Preparing thermoplastic polyurethane/thermoplastic starch with high mechanical and biodegradable properties

Abstract

This study focused on achieving biodegradable thermoplastic polyurethane (TPU) blends with good mechanical properties by the incorporation of starch. Thermoplastic starch (TPS) was developed by plasticizing corn starch in a twin screw. Polyolefin elastomer (POE) was used as the compatibilizer to improve the flexibility of the TPU/TPS system. The miscibility of the blends was investigated by Fourier transform infrared spectroscopy (FT-IR), scanning electron microscopy (SEM) and differential scanning calorimetry (DSC). Mechanical analysis (impact, tensile, and folding endurance tests), dynamic mechanical analysis (DMA), degradation tests and contact angle (CA) tests were employed to study the mechanical properties and hydrophilic sensitivity of the TPU/TPS blends. The results showed that POE improved the miscibility of the blends by enhancing the interfacial effects of the TPU/TPS blend, decreasing the interfacial tension between TPS and TPU, as well as increasing the inter-hydrogen bonds between TPU and TPS. The hydrophilic qualities of the TPU blends, such as water adsorption and CA, increased with increasing TPS content, resulting in improved biodegradation ability. TPU/TPS (20 wt%)/POE (10 wt%) endowed the blends with good folding endurance (>30.0 × 10³), notched impact strength (no break), elongation at break (>800%) and a 6.2% biodegradation rate after 7 weeks. The improved properties of TPU/TPS showed that these blends have high potential for use in environmentally friendly materials.

---

Introduction

Recently, thermoplastic polyurethane (TPU) has been widely applied in the fields of automotives, screens, roller system films, medicine, sports products, aerospace, electronics, etc. The molecular chain of TPU contains alternating hard and soft segments which endow it with good mechanical properties,¹ such as higher impact toughness and abrasive resistance. However, the drawbacks of TPU, such as its high price, non-biodegradability and low water sensitivity, limit its further application. Therefore, developing reinforced TPU composites is an effective method to improve the properties of TPU. Generally, fibers and nano-particles are commonly used to reinforce TPU. Layered silicates,^2–6 mica^7,8 and MMT⁹ have been reported to improve the mechanical, thermal, and barrier properties of TPU.

Recently, starch has attracted more and more attention from researchers and industrial companies due to its properties of renewability and biodegradability as well as its abundance in natural sources. During a thermal process, thermoplastic starch (TPS) is formed by incorporating plasticizers, such as glycerol, urea, formamide¹⁰ and water,¹¹ into corn starch. Some researchers have used polyurethane to react with TPS in order to improve the mechanical and hydrophobic performance of TPS;^12–17 however, the thermoplastic properties of PU/TPS were poor. Seidenstückert^18,19 developed thermoplastic and biodegradable materials based on TPU/TPS with good hydrolysis and biodegradation. They surveyed the effects of the TPU and TPS types and process designs, as well as their impact on the properties of the materials. However, as the starch content increased, the toughness of the TPU/TPS blends drastically decreased. Considering eco-friendly and economic requirements, the preparation of TPU/TPS blends that efficiently perform in these categories is necessary. However, the hydrophobicity of TPU and the hydrophilicity of TPS cause problems such as adhesion and wettability of the TPU/TPS blends. Furthermore, when TPS is blended with TPU, the introduction of rigid TPS will destroy the balance of microphase segregation and decrease the mechanical properties of TPU. Therefore, compatibilizers must be used in the starch-TPU matrix interface to improve the adhesion and compatibility. Chiu et al.²⁰ demonstrated that POE-g-MA improved the low viscosity and compatibility of PA and TPU during the blending process. POE-g-MA significantly improved the impact strength of PA to super-toughness grade (>800 J m⁻¹). TPU could achieve better mechanical properties because micro-phase segregation occurred between the hard and soft segments.¹² There are no reports on whether POE could rehabilitate the balance of the microphase segregation of TPU/TPS due to its excellent flexibility. Variations in the mechanical properties and biodegradability of TPU/TPS were not investigated when POE was employed as a toughening agent.

In this work, TPU/TPS blends were prepared by extrusion processing in order to obtain economically biodegradable materials with good mechanical properties. POE was used as the compatibilizer. In this study, the effects of POE loading on the performance of TPU/TPS blends are also described. FTIR, SEM and DSC were employed to study the interfacial interactions of the TPU/TPS blends. Dynamic mechanical analysis and tensile, impact and folding endurance tests were used to analyze the effects of TPS and POE content on the mechanical properties of TPU. Water adsorption, contact angle (CA) and degradation tests were performed to illuminate the relationship between the water sensitivity and degradation ability of the TPU/TPS blends.

