Mechanical properties and structure of mixtures of poly(butylene-adipate-co-terephthalate) (PBAT) with thermoplastic starch (TPS)

ABSTRACT

Mixtures of poly(butylene-adipate-co-terephthalate) (PBAT) with thermoplastic starch (TPS) in a broad concentration range from 70:30 to 30:70 wt.% were prepared in a twin-screw extruder. Mechanical properties were tested, showing that the addition of TPS leads to a substantial decrease in the tensile strength as well as in the elongation at break, while Young’s modulus is rising substantially. Additional data were collected by dynamic mechanical thermal analysis (DMTA) and scanning electron microscopy (SEM), enabling a more detailed discussion on relations between ultimate properties and structural parameters. The application potential of biodegradable PBAT/TPS blends consists mainly in the substitution of polyethylene in packagings, where the material investigated is also economically competitive. The main advantage of adding TPS to PBAT is increasing the biobased carbon content cost-effectively, since most PBAT on the market is fossil-based.

KEYWORDS:

• PBAT
• starch
• thermoplastic starch
• tps

Introduction

Biodegradable plastics (BDP) from renewable raw materials maintain the position among the most investigated topics in macromolecular science. In spite of this, the total volume of BDP covers hardly 1% of the world production of plastics. The main reason consists in the fact that BDP with acceptable physical properties are usually too expensive to compete efficiently with fossil–based high–volume plastics such as polyolefins retarding faster increase in the production volume.

An option to deal with this consists in the search for inexpensive biodegradable materials from renewable resources, among which especially starch is attractive from an economical point of view. Starch is highly favored as a renewable biodegradable polymer also because it is versatile, abundant [1] and its presence as an additive in another BDP may help to tailor the kinetics of biodegradation. To enable the starch processing by standard technologies, it must be transformed to thermoplastic form by introducing plasticizers (so-called thermoplastic starch, TPS). Water is the simplest option for TPS preparation, however, starch plasticized with water exhibits poor mechanical properties mainly due to extensive brittleness [2]. Glycerol is nowadays the most used agent, although a number of other plasticizers are also applied, such as polyols, glycols, formamide, urea, acetamide, citric or mellic acid, phosphorous compounds, and others [3]. The properties of prepared TPS depend on several factors. The most important ones are the structure of the starch determined also by the origin of the native starch and parameters of the preparation procedure of the TPS, but the composition of the final TPS is extremely important considering the additives present in the mixture [4]. However, the application of TPS itself is rare due to its brittleness and high sensitivity to moisture which limit the applications for packaging. Besides this, the retrogradation and crystallization of the mobile starch chains change its mechanical and barrier properties during storing [5,6].

As a result, TPS is often blended with other BDP such as polylactic acid (PLA) [7], usually with a compatibilizer, e.g. polylactic acid-grafted maleic anhydride [8]. Among the suitable blend partners of TPS, poly(butylene adipate-co-terephthalate) (PBAT) belongs to the most favored polymers since it combines biodegradability with other desirable physical properties, especially flexibility, unlike rather brittle PLA [9].

Several producers of PBAT [10,11], are offering the resin in various grades with an estimated current production capacity in the order of 100,000 tons per year. Typical properties of PBAT are shown in Table 1, and these are generally not far away from low-density polyethylene.

Table 1. Properties of PBAT. Reproduced with permission from [12]

Property	Unit	Standard	Value
Melt Flow Rate	g/10 min	ISO 1133	≤5	T = 190°C, w = 2.16 kg
Density	g/cm^3	ISO 1183	1.26
Tensile strength	MPa	ISO 527	>17
Melting temperature	°C	DSC	110–120
Elongation at break	%	ISO 527	≥600
Vicat softening point	°C	ISO 306	89

An obvious advantage of PBAT consists in the fact that it has been approved for food contact and compostability according to EN13432. Its main application is blown film [12], and another common application is represented by extrusion coatings [13]. Monolayer film is in general more common for biodegradable plastics than complex multilayer barrier film, where biodegradability of a single layer combined with other layers of non-biodegradable bioplastics would give marginal added value [14].

