Sage Journals: Discover world-class research

Abstract

Recently, cellulosic fiber/PLA biocomposite has been respected as a sustainable material, potentially for industrial applications. A compatibilizer is a key factor in determination of biocomposite performance. In this work, microfibrillated cellulose (MFC) was differently modified with (1) triethoxyvinylsilane and (2) 3-(amino propyl) triethoxy silane and then grafted to epoxidized natural rubber (ENR) as compatibilizers. In which they were compounded with PLA (PLA blend) and followed to wood pulp fibre (PLA biocomposite). The mechanical, thermal and surface properties of PLA blend and PLA biocomposite were investigated. The incorporation of ENR grafted with MFC modified by silane (ENR-g-MFCs) reduced the crystallinity of PLA due to the penetration of ENR into the PLA phase. A 5% loading of vinyl-silanized MFC grafted to ENR (ENR-g-Vinyl MFC) blended increased the flexural strength and impact strength of PLA blend by almost 100%. Furthermore, the ENR-g-Vinyl MFC compatibilizer improved the impact resistance and flexural properties of the PLA/wood pulp biocomposite. The 5% loading of ENR-g-Vinyl MFC reduced the surface energy of the composite by 33%.

Keywords

Polylactic acid ENR grafted MFCs compatibility physicochemical properties thermal analysis

Introduction

Polylactic acid-based composites with cellulose fibre are sustainable materials that can replace conventional plastic composites in consumer and industrial markets. However, incompatibility between the cellulose fibre and PLA matrix,^1,2 water swelling,³ the inherent brittleness of PLA,⁴ increasing melt viscosity during compounding,⁵ and low impact resistance^3,6 are recurring problems that reduce the durability and service life of these composites. These problems can be reduced by improving the compatibility of the materials with compatibilizers or by reducing the surface free energy of fibres with hydrophobic chemical compounds^7–11 to promote interphase bonding. Since chemical compounds are expensive and create chemical waste compatibilization is a more appropriate and suitable approach, especially for industrial processes. Many different methods of compatibilization have been proposed for PLA biocomposites¹² but most have been based on grafting chemicals to polymers.

PLA is a brittle material, and compounding PLA with a stiff filler like a short cellulosic fibre lowers impact resistance. In previous work,¹³ showed that silanized microfibrillated cellulose grafted to epoxidized natural rubber (ENR-g-MFCs) improved compatibility between the components of a PLA and pulp fibre composite. ENR-g-MFCs was accessible to the PLA phase due to the silane groups on MFC, while free hydroxyl groups on MFC produced hydrogen bonds with cellulose in the pulp. ENR-g-MFCs performed two functions guiding the ENR phase into the PLA phase, and enhancing interfacial adhesion between pulp fibre and PLA. The flexural strength, flexural modulus and impact resistance of the composite were improved.

Therefore, in this study, as a continuation of previous research,¹³ the new comparison of two different types of compatibilizers (1) triethoxy vinyl silane treated MFC grafted ENR (ENR-g-Vinyl MFC) and (2) 3-aminopropyltriethoxysilane-treated MFC grafted ENR (ENR-g-Amino MFC), were investigated. Effect of ENR-g-MFCs as blending with PLA, was analyzed in terms of phase dispersion, physical, and chemical properties. Subsequently, biocomposites of PLA/ENR-g-MFCs/wood pulp fibre were manufactured. Similarly, mechanical, thermal properties, including kinetic, morphology including physicochemical and hydrolytic stability were analysed.

Materials and Methodology

Materials

Polylactic acid (3052D grade, NatureWorks, USA) with a density of 1.24 g/cm³. Epoxidized natural rubber (ENR 25% of substitution, GSP, Thailand). Long-fibre spruce Kraft pulp was kindly provided by Siam Chemical Group (Thailand). (2,2,6,6-Tetramethylpiperidin-1-yl)oxyl or (2,2,6,6-tetramethylpiperidin-1-yl)oxidanyl (TEMPO) (purity of 98%), triethoxyvinylsilane (purity of 98%), and 3-aminopropyltriethoxylsilane (purity of 98%) were from Sigma Aldrich. Sodium bromide (NaBr) (purity of 99.5%) was from QReC.

Production of Microfribrillated Cellulose

Microfibrillated cellulose (MFC) was produced according to the method presented in previous work.¹³ Long-fibre Kraft spruce pulp was processed by oxidation with TEMPO to produce anionic charges before disintegrating using a high-pressure homogenizer with 1500 bar.

Modification of MFC with Silanes

The MFC was silanized with triethoxyvinylsilane and 3-aminopropyltriethoxylsilane according to a method proposed in previous work.¹³ The mole ratio of silane to MFC was 0.2:1. The reaction was carried out for 24 h at room temperature (28–30°C). The completed reaction was precipitated with acetone. The modified MFCs were labelled vinyl-silanized MFC and amino-silanized MFC.

