement of toughness. The evolution of impact-fractured surface morphology of the ternary blends was studied by scanning electron microscopy (SEM). Other mechanical properties, such as tensile and flexural properties were also characterized. Heat resistance of the ternary blends was evaluated by HDT. Attenuated total reflection infrared (ATR-IR) spectroscopy was used to determine the intermolecular interaction between each component of the ternary blends.

2 Experiment

2.1. Materials

PBA-g-SAN (Grade: HX-960) copolymer rubber powders with 60 wt% PBA content was friendly supplied by Zibo Huaxing addictives Co. Ltd., China. α-MSAN (Grade: NR-188) containing 30 wt% of acrylonitrile was procured from Yixin Lilai Chemical Co. Ltd., China. Polyester based TPU (Grade: IROGAN A80H4698) was obtained from Huntsman Polyurethanes Shanghai Ltd., China.

2.2. Sample preparation

PBA-g-SAN/α-MSAN/TPU blends with different weight ratios were prepared by melt blending using a two-roll mill at 180 °C. Additives (anti-oxidant and lubricant) were used to avoid thermal degradation during the blending process. The prepared blends were compression molded into sheets of approximately 2 mm and 4 mm thickness, respectively, at 180 °C under the pressure of 10 MPa in a flat plate vulcanization machine. The obtained sheets measuring 2 mm thickness were cut into dumb bell shaped pieces for tensile tests. Rectangular samples (80 × 10 × 4 mm³) cut from 4 mm thickness sheets were used for flexural, impact and HDT tests.

2.3. Characterization

2.3.1 Dynamic mechanical thermal analysis. The dynamic mechanical properties of the blends were measured in a dynamic mechanical thermal analyzer (DMTA) (MCR302, Anton Paar, Austria) at a heating rate of 3 °C min⁻¹ from −90 °C to 160 °C. The experiment was carried out under the extension mode at a fixed frequency of 1 Hz. The dimensions of samples prepared for testing were approximately 25 × 6 × 2 mm³.

2.3.2 Attenuated total reflection infrared analysis. Attenuated total reflection infrared (ATR-IR) spectroscopy was used to characterize the interaction among the components of the PBA-g-SAN/α-MSAN/TPU blends. The spectroscopy was obtained by an ATR-IR spectrometer (Nicolet IS5, Thermo Fisher, America) with resolution of 4 cm⁻¹ in the range of 4000–600 cm⁻¹ wavenumber.

2.3.3 Contact angle tests. Contact angle tests were conducted on a DSA 100 drop shape analysis system (Krüss, Germany) at room temperature. The specimens were prepared by compression molding. The contact angle was recorded at 30 s after 2.0 μL of solvent was dropped onto the surface of the specimen. At least five replicates were performed for each sample to ensure the reproducibility of the measured data.

2.3.4 Mechanical properties measurements. The toughening effect of TPU on PBA-g-SAN/α-MSAN blends was evaluated by the impact strength test. The test of notched Izod impact strength was carried out on an Izod impact tester (UJ-4, Chengde Machine Factory, China). The prepared rectangular samples were preserved at −30 °C, 0 °C and 25 °C for 12 hours before the test. Tensile and flexural properties of the blends were tested through a universal testing machine (CMT 5254, Shenzhen SANS Testing Machine Co. Ltd, China) at room temperature. The tests were carried out with a constant rate of 5 mm min⁻¹ for tensile tests and 2 mm min⁻¹ for flexural tests according to ISO 527 and ISO 178, respectively. At least three specimens were tested for each polymer system and the mean values were calculated.

2.3.5 Scanning electron microscope analysis. The morphology of impact-fractured surfaces of the PBA-g-SAN/α-MSAN/TPU blends was observed through a SEM apparatus (JSM-5900, JEOL, Japan) with an accelerating voltage of 15 kV. The fractured surfaces of blends were coated with a thin conductive layer of gold before scanning.

2.3.6 Heat distortion temperature measurements. The HDT of the PBA-g-SAN/α-MSAN/TPU ternary blends was determined by the Vicat/HDT equipment (ZWK1302-2, Shenzhen SANS Testing Machine Co. Ltd., China) at a heating rate of 120 °C h⁻¹. All the tests were conducted under the maximum bending stress of 1.80 MPa and 0.45 MPa respectively according to ISO 75-1.

