CHAPTER 5. BIO-COMPOSITES OF POLY(LACTIC ACID) WITH NANOCELLULOSE

BIO-COMPOSITES OF POLY(LACTIC ACID) WITH

NANOCELLULOSE

5.1 Introduction

Recently, increasing attention has been given to the preparation of nanocellulose (NC) reinforced polylactic acid (PLA) composites with the goal of obtaining fully biodegradable nanocomposites materials. However, to promote the applicability of these materials on large scale, the development of practicable methods of industrial processing is necessary. By means of traditional processing techniques, homogeneous dispersion of cellulose nanoparticles in polymers is very difficult to achieve, due to high tendency of NC to form aggregates [182, 183].

Nanocellulose can be prepared from a variety of sources e.g. microcrystalline cellulose, bacterial cellulose, cotton, hemp, tunicin etc. by (i) mechanical disintegration process such as high pressure homogenizing and grinding of microcrystalline cellulose, and (ii) chemical process such as acid hydrolysis of microcrystalline cellulose. Fig. 5.1 shows a flow diagram for the preparation of nanocellulose. Prepared cellulose nanoparticles may have rod, sphere, and network morphologies (Fig. 5.2).

Fig. 5.2 – Cellulose nanoparticles in the forms of (a) rods (SEM image), (b) spheres (SEM image) and (c) porous network (TEM image) [184]

For the final properties of composite, dispersion of cellulose nanoparticles and their chemical compatibility with polymer matrix are very important issues. The presence of hydroxyl groups on the cellulose nanoparticles surfaces and the high specific surface area may be the reason for their aggregation. At the same time, the stress transfer between the cellulose nanoparticles is facilitated by hydrogen-bonding and van der Waals interactions [183]. Moreover, the reinforcing effect of cellulose nanoparticles in polymer nanocomposites has been seen to follow a percolation model, suggesting that the stress–strain transfer process is dominated by the formation of a continuous network and strong interactions between the cellulose nanoparticles [185, 186]. However, it is expected that proper dispersion and distribution of nanoparticles in the polymer matrix, as well as in the processing solvent, is a prerequisite for creating polymer-cellulose nanoparticle nanocomposites that display a significant mechanical reinforcement [187].

Owing to the good stability of cellulose nanoparticles suspension in water, the mixing of aqueous polymer solution or emulsions with nanocellulose suspension and subsequent film casting has been the preferred method utilized in the production of polymer-nanocellulose nanocomposites. The development of the processing technologies in which non-aqueous polymers are used as matrix is one of the biggest challenges today and in the near future. In order to improve the dispersion of cellulose nanoparticles in non-aqueous polymer solution or suspension, several strategies have been adopted, including the use of surfactants [31, 188]and chemical modifications via reaction of the hydroxyl groups [24, 189-193]. In spite of great improvement in the dispersion of nanocellulose in non-aqueous polymer solution, the surface functionalization can reduce or even suppress the nanocellulose-nanocellulose interactions and thereby limit their reinforcing effect. Grunert and Winter [190] have reported for CAB-cellulose nanowhisker composites that native nanocellulose has a superior reinforcing capability as compared to silylated ones.

a

_b

Since, it is difficult to disperse them uniformly in a non-polar medium, the processing of cellulose nanocomposites was first limited to solvent casting [21-29, 194]. Recently, melt compounding processing method has been developed [30-35, 37]. The nanocelluloses have a very high surface area and a large tendency to aggregate when dried. To avoid aggregation during processing, Jonoobi et al [195] first mixed NC with a small amount of PLA in a solvent mixture of acetone/chloroform and then prepared composites by melt-mixing the dried mixture with PLA in the extruder. If the aqueous medium could be used in place of organic solvents then it might be the most economical solution.