Experiment

Materials

Corn starch (CS, amylose: 23–26 wt%; moisture: 12 wt%) was obtained from Wuhan Corn Starch Co. Ltd (Wuhan, China). Thermoplastic polyurethane (TPU), Utechllan UE-95A, was purchased from Bayer AG (Leverkusen, Germany). Polyolefin elastomers (POE) were obtained from Guangzhou Honest Chemical Co., Ltd (Guangzhou, China). Nitrile rubber (NBR) powder was purchased from Beijing Yudahang Industry & Trade Co., Ltd (Beijing, China). Disodium hydrogen phosphate (analytical grade, 99.5%) was purchased from China Specialty Products Co., Ltd (Jiangsu, China). Sodium dihydrogen phosphate (analytical grade, 99.5%) was purchased from Tianjin Fuchen Chemical Reagents Factory (Tianjin, China). Glycerol and other regents were of analytical grade (99.5%) and purchased from Longxi Chemical Reagent (Shantou, China). All the abovementioned materials were used as received.

Preparation of TPS and TPU/TPS blends

Prior to processing, dried corn starch, glycerol and water were mixed for 5 min in a mixer and then allowed to stand for 30 min at 80 °C. The ratio of corn starch : glycerol : water was 70 : 20 : 10. TPS was prepared by a thermal process using a twin-screw extruder (L/D = 40 : 1, Nanjing Chenmeng Machinery Co., Ltd, Jiangsu, China). The temperature profile along the extruder barrel was controlled to be 80–100–120–130–130–140–140–130 °C (from feed to die), respectively, with 100 rpm screw rotation speed; then the product was dried at 80 °C for 8 h.

The formulations of the TPU/TPS blends in this work are shown in Table 1. The mixtures of each formulation were fed into the same twin-screw extruder and extruded into strands, followed by pelletization. The temperature profile along the extruder barrel was controlled to be 120–140–145–150–150–150–150–150 °C (from feed to die), respectively. The screw rotation speed was 200 rpm. The products were dried at 80 °C for 4 h in a blast drying oven. Tensile strength (63.5 × 12.7 × 3.2 mm) and notched impact strength (127 × 13 × 4 mm) test specimens were injected (ARBUR 420M, Germany, ø = 25 mm, L/D = 28) with a 2-plate mold at 160–165–165–170–175 °C, respectively. The folding endurance (100 × 15 × 0.5 mm), dynamic mechanical thermal analyzer (DTMA) (30 × 6 × 0.5 mm) and water resistance (100 × 100 × 1.5 mm) test sheets were made using a hot press machine. Additionally, the specimens for the degradation tests were also obtained by compression-molding and were cut into 40 × 40 × 0.1 mm sized sheets.

Table 1 Composition and content of different samples

Sample	TPU	TPS	POE	NBR
TPU/20CS	80	20 (corn starch)	0	0
TPU/20TPS	80	20	0	0
TPU/10TPS/10POE	90	10	10	0
TPU/20TPS/10POE	80	20	10	0
TPU/30TPS/10POE	70	30	10	0
TPU/40TPS/10POE	60	40	10	0
TPU/20TPS/5POE	80	20	5	0
TPU/20TPS/15POE	80	20	15	0
TPU/20TPS/10NBR	80	20	0	10
TPU/20TPS/15NBR	80	20	0	15
TPU/20TPS/20NBR	80	20	0	20

---

Before the various characterizations, the extruded products were conditioned at 52% relative humidity (RH) for at least 10 days at ambient temperature in a closed chamber containing a Mg(NO₃)₂·6H₂O saturated solution of distilled water.²¹

Characterizations

Fourier transform infrared spectroscopy (FT-IR). FT-IR spectra were obtained from samples in potassium bromide pellets using a 170SX FTIR spectrophotometer (Nicolet, Madison, WI, USA).

Differential scanning calorimetry (DSC). DSC measurements were performed on a Netzsch DSC204 instrument under a nitrogen atmosphere. For every test, the sample was heated from −60 to 120 °C in the first scan, and the second heating scan was performed after cooling to −60 °C. Both the heating and cooling rates were 10 °C min⁻¹. The glass transition temperature (T_g) was recorded from the second heating curve to minimize the thermal history effect.

Scanning electron microscopy (SEM). The morphologies of the film cross-sections fractured under liquid nitrogen were investigated by scanning electron microscopy analysis (S-3700, Hitachi).

Mechanical analysis. After the samples were injection molded as standard bars, their tensile properties were measured on a Universal test machine (Reger, RGT-20A) according to ASTM D638-03. The notched impact strength was performed according to ASTM D256 using an impact tester (Sans ZBC1400-2). The folding endurance tests were performed using a folding endurance tester (PN-NZ135) according to ISO5626; the test temperature was 20 °C and the folding angle was 50° ± 1°. In all cases, a minimum of five specimens were tested to determine an average value.

Dynamic mechanical thermal analysis (DMA). The dynamic mechanical behavior of the specimens, maintained in a conditioning cabinet at 25 °C and 35% RH, was determined with a dynamic mechanical thermal analyzer (TA instrument DMA 2980-USA) with tensile mode at 1 Hz with a strain of 30 μm and a heating rate of 3 °C min⁻¹ in the temperature range from −70 to 100 °C.