3 Results and discussion

3.1. DMTA analysis

DMTA was used to measure the thermodynamic response of the ternary blends. Fig. 1 exhibits tan δ vs. temperature for the PBA-g-SAN/α-MSAN/TPU ternary blends. The peaks of the tan δ curves were chosen to characterize the T_g of the materials, specific values are shown in Table 1. There are two characteristic peaks for neat PBA-g-SAN at −41.9 °C and 119 °C, corresponding to the T_g for the PBA rubber core and SAN shell, respectively. α-MSAN has a much higher T_g at 146 °C. The T_g of TPU is measured as −33.9 °C. With respect to PBA-g-SAN/α-MSAN/TPU ternary blends, the T_g peaks of SAN shell are difficult to find from the curves because they are concealed by peaks of α-MSAN with stronger intensity. The T_g of the ternary blends at low temperature shifts between −42 °C to −44 °C.

Fig. 1 tan δ curves for PBA-g-SAN, α-MSAN, TPU and PBA-g-SAN/α-MSAN/TPU ternary blends (25/75/0-30) with different blend ratios.

Table 1 Storage modulus (25 °C) and T_g values of PBA-g-SAN/α-MSAN/TPU blends obtained from DMTA tests

PBA-g-SAN/α-MSAN/TPU	Storage modulus (Pa)	Tg,1 (°C)	Tg,2 (°C)	Tg,3 (°C)	Tg,4 (°C)	Tg,5 (°C)
100/0/0	1.29 × 10^8	—	−41.9	—	119	—
0/100/0	1.14 × 10^9	—	—	—	—	146
0/0/100	9.87 × 10^6	—	—	−33.9	—	—
25/75/0	1.03 × 10^9	—	−39.9	—	—	148
25/75/10	8.96 × 10^8	−44.1	—	—	—	148
25/75/20	8.02 × 10^8	−44.4	—	—	—	147
25/75/30	7.89 × 10^8	−42.5	—		—	149

---

In PBA-g-SAN/α-MSAN binary blends, the T_g of PBA core shows a slight increase of 2 °C compared with neat PBA-g-SAN. In this study, the weight ratio of PBA-g-SAN and α-MSAN is determined as 25/75. So the dispersion phase of PBA-g-SAN is wrapped by the continuous phase of α-MSAN (as shown in Scheme 1). α-MSAN has stronger steric effect compared with SAN because of the additional –CH₃ group, the free volume for chain segment movement is relatively smaller in PBA-g-SAN/α-MSAN blends. In that way, the movement of the PBA chains becomes more difficult at its original T_g, which leads to a higher glass transition temperature. It is worth noting that at low temperature only one peak is found at the curves of the ternary blends with TPU. In general, polymer blends in which their components are fully miscible exhibit one single T_g, whereas for those immiscible polymer blends, two or more T_gs can be found.¹⁴ The single T_g of ternary blends at low temperature indicates that TPU and PBA-g-SAN are completely miscible with each other, for both the TPU soft segments and PBA core contain the ester group. It is also noted that the T_g of PBA-g-SAN/α-MSAN/TPU ternary blends at low temperature is slightly lower than that of neat PBA-g-SAN and PBA-g-SAN/α-MSAN blends. The probable reason may be attributed to good miscibility between PBA-g-SAN and TPU. PBA-g-SAN particles will be surrounded by TPU during the melt blending process (as shown in Scheme 1). On the contrary of the restriction effect of rigid α-MSAN, the soft segment of TPU can play a role like plasticizer for PBA-g-SAN. In this way, PBA-g-SAN will get expanding moving space and the T_g will get lower, which is beneficial for low temperature impact resistance of the blends to a certain extent.

Scheme 1 Schematic presentation for the reduced T_g of PBA-g-SAN/α-MSAN/TPU at low temperature.