An alternative solution can be to use an aqueous emulsion of poly(vinyl acetate) (PVAc), which is miscible with PLA in all over the compositions [147]. Moreover, aqueous emulsion of PVAc can be prepared easily by free radical polymerisation of vinyl acetate in aqueous nanocellulose suspension. The mixture of nanocellulose and PVAc emulsion can be dried and followed by melt-mixed with PLA in brabender or extruder. Thus, the optimum dispersion of nanocellulose in PLA can be obtained. It must be noticed that blends of PLA and PVAc are biodegradable [147]. The aim of the present work was to analyse the effect of different processing strategies of PLA/NC nanocomposites on the phase dispersion of NC in PLA as well as on the thermal and mechanical properties of these systems. To improve the dispersion of NC in PLA different approaches such as functionalization of PLA and NC by glycidyl methacrylate or inclusion of NC in poly(vinyl acetate) (PVAc/NC) were applied.

5.2 Experimental

5.2.1. Materials

Table 5.1 Molecular structures of the employed materials

Material Molecular structure

PLA O CH CH₃ C O n Vinyl acetate H₂C CH O C O H3C Glycidyl methacrylate C CH2 CH₃ C O O H2 C CH O H2C Benzoyl peroxide C O O O C O

Poly(lactic acid) (PLA) polymer 3051D, specific gravity 1.25, MFR = 10-25 g/10 min (210ºC/2.16Kg), crystalline melt temperature = 150 – 165 °C, glass transition temperature = 55-65 °C, tensile yield strength 48 MPa were supplied by NatureWorks. Microcrystalline cellulose, vinyl acetate (purity ˃99%), glycidyl methacrylate (GMA) (purity ˃97%) and di-benzoyl peroxide (DBPO) were purchased from Sigma Aldrich, Italy. List of the employed materials with their molecular structure are reported in Table 5.1.

5.2.2. Preparation

5.2.2.1 Nanocellulose preparation

Cellulose nanofibres were prepared by the acid hydrolysis of commercial microcrystalline cellulose. The acid hydrolysis was carried out with sulphuric acid (H2SO4) solution (60 wt.%) at 45 ºC under continuous stirring. The fibre to liquor

ratio was 1:10 (10 ml of sulphuric acid for 1 g of cellulose). After 30 minutes, the hydrolysis was quenched by adding 5 fold excess distilled water (50 ml) to the reaction mixture. The resulting mixture was cooled to room temperature and centrifuged. The fraction was washed with the addition of distilled water and centrifuged several times until the pH became 7. The suspension was then sonicated for 30 min in an ice-bath to avoid over heating. The newly generated suspension was stored in refrigerator at 4 °C.

5.2.2.2 Functionalization of PLA

2 grams of dried PLA was dissolved in 100 ml of toluene. After complete dissolution of PLA about 100 mg of initiator (DBPO) and 1 gram of GMA were added in solution and reaction was continued at 80 °C for 6 hours. The solution was cooled and poured into 300 ml methanol. The precipitation was filtered and washed with fresh methanol, finally dried at 90 °C for 12 hours under vacuum. Grafting was confirmed by FT-IR analysis. Grafting percentage (17.6%) and grafting efficiency (36%) were calculated based on weight gain. The GMA grafted PLA was named as PLA-GMA

5.2.2.3 Fucntionalization of nanocellulose

Cellulose was functionalized with GMA by modifying the procedure used by Stenstad et al. [196]. Required amount of HNO3 was added in the nanocellulose

aqueous suspension containing 1.3 gram of nanocellulose to obtain exactly of 0.1M HNO3 concentration in the mixture. The cerium grafting reactions were

performed in dilute nanocellulose suspensions, typically 0.2% (w/w) in 0.1M HNO3.

Nitrogen was bubbled through the samples for 15 min prior to addition of ammonium cerium (IV) nitrate to a total end concentration of 2 mM. Then the temperature was increased to 35 °C while nitrogen was bubbled through the samples under stirring for another 15 min. Glycidyl methacrylate was gradually added over a period of 30 min and the reaction was continued for another 30 min. The amount of glycidyl methacrylate was 0.01 mole/g of nanocellulose. After the reaction, the sample was washed twice with distilled water and finally twice with acetone. Grafting was confirmed by FT-IR analysis and grafting percentage (63%) and grafting efficiency (41%) were calculated based on weight gain. The GMA grafted nanocellulose was named as NC-GMA.