Grafting Modified MFCs to ENR

The ENR-g-MFCs used as compatibilizers were obtained by solution-blend casting.¹³ ENR was first dissolved in toluene and cast in a mold. The modified MFC was then added to the mold. The following three types of ENR-g-MFCs were produced: unmodified MFC grafted to ENR (ENR-g-MFC), vinyl-silanized MFC grafted to ENR (ENR-g-Vinyl MFC), and amino-silanized MFC grafted to ENR (ENR-g-Amino MFC).

Biocomposite Fabrications

Compounding of Biocomposites

PLA pellets were dried for 3 h at 80°C. The dried PLA pellets were placed in an internal mixer (Brabender, Germany) and melted for 3 min at 180°C with a rotor speed of 30–60 r/min. Then, 5 wt% of ENR-g-MFCs compatibilizer and 10 wt% of wood pulp fibre were added, and mixing continued for 10 min. The bulk biocomposite was then transferred to a grinding machine and granulated.

Biocomposite Sheet Fabrication

The granulated biocomposite was dried at 80°C for 3 h and transferred to a window mold measuring 10 × 10 × 0.2 cm³. Compression molding was carried out at 180°C with a pressure of 200 kg/cm² for 8 min. After compression, the molding was allowed to cool.

Characterization

Flexural Testing

The flexural strength and modulus of samples were determined using a universal testing machine (Instron machine Model 3365) at room temperature. Flexural testing was carried out in the three-point bending format in accordance with the ISO 178:2019 standard. All flexural tests were performed with a load cell of 1000 N, a support span of 64 mm and a crosshead speed of 5 mm/min. At least five samples were tested for each condition. Data were reported as average values with standard deviations.

Impact Resistance Test

Impact strength was determined in accordance with the ISO 179-1:2010 standard. The Zwick/Roell 5102 Pendulum Impact Tester was used with a 1 J pendulum to perform Charpy impact testing test with unnotched specimens. At least eight specimens were tested and data were reported as average values in kJ/m² with standard deviations.

Differential Scanning Calorimetry (DSC)

The PLA and the biocomposites were analysed by differential scanning calorimetry (DSC, Perkin Elmer DSC 7, USA) in nitrogen gas. Samples were first heated from 30 to 190°C at a rate of 10°C/min. Samples were then cooled from 190 to 30°C, at a rate of 10°C/min. In the final step, samples were heated in the same temperature range and at the same rate as in the first step. The percentage crystallinity (X_C) of samples was calculated according to the following formula Eq. (1):

% X c = \frac{Δ H_{m} x 100 %}{Δ H_{m}^{0} \times (1 - w) .}

(1)

where ΔH⁰_m was the theoretical melting enthalpy of 100% crystalline PLA (93 J/g),¹⁴ ΔH_m and ΔH_c were the melting and recrystallization enthalpy of the sample, while w was the total weight fraction of fibre and grafted ENR (0.15).

Thermogravimetric Analysis (TGA)

In the thermogravimetric analysis (TGA, Perkin Elmer TGA 8000), samples were scanned from 30 to 600°C under nitrogen gas. The kinetics of thermal decomposition were studied using heating rates of 5, 10, 15 and 20°C/ min. The activation energy of thermal decomposition (Ea) was calculated using the Flynn-Wall-Ozawa method (OWF) through a kinetic study of non-isothermal degradation, using the following maximum rate method Eq. (2):

\ln (\frac{β}{T m^{2}}) = \ln (\frac{A . R}{E a}) - (\frac{E a}{R . T m})

(2)

where β and Tm were the heating rate and temperature at maximum rate of decomposition, respectively, and Ea was calculated from the linear relationship of the slope.

Scanning Electron Microscope

The cross-sectional morphology of biocomposite samples was observed by SEM (FEI Quanta 400 and Hitachi High-Tech, SU3900, Japan) to investigate the effects of the different compatibilizers. Samples were coated with a thin layer of gold before analysis.

Contact Angle and Surface Energy Analysis

The surface energy of samples was determined using contact angle measuring apparatus (Dataphysics OCA 15EC). The measurements were taken at room temperature using droplets of distilled water, ethanol and dimethylformamide, placed on samples measuring 1 cm × 1 cm. Sessile drop volume was maintained at 5 μL using a microsyringe. The sample measurement was repeated six to nine times at different positions on the same sample.

Gel Permeation Chromatography Analysis

The molecular weight distributions of biocomposite samples were evaluated using gel permeation chromatography (GPC) to determine molecular weight losses in the PLA matrix. Samples were dissolved in tetrahydrofuran (THF) and analysed in the following condition: a Shodex column (GPC KF-806M, Japan) with an internal diameter of 8 mm and a length of 300 mm, a mobile phase of THF at a flow rate of 1 mL/min, a temperature of 40°C. Polystyrene standards were used for calibration.