Storage modulus is representative of the stiffness of materials. As shown in Fig. 2 and Table 1, the storage modulus of PBA-g-SAN at 25 °C is improved by an order of magnitude after blending with α-MSAN (from 1.29 × 10⁸ Pa to 1.03 × 10⁹ Pa). It can be observed that the PBA-g-SAN/α-MSAN blends can retain a high storage modulus before 125 °C, which is about 25 °C higher than neat PBA-g-SAN. It means the introduction of rigid α-MSAN can enhance the stiffness and heat resistance of the blends simultaneously. As listed in Table 1, the storage modulus of the blends decreases slightly from 1.03 × 10⁹ Pa for PBA-g-SAN/α-MSAN to 7.89 × 10⁸ Pa for PBA-g-SAN/α-MSAN/TPU (25/75/30) at 25 °C. The decrease in storage modulus indicates the addition of TPU will reduce the stiffness of the blends, which is attributed to the low modulus of TPU.

Fig. 2 Variation in the storage modulus of PBA-g-SAN, α-MSAN, TPU and PBA-g-SAN/α-MSAN/TPU ternary blends with different blend ratios.

3.2. Attenuated total reflection infrared spectroscopy

In order to determine the chemical composition and intermolecular interaction of PBA-g-SAN/α-MSAN/TPU ternary blends, ATR-IR analysis was carried out at room temperature. The spectra is shown in Fig. 3. The bands at approximately 1733 cm⁻¹ and 2235 cm⁻¹ are attributed to stretching vibration of carbonyl (C=O) and cyano (C≡N), respectively. The absorption peaks at 699 cm⁻¹ and 1601 cm⁻¹ are characteristic peaks of single substituted phenyl ring and vibrations of benzene skeletal ring, respectively. In addition, bands at 2930 cm⁻¹ and 1388 cm⁻¹ are indicative of C–H asymmetric stretching vibration of –CH₂– and bending vibration of –CH₃. Peaks at 3336 cm⁻¹ and 1532 cm⁻¹ are attributed to symmetric stretching vibration and deformation vibration of N–H, respectively.

Fig. 3 ATR-IR spectroscopy of (a) TPU, and PBA-g-SAN/α-MSAN/TPU (25/75/0-30) ternary blends with (b) 0, (c) 10, (d) 20 and (e) 30 phr TPU.

The exact positions of the carbonyl peaks are listed in Table 2. It can be found that the carbonyl peak of the ternary blends shifts between neat-TPU and PBA-g-SAN/α-MSAN. If a certain interaction between groups exists in polymer blends, the peaks position of IR spectra for the blends will change compared with the unblended components.²⁴ According to the peaks shift of carbonyl group in ternary blends, a carbonyl–carbonyl dipolar interaction is presumed to occur between the chains of PBA-g-SAN and TPU.^25–27 It demonstrates that the carbonyl of TPU and the carbonyl of PBA-g-SAN will interact with each other in the ternary blends. ATR-IR results provide evidence for the existence of the structure shown in Scheme 1 and support the explanation of decreased T_g at low temperature.

Table 2 Positions of the carbonyl peaks of neat-TPU, and PBA-g-SAN/α-MSAN/TPU blends (25/75/0-30) with different compositions

PBA-g-SAN/α-MSAN/TPU	Wavenumbers (cm^−1)
0/0/100	1729.2
25/75/0	1733.9
25/75/10	1733.1
25/75/20	1731.7
25/75/30	1732.6

---

3.3. Contact angle tests

To furtherly confirm the formed structure of ternary blends shown in Scheme 1 from the viewpoint of thermodynamic factors, contact angle tests were carried out to evaluate the interfacial tension between the components. The results of contact angle and surface tension of each component are shown in Table 3. The Owens–Wendt method²⁸ is used herein to calculate the dispersive and polar surface tension.

Table 3 Contact angle and surface tension of different components in PBA-g-SAN/α-MSAN/TPU ternary blends

Sample	Contact angle (°)		Surface tension (mJ m^−2)
	Distilled water	Diiodomethane	γ	γ^d	γ^p
TPU	89.8	49.4	35.01	32.49	2.52
PBA-g-SAN	96.1	43.9	37.59	36.92	0.67
α-MSAN	97.9	39.7	39.99	39.70	0.29

---

The interfacial tension between different components can be calculated based on the methods of harmonic mean eqn (1)²⁹ or geometric mean eqn (2):³⁰

where γ_AB is interfacial tension between component A and component B, γ_A and γ_B are surface tension of component A and component B, respectively. The superscript of d and p are representative for dispersive and polar terms, respectively.

