Abstract

The effect of plasticizer (polydimethylsiloxanol) and neat (SiO2) or modified (having amine functional groups) silica (A-SiO2) on morphology, thermal, mechanical, and rheological properties of PLA/TPS blends compatibilized by maleated PLA (MPLA) was investigated. Toughened PLA/MPLA/TPS (60/10/30) blend containing 3 wt.% of plasticizer and various contents (1, 3, or 5 wt.%) of silica were prepared in a corotating twin-screw extruder. From SEM, it is clear that plasticized PLA/MPLA/TPS blend continuous porous structure is highly related to the silica content and its functionality. The results indicate that polydimethylsiloxanol enhances ductility and the initial thermal stability of the plasticized blend. DSC and DMTA analyses show that nucleation ability and reinforcing effect of A-SiO2 on plasticized blend crystallization are much better than those of SiO2. Silica practically had no effect on the thermo-oxidative degradation. However, the composites with A-SiO2 had better thermal stability than those with SiO2. Moreover, silica significantly improved the elongation at break.

1. Introduction

Polylactide (PLA), a biodegradable linear aliphatic polyester, is one of the most potential materials with great environmental benefits since it offers very low toxicity and high mechanical performance. Therefore, PLA can be broadly applied in the fields of drug delivery systems, sutures, orthopaedic implants, tissue engineering, and packaging materials [14]. However, a low glass transition temperature (about 60°C), relatively high brittleness, and higher cost, if compared to conventional polymers, limit its applications [5]. There are many ways of improving PLA toughness including copolymerization and melt blending with other more flexible commodity polymers such as linear low-density polyethylene or natural rubbers, as well as plasticizers [610]. Generally, an efficient plasticizer decreases the glass transition temperature (Tg) and melting point (Tm) of polymers. It was reported that the addition of plasticizers such as polyethylene glycol, citrate, oligomeric lactic acid, and triacetine could overcome PLA’s brittleness [1114]. Unluckily, plasticization decreases PLA stiffness, limiting its constructional applications. Therefore, an efficient impact modifier is required to improve toughness without extensive loosening of PLA stiffness.

Thermoplastic starch (TPS) is one of the well-known biopolymers and is often used in order to lower the cost of the final product and enhance the biodegradable characteristics of polymer composites [1517]. Świerz-Motysia et al. reported that TPS increased the biodegradation rate of PLA in so-called soil test, which was enhanced as a function of TPS [18]. Thus, melt compounding TPS with PLA is one of the most promising methods to solve PLA limitations or to lower its high price [19]. Nevertheless, incompatibility between hydrophobic PLA and hydrophilic starch results in poor mechanical properties, due to inadequate interfacial interaction between these polymers. To improve the adhesion of PLA and TPS, chemical modification of the starch before blending is performed and compatibilizers such as maleic anhydride (MA) and epoxy resins have been used [4, 2024].

It has been previously suggested that the intensive interfacial interactions among hydrogen bonds of the anhydride groups of the maleated polylactide (MPLA) and the hydroxyl groups of the starch are responsible for the increase in tensile, flexural, and impact properties of PLA and thermoplastic potato or corn starch blends [25].

Recent results show that the nanoscale distribution of montmorillonite [18, 2628], fumed silica (SiO2) [2932], and calcium carbonate [33] within the PLA matrix significantly enhances their thermal, mechanical, and rheological properties, as well as reduces permeability and flammability, compared to conventional micro- and macrocomposites. In order to reach the nanoscale distribution, the naturally hydrophilic nanofiller has to be organically modified to be more compatible with the organic polymer matrix. However, only few studies have been conducted in the development of plasticized PLA/thermoplastic starch nanocomposites recently [3436].

In a previous work [37], we have studied the influence of the core-shell polymeric nanofiller and maleated polylactide, used as a compatibilizer, on the structural, mechanical, and barrier properties of PLA and TPS blends. The addition of the core-shell polymeric nanofiller in the presence of MPLA improved adhesion between PLA and starch. Melting point and crystallinity temperature of PLA were found to decrease by adding TPS and the polymeric nanofiller. The nanocomposites showed significantly lower stiffness compared to neat PLA. However, the storage modulus increased as a function of nanofiller content. Moreover, the addition of the core-shell polymeric nanofiller strongly improved impact strength, the elongation at break and barrier properties of PLA/TPS blends.