Contact angle (CA) test. The surface energy was calculated by measuring the contact angle (CA) with a CA meter (Contact Angle System OCA20, Germany). The testing liquids were purified water and diiodomethane. Five microliters of testing liquid was deposited on the solid sheet surface. The CAs were measured with a CCD camera and processed using an image analysis video card which calculated the CAs automatically using an image analysis setup. The average values of five measurements were taken. The surface energy was calculated according to eqn (1)–(3):

γ_s = γ^D_s + γ^P_s (3)

where γ_w is the surface energy of water, and γ^D_w and γ^P_w are its dispersion force and its polar force, respectively. γ_d is the surface energy of diiodomethane, and γ^D_d and γ^P_d are its dispersion force and polar force, respectively. γ_s is the surface energy of the solid and θ is the CA.

Water resistance tests. The samples were dried at 100 °C for 24 h, cooled to 25 °C in a desiccator, then weighed immediately (M₀). All the samples were immersed in deionised water for 72 h. They were then removed at specific intervals and gently blotted with tissue paper to remove the excess of water on the surface. Finally, the samples were weighed (M_x). Five specimens were tested to determine an average value. The amount of water absorbed by the samples was calculated by eqn (4):

Degradation properties. First, 0.2 mol L⁻¹ Na₂HPO₄ solution and 0.3 mol L⁻¹ NaH₂PO₄ solution were prepared. Then, the solutions were combined to form a degradation solution of pH 7.4 with a formulation of Na₂HPO₄ : NaH₂PO₄ = 81 : 19. Before the tests, the degradation samples were dried at 100 °C for 36 h, cooled to 25 °C in a desiccator, and then weighed immediately (M₀). Finally, the samples were immersed in the degradation solution for seven weeks. Each week they were removed from the solution, any water adhering to their surface was removed, and they were weighed quickly to obtain M_x. The weight loss ratios of the samples were calculated by eqn (5):where M_x is the weight of sample x at time t in the degradation solution.

Results and discussion

Interfacial interaction of TPU/TPS blends

Fourier transform infrared spectroscopy (FT-IR) is a useful tool to characterize the changes in functional groups during chemical reactions. In this segment, the FT-IR spectra of TPS, TPU and the blends are presented in Fig. 1.

Fig. 1 FT-IR spectra of (1) CS, (2) TPS, (3) TPU, (4) TPU/20CS, (5) TPU/20TPS, (6) TPU/20TPS/10POE, (7) TPU/20TPS/5POE, (8) TPU/20TPS/15POE, and (9) TPU/20TPS/10NBR.

In curves 1 and 2, both CS and TPS had a characteristic peak around 1645 cm⁻¹ which was attributed to ν-OH.²¹ Another band at 2927 cm⁻¹ was attributed to the asymmetric stretching vibration of –CH₂– in the glucose unit.²² As for TPU (presented in curve 3), the peak at around 3334 cm⁻¹ was assigned to the inter- and intra-molecular hydrogen bonds of the NH groups (ν-NH, H-bonded) in the urethane carbonyl groups. Additionally, two peaks present at 1732 cm⁻¹ and 1701 cm⁻¹ were assigned to the stretching of the free carbonyl groups (ν-C=O, free) and the hydrogen bonded carbonyl groups (ν-C=O, H-bonded),²² respectively. Comparing the FT-IR spectra of TPU and TPU/TPS, it was found that the peak intensity at 1702 cm⁻¹ of TPU/20TPS was stronger than that of TPU. This was attributed to the formation of hydrogen bonds between TPS and TPU. When POE was added to the TPU/20TPS blend, the peak at 3334 cm⁻¹ was weaker and broader than that of the TPU/20TPS blend. Furthermore, the hydrogen bond band shifted to 1706 cm⁻¹, and a slight decrease in peak intensity was observed. Compared with TPU and TPS, the flexibility of the POE molecular chain was greater, and it acted as a compatibilizer in the TPU/20TPS blend. As a result, the inter- and intra-molecular hydrogen bonds of the NH groups and OH group decreased, respectively. Therefore, POE improved the compatibility between TPU and TPS, which is attributed to the increase of the inter-hydrogen bond bands of TPU and TPS. Curve 6, 7 and 8 show that the peak intensities at 3334, 1732 and 1706 cm⁻¹ decreased with increasing POE content, respectively. It is possible that POE could act as a soft segment in the TPU/TPS blends and dilute the intensity of the hydrogen bonds. Moreover, the addition of POE would shield the hydrogen bonds. These two reasons were attributed to the decrease of peak intensity at 1706 cm⁻¹. However, when NBR was mixed with TPU/20TPS, the peak at 3321 cm⁻¹ in curve 9 was sharper than that of TPU/20TPS/10POE. This indicates that the hydrogen bonds formed by the NH groups in the hard segment of TPU increased, and the addition of NBR corresponded to the decrease in miscibility of TPU and TPS.