Interfacial tension of TPU/PBA-g-SAN and TPU/α-MSAN are shown in Table 4. It is found that the interfacial tension between TPU and α-MSAN is larger than that between TPU and PBA-g-SAN. Based on the thermodynamic rule that lowest free energy is preferred,³¹ TPU tends to interact more with PBA-g-SAN than with α-MSAN. Therefore, TPU particles are more likely to locate around the PBA-g-SAN during the blending process, which proves the mechanism proposed in Scheme 1.

Table 4 Interfacial tension of TPU/PBA-g-SAN and TPU/α-MSAN

Component couple	Based on harmonic mean equation (mJ m^−2)	Based on geometric mean equation (mJ m^−2)
TPU/PBA-g-SAN	1.36	0.73
TPU/α-MSAN	2.49	1.46

---

3.4. Mechanical properties

Impact strength of PBA-g-SAN/α-MSAN/TPU ternary blends tested at three different temperatures (25 °C, 0 °C and −30 °C) are shown in Fig. 4. It shows that the impact strength of the blends increases with the TPU content at all temperatures. Soft segments of TPU undergo large deformation during impact test because of its low elastic modulus, inducing crazes and shear band of the matrix. As a result, much impact energy will be dissipated. Simultaneously, the TPU particles can control or stop the crazes developing into failure cracks during the impact progress.³² The cavitation of TPU particles under the impact load also leads to the absorption of energy.^32,33 So the rigid matrix polymer will be toughened by incorporating TPU. But the increasing tendency of different temperatures exhibits some differences. At room temperature, the value and increment of the impact strength of the blends are relatively larger compared with that of the blends at 0 °C and −30 °C. At the curves of 25 °C, the ternary blends present a brittle-tough transition behavior. The transition content, where is the sharp shift in failure mode from brittle fracture with low impact toughness to ductile fracture with excellent impact toughness (21.9 kJ m⁻²), is around 15–20 phr TPU content. With further addition of TPU content to 30 phr, the impact strength value increases to 26.5 kJ m⁻², which is almost 5 times to that of blends without TPU added (5.3 kJ m⁻²). With respect to the curves of 0 °C, the impact strength of the blends increases slightly from 2.9 kJ m⁻² to 7.6 kJ m⁻² after 20 phr TPU was incorporated. With further addition of TPU to 30 phr, the value increases significantly to 18.7 kJ m⁻², which is improved by 6.4 times compared with blends without TPU added. When the test temperature drops to −30 °C, the impact strength increases slightly from 2.2 kJ m⁻² to 8.2 kJ m⁻² with addition of 30 phr TPU. According to Wu's model of interparticle distance, when the TPU particles are sufficiently close together, the stress field around neighbouring particles will interact considerably.³⁴ This will result in enhanced matrix yielding and a transition to tough behavior. The increasing TPU content in blends leads to smaller interparticle distance, causing higher impact strength of the blends and an obvious brittle-tough transition at 25 and 0 °C. The impact strength results reveal that the toughening effect of TPU on PBA-g-SAN/α-MSAN is remarkable at 25 °C and 0 °C, but not obvious at −30 °C.

Fig. 4 Impact strength of PBA-g-SAN/α-MSAN/TPU (27/75/0-30) ternary blends at different temperatures.

According to the DMTA analysis, the low temperature T_g of the ternary blends is far below 0 °C, which means rubber phase of TPU soft segments and PBA cores are both flexible at 25 °C and 0 °C. So the impact energy can dissipate effectively and toughening efficiency of TPU is high. At −30 °C, the ambient temperature is close to the T_g of ternary blends at low temperature. The chain segments of TPU and PBA core are nearly frozen, which will restrict the movement of macromolecular. So the TPU particles exhibit relatively low toughening efficiency at −30 °C. However, TPU still can exert toughening effect at −30 °C according to the increasing impact strength. Despite the toughening mechanism of TPU mentioned above, the decreased low temperature T_g of the ternary blends by incorporating TPU makes the chain segments more flexible at −30 °C compared with PBA-g-SAN/α-MSAN, which also attributes to the increasing toughness.

