1 Introduction

The used of plastics as packaging materials has been used since nineteenth century and becomes the main packaging materials for food and non-food applications due to low cost production, easily processable, light-weight and good mechanical properties. However, since plastic is durable and degrade very slowly, plastic wastes have become one of environmental issues. Since 1950, tons of plastic have been discarded and may persist for hundreds or even thousands of years. Subsequently, waste reduction is important for the future and environment, therefore, biodegradable plastics materials are good alternatives to solve these problems. The development of biodegradable plastics started more than 20 years ago. The first generation products were conventional plastics blended with starch causing slow disintegration of the products in nature while second generation products are emerging based on totally biodegradable polymer derived from synthetic or natural materials or blends of both (Groh et al. 2019). The adoption of biodegradable polymeric materials is increasing in food and consumer goods packaging application due to concern about the disposal and cost of petroleum-based polymers. Biodegradable plastics can be made from fully biodegradable polymer with hydrolysable backbones such as polylactic acid (PLA), polycaprolactone (PCL), polyamides, polyurethanes and polyureas, also, poly (vinyl alcohol) and poly (vinyl acetate) since their chemical structure have been found to be susceptible to biodegradation (Faludi et al. 2013; Yee et al. 2011; Seggiani et al. 2018). However, these biodegradable plastic are expensive and difficult to process using blown film technique. The development of degradable plastic packaging from polyolefin and natural biodegradable polymer of polysaccharides such as starch is the key factor to reduce the products cost and enhances biodegradation properties. Today, there is a number of polyethylene grades relevance to packaging but the most widely used is low-density polyethylene (LDPE). Starch can be easily incorporated into LDPE film to impart biodegradation. Various starches can be incorporated into LDPE such as potato starch, tapioca starch, cassava starch and corn starch in order to enhance the biodegradability of LDPE. Previous study showed a reduction in mechanical properties such as tensile strength and elongation at break after the addition of starch into LDPE. Therefore, mechanical properties limit the usage of unmodified starch in polyolefin (Nawang et al. 2001; Gonçalves de Moura et al. 2017). The solution of these drawbacks of incorporating starch into polyolefin, compatibility between the two components needs to be improved. Starch and LDPE has low compatibility due to difference in polarity. The hydrophilic structure of starch is not compatible with highly hydrophobic criteria of LDPE. There are two ways to overcome this problem, whether by modify the starch before blending or by adding compatibilizer such as ethylene-maleic anhydride (MAH), ethylene vinyl alcohol (EVA) and maleic anhydride grafted PE (PE-g-MAH) (Huang et al. 2005; Khalf et al. 2010). PE-g-MAH is considered the most effective compatibilizer between starch and LDPE (Nakamura et al. 2005). PE-g-MAH is used to enhance the compatibility between LDPE and tapioca starch besides improves mechanical, morphology, thermal and rheology properties (Nakamura et al. 2005; Wang et al. 2005). Even though there was a number of studies reported on the using of LDPE-co-glycidyl methacrylate copolymer, LDPE-g-dibutyl maleated (LDPE-g-DBM) and LDPE-g-itaconic acid as compatibilizer (Sailaja and Seetharamu 2008; Sailaja 2005; Girija and Sailaja 2006). However, these compatibilizers contribute to higher cost of manufacturer compare to PE-g-MAH. Commercially, filled polymers are becoming very important due to high stiffness, strength, reinforcement and heat distortion temperature over the unfilled ones (Sobhanie et al. 1997). Calcium carbonate, CaCO3 is one of the most used filler in polyolefin, which is considered as inert filler chemically. CaCO3 was traditionally used in plastic as bulking agents to substitute the expensive polymers. CaCO3 particle of an aspect ratio close to unity are expected to modify the viscosity of the polymer melts. Their good thermal conductivity contributes to the homogeneity of the melt and good dispersion in the polymer matrix (Gorna et al. 2008). CaCO3 is hydrophilic, which often modified to provide a hydrophobic surface to increase adhesion with hydrophobic polymer matrix. Fatty acids such as stearic acid are commonly used as coating agents for CaCO3 particles. The incorporation of CaCO3 filler into PE not only lower the cost but also improve the tensile strength, stiffness, thermal property and shrinkage. Recently, most researchers have expanded their studies on producing fully biodegradable polymer for plastic packaging. In order to be more economic, the incorporation of starch into LDPE to produce biodegradable plastic packaging film was conducted using hot pressed melt (Arvanitoyannis et al. 1998a), compression molding (Nawang et al. 2001), two roll mills (Kiatkamjornwong et al. 2001), extrusion (Huang et al. 2005), and melt blending (Sailaja and Seetharamu 2008), to produce the LDPE/starch blends. Nawang et al. (2001) used sago starch filled linear low-density polyethylene (LLDPE) to study their mechanical properties. Prior to analysis, the blended samples were compression molded on a compression molding machine at 150 °C for 15 min (Nawang et al. 2001). Kiatkamjornwong et al. (2001) used cassava starch to produce plastic sheet, blended it with LDPE using two roll mill at 165–1700 °C for 20 min while, Huang et al. (2005), used tapioca starch and mentioned that the compounded sample have been prepared using single screw extruder with controlled at four step temperature of 125,130,135 and 1300 °C (Huang et al. 2005; Kiatkamjornwong et al. 2001).