Morphology

In this segment, the influence of TPS and POE loadings on the micro structures was characterized by scanning electron microscopy (SEM). The morphologies of the fractured surfaces of the TPU, TPU/CS and TPU/TPS samples are presented in Fig. 2.

Fig. 2 SEM images (2.0k×) of (A) TPU; (B) TPU/20CS; (C) TPU/20TPS; (D) TPU/20TPS/5POE; (E) TPU/20TPS/10POE; (F) TPU/20TPS/15POE; (G) TPU/30TPS/10POE; (H) TPU/40TPS/10POE; and (I) TPU/20TPS/10NBR.

It can be seen that TPU displayed a continuous and smooth fractured surface with slight coarseness, as shown in Fig. 2(A), which was attributed to the microphase segregation that occurred between the hard and soft segments.¹² Fig. 2(B) shows that significant phase separation occurred at the surface of TPU/20C. Because the corn starch could not achieve thermoplastic properties during the thermal process, it dispersed in the TPU matrix in particle form. A clear interface and gaps were found between the corn starch particles and the TPU matrix, indicating that the miscibility of TPU/20CS was poor. The results were also confirmed by the FTIR spectra (Fig. 1, curve 4; there was no bond formed between TPU and CS). When the corn starch was plasticized by glycerol in the thermal process, the corn starch granules were destroyed and TPS displayed a homogenous morphology.¹³ In Fig. 2(C), no starch granules could be found in the fractured surface, indicating that the TPS was well embedded into the TPU matrix with some cracks. When compared with the surface of TPU/20CS, it could be concluded that the TPS was better dispersed in the TPU matrix, and the compatibility and interfacial adhesion between the TPU and starch phases improved.²³

Fig. 2(D)–(F) present a miscible morphology; no cracks appeared in the TPU/TPS blend after the incorporation of different amounts of POE. Moreover, the TPS was well dispersed in the TPU matrix and the “sea–sea” structure was formed, which was favorable for the delivery and dispersion of stress.²⁴ Fig. 2(G) reveals that no distinct interface was visible between the TPU and TPS phases, even when the TPS content was as high as 30 wt%. The morphology of TPU/30TPS/10POE indicated that TPS could disperse well and embed in the TPU matrix. It was difficult to find cracks in the boundary of TPS/TPU. However, TPU/40TPS/10POE presented a deformed morphology with TPS aggregation, as shown in Fig. 2(H). This was attributed to the decrease of compatibility and interfacial adhesion between the TPU and starch phases with increasing TPS content. The morphology of TPU/20TPS/10NBR (I) was disconnected and the TPU matrix was highly deformed. The spherical domains of the dispersed TPS phase were large, and the interface between TPU and TPS was smooth and clear. This indicated a poor interfacial adhesion between the two phases. The reason was that the addition of NBR decreased the miscibility of the TPU/TPS blends.

Thermal analysis

DSC is an effective tool to investigate the effect of compatibilizer on the miscibility of blends. The DSC thermograms of the TPU, TPS and TPU/TPU blend systems are shown in Fig. 3, and the T_g values are summarized in Table 2. The glass transition temperatures (T_g) of TPU and TPS are −33.6 °C and 82.2 °C, respectively. TPU/CS only displayed one T_g at 34.1 °C because the CS could not show the thermoplastic properties. Two distinct T_g values appeared at −28.9 °C and 79.6 °C for TPU/TPS; these values are close to those of TPU and TPS. This indicated the miscibility of the two components and coincided with the results of FTIR and SEM. When POE was added to the TPU/20TPS blend, the two T_g values corresponding to the blend components shifted toward each other. Therefore, the compatibility of TPU and TPS was enhanced through the interpenetration of TPU and TPS at the interface.²⁵ The reason was that the addition of POE could cause a decrease in the interfacial tension of TPU/TPS and an increase in the thickness of the interface of TPS/TPU.²⁶ When NBR was added to the TPU/20TPS blend, the two T_g values were −36.4 °C and 85.9 °C; the difference is greater when compared with those of TPU/TPS. This result indicated that the miscibility of the TPU and TPS components could not be improved by NBR.

Fig. 3 DSC thermograms of TPU and TPU/TPS blends. (1) TPU; (2) TPS; (3) TPU/20CS; (4) TPU/20TPS; (5) TPU/20TPS/10POE; and (6) TPU/20TPS/10NBR.

Table 2 T_g values of TPU, TPS and TPU/TPS blends

Sample	TPU	TPS	TPU/20CS	TPU/20TPS	TPU/20TPS/10POE	TPU/20TPS/10NBR
Tg1 (°C)	−33.6	—	−34.1	−28.9	−23.5	−36.4
Tg2 (°C)	—	82.2	—	79.6	69	85.9

---

The FTIR, SEM and DSC results demonstrated that the addition of POE enhanced the miscibility of TPU and TPS through the interpenetration of TPU and TPS.

