Polymer nano-biocomposite based on poly (ethylene-co-methyl acrylate) (EMA)/cellulose nanocrystals (CNC); preparation and properties

Abstract In addition to being bio-degradable, cellulose nanocrystals (CNC) provide high mechanical strength to polymer composites. They exhibit outstanding properties like high strength-to-weight ratio, abundant surface functional groups and sustainability. In this study, we prepared nanocomposites of poly (ethylene-co-methyl acrylate) (EMA) and CNC using the melt-mixing technique. The mechanical properties of EMA were improved after adding CNC to it. The ultimate tensile stress and maximum elongation increased from 6 MPa and 860% (for pristine EMA) to 9.5 MPa and 1090% (for 2 wt.% EMA/CNC composite). Thermal properties of the EMA/CNC composites were analyzed by using DSC and TGA analysis. The dispersion of CNC in the composite was analyzed by SEM and XRD analyses. The EMA/CNC composites showed interesting birefringence properties on analysis via polarized optical microscopy (POM). The POM analysis indicated that the CNC in composites aligns themselves in the direction of applied stress, enhancing the mechanical strength of nanocomposites. Biodegradation of the bio-composite was also studied via soil-burial test. Graphical Abstract


Introduction
Polymer nanocomposites are an important class of materials that have improved mechanical, thermal, barrier and optical properties. In recent years, there has been significant interest in developing new polymer nano-biocomposites to make the earth greener and more sustainable. Cellulose is vital biofiller abundantly available in nature. Cellulose crystals or fibers having one dimension in the 100 nm range are considered as nanocellulose. [1] Cellulose nanocrystals (CNC) are highly crystalline, rod-like morphology with diameter and length in the range of 5-50 nm and 100-500 nm, respectively. [2] Generally, it is extracted by hydrolysis of cellulose with 6-8 M H 2 SO 4. By hydrolysis, the non-crystalline regions are hydrolyzed leaving behind crystalline parts having dimensions in nano-sizes. [3,4] Due to very high inter and intra-molecular hydrogen bonding, it is difficult to dissolve or disperse CNC in any organic solvent. In addition, CNC requires special handling owing to its hydrophilic nature and poor thermal stability. [5][6][7][8] Due to crystalline structure of CNC, after addition in polymer matrix it reduces the oxygen and water barrier properties which makes it suitable for food packaging applications. [9,10] Superabsorbent fibers of CNC are also studied as commercial acrylic acid based superabsorbent materials are petroleum-based products and non-bio degradable in nature. [11][12][13] Nanocellulose has extensively studied as a separator, electrolytes, binders, electrode materials, solar cells and substrate materials for the energy storage devices. [14][15][16] Other areas of applications include drug delivery, tissue scaffolding and sustainable reinforcing agents for polymer matrix. [17] Nanocellulose as a reinforcing agent in the polymer matrix is widely studied. With the incorporation of the nanocellulose in the polymer matrix, there is an increment in the tensile strength, young's modulus with a decrement in elongation at break with increment of filler loading. Being bio-degradable in nature, it also enhances the biodegradability of the product. [18,19] Generally, CNC is mixed with polymer either by melt mixing or during in-situ polymerization, as it is challenging to find a common solvent for both the polymer and nanocellulose due to very high internal hydrogen bonding in CNC. So, for thermoplastics, melt mixing is generally carried out using extruders or internal mixers. Water forms hydrogen bonding with CNC and makes a stable suspension in water. Therefore, watersoluble polymers are well explored with CNC by dissolving both in water and casted as a sheet. Most of the work on nanocellulose based composites is done by solution mixing. But for economic reasons and environmental safety, it is not so much feasible. Lee et al. synthesized the CNC using acid hydrolysis from commercially available microcrystalline cellulose (MCC). With the addition of 1 wt.% CNC into PVA they achieved 43% increment in tensile modulus and 49% increment into tensile strength as compared to pristine PVA. [20] Savadekar et al. prepared sheets of thermoplastic starch (TPS) and CNC via solvent casting method using water as a solvent. Even at a low CNC concentration (i.e., 0.4 wt. %), they achieved a 46% improvement in tensile strength with a reduction in water transmission rate and oxygen permeability. [21] Since several polymers or elastomers (like NR, SBR, NBR, etc.) are commercially available in the latex form, hence, CNC can be directly dispersed in the polymer latices. [22] Phomrak and coworkers added bacterial nanocellulose (BC) directly to NR latex and prepared composite films. They observed a remarkable increase in tensile strength as well as tensile modulus indicating the reinforcing effect of CNC. [23] EMA is a random copolymer of ethylene and methyl acrylate monomers. The incorporation of the methyl acrylate monomer into the polyethylene chain disturbs the symmetrical structure of polyethylene (PE) leading to a highly amorphous elastomeric polymer. Due to lower crystallinity, its tensile strength is lower but elongation at break is higher than PE. EMA remains stable at high temperatures making it suitable for making sheets, films and molding components using conventional thermoplastic machinery. Due to its high toughness, it is used commercially as a toughening agent in polyethylene terephthalate (PET), PE and PP based films. Other areas of applications of EMA are in the field of packaging, coatings, bio-medical and as a compatibilizer for immiscible blends. To the best of our knowledge, the preparation of EMA/CNC composites and CNC's effect on the mechanical properties of EMA has not been explored. In this study, we studied the preparation of EMA/CNC composites via the melt-mixing process and evaluation of mechanical, thermal and optical properties. To highlight the bio-degradability of CNC, the bio-degradation studies of EMA/CNC nanocomposites were also conducted using the soil-burial method.