Fig. 5 shows the tensile properties of PBA-g-SAN/α-MSAN/TPU ternary blends. Compared with pristine PBA-g-SAN/α-MSAN, the tensile strength of the blends decreased slightly after 5 phr TPU was incorporated (from 40.2 MPa to 38.2 MPa). With further addition of 10 phr and 15 phr TPU, the tensile strength keeps almost unchanged around 38 MPa. A steeper drop of tensile strength appears when 20 phr and 30 phr TPU is added (34.9 MPa and 32.0 MPa, respectively). The decrease of tensile strength is mainly due to the relatively lower strength and modulus of TPU compared with α-MSAN. However, the decrement can be acceptable considering the retention rate of tensile strength is nearly 80% even with addition of 30 phr TPU. Elongation at break is also used to characterize the toughness of materials. In Fig. 5, the elongation at break of the ternary blends increases continuously with TPU content. The large scale enhancement of the elongation at break (from 4.2% to 96.2%) also suggests that the toughness of the blends improves significantly, which is consistent with the results of impact strength.

Fig. 5 Tensile properties of PBA-g-SAN/α-MSAN/TPU (25/75/0-30) ternary blends.

With respect to the flexural properties vs. TPU content (Fig. 6), both flexural strength and modulus decrease as a linear trend with the increasing TPU content. Specifically, the flexural strength decreases from 68.5 MPa for pristine PBA-g-SAN/α-MSAN to 45.5 MPa for blends with addition of 30 phr TPU. The flexural modulus decreases from 2547 MPa for pristine PBA-g-SAN/α-MSAN to 1445 MPa for blends with addition of 30 phr TPU. Flexural properties are usually used to evaluate the stiffness of materials, and the decline in flexural strength and modulus means the reduction of stiffness of the blends. It can be concluded that the increasing content of TPU can improve the toughness by large scale but sacrifices the stiffness of the blends to some degree.

Fig. 6 Flexural properties of PBA-g-SAN/α-MSAN/TPU (25/75/0-30) ternary blends.

3.5. SEM analysis

To further support the toughening results, SEM is used to observe microstructure of impact-fractured surfaces of the blends as shown in Fig. 7. TPU has good compatibility with PBA-g-SAN/α-MSAN since no obvious phase separation is observed in all SEM images, which is consistent with the single T_g of the ternary blends at low temperature. The weight fraction of TPU contained in the blends and the impact temperature have significant influence on the morphologies of the blends. In general, the impact-fractured surfaces with thread-like morphology indicate good toughness and high impact strength of the testing samples, while the smooth and flat surface represents for brittle fracture. At 25 °C, the impact-fractured surfaces transform into rough and thread-like morphology when the addition of TPU is more than 20 phr (Fig. 7c and d), which indicates the ductile fracture. At 0 °C, the thread-like morphology appears when 30 phr TPU has been added (Fig. 7h). In comparison, all the impact-fractured surfaces show clear-cut morphology when the test temperature is −30 °C (Fig. 7i–l), proving the blends failed in a brittle manner during impact test. Only the blends with the addition of 30 phr TPU exhibit relatively rougher morphology (Fig. 7l). The morphology of the blends with same composition shows continuous transformation along with different temperature (such as Fig. 7a, e and i).

Fig. 7 SEM micrographs of impact-fractured surfaces of PBA-g-SAN/α-MSAN/TPU (25/75/0, 10, 20, 30) ternary blends, (a)–(d): impact at 25 °C, (e)–(h): impact at 0 °C, (i)–(l): impact at −30 °C.

In a higher magnification, some separate small spherical holes can be observed on the fracture surfaces of the blends with low TPU content (Fig. 7a, e and i), which was caused by cavitation of TPU particles. It indicates that with low TPU content, the stress field around dispersed particles cannot interact considerably under impact condition and TPU particles work separately. So blends with low TPU content fail in brittle manner. However, the increasing TPU content will shorten the distance between particles. In that way, the stress field around particles can be connected and interact effectively, based on which the brittle-tough transition occurs, simultaneously the morphology with separated holes is replaced by the rough thread-like morphology (Fig. 7c, d and h). With respect to −30 °C, the morphology with holes still can be observed even when the addition of TPU is 30 phr (Fig. 7l). At a low temperature, TPU particles are more like rigid fillers and lose the rubbery nature. The deformation of TPU under impact condition becomes more difficult. Therefore, cavitation of TPU particles at −30 °C cannot improve the toughness of the blends significantly as 0 °C and 25 °C, the morphology also has no obvious change. SEM analysis establishes the relationship between microstructures and mechanical properties of the blends, and all morphologies coincide well with the results of the impact test.