Dynamic Mechanical-Thermal Analysis

Dynamic mechanical thermal analysis (DMTA 1, Mettler Toledo, Switzerland) using the tension method was conducted to measure the storage and loss modulus. The specimens were heated from 25°C to 140°C at a rate of 3°C/min, with a displacement of 5 µm and a frequency of 1 Hz.

Assessment of Hydrolytic Stability

The resistance of the PLA biocomposite to water was determined from a hydrolytic stability analysis. The biocomposite was immersed in water at 35°C for 24 h. The stability of the saturated samples was analysed using DMTA. The viscoelastic behaviour and glass transition temperature of the biocomposite were analysed before and after water uptake.

Results and Discussion

Interaction Analysis of PLA/ENR-g-MFCs via DMTA

DMTA profiles of all the PLA/ENR-g-MFCs blends bases (Figure 1(a)) exhibited tan δ peaks at roughly 60–65°C, indicating the glass transition temperature (Tg) of PLA in each blend. However, there were tan δ peaks at around 80°C–110°C, indicating the recrystallization of PLA temperature increased. The height of a tan δ peak is known as the damping factor, which is derived from E″/E′ (loss modulus/storage modulus), and predicted the behaviour of motion by comparing the damping factor of materials. The damping factors of the PLA/ENR-g-MFCs at the tan δ peaks of recrystallization were compared. The ENR-g-Amino MFC base had the high damping factor, indicating the most viscous phase motion, due to the presence of an amorphous phase of PLA that could be a result of chain scission degradation during the aminolysis of silane amino groups to COOR of PLA as compounding between PLA and ENR-g-Amino MFC base. The ENR-g-MFC base showed the lowest damping factor, indicating elastic motion due to the crystalline phase of PLA. The ENR-g-MFC and ENR-g-Vinyl MFC bases should be more stable than the ENR-g-Amino MFC base as they displayed fewer signs of degradation. This issue will be discussed again in the activation energy analysis.

Figure 1.

Tan δ and storage modulus of PLA/ENR and PLA/ENR-g-MFCs blended.

At sub-zero temperatures, the MFCs as grafted on ENR restricted the molecular motion of the ENR phase. This behaviour was indicated by the lower damping factors of these blends compared with the ENR bases. Interestingly, the Tg of the ENR phase for all ENR-g-MFCs blends were decreased to lower temperature, roughly – 50°C. Probably, this was due to the plasticizing effect of the grafted MFCs increasing free volume of ENR phase. Also, the storage modulus of the PLA/ENR-g-MFCs were higher than the storage modulus of the ENR bases (Figure 1(b)), confirming the restricting and reinforcing effect on the elasticity of the grafted MFCs and, therefore, their assimilation into the ENR phase.

Effect of ENR-g-MFCs as Compatibilizers for Mechanical Properties

Effect of compatibilizer base ENR-g-MFCs to mechanical properties for PLA blends and biocomposites, is presented in terms of comparative correlation to neat PLA as Figure 2. The impact strength of the ENR-g-MFCs as blended with PLA showed notable improvements compared to the impact strength of the ENR base; in particular, the ENR-g-Vinyl MFC base (Figure 2(a)). ENR improved the flexural strain of all blends. The ENR-g-Vinyl MFC base showed the highest flexural strength and strain, possibly due to the CH = CH₂ bond on the chain ends of vinyl silane, which could engage with PLA molecules. The double bond was cleavage releasing radicals that bonded to PLA at the α-carbon as illustrated in Scheme 1(a).¹⁵ The ENR-g-MFC base showed the second best mechanical properties. The poor performances of ENR-g-Amino MFC could have been caused by the shortening of PLA molecules during aminolysis would be described later, but the effectiveness of this compatibilizer would be further investigated when used with the PLA/wood pulp fibre biocomposite.

Figure 2.

Comparative analysis of mechanical properties, effect of compatibilizer base ENR-g-MFCs for (a) PLA blended, (b) PLA biocomposite.

Scheme 1.

Proposed bonding interactions between cellulose and PLA via the compatibilizers; (a) ENR-g-Vinyl MFC and (b) ENR-g-Amino MFC.

Figure 2(b), presents the capability of ENR-g-MFCs as functioning of compatibilizer for the fibre reinforcing in biocomposites. Notably, only the ENR-g-Vinyl MFC and ENR-g-Amino MFC bases could improve the flexural strength up to 37% and 35%, respectively (Table 1). Whilst the ENR and ENR-g-MFC bases showed the negative effect by reduction of flexural strength down to −25% and −15%, respectively. This crucially evidenced indicating the high efficiency of ENR-g-Vinyl MFC and ENR-g-Amino MFC as the compatibilizers for short fibre biocomposites. Additionally, the substantial increase of impact resistance observed in ENR-g-Vinyl MFC and ENR-g-Amino MFC base biocomposites up to 107% and 67%, respectively, explicitly indicated the great bonding improvement among PLA, ENR, and wood pulp fibre phases.

Table 1.