PBAT has been blended with PHB and natural fibers [15,16], and with PHBV [16,17]. Also, PBAT blended with PLA has been described [18–20], with polylactide- poly(butylene adipate-co-terephthalate)-polylactide tri-block copolymers incorporating PLA blocks (LPB and HPB) as compatibilizers [18]. Also, PBAT/kraft lignin blends have been prepared [21], as well as PP/PBAT blends [22]. In formulations with biodegradable and non-biodegradable constituents, one has to bear in mind the target application. In the past mulching film made from TPS and PE was used, but discontinued, due to the PE microplastics particles left behind in the soil. A good application for PBAT in PE, for instance, is to enhance its mechanical properties, e.g. the elongation at break, where biodegradability is no requirement. PBAT nanocomposites [23,24] reinforced e.g.. by functionalized cellulose nanocrystals or nanoclay are described in [25] and [26], respectively. The foaming behavior of PLA-PBAT blends is discussed in [27]. PBAT is an attractive material for blown film, to replace LDPE, e.g. for thin foils for the production of bags [12].

PBAT/TPS blends are a good means of achieving biobased carbon content in PBAT-formulations. Manufacturers BASF and Jinhui Zhaolong, for instance, are selling their pure PBAT under the trade names Ecovio™ and Ecoworld™, and their PBAT/TPS compounds under the trade names Ecoflex™ and Ecowill™, respectively.

Thermoplastic starch (TPS) is considered to be the cheapest biopolymer. However, it has poor mechanical properties, so it is an interesting blend partner for other bioplastics formulations. Maleic anhydride (MA), citric acid (CA) and maleated poly(butylene adipate-co-terephthalate) PBAT (PBAT-g-MA) have been described as compatibilizers for PBAT/TPS blends [28]. Table 2 shows PBAT and starch blends compared to LDPE.

Table 2. Commercial PBAT-based grades. The table lists the properties of 30 µm blown film: PE-LD, Ecoflex®, compounds of granular starch and Ecoflex®, compounds of thermoplastic starch and Ecoflex®. Reproduced with permission from [29]. MD = machine direction, CD = cross direction

	Test	Standard	PE-LD carrier bags (mean)	Ecoflex®	Ecoflex® + granular starch compound	Ecoflex® + thermoplastic starch compound
General	Transparency	-	Opaque	Translucent	Opaque	Opaque
	Printability	-	Eight colors flexoprint	Eight colors flexoprint	Poor printability	Eight colors flexoprint
Mechanics (MD, CD)	Modulus of elasticicty [MPa]	ISO 527	330/270	110/100	150/140	270/205
	Tensile strength [MPa]	ISO 527	32/25	35/40	23/22	21/20
	Ultimate elongation [%]	ISO 527	460/640	640/750	390/590	490/540
	Puncture resistance [J/mm]	DIN 53,373	17	26	9	19
Barrier properties	Oxygen [ml/(m^2*day*bar)]	ASTM D 3985	4800	2000	-	-
	Water vapor [g/(m^2*day)]	ASTM F 1249	3	240	-	-
Specific	Food contact	2002/72/EC	Not limited	Not limited	Dry food	Dry food
	Biodegradability		No	Yes	Yes	Yes

As can be seen from Table 2, blends containing TPS yield significantly better mechanical properties than granular starch. Among the frequently mentioned BDP which are mixing with TPS, polybutylene-adipate-co-terphthalate (PBAT) is often described, mainly due to its high flexibility. The deterioration of properties caused by the incompatibility of PBAT with TPS is dealt with by using compatibilizers such as citric acid and maleic anhydride [30], maleated PBAT [31], or a styrene–maleic anhydride–glycidyl methacrylate (SMG) terpolymer [32]. Modification of starch is also described as synthesizing a compatibilizer based on maleic anhydride grafted polyethylene glycol grafted starch (mPEG-g-St) [7]. Another approach is reactive extrusion of PBAT in the presence of a chain extender [32]. Mixtures of PBAT with TPS and PLA were also described using compatibilizers with anhydride functional groups [33]. Other approaches are also mentioned, e.g. oxidation of TPS was reported to tailor the rheological properties of the PBAT/TPS mixture [34].

Experimental

Materials and sample preparation

For preparation of mixtures, the polybutylene-adipate-co-terephtalate (PBAT, supplied by JinHui Zhaolong, China, grade Ecoworld™ 003 with an MFI < 5 according to DIN ISO 1133 at 190°C and 2.16 kg) and thermoplastic starch (supplied by Agrana, Austria) containing starch and glycerol as plasticizer in weight ratio 1:1 were used.