5.2.2.4 Processing of nanocomposites

Fig. 5.3 shows the flow diagram for processing of PLA nanocomposites with NC. Nanocomposites were prepared following two different routes- (i) NC was first dispersed in PLA by solvent casting (organic) (masterbatch 1) and followed by melt mixed with PLA, (ii) NC was first dispersed in PVAc (aqueous) (masterbach 2) and followed by melt mixed with PLA.

Fig. 5.3 - Flow diagram for processing of PLA nanocomposites with NC.

Masterbatch 1:

Nanocellulose (or NC-GMA) aqueous suspension was solvent-exchanged to acetone by a series of centrifuging and re-dispersing steps, then different mixtures of PLA and nanofibres were prepared by mixing the acetone based nanofibres suspension with dissolved PLA (5 gram/150 ml solvent mixture) in acetone-chloroform 9:1 w/w mixture. Similarly, a masterbatch for PLA-g-GMA film with nanocellulose was prepared. The masterbatches were poured on flat plate, then solvent was evaporated and finally dried at 60°C under vacuum for 10 hours.

Masterbatch 2:

Nanocellulose aqueous suspension (0.4 g NC in 400 ml water) was mixed with previously prepared 100 ml of PVAc aqueous emulsion - in which PVAc content was about 2 gram (5 times of nanocellulose) - by stirring for 20 min at 40 °C. The masterbatches were poured on a flat plate, dried under aspirator and followed by immersed in water to remove water extractable components. Finally the materials were dried at 70°C under vacuum for 10 hours.

Melt mixing:

PLA nanocomposites were prepared by the addition of masterbatch 1 or masterbatch 2 into required amount of bulk PLA in Brabender plasti-corder internal mixer at 170 °C for 5 min setting the rotor speed at 60 rpm. The prepared samples are enlisted with their code in Table 5.2.

Table 5.2 list of samples with their code

Materials Compositions (wt.%) Sample code PLA/NC 99/1 NC1 PLA/NC 97/3 NC3 PLA/NC 95/5 NC5 PLA/PLA-g-GMA/NC 94/3/3 PLA-GMA3 PLA/NC-GMA 97/3 NC-GMA3 PLA/PVAc/NC 95/4/1 NC*1 PLA/PVAc/NC 80/17/3 NC*3 PLA/PVAc/NC 66/28/6 NC*6

5.3 Results and discussion

5.3.1. Characterization of NC, functionalized NC and PLA

Fig. 5.4 shows a TEM micrograph of a dilute suspension of NC obtained from microcrystalline cellulose. It can be seen that the nanocellulose rods appear mostly separated while in some places they are agglomerated. The distribution range is found to be wide but the majority of the overall size lies in the nanometric range. In the case of cellulose fibres, it has been reported by Samir et al. [27] that the sulfuric acid hydrolysis usually could cleave the amorphous region of microfibrils longitudinally, resulting in a diameter reduction of fibres from micron to nanometres. Fig. 5.5 shows the AFM amplitude image of a suspension of nanocellulose prepared by sulphuric acid hydrolysis. At least two modes were used to record data during scanning, one for the height image and the other for the amplitude image. The height image displays topographical detail by tracking the surface with the probe and the amplitude image gives the contrast between soft and hard polymer segments. It is obvious from the amplitude image (Fig. 5.5) that the cellulose obtained after acid hydrolysis under the present experimental set up contains particles and rods in the nano dimension. Fig. 5.6 shows diameter distribution of cellulose nanofibres. The average diameter was found to be 23.6 ± 7.34 nm.

Fig. 5.4 – Transmission electron micrograph from a dilute suspension of nanocellulose

Fig. 5.5 – Atomic force microscopy image of nanocellulose isolated from microcrystalline cellulose

Diameter (nm) 0 10 20 30 40 50 C o u n t (% ) 0 10 20 30 40 50 Experimental data Gaussian regression

Fig. 5.6 – Distribution of diameter of nanocellulose isolated from microcrystalline cellulose Wave number (cm-1) 1000 2000 3000 4000 T ra n s m it ta n c e Nanocellulose (NC) NC-GMA 1741 C=O 845 2933-CH22872 -CH Epoxy group

Fig. 5.7 – FT-IR spectra of nanocellulose and nanocellulose grafted GMA.