Information of mechanical: effect of ENR-g-MFCs as compatibilizers on the properties of PLA blends and biocomposites.

PLA/ENR-g-MFCs	Pulp fibre (wt%)	Flexural modulus (MPa)	Changed (%)^a	Flexural strength (MPa)	Changed (%)^a	Flexural strain (%)	Changed (%)^a	Impact strength (kJ/m²)	Changed (%)^a
Neat PLA	0	2385 ± 110	N/A	32.20 ± 0.77	N/A	1.39 ± 0.05	N/A	3.78 ± 0.21	N/A
ENR	0	2287 ± 12	+18	37.34 ± 2.33	−35	1.79 ± 0.16	−49	2.93 ± 0.76	+6
ENR	10	2698 ± 125	+18	24.30 ± 1.35	−35	0.92 ± 0.04	−49	3.11 ± 0.64	+6
ENR-g-MFC	0	2283 ± 35	+22	40.11 ± 2.08	−32	1.97 ± 0.10	−49	5.54 ± 0.57	−30
ENR-g-MFC	10	2782 ± 102	+22	27.29 ± 3.29	−32	1.01 ± 0.15	−49	3.88 ± 0.81	−30
ENR-g-vinyl MFC	0	2233 ± 19	+14	42.02 ± 2.19	+5	2.12 ± 0.20	−8	7.37 ± 0.41	+4
ENR-g-vinyl MFC	10	2551 ± 28	+14	44.01 ± 2.13	+5	1.96 ± 0.13	−8	7.65 ± 0.67	+4
ENR-g-amino MFC	0	2330 ± 23	+8	34.81 ± 0.89	+24	1.61 ± 0.07	+20	4.27 ± 0.26	+48
ENR-g-amino MFC	10	2519 ± 51	+8	43.33 ± 2.48	+24	1.93 ± 0.11	+20	6.30 ± 0.74	+48

^a Changed (%)' indicates the percentage of increment (+) or decrement (−) of mechanical properties of biocomposite (fibre 10 wt%) relatively to blended (fibre 0 wt%).

In overview, ENR-g-Vinyl MFC greatly performs as a compatibilizer in combination of pulp fibre – PLA phases as increasing of flexural properties and also in particular, impact resistance. Additionally, it also functions an impact modifier for PLA.

Compatibility Analysis of PLA/ENR-g-MFCs/Wood Pulp Fibre via SEM Images

The dispersion of ENR and ENR-g-MFCs in the PLA phase was considered to be the principal indicator of compatibility between the components and to influence mechanical and thermal properties along with water swelling, and eventually durability.

Phase dispersion of ENR in PLA could be investigated with SEM. The contrasting images of surface morphologies identified the particular characteristics of each formulation. In Figure 3(a), the ENR base exhibited a notably rougher cross-sectional surface, with a larger average particle size of 0.617 μm. In contrast, ENR-g-Vinyl MFC and ENR-g-Amino MFC, shown in Figures 3(c) and 3(d), respectively, presented smoother surfaces with reduced particle sizes of 0.543 μm and 0.478 μm. In a previous work,¹³ cellulose nanofibre or MFC grafted to ENR were reported to act like guiding molecules bringing ENR into the PLA phase. Hence, the improvement of compatibility seen in Figures 3(c) and 3(d) indicated improved engagement and interaction between PLA and ENR phases.

Figure 3.

SEM images show cross-sections of PLA blend and biocomposite: (a) ENR, (b) ENR-g-MFC, (c) ENR-g-Vinyl MFC, (d) ENR-g-Amino MFC, (e) ENR/fibre, (f) ENR-g-MFC/fibre, (g) ENR-g-Vinyl MFC/fibre, and (h) ENR-g-Amino MFC/fibre.

SEM images of Figures 3(e)–3(h) revealed interfacial adhesion of wood pulp fibre in matrix phase of biocomposites. The ENR/fibre (PLA/ENR/wood pulp fibre biocomposite) failed, as shown by fibre pull out in Figure 3(e) indicating the least interfacial shear strength of fibre – matrix. Meanwhile, the failure surface of the ENR-g-MFC/fibre biocomposite as demonstrated in Figure 3(f), appeared slightly fibre-matrix apart. This confirmed the better interfacial adhesion in the biocomposite due to ENR-g-MFC despite without silanizing. However, the voids visible in both PLA matrices indicated a discontinuous phase, which weakened the mechanical properties of the biocomposite (Figure 2(b)). The wood pulp fibre and PLA bonded with strong interfacial adhesion in the ENR-g-Vinyl MFC/fibre biocomposite (Figure 3(g)). Despite the hydrophilicity of the wood pulp fibre, fewer and shallower voids were present, indicating better compatibility between all phases. This supported the reinforcement mechanism that efficiently transferred loads from the matrix to the fibre, leading to the improved flexural and impact strength of the material. Similarly, the ENR-g-Amino MFC/fibre showed improved interface and mechanical properties (Figure 3(h)). Amino groups at chain end of silane could also covalently bond via aminolysis to carbonyl groups of PLA. By a dual linkage,¹⁶ could provide tight bonds between PLA and wood pulp fibre by the mechanism proposed in Scheme 1(b).