Mechanical properties

The effects of TPS content and compatibilizer on the mechanical properties of the TPU/TPS blends were studied by tensile, impact and folding endurance tests; the data are listed in Table 3. First of all, TPU exhibited excellent mechanical performance. Its notched impact strength was too high to determine; also, the folding number could be higher than 30.0 × 10³. When corn starch was blended with TPU, a decrease in the mechanical properties was observed. The notched impact strength and folding number of TPU/20CS drastically decreased to 376.3 J m⁻¹ and 0.58 × 10³, respectively. The deterioration in the mechanical properties of TPU/20CS were caused by the poor compatibility between TPU and corn starch. Compared with those of TPU/20CS, the notched impact strength and folding number of TPU/20TPS increased significantly; the values were 872.5 J m⁻¹ and 4.5 × 10³, respectively. This was attributed to the increase in compatibility and interfacial adhesion between TPU and starch, though the values were far lower than those of TPU.

Table 3 Notched impact strength, tensile strength, elongation at break and folding endurance for samples

Sample	Notched impact strength (J m^−2)	Elongation at break (%)	Tensile strength (MPa)	Folding endurance (times (×10^3))
TPU	No break	>800	12.5	>30.0
TPU/20CS	376.3	575.6	12.2	0.58
TPU/20TPS	872.5	>800	12.2	4.5
TPU/10TPS/10POE	No break	>800	11.5	>30.0
TPU/20TPS/10POE	No break	>800	11.2	>30.0
TPU/30TPS/10POE	587.3	>800	9.5	16.0
TPU/40TPS/10POE	432.1	>800	8.5	8.6
TPU/20TPS/5POE	No break	>800	11.2	15.0
TPU/20TPS/15POE	No break	>800	11.0	>30.0
TPU/20TPS/10NBR	575.4	422.3	8.2	1.7
TPU/20TPS/15NBR	632.3	434.5	8.6	1.5
TPU/20TPS/20NBR	786.4	572.1	9.0	0.8

---

The introduction of POE could play an important role in increasing the flexibility of TPU/TPS. This corresponded to the increase of the interfacial force of TPS and TPU and increased the mechanical properties of the TPU/TPS blends. The notched impact strength of TPU/TPS/10POE depended on the TPS loading. With increasing TPS content, the mechanical properties of the blends decreased. When the TPS content was lower than 30%, the notched impact strength was too high to be measured. Meanwhile, when the loading of TPS reached 30%, the mechanical properties of the blends deteriorated. The interfacial adhesion between the TPU and starch phases decreased as the TPS content increased. As a result, the overloading of TPS reduced the flexibility of TPU and lowered the folding endurance due to the brittleness of TPS. However, it was clear that the flexibility of TPU/TPS greatly improved after it was composited with POE. The flexibility of TPU/20TPS/10POE and TPU/20TPS/15POE increased, and their notched impact strengths, elongations and folding numbers were comparable to those of TPU. Therefore, TPU/20TPS/10POE was considered to be a potential material to substitute for TPU. However, when NBR was added, the mechanical properties of all the TPU/20TPS/NBR blends decreased. Their folding number was reduced from 1.7 × 10³ to 0.8 × 10³ when the NBR content ranged from 10% to 20%. It was revealed that the addition of NBR to TPU decreased the flexibility. The reason was that NBR exhibited greater rigidity than the soft segments in TPU. Furthermore, NBR could not achieve thermoplastic properties, which restricted it to reducing the interfacial tension through the interpenetration of TPU and TPS at the interface. This corresponded to the drastic reduction in folding endurance and notched impact strength. Moreover, according to the FTIR, SEM and DSC results, incorporating NBR led to a weakening in the interfacial force of TPU and TPS, resulting from the severe phase separation, and caused a deterioration in the mechanical properties of TPU/20TPS.

Dynamic mechanical thermal analysis (DMA)