2 Preparation and Testing

2.1 Materials

Low density polyethylene (LDPE) was supplied by Polyolefin Company, which was used as polymer matrix. The properties of LDPE are: density of 0.923 g/cm3, with melt flow rate (MFR) value of 5 g/10 min, stiffness of about 240 MPa, tensile strength at break of 13.7 MPa and melting point of 111 °C. Calcium Carbonate (CaCO3) was supplied by Imerys Minerals Malaysia Sdn Bhd with the trade name Imercarb ®1 T. Typical properties of Imercarb ®1 T include a specific gravity of 2.7, Mohs hardness of 3 and moisture contain about 0.2%. Tapioca starch food grade was obtained from local vendor with purity degree of 99.4% (dry weight basis) and moister content of 12.6%. Polyethylene-g-Maleic Anhydride (PE-g-MAH) was obtained from The Dow Chemical Company with trade name Amplify™ GR 207 and used as hydrophilic-hydrophobic compatibilizer. Palm cooking oil was used as processing aid is product of Lam Soon Edible Oil. PCO was used to enhance the lubricating effect by decreased the resistance to flow in the extruder hence reduced the melt viscosity of the mixing blend (Min et al. 2008).

2.2 Preparation of LDPE/TS/CaCO3 Films

The LDPE/TS/CaCO3 dry blending formulations of the samples are shown in Table 1. The loading of tapioca starch (TS) remains unchanged in the formulations while the content of CaCO3 was changed throughout the compounding. The percentage of PE-g-MAH and Palm Cooking Oil (PCO) were kept constant as 20 and 15 wt.% respectively. The composition of LDPE/TS/CaCO3 blends was compounded using twin-screw extruder at screw speed 80 rpm. The extrusion process temperatures were 170, 180, 190 and 200 °C for feed, compression, metering and die zone respectively. Finally, the blends were blowing into the film sheets using blown film extrusion machine at screw speed of 500 rpm. The processing temperature of the blown film extrusion machine were 150, 137, 160, 160, 140, 130 °C for die head heater 1, die head heater 2, bend heater, extruding drum heater 3, extruding drum heater 2, extruding drum heater 1 respectively.

Table 1 Samples formulation

2.3 Characterization Tests

2.3.1 Melt Flow Index (MFI) Measurement

Melt flow index measurement was conducted to examine the flow property of the blended formulation before processing. MFI value was determined according to ASTM D1238 method using Zwick 4103 Extrusion Plastometer. The mass of extruded material was determined and the results were expressed as g/10 min.

2.3.2 Fourier Transform Infrared Spectroscopy (FTIR) Analysis

LDPE/TS, CaCO3, LDPE/TS/C5 and LDPE/TS/C10 were characterized using FTIR-Perkin Elmer spectrum. The test was conducted according to ASTM E1252–98 (2007). Approximately, 3 to 10 mg of sample was mixed with dried KBr powder and placed in boron carbide mortar, ground to a particle size less than 2-μm diameter. The mixture then transfers to a manual compression device to prepare a disc of uniform thickness of about 1 mm. The samples were scanned 15 times and FTIR spectra were recorded in wavelength range of 4000 to 400 cm−1.

