3.1 Preparation of alkali-treated LCNF and LCNF reinforced bioplastic
The lignin content in the raw lignocellulosic fibre from corn cob were first identified using the standard test method for acid-insoluble lignin D1106-96. The lignin content was found to be 18.86%, which is comparable to the previous literature data, where Takada, et al.  and Abdul Khalil, et al.  obtained 15.7% and 15% of lignin content, respectively. Since the amount of lignin was sufficient, soda process was used to prepare alkali-treated LCNF. To evaluate the effect of silane treatment, a portion of alkali-treated LCNF were treated with selected silane coupling agent, APS. Following that, a starch-based plain bioplastic and LCNF reinforced bioplastic contained different loading of LCNF (8 wt%, 10 wt% and 12 wt%) were prepared and subjected to characterization studies.
3.2 FT-IR analysis
FTIR technique was used to identify the functional groups of corn starch-based plain bioplastic and the APS-treated LCNF reinforced bioplastics. This is to evaluate the interactions between the components of silane coupling agent and the LCNF reinforced bioplastics
3.2.1 Corn starch-based plain bioplastic and LCNF reinforced bioplastics
The FT-IR spectra for corn starch-based plain bioplastic, 8wt%, 10wt% and 12wt% of LCNF reinforced bioplastics are presented in Fig 2. Broadly, both plain bioplastics and LCNF reinforced bioplastics showed similar FT-IR trends as they have the same matrix. For the spectrum of plain bioplastics, the broadband at 3265.15 cm-1 is due to the OH stretching, indicating an increase in the number of hydrogen bonds between the plastics components. On the other hand, the absorption band at 2929.68 cm-1 and 1334.38 cm-1 were also observed, suggesting the presence of C-H stretching (symmetric) and C-H deformation groups, respectively. The spectrum also shows a peak at 1647.48 cm-1 which corresponds to the C=O stretching group in acetic acid and the band of 1006.38 cm-1 that corresponds to the C-O bond stretching. The observed functional groups confirm the characteristic of starch-based plain bioplastics, as evidenced by Patnaik, et al. , Amin, et al.  and Domene-López, et al. .
Transmission peaks of 3421.69, 3175 and 3280 cm-1 were observed for 8 wt%, 10 wt% and 12 wt% of LCNF reinforced bioplastics, respectively. These are due to the presence of OH stretching. This is also due to the presence of phenols of lignin and hydroxyl group of cellulose and hemicellulose from the fibres . Peaks observed at 3063.97, 2922.23 and 2922.00 cm-1 for 8 wt%, 10 wt%, and 12 wt% of LCNF reinforced bioplastics, respectively are corresponds to the symmetrical C-H stretching/vibration of CH and CH2 in cellulose and hemicellulose from the added fibres . Additionally, peaks observed at 1505.84, 1364.20 and 1364.22 cm-1 for 8 wt%, 10 wt% and 12 wt% of LCNF reinforced bioplastics, respectively are relate to the C-H stretching (deformation), which due to the CH2 stretching of hemicellulose. The peaks at 1796.57, 1654.00 and 1654.94 cm-1 for 8 wt%, 10 wt% and 12 wt% of LCNF reinforced bioplastics, respectively are attributed the C=O group which can be explained by the presence of lignin [34, 35]. Moreover, the transmission peaks at 1155.40, 1006.38 and 998.93 cm-1 for 8 wt%, 10 wt%, and 12 wt% of LCNF reinforced bioplastics respectively show the presence of C-O bond in the composite. This confirms the presence of acetyl group in the lignin and hemicellulose from the LCNF .