The two components were blended by mass at 0%, 30%, 40%, 50%, 60%, and 70% of TPS in a laboratory-scale twin-screw extruder. No additional plasticizer, compatibilizer, or additives were used. Prior to compounding and processing, PBAT was dried for 4 hours at 80°C. The blending was performed as a single pass extrusion on a twin-screw extruder (ZSK18) with L/D = 40 at 150 rpm with vacuum degassing and stand pelletizing. The temperature profile during the extrusion was between 180 and 220°C, hotter at the die plate.

Feeding of the starch was done via side feeder, PBAT through the main feed hopper.

Testing procedures

Mechanical properties were made in tensile mode. The procedure corresponds to the standard ASTM D638. Seven dog-bone testing specimens for each material were stamped out from the compression molded slabs with the dimensions of the deformed area 3.5 mm * 30 mm (thickness 1 mm). A universal testing machine (Instron 4301) was used with a clamp distance of 35 mm at a deformation rate of 1 mm/minute up to a deformation of 1% (for determination of Young’s modulus) and 50 mm/minute at higher deformations. The mean values and standard deviations were calculated from seven specimens for all parameters. Scanning electron microscopy (SEM) was carried out on samples that were first broken by a brittle fracture after cooling down in liquid nitrogen, and the surface was covered with gold. The micrographs were obtained on a FIB Microscope Quanta 3D 200i in a secondary electron mode. Alternatively, for the investigation of the surface of the samples after ductile fracture, the specimens were used after mechanical tests up to fracture in Instron at room temperature.

The preparation of surfaces and the procedure were identical as for those for brittle fracture investigation.

Dynamic mechanical analysis (DMTA) was performed as an additional tool for analysis of the structure and specific mechanical properties. Specimens of 1 mm thickness were tested in the instrument DMA Q800 (TA Instruments, Germany) in tensile mode and in a temperature range of −50 to 100 °C, with a heating rate of 2°C/minute at an amplitude of 20 µm and a frequency of 10 Hz.

Results and discussion

Standard mechanical properties are usually the most important data regarding any application of plastics-based materials. Tables 3 and 4 show the basic ultimate properties of the authors‘ mixtures.

Table 3. The effect of rising content of thermoplastic starch (TPS) in the mixture with PBAT on mechanical properties of blends mixed in a twin screw extruder: σ_M – ultimate tensile strength, ε_b – elongation at break σ_y yield strength ε_y elongation at yield, E – Young’s modulus, SD – standard deviations. The PBAT was Ecoworld™ 003, and for the sample 50/50EF, Ecoflex™ (BASF) was used for comparison

PBAT/TPS	^σM (MPa)	SD (MPa)	^εB ^(%)	SD (%)	E (MPa) DMTA	E (MPa)	SD
100/0	19.1	1.8	985	96	97	50.7	1.9
70/30	3.9	0.1	14	4	251	173.1	8.8
60/40	4.6	0.4	8	3	269	201.7	13.7
50/50	4.4	0	13	2	198	131.3	3.7
40/60	8	0.5	3	1	591	466.1	31.3
30/70	7	2.1	2	0	873	672.4	168
50/50(EF)	9.6	0.6	55	19	-	172	17.7

Table 4. The effect of rising content of thermoplastic starch (TPS) in the mixture with PBAT on mechanical properties of blends mixed in an extruder and additional mixing in a laboratory mixer (Brabender) as an additional homogenizing step: σ_M – ultimate tensile strength, ε_B – elongation at break σy yield strength ε_y elongation at yield, E – Young’s modulus, SD – standard deviation. The PBAT was Ecoworld™ 003, and for the sample 50/50EF, Ecoflex™ (BASF) was used for comparison

PBAT/TPS	^σM (MPa)	SD (MPa)	^εB ^(%)	SD (%)	^σY (MPa)	^εY ^(%)	E (MPa)	SD
100/0	19.1	1.8	985	96	-	-	50.7	1.9
70/30	7	0.2	296	12	5.4	13	125	4.6
60/40	6.1	0.2	291	27	5.1	9	174.2	11.7
50/50	7.5	0.4	325	26	-	-	115.9	5.3
40/60	9.1	1.3	6	1.5	-	-	320.2	31.2
30/70	n/a	n/a	n/a	n/a	n/a	n/a	n/a	n/a
50/50(EF)	10.9	0.1	46	6	-	-	198.8	6.3