Materials
A commercial grade of EMA (ElvaloyV R 1330) having 30% by weight methyl acrylate content, the density of 0.95 g/cm 3 and MFI (ASTM D1238) of 3 g/10 min was procured from DuPont, Mumbai. The mentioned EMA was used as a matrix. CNC powder of mean diameter < 100 nm and a specific gravity of 1.5 g/cm 3 with 98% purity was provided by ICAR-CIRCOT, Mumbai and used as a reinforcing bioadditive.

Nanocomposite preparation
For the preparation of nanocomposite, EMA and CNC were dried at 100 C for about 4 h to remove moisture and then cooled in a desiccator to room temperature. Both components were melt-mixed in a Haake-Rheomix internal mixer at 110 C and 60 rpm using cam-type rotors. [24][25][26][27][28][29] The mixing sequence adopted was melting of EMA for 2 min, then the CNC was added to the polymer melt, followed by mixing for an additional 6 min. The total mixing time of 8 min was kept constant for all compositions. The mixed hot lump was taken out from the internal mixer and sheeted-out through a two-roll mill. The sheeted-out composite was molded using a compression molding machine at 120 C and 5 bar pressure. [24,25,[30][31][32][33] A thickness of 2 mm was maintained for all nanocomposite sheets. The samples were designated as EMAX, where 'X' shows the loading level of CNC in wt. % as shown in Table 1.

Functional group analysis
For recording FT-IR spectra, Perkin Elmer-Spectrum-II was used in the ATR mode. The samples were scanned in a range of 600 cm À1 to 4000 cm À1 . Before taking FT-IR spectra, samples were pre-dried and cooled in a desiccator.

CNC dispersion
The dispersion of CNC in the matrix was studied with the help of Scanning Electron microscopy (SEM) and X-ray Diffraction (XRD) analysis. The surface of the different nanocomposite sheets was analyzed using a field emission scanning electron microscope (FESEM, JEOL JSM-7610F) under an accelerating voltage of 5 kV. All the samples were gold sputter-coated before the test to remove the effect of static charging. To check the effect of nanocellulose on the crystallinity of the matrix, X-ray studies were conducted using Panalytical X'Pert3 Powder instrument. These measurements were performed using Cu Ka radiation having a wavelength of 1.5406 Å with a step size of 0.02 and 2h scanning from 7 to 40 .

Thermo-mechanical analysis
Thermal transitions of nanocomposites were studied using a TA DSC Q20 instrument. About 5 mg sample was sealed in an aluminum pan and heated at a rate of 5 C/min from À 60 to þ100 C. The second heating cycle was considered for evaluation. TGA analysis of EMA, CNC and nanocomposites was carried out using TA TGA Q500 at a heating rate of 10 C/min from 25 C to 700 C in a nitrogen atmosphere.
The tensile properties (tensile strength, % elongation at break and modulus) of all the composite samples were determined by Hounsfield Universal Testing Machine. The test was performed according to the ASTM D412-98a at room temperature using dumbbell-shaped specimens with a cross-head speed of 500 mm/min. Tear strength was also determined according to the ASTM D624-00 (2012) at room temperature.
Dynamic Mechanical analysis was carried out using Gabo Eplexor 8 DMA using 50 Â 10 Â 2 mm samples. Temperature sweep was done in tensile mode from À80 C to 130 C at 1 Hz frequency with a temperature scanning rate of 2 C/min.