3.2.2 APS-treated LCNF reinforced bioplastics
FT-IR was also used to identify the chemical functional groups of APS-treated LCNF reinforced bioplastics at different LCNF loading, namely LCNF8/APS10, LCNF10/APS10, LCNF12/APS10. Fig 3 shows the spectrum for 10 wt% APS-treated LCNF reinforced bioplastics at different fibre adding ratios. The peaks at 3250.24, 3272.60 and 2922.00 cm-1 for LCNF8/APS10, LCNF10/APS10, LCNF12/APS10, respectively are due to the presence of O-H stretching group. Difference of band between the untreated and APS-treated bioplastics can be observed, specifically in the case when higher LCNF amounts were added, the difference is mainly due to an increase of hydrogen bonding interactions between the APS reagent and hydroxyl groups of the LCNF components . Small peaks at 2937.14, 2929.69 and 2922.33 cm-1 for LCNF8/APS10, LCNF10/APS10, LCNF12/APS10, respectively are attributed to the presence of C-H stretching (symmetric). The peaks at 1654.93, 1640.03 and 1647.48 cm-1 for LCNF8/APS10, LCNF10/APS10, LCNF12/APS10, respectively are attributed to the C=O bond. On the other hand, the presence of C-H stretching (deformation) was represented by the peaks at 1364.21, 1349.29 and 1341.84 cm-1 for LCNF8/APS10, LCNF10/APS10, LCNF12/APS10,respectively, confirming its link with the fibre’s surface . Also, the absorption peak intensity of C-O at the wavenumbers of 998.92, 991.47 and 998.92 cm-1 for LCNF8/APS10, LCNF10/APS10, LCNF12/APS10, respectively can be noticed.
The effects of silane treatment can also be observed by the formation of Si-O-Si bonds or SiO string groups, Si-O symmetric group, Si-C bonds and N-H bending group, as depicted as Fig 4. In all APS-treated sample, weak peak at 1080 cm-1 which represents the asymmetric stretching of Si-O-Si bonds or SiO string groups (siloxane) can be observed. In addition, peak at wavelength around 760 cm-1 and 857 cm-1 are related to the Si-O symmetric group and Si-C bonds . The presence of N-H bending group also resulted in the appearance of peak at around 1543.12, 1543.00 and 1565.48 for LCNF8/APS10, LCNF10/APS10, LCNF12/APS10, respectively. The formation of these groups confirms the reactions between the hydroxyl groups of LCNF components and the hydrolyzed silane group of APS reagent.
3.3 Mechanical strength test
3.3.1 Ultimate tensile strength
The tensile properties determination was performed referring to ASTM D882 standard . Fig 5 shows the tensile strength of the corn starch-based plain bioplastic, LCNF reinforced bioplastics with 8wt%, 10wt% and 12wt% fibre loading and APS-treated LCNF reinforced bioplastics.
From the results, it was observed that the ultimate tensile strength of the composite increases with LCNF fibre loading, where a 12wt% of LCNF loading results in a mean tensile strength of 2.983 MPa, which represents a 108% of increment of tensile strength compared to the plain bioplastics. The enhancement is possible as the bioplastic components are from same biological origin, and this results in high structural compatibility between corn starch and lignin-containing cellulose fibre that derived from the corn cobs . Additionally, good dispersion of LCNF in corn starch-based bioplastic matrix also promotes the interfacial interaction between the matrix and reinforcement filler which significantly improve the stress transfer . In our case, ultrasonic homogenization was used to promote the dispersion of LCNF in the corn starch-based bioplastic matrix [39, 40]. The ultrasonic homogenization allows good dispersion and it also helps to credit in high ultimate tensile strength.