3.6. Heat distortion temperature analysis

Fig. 8 shows the HDT of PBA-g-SAN/α-MSAN/TPU ternary blends, which is usually used to evaluate the heat resistance of materials. In general, the HDT of the blends shows an inversely proportional to the increasing content of TPU. The HDT decreases from 115 °C for pristine PBA-g-SAN/α-MSAN to 107 °C for blends with addition of 30 phr TPU (measured under the maximum stress of 0.45 MPa), and the HDT can be decreased from 104 °C for PBA-g-SAN/α-MSAN to approximately 95 °C for blends with addition of 30 phr TPU (measured under the maximum stress of 1.80 MPa), respectively. Generally, the incorporation of rubber phase may have unfavorable effect on heat resistance. The rubbery nature of TPU soft segments improves the toughness of the blends but sacrifices the heat resistance. However, the decrement of HDT (approximately 10 °C) is not very large, which is acceptable.

Fig. 8 HDT of PBA-g-SAN/α-MSAN/TPU (25/75/0-30) ternary blends.

4 Conclusion

PBA-g-SAN/α-MSAN/TPU ternary blends of different blend ratios were prepared via melt blending. According to the results of notched Izod impact test, the addition of TPU can significantly enhance the toughness of the blends at 25 °C and 0 °C. Compared with pristine PBA-g-SAN/α-MSAN, the impact strength of the ternary blends with addition of 30 phr TPU increased by 5.0 times and 6.4 times at 25 °C and 0 °C, respectively. At a lower temperature for −30 °C, the toughening efficiency of TPU was not notable like 25 °C and 0 °C, but the impact strength still increased with addition of TPU. The temperature dependence of toughening efficiency was mainly accounted for the material's T_g at low temperature. According to the results of DMTA analysis, only one T_g was found at low temperature in the ternary blends, which meant the PBA core and soft segments of TPU were completely miscible with each other. No phase separation observed in SEM also suggested good miscibility between the components. In addition, the T_g of the PBA-g-SAN/α-MSAN/TPU at low temperature shifted slightly to a lower one (between −42.5 °C and −44.4 °C) compared with pristine PBA-g-SAN/α-MSAN (−39.9 °C). ATR-IR analysis and contact angle tests revealed that TPU interacted more with PBA-g-SAN than with α-MSAN. Therefore, the unique structure formed in ternary blends (Scheme 1), which was caused by good miscibility and interaction between TPU and PBA-g-SAN, was responsible for the decreased T_g at low temperature. The SEM micrographs of the impact-fractured surfaces were in good agreement with impact strength. In tensile test, elongation at break increased significantly from 4.2% to 96.2% when 30 phr TPU was introduced. Retention of tensile strength of the blends kept above 80%. But the improvement of toughness was at the cost of the stiffness of the blends, such as flexural strength and modulus. Heat resistance of the blends also decreased to some extent with the increasing TPU content.

Acknowledgements

This work was supported by the Innovation Priority Academic Program Development of Jiangsu Higher Education Institutions (PAPD).