Fig. 5.7 shows the FT-IR spectra of NC and NC-GMA. The peak at 1741 cm-1 in the FTIR spectrum of NC-GMA in Fig. 5.7 indicates that a large amount of ester groups were introduced, indicating that polymer chains of GMA were grafted on NC. Further more, the peak at 845 cm-1 also confirms the presence of epoxy group

on the NC. Similar results have been reported by Stenstad et al [196] for nanocellulose functionalization with GMA.

Fig. 5.8 shows the FT-IR spectra of PLA, PLA-GMA and GMA. In the spectrum of PLA-GMA three peaks at 995, 908 and 845 cm-1 were appeared, which are also present in the spectrum of GMA, indicating the grafting of GMA onto PLA chain. Moreover, the peaks at 908 and 845 cm-1 show that epoxy groups were introduced on the PLA chains.

Wave number (cm-1) 700 800 900 1000 T ra n s m it ta n c e PLA PLA GMA GMA 995 908 845 Epoxy groups

Fig. 5.8 – FT-IR spectra of PLA, PLA-GMA and GMA

5.3.2. Characterization of nanocomposites 5.3.2.1 Morphology of PLA/NC nanocomposites

Fig. 5.9 shows the SEM images of fractured surface of the PLA nanocomposites with different NC content. The dispersion of nanocellulose into PLA and the fractured surface morphology of the nanocomposites were examined by SEM microscopy. In the case of NC1 and NC3 (prepared in organic solvent followed by melt mixing), it can be seen that the cellulose nanoparticles are homogeneously dispersed inside the PLA matrix, with limited formation of agglomerates. These agglomerates increased as NC content increased. Better dispersion of nanofibres into PLA was found in the case of PLA nanocomposites containing either functionalized components (PLA-GMA3 and NC-GMA3) or nanocellulose dispersed in PVAc (NC*3, prepared in aqueous media followed by melt mixing) as compared to NC3.

Fig. 5.9 – Scanning electron micrographs of PLA/NC nanocomposites (NC content, 1-3 wt.%)

5.3.2.2 DSC analysis

DSC thermal parameters of PLA and its nanocomposites are reported in Table 5.3. Fig. 5.10 and Fig. 5.11 represent the DSC first heating and second heating thermograms of PLA and its nanocomposites containing NC and PVAc with different compositions. Before analysis, PLA and nanocomposite samples were annealed at 90 °C overnight. DSC analysis showed that the phase behaviour and crystallinity of PLA in composites were affected by the components functionalization and nanocellulose content. The crystallinity of PLA was decreased by increasing fibre content and in presence of PVAc.

Temperature (°C) 0 20 40 60 80 100 120 140 160 H e a t fl o w e n d o u p ( W /g ) PLA PLA/PVAc/NC1 PLA/PVAc/NC3 PLA/PVAc/NC5 62.3 °C 59.3 °C 57.4 °C 44.0 °C

Fig. 5.10 - DSC 1st heating thermograms of PLA and its nanocomposites with NC dispersed in PVAc

Temperature (°C) 0 20 40 60 80 100 120 140 160 H e a t fl o w e n d o u p ( W /g ) PLA NC*1 NC*3 NC*5 58.2 °C 58.2 °C 55.8 °C 52.8 °C

Fig. 5.11 - DSC 2nd heating thermograms of PLA and its nanocomposites with NC dispersed in PVAc

PVAc content in composite (wt.%) 0 5 10 15 20 25 30 Tg ( °C ) 40 45 50 55 60 65 70 Tm ( °C ) 140 145 150 155 160 Tg1 Tg2 Tm1

Fig. 5.12 - Effect of PVAc content in composites on melting and glass transition temperature of PLA