Interestingly, we also studied the effects of APS concentration on the mechanical strength and the results are depicted as Fig 6. It was noticed that the concentration of APS and fibre loading play an important role in determining the mechanical strength of LCNF reinforced bioplastics. From Fig 6, it was observed that the mechanical strength of the high LCNF loaded bioplastics was dramatically weakened when the concentration of APS increased, where for the case of 12 wt% LCNF loading, the tensile strength reduced as the concentration increased. It is also important to point out that a lower concentration of APS is more preferable in all the cases, where for 8wt% LCNF loading, the bioplastic treated with lower concentration (10 wt% APS) demonstrated an improved mechanical strength for about 12.1%. For the case of 10 wt% LCNF loading, the bioplastic treated with a 10 wt% APS concentration showed an enhancement of 0.7% in tensile strength. Therefore, it can be deduced that silane treatment is more effective for low fibre loading composite. This is reasonable as a higher concentration of fibre loading has higher tendency to induce agglomeration of fibres. Our finding was supported by a recent study reported by Chotiprayon, et al. , where they also found that at higher concentration of fibre loading in the biocomposite matrix exhibited slightly decreased tensile strength.
Even so, it is important to point out that all the 10 wt% APS-treated LCNF reinforced bioplastics showed better tensile strength as compared to untreated corn starch-based plan bioplastics. The enhancement of tensile strength is mainly due to the removal of hemicellulose from the LCNF surface, which leads to improvement of the crystallinity index of the bioplastic composite . Moreover, the silane treatment also reduced the polarity by forming siloxane (Si-O-Si bonds) on the fibre surface which facilitate the formation of stable covalent bonds between fibre and the bioplastic matrix and makes load transfer efficient .
3.3.2 Elongation at break
Elongation at break indicates the flexibility and stretching ability of the bioplastics. It is the ratio between the elongation and initial length. The elongation at break for plain corn starch-based bioplastic, LCNF reinforced bioplastics and APS-treated LCNF reinforced bioplastics are presented in Fig 7. As shown in Fig 7, LCNF reinforced bioplastic showed lower elongation at break as compared to the plain bioplastic. The reduction is possible due to the increment of compactness of the bioplastic structure. It is known that the presence of lignin and cellulose in LCNF allows more interactions between the molecules in the bioplastic. In that case, the restructure of the bioplastic matrix network results in the reduction of flexibility of bioplastic as the chain movement was greatly hindered . This finding was supported by Pelissari, et al. , who reported the elongation at break of cellulose nanofibres (from banana peel) reinforced banana starch plastics. Likely, they also found that the plain bioplastic demonstrated greater elongation at break as compared to the LCNF reinforced composite.
3.4 Thermal gravimetric analysis
3.4.1 LCNF reinforced bioplastic
Through the above analysis, the 10 wt% APS-treated LCNF reinforced bioplastics showed excellent mechanical properties. In this section, we emphasize on the analysis of the thermal stability of the untreated and APS-treated LCNF reinforced bioplastics. The TGA curves for corn starch-based plain bioplastic and LCNF reinforced bioplastics are presented as Fig 8. As shown in Fig 8, three thermal events occur during degradation. Each event correlates with the degradation temperature as well as the weight loss in the TGA curve. The first thermal event (weight loss) occurred from ambient temperature to around 290 . This results in a weight loss of around 28%. A slight and gradual weight loss occurred from 80 which corresponds to the water dehydration, elimination of some low molar weight components and volatilization of fructose fragments in the composite . The degradation was more obvious in plain corn starch-based bioplastic as compared to LCNF reinforced bioplastics. This attributes to the presence of lignin in the LCNF in reducing the hydrophilicity of the fibril surface and the plastic matrix. The degradation curve for LCNF reinforced bioplastics was relatively flatten during this thermal event, indicating an improved thermal stability property.
After this stage, a decomposition stage from 290 to 330 occurred and contributed to the sharp weight loss of approximately 40%. This stage is known as depolymerization. It attributes to starch and glycerol decomposition due to breaking bonds of hydroxyl groups and starch carbon chains . Interestingly, it can be seen that the degradation process was slower when higher fibre is loaded in the bioplastics. This indicates that fibres helps to reduce the weight loss of the bioplastic, which is possible by strengthening the carbon chains of the matrix . The strengthening effect is credited to the presence of lignin as crosslinkers between the cellulose content in LCNF and starch-based plastic. It connects the cellulose through hydrogen bonds and dipole-dipole interactions while being linked to the plastic matrix through van der Waals and hydrogen bonds . Thus, it can be said that at higher fibre loading, more bonding between the fibre and bioplastic component can happen and as a result, the bioplastic exhibits stable thermal behaviours.