2.3.3 Differential Scanning Calorimeter (DSC) Analysis

DSC scanning was conducted according ASTM D3418 using Perkin Elmer, DSC 7 machine. DSC measurements were performed under controlled nitrogen gas flow with heating and cooling rate of 10 °C /min, temperature of 30–150 °C, and samples weight ranged from 5 to 10 mg.

2.4 Mechanical Tests

The tensile strength, elastic modulus and elongation at break were determined according to ASTM D882 method using Llyord Instrument LRX 2.5 KN testing machine. The films were cut out into uniform shape according to ASTM D882–02. The test was performed in both machine direction and cross direction at an ambient temperature with a crosshead speed of 50 mm/min. Five measurements were taken for each sample and the average value was reported. The tearing resistance test, also commonly called the Elmendorf tear test, was used to measure the internal tearing resistance of the film. Tear strength of the films were investigated using HT-8181 Elmendorf Tearing Strength Tester according to ASTM D1922. Five measurements were taken for each sample in both the machine and cross direction and the average value was reported.

2.5 Water Absorption Test

Water absorption test is important for packaging materials in order to prevent the food from environmental moisture, which can contribute to the growth of microorganism. Therefore, the water absorption of LDPE, LDPE/TS, LDPE/TS/C5 and LDPE/TS/C10 was conducted according to ASTM D 570–98 at 25 °C. The samples were cut into bars form of 76.2 mm long and 25.4 mm wide. Each sample was weighed prior to be immersed in distilled water. Subsequently, the samples were wiped, and reweighed. Results were recorded every 3 days for 30 days. Three samples were used for each formulation and the average value was reported. The percentage of water absorption was calculated according to eq. 1 at room temperature.

$$ {M}_t\%=\frac{W_s+{W}_0}{W_0}\times 100 $$
(1)

Where:

Mt%:

percentage of total water absorption,

Ws:

Weight of sample after immersion in water

W0:

Original weight of sample before immersion in water

2.6 Morphological Analysis

Surface morphological study of LDPE, LDPE/TS, LDPE/TS/C5 and LDPE/TS/C10 film was examined using Philip XL40 Scanning Electron Microscope (SEM). The films have been sputter coated with a thin layer of gold before observation. All samples were observed under 100× and 500× magnification.

3 Result and Discussion

3.1 Melt Flow Index (MFI) Measurement

The results in Table 2 show that the MFI of the sample is declined when the content of LDPE matrix is decreasing in the blend. Similar results were reported in previous research that MFI of filled LDPE decreased as the filler loading increased (Thakore et al. 1999). Reduction in the MFI values indicates that the viscosity of the blend decreased. MFI values are used to measure the melt strength of the blend during the processing. For blend that is process via blown film, their melt strength must be sufficiently high, which can lead to increase the blow-up ratio of the films and expand into film with desired thickness. In this study, the blend samples with 15 and 20 wt.% CaCO3 couldn’t be blown into films as a result of low melt strength due to the reduced the volume fraction of LDPE. The melt strength was dropped significantly as the filler concentration was increased. The extruded strand become too brittle to draw since the volume fraction of the polymer decreased due to the increase in the volume fraction of the CaCO3 filler. The draw-ability is depended entirely on the matrix, not on the filler particles, and since there was less volume of matrix in the continuous phase to be drawn, the melt strength was reduced (Kao et al. 2002).