To the best of our knowledge, no systematic studies have been done so far to investigate the properties of polydimethylsiloxanol toughened PLA/TPS/silica nanocomposites. Therefore, the objective of this paper is to investigate the thermal, mechanical, rheological, and morphological properties of polydimethylsiloxanol toughened PLA/TPS/silica nanocomposites in the presence of MPLA used as a compatibilizer. In particular, the effects of plasticizer, spherical silica content, and its functionality were also studied.

2. Materials and Methods

2.1. Materials

Polylactide polymer (PLA, 2003D) was provided by Nature Works, USA, with melt flow rate of 6.9 g/10 min (ISO 1133 at 210°C and 2.16 kg). The characteristics of PLA are described in Table 1. Maleated polylactide (MPLA) containing 0.68 wt.% of grafted maleic anhydride, with melt flow rate of 3.8 g/10 min (ISO 1133 at 210°C and 2.16 kg), was obtained according to the procedure published elsewhere [18, 38], and it was used as a compatibilizer for PLA/TPS blends at the concentration of 10 wt.%. Maleic anhydride (MA) as coupling agent and dicumyl peroxide (DCP) as initiator were purchased from Sigma-Aldrich, Italy.

Thermoplastic corn starch (TPS) was obtained by melt blending according to the procedure published elsewhere [18, 39] and used as a biodegradable additive. Standard corn starch with about 23% of amylase, 11–13% humidity, and pH 6.5 was purchased from Cargill, Germany. The glycerol with 99.5% purity was the product of Rafineria Trzebinia, Poland. The characteristics of TPS are shown in Table 1.

Plastosil M-2000, the commercial name of polydimetylsiloxanol, containing 0.04 wt.% of reactive silanol groups, η = 2000 ± 500 cP, d = 0.97 g/cm3, supplied by Chemical Plant Polish Silicones Ltd., was used as a plasticizer (P) for the PLA/MPLA/TPS (60/10/30) blend.

Neat (SiO2) and modified (having amine functional groups) spherical silica (A-SiO2) nanoparticles synthesized according to the previously reported sol-gel process [4042] were used as nanofillers. Tetraethoxysilane (TEOS), technical grade, commercial product name TES 28, was supplied by Wacker Chemie (Germany). Ethyl alcohol (reagent grade) and aqueous ammonia (reagent grade, 25 wt.%, d = 0.91 g/cm3) were supplied by POCh S.A. (Poland) and used as received. The characteristics of silica are described in Table 2.

2.2. PLA Functionalization

In order to improve compatibility between PLA and TPS, PLA was grafted with maleic anhydride via reactive extrusion to obtain a grafting yield of 0.68 wt.% according to the procedure published elsewhere [18, 38]. MPLA was prepared as follows: PLA, MA (2 wt.%, PLA basis), and DCP (18 wt.%, MA basis) were mixed in the internal mixer and, subsequently, the mixture was extruded using a corotating twin-screw extruder (Berstorff, Germany) with a screw diameter of 25 mm and length to diameter ratio of 33. To prevent the degradation of the polymer, during the processing time, a nitrogen purge flow was used. Different screw elements along the screw worked in order to induce polymer melting [43]. The three mixing sections enhanced the compounding and increased the residence time of the mixture in the barrel. The barrel pressure in these parts, as well as at the section before the die, could be increased. The extruder also had a vacuum degassing port to remove any moisture traces or other volatile products formed during compounding. The process was carried out at a barrel temperature 130–140°C. The output was 4 kg/h, and the screw speed was 100 rpm.

2.3. TPS Preparation

To obtain TPS, destructurization and plasticization of native corn starch were performed in one-step extrusion process in the earlier mentioned Berstorff twin-screw corotating extruder, using 30 wt.% of glycerol as a plasticizer [18, 39]. The process was carried out at a barrel temperature 130–170°C and screw speed of 80 rpm. The TPS was cooled in ambient air and then was pelletized. Figure 1 shows SEM micrographs of native and thermoplastic corn starch. It is obvious from Figure 1 that the starch particles grains disappeared after compounding with a glycerol, suggesting that thermoplastic starch was successfully plasticized and TPS presented a homogeneous morphology.

2.4. Synthesis of Neat and Modified Spherical Nanosilica

Neat (SiO2) and modified (heavy amine functional groups) spherical silica (A-SiO2) were synthesized according to the developed sol-gel process using tetraethoxysilane (TEOS) as alkoxysilane precursor at room temperature (23°C), with a stirring speed of 250 rpm [4042]. The process was carried out in an aqueous ammonia-ethyl alcohol reaction mixture using molar ratio of TEOS/EtOH/H2O as 0.023/0.500/0.477, and the initial pH ranged from 10.4 to 11.3. The final pH range was 7.5–10.8. The pH of the reaction mixture was measured using a pH meter (Schott Instruments LAB 850). Modified silica was synthesized by adding drop by drop γ-aminopropyltriethoxysilane (Momentive Performance Materials, USA) to the reaction mixture, according to the reaction (Figure 2).