Dynamic mechanical thermal analysis (DMA) measures mechanical properties (storage modulus, loss modulus and tan σ) as a function of temperature. In this segment, Fig. 4 describes the DMA analysis results of the TPU/TPS blends. Fig. 4(a) presents the storage modulus (E′) vs. the temperature curve of the TPU/TPS blends. A plateau was observed when the temperature ranged from −60 °C to −30 °C. The E′ of TPU suffered a sharp decrease around its glass transition temperature (T_g), which was attributed to the thermoplastic nature of TPU.²⁷ Because the TPU resin was dominated by the amorphous part, it could not achieve a rubbery plateau when heated from T_g to 80 °C. The E′ of TPU decreased to a rather low value when the temperature increased to 100 °C. It was noted that when TPS was introduced into the TPU matrix, the storage modulus of the TPU/TPS blends increased significantly at the glass stage. A rubbery plateau could not be observed in the temperature range from 10 °C to 80 °C; however, the E′ of TPU/TPS achieved a rather stable value when the temperature ranged from 80 °C to 100 °C. This behavior was attributed to the stiffness and the amorphous domain of TPS. When POE was added to the TPU/TPS blends, the E′ of TPU/20TPS/10POE was comparable to that of TPU/20TPS at lower temperature. This result was attributed to the stiffness of TPS dominating the E′ of the blends at low temperature. The storage modulus decreased significantly when the temperature ranged from −20 °C to 100 °C. The penetration degree of POE at the interface of TPU and TPS increased with increasing temperature. As a result, the soft compatibilizer could increase the compatibility and interfacial adhesion and decrease the E′ of TPU/TPS; however, Fig. 4(a) also reveals that the E′ of TPU/TPS decreased when the TPS content increased from 10% to 40% below −30 °C when POE was fixed at 10%. Fig. 4(b) shows the tan δ curves of the TPU/TPS blends. This figure reveals that TPU achieved T_g at the tan δ peak for −3.1 °C; this was attributed to the soft segment of TPU. There were two T_g values in TPU/20TPS; T_g1 shifted to a lower temperature (about −11.4 °C) than that of TPU, while T_g2 was about 64.5 °C (attributed to TPS). After the incorporation of POE, T_g1 and T_g2 of TPU/20TPS/10POE increased to −6.48 °C and 71.5 °C, compared with TPU/20TPS. The increase in the T_g of the blends resulted from the penetration of the TPS into the TPU domains, promoted by POE.^25,26 As a result, the TPS could restrict the molecular motion of the soft segment of TPU, and the hard segment could retard the molecular motion of TPS. The introduction of POE can increase the compatibility and interfacial adhesion and increase the formation of inter-hydrogen bonds between TPU and TPS. When the TPS content ranged from 20% to 40%, there was no significant difference in the tan δ curves of TPU/TPS/10POE. This was also attributed to POE acting as a soft compatibilizer and improving the miscibility of TPU and TPS.

Fig. 4 DMA curves of TPU/TPS/POE with various TPS content, (a) storage modulus (E′) and (b) dissipation factor (tan δ) as a function of temperature: (1) TPU, (2) TPU/20TPS, (3) TPU/20TPS/10POE, (4) TPU/30TPS/10POE, (5) TPU/40TPS/10POE.

Water resistance

One of the fatal defects of TPS is that it imbibes water at high humidity, which decreases the mechanical properties of TPU/TPS blends.²⁸ In this segment, the effect of TPS content on the hydrophobic abilities of the TPU/TPS/10POE blends was evaluated by contact angle (CA) tests. In order to study the hydrophilicity of TPU/TPS in detail, the CA and surface energy (SE) of the TPU/TPS/10POE blends were measured. Fig. 5 presents the effects of TPS content on the CA of the blends, and the data are listed in Table 4.

Fig. 5 Dependence of water contract angle on TPS content for TPU/TPS blends. (A) TPU, (B) TPU/20TPS, (C) TPU/10TPS/10POE, (D) TPU/20TPS/10POE, (E) TPU/30TPS/10POE, and (F) TPU/40TPS/10POE.

Table 4 Effect of TPS content on water adsorption and contact angle test of TPU

Samples	Water adsorption (%)	Contact angle (°)	Surface energy (nN m^−1)
TPU	0.53	85.6	42.5
TPU/20TPS	10.3	57.1	44.5
TPU/10TPS/10POE	3.5	79.6	31.4
TPU/20TPS/10POE	6.5	77.2	33.6
TPU/30TPS/10POE	13.1	71.6	36.1
TPU/40TPS/10POE	15.7	65.6	38.8

---

Compared with that of TPU, the water CA of TPU/20TPS decreased significantly from 85.6° to 57.1°, while its SE increased from 42.5 to 44.5 nN m⁻¹. The presence of polar OH groups (from both starch and glycerol) increased the hydrogen bonding interactions; therefore, the hydrophilicity of TPU/TPS increased. When POE was added, the CA of TPU/20TPS/10POE increased, while its SE decreased. The addition of POE could improve the miscibility of TPU and TPS through enhancing the interpenetration of TPS into the TPU matrix.²⁶ Some of the hydroxyl groups in the starch would be shielded, resulting in an increment of the CA. When POE was added to the TPU/TPS blends, it caused a decrease in the interfacial tension of TPU/TPS, which corresponded to the decrease of SE. The FTIR, SEM and DSC results revealed that POE could eliminate the interfacial cracks of the TPU/TPS blends and shield the inter-hydrogen bonds of TPU and TPS. When the TPS content increased, the CA of TPU/TPS/10POE decreased to 65.6° and the SE increased to 38.7. This result could also be ascribed to the introduction of higher amounts of hydroxyl groups from starch and glycerol into the TPU, which led to an increase in hydrophilicity. The increase in surface energy can be explained by the introduction of polar functional groups. Variations in surface chemical composition and surface structure resulted in the varied CA of TPU/TPS/10POE.