Another issue should be addressed here. It is known that PLA and short fibre PLA biocomposites behave as brittle and stiff materials. The higher flexural strain with lower elastic modulus of the ENR-g-Vinyl MFC/fibre and ENR-g-Amino MFC/fibre biocomposites indicate increased toughness, which is a desirable property for engineering applications where fatigue may be a concern.

The ENR-g-Vinyl MFC/fibre and ENR-g-Amino MFC/fibre base showed different levels of improvement in mechanical properties. Although the ENR-g-Vinyl MFC/fibre base showed the highest impact strength (7.65 kJ/m²), ENR-g-Amino MFC base increased the flexural and impact strength of the biocomposite by 24% and 48%, respectively (Table 1). The ENR-g-Vinyl MFC could support the fibre reinforcement only 5% and 4%, respectively. As described earlier, the bonding between PLA (COOR) on main chains and amine (NH₂) at the chain ends of silane was achieved via aminolysis but the resulting reduction of molecular weight (Table 3) weakened mechanical properties.

Effect of ENR-g-MFCs as Compatibilizers for Thermal Properties

The results of the DSC analysis of PLA/ENR-g-MFCs blends were presented in Table 2. The glass transition temperatures of the samples were in a narrow range of 56.6 °C–58°C. The crystallization temperatures (Tc) of the ENR-g-MFCs bases were higher than the Tc of the ENR base due to the penetration of ENR between PLA molecules. Since the ENR molecules could retard crystallization, more energy was needed to perfect the crystals and %crystallinity among the ENR-g-MFCs bases was 6−10% compared with 21% of the ENR base. Therefore, the grafting of ENR with MFC, Vinyl MFC, and Amino MFC altered the mechanical and thermal properties of the PLA/ENR blend by bringing ENR into the PLA phase.

Table 2.

Information of thermal analysis: effect of ENR-g-MFCs as compatibilizers on the properties of PLA blends and biocomposites.

PLA/ ENR-g-MFCs	Pulp fibre (%wt)	T_g (^oC)	T_c (^oC)	T_m (^oC)	ΔH_m (J/g)	X_c (%)	Increment of %crystallinity (%)
ENR	0	57.6	125	150	20	21	5
ENR	10	54.1	122	154	29	26	5
ENR-g-MFC	0	57.7	130	153	6	6	23
ENR-g-MFC	10	54.9	121	154	32	29	23
ENR-g-vinyl MFC	0	57.6	129	153	7	7	11
ENR-g-vinyl MFC	10	57.4	127	152	10	18	11
ENR-g-amino MFC	0	56.6	130	152	10	10	13
ENR-g-amino MFC	10	57.1	126	152	25	23	13

Interestingly, incorporating the wood pulp fibre into all PLA/ENR-g-MFCs for biocomposite production, this could considerably increase %crystallinity over the PLA/ENR. By the hypothesis, grafting MFCs on ENR not only toughened PLA but the free hydroxyl of MFCs could produce a strong hydrogen bonding with pulp fibre as a reinforcement fibre for the biocomposite as well. The improved interfacial adhesion due to ENR-g-MFCs (Figures 3(f–h)) as the compatibilizers over ENR base (Figure 3(e)), was prominently subsequent to increase % crystallization, especially, the ENR-g-MFCs/fibre biocomposite showing the maximum of 23%. Theoretically, the cellulose fibre could perform like a nucleating agent,¹⁷ due to a contrasting surface property between cellulose fibre (hygroscopic) and PLA (hydrophobic), this could activate crystallization for PLA molecules. This means that the improvement of compatibility causing fibre closer to matrix and activating crystallization for PLA molecules.

Thermal Activation Energy Analysis

Thermal activation energy could be exploited as an index for evaluation the stability of materials like biocomposites,¹⁸ in which it is always dependent on molecular weight.¹⁹ Hence, stabilization of molecular weight of PLA matrix is essentially considered. However, compounding PLA to cellulose fibre production of composite, always reduces themolecular weight of PLA due to hydrolytic degradation or thermo oxidative degradation.^20,21 Therefore, in this research, the Ozawa-Flynn-Wall model was employed to investigate efficiency of compatibilizer based ENR-MFCs on molecular weight of PLA matrix as relative to the activation energy (Figure 4).

Figure 4.