The final product of neat or modified silica was dried in a spray dryer for 2 hours at 50–90°C. Particle size and particle size distribution in resulting sols were measured by photon correlation spectroscopy (PCS) using a Malvern apparatus (Zetasizer Nano ZS, UK). The monomodal particle size distribution and very low polydispersity of particle size were observed for homogeneous sols obtained by the sol-gel process [40]. It is clear from Figure 3 that spherical shape and uniform sized (about 30 nm) silica nanoparticles were obtained. The amine groups content was determined based on nitrogen content measurement by the Kjeldahl method. The characteristics of the neat and modified silica are presented in Table 2.

2.5. Blend Preparation

The PLA/MPLA/TPS 60/10/30 blends without or with 3 wt.% silicone rubber plasticizer were melt blended using the earlier mentioned twin-screw corotating extruder. PLA, MPLA, TPS, and plasticizer (P) were fed into the throat of the extruder using separate gravimetric feeders. The temperatures of the ten zones were 35°C/160°C/175°C/180°C/180°C/180°C/180°C/185°C/185°C, and 190°C for the sequential heating zones, from the hopper to the die. The screw speed was 100 rpm. The melt temperature and pressure were continuously recorded during compounding. The extrudate was immersed immediately in a cold-water bath (20°C) and pelletized with an adjustable rotating knife located behind the water bath, into 5 mm pellets.

2.6. Composite Preparation

The various composites containing 1, 3, or 5 wt.% of SiO2 or A-SiO2 silica nanoparticles were prepared through melt-direct dispersion by using a conventional polymer extrusion process and optimized parameters. Firstly, PLA pellets with 10 wt.% of silica were blended using the earlier mentioned twin-screw corotating extruder, based on the reported method [44]. Prior to the melt processing, this masterbatch was dried for 12 hours at 80°C and then melt blended with pure PLA, MPLA, TPS, and plasticizer. The process was carried out using a screw speed of 150 rpm and also a temperature profile of 35°C/165°C/175°C/180°C/180°C/180°C/175°C/185°C/185°C, and 200°C for the sequential heating zones, from the hopper to the die. Then, the material was cooled in water and pelletized.

2.7. Fourier Transform Infrared (FTIR) Spectra and the Grafting Degree

Fourier transform infrared spectroscopy (FTIR) spectra were recorded using a Thermo Scientific spectrometer, model Nicolet 6700 for a frequency range between 4000 cm−1 and 500 cm−1. An average of 64 scans at a resolution of 2 cm−1 was conducted at room temperature. To collect the spectra of the polymers and composites, thin films were prepared in a hydraulic hot press. The nanosilica spectrum was taken using KBr pellets. The spectra presented were baseline corrected and converted to the absorbance mode.

The MA grafting degree was measured by Fourier transform infrared (FTIR) spectroscopy. FTIR was performed on a spectrometer PerkinElmer System 2000 on films 0.07 mm thick. The spectra were obtained by collecting 64 scans between 500 and 4000 cm−1 with a resolution of 2 cm−1. The absorbance of the analytical band was determined, and the MA content was calculated from the calibration curve. The measured grafting yield was 0.68 wt.% [18].

2.8. Scanning Electron Microscopy

The microstructures of silica nanoparticles, TPS, blends, and composites were examined using a JEOL JSM-6490LV scanning electron microscope (SEM). Fractured surface of TPS, blends, and composites were gold-coated prior to observation to avoid electrical charging and to increase image contrast.

2.9. Differential Scanning Calorimetry

Thermal analysis was performed by differential scanning calorimetry (DSC) using a DSC-7A apparatus of PerkinElmer (Switzerland) under nitrogen. All measurements were carried out according to the following cycle: heating from 20 to 180°C at a rate of 10°C/min, 3 min isothermal step at 180°C, cooling down to 0°C at a rate of −10°C/min, 3 min isothermal step at 0°C, and final heating up from 0 to 180°C at a rate of 10°C/min. The amount of sample placed in the DSC aluminium pans was about 6 mg. An empty pan was used as a reference. The glass transition temperature (Tg), crystallization temperature (Tc), and melting temperature (Tm) were determined from the second heating scans. The degree of crystallinity () of PLA, blends, and composites was evaluated from the melting enthalpy results (ΔHm) of each sample using (1), where is the experimental melting enthalpy and is the melting enthalpy for 100% crystalline PLA, 93 J/g [45]:

2.10. Thermogravimetric Analysis

Thermogravimetric analysis (TGA) was performed using a thermogravimetric analyzer (TGA/SDTA 851e Mettler Toledo) at a rate of 10°C/min from 20 to 600°C under flowing air (50 ml/min). The precision on temperature measurements is ±0.5°C.