Fig. 6 shows the water absorptions of the blends at different times after the samples were submerged in water for 72 h. TPU exhibited good hydrophobicity, and its WA at 72 h was 0.6%. The WA of TPU/20TPS increased to 12.5% due to the hydrophilicity of TPS. When POE was added to TPU/20TPS, the WA of TPU/20TPS/10POE decreased to 8.3%. With increasing TPS content, the water absorption ratio of the TPU/TPS/10POE blends increased from 4.8% to 17.0%. Fig. 6 displays the water absorption of the blends with different TPS contents after 72 h of submersion in water. The dashed line is the theoretical value of equilibrium water uptake; WA (theory) can be calculated by eqn (6) (water absorption of POE was neglected):

WA (theory) = ø_TPS·WA_TPS + ø_TPU·WA_TPU (6)

where WA and ø are the equilibrium water absorption and mass fraction in the blends, respectively. The WA of TPS was 50.5% according to the moisture test, because TPS dissolves in water.²⁶ By eqn (6) and Fig. 7, it can be seen that the test WA values of the TPU/TPS blends were lower than the theoretical values. In the case of the TPU/TPS/POE blends, TPS dispersed well in the TPU matrix, the interfacial tension of TPU/TPS decreased and cracks at the interface were eliminated by POE. As a result, TPS was isolated from water by TPU and POE, which retarded the water uptake ability of TPS. Moreover, due to shielding of the hydrogen bonding interactions between TPU and TPS¹³ by POE, the CA of TPU/20TPS/10POE increased and led to a decrease in the water absorption ability.

Fig. 6 Dependence of water absorption on TPS content for TPU/TPS blends. (1) TPU, (2) TPU/20TPS, (3) TPU/10TPS/10POE, (4) TPU/20TPS/10POE, (5) TPU/30TPS/10POE, (6) TPU/40TPS/10POE.

Fig. 7 Dependence of water absorption after submersion in water for 72 h on TPS content for TPU/TPS/10POE blends: (1) theoretical value; (2) measured values.

Degradation properties

Fig. 8 shows the weight loss of the TPU and TPU/TPS/10POE blends in degradation solution over 7 weeks. The weight loss of TPU in 7 weeks was only 0.887%, which meant that TPU did not readily degrade under the test conditions. It has been reported that the biodegradable properties of TPU are dependent on its hard segment.¹⁸ However, the hard segment of TPU mostly consists of aromatic groups, which are difficult to hydrolyze. As a result, TPU is only slightly affected by hydrolysis and does not demonstrate biodegradable properties. In pH 7.4 solution, starch can be hydrolyzed at the ether linkages on its polysaccharide chains;²⁷ thus, starch is sensitive to hydrolysis. Fig. 8 shows that the degradation rate of TPU/TPS/10POE increased significantly compared with that of TPU, and the degradation rate was dependent on the TPS content. The weight losses of TPU/30TPS/10POE and TPU/40TPS/10POE after 7 weeks reached 9.8% and 11.3%, respectively. No plateaus appeared in any the curves within the test period, indicating that the biodegradation was not complete; it could be expected to reach completion with prolonged tests and if powder were used instead of sheets. With increasing TPS content, the morphology of TPU/TPS/POE (10%) exhibited a “sea-island” construct and holes in the interface. Moreover, the hydrophobicity of TPU/TPS/POE (10%) decreased, which resulted in a decreased CA. This meant that the pH 7.4 solution could easily make contact with the starch, which improved the hydrolysis rate of the blend. Moreover, the possible biodegradability due to starch hydrolysis would induce spontaneous attack and adhesion of fungi on the TPU/TPS blends.^18,29 This result demonstrated that TPS improved the degradation rate of TPU/TPS, and this improvement could be controlled by varying the TPS content. To achieve the systemic biodegradation of TPU/TPS blends, it is expected that the test time should be prolonged. Due to their biodegradable properties, these TPU/TPS blends are favorable for fabricating products such as sports shoes, disposable diapers, sanitary towels, and sheets used for agricultural purposes. They could degrade in the natural environment once they were used up and discarded. The motive for the development of biodegradable TPU/TPS blends lies in their potential to contribute to the decrease of modern waste.

Fig. 8 Effect of TPS content on weight loss ratio at different weeks of TPU/TPS/10POE blends: (1) TPU, (2) TPU/10TPS/10POE, (3) TPU/20TPS/10POE, (4) TPU/30TPS/10POE, and (5) TPU/40TPS/10POE.

Conclusions

In this work, TPU/TPS blends were prepared to improve the degradation ability of TPU and decrease its price. To achieve good mechanical properties, POE was chosen to be the optimized compatibilizer. The FTIR, SEM and DSC results demonstrated that compared with TPU/TPS, the TPU/TPS/10POE blends exhibited improved physical properties. This was attributed to the fact that POE could enhance the interfacial effects of the TPU/TPS blends, improve their miscibility and decrease the interfacial tension between TPS and TPU. As a result, the TPU/TPS blends retained good mechanical properties, while their degradable properties improved greatly. In addition, the mechanical properties of the TPU/TPS/10POE blends were strongly influenced by the TPS content. The hydrophilicity and hydrolysis properties of starch in pH 7.4 solution could induce the TPU/TPS blends to degrade and increase the biodegradable properties of the TPU/TPS blends. In summary, incorporating TPS with 20 wt% and 10 wt% POE in TPU resulted in blends with good folding endurance (>3.0 × 10⁴), notched impact strength (no break) and elongation at break (>800%), as well as a 6.2% biodegradation rate over 7 weeks. The improved properties of these TPU/TPS blends show that they have high potential for use in environmentally friendly materials.