Polarized optical microscope (POM)
Distribution and orientation of CNC before and after stretching were determined by Olympus BX 51 under 10Â magnification. Film samples were placed on glass slides and then analyzed.

Bio-degradation studies
Bio-degradation test was carried out by the soil-burial method. [34][35][36][37][38][39][40][41] A potting soil was prepared by mixing garden soil and vermicompost in an equal weight ratio. This potting soil was then filled into rectangular containers with dimensions of 30 Â 25 Â 10 cm, as shown in Figure S1 in the supporting information. These containers were porous for proper air circulation. Triplicates of each composition film having dimensions of 50 Â 40 mm and an average thickness of 0.45 mm were used for testing. Before burying into the soil, the samples were first pre-conditioned at 50 C for 24 h. Then, they were cooled in a desiccator, and the initial weight of the samples was recorded. The films of different compositions were then buried into potting soil. Water was added during the period of the bio-degradation test to keep the soil moist. After the study period, samples were taken out, washed with water to remove the loosely bound soil, dried in an oven for 24 h at 50 C, and then cooled in a desiccator.
The weight of these degraded samples was taken, and % loss of the weight is calculated by using Equation 1: (1)

Results and discussion
Polymer-CNC composites were prepared using EMA as base polymer and taking different content of CNC as a bio-nanofiller via melt mixing method in Haake -Rheomix internal mixing at 110 C.

FT-IR analysis of EMA and EMA/CNC composites
FT-IR spectra of EMA and nanocomposites are shown in Figure 1. In EMA, due to the absorption of moisture, a broad absorption band at 3359 cm À1 observed due to different stretching vibration of -OH group. The absorption bands at 2919 cm À1 and 2851 cm À1 are due to the C-H asymmetric and symmetric stretching vibrations, respectively. Carbonyl group > C¼O shows an intense peak $1735 cm À1 due to ester functionality in EMA. Peaks located at 1462 cm À1 and 1373 cm À1 are for -CH 2 and -CH 3 bending. O-C bond stretching in methyl ester group can be seen at 1195 cm À1 accompanied by 1159 cm À1 stretching peak. The -CH 2 rocking vibration of the large chain polyethylene segment is located at 721 cm À1 . [42] In FT-IR spectra of CNC, a broad absorption band at 3333 cm À1 corresponds to the hydrogenbonded O-H group stretching, and the peak at 2888 cm À1 for the C-H stretching. Peaks at 1425 cm À1 , 1363 cm À1 and 1316 cm À1 are for the -CH 2 scissoring, -OH bending or C-H bending and C-H asymmetric deformation, respectively. The characteristic C-O-C stretching of glucosidic linkage of cellulose is present at 1161 cm À1 and 899 cm À1 . Strong peaks at 1060 cm À1 and 1028 cm À1 are responsible for the C-O stretching at C-3, C-C stretching and C-O stretching at C-6. [43,44] The peak corresonding to the ester functionality at $1735 cm À1 in EMA broadens with the addition of CNC. This broadening confirms the hydrogen bonding interactions between the EMA and CNC, which also contributes to the reinforcing effect of CNC in EMA. Figure 2 shows the FESEM images of EMA and EMA/CNC composites. Figure 2(a) shows the dimensions of CNC used as a reinforcing agent in the polymer matrix are in nanometers. Figure 2(a-c) are cross-sectional images of brittle fractures samples. Figure 2(c) indicates that CNC at 2% loading is well-dispersed in the EMA matrix. But at 10% loading, it undergoes agglomeration, as observed in Figure  2(d) which is encircled in red. XRD studies were carried out to study the morphological changes with the addition of CNC in EMA. Figure 3(a) shows the XRD pattern of EMA, CNC and nanocomposites.  110) and (400) lattice planes. [45] While EMA shows a broad hallow corresponding to the amorphous phase and small crystalline peaks at 2h ¼ 21.5 and 23.5. [46] In the XRD spectra of the composite EMA2 (i.e., with 2% CNC), there are no characteristic peaks for CNC, indicating a better dispersion of the CNC in the EMA matrix. As the amount of the CNC in composite is increased to 15% loading, the peak at 2h ¼ 21.5 was shifted toward the crystalline peak of CNC at 2h ¼ 22.3. [42] This indicates the agglomeration of CNC particles at higher filler loading. This agglomeration was also confirmed by the FESEM analysis (Figure 2), and evaluation of mechanical properties as explained afterwards.