2.11. Tensile Properties

The test specimens were prepared by injection moulding using an Arburg 420 M single screw injection machine (Allrounder 1000-250, Germany) containing five different heating zones. The temperatures of these were 180/190/195/195/200°C, from the feeding zone to the die, when the mould was cooled with water at 25°C.

Tensile strength and elongation at break were determined using an extensometer clip-on incremental (Instron series 5500 R, UK) at a cross-head speed of 5 mm/min, whereas tensile modulus was measured at the speed of 2 mm/min. All tests were performed at standard atmosphere conditions (23°C and 50% HR). Prior to testing, the samples were stored at 23°C and 50% RH for 48 h, according to ISO 527 and ISO 179 standards. All the results represent an average value of a minimum 5 tests.

2.12. Dynamic-Mechanical Thermal Analysis (DMTA)

The dynamic mechanical properties of samples were tested using a dynamic mechanical analyzer, model Rheometrics RDS 2. The torsion method was used with a frequency of 1 Hz, a strain level of 0.1% in the temperature range of −150°C to 100°C. The heating rate was 3°C/min. The testing was performed using rectangular bars measuring approximately 38 × 10 × 2 mm, prepared by injection moulding.

3. Results and Discussion

3.1. FTIR Analysis

Figure 4 shows FTIR spectra of the pure polymers (PLA, MPLA, and TPS), plasticizer, A-SiO2, and composite with 5 wt.% A-SiO2. The strong absorption band, which appears at around 1750 cm−1 in the spectrum of both PLA and MPLA, is assigned to C = O carbonyl stretching vibration. The spectrum of MPLA shows the new relatively weak absorption band at around 1850 cm−1, which is ascribed to the carbonyl group (C = O) stretching of the succinic anhydride ring (or saturated cyclic anhydride ring) [4, 18, 46]. The new absorption band indicated that MA was grafted onto the PLA backbone, during which MA was transformed to a saturated anhydride (succinic anhydride). The typical anhydride band at 1780 cm−1 could not be observed because of overlapping of the intense PLA band at 1750 cm−1 [47]. In the spectrum of TPS, the peaks at 1020 cm−1 and 1075–1150 cm−1 were attributed to C–O stretching of the C–O–C group and C–O stretching of the C–O–H group, respectively [48]. The peak at 1650 cm−1 was due to the bound water present in the starch. A broad band, due to the hydrogen bonded hydroxyl group (O–H), appeared at 3040–3640 cm−1 and is attributed to the complex vibrational stretching, associated with free, inter- and intramolecular bound hydroxyl groups [49]. The spectrum of the plasticizer shows a strong absorption band at 1260 cm−1 and a weak absorption band near 860 cm−1 attributed to the Si–CH3 stretching. Moreover, the peaks at 2910 cm−1 and 2960 cm−1 were attributed to C–H stretching of the methyl group. However, the typical silanol groups (Si–OH) band at 3500 cm−1could not be observed because they were in very small amount (0.04%). In the spectrum of A-SiO2, there is a strong band at 1060 cm−1 attributed to the Si–O groups and a broad peak with a maximum at 3435 cm−1 corresponding to the surface hydroxyl groups [50]. However, the peak of amine functional groups was not recorded, due to their presence in very small amount (0.35%).

From Figure 4, it is clear that the characteristic peak assigned to the anhydride group (1850 cm−1) in MPLA is diminished in the spectrum of composite, maybe due to the very small amount that the composite contains (0.068 wt.% MA groups of the whole composite). The new weak absorption band at 1260 cm−1 attributed to the Si–CH3 stretching from the plasticizer was observed. The typical Si–O groups band at 1060 cm−1could not be observed because of overlapping of the intense PLA band at 1180 cm−1, TPS band at 1120 cm−1, and plasticizer band at 996 cm−1. The intensity of the peak centred at 3310 cm−1, corresponding to OH groups of starch, is suppressed in the spectrum of the composite. This may be due to reaction between the OH groups of starch and anhydride groups of MPLA. The formation of hydrogen bonds between the hydroxyl groups of the carbonyl groups of TPS and PLA is one of the possible reactions [48]. The complex interactions between A-SiO2, plasticizer, MPLA, and TPS also could have occurred.