Acknowledgements

The authors wish to acknowledge the financial support of the National Science Fund for 50903032 and 11172105, Fund Research Grant for Science and Technology in Guangzhou, China (2014J4100038), Fundamental Research Funds for the Central Universities (2015ZZ020) and the Opening Project of The Key Laboratory of Polymer Processing Engineering, Ministry of Education, China.

Notes and references

1. C. Hepburn, Polyurethane elastomers, Elsevier Science Publishers, London, 2nd edn, 1992.
2. B. Finnigan, D. Martin, P. Halley, R. Truss and K. Campbell, Polymer, 2004, 45, 2249–2260.
3. A. Pattanayak and S. C. Jana, Polym. Eng. Sci., 2005, 45, 1532–1539.
4. A. Pattanayak and S. C. Jana, Polymer, 2005, 46, 5183–5193.
5. C. H. Dan, M. H. Lee, Y. D. Kim, B. H. Min and J. H. Kim, Polymer, 2006, 47, 6718–6730.
6. B. Jaúregui-Beloqui, J. C. Fernández-García, A. C. Orgilés-Barceló, M. M. Mahiques-Bujanda and J. M. Martín-Martínez, J. Adhes. Sci. Technol., 1999, 13, 695–711.
7. D. Baral, P. P. De and G. B. Nando, Polym. Degrad. Stab., 1999, 65, 47–51.
8. U. A. Pinto, L. L. Y. Visconte and R. C. R. Nunes, Eur. Polym. J., 2007, 37, 1935–1937.
9. F. Chavarria and D. R. Paul, Polymer, 2006, 47, 7760–7773.
10. X. F. Ma and J. G. Yu, J. Appl. Polym. Sci., 2004, 93, 1769–1773.
11. S. Ai, T. Is and T. Ei, Russ. Chem. Rev., 2000, 69, 451–459.
12. Q. X. Wu and L. N. Zhang, Ind. Eng. Chem. Res., 2001, 40, 558–564.
13. Y. S. Lu, L. Tighzert, P. Dole and D. Erre, Polymer, 2005, 46, 9863–9870.
14. M. Barikani and M. Mohammadi, Carbohydr. Polym., 2007, 68, 773–780.
15. Q. X. Wu, Z. S. Wu, H. F. Tian, Y. Zhang and S. L. Cai, Ind. Eng. Chem. Res., 2008, 47, 9896–9902.
16. Y. M. Leng, Y. Zhang, X. X. Chen, C. Yi, B. B. Fan and Q. X. Wu, Ind. Eng. Chem. Res., 2011, 50, 11130–11135.
17. Y. Zhang, Y. M. Leng, M. Zhu, B. B. Fan, R. X. Yan and Q. X. Wu, Carbohydr. Polym., 2012, 88, 1208–1213.
18. T. Seidenstückert and H. G. Fritz, Polym. Degrad. Stab., 1998, 59, 279–285.
19. T. Seidenstücker and H. G. Fritz, Starch/Staerke, 1999, 51, 93–102.
20. H. T. Chiu and C. Y. Chuang, J. Appl. Polym. Sci., 2010, 115, 1278–1282.
21. J. M. Fang, P. A. Fowler, J. Tomkinson and C. A. S. Hill, Carbohydr. Polym., 2002, 47, 245–252.
22. J. L. Sormana and J. C. Meredith, Macromolecules, 2004, 37, 2186–2195.
23. B. Y. Shin, S. I. Lee, Y. S. Shin, S. Balakrishnan and R. Narayan, Polym. Eng. Sci., 2004, 44, 1429–1438.
24. F. X. Yang, R. N. Du, B. Li, L. Chen, J. H. Yang, Q. Fu and Q. Zhang, Acta Polym. Sin., 2010, 3, 366–371.
25. S. Q. Xu, B. Chen, T. Tang and B. T. Huang, Polymer, 1999, 40, 3399–3406.
26. H. M. Jeong, B. K. Ahn and B. K. Kim, Eur. Polym. J., 2001, 37, 2245–2252.
27. X. L. Yao, X. D. Qi, Y. L. He, D. S. Tan, F. Chen and Q. Fu, ACS Appl. Mater. Interfaces, 2014, 6, 2497–2507.
28. S. D. Zhang, Y. R. Zhang, X. L. Zhang and Y. Z. Wang, Starch/Staerke, 2009, 61, 646–655.
29. A. J. F. Carvalho, M. D. Zambon and A. A. D. Curvelo, Carbohyd. Polym, 2005, 62, 387–390.

---