Study of CNC dispersion via FESEM and XRD analysis
The effect on the XRD spectra of EMA2 after stretching can be seen in Figure 3(b). A 40 Â 25 Â 2 mm sample was stretched at a very slow rate of 2 mm/min to 400%  elongation and then the sample was relaxed for 1 h at room temperature, after that the XRD spectra of the stretched sample was recorded. The crystalline peak intensity of EMA sharply increased with the stretching of the sample. This showed that there is an increment in the crystallinity of the EMA due to the alignment of the CNC in stretching direction which leads to improved mechanical properties.

Mechanical properties
The mechanical properties of EMA and EMA/CNC composites were evaluated in terms of tensile and tear properties, as summarized in Table 1. The stress-strain plots of EMA and EMA/CNC composites at different CNC loading are shown in Figure 4. It indicates that the pristine EMA has an ultimate tensile stress of about 6 MPa and elongation at break of about 860%. On CNC loading, the tensile properties are improved. It indicates that there is an enhancement in both the tensile stress and elongation at the break with increase in CNC loading. Table 1 indicates that with 2% CNC loading, EMA/CNC nanocomposite showed maximum tensile stress of 9.5 MPa and maximum elongation of 1090%. But as the CNC content increases above 5%, its tensile properties started to decrease. This reduction is due to poor dispersion of CNC in EMA, leading to agglomeration which was corroborated by XRD and SEM analyses, as already discussed in Section 3.2. Compared to pristine EMA, the EMA/CNC composites showed a sharp increase in tensile stress, especially after about 600% elongation. It may be due to the change in orientation of CNC in the direction of applied stress which was visible in microscopic studies using polarized microscopy, as explained in the later section. Good polar-polar interaction between CNC and EMA also contributes to the reinforcing behavior of CNC. Table 1 also shows that the tear strength of EMA/CNC composite is maximum with 2% CNC content. However, with the CNC content above 5%, the tear strength gradually decreased, indicating the reduced interaction between EMA and CNC due to agglomeration of CNC.
Dynamic mechanical analysis is an important characterization technique to assess the viscoelastic behavior of composite under dynamic loading at different temperatures. The properties are expressed in the terms of storage modulus (E 0 ), loss modulus (Eʺ) and loss factor or tan d (ratio of loss modulus to storage modulus). Storage modulus express the elastic behavior of a viscoelastic material during the dynamic loading. It is the measure of the stiffness and elasticity of the material. Figure 5(a) shows the storage modulus behavior of composites with temperature. The E 0 values increased monotonically with increase in the CNC amount in EMA due to high stiffness and reinforcement behavior of CNC in the polymer matrix, the values are tabulated in Table 2. With increase in temperature, the E 0 values of EMA and all composites suddenly decreased at around À24 C which is a transition from glassy region to rubbery region due to increased polymer chain mobility in this region. After a rubber plateau, there is another sudden drop of E 0 around 97 C for melting of EMA. [47,48] Figure 5(b) shows the loss modulus (Eʺ) versus temperature behavior of EMA and its composites. Eʺ is the measure of energy lost or dissipated in dynamic loading. For all the compositions, the loss modulus increased with increase in temperature and value peaked at glass transition temperature of EMA, i.e., À24 C. This is due to increased chain movements at T g . Loss modulus peak gives a better representation of determining the T g . With CNC addition there is an increase in the loss modulus. The value of Eʺ at 25 C is highest for EMA15, which is due to poor dispersion at higher loading. [49,50] The tan d is the  ratio of loss modulus to the storage modulus or it is also called loss factor. The tan d determines how much a material is effective in damping. The tan d peaked at T g of EMA, i.e., À24 C and T m of EMA, i.e., 97 C. There is subtle change in the damping behavior of EMA with addition of CNC. At higher loading of CNC in EMA i.e., EMA15, the tan d value at 25 C is highest due to poor dispersion at higher loading of CNC which is also corroborated with mechanical properties, SEM and XRD analysis. [26]