Table 2 MFI result of the samples

3.2 Fourier Transform Infrared Analysis (FTIR)

FTIR analysis was used to identify the component present in the film, especially CaCO3 after the processing. FTIR also can be used to characterize the compatibility of the polymer blends. For the compatible polymer blends, the FTIR spectrum band would deviate greatly displace (the band’s frequency shifting and the peak style asymmetrically widening) compared with that of the single polymer component (Wang et al. 2005). According to Gulmine et al. (2002), neat LDPE shows strong absorption band at 2919 cm−1 and 2851 cm−1 which is corresponding to asymmetric stretching of CH2; and strong bending deformation at 1473 cm−1 and 1463 cm−1 (Gulmine et al. 2002). Infrared spectrum of all films obtains in Figs. 1 and 2 showed there is absorption at 2915–2917 cm−1 that is corresponding to the ethylene group presented in the film. Infrared spectrum of neat CaCO3 shows there is absorption at 876, 1424 and 3735 cm−1 (Gorna et al. 2008). Kiatkamjornwong et al. (2001) stated that cassava starch give strong absorption band at 1727 cm−1 because of the presence of carbonyl group. FTIR spectrum showed absorption at 1747 cm−1 for each blown film, which is due to carbonyl group from starch (Kiatkamjornwong et al. 2001). Incorporation of CaCO3 into LDPE/TS system does not alter much on the absorption band of the blend. However, from the graph, it can be seen that the peak shape and location of the band change distinctly. The peak location shifted from 1424 cm−1 in CaCO3 spectrum to 1435 cm−1 in blend with 5 and 10 wt.% CaCO3. According to the mechanism of compatibility, the better the compatibility of the polymer blend, the more correlative peaks shifted and the peak shapes changed. This phenomenon indicated that CaCO3 are compatible with LDPE/TS matrix due to the usage of PE-g-MAH. LDPE matrix physically interacts with the ethylene group of the compatibilizer while carbonyl group from tapioca starch and carbonate group from calcium carbonate interact chemically with anhydride group of the compatibilizer (Arvanitoyannis et al. 1998b; Tabasum et al. 2019).

Fig. 1
figure 1

FTIR spectrum of LDPE/TS film, CaCO3 and LDPE/TS/C5 films

Fig. 2
figure 2

FTIR spectrum of CaCO3, LDPE/TS/C5 and LDPE/TS/C10 films

3.3 Differential Scanning Analysis

DSC was used to distinguish thermal transition, especially melting temperature and enthalpy of the obtained film. The crystalline melting temperatures and enthalpy of neat and blends are shown in Table 3. The endothermic peak temperature of LDPE was reduced after the incorporation of TS and 5% CaCO3 into the blend. However, addition of 10% CaCO3 increased the endothermic temperature to about 111 °C. Figure 3 showed that the second endotherm peak around 110–120 °C is representing the starch’s melting temperature. DSC curve of LDPE/TS film showed that TS has primary and secondary melting temperature. As CaCO3 was added, the primary melting temperature of the starch was shifted. The more CaCO3 was added, the more this shifted occurred. Hence, it can be concluded that CaCO3 lower the melting temperature of the films. The decreased in melting endotherm in the blends seems to indicate that there is some degree of compatibility between the LDPE, TS and CaCO3 (Prusty and Swain 2016).

Table 3 Thermal properties of blended samples
Fig. 3
figure 3

DSC results of LDPE, LDPE/TS, LDPE/TS/C5 and LDPE/TS/C10 blends

3.4 Tensile Test

3.4.1 Tensile Strength

Tensile strength of LDPE, LDPE/TS, LDPE/TS/C5 and LDPE/TS/C10 films in both machine direction and cross direction are shown in the Fig. 4. The results indicated that tensile strengths along machine direction are higher than transverse direction due to higher molecular alignments at machine direction, which led to higher strength. Previous studies (Arvanitoyannis et al. 1998a; Matzinos et al. 2001; Raj et al. 2001) reported that, starch reduces the tensile strength of LDPE blends film due to the poor interfacial adhesion between the two components. The result in Fig. 4 shows a small reduction (less than what was reported by previous studies) in tensile strength of LDPE after the addition of TS due to the function of PE-g-MAH compatibilizer that enhanced the interfacial adhesion between the matrix and the starch. However, the coupling efficiency of PE-g-MAH may be influenced by the structure of the main PE phase. Branched chain polymers usually have fever entanglement than linear polymer for a given molecular weight, resulting in lower tensile strength and elongation at break. Molecular orientation imparted during the blown film processing from the shearing and biaxial stretching action is also known to have major effect on the physical properties of the film (Majumder et al. 2007). After the addition of CaCO3, tensile strength of the film was increase in contrast to neat LDPE and LDPE/TS films. Previous study by Gorna et al. (2008), reported that the incorporation of CaCO3 into polyolefin reduced the tensile strength since the presence of filler influences the crystallization process and leads to an increase of the amount of imperfect crystalline phase and reduces the size of the crystallite (Gorna et al. 2008). It is obvious that fillers interfere with polymer chain mobility thus, reduced the stress transfer. However, in this study, addition of CaCO3 loading in the blend increased the tensile strength of the films which means that CaCO3 acted as reinforcement in the blends with very good interfacial adhesion. The respectable interaction of the component in the blend that showed by the low distinct of phase separation of CaCO3 in the surface morphology of LDPE/TS/C10 is the attributer to improve the tensile strength of the blend.