References

1. D. Bourry and B. D. Favis, J. Polym. Sci., Part B: Polym. Phys., 1998, 36, 1889–1899.
2. I. Vinckier, P. Moldenaers and J. Mewis, J. Rheol., 1997, 41, 705–718.
3. G. Wildes, H. Keskkula and D. R. Paul, Polymer, 1999, 40, 7089–7107.
4. N. Kitayama, H. Keskkula and D. R. Paul, Polymer, 2000, 41, 8041–8052.
5. Y. Han, Z. X. Tai, C. Zhou, M. Y. Zhang, H. X. Zhang and F. Q. Liu, Polym. Bull., 2009, 62, 855–866.
6. J. An, C. Kim, B. H. Choi and J. M. Lee, Polym. Eng. Sci., 2014, 54, 2791–2798.
7. N. K. Shrivastava, S. Suin, S. Maiti and B. B. Khatua, RSC Adv., 2014, 4, 24584–24593.
8. C. K. Radhakrishnan, R. Alex and G. Unnikrishnan, Polym. Degrad. Stab., 2006, 91, 902–910.
9. Y. Zhang, Y. Xu, Y. Song and Q. Zheng, J. Appl. Polym. Sci., 2013, 130, 2143–2151.
10. Z. G. Liu, Y. J. Deng, Y. Han, M. Chen, S. L. Sun, C. L. Cao, C. Zhou and H. X. Zhang, Ind. Eng. Chem. Res., 2012, 51, 9235–9240.
11. Y. G. Du, J. G. Gao, J. B. Yang and X. Q. Liu, J. Polym. Res., 2012 DOI:10.1007/s10965-012-9993-3.
12. C. M. Benson and R. P. Burford, J. Mater. Sci., 1996, 31, 1425–1430.
13. W. Zhang, S. Chen and J. Zhang, J. Thermoplast. Compos. Mater., 2013, 26, 322–335.
14. W. Zhu and J. Zhang, J. Vinyl Addit. Technol., 2014 DOI:10.1002/vnl.21403.
15. Z. Zhang, S. Chen and J. Zhang, J. Vinyl Addit. Technol., 2011, 17, 85–91.
16. Q. Cao, Y. L. Cai, B. Jing and P. S. Liu, J. Appl. Polym. Sci., 2006, 102, 5266–5273.
17. F. Feng and L. Ye, J. Appl. Polym. Sci., 2011, 119, 2778–2783.
18. L. C. Tan, Q. X. Su, S. D. Zhang and H. X. Huang, RSC Adv., 2015, 5, 80884–80892.
19. X. Jing, H. Y. Mi, X. F. Peng and L. S. Turng, Polym. Eng. Sci., 2015, 55, 70–80.
20. R. L. Yu, L. S. Zhang, Y. H. Feng, R. Y. Zhang and J. Zhu, Chin. J. Polym. Sci., 2014, 32, 1099–1110.
21. H. Hong, J. Wei, Y. A. Yuan, F. P. Chen, J. Wang, X. Qu and C. S. Liu, J. Appl. Polym. Sci., 2011, 121, 855–861.
22. Q. W. Lu and C. W. Macosko, Polymer, 2004, 45, 1981–1991.
23. A. Frick and A. Rochman, Polym. Test., 2004, 23, 413–417.
24. S. W. Peng, X. Y. Wang and L. S. Dong, Polym. Compos., 2005, 26, 37–41.
25. B. D. Zhang, M. H. Lee, A. N. Chakoli, W. C. Zang, K. Zhang, Y. Zhang, G. L. Song, C. H. Chen, X. D. Li and Y. Li, RSC Adv., 2014, 4, 14024–14030.
26. S. K. Murase, N. Haspel, L. J. del Valle, E. A. Perpete, C. Michaux, R. Nussinov, J. Puiggali and C. Aleman, RSC Adv., 2014, 4, 23231–23241.
27. X. F. Ma, J. G. Yu and N. Wang, J. Polym. Sci., Part B: Polym. Phys., 2006, 44, 94–101.
28. H. Xiu, H. W. Bai, C. M. Huang, C. L. Xu, X. Y. Li and Q. Fu, eXPRESS Polym. Lett., 2013, 7, 261–271.
29. S. Y. Hobbs, M. E. J. Dekkers and V. H. Watkins, Polymer, 1988, 29, 1598–1602.
30. L. As'habi, S. H. Jafari, B. Baghaei, H. A. Khonakdar, P. Potschke and F. Bohme, Polymer, 2008, 49, 2119–2126.
31. Z. Zhang, X. J. Zhao, S. C. Wang, J. Zhang and W. Zhang, RSC Adv., 2014, 4, 60617–60625.
32. M. Mehrabzadeh and D. Rezaie, J. Appl. Polym. Sci., 2002, 84, 2573–2582.
33. Y. Y. Shi, W. B. Zhang, J. H. Yang, T. Huang, N. Zhang, Y. Wang, G. P. Yuan and C. L. Zhang, RSC Adv., 2013, 3, 26271–26282.
34. S. H. Wu, Polymer, 1985, 26, 1855–1863.

---