3.2. Morphological Analysis

It is well known for polymer blends that the morphology control of the respective phases is a key factor in achieving the desired material properties [26, 34, 37]. Figure 5 shows cross-sectional images of the nonplasticized and plasticized PLA/MPLA/TPS (60/10/30) blends. As expected, the PLA and thermoplastic starch blend in the presence of compatibilizer (MPLA) shows individual grains of starch forming dispersed phase in the PLA matrix. From those images, it is obvious that the phase morphology of the samples can be depicted as continuous and porous. Moreover, the addition of the plasticizer results in less porous structure, suggesting improved toughness. It is well known that filler dispersion and adhesion to the polymer matrix are of great importance for the mechanical properties of composites improvement. Good control of the interface morphology of the composite is one of the most critical parameters to achieve the desired mechanical properties of such materials [26, 34, 37]. Scanning electron microscopy was performed to explain the behaviour of the silica-filled PLA/MPLA/TPS/P composites.

Figure 6 shows the dispersion state of silica-filled PLA/MPLA/TPS/P composites.

In our previous work, we found that silica nanoparticles were agglomerated at higher content, thus reducing their available surface area for reinforcing effect [51]. Moreover, A-SiO2 shows bigger tendency to form agglomerates than SiO2.

From the SEM images (Figure 6), it is clear that the PLA/MPLA/TPS/P porous structure is highly related to the silica content as well as its functionality. However, pores size distribution could not be analyzed because of their irregular shape. The neat silica results in more porous structure than the modified one. Furthermore, the composites with of A-SiO2 had more regular shape and smaller size of pores than those with SiO2. This may be due to the result of complex chemical reactions that occur mainly among hydroxyl groups of neat silica, MPLA and TPS, which seems to be more favoured than reactions among amine functional groups of modified silica, MPLA and TPS. These behaviours are highly proportional to the trend observed in the stress-strain behaviour of the composites with neat or modified silica as depicted in Figure 7.

3.3. Thermal Properties

The second heating of differential scanning calorimetry (DSC) thermograms of neat PLA, TPS, nonplasticized, and polydimethylsiloxanol plasticized PLA/MPLA/TPS (60/10/30) blends are displayed in Figure 8. It is clear from the above figure that the temperature according to the endothermic peak for each sample is considered to be the glass transition temperature (Tg) of PLA. Moreover, all the samples show an exothermic peak that can be correlated to the crystallization of PLA; the corresponding temperature is known as crystallization temperature (Tc). The neat PLA showed a clear glass transition temperature at 62.1°C, crystallization temperature at 112.5°C, and melting point (Tm) at 153.6°C, corresponding to residual crystallinity, as also discussed by Martin and Averous [14]. It is interesting to know that the Tc peak of PLA did not appear during cooling of PLA and its blends. It is believed that the main reason for this occurrence was due to a very slow crystallization rate of PLA during cooling [52]. The crystallinity (Xc) of pure PLA is only 3.3% after melt blending, which indicates that the material is almost amorphous. It can be also observed in Figure 8 that there are no discernible changes in the DSC thermograms of TPS, suggesting that the thermoplastic starch is in the amorphous phase.

The DSC data for various blends and composites are summarized in Table 3. It is well known that the value of Tg depends primarily on chain flexibility, molecular weight, branching/crosslinking, intermolecular attraction, and steric effects. The glass transition temperature of PLA (62.1°C) was reduced to 60.3°C with the introduction of TPS and MPLA. This emphasized that the chain mobility of PLA has been increased, owing it to the plasticizing effect brought by gelatinized starch with glycerol [53]. Moreover, a small reduction of melting temperature of PLA was observed (from 154°C to 153.2°C), and there was a significant increase in crystallization temperature (from 112.5°C to 128°C) as well as in the degree of crystallinity (from 3.3 to 7.2%). The addition of plasticizer to the PLA/MPLA/TPS blend further decreased the Tg value from 60.3°C to 59.6°C. The reduction of Tg affected other two temperatures, that is, Tm and Tc. The addition of plasticizer decreased the Tm value from 153.2°C to 148.9°C, while the Tc value was reduced from 128°C to 126.7°C resulting in significantly lower crystallinity of the PLA phase. This may have occurred due to the preferable interaction between plasticizer silanol groups and TPS hydroxyl groups and carboxyl groups of the PLA chains [54]. Hence, the thermal characteristics of plasticized PLA/MPLA/TPS blend compared to that unplasticized demonstrated that polydimethylsiloxanol could be regarded as the efficient plasticizer for the PLA/MPLA/TPS blend. These results are in agreement with their tensile properties. Figure 8(b) shows DSC traces for the PLA/MPLA/TPS/P blend and composites differing in A-SiO2 content. It can be observed from Table 3 that the addition of neat as well as modified silica had only little influence on the glass transition temperature of the composites. However, the Tg of PLA increased from 59.6 to 61.5°C at 5 wt.% of A-SiO2. It should be noted that Tc strongly depends upon the content as well as functionality of silica. Table 3 also shows that the Tc decreases with increasing loading of silica and is lower for the composites with modified silica. Moreover, the degree of crystallinity increases with silica content and is significantly higher for the composites with modified silica. This behaviour indicates that the large surface of the dispersed silica acts as a nucleating agent for the PLA phase crystallization. It should be noted that modified silica is a more efficient nucleating agent for the toughened PLA/MPLA/TPS blend than neat silica. Wu et al. [55] also reported similar results, where the Tc of PLA decreased with increasing MMT loadings. They also suggested that the difference in the dispersion state of MMT might also be an important factor influencing the crystallization behaviour of PLA nanocomposites.