Table 5.3 DSC thermal parameters of PLA and its nanocomposites

Sample Tg1 (°C) Tg2 (°C) Tcc2 (°C) Tm1 (°C) Crystallinity. (%) PLA 62.3 58.2 128.4 149.3 27.0 PVAc 28.6 39.7 - - - NC1 62.9 58.3 128.4 150..3 26.5 NC3 60.1 58.7 128.4 149.1 24.3 NC5 59.8 58.6 128.4 149.1 23.7 PLA-GMA3 58.5 58.7 - 147.9 22.4 NC-GMA3 60.7 58.7 - 149.1 23.8 NC*1 59.3 58.2 - 148.1 22.7 NC*3 57.4 55.8 - 144.1 22.6 NC*6 44.0 52.8 - 143.8 20.5

Fig. 5.12 shows the effect of PVAc content in composites on the glass transition temperature (Tg) and melting temperature (Tm)of PLA. In Fig. 5.12, it can be seen

that the glass transition temperature (Tg2) of nanocomposites follow a linear

relationship with PVAc concentration that can be attributed to the miscibility of PVAc with PLA. The miscibility of PLA and PVAc was reported by Gajria et al [147]. No cold crystallization peak temperature was appeared in the second heating run in the case of PLA nanocomposites containing PVAc, which probably due to the impedance of PVAc chain on the crystal growth of PLA.

5.3.2.3 Thermogravimetric analysis

Fig. 5.13 – TGA and DTG curves of PLA, PVAc, NC and their nanocomposites

Fig. 5.13 shows the TGA and DTG curves of PLA, NC, NC1, PVAc and NC*1 analyzed at 10 °C min-1 in nitrogen. In the Fig. 5.13, it is seen that nanocomposite of PLA with NC displayed higher thermal stability than NC. It is also seen that PLA nanocomposite (NC*1) in presence of PVAc displayed higher thermal stability (about 13 °C) than NC1. PVAc started to degrade at 218 °C showing maximum degradation rate at 337 °C. On the other hand PLA exhibited maximum degradation rate at 331.6 °C, suggesting that this PLA (PLA 3051D) has lower thermal stability than other (Hycail 1011), which diplayed maximum degradation rate at 352 °C. Generally PLA undergoes thermal degradation above 200 °C by hydrolysis, lactide reformation, oxidative main chain scission, and inter or intramolecular transesterfication reactions [165]. The cellulose nanocrystals also exhibited significantly different thermal behaviour than the original cellulose. The original cellulose showed the typical decomposition with onset temperature above 350 °C. The cellulose nanocrystals, on the other hand, showed more gradual thermal transitions that started at a lower temperature around 150 °C. The NC nanocrystals lost nearly 40 % of their mass in the 150-300 °C range showing maximum degradation rate at 279.5 °C, which is about 73 °C lower than the thermal degradation rate of microcrystalline cellulose fibres (352 °C). Similar thermal stability of NC was reported by Lu et al [184]. They mentioned that higher thermal conductivity of cellulose nanocrystals might be ascribed to smaller phononscattering in the bundle of crystallized cellulose chains in the cellulose nanocrystals than the amorphous random chains in cellulose.

5.3.2.4 Mechanical properties 0 1000 2000 3000 PLA NC1 NC3 NC5 E la s ti c m o d u lu s ( M P a ) 0 20 40 60 80 T e n s il e s tr e n g th ( M P a )

a

Fig. 5.14a – Elastic modulus (diagonal pattern) and tensile strength (solid bar) of PLA and its nanocomposites with different NC contents.

Tensile mechanical tests results of PLA and its nanocomposites with NC are reported in Table 5.4. Fig. 5.14a represents the elastic modulus and tensile strength of PLA and its nanocomposites with NC of different concentrations. It is

seen that the addition of cellulose nanofibres into PLA increased the elastic modulus of nanocomposites due to the higher modulus of NC [197]. It is also seen that the elastic modulus of NC3 and NC5 samples did not improve more with increasing the NC content. This is probably, due to agglomeration of NC at higher content into PLA, which was observed by SEM analysis (Fig. 5.9). Although, elastic modulus of nanocomposites (NC1, NC3 and NC5) was improved by the addition of NC, no improvement was found in tensile strength. To increase the dispersion of NC into PLA, both NC and PLA were functionalized by GMA and also a different approach - such as premixing of the aqueous emulsion of PVAc with the NC suspension followed by melt mixing with PLA – was used.