The non-isothermal decomposition; the relationship between $\ln (\frac{β}{T m^{2}})$ and $(\frac{1}{T m})$

Ea was calculated from a kinetic study of non-isothermal degradation via Arrhenius plots (Ozawa-Flynn-Wall analysis), constructed from Eq. (2): It always indicates the ability of polymers or composites to resist thermal decomposition. Ea is correlated with bond energy, molecular weight, the mechanism of thermal decomposition and thermal diffusion. The Ea of the ENR-g-MFC/fibre was the highest, while the Ea of the ENR-g-Amino MFC/fibre was the lowest (Table 3). This could be a consequence of the low molecular weight of the ENR-g-Amino MFC/fibre. Interestingly, the improved compatibility of the biocomposite due to ENR-g-Vinyl MFC/fibre and ENR-g-Amino MFC/fibre reduced Ea compared with the ENR/fibre and ENR-g-MFC/fibre. ENR mainly consists of ethylene groups (CH = CH), and due to the richness of π bonds in ENR, it is readily cleavable and releases radicals into the occupied phase at decomposition temperatures. Since interfacial adhesion between PLA, ENR-g-MFCs and wood pulp fibre was good, those radicals would be able to diffuse rapidly through the composite, accelerating the propagation of thermal decomposition. ENR-g-Amino MFC/fibre had the lowest molecular weight, which was a main factor in the rapid decomposition of the corresponding biocomposite base to a more amorphous phase which facilitated the diffusion of radicals through the composite.

Table 3.

Activation energy and molecular analysis of PLA biocomposites.

PLA/ ENR-g-MFCs /Pulp fibre	Activation energy (Ea, kJ/mol)	Mn (g/mol)	Mw (g/mol)	MWD
ENR/fibre	133	61,939	122,762	2
ENR-g-MFC/fibre	148	71,781	133,798	1.9
ENR-g-vinyl MFC/fibre	126	70,283	116,964	1.7
ENR-g-amino MFC/fibre	99	50,196	73,020	1.5

The possible reaction between PLA and ENR-g-MFCs could be explained by analysis of the molecular weight using GPC (Table 3). The ENR-g-Amino MFC/fibre had the lowest both Mn and Mw, which was attributed to the hydrolysis of amino groups (aminolysis) at chain ends (Scheme 1(b)). Since the amino groups of 3-aminopropyl triethoxysilane were more susceptible than the vinyl groups of vinyltrimoxysilane to carbonyl groups of PLA, degradation via hydrolysis was more advanced in the ENR-g-Amino MFC/fibre. Possibly, aminolysis could occur randomly in main chains of PLA (endo-hydrolysis), which would diminish the mechanical properties of the material (Figure 2). The Mn of the ENR/fibre was lower than the Mn of the ENR-g-MFC/fibre and ENR-g-Vinyl MFC/fibre. Possibly, the lower Mn of the ENR/fibre was caused by shear forces during compounding. The friction generated heat, and PLA molecules were degraded through chain scission. In extreme conditions, the π bond of the ethylene groups of ENR (75% by mole) could be cleaved to release high-energy radicals that shortened PLA molecules by thermal oxidation. In another hand, the molecular structures ENR-g-MFC/fibre consisted of the long chains of MFC, which might act as a barrier between PLA and ENR. Subsequently, molecular friction between ENR and PLA was reduced during processing, and thermal oxidation was mitigated. In the case of the ENR-g-Vinyl MFC/fibre, PLA and ENR-g-Vinyl MFC base (Scheme 1(a)) could engage since the CH = CH₂ groups at chain ends of vinyl silane could act like guiding molecules due to hydrophobicity. Furthermore, as ascribed previously, the π bond (CH = CH₂) of vinyl silane could be cleaved to radicals building covalently bond to α-carbon of PLA, thus taking ENR to be more compatible with PLA. This could confirm the compatibilizing potential of vinyl silanized MFC.

The relationship between thermal activation energy and molecular weight of the biocomposites is illustrated in Figure 5. The number average molecular weight (Mn), weight average molecular weight (Mw), and thermal activation energy of the biocomposites are summarized in Table 3. A linear relationship is observed molecular weight and thermal activation energy, particularly for Mw, with an R² value of 0.96. The addition of ENR-g-MFCs and pulp fibres further increased the molecular weight and simultaneously promoted crosslinking, resulting in a more rigid structure. Consequently, the thermal activation energy increased with rising molecular weight. In thermal degradation, the polymer chains break down into smaller fragments. The energy required for this process, or the activation energy, is influenced by the polymer’s molecular weight. For higher molecular weight polymers, the activation energy tends to be higher due to the increased bond strength and chain entanglements. According to research by Behera et al.,²² the thermal activation energy of PLA increases with molecular weight. This makes high-molecular-weight PLA more suitable for applications requiring thermal resistance. Welle and Franz²³ also demonstrated a similar trend, showing that higher molecular weight of Polyethylene Terephthalate (PET) correlates with greater thermal stability, as a higher activation energy is required to degrade the polymer chains. Some studies suggest a direct proportionality between activation energy and molecular weight in certain polymers. In other cases, the relationship may be nonlinear, depending on the polymer type, the presence of additives, and other factors.

Figure 5.