Fig. 4
figure 4

Tensile Strength of LDPE, LDPE/TS, LDPE/TS/C5 and LDPE/TS/C10 films in machine (MD) and cross direction (CD)

3.4.2 Elongation at Break

The result of Fig. 5 shows a decline in elongation at break of the films with TS in both machine direction (MD) and cross direction (CD) due to the low elongation at break of starch in contrast to LDPE. Physically, incorporation of starch/filler in the matrix of LDPE weakens the London forces between LDPE layers and reduce the possibility for chemical interaction between starches and LDPE, incorporation of starch produced discontinuity in the film matrix, and led to further decrease in elongation values. (Raj et al. 2001). Elongation at break of LDPE film at MD reduced after TS was added and further decreased was recorded at 5% of CaCO3. However, elongation percent was amended at 10% of CaCO3 from 66.361 to 71.203% but still lower then elongation of LDPE film. A biaxial molecular orientation is imparted into the materials produced by the film-blowing process. The two axes of orientation are the axial direction due to the drawing of the tube and the circumferential direction due to the air pressure existing inside the tube. The degree of molecular orientation existing in the main phase affects the mechanical properties of the specimens. The higher the degree of molecular orientation is the lower the extent of chain entanglement and the associated intermechanical locking. A higher degree of chain entanglement led to more effective transfer of stresses from matrix to starch particles (Matzinos et al. 2001).

Fig. 5
figure 5

Elongation at break of LDPE, LDPE/TS, LDPE/TS/C5 and LDPE/TS/C10 films in machine (MD) and cross directions (CD)

3.4.3 Tensile Modulus

Tensile modulus of LDPE and LDPE/TS blends film shown in Fig. 6 indicated that the moduli of LDPE films were enhanced by the addition of TS in both MD and CD. After the addition of CaCO3 the moduli of sample LDPE/TS/C5 were improved in both CD and MD. This results means that the stiffness of LDPE/TS film was enhanced after the addition 5 and 10 wt.% of CaCO3 filler due to the good physical interaction of the components in the blend as revealed by morphological study. The result indicated that TS and CaCO3 reinforced LDPE film and enhanced its stiffness. Similar finding was reported in former study which found that the modulus of LDPE film increased with the addition of starch (Matzinos et al. 2001).

Fig. 6
figure 6

Tensile modulus of LDPE, LDPE/TS, LDPE/TS/C5 and LDPE/TS/C10 films in machine (MD) and cross direction (CD)

3.5 Tearing Resistance

Resistance to tear is a key property of packaging film. Internal tear resistance is of significance for high capacity pouches and bags and many other industrial applications. Tearing resistance of LDPE, LDPE/TS, LDPE/TS/C5 and LDPE/TS/C10 films in machine and cross directions are shown in Fig. 7. The result reveled that tearing resistance of LDPE film was gradually improved after the blending with TS and the addition of CaCO3 in both directions. Tearing resistant was slightly improved when TS and 5 wt.% of CaCO3 were add in both direction (MD and CD) and a 30% improvement at 10 wt.% of CaCO3 in machine direction. Earlier study has reported a decrease in tear strength of LDPE when starch is add (Raj et al. 2004). The enhancement of tear strength of LDPE after the addition of TS and CaCO3 films is due to the high interaction between the matrix and the fillers as a result of using PE-g-MAH compatiblizer. Natural starch is a strong and stiff material (Matzinos et al. 2001), which imparts its properties of stiffness to LDPE films when the two components are compatible. This is the main motive behind the improvement of modulus of LDPE matrix after the addition of TS and CaCO3.