Table 3 also shows that the melting temperature of the composites with neat silica is higher (2.8–3.3°C) when compared to the PLA/MPLA/TPS/P blend. However, Tm decreases in SiO2 function. In the contrary, Tm increases with A-SiO2 content. The melting temperature of the composites with modified silica is 2.3–4.1°C higher than that for the blend without silica. However, Tm of PLA/MPLA/TPS/P/A-SiO2 composites is 1–3°C lower as compared to that of pure PLA.

Figure 9 shows the TGA curves of neat PLA, TPS, nonplasticized, and plasticized PLA/MPLA/TPS (60/10/30) blends without or with 5 wt.% A-SiO2. The initial thermal stability is characterized by the temperature that occurred at 10% weight losses, referred to as T10 (see Table 4). As shown in Figure 8, TPS dramatically reduces thermal stability of PLA. Petinakis et al. [56] reported that small molecules such as CO, CO2, H2O, CH4, C2H4, and CH2O were produced when starch was decomposed. These molecules could break down the PLA chain resulting in lower thermal decomposition temperatures of PLA. During thermal degradation, the TGA curves display triple-step degradation processes for all the blends. Addition of polydimethylsiloxanol resulted in significant improvement in the initial thermal stability of the PLA/MPLA/TPS blend. As shown in Table 4, the T10 increases dramatically from 236°C to 296°C with the incorporation of 3 wt.% of plasticizer. However, there is practically no effect of the plasticizer on the other degradation temperatures (at 50% weight loss, Tmax1 and Tmax2). Moreover, the addition of silica practically had no effect on the thermal degradation of the plasticized PLA/MPLA/TPS blend. However, the composites with A-SiO2 had better thermal stability than those with SiO2. The best thermal stability showed the composite with 5 wt.% of A-SiO2 (Figure 8), where T10 reached 296°C (an increment of 3°C). Moreover, the Tmax2 was 5°C higher when compared to neat PLA.

3.4. Mechanical Properties

Neat PLA has high tensile modulus (3930 MPa) and tensile strength (67 MPa). However, it is a brittle material with the elongation at the break of 6%. Thus, it is needed to be improved by some additives. To investigate the effect of interfacial modification on the mechanical performance of the polydimethylsiloxanol toughened PLA/MPLA/TPS blends, the tensile stress-strain behaviour was characterized. It is clear from Table 5 that the addition of plasticizer caused a significant increase in an elongation at break indicating that the polydimethylsiloxanol is an efficient plasticizer for PLA/MPLA/TPS blends. Based on the abovementioned results, it can be supposed that using plasticizer can improve intermolecular interactions among the blend components through the reactive =SiOH groups.

Moreover, tensile strength and modulus increased compared to the blend without plasticizer.

The similar results and the dependency of poly(ethylene glycol) (PEG) on mechanical properties of PLA/TPS blends were reported by other researchers [57]. The optimized mechanical properties were obtained for the blend with 3 wt.% PEG. Szadkowska et al. [54] used two types of reactive plasticizers with silanol groups to obtain blends of PLA, maleinated PLA (MPLA), and thermoplastic corn starch (TPS). The incorporation of plasticizer with silanol groups into PLA/MPLA/TPS blends resulted in enhanced mechanical properties (especially elongation at break and impact strength) compared to the nonplasticized blend. This is a consequence of the chemical reactions that occurred between silanol groups of the plasticizer and functional groups of TPS and MPLA, which improved compatibility between PLA and TPS.