Relationship between activation energy and molecular weight of biocomposites.

Physicochemical Analysis

The contact angle and surface energy analysis of the PLA/wood pulp biocomposite showed that polarity was affected by the type of ENR-g-MFCs used to compatibilizer the components. The ENR-g-Vinyl MFC/fibre produced the highest contact angle, followed by ENR-g-Amino MFC/fibre, ENR/fibre and ENR-g-MFC/fibre (Table 4). Meanwhile, the dispersive force was highest in the ENR/fibre. Therefore, the ENR-g-Vinyl MFC/fibre could best compatibilizer the fibre and PLA in the PLA/wood pulp biocomposite. Free hydroxyl groups on cellulose molecules of wood pulp fibre could form strong hydrogen bonds with free hydroxyl groups in ENR-g-Vinyl MFC while CH = CH₂ groups on ENR-g-Vinyl MFC could link with PLA. The mechanism of linkage was presented in Scheme 1(a).

Table 4.

Contact angle and surface energy of biocomposite base different ENR-g-MFCs.

PLA/ENR-g-MFCs/pulp fibre biocomposite	Contact angle (degree)	Surface energy (mN/m)	Polar (mN/m)	Dispersive (mN/m)
ENR/fibre	70	34	17	17
ENR-g-MFC/fibre	69	33	22	12
ENR-g-vinyl MFC/fibre	89	23	7	16
ENR-g-amino MFC/fibre	80	27	12	14

Hydrolytic Stability

The PLA/wood pulp biocomposites usually absorb ambient moisture or water due to the inherently hygroscopic nature of cellulose fibre. The water molecules cause the composite to swell, cracking is initiated and further damage follows. Additionally, due to hydrolysability of PLA, water–matrix interactions accelerate hydrolytic degradation, reducing molecular weight. Then, the degraded composite would fail.

In order to investigate hydrolytic resistance, the biocomposite was immersed in water at 35°C for 24 h. Dynamic mechanical–thermal analysis (DMTA) was deployed to study the viscoelastic properties of the samples. The peak of tan δ of the ENR/fibre shifted from 65.35°C to 60.85°C after immersion (Figure 6). This shift could indicate the phase behavior as occupying in the composite. The ENR/fibre was significantly affected by water. The reduction of Tg with increasing of damping factor (peak height) was evidence of the strong interaction between water and the PLA matrix. The contact angle and surface free energy analysis (Table 4) showed that the ENR/fibre was more hydrophilic than the ENR-g-Vinyl MFC/fibre and ENR-g-Amino MFC/fibre. Absorbed water could activate hydrolysis in the greater amorphous phase. Furthermore, ENR behaved like a plasticizer to increase the amorphous phase as well. The viscosity of PLA was then increased by the damping factor. On the other hand, the ENR-g-MFC/fibre, ENR-g-Vinyl MFC/fibre and ENR-g-Amino MFC/fibre all showed better stability as their tan δ profiles hardly changed.

Figure 6.

Tan δ of PLA/wood pulp biocomposites blended with ENR-g-MFCs before and after immersion in water for 24 h.

The ENR-g-MFC/fibre, ENR-g-Vinyl MFC/fibre and ENR-g-Amino MFC/fibre also demonstrated higher glass transition temperatures than the ENR/fibre, especially the ENR-g-Vinyl MFC/fibre (69.45°C before immersion and 69.35°C after). The almost unchanged Tg of this biocomposite could be attributed to the restriction of PLA by the interaction with cellulose, probably both cellulose and nanocellulose. Furthermore, the hardly changed Tg after immersion confirmed the resistance of PLA to water. The penetration of hydrophobic ENR molecules also protected PLA from water. The water resistance of the ENR-g-Vinyl MFC/fibre biocomposite could prolong the service lifetime of final products.

Conclusions

Vinyl-silanized MFC grafted to epoxidized natural rubber (ENR-g-Vinyl MFC) was an effective compatibilizer for a biocomposite of polylactic acid and wood pulp. The compatibilizer acted as a guiding molecule, facilitating compatibilization between PLA, ENR and cellulose phases during blending. Improved interfacial adhesion was a primary consequence. Although amino-silanized MFC grafted to ENR showed the best compatibilizing effects with 10% wt of wood pulp fibre filler, due to aminolysis, this formulation had the lowest molecular weight PLA phase and the lowest activation energy. However, both modified MFCs grafted to ENR, improved the hydrophobicity of the PLA/wood pulp biocomposite and could enable the use of this biocomposites in areas of high relative humidity.

Footnotes

Acknowledgements

The authors gratefully acknowledge Faculty of Science, Prince of Songkla University (PSU), Thailand for the provision of equipment and facility support.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

Thorsak Kittikorn

Suding Kadea

Rattanawadee Hedthong

Data Availability Statement

The data will be made available on reasonable request.*

References

Liu

Jiang

, et al. Effects of modifications of bamboo cellulose fibers on the improved mechanical properties of cellulose reinforced poly(lactic acid) composites. Compos B Eng 2014; 62: 191–197.