As seen, for blends mixed in the twin – screw extruder, the tensile strength, and especially the elongation at break exhibit a decrease after the addition of 30 wt.% of TPS. A further increase of the TPS amount results in a small increase of strength while the elongation at break continues to decrease up to very brittle material if the TPS content rises up to 70 wt.%. An increase of TPS content leads to a monotonous increase of Young’s modulus; This behavior is somewhat unexpected due to increase of plasticizer content together with starch, but, on the other hand, it corresponds with both the monotonous increase in Young’s modulus as well as a slight decrease in T_g determined by DMTA as shown in Figure 2, indicating that the presence of glycerol is overruled by the starch, behaving either as small reinforcing particles or, more probably, increasing the stiffness via formation of hydrogen bonds within the blend structure.

The tendency of modulus increase is confirmed also by data from DMTA which were taken from curves of storage modulus G’ at a temperature of 20°C corresponding roughly to the conditions of tensile properties determination. It is worth mentioning that for both dependencies (Young’s modulus E as well as storage modulus G’) a shallow minimum appeared at 50% of TPS which most probably would be considered as experimental scatter.

By comparing the data in Tables 3 and 4 where samples were prepared by two alternative procedures (compression molding from pellets prepared by extrusion, Table 3, or compression molding from pellets reprocessed in a Brabender mixer, Table 4) the overall trends are the same but the samples after reprocessing exhibit higher strength and elongation while having slightly lower moduli. In both cases, the samples with TPS content over 50 wt.% are very brittle, even

so that reprocessed samples with 70 wt.% of TPS were impossible to stamp the testing specimens and samples for mechanical testing. Also, the DMTA testing got broken during fastening in the instrument clamps.

Another feature worth mentioning is a difference in mechanical parameters for samples with TPS content 50 wt.% prepared with either Ecoflex™ or Ecoworld™ matrix. Here the reason most probably consists in different variables of basic PBAT, where the difference is seen in molecular weight; The MFI for Ecoflex™ is claimed to be 3.8 g/10 minutes while for Ecoworld™ rather vague information is given, namely, below 5 g/10 minutes. DMTA data are shown in Figures 1 and 2. Figure 1, showing the courses of storage moduli on temperature, was mentioned above concerning the comparison of moduli measured by DMTA and by the standard procedure of mechanical properties determination. As stated above, the data of both methods correspond to an acceptable extent.

Figure 1. Dependences of Young’s moduli values on temperature for varying ratios of PBAT/TPS. The weight portion of TPS in the mixture with PBAT is shown in the Figure

Figure 2. Dependences of tan delta values on temperature for varying ratios of PBAT/TPS. The weight portion of TPS in the mixture with PBAT is shown in the Figure

Corresponding courses of loss tangent tgδ on temperature are shown in Figure 2, where also temperature maxima are indicated, corresponding to glass transition temperatures (T_g) of the samples. The changes of T_g of the PBAT matrix are small but statistically significant, as seen in Figure 2. Generally, a small decrease in T_g was observed with the increase of TPS content in the mixture. This decrease is somewhat surprising since peak maximum for TPS plasticized by glycerol is around 25°C [35], so that if at least partial miscibility of PBAT and TPS would occur, the two T_gs should move closer to each other (at complete miscibility, just one T_g peak should appear at the temperature corresponding to the rule of mixtures). Obviously, the mixture of PBAT with TPS represents an immiscible pair and the addition of TPS makes the matrix polymer PBAT stiffer, perhaps due to slight interactions between starch hydroxyl groups and ester functional groups of the PBAT. Therefore, the reason for a decrease in T_g might be seen in a certain solubility of glycerol in PBAT and to some extent in the plastification of the polymeric matrix. However, the total effect of the introduction of TPS is highly antiplasticizing, as seen in mechanical properties (Tables 3 and 4) where TPS presence results in an increase of brittleness of the mixture. Thus, it has to be stated that the small plasticizing effect of glycerol penetrating in a small amount into the PBAT matrix (the T_g decrease is marginal) is highly overruled by the presence of rather polar starch molecules representing defects in the blend structure being more vulnerable to fracture at low both force and deformation.