For LCNF reinforced bioplastics, the last stage corresponds to the additional degradation of lignocellulosic component of LCNF from corn cob, namely the cellulose, hemicellulose and lignin. A smooth degradation curve was observed at the temperature range from 320 to 360 , which can attributed to the degradation of cellulose and hemicellulose under the temperature of 300 to 450 , agreed by Liu, et al. , and degradation of lignin which generally occurred at the temperature range of 270 to 370 . Overall, the TGA analysis confirm that the addition of LCNF as reinforcement filler improved the thermal stability of bioplastic.
3.4.2 APS-treated LCNF reinforced bioplastics
The effect of silane treatment on the thermal behaviour of bioplastics was also evaluated. The TGA curve for the 10 wt% APS-treated LCNF reinforced bioplastic was presented as Fig 9. The first thermal event for APS-treated LCNF reinforced bioplastic occurred from ambient temperature to around 285 , inducing preliminary weight loss. Second stage occurred at around 320 and third stage occurred at around 340 . Similar trend was observed as the LCNF reinforced bioplastics, shown in Fig 8. At 150 , the APS-treated LCNF reinforced bioplastic tend to retain more residues as compared to corn starch-based plain bioplastic, indicating that the LCNF reinforced bioplastic showed a greater thermal stability at relatively low heating temperature. This could be attributed to the formation of siloxane layer on the surface of the fibres that form a linkage to the hydroxyl group in the bioplastic matrix which helps to reduce the vaporization of moisture and fructose components in the bioplastic composite . However, it is worth highlighting that the APS-treated LCNF reinforced bioplastic could not retain as much residues at the temperature range of 200 to 300 as compared to untreated LCNF reinforced plastic and plain bioplastic. The possible reason may be due to the excessive hydroxyl group crosslinked network which induced greater degradation at its favourable degradation temperature. In addition, large removal of non-cellulosic components of the fibres through silane treatment may also significantly affect its thermal behaviours [47, 50]. For 300 and above, the APS-treated LCNF bioplastics were able to retain more residues, where both the APS-treated LCNF reinforced bioplastics, at 10 wt% and 12 wt% LCNF loading were able to retain 8% more residues as compared to the corn starch-based plain bioplastics, suggesting silane treatment could improve the thermal stability of bioplastic at high temperature.
3.5 Opacity analysis
Opacity result of the bioplastic could provide information on the size of added materials dispersed within the bioplastic and could be an indicator of the number of pores within the bioplastic . If the dispersed fibres in the bioplastic has a greater particle size, it will fill up the pores of starch matrix, reduce the light pathway and results in a high opacity . The opacity of corn starch-based plain bioplastic, LCNF reinforced bioplastics and APS-treated LCNF reinforced bioplastic are shown in Table 1. By referring to Table 1, it can be seen that the addition of LCNF in the bioplastic increased the opacity to 1.39, 1.42 to 1.54 for 8 wt%, 10 wt% and 12 wt% LCNF reinforced bioplastics, respectively. Similar trend was observed for APS-treated LCNF reinforced bioplastics, where the opacity was incresed to 1.45, 1.46 and 1.47 for APS-treated 8 wt%, 10 wt% and 12 wt% LCNF reinforced bioplastics, respectively. The increased opacity is attributed to the compactness of LCNF reinforced bioplastic that induced by the formation of cross-linked intermolecular bonding between the starch chains within the fibre. This compactness would cause a reduction in the pathway for light transmission through the bioplastic, and this lead to an increase in the opacity. In addition, it was observed that all the LCNF reinforced bioplastics were in yellowish color and the intenstity of colour increased when higher amounts of fibre was added. The yellowish color may also influence the opacity .