0 1000 2000 3000

PLA NC3 PLA-GMA3 NC-GMA3 NC*3

E las ti c mo d ul us ( M P a ) 0 20 40 60 80 T e n s ile s tr e ng th ( M P a )

b

Fig. 5.14b – Elastic modulus (diagonal pattern) and tensile strength (solid bar) of PLA and its nanocomposites with and without functionalization.

Table 5.4 Tensile properties of PLA and its nanocomposites

Sample Modulus (MPa) Strength (MPa) Elongation at break (%) PLA 2188 ± 348 48.08 ± 1.62 4.04 ± 0.77 NC1 2425 ± 168 48.82 ± 2.65 3.45 ± 0.14 NC3 2240 ± 69 47.83 ± 0.59 2.86 ± 0.32 NC5 2331 ± 270 46.07 ± 1.30 3.56 ± 0.28 PLA-GMA3 2417 ± 84 53.69 ± 2.19 3.66 ± 0.28 NC-GMA3 2547 ± 122 52.35 ± 1.77 2.91 ± 0.32 NC*1 2545 ± 61 54.28 ± 2.04 2.91 ± 0.06 NC*3 2621 ± 80 55.09 ± 0.99 2.36 ± 0.05

0 1000 2000 3000 PLA NC1 NC*1 E la s ti c mo d u lu s ( M P a ) 0 20 40 60 80 T e n s il e s tr e n g th ( MP a )

a

Fig. 5.15a – Elastic modulus (diagonal pattern) and tensile strength (solid bar) of PLA and its nanocomposites with and without PVAc.

0 1000 2000 3000 PLA NC*1 NC*3 NC*6 E la s ti c m o d u lu s ( M P a ) 0 20 40 60 80 T e n s il e s tr e n g th ( M P a )

b

Fig. 5.15b – Elastic modulus (diagonal pattern) and tensile strength (solid bar) of PLA and its nanocomposites in presence of PVAc.

Fig. 5.15a shows a comparison between two processing procedures such as masterbatch 1 and masterbatch 2 based on elastic modulus and tensile strength of nanocomposites with 1 wt.% of NC content. Higher values of elastic modulus and tensile strength of nanocomposites were found for masterbatch 2. It was also found that both elastic modulus and tensile strength were increased by the functionalization of components (NC-GMA3 and PLA-GMA3 in Fig. 5.14b). For the same concentration of NC (3 wt.%), higher values of elastic modulus and tensile strength were found for the masterbatch 2 (using PVAc aqueous emulsion

technique). This increment could be attributed to the better dispersion of NC in PVAc followed by in PLA. Further increase of NC using PVAc emulsion, the elastic modulus value of nanocomposite (NC*6) was not improved (Fig. 5.15b), which is probably due to the presence of agglomeration of NC at higher concentration.

5.4 Conclusion

The effect of different processing strategies of PLA/NC nanocomposites on the phase dispersion of NC as well as on the thermal and mechanical properties of these systems was investigated. For improving the dispersion of NC in PLA matrix, different approaches such as functionalization of PLA and NC by glycidyl methacrylate or dispersion of NC in poly(vinyl acetate) (PVAc/NC) were applied. SEM analysis revealed better dispersion of nanofibres for PLA nanocomposites containing PLA or NC functionalised with GMA or PVAc/NC system. The mechanical properties generally increased as a result of the addition of functionalized PLA or functionalized NC into the PLA matrix. Higher tensile modulus and strength were found for PLA nanocomposite with PVAc/NC. DSC analysis showed that the phase behaviour and crystallinityof PLA in composites was affected by the components functionalization and NC content. The crystallinity of PLA was decreased by increasing fibre content and in presence of PVAc. The Tg

and Tm of PLA in composites were also decreased by the addition of PVAc/NC.

Moreover, it was found that the nanocomposite of PLA with PVAc/NC exhibited higher thermal stability than other composites, as compared to NC and PLA single components.