Fig. 7
figure 7

Tearing resistance of LDPE, LDPE/TS, LDPE/TS/C5 and LDPE/TS/C10 films in machine (MD) and cross directions (CD)

3.6 Water Absorption

A water absorption property of LDPE film is related to their degradable properties. The higher water absorption intake is the better degradation. The synthetic component appeared to be biodegraded by surface absorption of moisture and microorganisms, such as fungi and bacteria, because water absorption on the material allows the microorganism to grow and consume the surface as a carbon source (Kiatkamjornwong et al. 2001; Siracusa et al. 2008). Figure 8 reveal the results for the water absorption of LDPE, LDPE/TS, LDPE/TS/C5 and LDPE/TS/C10 films for 30 days. Water absorption of LDPE films was zero, due to high hydrophobicity in contrast to other films with percentage of water absorption of less than 7%. The trend shows in this study is in agreement with previous studies which found that, the percentage of water absorption of LDPE/starch film increased gradually then becomes constant and then decreased again after certain time of immersion (Kiatkamjornwong et al. 2001; Charoenkongthum and Aht-ong 2002). For the first 3 days, LDPE/TS films have higher percentage of water absorption of 6% due to the nature of starch, which is susceptible to water absorption. The affinity LDPE/TS blend films to water is due to hydrogen bonds formation between hydroxyl groups of starch and water molecules (Charoenkongthum and Aht-ong 2002). From the graph, it was obvious that incorporation of CaCO3 into LDPE/TS blend reduced the water absorption of the films as a result of hydrophilic property enrichment of the film by CaCO3. The decreasing in percentage of water absorption after some period of immersion implied that the film has become saturated with water (Kiatkamjornwong et al. 2001). Hence, it was concluded that the non swelling penetration (Fickian Diffusion) of water is reduced be the present of CaCO3 in the LDPE/TS blends, which can be considered an advantage for packaging materials.

Fig. 8
figure 8

Percentage of water absorption of LDPE, LDPE/TS, LDPE/TS/C5 and LDPE/TS/C10 films for 30 days

3.7 Morphological Study

The final properties of polymer blends are mainly depending on the morphology of the components. Therefore, it is necessary to study the morphology of the final blends product. In most cases, the major component of the blend forms the matrix, whereas the minor component is the dispersed phase. Figures 9 and 10 shows the surface morphology of the obtained films observed at 100× magnification and 500× magnifications. Morphology of LDPE film contained 20 wt.% TS showed that starch particle dispersed well in the polymer matrix. When CaCO3 was added to the blend, the morphology of the film remains unchanged when observed under 100× magnifications, where the particle size of starch and CaCO3 in the films appears to be the same in size. However, at 500× magnification, it was observed that the particle size of tapioca starch is bigger than the particle size of CaCO3. The size of tapioca starch particle seems to be ~20–22 μm in diameter while CaCO3 particle size appear to be of ~3–6 μm in diameter. The morphology of TS and CaCO3 in the films showed that there is good dispersion of the fillers (TS and CaCO3) in the matrix of LDPE. The good dispersion of the filler in the matrix was believed due to the used of PE-g-MAH as compatibilizer and also the used of stearic-coated CaCO3. SEM morphology also revealed that the physical interaction of all the components in the blends was good especially blend contain 10% CaCO3. Comparing the morphology of the films at 500× magnification reveals that as loading of CaCO3 increased, the phase separation become less distinct. With 5% CaCO3, starch particle and CaCO3 particle was clearly seen in the images. Good physical interaction in films contain 10% CaCO3 verified by the increased in tensile strength of this films compared to films with 5% CaCO3. Morphological study of LDPE/starch shows that all films exhibit a similar morphology with the starch droplet-like particle being homogeneously dispersed throughout the polymeric matrix similar observation was found in another study (Matzinos et al. 2001).

Fig. 9
figure 9

SEM Morphology of a LDPE, b LDPE/TS, c LDPE/TS/C5 and d LDPE/TS/C10 films at 100× magnifications

Fig. 10
figure 10

SEM morphology of a LDPE, b LDPE/TS, c LDPE/TS/C5 and d LDPE/TS/C10 films at 500× magnifications

4 Conclusion

LDPE/TS blends are problematic to be processed with 15, 20 wt% of CaCO3 to produce LDPE/TS/CaCO3 films using blown machine and only blends contain 5 and 10 wt% of CaCO3 can be blown into film. Incorporation of more than 10% CaCO3 into LDPE/TS blends reduced their melt strength. In addition, introducing CaCO3 into LDPE/TS blend led to slightly decline of melting temperature and enthalpy of the films. Mechanical properties of LDPE were marginally decreased after the blending with TS, and enhanced after the addition of calcium carbonate. The water uptake of the films is less than 9% after 30 days of immersion in water and CaCO3 found to reduce the water uptake of LDPE/TS films.