Silica further increased the elongation at break of the material. However, the improvement strongly depends on silica content as well as its functionality. For neat silica, elongation increased whereas for the modified one, it decreased as silica content increased. Significantly higher value was observed for 1–3 wt.% of the chemically modified silica, indicating, most probably, intensive interfacial interactions among the hydrogen bonds of the anhydride groups of the MPLA, the hydroxyl groups of the starch, and amine groups of the silica. The improvement in elongation was also observed for plasticized PLA and organically modified montmorillonite nanocomposites [27]. However, Arroyo et al. reported an opposite behaviour for PLA/thermoplastic starch/montmorillonite nanocomposites [28]. Moreover, composites show lower tensile strength and modulus than that of the plasticized PLA/MPLA/TPS blend, which increased with increasing silica content. This anomalous behaviour may be resulting from preferential or virtually unpredictable distribution characteristics of silica around the micropores within the toughened PLA/MPLA/TPS structure with respect to weight content and functionality [18]. Recalling DSC findings, we should also take into account the PLA crystallization behaviour as well. In polymers, surfaces are known to act as catalysts for the nucleation of crystals. In polymers patterned with pores, as in our case, it is possible that the shape of the pores, sizes of which shows dependence on silica content and functionality, can control the kinetics of surface-induced crystal nucleation. Moreover, it is well known that the degree of crystallinity may significantly influence on the mechanical properties, since it affects the extent of the intermolecular secondary bonding. Theoretically, higher crystallinity could reduce the elongation at break. But the kind of physical crosslinking formed through physical hypobonds made the composite elastic and have a high elongation at break [58]. For crystalline regions wherein molecular chains are packed in an ordered arrangement, wide-ranging secondary bonding occurs between adjacent chain segments. These bonds lead to significant increase of polymer tensile modulus with the growing degree of crystallinity [26, 51].

3.5. Dynamic Mechanical Thermal Analysis (DMTA)

DMTA measures the response of a given material to an oscillatory deformation (here in torsion mode) as a function of temperature. DMTA results are expressed by two main parameters: the storage modulus (G′), corresponding to the elastic response to the deformation, and tan δ, that is the G″/G′ ratio, useful for determining the occurrence of molecular mobility transitions such as the glass transition temperature. Figure 10 shows the temperature dependence of G′ and tan δ of pure PLA, TPS, and unplasticized and plasticized PLA/MPLA/TPS blends. As can be seen, the storage modulus of the toughened blend was lower than that of the blend without a plasticizer. It is known that the storage modulus detected by DMTA relates to composite stiffness. The stiffness of the PLA/MPLA/TPS blend decreased with the addition of the plasticizer (Table 6). This is a typical behaviour for plasticized thermoplastics.

In general, the storage modulus decreased as the temperature increased. However, in the region corresponding to the maximum of tan δ plot, the decrease in storage modulus was usually rapid. Figures 10(b) and 10(c) show the curves of the loss factor (tan δ) as a function of temperature for PLA, TPS, and PLA/MPLA/TPS (60/10/30) blend without or with plasticizer. The loss factors were sensitive to molecular motion, and their peak was related to the glass transition temperature. The curve of TPS revealed one thermal transition located at −42°C corresponding to a glycerol-rich phase of TPS and represented the glass transition temperature of glycerol [18, 37, 59].

It can be noted that the tan δ curves of PLA/MPLA/TPS blends revealed three thermal transitions (α, β, and γ). In α transition, for the blend without the plasticizer, a tan δ peak located at about 22°C could be ascribed to the glass transition temperature of the starch-rich phase [13, 18, 60], whereas that of the blend containing plasticizer occurred at 5°C. This indicates that the Tg value of the TPS phase has shifted to significantly lower temperatures after the plasticization due to the enhanced chain mobility of the starch molecules. Another tan δ peak, in α transition, located at 64°C, represented the glass transition temperature of PLA. The value is slightly lower than that of the pure PLA (67°C). This is a typical behaviour of plasticized thermoplastics, where plasticizers can reduce the Tg by increasing the free volume, and thus chain mobility of the polymeric molecules may also promote their crystallinity due to the enhanced chain mobility. These results are in agreement with thermal and mechanical properties (Tables 35).

