Effect of ball milling
Ball milling is low cost and environmental friendly, and is used worldwide in industrial scale. Planetary ball milling is a powerful and reliable method for reducing the size of fillers to the nanoscale. This mechanism is largely based on centrifugal accelerating force rather than gravitational accelerating force, resulting in a short time requirement for the process [34]. Moreover, several others parameters such as time of each cycle, total number of cycles, pause time between each cycles, milling balls size speed (RPM), material to ball weight ratio, nature of the fillers etc. can affect the final properties of the fillers. Therefore, understanding overall procedure of ball milling is complex and challenging in order to get desired shape and size of the fillers. This conditions (15 minutes’ cycles—15 minutes’ pause time between each cycle—total 99 cycles and 300 RPM) have been chosen after extensive trials of the experiment (data not shown).
Morphological studies
The morphology of nanofillers is illustrated in Figure 2 by TEM examination. Those nanoparticles revealed in a heterogeneous mix of quadrilateral and irregular shapes. This evidenced the ball milling treatment had effectively disintegrated the micro-fibers into varying nano-sized feature. Apparently, the pack-like structure of quadrilateral-shaped particle may impart them with highly preserved rigidity for advantageous application in reinforcing composite materials. In size measurement from TEM image, they presented a ranging size of about 30-110 nm in both width and length dimensions. It was similar to the findings from Saba et al. [34], who had obtained a nanofiller size up to 100 nm from oil palm empty fruit bunch fibers in a previous study, by treating the ground fiber material with bromine water and tin chloride solution, followed by cryocrushing and ball milling processes. As from Zetasizer analysis (Figure S2), those particles exhibited relatively consistent size at around 300 nm. Despite this size is far larger than the size shown in TEM image, the results is still considerably valid for Zetasizer analysis since this equipment is more accurately used in measuring round shape particles rather than nanofillers particles with diverse shapes. Another explanation was likely due to the poor dispersion of nanofillers in the aqueous solution and ultimately resulted in the agglomeration into larger particles. This was in agreement with the Nanoplus analysis results (Figure S3), where the particles showed three distinct sizes at 81.3 nm, 159.0 nm, and 850.2 nm, respectively, indicating the fillers have a diameter of 81.3-159.0 nm, while with a length at around 850.2 nm approaching micron meter.
As viewed by SEM microscopy (Figure 3), the microfibers are presented as scattering particles, as a result of the deposition on carbon tape through ethanol solvent evaporation process. From surface morphology, the particles revealed a smooth texture, which promoted by the extensive impact and shear forces generated from mechanical ball milling. Meanwhile, the microfibers also showed random shape feature that could aid in promoting its fire retarding behavior as filler in composite materials due to non-uniform transmission/conduction of heat at each point in time [34,35]. Moreover, a few microns in diameter and while tens of micrometer in length was displayed by the microfibers, suggesting the fiber is suitable acting as a micro-filler in composite matrix fabrication. For elemental analysis (Figure 4), microfibers revealed carbon and oxygen as their major compositional elements, implying the fiber is mainly composed of hydrocarbon compounds. In addition, the traces for magnesium, silica, chloride, calcium and iron, indicating the typical residual elements for a plant biomass fiber [36].
Thermal analysis
The thermo-stability of the samples is evaluated by TGA and DSC curves as depicted in Figure 5 and Figure 6, respectively. From TGA curve (Figure 5), four distinct phases were revealed by both samples, i.e. moisture evaporation, hemicellulose decarboxylation, cellulose decomposition, and lignin pyrolysis. At beginning, both samples showed weight loss in 70-130°C temperature regions, in corresponding to the evaporated moisture content. However, the microfibers lost more weight at this region when comparing to nanofillers. It was probably caused by the greater free volume in microfibers that had higher affinity in entrapping water molecules. In subsequent weight loss stage at 250-350°C, it involved the decomposition process for hemicellulose and cellulose. This was attributing to their closely similar structural feature built-up with adjacent intermolecular linkage via physically twisting and hydrogen bonding [37]. Additionally, the microfibers exhibited slightly higher decomposition temperature at 263.2°C, when compared to nanofillers at 260.8°C temperature. It was possibly contributed by the lower crystallinity as well as the smaller nanoscale size of nanofillers that induced it for earlier thermal degradation. Between 350-500°C, the liquefaction and gasification process happened for lignin since this compound had strong fire retardant behavior. At final phase, microfibers sample showed constant weight beyond 550°C, whereas nanofillers continued to lose weight and only presented constant weight after 620°C. This evidenced the thermal degradation is relatively consistent in nanofillers, which may endow it with tunable properties for high temperature fabrication process.
As for DSC analysis (Figure 6), both samples showed broad endotherms extending from 70°C to 140°C temperature, which basically correlated to the water vaporization process [38]. The enthalpy heat gained by microfibers at this temperature range was 65.09 J/g, which remarkably larger than the nanofillers sample with 14.82 J/g. This showcased the microfibers required higher heat energy for evaporating larger water content that in line with the weight loss shown in TGA curve. Besides this, the nanofillers revealed second wide endotherm at around 230°C with 25.33 J/g enthalpy heat, in correlating to the energy required to decompose cellulose-based compounds. But for microfibers, it was presented as horizontal curve rather than in endothermic band. This was possibly due to the ball milling effect that had gradually decreased the fiber compactness of nanofillers after numerous cycles of treatment and subsequently reduced its thermal resistance in withstanding high temperature. Beyond this point, both curves rising up to likely form an exothermic band for heat release in order to break down the bonds of cellulose and hemicellulose components. Thus, the analyzed DSC results herein agreed well with the TGA curves.
X-ray diffraction (XRD) analysis
Figure 7a and 7b shows the XRD spectra for microfibers and nanofillers, respectively. Both samples exhibited their main crystalline peaks at 16.8°, 22.6° and 34.9°, which corresponding to the (110), (200), and (004) planes crystallography. It was in response to the distinctive lignocellulosic material with Iβ native cellulose structure. Meanwhile, the peak at 22.6° was observed broadening for nanofillers and had somehow overlapping the peak at 16.8° when comparing to microfibers sample. This indicated the crystals domain in nanofillers sample was decreased and the fiber bonding structure became homogeneous possibly resulted by the amorphization effect of ball milling [34,39]. From crystallinity measurement, the microfibers gave a crystallinity index of 71.8%, which slightly higher than the crystallinity of nanofillers with 68.9%. Nonetheless, the obtained crystallinity in this work is considerably high for nanofillers fiber (68.9%), evidencing its highly crystalline structure is still preserved although undergoing harsh mechanical grinding process. It would endow it with great rigidity for acting as load-bearing agent in material reinforcement application.