Figure 3. SEM images of fracture surfaces of mixtures of PBAT with thermoplastic starch (PBAT/TPS ratios in 70/30, 50/50 or 30/70 wt.%), prepared either by hammer hit after cooling the sample in liquid nitrogen (right column) or taken from parts of samples broken during tensile tests at room temperature (left column). The composition PBAT/TPS 50:50 is shown for samples either prepared from pellets (“unmixed”, after twin screw compounding) or after reprocessing in Brabender

Figure 3. (Continued)

The SEM images indicate that the surface of samples prepared by brittle fracture does not differ substantially. It is worth mentioning that no cracks were observed on the surface unlike as reported in [36] for PBAT/TPS plasticized by glycerol and mixed in an internal laboratory mixer. The samples prepared by compression molding after reprocessing in the Brabender mixer seem to have a somewhat finer structure compared to the sample compression molded directly from pellets (only samples of 50:50 composition are shown here as an example), but the difference is not large. On the other hand, quite substantial differences are revealed by the SEM images taken from samples prepared by ductile fracture during tensile tests. The sample composed to a majority of the PBAT matrix (70 wt.%) has formed a number of rather fine fibers, while for samples containing PBAT/TPS ratio 50:50, the fibers were much thicker. Moreover, the patterns from samples after reprocessing show shorter fibers compared to samples compression molded from pellets.

The sample containing 70 wt.% of TPS exhibits the fracture surface of features resembling brittle fracture even for specimen broken during tensile test at room temperature. The elongation at break confirms high brittleness of this composition. All these observations correlate with increased brittleness of PBAT with rising content of TPS.

Conclusions

Adding TPS to PBAT can reduce raw material costs and increase the biobased carbon content. Although (partly) biobased PBAT is available on the market, quantities are low and prices are high, so achieving a certain biobased carbon content from TPS addition is a viable option. The investigation of mixtures of PBAT with thermoplastic starch in a broad concentration range indicates that TPS addition results in a substantial decrease in tensile strength and an increase in the brittleness of the blends with rising TPS content. On the other hand, the tensile strength at a TPS content of 50 wt.% and more is similar to values for LDPE. This mixture may be attractive especially because of its high biodegradability (both PBAT and TPS are biodegradable) and competitive price of the blend.

Acknowledgments

The authors (FI and IC) are grateful for support of this research by the Slovak grant agency VEGA (projects No. 2/0570/17 and 2/0019/18).

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Notes on contributors

Maximilian Lackner

PD DI Dr. Maximilian Lackner MBA is study programme director at the University of Applied Sciences Technikum Wien, Austria. His research interests include polyhydroxyalkanoates (PHA), particularly their manufacturing from CO2 using cyanobacteria and from CH4 using methanotrophic bacteria.

František Ivanič

Ing. František Ivanič (PhD) works at the Centre for Functional and Surface Functionalized Glass, Alexander Dubček University of Trenčín. His research interests are polymer composites and bioactive glasses.

Mária Kováčová

Mgr. Mária Kováčová (PhD) is a research scientist at Polymer Institute SAS in Bratislava, Slovakia. Her research interests include polymer (nano)composites especially with carbon (nano)fillers, innovative antibacterial materials based on hydrophobic carbon quantum dots and their use in biomedicine, materials for 3D printing.

Ivan Chodák

Prof. Ivan Chodák (PhD, DSc) is a senior scientist at Polymer Institute SAS in Bratislava, Slovakia. His scientific interests cover crosslinking of polyolefins (patented extremely efficient system for polypropylene crosslinking), multiphase systems, i.e. polymer blends and composites with polymeric matrices including nanocomposites and electroconductive composites. Investigation of biodegradable polymers belongs to his most successful topics (patented material based on PHB and PLA with very high toughness, the commercial production is under preparation). He is the author of over 150 scientific publications, 15 patents (4 of them applied in industry). Frequent and long term cooperation with industry (DSM, BASF, Biomer, National Power, GE Plastics etc.). The research interests of Prof. Chodak include multiphase materials with polymeric matrices, especially blends of biodegradable polymers, polymeric clay-containing nanocomposites and electroconductive composite materials.