As shown in Figures 11(a) and 11(b), the storage modulus for the various composites increased with the addition of silica nanoparticles. However, the modified silica is more efficient reinforcement for the investigated blend comparing with neat silica, due to the higher storage modulus (Table 6). Figures 11(c)11(f) show the curves of the loss factor (tan δ) as a function of temperature for the composites containing neat or modified silica, respectively. The addition of silica significantly increases α and β-relaxation temperatures of the TPS phase, due to the improved intermolecular interaction of TPS in both the starch-rich and starch-poor phases. However, the values of the above temperatures strongly depend on the silica functionality and content. It can be observed from Table 6 that the Ta1 of TPS increases dramatically from 5 to 20°C and to 30°C with 3 wt.% of neat or modified silica, respectively. Since modified silica particles were more reactive than neat silica, A-SiO2 increased the glass transition temperature higher than that of SiO2.

Moreover, Table 6 shows that the Ta of the PLA phase of the composites with neat silica is slightly higher as compared to that of the PLA/MPLA/TPS/P blend as well as to the composites with A-SiO2. However, there is practically no effect of the silica content and functionality on the α relaxation temperature.

Simultaneously, the β-relaxation of a glycerol-rich phase of TPS can be observed as a clear maximum in tan δ (Figure 11(c)) for the blends and represents the glass transition temperature of glycerol [59]. The β-relaxation appears as a maximum at −42°C in tan δ for the blend without plasticizer, whereas β-relaxation of the plasticized blend Tβ is about 10°C lower. This is related to increased interfacial interactions between the TPS phase and plasticizer.

Moreover, for the blend without the plasticizer, a tan δ peak located at about −107°C could be ascribed to the β transition temperature of the PLA, whereas that of the blend containing the plasticizer occurred at about −121°C. This indicates that the Tβ value of the PLA has shifted to lower temperatures after the plasticization. Functioning like a physical joint, the plasticizer enhanced the chain mobility of the PLA molecules and hence increased the free volume and reduced glass transitions of the blends. These results indicate that the polydimethylsiloxanol played a significant role as a plasticizer in the PLA/TPS blends.

4. Conclusions

Toughened PLA/MPLA/TPS/silica composites were successfully prepared by melt blending in the twin-screw corotating extruder. The morphological analysis of the plasticized blends showed that the addition of plasticizer resulted in less porous structure with increase in its content, suggesting improved toughness. More porous structure was observed for neat silica than that for the modified one. Moreover, at the same contents of silica, the shape of pores was more regular and the size was smaller in the composites with modified silica than with the neat one. Thermal and mechanical properties demonstrated that polydimethylsiloxanol can be used as the efficient plasticizer for the PLA/MPLA/TPS blend. DSC analysis showed that glass transition temperature and melt temperature were not significantly affected by the addition of silica. However, the strong effect of silica content and functionality on crystallization temperature and crystallinity of PLA/MPLA/TPS/P blend was observed. From TGA analysis, it is clear that the plasticizer greatly improved the initial thermal stability of the PLA/MPLA/TPS blend. However, the addition of silica had no effect on the thermal degradation of the plasticized blend. The tensile strength and tensile modulus of plasticized PLA/MPLA/TPS blends decreased as the content of plasticizer increased. However, the plasticized blend had higher tensile properties as compared to the unplasticized one. Meanwhile, the elongation at break reached up to 26%. The addition of silica significantly increased the elongation at break of the material (up to 91%). However, the improvement strongly depends on the silica content and functionality. For neat silica, elongation increased, whereas for the modified one, it decreased as silica content increased. The highest value was observed for 1 wt.% of the A-SiO2. This was due to the intensive interfacial interactions among the hydrogen bonds of the anhydride groups of the MPLA, the hydroxyl groups of the starch, and amine groups of the silica. Moreover, the composites showed lower tensile strength and tensile modulus than that of the plasticized PLA/MPLA/TPS blend, which increased with the increasing silica content. This was resulting from preferential or virtually unpredictable distribution characteristics of silica around the micropores within the toughened PLA/MPLA/TPS structure with respect to concentration and functionality. From DMTA, it was found that the modified silica is more efficient reinforcement than the neat one, due to the higher storage modulus of the composites.

The plasticized PLA/MPLA/TPS/silica composites can be one of the good candidates in the potential application, such as packaging materials, disposable goods, electronics materials, tissue engineering materials, surgical sutures, and drug delivery systems.

Data Availability

The presented research results are protected by patents PL 216 930 and PL 216 295.

Conflicts of Interest

The authors declare that there are no conflicts of interest regarding the publication of this paper.

Acknowledgments

The authors are very grateful to Professor Maria Zielecka for silica modification and the evaluation of the results and to Maciej Studziński, MSc, for providing DMTA analyses. The work was fonded by the statutory resources of the Industrial Chemistry Research Institute.