Polarized optical microscopy (POM) analysis of EMA/CNC composites
We carried out polarized optical microscopy (POM) analysis of EMA/CNC composites. EMA2 sample was chosen for this analysis as it displayed the best mechanical properties due to the excellent CNC dispersion. POM images of EMA2 composite film before and after stretching are shown in Figure 6. Micro-domains of CNC are observed in the matrix, such relatively large micro-domains of aligned CNC have been reported for PVA-CNC composites. [51] Before stretching there was no birefringence observed under polarized light. After stretching the film, CNC got oriented in the direction of the stress. This anisotropic effect showed birefringence property under polarized light. As explained earlier, the CNC orientation in the stress direction leads to the improvement of the mechanical properties of composites. Such an orientation of CNC in polymer matrices in applied stress direction is well reported by several authors. [52][53][54][55][56][57][58][59][60] The orientation of the other nanoparticles such as silver and gold nanoparticles with stretching in PVA and PE matrix is also reported in literature. [61,62] 3.5. Thermal analysis of EMA and EMA/CNC composite TGA and DSC were used to study the thermal properties of EMA, CNC and nanocomposites. The TGA and DTG curves indicate that CNC shows three stages of weight loss (Figure 7). Initially, there is a weight loss at $100 C due to absorbed moisture evaporation, due to its hydrophilic nature. The second weight loss occurs at 200-260 C, corresponding to the degradation of cellulose low molecular weight chains. The prominent cellulose degradation peak corresponds to the glycosidic linkage breakage observed from 260-400 C with a T max of 328 C (as evident from DTG curve) and a weight loss of 60-65%. [63,64] In EMA, the degradation starts at 326 C and continuous to 480 C. The DTG curve showed a sharp peak at 452 C indicating the maximum degradation. This degradation occurs due to the homolytic cleavage of methoxy carbonyl  side group, followed by b scission in the polymer chain. [42] Table 3 shows that with the addition of CNC, there is a decrease in onset degradation temperature (T onset ) in all the EMA/CNC composites. This is due to the early degradation of CNC. Notably, the T max of the composite is unaltered w.r.t. to the pristine EMA. All the nanocomposites showed two-step degradation behavior; the former corresponds to the degradation of the CNC and the latter for the EMA.
T onset and T max for CNC, EMA and their composite are compiled in Table 3. Char content for EMA was almost negligible, but it was 14.11% for CNC. As the CNC content was increased in the EMA, the char content increased from 0.07% (EMA2) to 1.48% (EMA15). DSC traces for EMA and its nanocomposites are shown in Figure S2 and the values are tabulated in Table 3. Pure EMA has a T g at $ À34 C and T m at $84 C. [48,65] With the addition of the CNC, no significant changes in T g and T m are observed.

Bio-degradation studies of EMA and EMA/CNC composites
As explained in the experimental section, the bio-degradation of EMA and EMA/CNC nanocomposites was studied   via the soil-burial method. Figure 8 shows the visual appearance of the films before and after 2 months of bio-degradation time. As CNC is biodegradable, after 2 months, fungus growth can be seen with naked eyes in EMA10 and EMA15. Lighter and fewer spots are visible in EMA2 and EMA5 samples, due to lower concentration of CNC in the polymer matrix. There is a continuous loss in the weight of all the nanocomposite compositions with the increase in soil burial time ( Figure 9). The compositions having a high amount of CNC showed high bio-degradation behavior. EMA is a synthetic polymer and non-biodegradable in nature, so there was no substantial change in the weight after 2 months of degradation time. EMA15 containing 15% of the CNC showed the highest weight reduction.

Conclusion
A bio-nanocomposite based on EMA and CNC was prepared by melt-mixing technique The concentration of CNC was varied in EMA, and the nanocomposites properties were evaluated. At a 2 wt.% loading of CNC, nanocomposite showed the best mechanical properties with an enhancement of 55.75% and 14.19% in tensile and tear strength, respectively. At loading more than 5 wt.% of CNC, there is a decline in the mechanical properties due to the poor dispersion of CNC in EMA leads to agglomeration, as evident by SEM and XRD analysis. These agglomerates act as a weak point in the matrix and mechanical failures start from such accumulating points. Notably, the POM images of EMA/CNC composite showed birefringence property after stretching, indicating that CNC orientation happens in the direction of the applied stress. The CNC orientation also accounts for the increase in mechanical properties. The biodegradation of EMA/CNC nanocomposite was carried out using the soil-burial test method and was compared with the pristine EMA. It showed a gradual weight loss in the EMA/CNC composite with burial time, whereas, EMA had no weight loss. Interestingly, there was gain in weight loss in EMA/CNC composite with an increase in CNC loading. The prepared EMA/CNC nanocomposite can have several potential applications owing to improved mechanical strength, interesting optical properties and bio-degradability.