. Mechanical and biodegradable properties of l-lactide-grafted sisal fiber reinforced polylactide composites. J Reinforc Plast Compos 2014; 33: 2034–2045.

Oliver-Ortega

Tarrés

Mutjé

, et al. Impact strength and water uptake behavior of bleached kraft softwood-reinforced PLA composites as alternative to PP-based materials. Polymers (Basel) 2020; 12: 2144.

Farah

Anderson

Langer

. Physical and mechanical properties of PLA, and their functions in widespread applications-A comprehensive review. Adv Drug Deliv Rev 2016; 107: 367–392.

Teixeira

Covas

Hilliou

. In-process assessment of clay dispersion in PLA during melt compounding: effects of screw speed and filler content. Polym Degrad Stabil 2020; 177: 109190.

Gao

Qiang

. Fracture surface morphology and impact strength of cellulose/PLA composites. Materials (Basel) 2017; 10: 624.

Spiridon

Darie

Kangas

. Influence of fiber modifications on PLA/fiber composites. Behavior to accelerated weathering. Compos B Eng 2016; 92: 19–27.

Peltola

Immonen

Johansson

, et al. Influence of pulp bleaching and compatibilizer selection on performance of pulp fiber reinforced PLA biocomposites. J Appl Polym Sci 2019; 136: 47955.

Debeli

Tebyetekerwa

Hao

, et al. Improved thermal and mechanical performance of ramie fibers reinforced poly(lactic acid) biocomposites via fiber surface modifications and composites thermal annealing. Polym Compos 2018; 39: E1867–E1879.

10.

Mohanty

Nayak

. Effect of poly (lactic acid)‐graft‐glycidyl methacrylate as a compatibilizer on properties of poly (lactic acid)/banana fiber biocomposites. Polym Adv Technol 2016; 27: 515–524.

11.

Masek

Cichosz

Piotrowska

. Biocomposites of epoxidized natural rubber/poly(lactic acid) modified with natural fillers (Part I). Int J Mol Sci 2021; 22: 3150.

12.

Olonisakin

Fan

Xin-Xiang

, et al. Key improvements in interfacial adhesion and dispersion of fibers/fillers in polymer matrix composites; focus on PLA matrix composites. Compos Interfaces 2022; 29: 1071–1120.

13.

Kittikorn

Kadea

Hedthong

. Effect of epoxidized natural rubber-grafted-modified microfibrillated cellulose to compatibility in wood pulp/polylactic acid biocomposite: mechanical–thermal and durability analysis. J Polym Environ 2023; 31: 2473–2485.

14.

Fischer

Sterzel

Wegner

GKZZ

. Investigation of the structure of solution grown crystals of lactide copolymers by means of chemical reactions. Kolloid-Z u Z Polymere. Kolloid-Z u Z Polymere 1973; 251: 980–990.

15.

Schneider

Bourque

Narayan

. Moisture curable toughened poly (lactide) utilizing vinyltrimethoxysilane based crosslinks. Express Polym Lett 2016; 10: 799–809.

16.

Meng

Nguyen

Tekinalp

, et al. Supertough PLA-silane nanohybrids by in situ condensation and grafting. ACS Sustainable Chem Eng 2018; 6: 1289–1298.

17.

. Nucleation roles of cellulose nanocrystals and chitin nanocrystals in poly (ε-caprolactone) nanocomposites. Int J Biol Macromol 2022; 205: 587–594.

18.

Oza

Ning

Ferguson

, et al. Effect of surface treatment on thermal stability of the hemp-PLA composites: correlation of activation energy with thermal degradation. Compos B Eng 2014; 67: 227–232.

19.

Wang

Liu

Apostolidis

, et al. Rheological behavior and its chemical interpretation of crumb rubber modified asphalt containing warm-mix additives. Transp Res Rec 2018; 2672: 337-348.

20.

Limsukon

Rubino

Rabnawaz

, et al. Hydrolytic degradation of poly (lactic acid): unraveling correlations between temperature and the three phase structures. Polym Degrad Stabil 2023; 217:110537.

21.

Cuadri

Martín-Alfonso

. Thermal, thermo-oxidative and thermomechanical degradation of PLA: a comparative study based on rheological, chemical and thermal properties. Polym Degrad Stabil 2018; 150: 37–45.

22.

Behera

Chang

Chiu

, et al. Characterization of poly (lactic acid) s with reduced molecular weight fabricated through an autoclave process. Polym Test 2017; 60: 132–139.

23.

Welle

Franz

. Diffusion coefficients and activation energies of diffusion of low molecular weight migrants in poly (ethylene terephthalate) bottles. Polym Test 2012; 31: 93–101.

Comparative evaluation of different compatibilizers based ENR-g-silanized MFCs for wood pulp/PLA: Analysis of compatibility and stabilities